DOL memo 12-3-07

Office of Labor-Management Standards

Paperwork Reduction Act

On July 2, 2007, the Department of Labor published a Final Rule revising the Form LM-30 (Labor Organization Officer and Employee Report). 72 Fed. Reg. 36105. In Section F, the Department conducted an analysis under the Paperwork Reduction Act of 1995, 44 U.S.C. 3501 (“PRA''). See 5 CFR 1320.9.

Approximately 75 days after publication of the final rule, the Office of Management and Budget began receiving comments on the information collection requirements in the Final Rule. A letter from the AFL-CIO, dated September 21, 2007, asserted that the “projections severely underestimate the impact of the new rule.” The AFL-CIO focused on the portion of the rule providing that union stewards who work for more than 250 hours for the union while being paid by the employer, under a union-leave or no-docking policy, are union employees who must report these earnings. It said this change alone “will increase the number of annual LM-30 filings from just over 4,000 to over 80,000.” The AFL based this number on its belief that the rule required that union stewards who earn even an hour’s pay under a union-leave or no-docking policy would be subject all of the rule’s requirements, meaning that they would “need to file an LM-30 for all other covered transactions.” This error may explain why its estimate of the number of filers is overstated.¹

The AFL-CIO letter also contended that the Department “significantly underestimated” the burden on individual filers because it did not “specify precisely” which financial transactions are reportable. The letter asserted that the Department was still considering whether “personal financial transactions, such as credit card bills, mortgage loans, or other consumer loans” are reportable. The letter did not provide burden figures relating to these transactions, and subsequent to this letter, the Department clarified that credit card balances and interest on checking and savings accounts and certificates of deposits were not reportable. See Frequently Asked Question (FAQ) 70.

The letter stated that the Department’s estimate that the form would take 120 minutes to complete “cannot withstand scrutiny.” The letter said the process of determining whether a business is one whose payments are reportable would be “time consuming” but it did not offer an estimate of the amount of time it would consume.

Finally, the AFL-CIO’s letter states that the rule imposes a burden on all stewards to keep track of the time they devote to union business while being paid by the employer to determine whether they have exceeded the 250 hour threshold. The letter provides that “thousands of individuals may determine ultimately that they have no reportable transactions but only after they have spent many hours keeping records and doing research.”

A letter dated September 20, 2007, from Senator Edward M. Kennedy and Chairman George Miller was sent to OMB, stating that the Department “seriously underestimate[d]” the number of filers and the time required to complete the revised form. The letter posits that “tens of thousands” of individuals will have to file. The letter asserts that the rule “requires officers and employees to report personal financial transactions – such as credit card balances, consumer loans, and mortgages – with financial institutions that do any business with the union or that do more than 10 percent of their business with employers whose employees are represented by the union.” The letter states that determining whether a payment is reportable “will clearly take far longer than the 20 minutes allocated for recordkeeping or the 5 minutes allocated for reporting.” Like the AFL-CIO’s, this letter did not account for the Department’s guidance that credit card balances and interest and dividends from bank deposits were not reportable.

On October 26, 2007, Senator Kennedy and Chairman Miller submitted a second letter to OMB. The letter relied on a study by Professor John Lund for the proposition that “at least 120,000” union stewards work under no-docking or union leave policies, and would be required to read the Form LM-30, its instructions, and the Frequently Asked Questions (FAQs) posted on the OLMS website. The letter also relies on Lund’s conclusion that 57,556 of these stewards would work more than 250 hours and would thus be required to file a report. Professor Lund is a professor at the School for Workers, University of Wisconsin-Extension.

The Congressional letter states that the Department’s burden estimate failed to include the time each union steward would devote to keeping records of time spent doing union work while receiving payment from the employer. This requirement will “impose an enormous time burden on more than 100,000” union stewards. The letter states that for “active stewards who do more than 250 hours of no-docking work annually, keeping these detailed time records could take several hours.”

The letter observes that the burden estimate failed to take into account the time necessary to read the FAQs, and estimates that “reading this FAQ document alone will require at least one to two hours.” The Department also underestimated, according to the letter, “the time required for union members to (1) contact financial institutions to ask them how much business they do with other unionized employers, (2) contact the Department of Labor for assistance, and make good faith estimates.” This time needed, the letter estimates is “well above” 20 minutes.

