Rockwell Automation/Dodge

ROCKWELL AUTOMATION/DODGE 547 Rockwell Automation/Dodge (Formerly Reliance Electric) and International Association of Ma- chinists and Aerospace Workers, AFL–CIO. Case 10–CA–29818 January 26, 2000 DECISION AND ORDER BY MEMBERS FOX, LIEBMAN, AND BRAME On November 3, 1998, Administrative Law Judge Howard I. Grossman issued the attached decision. The Respondent filed exceptions and a supporting brief. 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 brief and has decided to affirm the judge’s rulings, findings, and conclusions only to the extent consistent with this Decision and Order.1 A. Facts The sole issue in this case is whether the Respondent violated Section 8(a)(3) and (1) of the Act by discharging Gregory Silvers on October 16, 1996. In February 1990, Silvers began working as a grinder operator for the Re- spondent, which manufactures high-quality ball bearing products at its Rogersville, Tennessee facility. Compo- nents of these products include inner and outer rings, which must be produced to a tolerance of 100 millionths of an inch. Grinder operators, such as Silvers, who manu- facture these rings, work in “cells” consisting of two banks of three machines each. Such operators are re- quired to use a “waveometer” to test samples of rings every 2 hours for “race waviness.” In the spring of 1995, the International Association of Machinists and Aerospace Workers, AFL–CIO (the Un- ion), began an organizing campaign at the Respondent’s plant. Silvers testified that he passed out union authoriza- tion cards, which he kept in his toolbox, but did not pass out cards during “company hours.” On or after July 12, 1995, Silvers visited Supervisor Carl Ogle at Ogle’s house. Ogle told Silvers that he had just been discharged by Plant Manager Terry Singleton in connection with a charge of sexual harassment against him. Ogle also stated that, while in Singleton’s office, he had noticed a list on Singleton’s desk that had Silvers’ name on it with the notation “for union activity” next to it. Ogle told Silvers to “watch [his] behind” because “they were after [him].”2 1 In light of our disposition of this case, we find it unnecessary to pass on the Respondent’s Exception 1. 2 While the judge placed this conversation as occurring on about July 1, 1995, the parties stipulated that Ogle was a supervisor until July 12, 1995. Silvers was the only witness regarding his conversation with Ogle. According to Silvers’ account, Ogle told Silvers about Ogle’s discharge and the list on Singleton’s desk in the same conversation. Accordingly, this conversation could not have occurred until on or after July 12, after Ogle had been discharged and was no longer a supervisor. About 3 days later, Silvers told Foreman Eddie Merrill that he wanted to take a vacation day but was afraid he would be fired because he had heard that he was on what he termed a “hit list” in Singleton’s office. Merrill told him that he could not be fired for taking a vacation day, and Silvers took a day of vacation. A few days later, Sil- vers was called to Supervisor Ken Ball’s office. Ball told Silvers that he had heard that Silvers had said that there was a list on Singleton’s desk, and Ball wanted to know who had told Silvers that there was such a list. Silvers responded that he could not give this information, but Ball replied that, if he valued his job, he would tell. Fi- nally, Silvers stated, untruthfully, that his former wife had told him about the list. On August 28, 1995, about a month after Silvers was questioned by Ball, Silvers and two other employees working on Cell 5 received disciplinary counseling for producing outer rings that failed inspection “due to op- erators’ lack of attention to mach[ine] and SPC checks.” Silvers previously had been counseled on May 3, 1994, for failure to make process control checks, and he also received a May 8, 1995 warning for talking to employees in other departments when he was not on break.3 The Union’s organizing campaign resulted in an elec- tion held November 12, 1995, which the Union lost.4 About a week before the election, Silvers began wearing a union hat to work. No other events pertinent to this case took place until almost a year later. When Silvers, who worked on shift B, arrived at work at 7 a.m. on September 30, 1996,5 he found that Donald Davis, the operator on his work cell “team” on the preceding shift, shift D, had set up produc- tion of a new outer ring and had produced 30 rings. Sil- vers proceeded to produce 174 additional outer rings on the same machine during his own 12-hour shift. Silvers testified that he had intermittent problems with the “dresser unit” that produced the rings for the entire shift and that he stopped several times and made adjustments to the machine. He also testified that he told Davis, when Davis returned at 7 p.m. for the evening shift, that he was having trouble with the dresser unit. Nevertheless, Sil- vers filled out a work report form6 showing that he had performed “race waviness” tests seven times during the shift and that all the tests had shown the outer rings to be acceptable. Silvers testified that he performed these tests at the start of his shift and at approximately 2-hour inter- vals thereafter. 3 The General Counsel did not allege that any of the discipline re- ceived by Silvers prior to his discharge violated the Act. 4 Silvers testified that the result was “about two to one for the com- pany.” We take administrative notice that the representation proceeding in question was Case 10–RC–14662 and that the tally of ballots showed 140 votes for the Union, 327 against the Union, and 5 challenged bal- lots. 5 All dates hereafter are in 1996 unless otherwise indicated. 6 This form is referred to by the parties variously as an “ISO sheet,” an “RQC-228” form, an “RQC” sheet, or an “SQC” form. 330 NLRB No. 82 DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 548 After the shift change at 7 p.m., Davis, who replaced Silvers, continued production of the outer rings but soon found that the machine was making bad rings. Davis and set-up person Benny Brooks checked the rings and found that all 15 that Davis had completed on that shift were bad. They then “purged” additional rings that were still in the machine, approximately another 15 rings. All the rings that had been produced on the machine since Davis had set up the production run on his prior shift were be- ing placed into a “stubben” or basket. This included 30 rings Davis had produced on his prior shift, the 174 rings that Silvers produced on his shift, and the approximately 30 rings that Davis produced at the start of his succeed- ing shift—a total of approximately 234 rings. Davis and Brooks then reset the machine. No additional rings were placed in the stubben. The stubben was set aside outside the foreman’s office. At the direction of Shift Supervisor Ken Davis, lead grinder operator Billy Joe Frost performed waveometer tests on all the rings in the stubben. The tests showed that, of the approximately 234 rings, 189 were bad and approximately 45 were good. The rings shown to be good were the ones in the bottom of the pile in the stub- ben, from which the Respondent inferred that they were the first ones that had been produced. The good rings that were identified were removed from the stubben, and the bad rings left in it. When Silvers’ supervisor, Perry Price, who had been on vacation on September 30, returned to work on October 1, Ken Davis informed him about the bad rings. Price checked about 15 to 20 of the bad rings with a testing device called an “Anderometer” to deter- mine whether they could be salvaged. He determined that they could not be salvaged and would have to be scrapped. Ultimately, quality control inspectors also con- cluded that all of the 189 bad rings would have to be scrapped and they issued a “scrap ticket” for the rings. The rings were discarded on October 15.7 When he reported for work on October 1, Silvers saw that the stubben containing the rings that he had pro- duced was sitting in front of the supervisor’s office. Frost told Silvers that all but about 50 of the rings that had been made on that production run were bad. Price asked Silvers what had caused the bad rings to be produced. Silvers replied that he did not know what had happened because all the rings he checked were good. According to Price’s uncontroverted testimony, Silvers did not ques- tion the determination that the rings were bad and did not ask to check the rings himself. Price subsequently informed Production Manager Alan Annis about the large number of bad rings resulting from 7 The judge’s erroneous statement, in sec. II,B,2,b of his decision, that the rings were discarded on September 30, is contrary to Price’s uncontroverted testimony that the rings were discarded on October 15, and contrary to the uncontroverted testimony of Silvers, Price, and Production Manager Annis that they saw or examined the rings on various dates after September 30. the production run. Annis went to the grinding depart- ment and inspected some of the bad rings. At Annis’ request, Price checked five or six rings on the waveome- ter and showed him why the rings were bad. Annis and Price reviewed the operator log sheet and other forms in which Silvers made entries during his September 30 shift and saw that he had recorded that he had made the re- quired “race waviness” tests and that all the rings tested had passed the tests. Neither Price nor Annis could think of any possible explanation of how all the “race wavi- ness” tests could be satisfactory and yet so many bad rings could be produced. They reached the conclusion that Silvers must have falsified his test result form. In the meantime, the Respondent’s human resources manager, John Pinkerton, who conducted periodic meet- ings with