Union Spending in Michigan: A Review of Union Financial Disclosure Reports

Administrative and overhead costs for unions are unusually high, and spending on workers' representation — the core task of labor unions — is correspondingly low, according to the findings in "Union Spending in Michigan: A Review of Union Financial Disclosure Reports." The study examined national, state and local data from federal LM-2 reports for six prominent unions and used the information to create estimated budgets. Director of Labor Policy Paul Kersey's review of union financial reports shows that less than half of a typical union member's dues go into representation which is nearly matched by the cost of overhead and administration. In contrast, a typical nonprofit will usually try to keep overhead and administrative costs at around half of what is spent on its core programs. Overall, six unions, all with a significant presence in Michigan, were examined: Teamsters, National Education Association/Michigan Education Association, United Auto Workers, Service Employees International Union, American Federation of State, County and Municipal Employees, and United Brotherhood of Carpenters and Joiners of America.

Executive Summary

It has long been Michigan's policy to promote private- and public-sector unionization and to ensure that labor unions have a source of funding. One mechanism for achieving this has been the enforcement of "agency fee" clauses in collective bargaining agreements. The advantage of a guaranteed revenue stream, however, ought to be coupled with some basic responsibilities. Union officials have an obligation not to abuse the privilege of collecting mandatory dues, and a responsibility to run operations that are efficient, transparent and focused on providing effective representation.

The residents of Michigan, home to some of the nation's most powerful unions, are in a key position to determine for themselves if unions are using their dues and agency fees appropriately.  Are unions acting as responsible stewards ofthe mandatory dues they collect from the men and women they represent? We examined fiscal year 2006 federal LM-2 forms from six unions with a strong presence in Michigan and used the data to create rough estimates of their spending on representation, politics, charity, overhead and administration.

Our examination indicates that a relatively small amount of union dues money is actually used to represent workers. At best, the picture that emerges from many LM-2 forms is one of bloated, directionless union organizations with excessive overhead and administrative costs. These findings call into question the entire rationale behind the state's policy of allowing unions to negotiate for agency fee clauses in collective bargaining agreements. Even if one accepts the premise that individual workers should be forced to support a union — regardless of how they feel about it — in order to ensure that the union has adequate resources for effective representation, it is not evident that Michigan unions need all the money they receive in mandatory dues in order to accomplish this.

The LM-2 forms are not without shortcomings: loose definitions and the absence of independent  verification, combined with a strong motivation for union officials to obscure political activity and inflate representation, have led to a number of irregularities. For instance:

• Three Teamsters locals in Michigan reported between $60,000 and $110,000 in "Steward Dues" paid to themselves and listed as a cost of representation. These steward dues payments made up close to 10 percent of representation spending for all three locals.

• The National Education Association claimed that contributions to the Economic Policy Institute, The Campaign for America's Future, The National Coalition on Health Care and America's Agenda were representational in nature.

• One Michigan UAW local listed $10,000 in softball field rentals and over $30,000 in athletic uniforms as representation expenses. (For this and similar reasons, the LM-2 form from this local was not used in our estimates.)

• SEIU National Headuarters claimed payments to ACORN International, sponsorship of Air America and a donation to the Michigan Republican Party as representation expenses.

• One AFSCME local in Michigan claimed that 100 percent of its officer and staff time was spent on representation, with no time dedicated to internal union administration. Two other Michigan locals claimed no money or significant staff and officer time was spent on representation. (LM-2 forms from these locals were not used in our estimates.)

Our findings are subject to the limitations of our data; we do not claim that our estimates are particularly precise. But our review of LM-2 reports does allow us to make rough estimates of what a typical union budget might look like, and of the percentage of membership dues that go to various categories.

Among the six unions that we reviewed, representation made up as much as 55.2 percent and as little as 29.5 percent of total spending. On average, the unions we examined spent less than half of their funds on representation. General overhead took up more than a quarter; overhead and administrative costs combined made up nearly as large a portion of union expenditures as workplace representation.

By comparison, an average non-profit's spending on administration and overhead should be roughly half of its spending on core programs. LM-2 forms suggest the union movement is a fairly bloated one, with overhead and administrative costs far out of line with what is needed to support the core union mission of worker representation.

While the Department of Labor deserves considerable credit for updating LM-2 forms and rules, our review revealed numerous weaknesses in the system. An even greater oversight, however, is the failure of the state of Michigan to establish any financial reporting requirements for unions representing state or local government employees.

For decades, the policy in Michigan and other states without right-to-work laws has been to allow  unions to negotiate for and collect mandatory union membership dues or agency fees from the workers they represent. This significant curtailment of workers' rights of free association has been tolerated out of a belief that unions require a reliable stream of funds to represent workers effectively, and that it is unfair to allow individual workers to be "free-riders" who avoid bearing the costs of representation. These assumptions are essential to understanding union objections to right-to-work laws; if unions do not need these funds, the moral case against free-riders loses whatever force it might otherwise have.

A review of LM-2 forms filed by various unions undermines that rationale. Based on the information in these reports, it appears that the typical Michigan union could effectively represent workers on just over half of its current revenue. This could be accomplished by reducing political activism and reining in excessive overhead and administrative costs.

Were Michigan to enact a right-to-work law, empowering workers to withhold financial support from a workplace union, the effect on union revenue would probably be substantial. But unions would be unlikely to see a cut in revenues that approaches 50 percent. A revenue reduction of 10 to 25 percent would be far more likely. This amount should still provide unions adequate resources for representing their workers.