Finally, the Congressional letter states that the Department “significantly increase[d] the burden on union members” when the Department explained that the 250 hour threshold must include hours for which money was paid directly to an individual by the union to cover wages lost when the individual left work to perform union duties.

As mentioned, the letter attached a document prepared by Professor Lund. Lund submitted comments during the Form LM-30 rulemaking’s comment period (Document 993), but he has never provided the Department any burden estimates concerning the Form LM-30.

In his post-comment submission to OMB, Lund estimates that “at least 121,923 shop stewards” are newly covered by the revised Form LM-30, that between 55,689 and 73,481 stewards receive payments from employers for union work under a collective bargaining provision, and that 57,556 stewards would work 250 hours or more for the union while being paid by the employer or the union.

Lund begins with a 1980 study from the Department’s Bureau of Labor Statistics, which reports the number of collective bargaining agreements that cover more than 1,000 union members and provide “some or all” stewards employer pay for “some or all” of their union work. Major Collective Bargaining Agreements: Employer Pay and Leave for Union Business, U.S. Department of Labor, Bureau of Labor Statistics, October 1980. He also relies on Leo Troy and Neil Sheflin’s Union Sourcebook: Membership, Structure, and Finance Directory, 1^st edition, 1985, for information on union membership numbers and union density in various industries in 1980. Finally, he relies on The Outlook for Industry Output and Employment through 1990, The Monthly Labor Review, U.S. Department of Labor, Bureau of Labor Standards, August 1981, for employment levels in various industrial sectors in 1979.

Preliminarily, the Department observes that these documents do not contain data more recent than 1980 and no relevant predictions for conditions subsequent to 1980. Nor do they contain information, old or current, on the number of stewards, amount of union work performed by stewards, the number of stewards who work under a collective bargaining agreement, and the number who work outside of, a collective bargaining agreement. As Lund acknowledges, his study relies on several assumptions and an extrapolation from incomplete information compiled in 1980.

A summary description of Professor Lund’s analysis is necessary to explain why the Department largely rejects the study as a basis for accurate burden estimates.

Lund begins with the 1980 BLS report, Major Collective Bargaining Agreements, which he relies on to establish the number of workers covered by a major collective bargaining agreement (covering more than 1,000 employees) in various industries. This figure is 661,400. He also uses the study to determine how many of these workers work under CBAs that provide pay to “some or all” stewards for “some or all” time spent in grievance handling.² He finds that, in 1980, 3,784,800 of the 6,610,400 workers did so, concluding that “57.3 % of the union members were covered by CBAs that required at least some employer payment for shop stewards.” Although he acknowledges that “there are no data for this,” he assumes that 57.3% of all collective bargaining agreements, not just major ones, have provisions providing for employer pay for union work.

Lund then determines the number of workers covered by a CBA that contains a specific provision that the employer will not pay stewards for time spent working for the union. He states that 7.2% of these workers are in this category. Based on these assumptions, Lund concludes that the remaining 35% of the workers have a silent CBA, with no terms allowing or prohibiting employer payment for time spent on grievances.

Lund then seeks to derive the total number of union stewards in 2006. He totals 2006 BLS data on union membership in various industries, determining that there are 6,417,000 union members in these industries. Lund Study, figure 3, p. 9. He then assumes that unions have one steward for every 50 employees, resulting in 128,340 stewards in these industries in 2006. Lund Study, pp. 5, 9.

Lund then seeks to estimate the number of stewards in 2006 who will work under a CBA that provides for employer pay for union work. Lund Study, pp. 10-13. Again, he uses an assumption that the percentage of major CBAs with these provisions in 1980 would mirror the percentage of major CBAs in 2006, and that non-major CBAs would contain these same terms in the same proportions. Lund Study, pp. 4, 10.

Because he determined that 57.3% of 1980 workers under the major CBAs studied by BLS worked under these provisions, he prepared to apply this percentage to the number of 2006 workers under major CBAs in 2006. Lund Study, p. 10. As he does not have data on the number of 2006 workers under a major CBA, he makes his own estimate by determining the number of workers in 1980 in each industry, the percentage of such workers that were in union worksites, and the number, provided as a percentage, of workers in those union worksites under a major CBA. Lund Study, p. 11. He concluded that there were 128,326 stewards. Applying his 57.3% reduction, he concluded that there were 73,481 stewards “covered by CBA-mandated employer paid time.” Lund Study, p. 9.

