Painting Co.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 351 NLRB No. 6 42 The Painting Company and Tri-State Building & Construction Trades Council, affiliated with Na- tional Building & Construction Trades Depart- ment, AFL–CIO–CLC. Cases 9–CA–33482, 9– CA–33665–1–2, 9–CA–33674, and 9–CA–33723– 1–2 September 27, 2007 SUPPLEMENTAL DECISION AND ORDER REMANDING IN PART BY CHAIRMAN BATTISTA AND MEMBERS LIEBMAN AND SCHAUMBER On March 23, 2000, the Board issued a Decision and Order finding that the Respondent, among other things, had violated Section 8(a)(1) and (3) of the Act by dis- charging certain employees, including Charles Crisp, Warren Hull, Robert Meade, and Mark Pratt. The Order directed the Respondent to make these employees whole for their losses. On February 6, 2002, the United States Court of Appeals for the Sixth Circuit enforced the Board’s Order.1 On January 16, 2003, the Regional Director issued a compliance specification setting forth the amount of backpay due the claimants. A hearing on the specifica- tion took place on May 19–20, 2003, before Administra- tive Law Judge Joseph Gontram. On September 8, 2003, Judge Gontram issued the attached supplemental deci- sion adopting the Region’s compliance specification, as modified by the parties’ posthearing joint stipulation.2 The Respondent filed exceptions and a supporting brief, the General Counsel filed an answering brief, and the Respondent filed a reply brief. The National Labor Relations Board has delegated its authority in this proceeding to a three-member panel. The Board has considered the supplemental decision and the record in light of the exceptions and briefs and has decided to affirm the judge’s rulings, findings, and conclusions3 only to the extent consistent with this Deci- sion and Order.4 1 Painting Co., 330 NLRB 1000 (2000), enfd. 298 F.3d 492 (6th Cir. 2002). 2 The joint stipulation was based on revised earnings information that the Respondent produced at the hearing. Although the Respondent does not dispute the General Counsel’s arithmetic calculations, it does not agree with the underlying theory for the backpay calculation. 3 The Respondent has requested oral argument. This request is de- nied because the record, the exceptions, and briefs adequately present the issues and positions of the parties. 4 As the judge noted in his decision, the Respondent conceded that the backpay sought in the compliance specification for unlawfully discharged employees David Dunn and Rena Lawson is correct. Ac- cordingly, we shall order that the Respondent pay those discriminatees the backpay amounts that the judge recommended. Background The primary issue in this compliance proceeding is the reasonableness of the gross backpay calculation for dis- criminatees Charles Crisp, Warren Hull, Robert Meade, and Mark Pratt. The Respondent is a painting contractor based in the Columbus, Ohio area. The Respondent typically employs painters for specific jobs, rather than on a permanent basis. A small number of painters re- main with the Respondent over various periods of time and are transferred or reassigned from job to job. Most of the Respondent’s painters work on more than one job for the Respondent in a given year. The discriminatees began working for the Respondent on a painting job in Franklin Furnace, Ohio, on Decem- ber 4, 1995. The Respondent discharged them on Janu- ary 2, 1996. The backpay period ran from the date of discharge (January 2, 1996) to the valid offer of rein- statement on April 1, 2002.5 Because the backpay period lasted for more than 5 years, and because few of the Re- spondent’s painters worked for the Respondent through- out those years, the compliance officer extrapolated yearly earnings for the discriminates by looking to the actual annual earnings of the Respondent’s painters, most of whom worked less than a full calendar year. The compliance officer first divided each painter’s actual earnings for the year by the number of calendar days the painter worked to arrive at an average daily wage. This daily average was multiplied by 365 to arrive at a yearly wage. The yearly wages for all painters were then to- taled, and that figure was divided by the number of painters to arrive at an average yearly wage. The average yearly wage was deemed to represent the lost earnings for the discriminatees for that year. The judge adopted the Region’s gross backpay for- mula, finding it to be reliable and accurate. For the rea- sons stated below, we disagree and conclude that the gross backpay calculation was not reasonable. Accord- ingly, we reverse the judge’s findings and recompute the amount of backpay so that it more closely approximates the amount the discriminatees would have earned had they not been unlawfully discharged.6 5 The judge found, and we agree, that the Respondent’s July 3, 2000 letters to the discriminatees advising them of available positions did not constitute valid, unconditional offers of reinstatement that would termi- nate the backpay period as of that date. 6 Discriminatee Warren Hull’s whereabouts have been unknown since 1996. The specification calculated backpay for him using the same formula and offsetting interim earnings based on the average interim earnings of the other discriminatees. The judge adopted the specification backpay calculation for Hull and ordered the Respondent to place that amount in escrow with the Regional Director for 1 year. Starlite Cutting, Inc., 284 NLRB 620 (1987). As explained below, we do not resolve the backpay issues related to Hull in this decision. PAINTING CO. 43 Analysis Both the Board and the courts have applied a broad standard of reasonableness in assessing methods for calcu- lating gross backpay. Any formula that approximates the amount the discriminatees would have earned absent the discrimination is acceptable if not unreasonable or arbi- trary under the circumstances. La Favorita, Inc., 313 NLRB 902, 903 (1994), enfd. mem. 48 F.3d 1232 (10th Cir. 1995). The Board is required only to adopt a formula that will reasonably approximate the amount due; it need not find the exact amount. NLRB v. Overseas Motors, 818 F.2d 517, 521 (6th Cir. 1987), citing NLRB v. Brown & Root, Inc., 311 F.2d 447, 452 (8th Cir. 1963). Notwith- standing this leeway, “the objective is to reconstruct as accurately as possible what employment and earnings the discriminatee would have had during the backpay period had there been no unlawful action.” Performance Friction Corp., 335 NLRB 1117 (2001), citing American Mfg. Co. of Texas, 167 NLRB 520 (1967); CHM Section 10540.1. Where the employment of the work force as a whole is “intermittent, the fact of intermittency must be taken into account unless there is something in the record which jus- tifies a finding that, for some reason, the employee in- volved would not have been affected by the fluctuations that affected the group as a whole.” NLRB v. Ironworkers Local 378, 532 F.2d 1241, 1243–1244 (9th Cir. 1976) (rejecting compliance specification that treated discrimina- tee as comparable to elite group of steadily employed ironworkers absent sufficient justification), on remand 227 NLRB 692 (1977), supplemented by 262 NLRB 421 (1982). “It is the General Counsel’s burden to establish gross backpay amounts that are reasonable, not arbitrary.” Parts Depot, Inc., 348 NLRB 152, 153 (2006). Here, the General Counsel failed to establish that the Region’s backpay calculation