For decades unions have enjoyed privileges rarely extended to other private organizations. This report indicates that Michigan unions have been negligent toward, and in some cases may have abused, these privileges. There does not appear to be a compelling reason for these practices to continue. Michigan unions do not really need all the money they have been given.

Introduction

Unions remain a force to be reckoned with in Michigan's economy and government. Granted, union membership in Michigan (and the rest of the country), has been slipping, with the number of Michigan workers covered by union contracts falling from more than 1 million in 1997 to just 878,000 or so in 2006. Still, the 2006 figure represents slightly more than one-fifth of Michigan's work force, and our state remains one of the most heavily unionized in the country. Only Alaska, Hawaii, New Jersey, New York and Washington State have higher percentages of their work force covered by union contracts.[1]

Michigan workers have the right to expect that their unions will use their dues money to provide them with quality representation in the workplace. Whether through collective bargaining or pursuing grievances, unions are expected to protect workers and win for them the advantages of fair wages, generous benefits and safe working conditions.

Michigan's policy for decades has been to promote unionization and to ensure that unions have a steady stream of resources, in particular by allowing for the enforcement of "agency fee" clauses in collective bargaining agreements. These clauses mandate that all workers covered by the contract must either join the union or pay an agency fee equivalent to regular membership dues. While individuals can legally protest their agency fee payments that exceed a pro-rata share of the cost of representation, this process has proved to be convoluted and difficult to enforce. As a consequence, a union will generally receive the full dues amount from all workers it represents, whether individual workers are satisfied with the union's representation or not.[2]

At a minimum, this provision for a guaranteed revenue stream ought to have some basic responsibilities attached to it — namely that union officials will not abuse the privilege of collecting mandatory dues, and that their operations will be efficient and focused on providing effective representation. This responsibility to use mandatory dues wisely is grounded not in free-market theory — forced support of a private organization is contrary to a free economy — but in basic morality that should be recognizable to all regardless of ideology. Agency fees are permitted because it is widely believed  that unions need these funds to represent workers in the workplace; unions bear the responsibility of using these funds for that end and doing so thriftily. The same basic principle applies to government officials and the use of tax funds; tax money is to be used for public purposes, not the personal interests of government agents. Whether the source is taxes, dues or agency fees, misuse or excessive waste is a betrayal of trust.

For most of the period in which the National Labor Relations Act has been in effect, lax union financial reporting standards made it difficult to assess whether or not union officials met their ethical obligation to use union dues and agency fees, misuse or excessive waste is a betrayal of trust.

For most of the period in which the National Labor Relations Act has been in effect, lax union financial reporting standards made it difficult to assess whether or not union officials met their ethical obligation to use union dues and agency fees for the benefit of the men and women they represent. The LM-2, LM-3 and LM-4 forms required by the Department of Labor under the Labor-Management Reporting and Disclosure Act contained minimal financial information and, aside from officer compensation, said almost nothing about how unions spent their dues money.

This flaw was at least partially corrected by new LM-2 forms and rules issued by the U.S. Department of Labor in 2003. These new LM-2 forms, which cover the operations of nearly all international, national and regional union organizations, as well as many larger union locals, call for the itemization of union expenditures larger than $5,000, as well as a categorization of all union spending into functional categories such as representation, politics and lobbying, grants and charitable programs, overhead and administrative expenses. In addition, union officials and staff are expected to account for their own time using these same functional categories. While the new LM-2 forms and rules are not perfect, they do provide workers, employers and citizens with valuable knowledge of union operations and spending.

As a consequence, the residents of Michigan, a state with some of the most powerful unions in the nation, are in a better position to determine for themselves if unions are using their dues and agency fees appropriately. Are unions acting as responsible stewards of the mandatory dues money that they collect from the men and women they represent? Are unions in Michigan and around the nation focused, as they ought to be, on workplace representation? Are unions efficient or wasteful? In an attempt to answer these questions, we have carried out a systematic review of LM-2 reports filed for fiscal year 2006 by unions with a significant presence in Michigan.

Our examination of LM-2 reports indicates that a relatively small amount of union dues money is actually used for representing workers. While we cannot say for certain that a large proportion of union dues goes into political activism, there are indications that funds supposedly allocated for representation or charitable activities are instead redirected towards political or social activism on matters far removed from their members' workplaces. At best, the picture that emerges from many LM-2 forms is of bloated, directionless union organizations with excessive overhead and administrative costs.

These findings call into question the entire rationale behind the state's policy of allowing unions to negotiate for agency fee clauses in collective bargaining agreements. Even if one accepts the argument that it is appropriate for individual workers to be forced to support a union in order to ensure that said union has the resources it needs to be an effective representative, it is not at all clear that Michigan unions actually need all the money they receive in mandatory dues in order to represent those workers well.

Background: Agency Fees and the Duty of Fair Representation

Wherever collective bargaining takes place in Michigan, it is governed by one of three sets of laws. The National Labor Relations Act is a federal statute passed in 1935 covering nearly all private-sector employees.[3] Collective bargaining between counties or municipalities and their unionized employees is done under Michigan's Public Employee Relations Act.[4] The State of Michigan also bargains collectively with most of its employees under rules promulgated by the state's Civil Service Commission.[5]

All three labor laws share a common framework of obligations for unions, management and

individual workers, which can be summed up as a series of four mandates: mandatory recognition, mandatory representation, mandatory bargaining and mandatory support.[6]

Mandatory recognition means that when a union proves that it has the support of a majority of workers in a bargaining unit, the employer must recognize that union. An employer may recognize a union voluntarily, or the union's support may be confirmed after a secret-ballot election in which the union wins a majority of votes.