In conducting this analysis, Lund recognized that the industry sectors for which he had 1980 CBA data were not the same as the sectors for which he had 2006 union membership data. He was required to determine which 2006 industry sectors were comparable to the 1980 sectors, and make adjustments. His explanation defies easy summary; he writes, “Note that because of the lack of compatibility between 2006 and 1980 data, it was necessary to use the percentage of CBAs in “other services” in the “information” and “education and health services” sectors. “The total number of stewards who would now have mandatory CBA provisions for employer-paid grievance time is 55,689.” Lund Study, p. 9.

Having reached the conclusion that the range for the number of stewards working under CBAs with union-leave or no-docking provision is between 55,689 and 73,481, Lund seeks to estimate the number of these stewards who would work 250 or more hours per year in 2006. In essence, his theory is (1) the percentage of union members covered by a major CBA in 1980 would mirror the percentage of members in 2006, (2) the percentage of CBAs with employer pay for union work in 1980 would mirror the percentage of CBAs with comparable provisions in 2006, (3) unions had one steward for every fifty members in both 1980 and 2006, and (4) all workers covered by a major CBA would be likely to have stewards who work for the union while receiving employer pay for at least five hours a week.

As Lund did not have data on the percentage of union members in 1980 covered by a major CBA, he extrapolated a figure. He used BLS data on employment in various industrial sectors, used data from the Union Sourcebook on the percentage of employees unionized (union density), multiplied these figures to derive the number of unionized workers, and compared that total to the number of members covered by the major CBAs studied in the 1980 BLS report. Lund then assumed that the resulting percentage (members covered by major CBAs to all unionized workers in 1980) reflected the 2006 composition of unionized workers under a major CBA to all unionized workers. In making these calculations, he again adjusted for the fact that his 2006 data on industry sectors was not the same as his 1980. He concluded, however, that 43.5% of all unionized workers in the identified sectors were covered by a major CBA in 2006.

Lund completes his analysis by calculating the number of union members covered by a major CBA in 2006. He takes the percentage of union members under a major CBA in 1980 to all unionized workers in 1980, identified as 44.8%, and applies that percentage to the total number (6,417,000) of union members in the various industries in 2006. This figure is 2,877,814 (6,417,000 / 0.448 = 2,877,814). Using his formula of one steward to each 50 members, he concludes: “we get 57,556 stewards who would have 250 or more hours per year to report.” Lund Study, p. 13.

The Department is unable to accept the Lund analysis. As the preceding paragraph demonstrates, Lund has erroneously characterized the estimated number of stewards covered by a major bargaining agreement (57,556) as the number who would receive payment under the agreement. But as the 1980 BLS study shows (p. 32, Table 1), and Lund himself appeared to earlier recognize (p. 7), only a fraction of the major collective bargaining agreements have a provision calling for employer pay for union work. Yet Lund treats every steward in 2006 under a major CBA as being under such a provision. Lund Study, p. 13. One following his train of thought might have predicted that the 57,556 figure would be reduced by 57.3%, reflecting the composition of contracts with employer pay for union work to all contracts, but that would be guesswork.

Even had Lund followed through with the logic of his suppositions, the analysis would still have been wrong. A second, equally serious error concerns his understanding of the employer-payment provision in the CBAs. The collective bargaining agreements identified by BLS as providing for employer pay for union work provide for “full or partial pay to some or all union representatives.” BLS Study, p. 32, Table 1. Lund recognizes this (p. 6), but nevertheless makes determinations premised on the ground that all stewards under such an agreements would have a right to such pay (and would receive it for more than 5 hours of work per week). Lund Study, pp. 12-13. The BLS study cannot reasonably be read this way. A CBA characterized as permitting some or all stewards to receive compensation for some or all of their union worksite activity does not justify an assumption that all such stewards would receive any such compensation nor an assumption regarding the amount of compensation that would be available either collectively or on a per steward basis. Indeed, the BLS study shows that the provisions granting employer pay for union work were not as generous as Lund projects. It quotes one agreement as stating “the company will pay stewards for lost working time … but not to exceed three hours per week” and another as providing that “[t]he company will also pay each of 20 stewards for time spent during their regular shift-hours, not to exceed 2 hours per week.” BLS Study, p. 8.