is reasonable. The Region’s backpay formula is unreasonable because very few painters worked the equivalent of a full year for the Respondent in any given calendar year, yet the specifi- cation assumes that to be the standard. The Respondent’s evidence shows that, between 1996 and 2001, the Respon- dent employed 350 painters for various periods of time. Only a handful of those painters—between 5–12 painters in a given year—worked what could reasonably be charac- terized as a full year for the Respondent. Over the entire 6-year period, the average annual number of days worked by painters was just 91 workdays, ranging from an average of 63 workdays in 1996 to an average of 130 workdays in 2001. The majority of Respondent’s work force did not In Member Liebman’s view, Hull’s gross backpay and the Respon- dent’s matching FICA were approximately determined under extant law and should be placed in escrow. See Starlite Cutting, Inc., above. work the average number of workdays in the year. Conse- quently, the Region’s calculations do not yield “a close approximation” of the employment and earnings the dis- criminatees would have had during the backpay period. Absent some evidence to suggest that a given discrimi- natee would have exceeded the average number of work- days each year, the most accurate method for determining the amount of backpay due is to assume that each dis- criminatee would have worked the annual average number of workdays and earned the same annual wages as the average painter employed by the Respondent during that year. We will further presume that the average number of workdays for a given year would have been worked during one or two consecutive quarters.7 It is also reasonable to assume that 65 workdays (13 weeks times 5 days) repre- sents the number of workdays in a full calendar quarter.8 We will now recalculate backpay for Crisp, Meade, and Pratt pursuant to our revised formula. To compute net backpay, we subtract the interim earnings of each dis- criminatee. Specifically, for the years 1996 and 2002, when the average number of workdays was respectively 63 and 65, we subtract interim earnings for one full quar- ter.9 In 2001, when the average number of workdays was 130 days, the offset is two full quarters. In the other years, when there are more than 65 workdays but less than 130 workdays, the offset amount equals one full quarter plus a percentage of the discriminatee’s quarterly interim earn- ings.10 Applying this formula, we find that the net back- pay amounts due the discriminatees are as follows: Charles Crisp–$36,430; Robert Meade–$1,727; and Mark 7 Most of the Respondent’s painters worked in consecutive periods of time during a year. 8 Even though the Respondent’s work force worked some weekends, using 65 workdays is not unreasonable given that the average annual number of days worked by Respondent’s painters was so few. Further, the average annual earnings attributed to the discriminatees incorpo- rates payments for all hours worked and pay for weekends and over- time received by the Respondent’s painters. 9 In the compliance specification, the quarterly interim earnings for discriminatees Crisp and Pratt were calculated by dividing their annual earnings by four. This results in an even distribution of income over the year. Although the compliance specification used discriminatee Meade’s actual quarterly earnings, we have, for consistency, evenly distributed his annual earnings in determining his quarterly interim earnings. We agree with the judge that Meade’s backpay is properly calculated on a quarterly basis, consistent with F. W. Woolworth Co., 90 NLRB 289 (1950). 10 The percentage is computed by dividing the number of days over 65 by 91 days. The divisor is 91 days because the specification spread the annual earnings equally over each calendar quarter. To illustrate the calculation, in 1999 the average number of workdays was 79 (or 65 plus 14) workdays. The total offset is one quarter’s interim earnings plus 14 divided by 91 (15 percent) of the quarterly interim earnings. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 44 Pratt—$25,365. Appendix A sets out the complete back- pay calculations for each discriminatee.11 ORDER The National Labor Relations Board orders that the Re- spondent, The Painting Company, Plain City, Ohio, its officers, agents, successors, and assigns, shall make whole the following discriminatees by paying them the following amounts, with interest as set forth in New Horizons for the Retarded, 283 NLRB 1173 (1987). David Dunn $ 168 Rena Lawson 168 Charles Crisp 36,430 Robert Meade 1,727 Mark Pratt $ 25,365 IT IS FURTHER ORDERED that the issue of backpay due discriminatee Warren Hull shall be severed and remanded to the Regional Director to further investigate Hull’s whereabouts, resolve the backpay issue relating to him pursuant to the terms of this Supplemental Decision, and to take further appropriate action. APPENDIX A Charles Crisp YEAR WORKDAYS & OFFSET AVG. ANNUAL EARN. CRISP-Q. INTERIM EARN. NET BACKPAY 1996 63-1Q $ 7,584 $ 4,425 $ 3,159 1997 75-1Q +11% (10÷91) $ 8,266 $ 3,581 (3226 X 1.11) $ 4,685 1998 90-1Q +27% (25÷91) $ 10,539 $ 5,657 (4454 X 1.27) $ 4,882 1999 79-1Q +15% (14÷91) $ 10,148 $ 3,696 (3214 X 1.15) $ 6,452 2000 107-1Q +46% (42÷91) $ 13,676 $ 8,087 (5539 X 1.46) $ 5,589 2001 130-2Q $ 17,407 $ 14,448 (7224 X 2) $ 2,959 2002 65-1Q $ 8,704 0 $ 8,704 TOTAL $ 36,430 11 Our decision today resolves backpay for all the discriminatees ex- cept Hull, whose whereabouts is unknown. Chairman Battista and Member Schaumber find that Hull’s backpay raises significant issues of law and policy. Those issues include the question of which party has the burden of proof concerning whether a discriminatee has reasonably searched for work during the backpay period. In these circumstances, an Order will be entered only as to the five discriminatees listed, with the backpay issues relating to Hull to be severed and resolved as soon as possible. Parts Depot, Inc., 348 NLRB 152, 155 fn. 19 (2006). Mark Pratt YEAR WORKDAYS & OFFSET AVG. ANNUAL EARN. PRATT-Q. INTERIM EARN. NET BACKPAY 1996 63-1Q $ 7,584 $ 4,425 $ 3,159 1997 75-1Q +11% (10÷91) $ 8,266 $ 3,875 (3491 X 1.11) $ 4,391 1998 90-1Q +27% (25÷91) $ 10,539 $ 4,459 (3511 X 1.27) $ 6,080 1999 79-1Q +15% (14÷91) $ 10,148 $ 5,874 (5108 X 1.15) $ 4,274 2000 107-1Q +46% (42÷91) $ 13,676 $ 7,618 (5218 X 1.46) $ 6,058 2001 130-2Q $ 17,407 $19,232 (9616 X 2) $ 0 2002 65-1Q $ 8,704 $ 6,983 $ 1,721 TOTAL $25,365 Robert Meade YEAR WORKDAYS & OFFSET AVG. ANNUAL EARN. MEADE-Q. INTERIM EARN. NET BACKPAY 1996 63-1Q $ 7,584 $ 5,857 $ 1,727 1997 75-1Q +11% (10÷91) $ 8,266 $ 12,460 (11,225 X 1.11) $ 0 1998 90-1Q +27% (25÷91) $ 10,539 $ 13,958 (11,012 X 1.27) $ 0 1999 79-1Q +15% (14÷91) $ 10,148 $ 18,132 (15,767 X 1.15) $ 0 2000 107-1Q +46% (42÷91) $ 13,676 $ 20,379 (13,958 X 1.46) $ 0 2001 130-2Q $ 17,407 $ 27,608 (13,804 X 2) $ 0 2002 65-1Q $ 8,704 $ 11,973 $ 0 TOTAL $ 1,727 Mark Mehas, Esq., for the General Counsel. William C. Moul, Esq. and Craig Pederri, Esq. (Thompson Hine LLP), of Columbus, Ohio, for the Respondent. DECISION STATEMENT OF THE CASE JOSEPH GONTRAM, Administrative Law Judge. This backpay case was tried in Columbus, Ohio, on May 19 and 20, 2003. On March 23, 2000, the Board issued its Decision and Order find- ing that the Respondent (or TPC) had violated Section 8(a)(1) and (3) of the National Labor Relations Act. The Board