Lovejoy Industries

1085 309 NLRB No. 173 LOVEJOY INDUSTRIES 1 The Respondent contends in its answering brief that the General Counsel has failed ‘‘to comply with the Board’s Rules and Regula- tions regarding the filing of exceptions.’’ Contrary to the Respond- ent’s assertion we find that the General Counsel’s exceptions sub- stantially comply with Sec. 102.46(b) of the Board’s Rules and Reg- ulations. 2 The General Counsel and the Respondent have excepted to some of the judge’s credibility findings. The Board’s established policy is not to overrule an administrative law judge’s credibility resolutions unless the clear preponderance of all the relevant evidence convinces us that they are incorrect. Standard Dry Wall Products, 91 NLRB 544 (1950), enfd. 188 F.2d 362 (3d Cir. 1951). We have carefully examined the record and find no basis for reversing the findings. 3 Member Oviatt agrees that the warning and suspension of em- ployee Valenzuela for low production on October 24, 1988, violated Sec. 8(a)(5) of the Act, but only because the Respondent departed from its past practice by relying on a 10-minute observation of Valenzuela to establish the production standard for disciplinary pur- poses. Member Devaney finds it unnecessary to pass on whether the March 17, 1988 remarks of Plant Manager Peter Balma regarding a wage increase violated Sec. 8(a)(1) of the Act, in view of the Board’s adoption of the judge’s findings that the April 14 and May 2, 1988 remarks of President Robert Novak regarding a wage in- crease violated Sec. 8(a)(1). He also finds it unnecessary to pass on the 8(a)(3) findings regarding the November 11, 1987 low produc- tion warnings to employees Ramirez and Olivares and the October 24, 1988 cafeteria warnings of employees Aguirre, Banales, Canales, Garcia, Mombela, Paz, and Valenzuela, as these findings are cumu- lative of other unfair labor practices found by the Board and would not materially affect the remedy. 4 We adopt the judge’s recommended dismissal of the complaint allegation that Plant Manager Balma’s June 16, 1988 warning to Valenzuela violated the Act. The dismissal of this complaint allega- tion does not affect our decision regarding whether the T-shirt alle- gation is closely related to the warning allegation. 5 Contrary to his colleagues, Member Devaney would affirm the judge’s finding that the complaint allegation regarding the T-shirt in- cident is not barred by Sec. 10(b) of the Act. In his view, the ‘‘closely related’’ test of Redd-I, Inc., 290 NLRB 1115 (1988), and Nickles Bakery of Indiana, 296 NLRB 927 (1989), has clearly been met, and this 8(a)(1) allegation is properly before the Board for de- termination on the merits. In particular, Member Devaney does not agree with the majority that Southwest Distributing Co., 301 NLRB 954 (1991), is distin- guishable from the present situation. First, in this case, as in South- west, supra, the Respondent assertedly engaged in an unlawful cam- Continued Acme Die Casting, a Division of Lovejoy Industries Incorporated and United Electrical, Radio & Machine Workers of America. Cases 13–CA– 27619, 13–CA–27788, 13–CA–27941, 13–CA– 28033, and 13–CA–28118 December 16, 1992 DECISION AND ORDER BY MEMBERS DEVANEY, OVIATT, AND RAUDABAUGH On May 23, 1991, Administrative Law Judge Rich- ard J. Linton issued the attached decision. The Re- spondent filed exceptions and a supporting brief, and the General Counsel filed exceptions, a supporting brief, and an answering brief. The Respondent filed an answering brief to the General Counsel’s exceptions.1 The National Labor Relations Board has delegated its authority in this proceeding to a three-member panel. The Board has considered the decision and the record in light of the exceptions and briefs and has de- cided to affirm the judge’s rulings, findings,2 and con- clusions as modified and to adopt the recommended Order as modified.3 1. The judge found that the Respondent permitted employees to wear nonunion T-shirts in hot weather without wearing a long-sleeved uniform shirt, although requiring the long-sleeved uniform shirt to be worn with union T-shirts. The judge concluded that this dis- parity of treatment violated Section 8(a)(1) of the Act. The Respondent excepts, arguing that the complaint al- legation regarding the T-shirts must be dismissed as untimely under Section 10(b) of the Act because it was neither contained in the charge nor closely related, fac- tually or legally, to any allegation in the charge. The charge filed in Case 13–CA–27941, the charge at issue, dated August 15, 1988, alleges, inter alia, ‘‘In or around June of 1988 the employer issued written warnings to Juan Avila, Nelson Diaz, and Jorge Nico- las Valenzuela.’’ The complaint in Case 13–CA– 27941, dated September 23, 1988, contained, inter alia, these two allegations: (1) In violation of Section 8(a)(1) and (3), ‘‘[O]n or about June 16, 1988 Re- spondent issued a disciplinary warning to its employee, Nicolas Jorge Valenzuela’’;4 and (2) In violation of Section 8(a)(1) ‘‘In or about early June 1988 . . . Re- spondent acting through Pete Balma, at Respondent’s facility, told its employees that they would not be per- mitted to wear tee shirts with union insignia during working hours, while it allowed said employees to wear similar garments without union insignia.’’ The evidence established that on June 8, 1988, em- ployee Nicolas Jorge Valenzuela distributed T-shirts with the Union’s emblem to about 60 employees. Later that morning Valenzuela was told by Plant Manager Balma that the union T-shirt, which Valenzuela was wearing, had to be worn either over or under the uni- form shirt. The judge found that the allegation regarding the T- shirt incident was closely related, factually and legally, to the June 16, 1988 warning. He reasoned that the T- shirt incident could show animus by Balma against Valenzuela’s union activities. Considering the allega- tion on its merits, the judge concluded that by requir- ing the uniform shirt to be worn with the union T- shirts but allowing other T-shirts to be worn without the uniform shirt the Respondent had violated Section 8(a)(1). For the reasons that follow we find that the charge allegation regarding the written warning for low production and the complaint allegation regarding the T-shirt incident are not closely related.5 1086 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD paign to undermine employee support for the Union; here, the Re- spondent hoped to eventually rid itself of the Union after the 1-year certification period expired. The Respondent’s campaign allegedly occurred generally after the Union’s election victory in October 1987 and continued through November 1988 while the Respondent was refusing to bargain with the Union. Second, in Southwest, the Board also considered that the additional 8(a)(1) complaint allegations arose from the same factual situation or sequence of events as the allegations in the timely pending charges. Member Devaney would treat the T-shirt incident allegation in the same manner that the Board treated the additional 8(a)(1) allegations in Southwest. Here, the General Counsel’s theory for employee Valenzuela’s June 16, 1988 low production warning was pretext, i.e., discipline based on his union activities, including his earlier involvement in the T-shirt incident which had occurred during the same time period and while his production levels were under scrutiny. Contrary to his colleagues, and consistent with Southwest (see 301 NLRB at 956 fn. 7), Member Devaney notes that the fact that the timely 8(a)(3) charge allegation and the otherwise untimely 8(a)(1) complaint allegation invoke dif- ferent sections of the Act does not preclude a finding that they are based on essentially similar legal theories. Finally, Member Devaney further observes that the T-shirt incident could show union animus with respect to the alleged 8(a)(3) involving Valenzuela, a factor also considered important by the Board in Southwest. 6 296 NLRB 927 (1989). 7 Cf. Well-Bred Loaf, Inc., 303 NLRB 1016 (1991), (charge allega- tions of unlawful discrimination were closely related to complaint al- legations of impression of surveillance and a warning to an em- ployee where all of the allegations ‘‘occurred within the same gen- eral time period’’ and was part of ‘‘an overall plan to resist the Union’’); Southwest Distributing Co., supra (discharges alleged in the charge and discharges alleged in the complaint to have violated Sec. 8(a)(3) and (1) were closely related as they were all based on the same legal theory and the same factual situation, and occurring during the 4-month period after the election and for the alleged pur- pose of ridding the employer of union supporters). 8 Contrary to the judge, we conclude that a complaint allegation is not closely related to a charge allegation merely because the com- plaint allegation may show animus to bolster the charge allegation. When the complaint in this case was issued Board precedent allowed the issuance of complaint allega- tions of independent violations of Section 8(a)(1) based solely on the pre-printed ‘‘other acts’’ language on the charge form. The Board subsequently held in Nickles Bakery of Indiana,6 that the preprinted ‘‘other acts’’ language on the charge form may not be relied on to support complaint allegations of independent vio- lations of Section 8(a)(1). Rather, there must be a showing of factual relatedness between the specific al- legations of the charge and the complaint’s 8(a)(1) al- legations. To determine whether a charge adequately supports a complaint allegation the Board considers: (1) whether the charge and complaint allegations in- volves the same legal theory, and (2) whether they arise from the same factual circumstances. The Board also may look at whether a respondent would raise similar defenses to both allegations. Applying the principles of Nickles to this case, we find initially that the General Counsel has not estab- lished a factual nexus between the allegation of the charge regarding the Respondent’s written warnings to employees Avila, Diaz, and Valenzuela, and the com- plaint’s allegation regarding the T-shirt incident. There is no evidence or contention that the T-shirt incident played any role in the Respondent’s decision to issue the written warnings. The allegations also involve separate legal theories. Thus, the allegation regarding the T-shirt incident is based on an 8(a)(1) theory of disparate treatment but the allegation regarding the written warnings is based on an 8(a)(3) theory of unlawful motivation. Further, the General Counsel does not argue that the allegations are connected, either factually or legally, as a part of an overall plan by the Respondent to resist or under- mine the Union.7 Finally, in light of the General Counsel’s failure to establish the requisite link between the charge allega- tion regarding the written warnings and the complaint allegation regarding the T-shirt incident,8 we cannot find that the Respondent would raise similar defenses to these allegations. See Nippondenso Mfg, U.S.A., 299 NLRB 545 (1990). We therefore find that the complaint allegation is not closely related to the charge allegation. The com- plaint allegation that the Respondent violated Section 8(a)(1) regarding the T-shirt incident is accordingly dismissed. 2. The judge found, and we agree, that the Respond- ent violated Section 8(a)(3) and (5) of the Act by fail- ing to give raises in 1988. The judge further found that raises of 25 cents an hour would have been given in February and July 1988. The Respondent excepts. We affirm the judge’s findings of the violations and his findings that raises would have been given in February and July 1988. In our view, however, the evidence pre- sented at trial was insufficient to determine the appro- priate amounts of these wage increases. We therefore find that a determination of the amounts of the Feb- ruary and July 1988 wage increases should be left to the compliance stage of this proceeding. ORDER The National Labor Relations Board adopts the rec- ommended Order of the administrative law judge as modified below and orders that the Respondent, Acme Die Casting, A Division of Lovejoy Industries, Incor- porated, Northbrook, Illinois, its officers, agents, suc- cessors, and assigns shall take the action set forth in the recommended Order, as so modified. 1. Delete paragraph 1(b) and reletter the subsequent paragraphs accordingly. 2. Substitute the following for paragraph 2(e). ‘‘(e) Make whole the unit employees for any mone- tary loss they may have suffered as a result of the Re- spondent withholding the general wage increases in February and July 1988 in the manner set forth in this decision.’’ 1087 LOVEJOY INDUSTRIES 3. Substitute the attached notice for that of the ad- ministrative law judge. APPENDIX NOTICE TO EMPLOYEES POSTED BY ORDER OF THE NATIONAL LABOR RELATIONS BOARD An Agency of the United States Government The National Labor Relations Board has found that we violated the National Labor Relations Act and has or- dered us to post and abide by this notice. Section 7 of the Act gives employees these rights. To organize To form, join, or assist any union To bargain collectively through representatives of their own choice To act together for other mutual aid or protec- tion To choose not to engage in any of these pro- tected concerted activities. WE WILL NOT interfere with these rights. WE WILL NOT threaten to cancel established wage increases because employees elect as their collective- bargaining representative, or support, United Electrical, Radio & Machine Workers of America (UE) (the Union) or any other labor organization. WE WILL NOT refuse to recognize the Union as the exclusive collective-bargaining representative of the employees in the bargaining unit described below. WE WILL NOT change or eliminate working condi- tions of bargaining unit employees because they sup- port the Union, nor unilaterally do so without affording the Union notice and opportunity to bargain concern- ing any proposed changes. WE WILL NOT withhold twice-yearly general wage increases from unit employees because the bargaining unit employees elect or support the Union, and WE WILL NOT withhold established wage increases without affording the Union notice and an opportunity to bar- gain. WE WILL NOT issue written warnings to employees, suspend employees, or otherwise discriminate against employees in order to discourage them from support- ing the Union, or any other labor organization. WE WILL NOT issue written warnings to employees, suspend employees, or otherwise discipline employees pursuant to unilateral changes we make in established working conditions without affording the Union notice and opportunity to bargain over such proposed changes. WE WILL NOT in any like or related manner interfere with, restrain, or coerce employees in the exercise of the rights guaranteed you by Section 7 of the Act. WE WILL, on request, bargain with the Union and put in writing and sign any agreement reached on terms and conditions of employment for our employees in the bargaining unit: All full-time and regular part-time production and maintenance employees employed by Acme Die Casting at its facility located at 3610 Commercial Avenue, Northbrook, Illinois 60062; excluding all office clerical employees, technical employees, tool and die makers, managerial employees, pro- fessional employees, confidential employees, and guards and supervisors as defined in the Act. WE WILL restore the working conditions prevailing before the October 16, 1987 election by which you (1) in cold weather were permitted to start your cars be- fore the end of the shift, (2) were permitted to warm food in cafeteria microwaves before breaks and lunch, (3) were permitted to consume food and drinks, includ- ing coffee, at work stations to the extent such practice does not constitute a safety hazard, and (4) had free access to restrooms without the restriction of first hav- ing to give notice to your supervisor. WE WILL remove from our files any reference to the following disciplinary actions, and WE WILL notify each of these employees in writing that such discipline will not be used against these employees in any way: 11–11–87 written warning Antonio Ramirez 11–11–87 written warning Fidencio Olivares 1–07–88 written warning Mauricio Aguirre 10–24–88 written warning Nicolas Valenzuela 10–24–88 written warning Jose Aguirre 10–24–88 written warning Rodolfo Banales 10–24–88 written warning Marcial Canales 10–24–88 written warning Mario Garcia 10–24–88 written warning Francisco Mombela 10–24–88 written warning Hugo Paz 10–24–88 written warning Nicolas Valenzuela 10–24–88 suspension Rodolfo Banales 10–24–88 suspension Nicolas Valenzuela WE WILL make whole, with interest, Rodolfo Banales and Nicolas Valenzuela for any loss of pay or benefits they may have suffered because we suspended them on October 24, 1988. WE WILL restore, retroactive to February 1988, twice-yearly general wage increases for bargaining unit employees, and WE WILL continue to grant semiannual general wage increases until Acme and the Union agree otherwise, until we bargain to a good-faith im- passe, or until the Union refuses to bargain in good faith over that condition of employment. WE WILL make whole all bargaining unit employees and former employees for any monetary loss you suf- fered by our withholding the wage increase in both February and July 1988. 1088 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 1 All dates are for 1988 unless otherwise indicated. 2 Exhibits are designated as G.C. Exh. for the General Counsel, R. Exh. for Respondent Acme, and Jt. Exh. for the single joint exhibit. The Union did not offer any exhibits. References to the 26-volume transcript of testimony are by volume and page. Included for con- venience, annotations to the record are not intended to be exhaustive. 3 At the close of the Government’s case-in-chief, the General Counsel withdrew an allegation on the June 9, 1988 discharge of Juan Lopez. (14:2040.) 4 The transcript of testimony contains many errors, mostly mis- spelled names and words. The parties cooperated in compiling two exhibits. The first, R. Exh. 131, is a 55-page list of corrections by page, line, error, and correction. As they describe on the last page of that exhibit, the parties disagree on the correct way one sentence should be rendered. It is a minor point. Agreeing with the General Counsel’s rendering, I approve all the corrections set forth on R. Exh. 131. The second exhibit, R. Exh. 132, is an alphabetized list of 78 names, correctly spelled, which appear in the transcript. Not all names in the transcript are on the list. WE WILL restore the working condition prevailing before October 21, 1988, when you were permitted during worktime to get coffee in the cafeteria and re- turn with it to your work stations. WE WILL rescind the October 24, 1988 written no- tice which eliminated the established privilege of your being permitted during worktime to get coffee in the cafeteria and return with it to your work stations. WE WILL restore the Saturday overtime schedule of 6 a.m. to 12 noon, with one 20-minute break, prevail- ing before we unilaterally reduced the hours and breaktime on January 31, 1988. WE WILL make whole, with interest, all the bargain- ing unit employees (including employees no longer employed) who suffered a loss of pay as a result of our unilaterally reducing the hours scheduled for Satur- day overtime. ACME DIE CASTING, A DIVISION OF LOVEJOY INDUSTRIES INCORPORATED Dawn Miller Scarlett, Esq. and Emilie Fall Schrage, Esq., for the General Counsel. Larry G. Hall, Esq., James J. Salzman, Esq., and Elizabeth M. McDowell, Esq. (Matkov, Salzman, Madoff & Gunn), Chicago, Illinois, for the Respondent Company. Terry Davis, Field Org. and Timothy P. Curtin, International Representative (UE), for the Charging Union. DECISION STATEMENT OF THE CASE RICHARD J. LINTON, Administrative Law Judge. The prin- cipal issue here is whether Acme Die Casting withheld semi- annual general wage increases from its employees in 1988 because, in a Board-conducted election held October 16, 1987, the employees selected the Union to represent them. Finding that Acme was so motivated, in violation of 29 U.S.C. § 158(a)(3), and that the withholding was a unilateral change violating 29 U.S.C. § 158(a)(5), I order Acme to make whole its employees with interest, to resume its prac- tice of granting a general pay increase twice a year, and to bargain with the Union before making such changes. I presided at this hearing in Chicago, Illinois, on 26 days beginning January 23, 1989, and closing October 2, 1989. The first pleading in the case is the June 1, 1988 complaint issued in Case 13–CA–27619 by the General Counsel of the National Labor Relations Board through the Regional Direc- tor for Region 13 of the Board. The complaint is based on a charge filed March 18, 1988, by United Electrical, Radio & Machine Workers of America (UE) (the Union, UE, or the Charging Party) against Acme Die Casting, a Division of Lovejoy Industries Incorporated (Respondent or Acme).1 On March 23, 1988, NLRB Region 13 served the charge on Acme. Other charges and complaints followed. Although technically consolidated by NLRB Region 13, the various complaints and their amendments were not physically com- bined into a single document until the Government moved for that purpose during a lengthy adjournment following the fourth day of the hearing in January. Even then the General Counsel, a few days later, moved to amend the proposed consolidated complaint. (G.C. Exh. 1vv, Apr. 26, 1989.)2 Over certain objections, I granted the motion by order dated May 14, 1989. (G.C. Exh. 1yy.) At the resumption on June 26, 1989, the fifth day of the hearing, the Government finally submitted a single document incorporating all allegations. This was when the Government offered, and I received (5:741–743), the amended consolidated complaint (G.C. Exh. 1zz). Unless otherwise indicated, either specifically or by context, ‘‘complaint’’ herein refers to the single document of June 26, 1989—the trial complaint. In the complaint the General Counsel alleges that Acme violated Section 8(a)(1) of the Act by: (1) threats uttered dur- ing March–June 1988 that the employees would not receive scheduled wage increases because they had selected the Union as their bargaining representative; (2) a threat in June 1988 that Acme would close its facility because of the Union; (3) telling employees in June 1988 that they could not wear T-shirts bearing union insignia during working hours, while allowing employees to wear similar garments without insignia; and (4) on September 15, 1988, causing a supervisor’s car to injure a striker because of the employee’s participation in a strike. The complaint alleges that Acme violated Section 8(a)(3) of the Act by: (1) discharging Raymundo Aquirre on Feb- ruary 25, 1988, warning Jorge Nicolas Valenzuela on June 16, 1988, and suspending, for 2 days, Sacramento Olivares on August 12, 1988; (2) making a series of unilateral changes in working conditions on October 19, 1987, Feb- ruary 12, and October 24, 1988; (3) failing since December 2, 1987, to grant scheduled wage increases; and (4) issuing written warnings to several employees, and suspending two others, based on the alleged unilateral changes.3 The complaint also alleges that Acme, by the unilateral changes, violated Section 8(a)(5) of the Act. By its July 11, 1989 answer (R. Exh. 36) to the trial com- plaint, Acme admits certain factual allegations, denies others, and denies violating the Act. As affirmative defenses, Acme alleges that Section 10(b) of the Act bars certain allegations, and it raises certain procedural objections to various allega- tions. On the entire record,4 including my observation of the de- meanor of the witnesses, and after due consideration of the 1089 LOVEJOY INDUSTRIES 5 Although lengthy (General Counsel’s 153 pages plus a proposed order and notice to employees; Acme’s 469 pages), the detailed briefs, usually well annotated with references to the 3997-page tran- script, were helpful to me. 6 For years Acme, bearing ‘‘Corporation’’ at the end of its name, was a separate and, possibly, an independent corporation. See Acme Die Casting Corp., 262 NLRB 777, 778 (1982). Some years ago Acme became a division of Lovejoy Industries Incorporated. What is unclear is whether Acme dissolved its corporate status and con- verted to an unincorporated division of Lovejoy, as its current name would suggest, or whether, retaining its corporate form, it actually is a wholly owned subsidiary of Lovejoy. 7 In that litigation the name most commonly used as Valenzuela’s given name, Nicolas, is misspelled as Nicholas. briefs5 filed by the General Counsel and the Respondent, I make the following FINDINGS OF FACT I. JURISDICTION Although its corporate status is unclear,6 Acme manufac- tures die castings and related products at its Northbrook (Chicago), Illinois plant. During 1988 Acme sold, at non- retail, and shipped from its Northbrook plant goods valued at least $50,000 directly to points and customers outside Illi- nois. Acme admits, and I find, that it is an employer within the meaning of Section 2(2), (6), and (7) of the Act. II. LABOR ORGANIZATION INVOLVED Respondent admits, and I find, that United Electrical, Radio & Machine Workers of America (UE) is a labor orga- nization within the meaning of Section 2(5) of the Act. III. THE ALLEGED UNFAIR LABOR PRACTICES A. Background 1. Union elections and litigation Hired in 1971, Jorge Nicolas Valenzuela, one of the al- leged discriminatees, had been working over 17 years for Acme at the time of his January 1989 testimony. (3:460; 4:646, 688; 6:812.) That service is longer than that of most of the other witnesses. From Valenzuela we learn that during his tenure there have been previous union campaigns at Acme. The first began in 1980 with Local 456, Service Em- ployees International. (3:461; 6:812.) Acme laid off (fired, in effect) Valenzuela in that campaign. (3:461.) Charges were filed, litigation followed, and the Board, adopting (with slight modification) the order of Administrative Law Judge Robert T. Wallace, directed Acme to reinstate Valenzuela (one of the leading supporters of Local 456) and to pay him back- pay. Acme Die Casting Corp., 262 NLRB 777 (1982). Acme did so in 1984 (6:814–815) following enforcement (with modification on a separate point) of the Board’s Order. NLRB v. Acme Die Casting Corp., 728 F.2d 959 (7th Cir. 1984).7 A Board-conducted election was held November 14, 1980. Acme, 262 NLRB at 778 fn. 1. The union lost. (23:3516.) Robert Novak became Acme’s president on February 24, 1987, having started as a sales engineer in 1974. (23:2511.) Novak reports that there have been two union elections be- fore the third one of October 1987. The first was in 1980 and the second in 1985 and, Novak asserts, Acme won both. (24:3516.) Novak persuaded Peter P. Balma, a former man- ager at Acme, to become Acme’s plant manager effective April 1, 1987. (24:3699, 3704, 3712.) In the spring of 1987 the UE began talking to Acme’s em- ployees, and the card-signing drive began in June. (2:388; 3:423–424, 464.) Timothy P. Curtin, an International rep- resentative of the Union, was in charge of the organizing campaign. Curtin’s principal assistant was Terry Davis, a field organizer for the UE. (2:387–389; 16:2243–2245.) Eventually the Union prepared a letter (G.C. Exh. 31), dated August 13, 1987, demanding recognition. About that date, Curtin, accompanied by some 60 to 70 of Acme’s employees waiting just outside the building, delivered the letter to Acme’s president, Robert Novak, who declined to accept the letter or to recognize the Union. (2:389–391; 3:489–90; 4:714–716.) Shortly thereafter, on August 17, the Union filed a petition (G.C. Exh. 33) for a representation election in a unit of Acme’s production and maintenance employees. (2:392–393.) The ensuing Board-conducted election was held on October 16, 1987. Valenzuela served as the Union’s observer. (2:402.) The tally of ballots (G.C. Exh. 34) reveals that, of 113 eligible voters, 2 ballots were challenged, 69 employees voted yes, and 39 voted no. The Union had won. (2:400.) Following the election the unit employees elected stew- ards, and Curtin sent a letter (R. Exh. 1) to Acme’s Peter Balma notifying him of those elected. (2:200–201, 401–402.) The letter named Marcial Canales as chief steward, Mauricio Aguirre, Raymundo Aguirre, and Jorge Serrato as stewards on the first shift, Paul Magee as the second-shift steward, and Angel Otero as Magee’s assistant steward. In the letter Curtin requested that Acme notify Canales ‘‘on all matters concerning discipline of any Acme Die worker.’’ Canales hand delivered it to Supervisor Ron Adamczyk because Novak, although in his office, could not or would not receive Canales. (13:1775.) Acme filed objections to the conduct of the October 16, 1987 election, but on April 13, 1988, the Board issued its decision (G.C. Exh. 35, not included in bound volumes) cer- tifying the Union as the exclusive collective-bargaining rep- resentative of Acme’s employees. (2:402–404.) As reflected in the certification, and admitted in the pleadings, the cer- tified unit is: All full-time and regular part-time production and maintenance employees employed by the Employer at its facility located at 3610 Commercial Avenue, North- brook, Illinois 60062; excluding all office clerical em- ployees, technical employees, tool and die makers, managerial employees, professional employees, con- fidential employees, and guards and supervisors as de- fined in the Act. On Monday, April 18, the Union’s Terry Davis telephoned Acme and told Novak that the Union wanted to begin con- tract negotiations as soon as possible. Novak said Acme’s lawyer would contact Davis. (2:405.) Davis sent a confirm- ing letter (G.C. Exh. 36) of the same date with copies to four employees who had been elected as the employee-members of the Union’s negotiating committee (2:401, 406; 3:431; 13:1774): Antonio Aguilera, Jesus Lopez, Antonio Ramirez, 1090 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD and (Jorge) Nicolas Valenzuela. By letter (G.C. Exh. 37) dated April 21, Acme’s attorneys, by Larry G. Hall, ac- knowledging receipt of the Union’s April 18 letter, citing the pending objections, and asserting a good-faith doubt, denied the request to begin negotiations because Acme ‘‘cannot en- gage in collective-bargaining negotiations with a labor orga- nization that does not represent an uncoerced majority of the employees.’’ When the Union received Attorney Hall’s rejection letter, Terry Davis and a couple of union staff assistants, including Lydia Sanchez Bracamonte, led a group of 40 to 50 employ- ees to the office, during the lunch period, to see Novak. Plant Manager Peter P. Balma advised the group that Novak was not there. To Davis’s question of why Acme would not negotiate, Balma said he did not know. (2:408.) Following this confrontation Davis, on April 25, filed a charge (G.C. Exh. 38) by the Union against Acme alleging a refusal to bargain. (2:409.) In the ensuing summary judgment proceed- ing the Board, by its decision and order of August 31, 1988, found Acme in violation of Section 8(a)(5) and (1) of the Act by refusing to bargain with the Union. Lovejoy Indus- tries, 290 NLRB No. 127 (Aug. 31, 1988) (not reported in Board volumes). By letter (G.C. Exh. 40) dated September 9 to Novak, Terry Davis renewed the Union’s request that Acme begin negotiations. Both the General Counsel (Br. at 6) and Acme (Br. at 177) refer to this letter, although Respondent Acme does so by observing that the letter ‘‘allegedly was delivered to Novak by the members of the negotiating committee.’’ Terry Davis testified that some letters were mailed, and oth- ers hand delivered. Because she could not recall as to the September 9 letter, she testified that she must rely on her handwritten note on her copy of the letter that members of the negotiating committee submitted the letter to Novak on September 9. I sustained Respondent’s objection to receipt of the letter. (2:412a–414.) No argument was made that the ex- hibit should be received for the truth of the note pursuant to Rule 803(6), Federal Rules of Evidence (FRE). Although the General Counsel never reurged the Govern- ment’s offer of General Counsel Exhibit 40, the court re- porter included the letter in the folder of General Counsel’s exhibits. Valenzuela, one of the members of the Union’s ne- gotiating committee, testified that on September 9 he and two other members spoke to Novak. The visit preceded a strike vote by the workers. The Valenzuela group asked, among other matters, that the Union be recognized. (5:751.) Valenzuela does not specifically assert that his group deliv- ered General Counsel’s Exhibit 40 to Novak. Actually, Davis testified that because the group reported to her that Balma said, ‘‘mail it,’’ that she mailed it to Acme. (2:412.) Delivery is presumed from mailing. Accordingly, I now receive Gen- eral Counsel’s Exhibit 40 in evidence. No recognition followed. Indeed, a 2-day strike ensued on September 14–15, 1988. The first day of the strike members of the negotiating committee, Davis testified, were similarly unsuccessful in delivering another letter to Acme. Davis mailed this one also. (2:412, 414.) Dated September 13, the letter (G.C. Exh. 41) asserts that the employees are striking ‘‘because of the Company’s unfair labor practice of refusing to negotiate on wages.’’ The employees participated in another strike weeks later, a 1-day event on October 26. A one-sentence letter (G.C. Exh. 42) from the Union’s Terry Davis to Robert Novak in- formed Acme that the workers ‘‘are on an unfair labor prac- tice strike for the Company’s unilateral withdrawal of the right to take food and coffee to the work stations.’’ (2:415– 415a.) After the October 2, 1989 close of the hearing, and follow- ing the April 23, 1990 receipt of the briefs, the Seventh Cir- cuit, on June 14, 1990, enforced the Board’s August 31, 1988 order that Acme bargain with the Union. NLRB v. Lovejoy Industries, 904 F.2d 397 (7th Cir. 1990). 2. Animus Should the animus inherent in the unfair labor practice finding respecting Valenzuela’s 1980 layoff carry forward and be applied to Acme in this case? The General Counsel contends that it should. (3:464; 4:709–710; Br. at 114–115.) Arguing the contrary, Acme asserts that both ownership and management are different now from that in 1980. (4:710– 711; Br. at 67 fn. 40.) For some reason Acme, in its posthearing brief, although pointing to asserted differences in Acme’s management, does not refer to any differences in ownership. Differences in management and ownership are described shortly. The differences are significant, although not complete. I find that some of the taint from the earlier case should attach, although the effect, or weight, I give to this factor is slight. B. Acme Die Casting 1. Products and customers As described by Plant Manager Peter P. Balma, Acme, op- erating as a ‘‘job shop,’’ manufactures die castings made from aluminum or zinc. The castings weigh anywhere from 10 grams to 10 pounds. The casting is produced by injecting aluminum or zinc, as a heated liquid, into a mold. Secondary operations, such as drilling, tapping, and painting form the castings into finished products. Acme’s customer base consists primarily of companies in the telecommunications and automobile industries. Acme fo- cuses particularly on supplying the telecommunications in- dustry, manufacturing switching components for AT&T, and producing radio and miscellaneous components for Magna- vox. (1:75–77; 24:3699–3700.) 2. Managers, supervisors, and departments As mentioned earlier, Robert Novak has been president of Acme since February 24, 1987, replacing Leroy Hagner who had held the position for about 10 years. (23:3507, 3513.) Novak testified that Hagner actually did not leave until April 30, 1987. During those 2 months, Hagner would come to the plant for certain hours during the day. (24:3611.) As with most other matters in the record, definitive facts are elusive. Thus, in the batch of correspondence with Acme’s cus- tomers, Hagner, as ‘‘General Manager,’’ wrote to AT&T by letter dated March 6, 1987. (R. Exh. 123–16.) Elsewhere, on May 20, 1987, Novak tells AT&T that he, Novak, became Acme’s president and general manager ‘‘effective May 1, 1987.’’ (R. Exh. 123–21.) On June 29, 1988, Novak in- formed Motorola that he was appointed Acme’s president ‘‘in April of 1987.’’ (R. Exh. 124–17.) At one point Balma testified that Hagner was still acting as president when he ar- 1091 LOVEJOY INDUSTRIES rived on April 1 even though Novak was president. (24:3770–3771.) The precise date is unimportant, for it does not appear that Hanger’s 2-month layover had any impact on actions Novak and Balma took, or did not take, respecting employees and work rules during that timeframe. Starting as a sales engineer in 1974 at Acme, Novak worked his first 2 years in the toolroom, followed by several years as a sales engineer, and then for a couple of years as the sales manager before becoming Acme’s president in late February 1987. In his sales positions, Novak was not in- volved in the union elections of 1980 and 1985, and before becoming president, Novak had not gained any direct experi- ence in personnel or employee relations. (23:3511–3516.) Hagner was Acme’s president during the 1980 and 1985 union campaigns. Peter P. Balma has been Acme’s plant manager since April 1, 1987, replacing Harold Georgeson who became vice president of engineering at Acme. Starting at Acme in about 1967 as maintenance manager, Balma worked in that capac- ity for about 6 years before serving the next 4 years, to about 1977, as vice president of manufacturing or operations. For the next 10 years Balma worked elsewhere. He returned to Acme on April 1, 1987, as plant manager reporting to Novak. (1:198–199; 24:3699, 3704.) As plant manager in 1980, Georgeson is the person who laid off Valenzuela. Acme Die Casting Corp., 262 NLRB 777, 778 (1982). Directly or indirectly, everyone at Acme reports to Novak. Reporting to him directly from the production area is Balma (1:75; 23:3509; 24:3699), and from the staff area are Georgeson, the chief engineer (23:3509), Mary Compton, the current quality control (or assurance) manager (23:3509; 24:3703), Catherine ‘‘Kay’’ Mooney, the controller (1:52; 23:3509), and the sales manager, Dick LaRue. (23:3508; 24:3588, 3705.) Serving as office manager, Mooney super- vises four clerks who handle payroll, accounts payable and receivable, and insurance. (1:52; 23:3510.) Acme does not have a personnel (employee relations/human resources) man- ager or department. That function is handled by Novak, Balma, or the office clerks. (23:3510.) As plant manager, Balma oversees the production and maintenance operation. (24:3699.) That operation is divided into several departments. Jim Scott supervises the aluminum die casting (ADC) department, including the trim presses, a position he has held since July 1985. (1:78; 2:246; 21:3184; 24:3701.) Since February 15, 1988, Scott has been aided by Juan San Roman, formerly one of the three setup operators in ADC. On that date Sam Roman became assistant super- visor. (20:2910, 2957; 21:3215.) Zinc die casting (ZDC), a smaller function than ADC, is supervised by Santi Greco. (1:86; 24:3703.) Larry Stoner supervises the secondary machining depart- ment (SMD), a position he has held for 21 years. Aiding Stoner are assistant supervisors Dan Basgall and Faustino ‘‘Tino’’ Ontiveros. (1:89; 21:3077; 24:3702.) Hired by Acme in 1973, Basgall was promoted to Assistant Supervisor in April 1987. (20:2990–2991.) Ronald Adamczyk supervised the precision machining- waveguide department from 1980 until April 1988 when Dan Basgall assumed that responsibility and Adamczyk became supervisor of the shipping and receiving department. (2:230– 231; 20:2831; 24:3702–3703.) In October 1987 Tom Malleck, now no longer at Acme, supervised shipping and receiving. (24:3703.) Before March 20, 1987, San Murchison was the quality control (QC) manager. On that date Novak posted a notice (R. Exh. 27) informing employees that effective ‘‘imme- diately’’ Murchison was assigned ‘‘special duties’’ and that Robert Ferguson was the new QC manager. (9:1355–1356, 1406.) The status of Murchison as a statutory supervisor or agent after March 20, 1987, is in dispute. As already noted, Mary Compton has replaced Ferguson as the QC manager. When Balma returned to Acme on April 1, 1987, Gus Hauser was the maintenance supervisor. In about August 1989 Acme terminated Gus Hauser. Balma currently oversees that department. (2:193; 24:3704.) Acme operates a night shift for the ADC and the west side of secondary machining (SMD) under the supervision of Howard A. McArtor Jr. (1:78, 104–105; 2:208; 19:2754a– 2755; 24:3703.) McArtor has worked intermittently for Acme as a supervisor since July 1, 1968. His most recent assign- ment as night supervisor began January 1, 1986. (19:2754.) Assisting McArtor as lead or setup persons are John Selwitschka in ADC and Samuel Aguirre in SMD. (2:216; 19:2764, 2789.) As already mentioned, the tally of ballots (G.C. Exh. 34) puts the number of eligible voters at 113. When he first testi- fied on the opening day (Jan. 23, 1989), Balma placed the number of unit employees at 101. (1:77.) Testifying on Sep- tember 27, 1989, toward the end of the hearing, Balma testi- fied that the number of hourly employees has dropped from about 120 in October 1987 to the current level of about 95 hourly employees. Balma testified that the drop has come from attrition, rather than layoffs, in a period of declining sales. (24:3700–3701.) The difference in the October 1987 figures, 113 versus 120, is explained on the basis that the 120 hourly employees would include the tool and die makers excluded from the unit count. Balma testified (Sept. 28, 1989) that Acme’s toolroom has six employees plus a super- visor, Hank Bollen. (25:3903–3904.) Return now to the question of animus and the different management. As a reading of the Valenzuela case (262 NLRB 777) and the instant record discloses, the president has changed (Leroy Hagner then, Robert Novak now) as has the plant manager (Harold Georgeson then, Peter Balma now, although Georgeson remains at Acme, but in engineer- ing). Supervisor Larry Stoner remains, and Dan Basgall is now an assistant supervisor rather than a ‘‘boss set-up man’’ (262 NLRB at 780). Ownership of Acme in 1980–1981 is not shown. In the earlier case Acme is described as a Delaware corporation, with ‘‘Corporation’’ as part of its name. Ownership is not given. In the instant case the complaint alleges, and the answer admits, that Acme is a corporation. At one point in the record before me Respondent’s counsel suggests that in 1980 Acme was owned by ‘‘Beatrice Foods.’’ (4:710.) There is no supporting evidence. At some point, year unspecified, Lovejoy Industries Incorporated acquired Acme. Not only does counsel so indicate (4:710), but Novak explains that he reports to one Tony Girone at Lovejoy, although he is unsure 1092 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 8 In its posthearing brief Acme refers to Girone as Lovejoy’s presi- dent. (Br. at 62, 222.) Based on that representation, I shall do like- wise. whether Girone’s title is vice president. (24:3576.)8 Novak is quoted (16:2323, Jose Sandoval) as referring to Walter Lovejoy as the ‘‘owner’’ of Lovejoy Industries, and Balma so identifies Walter Lovejoy. (25:3822.) Lovejoy Industries Incorporated (Lovejoy) apparently is a privately owned com- pany. As earlier stated, I find that some of the taint should carry forward from the earlier case. Because of the significant dif- ferences in management and ownership, however, I attach lit- tle weight to this source of animus. C. Overview Before the October 16, 1987 election, contends the Gen- eral Counsel, Acme’s employees long had enjoyed certain freedoms in the workplace notwithstanding posted rules cov- ering some of the matters. Acme’s relaxed approach to the concept of plant rules, the Government suggests, elevated those freedoms to terms and conditions of employment which Acme could not change unilaterally after the Union won the election. Yet change them Acme did, and to retaliate against the employees because they voted for the Union, the General Counsel alleges and argues. Consistent with its de- termination to punish the workers for so voting, Acme de- nied them a general wage increase and issued written warn- ings and suspended and discharged employees. Acknowledging that enforcement of rules had been spotty and without centralized direction and support until April 1987, Acme argues that what changed was not its rules but its top management. Robert Novak became president on Feb- ruary 24, 1987, and he brought Peter P. Balma back as plant manager about a month later, on April 1. When he arrived, Balma testified (24:3710, 3715) conditions were ‘‘deplor- able,’’ in ‘‘chaos,’’ with employees walking throughout the plant as they wished. Except for efforts to maintain discipline by Jim Scott in ADC and Gus Hauser in maintenance, rules were not being enforced. (24:3715–3716.) After Balma met with the supervisors and told them to enforce the rules, mat- ters improved. Then the Union entered the picture and de- manded recognition. About a month before the election employees, principally supporters of the Union, began disregarding rules again, and that continued in the first days after the election. About a week after the election, Acme’s supervisors met with their employees and informed them that the rules must be fol- lowed. Acme argues that when the Union realized the com- pany would not bargain pending resolution of Acme’s objec- tions to the election, it switched from being merely aggres- sive to using a tactic of suggesting, through Terry Davis and the employee leaders, that employees could adopt unlawful means of forcing Acme to the bargaining table. Davis and the leaders deny any such tactic or suggestion. D. Changes in Work Rules—October 19, 1987 1. Allegations, posted rules, and past practice Complaint paragraph 11(a) alleges that on October 19, 1987 (the Monday following the Union’s Friday election vic- tory), Acme unilaterally implemented several changes in working conditions, as follows: (1) Use of the bathroom without permission was pro- hibited. (2) Leaving the facility to start car engines before the end of the shift was prohibited. (3) Consumption of food and beverages in work areas was prohibited. (4) Employees were no longer permitted to warm up food prior to their lunchbreaks. (5) Work station clean-up time was changed from 4:10 p.m. to 4:15 p.m. for the first shift and from 2:40 a.m. to 2:45 a.m. on the second shift. (6) Personal clean-up time was changed from 4:20 p.m. to 4:25 p.m. on the first shift and from 2:50 a.m. to 2:55 a.m. on the second shift. (7) Employees received written warnings for conduct that was previously tolerated, and for conduct that pre- viously only warranted an oral reprimand. Complaint paragraph 13 alleges that these unilateral changes violate Section 8(a)(3) and (1) of the Act, and com- plaint paragraph 14 alleges that they also violate Section 8(a)(5) and (1) of the Act. There is no dispute that for many years Acme has had posted a schedule for breaks, lunch, and cleanup. The sched- ule posted as of the election, and until January 1988, is dated December 1, 1986, effective beginning the following day, and reads (G.C. Exhs. 9–3): Because you are working nine hours a day, we are changing the break times and lunch times as follows: 7:00 AM—Start 9:10 AM to 9:20 AM—Break 11:30 AM to 12:00 PM—Lunch 2:00 PM to 2:10 PM—Break 4:15 PM–4:30 PM—Clean-Up 4:30 PM–Home As described by Howard McArtor, the night-shift super- visor, times for the night shift for the same period, were (19:2755–2761): 4:30 p.m.—Start 7:00 p.m.–7:10 p.m.—Break 10:00 p.m.–10:30 p.m.—Lunch 12:30 a.m.–12:40 a.m.—Break 1:45 a.m.–1:55 a.m.—Machine Area Cleanup 1:55 a.m.–2:00 a.m.—Personal Cleanup 2:00 a.m.—Punch Out The break and lunch periods actually have second break/lunch periods to accommodate the trim press operators who relieve the die cast machine operators. Moreover, be- cause they have a larger and messier area to clean, the die cast operators begin their machine cleanup 5 minutes earlier than the stated time. The General Counsel contends the evidence shows that be- fore the Union’s October 1987 election victory Acme did not strictly enforce its time schedule. Moreover, employees could put their lunches on the furnaces to heat and return for them at lunch. They could go to the cafeteria, get coffee, and bring 1093 LOVEJOY INDUSTRIES the coffee to their work stations during worktime. They also needed no permission to go to the restroom/washroom. Fol- lowing the election, however, Acme announced that hence- forth the scheduled times must be complied with, that em- ployees could not bring food or drink to their work areas, and they must ask permission before going to the washroom. (Br. at 7–11.) Citing the testimony of its supervisors, plus that of Plant Manager Balma, and President Novak, Acme seeks findings as follows: (1) that in April and May 1987 Balma and Novak tightened up enforcement of the rules; (2) that employees generally complied until about 3 to 4 weeks before the elec- tion when many (particularly the union leaders) began abus- ing the rules; and (3) when the abuse continued after the election Acme announced to employees in group meetings that the rules—all of which preexisted—would have to be obeyed. Acme’s pre-Balma practice (that is, its practice before April 1987) consisted of a rather mixed approached to en- forcement of the scheduled worktimes. Written warnings were issued, however. When they issued, it usually was after the employee had been orally warned to comply with the schedule. For example, Jim Scott, who was hired in July 1985 as a supervisor over the ADC (aluminum die casting) to replace Ernesto Vega (20:2913), held a group meeting with his employees in July 1985 and informed them that the time rules would have to be obeyed. He orally warned eight employees in August 1985 against taking unauthorized breaks. When they persisted, he issued them written warn- ings (G.C. Exhs. 5hhhhh, 7-13, 7-14, 7-31, 7-33, 7-34, 7-35, 7-37) in September 1985. (20:2914–2918, 2940–2941, 2946, 2977; 23:3401.) Scott was not the only supervisor to issue written warnings in 1985 over abuse of the scheduled times of work and breaks. In June 1985 Thomas Malleck issued warnings to several employees for repeatedly stopping work early at shift’s end. (G.C. Exhs. 5iiiii–sssss.) It was one of three list- ed factors in the July 3, 1986 discharge of Tyrone Newson. (G.C. Exh. 5aaa.) When Malleck gave a 3-day layoff to Rosendo Lopez on October 22, 1985, one of the three listed infraction-reasons was (G.C. Exhs. 7–10): ‘‘Failure to obey instructions when told to keep breaks and lunch periods to scheduled time spans—10 min & 30 min.’’ On July 30, 1986 H. Georgeson, apparently the then plant manager, Harold Georgeson, imposed a 2-day layoff on Salvador Ontiveros, an employee in Ron Adamczyk’s department, because (G.C. Exh. 7–1): ‘‘Employee left the building 10 minutes before quitting time and did not punch timecard.’’ Failure to punch the timecard is an additional factor, of course. Nevertheless, failure to observe the schedule was one of the stated reasons. Adamczyk does not recall the incident. (20:2902–2904.) Despite Georgeson’s action respecting Salvador Ontiveros in July 1986, it appears that Georgeson normally was not supportive of supervisors respecting discipline of employees. As mentioned above, when Scott arrived in July 1985 and observed a lack of discipline in his department and in the plant, with employees taking breaks every hour and wander- ing throughout the plant, he went to Georgeson only to learn that Georgeson’s style was not to make waves. (23:3398.) Georgeson may well have been taking his cue from then President Leroy Hagner, whose philosophy toward problems, we learn from Balma, was to ignore them on the hope they would go away. (24:3708–3709.) Hagner’s ostrich-like phi- losophy was one reason Balma left Acme in 1977 after hav- ing been there 10 years. (24:3709–3710.) 2. Robert Novak and Peter Balma arrive in February– April 1987 Acme’s problems did not disappear. Letters (R. Exhs. 123 and 124) from customers suggest that Hagner’s strange style of business management brought the Company to near col- lapse by early 1987 when Tony Girone, Lovejoy’s president, promoted Robert Novak, Acme’s sales manager, to replace Hagner as president. Novak inherited massive problems with product quality and with manufacturing (production delays, disrepair of tools and trim dies). Personnel problems on the plant floor simply compounded the unacceptable situation. Needing help, Novak reached out for Balma. As Novak had started at Acme in 1974 (23:3511), he and Balma had worked together on special projects. (24:3711–3712.) Novak (in 1987) told Balma he could not get a handle on the situation. (24:3712.) When Balma arrived at Acme on April 1, 1987, he found employee discipline ‘‘deplorable,’’ production ‘‘bad,’’ product quality ‘‘atrocious,’’ Novak up to his ears in alligators trying to respond to customers’ com- plaints, and only three supervisors (Jim Scott, Howard McArtor, and Gus Hauser) trying to maintain control of their departments. (24:3710–3711, 3716.) Most of production was 5 to 6 months late, with one data sheet listing customers whose orders were 18 months late. Balma explains that the dollar value of the production backlog exceed $8 million. (24:3714.) As for permission to go to the restroom, no party claims that was required before Balma’s April 1, 1987 arrival. Su- pervisors handled the matter differently. Although Scott tech- nically included notice as part of an employee’s duty to in- form Scott when the employee was leaving his work area, Scott disregarded infrequent quick trips to the restroom even when he received no notice. (20:2915, 2923, 2944–2945, Roman; 23:3431–3436, Scott.) On McArtor’s night shift, when the die cast operator needed to go to the restroom he simply signaled his trim press operator who took over the die cast machine for those few minutes. Permission from McArtor was never required. (19:2765, 2799.) Stoner asserts that notice only was required, and that could be simply a wave of the hand. (21:3096–3097.) Before October 1987, employees in practice would leave their work stations a few minutes before shift’s end to start and warm car engines, usually during cold weather, and re- turn to work to punch out. Several employees testified that such was the common practice. Most of these testifying em- ployees worked, before 1987, either for Larry Stoner (em- ployees Nicolas Valenzuela and Antonio Ramirez) or Ron Adamczyk (Mauricio Aquirre), although Antonio Aguilera (9:1350, 1380) worked for Sam Murchison, the quality con- trol manager before 1987 (8:1288; 9:1356; R. Exh. 27), and Sacramento Olivares (11:1499), was working for Jim Scott. Ron Adamczyk, supervisor over the precision machining and packing departments for the relevant time (2:231), confirms that until October 1987 employees in his department freely did so before the shift’s end. (2:237–238.) Stoner simply says that the last 5 minutes were for the em- ployees to wash their hands, change clothes, or go outside and start car engines, and that he did not police that 5-minute 1094 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 9 Roman is married to a sister of Banales. (20:2937.) 10 Canales did not transfer to Scott’s ADC until November 1988. Before that he worked in the drilling department under Robert Fer- guson. (13:1770.) Roman apparently merges the timeframe as to Canales. 11 Roman and Loza are first cousins. (20:2929.) 12 Located generally in the center of the plant (R. Exh. 13), the cafeteria is up a flight of stairs. period. (21:3092, 3105–3106.) With one of the largest de- partments, Stoner’s secondary machining department has some 12 to 20 regular employees plus temporary transfers from other departments. (1:89.) Valenzuela (3:544–546) and Antonio Aguilera (9:1350– 1351) assert that employees would go out to their cars as much as 10 minutes before shift’s end, while Antonio Rami- rez describes the practice as the last 5 minutes, which is scheduled for personal cleanup. (10:1434, 1436.) Howard McArtor, the night-shift supervisor, described his pre-October 1987 practice as more restrictive. He permitted employees to go start their cars during their personal wash- up time, the last 5 minutes, on those nights when the tem- perature dropped below zero. (19:2765, 2795.) McArtor’s employees number between 20 to 35 employees, generally around 22 to 25, but about 35 in late 1987 and early 1988. (2:208; 19:2754a.) Jim Scott’s ADC, with some 26 employees, is one of the three largest departments. (1:78; 21:3214.) As early as the July 1985 meeting with his employees, then numbering 32 (23:2399), Scott told his employees not to leave the building before 4:30 p.m. So testified Juan San Roman, one of Scott’s setup leaders and, beginning February 15, 1988 (20:2910, 2957), his assistant supervisor. (20:2922.) Scott describes his control after September 1985 as rather successful, with his being required merely to talk to any employee who occasion- ally would disobey one of the rules. (23:3402–3403, 3404, 3432.) Roman describes things a bit differently. Roman testified that conditions temporarily improved, with the warned employees no longer leaving as a group, but that the usual stretching of the rules continued with Scott ‘‘en- forcing’’ the rules by talking to the violators. (20:2920, 2946, 2978.) At one point Roman asserts that conduct even got worse after the warnings. (20:2979.) Although initially de- scribing violations of time rules as ‘‘common practice’’ and a ‘‘problem’’ in the 2 years after July–September 1985 (20:2943–2944, 2947), Roman later named three employees as the principal violators: Rodolfo Banales,9 Marcial Canales,10 and Marco Antonio Preciado Loza.11 (20:2948– 2952.) Scott would orally warn them (20:2949, 2952), but apparently without effect. From September 1985 to after the October 16, 1987 election, Scott issued no written warnings. When asked what good it does to repeat oral warnings never imposing progressive discipline for 2 years, Roman answered that he is not sure. (20:2952.) Before Balma’s April 1, 1987 return, employees appar- ently were substantially free between breaks to obtain coffee in the cafeteria and take it to their work stations. Similarly, some would leave their work stations to place their lunches on the furnaces, before breaks or lunch, and then return to work. Balma testified that when he arrived in April 1987 he observed that employees came and went as they pleased, fre- quently ignored admonitions from supervisors, and made fre- quent trips to the cafeteria during worktime to take food and drink to their work stations. It was ‘‘chaos’’ with day em- ployees from the departments of Larry Stoner (secondary ma- chining), Ron Adamczyk (precision machining, waveguide), and Santi Greco (zinc die casting) being the principal abus- ers. (24:3710, 3715–3716.) Employee testimony is that em- ployees would stop work and begin cleanup, not at the post- ed time of 4:15 p.m., but at 4:10 p.m. or even 4:05 p.m., leaving them standing by the clock for several minutes be- fore punching out at 4:30 p.m. Indeed, Balma testified that employees would begin standing by the timeclock at 4:20 p.m. waiting to punch out. (24:3715; 25:3847–3848.) During his first week back, he testified, Balma spoke to the supervisors individually about the disregard of posted times. (24:3726.) He told them to have their employees let them know when they were leaving the area, even to go to the restroom/washroom, although he did not tell them to re- quire employees to request permission to go to the restroom. He also said they should tell employees not to leave the plant early to start car engines. (24:3736–3738; 25:3849.) Over the next few weeks Balma apparently concluded that his talks with the individual supervisors were not yielding the results he desired, so he raised the subject in a general meeting he had with the supervisors on April 28 in the cafe- teria. That was a general meeting covering several topics, es- pecially safety. The specific date of the meeting derives from a memo (R. Exh. 21) of that date which Balma posted to all plant employees respecting the necessity to wear safety glasses. A steady parade of employees going up to the cafe- teria12 to get coffee prompted Balma to reurge the super- visors to enforce the rules. (24:3726–3729.) Observing thereafter that improvement was only slight, Balma went to Novak about the matter. (24:3730–3731.) As- sistant Supervisor Don Basgall testified that in May 1987 Balma and Novak held a group meeting in the front office with the supervisors and assistant supervisors. Balma and Novak alternated in telling the supervisors to enforce the rules. (20:2998–3001.) Basgall went to each of his (4 to 10) employees and informed them that Balma and Novak wanted the rules enforced. This seemed to have a beneficial effect, Basgall testified, but when occasionally one of his employees deviated from the time schedule Basgall would remind him to comply. Basgall issued no written warnings. (20:3001– 3003.) Around late May to early June 1987, Balma testified, Tony Girone, Lovejoy’s president, visited Acme. In his tour of the plant with Balma, Girone observed cans and coffee cups, a hot plate, and two old refrigerators (where employees kept their food) near the maintenance area. As a result of Girone’s visit, Balma told the supervisors Girone did not like what he saw. Acme replaced the refrigerators with a new one, which was put in the cafeteria. Acme ordered two new microwaves, paying for them in July (R. Exh. 129), and placed them in the cafeteria for employees to heat their food. (24:3744– 3749.) One factor which creates some confusion in the record is the timing of a visit by an OSHA inspector, Nars Serges. Several witnesses place the visit in 1987. Balma had brief difficulty with the year, whether 1987 or 1988, before fixing it as 1988. The inspector’s visit was in May 1988 with Balma prevailing at a subsequent informal hearing in OSHA’s district office. (24:3739–3743; 25:3842–3843, 1095 LOVEJOY INDUSTRIES 13 The die cast furnaces contain molten metal simmering at 1220 degrees Fahrenheit. (21:3199, Scott.) Moisture from drink or food falling into the molten metal can cause an explosion, and Scott de- scribes two bad injuries he saw happen at a previous employer when moisture reached the molten metal. (34:3405–3406.) 