Mandatory representation means that once a union is recognized, whether voluntarily or after a secret-ballot election, all workers in the bargaining unit, including those who opposed unionization or supported another union, are legally obligated to accept that union's representation. At the same time, the union is legally obligated to represent all workers in that bargaining unit both in collective bargaining and in pursuing workplace grievances; this is referred to as the "duty of fair representation."

Mandatory bargaining refers to the obligation of both union and employer to bargain in good faith. This good-faith bargaining is more than ordinary honesty in making offers or fulfilling the terms of a contract in force: Both union and employer must be ready to demonstrate that they are bargaining with the intention of reaching an agreement.

Finally there is mandatory support. In the absence of a state right-to-work law, a collective bargaining agreement may include a provision for what is called "agency fees," "union security," or (by its critics) "forced dues." All three terms refer to the same basic thing: All workers covered by the contract must either join the union and pay the associated dues or remain out of the union and pay an agency fee, typically equal to regular membership dues.

Under a right-to-work law, a collective bargaining agreement cannot require an agency fee for nonmembers covered by a union contract. Opponents of right-to-work laws argue that this puts unions in the untenable position of being obligated to represent workers who may choose not to contribute to the costs of providing that representation. Union advocates claim that without the mandate of an agency fee clause, workers will be tempted to forgo paying union dues, leaving the union without the resources to represent workers effectively at the bargaining table or to pursue grievances aggressively. The extent to which this situation represents a burden on unions is dubious, however; union officials have yet to pursue legislation that would relieve them of the duty of representing non-dues-paying employees in right-to-work states.

Other Sources of Information on Union Spending

In theory at least, workers who refuse to join the union outright may have their union dues limited to a pro-rata share of the union's expenditures that are directly attributable to representation, such as collective bargaining, contract administration and grievances. Of particular importance, nonmembers are not obligated to pay for the costs of union political activity. This right of nonmembers to limit their dues payments in this way was established in a series of court cases which culminated in the U.S. Supreme Court's decision in Beck v. Communication Workers of America.[7]

While information on union spending patterns had been extremely limited for decades prior to the 2003 update of union LM-2 reports, union members and interested citizens had some glimpses at what lay beyond the curtain. The little that was visible provided grounds for concern.

The rights of nonunion members to limit their dues under Beck created one source of information on union spending; as part of the process of determining the correct reduction in dues, unions are expected to give nonmembers who invoke their rights under Beck an accounting of union spending on "chargeable" expenses (related to representation) and nonchargeable expenses (political, social and charitable). On rare occasions, nonmembers have successfully challenged a union's reckoning in court, and in the process more information was revealed as part of the discovery process.

It is very likely that the majority of nonmembers do not receive a refund that comes close to the amount to which they are entitled, however. Relinquishing union membership and requesting a dues reduction are separate steps, and while they can be taken simultaneously, not all workers who leave the union do so. The reductions that unions offer are likely to overstate the amount that they spend on representing workers: The National Education Association and its affiliates, including the Michigan Education Association, typically offers workers who object to paying for nonrepresentational activities a reduction of between 20 percent and 33 percent.[8] Our examination of MEA and NEA spending raises questions as to whether the NEA spends much more than 50 percent of its funds on representation. There is reason to believe that the actual amount could be even lower than that. In one case involving an MEA affiliate at Ferris State University, where an agency fee payer was able to litigate to the point where the court examined specific union spending items, "chargeable" expenses were limited to around 10 percent of regular NEA/MEA dues.[9]

In addition, there have been numerous accounts - sometimes anecdotal, sometimes based on union officials' own statements - indicating extensive union political involvement. In 2007, for instance, the Change to Win coalition, a group of seven unions that broke away from AFL-CIO two years earlier, announced that it was establishing a $14 million fund dedicated to "education campaigns centered around politics and several issues, notably including health care and the right to organize." The resolution creating the fund specified that the money would be used to "build a state-of-the-art coordinated political program to ensure the election of a pro-labor president in 2008 and pro-labor majorities in the Senate and House in order to pass the Employee Freedom of Choice Act."[10] In another case, in 2001 an IRS review of union finances showed that unions had failed to report millions of dollars of taxable political spending, including voter drives and advertisements. The AFL-CIO allegedly spent $35 million on political activities in 1996 but reported none of this to IRS.[11]

The Ethical and Legal Basis for Union Financial Disclosure

Because worker support for unions, or at least their collective representation operations, may be compelled, union spending is an entirely appropriate subject of public discussion and inquiry.

The collection of forced dues is not the only prerogative unique to union officials: Under state and federal law, union officials have authority to bargain wages and terms of conditions of employment for all workers in a bargaining unit, including union opponents, and also have the power to pursue or disregard their grievances. While unions are subject to a legal duty of fair representation and are expected to represent all bargaining unit members — not just union members — with equal diligence, courts and regulatory agencies tend to give union officials broad discretion in deciding what contractual demands to make and which grievances to take up. A worker may find that a representative has been chosen for him over his strenuous objections. That worker cannot opt out of an agreement reached by that union and cannot have the union replaced.

In no other situation are competent adults forced to accept the services of a representative without an individual right to choose or replace that representative. Unions have a remarkable level of legal power over bargaining unit members. It is only reasonable that unions should give a thorough accounting of how they use that power.

And because much of the authority wielded by unions flows directly from state and federal law, rather than from the voluntary support of workers, unions are accountable not only to their members but to society as a whole, for the manner in which they use these powers. This naturally includes an accounting of how the union spends funds from union dues and agency fees.