Lund’s study also relies on other improbable assumptions, including the assumption that large unionized companies are comparable to small unionized companies. Lund Study, p. 4 (“a second required assumption deals with extrapolating the percentage of CBAs with employer-paid time from major CBAs to all CBAs in 1980.”) BLS data indicate that, in fact, smaller companies overall provide fewer benefits and lower pay scales than large companies. Although the publicly available BLS data does not tabulate the data by union and size, the differentials reported by BLS by company size suggest that the assumption made by Lund is unsound. In this same vein, it seems more probable than not that employer-paid steward time would have been reduced or eliminated in smaller workplaces over the last 25 years as competition has forced cutbacks in labor and associated costs, not remained static as Lund’s study would suggest.

Other aspects of the study do not necessarily suggest major flaws, but they do raise questions as to the carefulness, thoroughness or reliability of the study. Economic studies have shown large changes in the structure of the U.S. economy, from industrial to service. Lund does not account for this shift, and to compound the problem, he makes unexplained and dramatic changes to his data purportedly in order to make information about 1980 industrial sectors comparable to the sectors tracked in 2006. Lund Study, p. 12. Indeed, one chart apparently lost track of 194,100 hotel and restaurant workers. Compare figure 2, p. 8 with figure 5, p. 11. The study also appears to confuse data on workers with data on members. For example, Figure 1, p. 7 states that there are 6,610,400 “workers” under a CBA with employer paid grievance time while Figure 6, p. 12, states that there are 6,610,400 “covered union members” under such a provision. It is, of course, not safe to assume that workers within a bargaining unit are necessarily union members. Finally, the Lund Study erroneously assumed that a steward who works 250 hours must report when the rule requires those who exceed 250 hours to report. Lund Study, p. 2; Form LM-30 Instructions, p. 10. Had Lund correctly understood the 250 hour threshold, his analysis would surely not be much changed, but his misapprehension of this fundamental point casts doubt on his estimate of other numbers that cannot be so easily checked.

Lund relies primarily on the 1980 BLS study, a document that the Department discussed at great length in its Paperwork Reduction Act Analysis and found to be of little use. 72 Fed. Reg. 36155. Lund is notably silent on this point, allowing the reader to assume both that the Department was unfamiliar with existing data on the burden issue and that there is no counterargument to relying on data more than 25 years old to predict contemporary practices. As mentioned above, the Department did not receive comments during the comment period on which it could rely to determine the number of union stewards affected by the rule and the number that would file. Faced with this dearth of data, but recognizing that the rule would increase the recordkeeping and reporting burden, the Department tripled the overall burden estimate from that contained in the Notice of Proposed Rulemaking. 72 Fed. Reg. 36156.

Although the Department has concerns about many aspects of the Lund study, as discussed above, it may be useful to estimate how the Department might have used the Lund study had Lund submitted timely comments to the Department on the burden associated with the changes to the Form LM-30. This analysis follows and is used to estimate the burden on stewards in the spreadsheet requested by OMB:

As mentioned, Lund acknowledges that the 1980 BLS study is silent with regard to the number of stewards in 1980 (much less in 2006), but he assumes, based on his experience as a labor academic and a stewards trainer (but without providing a basis to evaluate the breadth of his experience on this point), that one steward exists for every fifty members. He further assumes (again without any basis to evaluate his assessment) that this ratio prevailed in both 1980 and 2006. Moreover, he fails to mention that the BLS document on which he chiefly relies quotes a representative CBA as stating “there shall be a maximum ratio of 1 steward for each 75 employees acting within the bargaining group.” BLS Study, p 7.

The Department assumes, for this exercise, Lund’s estimate of 6,417,000 union members in the pertinent industries, and his calculation that there is one steward to every 50 members, resulting in 128,340 stewards. The Department also accepts his proposition that 57.3% of these stewards work under provisions permitting employer pay for union work, resulting in 73,841 stewards. Lund Study, p. 9.