ordered the Respondent to make whole certain employees, including Charles Crisp, Warren Hull, Robert Meade, and Mark Pratt, for losses resulting from the Respondent’s discharge of these em- ployees in violation of the Act. Painting Co., 330 NLRB 1000 PAINTING CO. 45 (2000).1 On February 6, 2002, the United States Court of Ap- peals for the Sixth Circuit entered a judgment enforcing the Board’s Order. Painting Co. v. NLRB, 298 F.3d 492 (6th Cir. 2002). A controversy having arisen over the amount of back- pay due to the discriminatees under the terms of the Board’s Order, the Acting Regional Director for Region 9 of the Board issued a compliance specification and notice of hearing on January 16, 2003. The issues in this backpay proceeding are (1) whether the General Counsel’s method of calculating the backpay obligation is unreasonable or arbitrary, and (2) when does the backpay period end. On the entire record, including my observation of the de- meanor of the witnesses, and after considering the briefs filed by the General Counsel and the Respondent, I make the follow- ing FINDINGS OF FACT I. THE UNDERLYING UNFAIR LABOR PRACTICES CASE The Respondent is a painting contractor headquartered in Plain City, Ohio. It is wholly owned by three brothers: Jeff, David, and Terry Asman. TPC generally employs about 30 painters, although its work force fluctuates between 10 and 50 painters. Crisp, Hull, Meade, and Pratt started working for TPC at its Franklin Furnace job on December 4, 1995, as journey- man painters. The Respondent notified them on January 2, 1996 that they were fired, and their last day of employment with TPC was December 29, 1995. The Board affirmed the finding and conclusion of the administrative law judge that they were dis- charged in violation of the Act, and it ordered TPC to offer them reinstatement and make them whole. In his decision, the administrative law judge found that the Respondent presented “a demonstrably false explanation for the termination of the employment of four employees [the dis- criminatees herein] who actively supported the Union.” The judge specifically discredited David Asman’s testimony that the discriminatees were discharged because they were substandard workers who were not well suited for the Franklin Furnace job. 330 NLRB at 1011. The judge also found that the Respondent did not have a policy of replacing current employees based on seniority. The Board’s Order in this case directs the Respondent to of- fer the discriminatees full reinstatement to their former jobs or, if those jobs no longer exist, to substantially equivalent jobs, and to make the discriminatees whole for any loss of earnings and other benefits, computed on a quarterly basis from the date of discharge to the date of a proper offer of reinstatement. 330 NLRB at 1014–1015. II. GENERAL COUNSEL’S BACKPAY CALCULATION Jon Grove, compliance officer for Region 9 of the Board, computed the backpay and described the bases for his calcula- tions. The data he used in making his computations were ob- tained from the Respondent’s records. The computations them- 1 The Respondent concedes that the backpay sought in the compli- ance specification for unlawfully discharged employees David Dunn and Rena Lawson is correct. Accordingly, an appropriate order will be entered. selves are reflected in the appendices to the compliance specifi- cation.2 The compliance officer used a backpay period of Janu- ary 2, 1996, to April 1, 2002. The backpay calculations are based on Formula Two of the Board’s Casehandling Manual- Compliance Proceedings, § 10532.3, which uses the earnings of comparable employees to calculate the lost earnings of dis- criminatees. The Respondent agrees that Formula Two is the correct method of calculating the backpay in this case. Because the backpay period lasted for more than 5 years, and because few of the Respondent’s painters worked for the Re- spondent throughout those years, the compliance officer ex- trapolated yearly earnings from the actual earnings of the paint- ers who did not work a full calendar year. (The actual annual earnings were used for all painters who worked a full calendar year. For the years 1996 through 2001, there were 5, 5, 8, 6, 8, and 12 painters, respectively, who worked the entire year.) The method by which such yearly earnings were extrapolated was as follows: each applicable employee’s actual earnings were divided by the number of days the employee worked to arrive at an average daily wage. This daily average was multiplied by 365 to arrive at a yearly wage. The yearly wages for all painters were then added, and divided by the number of painters, to arrive at an average yearly wage.3 The compliance officer checked the reliability of his extrapo- lation method by comparing the average actual wages of the painters who worked a full year against the extrapolated wages for the other painters. In 1996, the average actual wages of the painters who worked a full year was $35,348. The extrapolated wages for the employees who worked less than a full year was $34,409. This comparison shows that the extrapolated wages were sufficiently close to the actual wages to be reliable. More- over, because the extrapolated wages were less than the actual wages, the extrapolation method likely benefited the Respon- dent by reducing the average wage on which backpay was cal- culated. The Respondent objects to this calculation method and main- tains that the earnings of the Respondent’s painters should not be extrapolated to calculate backpay, but that only actual, unex- trapolated earnings should be used. The actual, unextrapolated wages of its painters was much less than the yearly earnings calculated under the extrapolation method. For example, the actual average earnings of the Respondent’s painters for 1996 was $7583, whereas their extrapolated earnings were $34,409. 2 At the hearing, the Respondent produced additional records result- ing in slightly revised backpay calculations. The parties have submitted these revised calculations in a joint stipulation, together with appendi- ces 1–4. The Respondent does not dispute the mathematical calcula- tions, but does dispute the theory on which the calculations were made, as well as the General Counsel’s contention that it owes the amounts set forth in the joint stipulation. The General Counsel’s calculations as set forth in the joint stipulation and appendices are used throughout this decision, rather than the calculations set forth in the original compli- ance specification. 3 Consistent with F. W. Woolworth Co., 90 NLRB 289 (1950), this yearly wage was divided by four to arrive at quarterly earnings. The Respondent does not specifically challenge this proper use and calcula- tion of quarterly earnings, except for the use of the extrapolated yearly earnings described above. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 46 The obvious explanation for this difference is that most of the Respondent’s painters worked for the Respondent for much less than a full year. But the compliance officer calculated lost wages for the full years of 1996 to 2001, plus the first quarter of 2002. Where a discriminatee has been denied earnings for a full year, one does not calculate his lost earnings based on 2 or 3 months of work, but on 12 months of work. And in determin- ing a discriminatee’s lost earnings for a full year, it is proper and necessary to use the earnings and the extrapolated earnings of comparable employees for the full year. Similar reasoning would apply to any period of time for which backpay is being calculated. Thus, if only one quarter of a year were involved, one would not use as comparative earn- ings the wages of employees who worked for only 1 month. The earnings of such employees could either be excluded from the calculations or, if there were many such employees, the properly comparable earnings would be extrapolated to arrive at yearly earnings and then reduced to quarterly earnings. In the present case, many of the Respondent’s painters worked for periods substantially less than 1 year. Accordingly, it was proper to include the wages of these employees and to extrapolate the wages these employees would have earned if they had worked for the entire year. In the alternative, the com- pliance officer could have excluded all painters who worked for periods of less than 1 year and used, instead, only the actual earnings of painters who worked for each full year. As noted above, this would have resulted in a higher backpay obligation, something the Respondent does not seek and the General Coun- sel does not advocate. However, it would not be proper to adopt the Respondent’s argument and to use directly, without ex- trapolation, the earnings of workers who worked 1, 2, and 3 months as comparable absolute earnings for backpay periods greatly exceeding 1, 2, and 3 months. To use a trite but applica- ble phrase, the Respondent’s contention would compare apples with oranges. The General Counsel’s method compares apples with apples. The Respondent also contends that even if the calculation of backpay were to include extrapolated yearly wages, a multiplier of 260 rather than 365 should be used because its painters did not often work weekends and the former figure excludes all weekends. Nevertheless, the Respondent’s painters did work some weekends, and to exclude all weekends from the calcula- tion would lead to the same, possible imprecision as the inclu- sion of all weekends. The Respondent has not demonstrated that a multiplier of 260 would lead to a more reliable figure than a multiplier of 365. In the end, the reliability and accuracy of the extrapolation method used by the Government is demon- strated, as noted above, by the closely comparable amounts between the actual and the extrapolated earnings. The Respondent’s objection to the Government’s backpay calculation is primarily based on its objection to the backpay period of approximately 5–1/2 years. Recognizing this helps to explain the inapt comparisons asserted by the Respondent. For example, the Respondent’s contention that actual, rather than extrapolated, earnings of its painters is the proper measure for backpay might be defensible if, but only if, the Respondent also proves that the discriminatees would have worked the same amount of time as the average time worked by its other paint- ers. Accordingly, I will now turn to the calculation of the back- pay period. III. THE BACKPAY PERIOD The parties agree that the backpay period begins on January 2, 1996. The question is, when does the backpay period end? By identical letters dated July 3, 2000, the Respondent ad- vised each of the discriminatees, “The Painting Company cur- rently has positions available for painters as advertised in the Columbus Dispatch and the Orlando Sentinel newspapers. If you are interested in employment with The Painting Company, please respond by phone . . . by fax. . . or by mail. . . . Your response is requested by July 7, 2000 and 5:00 p.m. as we will be filling these positions as soon as possible.”4 Although the letters to these discriminatees were dated July 3, the envelopes bear postmarks of July 5. Pratt testified that he received the letter on July 5, and that he telephoned the Re- spondent on the date he received the letter, which was also the date the letter directed him to call the Respondent. (Pratt was apparently confused about this date because the letter directed him to call the Respondent no later than July 7.) As soon as Pratt received the letter, he telephoned the Respondent and talked to David Asman. He asked Asman what he would be paid, and Asman told him between $7 and $8 an hour. Pratt then asked if Asman had any prevailing wage jobs. Asman replied yes, but they were full and he could not offer Pratt a job at the prevailing wage. The Respondent had paid Pratt and the other discriminatees $19.12 an hour on the job from which they had been unlawfully fired in 1996.5 David Asman testified that his telephone conversation with Pratt occurred on July 17, 2000, and that in this telephone con- versation, he offered to pay Pratt $14.50 per hour, which was allegedly consistent with what the Respondent was paying on its jobs at the time. I do not credit this testimony because I find that Asman was confused about the date of the conversation and what letter the conversation was in response to. The Re- spondent was paying its painters $14.50 per hour in 2001,6 not in 2000 when the conversation allegedly occurred. In fact, the Respondent did offer to pay Pratt and the other discriminatees $14.50 per hour in its 2002 letter to them. But since the Re- spondent’s own records show that it was not paying its painters this hourly wage in 2000, I find that Asman was not credible when he testified that on July 17, 2000, he offered to pay Pratt $14.50 per hour. Charles Crisp received his letter on July 8, 2000. The dead- line set forth in the letter for contacting the Respondent had already passed by the time he received the letter. Accordingly, he did not contact the Respondent after he received it. By identical letters dated April 1, 2002, the Respondent of- fered to reinstate each of the discriminatees to their former jobs as journeyman painters.7 As noted above, these letters listed a starting wage of $14.50 per hour, the existing average wage paid to its painters.8 The letters listed a specific job to which the 4 GC Exh. 3. 5 R Exh. 8 6 R Exhs. 5 and 14. 7 GC Exh. 2. 8 R Exh. 14. PAINTING CO. 47 discriminatees would be assigned, and they were allowed a reasonable time of 30 days within which to respond. The Respondent contends that backpay should end at the conclusion of, or shortly after, the job from which the discrimi- natees were fired, the Franklin Furnace job, because the dis- criminatees would not have been transferred to any of the Re- spondent’s other jobs. The Franklin Furnace job ended on March 17, 1996, and the Respondent contends that there was no likelihood that the discriminatees would have been transferred or reassigned after the conclusion of that first and only job they worked for the Respondent.9 A. Transfer or Reassignment The Respondent has a permanent core of painters who re- main with the Respondent over various periods of time and who are transferred or reassigned from job to job. There are a large number of transfers and a large number of painters who transfer to different jobs. For example, on December 31, 1995, there were 34 painters, excluding the discriminatees, working for the Respondent.10 During 1996, the Respondent transferred 29 of these painters to other jobs from the job on which each of the painters started.11 Moreover, the Respondent transferred these painters 174 times in 1996, an average of 6 transfers per painter. On the other hand, about five painters, representing about five percent of the Respondent’s work force