Under the Labor-Management Reporting and Disclosure Act of 1959, every labor organization covered by the act "shall file annually with the Secretary [of labor] a financial report," which is to include assets and liabilities, receipts, salaries and other compensation, loans and other disbursements.[12] The requirement covers any labor organization that represents private-sector employees, even if that number is modest. Consequently, the vast majority of union organizations are at least partially covered, although most locals of government employee unions are not required to file under LMRDA. The statute gives to the U.S. secretary of labor the authority to draft and revise the forms and rules used for this financial disclosure.[13]

This provision gave rise to the creation of the first LM-2, LM-3 and LM-4 forms. Depending on their total disbursements, union organizations, from locals up to international federations, have been and remain obligated to file one of these forms. The largest groups, with annual disbursements of more than $200,000, were required to file form LM-2.

For most of the period since LMRDA was passed, the demands of union financial disclosure were fairly limited. The forms themselves were largely unchanged for 40 years.[14] As recently as 2004, the LM-2 form included of a series of basic financial questions, many of which were yes-or-no. These were followed by a statement of assets and liabilities, a one-page summary of receipts and disbursements, schedules for loans, assets and investments, and payments to union officers and staff. The form broke expenditures down in a very general way, with "Schedule 12" covering contributions, gifts and grants; "Schedule 13" covering office and administrative expenses; and "Schedule 15" covering "other disbursements." Other than officers and staff, the LM-2 did not require the union to give the names of any payees or itemize any individual transactions.

The Current Union Disclosure Law

In 2003, the U.S. Department of Labor released new rules and forms for union financial disclosure. The most substantial and controversial changes were made to form LM-2. After a federal court rejected a union challenge to DOL's revised LM-2, the new regulations took effect on July 1, 2004; union organizations were required to file the revised form for fiscal years beginning on or after that date.[15] The revised LM-2, which now applies to unions with disbursements of $250,000 or more, requires that vendors receiving more than $5,000, or individual disbursements greater than that amount, be itemized. Most expenditures must now be allocated to one of five functional categories:

Representation. This covers collective bargaining, administration and enforcement of contracts, and attempts to organize new bargaining units. Strike benefits are a separate item.

Political Activities and Lobbying. This covers all attempts to affect elections, or to influence legislation or regulatory actions. Get-out-the-vote drives, voter education and PAC expenses are included as well.

Contributions, Gifts and Grants. This mainly consists of charitable activities.

General Overhead. This involves operational costs, such as building maintenance and security, that cannot be attributed to another function.

Union Administration. This category covers the costs of union membership meetings, union conventions, officer elections, disciplinary proceedings and trusteeships. (Readers should bear in mind that although "overhead" and "administration" are often conflated in nonprofit finance, they cover very distinct aspects of union operation under the LM-2 rules.)

The form also has separate entries for purchases of investments and fixed assets, goods for resale, and benefits for employees and officers, as well as the payment of strike benefits. In addition, union officers and staff are required to estimate the percentage of time that they spend on each of the five functional categories. These are then used to allocate officer and employee salaries into an overall calculation of union spending for each category.

The revised LM-2 forms are not without shortcomings. The statute does not allow DOL to impose fines or administrative penalties when a union fails to file a financial disclosure report or when a union files a report deemed unresponsive: DOL instead must file a civil action against those responsible, a clumsy and time-consuming enforcement mechanism.[16] The threshold for itemization is higher than necessary and still allows important transactions to remain hidden.

The definitions for functional categories are perfunctory and give little guidance on specific cases in gray areas.[17] For instance, if a union buys office supplies and equipment for an employee who spends most of his time on collective bargaining and the rest on internal union discipline, the rules give little guidance as to whether this purchase should be considered a representation expense or overhead. The definition for bargaining does not mention overhead costs, while the definition for general overhead reads as follows:

SCHEDULE 18 - GENERAL OVERHEAD

Report the labor organization's direct and indirect disbursements to all entities and individuals during the reporting period associated with general overhead that cannot be allocated to any of the other disbursement categories in Statement B.

Some disbursements for overhead do not support a specific function, so these disbursements should be reported in this schedule. Include support personnel at the labor organization's headquarters, such as building maintenance personnel and security guards, and other overhead costs. Not all support staff should be included in General Overhead. For instance, the salary of an assistant, whenever possible, should be allocated at the same ratio as the person or persons to whom they provide material support.[18]

A union official apportioning expenses among the various functions may find it difficult to determine just how closely a particular item should relate to representation, politics, charitable activities or union administration in order to qualify for inclusion under these categories, as opposed to general overhead. In addition, the rules give no guidance on such issues as how costs of a union publication with articles about both political issues and collective bargaining should be allocated, or the treatment of a training program that covers topics ranging from grievances to the procedural rules for union elections.

Though there are shortcomings, revisions to the forms overall have made union spending more transparent. The functional categories were introduced as part of the 2003 LM-2 revisions. A certain amount of trial-and-error in clarifying the definitions is to be expected. It would be unrealistic to expect DOL to anticipate where every gray area would be or develop practical rules for handling them until the department had the chance to see what the itemized disbursements look like and what sorts of disbursements are most likely to cause confusion.