various groups of employees, had just held a meeting with grinding department employees on shift D, the shift that followed shift B on which Silvers worked. At the meeting, an employee commented that operator Adam Williams had bid off of shift D to get away from Silvers, who had been his shift B team counterpart, be- cause Silvers left his machine in poor condition when going off shift. Pinkerton held a meeting with Annis and Plant Man- ager Singleton about morale problems on shift D, at which Williams’ problem with Silvers was discussed. Annis, in turn, informed Pinkerton about the issue of Silvers’ apparent falsification of test records regarding the production of bad rings. Thereafter, at Annis’ re- quest, Williams’ supervisor, Jerry Turner, asked Wil- liams about his experience working as a grinder operator following Silvers. Williams told Turner that Silvers did not maintain the machines during his shift, which caused Williams extra work, and that Williams experienced some “crashes” caused by parts being loaded backwards. Williams’ statement was put in written form, which he signed. Annis talked to other D shift employees and found that another one, Scott Stapleton, was also un- happy about his experience working after Silvers. There- after, on October 14, Annis wrote a memorandum to Singleton and Pinkerton recommending that Silvers be discharged for intentionally falsifying company quality documents and for “displaying disruptive, anti-team be- havior.” After receiving Annis’ memorandum recommending Silvers’ discharge, Pinkerton reviewed Silvers’ personnel file in which he noted, among other things, Silvers’ Au- gust 28, 1995 written warning for lack of attention to machine and SPC checks. Pinkerton testified that, when an employee has engaged in conduct for which he might be terminated (such as intentional falsification of quality control documents), the Respondent reviews his entire personnel file, including prior warnings, “to determine a pattern or situation.” In other circumstances, the Respon- dent does not consider warnings issued more than a year prior to the incident in question. Pinkerton also examined ROCKWELL AUTOMATION/DODGE 549 what actions the Respondent had previously taken in similar situations. He found records of two former em- ployees, Kent Seals and Alan Bernard, who, like Silvers, previously had engaged in falsification of records and had been the subject of prior corrective actions for qual- ity or performance. In each instance, the employee had been terminated. Pinkerton then decided to talk to Silvers himself. On October 15, Price and Annis met with Silvers in Annis’ office. Annis asked Silvers to explain how 189 bad rings could have been produced in light of his form indicating that he had made seven sets of acceptable waveometer tests. Silvers stated that he had experienced problems with the dresser machine for the entire shift but that the rings tested good on the waveometer neverthe- less. Shortly thereafter, Silvers met with Pinkerton, as well as Annis and Price, and they essentially repeated the discussion. Silvers never denied that he had made bad rings. Indeed, Silvers testified that Pinkerton said to him, “I see that you have run some bad parts,” and Silvers replied, “Yes, I suppose I have.” Moreover, Silvers con- ceded that, if roles were reversed and he were viewing the situation from the personnel manager’s point of view, he would probably think that the test result document had been falsified. Pinkerton suspended Silvers for the rest of the day and told him that they would review his situation with Plant Manager Singleton. They then met with Sin- gleton, who agreed that termination was appropriate, in view of the falsification of records, Silvers’ prior warn- ings, and the adverse reports about Silvers that they had received from shift D employees. The next morning the same supervisors again met with Silvers. They told him that they had concluded that he had falsified his waveometer reports and that it had been determined that he should be terminated. Pinkerton told Silvers that, under the Respondent’s procedures, he could elect “peer review” of the termination decision or he could resign. If he resigned, the Respondent would not oppose any claim for unemployment benefits. Silvers chose to resign. B. Analysis and Conclusions The judge found that Silvers’ discharge was unlawful. He found that Silvers had engaged in protected union activity by passing out union cards during the election campaign and by wearing a union hat during the week before the November 12, 1995 election. He noted that just-discharged Supervisor Ogle had told Silvers in July 1995 that his name was on a list on Plant Manager Sin- gleton’s desk with the notation “for union activity” next to it and that Silvers shortly thereafter had been ques- tioned by Supervisor Ball about how he had learned about the list. The judge also found that there was evi- dence of disparate treatment in that other employees who had falsified records had not been discharged. The judge found evidence of discriminatory motivation in that the Respondent had not followed its own policies when it took into account the August 1995 warning in deciding on Silvers’ discharge, because the warning had been is- sued more than 12 months before the discharge. In reject- ing the Respondent’s contention that it would have dis- charged Silvers in any event regardless of his union ac- tivity, the judge found that the Respondent had failed to establish that Silvers had produced bad parts or had falsi- fied records, and he found unconvincing the negative reports about Silvers from other employees. In sum, the judge concluded that, under Wright Line,8 the General Counsel had established a prima facie case and that the Respondent had failed to rebut it, as all the reasons that the Respondent advanced for discharging Silvers were, in the judge’s view, pretextual. The Respondent argues that the General Counsel failed to present substantial evi- dence to support a prima facie case of unlawful discrimi- nation and, even assuming that a prima facie case was presented, the overwhelming evidence establishes that Silvers falsified company records and his termination was consistent with company practice. The judge found that the General Counsel met his bur- den of establishing by a preponderance of the evidence that Silvers’ union activity was a motivating factor in the Respondent’s decision to discharge him. The judge based this finding, in part, on the following facts. Before Sil- vers’ discharge, recently terminated supervisor, Carl Ogle, told Silvers that Silvers’ name was on a list on the plant manager’s desk, along with the notation, “For un- ion activity.” Ogle told Silvers to “watch your behind” because “they’re after you.” When Supervisor Ken Ball later confronted Silvers about this list and asked Silvers who told him about it, Silvers said that he could not tell him, and Ball told him, “if you value your job, you’ll tell me.” Our concurring colleague would conclude that these facts and other factors cited by the judge do not support his conclusion that Silvers’ union activity was a motivat- ing factor in his discharge. We find it unnecessary to reach this issue, because, even assuming arguendo that the General Counsel has met his burden, we find that the Respondent has met its Wright Line burden of showing that it would have discharged Silvers even in the absence of his protected activity.9 Contrary to the judge, we find that the preponderance of the evidence establishes that the Respondent reasonably believed that, during his shift on September 30, Silvers produced a large number of bad rings and falsified his report form by indicating that he had conducted seven sets of “race waviness” tests, all of 8 251 NLRB 1083 (1980), enfd. 662 F.2d 899 (1st Cir. 1981), cert. denied 455 U.S. 989 (1982), approved in NLRB v. Transportation Management Corp., 462 U.S. 393 (1983). 9 As we find that the Respondent has established its Wright Line de- fense, we find it unnecessary to pass on our concurring colleague’s position that the General Counsel failed to establish a prima facie case. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 550 which showed the tested rings to be acceptable.10 Indeed, Silvers himself conceded to Human Resources Director Pinkerton that, if he were viewing the situation from Pinkerton’s perspective, he would conclude that the re- port had been falsified.11 Moreover, uncontroverted evi- dence shows that the Respondent previously had dis- charged two employees, Seals and Bernard, who, like Silvers, engaged in falsification of records and had re- ceived prior disciplinary actions.12 Consequently, we find that the Respondent has demonstrated that, under the circumstances, it would have discharged Silvers even in the absence of his protected conduct.13 The judge concluded that the Respondent’s asserted reasons for discharging Silvers were pretextual. We dis- 10 See Goldtex, Inc., 309 NLRB 158 fn. 3 (1991) (Wright Line de- fense established by employer’s reasonable belief that employee forged magazine subscriptions), enfd. mem. 16 F.3d 409 (4th Cir. 1994); GHR Energy Corp., 294 NLRB 1011, 1013 (1989) (Wright Line defense established by employer’s reasonable belief that employees engaged in serious misconduct), enfd. 924 F.2d 1055 (5th Cir. 1991) (unpub- lished); American Thread Co., 270 NLRB 526, JD at 532 (1984) (“I do not find it necessary to conclude that Ramey actually removed an oil can from Respondent’s premises. It is only necessary to decide whether Respondent’s actions against Ramey were predicated on a genuine belief that he was involved in the unauthorized removal of Respon- dent’s property.”) 