in 1996, worked the entire year, and these painters were generally from the Co- lumbus-Central Ohio area.12 These numbers remained the same in 1997. In 1998, 8 painters, representing 12.3 percent of the Respondent’s work force; in 1999, 6 painters or 6.7 percent of the Respondent’s work force; in 2000, 8 painters or 8.5 percent of the Respondent’s work force; and in 2001, 12 painters or 16.2 percent of the Respondent’s work force, worked the entire year. Just as all of the Respondent’s painters frequently transfer to other jobs of the Respondent, the painters who worked on the Franklin Furnace job also transferred with regularity. There were 19 painters who worked on the Franklin Furnace job, excluding the discriminatees, and 11 of these painters trans- ferred to other jobs of the Respondent. All of these painters were transferred to the Respondent’s other jobs after the termi- nation of the Franklin Furnace job, and there were three paint- ers who each were transferred to more than 20 jobs.13 If one of the Respondent’s painters becomes unemployed for any reason, the Respondent does not contact that painter when another job for which the painter is qualified becomes avail- able. Rather, in order for that painter to be rehired, the painter is 9 Alternatively, the Respondent contends that the July 17, 2000 let- ters to the discriminatees were valid and proper offers of reinstatement and that backpay should not extend beyond July 17, 2000. The letters were described above and the contention is considered in the analysis section below. 10 R. Exh. 2. 11 These transfer records only tracked the transfers of workers who were working on December 31, 1995, and do not include the record of transfers for painters hired after the first of the year. 12 R. Exh. 9, p. 4; R. Exh. 18. 13 R. Exh. 7. generally required to contact the Respondent. When the Re- spondent recruits painters for any job, it primarily advertises in the Columbus Dispatch newspaper. While this recruitment practice did not limit the number or residences of the painters who were eligible to work for the Respondent, it did result in most of the Respondent’s painters being from the Central Ohio area. When the Respondent worked on projects outside of the Central Ohio area, it generally used painters who were hired from its Columbus, Ohio office. And, when local painters were hired for such projects, these workers rarely stayed on to work other jobs for the Respondent in the Central Ohio area.14 But these statistics, like other evidence presented by the Respon- dent, only show that some painters from other areas chose not to work on the Respondent’s other projects or not to transfer to other projects of the Respondent. The statistics do not show that the discriminatees were in similar circumstances as these other workers or would have made the same choice as these other workers. The Respondent was working on at least 3, and possibly as many as 17 different projects during the same time that it was also working on the Franklin Furnace job.15 One of these jobs was in Florida and one was in Tennessee. The remaining pro- jects were in Ohio. All of the jobs, but especially the Ohio jobs, were, at least potentially, projects to which the discriminatees could have transferred. Moreover, the Respondent told the dis- criminatees, before they were unlawfully fired, that the Re- spondent had plenty of work and that if they did a good job, the Respondent could and would keep them working. The Respon- dent also told the discriminatees that it had weekend work for them and that it always had prevailing wage work. No one at the Respondent ever told the discriminatees that they were not qualified to do certain painting work. The Respondent asserts there is little or no likelihood that the discriminatees would have been transferred to another of the Respondent’s projects during or after the completion of the Franklin Furnace project. The evidence does not support this assertion. First, there were jobs to which the discriminatees could have transferred. Second, 11 of the 19 painters who worked on the Franklin Furnace job did transfer to other pro- jects of the Respondent from the Franklin Furnace job. Third, the Respondent did not transfer employees based on seniority, so the fact that the discriminatees had recently been hired in December 1995 would not have adversely affected their trans- fer opportunities. Fourth, the Respondent told the discrimina- tees that it had plenty of work and could keep them working. Asman testified that the discriminatees were poor workers and if they had been able to perform the work they were hired to perform, they would not have been discharged.16 The Re- spondent made this same claim before Administrative Law Judge Stephen Gross. Judge Gross found the claim incredible and concluded that the Respondent did indeed fire the discrimi- 14 See R. Exh. 11. 15 R. Exh. 10. In spite of the beginning and ending dates for the jobs as listed in this exhibit, the Respondent was working on these jobs at the same time as it was working on the Franklin Furnace job. (Tr. 168.) 16 Tr. 226. DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 48 natees because of unlawful discrimination. I am bound by the finding of Judge Gross, but if I were not, I would similarly and independently find that Asman’s assertion is not credible. As- man is the vice president of TPC, and his duties are limited to estimating jobs. He finds jobs from reports and he estimates jobs from blueprints or from visiting the sites. There is no evi- dence that Asman supervises the painters or has any firsthand knowledge of the painters’ abilities. Moreover, there is no evi- dence in the present case that Asman was told by anyone else about the capabilities of the discriminatees. In addition, Asman acknowledged that the discriminatees were qualified to do ar- chitectural painting and he further acknowledged that the ma- jority of TPC’s jobs involve architectural painting. Accord- ingly, the Respondent’s contention that the discriminatees were poor workers and that the jobs to which they could be trans- ferred were limited because they were poor workers is not credible and is rejected. Pratt and Crisp testified at the compliance hearing. Pratt pre- fers to work within a 2-hour drive from his home in Kentucky. Columbus is approximately 120 miles from Pratt’s residence. He was prepared to continue his work for the Respondent, in- cluding making “trips up here to Columbus, you know, to do some painting up here, and you know, I thought I had me a pretty good job at the time.”17 Crisp did not state any prefer- ence for working close to his residence, although he has fre- quently worked for an employer in Louisville, Kentucky. Also, Crisp failed to apply for a job with a prospective employer, BCI Construction and Engineering, Ltd., and noted on the State unemployment compensation form that the company “would hire for out of state work.”18 The evidence does not disclose the type of work involved or the state in which this prospective job was located. The Respondent claims that there is little likelihood the dis- criminatees would have accepted work in the Columbus area at the end of the Franklin Furnace job. I find the evidence sup- ports the opposite conclusion. Columbus is within the general area Pratt prefers to work. Moreover, the locations of jobs he has held since the Respondent fired him shows that he is will- ing to travel as far as Indiana and Georgia for work. He was willing to work in the Columbus, Ohio area and was looking forward to it. The locations of the jobs held by Crisp after the Respondent fired him were in such out-of-state places as Indi- ana and Illinois, and in other locations in Ohio such as Cincin- nati, Portsmouth, Long Bottom, Chesapeake, and Cleveland.19 Interim jobs held by Meade were in Illinois, Kentucky, and Ohio.20 I find there is a substantial likelihood that the discrimi- natees would have accepted work from the Respondent in the Columbus area. The Respondent claims that the work history of Pratt and Crisp after their discharge shows that they both changed em- 17 Tr. 267. 