Given the lack of specificity in the rules, combined with the political and legal incentives facing union officials, one can reasonably expect that they will tend to categorize borderline items as representational where possible and avoid treating disbursements as political or lobbying expenses where they can plausibly be considered something else. Union political activism has been the subject of controversy for many years, with the Landmark Legal Foundation actively pursuing an investigation of the National Education Association for failing to report extensive political activities to the IRS.[19] The extent of representational spending, meanwhile, is at the crux of litigation based on Beck v. Communications Workers of America. It is in a union's best interest to have worker representation appear to be as large a portion of union spending as possible in order to discourage workers from invoking their rights under Beck and to avoid undercutting their position in various lawsuits.[20] Consequently, we are reasonably confident that the LM-2 figures for representation effectively capture all or nearly all costs necessary for a union to carry out its duties as collective representative in the workplace. If anything, the representation expenses on LM-2 forms are likely to be inflated.

Finally, the rules do not require any external audit or other independent verification of the information in the report. This omission distinguishes union financial disclosure from the filings required of corporations under the Sarbanes-Oxley Act of 2002 and from the regular practices of most private-sector organizations.

Unions and their supporters have also been critical of the LM-2 requirements. In its suit to prevent the implementation of the 2003 revisions, the AFL-CIO argued that the new rules were more extensive than allowed under the statute and that union financial disclosures should be limited to a balance sheet and income statement.[21] In a longer critique of the LM-2 rules which was highlighted by the AFL-CIO's blog, Scott Lilly of the Center for American Progress argued that the DOL was guilty of "rigorous and in fact pernicious regulatory enforcement," and that the LM-2 forms were part of a DOL effort designed to, "Greatly increase the time, effort, and expense to labor union, their officers, and employees with department reporting requirements" and "Use information gleaned from Labor Department investigations of union officials and employees along with data from expanded union reporting requirements to launch a public relations effort to discredit unions and weaken their ability to organize and act on behalf of their members."[22]

Lilly also observed that the five functional categories used by the Department of Labor did not line up with the budgeting categories used by unions. According to Lilly, this meant that the new regulations "created an entire second layer of accounting simply for the purpose of filing the reports."[23]

There is little doubt that the new regulations create some burdens for union officials. Certainly the filings are longer: UAW Local 600's LM-2 for 2004, the last under the prior rules, covers 21 pages, while the LM-2 for 2005 covers 30 pages of somewhat smaller type. The DOL has created software that might ease things somewhat, but installing software and entering data have costs, too.

But unions are not alone in bemoaning the burden of government paperwork or even of disclosure forms. Groups such as the U.S. Chamber of Commerce have been sharply critical of the financial reporting requirements for corporations established by the Sarbanes-Oxley Act. Estimates of the cost of complying with Sarbanes-Oxley are as high as $6 billion annually and likely to increase.[24] The experience of filling out detailed financial reports has not made the union movement more sensitive to the regulatory burdens confronted by others, however; the AFL-CIO was opposed to regulatory modifications of Sarbanes-Oxley intended to lower the burden of corporate financial reporting.[25]

The other objections raised to the higher standard of financial reporting under the 2003 LM-2 rules are ultimately unpersuasive. To the extent that public confidence in unions might be undermined by the manipulation of data found in LM-2 reports, the proper response for unions is more disclosure - telling "the rest of the story," and letting their members and the general public come to their own conclusions.

As for the incompatibility of the LM-2 functional categories and union budgeting procedures, Lilly does not identify where the mismatches are, but if union budgeting procedures cannot be resolved with the functional categories established under the LM-2 forms, it may be an indication that the unions are out of touch with their membership. The creation of the "representation" function in particular seems common-sensical. A 2004 survey of union members undertaken by Zogby International and the Mackinac Center for Public Policy found that the vast majority of union members, 73 percent, consider the primary responsibility of unions to be that of "bargaining for better wages, benefits, and working conditions for its members." Separating the representation function from political activism or internal union governance is very much in line with the priorities of union members.

Unlike shareholders of the publicly traded corporations covered by Sarbanes-Oxley, union members cannot sever their relationship with a union by selling shares if they are unsatisfied with the union's performance or with information the union discloses. Individual employees must be prepared to change jobs to avoid the injuries they might suffer because of irresponsible union leadership. If anything, the case for stringent financial reporting is stronger in the case of unions than for private companies.

A Review of Union Financial Disclosure Forms

The Mackinac Center has undertaken a systematic review of union spending based on LM-2 reports. Six unions were examined:

• International Brotherhood of Teamsters


• National Education Association /Michigan Education Association


• United Auto Workers


• Service Employees International Union


• American Federation of State, County, and Municipal Employees


• Carpenters (United Brotherhood of Carpenters and Joiners)

For each of these organizations, we reviewed both the international union and intermediate organization in Michigan. In most cases, we also reviewed up to three local unions located in Michigan, although in the case of the National Education Association, we were obligated to look at a local report from another state.

Our analysis began with each organization's total spending on the five basic functional categories of representation; political activities and lobbying; contributions, gifts, and grants; general overhead; and union administration. We also included fixed assets and purchases of investments (these figures are found on "Schedule B" of an LM-2 form). Benefits for employees and officers were allocated among the five functional categories on the same percentage as wages, and the payment of strike benefits was treated as a representation expense. "Per-capita taxes" - meaning funds that are collected and then passed on to other organizations in the union's structure - were disregarded to avoid double-counting. Other spending categories were also disregarded, since they tended to be minimal.

After developing a spending profile for international, intermediate and local levels of a union, we calculated a weighted average based on spending per member that should roughly indicate the percentage of workers' dues that are spent on representation and other union functions.

In addition, we examined itemized expenditures attributed to representation for the international headquarters of several of the unions listed above. From that, we developed a profile of union "representational" spending.