3851.) After checking the papers, Respondent’s counsel rep- resented that the (informal) hearing was conducted in July 1988. (25:3841–3842.) Other witnesses, erroneously it ap- pears, link some of their testimony to the OSHA inspection, but place the OSHA visit in May-June 1987. It was right be- fore Balma talked to them in about June 1987, or shortly after his arrival, on not allowing employees to have food or drink at the furnaces because of the danger. (23:3407–3408, Scott.)13 Balma is a bit unclear respecting results after the Novak- Balma meeting with the supervisors. He asserts there was some improvement. (24:3732.) Stoner, however, would only talk to employees, rather than issue them written warnings, and Balma thinks Stoner therefore was unable to correct many of the problems in his department. (24:3732.) Indeed, Stoner concedes that he has never issued a written warning for abusing the time schedules. Explaining that his style is to talk with the employees, Stoner testified that any em- ployee he would have to talk to usually would straighten out. (21:3089–3090.) Ron Adamczyk, who then supervised the precision ma- chining department (PMD), gave rather confusing testimony concerning the months after Balma’s arrival. Initially, Adamczyk claimed that his employees generally observed the rules, with only an occasional need for him to remind any- one. (20:2838.) Balma apparently saw things differently be- cause after Balma complained that Adamczyk’s employees were abusing the rules, Adamczyk concluded that his em- ployees were getting out of hand. According to Adamczyk, he began to clamp down, telling his employees to comply with the rules, to let him know where they were going, and that they were not to leave the building (to start their cars) before 4:30 p.m. (20:2840–2841, 2845.) Adamczyk’s testimony is very mushy. Nevertheless, be- tween Novak’s arrival and the October 16, 1987 election, only Adamczyk, so far as the record shows, issued a written warning for failure to comply with the rules. Dated October 7, the warning is to Juan Lopez for ‘‘Leaving work without permission,’’ and reads in the ‘‘Remarks’’ section (R. Exh. 3): On October 6, 1987 you left your work area, and were seen sitting in your car at 4:20 p.m. You are not to leave the plant until you punch out at 4:30 p.m. If this infraction should happen again you will be suspended for three days without pay. It was Novak who observed Lopez as Novak returned from a sales meeting around 4:10 to 4:15 p.m. He called Adamczyk and instructed him to go advise Lopez that a writ- ten disciplinary warning would issue (24:3615–3616). Adamczyk complied. (2:243; 20:2847.) Lopez offered no ex- cuse or claim that he was exercising a practice elevated to a recognized privilege. (20:2848.) The Lopez matter was but one incident in what Balma and the supervisors observed in the weeks before the October 16 election as a renewed disregard of the rules, principally by union supporters such as Nicolas Valenzuela and Marcial Canales. Recalled that the Union’s letter demanding recogni- tion is dated August 13 (G.C. Exh. 32) and that the Union filed its representation petition on August 17 (G.C. Exh. 33). Balma told the employees that they were spending too much time away from their work stations. (25:3834.) During this same general period Balma observed that Valenzuela was frequently away from his work area. When Balma told Valenzuela he should work at his machine rather than wander over all the plant, Valenzuela merely shrugged and laughed. Although he did not give Valenzuela a written warning, Balma told Stoner to watch Valenzuela. (24:3775; 25:3876.) Stoner did so, observing that Valenzuela was walk- ing around more than was necessary. (21:3093.) Con- sequently, Stoner and Balma called Valenzuela in on Sep- tember 28, 1987. To Valenzuela’s claim that he was looking for parts, Stoner said that such excuse was unacceptable be- cause he too often was away from his work area. Stoner warned that corrective action would follow if Valenzuela did not cease his wandering. Stoner prepared an internal memo (R. Exh. 79) covering the incident. As the memo asserts, and Stoner testified (21:3126), Valenzuela was told (by Balma) that the oral warning had nothing to do with his union activi- ties. Balma testified that Acme, under former President Hagner, had developed a strong reluctance to correct Valenzuela for fear he would again go to the Labor Board and embroil the Company in another litigation hassle. (24:3770.) The September 28, 1987 oral warning (documented by file memo) was Acme’s first real effort since April 1987 to cor- rect Valenzuela’s wanderings. However, only a few days after Balma’s arrival he issued Valenzuela a written warning (R. Exh. 12, Apr. 6, 1987) for ‘‘Reporting to work under the influence of alcohol.’’ Balma asserts that he actually cannot say Valenzuela was intoxicated, but he found Valenzuela in the washroom. (24:3770; 25:3875.) Valenzuela asserts he had a hangover for Sunday overtime and extended his break by 10 minutes that Sunday morning. (4:695; 7:997.) Acme pe- nalized Valenzuela by removing his eligibility for weekend overtime the next 4 weeks. (4:696; 7:996.) Around early October, Adamczyk, observing the renewed disregard by employees, told his employees that they were to comply with the rules, let him know where they were going, and to observe the posted time schedule. (20:2848– 2849.) In the last 4 weeks or so before the election, Scott ob- served that employees from other departments renewed com- ing into ADC and interrupting the work of the employees there by talking with them. He particular noted, as frequent visitors, Valenzuela and Canales. Scott sent them out and re- ported the matter to Larry Stoner. Eventually Scott was suc- cessful in keeping them out during working time. (23:3409– 3410, 3495–3498.) Juan San Roman recalls that during the same time period several employees were walking around. From other depart- ments he named Mauricio Aguirre (Adamczyk’s packing, 11:1608–1611), Marcial Canales (waveguide under Adamczyk, 18:2627; 20:3859), Antonio Aguilera (packing, 1096 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 14 The supervisory status of Sam Murchison after March 20, 1987, when Ferguson replaced Murchison as quality control manager, is disputed. I need not resolve that dispute here. 15 ‘‘Wise chieftans realize that unduly harsh or unnecessarily lax discipline will undo the morale of their Huns.’’ W. Roberts, Leader- ship Secrets of Attila The Hun 37 (1985). Sam Murchison/Robert Ferguson 8:1292; 9:1355–1356),14 Nicolas Valenzuela (Stoner, 3:460), plus Rodolfo Banales from Jim Scott’s ADC. (20:2925–2926.) Scott testified that for 2 to 3 weeks before he issued a written warning (R. Exh. 96) on October 23, 1987, to Banales (or since about 2 weeks before the election), Banales would go to the vending ma- chines, obtain food and return with the food to his machine. Banales would do this three to four times a day, even after returning from breaks. Scott confronted him on several occa- sions. Banales would just smile, laugh, and say, ‘‘Cool down. Take it easy.’’ (22:3274–3279; 23:3461–3470.) Scott asserts that Banales, in this timeframe, was the only em- ployee who persisted in ‘‘excessively’’ leaving his machine to buy food and drink, the only one in the department ‘‘abus- ing it that much.’’ (23:3472.) Scott insists that the election a week earlier had nothing to do with the warning, and that what motivated the warning was Banales’s lack of coopera- tion. (23:3741.) Balma testified that, before the election, Scott had come to him about Banales. Balma went to Banales and spoke to him about his leaving his machine. He asked Banales how he would like it if Acme docked his pay 15 to 20 minutes a day. Banales said to go ahead, he did not care. Balma told Banales that he had to obey instructions, and if he did not do so he would get another warning and be suspended. (24:3750–3752; 25:3852–3855.) The General Counsel neither alleges nor argues that the October 23, 1987 warning to Banales violates the Act, and during his own testimony Banales is not asked about this warning or the circumstances. Despite Balma’s reference to a possible suspension, there is no evidence Banales was sus- pended on this occasion, and the warning document makes no mention of a suspension. It does state that Banales is being warned for a ‘‘Failure to obey instructions’’ and for ‘‘Excessive breaks.’’ Scott and Balma signed the form, but Banales refused to sign. In the remarks section Scott wrote (R. Exh. 96): Having been verbally warned many times, Rodolfo con- tinues to leave his work station between break periods. Many times this occurs less than 1 hour after break time is over. Employees clearly understood the posted time schedules. For example, Antonio Ramirez testified that before the elec- tion the two breaks were (scheduled for) 10 minutes each but that some employees (in Stoner’s department) sometimes would leave 1 to 2 minutes early. (10:1434, 1452.) Leon Bonner (McArtor’s night shift) testified that before the elec- tion breaks were 10 minutes, but little fuss was made when an employee returned 3 minutes late. (14:2034.) Even Valenzuela acknowledges the two breaks were for 10 min- utes each. (19:2742.) Indeed, in one of the IUE’s campaign leaflets, an August 25, 1987 (3:466; 7:714) listing of 18 items the employees wanted, the 10th reads (R. Exh. 4): 10. We want time to clean up/wash up and more break time. The penultimate statement near the bottom asserts, ‘‘AFTER WE WIN THE ELECTION, WE CAN NEGOTIATE FOR THE CHANGES WE NEED.’’ 3. Factual conclusions about preelection status of rules Resolving all the evidentiary conflicts and inconsistencies about the status of the rules probably would be impossible. Witnesses for both sides overstate in promoting their view of the facts. When the exaggerations are disregarded, however, some basic facts are clear. First, Acme had work rules based on a posted time schedule. Second, through the years, en- forcement of the rules has been inconsistent, largely depend- ing on individual supervisors, with no positive enforcement policy coming from the plant’s senior management. Third, because management’s lax attitude toward enforcement tend- ed to undermine the efforts of individual supervisors, even those supervisors, such as ADC’s Jim Scott, who tried to en- force the rules had to accept violations which did not become ‘‘excessive’’ or abusive.15 Fourth, accepted conduct (violations) include employees’ going to the cafeteria between breaks to get coffee and to carry that coffee back to work stations; to leave work sta- tions a few minutes early to heat food in the microwaves (principally on the furnaces before July 1987); and to bring heated or other food to the work stations. To a lesser extent, and with more limitation among individual supervisors, em- ployees could leave the building 5 or even 10 minutes before shift’s end in cold weather to warm up their car engines, and then return to punch out at shift’s end. To an even lesser ex- tent, particularly after Novak and Balma spoke to the super- visors in May 1987, some employees continued to stop work and begin cleanup as early as 4:05 p.m. Finally, permission was never required to go to the rest- room. Although some supervisors had asked their employees to let them know when they were going to the restroom, or otherwise leaving the work area for parts, that was honored in the breach rather than in the observance, so long as the absences did not become lengthy or frequent. As some employees, particularly Nicolas Valenzuela (Stoner, supervisor), but also Marcial Canales (Adamczyk, supervisor) and, in October, Rodolfo Banales (Scott, super- visor), as well as others, abused the rules, especially after mid-August 1987, why not did Novak and Balma crack down? Leroy Hagner, Acme’s former president, and Harold Georgeson, the former plant manager, may have followed an ostrich-like philosophy, but Novak and Balma profess a law- and-order philosophy. It surely went against their grain to see abuses and not decisively stop them. I credit Balma only partially respecting his claim that em- ployees substantially followed the rules after May–June 1987. Instead, I find that while most employees had im- proved by July 1987, there were still abusers, particularly so after the Union filed its representation petition in mid-Au- gust. Even by Balma’s own account, abuse—particularly by some of the union leaders and their supporters—became pro- nounced by mid-September. Yet during those last 4 weeks, or during the whole summer for that matter, the only written warning (R. Exh. 3) that issued by manager or supervisor for 1097 LOVEJOY INDUSTRIES 16 Stoner’s presence at the meeting can be inferred based on his later telling his employees that Novak did not want employees leav- ing the building before 4:30 p.m. (21:3104, 3153.) violation of the work ‘‘rules’’ came at Novak’s instruction on October 7 when Novak observed Juan Lopez (Adamczyk, supervisor) sitting in his car around 4:15 p.m. On September 28 Balma and Stoner had merely orally warned (R. Exh. 79) Valenzuela against wandering through the plant during worktime. Why no general crackdown in September and October 1987? The answer must be inferred from the record. First, as Novak tells us, Acme won the two previous elections. Acme’s obvious strategy was to avoid rocking the boat un- necessarily—particularly if to do so (antagonizing employ- ees) by issuing written warnings, even to abusers, would be to fall into a calculated trap. Instead, remain calm, avoid an- tagonizing the employees, and win the third election. Second, any discipline respecting Valenzuela would have to be well documented because of the prior litigation which Acme lost. Therefore, a documented oral warning—rather than a written warning—to Valenzuela was appropriate. Acme lost the elec- tion. Shortly thereafter (the time is disputed), Balma or the supervisors spoke to the employees assembled by depart- ments. 4. The meetings of Monday, October 19, 1987 a. Introduction The election was conducted on Friday, October 16, 1987. Although the election’s hour is not disclosed in the record, presumably it would have covered the shift change (thus, about 4 to 5 p.m.) per the General Counsel’s guidelines. 2 NLRB Casehandling Manual 11302.3 (Apr. 1984). At some point in the days following the election, the par- ties agree, Acme’s supervisors, in some instances with Balma, spoke to each department about enforcement of the posted schedule and work rules. Principal issues in dispute are (1) the day this was done and (2) what was said about certain matters such as whether permission was needed to go to the restroom. Timing is of some significance. Most of the General Counsel’s witnesses who testified on these depart- ment meetings assert they were held early the Monday morn- ing following the election, or October 19, 1987. Acme’s wit- nesses on the subject offer a wide range of possibilities— from 1 to 2 days up to 2 weeks later—but generally focus on about a week later. Clearly Acme’s meetings announce an enforcement of the work rules. The Government contends that the early Monday timing was calculated to inform the employees they were being punished for voting for the Union. (Br. at 14.) Re- spondent Acme argues (Br. at 79, 300) that the announced reiteration of the preexisting rules was simply a response to unbridled aggression of a group of union supporters which, before work in the first days of the week after the election, strutted through the plant chanting, ‘‘We are in control! We are in control!’’ The group’s aggressive behavior was in ad- dition to a marked increase in employee disobedience of the work rules. There were two types of meetings, actually. First, Novak and Balma met with supervision. Whether that meeting oc- curred on election day following the tally of ballots, over the weekend, or before the 7 a.m. start of the Monday day shift, I need not resolve. The inconsistent testimony of Respond- ent’s witnesses, and the failure of some of the supervisor witnesses to address the topic, suggest that the meeting oc- curred early that Monday, October 19, and in the presence of only some supervisors and perhaps no assistant super- visors. This is so because, I find, the supervisors met with their employees that Monday morning, October 19, and Novak and Balma had already held their own meeting with the supervisors. b. Novak and Balma meet with the supervisors Novak testified that he and Balma met in the inner con- ference room with the supervisors and assistant supervisors the Thursday (Oct. 22) following the election to ‘‘regroup,’’ to let the supervisors know that the loss was not the end of the world, and to enforce the rules in the face of the ‘‘We are in control! We are in control!’’ strutting and chanting. (23:3526–3528; 24:3662–3664, 3696.) According to Novak, he held off until that Thursday before holding the meeting (he does not recall the hour) because he did not want to overact. (23:3527; 24:3696–3697.) Three of the four produc- tion supervisors who testified did not address this meeting: Ronald Adamczyk, Howard A. McArtor Jr., and Larry Ston- er.16 Two assistant supervisors testified, Daniel Basgall and Juan San Roman. But Roman was not promoted to assistant supervisor until February 15, 1988. (20:2910, 2957.) Before that he was a setup person. (20:2924.) Basgall, who became an assistant supervisor in April 1987 (20:2990), testified that he did not attend. (20:3005.) Balma testified that in the first couple of workdays after the election (apparently meaning Monday and Tuesday), a group of up to 15 employees, including Marcial Canales, Nicolas Valenzuela, and Mauricio Aguirre, before start of the shift, shouted over to Balma, Scott, and another supervisor, ‘‘We have got control now!’’ and ‘‘We are in command now!’’ Balma recalls Canales shouting the words. During these days Balma also observed a marked deterioration in ob- servance of the rules. (25:3835–3836, 3899–3901.) ADC Supervisor Jim Scott testified that in the 3 to 4 workdays following the election employees would march around chanting ‘‘We won.’’ (23:3411.) Juan San Ramon, a setup person in 1987, testified that after the election employ- ees, mostly from ADC’s trim press area, began leaving their work stations more often, remaining away longer, spending time in the cafeteria and in the aisles, generally talking and wandering around and cleaning up early. (20:2953, 2956– 2957, 2982.) In fact, some employees told Roman they want- ed to sabotage Acme by having friends and relatives call the Company so as to tie up the telephone lines and to keep Acme from receiving orders. (20:2927, 2955.) Nicolas Valenzuela testified that after the election union supporters went to the home of Canales where they cele- brated. When they returned to work on Monday, they were still celebrating their victory. (19:2743–2748.) Although he denies the ‘‘We are in control’’ accusation, Valenzuela con- cedes someone could have said it. (19:2748–2750.) Marcial Canales denies that it happened. (18:2672.) Balma testified that he went to Novak about the problem in the latter part of the week (after the election), and that they then called in the supervisors for a meeting. (25:3836– 3837,3901.) The supervisors were told not to be intimidated, 1098 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 17 The exception is Leon Bonner who testified that Balma spoke to McArtor’s second-shift employees about a week after the election. (14:2032.) that business had to continue, to bring order to their depart- ments, and to enforce the existing rules. (25:3837–3838, 3902.) Supervisor Adamczyk is uncertain about the time of the meeting, placing it anywhere up to 2 weeks after the elec- tion, and he is unsure of whether Novak was there. (20:2893–2894.) c. Supervisors meet with employees (1) On Monday, October 19, 1987 Balma asserts that the supervisors were not told to hold group meetings with employees in their departments, that it was left to their discretion. (25:3901.) Most exercised their discretion to hold such a meeting. Whether McArtor elected not to do so and Balma then stepped in, or whether Balma preempted McArtor, is unclear. Balma did meet with McArtor’s employees at the beginning of their second shift in the first day or two in the week following the election. Balma, McArtor testified, told the group he had talked with the day shift. (19:2768, 2800.) The day-shift employees Balma met with were Scott’s ADC employees. Scott testified that he met with his ADC employees immediately after the meeting with Novak and Balma. (23:3412.) Balma recalls Scott’s meeting as being 1 or 2 days later and that he participated at Scott’s invitation. (25:3902.) In contrast to the generalized timeframe given by the man- agers and supervisors for the enforce-the-rules meeting with the departments, the testifying employees are nearly unani- mous in fixing the date as the Monday after the election, or October 19, 1987.17 Mauricio Aguirre testified that Adamczyk held it for the packing employees about 7 a.m. that Monday. (11:1610–1611, 1636, 1649.) Raymundo Aguirre (8:1172) and Sacramento Olivares (11:1495) so date the Scott-Balma meeting. Larry Stoner spoke that Monday morning to his secondary machining department. So testified Nicolas Valenzuela (3:528; 7:1059–1060) and Antonio Rami- rez. (10:1427, 1448.) Antonio Aguilera testified that in October 1987 he and four other employees worked as packers under Sam Murchi- son in the quality control department (8:1288–1289, 1292, 1301, 1303; 9:1345), and that on Monday, October 19, Mur- chison held a meeting with the packers around 7:05 or 7:10 that morning. (8:1288–1289; 9:1383, 1387.) Robert Ferguson was not present, and Murchison spoke without referring to any notes. (9:1411.) Recall that on March 20, 1987, Acme posted a notice (R. Exh. 27) that effective ‘‘immediately’’ Sam Murchison ‘‘had been assigned to special duties,’’ and ‘‘Robert Ferguson is our new Quality Control Manager.’’ The status of Murchison after that notice as a statutory super- visor or agent is in dispute, with the Government alleging and arguing that he is and Acme that he is not. Balma testi- fied that Murchison was given some menial jobs to keep him busy, and that he had no responsibility over employees after Balma became plant manager. (24:3705–3706.) Murchison, who apparently retired about June 1988 (9:1404), did not tes- tify and neither did Ferguson. I need not address the question of supervisory status if the evidence shows that Murchison was an agent. I find that the evidence does show apparent agency. Even after the March 20 notice Murchison, not Ferguson, continued to give work orders in the department. After his April 1 arrival, Balma would issue some orders to the employees there, but, as Aguilera testified, the bulk of the orders came from Murchi- son. (8:1292, 1298; 9:1406–1407.) Through Murchison also worked alongside Aguilera and the other employees (8:1302– 1303; 9:1416), he continued wearing street clothes as Balma and Novak did rather than a uniform as the employees and supervisors such as Scott did. (8:1313–1315.) When Aguilera requested a vacation, or leave of absence in June 1987, he went to Murchison. (8:1310–1312; 9:1407.) Balma testified that employees usually leave their vacation slips on his desk or forward the slips to the payroll clerk through the super- visor. Balma concedes that on this occasion Murchison brought him Aguilera’s request and that he did not tell Mur- chison he should not be involved in the matter. (25:3846.) Balma testified that Murchison was not at the meeting Novak and Balma held with the supervisors (25:3837), that Murchison worked directly for Balma, that, if anyone, he would have been the one to tell Murchison to hold a discus- sion with the employees he works with, that he gave no such instructions (25:3838), and that he knows of no reason Mur- chison would have held a meeting with a group of employ- ees. (25:3903.) Crediting the testimony that Murchison did hold such a meeting, I find that Murchison did so under ap- parent authority and that the employees reasonably would view Murchison as speaking for Acme. For months Acme permitted Murchison, after his removal as the QC manager, to function in the capacity of an agent either for generating or relaying work orders to department employees. Having clothed Murchison with the cloak of apparent authority, and having for months accepted the benefits of Murchison’s ap- parent agency, Acme is in no position now to cry ‘‘unfair.’’ Accordingly, I find that on October 19, 1987, Murchison spoke as the apparent agent of Acme. (2) Conflicting versions Respondent’s evidence is that in general Balma and the su- pervisors simply reminded employees of the preexisting rules observed that employees had been violating the rules, told them that the violations must cease, and at least in one meet- ing, informed them that written warnings would issue to those who violate the rules. Serving as an interpreter for Ronald Adamczyk in both his packing and waveguide sec- tions was Salvador Ontiveros. (20:2852, 2877.) For inter- preters in his department Larry Stoner used Faustino ‘‘Tino’’ Ontiveros (20:3006; 21:3103) who, as Dan Basgall, is an as- sistant supervisor for Stoner. (21:3077.) Balma and Scott, ac- cording to Sacramento Olivares (11:1495), enlisted Juan San Roman, setup, as their interpreter. When Balma spoke later that day to McArtor’s second-shift employees, no interpreter was used, McArtor testified. (19:2769.) The General Counsel’s witnesses testified either that Balma or his supervisors spoke of enforcing rules or else an- nounced that new rules were being imposed. Called as the last witness during the General Counsel’s case-in-chief, Leon Bonner, one of McArtor’s night-shift employees, testified that Balma told their group that the posted rules would now be enforced. Balma ran down the list: breaks were 10 min- 1099 LOVEJOY INDUSTRIES 18 The complaint erroneously lists the March 24 delivery date as the service date. Service of a mailed charge is effective on mailing. 29 CFR § 102.112, 113(a). 19 As earlier noted, although NLRB Region 13 had issued an order consolidating the outstanding complaints, no consolidated complaint, as a single document, issued. utes with no extra minutes, no more warming food in the cafeteria (on the microwaves) before the lunch period, no be- ginning cleanup before 2:45 a.m., or going out to start cars before the 3 a.m. end of the shift. (14:2033.) Antonio Aguilera testified that Sam Murchison told that small group of employees that from then on if anyone had to leave the area, including going to the washroom, he must first notify Murchison. Food no longer could be brought to the work station and food could not be heated before lunch. Cleanup must start at 4:15 p.m., and employees no longer could leave before 4:30 p.m. to start their cars. An oral warning would be given for a first violation, a written warn- ing for the second, and a 3-day suspension for the third. (8:1288; 9:1345.) For awhile Aguilera informed Murchison whenever he left the area, but then he ceased the practice (9:1400–1402), apparently with no repercussions. Mauricio Aguirre testified, essentially, that Adamczyk went down the same list. A wrinkle here is that Aguirre claims Adamczyk said the employees would have to obtain permission before going to the restroom. Before the election the employees under Adamczyk did not even have to notify anyone before leaving for the restroom. (11:1611–1614.) Adamczyk testified that he ‘‘brought up’’ that the employee were not to leave the area without notifying him, and that he had told them of this and the other rules in early October. (20:2848–2851.) Adamczyk does not specifically address the matter of going to the restroom, or asking for permission to go there, except to say that after the election employees were still leaving their work stations, walking around, and going in groups to the restroom/washroom. (20:2891–2892.) Raymundo Aguirre testified that Balma told Scott’s group they must get permission before going to the restroom/washroom. (8:1173.) After Scott went through the hassle of that for about a month before employees just stopped asking, Scott said to forget it. (8:1206–1207.) Sac- ramento Olivares testified that Scott told the group on Octo- ber 19 that ‘‘new’’ rules were being imposed since the Union had won the election. In listing the now strict enforcement of all posted times, Scott said permission was needed before going to the washroom. Employees were not even to get a drink of water before breaks. Balma added that if anyone did not like the rules he could leave. Before the election no per- mission was needed for getting a drink or going to the rest- room, Olivares testified. (11:1495–1497.) Scott testified that he gave no change in the rules, but sim- ply told his group that even though the Union had won the election there still were rules to follow, that the rules had not changed, and to let him know when they were leaving their work stations or the area. Balma, Scott said, repeated this, adding that employees were going to the washroom in groups and staying there a long time, and to let Scott know when they are going to the washroom. (23:3412–3413, 3435.) Balma denies telling any employees or supervisors that an employee first needed permission to go to the rest- room. (24:3737; 25:3840.) Beginning in April, however, he had begun telling the supervisors to have their employees tell them when they were going to the restroom. This was neces- sitated, Balma explains, by the number of employees walking away from their areas, and throughout the plant, leaving their machines running, and the supervisors unable to tell Balma where a missing employee was. (24:3737.) A setup person in October 1987, Juan San Roman, assist- ant supervisor since February 15, 1988 (20:2910, 2957), tes- tified that Balma and Scott spoke to Scott’s ADC employees (20:2927, 2956), Balma told the group that they were wan- dering too much and spending too much time away from work drinking coffee. Balma said the employees were to obey the rules. He reminded them that if they had to go to the washroom or to the cafeteria they were to let Scott know. Starting cars would have to be after shift’s end at 4:30 p.m., and cleanup must start at the 4:15 p.m. time as posted. (20:2928–2929.) Nicolas Valenzuela claims Larry Stoner told his group Acme had ‘‘new’’ work rules which included obtaining per- mission before going to the restroom. (3:529; 7:1061–1062.) Antonio Ramirez does not report Stoner describing any rules as ‘‘new.’’ Although Ramirez, on direct examination, quotes Stoner as telling the employees they must ask for permission when going to the restroom (10:1431), on cross-examination he asserts that Stoner said they would have to tell either him, Faustino Ontiveros, or Dan Basgall before leaving for the restroom. (10:1451.) Dan Basgall testified that Stoner said nothing more than what Basgall told his section after the May 1987 meeting between Novak/Balma and the super- visors. (20:3007.) Even so, Stoner concedes that the matter of not being able to warm up cars before shift’s end was a change, although the only change, and that he told the em- ployees Novak wanted that stopped. (21:3104–3105.) Novak, Stoner testified, had been harping to the super- visors about the car-warming matter ever since he took over months earlier. Nevertheless, Stoner concedes, this meeting was the first time he had told his employees the practice had to stop. (21:3153.) 5. Acme’s 10(b) defense The original charge (G.C. Exh. 1a) in Case 13–CA–27619, filed March 18, 1988, and served March 23,18 alleges viola- tions of Section 8(a)(1), (3), and (5) of the Act. The rules allegation charges that on or about October 19, 1987, and at subsequent dates, Acme ‘‘instituted unilateral changes in work rules in order discriminate against employees for union activity and for voting for unionization in an NLRB election. The Employer failed to bargain over these changes and con- tinues to fail to bargain.’’ (G.C. Exh. 1a.) The June 1, 1988 complaint in Case 13–CA–27619 (G.C. Exh. 1d, the first complaint) alleges, in paragraph viii(a), as unilateral changes of October 19, 1987, substantially as is done in paragraph 11(a) of the trial complaint (G.C. Exh. 1zz), quoted earlier. One difference is that the fourth item of the first complaint (meal breaks reduced from 40 minutes to 30 minutes) was deleted along the way and replaced by the current fourth item (employees no longer permitted to warm up food prior to lunchbreaks). That substitution came by way of a January 11, 1989 set of ‘‘Further Amendments To The Consolidated Complaint.’’ (G.C. Exh. 1oo.)19 1100 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 20 The written warnings to Ramirez and Olivares, complaint pars. 11(f)(1) and 11(g), are alleged as violations of Sec. 8(a)(3) and (5)— complaint par. 13 and 14. The seventh alleged change, quoted earlier, pertains to written warnings. In the first complaint the allegation is con- cise ‘‘(vii) Written warnings were issued.’’ (G.C. Exh. 1d.) The ‘‘Further Amendments’’ of January 11, 1989, expanded that allegation to read, as later incorporated into the trial complaint as paragraph 11(a)(7), ‘‘Employees received writ- ten warnings for conduct that was previously tolerated, and for conduct that previously only warranted an oral rep- rimand.’’ Acme’s January 16 answer to the ‘‘further amendments’’ listed Section 10(b) as its first affirmative defense and, as a second affirmative defense, a concept of excessive delay in making the amendments ‘‘when this information has been known and available to the Charging Parties and the General Counsel for many months.’’ (G.C. Exh. 1qq at 3.) Acme reit- erated its position at the beginning of the hearing (1:12–27) on these and several other allegations, and in its July 11, 1988 answer (R. Exh. 36 at 9; 11:1605–1606; 12:1662) to the trial complaint. (G.C. Exh. 1zz; 5:41; 6:797.) When brief- ing this general subject (Br. at 5–6, 289–291), Acme men- tions the food warning allegation but not the warnings allega- tion. The Government does not pause to brief the matter, but at the hearing the General Counsel observed that the allega- tions are not significantly different, arise from the same facts, and merely clarify. (1:27.) At the hearing I rejected the lateness, or excessive delay, ground on the basis Acme would have ample time to prepare its defense. The hearing opened on January 23, 1989, and the General Counsel did not rest the Government’s case-in-chief until August 1, 1989. (14:2040.) Acme cites no authority that delay alone should bar an otherwise proper amendment. Ad- hering to that ruling, I deny Respondent’s motion to strike or dismiss these two allegations based on the lateness de- fense. Turning to the first affirmative defense, Section 10(b), I note it is unclear whether I reserved ruling on that defense as to these two allegations as I did to others. I treat the mat- ter as if I did. Acme contends that the two amendments (I shall treat the matter as if Acme had not abandoned the warnings amendment) should be dismissed under the stand- ards of Nickles Bakery of Indiana, 296 NLRB 927 (1989), and Redd-I, Inc., 290 NLRB 1115 (1988). I find this defense to have no merit. The original charge specified unilateral changes on October 19, 1987. The ‘‘warm up food’’ amend- ment addresses the same group of facts, is based on the same legal theory, and Acme’s legal defense will be the same. Ac- cordingly, I deny Respondent Acme’s motion to strike or dis- miss this amendment. For the same reasons I make the same ruling regarding the warnings amendment. That amendment actually assists Acme by clarifying and focusing the allegation. I likewise deny the motion to strike that amendment. 6. Conclusions postponed Discussion and conclusions about the announcements of October 19, 1987, must be postponed until I summarize relat- ed allegations. Drinking coffee in the work area is indirectly related to an alleged prohibition against going to the cafeteria for coffee ‘‘during working hours,’’ and the January 1989 amendment about written warnings is directly related to a batch of allegations about specific warnings. I now turn to these other matters. E. Discipline Imposed Mid-November 1987 to Mid- March 1988 1. Introduction Additional incidents serve as the basis for allegations of both unilateral changes (8(a)(5) violations alleged) and dis- crimination (8(a)(3) violations alleged). In terms of time se- quence, the first incident after the election was written warn- ings in November 1987 for low production. A series of warnings, stretching into November 1988, is alleged under paragraph 11(f). Separately alleged are matters such as a fail- ure to grant scheduled general wage increases and obtaining or drinking coffee in the cafeteria. 2. Low-production warnings of November 11, 1987, to Antonio Ramirez and Fidencio Olivares a. Facts Machine operators Antonio Ramirez and Fidencio Olivares work in Larry Stoner’s secondary machining department (SMD) under, apparently, Assistant Supervisor Faustino Ontiveros. Ramirez and Stoner testified, but Olivares and Ontiveros did not. Ramirez put his hire date in about 1985. (10:1424.) The employment record (G.C. Exh. 3c) of Olivares discloses a hire date of April 30, 1985. Stoner thought they had been there longer, with Ramirez employed longer than Olivares. (2:364.) If lower clock numbers indi- cate longer seniority then Ramirez, with clock number 97 (G.C. Exh. 29), has worked at Acme longer than Olivares, with number 136 (G.C. Exh. 28). On November 11, 1987, Stoner issued written warnings to Ramirez (G.C. Exh. 29) and Olivares (G.C. Exh. 28) in a joint interview, with Faustino Ontiveros present.20 (21:3107– 3111.) Brief and direct, the warnings, with item 13 ‘‘Other’’ checked, read under Stoner’s remarks: Production not up to standard compared to other em- ployees working same machines under same working conditions. In his 21 years (as of his September 1989 testimony) at Acme, all as a supervisor over SMD (21:3077), the only written warnings for low production Stoner has ever issued were these November 11 warnings to Ramirez and Olivares (21:3116, 3152). Indeed, the only other written warnings Stoner has ever issued have been for some kind of physical violence (21:3153). In turn, these written warnings are the only ones either Ramirez or Olivares has ever received. (2:365–367; 10:1439, 1455.) Stoner prides himself on resolving employee production problems by talking to the individual and finding solutions to these problems. Sometimes he finds that an employee does not have the proper coordination for a type of machine. (21:3152.) At other times he uses innovative motivation, as with an employee making frequent trips to the restroom. For that employee, Stoner offered to move his machine into the restroom if he had a problem Stoner was not aware of. (21:3117.) Thus, over the years Stoner has had employees 1101 LOVEJOY INDUSTRIES 21 Stoner refers at one point to four machines needed to produce the part. (21:3138.) The reference is unclear. 22 Ramirez worked part of November 13, 1987, on a different op- eration and the hours are not allocated. Thus, the 899 is of limited value. 23 Although the figures shown for Ramirez are for drilling, the op- erations shown for Olivares switch from drilling, to milling, to tap- ping (threading). 24 Olivares worked from 7 a.m. to 2 p.m., first at milling and then at drilling, all on part 791. The hours are not allocated. who produced at significantly lower rates than rates Stoner considered standard. (21:3152.) Stoner asserts that he had his ways of motivating these employees to produce. His feigned sincerity in offering to move an employee’s machine to the restroom is one of his methods. (21:3117.) If there is a prob- lem, Stoner asks questions. ‘‘And if I can handle it my way, that is the way I do it.’’ (21:3117.) For Ramirez and Olivares, Stoner used written warnings because it was ‘‘just too obvious, too obvious.’’ (21:3117.) The parts were not flowing, so Stoner knew something was wrong because he has been around machines for so many years. (21:3117.) In fact, Stoner had noticed for several days that the production of Ramirez and Olivares was down and that parts were not flowing as they should have been. (21:3108, 3131–3132.) Stoner checked the production records of the other 75 per- cent of the day shift doing the job, plus that of the two night-shift employees doing it, and found that the night shift was out producing the day shift. Stoner identifies Robert Burris and Eusebio Hernandez as the night-shift employees. (2:366–367; 21:3140–3142.) Stoner testified that the job was ‘‘new,’’ employees had been working it about a month, no production standards were set, and, indeed, Acme was in the process of watching the job to set production standards. (2:265–366; 21:3108, 3138.) On this new part Ramirez and Olivares worked consecu- tively. That is, Ramirez would drill two holes on the front end, and pass it to Olivares who would drill two holes on the other end. Their production was linked.21 (2:365; 21:3137–3138.) The only production reports introduced are a few for Ramirez and Olivares. According to Stoner, the pro- duction reports are put into boxes and hence to he knows not where. (21:3132, 3143.) He approximates the production of Burris and Hernandez from memory. (21:3143, 3158.) He se- lected Burris and Hernandez for specific comparison because they operated, in the evenings, the same machines Ramirez and Olivares operated on days. (21:3144.) Stoner concedes that the night shift has better conditions and perhaps operates smoother than the day shift and it therefore is easier to produce more at night. (21:3145.) Moreover, Hernandez is one of Stoner’s top producers, perhaps the best at around 1200 to 1300 pieces a shift, with Burris apparently not far behind. (21:3143–3144.) The records introduced reflect the following production of good parts for Ramirez (R. Exhs. 28–32) and Olivares (R. Exhs. 80–1 to 80–5) for part 791: Ramirez: 670 785 976 1108 22899 Olivares:23 901 772 1024 1111 24961 Stoner testified that the production of Ramirez in the days before November 9 ranged around 670 to 700. (21:3132– 3133.) After observing the poor production of Ramirez and Olivares, and before analyzing production records, Stoner asked Faustino Ontiveros whether there was some problem with the machines. Ontiveros assured Stoner there was no setup or machine problem. In the opinion of Ontiveros, Ston- er testified, Ramirez and Olivares simply were not producing up to their capabilities. (21:3109, 3135, 3157–3158.) Com- puting from the figures on the table of production rates, for the 2 days of November 9 and 10 the production average for Ramirez was 728 and Olivares, 837. The morning of Wednesday, November 11, 1987, Stoner, with Ontiveros present to interpret, called in Ramirez and Olivares. Ramirez testified that Stoner said he was giving them a warning for low production. (10:1439–1440.) Stoner testified that he first mentioned their production problem, spread their production records out for them to see, and asked if there was some problem. Ramirez, who spoke in English without need for interpretation (10:1460; 21:3109, 3149), said that perhaps they had experienced a bad day. (21:3109, 3149.) Stoner said that was not a good excuse. (21:3109.) According to Stoner (21:3144), he also showed them the records of Burris and Hernandez. Ramirez testified that Stoner merely said that night-shift employees were pro- ducing more, and when Ramirez asked to see their records, Stoner declined. (10:1441.) I credit Ramirez concerning the matter of the night-shift records. Stoner told the two he was giving them a warning for low production. According to Stoner (21:3110), they said nothing. Ramirez, whom I credit on this point, protested that Stoner had not told them how many parts they were to produce, and that the number of parts expected should be posted on the machine. Stoner said the expected number would be pub- lished soon. (10:1440.) Ramirez testified that Stoner said the warning was not being given to him because of the Union. (10:1440.) At first denying that, Stoner eventually concedes he could have said it, although he asserts that the warning is unrelated to any of their union activities. (21:3116, 3150– 3151.) Stoner concedes he suspected that Ramirez was sup- porting the Union, although he had no idea about Olivares. (21:3116). I credit Ramirez that Stoner did say his warning was not for any union activities. Ramirez was active in the organizing (2:396; 10:1425), al- though it is not clear that he openly was so in the presence of any supervisor. Ramirez concedes that he never told Ston- er he supported the Union. (10:1450.) Following the election, on November 7, Terry Davis testified, the employees elected stewards and a negotiating committee. Ramirez was one of the five selected to the negotiating committee. (2:401, 3:431.) 1102 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD A November 17 letter (R. Exh. 1) from the Union to Acme named the stewards but not the negotiating committee. There is no evidence that Acme learned before the November 11 warning of Ramirez’ election as a negotiating committee member. QC inspector Nelson Diaz testified that before the election Olivares always wore a union cap even inside the plant. (14:2027.) Conceding that he did not give Ramirez and Olivares a specific number of these parts to produce (2:370), Stoner as- serts that employees know what to produce by simply work- ing all day. (2:370; 21:3139, 3141.) Although never commu- nicated to his employees, Stoner’s rule-of-thumb level of ac- ceptability is 80 percent of standard. (2:3139.) As no stand- ard had been established for this job, Stoner looked to the production count of others, particularly Burris and Hernan- dez. (21:3140.) According to Stoner, Burris and Hernandez were producing 1200 to 1300 pieces, or parts, in a 7 a.m. to 4:30 p.m. shift. (21:3134.) For purposes of figuring the 80-percent level in this instance, Stoner, apparently recogniz- ing the Burris/Hernandez level as one exceeding standard, pegs standard at close to 1100, or 80 percent of standard at about 880. (21:3247–3148.) Stoner testified that he could have lived with the 900 (R. Exh. 80–1) which Olivares pro- duced on November 9. (21:3145.) In the days before the November 11 warning, Stoner con- cedes he never asked Ramirez about his production and whether he had any problems. (21:3134.) Why not? Because, Stoner testified, he asked Ontiveros who said there were no problems and who expressed the belief that Ramirez and Olivares could do better. (21:3135, 3157.) Stoner therefore did not ask Ramirez because (21:3135): Because I felt there was something very belligerently happening. Now whether they had instructions of slow- ing down or something like this, I don’t know. Asked whether he felt Ramirez was part of the ‘‘bellig- erence’’ that was happening, Stoner replied yes. Asked why he thought that, Stoner refers to the low production figures. (21:3135–3136.) In short, Stoner concluded that both Rami- rez and Olivares intentionally were slowing their production. Oddly, Stoner asserts at this point that he does not believe the intentional slowdown was union related. (21:3150.) At the warning meeting Stoner did not ask Ramirez and Olivares if they were engaged in an intentional slowdown be- cause ‘‘I don’t need the hostility.’’ (21:3150.) Ramirez de- nies having any prewarning discussion with the Union about reducing his production level, denies that he and Olivares discussed such, and denies saying anything to Olivares about the topic. (10:1463–1464.) As the above table of production figures reflects, the pro- duction of Ramirez and Olivares increased substantially after Stoner issued the warnings the morning of Wednesday, No- vember 11. Ramirez concedes that conditions for that job re- mained the same. (10:1461.) Notwithstanding the figures, Ra- mirez asserts that he did not speed up after the warning. (10:1443.) Although Stoner observes that the figures would be even greater after November 11 by virtue of the 45 min- utes to an hour spent in the warning meeting (21:3147– 3148), Ramirez insists that the warning meeting lasted about 10 minutes. (10:1460.) As I already have found that Stoner showed no comparison production records to Ramirez and Olivares, the figure of 10 minutes fits better than the longer time given by Stoner. I credit Ramirez respecting the 10 minutes. On the other hand, I find that Ramirez and Olivares did increase their speed after they received their written warnings from Stoner the morning of November 11, 1987. b. Discussion (1) Section 8(a)(3) Outlining the Government’s prima facie case as to motiva- tion, the General Counsel lists these factors. First, Stoner ‘‘was aware’’ (suspected, actually) that Ramirez supported the Union. I credit Diaz that Olivares openly wore a union hat inside the plant before the election. I find that Stoner knew of this. Second, the General Counsel observes that the warnings issued within a month of the election. (Br. at 75.) Standing alone, that means nothing in view of the fact the job began about the time of the election or even a few days later. Third, the warnings not only involve a departure from Stoner’s normal supervisory practice of never issuing a writ- ten warning for low production but Stoner—contrary to his established practice—did not even talk with Ramirez and Olivares before he called them in. On this point Stoner gave confused and seemingly contradictory testimony. Stoner ini- tially claims that Ramirez and Olivares were working below capacity. Stoner concluded that the low production was relat- ed to the ‘‘belligerence’’ and, because of the low figures, Stoner concluded the low production was intentional. Stoner does not explain what he means by the ‘‘belligerence,’’ but I find it refers to the Union. However, I further find that the reference is not to protected activities, but to supposed un- protected activities of a work slowdown. For example, Stoner asserts that he does not know whether Ramirez had received any ‘‘instructions’’ to slow down. ‘‘Instructions,’’ I find, re- fers to the Union. Indeed, at the warning meeting Stoner felt a need to disclaim that he was issuing the warnings—the first for low production in his 21 years—because of any union ac- tivities by Ramirez or Olivares. Yet later in his testimony Stoner asserts that he did not think the slowdown, which he believed was intentional, was related to the Union. (21:3150.) Fourth, other factors in the prima facie case are the past capabilities of Ramirez and Olivares and this being their very first written warnings for low production. Indeed, Ramirez had never been given a written warning for any reason. Ston- er admits that Olivares is quiet, pleasant, and always smiling (21:3115). The probability is high that Olivares also had never received a written warning of any kind. Fifth, Acme’s failure to produce the production records of Burris and Hernandez, and the other day-shift employees doing the work, is darkly suspicious. According to Stoner he does not know what happened to the records after he placed them in a box. Not believing Stoner, I find that he knows exactly where the records are and that he testified falsely. According to Stoner, he counted as most comparable the production of Burris/Hernandez because they operated at night the machines Ramirez/Olivares operated on days. (21:3144.) Stoner nevertheless concedes that Burris/- Hernandez are at the top of the curve and that no one can do more than Hernandez. (21:3143.) In the face of this, Ston- er supposedly discounted, perhaps even disregarded, the records of the other 75 percent of the day shift who also 1103 LOVEJOY INDUSTRIES 25 Percentages are computed on the basis of 1108 for Ramirez, being an increase of 380 over his 728 2-day average (then dividing 380 by 728), and on the basis of 1111 for Olivares being an increase of 274 over his 2-day average of 837 (then dividing 274 by 837). 26 Although Stoner asked for their explanation at the beginning of the November 11 meeting, that was merely a formality to guarantee he had not overlooked something. It was not made as part of his 21-year practice of seeking a solution. worked on the job, on the asserted basis that Stoner did not consider the records of the other day-shift employees to be comparable because the machines of Ramirez and Olivares were not used. Nonsense! When Acme sets a production standard, it is set for an operation—not individual machines. In any event, Stoner testified with an unfavorable demeanor, and I do not believe him. Finding that Stoner testified falsely to hide the true facts, I find the truth to be that the records of the other 75 percent of the day shift would reveal produc- tion rates closer to that of Ramirez/Olivares than to those of Burris/Hernandez. The General Counsel established a prima facie case. Did Acme carry its burden, on the motivation issue, to persuade by a preponderance of the evidence that it would have taken the same action through Stoner even if Ramirez and Olivares had not been union supporters? No. Stoner, I find, would have followed his usual procedure of talking with the individuals, Ramirez and Olivares here, to ascertain if they were having a problem (if in fact they had a lower production rate than the other 75 percent of the day workers) and to seek a solution if they were. To Stoner, with his past practice of 21 years, it would have been unthinkable to have issued a written warning for low production without first try- ing to work with the individuals to solve any problem. Stoner, however, did issue a warning to each. Immediately the production of Ramirez and Olivares improved substan- tially. Thus, by the following day they were producing at the 1100 mark—a 52.2-percent increase for Ramirez and a 32.7- percent increase for Olivares over their 2-day averages for November 9 and 10.25 Why did it improve? Was it because they had been engaged in a work slowdown and now began producing normally? Ramirez denies this. Was it because they were frightened by their first written warnings and, in- nocent of any slowdown suspicion, began working at ram speed? Ramirez denies changing his speed and concedes that conditions did not change. (10:1443, 1461). Not crediting Ramirez, I find that he did increase his speed in order to raise his production. Whether the increase was beyond his normally acceptable effort, or merely up to his usual effort, is not shown by the record. These are among the matters which, I find, Stoner would have discussed with Ramirez and Olivares had he talked with them. He did not talk with them, however, because he believed they were in- tentionally slowing down. Although his explanation thereafter becomes confusing, seemingly linking the slowdown to the Union (‘‘belligerence’’), yet disclaiming belief that it was part of any union activities by Ramirez, the fact remains that Stoner issued the warnings without first exploring the matter with them, assertedly because he believed they were inten- tionally slowing down.26 I find that Stoner believed that the slowdown was connected to their support of the Union. The question then becomes whether, absent the perceived union connection, Stoner would have proceeded as he did or would he have conferred with Ramirez and Olivares without issuing warnings to them. Although no similar case is shown in the record, the evi- dence demonstrates that Stoner is willing to go to great lengths to motivate his employees. If Stoner, the morning of November 11, had believed that Ramirez and Olivares had been guilty of slowing their production as a misguided meth- od of expressing their desire for a pay raise, and had there been no union on the scene, what would Stoner have done? That scenario is the critical question. Would he have issued his first low production warnings in 21 years, or would he have orally warned them that their method was wrong and that they would have to bring their production up to stand- ard? Acme has failed to demonstrate that Stoner would have proceeded, absent union considerations, to issue written warnings. If anything, the evidence shows that Stoner would first have tried talking and discussing in keeping with his 21- year record of no written warnings outside of physical vio- lence. I find that Acme has failed to carry its burden. Ac- cordingly, I find that Acme, as alleged, violated Section 8(a)(3) and (1) by issuing the written warnings on November 11, 1987, to Antonio Ramirez and Fidencio Olivares. I shall order Acme to remove the warnings from their files and to notify them, in writing, that such warnings will not be used against them in any way. (2) Section 8(a)(5) As noted earlier, the complaint also alleges these written warnings as violations of Section 8(a)(5) and (1) of the Act. Complaint paragraph 11(a)(7) alleges that on October 19, 1987, Acme unilaterally implemented changes so that Acme issued written warnings for conduct which previously was tolerated or for which only an oral reprimand issued. That alleges, of course, an overall change respecting warnings. Paragraphs 11(f)(1) and 11(g) allege a unilateral change re- specting the November 11, 1987 warnings to Ramirez and Olivares. Defending against the unilateral change allegations, Acme points to warnings issued before the election. (Br. at 325.) The first one (R. Exh. 130 at 1) issued on January 12, 1983, to Emilio Mora from Supervisor Thomas J. Malleck. Malleck (R. Exh. 132 at 2) was the shipping and receiving supervisor. (24:3590, 37033.) The second cited warning (R. Exh. 74) issued November 18, 1985, to Marcial Canales from Ronald Adamczyk, and the third (G.C. Exh. 5zz), on June 30, 1986, to Tyrone Newson from Howard McArtor. These warnings are in a printed format with Acme’s letterhead. None of the 12 numbered grounds specifies ‘‘low production.’’ Checks are placed at ‘‘Other’’ (as for Mora), ‘‘failure to obey in- structions’’ (as for Canales), or both (as for Newson). The November 1985 warning to Canales actually is more a stress on a failure to obey instructions which resulted in poor pro- duction. (20:2859–2863.) Nevertheless, Adamczyk expressly includes the concept of poor production (R. Exh. 74; 20:2863), and I agree with Acme that the warning is properly listed here. Even if it is true that written warnings were the exception rather than the rule, it nevertheless is clear that Acme had a written warning system which its supervisors used with dif- ferent degrees of frequency. Indeed, as we have seen, Stoner had never issued a written warning for low production, pre- ferring instead to motivate his employees by psychological 1104 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 27 Also appearing as G.C. Exhs. 7–16. The record contains several premarked exhibits which are duplicated later under different exhibit numbers. Although the warning bears the date of January 6, 1988, at the top, Adamczyk asserts that such date—his mistake—really should be January 7, 1988. (2:235; 20:2859.) persuasion rather than by authoritarian punishment. Acme’s preexisting disciplinary system allowed for that flexibility. Complaint paragraph 11(a)(7), alleging the issuance of written warnings for conducte previously tolerated, is an overall allegation. Respecting that, plus the specific allega- tion of unilateral change on the November 1987 warnings to Ramirez and Olivares, paragraphs 11(f)(1) and 11(g), I find a violation of Section 8(a)(5) and (1). This is so because I have found that Stoner, absent union considerations, would not have issued the warnings. Thus, through Stoner, Acme implemented a new system of imposing written warnings in situations where before the election Acme, through Stoner, would, at most, have orally warned the employees. Whether termed a new system engrafted over the old, or simply a stricter enforcement of the old system, Acme was not at lib- erty to make unilateral change in the disciplinary system fol- lowing the Union’s election victory, and any such unilateral change violated Section 8(a)(5) and (1) of the Act. Hyatt Re- gency Memphis, 296 NLRB 259 (1989). Acme must rescind the warnings, cease the unilateral change, and offer the Union the opportunity to bargain before Acme implements any such change. Although the change was made after the election and be- fore the Union’s certification, any unilateral changes an em- ployer makes during that period are at the employer’s risk because the duty to bargain, although not enforceable until certification by the Board, relates back and attaches as of the date of the election once the certification issues. Celotex Corp., 259 NLRB 1186, 1193 (1982). Unilateral changes fol- lowing an election victory by a union are therefore a gamble by the employer based on an assumption that it will succeed in overturning the election results. If the employer succeeds in its overturning effort, it wins the gamble. But if it fails— as Acme did here—it loses the gamble and must suffer the consequences. 3. Warning of January 7, 1988, to Mauricio Aguirre for leaving work area without permission to use restroom a. Facts Complaint paragraph 11(f)(2) alleges that Acme issued a written warning on January 6, 1988, to Mauricio Aguirre ‘‘for leaving his work area without permission to use the bathroom.’’ In its answer (R. Exh. 336 at 4), Respondent Acme admits the truth of the allegation. Complaint paragraph 11(g) alleges that the discipline imposed on Aguirre (and on the others named in the several subparagraphs of 11(f) was done ‘‘pursuant to unilateral changes in working conditions’’ implemented, as alleged in paragraph 11(a), on October 19, 1987. Acme denies this allegation. (R. Exh. 36 at 4.) The complaint alleges that Acme violated Section 8(a)(3) and (5) by issuing the January 6, 1988 warning to Aguirre. Acme de- nies any violation. On Thursday, January 7, 1988, Ronald Adamczyk issued a written warning (G.C. Exh. 15)27 to Mauricio Aguirre. (2:232; 11:1625; 20:2856–2857.) Checked at items 7, ‘‘Fail- ure to obey instructions,’’ and 9, ‘‘Leaving work without permission.’’ The ‘‘Remarks, in Adamczyk’s hand (2:233), read: On Jan. 7th 1988 at 11:02 a.m. you left your work area, without reporting to your immediate supervisor, to go to the restroom. On Jan. 5th 1988 you were just re- minded about this same condition, that you should re- port to your supervisor before leaving your work sta- tion. I therefore am giving you a written warning, and if this or any other violation of work rules are dis- obeyed, the next warning will be three days off without pay. [Emphasis added.] Mauricio Aguirre has worked for Acme since about July 1979. (11:1607–1608.) During 1987 and until about January 1988 Aguirre worked in the packing department. (11:1608– 1609.) Adamczyk was supervisor of the precision machining and waveguide departments from 1980 until April 1988 he was reassigned from those departments to become supervisor of the shipping department. (2:231; 10:2831–2835.) Adamczyk had seven packing employees and 10 employees in the PMD and waveguide. (20:2834–2835.) Adamczyk re- fers to the departments, either jointly or separately, as a small group or small crew. (20:2838, 2867, 2878.) Aguirre was active for the Union before the October 1987 election (11:1609–1610), and recall that he is named in the Union’s November 17, 1987 letter (R. Exh. 1) as the first- shift steward for packing and several other departments. Aguirre testified that before the election he went to the restroom, as necessary, without even once asking permission or even notifying his supervisor because such a preliminary was not required. (11:1612–1614, 1654.) According to Aguirre, in the meeting Adamczyk held right after the elec- tion, Adamczyk told his employees that they were not to go to the restroom without first obtaining ‘‘permission.’’ (11:1611.) Adamczyk testified that permission was never re- quired, but that the employees were supposed to notify him before leaving the work area for the restroom. (20:2859.) Adamczyk initially testified (Jan. 24, 1989) that before October 1987 there was no problem of employees disappear- ing, that employees would go to the restroom and come right back, and that such good order was why he had never issued any (written) warnings before the election for an employee not notifying him. (2:234.) His testimony 8 months later (Sept. 19, 1989), however, describes employees frequently leaving without notifying him (going to the cafeteria and elsewhere) before he met with Balma and then held a meet- ing about July 1987 with each of the departments and told the employees they must tell him. (20:2839–2841, 2869– 2878, 2883–2884.) According to Adamczyk, he even told his employees he was going to start giving (written) warnings. (20:2879.) Once again, however, later in his testimony Adamczyk twists and asserts he made no special announce- ment that the rules would be enforced. Twisting again, Adamczyk then asserts that he did tell the employees the rules would be enforced. (20:2885–2886.) Despite his purported July 1987 notice that the rules would be enforced, and with warnings yet, Adamczyk con- cedes that the employees’ freewheeling conduct continued. And although Adamczyk ‘‘tried’’ to put a stop to it, he issued no written warnings. (20:2891–2892.) 1105 LOVEJOY INDUSTRIES Because the conduct continued, particularly in the 3 to 4 weeks before the election, Adamczyk testified, he held an- other meeting with his employees in early October in which he again reminded them of the rules, including that they were to let him know before leaving their work area. (20:2848–2849.) Adamczyk has difficulty recalling whether this is the same meeting held the week after the election. (2:241–242). However, he recalls that it was pursuant to in- structions from President Novak and that the other super- visors were holding meetings at about the same time. (2:240.) It seems clear that Adamczyk held only one October meeting—on October 19, 1987. There is no dispute that in the October 19 meeting Adamczyk, among other rules, addressed the matter of going to the restroom or leaving the area. Aguirre claims Adamczyk told them they must have permission. (11:1611.) Adamczyk asserts he said they simply should tell him they were going so he would know where they were. (2:233, 238, 240; 20:2850.) Unable to recall exactly what he said at the meeting, Adamczyk testified in the rather mushy language of ‘‘I brought up’’ the topic. Thus, ‘‘I brought up about every- thing, work rules, you know, leaving early, walking around and leaving their work area. I wanted to know where they were going, and lunch areas, you know.’’ Adamczyk told them he was talking to them about these issues again ‘‘be- cause they were abusing them. I wanted to put a stop to it.’’ (20:2850.) Asked by counsel whether he could remember the specific rules or issues he addressed, Adamczyk answered, ‘‘Not real- ly, they were the rules that we always had. We just wanted to enforce them.’’ Adamczyk testified (20:2851): JUDGE LINTON: Do you recall what you said? THE WITNESS: I brought up about leaving their work areas, the time limits. JUDGE LINTON: Just what did you say? THE WITNESS: Well, not to leave their work areas. If they wanted to leave their work area, I wanted to know about it. Later on the same page, in answer to a question by Acme’s counsel, Adamczyk further stated, ‘‘I brought it up that they were not supposed to leave their work area unless they notify me.’’ Although Adamczyk seemed to have a compulsion to preface his responses with ‘‘I brought up,’’ in view of the balance of his answer to each question, I inter- pret the ‘‘I brought up’’ as the equivalent of ‘‘I said.’’ Thus, he told the employees they were to notify him when leaving their work areas. Despite this meeting and the warning of enforcement, Adamczyk testified, employees continued to leave their work stations without notifying him. Still he issued no written warnings to enforce the rules. (20:2899.) Because of this and other abuse of the rules, Adamczyk testified, he held an in- formal ‘‘reminder’’ meeting with his employees on January 5, 1988. (20:2857, 2898, 2905.) Among the items he stressed was for them to tell him if they were leaving the work area. (20:2858, 2905.) This meeting lasted 5 to 10 minutes. (20:2898.) Conceding, on cross-examination, the fact of the January 5 meeting, Aguirre answers ‘‘yes’’ to a question whether on January 5 Adamczyk reminded employees to report to Adamczyk before leaving the work station. (11:1643.) On re- direct examination by the Government, when asked what Adamczyk said, Aguirre renders the reminder as requiring permission to go to the restroom. (11:1650–1651.) Aguirre demonstrated, on recross-examination, that he in- terprets the requirement to report, as stated on the face of the written warning, as meaning that he must obtain permis- sion. When the text is read to him and Aguirre asked where it says he failed to receive permission, Aguirre answers, ‘‘It is very clear here.’’ (11:1652) and (11:1653): By Mr. Salzman: Q. What is your understanding of what this warning says? A. I understand that he gave it to me for disobeying the rule, asking for permission to go to the bathroom. Asked how he interprets the ‘‘reporting’’ phrase to mean ‘‘permission,’’ Aguirre answers, ‘‘That one should ask for permission first before going to the bathroom.’’ Q. And that is what Ron Adamczyk told you? A. He told me he was going to give me a warning for having disobeyed the rule that says you have to ask for permission to go to the bathroom. (11:1653–1654.) Aguirre never testifies that the words on the warning document incorrectly substitute ‘‘report’’ and ‘‘re- porting’’ for ‘‘permission.’’ To Aguirre they mean the same. As the document is consistent with Adamczyk’s version, I find that in the meetings of July and October 1987 and Janu- ary 5, 1988 (and the to-be-described warning meeting of Jan. 1988), Adamczyk specified notice, not permission, as the re- quirement for leaving work stations to go to the restroom or anywhere else. Turn now to Thursday, January 7—2 days after the Janu- ary 5 ‘‘reminder.’’ Adamczyk testified that he turned and Aguirre was gone from his work area without notifying Adamczyk. (20:2858, 2897.) Not disputing that he failed to give notice, Aguirre testified that, feeling the need, he went to the restroom without first obtaining permission because he forgot about the new rule imposed after the election. (11:1625, 1654.) As he emerged from the restroom, Aguirre testified, Adamczyk beckoned Aguirre with his finger and told Aguirre he was going to give him a written warning for failing to obey the rules. Protesting that he was not a slave to be treated that way, Aguirre said he had just forgotten. Adamczyk adhered to his intention. (11:1625–1626.) Accord- ing to Adamczyk, he did not learn that Aguirre had gone to the restroom until he questioned Aguirre on the latter’s return (20:2858). I credit Aguirre’s more specific version. Adamczyk testified that he decided to give Aguirre a writ- ten warning because just 2 days earlier he had reminded Aguirre and the others and this violation angered Adamczyk, ‘‘got me a little ticked.’’ (20:2858, 2906.) Delivery of the written warning took place later that day in the supervisor’s office. When Adamczyk gave him the written warning Aguirre testified, Adamczyk said Aguirre was no different from the other workers. Aguirre replied that he was different because he was a union steward. Adamczyk did not respond. (11:1626–1627, 1639.) According to Aguirre, he meant that the thought he was being discriminated against because of his status as a union steward. (11:1639.) Other than knowl- 1106 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD edge imputed to Adamczyk by the Union’s November 27, 1987 letter (R. Exh. 1, received by Acme the next day) to Acme naming Aguirre as a steward, there is no direct evi- dence that Adamczyk was personally aware of Aguirre’s steward status. Aguirre, Marcial Canales, and Antonio Ramirez signed a December 21, 1987 grievance form (G.C. Exh. 49) protesting a written warning issued to Baldemar Corral by Maintenance Supervisor Gus Hauser. Aguirre and Canales presented it to Hauser. (11:1621–1624.) There is no further evidence about the Corral grievance and there is no direct evidence Adamczyk learned of it. At the close of the January 7 warning conference Adamczyk told Aguirre that if it happened again he would be suspended for 3 days without pay. (20:2857; G.C. Exh. 15.) Adamczyk denies that Aguirre’s union activities had anything to do with his decision to discipline Aguirre. (20:2859.) Following Aguirre’s receipt of the January 7 warning, Aguirre and Canales, the chief steward, presented a written grievance (no copy of record) to Adamczyk protesting the January 7 warning. (11:1627.) Without contradiction, Aguirre testified that when the grievance was presented to Adamczyk, Adamczyk declared there was no union at Acme. Perhaps he just did not see the Union there, Aguirre re- sponded. That apparently ended the conversation. (11:1627– 1628.) There is no further evidence concerning that grievance or of any reaction to it by Adamczyk or Acme. b. Discussion (1) Section 8(a)(3) For the Government’s prima facie case as to motivation, the General Counsel’s argument is unclear. As to knowledge, the Government apparently relies on imputed knowledge, or assumed from Aguirre’s general union activities (no showing that any took place in the presence of Adamczyk). Respect- ing intent, the General Counsel appears to argue disparity based on no warnings for admitted violations of others be- fore this event but no written warnings to them. (Br. at 86– 87.) Respecting complaint paragraph 11(f)(2), the specific alle- gation as to Aguirre, I find that the evidence falls short of establishing a prima facie case. Assuming, as I do, that knowledge of Aguirre’s union steward status is imputed to Adamczyk, there is no evidence that Adamczyk, after the January 5 reminder meeting, disregarded a violation by someone either openly opposed to the Union or even some- one who was openly union but not a steward. The evidence is that Adamczyk after three meetings with employees, the last one just 2 days earlier, finally got ‘‘ticked’’ about the violation. That the violation, the first one after January 5 so far as the record shows, was committed by Union Steward Aguirre shows coincidence, but not motive. I shall dismiss paragraph 11(f)(2). (2) Section 8(a)(5) Although discussing the general unilateral change allega- tion of complaint paragraph 11(a)(7), the General Counsel (Br. at 66–70) does not address paragraphs 11(f)(2) and 11(g) which specifically allege that Aguirre’s no-permission warning of January 7 resulted from the October 19, 1987 unilateral changes. Acme discusses various preelection warn- ings for years as far back as 1981. (Br. at 326–328.) The 1981 warning (R. Exh. 130 at 11) which Acme cites does not qualify as a warning for leaving without permission. Indeed, that item, number 9 of the form, is not marked. The remarks describe several incidents of ‘‘improper conduct’’ (item 11) by Miguel Paiz. Leaving the work area and wan- dering around the plant for 25 minutes is noted, but it is not clear that this was an independent factor in light of the more serious incidents described in the warning. Although item 9, ‘‘Leaving work without permission,’’ was checked on the May 25, 1983 warning/termination (R. Exh. 130 at 2) of Fernando Chavez, the matter has little rel- evance. Chavez simply punched out early, left without word, and was terminated ‘‘due to past record of this worker.’’ An- other with item 9 marked is for ‘‘walked out of his duties’’ without a word to anyone. That March 1985 warning (G.C. Exh. 7–2) to Wayne Daniels also has item 4 marked for in- subordination to the supervisor, Ernesto Vega. Neither is for going to the restroom without permission nor is the Novem- ber 18, 1985 warning (G.C. Exh. 7–9) which Scott issued to Hector Martinez. Although item 9, leaving work without per- mission is marked, the text of the remarks discloses that Scott issued the warning because he found him in a car out- side the building 1 hour and 15 minutes after he was sched- uled to return from lunch. Several of the other warnings cited by Acme have only re- mote relevance, as such as the insubordination/unauthorized break warnings which Jim Scott issued in September 1985. And it is frivolous for counsel to cite (Br. at 327) the August 1, 1985 warning (G.C. Exh. 5 ddddd) to Eldiberto Garcia from Scott for item 1, ‘‘Inexcused absence’’ (missing 5 working days without calling in during the previous 2 weeks to notify supervisor). Aside from their relevance on the general issue of whether Acme had, and used a written warning system before the election, not one of the the preelection written warnings of record issued for going to the restroom—whether without no- tice or without permission. Perhaps overly fine distinctions should not be drawn. Conceptually, leaving the work station early to go to the cafeteria can be classified, as Scott did it in 1985, as item 4, insubordination, for taking unauthorized breaks against prior orders. He easily could also have checked item 9, leaving work without permission. Still, the specific matter here is going to the restroom—something far more basic and necessary than getting coffee in the cafeteria. Acme’s enforcement, through Adamczyk, of the preexist- ing rule against leaving the work area without notice is a marked departure from its preelection practice of never ap- plying, much less enforcing, this rule to restroom visits— even if the rule is interpreted as requiring only notice and not permission. I credit Aguirre when he testified that before the election he had never notified the supervisor when he went to the restroom. By October 1987, the practice of free access to the rest- room, without the restriction of first giving notice to the su- pervisor, had become an established term and condition of employment. I so find. I also find that the preelection prac- tice was a significant term of employment. Following a union’s election victory and the union’s certification, an em- ployer is not at liberty unilaterally to change an established an significant benefit, term or condition of employment. Ven- 1107 LOVEJOY INDUSTRIES 28 In polite society courtesy dictates that roman numerals be used sparingly. See rule 6.5, NLRB Style Manual at 22 (1983). The switch to arabic numerals was discussed in a conference call on May 12, 1989. Parentheses around the paragraph numbers were removed by amendment. (8:1285–1286.) ture Packaging, 294 NLRB 544 (1989); Adair Standish Corp., 290 NLRB 317, 319, 329 (1988) (Adair II), enfd. in pert. part 912 F.2d 854 (6th Cir. 1990). The prohibition in- cludes enforcing more stringently rules which, before the election, received only lax or sporadic enforcement. Hyatt Regency Memphis, 296 NLRB 259 (1989). Thus, an em- ployer cannot be lax before the election and unilaterally switch after the election to a system which goes ‘‘strictly by the book.’’ Celotex Corp., 259 NLRB 1186, 1194 (1982). Yet that is what Acme did here. Maurico Aguirre suffered a written warning as a result. Accordingly, by this unilateral change, Respondent Acme, I find, violated 29 U.S.C. § 158(a)(5) and (1). Hyatt Regency Memphis. 4. Warning of February 10, 1988, to Antonio Loza for tardiness/coffee a. Procedure Some background of the tortured procedural trail is re- quired. In the first (June 1, 1988) complaint, Case 13–C–- 27619, paragraph VII(d) alleged, as unilateral changes, warn- ings to several employees including: (v) In or about December or January 1988, Marco An- tonio Preciado Loza, for tardiness. By its June 16, 1988 answer, Acme denied the allegation. In the Government’s ‘‘Further Amendments’’ of January 11, 1989, the wording remains the same except ‘‘to’’ is placed before Loza’s name. (G.C. Exh. 1oo). Acme repeated its de- nial in its January 16, 1989 answer. (G.C. Exh. 1qq.) Following the January 26, 1989 adjournment sine die (4:732) the General Counsel, on April 14, 1989, moved (G.C. Exh. 1ss) to consolidate the five outstanding com- plaints into a single document—a consolidated complaint of even date, attached to the motion. Paragraph XI(f)(iii), now carrying the Loza allegation, changed the date to February 10, 1988, but retained the reason as tardiness. On April 26, 1989, Acme denied the modified allegation. (G.C. Exh. 1uu at 4.) In its third affirmative defense of April 26, Acme lists subparagraphs (i)–(vii) of paragraph XI(f) as time barred by 29 U.S.C. § 160(b). The thrust of Acme’s limitation conten- tion is that Case 13–CA–27619 has been pending a long time and therefore the General Counsel should be prohibited from such a late amendment ‘‘when this information has been known and available to the Charging Parties and the General Counsel for many months.’’ (G.C. Exh. 1uu at 6.) Soon after those pleadings were exchanged, the General Counsel filed a new motion (G.C. Exh. 1vv), dated April 26, 1989, to amend the consolidated complaint (as proposed on April 14) so that, among other amendments, paragraph XI(f)(iii) would read: (iii) On or about February 10, 1988, to Marco Antonio Preciado Loza for getting coffee during working hours. On May 8, 1989, Acme denied the amended Loza, allega- tion (G.C. Exh. 1xx at 1) and, as an affirmative defense to this allegation, pleaded (id. at 3): With regard to Paragraph 1 of the General Counsel’s Motion, the unfair labor practice charge which formed the basis for this allegation, Case No. 13–CA–27619, was filed on March 18, 1988. On June 1, 1988, presum- ably after conducting an investigation into the allega- tions, the General Counsel issued a Complaint alleging, inter alia, that Respondent had violated the Act by dis- ciplining Loza for tardiness. (See Paragraph VII(d)(v) of the Complaint in Case No. 13–CA–27619). After further investigation this Complaint was amended on July 28, 1988 and then again on January 11, 1989, vir- tually on the eve of the trial. As of this last amendment, the General Counsel still alleged that Respondent vio- lated the Act because it disciplined Loza for tardiness. At the trial, which commenced on January 23, 1989, the General Counsel discovered facts which proved that this allegation was unfounded and meritless. The Gen- eral Counsel now wishes to make a belated effort at continuing its ‘‘fishing expedition’’ using the ‘‘shotgun approach’’ hoping that one of the numerous allegations will find a target. This requested amendment should be barred by Section 10(b) of the NLRA and, in any event, due to the length of time that Case No. 13–CA–27619 has been pending, the amendments that have already been investigated and made, the late date that this amendment was filed and the circumstances of its fil- ing, the General Counsel should be prohibited from Amending the Consolidated Complaint when this infor- mation has been known and available to the Charging Parties and the General Counsel for over a year. By order dated May 14, 1989, I granted the General Coun- sel’s April 14 motion to consolidate the complaints into a single document and the April 26 motion to amend the single document, but without prejudice to Acme’s renewing its ob- jections on resumption of the hearing. (G.C. Exh. 1yy.) At the resumption, colloquy indicated that I would resolve such issues in this decision. (5:737–739.) The June 26, 1989 amended consolidated complaint, the ‘‘trial complaint,’’ finally incorporated all allegations into a single document. Numerous pleadings and amendments cre- ate the danger of confusion and oversight. Despite the April 14, 1989 ‘‘getting coffee’’ version of paragraph XI(f)(iii), the June 26, 1989 trial complaint version, now appearing in the more readable form of paragraph 11(f)(3),28 reverted to the original tardiness reason, although retaining the February 10, 1988 date. By its July 11, 1989 ‘‘trial answer,’’ Acme admits that Loza was given a written warning on February 10, 1988, but denies that the reason was tardiness. (R. Exh. 36 at 4.) For its affirmative defense, Acme repeats the lengthy quotation (‘‘With regard to Paragraph 1 . . .’’) which I set forth above concerning the General Counsel’s motion of April 26, 1989. The one modification made is to show the date of April 26 for the General Counsel’s motion. (R. Exh. 36 at 6–7.) 1108 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 29 Raymundo Aguirre testified that all employees had trouble get- ting to work that day because of the snow storm. (8:1200.) No party offered certified copies of any weather reports for the date, nor even copies of newspaper reports of weather conditions for the date. 30 The August 25, 1987 leaflet (R. Exh. 4) distributed by the Union and its supporters is a list of changes in pay, benefits, and working conditions the employees desire to see effected at the plant. (3:445–446, Davis.) Item 17 reads: ‘‘It is not fair to cut our pay 15 minutes if we are only a few minutes late.’’ (R. Exh. 4; 4:714; 8:1203.) 31 Actually, the timecard (the original is in evidence) requires in- spection and interpretation because it is over-punched by the stamp of the punching-out time at the close of the shift. Over the General Counsel’s objection, Scott testified that the first punch is for 6:52 a.m. He did not explain and identify each digit. (21:3228–3231.) The hours seem clear enough, a 6 (for the morning) and a 4 (for the afternoon). The problem is the minutes. Although it is reasonably clear that the four digits are 2, 3, 5, and 7, it is too difficult for me to ascertain whether the correct combination is 6:52 or 6:57 for the morning, and 4:32 or 4:37 for the afternoon. I find, however, that it is one or the other. 32 There is no controversy over this card or the name. Presumably Arias is Francisco Arias Pantoja. In the Hispanic culture, the family, or father’s, surname is place inside, with the mother’s name outside. (2:261.) Thus, ‘‘Loza’’ apparently is the American version of Loza’s family surname of Presiado. b. Facts Notwithstanding the twisted turns in the procedural path, once here, the traveler finds the basic factual dispute easy to resolve. Marco Antonio Preciado Loza began work at Acme at least as early as 1980. Since about 1982 he has worked as a trim press operator in the ADC. (12:1663–1664.) Recall that Jim Scott has supervised the ADC since July 1985. Be- fore and after the October 1987 election Loza has been ac- tive for the Union. Indeed, Loza even received a written warning (G.C. Exh. 50) on September 15, 1987, for insubor- dination (item 4), improper conduct (item 10), and ‘‘other’’ (item 13). The text reads: Employee was seen passing out union literature during working hours. When he was approached by super- vision he threw them into the air. When told to pick them up he refused. There is no dispute about the event described. I need not re- solve it because the complaint does not allege a violation concerning the union literature warning. The pertinent point is that at least as of that event Respondent Acme was aware of Loza’s union sympathies. In February 1988 Loza rode to work with Juan Ornelas and several other Acme employees, including Baldemar Cor- ral. (12:1672–1673.) Loza’s normal procedure on entering the plant is (and was in February 1988) to clock in, change his clothes in the locker room, proceed upstairs to the cafe- teria where he gets a cup of coffee, then take the coffee downstairs to his work station, all before the 7 a.m. start of the shift. (12:1679, 1718–1720.) According to Loza, on February 10, 1988, most of the em- ployees, including his group, arrived late because it was snowing hard and traffic was slow.29 (12:1675–1676). Loza testified that he punched in at 7:03 a.m. and in the next 3 minutes changed clothes, got his coffee and, knowing that he was late, was rushing to his work station when he met Jim Scott. Understanding that Scott wanted him to pour out his coffee and go to work, Loza replied that Scott could simply dock him 15 minutes. When Scott persisted, Loza complied. (12:1676–1678; 1721, 1725, 1730–1733.) Loza testified that when he was as much as 15 minutes late in the past he would be docked 15 minutes, and on this occasion he had expected to be so docked.30 Loza insists that he punched in at 7:03 a.m., that the others he rode with, although not all were at the timeclock when he was, also punched in after 7 a.m. He denies punching in at 6:52 a.m. (12:1733–1734, 1736.) About 5 minutes after arriving at his machine, Loza testi- fied, Scott brought him a written warning. Loza refused to sign because, he testified, he considered it unfair. On other occasions, including three to four times in the year before the election (he does not recall the dates), he had arrived late for work, and still had gotten coffee, yet was not warned or even orally reprimanded. (12:1674, 1678, 1734–7135.) He simply was docked on those occasions and, as noted, on this occa- sion he expected to be docked 15 minutes. The warning (G.C. Exh. 17) reflects that Scott did not check item 2, Tar- diness, but checked item 13, ‘‘Other.’’ The text of ‘‘Re- marks’’ reads: Antonio came out of cafeteria 10 min. after beginning of shift with a cup of coffee in his hand. He has been warned that next offense will be cause for a 3 day dis- ciplinary layoff. Before summarizing Scott’s version, I need to list the names of those riding that February 10 in the station wagon of Juan Ornelas. Loza testified that the other riders were Jesus ‘‘Chuy’’ Arrendondo, Baldemar Corral, Francisco Arias Pantoja, Antonio Sanchez, and Agustin Urvieta. (12:1672–1674, 1721–1722.) Certain that Baldemar Corral rode that day, Loza describes Corral as sitting to Loza’s left on the same seat. Although Loza did not see Corral punch in that morning, or see him at his work station, he did see him at the morning break and at lunch. Corral rode home with the group that day, Loza testified. (12:1722–1725.) Of the group, apparently only Loza received a written warning for events that day. Even before considering Scott’s version, the documentary evidence discloses a two-fold problem with Loza’s version of arriving late. First, Loza’s timecard (R. Exh. 81) for the week ending February 13, 1988 (Saturday), reflects that on Wednesday (February 10) Loza punched in before 7 a.m.31 Loza testified that Chuy Arrendondo and Agustine Urvieta punched in about a minute before he did (12:1733). Urvieta’s card (R. Exh. 84) bears a clear punch-in time of 6:51 a.m., as Scott (21:3234) testified. Arrendondo’s card (R. Exh. 82), however, shows a punch-in time of 6:57 a.m. (21:3231– 3232), as does that (R. Exh. 86) of Juan Ornelas (21:3236) and the card of Francisco Arias.32 As his card (R. Exh. 88) reflects, Antonio Sanchez punched in at 6:39 a.m. (21:3237.) As Sanchez rode with the group, and punched his card sev- eral minutes before the others, Sanchez’ punch-in time is strong evidence that Juan Ornelas and his riders arrived well before 7 a.m. that morning. Of the group, only Loza testified. Recall the testimony of Loza that Baldemar Corral rode with the group that morning, seated to Loza’s left. Corral’s 1109 LOVEJOY INDUSTRIES timecard (R. Exh. 85) reflects that he was off work Wednes- day through Friday of that week as the penalty of a discipli- nary suspension. (21:3235.) I find Loza to be an unreliable witness. Supervisor Jim Scott testified that Loza was not late that February 10, that Scott saw Loza about 6:50 a.m. on the stairs leading to the cafeteria, that when he saw Loza again it was about 7:10 a.m., and he had been looking for Loza for 10 minutes respecting a job assignment. He told Loza he had been looking for him and to pour out his coffee because he had a job assignment for him. (2:249–250; 21:3225– 3226.) I credit Scott. Scott testified that he decided to give Loza a written warn- ing for two reasons. First, Scott considered Loza’s action a gross misuse of the time schedule. Second, even though Scott had cautioned Loza previously during the last 3 to 4 weeks against similar conduct, Loza seemed to be falling into a habit of ‘‘trying to come downstairs a few minutes late after the beginning of the shift.’’ After reading the warning to Loza, Scott reminded him that he had cautioned Loza ear- lier and that Scott now had to issue this warning because Loza did not seem to care what Scott said. Scott notified Loza that if he continued to do this Scott would impose a 3-day layoff on Loza. Loza did not respond. (21:3226–3227.) When he was called as a rebuttal witness on the last day of the hearing, Loza did not address the subject of this warning. c. Discussion (1) Section 8(a)(3) Contending that Scott’s warning for this event is unprece- dented, observing that the warning text says nothing of any earlier oral caution or cautions, pointing to Loza’s unrebutted testimony of docking as a past practice, citing Loza’s open union activities, and pointing to the the timing of the warn- ing ‘‘within a few months of the Union’s election victory,’’ the General Counsel argues that these facts ‘‘warrant the in- ference that the issuance of this written warning would not have occurred absent the Union victory and Loza’s own in- volvement in Union activities.’’ (Br. at 88.) As part of its argument, Acme lists the several written warnings Scott has issued in the past. (Br. at 330.) Most of the written warnings Scott has issued are the ones he im- posed in September 1985. They specifically recount a prior oral warning. For example, the September 9, 1985 warning (G.C. Exh. 7–11) to Eliaza Unzueta states, ‘‘Eliaza was ver- bally warned not to take unauthorized breaks but continues to do so (11:30 a.m.) (9 a.m.).’’ Scott’s October 23, 1987 written warning (R. Exh. 96), discussed earlier, to Rodolfo Banales, begins, ‘‘Having been verbally warned many times, Rodolfo continues to leave his work station between break periods.’’ But Scott’s November 6, 1987 warning (R. Exh. 23) to Raymundo Aguirre for con- tinuing to leave his work station several minutes before the break period says nothing of prior orals other than that which may be implied by his use of ‘‘Raymundo continues.’’ Scott’s general practice is to warn employees orally before he issues written warnings to them. I find that Scott earlier had orally warned Loza not to stay in the cafeteria so as to delay being at his work station at 7 a.m. When Loza again delayed reporting for work, even giving the appearance of a deliberate delay (Scott had seen him going into the cafeteria some 20 minutes earlier), Scott decided on a written warn- ing. Regardless of whether the Government established a prima facie case, I find that Scott would have issued the written warning to Antonio Loza on February 10, 1988, even absent the October 1987 election or Loza’s union activities. I therefore shall dismiss complaint paragraph 11(f)(3) insofar as it alleges a violation of 29 U.S.C. § 158(a)(3). (2) Section 8(a)(5) Before the October 1987 election Scott issued written warnings. In issuing the February 10, 1988 written warning to Antonio Loza, Scott responded to Loza’s abusive dis- regard of Scott’s earlier oral warnings. No unilateral change has been shown. I therefore shall dismiss complaint para- graph 11(f)(3) in its entirety. (3) Section 10(b) The different wordings of the allegation (complaint par. 11(f)(3)), whether ‘‘tardiness’’ or for ‘‘getting coffee during working hours,’’ do not violate the limitations provision. Al- though the ‘‘getting coffee during working hours’’ can be construed as describing a working privilege, that is just the positive side of the ‘‘tardiness’’ concept. Only a single event is in issue. Nor has Acme shown that it was in any way prej- udiced by the changes in the allegation. In any event, I find no violation of Acme’s right to due process. For the same reasons, I find that the correction of the date of the incident to February 10, 1988, a date Acme well knew, caused Acme no prejudice. I overrule Acme’s objections in this respect and I deny its motion to strike. 5. Warning of February 25, 1988, to Antonio Loza for producing defective parts a. Allegations Recall that paragraph 11(a)(7) alleges that, as a unilateral change on October 19, 1987, Acme began issuing written warnings for conduct which it ‘‘previously tolerated, and for conduct that previously only warranted an oral reprimand.’’ Complaint paragraph 11(f)(4) alleges that on or about Feb- ruary 25, 1988, Respondent Acme issued a written warning to Marco Antonio Preciado Loza for producing defective pieces. Paragraph 11(g) alleges that Acme imposed the dis- cipline described in paragraph 11(f) pursuant to the unilateral changes described in paragraph 11(a). Concluding paragraphs allege that these actions violated Section 8(a)(3) and (5) of the Act. I shall dismiss these allegations. b. Facts Since about 1982 Loza has worked in the punch press, or trim press, section of the ADC. His supervisor is Jim Scott. (12:1664.) In February 1988 Loza trimmed large (about 22 inches by 10 inches) castings for AT&T. As Loza describes his operation, using his punch press he trims flashing from the parts, and puts the trimmed parts into a basket. Later the basket is taken to another department for additional work. (12:1681–1683.) As described by Scott (21:3210–3214, 3243–3244; 23:3420–3422, 3499–3500), and Roman to some extent (20:2961–2962, 2968–2969), a trim press has a stationary bottom and a movable top. A two-piece die is mounted in- 1110 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 33 On the particular casting Loza was trimming, Scott testified that Loza was not to look for nicks and scratches, but broken rails mak- ing the grids or windows. (23:3422.) 34 The last three digits of the part number. (21:3246.) side, one part on the bottom, or nest, and the second on the top or movable platen. Guide pins in the bottom die fit into corresponding holes in the upper die when the press is closed. Rows of punches are set in the top die to punch win- dows in the casting, and cutting edges trim the flashing, all to shape the casting per the specifications. A trim operator takes a casting from the adjoining table, places it in the nest, then activates two switches. At that acti- vation, the top of the press comes down. As the press closes, the die punches and trims to shape the casting. When the op- erator releases the switches, the top opens, and the operator removes the casting, checking it visually for nicks, scratches, or casting defects.33 He places good castings in a nearby bas- ket. Scrap castings and excess metal are taken to the remelt furnace. After every cycle he blows cuttings and pieces from the bottom die or press, and from the top when necessary, with an air hose. Every few cycles, depending on the part, he brushes oil over the top and bottom of the two-part die. If those parts are not oiled they become dry. The top then can stick to a casting and rip out chunks when the press opens. Loza agrees that he is to clean out the die, or press, every cycle. (12:1683, 1736.) His testimony thereafter diverges from Scott’s and becomes rather confusing. Initially asserting that he checks about one part in every 25 (12:1691), subse- quently he testified that on February 24 he inspected his parts when he first started, again about 4 to 5 hours later, and once more at the end of his shift. These checks showed all pieces to be fine, with everything normal. (26:3971– 3974.) This was his standard procedure, Loza testified. (26:3975.) Moreover, the inspector came by and checked the parts about three times. When the inspector finds a problem he says so. He said nothing on these three inspections on February 24, Loza testified. (12:1693.) What makes Loza’s testimony confusing is that his pro- duction report (R. Exh. 89) for February 24 reflects that he produced a total of 575 pieces of the part, no. 791,34 in ques- tion. If Loza had inspected every 25th part, he would have checked 23 parts. Loza possibly means that, in addition to checking every 25th part on February 24, he also checked the basket of parts on three other occasions—at the start of his shift (to make sure, he testified—12:1689—that the night shift had not left any bad parts in the basket), 4 to 5 hours later (a general check, apparently), and at the end of his shift to satisfy himself that all is well. (12:1689.) If so, Loza mis- stated when he answered government counsel’s question, at the rebuttal stage (26:3971). Q. Now on February 24, 1988, how often did you look at the pieces you were making? A. When I first started working. Then later, about four or five hours later. And then toward the end of the day I would look through all of the parts and see what I had done. There is no dispute that at the beginning of the shift on February 25, 1988, Supervisor Jim Scott and his assistant su- pervisor, Juan San Roman, gave Loza a written warning (G.C. Exh. 18) for work attributed to Loza on the previous day. The warning, dated February 25, 1988, has item 13 marked (‘‘Other’’) and is signed by Scott. Loza also signed under a printed caption which states he has read the report. After ‘‘Other’’ at item 13 Scott (2:258) wrote: ‘‘Carelessness resulting in damaged work product.’’ The ‘‘Remarks’’ sec- tion reads: Antonio was trimming part no. 844209791 shelf. A piece of one rail broke off and stuck in the top of the trim die. He continued to work and 4 more pieces came off the castings. In all 79 pieces were damaged and had to be scrapped. There is some difference as to the warning meeting itself. According to Loza (12:1687, 1739), Roman interpreted. Roman recalls that Valenzuela was there interpreting for Loza and that Roman interpreted for Scott. (20:2930, 2982– 2983.) Loza does not place Valenzuela at the meeting. Scott testified he spoke in English and he implies that Roman, al- though present, did not need to interpret because Scott has no problem communicating in English with Loza. (2:261– 262.) No one testified that he saw Loza produce the damaged parts. The evidence is circumstantial. Loza worked the day shift. Recall that Howard McArtor is the supervisor for the evening shift. McArtor describes the sequences as follows. As no operator wants to be blamed for defective work by someone else on a previous shift, at the beginning of the evening shift the operator (name not given) following Loza discovered broken parts in the basket and a piece stuck in the die. He reported this to the inspector, Ramon Topete, who counted 79 or 80 broken parts. Topete reported this in- formation to McArtor who directed that the broken parts be tagged for the day shift and that the die, now impaired, not be used. The machine was not used that evening shift. (19:2784–2785, 2810–2811.) Neither the operator nor Topete testified. I consider McArtor’s description only on the basis of a report made to him, rather than for the truth of the mat- ter reported. Apparently arriving for work on February 25 well before the 7 a.m. start of the first shift, Scott found a note from McArtor describing the problem. The basket of castings had been moved to the inspection office. Scott counted 79 or 80 defective castings. (21:3238–3240; 23:3437–3438.) Roman saw the 79 broken parts. (20:2930.) Before Loza arrived at his work station, Scott and Roman went to Loza’s machine. There they found, Scott testified, several pieces of casting stuck in the top portion of the trim die. The pieces were mashed on top of one another between the rows of punches. The pieces, all wedged in one spot, had to be chiseled out. They were pieces of the castings and matched the torn gaps in the castings. (21:3242–3245; 23:3451.) A necessary impli- cation of Scott’s description is that the damage was done, to the damaged castings, at the same section on each casting. Indeed, Scott later confirms this implication. (23:3440.) Because the obvious nature of the defect and the high number of broken parts indicated that Loza had not cleaned or oiled the die, and had not inspected the parts or the die, Scott formed the opinion that Loza—who had several years’ experience in trimming this type of casting—had intended to damage the castings. Scott therefore decided to issue a writ- 1111 LOVEJOY INDUSTRIES ten warning to Loza. (2:252–254, 264–265; 21:3241; 23:3442, 3452.) Scott does not know what motivated Loza to do this. (23:3454.) Roman apparently agrees that it was intentional. (20:2966, 2970.) One factor that influenced both Roman (20:2970) and Scott (21:3245) in so concluding is the abnormal noise which would have been made when each casting broke. Aside from hearing each break, Loza, Scott testified, would had to have seen the break when he removed each casting to inspect it and place it in the basket. (23:3452–3453.) Asked why, since he had put ‘‘carelessness’’ on the warn- ing reason, he now thought the work was intentional, Scott testified that it was ‘‘intentional carelessness,’’ that Loza did not stop but continued to damage the castings, and that in his opinion Loza knew what was happening. (23:3455.) Loza admits that at the warning meeting he did not deny producing the damaged partsal. Neither did he affirm it, Loza testified. (12:1739.) According to Loza, when Roman showed him the broken parts, Loza said those were not his parts, that he would had to have been completely blind not to have seen such a defect. (12:1694.) Even an employee of only 1 or 2 months could tell the difference, Loza concedes. (26:3973.) Roman testified that Loza ‘‘admitted’’ he had done it be- cause he just ‘‘shook his arms like that.’’ (20:2960.) Later in his testimony Roman became more affirmative, asserting that Loza said, ‘‘I probably did’’ it; ‘‘I did it,’’ (20:2976) and ‘‘Okay, I ruined the castings.’’ (20:2983–2984.) Scott testified that Loza, not denying it (2:260), simply shrugged his shoulders when Scott, displaying the sample broken cast- ings, asked for an explanation. (21:3248–3249.) Crediting Scott, I find that Loza merely shrugged his shoulders, did not make the statements Roman claims, and that Loza did not say the parts shown there were not his. At the hearing Scott identified a photograph (R. Exh. 24) as a picture of one of Loza’s broken parts. (21:3239; 23:3440.) Denying that it is one of the parts he worked on or was shown on February 25, Loza says it is similar but dif- ferent in two ways. First, his parts had small holes in the interlacing rails. Second, whereas the part in the photo has a break, a gap in an outside rail one square from the corner square, the ones Scott and Roman showed him on February 25 had a corner broken off rather than the break shown in the photo. (12:1737–1739; 26:3970, 3972.) According to Loza, he looked at his parts when he first started work on February 24, then about 4 to 5 hours later, and again at the end of his shift. All the pieces were fine, and everything was normal, Loza testified. (26:3971–3974.) Scott does not know why Loza did not show the 79 broken pieces as scrap on his production report. (R. Exh. 89; 2:263; 23:3443.) Although at one point (2:258) Scott speculates that Loza did not do so in order to avoid incriminating himself, Scott later concedes (2:264) that it is not unusual for employees to forget to insert the number of scrap pieces. Loza concedes that before the October 1987 election the most scrap parts he produced in any 1 day would have been seven or eight. (12:1685.) Loza testified that a basket would hold about 340 castings of part 791, and that at the end of his day shift on February 24 he left that basket about 3/4ths full (12:1695–1697)—or containing about 255 castings of part 791. As he descended from the cafeteria with coffee in hand the morning of Feb- ruary 25, Loza testified, Scott and Roman intercepted and es- corted him to the supervisors’ office where the warning was given. (12:1686; 26:3974.) He was not shown the basket of parts in question. (12:1688.) Thus, Loza never got to check the basket or parts in question. Supervisor McArtor concedes that the basket was not tagged to indicate which operator produced the parts. (19:2811.) That takes us back to the note (not in evidence) which Scott testified McArtor left. The note, Scott testified, said that the basket came from the press for die cast machine number 18. (21:3240; 23:3438.) Presses, Scott testified, do not have numbers because they are identified by the die cast machine they adjoin and serve. (23:3438.) Why, then (asked the General Counsel on cross-examination) did Loza mark ‘‘14’’ in the machine number column of his production re- port? Scott is not sure, but speculates that perhaps the press has a number 14 on it somewhere to cause Loza to place that number on the report. (23:3438–3439.) Strangely, Loza, the person who would know, is never asked, not even when re- called at rebuttal, what the 14 refers to. In any event, Loza apparently worked on the same press, number ‘‘14,’’ on Feb- ruary 25, after receiving the warning, for his production re- port (R. Exh. 92) that day reflects that he produced 520 good castings of part 791 on machine (press) number 14. Scott tes- tified that he knew from observation, as well as production reports, what press Loza operated on February 24. (21:3241.) Whatever the correct number is for the press, I find that Loza operated the press where Scott found the pieces stuck in the die. Turning now to the parts, I find that Loza only shrugged at the warning interview when Scott showed him the rep- resentative castings and asked how he could have continued to operate, producing 79 castings with such obvious breaks. Thus, I do not credit Loza when he testified that he told Roman the broken castings displayed were not his. Neither do I credit Loza’s version that the broken casting in the pho- tograph (R. Exh. 24) is not one of his and that the casting Roman showed him the morning of February 25 had a corner broken. In short, I find that Loza produced the 79 broken castings which Scott counted the morning of February 25, 1988. The day QC inspector may well have checked Loza’s bas- ket three times on February 24 without finding any damaged goods. However, as the damaged parts were on the top of the basket it means, as Scott testified (2:255; 23:3439, 3442, 3452), that the items were produced in about the last 90 min- utes of the shift. There is no evidence that the QC inspector came by during that last 90 minutes. Roman did not check the parts at all that day because he was busy elsewhere. (20:2961.) It is unclear whether the noise produced by the breaking of the casting, especially 79 in sequence, would have been heard by other employees nearby. Actually, the record does not reflect whether any employees were working nearby or, if any were, whether one would have said any- thing. Loza denies intentionally damaging any castings in an at- tempt to force Acme to grant employees a pay increase, and he denies that any representative of the Union ever told him to slow down or to make bad parts as a tactic to secure such a pay increase. (26:3964.) Whatever Loza’s motive, I find that on February 24, 1988, he deliberately damaged the 79 castings. Aside from the preelection warning (G.C. Exh. 50) issued to Loza by Scott over the union literature incident, the time- 1112 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 35 Two production reports of Jose Aguirre do not appear to gen- erate any weight. The first (G.C. Exh. 21–9) is a March 3, 1988 re- port by Aguirre showing that he reported 186 scrap pieces. Scott ex- plained that the scrap was produced earlier in the production line, with Aguirre, in this respect performing his duty to inspect, merely reporting what he found. (2:270–272, 283.) Apparently the same condition applies to Aguirre’s March 9, 1988 report (G.C. Exh. 21– 6) of 688 scrap pieces, although Scott did not address that specific exhibit beyond expressing the view that Aguirre probably wrote the figures. (2:282.) frame of other union and concerted activities Loza describes (12:1699–1671), other than being postelection, is not entirely clear. He does describe wearing a union cap two or three times a week after the election. (12:1670.) Although that pre- sumably was in the days after the election, Loza does not de- scribe even one instance of wearing the cap in the presence of Scott, Roman, or any other supervisor. Scott denies seeing Loza wearing any such emblem. (21:3250.) Nevertheless, Scott admittedly suspected that Loza supported the Union. That suspicion apparently derives from the literature incident plus employees (union supporters) Loza associated with such as Jose Aguirre and Rodolfo Banales. (21:3250–3251.) Scott asserts that neither any union sympathies of Loza nor his as- sociation with (union supporters) played any part in his deci- sion to issue the written warning to Loza. (21:3249, 3252.) c. Discussion (1) Section 8(a)(3) Describing the Government’s prima facie case of unlawful motivation, the General Counsel advances three arguments. First, timing (‘‘during the first months following the Union’s election victory’’). Second, Acme has failed to prove that the 79 castings were in fact damaged by Loza (rather than by someone else) thereby warranting the inference that Loza would not have received the warning but for his union activ- ity. (Br. at 94–95.) Third, disparity. (Br. at 68, 97.) The timing contention is misapplied. If the castings had been damaged days or weeks before the election, and the warning delayed until after the election (or until after some significant union or protected act by Loza), then that would be the kind of timing the Government could cite in support of its prima facie case. But here the timing factor cuts the other way because the warning issued within an hour or so after Supervisor Jim Scott learned of the damaged castings. Respecting the Government’s second argument (Acme’s failure to prove), the General Counsel confuses the initial burden. It was not Acme’s burden to show that Loza dam- aged the castings. Instead, at the prima facie stage it was the General Counsel’s burden to prove that Loza did not damage the castings, and either that Acme knew he had not done so or had no good-faith belief he did or could not reasonably have so believed. In any event, the evidence indicates that Acme reasonably believed that Loza is the person who pro- duced the 79 defective parts. In the foregoing second basis of the Government’s conten- tion of unlawful motivation, the General Counsel apparently is not arguing the applicability of NLRB v. Burnup & Sims, 379 U.S. 21 (1964). Indeed, the Burnup analysis applies to an alleged violation of Section 8(a)(1) of the Act, 29 U.S.C. § 158(a)(1). The allegation here is of unlawful motivation— alleged (in par. 13) as a violation of ‘‘Section 8(a)(1) and (3) of the Act.’’ But the Section 8(a)(1) there is only derivative of the Section 8(a)(3). The independent 8(a)(1) paragraph, number 12, does not reference any of the warning allega- tions. Thus, I treat the General Counsel’s argument, of no good-faith belief by Acme that Loza is the person who pro- duced the damaged pieces, as an argument for an 8(a)(3) and (1) finding. Magnolia Manor Nursing Home, 284 NLRB 825 fn. 1 (1987). Disparity. An initial showing of disparity appears. The warning (G.C. Exh. 18) to Loza states that it is for ‘‘care- lessness.’’ Scott concedes that Loza and others in the past have only been orally warned for damaging parts.35 (2:253– 254). Whether that surface disparity serves to establish a prima facie case of unlawful motive is highly questionable. Assuming, however, that it does, I find that Acme carried its burden of demonstrating that it would have taken the same action against Loza even absent his union activities. Scott credibly testified that although Loza and others in the past have damaged pieces and received only oral warn- ings, never had employees damaged castings in this manner. (2:253–254.) What determined Scott’s decision here was not simply the large number, but also the obvious nature of the damage, the pieces stuck in the upper die, the fact that Loza would have heard, seen, and ignored the damage, and Loza’s years’ of experience with this type of work. All these factors caused Scott to believe that Loza had deliberately damaged the parts. (2:254; 21:3241, 3245; 23:3442–3443, 3452–3455.) There is no evidence that Scott or Acme has ever re- stricted discipline to oral warnings for suspected saboteurs. Although the written warning Scott issued on February 25, 1988, cited ‘‘carelessness’’ rather than the ‘‘intentional care- lessness’’ Scott expressed at the hearing (23:3455), it is clear, and I find, that in February 1988 Scott believed that Loza had acted, in essence, as a saboteur. At the hearing Scott declined to speculate on Loza’s motive. I credit Scott’s testimony that Loza’s union activities played no role in Scott’s decision. (21:3249–3252.) Contrast this case with NLRB v. Lakepark Industries, 919 F.2d 42 (6th Cir. 1990). In summary I find that even if the General Counsel estab- lished a prima facie case of unlawful motivation, Acme car- ried its burden of demonstrating that Scott would have issued the written warning to Loza even absent Loza’s union activi- ties. Accordingly, I shall dismiss complaint paragraph 11(f)(4) respecting the alleged violation of 29 U.S.C. § 158(a)(3). (2) Section 8(a)(5) I also shall dismiss the 8(a)(5) allegation. Although the written warning expressly is for ‘‘carelessness,’’ it is clear from the face of the warning (continuing to produce a total of 79 broken parts after one broken piece stuck in the die), and Scott’s comments to Loza at the warning meeting, that he could ‘‘not believe’’ Loza would make that mistake and not see it (2:267). There was ‘‘no excuse’’ (21:3249) for the damage, Scott told Loza. Scott’s comments demonstrate that Scott considered Loza’s conduct to be gross negligence at the very least. Even without considering Scott’s testimony at the hearing, it is clear that Scott’s warning to Loza addressed conduct going beyond a brief inattention to duty which might result in damage to two or three pieces. In the years before the October 1987 election, Acme issued numerous warnings for a variety of infractions. Al- 1113 LOVEJOY INDUSTRIES though none appears for gross negligence, so far as appears from the record, the Loza incident would be the first occa- sion of gross negligence or, worse, sabotage. Thus, the warn- ing here was merely one more exercise by Acme of its warn- ing system as applied to a situation more akin to deliberate misconduct than to simple negligence. Finding no unilateral change, I shall dismiss complaint paragraphs 11(a)(7), 11(f)(4), 11(g), and 13 to the extent they allege that the Feb- ruary 25, 1988 warning to Marco Antonio Preciado Loza violates Section 8(a)(5) of the Act, 29 U.S.C. § 158(a)(5). 6. Raymundo Aguirre fired February 25, 1988 a. Facts Complaint paragraph 9(a) alleges that Acme discharged Raymundo Aguirre on February 25, 1988. Acme so admits. (R. Exh. 36 at 3.) The complaint also alleges that Acme vio- lated Section 8(a)(3) and (1) of the Act by that discharge. Acme denies any violation. Signed by Scott and Balma, the discharge report (G.C. Exh. 24), with items 4 (insubordination) and 10 (improper conduct) checked, reads: Raymundo asked if I had disciplined an employee of my dept. When I answered yes, he said to me, ‘‘Do what you want to do, but I am going to make a move on you.’’ He also made a gesture to the employee who had been disciplined. When the plant manager asked him if he had threatened his supervisor, he answered yes. The event giving rise to Raymundo Aguirre’s discharge occurred shortly after Scott gave Antonio Loza the written warning which I described in the previous section. Recall that Raymundo Aguirre, a trim press operator in Scott’s ADC, had been elected the day-shift steward for Scott’s ADC. The Union’s November 17, 1987 letter (R. Exh. 1) to Balma and Acme had so informed Acme. Thereafter, Scott and Aguirre had even conferred about potential discipline matters on a couple of occasions. At the beginning of the shift on February 25 Scott, Aguirre testified, told him that he was going to give Loza a (written) warning for bad work the day before. Shaking his head in a negative fashion, Aguirre went to work without responding or inquiring about the cir- cumstances. (8:1153–1154), 1222–1223.) About 20 minutes later, around 7:20 a.m., Aguirre testi- fied, he observed Scott and San Roman escorting Loza to- ward the front office. (8:1158, 1224, 1244.) About 30 min- utes later Aguirre saw Loza returning to his work station. (8:1159, 1228.) The evidence thereafter is extensive and dis- puted. Two points are important. First, it is undisputed that, as Loza passed within 25 to 45 feet of Aguirre, Aguirre com- municated with Loza by gestures over the distance and ma- chine noise. According to Aguirre he initially inquired by head and hand gesture as to what had happened. Aguirre tes- tified (8:1160) that Loza answered yes by nodding his head up and down. (8:1160.) But Loza testified that his own ges- ture, palms down, meant nothing had happened and that ev- erything was fine. (12:1699–1770, 1743.) This was followed by Aguirre’s pointing to the cafeteria and, with his hand, in- dicating that they would discuss it in the cafeteria at break. Aguirre describes opening and closing his fingers as the opening and closing of lips. (8:1160–1161.) Loza asserts that Aguirre indicated 11:30 a.m. by first holding up 10 fingers, then 1 finger with another finger crossing it to indicate the half hour, followed by opening and closing the hand in the fashion of a quacking duck. Loza nodded yes. (12:1700– 1702; 1743–1747.) Scott (22:3335) and San Roman (20:2931) testified that Scott told Roman to accompany Loza back to his work sta- tion. Roman testified that, walking on Loza’s right, he heard Aguirre, to Loza’s left, shout something to Loza. Roman did not observe any gestures other than that Loza ‘‘shook’’ his head in a ‘‘positive’’ fashion left to right. Roman looked back and saw Scott approaching. (20:2933.) Aguirre (8:1228, 1233, 1238) and Loza (12:1750) assert that Roman was not with Loza. Although Aguirre gives an elastic timeframe for Scott’s appearance on the scene (8:1238–1239), Loza clearly states that Scott observed at least Loza’s gestures. (12:1701.) Scott testified that as he proceeded some 25 to 30 feet behind Loza and Roman he heard Raymundo Aguirre holler something to Loza who, by nodding his head up and down, signaled yes. (2:302, 349; 22:3337–3338.) Hand signals, Scott acknowl- edges, are sometimes used on the noisy shop floor by em- ployees and supervisors (2:312), but they are the common variety of arm or hand motions. (2:356.) Aguirre asserts that hand signals such as for beckoning someone, are used daily. (8:1162–1163.) When Loza nodded yes, Scott testified, Aguirre, in an agi- tated state, jumped out of his chair and gave a hand signal Scott had never seen. (2:355; 22:3343.) Holding his left index finger and left middle finger in a ‘‘V’’ shape, Aguirre began thrusting his right forefinger through the ‘‘V’’ so formed. (2:301, 350); 22:3338.) Scott asserts that Aguirre was not pointing upward (2:360). Aguirre denies making any such sign, and he denies ever having seen such a sign. (8:1171–1172.) Not believing either denial by Aguirre, and finding that Scott testified in a believable fashion, I find that Aguirre did as Scott describes. Seeing the gesture, but not knowing what it meant, Scott assumed it meant he was going to try to do something against Scott for issuing a written warning to Loza. (2:303–304.) Observing Aguirre so agi- tated, Scott intended to pass by and to let Aguirre cool down before talking with him. (2:351; 22:3338.) Scott could see that Aguirre appeared angry. Aguirre was red in the face with veins in his forehead bulging, pacing back and forth, slamming equipment, and waving his arms. (2:352, 360; 22:3339.) Whether Aguirre screamed to Scott, ‘‘Hey come over here’’ (2:311, 351; 22:3338, Scott) or whether Scott, being only 4 to 5 feet away, walked over to Aguirre when Aguirre asked his question (8:1249, Aguirre), both agree that Aguirre asked Scott if he had issued a written warning to Loza and that Scott answered yes. Scott, whom I credit, testified that Aguirre yelled at him, ‘‘You can’t do that,’’ that Loza had worked at Acme for 9 years. (2:298, 352; 22:3339.) Asserting that a warning to Loza was ‘‘ridiculous’’ (8:1165, 1249, 1253), Aguirre con- cedes (8:1165, 1261) that he told Scott (2:300, 353; 22:3339, Scott): ‘‘I am going to make a move on you.’’ Scott asked Aguirre what he meant, but Aguirre, Scott testified, was so angry he could not or would not answer. (2:308; 22:3339.) When Aguirre made his statement about moving on Scott, he 1114 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 36 Normally, Scott testified, Raymundo Aguirre was a quiet, good employee. (2:295.) Balma acknowledges that there had been no pre- vious problem with Aguirre. (2:195–196.) Nevertheless, on Novem- ber 6, 1987, Scott issued Aguirre a written warning (R. Exh. 23) for ‘‘failure to obey instructions’’ over Aguirre’s persistently leaving his work station several minutes before the break period. When Scott warned Aguirre that the next offense could result in a 3-day suspen- sion, Aguirre said he did not care. (R. Exh. 23; 22:3343–3345.) 37 Scott (2:354; 22:3341, 23:3504), Raymundo Aguirre (8:1167, 1267–1268, 1277), and Balma (2:194–195; 24:3757) are the three witnesses to the conversation. Jorge Serrato supposedly saw them talking (14:1952–1953) but could not hear the conversation over the noise and distance. (14:1938.) 38 Scott (2:355; 22:3342) and Balma (24:3757) do not describe what they did when Aguirre left the washroom and went up to the cafeteria. I find that they used these minutes to prepare the dismissal report (G.C. Exh. 24). Balma also prepared his own report (R. Exh. 25; 24:3756). However, that possibly was not until Aguirre left be- cause Balma suggests that it was not until it was all over that he asked Scott about the whole episode. (25:3857.) 