11 The judge, in the last paragraph of sec. II,B,2,b of his decision, discounted this admission as based on a hypothetical question that, implicitly, he found unsupported by the facts. We do not agree with his paraphrase of the question and find the evidence supports the facts set forth in the question. 12 Contrary to the judge, we find that the fact that the Respondent did not introduce into evidence the records that Seals and Bernard had falsified provides no basis for disregarding Pinkerton’s uncontroverted testimony about their discharges. Thus, unlike the judge, we do not find that the Respondent’s discharge of Silvers constitutes disparate treat- ment. See Merillat Industries, 307 NLRB 1301, 1302–1303 (1992) (no disparate treatment found where employee’s discharge was consistent with employer’s treatment of others who had committed similar of- fenses). The 1997 examples, cited by the judge in sec. II,B,5 of his decision, of employees Fields, Newman, and Hatley receiving disci- pline short of discharge for falsification of documents, all involved employees who, unlike Silvers, had received no prior disciplinary ac- tions. See Hoffman Fuel Co., 309 NLRB 327, 329 (1992) (disparate treatment not shown by employer’s failure to discipline employee who, unlike alleged discriminatee, had prior “clean record”). Additionally, the discipline of employee LeBlanc, also cited in sec. II,B,5 of the judge’s decision, for allowing his machine to run unattended in De- cember 1996 was for a different infraction than that committed by Silvers. Moreover, all four instances of discipline on which the judge relied occurred subsequent to Silvers’ discharge. At the time of Silvers’ discharge, however, Seals’ and Bernard’s prior discharges for falsifica- tion of records were the Respondent’s precedent for this type of infrac- tion. 13 Unlike the judge, we do not find that the Respondent’s reviewing Silvers’ entire personnel record, including a warning issued 14 months before his falsification of test records, was contrary to the Respondent’s “12-month rule” and, thus, discriminatory. As shown by Pinkerton’s testimony, when the Respondent is determining what action to take regarding an infraction that may result in discharge, the Respondent’s policy is to review the employee’s entire personnel record. Falsification of documents is an infraction that may result in discharge, even though discharge is not imposed in all instances. The Respondent’s practice of not considering prior disciplinary actions more than 12 months old applies to lesser infractions. agree. The judge focuses, in part, on the Respondent’s determination that the 189 rings were not merely bad, but were “scrap” and not salvageable. The question of whether the rings were “merely” bad or were so defective that they were scrap is irrelevant. Silvers was discharged for falsifying his test result form. The Respondent had concluded that, since so many bad rings had been pro- duced, Silvers’ report, which represented that he had repeatedly conducted the required tests and that all tested rings were acceptable, could not be true. The Respon- dent’s determination that Silvers had falsified his test result form did not turn on whether the bad rings he pro- duced were reworkable or were scrap. Thus, the judge’s focus on these latter points is entirely irrelevant. The judge also finds fault with the Respondent’s de- fense in that neither Frost, who performed waveometer tests on all the rings in the stubben,14 nor the quality con- trol inspectors who ultimately determined that 189 of the rings were scrap, testified about their examination of the rings.15 Nevertheless, Price and Annis gave ample, un- controverted testimony about the testing process and the finding that 189 rings were bad.16 Price, as Silvers’ su- pervisor, and Annis, as manufacturing manager, were familiar with the facts concerning the examination of the rings and had been personally involved in this process. Additionally, the Respondent introduced into evidence the “scrap ticket” documenting that the 189 rings, valued at $1,373.80, had been determined to be scrap. Equally important, at no point in the entire process did Silvers ever take issue with the Respondent’s determination that the rings were bad. In fact, Silvers testified that he admit- ted to Pinkerton that he had made bad parts. Thus, unlike the judge, we find, in these circumstances, no shortcom- ing in the Respondent’s choice not to present every pos- sible witness who was involved in examining the rings in question. We also do not agree with the judge’s characterizing as “scarcely comprehensible” and “unpersuasive” Price’s testimony concerning how the Respondent determined which rings in the stubben were produced by Silvers and 14 While the judge noted that there was no evidence that Frost’s job functions included quality control inspection, there was no question raised at the hearing of Frost’s competence to perform waveometer tests. Moreover, as all grinder operators were required to perform waveometer tests and Frost was a lead grinder operator, it may rea- sonably be inferred that he was competent to perform such tests and that occasionally testing parts produced by other operators was within the scope of his responsibility. 15 None of these individuals were called as witnesses, except for Quality Control Inspector Gardner, who testified briefly about other matters. 16 In addition to testing 15 to 20 rings on October 1 or 2, Price ran “forms scans” on five rings at a later date. The scans showed that the five tested rings were not reworkable. The judge failed to distinguish between the two groups of tests and mistakenly inferred that, because the forms scans showed that five rings could not be reworked, there were 10 to 15 other rings that could have been reworked. This conclu- sion is not supported by the evidence. ROCKWELL AUTOMATION/DODGE 551 which were those of Davis.17 Price’s testimony clearly indicates that the Respondent believed that the rings that were produced first were placed in the bottom of the stubben and that additional rings, as they were produced, were piled on top of the rings that already had been made. Thus, the rings at the bottom were the ones that had been made first, i.e., the ones made by Davis when he started the new production run of rings prior to the start of Silvers’ shift on September 30. It was these rings on the bottom that the Respondent determined were good. Although all the rings in the stubben presumably would mix together to some extent, we find reasonable the Respondent’s assumption that the ones at the bottom of the stubben were the first ones produced, i.e., the ones made by Davis. Moreover, even assuming that the rings initially produced by Davis could not be identified, Sil- vers made 174 of the 234 rings placed in the stubben, and 189 of the 234 rings (or 81 percent) were found to be bad. Simply based on these figures alone, it was reason- able for the Respondent to conclude—contrary to Sil- vers’ waveometer reports—that most of the rings that Silvers made on September 30 were bad.18 In sum, con- trary to the judge, we find that the Respondent reasona- bly concluded that Silvers had falsified his test results report, based on the disparities between the rings that were produced and the information reflected in Silvers’ report. Accordingly, we reject the judge’s conclusion that the Respondent’s discharge of Silvers violated the Act. We find that the Respondent met its Wright Line burden by establishing that it would have discharged Silvers even in the absence of his protected activity. Therefore, we shall dismiss the complaint. ORDER The complaint is dismissed. MEMBER BRAME, concurring. I join in the dismissal of the complaint, but only for the following reason. I find that the Respondent’s discharge of grinder operator Gregory Silvers did not violate Sec- tion 8(a)(3) and (1) of the Act because the General Coun- sel failed to establish a prima facie case of union dis- crimination. Therefore, unlike my colleagues, I do not consider whether the judge correctly concluded that the Respondent would not have discharged Silvers in the absence of his protected union activity. Briefly, the salient facts are as follows. A union orga- nizing campaign began at the Respondent’s facility in the spring of 1995. Silvers testified that his only activity in 17 Price’s testimony in question appears in the text following fn. 16 of the judge’s decision. 18 Indeed, even assuming that Silvers had made all of the good rings, he still would have produced 129 bad rings, as only about 45 of the rings in the stubben were good. Thus, even viewing the facts in the light most favorable to Silvers, 74 percent of Silvers’ production still would have consisted of bad rings. support of the organizing campaign consisted of passing out cards. He testified that he did not pass out cards dur- ing “company hours” but that he kept the cards in his tool box and that employees “knew that they could get them whenever they wanted.” There is no evidence con- cerning specifically where, when, how often, or to whom Silvers passed out cards in this fashion. Additionally, there is no evidence that the Respondent was aware of Silvers’ card activity. On about July 12, 1995, freshly discharged former Su- pervisor Carl Ogle told Silvers that he had seen on Plant Manager Terry Singleton’s desk a list with Silvers’ name on it followed by the notation “for union activity.” Ogle cautioned Silvers to “watch [his] behind” because “they were after [him].” A few days after Silvers mentioned this list to Foreman Eddie Merrill, Supervisor Ken Ball questioned Silvers concerning who had told him about this list. In his testimony, Silvers characterized this list as a “hit list.” About a month later, on August 28, 1995, Silvers re- ceived disciplinary counseling for producing outer rings that failed inspection “due to operators’ lack of attention to mach[ine] and SPC