We should note that a great deal of union financial activity is not covered by LM-2 reports. LM-2 forms mainly cover the receipt and disbursement of union dues and per-capita taxes (i.e. union funds that are collected by one level of a union's hierarchy, but forwarded to another level.) The LM-2 does not cover contributions to candidates or ballot initiative campaigns by union-controlled PACs, nor does it cover union pension or health care benefit funds; the employee benefits listed on an LM-2 form belong to the union's own officers and/or staff.

The reader should bear in mind that our findings are subject to the limitations of our data; we do not claim that our estimates are particularly precise, and the results will be affected by the extent to which unions have misallocated spending, whether accidentally or intentionally. The purposes of many itemized disbursements remain unclear, and since we do not have access to original records, we cannot verify any individual items.

The approach taken by the DOL in creating the LM-2 forms presents two particular problems, both of which are probably unavoidable. Non-itemized expenditures cannot be verified in any manner, and neither can employee time. Employee wages and benefits make up a major part of a union's budget, as they do in most nonprofit organizations, but the LM-2 forms give us no documentation or explanations to back up the time allocations.[*]

In addition, payments among affiliates within a union create an unavoidable double-counting problem. A common practice of union national organizations, and to some extent among intermediate organizations as well, is to make payments to other union affiliates' local organizations. These payments are treated as part of that union's functional spending. For instance, a national organization may make a payment, denoted as "bargaining assistance" to a local. Those funds are prone to being counted twice when the union's overall spending is tallied up, first at the national level and then at the local level.[†]

The double-counting effect caused by intra-union transfers appears to be modest for most unions, but in some cases may involve as much as 10 percent of a union's overall revenues and spending.

With those caveats in mind, our review of LM-2 reports does allow us to make rough estimates of what a typical union budget might look like and of the percentage of membership dues that go to representation, politics, charitable work, overhead and administration.

International Brotherhood of Teamsters

The International Brotherhood of Teamsters represents a fairly typical spending profile and its spending on the core representation function is higher than average. Our review of itemized expenditures found no unusual or questionable patterns.

We examined spending at three separate Michigan-based locals: No. 337 in Detroit, No. 243 in Plymouth and No. 406 in Grand Rapids.[*] Total expenditures ranged from $1.3 million to $2.7 million, and spending per member for all three was fairly close, ranging from $326.63 to $384.91, with an average per-member figure of $357.57. The Teamsters union appears to have most of its spending concentrated at the local level; going by per-member spending figures for local, intermediate and national headquarters, 67.5 percent of Teamsters union dues (out of a total of $529.91 per member for local, joint council and national headquarters) are spent at the local level.

Compared to other unions we studied, Teamsters locals in Michigan appear to spend a respectable amount on representation: 61.9 percent. This number was partly due to Local 337, where better than 71 percent of dues went into the representation function. The figures for the other two locals were somewhat lower, but all three claimed representation spending in excess of 55 percent. Overhead was the second-largest item for all three locals, averaging 25.3 percent overall. Local spending on the "administrative" function — which in this context refers to union internal governance, including conventions and internal discipline — was 8.4 percent, a relatively modest amount compared to that of other union groups.[26]

One odd feature of all three Teamsters locals was that they reported between $60,000 and $110,000 in "Steward Dues" paid to themselves and listed as a cost of representation. These steward dues payments made up nearly 10 percent of representation spending for all three locals. These locals apparently pay the cost of union dues for at least some stewards - a perfectly reasonable policy - but treating all of the forgone dues as a representation cost may overstate the case.[27] We do not have time allocations for any of the stewards, so we have no verification that 100 percent of their forgone dues payments should be treated as representation.

Teamsters Joint Council 43, headquartered in Detroit, is the statewide Teamsters affiliate for Michigan. According to its LM-2 form, Joint Council 43 takes up a modest portion of Teamsters spending in Michigan, with total spending of a little more than $1.8 million for nearly 51,000 members. That breaks down to $36.83 per member or just 7 percent of overall spending per member. Joint Council 43 claims to do less representation work than any of the Teamsters locals we reviewed, with just 47.4 percent of spending recorded under this category. Representation is still the largest single item, but political activism becomes more prevalent at this level, making up 21.4 percent of the overall budget for Joint Council 43.[28]

Teamsters National Headquarters in Washington D.C., accounts for $135.51 of spending per member, 25.6 percent of the total. At this level, representation makes up a relatively modest portion of overall spending, 36.6 percent, although representation remains the largest functional category. Union administration is the second largest item, taking up 26.5 percent of spending.[29]

At the national headquarters level, all of the unions reviewed submitted reports of more than 100 pages, with extensive listings of itemized disbursements. We took advantage of this level of detail to perform a second analysis of union itemized expenditures, focusing on the representation category only. Itemized representation expenditures were broken down into the following subcategories: overhead, legal, research and staff training, printing, entertainment, travel, transfers to subsidiaries, suspect (possibly  political), organizing and unclassifiable. These subcategories provide a better sense of what materials and services unions purchase under the representation rubric. We also factored in personnel costs - covering compensation costs for officers and staff, and including benefit costs - according to the percentage of staff time dedicated to representation, as well as non-itemized expenditures ("other") attributed to representation.[†] The results of this are as follows:

Wages and benefits for officers and staff made up the majority of representation by the Teamsters National Headquarters, with transfers to locals and joint councils making up the largest part of the remainder. Unfortunately, while we can identify specific Teamsters subsidiaries that received assistance from the Teamsters national headquarters, there is little we can tell about what form that assistance took - that is, whether it was legal or consulting work, or whether it pertained to collective bargaining, a grievance or some other aspect of representation.