The Department does not, however, accept Lund’s unsupported premise that more than a negligible amount of union stewards take employer pay that is not pursuant to a collective bargaining agreement. A steward who does not have his or her union-leave or no-docking rights secured by a binding collective bargaining agreement is in a position of weakness vis-à-vis the employer. As the benefit is not a contractual right, an unscrupulous employer can at any time pressure the steward, for example, to withdraw a grievance or overlook a violation of the contract because the employer can unilaterally take away the steward’s right to this valuable leave. For this reason, unions are likely highly motivated to secure this right in a collective bargaining agreement. Furthermore, there appears to be in the labor movement a concern that these arrangements – employer pay for union work not pursuant to a collective bargaining agreement – raises legal issues. The AFL-CIO’s comment asserted that “to be lawful under LMRA [Labor Management Relations Act, § 302(c)(1)], union-leave and no-docking payments must be made pursuant to a collective bargaining agreement.” Comments of the AFL-CIO, p. 25. The Department takes no view on this issue, and only the Department of Justice is authorized to speak on the scope of section 302 of the Labor Management Relations Act. But if a leading federation of labor organizations believes the arrangements raise legal issues, it is likely that the practice is not widespread.

As the BLS study holds that contracts that permit employer pay for union work “include full or partial pay to some or all union representatives,” BLS Study, table 1, p. 32, the Department does not assume, as Lund did, that all stewards will receive such pay. To give effect to the contract terms that provide some or all pay to some or all stewards, the Department estimates that one half of the stewards will be covered and that one half of those stewards will receive payment, reducing the number of stewards to 18,460.

The Department acknowledges that this figure is an estimate, as are additional figures discussed below. As mentioned, in the Paperwork Reduction Act analysis in the Final Rule, the Department carefully scrutinized each relevant comment, studied all relevant publications brought to its attention, and conducted its own research. The Department has also closely studied the post-final rule submission to OMB.

In this instance, and in others, the Department has relied on its judgment and expertise, based on its 47-year history working closely with unions. In fiscal years 2006 and 2007 alone, the Department’s Office of Labor-Management Standards (OLMS) audited 1,512 local, intermediate and international unions, investigated 247 complaints of undemocratic union officer elections, supervised 59 rerun elections, and investigated 745 allegations of criminal violations involving unions. Personnel in each of its 20 district offices have developed personal relationships with union officer, employees and members in their jurisdiction. OLMS has been administering the Form LM-30 reporting requirements since 1963 and is deeply familiar with reporting levels, activities, and practices.

In this regard it is also important to note that although the Paperwork Reduction Act allows OMB to approve a collection of information for up to three years (See 5 CFR 1320.12(e)(1)), the Department is requesting PRA approval for the revised Form LM-30 only until November 30, 2009. Virtually all, if not all, union officers and employees use the calendar year as their fiscal year. Therefore, the first revised Form LM-30 reports (covering 2008) are required to be filed between January and March 31, 2009. This will result in a full year of filing the revised Form LM-30 before the Department prepares a new paperwork package for public and OMB review. At the end of that year, for the first time, the Department will have information on the number of reports filed, the number and sources of transactions reported, the number of reports from stewards, comments from filers and others on their experiences with the Form and a documented level of burden. This will provide the Department and the public with a full year of filing experience with the revised Form LM-30 before the Department prepares a new paperwork package for public and OMB review. At that time, the burden estimate, if it is proved inaccurate, may be adjusted.

As mentioned, the Department estimates that there are 18,460 stewards who will receive some employer pay for union work. Of this universe, the Department predicts that 90% will work 250 hours or fewer, resulting in 16,614who will have no filing obligation and 1,846 who will. The Department finds it unlikely that more than 10% of this population works for more than 250 hours. The BLS study shows that the amount of paid union work is often limited, and the expense involved in paying a union steward for more than six weeks a year away from the job for which he was hired militates against a higher number. ³