checks.” The disciplinary counsel- ing was not alleged to be unlawful. During the week before the November 12, 1995, elec- tion, Silvers wore a union hat at work. The Union lost the election, and there is no evidence that the Respondent demonstrated union animus or made any unlawful anti- union remarks prior to the election. There is also no evi- dence that the Respondent’s employees engaged in any union activity subsequent to the election. Thereafter, on October 14, 1996, Production Manager Alan Annis recommended to Singleton and Human Re- sources Director John Pinkerton that Silvers be dis- charged.1 Silvers was discharged on October 16, 1996, based on his falsification of company records on Sep- tember 30, 1996. Thus, Silvers’ discharge did not occur until 15 months after Silvers was told that his name was on what he characterized as a “hit list” and more than 11 months after the election campaign had ended and all union activity had ceased. In the absence of any open hostility to the Union on the part of the Respondent, I find that the General Coun- sel has failed to show that the Respondent’s discharge of Silvers in 1996 was sufficiently linked to Silvers’ limited 1995 union activity described above.2 First, the record is simply devoid of any specifics concerning Silvers’ card distribution, except for the fact that he kept union au- thorization cards in his toolbox at work. Indeed, the evi- dence does not establish that Silvers had more than token 1 Both Annis and Pinkerton began their employment at the Respon- dent’s Rogersville plant after the November 12, 1995 election had been held and the Union’s organizing campaign had subsided. 2 See Wright Line, 251 NLRB 1083 (1980), enfd. 662 F.2d 899 (1st Cir. 1981), cert. denied 455 U.S. 989 (1982), approved in NLRB v. Transportation Management Corp., 462 U.S. 393 (1983). DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 552 involvement in the distribution of union cards. Nor does the record show that the Respondent knew of Silvers’ participation in card distribution, however minimal. Sec- ond, Silvers’ only other union activity was the wearing of a union hat during the week before the election. As em- ployees other than Silvers also wore union hats, there is no evidence that Silvers’ wearing a union hat made him stand out.3 Thus, the evidence shows that Silvers was a peripheral player in the union campaign and, hence, an unlikely target for antiunion retaliation.4 Additionally, in August 1995, shortly after the period in which Silvers assertedly passed out union cards and only weeks after Ogle told him that he was on what Sil- vers called a “hit list,” Silvers engaged in producing outer rings that failed inspection “due to operators’ lack of attention to mach[ine] and SPC checks.” Yet, Silvers received only disciplinary counseling for this infraction. That the Respondent failed to seize on this infraction, committed in the midst of Silvers’ union activity, as a basis to discharge Silvers underscores the Respondent’s lack of antiunion motivation in its dealings with Silvers. The Respondent’s failure to take this opportunity to im- pose harsher punishment or discharge Silvers makes even more improbable the notion that the discharge of Silvers over a year later was due to antiunion animus. Indeed, Silvers’ October 1996 discharge was far too at- tenuated from the November 1995 election and the spring and summer 1995 union card-signing campaign to have reasonably been carried out for antiunion reasons.5 The judge’s principal grounds for finding Silvers’ dis- charge motivated by antiunion animus appear to be Ogle’s statement that there was a list on Singleton’s desk bearing Silvers’ name with the notation “for union activ- ity” after it, Ball’s questioning Silvers concerning how he learned about this list, and Ogle’s warning to Silvers to “watch [his] behind” because “they were after [him].” The judge, without analysis, accepts Silvers’ characteri- zation of Singleton’s list as a “hit list.” However, Ogle, 3 According to his uncontroverted testimony, General Foreman Jerry Turner saw some employees other than Silvers wearing union hats prior to the election. 4 Cf. Asarco, Inc. v. NLRB, 86 F.3d 1401, 1409 (5th Cir. 1996) (no inference of antiunion animus arose from fact that discharged employee was an aggressive union president) denying enf. in relevant part 316 NLRB 636 (1995). 5 See General Electric Co. v. NLRB, 117 F.3d 627, 638 (D.C. Cir. 1997) (discharge of employee 2 years after Board election and 10 months after he testified at a Board hearing “did not follow the election or proceedings so closely as to be remotely suspect”), affirming in relevant part 321 NLRB 662, 674–677 (1996); MECO Corp. v. NLRB, 986 F.2d 1434, 1437 (D.C. Cir. 1993) (8-month gap between employ- ees’ last concerted activities and their discharge “strongly militates against any inference of anti-union motivation”), denying enf. 304 NLRB 331 (1991); Geo. V. Hamilton, Inc., 289 NLRB 1335, 1340– 1341 (1988) (employee’s participation on union negotiating team found “too remote in time to be linked to” his layoff 11 months later); Rock- land-Bamberg Print Works, Inc., 231 NLRB 305. 306 (1977) (em- ployee discharge was “too remote in time” from Board election held 5 months earlier), enfd. mem. 566 F.2d 1173 (4th Cir. 1977). who told Silvers about the list, did not state that it was a “hit list.” Rather, the “hit list” terminology was Silvers’ characterization. More importantly, the evidence does not show the list to have been used as a “hit list.” During a union election campaign, it is not unlawful or objec- tionable for a company’s management internally to pro- ject the likely outcome of a possible election, assuming that no unlawful interrogation or polling is used for this purpose. Where such management projections are not communicated to employees, it cannot be contended that they interfere with, restrain, or coerce employees in the exercise of their Section 7 rights. To make such a projec- tion, an employer might well go down a list of employ- ees in the unit, write next to each name a guess regarding how that employee would vote, and then tally the hypo- thetical votes. Additionally, an employer might try to guess each employee’s view on union representation in order to determine where its election campaign should be focused. Thus, the mere fact that Singleton had a list with Silvers’ name on it and the notation “for union activity” next to his name falls well short of proof that the Re- spondent had singled out Silvers for retaliation. Addi- tionally, that the list was not a “hit list” is further borne out by the fact that the only other employee whose name was reported to be on the list, John McCoy, was not dis- charged and remained employed by the Respondent at the time of the hearing 2-1/2 years later. Nor does Ball’s questioning of Silvers concerning who told him about the list show antiunion motivation. The Respondent, no doubt, considered the list to be confiden- tial.6 It was reasonable, then, for the Respondent, having learned that the confidentiality of the list had been breached, to wish to ascertain how this breach had oc- curred. Thus, Ball’s questioning Silvers regarding who told him about the list reflects merely the Respondent’s legitimate concern for maintaining the confidentiality of management documents. Finally, Ogle’s statement to Silvers that he should “watch [his] behind” because “they were after [him]” also fails to show antiunion motivation. In relying on this statement, the judge overlooked the fact that it cannot be attributed to the Respondent, as Ogle, who had just been discharged, was no longer a supervisor at the time that he made this statement.7 Consequently, Ogle’s statement was nothing more than the view of a disgruntled, recently discharged former supervisor. Thus, I find that Silvers’ discharge was too remote in time from his limited union activity and appearance on Singleton’s list to support the inference that Silvers’ un- 6 Ogle saw the list on Plant Manager Singleton’s desk in his office. There is no evidence that Singleton discussed the list with Ogle or circulated the list to other individuals. 7 See Electronic Data Systems Corp., 305 NLRB 219, 239 (1991), enfd. in relevant part 985 F.2d 801 (5th Cir. 1993) (three dispatchers, assumed to be supervisors, could no longer speak for the employer after it had terminated them). ROCKWELL AUTOMATION/DODGE 553 ion activity was a motivating factor in the Respondent’s decision to discharge him for falsifying his test results report.8 Therefore, I concur in dismissing the complaint, because the General Counsel has failed to establish a prima facie case of a violation of Section 8(a)(3) and (1) of the Act. Elaine Robinson-Fraction, Esq., for the General Counsel. James C. Hoover, Esq. (Ford & Harrison), of Atlanta, Georgia, for the Respondent. Paul Randolph, Business Representative, of Chattanooga, Ten- nessee, for the Charging Party. DECISION STATEMENT OF THE CASE HOWARD I. GROSSMAN, Administrative Law Judge. The charge was filed on December 9, 1996, by International Asso- ciation of Machinists and Aerospace Workers, AFL–CIO (the Union). Complaint issued on November 21, 1997, and alleges that Rockwell Automation/Dodge (formerly Reliance Electric) (Respondent or the Company), on October 16, 1996, discharged Gregory Silvers because of his assistance to the Union, and in order to discourage protected activity by other employees, in violation of Section 8(a)(3) and (1) of the National Labor Rela- tions Act (the Act). A hearing on these matters was held before me in Rogers- ville, Tennessee, on February 12 and 13, 1998. Subsequent to the hearing, it appeared that the transcript omitted testimony of several of Respondent’s witnesses. Efforts to secure the miss- ing pages were unsuccessful. On August 