39 Balma’s version excludes the question that Scott reports. was ‘‘in Scott’s face’’ but did not touch Scott, who tried to clam Aguirre. (2:353, 358–359.) At work that day in Stoner’s secondary department, Jorge Serrato, a brother-in-law of Aguirre, was situated about 25 feet from where Aguirre was working. Serrato testified that he observed Aguirre and Scott on this occasion and, although he could not hear what they were saying over the noise and distance, he observed that the discussion was calm with no loud talk or moving of hands. (14:1935–1939, 1950.) Serrato concedes that he continued working his own machine, and that only at intervals could he look back over his left shoul- der and see the pair, and even then he could see only the left side of Aguirre’s face and Scott hardly at all because Aguirre blocked Serrato’s view of Scott. (14:1944–1945, 1950–1952.) I attach little weight to Serrato’s testimony. First, at best he could look only at intervals. Second, in any event, Serrato, who testified in a sarcastic fashion on cross- examination, testified with an unpersuasive demeanor. Be- yond Serrato perhaps noticing the pair, I do not credit any of Serrato’s testimony. Suspecting that the hand gesture had something to do with harming somebody, and in light of Aguirre’s wild and strange behavior,36 Scott was alarmed by the incident. (2:301, 304; 22:3343.) Proceeding to a production control meeting, Scott related the incident to the supervisors and oth- ers in attendance. Greg Brown, the production control man- ager (but not employed by Acme at the time of the hearing and not a witness), remarked that he thought the gesture was a sign by one of the Latin gangs meaning to stick someone with a knife. (2:304, 307; 22:3339–3340.) Shortly thereafter Balma joined the meeting and Scott reported the matter to him. Balma said they would go talk to Aguirre. There is no dispute that only minutes after the gesture in- cident Balma and Scott approached Raymundo Aguirre as Aguirre was at work. Aside from slight differences in the wording of the question by Balma, as recalled by the wit- nesses,37 the witnesses agree that Balma asked whether Aguirre had threatened Scott. All three witnesses—including Aguirre—agree that Aguirre answered, ‘‘Yes.’’ All agree that Balma repeated his question, and that Aguirre again an- swered, ‘‘Yes.’’ More differences follow, but all agree that Balma, in effect, fired Raymundo on the spot, told him to punch out and leave, and that Aguirre said Balma could punch the timecard. Balma did so. According to Aguirre, he was about to explain his mean- ing, that he meant filing a grievance over the warning to Loza and did not mean a physical threat to Scott, but that Balma never gave him a chance. (8:1168–1169.) Yet Aguirre concedes that nothing was said in the 20 to 30 seconds it took the three to walk to the timecard rack. (8:1269–1270.) I do not credit Aguirre’s testimony that he desired to give Balma, in February 1988, the testimonial explanation which he offered to me in June 1989. When Balma punched Aguirre’s timecard, Scott, Aguirre testified, got ‘‘right in my face’’ and, laughing, said, ‘‘Now, Raymundo, you can leave.’’ (8:1169, 1269, 1270.) Initially testifying that he did not recall such an event, Scott went on to deny it. (23:3505.) I credit Raymundo Aguirre on this point. Contrasted to Scott’s initial uncertainty, Aguirre’s de- livery and demeanor were fervid in describing Scott’s scorn- ful laughing. I have no doubt the incident occurred as Aguirre describes. After Balma punched his timecard, Aguirre then went to the locker room, changed his clothes, and from there he went to the cafeteria to wait for the break to make arrangements for a ride home. Shortly thereafter, Balma and Scott entered the cafeteria with the dismissal report (G.C. Exh. 24).38 Scott testified that he handed the paper to Aguirre. (2:315.) Aguirre first recalls that Scott had the paper, then recalls it was Balma. (8:1271.) Aguirre testified that he read the dis- missal report, including the reference there to a gesture. Ac- cording to Aguirre, he asked Balma why he was being fired when all he was trying to do was to help Loza. Supposedly, Balma answered by saying that he did not care what Aguirre had done, but whenever his supervisor spoke he was to re- main quiet. (8:1170, 1277, 1279, 1282.) Both before and after reading the report, Aguirre testified, he had no idea why he was being fired or told to leave, that neither Balma nor Scott asked about any gestures (8:1171), and, thinking that Balma wanted no explanation from him, Aguirre told Balma to forget it. Taking his coat, he then left to wait outside in the cold. (8:1170–1171, 1277–1278, 1283.) Balma testified that he saw Aguirre leave the plant in Francisco Mombela’s car. (25:3907.) Although not expressly disputing, or denying, Aguirre’s version of the cafeteria conversation, and not describing Aguirre’s reading of the dismissal report, the version of Scott (2:355; 22:3342) and Balma (24:3757; 25:3856) is different. They testified that Balma told Aguirre that wherever he worked he could not threaten people, and asked Aguirre why he did it.39 ‘‘I had to put a stop to this somehow,’’ Aguirre replied. Balma apparently did not ask what Aguirre was re- ferring to. Aguirre denies that Balma said he could not threaten people. (8:1277.) Aguirre was not asked at the hear- ing whether he made the statement about needing ‘‘to put a stop to this somehow.’’ Although Scott first described the statement when he was called by the General Counsel the second day of the hearing (2:355), Aguirre did not testify on rebuttal. As with Aguirre’s asserted claim of desiring to explain earlier to Balma that he had been referring only to filing a grievance for Loza, Aguirre’s testimony about his cafeteria explanation, not directly answered by Balma, has a contrived 1115 LOVEJOY INDUSTRIES 40 Suspicion is not enough. There must be some specificity in the record linking particular employees to the misconduct alleged. Gen- eral Telephone Co. of Michigan, 251 NLRB 737, 739 (1980), enfd. mem. 672 F.2d 895 (T) (D.C. Cir. 1981). There must a ‘‘sufficient nexus.’’ Columbia Portland Cement Co. v. NLRB, 915 F.2d 253, 257 (6th Cir. 1990). 41 Columbia Portland Cement Co., 294 NLRB 410 (1989), enfd. on point 915 F.2d 253, 257 (6th Cir. 1990); Magnolia Manor Nurs- ing Home, 284 NLRB 825, 829 (1987). 42 Champ Corp., 291 NLRB 803 fn. 13 (1988), enfd. 913 F.2d 639 (9th Cir. 1990). 43 Magnolia Manor Nursing Home, 284 NLRB 825 fn. 1 (1987). 44 Champ Corp., 291 NLRB 803. 45 Aztec Bus Lines, 289 NLRB 1021, 1027 (1988). 46 Called by the Government early in the hearing, Balma already had testified that he did not ask Aguirre how or in what way he had threatened Scott. (2:195.) ring. Finding Aguirre’s version to be an afterthought, I find that the cafeteria exchange occurred as Scott and Balma de- scribe. b. Analysis (1) Legal principles Although the Government argues the issues on the tradi- tional motivation theory, as does Acme, the General Counsel first argues that Raymundo Aguirre’s case is controlled by NLRB v. Burnup & Sims, 379 U.S. 21 (1964). Unfortunately, Acme does not brief the Burnup & Sims question. In Burnup & Sims the Supreme Court affirmed the Board’s ruling that Section 8(a)(1) of the Act is violated despite the employer’s good faith if it is shown that an employees was discharged for alleged misconduct while engaged in a protected activity, and that the employee was not in fact guilty of that mis- conduct. Once it has been established that an employee is engaged in union or other protected activity, the burden shifts to the employer to demonstrate an honest belief that the employee was engaged in misconduct.40 If the employer establishes an honest belief that the employee engaged in misconduct, the burden shifts back to the General Counsel to prove either that the employee did not in fact engage in the misconduct41 or that his conduct was not serious enough to warrant dis- charge.42 Finally, even if the misconduct is otherwise serious enough to warrant discharge, the employer violates Section 8(a)(3) of the Act if the employer acts without a good-faith belief43 (that is, advances the asserted misconduct as a pre- text), such as disparate treatment of those engaging in the protected conduct as contrasted with those refraining from (or opposing) protected conduct.44 In short, an employer may not use a double standard as a means of punishing employees engaged in protected conduct.45 When the discussion is of motive considerations, such as disparity and pretext, the focus shifts to the analytical structure of Wright Line, 251 NLRB 1083 (1980). See Mark Industries, 296 NLRB 463 fn. 2 (1989). (2) Discussion (a) Section 8(a)(1)—Burnup & Sims (i) Protected activity Under Burnup & Sims, the first question is whether Raymundo Aguirre’s conduct occurred in the midst of pro- tected activity. The answer is yes. The postcertification pe- riod still constituted part of the ongoing effort by the em- ployees and the Union to obtain recognition and to begin bargaining. Moreover, even without recognizing the Union, Acme accepted grievances from the stewards as a means of resolving disputes. Thus, protected activity was occurring, and Acme was well aware of it. Although Acme and Scott knew of Raymundo Aguirre’s status as a union steward, I need not go beyond Acme’s knowledge of the general union activity. That is so because employer knowledge of the ac- cused employee’s participation in the protected conduct is not a prerequisite for application of the Burnup & Sims test. Ideal Dyeing & Finishing Co., 300 NLRB 303 (1990). (ii) Honest belief The second question is whether Acme honestly believed that Raymundo Aguirre had engaged in serious misconduct warranting discharge. Plant Manager Peter P. Balma made the decision to discharge Aguirre. (2:194.) Thus, the question is ‘‘What did Balma know, and when did he know it?’’ Scott initially testified that at the meeting in the produc- tion control office (where Greg Brown suggested the knifing explanation for the hand signal), when Balma arrived, he re- ported the details of the incident to Balma. (2:307, 354.) He told Balma the ‘‘whole story’’ from the Loza warning to the gesture and the ‘‘move on you’’ statement. (2:307, 354.) ‘‘Let’s go talk to Raymundo,’’ Balma exclaimed. (2:354; 22:3341.) According to Balma, it was not until after Aguirre’s depar- ture that he learned from Scott what had happened and that Aguirre had been protesting Loza’s warning. (25:3857.) When Balma arrived at the production control meeting, he observed that Scott was ‘‘visibly shaken.’’ Scott told Balma, ‘‘I just got threatened by Raymundo Aguirre.’’ ‘‘You can’t be serious,’’ Balma replied. ‘‘Yes,’’ Scott assured him. To Balma’s question of why, Scott said he did not know. ‘‘At that time he was so shaken, I didn’t know what the whole circumstances was at that time, besides him making a threat. I said, Okay, let’s go immediately.’’’ (24:3756.) They went direct to Aguirre where, Balma testified, he ob- served that Aguirre was visibly angry. (24:3757.) As I have summarized, Balma then inquired whether Aguirre had threatened Scott. To Balma, as soon as he heard that a super- visor (Scott) had been threatened, ‘‘it was over.’’ (25:3857.) All that remained was the confirmation of the fact. When Aguirre admitted making a threat, ‘‘as far as I was con- cerned, it was all over with.’’ (25:3857–3858.) And on cross- examination (25:3858): Q. Did you ask him [Aguirre] how he had threat- ened? A. No, I did not.46 Q. Did you ask Mr. Scott how he was threatened? A. Not at that time, no. Q. So, he could have been threatening to do any- thing, right? A. Not with the look on his face. 1116 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD As noted earlier, after Aguirre had left, Balma assertedly asked Scott what had made Aguirre so mad. Scott then de- scribed the whole incident. (25:3857.) Even that version re- quires some interpretation, for Balma signed the dismissal notice (G.C. Exh. 24) given to Aguirre, and quoted earlier. There the ‘‘move on you’’ statement is quoted plus reference to a gesture. Presumably Balma read the notice before he signed it. As Balma’s own report (R. Exh. 25) of even date describes the cafeteria conversation, it is clear that Balma completed his report after Aguirre had left the plant. Even so, Balma’s text begins, ‘‘Jim Scott came to me & said Raymundo had threatened him with a physical sign & ver- bally told him I going to make a move on you.’’ If just these lines were considered, it would appear that Scott, in the pro- duction control office, at least had mentioned the sign plus the ‘‘move on you.’’ According to Balma, however, he had no details to support the generalized ‘‘threat’’ reference, plus Scott’s ‘‘visibly shaken’’ appearance, plus the (still) angry expression of Aguirre. There is no indication that Balma and Scott conversed as they walked out to confront Aguirre. Balma concedes that in his confrontation of Aguirre, Aguirre never said anything to indicate that he was threatening Scott with bodily harm. (25:3860.) Nevertheless, to Balma, a threat means ‘‘that you are going to hurt somebody, vio- lence.’’ (25:3859.) First, crediting Scott, I find that Scott made a full report to Balma before they went to confront Aguirre. Balma con- cedes he asked Scott why Aguirre threatened him. According to Balma, Scott answered that he did not know. I find that Balma is confused on this point. The reason Balma, in the cafeteria, did not ask Aguirre what he meant about putting a stop to ‘‘it’’ is that Balma already knew. On these facts I find that Balma understood from Scott that Raymundo Aguirre had just ‘‘threatened’’ Scott by the hand gesture and the ‘‘move on you’’ remark. Based on Balma’s definition of ‘‘threat’’ as meaning personal violence, plus Balma’s seeing Scott’s ‘‘visibly shaken’’ appearance, plus the other details, I find that Balma decided on the spot—in the production control office—to fire Aguirre, sub- ject to Balma’s confirming the threat. When Aguirre freely answered ‘‘yes’’ to Balma’s question, and ‘‘yes’’ again when Balma (perhaps a bit surprised at the admission) asked again, Aguirre’s fate was sealed. I find that Balma acted on his honest belief Aguirre had threatened a supervisor, Jim Scott, with physical violence. A threat of physical violence against a supervisor is serious misconduct. (iii) Serious misconduct Did the General Counsel show that the misconduct did not occur, or that it was done by someone other than Raymundo Aguirre? Unlike Ideal Dyeing & Finishing Co., 300 NLRB 303, where it was found that the accused employee (Juan Perez) was not the person who made the threat, here there is no question of identity. The question is whether the thrust- ing ‘‘V’’ signal, coupled with the ‘‘move on you’’ remark, with both made in the context of an angry Aguirre, constitute serious misconduct. I find that they do. Moreover, I find that they occurred. Thus, the General Counsel has failed to show that the misconduct did not occur. (iv) Conclusions On these findings I conclude that the evidence fails to sup- port the Government’s Burnup & Sims argument. That being so, I need not discuss whether Burnup & Sims would be in- applicable in any event inasmuch as the complaint fails to enumerate the Raymundo Aguirre discharge paragraph (par. 9,a) in paragraph 12 with the other numbers alleged to be violations of Section 8(a)(1) of the Act. The discharge is enumerated in paragraph 13 with the others alleged to be violations of ‘‘Section 8(a)(1) and (3) of the Act.’’ As that is the discrimination paragraph, it is clear that the ‘‘8(a)(1)’’ listed there derives from the 8(a)(3) allegation, and does not stand independent of the 8(a)(3) allegation, and that the prop- er allegation for the derivative 8(a)(1) allegation in complaint paragraph 13 would be ‘‘Section 8(a)(3) and (1) of the Act.’’ (b) Section 8(a)(3)—Wright Line (i) The prima facie case That brings us to the standard motivation analysis under Wright Line, 251 NLRB 1083, enfd. 662 F.2d 899 (1st Cir.), approved in NLRB v. Transportation Management Corp., 462 U.S. 393 (1983). The General Counsel argues pretext as shown by (1) no investigation of the facts by Balma and (2) disparity. Balma, as I have found, acted in good faith. His understanding and use of the term ‘‘threat’’ equates with physical violence. Scott’s description of the gesture plus the ‘‘move on you’’ statement suggest physical violence. Balma observed a ‘‘visi- bly shaken’’ Scott who claimed that Aguirre had threatened him. When Balma confronted Aguirre, he found a visibly angry Aguirre. With those corroborating facts, combined with Aguirre’s repeated admission that he had threatened su- pervisor Scott, Balma fired Aguirre. As the General Counsel concedes (Br. at 53), there is no evidence of animus directed at Aguirre. Indeed, he and Scott previously had discussed a couple of potential disciplinary matters. Even here Scott advised Aguirre in advance that he intended to warn Loza. Rather than seeking to discuss it then, Aguirre shook his lead and turned away. On later learn- ing that Loza had been warned, Aguirre exploded in outrage at Scott. I find no merit to the first ground, particularly where the details, reported later by Scott, demonstrate that Aguirre made a hand signal which, coupled with his angry ‘‘move on you’’ statement, reasonably can be interpreted as a threat that Scott must be stabbed. Arguing disparity, the General Counsel cites two incidents in contending that Acme has tolerated conduct equal or worse in nature. First, Acme tolerated conduct by Gus Hauser, the maintenance supervisor, during the first (Septem- ber 1988) strike. Later I summarize the strike incidents. At this point it is sufficient to note that on September 15, the second day of that 2-day strike, Gus Hauser drove into the east driveway to Acme’s property. Strikers were in the drive. Complaint paragraph 8 alleges that Hauser deliberately caused his car to hit and injure a striker. Summary of that evidence will show that as Hauser drove in, strikers hit, beat, or slapped his car. Hauser pulled for- ward, stopped, got out, and told striker Valenzuela that if he again banged on his car would hit him in the face. Arguing ensued and Balma heard striker Nelson Diaz offer to do bat- 1117 LOVEJOY INDUSTRIES 47 Acme also objected on the basis that the matter did not con- stitute impeachment evidence as to Balma because there was no showing that Balma was even aware of the incident. (26:3990.) tle with (‘‘take care of’’) Hauser. Balma intervened to end the incident. Balma did not discipline Hauser (nor, appar- ently, Diaz or Valenzuela, either) and considers that Hauser did no wrong on that occasion even though it is a violation of company policy for anyone, including a supervisor, to threaten another individual under normal circumstances. Balma did not consider the incident to be normal cir- cumstances. (25:3892–3894.) Reacting to violence against his vehicle, Hauser threatened violence. While such conduct, by both sides, must be de- plored, it is not comparable to a threat to stab a supervisor because the supervisor (peaceably) issued a written warning to another employee. I find no disparity in the Hauser inci- dent. The second incident concerns Leon Bonner and John Selwitschka. Hired in September 1985, Bonner works the evening shift in Howard McArtor’s secondary department. (14:2031–2032.) Selwitschka is McArtor’s die casting setup person. (19:2764.) Bonner testified, but Selwitschka did not. Bonner describes a threat of physical violence by Selwitschka against him. Selwitschka was neither fired nor, so far as the record shows, even issued a written warning. The incident, Bonner testified, occurred about October 1, 1987, following a meeting at which President Novak spoke to employees, assembled in the cafeteria at the start of the second shift, about the advantages the employees had without a union. (26:3977, 3979–3980.) During the course of his re- marks, Novak described Acme’s profit-sharing plan. At some point Bonner, speaking up, inquired how Novak determined the shares to be distributed. Novak answered that everyone received 10 percent of his earnings. Bonner said that as Novak earned more than ‘‘we’’ did his share was much larg- er than ‘‘ours,’’ and that Bonner did not think that was fair because ‘‘we’’ do most of the work. (26:3980.) Bonner’s comments to Novak apparently upset Selwitschka. As the employees left the meeting, Selwitschka approached Bonner downstairs. Selwitschka asked Bonner where he was before coming to Acme. (Supervisor McArtor, who apparently had been walking with or beside Selwitschka, split off and proceeded elsewhere at that point.) Bonner replied that it did not matter because he is at Acme now. Bonner and employee Jerry Stewart, now followed by or accompanied by Selwitschka, walked without talking about 30 feet to the glove cabinet. (26:3977, 3983–3985.) As the three reached the glove cabinet Selwitschka, in sexually explicit terms, told Bonner that he was spoiling Selwitschka’s job. Bonner replied that such was not Bonner’s concern. Selwitschka then invited Bonner to step outside where Selwitschka would ‘‘kick [Bonner’s] ass’’ all over the plant. ‘‘Let’s go,’’ Bonner replied. As the two, both now angry, eyeballed each other, Stewart stepped between them and the confrontation ended without any physical contact. (26:3978, 3982, 3984–3987.) A few minutes later, at Stewart’s suggestion, Bonner went upstairs to where Novak was seated in the cafeteria. Bonner described the incident to Novak and said he had a witness. At Novak’s direction, Bonner summoned Stewart who, with Bonner present, also described the incident to Novak. When Stewart finished, Novak began telling Bonner about the hard times Selwitschka had experienced before coming to work at Acme. Bonner said that was not his concern. Novak said he would talk to Selwitschka but that he would not fire him. Bonner said that if Selwitschka acted, Bonner would have to defend himself. Novak said Bonner had a right to defend himself. About 15 minutes later Bonner observed Novak talking with Selwitschka in front of the supervisor’s office. Selwitschka continued to work thereafter, and was still em- ployed when Bonner testified. (26:3978–3979, 3987–3988.) Neither Selwitschka nor Stewart testified. Novak was not re- called to address the incident. I credit Bonner’s uncontradicted testimony. Bonner had testified on the 14th day of the hearing about other matters. His testimony about the Selwitschka incident came on the final day of the hearing as the General Coun- sel’s last witness in the rebuttal stage. When Respondent ob- jected that Bonner’s testimony on this matter was not proper rebuttal, the General Counsel explained that the Government did not view the incident as direct evidence of disparity for the case-in-chief because it involved two employees rather than an employee and a supervisor. The evidence was of- fered, explained the General Counsel, to impeach Balma who testified during Acme’s case-in-chief that it was Acme’s pol- icy to discipline employees, including supervisors, who threaten other employees, and that all of this is relevant to the discharge of Raymundo Aguirre. (26:3988–3991.) Ex- pressing uncertainty about the matter, I nevertheless over- ruled Respondent’s objections.47 (26:3991.) Respondent made no motion to strike. Although the evidence came just a few minutes before the close of the hearing (Balma gave two pages of surrebuttal testimony on another matter), Respondent did not request time (the hearing closed on a Monday) to bring in evidence showing that Acme at least issued a written warning to Selwitschka over his threat of physical violence against Bonner. Based on Acme’s failure to produce evidence show- ing that it disciplined Selwitschka with a written warning, I infer that Acme did not, and I find that at most Acme, through Novak, did no more than orally warn Selwitschka against making invitations to fight accompanied by threats of physical violence. Earlier Balma testified that, under Acme’s policy, a threat by anyone, including a supervisor, is a dischargeable of- fense—indeed, automatic dismissal! (1:93–96; 25:3893, 3895–3896.) Balma even testified that if one employee threatens another, he immediately would discharge both. (25:3895.) If that is his policy, Balma did not explain why he did not fire Scott for getting threatened by Raymundo Aguirre. I do not credit Balma’s ‘‘both get fired’’ testimony. At the hearing the Government limited its offer of the Selwitschka incident to impeachment of Balma’s testimony about company policy specifying discharge for threats. (26:3990.) Evidence offered only to impeach is considered only for resolving credibility, Balma’s here, on the point in issue. That is, Balma claims Acme has a policy of automatic dismissal for anyone making a threat. If Balma is impeached as to that (by the Selwitschka incident), then the impeach- ment serves to support a finding that I do not find Balma credible on that point. But as impeaching evidence, the Selwitschka incident would not constitute substantive evi- dence of disparity in support of the Government’s burden to 1118 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 48 Under cross-examination, Banales increases the number to 80 and then concedes his count was 100 to 125. (14:1996–1997.) The latter is the count shown in the warning and conforms to Scott’s count. (21:3255.) make a prima facie showing of unlawful motivation. Never- theless, on brief the General Counsel apparently advances the Selwitschka matter, along with the Hauser picket-line inci- dent, as substantive evidence of disparity. (Br. at 49–50, 54– 55.) As noted earlier, the General Counsel also explained at the hearing that the evidence is relevant to Raymundo Aguirre’s discharge. (26:3989.) It therefore appears that the General Counsel’s offer of this matter, although ambiguous, in fact was intended to serve as substantive evidence of disparity. In view of that, and my overruling of Acme’s objections, I shall consider the Selwitschka’s matter for substantive purposes as well as for any impeaching effect. (ii) Conclusions I conclude that the General Counsel has failed to make a prima facie showing of unlawful motivation in Acme’s dis- charge of Raymundo Aguirre. It must be remembered that it was Peter P. Balma—not Robert Novak—who made the de- cision to discharge Aguirre. Acme apparently has internal (not distributed to employees) written policies on discharging for threats (1:91; 25:3895), and presumably (no copy is of record) Balma’s description of Acme’s policy is a reflection of the written policies. Thus, there is an institutional policy, and not simply a Balma policy and a separate Novak policy. There is not, however, one person, such as a personnel man- ager, coordinating all discipline to assure that each discipli- nary decision conforms to the institutional policy. As a re- sult, we have at least one example of disparity from the standpoint of an Acme policy, with Robert Novak himself deviating from that policy. But it was not Novak who fired Aguirre. Perhaps it could be argued that Acme’s institutional policy applies for disparity purposes. That is, any disparity analysis would consider motive by comparing the discipline imposed with the Acme policy, and motive of the deciding official would not be controlling. Older written warnings are so eval- uated when we treat them as the employer’s past practice without inquiring into the motive of the issuing supervisor. There is no question about the relevance of the disparity. Besides voting in the October 1987 election (14:2032), Bonner did not otherwise distinguish himself, before Novak’s called meeting, as a union supporter. His remarks at Novak’s union issues meeting, however, would reasonably be inter- preted as aligning himself with the prounion faction. This is so even though the Union’s August 15, 1987 leaflet (R. Exh. 4) asserts, at item 14, ‘‘We want to continue the profit-shar- ing plan.’’ Continue the plan, yes, but improve the workers’ share, per Bonner. Indeed, when the Union prepared its list of contract demands (16:2195) dated January 16, 1988, item 14 was changed to read (R. Exh. 66): 14. Improve the profit- sharing plan. I have found that Balma acted with a good- faith belief that Raymundo Aguirre had threatened Scott with a stabbing. The instant question is whether Balma’s predecision knowledge from Scott’s report that Aguirre’s threat was associated with his protest of Loza’s warning also was a motivating reason for the decision to discharge Aguirre. I find the answer to be no. As the record reflects, concerted protests were numerous and without reprisals. There is no evidence that Balma felt animus against Aguirre simply because Aguirre was protesting the Loza warning. Balma denies being motivated by any union activities of Aguirre. (25:3832.) Balma’s sole motivation, I find, was his perception, shared by Scott, that the combination of cir- cumstances (Aguirre’s anger, his gesture, his ‘‘move on you’’ statement, and Scott’s shaken appearance, plus Aguirre’s admission) demonstrate that Aguirre had threatened Scott with physical violence. Balma himself grounded his decision on a lawful consider- ation—automatic dismissal of anyone making a threat of physical violence. I find that neither union nor concerted ac- tion considerations were a motivating reason for Balma’s de- cision. As motivation of the decision maker is the issue, I therefore find that the General Counsel has failed to establish a prima facie case. Accordingly, I shall dismiss the Raymundo Aguirre discharge allegation, complaint paragraph 9(a). 7. Warning of March 10, 1988, to Rodolfo Banales for producing defective parts a. Pleadings As it did with Antonio Loza on February 25, Acme issued a written warning on March 10, 1988, to Rodolfo Banales for producing defective parts. Complaint paragraph 11(f)(6) so alleges. The complaint also alleges that Acme violated Sec- tion 8(a)(3) and (5) by issuing the warning. Admitting the fact of the warnings, Acme denies violating the Act by so warning Banales. (R. Exh. 36 at 5.) b. Facts Hired by Acme around February 1985, Rodolfo Banales works the day shift as a punch (trim) press operator in Jim Scott’s ADC. Before and after the October 1987 election Banales wore union insignia such as buttons and caps. (14:1955–1957.) About 4 of his years at Acme have been as a punch press operator. (14:1993–1994.) As summarized ear- lier, on October 23, 1987, Scott issued a written warning (R. Exh. 96) to Banales for taking excessive breaks. The General Counsel did not allege that warning to be unlawful. Before and after the election, Banales wore union buttons and caps. In about February 1988, on one of the occasions when the employees gathered before President Novak to press for recognition, Banales, with Scott and San Roman also present, told Novak that he had a good business but that he did not care about his employees. Novak replied, ‘‘Busi- ness, which business, I don’t see no business.’’ (14:1955– 958.) According to Banales, at the 7 a.m. start of his March 10, 1988 shift, the QC inspector informed him that parts from his work the day before, March 9, were bad and that he had notified the supervisor. These were parts for outside lamps. Scott came and instructed Banales to separate out the defec- tive parts. On direct examination Banales claims that he counted about 75 defective pieces.48 He discovered that the parts had a short crack (one-quarter to one-half inch) on the inside of the side which would be opposite from him as he worked. In the supervisor’s office at the end of the shift, 1119 LOVEJOY INDUSTRIES 49 Triplicated copies are in the record as G.C. Exhs. 5w, 5xx, and 19. 50 A photo (R. Exh. 90) of an undamaged one is in evidence. No damaged one was identified nor photo of such offered. Scott asserts that none was taken. (23:3473.) 51 In his pretrial affidavit, Banales states that the ‘‘operator’’ (ap- parently a reference to the trimmer rather than die cast operator) ‘‘has the responsibility for spotting damaged pieces that come out of the machines.’’ (14:2002.) Scott with San Roman present, issued a written warning49 to Banales for his work on March 9. (14:1960–1963.) The warning, with item 13 ‘‘Other’’ marked and followed by the hand printed word ‘‘Carelessness,’’ reads (G.C. Exh. 19): Rodolfo was trimming part no. 420304. A piece of one casting broke off and stuck in trim die. He continued working until the end of shift even though every casting was cracking. He destroyed 100–125 castings. Next of- fense will be cause for a 3-day disciplinary warning as he has been verbally warned about problems with his work several times. At some point on March 10, ‘‘earlier in the day’’ than the time of the warning (2:269–270), Scott, as he testified, was making his rounds. As he came to where Banales was trim- ming, Scott stopped, reached into Banales’ basket of trimmed parts, and pulled out one of the castings. Scott observed that the casting had suffered extensive damage. Indeed, it was squashed and broken at the top. Digging into the almost-full basket, Scott saw that other castings were similarly damaged. The whole basket appeared to be damaged. (2:269; 21:3253, 3259; 23:3457.) Banales was trimming a relatively large item, an aluminum alloy hood for a gas lamp. Measuring some 10 inches square and about 6 inches high, the hood has a collar at the top forming a round opening spanned by two bars crossing at right angles.50 Cross bars were either broken or bent inward, and the collars of several casting were cracked or bent. In this trim operation, the operator places the hood casting on the nest (the bottom half of the die). With the casting in that position, when the press is activated the top half of the die closes. The first part of the casting touched as the top die closes is the area with the cross bars. (21:3253–3258; 23:3457–3460.) When Scott saw the damaged castings, he told Banales to stop. Inspecting the top die (or top half), Scott found large broken pieces of the cross bars and top section of the hood jammed in the die. (21:3256–3257; 23:3459.) Ultimately, the die had to be sent out for disassembly in order to remove the casting pieces, so tightly were they wedged in the die. (21:3261.) After his initial inspection of the castings in the basket and the pieces stuck in the top die, Scott asked Banales, ‘‘Don’t you see the stuck pieces? You are breaking all the castings.’’ Nodding his head yes, Banales answered (21:3259–3260): I don’t get paid enough to be an inspector. I am just here to run the machine. Banales claims he did not know the parts were damaged and did not learn until the morning of March 10 when the inspector told him. (14:1961.) Scott testified that he decided to give Banales a written warning because the problem was so obvious he was sure Banales was aware of it, yet, in total disregard for company property, Banales continued producing damaged castings. Moreover, Banales’ nodding ‘‘yes’’ to the question of his awareness, plus his answer that he was paid to run the ma- chine and not to be an inspector, also confirmed for Scott that Banales was fully aware he was breaking the castings but did not care. (2:269; 21:3260–3261.) Scott does not list the fact that Banales had to remove each part by hand. On rebuttal Banales disavowed any intentional damage on March 10 or any other date. (26:3962–3963.) Banales never ex- pressly denies giving the nodding ‘‘yes’’ answer and making the inspector comment which Scott attributes to him. Banales asserts that he made quick visual checks of the castings before and after he trimmed them. The ‘‘before’’ check is, of course, to spot any defects before he begins his trimming operation. Indeed, he found some with ‘‘blisters.’’ Banales claims he was never told he must inspect the pieces or do perfect work, nor was he ever told that the trimmers should check every 10 to 15 pieces.51 Nevertheless, on March 9 he supposedly made a quick visual check of each part, before and after. (14:1964–1966, 1999–2000, 2015.) In- deed, Banales concedes that a trimmer should follow that procedure, plus using the airhose to clean out the die. (14:1994.) Banales’ production report (R. Exh. 95) for March 10 is in evidence. It records 544 good castings of the lamp hood, with no scrap shown, during his 7 a.m. to 4:30 p.m. shift. (22:3273.) Normally, Banales testified, he would trim about 700 of these units in a 9-hour shift. (14:1964, 2017.) Neither the General Counsel nor Acme offered, or even identified, the production report for March 9. Indeed, in neither evi- dence nor briefs do the parties seek to reconcile the dif- ference in dates. One side (the General Counsel) focuses on March 9, the other (Acme) on March 10. The General Counsel, through Banales, offers evidence of instances in which Banales and others produced many defec- tive parts without receiving a written warning. His own prior instance, occurring about a month before the incident at issue, resulted in some 100 defective castings. However, the defect was caused by a defective die, as described in his pre- trial affidavit. (14:1998–1999.) On redirect examination Banales claims it was the same as here—pieces of broken castings had become lodged in the die. (14:2013.) All during this timeframe after the election Banales openly wore union insignia. During those weeks he sometime produced damaged parts, including the 100 bad parts in early February. He re- ceived no warnings. (14:1995–1996.) According to Banales, within days after his March 9 work he observed Scott, on separate occasions, tell setup persons Guiseppe Genualdi and W. J. McCullough to correct their dies because the dies were producing bad castings. Genualdi produced about 400 defec- tive parts and McCullough about 500. Neither Genualdi nor McCullough wore any union insignia. Scott did not take ei- ther to the office. Banales concludes that neither received a written warning. (14:1967–1971, 2005–2007.) Banales claims that his own March 9 problem was the result of a defective die apparently resulting from a broken piece sticking in the die. (14:2005, 2006.) Scott asserts that Banales had never previously caused the type of damage that he did on March 10. (21:3260.) According to Banales, at the warning meeting on March 10, Scott said that he had no complaint about Banales in all 1120 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD the time Banales had worked for him, but that this time Scott was giving him a written warning. (14:1960–1961.) Banales does not tell us whether he makes some mental distinction between complaints about work quality as compared to his October 1987 written warning (R. Exh. 96) from Scott for excessive breaks. Scott does not describe the warning meeting, but San Roman does. Although Roman places Banales and Valenzuela there, no one confirms this. Testifying that Scott displayed the damaged castings and, it appears, charged that Banales had failed properly to check the parts and to clean the machine, Roman asserts that Banales neither denied caus- ing the damage nor offered any excuse. Roman explains that he is married to a sister of Banales. (20:2936–2937.) Scott testified that Banales should have oiled the trim die no less often than every six or seven cycles. (21:3257.) Roman ex- pressly attributes the damage to the failure of Banales to oil the castings and to check them frequently. Banales should have checked about every 10 parts. Had he done so, he would have prevented the extensive damage. Roman con- cluded that Banales intentionally damaged the castings. (20:2974.) c. Discussion (1) Section 8(a)(3) The General Counsel failed to establish, prima facie, that Acme—and particularly Supervisor Jim Scott—was unlaw- fully motivated in issuing the March 10 written warning to Rodolfo Banales. As Banales admits, he was openly active for the Union when, a month before the instant event, he damaged 100 castings yet received no warning. Although his remark to Novak possibly (but not necessarily) occurred later, Novak’s purported reply indicates ambiguity but not animus. Banales begins his account as of 7 a.m. on March 10. Scott’s version clearly is later in the day. Although Banales begins with a count of 75 damaged castings, he reluctantly agrees, on cross-examination, that he counted 100 to 125— the number Scott specified in the warning. Similarly begin- ning with a description of the damage as being a small crack on the inside of the collar opposite from him, Banales even- tually describes about the same cause as did Scott—pieces jammed in the top die. All circumstances considered, includ- ing the failure of Banales to deny the nodding-yes answer and his not-paid-to-be-an-inspector remark, it is clear that Scott’s version is the more credible. That being so, it means that Scott did not seize on a pretext, that there is no dispar- ity, and that no improper motivation is shown. This is not to say that Scott’s version does not have a question or two. Thus, Scott never explains (he was not asked) why the written warning does not include Banales’ ‘‘inspector’’ remark. One wonders why one of the damaged castings was not preserved and photographed, and why the production report for March 9 was not introduced (although that seems more the Government’s burden). Why does the warning (G.C. Exh. 19) state that Banales continued the damage until the end of the shift when Scott in fact stopped him ‘‘earlier in the day?’’ Finally, why did Scott specify ‘‘Carelessness’’ rather than, for example, ‘‘apparently inten- tional damage?’’ The March 10 production sheet (R. Exh. 95) lends some support to Scott’s version. Thus, it shows 544 castings rather than the 700 Banales usually produces of that part. That suggests Banales stopped production earlier than 4:30 p.m.—an explanation consistent with Scott’s testi- monial sequence. This fact apparently overrides the otherwise inconsistent time of ‘‘until the end of the shift’’ specified in the warning. As for Scott’s reason in restricting the itemized reason to ‘‘Carelessness,’’ Scott is not asked. In any event, it is not a critical aspect. Crediting Scott and Roman over Banales, I find Scott rea- sonably believed that Rodolfo Banales had acted with delib- erate carelessness for the purpose of damaging castings which Acme had entrusted to his care. Finding, therefore, that the General Counsel has failed to establish, prima facie, that Acme was unlawfully motivated in issuing the March 10, 1988 written warning to Rodolfo Banales, I shall dismiss the 8(a)(3) aspect of complaint paragraph 11(f)(6). (2) Section 8(a)(5) I shall dismiss the 8(a)(5) allegation. Acme had issued preelection warnings for production matters, although none because of intentional damage, or even for ‘‘Carelessness.’’ Examples of other extensive damage to parts, by Banales and others, actually were caused by defective dies. Although the evidence fails to express it clearly, it appears that, on those occasions, the dies either were improperly aligned (producing a dimensional defect) or themselves miscast or broken. Aside from Banales’ testimony (14:2013) that his February damage also was caused by broken pieces jamming in the die, there is no evidence that the February damage had the same nature as the March damage. Not crediting Banales, I find that his February defects were of a nature indicating no fault on his part. Accordingly, as the March 10, 1988 written warning to Banales was consistent with Acme’s past practice, did not in- stitute a new system, and was not a harsher enforcement of its rules so that conduct previously tolerated before the elec- tion was now subjected to discipline, I now shall dismiss complaint paragraph 11(f)(6), as to the 8(a)(5) as well as the 8(a)(3) violations. 8. Warning of March 10, 1988, to Sacramento Olivares for producing defective parts a. Pleadings As with Antonio Loza and Rodolfo Banales, Acme issued a written warning to Sacramento Olivares for producing de- fective parts. Complaint paragraph 11(f)(5) alleges that Acme issued the written warning on March 10, 1988. The com- plaint also alleges that Acme, by issuing the warning, vio- lated Section 8(a)(3) and (5) of the Act. Admitting the fact of the warning, Respondent Acme denies that it violated the Act by issuing the warning. (R. Exh. 26 at 4, 5.) b. Facts Sacramento Olivares is a punch (trim) press operator in Jim Scott’s ADC. As of his July 1989 testimony, Olivares had been working at Acme for 10 years. (11:1479, 1520.) Before the October 1987 election, Olivares actively sup- ported the union organizing. His support included distributing cards and wearing union insignia. His wearing of the union insignia continued after the election, and he assisted the 1121 LOVEJOY INDUSTRIES 52 Apparently a reference to the six stewards named by the Union in its November 17, 1987 letter (R. Exh. 1) to Acme’s Balma. 53 Recall that San Roman was promoted to be one of Scott’s as- sistant supervisors on February 15, 1988. (20:2910, 2957.) 54 The transcript actually reads [emphasis added] ‘‘He told me that he knew already the hours of the steward of that department.’’ The parties do not show this error on the list of corrections. (R. Exh. 131 at 47.) Testifying in English at that point (11:1482), Olivares spoke with an accent (11:1487) which no doubt explains the error. I now correct the error to read ‘‘already that I was the steward.’’ 55 Some exhibits are duplicated. This one, G.C. Exh. 20, also ap- pears in the record as G.C. Exh. 5nn. (11:1485.) 56 San Roman asserts that Scott presented the warning in the su- pervisor’s office in the presence of himself and Marcial Canales (20:2934–2935). Scott does not tell us the location. Canales is not asked about the matter and, therefore, neither confirms nor denies his presence. Not even Olivares names Canales as being present, however. I find that Canales was not present. 57 Initially Olivares testified that Diaz first told him, and thereafter Scott came. (11:1490.) On cross-examination Olivares reports that Diaz and Scott came together. (11:1584.) When later called by the General Counsel, Diaz was asked about other matters. 58 Cast from an aluminum alloy (23:3474), the part, shaped like a frame, is easy to pick up because it measures about 1 inch in height, about 7 inches in width, and about 9 inches length, with the frame being about 3/8 inch in thickness. (22:3296; 23:3474.) A photo (R. Exh. 99) in evidence is of a similar, but not identical, casting. (22:3298–3299.) 59 Several deep nicks and gouges along the bottom edge of the front side of the casting. (22:3295; 23:3473.) ‘‘seven leaders’’52 of the Union. Balma acknowledges that Olivares wore a union button, as did a majority of the em- ployees. (25:3907.) Balma, in the next breath, asserts that he only recently learned that Olivares is a union supporter (per- haps meaning he recently learned that Olivares is a steward). I construe Balma’s testimony to mean he distinguishes the general support shown by the majority of employees in their wearing of union insignia from the more active role per- formed by an elected and active steward. While that distinc- tion possibly can be relevant on the subject of motivation, knowledge of open support of the Union is just that— knowledge. After Raymundo Aguirre’s February 25 discharge, Olivares was elected to replace him as steward for the ADC. (11:1480–1481, 1485.) A week later San Roman53 told Olivares that he knew ‘‘I was’’ the steward in the depart- ment, that Roman would like to get along with Olivares, work together, and understand each other.54 (11:1482–1483.) When subsequently addressing events pertaining to Olivares, San Roman does not deny this assertion of Olivares. Credit- ing Olivares, I find that Roman did so remark to Olivares in early March. There is no evidence, however, that San Roman ever informed Scott that Olivares was now a steward. As of the hearing, Balma testified that he was not aware that Olivares was a steward. (2:194; 25:3907.) I credit Balma. If Balma was not aware, that would suggest that Scott was not aware. On Thursday, March 10, 1988, Scott issued a written warning (G.C. Exh. 20) to Olivares.55 (2:272; 11:1485; 20:2934; 22:3294.) Scott checked item 13, ‘‘Others,’’ writing beside it the word, ‘‘Carelessness.’’ For remarks Scott wrote: Sacramento was performing the first trim operation on part no. 844192732. The castings became heavily dam- aged on face area. He continued trimming and 300+ castings had to be scrapped. Olivares testified that Scott, accompanied by San Roman, came to his work station56 on March 10, presented the warn- ing to him, and said that it was for bad castings he had made a week earlier. Olivares asked why Scott was now telling him this when a week earlier he said he was not going to do anything about them. Scott, Olivares testified, said he had given warnings to two other employees for the same reason and it would not look good if he did not issue this warning to Olivares. Scott asked Olivares to sign. Olivares refused. (11:1486–1489.) On the day he produced the bad parts, Olivares testified, he learned of the defective parts when Scott and inspector Nelson Diaz came to his machine.57 Scott said bad castings were being produced, but he did not know why. They showed him some bad castings, and Olivares looked in his basket, which contained about 400 of the parts. Checking about 50, he found about 30 of them defective. (11:1490, 1590.) After Scott and Diaz left, Olivares began analyzing the manner of his operation. His operating method admittedly in- cluded the following procedure when a piece of scrap would become lodged in the die. Taking the casting part in hand, he would use it to hit the scrap piece in order to dislodge it.58 As the part hit the scrap, the face of the casting would hit the steel mold, or die. As Olivares concedes, that dam- aged the face of the casting. (11:1491, 1577–1578, 1589.) As Scott explains, the damage consisted of deep gouges in the soft face of the casting. Because the casting face is painted and letters are applied by silk screening (on a smooth sur- face), the gouges ruined the castings. (22:3294–3295; 23:3473–3474.) According to Roman, Scott showed Olivares the damaged castings, told him he had ruined about 300, gave him the written warning, and said he must be more careful. Olivares made no response. (20:2935.) Later, apparently, Olivares told Roman that he had taken the casting and hit the die in the process of dislodging a piece of scrap from the die. (20:2965, 2971.) In Roman’s opinion, based on what Scott and Olivares told him, Olivares did not properly check the cast- ings. Based on his own experience, Roman testified, Olivares would have found the defect had he checked the parts. Roman acknowledges that Olivares did not say anything to indicate the damage was intentional. (20:2967.) Nevertheless, in San Roman’s opinion, the damage was intentional. (20:2964, 2966, 2972–2973.) Scott testified that a QC inspector alerted him to the prob- lem. (2:273; 22:3296; 23:3474.) When Scott arrived at Olivares’ machine Olivares had a casting in his hand. Taking the part, showing the damage59 to Olivares, and asking Olivares whether he could not see the damage, Scott told Olivares ‘‘You know this is no good.’’ ‘‘Yes,’’ Olivares re- plied, ‘‘But I don’t get paid enough to look that close at them.’’ (2:273; 22:3297, 3300.) Scott testified that the opera- tor should oil the die for this part about every 10 cycles of the press and blow out the pieces of flashing that had fallen into the nest. (22:3297.) In Scott’s opinion, the damage reflects extreme careless- ness (2:277; 22:3294) by not removing scrap that had fallen 1122 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 60 Balma explains that ‘‘shots’’ is a term which refers to a cycle of the machine. In die casting, one shot, or cycle, may produce sev- eral parts. (1:118.) At the trim press it appears that only one part is trimmed at a time. 61 At the hearing Scott, I find, confused the Olivares case with Banales’ and therefore reported similar circumstances for both. into the nest, or bottom die (22:3297). Scott issued the writ- ten warning to Olivares because he felt that the damage was intentional. (22:3300.) There is no dispute that Olivares is an experienced trim press operator. Scott usually assigns that part to Olivares because he generally is very quality con- sequence (2:272), careful, and knows what to look for. (22:3298.) Conceding that he has worked on the trim operation for the casting part some 8 to 10 times previously (11:1588), Olivares asserts that on the previous occasions there were bad parts, too (11:1492, 1600), but that he never had re- ceived a written warning for bad castings (11:1492, 1576) or for anything else. (11:1485.) Admitting that he knew the casting face had to be painted after he trimmed it, Olivares claims no one ever told him the face could have no flaws. He asserts he did not understand how critical the surface of the face was and that no one had ever explained it to him. (11:1586, 1596.) When he is working, Olivares explains, he does not look at the face because he is working fast, and the face is opposite from him when he removes it and puts it in the basket. (11:1584.) Olivares explains that his inspection process was a quick look, first at the beginning and then at about one or two parts every hour and a half. (11:1587, 1596, 1602–1604.) Olivares appears to estimate about 3.33 hours to produce 300 of these parts, at 40 seconds a cycle (11:1583–1584). Assert- ing that a check should be made after every 15 to 20 parts, Roman estimates that it took Olivares about 2 hours to produce the defective pieces. (20:2971.) Olivares denies saying anything to Scott about pay. What he said, Olivares explains, is that he could not be checking every casting. (11:1581.) Olivares asserts that he did not know he was producing bad parts, and had he known he would have tried to correct the problem. (11:1582.) At the rebuttal stage, Olivares denies that anyone, union representa- tive or other person, suggested that he deliberately make bad parts as a way of forcing Acme to give the employees a pay increase. He further denies deliberately making defective parts for any reason. (26:3954–3955.) As mentioned, when Olivares trimmed this part on earlier occasions he also produced some scrap. He would just be shown the damaged parts and the scrap would be taken to be remelted. Before the election he simply would be told to pay attention. Some 5 to 6 months earlier than the March in- cident, or about September-October 1987, Olivares even damaged a die, yet Scott did not issue him a written warning. (11:1492–493, 1600–1601.) Olivares asserts that he did not count the scrap parts on the previous occasions that he made this part because the employees do not count scrap or enter scrap on a report. (11:1601.) This bit of testimony is puzzling because the pro- duction reports have a column for good castings and one for scrap. These are in addition to the column for the gross num- ber, or ‘‘Shots.’’60 On his reports for the first 11 days of March 1988, for example, Olivares filled in the ‘‘Shots’’ col- umn, but entered no figures in either of the columns for good and bad castings. (G.C. Exh. 22.) Reliability of these reports as to scrap is questionable. Already we have seen that the scrap reported can be caused from the die cast operator rath- er than by the trim press operator. Scott tells us that the pur- pose of the production sheets is to determine and monitor production standards and performance rather than to focus on scrap produced by an operator. (2:276–277.) Indeed, Balma explains that Acme (as of the opening of the hearing) is in the process of implementing a new system which, it appears, will record the scrap count reliably. (1:142.) Scott (Acme, actually) failed at the hearing to resolve cer- tain discrepancies. One discrepancy, for example, is Scott’s confusion over the dates. When called under FRE 611(c) early in the hearing, Scott stated that the March 10 warning to Olivares had issued for parts which Olivares had produced on March 10. (2:272.) Scott later corrected the date of dam- age to be March 4. (22:3300–3301.) However, Scott never explains why the date of the warning remains March 10. I therefore accept the version of Olivares regarding Scott’s comments to him on March 10—that he had issued warnings to two others (Loza and Banales, apparently) and had to be consistent. Correspondingly, I do not credit San Roman’s version that the warning was delivered in the supervisor’s of- fice. Scott also fails to address why the warning fails to record the purported remark of Olivares, that he was not paid enough ‘‘to look that close at them.’’ Scott, I find, here con- fuses a similar comment by Banales and transfers it, by men- tal error, to Olivares as well. What assisted in causing Scott to make this error was the admitted comment of Olivares that he could not check every part. I find Scott’s mental error to have been caused by factors other than animus. In short, although Scott initially (on March 4) told Sac- ramento Olivares that he was not going to do anything about the damage, 6 days later Scott, to appear consistent, issued the warning. c. Discussion (1) Section 8(a)(3) I shall dismiss the 8(a)(3) allegation. Skipping for the mo- ment the question of knowledge, and addressing the issue of motivation, I note the lack of evidence showing animus to- ward Sacramento Olivares. Animus, and unlawful motivation, must be inferred from the circumstances. The General Coun- sel apparently relies principally on much of the same pur- ported disparity evidence argued in the case of Banales. However, those other incidents are inapposite because they involved damaged or misaligned dies. Here the die was not damaged. Accepting the version of Olivares and San Roman on the cause of damage,61 I find that the damage here resulted from the deliberate careless- ness displayed by Olivares. Initially (March 4) not knowing what had caused the problem, but later, I find, learning from San Roman how Olivares had used the soft metal castings as a hammer to knock out scrap pieces stuck in the die, on March 10 Scott, I find, decided that Olivares had intended the carelessness and that a warning should be issued to him to be consistent with the warnings issued to Loza and Banales. Certainly the ‘‘not paid enough to look that close’’ 1123 LOVEJOY INDUSTRIES remark by Olivares supports Scott’s conclusion of deliberate damage. The General Counsel’s evidence fails to show, prima facie, that union animus was a moving reason for Scott’s issuance of the March 10 warning to Sacramento Olivares. Scott ap- parently was confused at the hearing and therefore, it ap- pears, described the Olivares circumstances as being similar to that of Banales. Despite Scott’s confusion at the hearing, the evidence fails to show disparity, demonstrates that Olivares acted with deliberate carelessness, and supports the finding, which I make, that Scott decided that Olivares had intended his carelessness. To the extent Scott was persuaded to so act in order to be ‘‘consistent’’ with his issuing warn- ings to Loza and Banales, the fact remains that the basis of Scott’s action was deliberate carelessness by Olivares that prompted the action. Olivares testified that carelessness or inattention in the past, before the election, resulted in nothing more than an oral caution or reprimand. (11:1492–1494.) Scott testified that never before had Olivares created scrap in the quantity he did on this occasion. (22:3301.) Crediting Scott on this point, I find that, even if the General Counsel has shown a prima facie case, Acme demonstrated that it would have issued the warning to Olivares for this deliberate carelessness regardless of his union activities. Accordingly, I shall dismiss the 8(a)(3) aspect of complaint paragraph 11(f)(5). (2) Section 8(a)(5) Scott, as I have found, confused the case of Sacramento Olivares with that of Rodolfo Banales. Nevertheless, the evi- dence demonstrates that Olivares, from malicious intent, damaged the soft metal castings when they hit the steel die as he used the castings to knock scrap pieces loose from the die. Belatedly concluding (as I have found) that Olivares in- tentionally damaged the pieces, Scott issued the warning in order ‘‘to be consistent.’’ Scott’s ‘‘consistent’’ remark is am- biguous. Among several possible meanings, one is that Scott, despite thinking that Olivares had not been at fault, or at most inattentive, had decided to issue the warning because the numbers damaged were so many and Loza and Banales had been warned for damaging large numbers of parts. That is a strained interpretation. The more likely meaning, and the one I infer, is that Scott, initially unaware how the damage occurred, and later learning from San Roman how Olivares had used the castings as a kind of poking stick or hammer, had decided that Olivares had intended the damage. The not- paid-enough-to-inspect remark by Olivares is certainly con- sistent with intentional misconduct. Having recently issued warnings to Loza and Banales for similar misconduct, Scott issued a written warning to Olivares to be consistent. Acme argues that the Olivares warning is similar to preelection warnings to other employees for displaying a bad attitude about doing a good job. (Br. at 339.) Those warnings are inapposite because they issued for ‘‘insubordination’’ and/or a ‘‘failure to obey instructions.’’ (G.C. Exh. 7–10, R. Exh. 130–10.) Before the election, routine carelessness did not trigger written warnings. Acme was not at liberty after the Union’s election victory to begin issuing warnings for carelessness or mere negligence. Although Scott wrote ‘‘carelessness’’ on the warning form, in fact Scott issued the warning for intentional carelessness. That is, for deliberate damage. A written warning in such circumstances was not a departure from past practice, did not institute a new warning system, and was not a harsher enforcement of its rules on conduct previously tolerated. Accordingly, I shall dismiss the 8(a)(5) allegation. 