The Department estimates that 90% of the 16,614 non-filing union stewards will have no additional recordkeeping beyond the records they must already keep. Although the absence of data has made prediction difficult, it is the Department’s judgment that employers do not ordinarily permit union stewards to perform union work on employer time without any accounting. Assuming some present-day relevance of the 1980 BLS study, collective bargaining agreements regularly limit the amount of time a steward could perform union work on company time. See BLS Study, Table 1, p. 32. These terms could not be enforced absent recordkeeping and employers have a direct financial interest in ensuring that the limits are honored. The profit motive is likely to create an incentive by employers to require recordkeeping in order to monitor the time spent in such activities and to ensure that it does not exceed the contractual limit. Further, some stewards, whether they are already keeping records or not, will know from experience at the beginning of the year that they will not work more than 250 hours (i.e., that they rarely spend a half-day every week on union business), and that no records will need to be maintained for Form LM-30 purposes. Thus, 90% of the union stewards likely are already keeping the necessary records and therefore, additional recordkeeping burdens are not properly included here.

The Department estimates that 5% of the 16,614 non-filing stewards (831 stewards) will need 120 minutes each in keeping records in preparation for the possibility of filing, for a total of 1,662 hours (831 x 120 = 99,720 minutes or 1,662). These stewards will work for employers that do not require tracking of union work. At the end of each week, they will have to write down the hours they left their regular job and worked for the union. They will work short time periods of union work, requiring the tracking and adding of multiple entries.

The Department estimates that the remaining 5% of the 16,614 non-filing stewards (831 stewards) will need 60 minutes each in annual recordkeeping (831 x 60 = 49,860 minutes or 831 hours). These individuals will work fewer periods of union leave, requiring the tracking and adding of multiple entries.

In estimating the burden for both these groups, the Department has kept in mind that by definition they are tracking fewer than 250 hours. Further, they need not create any records of the kind of work, the involved union member, case or tracking numbers, the forum, the resolution of the matter, or any other facts. They need note only the amount of time worked.

The Department estimates that 100% of the 16,614 non-filing union stewards will need ten minutes reading time each to determine the sole fact relevant to them: that stewards who receive fewer than 250 hours under a collective bargaining term need not file. This results in a burden of 166,140 minutes or 2,769 hours.

Of these 1,846 who will need to file, the Department believes that 90% (1,661 stewards) will keep records in the ordinary course of business. In addition to the reasons discussed above, union representative on full time union leave will have no recordkeeping burden, as they need report only their yearly salary for full time work, a record maintained by the employer and released to the steward for tax and other reasons. As result, these 1,661 filers will have no or negligible additional recordkeeping burden.

For the remaining 185 stewards who will file, the Department assumes that they will spend 120 minutes over the course of the year tracking time they would not otherwise have tracked, resulting in a yearly additional burden of 370 hours (185 stewards x 120 minutes = 22,200 / 60 = 370). The Department estimated 120 minutes of recordkeeping for these filers for the same reasons, discussed above, it determined that a percentage of non-filers would spend 120 minutes tracking time.

The reading time for the 185 stewards who will file is 55 minutes. They, like all other filers, will be required to review the form and the instruction in their entirety. See 72 Fed. Reg. 36157. This results in a yearly additional burden of 170 hours (185 stewards x 55 minutes = 10,175/ 60 = 170).

The Department did not include in this burden estimate the time spent reading the Frequently Asked Questions on Form LM-30. The FAQs were not issued at that same time the rule was. They were issued subsequently. The FAQs were created as compliance assistance and in all but a few instances were created by restating portions of the instructions into a question and answer format. FAQs can be more accessible to new filers, and they can guide individuals to their areas of concern quickly. The FAQs on the whole, however, provide no new information and, therefore, do not need to be read in order for a filer to complete an accurate and timely form. In a few instances, the FAQs make clear a point that was ambiguous in the rule. This is the sort of guidance agencies regularly provide and are not, to our knowledge, included in burden analyses, particularly when as here, they are issued after the Final Rule.

The additional yearly reporting burden relating solely to the special circumstances of stewards is then, in total 5,802 hours (1,662 +831 +2,769 + 370 + 170 =5,802.)