12, 1998, counsels for the General Counsel and Respondent submitted a Motion to Administrative Law Judge for the Admission of Proposed Stipulations into the Record and to Close the Record. The mo- tion includes stipulations as to the testimony which was con- tained in the missing transcript pages, and General Counsel’s agreement that this testimony need not be submitted to cross- examination. The motion further states the General Counsel’s agreement that an affidavit of one of Respondent’s witnesses be accepted as his testimony on direct, together with General Counsel’s cross-examination. Finally, the parties urge that I receive the stipulations and cross-examination in evidence, and close the record. On August 28, 1998, I issued an Order Reopening Hearing, Receiving Stipulation of the Parties, and Reclosing Hearing. Thereafter, the General Counsel and Respondent submitted briefs. After consideration of the entire record, including my observation of the demeanor of the witnesses, I make the fol- lowing FINDINGS OF FACT I. JURISDICTION Respondent is a Tennessee corporation with an office and place of business in Rogersville, Tennessee, where it is engaged in the manufacture of rollers and ball bearings. During the 12- month period preceding issuance of the complaint, Respondent sold and shipped goods valued in excess of $50,000 directly to customers located outside the State of Tennessee. Respondent 8 See New Otani Hotel & Garden, 325 NLRB 928 (1998) (the Gen- eral Counsel failed to meet his burden of proving that the discharge of three housekeepers were unlawful). is an employer engaged in commerce within the meaning of Section 2(2), (6), and (7) of the Act. The Union is a labor organization within the meaning of Sec- tion 2(5) of the Act. II. THE ALLEGED UNFAIR LABOR PRACTICES A. Gregory Silvers’ Employment History and Union Activity Silvers was employed as a grinder operator in February 1990. In October 1993, Supervisor Ken Ball issued a Depart- ment Update, in which he expressed the Company’s thanks to Silvers “for all his support and efforts he has contributed to- ward making the grinding department a safer and more efficient place to work.” Various improvements made by Silvers, with the “teamwork and cooperation from his co-workers,” are cited.1 On May 3, 1994, Silvers received a counseling for fail- ure to make “process control checks.”2 The Steelworkers began an organizing campaign in March or April 1995, but “phased out” according to Silvers. He found a Steelworker card on his toolbox and gave it to Supervisor Carl Ogle. Organizing efforts continued by the Machinists, the Un- ion herein, and a Board election was held on November 12, 1995. The Company distributed “Vote No” buttons and held employee meetings. Several supervisors asked Silvers what he thought of these buttons, and he gave ambiguous replies. Silvers passed out authorization cards for the Machinists. On May 8, 1995, he received a warning for talking to other employees in other departments when not on break.3 On about July 1, 1995, Supervisor Carl Ogle4 told Silvers that he, Ogle, had been in the office of Plant Manager Terry Singleton con- cerning a charge of sexual harassment against Ogle. Ogle re- ported to Silvers that, while in Singleton’s office, he saw a list of names on the plant manager’s desk. Silvers’ name was on the list followed by the words, “For Union activity.” “Watch your behind,” Ogle told Silvers—“they’re after you.”5 Ogle told management that Silvers was not involved in the Steel- workers’ campaign. About 3 days later, Silvers told Foreman Eddie Merrill that he wanted to take a day off (to which he was entitled). How- ever, somebody had told him that he was on a “hit list” in Sin- gleton’s office, and he was afraid of being fired. Merrill told him to take the day off. A few days after these events, Silvers was called to the office of Ken Ball, an admitted supervisor. Ball told Silvers that he had said that there was a list on Terry Singleton’s desk. Silvers agreed that this is what he had heard. “Who told you?” Ball asked Silvers. Silvers replied that he could not answer that question. Ball responded, “If you value your job, you’ll tell me.” Silvers finally told Ball that his ex-wife had told him that this is what she had heard, and that she was worried about Sil- vers’ ability to make child-support payments. Neither Singleton nor Ball testified in this proceeding. Sil- vers was a credible witness, and I accept his uncontradicted testimony as to these events. The next incident involves disputed evidence as to whether Respondent gave Silvers disciplinary counseling on August 28, 1 G.C. Exh. 9. 2 R. Exh. 7. 3 R. Exh. 4. 4 The parties stipulated that Ogle was a supervisor from January 1, 1990, until July 12, 1995. 5 The Company later discharged Ogle. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 554 1995, for running deficient parts on August 22, 1995. This alleged discipline was relied upon by the Company more than a year later when Silvers was discharged. The disciplinary form was first shown to Silvers by the General Counsel. It is hand- written and dated August 28, 1995, with the name “Greg Sil- vers” at the top, and the signatures “P. Price,” and “Ken Davis” at the bottom. In between is a description of the alleged defi- ciencies of “216 OR’s” (outer rings), by employees in “Cell 5.” Silvers averred that the first time he saw this document was the night before the hearing.6 During the events leading to his discharge over a year later, according to Silvers, Respondent showed him a typed document alleging that employees John McCoy and Benny Brooks had been written up for producing bad “15 inner rings.” Supervisor Perry Price told Silvers that he had also been written up in 1995, and Silvers denied it. No such typed document was introduced at the hearing. The handwritten document introduced by Respondent7 was shown to Price. He contended that he spoke to Silvers about this al- leged deficiency (in 1995), but denied that the writing on it was his. Instead, Price stated, Supervisor Eddie Merrill wrote the warning because two other employees were being given identi- cal warnings. On direct examination, Silvers testified that he could not have been included in this warning, because he was in “Cell 4” at the time, whereas the warning indicated “Cell 5.” However, on cross-examination Silvers was shown his timecard for Au- gust 22, 1995, which indicates that he was in “Cell 5,” and produced outer rings.8 Silvers then agreed that he had been working in “Cell 5” with at least one of the employees who was warned on that date. On the basis of this evidence, I conclude that Silvers was warned on August 28, 1995, for producing bad parts on August 22, 1995. Silvers became worried that Respondent was going to fire him. In order to make certain that it could not deny knowledge of Silvers’ allegiance to the Union, he started wearing a union hat the week before the election on November 12, 1995. The election was won by the Company. B. Silvers’ Discharge 1. The manufacturing and testing process Alan Annis was hired in December 1995, and was manufac- turing manager at the time of these events. He described the Company as a manufacturer of “mounted ballbearing prod- ucts.” According to Annis, the Company is known as a “Cadil- lac” in its field, and produces a high quality product. It is also a high volume producer, and makes 3-1/2 million bearings annu- ally. These products ultimately form parts of other machines, and the allowable tolerance for error is extremely fine. Thus, the tolerance for “roundness” of “inner” and “outer” rings is 100 millionths of an inch, less than the diameter of a human hair. The Company uses a variety of machines and procedures to check the accuracy of the manufacturing process. The sus- pected defect in this proceeding is “waviness.” If a ring has humps and valleys in its circumference, it will rattle and soon wear out in use. This defect can be described by imagining a 6 Since Silvers could not identify the document, the General Counsel declined to submit it as an exhibit. Whereupon Respondent did so, marking it R. Exh. 11. 7 R. Exh. 11. 8 R. Exh. 1. circular ring being changed to a straight path, like a road with hills and valleys which would provide an uneven ride. In Re- spondent’s product, these variations are so minute that they cannot be observed by the naked eye. Accordingly, a machine called a waveometer is used. It measures “waviness” in two ways. The first is an oscilloscope which shows the undulations of the waves, or “high” and “low” bands, as the parties describe them. The second method is a needle reading a gauge, with readings showing variations in millionths of an inch. Neither of these processes produces a film or other record of its results, according to Plant Manager Annis. Deficient parts are some- times “reworkable,” according to the plant manager. If not they are discarded, and sold as “scrap.” Annis asserted that the Company has a “scrap” rate of one to three percent. The scrap rate of a particular employee is rarely ascertained. Silvers testified that deficient parts are produced by every employee. Parts having minor defects are put back into the “bin,” rerun, and made into useable parts. This happens “every night,” according to Silvers. If a part cannot be reworked, an inspector sprays it with paint, and it is rejected. Because of the volume of production, the Company uses a sampling method for testing products. The sampling is done by a grinder operator, like Silvers. A company production sheet lists the various tests and their frequency, on which the grinder operator makes entries. Thus, “race waviness” is to be tested every 2 hours.9 If several sample parts produced during this 2- hour period test out satisfactorily, the entirety of that group is presumed to be adequate. Silvers worked in a “cell” consisting of two banks of three machines each. One bank produced outer rings and the other inner rings. When a machine produces a group of parts, the operator places them in a “pan.” A “washer” then puts them in a “buggy,” and they are ultimately placed in a basket, called a “stubben.” “Quality control” employees then determine the adequacy of the parts. 