There was little evidence that the Teamsters union attempted to classify political activity as representational, at least not in terms of purchases of goods and services.

Graphic 5 presents a summary of Teamster spending at the local, joint council and national level according to their LM-2 forms, as well as a combined figure that represents our best estimate of the percentage of union dues that go to the various union functions. As noted earlier, local spending is estimated by LM-2 statistics for Teamsters Locals 243, 337 and 406.

The National Education Association and Michigan Education Association

The LM-2 reports filed by the National Education Association and its Michigan affiliate, the Michigan Education Association, suggest the union is much less focused on representation than the Teamsters. Overall, a majority of union dues funds go toward overhead. In addition, a number of "charitable" contributions made by the National Education Association appear to have a distinctly political character to them.

Our examination of the NEA/MEA is complicated by the fact that the nation's largest public school employee union represents very few private-sector employees. While some smaller NEA locals in Michigan do file under the LMRDA, none of those are large enough to be required to file form LM-2. In fact, only three local affiliates of the NEA nationwide filed LM-2 forms during 2006, of which two represent teachers employed by schools run on military bases under the supervision of the Department of Defense for the purpose of educating the children of military personnel. The location and circumstances of these locals is peculiar enough that we considered their use inappropriate.[30] The third NEA local to file an LM-2 form is the United Faculty of Florida, located in Tallahassee. The UFF represents instructors and professionals at public colleges throughout Florida, as well as at two independent colleges — which triggers their obligation to file under LMRDA.

Using UFF as a stand-in for an NEA-MEA local in Michigan is not without problems: the UFF bargains with state colleges rather than local school boards, and Florida — aside from not being Michigan — is a right-to-work state. However the Michigan Education Association and its affiliates are among the most powerful unions in Michigan; a review of union spending in Michigan without a look at NEA/MEA would be incomplete. Given the limited options, using UFF as a stand-in seems to be the best route because its operations, while distinct from those of most MEA/NEA locals, are at least in line with that of a normal government employee union; it bargains with state entities rather than the Department of Defense. The reader should bear these distinctions in mind, but should also consider that the structure of the NEA/MEA appears to put the majority of spending at the state level rather than the local due to the unique role of Uniserv staff, who are hired by state affiliates and have a great deal of influence over representation at the local level. Local spending is significant, but unlike other unions the bulk of spending in NEA/MEA is not done at the local level.[*]

With that in mind, our review of the UFF's LM-2 report indicates that representation makes up 59.4 percent of UFF spending. Most of the remaining disbursements fall into the category of administration. UFF's operations cover members and chapters at college campuses across Florida, in contrast to most MEA-NEA locals that are dedicated to a single school district. It is possible that internal union decision-making is complicated by geography, which contributes to higher-than-usual administrative costs. UFF's total spending is over $823,621, for 3,990 members, which comes out to $206.42 per member, or 21.5 percent of the total of $958.30 that NEA/MEA appears to have spent per-member in 2006.[31]

The MEA is remarkably large for an intermediate union body. The MEA's LM-2 Form covers more than 200 pages; no other intermediate organization that we reviewed filed an LM-2 of even one-third that length. In direct contrast to Teamsters Joint Council 43, the MEA appears to be responsible for the majority of NEA/MEA spending in Michigan. Our review shows that MEA spent $632.44 per member, 66.0 percent of total per-member spending.

At first glance, the MEA comes off as an extremely inefficient representative organization, with only 22 percent of spending going into representation and 60 percent allocated to overhead. A closer look results in a slightly less severe appraisal, although MEA members still have considerable grounds for concern about their union's priorities.[32]

In light of the unusual level of detail contained in the MEA's LM-2 report, we were able to analyze subcategories of representation spending, a step we otherwise reserved for international headquarters of other unions. More than most union organizations, the MEA prefers to keep representation matters "in house." Looking at itemized disbursements attributed to representation (including personnel costs) on the MEA's LM-2, 85 percent of representation costs are taken up by personnel, with 11 percent of representation expenditures going to attorney fees and 2 percent going to staff training and research.[33]

All other subcategories that we used to break down representation spending received less than 1 percent of overall representation spending. That includes "overhead" costs attributed to representation, while for Teamster national headquarters this figure was 8.7 percent. While the rules allow some overhead costs to be shifted into representation and other functional categories, MEA appears to have done less of this than most other union organizations, which may make the MEA's commitment to representation appear less than it actually was. But this peculiarity of the MEA LM-2 report only mitigates matters to an extent. Even after adjusting for this oversight, it is unlikely that MEA's expenditures for representation would exceed 30 percent of its entire budget.

The National Education Association, based in Washington D.C., had more than 2.7 million members and total expenditures of over $330 million in 2006, which works out to $119.43 in per-member expenditures, 12.5 percent of the total per member expenditures for the UFF, MEA and NEA combined.

At this level, the NEA is largely disconnected from what most union members would probably consider the central purpose of unions; employee representation expenditures made up only 19.4 percent of NEA national headquarters spending. Spending on representation was exceed by spending on contributions, gifts and grants (23.0 percent) general overhead (23.1 percent) and administrative costs (24.6 percent).[34]

After breaking down itemized representation expenses, we find that transfers to locals and personnel costs make up the bulk of NEA headquarters representation spending. (44.0 and 38.4 percent of spending, respectively.) To a large extent the NEA's "representation" function consists of distributing funds to affiliates. As is described above, this creates a possible double-counting problem which may lead to overstating NEA's actual involvement in representation.