18 The Respondent states in its brief that Crisp “specified, in writing, that he was not interested in ‘out-of-state’ work. (See R Exh. 22.)” This statement misrepresents what is contained in that exhibit. Counsel would do better to check the accuracy of the statements made in its brief before it is sent to the Board for filing. 19 See joint stipulation, app. 1. 20 See joint stipulation, app. 3. ployers with frequency. The Respondent argues that an infer- ence should be drawn from these work patterns that the dis- criminatees would not have remained long with the Respondent even if they had not been fired. This contention is rejected. It is not at all clear that if the Respondent had not fired the discrimi- natees, their work histories since that date would be the same as they in fact turned out. The Respondent’s argument fails to account for or even consider the central employment experi- ence of the discriminatees during the backpay period of time— their unlawful discharge by the Respondent. It simply does not follow that a discriminatee’s employment history and experi- ence would have been the same without regard to his unlawful discharge by the Respondent. Moreover, the Respondent’s dis- charge of the discriminatees was the event that started the proc- ess and caused them to seek other employment in the first place. For all of these reasons, I do not infer from the discrimi- natees’ actual employment histories after discharge that their employment histories would have been similar or the same in the absence of such discharge. The Respondent claims that its total employment of painters was reduced to six in mid-July 1996, and that none of these painters were from outside Central Ohio. The Respondent ar- gues from these figures that, “It is a mathematical certainty that they [the discriminatees] would not have worked beyond mid- May 1996, at any location.”21 The reasoning that is used to arrive at this conclusion is simply fallacious. Moreover, the Respondent fails to address the fact that it did not hire or fire painters based on their residence. If, at any one or more times, all of the Respondent’s painters resided in Central Ohio, this was primarily a result of where the Respondent advertised for painters (viz., the Columbus Dispatch), and was not the result of any design on its part to employ only or primarily painters with such residences. Thus, the likelihood that the discrimina- tees would have continued working for the Respondent is not diminished or affected by the residences of the painters who actually worked for the Respondent in the absence of the dis- criminatees. In addition, the Respondent did not satisfy its bur- den of proof by showing that the discriminatees would not have held four of these six positions. The Respondent also contends that the discriminatees would not have been employed after the Franklin Furnace project because either Respondent or the discriminatees would have terminated the employment relationship.22 This argument as- sumes facts either not in evidence or not credited. First, there is no evidence in this case that the discriminatees would have terminated their employment relationship with the Respondent. Second, the only evidence that the Respondent may have ter- minated the employment relationship is from the testimony of Asman who based his conclusion on his asserted, but discred- ited, opinion of the discriminatees’ abilities. The Respondent claims that the backpay period should be no greater than the average length of employment of all of its painters. Alternatively, the Respondent argues that the statistics show there is no likelihood that the discriminatees would have worked beyond the end of the Franklin Furnace job. For exam- 21 R. Br., p. 9. 22 R. Br., p. 10. PAINTING CO. 49 ple, the Respondent states that it employed 350 painters over the 5-year period from 1996 to 2001. Of those, only 2 worked throughout that period, 9 worked a portion of 5 years, 7 worked a portion of 4 years, 19 worked a portion of 3 years, 64 worked a portion of 2 years, and 249 worked only during a single year. Of painters residing outside of Central Ohio, none worked more than 3 years, and only 28 worked more than a single year. An average of 7 painters worked all of each year during the period 1996–2001.23 The Respondent’s statistical argument assumes that the dis- criminatees would have been motivated by the same employ- ment considerations as motivated the Respondent’s other paint- ers, and that the discriminatees would have acted the same as the “average” painter (however that “average” painter is char- acterized) who has worked for the Respondent. The Respondent produced no evidence to show what factors motivated its other painters to leave their employment. The Respondent also pro- duced no credible evidence to show that the discriminatees would have been motivated to quit their employment with the Respondent if they had not been fired. The fact that other paint- ers may have left their employment with the Respondent after one, several, or many jobs does not prove that these discrimina- tees would have done the same. The mere fact that most of the Respondent’s painters elected to end their employment with the Respondent after a period of time, in the absence of an explana- tion of why those painters acted as they did and that the same considerations apply to the discriminatees, does not prove that the discriminatees would have acted similarly. If the Respon- dent seeks to establish that the discriminatees would have acted the same as other painters it has employed, it was the Respon- dent’s burden to prove that fact. The Respondent failed to do so. The Respondent has no policy against hiring painters from outside the Central Ohio area. The Respondent’s policy is to keep its painters working as long as each party desires that result. The Respondent had other work during and after the Franklin Furnace job, and it advised the discriminatees that it had plenty of work for them and would keep them working if they did a good job. The Respondent now seeks to disclaim and impeach this statement by its supervisor, but its own records support the statement that it did have work for them. The Re- spondent has not shown that it did not have, throughout the entire backpay period, jobs to which the discriminatees could have been transferred. Robert Meade has been employed with a single employer since late 1996. At some unknown date before the hearing in this compliance proceeding, and possibly as early as 1996, Meade moved to Louisiana to continue working for this em- ployer. The Respondent produced no evidence of any effort to obtain Meade’s appearance at the hearing and he did not ap- pear. The backpay due to Meade, as calculated by the compli- ance officer on a quarterly basis, is $10,598. If this backpay were calculated on a yearly rather than quarterly basis, as con- tended by the Respondent, the backpay would be $6,850. The compliance officer was unable to locate Warren Hull. The last address the compliance officer had for Hull was from 23 R. Exhs. 13 and 18. 