2. Silvers’ production on September 30, 1996 a. Summary of the evidence Silvers worked a 12-hour shift, from 7 a.m. until 7 p.m. An- other grinder operator, Donald Davis, worked from 7 p.m. until 7 a.m. Silvers testified that, when he arrived at 7 a.m. on Sep- tember 30, 1996, Donald Davis had already started a run of a new outer ring order, and had completed 30 parts. The machine was cut off, and Davis had five parts on a table. He stated that there were streaks in all the parts. Silvers said he thought it was the “dresser unit.” The parties stipulated that Donald Davis’ testimony if transcribed would have been that the ma- chine was working properly. Davis left, and Silvers checked the dresser unit. He then ran a few parts, and tested them on the waveometer, which was located in another room. The parts tested satisfactorily. Silvers started production again, but “streaks” soon appeared on the parts. Silvers loosened the dresser unit, and then produced five parts. He checked them on the waveometer, and they were again satisfactory. Silvers started production again, and, after awhile, the dresser unit again malfunctioned. This happened several times during Sil- vers’ shift. He recorded on his log sheet that he worked on the dresser unit.10 Silvers also testified that on 7 occasions he tested several samples out of 30 produced, every 2 hours on the 9 Jt.. Exh. 1; G.C. Exh. 7. 10 R. Exh. 3. ROCKWELL AUTOMATION/DODGE 555 waveometer, a total of about 12 to 20 parts. They were satis- factory.11 The log for September 30 by Silvers shows 30 “pre- vious” parts (run by Davis and in the machine), and 174 parts run by Silvers, for a total of 204 parts.12 Silvers testified that he told Davis when the latter returned for the evening shift that he was having trouble with the dresser unit. The parties stipulated that Davis’ testimony, if transcribed, would have been that Silvers said the parts were “running good,” and that he had been checking his bands. The parties stipulated that Davis’ testimony, if transcribed, would have been that he ran 15 parts, checked them, and found low bands. He and leadman Benny Brooks took the 15 parts out of the machine and placed them in the stubben where the parts from Silvers’ production had been placed. Davis’ testi- mony would have been that his parts and Silvers’ were “segre- gated.” Benny Brooks’ testimony, if transcribed, would have corroborated Davis. The log for the succeeding shift run by Donald Davis shows 204 “previous” parts, and 88 new parts produced. This log also records, presumably by Davis, “dresser unit” and “roundness” problems, and asserts that “shoes (were) reset.”13 The report of the quality waveometer checks shows nine acceptances and no rejects in the race waviness column.14 When Silvers returned for the morning shift on October 1, all the parts had been taken out of the stubben, and were near the foreman’s office. Separated from the group were 25 parts on the side. Billy Joe Frost, the lead grinder operator, told him that the stubben contained bad parts, but that there were 50 good ones. Foreman Ken Davis told Silvers that “they’d probably run them down or grind them to a “minus five,” which Silvers explained, meant that they would be placed in a scrap bin. Silvers asked why 25 parts had been set aside, and fore- man Ken Davis told him that these were Donald Davis’ parts. Silvers testified that there was no way of distinguishing his parts from Don Davis’ parts. “It was like throwing 50 paper clips in a pile, and then another 50 clips.” Perry Price was Silvers’ immediate supervisor. He was on vacation on September 30, and returned the next day, October 1. According to Price, foreman Ken Davis told him that some bad parts had been run, and had been pulled from the washer into the stubben. Price further averred that Davis told him that Billy Joe Frost had checked all the parts “the next day” and had gone “through them a hundred percent, and we got all the bad ones out.” Price stated that he later learned that were 234 parts in the stubben. On October 1 or 2, he took a “random sample” of 15 to 20 parts from the stubben, and tested them on a “form scan” to determine whether they could be salvaged. The form scan tests “roundness,” whereas the waveometer tests “waviness.” The maximum tolerance for being “out-of-round” is one hun- dred millionths of an inch. Respondent introduced five tests of “roundness” of the parts selected by Price. Four of them were out-of-round more than the accepted tolerance amount. Al- though one of the five was within that level, it was still unac- ceptable according to Price, because of excessive “peaks” and “valleys.” All of these test results are dated October 15.15 None of these parts could be salvaged, according to Price. 11 Jt. Exh. 1. 12 R. Exh. 3. 13 R. Exh. 3. 14 Jt. Exh. 1. 15 R. Exh. 12. Respondent also introduced test results of “roundness” run by another employee on February 11, 1998, which were within the tolerance level.16 Price asserted that he determined that there were 234 parts in the stubben, and that 189 were bad. He was asked how he de- termined which parts were the work of Silvers, and which were the work of Davis. Characterization of his answer is difficult. The transcript reads in relevant part: Q. So how was it possible to determine which parts be- longed to Donald Davis, and which parts belonged to Greg Silvers: A. And the ones that was on top would have been Greg’s (Silvers). Q. But they were all just thrown in there, isn’t that cor- rect? A. Well, how would you get the ones on top down on the bottom, if you already got parts in there? Q. I’m not understanding what you’re saying. A. There are parts already in the stubben, in the bottom of the stubben, and you place more parts on top of them parts. The ones on top are bad, the ones on bottom are good. Q. Is that after you ran the waveometer tests on all those parts, is that your determination? A. Yes Q. You ran the waveometer on all those parts? A. No, I did not run them all, all of the parts. Q. Then how do you know that? A. Because they were on the bottom? Q. Yes. A. The operator that checked the parts. Q. And who was that? A. Billy Joe Frost. Q. And what did he say? How many good parts did he say there were? A. He didn’t give an exact count. He said that there was about 45 parts that were good in there, and they were on the bottom. An official “scrap report” was issued on October 7, by Philip Gardner and Phillip Lloy. Gardner was a quality control super- visor at the time. The report shows 189 pieces of scrap from the run of part 043886 on September 30.17 Although Philip Gardner testified, he was not asked about the scrap report. Billy Joe Fox, Phillip Lloy, and Ken Davis did not testify. Gardner is trained in statistical analysis. He testified in re- sponse to a hypothetical question: If there are 204 parts of which 189 are bad, and 11 waveometer checks were made, what is the probability that all the waveover checks would show good parts?18 Gardner responded with three different statistical analyses. They all show that the probability of reach- ing this result is one in an astronomical figure. 16 R. Exh. 13. 17 Part number 043886 is the part number listed on the logs for the run on September 30/October 1 by Silvers and Davis. Jt. Exh. 1; R. Exhs. 3, 8. 18 As indicated, Silvers tested 12 to 20 parts on his seven visits to the waveometer. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 556 b. Factual analysis None of the individuals who assertedly tested the entirety of the September 30 production run testified about the matter. The quality control inspectors who produced the “scrap report,” Philip Gardner and Phillip Lloy, did not testify about it. Billy Joe Frost did not testify. Frost was a lead grinder operator, and there is no evidence that quality control inspection was one of his functions. Further the evidence of the date when he alleg- edly tested the parts is inconsistent. Foreman Ken Davis did not testify. Respondent’s evidence about the salvageability of the parts in the stubben does not warrant the conclusion that Price as- serted. He took a “random sample” of 15 to 20 parts from a total of 234, and tested them on a machine which the grinder operators were not required to use. The result showed parts—5 out of 15 to 20 randomly selected—which could not be re- worked. The remaining 10 or 15 from the “random sample” apparently could have been reworked. No conclusion is justi- fied about the salvageability of the total of 234 parts based on Price’s testimony. Silvers and plant manager Annis testified that some unsatisfactory parts usually could be reworked. Yet Respondent discarded the 189 parts on September 30. Price’s testimony about the assessment of responsibility for the assert- edly bad parts is scarcely comprehensible, and clearly unper- suasive. The inconsistencies, contradictions, and missing witnesses in this evidence, and the absence of any documentary record pro- duced by the waveometer, warrant an inference that Respon- dent has not presented credible evidence that Silvers made 189 bad parts on September 30.19 More precisely, it does not show that Silvers’ reports of seven visits to the waveometer, and its acceptance of 12 to 20 parts, were false. The fact that Silvers had trouble with the machine does not support Respondent’s argument—Donald Davis also had trouble, yet recorded nine satisfactory waveometer checks, and no rejects. It follows that the hypothetical question proposed to statisti- cal analyst Gardner was not grounded in fact. Accordingly, Gardner’s testimony has no probative value. Any argument based on probability would support Silvers rather than Respon- dent. The evidence shows that some deficient parts were nor- mally reworkable. Silvers testified that this happens every night. It is therefore improbable that all of 189 allegedly bad parts were scrap and that none could be reworked. Respondent relies upon Silvers’ answer to a hypothetical question by personnel manager John Pinkerton—how would Silvers take it if all the parts were bad, but all the reports showed that they were good? Silvers replied that he would take this as falsifying documents, but insisted that all his checks were good. Silvers’ answer to this hypothetical is no more an admission than Gardner’s testimony is probative. I credit Sil- vers’ testimony that he truthfully recorded 10 to 20 acceptances of parts by the waveometer on September 30. 