As the Final Rule’s total annual reporting and recordkeeping burden for both filers and those who review the form but determine that a report need not be filed will be 112,691 hours (13,832 (hours for filers) + 98,859 (hours for non-filers)), the addition of the 5,802 hours attributable to stewards alone would not materially change the analysis, particularly as DOL has already increased the burden due, in part, to the new rules governing stewards. 72 Fed. Reg. 36156. It should also be kept in mind that, although the stewards rule is new and the layout of the form is different, the vast majority of the Form LM-30 reporting requirements have not changed. Further, the Department added a $250 de minimis filing threshold, a rule that gifts or payments of $20 or less need not be counted when calculating the $250 de minimis threshold, and that hospitality received at widely attended gathering will ordinarily now no longer be reportable. All these changes will reduce burden. As a result, this rule will not bring about a dramatic increase recordkeeping or reporting burden.

Another issues, raised by the September 20, 2007, Congressional letter, but not Professor Lund, asserts that the rule “requires officers and employees to report personal financial transactions – such as credit card balances, consumer loans, and mortgages – with financial institutions that do any business with the union or that do more than 10 percent of their business with employers whose employees are represented by the union.” The letter states that determining whether a payment is reportable “will clearly take far longer than the 20 minutes allocated for recordkeeping or the 5 minutes allocated for reporting.”

The 20 minutes estimated for maintaining and gathering records is not unreasonable. See 72 Fed. Reg. 36157. Like the AFL-CIO’s letter, the congressional letter did not account for the Department’s guidance that credit card balances and interest and dividends from bank deposits were not reportable. After more than 40 years experience with Form LM-30, it is clear that most forms include few transactions. Because transactions that are required to be reported are rare most recordkeeping will be no more than 5 minutes, e.g., the amount of time needed to note the receipts of sports tickets, a cash payment, leather jacket, weekend golfing trip, and so forth. Transactions with vendors to the employer are especially rare, particularly in light of the guidance that credit cards, ordinary transactions, etc. are not reportable. It will also ordinarily be quite obvious – without any record gathering at all – that a financial institution is so large that it could not possibly be receiving 10% of its business with the employer. Furthermore, in OLMS’ experience, most reportable payments that do relate to a business dealing with the union or doing 10% of its business with the employer will involve businesses owned and operated by the union officer. In these instances, there will be no barriers to gathering records. Moreover, union officer and employers have been required to report payments from financial institutions and vendors that buy, sell, lease or otherwise deal in “substantial part” with the employer of the union members since the LMRDA was enacted in 1959. The Department in this rulemaking merely defined “substantial part” to mean that the vendor has to receive ten percent of its annual receipts from the employer. This minor clarification cannot reasonably be expected to raise the paperwork burden.

Further, if Lund’s estimate of 128,326 stewards is used without any reductions and assume, for the reasons discussed above, that 90% of this number would need 10 minutes or less to query employers and gather other records, while the remaining 10% would need to spend an hour doing so, the average is very close to the Department’s own estimate of 20 minutes. (128,326 x .10 = 12,833 x 60 minutes = 769,980 minutes; 128, 326 x .90 = 115,493 x 10 minutes = 1,154,934 minutes; 769,980 + 1,154,934 = 1,924,914/128,326 = 15 minutes.)

In sum, the Department remains of the view that it correctly and accurately estimated the burden associated with the final rule’s provisions relating to union stewards and that Professor Lund’s study does not present a credible challenge to the Department’s analysis. Furthermore, the information that can be gleaned from his study, properly viewed, is largely consistent with the Department’s own estimate.

1 The FAQs clarify the application of the rule to stewards who do not reach the 250 hour threshold. A steward who works fewer than 250 hours need report nothing, unless he or she is working under a union-leave or no-docking policy not authorized by a collective bargaining. The stewards who work fewer than 250 hours, but not pursuant to a collective bargaining agreement, need report only the union leave or no docking payments. See FAQs 40-43.

2 The BLS study explains that “pay and leave provisions may not become widespread in agreements because … the union prefers to compensate representatives to avoid any possible conflict of interest” Major Collective Bargaining Agreements, p. 4.

3. Although some argue that employers benefit from union stewards quickly resolving workplace grievances before they develop into full-blown disputes, union-leave and no-docking policies are demands that unions, not employers, bring to the bargaining table. The potential utility of union-leave and no-docking policies to employers does not suggest a higher percentage of stewards working more than 250 hours.

14