1996–1997. Hull’s backpay was calculated in the same manner as for the other three discriminatees, except that his interim earnings were the average of those discriminatees’ interim earnings. IV. ANALYSIS In compliance proceedings, the Board attempts to place the discriminatees, as nearly as possible, in the same financial posi- tion they would have enjoyed but for the illegal discrimination. Phelps Dodge Corp. v. NLRB, 313 U.S. 177, 194 (1941). A backpay award perforce involves some ambiguity and estima- tion, and therefore it is only an approximation, necessitated by the employer’s wrongful conduct. Cobb Mechanical Contrac- tors, 333 NLRB 1168 (2001), quoting Bagel Bakers Council of Greater New York v. NLRB, 555 F.2d 304, 305 (2d Cir. 1977). “The Board’s well-settled policy is that ‘[a backpay] formula which approximates what discriminatees would have earned had they not been discriminated against is acceptable if it is not unreasonable or arbitrary in the circumstances.’ . . . Further, it is also well-settled that any uncertainty in the evidence is to be resolved against the Respondent as the wrongdoer.” Id. (cita- tions omitted). The General Counsel bears the burden of prov- ing the amount of gross backpay. The Respondent bears the burden of proving any reductions in gross backpay. Florida Tile Co., 310 NLRB 609 (1993). Although the General Counsel is allowed wide discretion in selecting a method of computing backpay, its selection need not be accepted if the Respondent proposes an alternate formula. Rather, the formula that produces the most accurate method of calculating backpay will be accepted. Alaska Pulp Corp., 326 NLRB 522, 523 (1998). Again, any uncertainty in the attempt to reconstruct backpay must be resolved against the Respondent and in favor of the discriminatees. Employers in the construction industry are subject to the same rules and burdens of proof as employers in other indus- tries, notwithstanding the Board’s recognition of the distinctive employment practices in that industry. In Dean General Con- tractors, 285 NLRB 573 (1987), the Board stated, “Although we recognize that employment patterns in the construction industry have unique characteristics and jobs are frequently of short duration, these general characteristics, standing alone, do not justify a departure from our traditional make-whole remedy prior to compliance.” The Board described the Respondent’s burden as follows: If the Respondent establishes at compliance that [the discriminatee] likely would not have been transferred or reassigned elsewhere, the Respondent’s obligation toward [the discriminatee] will be to consider him eligible for em- ployment at future projects, on application, on a nondis- criminatory basis. Evidence pertaining to transfer or reas- signment may be considered both concerning the Respon- dent’s reinstatement obligation toward [the discriminatee] and the date when the Respondent’s backpay liability to [the discriminatee] may have terminated. 285 NLRB at 575. Similarly, in Laben Electric Co., 323 NLRB 428 (1997), the Board explained that the Respondent’s burden of proof in a construction industry compliance proceed- DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD 50 ing was proof that the discriminatees would not have been transferred to other projects. The Respondent did not prove that these discriminatees, Pratt, Hull, Meade, and Crisp, would not have been transferred to other projects. In fact, many of the painters on the Franklin Furnace job did transfer to the Respondent’s other jobs. How- ever, most of the Respondent’s painters, in general, do not transfer to its other jobs throughout a full year, much less a period of years. Also, most of the painters who do transfer to other jobs are painters who live in the seven county area con- tiguous to Columbus, Ohio. (The discriminatees did not live in this area.) But these statistics merely reflect what the Board recognized in Dean General Contractors as the typical em- ployment pattern in the construction industry. The statistics show that generally, with two exceptions, painters have not remained employed with the Respondent throughout the full 5 years encompassed in the backpay period. This employment pattern is not the result of a policy or practice of the Respon- dent to fire every painter after the completion of each job. On the contrary, the Respondent typically does transfer painters from one job to another. The Respondent’s statistics merely reflect an employment pattern that is typical of the construction industry, and they are clearly not sufficient, standing alone, to meet the Respondent’s burden to prove that these discrimina- tees would not have been transferred or reassigned to other jobs. Dean General Contractors, supra. The General Counsel’s method of calculating backpay uses earnings that the Respondent’s painters would have earned if they had remained employed with the Respondent over the course of a full year. Such yearly earnings are then broken down quarterly from which quarterly interim earnings are de- ducted. See F. W. Woolworth Co., supra. The Respondent’s alternate method of calculating backpay would compare its painters’ actual average earnings, without regard to such em- ployees’ actual length of employment, to arrive at what the discriminatees would have earned over a substantially longer amount of time. This alternate method is not a more accurate method of computing backpay. The Respondent’s alternate contention that backpay should end at the completion of the job from which the discriminatees were fired is rejected for the reasons set forth in this decision, including the failure of the Respondent to prove that the discriminatees would have been terminated at the end of that job. A. Robert Meade The Respondent maintains that over the 5-year backpay pe- riod, Meade earned more than he would have earned if he had remained with the Respondent, and that backpay should not be awarded to Meade because it would result in a windfall to Meade. The Respondent cites Ogle Protection Service, 183 NLRB 682 (1970), as the only authority in support of this con- tention. In Ogle Protection Service, the Board found that the Wool- worth formula for awarding backpay did not apply where the loss suffered by the employees did not arise from or involve a cessation of employment status or interim earnings. In the pre- sent case, those are precisely the bases on which the discrimina- tees’ backpay has been calculated. Accordingly, Ogle Protec- tion Service is inapposite, and the General Counsel’s calcula- tion of backpay on a quarterly basis, consistent with F. W. Woolworth Co., supra, is proper. The Respondent also contends that Meade moved to Louisi- ana, possibly as early as 1996, and would not have been avail- able to work for the Respondent since the time he moved. This claim is nothing more than a supposition by the Respondent. There is no evidence that Meade would not have returned if the Respondent had properly offered to reinstate him. The Respon- dent elected to wait over 5 years before offering to reinstate