3. Alleged complaints against Silvers by other employees Adam Williams’ testimony, if transcribed would have shown that he was a grinder operator on a night shift in 1995 and 1996, and that Silvers was on the preceding shift. Williams applied for a transfer because he wanted a day job and wanted 19 Respondent has presented a complicated argument dividing re- sponsibility for the alleged bad parts between Silvers and Davis. My conclusion above makes it unnecessary to consider this argument. to get away from Silvers. When following Silvers, he fre- quently found that the grinding machines were in poor condi- tion and had to be repaired. He gave Supervisor Jerry Turner a statement asserting the same position.20 Williams would also have testified that he was asked on cross-examination whether he transferred out because of a rela- tionship which he had with a woman on the night shift. He replied that he had dated the woman a few times, but she was not the reason he transferred out. He would have testified that he did not see this woman after transferring out of the shift, and that his “marriage was already disrupted” before he went out with the woman. Plant Manager Annis testified that Scott Stapleton voiced concerns similar to Williams’ complaints. Stapleton did not testify. Annis asked Supervisor Perry Price to question Donald Davis, who followed Silvers’ shift. According to Annis, Price “recorded some information about that discussion.” Although Davis’ account differed from Silvers’ testimony, as reported above, it does not state complaints similar to those of Williams. 4. The decision to discipline Silvers On October 14, 1996, Annis wrote a memo to Personnel Manager Pinkerton and Terry Singleton, who was plant man- ager at the time. Pinkerton had been at the Rogersville plant since May 1996. Annis recommended Silvers’ discharge. The reason was that Silvers had intentionally falsified company documents in an attempt to hide the fact that he had produced scrap products for a majority of his shift on September 10. Annis also noted a similar warning on August 28, 1995, and complaints from other employees about Silvers’ “disruptive, anti-team behavior,” and with complaints from “two employ- ees” that Silvers frequently left his machine with “wrecked” parts.21 On October 15, Annis, Ken Price, and Silvers had a brief conversation in Annis’ office. Annis told Silvers that they had 189 pieces of scrap, and Silvers’ log sheet showing seven ac- ceptable checks. Silvers had no “reasonable explanation” ex- cept to say that he had worked on the dresser unit according to Annis. A second conversation took place the same day with Silvers, Annis, Ken Price, and Personnel Director John Pinker- ton. Annis said that because of the quantity of the scrap, they were forced to conclude that Silvers had falsified his reports. Annis asked Silvers what he would believe if he were sitting in the supervisors’ shoes. According to Annis, Silvers replied that he would have to believe that the company documents had been falsified. Nonetheless, Silvers testified that all the parts he inspected on the waveometer were good. The only explanation he offered was that he had worked on the dresser unit. Respondent suspended Silvers for the balance of the day, and told him to come back the next morning. They told him that they had to review the matter with Plant Manager Terry Single- ton. The supervisors then met with Singleton. He concurred in their recommendation of discharge on the grounds of falsifica- tion of records. He also agreed that the discipline was further justified by Silvers’ work record, and the complaints about him by other employees. Pinkerton testified that discipline more than 1 year old was not considered. However, “when we have somebody who is going to be terminated, we review the entire 20 R. Exh. 17. 21 R. Exh. 9. ROCKWELL AUTOMATION/DODGE 557 file.” He testified that the Company considered Silvers’ similar offense in August 1995. Silvers returned to the office on October 16. The supervisors told him that they had concluded that he falsified his waveome- ter reports, and they had decided to terminate him for this rea- son. They informed Silvers that he had two options. He could submit the decision to a “peer review” panel consisting of five individuals chosen at random, three hourly paid employees, and two from management.22 The panel could modify the decision, rescind it or let it stand. There were about 70 volunteer panel- ists who had received training in how to review such matters. The peer review had upheld discharges in two cases, one a “salary exempt” employee according to Personnel Director Pinkerton, and the other a management employee. Silvers testi- fied that the supervisors told him that if the review decision upheld the discharge, it would go on his record as unsatisfac- tory work. The second option was that Silvers could resign. His record would show that he left by mutual agreement, and the Company would not contest any claim for unemployment benefits. Sil- vers selected this option, and signed a statement resigning from the Company.23 5. Comparative treatment of other employees Three employees falsified company documents in 1997, but were not given discipline as severe as that administered to Sil- vers. Billy Fields in June 1997 caused 69 percent of his pro- duction to be scrapped “due to large bores.” His report had been “intentionally filled out with inaccurate information . . . [f]alsification of a company document.” His discipline was a final written warning.24 J. T. Newman ran his entire shift with one “out of print” product. His report showed checks finding the product to be within tolerance. He was unable to clarify this matter, and was found to have completed his report form “with false information.” His discipline was a final written warning.25 Walter Hatley Jr. was found to have committed the same offense in June 1997, involving an “out of print” product, and also falsified his reports of the matter.26 His discipline was also a final warning. Dennis LeBlanc allowed a machine to run unattended in December 1996, producing 86 pieces of scrap. He was given a refresher course.27 Plant Manager Annis stated that the employees’ prior records of no discipline had been considered in determining the severity of the discipline in these cases. Personnel Director Pinkerton asserted that the Company had discharged two employees for falsification of records, because they had a prior record of performance problems. These in- volved “quality records” by Kent Seals, and a timecard of Alan Bernard. Neither of these documents was produced. 6. Factual and legal conclusions The General Counsel has the burden of establishing a prima facie case that is sufficient to support an inference that pro- tected conduct was a motivating factor in an employer’s deci- sion to discipline an employee. Once this is established, the 22 In the case of decisions pertaining to management employees, the panel consists of three management members, and two hourly paid employees. 23 R. Exh. 18. 24 G.C. Exh. 2. 25 G.C. Exh. 3. 26 G.C. Exh. 4. 27 G.C. Exh. 5. burden shifts to the Respondent to demonstrate that the disci- pline would have been administered even in the absence of the protected conduct. The General Counsel must supply persua- sive evidence that the employer acted because of unlawful mo- tivation.28 The evidence shows that the Union commenced an organiz- ing campaign in early 1995, and that Silvers passed out authori- zation cards. In May 1995, he received a warning for talking to other employees when not on break. In July 1995, Supervisor Ogle told Silvers that the former had seen a list of names on Plant Manager Terry Singleton’s desk. Silvers’ name was on the list, followed by the words “For Union activity.” “Watch your behind,” Ogle told Silvers. “They’re after you.” A few days later, Supervisor Ken Ball told Silvers that the latter had said there was a “hit list” on Singleton’s desk, and demanded to know from Silvers who had told him this. When Silvers said he could not answer this question, Ball said, “If you value your job, you’ll tell me.” Silvers finally gave Ball an answer that did not involve Ogle.29 Silvers started wearing a union cap shortly before a Board election in November 1995. He testified that he was afraid for his job, and wanted to nullify any claim that the Company did not know of his union activities. Respondent denies that Silvers engaged in any union activ- ity. Thus, Respondent asserts that Silvers testified that he never passed out union cards.30 The transcript reads in relevant part: Q. And what else did you do concerning the Union or- ganizing campaign while at the Employer? . . . . A. I passed out cards. I had them in my toolbox . . . I didn’t pass them out . . . during the company hours, but they were in the