9. Warning of March 11, 1988, to Jose Paz for starting car a. Pleadings Complaint paragraph 11(f)(7) alleges that on or about March 11, 1988, Acme issued a written warning to Jose Paz for leaving the facility to start his car engine before the end of the shift. This fact is alleged to violate Section 8(a)(3) and (5) of the Act. Acme admits the warning fact, but it denies violating the Act. (R. Exh. 36.) b. Facts In October 1987 Ronald Adamczyk was supervisor of the precision machinery department (PMD) and the waveguide department (WGD). During the latter part of 1987 Adamczyk also supervised the packing department. (2:231; 20:2833, 2835.) The packing employees included Juan Lopez and Jose Paz. (20:2865–2866.) Neither Lopez nor Paz testified. The testimony of Robert Novak, Acme’s president, and super- visor Adamczyk is undisputed. In summary, on October 6, 1987, several days before the election, Novak, returning to the plant from a business meet- ing, observed Juan Lopez sitting in his parked car on the parking lot. The time was around 4:15 to 4:20 p.m. Novak instructed Adamczyk to tell Lopez he would issue a warning to Lopez for being in his car before 4:30 p.m. (20:2847– 2848, 2886; 24:3615–3617.) Leaving immediately for the parking lot, Adamczyk found Lopez, who offered no defense. The following day, October 7, Adamczyk issued a written warning (R. Exh. 3) to Lopez. (2:237, 243; 20:2847.) The warning, with item 9 marked (‘‘leaving work without permis- sion’’), recites the facts, reminds Lopez that he is not to leave the plant until after punching out at 4:30 p.m., and warns that a repetition will result in Lopez’ being suspended for 3 days without pay. The complaint does not allege the Lopez warning to be unlawful. Imparting a sense of deja vu to the case, the situation re- occurred on March 10, this time involving Jose Paz. Novak notified Adamczyk who went to the parking lot and con- fronted Paz. Paz offered no excuse. Adamczyk issued Paz a written warning (G.C. Exh. 16) the following day, March 11. (2:236; 20:2855–2856, 2896–2897; 24:3616–3617.) With items 7 (‘‘Failure to obey instructions’’) and 9 (‘‘Leaving work without permission’’) checked, the March 11 warning to Paz reads: On March 10, 1988 at 4:25 p.m. you were out of the building warming up your car. You have been warned verbally that no one was to leave the building until 4:30 p.m. If this occurrence should happen again you will be suspended for three days without pay. The reference to prior verbal warnings appears to refer to Adamczyk’s talk to all his employees and not to some indi- vidual warning to Paz. (20:2888–2890.) Adamczyk testified that he issued the warning to Jose Paz without any knowl- edge of union activities by Paz, and that any union senti- 1124 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD ments by Paz played no role in Adamczyk’s decision to issue the written warning to Paz. (20:2856.) Novak did not address this topic. c. Discussion (1) Section 8(a)(3) Articulating no theory in support of an independent 8(a)(3) violation under complaint paragraph 11(f)(7), the General Counsel apparently relies on paragraph 11(a)(7) in arguing that the warning issued only because Acme launched a gen- eral crackdown against its employees. Faced with the Juan Lopez precedent, the Government now does not point to the election, but to the mid-August 1987 demand for recognition in arguing that the Juan Lopez warning ‘‘is consistent with counsel for the General Counsel’s theory that Respondent initiated a crackdown in discipline in response to the Union activities of these employees, in light of its issuance during the period after the demand for recognition in mid-August 1987.’’ (Br. at 79, 86, 88–89.) The General Counsel’s argument ignores the more likely fact that Acme, having won the two prior elections, was treading as lightly as possible before the election in order to avoid upsetting the employees. As I found earlier, this is why, despite all the bluster by Balma at the hearing, Balma and Novak actually did very little after midsummer by way of implementing Balma’s desires for order and discipline in the plant. Thus, Novak’s personal involvement resulted in a warning to Juan Lopez but not in any preelection reminder, by posting or group meetings, that employees are not to leave the plant before 4:30 p.m. The evidence is woefully insufficient to indicate that either Adamczyk or Novak acted from an unlawful motivation. In- deed, there is no evidence that Jose Paz supported the Union or that Adamczyk or Novak suspected that he did. I shall dismiss the 8(a)(3) aspect of paragraph 11(f)(7). (2) Section 8(a)(5) The Government fails to articulate its theory of an 8(a)(5) violation. The preelection warning to Juan Lopez stands as precedent. I shall dismiss the 8(a)(5) aspect of complaint paragraph 11(f)(7). F. Coercion Mid-March to Early June 1988 1. Alleged threats to deny scheduled wage increase a. Pleadings Complaint paragraph 5(a) alleges that Acme, through Peter Balma on March 17 and June 6, 1988, and Robert Novak on April 14 and May 2, ‘‘threatened employees that they would not receive scheduled wage increases because they had se- lected the Union as their bargaining representative.’’ These threats, the complaint alleges, violate Section 8(a)(1) of the Act. Acme denies the allegations. (R. Exh. 36 at 2.) b. Peter Balma (1) March 17, 1988 (a) Evidence The union leaders, including Marcial Canales and Nicolas Valenzuela, engaged in a continuing effort to persuade Acme to grant, per asserted past practice, general wage increases twice a year. Now coupled with that project was their new cause—seeking the reinstatement of Raymundo Aguirre. On March 17, Canales, Valenzuela, and a group of employees met with Balma on the plant floor. The conversation was in English. After attempting unsuccessfully to present Balma with a petition to reinstate Raymundo Aguirre, they asked for a general wage increase and for recognition of the Union. Balma’s response is in dispute. According to Valenzuela, whose English is limited, Balma replied that Acme could not give a raise because that was controlled by the Union, and as far as recognition, he knew nothing. (3:477–478; 6:851.) Valenzuela concedes that Canales explained to him and the others what they did not understand of the conversation. (6:850.) Explaining that he is the one who asked Balma, Canales testified that he asked Balma for a raise for the workers. Balma said no. When Canales asked why not, Balma replied, ‘‘Because of the Union, we ain’t going to have no raise. I told you guys.’’ (13:1799–1800, 1896.) Balma testified that in late 1987 and early 1988 groups of employees, led by Canales and Valenzuela, came to him and asked about a general pay increase. Balma told them he could do nothing about it, and he asked them why they bugged him. (25:3802–3803.) Eventually Balma complained to Novak. Novak consulted with Acme’s attorneys who, in turn, furnished Acme with a guideline answer (R. Exh. 119) for questions about a wage increase and about recognition. (23:3561; 24:3617–3618.) A copy of the guideline reached Balma around March 1. (25:3804, 3806.) The guideline reads (R. Exh. 119): Due to the unsettled case with the Union, the Com- pany is unable to discuss any general increases. The election was not recognized due to the unfair ac- tions on the part of the Union, therefore the Company filed a petition with the Labor Board. After he received the guideline, Balma, without reading the guideline verbatim, answered the wage increase questions by saying he was unable to discuss any general wage in- crease because of the unsettled case with the Union, that he had no authority as to that anyhow, and why do they bug him. (25:3806, 3880–3881.) Respecting the March 17 con- versation, when Canales, Valenzuela, and about 20 to 25 em- ployees approached him, Balma, at the hearing, relies more on his standard answer than he does on describing his answer on that specific occasion (although he is specific as to his answer on reinstatement of Aguirre). Although Balma does not specifically deny the remarks attributed to him by Valenzuela or Canales, he asserts that, after March 1, he re- 1125 LOVEJOY INDUSTRIES sponded only by giving his standard answer. (25:3804–3806, 3881.) (b) Discussion Balma testified more persuasively than did either Canales or Valenzuela, and I credit Balma’s version. Moreover, the different versions of Canales and Valenzuela, on references to the Union, appear to be garbled versions of Balma’s ‘‘Due to the unsettled case with the Union.’’ Even if Balma said, ‘‘I told you guys,’’ that would have been a reference to the numerous times he had told them he had no authority, rather than a reference to some earlier comment not described in the record. But even crediting Balma, the General Counsel argues, his answer is still unlawful because it ‘‘surely created the im- pression that the Union somehow stood in the way of the employees getting their wage increase.’’ (Br. at 20.) Under cases such as Atlantic Forest Products, 282 NLRB 856 (1957), the General Counsel observes, employers must avoid attributing to the union ‘‘the onus for the postponement of adjustments in wages and benefits, or disparaging or under- mining the union by creating the impression that it stood in the way of their getting planned increases or benefits.’’ Histacount Corp., 278 NLRB 681 (1986), cited (Br. at 442) by Acme, pertains to a statement of intention to file objec- tions if it lost the election. The more apposite cases are those dealing with statements about postponements of wage in- creases—such as Atlantic Forest Products. As the Board stated in Atlantic Forest Products, 282 NLRB at 858, the general rule is that an employer is re- quired to proceed with an expected wage or benefit adjust- ment as if the union were not on the scene. By exception to this rule, an employer may postpone such a wage or benefit adjustment so long as it makes clear to the employees that the adjustment would occur whether or not they select a union, and that the sole purpose of the adjustment’s post- ponement is to avoid the appearance of influencing the elec- tion’s outcome. In making such announcements, however, an employer must avoid attributing to the union the onus for the postponement of adjustments in wages and benefits, or dis- paraging and undermining the union by creating the impres- sion that it stands in the way of employees getting planned wage increases and benefits. As in Atlantic Forest, here Balma and the guideline stated that discussion of a general wage was prohibited because there was litigation with the Union. No explanation was given as to what litigation (objections to the election) with the Union had to do with the pay raise, or why litigation (processing objections) prevented discussion when no elec- tion was pending. The message conveyed, therefore, was that the Union stood in the way of a pay raise or even a discus- sion of one. At bottom, the underlying message was, ‘‘You voted in the Union. Now you can wait until the Company exhausts all legal appeals before we even discuss the wage increase you would have received before now had you not brought in the Union.’’ Accordingly, I find that Acme, by Balma’s March 17 statement, violated 29 U.S.C. § 158(a)(1). Hostar Marine Transport Systems, 298 NLRB 188 (1990). (2) June 6, 1988 Because this incident is based on remarks Balma allegedly made at a meeting convened on June 7, 1988, to discuss poor production by Nicolas Valenzuela, I postpone my summary of the remarks until I discuss that meeting in conjunction with the written warning issued to Valenzuela on June 16, 1988. c. Robert Novak (1) April 14, 1988 (a) Evidence Canales testified that on April 14 he and Valenzuela met with Novak and Balma in Balma’s office. The conversation was in English. After first inquiring about Raymundo Aguirre, Canales asked Novak for the 6-month raise for all the workers. The raise, Novak responded, was under the Union’s control, and he was sorry, but the employees were not going to get anything. (13:1801–1802.) Valenzuela confirms that Novak said Acme could not give a raise because the matter was controlled by the Union. Novak added, Valenzuela testified, ‘‘Ask the Union for the raise.’’ (3:502.) Valenzuela concedes that he could not un- derstand everything that was said at the meeting, and that what he did not understand, Canales interpreted for him. (6:854.) The portions Canales so interpreted are not specified in the record. The evening of April 14 (6:853) Valenzuela gave a four- page affidavit (R. Exh. 6), one of his six pretrial statements (4:638), to NLRB Region 13. Despite Valenzuela’s testimony that the wage increase matter was just as important to the employees as obtaining Raymundo Aguirre’s reinstatement, that he would have remembered the wage increase topic that evening if it had been discussed during the day (6:858–859, 862), the parties stipulated that Valenzuela’s April 14 affida- vit does not mention either the March 17 meeting (6:860) or the April 14 meeting (6:864). Valenzuela does not know why he did not mention them. Probably, he explains, he was fo- cusing on an employee (Raymundo Aguirre) who was out of work (6:862–863) and, in any event, there are many incidents which he does not describe in that affidavit (6:864). If the affidavit described the meetings but omitted impor- tant details, such discrepancies would be significant. As that is not the case, and as the affidavit apparently focused on Raymundo Aguirre, I attached no weight to the fact the affi- davit contains no mention of either the March 17 or April 14 meetings. Aside from the supplemental nature of the affi- davit, what a pretrial statement contains depends largely on the energy, skill, experience, and dedication of the investigat- ing Board agent, in light of the hour and conditions under which the affidavit is taken. Baker Mfg. Co., 269 NLRB 794, 815 fn. 72 (1984). Novak’s testimony is generalized because events on this topic admittedly merge in his memory after the first con- versation he had on this subject with employees on February 29. On that date, giving the substance of the prepared state- ment (R. Exh. 119), Novak said that no discussion could be held or wage increase given at the time because of litigation contesting the election in the unsettled case. Novak repeated 1126 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 62 This incident should not be confused with a similar confronta- tion occurring a few days earlier, as I described in the background section, preceding the Union’s filing the refusal-to-bargain charge. this position on April 14. (23:3562–3563; 24:3618–3620.) In his own testimony, Balma does not address this meeting. (b) Discussion Crediting Novak, I find that the ‘‘control’’ portion of Canales’s version is his shorthand understanding of what Novak said, and that the ‘‘not going to get anything’’ is the conclusion Canales reached based on Novak’s statement there would be no raise ‘‘at this time.’’ I find Valenzuela’s version unreliable because it is unclear how much actually is what Canales told him and what portion, if any, comes from Valenzuela’s own understanding. Even crediting Novak, however, I find, for the reasons stated respecting Balma’s March statement, that his statement violated Section 8(a)(1) of the Act. Telling employees that their pay raise is stalled because Acme is contesting the Union’s election victory puts the onus of delay on the Union’s very presence. Novak offered no explanation as to why Acme tied the raise to the litigation. Accordingly, I find Acme violated 29 U.S.C. § 158(a)(1), as alleged. (2) May 2, 1988 (a) Evidence During the lunch period on May 2 some 60 to 65 employ- ees went to the front of the building. Valenzuela and Canales along with some other employees entered the lobby area. Union Representatives Terry Davis and Lydia Sanchez Bracamonte accompanied them. In Novak’s absence Balma came out. Davis asked Balma about negotiations and, pos- sibly, a general pay increase. The employees began shouting their support of the Union and for negotiations. Balma re- plied that Novak was not there, that he knew nothing, and to go ask the lawyer. Balma went back in the front office and closed the door. The lunch period was over and employ- ees left to return to work. When Balma closed the door, and as the employees left to return to work, the employees shout- ed their support of the Union. (3:503; 6:866–872; 13:1802– 1803.)62 This incident, not alleged as a violation, is prelimi- nary to the second incident that day. Later that day, as the employees prepared around 4:25 p.m. to punch out, Novak and Balma entered the production area. About 80 to 85 employees, including some arriving second-shift employees, surrounded the managers. (3:504; 6:872–875; 13:1803.) Speaking in English, Canales, he testi- fied, asked Novak why he did not want to negotiate and to give the employees a contract. Novak, Canales testified, re- plied that it was in his lawyer’s hands. Canales again asked for a (general) wage increase. Novak said that was under the Union’s control. When someone else in the group asked the same question about a pay raise, Novak, Canales testified, re- sponded, ‘‘I told you guys not to bother with the Union be- cause that was going to happen, no raise.’’ (13:1803–1804.) Valenzuela’s account covers the standard three points the group was to address and does not include the quote which Canales attributes to Novak. (3:505–507; 6:875–877.) Valenzuela admittedly did not understand all the English. (6:875.) Although not fluent in English, Valenzuela speaks and un- derstands some English. On the pay raise topic, Valenzuela also spoke up, apparently repeating the request for a pay in- crease. Addressing Valenzuela Novak asked (3:507; 6:877– 878), ‘‘Oh, you want [a] raise, too?’’ ‘‘Yes, why not? I am working here too,’’ Valenzuela replied. To this, Valenzuela testified, Novak stated (3:507; 6:878), ‘‘You want the Union, go to the Union.’’ Valenzuela specifically testified that this exchange was not interpreted for him (by Canales), that he understood it. (6:878.) Admitting that he has no recollection of a May 2 con- versation, and asserting that he always followed the written guideline in answering these questions, Novak denies ever referencing the Union when stating his position for not giv- ing a pay raise. He specifically denies saying employees would not receive a wage increase because they had selected the Union. (23:3564–3567.) Novak does not expressly ad- dress Valenzuela’s testimony quoted above. (b) Discussion Crediting Canales and Valenzuela, I find that Novak made the remarks they describe. Novak not only lays the onus of no pay increase on the Union, he implies that Acme’s re- sponse was to retaliate against the employees for voting in the Union by indefinitely postponing or even canceling the pay increase. Accordingly, I find merit to complaint para- graph 5(a)(2). 2. June 6, 1988 plant closure threat by Peter Balma a. Pleadings Complaint paragraph 5(b) alleges that on or about June 6 Balma ‘‘threatened its employees that Respondent would close its facility because of the Union.’’ This statement, para- graph 12 alleges, violates Section 8(a)(1) of the Act. By its answer (R. Exh. 36 at 2, 5), Acme denies these allegations. b. Summary postponed As with the other June 6 remark attributed to Balma, I postpone summary of this allegation until I describe Valenzuela’s poor-production meeting convened on June 7. 3. Early June 1988 ban on union T-shirts a. Pleadings Complaint paragraph 6 alleges that about early June 1988 Peter Balma told employees they would not be permitted to wear T-shirts with union insignia during working hours, yet Acme allowed its employees to wear similar apparel which contained no union insignia. Paragraph 12 lists the allegation as one of Acme’s violations of Section 8(a)(1) of the Act. By its answer Acme denies the allegations. (R. Exh. 36 at 2, 5.) Additionally, as an affirmative defense Acme alleges that the allegation must be dismissed because it goes beyond the charge, and Section 10(b) of the Act prohibits any amendment or new charge. (R. Exh. 36 at 8–9.) b. Motion to dismiss denied Acme moves to dismiss (R. Exh. 36 at 8–9; 3:566–567) on the basis that the charge (G.C. Exh. 1q) in Case 13–CA– 17941 (the third of our consolidated cases) alleges written 1127 LOVEJOY INDUSTRIES 63 At the hearing (1:16) and on brief (at 450) Acme also asserts, as an additional ground, that it was not given the opportunity to re- spond to this allegation during the Region’s investigation. As Acme adduced no evidence in support of this naked assertion, I reject it. 64 The only independent 8(a)(1) allegations in other charges filed within 6 months of June 8, 1988, are two mid-September 1988 threats of discharge described in the September 18, 1988 charge filed in Case 13–CA–28033, our fourth case (but not included in the ensu- ing complaint). 65 Chairman Stephens concurred in the Nippondenso dismissal based on his partial dissent in Redd-I. 66 See generally S. Brown and A. Bass, Corporate Campaigns: Employer Responses to Labor’s New Weapons, 6 The Labor Lawyer 975 (No. 4 Fall 1990). Acme also introduced three articles, from dif- ferent publications, describing corporate campaigns. (R. Exhs. 52, 53, 54.) At some point after the October 1987 election, Davis con- cedes, she read the three articles. (15:2118.) 67 References to outside and inside ‘‘strategy’’ really are to tactics. warnings in June 1988 to Valenzuela, Diaz, and one other, the June discharge of employee Juan Lopez, and the August 1988 suspension of Sacramento Olivares, but makes no ref- erence to the T-shirts.63 (Br. at 449–451.) As the General Counsel observes (Br. at 63 fn. 25), a complaint may allege any matter closely related to or grow- ing out of charged conduct. Recent decisions by the Board emphasize that there must exist a factual and legal nexus be- tween the allegation in the charge and that in the complaint. Nickles Bakery of Indiana, 296 NLRB 927 (1989); Advertis- er’s Mfg. Co., 294 NLRB 740 (1989); and Redd-I, 290 NLRB 1115, 1116 (1988). The complaint which issued September 12, 1988, in Case 13–CA–17941, although not attacking any warning to Nelson Diaz, does allege that a June 16, 1988 warning to Valenzuela violates Section 8(a)(3) and (1) of the Act.64 Even though that June 16 warning to Valenzuela is for low production, and not for wearing a union T-shirt, the union T-shirt inci- dent is factually and legally related to the later warning to Valenzuela. This is so because the incident, if the General Counsel’s evidence is credited, could show animus by Balma against Valenzuela’s union activities. Such also relates to the type of legal defense Acme would be called on to present respecting the Valenzuela warning. Thus, this situation ap- pears distinguishable from that in Nippondenso Mfg. U.S.A., 299 NLRB 545 (1990), where the Board dismissed a com- plaint allegation (of interference with the distribution of union literature and the wearing of union insignia) for lack of a factual or legal nexus with the charge allegation of a discharge occurring in the same organizing campaign.65 Ac- cordingly, I deny Acme’s motion to dismiss. c. Evidence Before the morning shift began at Acme on June 8, Valenzuela distributed union T-shirts to about 60 employees. Bearing the UE emblem, the red and yellow T-shirts dis- played, in English and Spanish, ‘‘We want a contract.’’ (3:559–560.) There is no dispute that Acme provides uni- forms to the employees, uniforms of green pants and long- sleeved shirts. Although Acme, on brief, questions the testi- mony, witnesses credibly testified that in hot weather, both before and after the October 1987 election, employees occa- sionally would wear, in lieu of their uniform shirts, a variety of T-shirts, some displaying emblems, such as for sports teams, and others plain with no emblems. On this June 8, Valenzuela and Inspector Nelson Diaz were among employees who donned the UE T-shirts in lieu of their uniform shirts. The evidence is disputed whether Balma, later that morning, told Valenzuela to remove his T- shirt and put on his uniform shirt. Valenzuela and Canales (enlisted as an interpreter and witness by Valenzuela) insist that he did. Diaz testified that Balma simply told him to put on his green uniform shirt. (14:2024.) No warnings issued over the incident. Balma testified that he told the employees the uniforms were provided for safety reasons and that they had to wear them but could wear the union T-shirts either over or under their uniform shirts. He denies telling Valenzuela that he had to remove the union T-shirt. (24:3777–3778.) Thereafter, there is no dispute, Valenzuela and some others wore their UE T-shirts over their uniform shirts and continued to do so for some time. Indeed, Valenzuela testified that as of the hearing he and several employees continued to wear the T- shirts, apparently on Fridays. (6:844.) d. Discussion Crediting Balma, who testified persuasively, over the em- ployees, I nevertheless find a violation as alleged. There is evidence that in hot weather employees in different depart- ments occasionally wore T-shirts. Although no testimony shows that Balma or any specific supervisor observed this, knowledge can be presumed from the rather widespread na- ture of the practice. Finally, Balma did not deny awareness of the practice. The record, therefore, shows that Balma per- mits nonunion T-shirts to be worn without the need to wear the long-sleeved work shirt, but he prohibits the wearing of the Union T-shirts unless they are worn over the uniform shirt. In hot weather the union supporters must sacrifice, by bearing the heat, in order to display their union T-shirts. Such disparity of treatment inhibits the exercise of Section 7 rights, while favoring nonunion activity, and therefore vio- lates Section 8(a)(1) of the Act, as alleged. G. The Union’s Corporate Campaign Giving first hint in its opening statement (1:14), Acme de- fends against many of the discipline allegations on the basis that the (asserted) misconduct by the employees was inten- tionally performed on the advice of the Union. The mis- conduct conspiracy assertedly was designed to pressure Acme to abandon its objections in the representation case, to recognize and bargain with the Union, and to grant a general pay increase. As reflected by extensive briefing (G.C. Exh. at 135–151; Acme 239–266, 272–282), substantial evidence was adduced on this matter. Acme’s position, as described at the hearing (3:446–455; 15:2104, 2108), and in its brief (at 239, 273), is that the Union engaged in a ‘‘corporate campaign’’66 strategy. As implied by the names, an outside strategy refers to the Union’s efforts to mobilize support from groups and individ- uals outside the Company. The inside strategy refers to gen- erating action inside the plant.67 To that point, the parties are not in dispute. Davis acknowledges that in the summer of 1988 she consulted an International representative of the UE experienced in corporate campaigns with a view to develop- ing such a campaign for Acme. (16:2173–2174, 2181.) Sug- 1128 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 68 The next best offensive strategy, after attacking the enemy’s plans, is to disrupt his alliances. Sun Tzu, The Art of War 78 (S. B. Griffith translation, Oxford Univ. Press 1963). gestions from that source would be in the nature of expand- ing on tactics, such as mobilizing community support, which Davis already had initiated. Acme contends that the Union’s purpose in mounting its corporate campaign was to pressure the Company to do four things: (1) forgo its legal rights (abandon its objections to the election), (2) recognize the Union, (3) commence bargaining, and (4) grant a pay raise. To achieve this four-fold goal the Union, Acme argues, engaged in unlawful (such as second- ary) activities outside the plant,68 and suggested to employ- ees that inside the plant they adopt unprotected tactics (slow- downs, sabotage). Denying that the Union used or suggested the sinister tac- tics Acme advances, Terry Davis, the Union’s representative, describes only lawful goals and tactics. In the Union’s opin- ion, Acme was insincere with its objections and sought only the delay available in the legal system in a bad-faith effort to avoid recognizing and bargaining with the exclusive rep- resentative selected by a majority of unit employees. (15:2127, 2138.) At the hearing I ruled that evidence about the outside strategy (tactics) would be limited merely to de- scribing its existence, but that evidence as to the inside strat- egy (tactics) would be admissible because of its relevance to credibility of the witnesses concerning the discipline allega- tions of the complaint. (15:2108–2111, 2162; 16:2179–2180.) Davis concedes that the UE’s policy advocates taking a militant and aggressive stance in the workplace. (3:441; 15:2124.) She explains that this means seeking, as employee leaders, employees who will stand up to the ‘‘bosses’’ and display courage in the face of company intimidation. After the April 13 certification, this meant encouraging employees to make group demonstrations in the plant as a means of proving to Acme that a large majority of the employees sup- port the Union and insist that Acme recognize the Union and begin negotiations. (3:440–442; 15:2124.) The Union, inside strategy consisted of tactics such as group demonstrations on nonworktime, petitions, mass grievances filed by shop stew- ards, and wearing of union insignia. (15:2126–2128, 2136; 16:2203–2204.) Overtime at Acme is voluntary. (18:2633.) In the late April-early May 1988 timeframe the Union suggested to em- ployees the tactic of declining to work their hour of daily overtime in order to demonstrate solidarity and to pressure Acme. (15:2135, 2143, 2149; 16:2204, Davis; 16:2553, Curtin.) A large majority of the employees did just that the afternoon of May 10, gathering for a rally outside the plant attended, at the Union’s solicitation, by leaders of commu- nity organizations. (15:2128; 16:2206.) As Davis explains, however, the tactic of refusing to work overtime quickly be- came a divisive issue, and the Union abandoned the tactic. (15:2147, 2149.) By cutting overtime the employees were, of course, reducing their own income. Consequently, Valenzuela reported to Davis that Jose Ortega—an employee Davis de- scribes as being usually opposed to the Union—had sug- gested that the employees could avoid the pay loss associated with refusing overtime by simply slowing down in their work. Davis testified that at the union hall she explained to Valenzuela and other union leaders, in this time period, that such a tactic would be illegal, unprotected by the law, that it would jeopardize jobs, that it was not within the Union’s policy, and that he and the other leaders should explain this to the employees. (15:2142–2146; 16:2205, 2207, 2221, 2225–2226.) Davis denies ever suggesting to employees that they slow down in their production, and she asserts that it was never one of the Union’s tactics. (15:2135, 2147; 16:2242.) As for counseling employees to produce defective parts, Davis testi- fied that at no time did she suggest that employees produce bad parts as a form of protest. (16:2204, 2242.) Davis ex- plains that such a tactic would be counterproductive because it would result in discipline for the employees. (16:2204.) I need not resolve this overall dispute. H. Discipline Imposed June–August 1988 1. Nicolas Valenzuela warned June 16, 1988, for low production a. Pleadings Constituting one of its asserted 8(a)(3) violations, the com- plaint, in paragraph 9(c), alleges that on or about June 16, 1988, Acme issued a disciplinary warning to Jorge Nicolas Valenzuela. Acme admits the fact but denies the violation. The same June 16 warning (further identified as a warning for low production) is alleged, in complaint paragraph 11(f)(8), as being pursuant to the alleged October 1987 uni- lateral changes violating both Section 8(a)(3) (complaint par. 13) and Section 8(a)(5) (complaint par. 14). Again Acme ad- mits the warning but denies the rest. Additionally, I now cover the two independent 8(a)(1) al- legations pertaining to Plant Manager Balma which I earlier postponed. b. Facts As I mentioned in the background section of this decision, Jorge Nicolas Valenzuela began work at Acme in 1971. (4:646.) In 1977 he advanced to setup operator. (4:646, 688.) After his 1984 reinstatement, and until August 1987, he mainly did setups. (6:646.) Formal setup work involves pre- paring machines for operation with the necessary fixtures and at the proper settings, in accordance with blueprints and specifications, and checking tolerances. (3:570, 573–574; 4:652; 7:1068.) Informally, or in practice, there is no dispute that the duties of a setup person also include assisting other operators by securing drills or taps as needed, such as when a drill breaks. For this purpose Acme has given Valenzuela a key to the cabinet where drills and other parts are kept. (3:578; 4:662, 664, 671; 8:1127; 20:3064, 3068, 3070.) Valenzuela testified that in August 1987 Acme began gradually reducing his setup assignments and increasing his production work. (3:571; 4:648–650, 666.) No complaint paragraph alleges that Acme unlawfully discriminated against Valenzuela by reversing his assignments even though it ap- pears that setup work would be more desirable to an em- ployee than straight production work. In any event, by June 1988, Valenzuela testified, his setup to production ratio had been reduced to 20-percent setup and 80-percent production work (3:574; 4:668), or about 2 to 3 hours per day (7:976). At this point I shall defer further description of setups. How- ever, the topic has a bearing on the June 16 warning. 1129 LOVEJOY INDUSTRIES On Thursday, June 16, 1988, Acme issued a written warn- ing to Valenzuela. Item 13, ‘‘Other,’’ is checked followed by the explanation, ‘‘Slow down of production on your job.’’ (G.C. Exh. 30; R. Exh. 76.) The ‘‘Remarks’’ section reads, ‘‘See attached sheets.’’ A page and a half of daily notes are attached followed by (on Acme’s copy, R. Exh. 76) Valenzuela’s production sheets for June 6, 7, 8, and 10, the June 9 production sheet for Angel Otero, and the June 10 sheet for Robert Burris. The daily notes, typed, are for June 6 (a Monday), June 7, 8, 9, and 10. Assistant Supervisor Dan Basgall testified that the notes are his, although he did not type them. (2:380.) Valenzuela worked in Larry Stoner’s secondary machining department (SMD) on the day shift. During the relevant pe- riod Valenzuela worked for Basgall who assists Stoner. Ston- er had divided his department so that he and his two assistant supervisors (Basgall and Faustino Ontiveros) each supervised an area. Although Stoner, of course, had jurisdiction over the whole department, Basgall had primary charge of an area on the east side of the cafeteria, opposite from Stoner’s side of the plant. (4:654–655, 676; 20:2991–2992; 21:3118; 24:3758.) One of the parts Valenzuela produced in late May and early June was a curved, L-shaped part of a propeller blade. A photograph (R. Exh. 16) of the part is in evidence. The part is about 7 inches high, 7 inches across the wider end, and 6.5 inches long. (13:1916–1917.) Valenzuela testified that the aluminum part weighs about half a pound. (3:586.) Valenzuela had never worked on that part previously. (3:358; 7:954–958.) The part number, as shown on the production records, is AMBL 1517. (R. Exhs. 17a–c, 18a–d.) During this time Plant Manager Balma worked nearby 4 to 5 hours a day. Balma noticed that Valenzuela was seldom at his machine. Checking production records with Basgall, Balma determined that Valenzuela was producing far below the figure Acme had expected for the job. Balma checked company records and ascertained that Acme had quoted the job based on an estimated number of parts per hour. Al- though that number is not shown in the record, Balma testi- fied that, using that number, he figured that an operator could produce 3500 parts in a minimum of 4 days. (24:3760– 3763.) Balma did not, at this point, specify a daily or hourly figure. Out of a 9.5-hour shift, 30 minutes is for the lunch or meal break. As we see in a moment, Basgall’s time study formula would apportion 1 hour for the two 10-minute breaks, restroom visits, and material handling, leaving 8 hours devoted to actual production of the parts. Using that 8-hour figure, we see that the 3500 parts computes to 875 per 8-hour day, 109.375 parts per hour or 1.82 parts per minute. As Balma specifically said it would take at least 4 days (25:3870–3871), but at that point gave no maximum, we have no outside figure to give a range. All this is fuzzy, of course, because Acme did not offer its records, the Gov- ernment apparently did not subpeona them, and Balma was not asked the daily or hourly figure shown on Acme’s inter- nal quote sheet. Valenzuela, Basgall testified, had been producing about 300 to 350 of the parts in issue over a full shift (20:3014, 3049), or about 40 per hour. Valenzuela’s production records disclose that, working 1.25 hours on May 26, he produced 79 parts. (R. Exh. 17c.) At slightly over one part per minute, that would be about 63.2 per hour. Although Valenzuela worked a full shift on May 27 pro- ducing the part, he does not list on his production report (R. Exh. 17b) the number of parts he produced that day. For Tuesday, May 31, Valenzuela’s production sheet (R. Exh. 17a) shows that in the first 4.5 hours he produced 210 parts. After lunch and an hour on another job (20:3021), Valenzuela worked on the project 3.5 hours to 4:30 p.m., producing another 200 parts. Thus, for those 8 hours Valenzuela produced 410 parts, for 51.25 hour, or, at .85, fewer than one part per minute. Return now to Balma. After checking the records, as just described, Balma tried his hand at operating the machine and producing the part. This appears to have been about late Fri- day, June 3. Working 1 hour Balma produced three parts per minute. He suggested that Basgall see what he could produce. (24:3763; 25:3865.) Working at an ‘‘average’’ speed from 6 to the 7 a.m. start of the day shift on Monday, June 6, Basgall produced parts at the rate of two per minute. (2:378; 20:3012, 3044–3045, 3058–3059, 3066.) Shortly after the 7 a.m. shift began on Monday, June 6, Basgall told Valenzuela that his production was low and that he should be producing about two parts per minute for a total of 960 for the 9-hour shift. Taking the machine, Basgall pro- duced six parts in less than 3 minutes. Valenzuela said he could do it also, but that no one could keep up that pace for an entire shift. (3:58–584; 6:896; 20:3014–3015, 3045–3047, 3049, 3067.) Valenzuela testified that Basgall, who speaks Spanish, spoke to him in English. Valenzuela apparently re- sponded in English. (3:582; 7:1092.) The shift started at 7 a.m. and ended 9.5-hours later at 4:30 p.m., and included a 30-minute lunch period. Describing how he arrived at his piece-per-minute rate, Basgall testified that he allowed for items such as two 10-minute breaks, rest- room visits, and material handling amounting to 1 hour. The resulting 8 hours at two parts per minute would produce 120 per hour, or 960 for 8 hours during the 9.5 hour shift. (20:3014, 3048.) Although he does not specifically list fa- tigue as one of his allowance factors, Basgall concedes that usually one’s production rate will drop as the hours pass. In- deed, Basgall testified that he simply was trying to give Valenzuela a target figure to shoot for with his daily produc- tion. (20:3047.) Fatigue would be a problem with this part. Unlike other parts which are set and moved by the machine’s electric motor and compressed air, with this part Valenzuela had to hold each piece in place by applying muscle pressure. (3:585–587; 4:678; 7:1078; 8:1120.) Basgall admits he never produced the part for an entire day, and can only approxi- mate a day’s production. (20:3071.) Basgall agrees that Valenzuela expressed the view that the two parts per minute rate was unreasonable. (20:3015.) Ac- cording to Basgall, Valenzuela, displaying a belligerent atti- tude, also responded that he was working (at a good rate, ap- parently) and that he did not care what Basgall thought he should be producing. Although that remark took Basgall by surprise, he gave no (insubordination) warning to Valenzuela. (20:3013, 3015.) Basgall’s notes for June 6 also record that Valenzuela so commented. (G.C. Exh. 30–2.) At first saying he does not recall stating that he did not care, Valenzuela de- nies so replying to Basgall. (3:589.) Balma testified that when he asked Basgall later that day about his conversation with Valenzuela, Basgall reported that Valenzuela said he 1130 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 69 Although Canales apparently interpreted most of the meeting, as Valenzuela and Canales describe, Valenzuela also asserts that he (Valenzuela) sometimes spoke in English. (3:511.) 70 Earlier I postponed summarizing complaint par. 5(a)(1) which alleges that on or about June 6 Balma threatened loss of the sched- uled wage increase because employees had selected the Union. 71 On recross-examination Valenzuela quotes Balma as saying he could not give a raise because everything was controlled by the Union, and when everything with the Union is over with (no conclu- sion stated). However Valenzuela concedes that he does not remem- ber what Balma said at this meeting as opposed to another meeting. (8:1121.) 72 Complaint par. 5(b) alleges a threat by Balma on or about June 6 that Acme would close the plant because of the Union. would try to do better. (24:3765.) Valenzuela concedes that at some point in June, before the June 16 warning, Basgall told him that Balma wanted Valenzuela to quit walking around and to get back to work. (4:693.) That same day, June 6, Valenzuela, according to his pro- duction sheet (R. Exh. 18d), produced 355 parts for the full shift. The following day, Tuesday, June 7, Valenzuela was called into Balma’s office. Canales accompanied Valenzuela as a witness and interpreter. Balma and Basgall attended for Acme. The precise time is not given, but it appears to have been in the early afternoon. (6:897.) Balma asserts that Basgall did most of the talking for Acme (24:3765). Valenzuela (3:590) and Canales (13:807) state that Balma did the talking. In any event, all agree that Valenzuela was told his production was too low. There is general agreement on the topics mentioned there- after, although the sequence and contents differ. I find that the conversation went substantially as follows. When Balma said Valenzuela’s production was too low, Valenzuela through Canales,69 said he would try to make more produc- tion, but for Balma to try and give ‘‘the general raise.’’ (3:518, 590; 13:1807–1808.) Acme’s response is disputed. According to Valenzuela (3:518–519) and Canales (13:1808), Balma responded by saying they were not going to get anything until everything gets straight. To Canales’ question of whether Balma meant because of the Union Balma replied, ‘‘yes.’’70 On cross-examination Valenzuela, his memory refreshed by his calendar diary, concedes that Balma said there would be no raise unless he got more pro- duction.71 (6:911.) When Valenzuela and Canales brought up the raise ques- tion, Balma testified, Balma turned it aside by saying there was nothing they could do at the present and they were there to talk about his, Valenzuela’s, production, not about a pay raise. (24:3766, 3768.) Valenzuela said, ‘‘If I had more money, I could produce more parts.’’ Seeking for Valenzuela to repeat his statement, Balma asked, ‘‘What?’’ (20:3016, Basgall; 24:3765, Balma.) Canales, however, responded, ‘‘He didn’t say nothing.’’ (24:3766–3767.) Basgall recalls that Valenzuela also said he did not say it. (20:3016.) Basgall’s notes record Valenzuela as saying that if he had more money he could do a better job. (G.C. Exh. 30–2.) Canales denies that Valenzuela said this. (13:1897.) Finding Balma and Basgall the more persuasive, I credit them, and I therefore shall dismiss complaint paragraph 5(a)(1). I also find that Valenzuela, in English, in the course of the exchange about a pay raise, unwittingly revealed his thoughts when he stated that if he was paid more he could produce more. According to Canales, after Balma gave his answer that his comment about no raise meant because of the Union, Balma told Canales, ‘‘The plant is going to be closed be- cause of the Union, and the people are going to be mad at you and kill you.’’ (13:1810.)72 Canales said that Balma was wrong, that Acme would never close, and the employees would never give up trying to fix the place. (13:1808–1810.) Without addressing the specific words of Canales, Balma’s version is that Valenzuela and Canales again brought up the issue of money (after Valenzuela’s earlier statement that he could produce more if he had more money). Balma replied, ‘‘You know, with your attitude and the attitude of some of the employees in here, you guys aren’t going to be happy till you close this place down.’’ By this, Balma testified, he meant that if production continues to drop Acme would make no money and the plant would be closed. (24:3767–3768.) Crediting Balma, I find that he added, as Canales testified, ‘‘and the people are going to be mad at you and kill you.’’ I shall dismiss paragraph 5(b). Before leaving the meeting of June 7, I perhaps should note that in the course of the exchange about low production, Balma stated that he had observed Valenzuela missing about 30 minutes from work, all before 11 a.m. that morning, in making three visits to the restroom. Valenzuela said no it was only twice, with the third time being to adjust his glass- es. (3:590; 6:901; 20:3016.) When Valenzuela asked if Balma was checking only him, Balma replied Yes, that he had been watching him all morning and had recorded the number of times Valenzuela had visited the restroom.’’ Balma did not explain why he was doing this, Valenzuela testified. (3:590–592.) Without specifying it as that date, Balma concedes watching Valenzuela on many occasions be- cause he would see Valenzuela simply walking around. Valenzuela would laugh when cautioned to return to work. (24:3768–3769; 25:3868.) Basgall earlier had described Valenzuela to Balma as being arrogant. (25:3864.) That re- fers back to Balma’s testimony, described in the background section, that Acme seemed intimidated by the threat of po- tential litigation with the NLRB if management elected to discipline Valenzuela. Basgall testified that the meeting ending with Balma tell- ing Valenzuela he would have to increase his production, with Valenzuela saying he would try, and with Balma telling him that if he did not do so disciplinary action would be taken. (20:3016.) Basgall’s notes for that date so reflect. (G.C. Exh. 30–2.) Valenzuela’s production report for that June 7 reflects that he produced 340 parts during the 7 a.m. to 4:30 p.m. shift. (R. Exh. 18c.) The time devoted to the meeting in Balma’s office is not disclosed. The following day, Basgall testified, he again told Valenzuela his production was too low, that Valenzuela, with a ‘‘don’t bother me’’ attitude, said he was working. Basgall unsuccessfully tried to help Valenzuela by offering produc- tion suggestions but that Valenzuela rejected the assistance. (20:3017–3018; G.C. Exh. 30–2.) Valenzuela concedes it is possible that Basgall again told him that day that his produc- tion still was too low. (7:956.) Balma testified that Valenzuela continued to walk around and smirk when cau- tioned. (24:3769.) 1131 LOVEJOY INDUSTRIES 73 Balma testified that Otero left Acme in about July-August 1988. (2:184, 188–189.) 74 Valenzuela, inadvertently it is clear, entered the 365 in the scrap column rather than in the next column to the left for good castings. (8:1115–1116.) Also, there is a difference in the last two digits of the numbers of the parts Burris and Otero worked on compared to most of the ones Valenzuela worked on (AMBL 1519 compared to 1517). Valenzuela’s for June 10 was part 1519, however. Respond- ent’s counsel represented that the part numbers describe the same part and that Acme’s evidence would so establish. (7:955, 967.) No witness so states. The Government apparently agrees with counsel’s representation. At the end of that day June 8, Basgall testified, he checked Valenzuela’s production and found it unimproved. (20:3018.) The report (R. Exh. 18b) reflects that for the first hour Valenzuela did what Basgall describes (20:3020) as setup, and from 8 a.m. to 4:30 p.m. Valenzuela produced 301 of the parts in issue. Subtracting 1 hour from the 8 leaves 7 hours to produce the 301 parts—43 parts per hour, or .717 parts per minute. Valenzuela concedes that on June 9 Basgall again told him he should increase his production. (7:958.) To doublecheck Valenzuela’s production rate, and to get more production as well, Balma separately assigned two night-shift employees, Robert Burris and Angel Otero (then an assistant steward of the Union, R. Exh. 1), to produce ‘‘good’’ parts, stressing quality rather than speed.73 Balma spent less than 30 minutes training each (neither had pro- duced these parts before), and did not tell them his purpose. In their one night of production, each doubled Valenzuela’s production. (20:3018–3019; 24:3764–3765; 25:3871–3872.) Otero’s production report is dated June 9, showing a produc- tion of 714 parts on machine 79 (the machine Valenzuela used) after clocking in at 4:52 p.m. and out at 1:45 a.m., or a shift of less than 9 hours. (R. Exh. 18f.) Applying Basgall’s 8-hour formula, we see that Otero produced at the rate of 89.25 parts an hour, or 1.49 per minute. The report of Burris, dated (Friday) June 10 (R. Exh. 18e), shows that he produced 770 parts after clocking in at 4:30 p.m. and out at 1:45 a.m. Calculating by Basgall’s 8-hour formula yields an hourly rate of 96.25 or 1.6 parts per minute. Neither Burris nor Otero (no longer at Acme) testi- fied. Valenzuela testified, without objection, that when Otero reported for work on June 10 he came to Valenzuela and in- formed him that Balma had asked him the previous shift to produce the part. This was strange to Otero because Otero always gets his assignments from the night supervisor, How- ard McArtor. Moreover, Otero remarked, he was tired and sore from the work because Balma had told him to work as hard as he could in order to produce as many parts as he could. (7:972.) In his own testimony, Balma does not expressly address or deny telling Otero to produce his maximum. Only by im- plication from his reportedly telling Otero not to kill himself (24:3764) and that he was not looking for speed but for good parts (25:3871) does Balma deny the attributed assertion. I credit Balma’s implied denial over the unobjected to hearsay report of Otero that Balma told him to do his best. Otero was the assistant steward for the evening shift. By this time em- ployees were expressing their concerns to the Union’s Terry Davis that Acme was pushing them to work harder, to speed up, and was installing counting devices, new production charts, and issuing written warnings as pressure tactics. (16:2226–2236.) It seems unlikely that Balma would ask the assistant steward to work ‘‘as hard as he could’’ and to produce as many parts as he could. And had Balma done so, it seems likely that Otero and the Union would have filed a grievance over the matter. (The Union had filed other grievances.) There is no evidence that Otero filed a grievance or that the Union otherwise protested this request which Valenzuela claims Otero attributes to Balma. I find that it did not happen. The record has no production sheet, or record, of Valenzuela for June 9. Basgall’s notes (G.C. Exh. 30–2, 3) skip from June 8 to June 10. Valenzuela’s production sheet for June 10 (R. Exh. 18a) reflects that he produced 365 parts for the full shift of 7 a.m. to 4:30 p.m.74 Again Basgall’s 8- hour formula yields an hourly average of 45.63 and parts per minute of .76. Apparently on Monday, June 13, Balma went to Robert Novak, reported Valenzuela’s productivity problem, and said he planned to have Basgall issue a warning for a work slow- down. Novak testified that he agreed. (23:3548.) That day Novak telephoned Attorney Salzman, expressed concern over the ‘‘touchy’’ matter, and requested advice on the best way to handle it. Salzman dictated a statement to be read to Valenzuela. (23:3548; 24:3677.) For reasons unexplained, it was not until Thursday, June 16, that Valenzuela was called into Balma’s office. Present were Novak, Balma, Basgall, Valenzuela, and Canales. (7:968; 20:3022–3023; 23:3549; 24:3776.) At the meeting Balma, handing Valenzuela the written warning, said that he had been orally warned to increase his production, had not done so, and now was being given a written warning. (2:380; 20:3023, Basgall.) Novak then read his written statement and left. (23:3550; 24:3676.) As typed, the text of the statement reads (R. Exh. 117): You are participating in a work slow down on the job. (Here explain the number of parts Nicholas made on the dates involved. Then explain the number of parts the other employees made on the dates involved.) You continue to accept your regular wages from the company without providing the standard performance on your job. You’re slow down actions are not protected activity under the law. The company will not tolerate your work slow down. If you continue your work slow down you will be discharged for misconduct. The parenthetical instruction to Novak about the dates and parts prompts an observation about the written warning, dif- ferent versions of which are in the record. When called by the General Counsel, Basgall identified (2:378–381) the warning form with his 1.5 pages of attached notes as the warning issued. (G.C. Exh. 30.) So does Valenzuela. (3:569; 7:970–971.) Eight months later, Basgall identified the full package (R. Exh. 76, including production sheets for Valenzuela, Burris, and Otero) as the warning delivered. (20:3010–3011, 3023.) Valenzuela who, of course, recog- 1132 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 75 Balma testified that well before the hearing Acme fired Magee for failing to show up for work. (2:184.) 76 In conjunction with complaint par. 11(f)(1) and the November 11, 1987 warnings to Antonio Ramirez and Fidencio Olivares. 77 Lopez had transferred to the department some 2 months or so earlier. (2:213.) Even so, McArtor explains, because of the type of equipment and fixtures, experience by an operator at the machine does not generate greater production. (19:2808–2809.) Generating some confusion, however, is McArtor’s later testimony that before June Lopez had performed this same job more than 50 times produc- ing at least 40-percent more. (19:2791.) 78 At the end of a shift, McArtor makes personal notes about pro- duction problems (2:225; 19:2806–2807). Those notes as to Lopez nizes his own production sheets, asserts that he has never seen those for Burris and Otero. (7:971.) To the extent it matters (Basgall’s notes record the production figures for Burris and Otero even if the reports themselves were not at- tached), I find that only the warning notes (G.C. Exh. 30), were handed to Valenzuela. As for Novak’s statement, Valenzuela recalls some of the parts and concedes that Novak could have read the others. (7:968–970.) Accordingly, I find that Novak read the statement, and recited the different production figures (which I summarized earlier) for Valenzuela, Burris, and Otero which are reflected on the pro- duction sheets. Basgall testified that if Valenzuela had produced about 700 of these parts he, Basgall, could have lived with that number. However, Valenzuela never increased his production and it remained in the area of 300 to 350 for a full shift. (20:3049.) Respecting possible disparity of treatment, the General Counsel introduced (but does not argue) two production sheets (G.C. Exhs. 14–1 and 14–2) of Angel Otero and one (G.C. Exh. 14–2) of Paul Magee on unrelated parts. Com- ments written on the reports, apparently by leadmen, are to the effect that Otero had low production and Magee had ex- cessive scrap. Although none bears a date, Balma acknowl- edges that he frequently had to caution Otero about that par- ticular job, and apparently others, yet Otero was never given a written warning. (2:183, 188–189.) Balma initially states that he orally warned Magee about the excessive scrap, which Balma considers a production problem. (2:183.) How- ever, Balma then expresses uncertainty whether the two times he cautioned Magee (which occurred during the period of May to July 1988) pertained to General Counsel’s Exhibit 14-2. In any event, no written warning issued to Magee. (2:187–188.) As Respondent observes (Br. at 166 fn. 139), both Magee and Otero were elected stewards, with Magee75 being the second-shift steward and Otero his assistant. (R. Exh. 1.) Does the failure to issue Magee and Otero written warnings possibly show leniency toward union stewards? If anything, Acme’s past practice was one of leniency. Balma had cau- tioned or orally admonished Otero many times, and Magee twice, but he had done the same—no written warning—for Valenzuela on many occasions in admonishing him to cease wandering around and to return to work. As earlier de- scribed, and as recorded by file memo (G.C. Exh. 7–25; R. Exh. 79), on September 28, 1987, Balma and Larry Stoner, the SMD supervisor, orally warned Valenzuela to quit wast- ing time and to work more. So far as the record shows, before the election Acme had issued three written warnings respecting low production. The first (R. Exh. 130–1) was issued January 12, 1983, to Emilio Mora by Supervisor Tom Malleck; the second (G.C. Exh. 8– 20; R. Exh. 74) on November 18, 1985, to Marcial Canales by Supervisor Ronald Adamczyk; and the third (G.C. Exh. 5zz) on June 30, 1986, to Tyrone Newson by Supervisor Howard McArtor. Of these, the one to Mora gives the clear- est example of a warning for unsatisfactory production of parts based on a poor attitude. The November 18, 1985 warning to Canales, as I de- scribed earlier,76 actually is for a ‘‘failure to obey instruc- tions’’ (concerning the proper method) which failure Adamczyk testified (20:2859–2863, 2900), caused the poor production. There is no block for low production on Acme’s warning form. Malleck checked item 13, ‘‘Other,’’ writing ‘‘Unsatisfactory Production Quantities (Parts)’’ to the side. Adamczyk checked the item 7 box, ‘‘Failure to obey instruc- tions.’’ His remarks in the text, however, expressly list poor production as the resulting problem. McArtor checked both boxes, 7 and 13, on his June 30, 1986 warning to Tyrone Newson, remarking that Newson produces only half, or even less, that of the other workers on all his jobs. So far as the record discloses, the June 16 written warning to Valenzuela was his first written warning for a production problem. (Recall that the one, R. Exh. 12, Balma and Stoner gave him on April 6, 1987, was for reporting to work the day before under the influence of alcohol.) What of Basgall’s own past practice? This appears to be the first written warn- ing for Basgall to issue. That is of little significance, for, as I noted earlier in the section about Acme’s management, Basgall had been an assistant supervisor only since April 1987 (20:2990)—slightly over a year before the June 16 warning. In any event, it was Balma who called Basgall’s at- tention to Valenzuela’s absence from his machine and his ap- parent low production. (2:377, 379; 20:3062; 24:3760.) Most of the postelection warnings for low production are put at issue by the complaint. One of these should be men- tioned here. It is the written warning (R. Exh. 73) issued June 9, 1988, to Juan Lopez for low production. Although the Lopez warning was attacked by complaint paragraph 9(b), the General Counsel withdrew the allegation when the Government rested its case-in-chief. (14:2041.) Notwithstand- ing withdrawal of the allegation, Acme Introduced its evi- dence on the matter, over the General Counsel’s objection, because of the asserted relevance to Acme’s affirmative de- fense of sabotage by an intentional slowdown in order to force recognition and a pay increase. (19:2773, 2782–2783.) The Lopez matter is relevant to that defense—not for the low production warning (at least not here)—but for the discharge which occurred at the warning interview. What happened is this. In the days leading up to the June 9 event, Second-Shift Supervisor Howard McArtor observed that Juan Lopez, normally a very good and reliable worker (19:2777, 2784, 2790, 2796), in May suddenly began produc- ing much less of a certain part he recently had started.77 (2:219; 19:2776, 2791.) His first evening, May 17, on that part Lopez produced 369 pieces over a full shift. (R. Exh. 73–9.) On May 20 it dropped to 280. (R. Exh. 73–7.) At the end of the shift,78 McArtor told Lopez he had not produced 1133 LOVEJOY INDUSTRIES cover May 20 and 25, and June 7 and 9, 1988. (R. Exh. 2; 2:224– 225.) 79 That would suggest an expected rate of 420 parts. McArtor changes his numbers. On the June 9 warning form he asserts that ‘‘the normal number of pieces machined on machine 261 is 450 per 9 hrs of work.’’ (R. Exh. 73.) At the hearing McArtor stated that the minimum expected is 500, that if a person really works at it he can produce 600 of this part in 9 hours, and that when production drops to 475 a warning (oral or written is not specified) is justified. (19:2804.) Although he has had other employees drop below ex- pected production rates, McArtor concedes he had never previously issued a written warning for low production. 80 McArtor noted for May 25 that Lopez was warned for not wear- ing safety glasses, that Lopez has poor work habits, and had low production for the night. (R. Exh. 2.) No production report for May 25 is in evidence, and it is possible that Lopez worked on some other part that evening. 81 The date is uncertain, with both dates appearing in the same records. (R. Exhs. 2, 73.) Apparently, however, June 8 was the last night Lopez worked, and he was discharged at the beginning of the June 9 shift. 82 I agree with McArtor (19:2810) that there is no difference in the statement by Lopez and one saying he will not make the expected production unless Acme pays him $7.50 an hour. enough, achieving in 9 hours what should be done in 6 hours.79 (R. Exh. 2; 19:2806–2807.) Lopez’ next full shift on this part apparently was June 3. As shown by his production sheet (R. Exh. 73–6), Lopez increased his production to 360.80 According to his production report for June 7 (R. Exh. 73–5), Lopez produced 265 pieces. (The final digit appears to be a 5, a 9, or some other number.) McArtor checked the records of the first shift and learned that Robert Burris and Rafael Duarte were producing more of these parts. (19:2780.) Burris also was new to the department (4 months) and brand new on the machine. (2:215, 227, 229.) The April 26 produc- tion record of Robert Burris (R. Exh. 73–10) reflects that he produced, over a full shift, 577 castings of this part. (19:2780.) McArtor went to Lopez and, without giving a spe- cific expected number, told him that he was producing 60 percent of the number done on the day shift and that he ex- pected more parts from Lopez. Without saying anything, Lopez just glared at McArtor and then turned away. (2:217– 218, 222–223; 19:2803; R. Exh. 2.) The next evening June 8, Lopez produced 275 pieces.81 (R. Exh. 73–1, –4.) Although reluctant to assert that he thought Lopez was in- tentionally producing less, McArtor nevertheless concluded that he should issue a written warning to Lopez because he knew Lopez could produce more and others were producing more. (2:210, 220–221; 19:2773, 2782, 2790.) McArtor prepared the warning form for Lopez. (19:2776, 2778.) Balma and McArtor conferred. Although the intention was to give Lopez the prepared written warning for low pro- duction (2:227; 19:2776), they also were going to ask Lopez why he was having a problem. (19:2776; 24:3752.) It appears they thought Balma might be more successful in talking with Lopez than McArtor had been. In any event, at the beginning of the shift on Thursday, June 9, Balma and McArtor met with Lopez, Paul Magee, the Union’s steward for the second shift, and Angel Otero, Magee’s assistant steward. (2:227–228; 19:2774–2775; 24:3753.) Whether McArtor or Balma is unclear, but one gave Lopez only the warning sheet, without the production reports attached for the hearing. (19:2779, 2781–2782.) Balma asked Lopez why his production was so low. Becom- ing angry and upset, Lopez, earning about $5 an hour at the time (19:2774), told Balma in clear English: ‘‘Well, you pay me $7.50 an hour and I will get your pieces.’’ Immediately Balma said ‘‘We don’t need you if that is your attitude.’’ With Magee pleading for Balma to give Lopez another chance, that Lopez simply was angry and hot–headed, Balma escorted Lopez to the timeclock, punched his timecard out, telling Magee that Lopez would never change with that atti- tude. (19:2774, 2777, 2783, 2802, 2809–2810, 2812–2813; 24:3753–3754.) I credit the uncontradicted testimony of McArtor and Balma. Acme’s records (R. Exh. 73–2) reflect that it classified the discharge ground as insubordination. McArtor (19:2796, 2803) concedes that no one ever told him that Lopez had been heard telling others to slow their production. McArtor (19:2809) and Balma (25:3855) ac- knowledge that neither Lopez nor anyone else ever told them that Lopez had expressed an intention of slowing his produc- tion in order to force Acme to give him a pay raise.82 McArtor testified that after the discharge of Lopez he did not observe low production by any of his other employees (19:2803). I should mention that the Robert Burris, clock 172, used for the Lopez comparison (R. Exh. 73–10) is the same em- ployee used for the comparison (R. Exh. 76–9) in Valenzuela’s case. Moreover, Burris presumably is the same night–shift employee whose productivity figure Larry Stoner used (along with that of Eusebio Hernandez) in deciding to issue the November 11, 1987 warnings to Antonio Ramirez and Fidencio Olivares. As I summarized earlier, in effect Stoner rates Burris (and Hernandez) at the top of the produc- tion curve. (21:3143–3144.) c. Discussion (1) Section 8(a)(3) Contending the Government established a prima facie case, the General Counsel lists several factors. First, the back- ground of the prior case with Valenzuela’s unlawful dis- charge. That is entitled to some weight, but only slight weight of itself. Second, Acme’s hostility toward unionization continues notwithstanding there is no recent evidence of antiunion ani- mus directed toward Valenzuela. Here the General Counsel relies on two items. One, Robert Novak’s acknowledgement, on cross-examination, that he prefers not to deal with the Union, and his supervisors have told him they feel the same as he does. As Novak explains, because the UE is a militant union, he is afraid production and discipline would be dis- rupted by arguments if employees and the UE did not want to do something. (24:3661–3664.) The General Counsel cites no authority in support of the Government’s contention. The Board has held that statements of opposition to union rep- resentation, even though not independently violative of the Act, can serve as the basis for finding animus. Holo-Krome Co., 293 NLRB 594, 595 fn. 6 (1989), (Member Cracraft ex- pressly declining to join in the finding), vacated and re- manded on this point, 907 F.2d 1343 (2d Cir. 1990). I do not read the panel majority’s expression in Holo- Krome as broadly as did the Second Circuit. The Board’s ‘‘basis for’’ phrase seems to imply that there must be a con- 1134 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 83 One of these, before and after the election, is Gilberto Martinez. (4:672.) Stoner names Martinez as his setup person. (21:3096.) text indicating animus. On one hand, Novak’s bare pref- erence (not expressed to employees but elicited on cross–ex- amination), on its own and not tied to any expression of will- ingness to resort to illegal methods to maintain Acme’s non- union status, falls short of constituting animus or hostility. That is, Novak’s expressed preference, by itself, does not serve to establish the animus element of the General Coun- sel’s prima facie case. On the other hand, Novak’s expressed preference could be useful as a type of ‘‘background’’ which may be helpful in deciphering the motivations for manage- ment’s conduct. It is in the latter sense that I consider Novak’s expressed preference. The other item in this second point of the Government is that here there are independent violations of Section 8(a)(1) of the Act. Most of the findings of violations I have made seem more technical than motivational. Nevertheless, perhaps some slight weight should be added to the balance on this point, particularly as Valenzuela was involved respecting some of the incidents. Third, the General Counsel relies on the timing of (1) threats uttered by Balma and Novak in response to concerted requests for recognition and a general wage increase and (2) the T-shirt incident involving Valenzuela. (Br. at 116–117.) The timing factor is weak. If there is a surge of union activ- ity (here the activity occurred before and after the April 13, 1988 certification), followed by management’s seizing on previously tolerated conduct as a pretext to discipline an em- ployee for protected conduct, then timing is significant. But if the discipline, here on June 16, follows directly after an event not previously tolerated, then timing is merely coinci- dental. The timing point here seems to beg the question of whether Valenzuela’s conduct had been tolerated in the past. That leads to the fourth point, that Acme departed from its past practice. The General Counsel asserts that Valenzuela had never been given a written warning for low production since being hired in August 1971. That fact means little because there is no showing that his production in earlier years was bad yet he received no warning. More- over, Acme in essence relies on the same fact to show that the difference here is that, after learning that Acme had de- clined to recognize the Union following its certification by the Board, Valenzuela decided to do his part to persuade management to change its mind by adopting the tactic of sabotage–by–slowdown. As for the General Counsel’s assertion that Respondent had no preelection practice of issuing written warnings for production problems not resulting from a failure to follow specific instructions, that contention, as I summarized earlier, is wrong. Granted, however, there were only two (to Emilio Mora in November 1985 and to Tyrone Newson in June 1986), plus the low production-related warning of November 18, 1985, to Marcial Canales for his failure to follow instruc- tions on the proper work method. I find that the past practice factor does not add any weight to the prima facie side of the balance. The General Counsel does not list disparity of treatment. Based on my summary of potential disparity evidence, I find no disparity. Does the evidence indicate that Acme seized on a pretext? There is some difficulty in deciding this point. On the one hand Balma seems less than forthright in testifying that he merely asked Robert Burris and Angel Otero to produce good parts, and that he was not looking for speed. In select- ing Burris, Balma picked one of the top producers in the de- partment. On the other hand, I find that Valenzuela grossly exagger- ates the amount of time he spent during late May and early June 1988 performing setups or related work (such as replac- ing broken drills for other workers). As for actual or formal setups, Valenzuela’s production sheets in evidence, for May 26, 27, 31 and June 6, 7, 8, 10 (R. Exhs. 17a–c; 18a–d), cover the last 3 workdays of May and 4 workdays for the first week of June. Those are the days which Acme evalu- ated. Of those days, Valenzuela recorded only one (formal) setup—a single hour the morning of June 8. (R. Exh. 18b.) Basgall identifies the entry as being for setup. (20:3020, 3059.) Valenzuela momentarily testified that at (around) the time of his warning he did setup on one machine, but then modified that to say he did setups on all machines by helping others. (3:570–571.) Basgall testified that Valenzuela did setups once or twice a week only when he assigned Valenzuela to do such when Basgall could not get to it, and that Valenzuela did not take it upon himself to do setups. (20:2995, 3063–3064.) Basgall apparently is describing the work of setting up a machine for a production job in which another worker would operate the machine. For jobs assigned to him that might involve setting up the appropriate machine, Valenzuela handled that as a matter of course. (20:2995.) Although the record does not disclose the number of employees Basgall supervised in this May–June 1988 timeframe, Basgall testified that the number he supervised (during his supervisory career), and which Valenzuela assisted, ranged from 4 to 10. (20:2994, 3060.) Valenzuela asserts that ‘‘many’’ employees worked in Basgall’s east side area between August 1987 and October 27, 1988, but he names only five and one of them, Juan Car- los Garcia, he also describes as doing some setups. (4:672– 674.) Valenzuela names others who do setups,83 but he is not specific on who did setups in Basgall’s area. (4:671–672.) Basgall also names Juan Garcia as doing setups, apparently in Basgall’s area. (20:2995.) Valenzuela expressly concedes, however, that Basgall does many of the setups. (4:673.) It is the setup-related work of assisting others in getting drills, taps, and such that the General Counsel appears to suggest (Br. 112–113, 117) is what Valenzuela also was doing on the days in question. The problem with that is two- fold. First, Valenzuela never specifies what setup-related work he did on which of these days or whom he helped. His testimony is generalized and, in effect, simply describes whatever he did in that respect over a period of months or years. For example, at one point Valenzuela asserts that get- ting fixtures required that he leave Basgall’s area and go to the west (Stoner’s) side of the department and building. And, Valenzuela adds, fixtures are located on two floors there. The implication, of course, is that such work took Valenzuela well out of sight of Basgall, and Balma, and probably for several minutes at a time. (7:1072–1075.) Basgall, however, explains that such work did not require Valenzuela to leave Basgall’s area because fixtures for his employees are kept on shelves in the middle of Basgall’s area, and drills and taps are stored under lock, for which Valenzuela has a key, in the 1135 LOVEJOY INDUSTRIES supervisor’s office underneath the cafeteria. (20:3063, 3068– 3069.) The office is situated between Stoner’s area and Basgall’s area. (R. Exh. 13.) Second, Valenzuela asserts that anytime management spoke to him about low production, he reminded them that he did setup work as well as production. He immediately concedes, however, that he did not do so in the June 7 meet- ing with Balma and Basgall, nor even on June 16, because there was no necessity to mention it since he told them he would try to raise his production. (7:1030–1031.) Finally, Valenzuela lacks credibility. He testified that he did not remember the oral warning which Stoner and Balma delivered on September 28, 1987. (4:694; 7:1050.) I find that it was given. Respecting the June 7 meeting with Balma and Basgall, Valenzuela at first denies that Balma, when asked about a raise, responded that he could not give one until the production problems are over with. When shown his own calendar-notes (R. Exh. 14), Valenzuela conceded that Balma did say it. (6:908–911.) This is not to say that the versions of Balma and Basgall have no problems. Indeed, Balma and Basgall offer almost as much generalized testimony as does Valenzuela. I particu- larly note that Balma has a tendency to testify initially in ab- solute terms, only to modify and soften his position as his testimony progresses. For example, when describing work at- titudes on his April 1987 arrival, Balma testified none of the employees or supervisors, excepting Scott and perhaps McArtor, cared. (24:3710–3711, 3716–3717.) Later he modi- fied the ‘‘none’’ to a majority. Finally, he asserted that a ma- jority wanted to do their jobs, but that they violated the rules. (25:3874.) First testifying that in April 1987 he found Valenzuela totally drunk in the restroom (2:3770), Balma corrected that to say he never observed Valenzuela intoxi- cated. (25:3875.) Another example is Balma’s testimony that during the May–June 1988 timeframe he never saw Valenzuela at his machine. (24:3762.) The principal shortcoming in the versions of Balma and Basgall is that neither asserts he observed Valenzuela all day for even 1 of the days in question. Thus, neither is able to say, from personal observation, that Valenzuela in fact did no setup-related work. Both rely on intermittent observations, production records, generalized testimony, and only some specific testimony (such as Basgall’s detailing the location of fixtures and drills). Despite Basgall’s testimony that Valenzuela did only setups that Basgall assigned, Valenzuela testified that Stoner frequently assigned him setups between June 1987 and November 1988. (7:1052.) Indeed, Valenzuela asserts that it was Stoner who assigned him the production work which is the subject of the June 16 warning and, in so doing, did not give an expected number to produce. (7:1093.) Stoner does not deny either assertion directly, testifying simply that at the time covered by the June 16 warning Basgall was ‘‘pretty much’’ in control of the area with Valenzuela working for Basgall. (21:3118.) In the weeks thereafter to October 24, 1988, Stoner testified, Stoner worked through Basgall and only about 10 percent of the time would he suggest to Basgall a specific job for Valenzuela. (21:3119.) I credit Valenzuela on this, although I find that by May-June Stoner no longer ‘‘frequently’’ as- signed Valenzuela work. Instead, I find, such assignments were only occasional or infrequent. Is there a prima facie case? Probably so, particularly if it is deemed that some animus lingers from the earlier case. Assuming, for the moment, that the Government established a prima facie case, did Acme demonstrate that it would have issued the June 16 warning to Valenzuela even absent his union activities? I find that Acme did so. First, at the June 7 meeting with Balma and Basgall, Valenzuela, by stating that he should try for greater produc- tion and that Balma should try for the general pay raise, wafted the whiff of a possible slowdown as the reason for his production level. It was not lost on Acme despite Valenzuela’s declination to repeat his statement and Canales; ‘‘He didn’t say nothing.’’ Thus, 2 days later Balma fired Juan Lopez for linking greater productivity to his (Lopez’) receiving a substantial pay increase, and 4 days after the Lopez discharge Novak called Attorney Salzman and ob- tained the no-slowdown-will-be-tolerated statement which he personally read to Valenzuela at the June 16 warning meet- ing. Second, Basgall testified that he could have lived with an output of 700 by Valenzuela, suggesting that no warning would have issued had Valenzuela increased his production to that level. (20:3049.) But Valenzuela never increased his production at all. I find that Basgall, Balma, and Novak reached the good-faith conclusion that Valenzuela was hold- ing down his production intentionally. The June 16 written warning was based expressly on engaging in a slowdown. Although that usually implies a drop from previous levels, here Acme uses the term to describe an intentional effort by Valenzuela to produce less than what he could have done had he made a reasonable effort. The warning did not issue because Valenzuela failed to match Balma’s three units a minute (under Basgall’s 8-hour formula, 1440 units a day), Basgall’s two parts a minute (960 a day under Basgall’s formula) or even Otero’s 1-day total of 714. We shall never know whether Acme would have issued the warning had Valenzuela increased his production to, say, 550 parts. That circumstance would tend to detract from any concept of a slowdown by Valenzuela, and it would tend to persuade that a still-issued warning as being motivated for reasons other than those stated. Here, however, Valenzuela made no increase at all. That fact tends to sup- port a finding, which I make, that Acme acted in the good- faith belief that Valenzuela was intentionally producing much less than he could have been producing. To the extent that Valenzuela did any setup-related work during late May and, particularly, the first 10 days of June, I credit Basgall that he made allowance for any nonproduc- tion work Valenzuela may have done. (20:3020.) As de- scribed earlier, Valenzuela concedes (7:031) that at the June 7 Balma/Basgall meeting and the June 16 warning meeting he did not raise the matter of setups or setup-related work because, he asserts, it was unnecessary to mention it. I find that the reason he did not raise it is that he was doing prac- tically none, the exception being a 1-hour setup he did on June 8. To repeat, there is no allegation that Acme unlaw- fully reduced Valenzuela’s setup duties. In short, I find that even if there had been no union on the scene, Acme would have issued the June 16, 1988 writ- ten warning to Nicolas Valenzuela. Accordingly, I shall dis- miss complaint paragraph 9(c) which alleges a violation of Section 8(a)(3) of the Act. 1136 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 84 Balma testified that Acme discharged Hauser in about late Au- gust 1989. (24:3704.) Hauser did not testify. 85 Although no witness dates Hauser’s arrival, I note that a Feb- ruary 1986 disciplinary record reflects that Jeffrey D. Minghi was discharged, after several warnings, for hot tempered confrontations with maintenance supervisor Hauser. (G.C. Exh. 6–9.) 86 I did not receive Scott’s description of Hauser’s report for the truth of the assertions, but only for the basis on which Acme im- posed its disciplinary action on Olivares. (22:3327.) 87 This is a composite of the testimony of Scott and Olivares. Al- though Olivares first implies that he denied the accusation (11:1504), and eventually asserts that he told Scott he had said nothing to Hauser (11:1529), the better evidence is that he took the offensive. The difference is immaterial, however. 88 Balma concedes that he had not previously suspended anyone for making abusive remarks. (2:193.) There is no evidence of a simi- lar incident, particularly involving a supervisor, having been brought (2) Section 8(a)(5) As I have described, as early as 1983 Acme issued a writ- ten warning (to Emilio Mora) for low production. Only two others issued thereafter before the October 1987 election, and one or both sprang from a failure to obey instructions, result- ing in low production. There are no past practice (preelection) warnings for slowdowns. Neither is there evi- dence that Acme ignored any preelection slowdowns. Acme’s June 16, 1988 warning to Nicolas Valenzuela is consistent with its past practice. Accordingly, I shall dismiss this 8(a)(5) allegation, complaint paragraph 14’s inclusion of paragraph 11(f)(8). 2. Sacramento Olivares suspended August 12, 1988, for insubordination a. Pleadings Complaint paragraph 9(d) alleges that on or about August 12, 1988, Acme suspended Sacramento Olivares for 2 days. Acme admits the suspension fact but denies the conclussory allegations of paragraphs 9(e) and 13 that it violated Section 8(a)(3) and (1) of the Act by suspending Olivares. I shall dismiss this allegation. b. Facts Recall that I earlier dismissed the allegation attacking the March 10 written warning which Acme’s Jim Scott issued to Sacramento Olivares for producing over 300 damaged parts. Scott and San Roman testified that they believed Olivares in- tentionally damaged the parts. The August 12 (Friday) suspension attacked here responds to an incident that day involving Olivares and Gus Hauser, the then maintenance supervisor.84 Olivares worked as a punch press operator in Jim Scott’s ADC. (11:1479.) Olivares suggests that it was only a few months before Au- gust 1988 that Hauser either arrived at Acme or otherwise became the maintenance supervisor.85 Before Hauser took over, Olivares testified, Olivares and others would use a port- able fan to cool themselves when the weather was hot. Who- ever first got the fan would use it that day. When Hauser came, however, he began locking up the maintenance equip- ment, although he had not locked this portable fan. (11:1522–1524.) Before August 12 Olivares and Hauser had spoken only once or twice, concerning the repair of a ma- chine. (11:1520–1521.) Because August 12 was hot, Olivares went to the maintenance area to get the fan. This time Olivares found that the fan was chained and locked. (11:1501, 1522–1525.) Knowing that Hauser would have the key, Olivares went to Hauser, who was at his desk. It was in the late morning, before the 11:30 a.m. lunchbreak. Olivares asked, in English, if he could use the fan since no one else was using it and the weather was hot. (11:1502, 1521, 1526.) ‘‘No,’’ Hauser replied, because the fan be- longed to the maintenance department and no one else could use it. (11:1502, 1526.) Admitting that he became angry at this rejection, that he was standing very close to Hauser, and that, staring at Hauser, he raised his voice and loudly stated: What do you mean, nobody can use it? I just want it for a while and whenever you need it you can have it back, you can use it. ‘‘No,’’ Hauser replied. ‘‘Then,’’ Olivares concedes, ‘‘I gave him dirty looks,’’ whereupon Hauser told him to get out. (11:1502–1503, 1526–1527.) Scott testified that Hauser reported the event to him and asked that he issue a written warning to Olivares.86 Unlike Olivares, Hauser’s report included the assertion that Olivares had told him, ‘‘Fuck you, asshole.’’ Scott prepared a written warning, had Hauser check it for accuracy, then took it to Balma and explained the circumstances. (2:315–316; 22:3327–3328.) Leaving the drafted warning with Balma, Scott went to Sacramento Olivares. (22:3328, 3330.) Olivares places the time as shortly before 2 p.m. (11:1503.) Scott told Olivares that Hauser had told him what Olivares had said to Hauser, that Olivares should not have said it because Hauser is a su- pervisor, and that he had a warning for Olivares. He asked if Olivares had used that language. Olivares asked what he supposedly had said to Hauser. Pulling out a pocket note- book, Scott showed Olivares where Scott had written the ex- pletive, ‘‘asshole.’’ ‘‘Why do you believe him? Where are your witnesses?’’ asked Olivares. Scott said he believed Hauser. Olivares repeated that Scott had no witnesses, and asked why Scott believed Hauser. Replying that he had no reason to believe that Hauser fabricated the story and that Sacramento should not say that to anyone, Scott departed without issuing any written warning to Olivares.87 (2:316; 11:1503–1505, 1529; 22:3329–3331.) At the hearing Olivares denied calling Hauser an ‘‘asshole.’’ (11:1503.) In his pretrial affidavit of August 23, Olivares records that the slur name written in Scott’s pocket notebook was ‘‘sonofabitch.’’ That is the wrong name, Olivares asserts, because it was the product of his anger over the fact the fan was still locked up and no one was using it 11 days after the event. His anger interfered with his con- centration. (11:1530–1531.) (Thus, Olivares reported the slur term to the Board agent, who correctly recorded it, but Olivares erred by reporting the wrong slur for that which Scott had showed him.) In the meantime Hauser, Balma testified, reported the epi- sode to him. Balma considered the remarks of Olivares to be abusive and grounds for an automatic suspension. (2:193; 25:3896, 3897.)88 Later that afternoon, Scott testified, Balma 1137 LOVEJOY INDUSTRIES to Balma’s attention for which Balma did not impose discipline. In- deed, Balma fired Raymundo Aguirre for threatening Jim Scott. called him and said they would suspend Olivares for 2 days and for Scott to escort Olivares to the office toward the end of the shift. (22:3328.) Because Balma had observed Olivares walking around frequently, he and Scott checked the produc- tion records of Olivares and saw that his production was low. (23:3475–3478; 25:3807, 3810.) When Scott came to escort Olivares toward the end of the shift, Olivares asked that Canales accompany him. Scott said he could. With Balma, Scott, Olivares, and Canales in the of- fice, Balma testified that he questioned Olivares about the in- cident but that Olivares neither admitted nor denied Hauser’s accusation. (2:193; 25:3807.) Olivares testified that Canales (who does not address this meeting in his own testimony) asked Balma why he believed Hauser rather then them. Be- cause, Balma replied, Hauser was the supervisor. Without de- tailing any words, Balma said Olivares should not talk that way to supervisors. (11:1534.) The testimony of the witnesses is rather conclusory con- cerning the words spoken at the meeting about the Hauser incident. Nevertheless, there is no dispute that Balma sus- pended Olivares for 2 days and also reviewed his production, telling Olivares that his production was low. Balma testified that Olivares was suspended because of the Hauser matter and not for low production, a deficiency for which Olivares was not even given a written warning. (25:3806, 3811, 3897.) The disciplinary report (G.C. Exh. 25) which Scott prepared (2:318; 22:3326) checks boxes 4 and 10 for ‘‘In- subordination’’ and ‘‘Improper conduct’’ for using ‘‘very abusive language to Gus.’’ Scott and Hauser signed it. Scott wrote nothing about low production. (23:3479.) Balma added his own notes about reviewing the complaint, suspending Olivares for 2 days, discussing his production, and warning Olivares that the ‘‘next incident he will be terminated.’’ Balma denies that union support by Olivares had anything to do with his decision. (25:3832.) Scott testified that the disciplinary form (or a copy of it) was given to Olivares at the meeting. (22:3329.) Not only does Olivares deny that, and deny ever seeing the form be- fore the hearing, he asserts that he requested a copy at the meeting but Balma said no because after the 2-day suspen- sion he would forget everything and they would start new. Although the form has ‘‘Employee would not sign’’ handprinted by the space for the employee’s signature, Olivares asserts he was not asked to sign it. He concedes that he would have refused, had he been asked, because the accu- sation is untrue. (11:1508, 1532–1533.) As the warning or disciplinary form contains nothing in conflict with the undis- puted core facts, I need not resolve whether Olivares was asked to sign, given a copy, or both. Olivares testified that he served his 2-day suspension on August 15–16. (11:1508.) That would be the following Monday–Tuesday. There is no evidence that Olivares had received any discipline before the October 1987 election. Scott is not aware of any having been imposed after his, Scott’s, arrival in 1985. (2:317.) Acme complains (Br. at 435 fn. 231) that I erroneously re- jected a proffered exhibit (R. Exh. 106) and associated testi- mony impeaching, it is argued, Olivares. Scott identified a note, dated (Monday) August 15, from Hauser which Hauser delivered to Scott. Hauser’s note is written on the face side of an Acme parts tag. (R. Exh. 106.) After the date, Hauser’s note reads: ‘‘Somebody cut the cord from the fan & left this note on top of it.’’ Scott testified that Hauser told him the same. (22:3332.) On the reverse or blank side of the tag ap- pears the following handprinted words: YU, NEST, GASS, and PUTO. Scott testified that ‘‘Puto’’ means ‘‘stupid,’’ although the evidence never reached the point of an interpretation being received by the official interpreter. (22:3332.) Acme suspects that either Olivares or a union supporter left the note, but conceded, when Scott was on the stand, that it had no proof. Respondent construes the note as a threat to Hauser, appar- ently on the basis that the anonymous person threatened, ‘‘You next, stupid Gus,’’ with the cut cord signifying that Hauser would be stabbed. I sustained the General Counsel’s objection and granted Acme’s request to place the document in the rejected exhibits folder. (22:3333–3334.) Acme had not sought, when cross-examining Olivares ear- lier, to question him about the note. After Acme rested its case-in-chief, the General Counsel recalled Sacramento Olivares as one of the General Counsel’s rebuttal witnesses. Olivares testified only briefly, denying any intent to slow his production or to produce bad parts. After a question or two on cross-examination, about any tactics suggested by the Union, Acme approached the subject of the rejected exhibit. After Olivares stated that he cannot write English very well, and has trouble spelling English, the General Counsel ob- jected as beyond the scope of direct examination when Olivares was asked to write the words, ‘‘You next, Gus.’’ (26:3957.) Describing the relevance of the requested writing specimen as showing, if it appeared Olivares had written the note, that Olivares had indeed uttered abusive words to Hauser on Au- gust 12, Acme admitted that the inquiry had nothing to do with the direct examination on rebuttal. Acme explains that it did not call Olivares for this purpose during Respondent’s case-in-chief because it did not think it was necessary to do so. I sustained the General Counsel’s objection. (26:3958– 3961.) Neither at the hearing nor in its posthearing brief does Acme articulate any theory of procedure or evidence in sup- port of its complaint that I ‘‘improperly precluded this testi- mony and line of questioning which would have shown that Olivares was responsible for this threat.’’ I reaffirm my rul- ings on this matter. c. Discussion Arguing that the Government established a prima facie case, the General Counsel concedes there is no evidence of animus directed toward Olivares. Notwithstanding that ab- sence, the General Counsel contends that animus is shown by Acme’s ‘‘persistent efforts to avoid its bargaining obligation, as well as the coercive statements about wage increases, made by Balma and Novak,’’ and Balma and Novak ‘‘made it clear that the employees would not receive their expected general wage increases because of the Union.’’ Also relying on timing, the General Counsel, contending that several warnings issued after the October 1987 election for conduct previously tolerated, argues that ‘‘Olivares’ sus- pension took place in the midst of a surge of disciplinary ac- tivity which began soon after the union election. These cir- cumstances raise a strong inference of discriminatory mo- tive.’’ (Br. at 32.) 1138 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD The General Counsel also asserts that an adverse inference should be drawn from Acme’s failure to call Hauser as a wit- ness. (Br. at 29 fn. 10.) As former Maintenance Supervisor Gus Hauser ‘‘was terminated by the Respondent prior to the hearing, however, it cannot reasonably be assumed that he is favorably disposed toward it. In these circumstances, it is well settled that an adverse inference will not be drawn.’’ Cine Enterprises, 301 NLRB 446 fn. 4 (1991). Although I am unimpressed by the General Counsel’s ar- gument, there is one factor which possibly suggests a bare prima facie case. That is the fact that both Scott and Balma projected the image of having accepted Hauser’s version without first asking Olivares his version. That is, Olivares seems to have been presented with the burden of disproving Hauser’s reported accusation. As Balma apparently said at the suspension meeting, Hauser was a supervisor. Scott’s ear- lier statements to Olivares reflect a similar presumption fa- voring a supervisor in any credibility dispute with a produc- tion worker. I am not persuaded that such a presumption by management can be used to tip the scales in favor of a prima facie case. But assuming a prima facie case was established, did Acme carry its burden of showing that it would have taken the same action even absent any union activities by Sac- ramento Olivares? I find the answer to be yes. Even under the version of Sacramento Olivares, insubordination is estab- lished. Standing ‘‘very close’’ to Supervisor Hauser, speak- ing ‘‘What do you mean.’’ in a loud voice, and staring into his face while giving Hauser ‘‘dirty looks’’ qualify as insub- ordination without regard to whether Olivares added, ‘‘Fuck you, asshole’’ or simply ‘‘Asshole.’’ There is no affirmative evidence showing that Olivares ac- tually uttered the obscenity to Hauser. Scott says Hauser so reported to him. I credit Scott. I also disbelieve Olivares’ de- nial that he uttered the obscenity. Without questioning Hauser, Balma testified that Hauser reported the remarks to him. I credit Balma. I also credit Balma when he testified that union considerations played no part in his decision to suspend Olivares for this obscenity. Balma clearly favors a supervisor’s word over that of a nonsupervisory employee, and a reported profanity or insub- ordination by the latter toward a supervisor merits, under Balma’s law of the shop, a suspension. (25:3896–3897.) Balma’s rule, hardly uncommon in industry, is a rational one. Balma’s decision here was consistent with his February 25, 1988 discharge of Raymundo Aguirre for the more serious matter of threatening Supervisor Scott (with violence). Re- call, too, that on June 9 Balma discharged Juan Lopez for an insubordinate attitude. There is no evidence Balma, before the election, tolerated similar conduct. As for Acme’s pre-Balma practice, disciplinary records in evidence reveal that in February 1986 Jeffrey Minghi was given a 2-day suspension for ‘‘insubordination,’’ ‘‘fighting on company premises,’’ and ‘‘improper conduct.’’ (R. Exh. 130–12.) The remarks section states: ‘‘Employee was very disrespectful of supervision. After using foul language, em- ployee was asked to stay home for two days. Disciplinary layoff for 2 days (2–19 and 2–20–86).’’ Minghi was fired when he returned, on February 21, for a history of causing problems, for which he was warned, ‘‘in hot-tempered con- frontations with our Maintenance supervisor, Gus Hauser.’’ (G.C. Exh. 6–9.) In October 1986 Acme fired Malcolm Price for ‘‘insubor- dination’’ and ‘‘improper conduct.’’ (G.C. Exhs. 6–4, 6–5.) Price reacted angrily when his brother was laid off. As Acme reported to the Illinois unemployment insurance office, ‘‘He made an angry scene, swearing repeatedly at his supervisor and ripping up his supervisor’s timecard. We were forced to discharge Malcolm Price for insubordination on 10-6-86.’’ (G.C. Exh. 6–6.) Howard McArtor was the supervisor. McArtor adds that Price also expressed a possible death threat. (19:2769–2772.) Clearly Price did more than Olivares, but, then, Price was fired and Olivares was merely suspended for 2 days. Although the pre-Balma practice has certain differences from the Balma practice, it also has some similarities. In short, I find that Balma’s policy is consistent with the pre- existing practice at Acme. Finding, therefore, that Acme has carried its burden of demonstrating that on August 12 it would have suspended Sacramento Olivares even absent his union activities, I shall dismiss complaint paragraph 13 to the extent it alleges paragraph 9(d), the suspension of Sac- ramento Olivares, as a violation of Section 8(a)(3) and (1) of the Act. I. The Gus Hauser/Sacramento Olivares Incident at the September 1988 Picket Line 1. Pleadings As established by the pleadings, Acme admits that about September 14 and 15, 1988, certain union-represented em- ployees ceased work concertedly, struck, and established and maintained a picket line at Acme’s plant. The complaint does not allege that the strike was an unfair labor practice strike, and the General Counsel disavowed any attempt to litigate for a positive finding in that respect. (3:456.) Complaint paragraph 8 alleges that about September 15 Acme, ‘‘acting through Gus Hauser, at the driveway entrance to Respondent’s plant, attempted to and inflicted bodily harm upon an employee by causing his automobile to hit said em- ployee, because said employee participated in’’ the Septem- ber 14–15 strike. Paragraph 12 includes paragraph 8 among the listed violations of Section 8(a)(1) of the Act. In its an- swer, Respondent denies these allegations. I shall dismiss complaint paragraph 8. 2. Facts Respecting the incident at the picket line on September 15, four witnesses testified: Nicolas Valenzuela and Sacramento Olivares for the General Counsel; Jim Scott and Peter Balma for Acme. Robert Novak also testified concerning his obser- vations of the physical condition of Sacramento Olivares later that morning. Recall that Acme discharged Hauser about late August 1989 (24:3704)—nearly a year after the in- cident at issue. As the General Counsel sought respecting the August 1988 suspension of Sacramento Olivares, here also the General Counsel requests that I draw an adverse infer- ence from Acme’s failure to call Gus Hauser, the discharged maintenance supervisor, to testify. (Br. at 36 fn. 15.) As I noted earlier, that would be inappropriate. Cine Enterprises, 301 NLRB 446 fn. 4 (1991). Rather than having essentially two versions of the incident, from two witnesses on each side, we seem to have four con- flicting accounts. Some differences are normal in eyewitness 1139 LOVEJOY INDUSTRIES observations of an event, particularly a picket line event. The difference here, however, are substantial. Before reaching the discrepancies, however, I shall begin with the physical lay- out. Acme’s plant faces south across Commercial Avenue. There is a parking lot on the west side of the plant (south- west actually), and a parking lot on the east side. The inci- dent here occurred at the entrance to the east parking lot, at the southeast driveway. A sidewalk runs in front of the building, and there is a grass area between the sidewalk and the street, Commercial Avenue. There is no dispute that the picketers at the southeast en- trance had moved aside the first day, September 14, and per- mitted the supervisors to enter. According to Sacramento Olivares, however, on September 14 Hauser had driven in at a fast rate of speed. (11:1544, 1547.) Police came to the scene several times on September 14. The morning of Sep- tember 14, Olivares testified, the police came twice and talked to some of the pickets. Although Olivares did not hear what the police said, the strikers told him the police wanted the pickets to keep moving and not block the entrance. (11:1538–1544.) Our events occurred between 6 and 6:30 the morning of September 15. Approximately 8 to 10 strikers were at this (southeast) driveway. The group included Marcial Canales (who did not testify about this incident, although I draw no adverse inference from that fact), Sacramento Olivares, Nel- son Diaz, and Nicholas Valenzuela. Plant Manager Balma was standing at the driveway not far from the southeast cor- ner of the building. Jim Scott was there, also. (3:558; 5:759; 23:3392; 25:3811.) Scott testified that, seeing employee Pietro DiFranco mo- tion for him, Scott walked toward the street. After conversing with DiFranco a moment, at or in the driveway, Scott started back toward Balma. Stepping onto the grass on the building, or west, side of the driveway, Scott observed Hauser driving on Commercial toward the southeast entrance. (23:3392– 3394, 3490, 3493.) Balma testified that he was positioned so he could also look toward the west parking lot. (25:3885.) Balma observed Gus Hauser driving east from the west, on Commercial, at about 25 to 30 miles per hour. (25:3812, 3885–3886.) Valenzuela and Olivares were in the drive. Valenzuela testi- fied that he first saw Hauser’s vehicle as Hauser was about to enter the drive. (5:764, 767, 770.) Olivares testified that he did not see the car until it was about 2 feet from him. (11:1569–1572.) From this point the descriptions of the wit- nesses diverge. Valenzuela (5:761–764) and Olivares (11:1515–1517, 1573) assert that they were talking and walking together from west to east (away from the building) in the driveway. Valenzuela was 1 or 2 feet to the left of Olivares and about 8 to 10 feet from the point where the drive meets the street. (5:761, 763–764, 771.) Olivares was about 6 feet from the street. (11:1535–1536.) At the street, the driveway entrance is about 15 feet wide. (5:767.) Olivares testified that his attention was directed to his right by people shouting and tires squealing. (11:1516, 1604.) At that point Hauser’s car was about 2 feet from him and Olivares realized that Hauser was driving very fast. Olivares first places himself as being closer to the eastern edge of the drive, but then asserts he was in the middle of the drive at that moment. (11:1570–1572.) Hauser applied his brakes and left skid marks 5 feet long, Olivares testified. From the length of the skid marks, Olivares estimates that Hauser was traveling about 35 miles per hour. (11:1516–1517, 1564– 1565, 1572.) As Hauser skidded, Olivares jumped forward twisting to his right. Hauser’s right front, or passenger, bumper hit the front of Olivares’ right knee. Olivares landed upright and did not fall. (11:1517, 1565–1567.) Valenzuela’s description is a bit different. According to Valenzuela, at the shouts and squealing he turned and saw Hauser’s car at the point of entering the drive. In Valenzuela’s opinion, Hauser accelerated his vehicle to about 30 miles per hour, charging past them without braking until he stopped after going beyond Valenzuela and Olivares. As Hauser charged past them, his vehicle hit Olivares and brushed Valenzuela. Valenzuela jumped backwards (west- ward, toward the grass) as did Olivares. Nevertheless, the left front bumper, on the driver’s side, hit the right leg of Olivares. (3:553–557; 5:767–775, 779.) Olivares fell as he jumped back toward the grass. (5:776.) Balma and Scott also have differences. Scott testified that as he stepped up onto the grass, after talking with DiFranco, he observed Hauser turn into the drive and suddenly apply his brakes and stop. Strikers divided to make way as Hauser entered. Scott was within 10 feet of Hauser’s vehicle, stand- ing on the driver’s side, as Hauser turned into the drive. Scott does not know why Hauser stopped. (23:3394, 3490– 3491, 3494.) When Hauser stopped, strikers began beating on his car. Scott did not hear them say anything, although they could have. Hauser started forward, stopped, got out of his car, came back and a yelling match ensued after Hauser told Valenzuela he would break his face if he touched his car. (23:3394–3395, 3492–3494.) Balma testified Hauser was driving about 25 to 30 miles per hour on Commercial. Hauser turned left from Commer- cial but stopped at the entrance to the drive, and still in the street. Valenzuela, Olivares, Hugo Paz, Nelson Diaz, and two to four others were standing in the middle of the drive block- ing Hauser’s entrance. It took up to a minute for them to move, and Hauser waited during that time. As Hauser started forward, the strikers began beating on his car and hollering in Spanish at Hauser. After he got up on Acme property (Balma does not know how fast Hauser was driving in this stretch) Hauser stopped, got out, and confronted Valenzuela with ‘‘You bang on my car again, I am going to hit you in the face.’’ Others joined the altercation and Balma intervened to break up this display of animal exuberance. At no time did Balma hear any squealing of tires. (25:3812–3814, 3885– 3893.) Robert Novak testified that later that morning, between 8 and 9, he observed Olivares, across the street perhaps 100 feet away, carrying what Novak thinks was a car battery from one car to another. As Olivares was carrying the object in front of him, Novak could not identify the object. (24:3658–3659, 3691–3692.) Olivares concedes that he worked all day at his job on September 16, but that he had to stop and rest at times because the pain in his knee made him tired. (11:1563.) According to Olivares, he went to see a doctor on September 15. The doctor was not in that day, so it was not until Saturday, September 17, that Olivares saw the doctor. Dissatisfied with the doctor’s treatment, he went to a hospital on September 30. No medical evidence for any 1140 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 89 ‘‘Play fair.’’ R. Fulghum, All I Really Need to Know I Learned in Kindergarten 6 (1989, Villard Books). of this was offered, although a hospital bill (R. Exh. 35) was identified to show the September 30 date and billing for treatment that date. (11:1517, 1567–1568.) Olivares denies that as of September 15 he was still angry with Supervisor Hauser over the portable fan incident a month earlier. (11:1531.) 3. Discussion In determining the likely sequence of events, I first note that I find Sacramento Olivares to be an unreliable witness. His demeanor was unpersuasive and I do not believe him. Valenzuela is hardly more credible. Moreover, I have de- scribed the substantial differences between their versions— contradicting each other, for example, as to which corner of the car struck Olivares, the direction Olivares jumped, whether Olivares fell, and whether Hauser even braked. Scott’s version has an element, the sudden stop, which could be consistent with a version that Hauser barely hit someone (neither he nor Balma is able to say he was watch- ing Olivares and that Olivares was not hit). The place of the stop seems inconsistent with Balma’s assertion that Hauser stopped in the street. Of the witnesses, Balma is the most persuasive and gave the more specific testimony. Notwithstanding Balma’s dif- ference with Scott, and the rather lengthy 1-minute wait by Hauser in the street (there is no evidence whether Hauser was blocking traffic on Commercial), I credit Balma. In doing so, and in discrediting Olivares and Valenzuela, I find that Olivares describes a fake injury. Any contact between Olivares’ knee and Hauser’s car, I find, was intentionally made by Olivares for the express purpose of crying ‘‘Foul!’’ Sacramento Olivares faked the hit, I find, because he was still angry at Supervisor Hauser over the August 12 portable fan confrontation.89 I shall dismiss complaint paragraph 8. J. Discipline Imposed October–November 1988 1. Nicolas Valenzuela warned and suspended on October 24, 1988, for low production a. Pleadings Complaint paragraph 11(f)(10) alleges that on or about October 24, 1988, Acme disciplined Nicolas Valenzuela for low production by (1) issuing him a written warning and (2) suspending Valenzuela for 3 days. This discipline is alleged as a unilateral change (par. 11,g) and as a violation of Sec- tion 8(a)(5) of the Act (par. 14). Note that there is no allega- tion the discipline was unlawfully motivated and, therefore, violative of 29 U.S.C. § 158(a)(3). By its answer Acme admits the fact of the discipline, but Acme denies the discipline was a unilateral change and de- nies that it was unlawful. b. Section 10(b) In its answer (R. Exh. 36 at 7–8) Acme possibly raises a limitations defense, 29 U.S.C. § 160(b), to the foregoing Oc- tober 24 warning and suspension discipline imposed on Valenzuela. Acme does not clearly address this at the hear- ing, nor does Respondent brief this point. To the extent Acme is asserting a 10(b) defense, the assertion is without merit. The charge (G.C. Exh. 1ee) filed October 28, 1988, in Case 13–CA–28118 (the fifth case), amended on Decem- ber 6, 1988, to add the cafeteria allegation (G.C. Exh. 1hh), alleges violations of Section 8(a)(3) and (5) respecting the October 24 warning and suspension as to Valenzuela. What Acme apparently seeks to argue is this. The Decem- ber 30, 1988 complaint in Case 13–CA–28118, the ‘‘fifth complaint,’’ alleges (respecting this) that (1) on October 24 Acme unilaterally implemented a rule prohibiting employees from going to the cafeteria for coffee during working hours, that (2) on October 24 Acme warned six employees (not in- cluding Valenzuela) for violating that unilaterally imple- mented rule, and that (3) on October 24 Acme suspended Valenzuela (and Rodolfo Banales) for violating the new cafe- teria rule. There is no allegation about Valenzuela being warned and suspended for low production. The first time Valenzuela’s October 1988 low production warning appears in a complaint allegation is in the January 11, 1989 amendments (G.C. Exh. 1oo) to the first complaint, Case 13–CA–27619, which issued June 1, 1988. The March 18, 1988 charge (G.C. Exh. 1a) in that case alleges (among other items) unilateral changes on October 19, 1987, and on subsequent dates, and generally alleges that written warnings were issued. Valenzuela’s October 1988 suspension is not mentioned in the January 11, 1989 amendments. When the testimony reached Valenzuela’s October 24 low production warning, the General Counsel moved to amend so as to make clear that no 8(a)(3) allegation was intended, but merely Section 8(a)(5). At this point Acme lodged its due- process objection because of the confusion caused by the several pleadings and amendments. As described earlier, I sustained the objection and that day we adjourned sine die to await service of a consolidated complaint. (4:619, 626, 730–732.) During the adjournment the General Counsel, by motion dated April 14, 1989 (G.C. Exh. 1ss), moved to file an at- tached consolidated complaint. The October 1988 low pro- duction warning to Valenzuela was repeated, but no allega- tion mentioned his suspension. Comes now April 26 and the General Counsel, for the first time, tacks Valenzuela’s Octo- ber 24 suspension for low production on the warning allega- tion. This is done in the General Counsel’s April 26 motion to amend (G.C. Exh. 1vv) the April 14 consolidated com- plaint. Respondent Acme objected on May 8 (G.C. Exh. 1xx) on the same grounds I have described, as carried forward into its subsequent trial answer of July 11. (R. Exh. 36.) By order dated May 24, I granted the General Counsel’s mo- tions, without prejudice to Acme’s renewing its objections on resumption of the hearing. (G.C. Exh. 1yy.) The trial com- plaint (the final single document) carried forward the low production-suspension allegation as the complaint paragraph 11(f)(10) which I summarized at the beginning of this sec- tion. At the beginning of the resumption, I assured Acme that its objections were preserved for ruling. (5:737–739.) Although the General Counsel’s complaint allegations have been rather confusing (chiefly because of the number of separate complaints), there clearly is no limitation problem under 29 U.S.C. § 160(b) because the various complaint alle- gations, and the trial complaint, are based on a charge filed October 28, 1988 (Case 13–CA–28118)—barely 4 days after the discipline and at the very start of the 6-month limitations 1141 LOVEJOY INDUSTRIES 90 The document is duplicated in the record as G.C. Exhs. 31 and 46. 91 Balma gave similar testimony on direct examination, including a specific example of Valenzuela’s being absent on one occasion in excess of 30 minutes before the lunch period. (25:3826.) On cross- examination Balma concedes he was describing an incident that oc- curred before the June 6 events. (25:3898.) As Balma apparently confused these events, I do not rely on his description of this Octo- ber 1988 matter. 92 Unlike his description of his work on the part in May–June as being quite physical, Valenzuela does not describe his work on this item. Basgall explains that it is a much smaller piece. Even so, much more is involved than pushing a button and watching the machine work. Here the drill press, while spring loaded for raising, must be manually pulled down for drilling and tapping (threading the drilled holes). The part is moved from the drill to a second machine for the threading. Care must be exercised in positioning the part and centering the drill press arm for the threading operation. Valenzuela would drill six holes in the part on one machine and thread the same holes on the second machine. (20:3027, 3052–055.) In short, it is clear that operator fatigue would be a factor as a workday passes, although not to the extent as the May–June operation. 93 Basgall’s formula, detailed in his notes (R. Exh. 77 at 2) and the hearing (20:3027–028, 3054), sets the time of a complete cycle of drilling and tapping one piece, as done by Valenzuela, at 2.5 min- utes. Dividing 60 minutes by 2.5 minutes yields a rate of 24 pieces per hour. period. To the extent Acme has moved to dismiss Valenzuela’s October 24 low production and suspension alle- gation based on Section 10(b) of the statute, I deny that mo- tion. c. Facts On October 24, 1988, Acme issued Valenzuela a written warning reading:90 On 10-20-88 Jorge [Valenzuela] was warned about his unacceptable level of production and work slow- down on 10-19-88. On this drilling and tapping job, Jorge should have produced 204 pieces in 8.5 hours. He made 82 pieces. On 10-21-88 Jorge did the same job for 6.0 hours and should have made 144 pieces. He made 77 pieces. Jorge was previously disciplined on 6-16-88 for a work slowdown due to unacceptable production levels. This is unacceptable behavior and will not be tolerated. Jorge will be given a disciplinary layoff of three days. Any further misconduct, including work slowdown, dis- rupting other employees, or leaving the assigned work station for improper reasons, will result in immediate termination. Valenzuela’s suspension culminated a series of events be- ginning mid-October when Daniel Basgall (2:384; 20:3025– 3026) observed that, rather than working, Valenzuela was calling out to other employees in the department and fre- quently wandering away from his machine.91 (20:3025– 3026.) Basgall made notes (R. Exh. 77) covering his obser- vations on October 19, 20, and 21. (20:3025, 3034, 3056.) As Basgall testified (refreshed to some extent by his notes), around 8:30 the morning of October 19 he timed Valenzuela for 10 minutes as Valenzuela worked at his operation of drilling and tapping (threading) six holes in a part.92 Basgall observed from the inspection office so that Valenzuela, un- aware of the observation, could set his own pace when he was actually working. Projected over the 9.5-hour shift, and allowing for lunch (30 minutes), two 10-minute breaks, and 10 minutes for restroom visits or personal time, Valenzuela’s rate would produce 204 parts by the end of the shift or 24 pieces per hour over the 8.5 hours of actual work.93 That day Valenzuela produced 82 pieces. (20:3026–028, 3033–034, 3052–056; R. Exhs. 19h, 77.) Unlike some of his production sheets on previous days in October (R. Exhs. 19a, b, e, f, g), Valenzuela’s production sheet for October 19 (R. Exh. 19h) has no mark for any setup work and, instead, shows production for the entire 7 a.m. to 4:30 p.m. shift. About 7:15 the next morning (October 20), Basgall testi- fied, Valenzuela was called into the supervisors’ office. Present were Basgall, Larry Stoner, Valenzuela, and Marcial Canales. Basgall began by telling Valenzuela that his produc- tion was too low and that he should be producing approxi- mately 204 pieces rather than the 82 he did on October 19. Becoming hostile, Valenzuela complained of being constantly watched by management, that he had a right to get coffee outside of breaktime. Basgall said breaks are the time to get coffee. Valenzuela responded that he needed the coffee to stay warm. (Basgall concedes it was cold that day in the shop, 20:3030.) Valenzuela was told that he was not produc- ing to his potential because he was spending time walking around, talking to others, and standing by his machine but not working. Valenzuela (in an ‘‘angry’’ tone per Basgall’s notes) replied, ‘‘I am working,’’ and ‘‘I am trying my best.’’ Valenzuela said he could do better if he was not hampered by members of management watching him. Stoner said a written warning would have to issue if Valenzuela did not improve. According to Basgall’s notes, the meeting ended with Valenzuela’s saying he would try to do better. (R. Exh. 77 at 4.) Basgall testified, however, that the meeting ended on an angry tone with Valenzuela and Canales ‘‘storming’’ out the door. (20:3031.) An hour later Valenzuela, Canales, and some others punched out and left the building. (2:384; 20:3028–3037.) Neither Canales nor Stoner addresses this meeting. In a limited reference on cross-examination, Valenzuela acknowledges the fact of the meeting and that Basgall complained about his production. (7:989.) The following day, Friday, October 21, is significant for the cafeteria incident which I summarize in the next section. Respecting Valenzuela’s production on October 21, Basgall testified that Valenzuela improved his output only slightly before being reassigned to another job at 2 p.m. (20:3032.) Basgall’s notes reflect that Valenzuela produced 78 pieces between 7 a.m. and 2 p.m. and state that he should have pro- duced 156 pieces had he produced at the rate of 24 pieces per hour for the 6.5 hours (of worktime). (R. Exh. 77 at 4.) Basgall does not explain the numbers, but it is clear that they would allow for the 30-minute lunch but not the 10 minutes for the morning break or 5 minutes for the morning wash- room earlier described as the formula in Basgall’s notes. (That oversight is apparently the basis for the difference in the numbers reported for the October 24 warning/suspension quoted toward the beginning of this section. The numbers there are 6.0 hours and 144 pieces.) Valenzuela’s production sheet for October 21 (R. Exh. 192) has numbers which differ from Basgall’s notes and 1142 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD Basgall’s initial testimony about Valenzuela’s working on something else after 2 p.m. (20:3032.) The production sheet shows, and Basgall concedes (20:3034), that from 7 to 8:15 a.m. Valenzuela did setup. From 8:15 a.m. to 2:30 p.m. Valenzuela did drilling and tapping on the job in question, producing 77 pieces. The sheet also shows that Valenzuela began the same operation on a second machine at 2:15 p.m., working there until 4:30 p.m. and somehow producing dif- ferent numbers, drilling 12 pieces and tapping 19. Basgall does not explain these discrepancies, and Valenzuela does not address the topic. Accepting Basgall’s testimony that the second operation was (somehow) different from the drilling and tapping in question, and focusing on the production of 77 pieces, we see that the time of 8:15 a.m. to 2:30 p.m. is 6.25 hours. Re- ducing that, under Basgall’s formula, by 30 minutes for lunch and 30 minutes the two breaks and restroom visits, and we have 5.25 hours. Basgall’s formula in his notes make no allowance for material handling (or fatigue or other vari- ables), but his testimony does. (20:3026.) Although the num- bers possibly could be rounded off to 5.0 to allow 15 min- utes for material handling, I nevertheless shall use the 5.25 hours to compute Valenzuela’s rate of 14.66 per hour. That would project to 124.66 for 8.5 hours of actual work out of a 9.5-hour shift. Compared to his 82-piece production count on October 19 (9.65 per hour for 8.5 hours), Valenzuela’s October 21 pro- jected output of 43 more pieces (125 over 82) reflects an in- crease in Valenzuela’s production of 52.44 percent! While that does not equal the figure Basgall wanted, it shows—con- trary to Basgall’s version—that Valenzuela produced at a substantially higher rate on Friday than he did on Wednes- day. Basgall testified that on Monday, October 24, he checked and observed that Valenzuela’s production had not increased from the previous week. (20:3038.) Valenzuela’s production sheet (R. Exh. 19j) for the date reflects that, working the drilling and tapping on the pieces in question from 7 a.m. to 4:30 p.m. with no setup time indicated, Valenzuela pro- duced 100 pieces. (7:1039.) At the end of the shift, Balma gave the warning/suspension form to Valenzuela in the pres- ence of Basgall and Antonio Aguilera. (4:615, 619; 7:1028– 1029; 20:3038–3039.) Valenzuela testified that Balma re- ferred to the production level shown on the form (quoted ear- lier). He recalls no reference to a slowdown. Valenzuela as- serts that at the meeting he protested that they should con- sider the setup work he does. (7:1030–1032.) At the hearing Valenzuela did not testify that he performed setup work not otherwise shown on his production sheets for October 19, 21, and 24. Aguilera does not address this meeting. d. Discussion Recall that the allegation here is one of unilateral change, not motivation. Asserting that Acme departed from a preelection practice of normally no discipline for poor work, the General Counsel argues that the discipline imposed Octo- ber 24, 1988, on Nicolas Valenzuela was harsher than past practice and therefore an unlawful unilateral change. (Br. at 121.) Countering that before the election warnings for low production have issued, Acme cites the warnings I discussed relative to the June 16 warning of Valenzuela. (Br. at 347.) The January 12, 1983 warning of Emilio Mora is the first of these. Finding a unilateral change here, I do so on the basis that here Valenzuela did show a substantial improvement after Basgall and Stoner orally warned him on October 20. That fact alone indicates that, especially in Stoner’s department, any discipline would have to await further developments. De- spite the ‘‘slowdown’’ allegation in the warning/suspension, Acme does not claim that Basgall, Stoner, or Balma either alone or tacking observation time, watched Valenzuela throughout his shift on October 19 or 21. Yet it was for those days that the discipline issued. (20:3038; G.C. Exh. 46.) Thus, the discipline in turn depends on the production formula Basgall derived from observing Valenzuela work in a single 10-minute span. Where work, as here, involves per- sonal and physical effort as well as mental attention and care, what an employee can produce early in his shift will be greater than his production rate in the latter part of the shift. In any event, there is no evidence concerning the stand- ard Acme had for the job, and early in the hearing Balma testified that Acme sets standards for expected units per hour on every job in order to make its job bids. (1:133, 137–138; 2:201–202.) I do not credit Basgall when he testified that he had similarly timed employee Juan Garcia at some earlier time but had discarded his notes when Garcia’s production appeared satisfactory. (20:3056–3057.) In short, I find that Acme departed from its past practice by (1) relying on a single 10-minute observation of Valenzuela’s work as the production standard for disciplinary purposes, and (2) imposing the discipline of a written warn- ing and 3-day suspension within a few days after an oral warning when the orally warned employee, Nicolas Valenzuela, showed immediate improvement in his produc- tion following the oral warning. Accordingly, I find that, as alleged, Acme violated Section 8(a)(5) and (1) of the Act by imposing the October 24, 1988 written warning and 3-day suspension on Nicolas Valenzuela. I shall order Acme to re- move the disciplinary notice from its files, to make Valenzuela whole, and to offer to bargain with the Union over any similar proposed changes. Moments after Balma, at the October 24, 1988 meeting with Valenzuela, handed Valenzuela the low production warning and 3-day suspension notice (G.C. Exhs. 31, 46), Balma also handed Valenzuela a written warning (G.C. Exh. 45) for congregating for coffee in the cafeteria before the morning break on October 21, and for associated conduct. (4:615–618; 7:1029; 20:3039–3040.) I turn now to the cafe- teria incident of October 21, 1988, and the warnings issued October 24 respecting that incident. 