Meade. Certainly, the longer an employer waits before offering to reinstate a fired worker, the less likely it will be that the fired worker would be willing or able to accept the reinstatement offer. When the Respondent finally did send an offer of rein- statement to Meade on April 1, 2002, Meade did not reply and, effectively, declined the offer. The Respondent could have sent the reinstatement offer at any time before or during the ex- tended proceedings in this case. Meade should not be penalized because the Respondent elected to wait absolutely as long as possible, thus helping to insure that the offer would not be ac- cepted. Nor should Meade be penalized because he elected to take a job in another state with his employer, a job that elimi- nated much of the backpay that would otherwise be due, and which has redounded to the benefit of the Respondent. There is no evidence in this case that Meade would not have accepted a proper reinstatement offer at any time before April 1, 2002. Accordingly, the backpay due to Meade should not be reduced because he moved from the area in order to pursue employ- ment. B. Warren Hull The Respondent maintains that because no representative of Region 9 of the Board has been in contact with Hull since 1996, and because there is no evidence that Hull made reasonable efforts to find interim employment, backpay should be denied. This claim misconstrues the respective burdens of the parties to this proceeding, and is rejected. The Respondent bears the burden of proving any reductions in gross backpay. Florida Tile Co., supra. Thus, the failure of proof cited by the Respondent was its own failure, not the Gen- eral Counsel’s. The calculation of Hull’s backpay was made in accordance with the Board’s Casehandling Manual, Compliance Proceed- ings, Section 10548.4. By using this section, rather than Section 10621.7, the compliance officer gave the Respondent the bene- fit of backpay reduced by average interim earnings. The com- pliance officer was not required to give the Respondent this benefit. On the other hand, the Respondent claims that it is not liable for any backpay, and this claim is rejected. In Starlite Cutting, Inc., 284 NLRB 620 (1987), the Board set forth the procedure to be followed where a discriminatee has not been found. The backpay due to such discriminatee is to be placed in escrow with the appropriate Regional Director and is to be held for a period not exceeding 1 year. C. Charles Crisp When Crisp applied to the painters union, the union was ac- cepting only apprentice applications. Accordingly, he accepted PAINTING CO. 51 the apprenticeship, but he was never classified as an apprentice and was never paid as one. The Respondent paid Crisp at its journeyman rate, but contends that Crisp’s backpay should be calculated at the apprentice rate. The Respondent contends that Crisp perpetrated a fraud on it by accepting pay at the journey- man level when, in fact, he was only an apprentice. This con- tention is rejected. The Respondent failed to produce any evidence to show or explain the basis for its own decision to pay a painter at the journeyman level. There is no evidence to show that the Re- spondent’s decision to pay Crisp at the journeyman level was improperly induced by Crisp or was otherwise inappropriate or contrary to company policy. The Respondent paid Crisp at the journeyman rate when it unlawfully fired him from his job. It paid him at the journeyman level based on its own criteria, and it has failed to articulate that criteria, much less demonstrate that Crisp’s rate of pay was improper or that its decision was fraudulently induced by Crisp. Whether Crisp was a journey- man painter in the union local where he was a member is ir- relevant without knowing what criteria the Respondent used in deciding to pay a painter at the journeyman level. The Respon- dent’s contention that Crisp’s backpay should be computed using the rate of pay for an apprentice is rejected. Crisp failed to apply for a job with a prospective employer, BCI Construction and Engineering, Ltd., and he noted that the job was out of state. The evidence does not disclose in what state this prospective job was located. The evidence also fails to disclose the type of work involved, the pay, or any other condi- tions or qualifications of this prospective job. A discriminatee is not required to go beyond reasonable exertions in an effort to mitigate backpay liability. Fabi Fashions, Inc., 291 NLRB 586 (1988). In the absence of any evidence showing the nature or location of the prospective job with BCI, the Respondent has failed to meet its burden of proof that Crisp’s failure to apply for the job demonstrates his failure to make reasonable exer- tions to mitigate. D. July 3, 2000 Letters The Respondent contends that the July 3, 2000 letters to the discriminatees were valid, unconditional offers of reinstatement which should terminate the backpay period. Backpay terminates or is tolled by a valid offer of reinstate- ment to a substantially equivalent position. Holo-Chrome Co., 302 NLRB 452 (1991). A reinstatement offer to a discriminatee must be specific, unequivocal, and unconditional in order to toll backpay. Id. In Holo-Chrome, the Board concluded that a letter from the employer stating, “I want to talk with you [i.e., the discriminatee] regarding a job opportunity for which you are qualified,” did not mention a specific job or job classification, and did not make an express offer of a job to the discriminatee. 302 NLRB at 454. Accordingly, the Board held that the letter did not end the backpay period. Similarly, the Respondent’s July 3, 2000 letter did not mention a specific job or job classifi- cation, and did not make an express offer of a job.24 Accord- ingly, the Respondent’s July 3, 2000 letters did not end or toll the backpay period. E. April 1, 2002 Letters By letters dated April 1, 2002, the Respondent offered to re- instate each of the discriminatees to their former jobs as jour- neyman painters. The General Counsel concedes, and I con- clude, that these letters terminated the Respondent’s backpay obligation to the discriminatees.25 Accordingly, I find and con- clude that the backpay period for the discriminatees covered the period January 2, 1996 to April 1, 2002. [Recommended Order omitted from publication.] 24 GC Exhs. 3 and 4. Because the July 3 letters so clearly fail to sat- isfy the requirements of a specific, unequivocal, and unconditional offer of employment that would toll backpay, I will not address the Respondent’s actions in the timing and mailing of its letters. The letters are dated July 3, 2000, obviously the day before a major holiday in the United States. However, the letters were not postmarked by the United States Postal Service until July 5, 2000. The letters imposed a deadline of July 7 for reply. Nor need I consider the Respondent’s statement to Pratt that the Respondent did not have prevailing wage work available to Pratt and that the job referred to in the letter paid between $7 and $8 per hour. 25 The General Counsel properly could have advocated a backpay ending date of May 1, 2002, because the offer of reinstatement afforded the discriminatees 30 days within which to make their decision. East- ern Die Co., 142 NLRB 601 (1963). However, I cannot say that the decision to advocate an ending date of April 1 was improper or that it results in an injustice under the circumstances of this case.