toolbox and they knew that they could get them whenever they wanted. I conclude that Silvers’ statement that employees knew the cards were in his toolbox does not negate his unequivoed lan- guage: “I passed out cards.” As Silvers testified, he was care- ful not to engage in card distribution during working time. Respondent points out that Silvers agreed that Ogle told Re- spondent that Silvers was not involved in the Steelworkers’ campaign.31 This does not negate the evidence that Silvers was involved in the Machinists’ campaign. The Steelworkers started first, and Silvers found a Steelworkers card on his tool- box, which he gave to Ogle. This clarification of the labor organization with which Silvers was allied does not diminish the evidence that he was involved in the Machinists’ campaign. Respondent argues that Silvers’ testimony that he was afraid for his job is “totally disputed” by his admission that he found out about the “hit list” in July 1995, but waited until a short time before the election in November to wear a union hat.32 Fear for one’s job is not a matter of statistical measurement. Ogle, who disclosed the “hit list” to Silvers, had been fired, and 28 Wright Line, 251 NLRB 1083 (1980), enf. 662 F.2d 899 (1st Cir. 1981), approved in NLRB v. Transportation Management Corp., 462 U.S. 393 (1983). In Manno Electric Inc., 321 NLRB 278 fn. 12 (1996), the Board noted that the District of Columbia Circuit had suggested that the General Counsel’s burden is one of persuasion, not merely produc- tion. “This change in phraseology does not represent a change in the Wright Line test,” the Board stated (id.). 29 Ogle was discharged at about this time. 30 R. Br. p. 13. 31 R. Br. p. 14. 32 R. Br. p. 15. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 558 the election was only a few days away. Respondent’s argument is a psychological non sequitur. Respondent notes that Annis and Pinkerton were employed after the 1995 election, and denied any knowledge of Silvers’ union activities. Since they were the “moving parties” in the discharge decision, the General Counsel has not shown knowl- edge of Silver’s activities on the part of a responsible official. Respondent cites Delchamp, Inc. v. NLRB, 585 F.2d 91 (5th Cir. 1978). In that case, several subordinate supervisors knew of the alleged discriminatee’s union activities, but the supervi- sor who discharged him did not have this knowledge. Respon- dent also cites NLRB v. McCullough Enviromental Services, 5 F.3d 923 (5th Cir. 1993). In that case a company official issued an order requiring the discharge of any employee who refused to sign a statement acknowledging that he had read a repri- mand. The court, disagreeing with the Board, found that there was no evidence that this official had knowledge of the alleged discriminatee’s nion activities. These cases are inapposite. Terry Singleton was the plant manager at the time of Silvers’ discipline, and had issued a “hit list” containing Silvers’ name with the explanation, “For Union activities.” Annis wrote Singleton a memo outlining the case against Silvers, and recommending his discharge. He and Pinkerton consulted with Singleton, and secured his agreement with the discharge decision. In similar circumstances, the same court of appeals has stated, “[t]hose in active management hav- ing an active voice in the discharge of (the employee) knew of (the employees’s) active work in behalf of the Union’s cam- paign. That . . . was ample to justify the inference of a Section 8(a)(3) antiunion discriminatory discharge.” NLRB v. Neuhoff Bros., Packers, Inc., 375 F.2d 372, 376 (5th Cir. 1967). The evidence shows that other employees who falsified re- cords were not discharged. Under long-established Board law, this is evidence of disparate treatment evidencing discrimina- tory motivation. Respondent, however, argues that these cases are distinguishable, because these employee had no prior re- cords of similar offenses, whereas Silvers produced bad parts in August 1995. Because Silvers’ discipline took place in October 1996, the August 1995 transgression occurred more than 12 months be- fore the October 1996 discipline. According to Pinkerton, an offense more than 12 months old is not considered in making a decision to discipline an employee. Faced with the conundrum presented by the time between the two infractions, Pinkerton explained that the entire file is examined when the Company has an individual who is going to be terminated. This is a clas- sic example of circular reasoning. How does the Company know that the case is one involving discharge and thus justifies review of the entire file if it has not already determined that the employee will be discharged? I conclude that Respondent’s failure to follow its own 12-month rule constitutes evidence of discriminatory motivation. I give no weight to Pinkerton’s testimony that the Company had discharged two employees for falsifying records when they had a history of prior offenses. The allegedly falsified records were not produced. Respondent’s disparate treatment of other employees who committed offenses similar to those attributed to Silvers constitutes evidence of discriminatory motivation. Respondent argues that, even if the General Counsel has es- tablished a prima facie case, the Company has shown that it would have discharged him in any event because of his produc- tion of bad parts and falsification of records on September 30, 1996. As shown above, Respondent has not established that Silvers engaged in the conduct which it alleges. The Company’s reliance on alleged bad reports about Silvers from other employees is not convincing. The only testimony presented was that from Adam Williams, who admitted two reasons other than Silvers for transferring to another shift—a desire for a day job, and an intention to terminate a relationship with a female employee. Although Donald Davis differed with Silvers as to what the latter said at the end of his shift, he does not repeat Williams’ accusations. Annis’ testimony about the alleged complaints from other employees is mere hearsay. Silvers’ preunion status in the Company is more accurately indicated in the commendation which he received in October 1993, for making the grinding department a safer and more efficient place to work, with the cooperation of his coworkers.33 I conclude that the General Counsel has established a prima facie case, and that Respondent has not rebutted it. All of the reasons advanced by the Company for discharging Silvers are pretextual, for the reasons given above. Finally, Respondent argues, Silvers was not discharged—he quit. This argument has no merit. First, the Company told Silvers that it had decided to discharge him. However, he might be able to retain his job if he could convince a peer re- view panel that the Company’s decision should be rescinded. The initial discharge decision was unlawfully motivated, and the condition imposed upon Silvers for continued employment required him to run the gamut of a peer review panel in order to save his job. This requirement deprived him of his statutory rights. The Board has held that employees who quit rather than work under conditions which deprived them of their statutory rights have been constructively discharged.34 I reach the same conclusion herein, and find that Respondent constructively discharged Silvers on October 16, 1996, because of his assis- tance to the Union, in violation of Section 8(a)(3) and (1) of the Act. In accordance with my findings above, I make the following CONCLUSIONS OF LAW 1. Rockwell Automation/Dodge (formerly Reliance Electric) is an employer engaged in unfair labor practices affecting commerce within the meaning of Section 2(6), (6), and (7) of the Act. 2. International Association of Machinists and Aerospace Workers, AFL–CIO is a labor organization within the meaning of Section 2(5) of the Act. 3. Respondent violated Section 8(a)(3) and (1) of the Act by discharging Gregory Silvers on October 16, 1996, because of his assistance to the above-named labor organization, and in order to discourage protected activity by other employees. 4. The unfair labor practice described above constitutes an unfair labor practice affecting commerce within the meaning of Section 2(2), (6), and (7) of the Act. REMEDY It having been found that the Respondent has committed an unfair labor practice, I shall recommend that it be ordered to cease and desist and to take certain affirmative action designed to effectuate the policies of the Act. 33 G.C. Exh. 9. 34 RCR Sportswear, 312 NLRB 513 (1993); Control Services, 303 NLRB 481, 485 (1991), enfd. 975 F.2d 1551 (3d Cir. 1992); White- Evans Service Co., 285 NLRB 81, 82 (1987). ROCKWELL AUTOMATION/DODGE 559 It having been found that Respondent unlawfully discharged Gregory Silvers on October 16, 1996, I shall recommend that Respondent be ordered to offer him reinstatement to his former position, discharging replacement employees if necessary, without prejudice to his rights and privileges previously en- joyed. It is further recommended that he be made whole for any loss of earnings he may have suffered by reason of Re- spondent’s conduct from the date of his discharge to the date of Respondent’s offer of reinstatement, in the manner prescribed in F. W. Woolworth Co., 90 NLRB 289 (1950), with interest as computed in New Horizons for the Retarded, 283 NLRB 1173 (1987).35 I shall also recommend an expunction order, and the posting of notices. [Recommended Order omitted from publication.] 35 Under New Horizons, interest is computed at the “short term Fed- eral rate” for the underpayment of taxes as set out in the 1986 amend- ment to 26 U.S.C. § 6621. Interest accrued before January 1, 1987 (the effective date of the amendment) shall be computed as in Florida Steel Corp., 281 NLRB 651 (1977).