2. The October 24, 1988 cafeteria warnings a. Pleadings Complaint paragraph 11(d) alleges that since on or about October 24, 1988, Acme ‘‘unilaterally implemented, and has since maintained, a working rule for its employees in the Unit that prohibits said employees from going to the cafe- teria for coffee during working hours.’’ Related to that allegation is paragraph 11(h)(1) which avers that on or about the same date, October 24, Acme issued disciplinary warnings to the seven following named 1143 LOVEJOY INDUSTRIES 94 Acme does not use the phrase ‘‘due process.’’ Instead, Respond- ent ‘‘objects to the General Counsel’s investigation methods, the timeliness of its actions and its attempt to resurrect issues which must have been previously investigated.’’ (R. Exh. 36 at 8.) 95 Valenzuela was suspended on October 24 over low production, not over the cafeteria incident. The General Counsel corrected the pleading error in the April 26 motion. That error affected no rights of Acme. employees ‘‘for going to the cafeteria for coffee during working hours:’’ Jose Aguirre Francisco Mombela Rodolfo Banales Hugo Paz Marcial Canales Jorge Nicolas Valenzuela Mario Garcia At the time he received the foregoing warning, Banales also was suspended for 3 days. Complaint paragraph 11(h)(2) states the fact. The warnings to the seven, and Banales’ sus- pension, paragraph 11(h)(3) alleges, flow from the rule of October 24 which, as paragraph 11(d) alleges, was imple- mented unilaterally. Paragraphs 13 and 14 allege that the posted prohibition, the warnings, and Banales’ suspension violated Section 8(a)(3) and (5) of the Act. Denying (R. Exh. 36 at 3) complaint paragraph 11(d)’s unilateral implementation allegation, Acme admits, as fact, the October 24 cafeteria warnings and the October 24 sus- pension of Banales. (R. Exh. 36 at 4.) Acme denies the alle- gations of violations of the Act. Acme also asserts (R. Exh. 36 at 8) that the inclusion of Valenzuela’s name in the enu- meration of those receiving the October 24 cafeteria warn- ings, and the allegation of Banales’ suspension, violate the limitations prohibition of 29 U.S.C. § 160(b). b. Section 10(b) Acme does not brief its 10(b) claim, nor did it include the allegations here as to Valenzuela and Banales when, after the General Counsel rested, Respondent moved to dismiss sev- eral complaint allegations. (15:2050–2082.) Acme may well have waived, or abandoned, its contention. Assuming no waiver or abandonment, I nevertheless deny any motion to dismiss the allegations on limitations grounds. The 8(a)(3) and (5) charge in the fifth case, Case 13–CA– 28118, filed October 28, 1988, specifically alleges that on or about October 24, 1988, Acme ‘‘discriminated against em- ployees Jorge Nicolas Valenzuela and Rodolfo Banales by giving them suspensions and final warnings as part of pro- gressive discipline, based on previous disciplinary actions which were in themselves unfair labor practices.’’ (G.C. Exh. 1ee.) That allegation is in a charge filed at the very beginning of the 6-month limitations period. Respondent’s position is unclear but, in any event, has no merit. To the extent Acme contends that it was denied due process because the names of Valenzuela and Banales were not added to the list of six others until the April 26, 1989 motion to amend (G.C. Exh. 1vv), I likewise reject that contention as having no merit.94 The lack of a consolidated complaint (single document), plus the frequent amendments, and the many interrelated allega- tions, combined to create confusion and misery. For those reasons and more I granted, on January 26, 1989, Respond- ent’s motion to adjourn sine die pending receipt of a single document. (4:626, 730–732.) During the long adjournment the General Counsel filed the consolidated complaint and, a few days later, the Government’s April 26 motion to amend. We did not resume until June 26. Respondent did not open its case-in-chief until September 11, 1989, the 15th day of the trial. Acme had more than enough time to prepare its de- fense. For all these reasons I also deny Acme’s motions (if they remain) to strike the allegations as to the October 24 suspen- sion of Banales. Indeed, the fifth complaint itself contains an allegation (par. VI d) attacking the October 24 suspension of Banales (and Valenzuela) over the new cafeteria rule.95 Re- spondent’s position as to Banales is, at best, unclear. c. Facts On (Monday) October 24, 1988, Acme posted (1:112; 23:3537) a notice (in English and Spanish), over Novak’s name and signature, reading (G.C. Exh. 10): NOTICE It has recently come to my attention that some em- ployees are again ignoring and abusing Company rules and the work schedules. Employees are reminded that they are not to leave their work stations unless they have a valid, work related reason or other recognized, legitimate reason (necessary and reasonable trip to washroom when this is not abused). Employees should not be in the cafeteria during working hours when they are not on a scheduled break or lunch periods. Enrique Monteon has been instructed to continue brewing fresh coffee so that it will be ready to drink during scheduled breaks and lunch periods. /s/ R. J. Novak ROBERT NOVAK Novak testified that Acme’s preelection policy permitted employees to go up to the cafeteria for a cup of coffee dur- ing worktime so long as the employee returned with the cof- fee to his work station. (23:3537; 24:3670.) Before the elec- tion, Novak testified, Novak repeatedly asked the supervisors to put a stop to employees walking around and (during worktime) staying upstairs rather than immediately returning with coffee in hand to their work stations. (23:3537, 3539.) Novak denies instituting any rules after the election prohibit- ing employees from going to the cafeteria, getting a cup of coffee, and immediately returning to work (with coffee in hand), asserts that such was permitted after the election as well, and denies that he intended any prohibition by his Oc- tober 24 notice. (23:3537–3538; 24:3672.) The purpose of his October 24 notice was to make sure that employees got their coffee and did not linger in the cafeteria. (23:3540; 24:3672.) Novak reads his October 24 notice as permitting the past practice but prohibiting any lingering for an additional work break. (24:3673.) Apparently reading it differently, Balma testified that, after employees were caught upstairs, the no- tice was posted that employees were not to go there during worktime. (1:110.) The incident prompting the (Monday) October 24, 1988 notice occurred 3 days earlier, Friday, October 21, around 1144 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 96 Valenzuela recalls it being either Banales or Mombela. (4:608; 7:1018.) Canales heard someone ask about a raise. (13:1819.) As we shall see, Scott, whom I credit, identifies the speaker as Banales. (22:3349.) 8:30 a.m. Although versions differ as to some details, there is no dispute that when Novak and Jim Scott entered the caf- eteria on that occasion, Valenzuela, Banales, Aguirre, Garcia, Mombela, and Paz already were there getting or drinking coffee, that it was their worktime, and that Canales entered shortly after Novak and Scott. Novak and Scott testified about the incident, as did Banales, Canales, and Valenzuela. A composite of the testimony of Rodolfo Banales, Nicolas Valenzuela, and Marcial Canales is that the employees ar- rived independently for coffee, and not pursuant to a plan, even though the time of their arrival is consistent with a gen- eral pattern of past practice. Several, including Valenzuela, were in line to get coffee, and others who had poured their coffee were preparing to return to work, and Mario Garcia already was leaving. All were engaged in conversation. After pouring his coffee Banales had removed his sack lunch from the refrigerator, sat at a table, and was removing a portion to take (downstairs, apparently) to a furnace to heat. At that point Novak and Scott entered. Novak was angry. He said they were supposed to be working because it is worktime. He asked for their clock numbers. One of the group (Banales, as I find based on Scott’s credited testimony) pro- vocatively asked whether the clock numbers were needed so that Acme could give them a pay raise.96 Someone asked Novak why he was angry, that they had been coming to drink coffee for a long time without his be- coming angry. Valenzuela (4:609; 7:1026) asked why he was angry with only them (the Latins/Hispanics) and not the Americans and Italians who also get coffee. Novak denied showing any favoritism. Canales, who had entered imme- diately after Novak and Scott, told Novak that the employees had a right to have coffee. Ignoring Canales, Novak contin- ued talking with the others and taking clock numbers. Canales admits that when Novak asked Canales for his clock number, Canales told Novak he could get it downstairs (pre- sumably meaning at the timecard rack or in the office). (13:1818.) Valenzuela turned to the others and spoke in Spanish. Valenzuela testified that he interpreted for them what he and Novak were saying. Valenzuela denies telling the employees they could continue drinking coffee and did not have to re- turn to work. When he spoke in Spanish to the others, Novak stated, ‘‘Oh, you refuse.’’ ‘‘Hey, Mr. Novak, nobody refuse,’’ Valenzuela replied. ‘‘Yes, you refuse,’’ Novak reit- erated. Valenzuela said they were not refusing to return to work. (4:611; 7:1014.) Novak and Scott then left, followed by Valenzuela and the others. As the employees followed Novak and Scott, Valenzuela testified, they called out to him, in a loud voice, ‘‘Mr. Novak, nobody refused to go back to work.’’ (7:1015.) Banales recalls that as the employees went down the stairs, and even after they reached the floor, they still tried to get Novak to listen but he would not do so. (14:1988–1989.) Nicolas Valenzuela testified on direct examination that for years the employees would get coffee, if they desired, be- tween 8 and 8:30 a.m.—that is, during worktime. Sometimes they even would remain in the cafeteria to drink the coffee. Supervisors such as Robert Ferguson, Jim Scott, and Daniel Basgall would see them getting coffee and never say that the employees were violating a rule. Moreover, Valenzuela would not even tell his supervisor that he was leaving for the cafeteria; he would just go. Sometimes employees would go more than once a day for coffee during worktime. After the election, restrictions were imposed, but even then Valenzuela on occasion went during worktime to get coffee. (4:600– 605.) Expanding, on cross-examination, his description of the practice before October 21, 1988, Valenzuela explains (7:990): It was allowed and it wasn’t. We all went there and they never said anything to us, but we were afraid there would be some repercussions, but they never said any- thing to us. But in early October 1988 Dan Basgall said to Valenzuela, ‘‘Another coffee; didn’t you drink coffee at 7?’’ ‘‘Yes,’’ Valenzuela answered, ‘‘but I wanted to have another cof- fee.’’ Basgall’s mentioning 7 a.m. presumably refers to the practice of getting coffee before, or at, the 7 a.m. start of the day shift. Valenzuela understandably took Basgall’s comment and question as an indication that Basgall was opposed to employees’ getting coffee during worktime. Not all employees went for coffee, Valenzuela testified, ‘‘but, as always, those of us who liked drinking coffee would go and get it. . . . Once, twice, three times a day.’’ (7:991– 992.) Balma never said anything directly, but ‘‘you could see by the expression on his face he didn’t like it when we went to get coffee.’’ (7:990–991.) Anger showed. (7:1096.) How- ever, Balma did not begin to frown or to display an angry expression until after the Union’s election victory. When Balma saw employees getting coffee before the election, Balma never did or said anything to express how he felt about their getting coffee. (7:1096–1097.) Novak’s testimony about the initial portion of the incident is unreliable. He describes seeing Valenzuela as he (Novak) was on his way to the supervisor’s office. Passing within 5 feet of Novak and ‘‘smirking’’ at Novak as he passed, Valenzuela went up the stairs to the cafeteria. It was 8:20 a.m. by Novak’s watch. Novak then ‘‘continued my busi- ness’’ (presumably inside the supervisors’ office), but (some- how) noticed that Valenzuela had not returned from the cafe- teria. Completing his business, the nature of which Novak does not recall, at 8:24 or 8:25 a.m., and observing that Valenzuela was not at his work station, Novak walked up to the cafeteria, entered, and saw Valenzuela and six really five because Canales makes six other employees, some sitting and some standing, drinking coffee. Novak said, ‘‘What are you doing up here? This is not your break period and you are supposed to be working.’’ He told them to return to work. Marcial Canales then entered, interrupted, and starting shout- ing about wages and employee rights. They all started shout- ing and defying Novak. Had they left when Novak directed them to do so, ‘‘the matter would have probably ended right there.’’ After unsuccessfully trying to get clock numbers, Novak went downstairs to find a supervisor. Finding Jim Scott in the supervisor’s office, Novak re- turned with Scott to the cafeteria. On their way up the stairs, Novak and Scott met Valenzuela, Canales, and the other five 1145 LOVEJOY INDUSTRIES 97 If, as Novak insists, employees traditionally were permitted to go get coffee to take back to the work station, then Novak’s question to Scott is inconsistent with that tradition. 98 During direct examination Canales testified that he was drinking coffee at the time. On cross-examination, Canales testified that it was a cup of water for taking some aspirins. (13:1821, 1899.) 99 On that occasion the supervisor found Canales there. Canales said he would return to work ‘‘pretty soon.’’ (13:1900.) coming down shouting ‘‘UE, UE, UE.’’ Novak directed the supervisors to issue written warnings to the seven employees because Novak felt that they were up there taking a (an extra) break rather than going for coffee and returning (with coffee in hand) to the job. On cross-examination Novak states that he returned the second time at 8:30 a.m., that Valenzuela was in the cafeteria from 8:24 to 8:30 a.m., and that Novak (and Scott) were in the cafeteria the second time about 10 minutes. (23:3540–3547; 24:3673–3676.) No doubt elements of Novak’s testimony are correct. It is not unusual for a witness, who must cover a wide range of events, to confuse some details or to merge elements of dif- ferent events. It appears that Novak’s memory played an early Halloween trick on him. ADC Supervisor Jim Scott, whom I credit, describes the incident as follows. Around 8:30 that morning he and Novak were outside the supervisor’s office in conversation. After about a minute, they saw Nicolas Valenzuela come from the washroom, smile at them as he passed, and climb the stairs to the cafeteria. ‘‘Why do you think,’’ Novak asked, ‘‘he is going up there?’’97 ‘‘I don’t know,’’ Scott responded. When Valenzuela had not returned after 3 or 4 minutes, Novak and Scott went upstairs to investigate. On entering the cafeteria they observed about seven employees, some sitting and some standing, drinking coffee, talking, and apparently taking a regular break. When Novak asked what was going on and why were the employees there, Valenzuela stepped forward and said, ‘‘We are getting coffee.’’ (22:3347–3348.) After confirming with Scott that the earliest breaktime was not until 9:10 a.m., some 35 minutes from then, Novak told Valenzuela, ‘‘This is not a breaktime. All you people need to go back to work.’’ ‘‘You let American people come up here all the time,’’ Valenzuela responded, ‘‘Why can’t we?’’ Saying that the rules are the same for everyone, Novak fur- ther stated, ‘‘This is not a breaktime, and I want you all to go back to work.’’ In the meantime, Canales had entered, poured his coffee, and was standing there. Novak told the employees to give him their clock numbers before returning to work. Banales asked when he was going to give them a raise. As Novak was still talking, he apparently did not hear Banales. Scott told his four employees—Aguirre, Banales, Garcia, and Mombela—to return to work. (22:3348–3350.) Later that morning Jose Aguirre called Scott over to his machine. ‘‘I am sorry about this morning, what happened, but I had to do something to try to get a raise,’’ Aguirre said. ‘‘I don’t understand how you think that doing that and breaking rules is going to get you a raise,’’ Scott replied. Aguirre simply said, ‘‘I need more money and I need to get a raise.’’ (22:3352–3352.) Aguirre denies this exchange. (26:3945.) I credit Scott who testified with a more persuasive demeanor. On Monday, October 24, 1988, Acme supervisors issued written warnings to the seven employees. As mentioned, four of the employees—Aguirre, Banales, Garcia, and Mombela— worked for Scott, two (Canales and Paz) worked for Robert Ferguson, and Valenzuela worked for Dan Basgall/Stoner. The warnings have 12 categories of violations plus ‘‘Other’’ for item 13. Each of the seven warnings here have boxes checked for infractions 7, 9, and 10. They read: 7. (x) Failure to obey instructions 9. (x) Leaving work [‘‘place’’ is inserted] without permission. 10. (x) Improper conduct. In the space for ‘‘Remarks,’’ Scott wrote as follows for Jose Aguirre (G.C. Exh. 7–17; R. Exh. 108): On 10/21/88 Jose was found upstairs in cafeteria drink- ing coffee and talking to other employees at 8:30 a.m. His first scheduled break time is at 9:10 a.m. The warnings for Mario Garcia (G.C. Exh. 7–20; R. Exh. 110) and Francisco Mombela (G.C. Exh. 7–21; R. Exh. 109) read the same except for the different name and clock num- ber, and Mombela’s breaktime is shown as starting at 9:20 a.m. Scott expanded the warning (G.C. Exh. 7–18; R. Exh. 107) to Rodolfo Banales to impose a 3-day suspension be- cause of other warnings, including written warnings on March 10 and August 31, 1988. (22:3351.) Banales concedes he ‘‘probably’’ could have been told that the reason for his suspension was the third written warning. (14:2011.) For his ‘‘Remarks’’ in the warnings of Marcial Canales (G.C. Exh. 7–19) and Hugo Paz (G.C. Exh. 7–22), QC Man- ager Robert Ferguson wrote (with the appropriate first name inserted): On October 21, 1988 [Marcial/Hugo] was found up- stairs in the cafeteria drinking coffee and talking to other employees at 8:30 a.m. His first scheduled break is at 9:10 a.m. [Marcial/Hugo] has been asked to adhere to the scheduled break periods. This is a written warn- ing. A second occurrence will result in progressive dis- cipline. A week or two earlier, Canales testified, Ferguson, finding Canales with a cup of coffee at his machine, told Canales not to get coffee in the cafeteria during worktime.98 Canales said the employees had the right to do so before the election so why not after the election. Ferguson did not answer. (13:1820–1821, 1898–1899.) Canales further testified that every workday from the Oc- tober 1987 election to October 21, 1988, he, and others, had gone to the cafeteria around 8 to 8:30 in the morning, during worktime, to get coffee. (13:1824.) He concedes it was not permissible to sit and drink coffee in the cafeteria during worktime and, except for one occasion before the election,99 he never did that. Instead, he took the coffee to his machine. (13:1900–1901.) Assistant Supervisor Daniel Basgall (assists Larry Stoner) filled out the warning (G.C. Exh. 7–23; G.C. Exh. 45) for Nicolas Valenzuela. (20:3039.) In the ‘‘Remarks’’ section Basgall, clearly obtaining his information from Novak, wrote: 1146 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 100 ‘‘Double, double toil and trouble; Fire burn and cauldron bub- ble.’’ IV, 1, 10. At approximately 8:24 a.m. on 10-21-88, Jorge was seen going into the washroom. He then went up to the cafeteria for at least 5 minutes before Bob Novak and Jim Scott found him talking to six other employees and drinking coffee. Jorge’s first scheduled break is not until 9:10 a.m. After being told to go back to work, Jorge began yelling and disrupting other employees. Although Basgall wrote out the warning, and notwithstanding Balma’s testimony that he was not involved in the cafeteria matter (25:3825), both Valenzuela (4:616) and Basgall (20:3039–3040) testified that Balma handed Valenzuela the cafeteria warning after giving him the low production warning/suspension. d. Discussion (1) Section 8(a)(3) Arguing that the evidence satisfies the Government’s bur- den (respecting the unlawful motivation allegation), the Gen- eral Counsel contends the following items establish a prima facie case. First, several of the seven employees, particularly Canales and Valenzuela, were known supporters of the Union. Second, several already had received unwarranted dis- cipline since the election. Third, the cafeteria incident fol- lowed a 2-day strike by a mere month. Fourth, Acme had expressed hostility toward the Union. (Br. at 59.) The first factor actually points both ways. The opposite version is that union supporters were warned because only union supporters were present. If union opponents had also been present, and only the union supporters had been warned, that would show disparity. The second factor, prior warnings, does add some weight because I have found favor- ably to the General Counsel on a few of those alleging un- lawful motivation. The third factor, passage of a month after a 2-day strike, argues timing. ‘‘Timing’’ is not the witches’ general incanta- tion from Macbeth.100 When uttered, it does not magically supply necessary, but missing, logic and causal connection to events not visible related. Had the cafeteria incident preceded the strike by a month, and had the warnings issued shortly after the strike, then timing would favor the Government. As the situation is the reverse, timing favors Acme. Hostility reflected in the findings of independent violations of Section 8(a)(1) adds some additional weight, although per- haps slight, to the General Counsel’s prima facie case. It ap- pears the General Counsel has established, prima facie, that Acme was unlawfully motivated in issuing the warnings over the cafeteria incident. Despite asserting it demonstrated that it would have disciplined the employees even in the absence of their union activities (Br. at 437), Acme does not brief the issue of whether it would have issued the warnings absent the October 1987 election and the union activities of the em- ployees. Credible evidence by the employees establishes that before the October 1987 election employees could go to the cafe- teria and pour coffee during work and return to their stations to drink the coffee. After the election Acme banned con- sumption of food and beverages in work areas. Even then employees, as Valenzuela testified, sometimes went for cof- fee despite a fear there would be repercussions. The first warnings for getting coffee did not come until Novak ob- served Valenzuela entering the cafeteria during worktime. (The warnings were not for insubordination over anything said in the cafeteria to Novak or Scott.) Novak, I find, seized on this occasion as a pretext to issue written warnings to union supporters, including Valenzuela (conqueror of Acme in the earlier case), and Canales, the Union’s chief steward, and to post the October 24 prohibition. Absent the October 1987 election and the union activities of these employees, Acme, I find, would not have issued written warnings to these employees. Accordingly, I find that, as alleged, Acme violated Section 8(a)(3) of the Act by issuing the written warnings and in suspending Banales. Acme must remove the warnings from its files and notify the employees it has done so. The posted October 24 prohibition must be rescinded. (2) Section 8(a)(5) The description and findings I have just summarized apply even more obviously to the refusal to bargain allegation. It is clear the warnings and posted prohibition unilaterally de- parted from past practice on a working condition of signifi- cance. I therefore find that, as alleged, Acme violated Sec- tion 8(a)(5) of the Act. For this additional reason the warn- ings must be removed and the posted prohibition of October 24, 1988, rescinded. 3. Marcial Canales warned and suspended in November 1988 for low production a. Pleadings Complaint paragraph 11(f)(9) alleges that around Novem- ber 14, 1988, Acme issued a written warning for low produc- tion to Marcial Canales. Other paragraphs allege that by this warning Acme violated Section 8(a)(3) and (5) of the Act. Admitting that it issued the warning Acme denies violating the statute. (R. Exh. 36 at 4, 5.) Complaint paragraph 11(i)(1) alleges that around Novem- ber 22, 1988, Acme suspended Marcial Canales for 3 days for low production. Admitting the suspension, Acme ‘‘denies that his low production was the only reason for his suspen- sion.’’ (R. Exh. 36 at 4.) The suspension paragraph 11(i)(2) alleges, resulted from the October 19, 1987 unilateral changes described in paragraph 11(a) and, paragraph 14 al- leges, constitutes a violation of Section 8(a)(5) of the Act. Acme denies these allegations. (R. Exh. 36 at 5.) Unlike allegations clearly labeling the warning as unlaw- fully motivated (par. 11j) and violative of Section 8(a)(3) of the Act (par. 13), suspension paragraph 11(c) is not listed in paragraph 11(j), the motivation paragraph, although it does appear in paragraph 13, the paragraph averring that the con- duct in the listed paragraphs violate Section 8(a)(3) of the Act. Is the inclusion of 11(i) in paragraph an inadvertence, or is its omission from the motivation paragraph an over- sight? In the Government’s brief, the General Counsel argues that the suspension violates Section 8(a)(5) of the Act (Br. at 132–133), but no contention is made that the suspension was unlawfully motivated and violative of Section 8(a)(3). In the fifth case, Case 13–CA–28118, neither the October 28 charge nor the December 6 first amended charge mentions either the October 14 warning to Canales or his November 1147 LOVEJOY INDUSTRIES 101 Recall that the 1-day strike of October 26, 1988, was called to protest, in the words of Terry Davis in her October 26 letter (G.C. Exh. 42) to Novak, ‘‘the Company’s unilateral withdrawal of the right to take food and coffee to the work station.’’ (2:415a.) 102 Johnnie’s Poultry Co., 146 NLRB 770 (1946). Salzman distrib- uted prepared disclosure/consent statements to the employees who, aided in part by Sanchez translating, signed them. (R. Exhs. 67, 68, 69.) 22 suspension. The first pleading to attack either appears in the January 11, 1989 amendments (G.C. Exh. 1oo) to the first complaint, Case 12–CA–27619. In the January 11 amendments, the warning and suspension are added in sepa- rate subparagraphs and alleged to be violative of both Sec- tion 8(a)(3) (including the motivation paragraph) and Section 8(a)(5). However, the April 14, 1989 consolidated complaint (G.C. Exh. 1ss) drops paragraph 11(i), the suspension, from the enumeration in the motivation paragraph, paragraph 11(j), although including it in both the conclusionary paragraphs as Section 8(a)(3) and (5). That situation is repeated in the April 26 amendments (G.C. Exh. 1vv) and, finally, carried forward to the trial complaint (G.C. Exh. 1zz). The questions remain. The complaint stresses, in paragraph 11(i)(2), that the sus- pension resulted from the unilateral changes described in paragraph 11(a). Even so, paragraph 11(a) is picked up in the motivation and the conclusionary paragraphs as 8(a)(3) con- duct as well as unilateral changes violative of Section 8(a)(5). Acme briefs the matter as including an 8(a)(3) alle- gation. (Br. 377, 379, 392–397.) Whatever question could have been raised about the pleadings, the issue was fully liti- gated at the hearing, briefed by Acme, and tried by implied consent. b. Background Hired in about early 1985, Marcial Canales worked in the drilling (waveguide) department under Robert Ferguson until early November 1988. (13:1770, 1816–1817, 1898.) Around November 4, about a week after the October 26 strike,101 Canales was transferred to Jim Scott’s ADC where Canales, for the first time at Acme, began operating a punch or trim press. (13:1769, 1838–1839; 18:2627.) Told by Ferguson that he was being transferred because Wave Guide had a lack of work, Canales informed Scott that he was entering ADC under protest. Canales was displeased because he felt that a less senior person, Antonio Diaz, should have been trans- ferred before he was selected. Scott did not respond. (13:1839, 1901–1902.) Canales was a central figure in the union organizing effort. After the Union won the election the employees elected Canales to be the chief steward. As I mentioned earlier in the background section of this decision, the Union notified Acme of the elected stewards, including the capacity of Canales as chief steward. The record reflects that Canales sought to be active in his position. Earlier I generally described Acme’s contentions that the Union engaged in a corporate campaign strategy which in- cluded encouraging employees to adopt certain ‘‘inside’’ tac- tics. One of these purported tactics was that employees lower their production—engage in a slowdown. Acme contends, and adduced evidence, and perhaps Terry Davis, began mak- ing the slowdown suggestion at least as early as May 1988. There is no dispute that in the summer of 1988 union lead- ers, particularly Canales and Valenzuela, would lead em- ployee discussions in the cafeteria during the lunch and break periods about ways of pressuring Acme to recognize the Union. Canales so admits (18:2682) as does Valenzuela. (19:2735–736.) Both deny that the tactic of lowering produc- tion ever was discussed. (18:2685; 19:2736.) Nevertheless, there is testimony that Canales and Valenzuela made this suggestion there and elsewhere and that this suggestion, among others, continued to be made in September and Octo- ber, including at a gathering on the west parking lot around the date of the second strike, October 26, 1988. I need not summarize all that testimony or seek to resolve all the dis- putes except for the specific instances which follow. In late October a group of six or seven employees, who remained outside during the October 26 strike, discussed their opinion that the strike was senseless, that they wanted a raise, that they were being insulted by the union leaders, and that they would be better off with the Union out. The group included then ADC employee Armando Escheverria who testified that the group elected Antonio Sanchez to ar- range a meeting with Plant Manager Balma to discuss these concerns. (17:2404, 2413–2414.) Sanchez did so. Responding that the group should speak with the Company’s lawyer, Balma said he would arrange the meeting. (21:3161–3163.) After the 4:30 p.m. end of the first shift on November 9, 1988, the group met in the front office conference room with Balma and Attorney James J. Salzman. Balma left after intro- ducing Salzman. After advising the employees of their John- nie’s Poultry102 rights, Salzman and the employees discussed events. It is not clear that a pay increase was mentioned. During the meeting Sanchez informed Salzman that these employees were having problems with union supporters. Salzman asked what problems. Escheverria and others then described certain problems with Canales, Valenzuela, and others. (17:2376–2378; 21:3164–3169.) Escheverria com- plained that Canales kept telling him to lower his production. (17:2377.) Of the group, at least Escheverria (17:2394–2395), Javier Navarez Carrasco (17:2500), and Jose Ortega (21:3176) signed statements about their complaints. Salzman wrote them by hand. The following day typed versions were presented. Sanchez translated for Ortega who signed his. (21:3177–3181, Sanchez.) Hired in 1979, Escheverria worked in the ADC until about April–May 1989 when he became a QC inspector. The job change entailed no pay increase. (17:2349, 2386–2387.) Carrasco a ZDC employee for several years, became a QC inspector in the May–June 1989 timeframe. Hearing that the position was open, Carrasco went to Balma. Telling Carrasco he could have the job, Balma also advised him that if he could not perform the work he would be sent back to ZDC. Carrasco testified that he took the QC inspector’s job, which involved no pay raise, in order to broaden his experience. He denies that he was given the QC inspector’s job as a reward for agreeing to testify for Acme. (17:2478, 2502–2504.) The General Counsel argues for an inference that these witnesses testified in exchange for a pay increase. (Br. at 150.) This argument is tied to the efforts of Escheverria’s group, as I describe later, seeking in November–December to obtain a general pay increase. Although considering these factors, I nevertheless reject the General Counsel’s argument. 1148 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD c. The November 14, 1988 warning (1) Facts When Canales was transferred to the ADC and the punch press, Scott showed him how to place castings, how to oper- ate the machine, and how to clean and oil the press and cast- ings. (23:3479–3480.) Conceding that the machine is acti- vated by buttons, Canales asserts that Scott only spent about 2 minutes with him (13:1841, 1873), but Scott credibly testi- fied that the work does not require a lot of skill because the machine does most of the work. (23:3484.) The trim press operator’s task is to trim castings coming from the operator of an adjacent die cast machine. A steel table, 20 feet or so in length, sits between each die cast ma- chine and its paired trim press. The die caster places the just- molded castings on the metal table where the castings are conveyed to the trim press. Scott has 12 to 13 trim presses operating at any one time. (21:3208–3210.) A trim press op- erator should be able to trim up to three times as many cast- ings as the die casting machine produces. (2:328.) On Canales’ first day, Scott told him that he was giving him an easy casting to trim, and that all he expected was for Canales to keep up with the die caster. (2:334.) November 4, 1988, was a Friday. There is no dispute that in that week, whether the single day of November 4 or as Canales suggests (18:2640), for a day of two earlier, Canales was paired with die caster Manuel Bueno. (18:2639; 22:3359.) The following Monday, November 7, Canales was moved to a different machine and paired with die caster Armando Escheverria. (18:2640; 22:3359.) Despite first testi- fying that Scott never told him he needed to increase his pro- duction (13:1876), Canales admits that in a pretrial affidavit given on November 22, shortly after the events, he reported that Scott, on the Monday (November 7) of that second week, told him he had to increase his production. (13:1877.) Canales acknowledges that Escheverria sent castings to him and they worked together a few times, although he does not fix the dates. (18:2640, 2644, 2647, 2667.) Escheverria testified that he knows one date to be November 9 because that was the day he met with Attorney Salzman and signed the pretyped Johnnie’s Poultry statement (R. Exh. 67), copies of which Salzman distributed to the small group. Because of that dated statement, Escheverria is able to fix the dates of the conversations that week as occurring of November 8 and 9. (17:2351, 2353, 2393.) Scott (23:3367–3369) matches pro- duction records on Magnavox part 660248 of Escheverria (R. Exh. 12–6) and Canales (R. Exh. 44) for (Thursday) Novem- ber 10. For the dates of November 7–9 we have only Escheverria’s records. (R. Exh. 112–3, 4, 4.) In any event, Escheverria testified that the first day, No- vember 8, on which they worked together that week, Canales approached him about an hour into the shift. Because they worked 15 to 18 feet apart, a conversation could be held only if one approached the other. Canales came to him, Escheverria testified. (17:2352, 2387–2389.) On this occasion, Escheverria testified, Canales said Escheverria should take it easy and not work so fast, that Acme was not paying him enough for that work. If they worked slower, Canales said, it would be a way to pressure Acme and to help the Union. Escheverria replied that he was working normally, that he liked his job, and that if he did not like working there he would be working elsewhere. Canales repeated these efforts three or four times each of the 2 days of November 8 and 9. (17:2352–2354, 2389–2393.) Escheverria never complied with this suggestion of Canales. (17:2400.) Unable to recall if he and Escheverria spoke on November 8 and 9, Canales acknowledges that they could have. Canales recalls no discussion about production, and asserts they prob- ably talked about sports or other matters. He denies suggest- ing that Escheverria slow his work, asserting that such a re- quest would be against the law. He denies ever suggesting that to anyone as a means of pressuring Acme. (18:2663– 2667.) Escheverria testified persuasively. Frequently exhibit- ing hostility throughout his testimony, Canales testified with an unfavorable demeanor. I credit Escheverria and disbelieve Canales. For similar reasons I credit Scott over Canales. Scott testified that Canales’ work on November 10 aver- aged 77 pieces per hour on Magnavox 248 (part 660248). (23:3372.) Experienced trim press operator Rodolfo Banales, records purportedly reflect (R. Exh. 113:3–6), averaged 120 parts per hour over 4 days in October, Scott testified. (23:3372–3373, 3374.) All the numbers on the documents are not explained, and the hours Scott was using as a formula are not specified. For example, taking the first of Banales’ sheets, October 10 (R. Exh. 113–3), Banales records 1018 pieces from 7 a.m. to 4:15 p.m. Using 9 hours (as Scott ap- parently does for Canales on his shift of 7 a.m. to 4:15 p.m.) as the work shift divisor (excluding the 30-minute lunch but including the breaks), computation yields a quotient of 113, not Scott’s 127. (But 1018 divided by 8 hours does yield 127.) The number of pieces Banales produced on October 11 are illegible on the copy. For October 12 Banales produced 940 pieces of the Mag- navox part from 7 a.m. to 4:15 p.m. (R. Exh. 113–5.) Scott calculated that as 118 per hour. But 940 divided by 9 is 104 (rounded) per hour. (If a formula for 8 hours were used, the hourly rate would be the 118 Scott inscribed on the form.) The figures for October 20 do not compute. Banales recorded 857 pieces trimmed from 7 a.m. to 4:15 p.m. Although Scott calculated that as 122 pieces per hour, 857 divided by 9 hours is, rounded, 95 and by 8 hours is 107. Other work is shown on the page (R. Exh. 113–6) as on other pages, but is unexplained and the implication by Scott’s testimony is that his hourly figure relates only to the single row where his figure is inscribed. Scott testified that Canales had been allowing the castings to pile up on the table because he was not keeping up with the die caster. Canales had to stop his trim press and move the castings because the table was full. This was happening everyday. (2:328–329; 23:3371.) On (Friday) November 11 Scott assigned Canales to a different trim machine to work on a job for Johnson Controls, part No. CST–70–4P. It was one of Acme’s easiest trim jobs, Scott testified. Nevertheless, Canales again had the pieces piling up so much on the table that the die caster had trouble finding space to put more cast- ings. (23:3370–3371, 3378, 3481, 3485–3486.) For that full day of November 11 Canales’ production sheet (R. Exh. 40) reflects a production figure of 400 pieces. (23:3371.) Scott inscribed ‘‘44’’ per hour on the sheet. (23:3374, 3375.) Scott testified that Jose Meza trimmed 102 castings per hour the first time he worked on the same part using the same machine. (3:332; 23:3373–3374.) Scott’s 102 per hour figure is inscribed on Meza’s early October (the precise date 1149 LOVEJOY INDUSTRIES is illegible, but perhaps October 4) production sheet. (R. Exh. 113–7.) Canales testified that Meza also was transferred from working with Canales in drilling to the ADC before Canales. (13:1880–1881.) Scott does not explain how he cal- culated the figure of 102 pieces per hour for Meza. Accord- ing to Meza’s production sheet, he worked the first 3 hours trimming 377 pieces of a different part. From 10 a.m. to 4:15 p.m. he trimmed 459 pieces of part CST–70–4P. Scott appar- ently used a 9-hour formula for his other calculations re- specting Canales, as on the 44 per hour for Canales for his 400 pieces from 7 a.m. to 4:15 p.m. on November 11. The figures for Rodolfo Banales, as I have shown, fit an 8-hour formula rather than a 9-hour one. If we count Meza’s work from 10 a.m. to the 4:30 p.m. end of the shift as 6.5 hours, then subtract 30 minutes for lunch but disregard the afternoon break, we have the 6 hours on the standard Scott applied to Canales. Dividing 459 by 6 hours yields a quotient of 76.5, or 77 pieces per hour—not 102! However, multiplying Scott’s 102 by 4.5 hours does yield the 459 pieces Banales produced. In theory, Scott could have made a mental error and arrived at 4.5 hours by count- ing from noon to 4:30. On the other hand, when Scott com- puted Canales’ rate for his December 13 work on part CST– 70–4P from 10 a.m. to 4:15 p.m., he obviously used a divi- sor of 6 hours (10 a.m. to 4:30 p.m. less lunch of 30 min- utes) in arriving at an hourly rate of 100. (12:3389; R. Exh. 43.) The comparison figures Scott shows for Sacramento Olivares, an experienced trim press operator, withstand com- putational scrutiny. Thus, on November 3 Olivares trimmed part CST–70–4P at the rate of 97 per hour and on November 4 at 104.5 per hour (Scott shows 104 rather than 105) as Scott inscribed on the documents. (R. Exh. 113–8, 9.) On November 5 Adan Pena, working on part CST–70–4P, trimmed 305 pieces between 8:05 and 11:45 a.m. Whether that time spanned his 30-minute lunch is not disclosed in the record. Scott inscribed the figure of 81 per hour. (23:3374; R. Exh. 113–10.) If 4 hours is used as a divisor the result is 76 per hour; if 3.5 hours, then 87 is the quotient. Scott’s figure of 81 for Pena appears to be substantially correct. Pena, Canales admits, normally works as a machine operator (die cast, apparently). (13:1881–1882.) The morning of Monday, November 14, Scott went to Canales and asked why Canales was not keeping up, that he was holding up production, and that he should be trimming about 100 pieces per hour. ‘‘You have enough production,’’ Canales replied, throwing up his hands and adding that the needed a helper. Moreover, if Scott did not like his produc- tion then, Canales said, ‘‘Get somebody else.’’ (2:329–30; 23:3370.) Scott’s note (R. Exh. 113) for that morning con- cludes by recording that Canales then said Scott should make the trim press operate faster. ‘‘I told him that he needed to work faster not the machine. He told me that’s all he could do.’’ Canales denies most of this, other than to say Scott said Canales should be trimming at least 600 pieces a day. (13:1842–843, 1875–877.) Canales concedes that in describ- ing the November 14 conversation in his pretrial affidavit he did not report Scott’s purported 600-a-day figure. (13:1877– 1878.) Canales’ mention of a helper apparently is a reference to the trim operator’s task of having to break off the extra flashing from the casting. According to Canales, he had re- ceived no training on how to do this, and did it by hand until after the November 14 conversation when Eduardo Vasquez, apparently a trim operator, showed him how to use a piece of metal to do it. (13:1843–1845, 1874.) Acknowledging that the extra flashing must be broken off by hand or a piece of wood, Scott asserts that San Roman assured him he had in- structed Canales in the proper method when training him on the Johnson controls part. (23:3487–3488.) San Roman does not address this in his testimony about other matters. Al- though it would seem obvious that breaking off the flashing could be done quicker and easier with a piece of wood or metal rather than by using the hand, I credit Canales that it was Vasquez who showed him the faster and easier tech- nique. Following his November 14 morning conversation with Canales, Scott reviewed production records, compared rates, and, as mentioned, inscribed a rate figure on the sheets. (23:3375.) Deciding that Canales’ production was unaccept- able and warranted a written warning, Scott went to Robert Novak who, agreeing with Scott’s decision, said he would participate in the meeting with Canales. (2:325, 339; 23:3375.) Scott does not explain why he went to Novak rath- er than the plant manager, Peter Balma. Novak asserts that Balma came to him about the slowdown and that he in- structed Balma and Scott to issue a disciplinary report. (23:3556.) Balma does not address the point. Canales sug- gests that around this time Balma was on vacation. (13:1852.) In any event, the afternoon of November 14 in the super- visors’ office the meeting was held. Present were Novak, Scott, possibly San Roman, Canales, and another employee Scott recalls as Mauricio Aguirre. (2:338; 23:3376.) Canales asserts that present to assist him was Antonio Aguilera. (13:1846.) Whether Aguirre or Aguilera, neither addresses this meeting in his testimony, nor does San Roman. According to Canales the meeting started with Novak screaming at him and ultimately giving him a written warn- ing after Canales unsuccessfully pleaded for a chance to learn the job. (13:1846–1847.) Scott testified that Canales said, ‘‘You people have enough production. I work hard enough and I need a helper.’’ (23:3377.) Novak said the job was for one person, that no one else had a helper, and he believed Canales was engaged in a slowdown. (23:3376.) Scott (23:3376) and Novak (23:3556–3557; 24:3676) testified that Novak read a prepared statement. The statement (R. Exh. 118) reads the same as the one (R. Exh. 117) which Novak read to Valenzuela in June. The statement Novak read to Canales reads: You are participating in a work slow down on your job. (Here explain the number of parts Marcial made on the dates involved then explain the number of parts the other employees made on the dates involved.) You continue to accept your regular wages from the company without providing the standard performance on your job. Your slow down actions are not protected activity under the law. The company will not tolerate your work slow down. If you continue your work slow down, you will be discharged for misconduct. 1150 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 103 Duplicated in the record as G.C. Exhs. 26–1 and 5q. The written warning (R. Exh. 113–1),103 with box 13 ‘‘Other’’ marked, provides as follows in the text of the sec- tion for remarks: On 11/14/88 Marcial Canales was warned about his unacceptable level of production. Compared to other employees who have run the same job in the same op- erations, he should produce 100 pcs. per hour compared to 44 pcs. per hour. On other work his output is no better; produces 77 pcs. per hour compared to 127, 122, 121, and 128 pcs. per hour by other employees. His attitude to his job is—if we don’t like it get somebody else. See other comments by his supervisor and attached statements read to him. This is second written discipli- nary report. The third report will be accompanies by a three day layoff without pay. The second group of numbers, comparing Canales’ 77 to Banales’ 127, 122, and 121, and possibly someone else’s 128 (although possibly the 128 was meant to be the 118 of Banales), as I have summarized, uses an 8-hour standard for Banales while applying a 9-hour standard to Canales. Such a comparison detracts from the integrity of the comparison for that job. The first set of figures, comparing a goal of 100 to Canales’ 44, picks Canales’ first day working that job, the Johnson Controls job, Park CST–70–4P. As I have com- puted, Meza trimmed 77 per hour (not Scott’s 102) on his first day with that part. Adan Pena, not normally a trim press operator, did approximately 81 per hour on November 5. Looking only at the figures, it seems that Canales was trim- ming substantially fewer than others of like training and ex- perience. Reference in the text that this is Canales’ second written disciplinary report is not explained by Acme. Scott asserts that it was his first to Canales because that of November 21 is only the second he signed. (2:343.) The reference possibly is to the November 18, 1985 warning (R. Exh. 74), men- tioned earlier, which Canales received from Ronald Adamczyk for failing to obey instructions, resulting in poor production. According to Canales he protested to the then plant manager, Harry Georgeson, who, telling Canales not to worry about it, threw away the warning. (13:1914–1915.) The reference could be to a 1987 warning which Canales re- ceived for failing to wear safety glasses. (13:1910–1911.) Scott testified that he believed Canales intentionally had trimmed at a slow rate based on his figures and his various remarks to Scott reflecting his attitude of opposition. (2:341.) (2) Discussion For some reason Novak does not testify that his slowdown statement, read to Canales, was in any way based on the No- vember 9 report Escheverria made to Attorney Salzman. Al- though the prepared statement came from Salzman in June, Novak easily could have been prompted in November not simply by the production figures, but also by Escheverria’s report. Indeed, as Acme cannot be said to have given Canales an overabundance of time before issuing him the November 14 warning, the factor which most persuades to- ward a finding of slowdown is the November 9 report of Armando Escheverria to Attorney Salzman. I infer that Salzman did not keep that information to himself but that he immediately conveyed it to Novak. Still, it is only by inference that I find Novak was moti- vated to read his prepared statement largely because of Escheverria’s report of Canales’ suggestion that Escheverria slow his production as a tactic for pressuring Acme. The pro- duction rate of Canales over those dates, including the early morning of November 14, prompted the action taken. In short, I find that there was more coordination between Novak and Scott than they describe. Actually, I have jumped ahead and bypassed a discussion of the General Counsel’s prima facie case, or lack of one. To show that prima facie case, the General Counsel argues: timing (soon after the cafeteria incident and the October 26 strike), no recent warnings, insufficient training instructions and time, and departure from past practice in tolerating low production. (Br. 129–132.) Timing could merit weight. If Canales’ testimony were accepted, lack of training and time could be accorded weight. The failure to grant more time to Canales ties to the past toleration factor. However, the Gen- eral Counsel offered no past production records on this point. Thus, there is no evidence showing that anyone ever trimmed as few parts as Canales did in his first few days without re- ceiving a warning. But there is no need to debate the prima facie case issue. Even if the General Counsel established a prima facie case it is clear, and I find, that Acme would have issued the No- vember 14, 1988 written warning to Marcial Canales regard- less of his union activities. Moreover, in view of my finding of a slowdown by Canales, there is no evidence of harsher discipline or unilateral change. Accordingly, I shall dismiss both the 8(a)(3) and (5) allegations respecting the November 14 warning, complaint paragraph 11(f)(9). d. The November 21, 1988 suspension (1) Facts Scott testified that after November 14 he assigned Canales to another trim press to give him a chance to improve his production on a different job, Magnavox 342 (part 660342), and immediately the table began filling with castings. It had been several months since Acme had run the job. (23:3378– 3379, 3488.) The part was no more complicated to trim than the Johnson Controls part, Scott testified. (23:3485–3486.) In describing work before November 14, Canales conceded that operation of the machines is essentially the same, but some parts are larger than others. (13:1842.) Canales does not say whether he considers a larger part more difficult. Scott sug- gests that a larger casting can be easier to handle, and that the Magnavox 342, although larger to begin with, has more cavities. When the trim press closes, it separates the cavities into six small castings. (23:3485–3486.) Canales concedes he received training from San Roman on how to place and oil every 10th part with a brush. San Roman did not tell him how many pieces Canales was ex- pected to trim. And according to Canales, during that week no one from supervision spoke to him about his production level. For that week, Canales trimmed some 110 pieces dur- ing the 9 hours of work each day, and he had no reason to believe his output was unsatisfactory. (13:1848, 1850–1852.) 1151 LOVEJOY INDUSTRIES 104 The dates have exhibits duplicated in the record. Canales iden- tified the exhibit for November 17 as R. Exh. 45 (13:1888), and Scott, R. Exh. 114–2, as an attachment to Acme’s copy of the No- vember 21, 1988 written warning, R. Exh. 114–1. (23:3381–3382.) For convenience, I shall use the numbers designating the documents as attached to Acme’s copy of the warning. 105 For all 3 days Canales showed 7 a.m. to 4:15 p.m. As dis- cussed earlier, cleanup time begins before the shift ends at 4:30. There is a 30 minute lunch. The 9 hours includes breaks in the morning and afternoon. 106 As with Scott’s computations for the November 14 warning, the General Counsel submits no analysis of these figures. 107 Balma testified that on (Monday) November 21 the decision had been made to suspend Canales. (25:3830.) Balma apparently means that the decision was made at the end of the November 21 shift. 108 Balma testified that Scott reported he had continuously warned Canales that without more production he would be suspended, and that, apparently, Scott had been keeping Balma informed of this. (25:3828.) As mentioned earlier, Balma was not present at the No- vember 14 warning, and Canales advises that, as of November 22, Balma had just returned (on that Monday, apparently) from vacation. (13:1852.) Scott testified that he ‘‘talked to’’ Canales a few times but saw little improvement and, after a few days, decided to write him up again after checking production records of oth- ers who (earlier) had trimmed on that job. (23:3379.) Scott makes no claim that he gave Canales a production goal for the Magnavox part 342. Although Scott asserts that the pre- vious week Canales would be very indignant and not want to talk with Scott (2:335), Scott does not tell us what either he or Canales said on these occasions after November 14 when he ‘‘talked to’’ Canales, nor does Scott claim he make any notes about such conversations. Skipping (for some reason) the next 2 days (November 15 and 16, 1988), production records in evidence for Canales begin with (Thursday) November 17 and cover November 18 and 21, a Monday. These records show that Canales trimmed Magnavox 342 as follows, with the rate per hour, as noted earlier, added by Scott: Date R. Exh.104 Pieces Hours105 Rate 11–17 114–2 1110 9 123 11–18 114–3 1111 9 123 11–21 114–4 1100 9 122 The comparison records in evidence are for Florentino Olivares on May 12 (R. Exh. 114–5), Sacramento Olivares on May 16 (R. Exh. 114–6) and June 16 (R. Exh. 114–7), and Alberto Martinez on June 21 (R. Exh. 114–8). Florentino Olivares and Alberto Martinez are die casters. (13:1890– 1891; 23:3382, 3384.) Canales asserts that they sometimes work on the trim press (13:1891), but Scott says such times are infrequent to rare. (13:3383–3384.) Hired October 13, 1976, Florentino Olivares has been classified as a die caster since October 2, 1978. (G.C. Exh. 3.) His employment record suggests he could have worked for a few months on the trim press as a relief operator from June 2 to October 2, 1978. (G.C. Exh. 3b.) Hired in 1979, Sacramento Olivares has been a punch press operator at Acme for much, perhaps all, of those years. (11:1479, 1520.) It is not clear that the records identified by Scott are the only ones for these three, or any other trim press operators, on Magnavox 342 from mid–May to late November 1988. In any event, the produc- tion sheets of the three disclose: Name R. Exh. Pieces Hours Rate F. Olivares 114–5 1272 6 213 S. Olivares 114–6 1400 9 175 Name R. Exh. Pieces Hours Rate 114–7 1316 9 165 A. Martinez 114–8 700 175 Scott’s computations for Sacramento Olivares and Alberto Martinez raise questions.106 On May 16 Sacramento Olivares recorded a full shift, 7 a.m. to 4:30 p.m., trimming 1400 parts, in those 9 hours. (R. Exh. 144–6.) But 1400 divided by 9 hours yields a quotient of 156 (rounded), not 175. If 8 hours is used as the divisor, however, the quotient is Scott’s 175. The same result obtains for the June 16 report (R. Exh. 114–7) of Sacramento Olivares. With 9 hours as the divisor the quotient is 146. Only by using an 8-hour divisor is the quotient Scott’s 165. On June 21 Alberto Martinez worked a partial day on Magnavox part 342, from 11:30 a.m. to the end of the shift, or 5 hours. The first problem is that the sheet does not show when he took his lunch. If at 11 a.m., then the correct divisor is 5 hours. If at 11:30 or later, the appropriate divisor is 4.5 hours. Using 5 hours yields a quotient of 140, and 4.5 hours an hourly rate of 156 (rounded). As the January 18, 1988 posting (G.C. Exh. 9–5) sets the lunch period (for most em- ployees, apparently) as 11:30 a.m. to 12 noon, I shall use 4.5 hours as the divisor, yielding the 156 rate. To replicate Scott’s 175, however, one must use a divisor of 4 hours! In making his computations, Scott erred either by carelessness or by fraud. I note that Scott’s errors always penalize Canales by using the smaller divisor for others while burden- ing Canales with a divisor of the full 9 hours. Having compared the records, and after reporting to Balma that Canales had not improved,107 Scott approached Canales shortly before the end of the shift on Tuesday, November 22. (2:345.)108 Scott told Canales to come to the office that Balma wanted to speak to him. Canales testified that he asked Scott if he (either Scott or Balma, apparently) was going to give Canales another warning or suspension. Who is talking about your production, Scott responded. Canales said he would be there after he cleaned up and changed his clothes. (13:1853.) Canales testified that he asked Scott about a warning or suspension because the only time Scott came to him was ‘‘to bother me,’’ and that he immediately became suspicious that he was going to be disciplined when Scott said Balma wanted to talk with him. (13:1904–1905.) On his way to the washroom Canales saw San Roman. Why, Canales asked, had San Roman never told him how many pieces the company wanted trimmed. San Roman said 1152 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 109 Based mostly on the version of Canales. (13:1855.) Scott re- calls Canales saying ‘‘because I am the Union’’ (23:3385), and Balma renders it as ‘‘because I’m a Union leader’’ (25:3829, 3831). The difference is immaterial. 110 Canales insists Balma told him he was fired. I need not resolve that immaterial dispute because if Balma said it, he relented. On his return from his suspension, Canales was given a written warning (R. Exh. 115), dated November 22, for insubordination and improper conduct in his refusal to come into the office and for insulting Scott. (2:356–357.) There is no allegation concerning the insubordination warning. he was afraid to talk to Canales. Why, asked Canales, for they used to talk. San Roman said the company probably wants more production from Canales. (13:1853–1854.) As- sistant Supervisor Juan San Roman does not address this conversation in his own testimony. I credit Canales. After changing his clothes, Canales passed by the office on his way to the timeclock. Many of the workers were there ready to punch out. Balma called out to Canales to come to the office. (13:1854; 23:3384.) ‘‘No,’’ Canales replied, ‘‘You guys come out here and say what you have to say in front of everybody. I’m not coming in there.’’ (1:346, 356; 13:1854; 23:3385.) Balma did so. With Scott present, Balma handed a written warning and 3-day suspension (G.C. Exh. 27) to Canales and told him he was suspended for 3 days for low production. (2:343–346; 13:1854, 1856; 23:3385.) About the same time Balma was handing the disciplinary notice to Canales, Canales asked why he was doing this, that Balma had been on vacation, that Balma knew Canales had been in the drilling department and had no experience on the punch press, and that Canales thought he had been doing a good job. Balma said he could make better production than Canales had shown. Canales admits that by now he was speaking in a loud voice and that he was upset. (13:1855, 1895–1896.) Scott (23:3385) and Balma (25:3829) character- ize it as yelling and screaming. In any event, I find, Canales told Balma, ‘‘You are afraid of me because I am a union man.’’109 Balma denied he was afraid of Canales. ‘‘Yes you are,’’ Canales said, and pointing to Scott, added, ‘‘and you sent this asshole to get me.’’ Balma told Canales he was sus- pended for 3 days and to punch out and go home.110 Canales did so. (2:356; 13:1859.) The warning which Balma handed to Canales on Novem- ber 22 is dated November 21 and has box 13 ‘‘Other’’ marked. The text of remarks reads (G.C. Exh. 27): Marcial continues to produce at an unacceptable level. He has recently received a written warning and has not improved his production. (Daily worksheets attached.) Scott testified that copies of the worksheets (production records) were not attached to the copy given to Canales. These are the sheets attached to Acme’s copy (R. Exh. 114.) (23:3381.) Scott testified that Balma is the one who decided on the suspension. (2:343–344.) To the written warning which Scott had prepared (23:3380) Balma wrote, at the bot- tom right, ‘‘Reviewed disciplinary reports. Marcial is being given 3 working days off without pay. 11-22-88. /s/ Pete Balma.’’ (G.C. Exh. 27; 2:343; 23:3380; 25:3830.) After the suspension/insubordination incident at the timeclock on No- vember 22, Balma wrote at the bottom left of his, Acme’s, copy of the November 21 warning/suspension notice (R. Exh. 114), a brief reference to Canales’ refusal to come in- side the office. (25:3830–3831; R. Exh. 114–.) When Canales returned from the suspension, Scott testi- fied, Scott reassigned him to the Johnson Controls job, part CST-70-4P, the second assignment involved in the November 14 warning. Recall that on his first day on that job, Novem- ber 11, Canales trimmed 400 parts in 9 hours, for an average of 44 parts per hour. (R. Exh. 40.) Recall also that after the November 14 warning, fellow trimmer Eduardo Vasquez showed Canales that he could work faster by removing the excess flashing from the castings with a piece of metal rather than by using his hands. Scott testified that Canales improved immediately after his return from suspension. No longer did castings pile up. (23:3388, 3489.) Presumably Canales’ first day back at work was Monday, November 28, as the 3 days of suspension would have been Wednesday to Friday, November 23–25. Not until December 8, the ninth workday, however, do we have a production record in evidence. That record (R. Exh. 41) shows that Canales produced 905 pieces over about 8.5 hours for, as inscribed by Scott, an hourly rate of 106. Recall that on November 14 Scott informed Canales he should be trimming close to 100 pieces per hour on part CST–70–4P. On Monday, December 12, Canales worked the first 30 minutes on another part, starting on CST–70–4P at 7:30 a.m. and trimming 873 castings before signing out at 4:15 (for cleanup). (R. Exh. 42.) Based on Scott’s earlier computa- tions, that would be a divisor of 8.5 hours, yielding an aver- age rate of 103. As Scott inscribed 109, it is obvious he used the more beneficial divisor of 8 hours. Such, of course, en- hances any theory that the suspension prodded Canales to work up to his capabilities. The last record, for December 13 (R. Exh. 43), reflects that in 6 hours Canales trimmed 600 pieces—exactly 100 per hour. (23:3389.) Scott testified that the doubling from 44 per hour on No- vember 11 to the 100 plus in December confirms what he previously suspected, that the 44 per hour was an intentional holding back on his production. Scott asserts that none of the factors such as skill, machines, or materials changed, but only time had elapsed with Canales’ intervening suspension. (23:3389.) Canales was not called on rebuttal to offer his view of why he was able to increase his production from 44 per hour to 100 plus per hour. (2) Discussion (a) Section 8(a)(3) Contrary to Scott’s version, I credit Canales that Scott did not ‘‘talk to’’ him about his production level between No- vember 14 and 21. Neither did San Roman. The last word of an expected rate was the 100-per-hour figure Scott had ex- pressed to Canales on November 14 about the Johnson Con- trols part. There is no evidence that Canales would have been unreasonable in also applying that expected rate to the Magnavox 342. Of course, Canales might have been sur- prised that he exceeded that goal by 200 when he trimmed 1100 or more parts a day. Then there is the matter of Scott’s arithmetic errors. I find these ‘‘errors’’ by Scott to have been intentional. Thus, they enhance the gap between the production rate of Canales and others by using a smaller divisor for others—a formula fraudulently calculated to enhance the production rate of oth- 1153 LOVEJOY INDUSTRIES 111 As the sage Amenemope instructed about 1250 B.C.: Do not make for yourself false documents, They are a deadly provocation. M. Lichtheim, 2 Ancient Egyptian Literature 146, 158 (1976, Univ. of Calif. Press). ers at the expense of Canales.111 Coming to Jose Meza, whom Scott especially compares (because Meza, like Canales later, was transferred from the drilling department) for his first day on the Johnson Controls job, Scott shrinks the divi- sor from the actual 6 hours to 4.5 hours. Regardless of whether Novak on the November 14 warning, or Balma on the November 21/22 warning suspension, was actually aware of, or had encouraged, the numbers fudging, Acme acted based in part on that distortion of the numbers. In short, I find that the Government established a prima facie case of unlawful motivation. Notwithstanding the prima facie case, I find that Acme would have warned and suspended Canales even absent his union activities. This is so, I find, because Acme—Novak, Balma, and Scott—believed that Canales was engaging in a work slowdown. Such a partial strike is unprotected activity. Acme’s overriding motivation, I find, was to discipline Marcial Canales because of that slowdown. To the extent there is any past practice at Acme for any- thing approaching a slowdown, Acme disciplined employees for it. Thus, Emilio Mora’s poor attitude resulting in low production earned him a written warning (R. Exh. 130–1) in January 1983. Similarly, Rosendo Lopez’ ‘‘negative attitude toward the job’’ figured in the written warning (G.C. Exh. 7–10) which he received in October 1985 for refusing to fol- low instructions. In June 1986 Tyrone Newson was warned (G.C. Exh. 5zz) for, in part, ‘‘Other,’’ making ‘‘less of the half of work of regular worker. He does half job in all jobs.’’ And Canales himself had received a written warning in No- vember 1985 for failing to following instructions, such fail- ure resulting in poor production. His warning concludes (R. Exh. 74): ‘‘If this should happen again, you will be sus- pended for 3 days without pay and another infraction of rules will call for immediate dismissal.’’ As for Novak’s own view toward Canales, the General Counsel elicited from Canales testimony that some 6 months after the election Novak, to an Acme customer in the pres- ence of Canales, praised Canales as being one of Acme’s fin- est workers. (13:1837–1838.) But it is clear that 7 months after that, when Novak received written evidence of Canales’ efforts to persuade die caster Armando Escheverria to engage in a slowdown, Novak’s opinion of Canales did a reverse. Finding that Acme has discharged its burden of dem- onstrating that it would have warned and suspended Marcial Canales on November 21/22, 1988, regardless of his union activities, I shall dismiss the complaint to the extent it al- leges that Acme violated 29 U.S.C. § 158(a)(3) by such dis- cipline. (b) Section 8(a)(5) The General Counsel merely begs the question in arguing that Acme’s November 21/22 discipline to Canales was harsher than its past practice of tolerating low production be- fore the election. Scott testified that in the past when an em- ployee’s production was below standard he would talk to the employee and thereafter he would see improvement. (2:331.) The General Counsel points to no specific evidence of low production which received no discipline. In any event, the slowdown by Canales was far more serious than mere neg- ligence, inattention, or laziness. It was unprotected conduct. Acme’s past practice, although lax at times, nevertheless had a history of imposing discipline for low production resulting from a poor attitude. The discipline here just addresses a more serious manifestation of a poor attitude. I shall dismiss complaint paragraphs 11(i)(1) and (2) in their entirety. 4. Jose Aguirre warned November 21, 1988, for low production a. Pleadings Complaint paragraph 11(f)(11) alleges that about Novem- ber 21, 1988, Acme issued a written warning to Jose Aguirre for low production. Acme admits this fact. The Government also alleges that the warning was a unilateral change (par. 11g) violative of Section 8(a)(5) of the Act (par. 14). Acme denies those allegations. (R. Exh. 36 at 4, 5.) There is no al- legation or contention that the warning to Jose Aguirre was unlawfully motivated. b. Facts Evidence is limited respecting this allegation, consisting of the warning (R. Exh. 111–1), 29 pages of production reports (attached to the exhibit but presumably not given to Aguirre), and Supervisor Scott’s brief, and general, testimony about the events and exhibit. Called by the General Counsel only as a rebuttal witness, Jose Aguirre addresses other points, not this allegation. Recall the evidence that following the cafeteria incident on October 21 Jose Aguirre apologized to Scott, saying that he had participated because he felt he had to do something to get a pay raise. (22:3352–3353, Scott.) I credited Scott over Aguirre’s denial. (26:3945.) Thereafter, Scott testified, Juan San Roman, Scott’s assistant, began reporting that Aguirre would stop working whenever he saw San Roman looking in his direction. Scott then observed that Aguirre repeated that conduct for him on several occasions, as if Aguirre did not care what his supervisor thought. (22:3353–3354.) Scott’s testimony is uncontradicted. Scott concluded that Aguirre was engaged in a slowdown. (22:3354.) Scott reviewed Aguirre’s production records. Ascertaining that Aguirre’s production had dropped substantially, Scott, apparently without orally cautioning or questioning Aguirre, issued a written warning, dated November 21, 1988, to Aguirre with box 13 ‘‘Other’’ checked. (22:3354–3356.) The text of Scott’s remarks section reads (R. Exh. 111–1): Jose’s production has taken a sharp drop since he was disciplined for taking an unauthorized break. He is hereby notified that the company will not tolerate a de- liberate slowdown of work. The 29 production sheets (R. Exh. 111–2 through 111–30) cover various dates from August 18 (R. Exh. 111–30) to Jan- uary 12, 1989 (R. Exh. 111–7). As he did on the production sheets of Canales, as described in the preceding section, Scott likewise computed the hourly rate and inscribed and circled it on certain sheets. (22:3357.) Scott perhaps was 1154 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 112 The production sheets are not attached in chronological order. Other than for three of the pages cited by the General Counsel (Br. at 134–135), the parties do not analyze the figures. 113 Scott never explains that he did so, but that is what is indicated by the numbers. Other than that, one would not know whether the scrap should be added or whether it should be considered part of the number for the shots. Recall that earlier Balma explained that ‘‘shots’’ is one cycle of the machine for die casting (1:118), and Scott testified that the scrap casting (on 1 day, at least, March 3) came from earlier in the line and that Aguirre did not create the scrap. (2:270–271; G.C. Exh. 21–9.) tired or rushed when he calculated these, for only a few of his averages are accurate, the others being substantially off. For the first sheet, August 18, 1988 (R. Exh. 111–30),112 Scott inscribed an average hourly rate of 85 working on part 660343 (the part number involved in the warning). That is a correct figure (within a fraction). To arrive at the figure, Scott obviously added the number in the ‘‘shorts’’ column and the number in the ‘‘scrap’’ column and divided by the hours.113 This was a partial day, but elsewhere it is clear Scott, as with Canales, bases his computation on a 9-hour formula. The next date (R. Exh. 111–29) appears to be October 7. Scott inscribed no number, but over the full shift Aguirre produced 770 shots on part 343 (660343). Scrap castings of 92 are shown. The total of those is 862, divided by 9 hours yields a rate (rounded) of 96 per hour. October 13 (R. Exh. 111–28), although probably a full day, does not show a clockout time. I shall not use that date. For Saturday, October 15, Scott inscribed an average of 81 for 345 shots and 87 scrap for hours of 7 a.m. to noon on part 343. (R. Exh. 111–27.) If Aguirre took no lunch period, then a 5-hour formula yields 86 (rounded), not 81. Scott inscribes a rate of 88 for the 7 hours (9 a.m. to 4:30 p.m.) on October 19 (R. Exh. 111–26), but 525 shots plus 76 scrap is 601, divided by 7 hours equals 86. On October 20 Aguirre worked on part 343 from 9 a.m. to 12:30 p.m., and Scott inscribed a rate of 88. (R. Exh. 111– 25.) With the 30-minute lunch period presumably part of that time, a 3-hour formula must be used. With 220 shots and 184 scrap totaling 404, the 3-hour divisor yields a quotient of 135! That seems high, and we have no testimony to ex- plain the jump. Two days later, working Saturday, October 22 from 6 a.m. to noon on part 343, Aguirre trimmed 600 shots and 149 scrap (749 total). (R. Exh. 111–24.) Scott did not inscribe a number, but a 6-hour divisor yields an hourly rate of 125. Scott inscribed a rate of 71 for Tuesday, October 25 on 535 shots, 105 scrap, totaling 640 over the full shift. (R. Exh. 111–2.) Scott’s 71 is correct using a 9-hour divisor. But his number of 70 for October 27 is off. On that day Aguirre, working the full 9.5-hour shift on part 343, trimmed 525 shots and scrap of what appears to be 737. That would produce a very high 151 rate. Multiplying Scott’s 70 by 9 hours produces 630—a number which cannot be derived from those on the production sheet. Possibly Scott assigned a lesser value to the 737 scrap, and added that number to the 525 shots to total 630. But that is speculation. The page for October 28 (R. Exh. 111–4) shows 530 shots, 71 scrap, and Scott’s inscribed 71 for the full shift. But 601 divided by 9 yields 67 (rounded), not 71. Scott’s 71 multiplied by 9 produces 639, a figure which cannot be de- rived from those on the page. I find that the correct hourly rate for October 28 is 67. October 31 (R. Exh. 111–5) was a full shift on part 343. Aguirre records 510 shots and 305 scrap (for a total of 815). Scott inscribed 68. But 815 divided by 9 is 90.55, or 91. Scott’s 68 multiplied by 9 matches nothing on the page. I shall use 91. Scott inscribed no rate on the November 1 record. Work- ing a full shift that day on part 343, Aguirre produced 550 shots and 57 scrap. (R. Exh. 111–23.) I compute that as an hourly rate of 67. Aguirre worked part 343 for another full shift on Novem- ber 2 trimming 505 shots and 187 scrap (totaling 692). Scott inscribed a rate of 67. But 692 divided by 9 is 77. Scott’s 67 multiplied by 9 hours produces 603—a number which matches nothing on the page. The page for November 4 contains a markover for the shots, and the copy is not clearly legible. The number is ei- ther 510 or 560 with 24 scrap for the full shift. (R. Exh. 111–22.) Scott inscribed no rate. At best the rate is 65, and 59 if the smaller number is used for the shots. Although Scott does not inscribe a number on any of the remaining pages, he asserts that he reviewed Aguirre’s pro- duction up to the date of the warning. (22:3357.) Using my calculations, and listing the rates above, the hourly rates for the 29 pages are: Date Rate Scott’s R. Exh. 111 8–18–88 85 85 30 10–7 96 — 29 10–15 86 81 27 10–19 86 88 26 10–20 135 88 25 10–22 125 — 24 10–25 71 71 2 10–27 151 70 3 10–28 67 71 4 10–31 91 68 5 11–1 67 — 23 11–2 77 67 6 11–4 59/65 22 11–7 75 21 11–8 64 20 11–9 64 19 11–18 69 18 11–29 92 17 12–1 81 16 12–6 82 15 12–7 82 14 12–8 80 13 12–9 109 12 12–20 92 11 1–9–89 110 10 1–10 82 9 1–11 88 8 1–12 84 7 If we disregard the rates over 100 in October 1988, we see that from August to mid-October Aguirre trimmed at a rate from 85 to 96. About the time of the October 26 strike his rate become erratic, bouncing from around 60 to 91, but fre- quently in the 60s through November 18, a Friday. On Mon- 1155 LOVEJOY INDUSTRIES 114 Employees date their first notice as being from January to around February 8 to 12. (3:547; 6:805; 8:1146; 11:1500, 1628.) Novak specifies January 30. (23:3520.) No timecards or other docu- ments were introduced on the point. I accept Novak’s more definite date of January 30. The General Counsel made no motion at the conclusion of the hearing to conform the pleading to the evidence under Rule 15(b) of the Fed.R.Civ.P. There is no fatal variance, however, because January 30 falls within the ambit of ‘‘on or about February 12.’’ In any event, the matter was tried by implied consent. day, November 21 Scott issued the warning. There is no pro- duction sheet in the record for that date. Thereafter, Aguirre’s rate returns to the 80s and 90s. The 109 recorded on December 9 was for a single hour. c. Discussion Fixing comparison percentage rates would be rather dif- ficult here in view of the variations, and lack of specific tes- timony. Nevertheless, using Scott’s four figures for August 18 through October 20 (my figures show a high rate on Oc- tober 20 which may or may not be correct), we see an aver- age rate of 86. The rates thereafter (Scott’s and mine) appear to be, roughly, about 20 points less—a drop of about 20 to 25 percent. Earlier Scott testified that he would consider a drop of at least 30 percent as indicating that a major problem had developed. (2:342.) By itself a production drop of about 25 percent certainly does not disclose any deliberate reduction by Aguirre. Scott testified, however, that when he looked in Aguirre’s direction Aguirre would stop working. This was in addition to iden- tical reports from San Roman. Aguirre does not address this testimony. Thus, the uncontradicted evidence shows deliberate conduct by Aguirre responsible for at least some of his drop in production. I therefore find that Scott, in issuing the November 21 warning to Jose Aguirre, acted on a reasonable and good-faith belief that Aguirre was slowing his work intentionally. Scott’s be- lief appears to have been well founded, for Aguirre’s produc- tion after the warning resumed an average rate of 89 for the 11 dates thereafter. Restricting the Government’s argument to the production reports, and contending that the warning departed from a past practice of not issuing written warnings for problems with quality or quantity of production, the General Counsel asserts that the warning was a unilateral change. That argument fails to address the evidence. Scott was faced here with what ap- peared to be a deliberate slowdown by Aguirre. As my sum- mary much earlier discussed, in September 1985, some 2 months after he arrived at Acme, Scott issued written warn- ings to several employees who were abusing breaktimes. Thereafter Scott had little problem until perhaps in the last few weeks before the October 1987 election. In any event, there is no evidence that before the election Scott tolerated conduct such as Jose Aguirre displayed here. Scott did not approach Aguirre and orally caution him that further work stoppages by him would result in a written warning. However, only a few days earlier Scott had given Aguirre a warning over the cafeteria incident and later that day Aguirre told Scott he had participated on a felt-need to get a pay raise. Although Balma (1:98–100) and Scott (2:336–337, 342) indicate that an employee is first orally cautioned or warned, Scott makes clear that each case is different (2:342; 21:3187) and that each supervisor decides whether a warning should be oral or written (21:3189). Balma confirms that the choice is within the supervisor’s discretion. (25:3897.) In view of the deliberate nature of Aguirre’s conduct, Scott’s issuance of a written warning to Aguirre without first cautioning him does not appear to constitute a departure from past practice. Similarly, it does not appear that Scott’s issuing a written warning in any event constitutes a unilateral change. Accordingly, I shall dismiss complaint paragraph 11(f)(11) and, to the extent they refer to this incident, para- graphs 11(g) and 14. K. Saturday Overtime Reduced 1. Pleadings Complaint paragraph 11(c) alleges that about February 12, 1988, Acme unilaterally (1) changed starting time for Satur- day work from 6 to 7 a.m. and (2) reduced Saturday breaktime from 20 to 10 minutes. The Government alleges that these changes were unlawfully motivated in violation of 29 U.S.C. § 158(a)(3) (complaint par. 11j and 14) and are unilateral changes in violation of 29 U.S.C. § 158(a)(5) (com- plaint pars. 11k and 14). Respondent Acme denies all allega- tions. (R. Exh. 36 at 3, 5.) 2. Facts There is no dispute that for years before January 1988 Acme worked overtime on many Saturdays during the year. Robert Novak testified that when he became president (in February 1987) Acme continued scheduling Saturday over- time every week in every department. (23:3517–3518.) Scott recalls Saturday work tapering off beginning with 1987. (23:3413–3414.) The Saturday schedule was, and for years had been, from 6 a.m. to noon with one 20-minute break at 10 a.m. (3:548; 6:806–807; 11:1500.) Novak asserts that the employees had two 10-minute breaks (23:3524), but the more credible version is the single 20-minute break. Some employ- ees, Novak testified, occasionally would work longer then 6 hours, and on some Saturdays a department might not be scheduled. (23:3519.) On January 30, 1988, Acme reduced the Saturday schedule by one hour by changing the starting time from 6 to 7 a.m. with the new 5-hour schedule, the single break was reduced from 20 to 10 minutes.114 Novak testified that Acme made the Saturday reductions because product demand was down from Acme’s two biggest customers, AT&T and Motorola. (23:3520–3524.) 3. Discussion a. Section 8(a)(3) The General Counsel’s one-sentence argument in favor of an 8(a)(3) finding is that the unilateral changes ‘‘had a tend- ency to discourage union activity and union support.’’ (Br. at 16.) Many things may ‘‘tend’’ to have that effect, but that does not make them illegal. The General Counsel apparently means that the changes were inherently destructive of impor- tant employee rights. See Morris, 1 The Developing Labor Law 195 et seq. (2d ed. 1983, ABA) and the 1982–1988 sup- plement at 96 et seq. (1989, ABA). The General Counsel does not articulate any theory in support of the Government’s 1156 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 115 Adair Standish Corp., 295 NLRB 985 (1989) (Adair IV) (in- creased frequency of asking for doctor’s notes from employees re- turning after illness not unlawful because not a material, substantial, and significant change); Chef’s Pantry, 274 NLRB 775 (1985) (re- quiring brief overtime to finish job not a change of established con- ditions). 116 Copies of the posted notices in evidence (G.C. Exh. 4) date from May 1985. All reflect the effective date to begin on a Monday. Contemporaneous entries on the employment records usually, but not always, give Sunday as the effective date. The discrepancy is imma- terial. argument. Rejecting that argument, I shall dismiss complaint paragraph 11(c) to the extent the complaint alleges that the changes violated Section 8(a)(3) and (1) of the Act. b. Section 8(a)(5) There is no dispute the Saturday changes were unilateral. The question is whether the unilateral changes were unlaw- ful. Citing cases,115 Acme argues that there is no violation because the changes were consistent with and comprehended within the past practice of a fluctuating schedule adapted to customer demand. (Br. at 283, 317.) Citing Venture Packaging, 294 NLRB 544 (1989), the General Counsel argues that Acme violated Section 8(a)(5) by the changes regardless of the asserted economic motive. The cited finding in Venture involved an admitted 1-hour change in shift starting time for the printing department em- ployees. The Company pleaded that the change was based on past practice. Hours of employment are a mandatory subject of bargaining, and the admitted change was found unlawful. The company’s past practice defense is not described. The Board also ruled that economic losses would not excuse the company from failing to offer to bargain with the union. Venture Packaging, 294 NLRB at 544 fn. 2. First, as changes of regular hours affect a mandatory sub- ject of bargaining, Venture Packaging, supra, so too do changes in rules affect overtime. Chef’s Pantry, supra at fn. 6. Second, Acme’s argument of no fixed overtime schedule because based on product demand is unavailing. In effect Acme argues that an employer which has laid off employees in the past during a business downturn can do the same after the employees vote in a union. But an employer may not make such changes respecting mandatory subjects. Adair II, 290 NLRB 317 (1988), and Adair III, 292 NLRB 890 fn. 1 (1989), enfd. on point, remanded on separate point 912 F.2d 854, 864, 867 (6th Cir. 1990). In any event, the Saturday schedule here had been 6 a.m. to noon, with a 20-minute break, for years. Fluctuations respected the number of em- ployees, or departments, but not the 6 a.m. to noon schedule. Finding that Acme’s unilateral change modified a manda- tory subject of bargaining, I find that, as alleged, Acme vio- lated Section 8(a)(5) and (1) of the Act by modifying the Saturday overtime schedule. I shall order Acme, on request of the Union, to rescind the change and to offer to bargain with the Union before implementing any new change. Acme must make whole employees adversely affected by the re- duced hours. L. Withholding Semiannual Pay Increases 1. Pleadings Complaint paragraph 11(b) alleges, ‘‘Since December 2, 1987, and continuing to date, Respondent has failed to grant its employees scheduled general wage increases.’’ Paragraphs 11(j) and 13 allege that this failure was unlawfully motivated in violation of 29 U.S.C. § 158(a)(3), and paragraphs 11(k) and 14 allege that it was a unilateral change in violation of 29 U.S.C. § 158(a)(5). (G.C. Exh. 1zz.) Acme denies the alle- gations. (R. Exh. 36.) As stated by the General Counsel (15:2074–2075), the De- cember 2, 1987 date appears to indicate the reach of the stat- utory limitations period. It is the charge for the second case, Case 13–CA–27788, which alleges a failure to make nor- mally scheduled wage increases. (G.C. Exh. 1c.) Filed May 27, 1988, that charge was served on June 2 (G.C. Exh. 1f)— a date 6 months after December 2, 1987. 2. Facts Because records document general wage increases back to 1980, I shall not dwell on testimony of the witnesses. The General Counsel’s employee witnesses testified that for many years Acme had given general, or across-the-board, pay in- creases twice a year. The amounts would vary, the months would vary, and frequently the employees, after 6 months had elapsed since the last raise, would have to ask about the general raise. Within 2 to 4 weeks they would receive it. (3:521–522, 527; 6:812–819, Valenzuela; 9:1351–1353, 1388, A. Aguilera.) Valenzuela recalls that one time before 1987 the manager said business was bad and there was no money for a pay increase. (6:820.) Copies of employment records (G.C. Exh. 3) and posted notices (G.C. Exh. 4) establish that Acme granted general pay increases effective on the following dates: 1980: 1–4 6–2 1981: 1–5 6–1 11–9 1982: 1–4 9–13 1983: 3–21 10–17 1984: 4–30 11–5 1985: 5–12116 12–2 1986: 6–30 1987: 2–16 9–28 ELECTION: 10–16–87 1988: None 1989: 1–2 The notices posted from May 1985 through February 1987 are by Leroy Hagner, the person preceding Robert Novak as Acme’s president. The October 2, 1987 notice (announcing the September 28 raise) and the December 16, 1988 notice (announcing the January 2, 1989 increase) are by Robert Novak. 1157 LOVEJOY INDUSTRIES 117 Escheverria disclaims being present. (17:2418.) Catherine Mooney, Acme’s controller since about 1983, testified that past general raises were given either as a per- centage or as a flat amount. (1:51, 59.) The implication is that the same rate or amount was given to all employees. That is not so for the September 28, 1987 wage increase. For that one the amounts varied by employees from 30 to 75 cents an hour because Novak, as he describes, tried to equal- ize some of the disparities. Novak first obtained the approval of Lovejoy’s president, Tony Girone. (24:3576–3580, 3678.) Because the September 1987 pay increase proved divisive with the employees, Novak abandoned that approach for the January 1989 increase, returning to the across-the-board for- mula. (24:3609, 3668–3669.) As shown in the table above, Acme granted no general pay increase in 1988. (24:3581.) The reason, Novak testified, is that sales were down and the profit margin did not justify a general increase. Sales losses were with Acme’s two big- gest customers, AT&T and Motorola. Business in 1987, Novak testified, was terrible, but expectations for 1988 were slightly better. (24:3581–3582.) Novak testified that no management person told him there was a time schedule for giving general wage increases, he has seen no document describing such a schedule, and, on becoming president in February 1987, he was unaware of any [particular] practice or history of granting pay raises to employees. (24:3574–3575.) Novak testified that the first time employees approached him in 1988 seeking a general pay raise occurred on Feb- ruary 29 (leap year), 1988. There were several such occa- sions thereafter. (23:3558, 3563; 24:3619–3620.) As dis- cussed earlier, even before this Novak had obtained from Acme’s attorneys, for Balma’s use in such situations, a pre- pared statement (R. Exh. 119) to use for reply to such re- quests. (23:3560.) The April 14, 1988 meeting, described earlier, is one of the subsequent occasions when employees asked Novak about a general pay increase. The April 14 meeting was held in Balma’s office. Novak testified that a group of employees had come to Balma asking to talk with Novak. The employ- ees, Novak testified, asked about a general wage increase. (23:3563.) Novak’s responded, as I summarized earlier. Marcial Canales testified that he first brought up the matter of reinstating Raymundo Aguirre. (13:1801) Novak testified that he began work on another general pay increase toward the end of August 1988 after some employ- ees had quit and the plant was experiencing difficulty in hir- ing employees at Acme’s new employee hire rate of $4 an hour. (24:3591, 3681.) After drafting a proposal (R. Exh. 120) which would grant a 4.5 percent general pay increase, averaging 31 cents per hour, and raise the starting hourly pay to $5, Novak presented his proposal to Girone by covering memo (R. Exh. 120) dated September 7. (24:3591–3593, 3682.) Girone rejected the proposal, Novak testified, because sales and the bottom line, the profit margin, did not justify it. As for employees leaving, Girone observed that shipments out were down so not as many employees were needed. (24:3593–3594, 3682–3684.) Going back to his drawing board, Novak drafted three slightly different options and a covering summary dated Oc- tober 25 (R. Exh. 121) which he submitted to Girone shortly after October 25. (24:3598–3599.) In the ensuing discussions with Girone, Novak finally recommended an across-the-board increase of 30 cents an hour which Girone, about mid-No- vember, approved. (24:3599–3601, 3683–3685.) Novak told only Balma. (24:3602.) Balma testified that Novak told him in late November that there would be a pay increase of 30 cents an hour in January 1989. (25:3820–3821, 3884.) Balma testified that the January 1989 pay raised officially increased the minimum hourly rate to $4.50, although infor- mally Acme had began hiring at that rate 3 to 4 months ear- lier. (25:3820, 3823–2824.) Indeed, Gustavo Navas is shown on Novak’s September 7 submission (R. Exh. 120 at 5) as being hired for part-time on August 4 at $4.50 (on the Octo- ber 25 proposal Novak had crossed him out as having quit, R. Exh. 121 at 5), and Jorge Gonzalez, another part-timer, is shown as being hired on June 13 at $4.50. (R. Exh. 120 at 5.) About Monday, December 5, 1988, Novak and Balma met with a group of employees in Balma’s office. (24:3602– 3603.) Mostly the same employees who met with Attorney Salzman on November 9, the group consisted of Antonio Sanchez, Armando Escheverria,117 Javier Carrasco, Jose Or- tega, Florentino Olivares, and Juan Ornelas. (17:2450–2451, 2466–2468; 24:3603; 25:3821–3822, 3881.) Balma testified that Sanchez requested to meet with him without giving a reason. Not until they came in and said they wanted to dis- cuss getting a pay raise did Balma know their purpose. Say- ing he had no authority over that, Balma called Novak who came to Balma’s office. Balma testified that he thought Novak might tell them of the pending January pay increase. (25:3881–3884.) When Novak arrived the employees asked for a (general) pay increase. Novak said he would talk to Walter Lovejoy and see what he could do. The employees also said they wanted to circulate a petition to get the Union out. Novak said they could not do that because it would be a year before anything could be done. (24:3604; 25:3822.) Balma testified he knew that a couple of the group did not want the Union, but he was surprised to learn that all of them—when they said they wanted the Union out—did not want the Union. (25:3882.) Although that may be so in light of the two (Olivares and Ornelas) who were not at the No- vember 9 meeting with Salzman, I find that Balma believed from the moment the employees walked in with Sanchez that the group generally was opposed to the Union. Novak testified that as of that December 5 date he was in the process of drafting a letter to employees respecting the January pay raise. He said nothing about it at the December 5 meeting. (24:3605, 3686.) The letter, dated December 16, 1988, not only was posted (24:3605), but also mailed to the homes of employees along with a copy of the December 16 notice (G.C. Exh. 4f) of the January 2, 1989 pay increase of 30 cents per hour. (Novak explains that the posted notice may have been enclosed in paycheck envelopes rather than with this letter. 24:3666.) Novak testified that he mailed it to employee homes because it was a letter to the families telling them of the raise and explaining why it was being given. He referred to the UE because he suspected the Union would comment as he describes. He wanted the families to be aware of what was happening at Acme and to hear that directly from him. (24:3666–3667.) The December 16 letter reads (R. Exh. 122): 1158 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 118 In its opening sentence Peat Marwick states that Lovejoy’s fis- cal year ended on April 30. 119 Note that the quarters do not coincide with calendar quarters. 120 As shown in Peat Marwick’s cover memo, net profit for the preceding five quarters, beginning with the quarter ending January 31, 1985, and ending with the quarter closing on January 31, 1986, are: 806.9, 1,130.8, 795.6, 1,086.9, and 787.3. DEAR ACME DIE CASTING PRODUCTION AND MAINTE- NANCE EMPLOYEES As most of you are aware, Acme’s business has not been good for some time. Customer orders are down; customer quality control demands have greatly in- creased, and the cost of producing our castings has in- creased. Even under these conditions a small group of employees have tried to further hurt the company’s business by purposely slowing down and performing less work. But, the company continues to strive to improve its business, and most employees continue to work hard to be as productive as possible. Due to Acme’s poor busi- ness, you have worked 15 months without a wage in- crease. Even though these poor business conditions con- tinue to the present, I do not feel you should have to wait any longer for a wage increase. Therefore, effective Monday, January 2, 1989, each production and maintenance employee will be given a 30¢ per hour wage increase by the company. The U.E. will likely tell you that this is not enough, that you de- serve more. However, the U.E. does not pay your wages; the U.E. is not responsible for keeping Acme competitive in the die casting industry; the U.E.’s criti- cisms will not sell one Acme casting or get or get us a new customer. Acme is giving the best wage increase that it can under the current business conditions. Only producing castings of the highest quality at the most competitive price will enable Acme to keep its cus- tomers and, thus, provide jobs for our employees. Thank you for your continued support and hard work. Merry Christmas, and Happy 1989. Sincerely /s/ R. J. Novak Robert Novak Confirming parts of the first and second paragraphs, Novak testified that business and profits were as bad at the end of 1988 as they were in previous months. (24:3609, 3680.) The employee group apparently proceeded with cir- culating the petition. Escheverria recalls that they presented it to management after the December 16 notice was posted announcing the January wage increase. (17:2404–2407, 2417.) Asked why the group continued circulating the peti- tion after the December 16 notice, Escheverria credibly ex- plained that it was done because the petition, besides request- ing a pay raise, also stated a desire not to be represented by the Union. (17:2420–2421.) As Joint Exhibit 1, the parties introduced a January 12, 1989 financial summary by the certified public accounting firm (Peat Marwick Main & Co.) for Lovejoy Industries, Acme’s parent.118 Expressly not an audit in accordance with generally accepted auditing standards, the financial statement was made, at Lovejoy’s request, ‘‘solely to assist you in con- nection with the current unfair labor practice trial before the National Labor Relations Board.’’ (Jt. Exh. 1–1.) Beginning with the quarter ending April 30, 1986, and continuing through the quarter ending October 31, 1988,119 the summary gives information for net sales, gross profit (net sales less cost of sales), and profit before interest and taxes (gross profit less expenses). Additionally, operating income before interest and taxes is listed for 16 quarters, from the quarter ending January 31, 1985, and extending through the quarter ending October 31, 1988. All figures given are in thousands. As the numbers for ‘‘operating income before in- terest and taxes’’ are identical to the numbers for ‘‘profit be- fore interest and taxes,’’ it is clear that these two categories are the same. For convenience, I shall refer to them, category three, as net profit. When one plots each of the three categories (counting op- erating income and profit before interest and taxes as one category) on a chart, one immediately sees that the graph lines are very similar in their ups and downs for the first two categories. For the third category, the lines—although taking the same climbs and falls—have more pronounced swings. Listing the three categories, in thousands of dollars, the num- bers are: Qtr. Ending Net Sales Gross Profit Net Profit 4–30–86 4,215.4 1201,366.3 801.8 7–31–86 4,028.5 1,012.5 532.3 10–31–86 2,716.5 381.0 6.8 1–31–87 3,100.1 628.9 233.7 4–30–87 3,549.0 1,065.8 406.0 7–31–87 3,119.2 694.5 166.3 10–31–87 2,855.3 518.1 104.8 1–31–88 2,860.5 594.6 216.4 4–30–88 3,841.8 845.0 278.9 7–31–88 3,265.0 652.0 197.8 10–31–88 3,228.3 591.8 193.0 From the foregoing we see that the first quarter, ending April 30, 1986, is the high point, with the low point reached two quarters later, on October 31, 1986. From there the graph lines and numbers climb through April 30, 1987, then drop again through October 31, 1987. The next quarter, end- ing January 31, 1988, has only a slight rise for categories one and two, while net profit doubles. The lines go up in the next quarter, ending April 30, 1988, with the third highest point for net sales. Over the next two quarters, ending Octo- ber 31, 1988, the lines and numbers move downward. I need to explain that the expenses deducted from gross profit (to yield profit/operating income before interest and taxes) consist of three items, as shown on the joint exhibit: selling, administrative, and corporate. As Peat Marwick ex- plains in its memo covering the figures, ‘‘Corporate expenses were allocated to the quarterly income statements proportion- ally on the basis of Acme’s quarterly sales to the quarterly sales of Lovejoy in total. We tested the mathematical accu- racy of this allocation and also agreed the total corporate ex- penses to the Lovejoy corporate general ledger.’’ That seems to say that Acme pays to Lovejoy the money shown as ‘‘cor- porate’’ expense. As noted earlier, gross profit is reached by deducting the cost of sales from net sales. Cost of sales is 1159 LOVEJOY INDUSTRIES shown to consist of material, labor, and overhead. The ‘‘labor’’ item presumably is for direct labor, principally, per- haps entirely, the production and maintenance employees. Recall Novak’s testimony that 1987 was a ‘‘terrible’’ year because of lost business from Acme’s two major customers, AT&T and Motorola. (24:3582.) If 1987 did not start with the net sales shown for the corresponding quarter in 1986 (quarters ending April 30), neither did 1987 have the steep decline 1986 reflects. Net profit was substantially less in 1987 than in 1986. Nevertheless, Acme granted two general pay increases in 1987 (although Novak’s September 28, 1987 increase was not an across-the-board increase). Sales and gross profits climbed in early 1988, as did net profit. The three categories dropped off, but not so much by October 31 as the corresponding figures for October 31, 1987. 3. Discussion a. Section 8(a)(3) Arguing that Acme refused to give general wage increases in 1988 in order to punish its employees for voting in the Union, the General Counsel advances several points. The first point is the alleged threats. I have found merit to some of these allegations. Observing that there is no evidence Acme, during 1988, ever pleaded poverty to its employees, the General Counsel contends that Acme’s economic defense, presented at the hearing, is a fabrication designed to mask its true and unlaw- ful motive. Recall that in the second half of 1986, when Acme’s numbers, especially its net profit, hit bottom, Acme gave no pay increase. This apparently was the occasion when, as Valenzuela testified, the manager informed employ- ees that there would be no pay raise because Acme had no money. In 1988 Acme told employees it could not discuss a general pay raise because of the objections case with the Union. Not once did management advance poor sales or low profits as a reason. The implication is that economics had nothing to do with it. I count this factor as weighing in favor of a prima facie case by the Government. Arguing disparate treatment, the General Counsel points to the fact that when Canales, Valenzuela, and the other prounion employees wanted to discuss a pay raise they were rebuffed, according to Novak and Balma, with the substance of a prepared statement (R. Exh. 119). But when the antiunion group came to Balma’s office on December 5 seek- ing a pay raise, Balma obligingly called Novak who scooted right over to Balma’s office to meet with the group. Not stopping with providing his personal curb service to the group by going to Balma’s office, Novak also assured the group that he would call no less than Walter Lovejoy, the owner of Acme’s owner, to press for a wage increase. (24:3604.) At no point did either Balma or Novak mention the substance of Respondent’s Exhibit 119, the prepared statement. I count this disparity as a factor weighing in favor of the General Counsel’s prima facie case. Acme’s financial records, the General Counsel argues, do not support Respondent’s claim of financial difficulties. Thus, Acme’s second poorest quarter of record, ending Octo- ber 31, 1987, is the same quarter Novak, and supposedly Lovejoy’s Girone, chose to give a pay raise effective Sep- tember 28. The numbers for early 1988 were not great, but the net profit in the quarter ending January 31, 1988, doubled that of the previous quarter and exceeded that of the quarter ending July 31, 1987. The next quarter, ending April 30, 1988, was higher still. And then when Acme, on December 16, did announce a pay raise, effective January 2, 1989, Acme had just experienced its worst quarter since the one ending October 31, 1987. Thus, the two pay raises under Novak bear little correlation to Acme’s financial data. Agree- ing with the General Counsel, I find that this factor weighs in favor of a prima facie case. Finally, the General Counsel contends that the timing of Girone’s approval was backdated by Novak, and that such approval really did not occur until after the December 5 meeting between Novak–Balma and the antiunion group in Balma’s office. That requires some review. Recall that Novak said Girone gave his approval in mid-November. (24:3685.) One problem with that date is that Novak kept changing the date. Initially he said it came about the first part of December, before he met with the employee group. (24:3602–3603.) Only later in his testimony did he date Girone’s approval as being about a month before Novak’s December 16 letter (R. Exh. 122) to all production and main- tenance employees discussing the 30-cent pay raise effective January 2, 1989. (24:3685.) And recall that as of the Decem- ber 5 meeting Novak supposedly had about finished drafting the December 16 letter. Agreeing with the General Counsel, I find that Novak fraudulently backdated Girone’s approval, and that the true date of Girone’s approval came shortly after Novak met with the antiunion group on December 5. At that meeting Novak learned that the group was circulating a petition to oust the Union. Although a decertification petition was premature, Novak and Girone, I find, recognized that the antiunion fac- tion could be supported by giving a general pay increase. Moreover, I find it unlikely that it took Novak (or Acme’s attorneys) 2 or 3 weeks to draft the December 16 letter. In- stead, a timeframe of drafting the letter between December 5 and 16 seems far more plausible. Finally, I do not believe Novak when he testified that Girone gave his approval in mid-November. Instead, I find that such approval came when Novak reported to Girone concerning the December 5 meet- ing with the antiunion group. Based on these factors, I find that the General Counsel has established prima facie that a motivating factor in Acme’s re- fusing to give pay increases earlier was the employees’ vot- ing for the Union in the October 1987 election. Acme defends on the basis of (1) it had financial difficul- ties and (2) it was forced to grant a pay raise because the company found it difficult to attract replacements for em- ployees who had left Acme. (Br. at 376.) The first ground I have found to be false. As for any difficulty in attracting replacements, Girone himself suggested to Novak in Septem- ber that Acme did not need any replacements because Acme was not shipping as much, and Novak agreed. (24:3682– 3683.) No doubt a beginner’s rate of $4 per hour would make hiring difficult. But Acme managed with that rate in 1987, when two raises were given. Moreover, by June 1987 Acme informally had begun hiring at $4.50 an hour. I find no merit in this ground, and I do not believe Novak. Acme, I find, has failed to carry its burden of demonstrating that it would have withheld its normal two general pay increases in 1988 even had there been no union activities or union election victory in October 1987. Finding that Acme with- 1160 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 121 The increase Novak devised for September 28, 1987, was not an across-the-board increase of a single amount. Novak testified that amounts varied from 10 to 75 cents. (24:3579–3580.) Jesus Arrendondo, for example, received 15 cents. (G.C. Exh. 3w.) held pay increases in 1988 in violation of Section 8(a)(3) of the Act, I shall order Acme to make whole its employees by granting the two pay increases retroactive to effective dates in 1988 and in amounts as described in the next section. I have not overlooked Novak’s testimony that Acme did not grant any general pay increases in 1987 or 1988 to its 17 or 18 managers and supervisors (although it did give five merit increases in 1987 and possibly some in 1988). (24:3582–3583, 3590.) The year 1987 is irrelevant because raises were given that year to production employees. Simi- larly, what Acme did respecting its managers and supervisors in 1988 also is irrelevant. Whatever hardships Acme forced its management to suffer in 1988 was simply a byproduct of Acme’s discrimination against the bargaining unit. b. Section 8(a)(5) Respecting the unilateral change allegation, it is clear that Acme had no fixed date, or even a fixed month, when it granted pay raises, and when given the amounts varied. Nev- ertheless, with two exceptions, Acme has given general pay raises twice a year from at least 1980. The exceptions are 1981 when Acme gave three general increases and 1986 when it gave only one. In 1986 Acme’s numbers dropped ex- tremely low in the quarter ending October 31, 1986. An employer may not unilaterally alter terms and condi- tions of employment without affording the union representing its employees a meaningful opportunity to negotiate in fact. NLRB v. Katz, 369 U.S. 736, 743 (1962). Pay increases or adjustments which are established and regular events are conditions of employment not subject to unilateral change. Central Maine Morning Sentinel, 295 NLRB 376 (1989); Venture Packaging, 294 NLRB 544. Raises which are spo- radic, with timing and amounts not given at traditional or regular intervals, are not conditions of employment. See Ithaca Journal-News, 259 NLRB 394, 395 (1981) (an 8(a)(3) case but the stated principle applies here). Our case approaches the borderline between the two posi- tions. Twice a year, almost without exception, Acme has granted general pay raises. They have ranged from about 15 cents per hour on February 15, 1987 (in the second quarter following Acme’s terrible quarter ending October 31, 1986), and 30 cents per hour on January 4, 1982, and January 1, 1989.121 Most of the increases were for 20 or 25 cents per hour. (G.C. Exh. 3; R. Exh. 116.) Timing of the general increases was static at the 1980 start of this period, but slippage began in 1982. Thus, for the first 2 years, 1980 and 1981, the increase was granted almost on the same dates in January and June. That held for January 1982, but the previous June timetable slipped to September. Another month was added to the slippage for both raises in 1983, and that pattern of slippage continued for 1984, 1985, and for the first (and only) raise of 1986. A shift back to the early part of the year occurred with 1987’s initial raise, and the first one under Novak was made in late September. For 1988 there was no raise, and 1989 saw the first raise back to the first week of January. A moving pattern, yes. However, except for 2 unusual years, one good (1981) and one bad (1986), Acme has given general wage increases twice a year of about 25 cents an hour. The ‘‘twice yearly’’ raises in All American Gourmet, 292 NLRB 1111 (1989), came in June and December ‘‘based on budgetary considerations.’’ Withholding of the December raise there was found to be an unlawful unilateral change. The remedial order fashioned by Administrative Law Judge Howard I. Grossman, and adopted by the Board, required the employer to make whole its employees ‘‘by granting them raises retroactively to December 1986 in such amounts as normally would have been granted to them with interest.’’ All American, supra. Fixing the month in 1988 that a first raise would have been given seems to be an exercise of relative certainty given the historical pattern. Considering the pattern, particularly the 1987 and 1989 return of the first raise to the early part of the year, the month of February 1988 appears to be a reason- able approximation. Similarly, the amount of 25 cents ap- pears to be the figure most nearly matching both the histori- cal pattern and the graph lines of Acme’s economic numbers. Thus, business started on an upswing for the quarter ending January 31, 1988 (Novak testified that expectations for 1988 were slightly better than for 1987) and continued to climb through the next quarter. On these considerations, I find that the February timing and 25 cents per hour were relatively fixed benefits which Acme unilaterally withheld. Turning now to the second raise for 1988, I find the month of July to be a reasonable approximation of when Acme would have granted a general pay increase but for Acme’s withholding that increase unilaterally. Acme had just finished its best quarter in 2 years for net sales and gross profit. However, the numbers started a slight decline in the quarter ending July 31, 1988. Perhaps a 20-cent raise is indi- cated. Despite the less than rosy numbers (recall Novak’s testimony that the numbers were no better at the end of 1988 than at the beginning, 24:3609, 3680), Acme somehow was able to set 30 cents for January 1989. That 30 cents matched Acme’s previous January 1982 high for an across-the-board pay increase. In light of that economic ability and corporate will, it seems reasonable to find, as I do, that in July 1988 Acme would have granted its employees a general increase of 25 cents per hour but for its unilateral withholding of that benefit. On these considerations I find that, as alleged, Acme uni- laterally withheld granting across-the-board pay increases to all bargaining unit employees in violation of 29 U.S.C. § 158(a)(5) and (1). But for Acme’s unlawful unilateral change of normally scheduled benefits, it would have granted general pay increases of 25 cents per hour in February 1988 and again in July 1988 to all employees of the bargaining unit. Fixing the exact effective dates, in relation to Acme’s pay periods, is a matter to be resolved at the compliance stage. I shall order Acme to make the bargaining unit em- ployees whole, with interest. CONCLUSIONS OF LAW 1. Acme Die Casting, a Division of Lovejoy Industries In- corporated (Acme) has violated Section 8(a)(1) of the Act by certain economic threats and by discriminatorily telling em- ployees they would not be permitted to wear T-shirts bearing union insignia; Section 8(a)(3) and (1) of the Act by unilater- ally changing work rules in 1987, by various disciplinary 1161 LOVEJOY INDUSTRIES 122 If no exceptions are filed as provided by Sec. 102.46 of the Board’s Rules and Regulations, the findings, conclusions, and rec- ommended Order shall, as provided in Sec. 102.48 of the Rules, be adopted by the Board and all objections to them shall be deemed waived for all purposes. warnings and suspensions, and by failing to grant normally scheduled general wage increases to bargaining unit employ- ees in February and July 1988; and Section 8(a)(5) and (1) of the Act by the same unilateral changes in work rules, dis- ciplinary warnings and suspensions, unilaterally reducing Sat- urday overtime beginning January 31, 1988, unilaterally withholding the February and July 1988 wage increases, and the October 24, 1988 suspensions of Rodolfo Banales and Nicolas Valenzuela. 2. Acme’s unfair labor practices affect commerce within the meaning of Section 2(6) and (7) of the Act. REMEDY Having found that Respondent Acme has engaged in cer- tain unfair labor practices, I find that it must be ordered to cease and desist and to take certain affirmative action de- signed to effectuate the policies of the Act. Acme has discirminatorily and unilaterally warned and suspended employees, changed or eliminated various work privileges, and withheld general pay increases of 25 cents from bargaining unit employees in February and again in July 1988, and has unilaterally suspended Nicolas Valenzuela on October 24, 1988. Consequently, Acme must revoke the discipline imposed, remove the disciplinary notices from its files, restore the previously enjoyed work privileges, make whole, with interest, the suspended employees and those who lost earnings as a result of the reduction in Saturday overtime and, retroactive to effective dates in February and July 1988, grant all employees then members of the bargaining unit gen- eral pay increases of 25 cents per hour, with interest. In making whole the employees for their loss of pay or other benefits suffered as a result of their suspensions and lost Saturday overtime, and sustained by employees as a re- sult of Acme’s withholding the two 25-cent per hour pay in- creases, Acme must compute the money due on a quarterly basis respecting the suspensions, less any net interim earn- ings, as in prescribed F. W. Woolworth Co., 90 NLRB 289 (1950), plus interest as computed in New Horizons for the Retarded, 283 NLRB 1173 (1987). Interest on the money due each employee for the withheld pay increases shall be computed in the same manner. The effective dates of the February and July 1988 general pay increases shall be deter- mined at the compliance stage in light of Acme’s payroll pay periods. As the language of many Acme employees is Spanish, I shall direct that the notice to employees be posted in English and in Spanish. Bacardi Corp., 296 NLRB 1220 fn. 2 (1989). On these findings of fact and conclusions of law and on the entire record, I issue the following recommended122 ORDER The Respondent, Acme Die Casting, A Division of Lovejoy Industries Incorporated, Northbrook, Illinois, its offi- cers, agents, successors, and assigns shall 1. Cease and desist (a) Threatening employees that they will not receive ex- pected wage increases because of the Union. (b) Discriminatorily telling employees they will not be permitted to wear T-shirts bearing union insignia inside the plant. (c) Refusing to recognize and bargain with the Union, United Electrical, Radio & Machine Workers of America (UE), as the exclusive bargaining representative of the em- ployees in the bargaining unit described below. (d) Changing or eliminating working conditions of bar- gaining unit employees because they support the Union, or unilaterally doing so without affording the Union notice and opportunity to bargain concerning any proposed changes. (e) Withholding twice-yearly general wage increases from unit employees because they voted-in or support the Union, or unilaterally doing so without affording the Union notice and an opportunity to bargain over such withholding. (f) Issuing written warnings to and suspending employees in order to discourage them from supporting the Union or doing so unilaterally without affording the Union notice and opportunity to bargain over such disciplinary action. (g) Issuing written warnings to employees and suspending them pursuant to unilateral changes made by Acme in estab- lished working conditions. (h) In any like or related manner interfering with, restrain- ing, or coercing employees in the exercise of the rights guar- anteed them by Section 7 of the Act. 2. Take the following affirmative action necessary to ef- fectuate the policies of the Act. (a) On request, bargain with the Union as the exclusive representative of the employees in the following appropriate unit concerning terms and conditions of employment and, if an understanding is reached, embody the understanding in a signed agreement: All full-time and regular part-time production and maintenance employees employed by Acme Die Cast- ing at its facility located at 3610 Commercial Avenue, Northbrook, Illinois 60062; excluding all office clerical employees, technical employees, tool and die makers, managerial employees, professional employees, con- fidential employees, and guards and supervisors as de- fined in the Act. (b) Restore the working conditions prevailing before Octo- ber 16, 1987, by which employees (1) in cold weather could start their cars before the end of their shift, (2) were per- mitted to warm food in cafeteria microwaves before breaks and lunch, (3) were permitted to consume food and bev- erages, including coffee, at work stations to the extent such practice does not constitute a safety hazard, and (4) had free access to restrooms without the restriction of first giving no- tice to their supervisor. (c) Remove from Acme’s files any reference to the follow- ing disciplinary actions: Date Action Employee 11–11–87 written warning Antonio Ramirez 11–11–87 written warning Fidencio Olivares 1–7–88 written warning Mauricio Aguirre 1162 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 123 If this Order is enforced by a judgment of a United States court of appeals, the words in the notice reading ‘‘Posted by Order of the National Labor Relations Board’’ shall read ‘‘Posted Pursuant to a Judgment of the United States Court of Appeals Enforcing an Order of the National Labor Relations Board.’’ Date Action Employee 10–24–88 written warning Nicolas Valenzuela 10–24–88 written warning Jose Aguirre 10–24–88 written warning Rodolfo Banales 10–24–88 written warning Marcial Canales 10–24–88 written warning Mario Garcia 10–24–88 written warning Francisco Mombela 10–24–88 written warning Hugo Paz 10–24–88 written warning Nicolas Valenzuela 10–24–88 suspension Rodolfo Banales 10–24–88 suspension Nicolas Valenzuela (d) Retroactive to February 1988, restore twice-yearly gen- eral wage increases for bargaining unit employees, and con- tinue to grant semiannual general wage increases until Acme and the Union agree otherwise, until they bargain to a good- faith impasse, or until the Union refuses to bargain in good faith over that condition of employment. (e) Make whole all bargaining unit employees, including those no longer employed, for any monetary loss they suf- fered by Acme’s withholding, as found, the 25-cent-per-hour general wage increase in both February 1988 and again in July 1988 in the manner set fort above in the remedy section of this decision. (f) Restore the working condition prevailing before Octo- ber 21, 1988, when employees were permitted during worktime to get coffee in the cafeteria and return with it to their work stations. (g) Rescind the October 24, 1988 written notice which, in effect, eliminated the pre-October 21, 1988 established privi- lege of employees being permitted during worktime to get coffee in the cafeteria and return with it to their work station. (h) Restore the Saturday overtime schedule of 6 a.m. to noon, with one 20-minute break, Acme unilaterally reduced the hours and breaktime on January 31, 1988. (i) Make whole with interest, in the manner set forth in the remedy section of this decision, all employees who have suffered a loss of pay as a result of Acme’s January 21, 1988 unilateral reduction of hours in the Saturday overtime sched- ule, including employees no longer employed. (j) Remove from its files any reference to the unlawful warnings and suspensions and notify the employees in writ- ing that this has been done and that the disciplinary actions will not be used against them in any way. (k) Preserve and, on request, make available to the Board or its agents for examination and copying, all payroll records, social security payment records, timecards, personnel records and reports, and all other records necessary to analyze the amount of backpay due under the terms of this Order. (l) Post in English and in Spanish at its facility in North- brook (Chicago), Illinois, copies of the attached notice marked Appendix.123 Copies of the notice, on forms pro- vided by the Regional Director for Region 13, after being signed by the Respondent’s authorized representative, shall be posted by the Respondent immediately upon receipt and maintained for 60 consecutive days in conspicuous places in- cluding all places where notices to employees are customar- ily posted. Reasonable steps shall be taken by the Respond- ent to ensure that the notices are not altered, defaced, or cov- ered by any other material. (m) Notify the Regional Director in writing within 20 days from the date of this Order what steps the Respondent has taken to comply. IT IS FURTHER ORDERED that the complaint is dismissed in- sofar as it alleges violations of the Act not specifically found.