On the other hand, NEA headquarters shifted relatively little of its overhead costs to representation, as is the case with its Michigan affiliate; only 2.0 percent of NEA's itemized representation disbursements were for items that could be considered overhead. We also found a number of questionable itemized expenditures attributed to representation:

$58,474 to the Economic Policy Institute

$30,000 to the Campaign for America's Future, which describes itself as "the strategy center for the progressive movement."[35]

$50,000 to the National Coalition on Health Care, a "representative alliance working to improve America's health care.

$50,000 to America's Agenda. According to the America's Agenda Web site, the organization's purpose is to "secure affordable, comprehensive health care for all Americans by first winning universal health care state by state.37

While entries like this, in which payments are made to groups with definite political or public policy agendas, make up a small portion of the representation expenditures made by NEA headquarters, they suggest that the NEA national office has a very generous definition of what constitutes workplace representation. The NEA's contributions, gifts and grants schedule contains even more entries of this sort:

Another $150,000 to the Economic Policy Institute

$8,000 to the National Council of La Raza

$10,000 to the National Women's Law Center

$10,000 to People for the American Way

$5,000 to the Rainbow/PUSH Campaign

$5,000 each to the Gay and Lesbian Alliance Against Defamation and the Gay, Lesbian, and Straight Education Network

$25,000 to the Democratic Leadership Council

In a memo to its affiliates, NEA headquarters attempted to explain these items. NEA's "talking points" on its grants and contributions asserted, "NEA also has contributed to outside organizations that work on issues related to NEA's core mission of protecting the basic right of every child to a quality public education."[38] But numerous grant recipients here are only tangentially involved in educational issues - from any perspective. While these particular line items by themselves make up small portion of overall NEA headquarters spending, they lend credence to NEA critics who allege that NEA is more deeply involved in political activism than is publicly acknowledged, to the detriment of public school teachers and support staff who rely on NEA and its affiliates for representation on matters of pay, benefits and working conditions.

Overall, we estimate that only 29.5 percent of NEA/MEA members' dues are ultimately spent on representation expenses. Representation costs are exceeded by overhead, which makes up 43.1 percent of overall NEA/MEA spending.[39] Our summary of combined local, MEA, and NEA spending is as follows:

United Auto Workers

While the MEA/NEA is arguably more powerful, no union is more associated with Michigan than the United Auto Workers. According to its LM-2 reports, the UAW is more focused on representation than many other unions, which is particularly noteworthy because the bulk of its members are employed by three companies - General Motors, Ford and (at the time) DaimlerChrysler - and the UAW did not negotiate master contracts with any of those companies during 2006.

Unlike most other unions, the UAW does not have statewide affiliates, an organizational distinction that is probably due to its members being geographically concentrated in Michigan and surrounding states. The UAW does, however, have a network of Community Action Programs, which are funded by per-capita "taxes" based on local membership in the CAP's geographical area. It appears that an individual worker can have a portion of his or her dues directed to more than one CAP.[40] While CAP funds are modest compared to the overall UAW structure, some of them are large enough to file LM-2 forms. The CAPs we reviewed appeared to have performed no representation function. The bulk of CAP spending was in the political and lobbying realm, with some spending directed to contributions, gifts and grants.

Our review of UAW LM-2 forms indicates that most UAW spending is eventually made at the national level. UAW locals vary wildly as to the quality of LM-2 filings, and a number of UAW LM-2 forms were disregarded because of allocations of expenditures and staff time that strained credulity. For instance, the LM-2 form submitted by Local 600 in Dearborn listed $10,000 in softball field rentals and over $30,000 in athletic uniforms as representation expenses.[41] Local 155 in Warren reportedly had a "Dues Clerk" spend 90 percent of his time on representation, an odd allocation of time for a union employee whose duties ostensibly involve tracking the receipt of membership dues. Local 155's itemized "representation" expenses also consisted mostly of overhead, catering, gift cards and insurance whose connection to representation was not elaborated upon.[42] Local 3000 in Woodhaven did not give job titles for most of its staff, making it impossible for us to evaluate their time allocations.[43] LM-2 forms for these locals were disregarded, even though they were among the largest in Michigan in terms of membership.

However, three UAW locals filed LM-2 forms that avoided particularly questionable allocations of expenditures and contained staff time allocations that we found plausible. Those locals were 652 in Lansing, 235 in Hamtramck and 228 in Sterling Heights. Our estimates of UAW spending are based on filings from these locals.

The portion of UAW local spending dedicated to representation varied widely, from less than 20 percent to nearly 50 percent, with 35.6 percent being average. Representation spending was exceeded by administration, which accounted for 41.1 percent of an average local's expenditures. The bulk of the remainder went to overhead. Overall per-member spending averaged $163.21, which makes up 23.9 percent of the combined per-member spending by locals, CAP funds and the international union.

The LM-2 reports do not fully break down the relationships and geographic coverage among the UAW's CAP funds, but it appears that most UAW members in Michigan have dues, in the form of "per capita taxes," sent to two separate CAP funds: a regional CAP fund and the Michigan CAP fund. The UAW also has a National CAP fund, but this appears to draw funds directly from the National UAW account rather than from per-capita taxes, so we did not include it in our estimates. Our estimates of UAW cap fund spending are based on LM-2 filings from the Michigan CAP and the Region 1 CAP. Because the two overlap, we added expenditures from both on a per-member basis rather than average the two. As mentioned before, the CAP funds made up a modest portion of UAW activity, spending a total of $13.88 per member in 2006, only two percent of the per-member total.

None of that CAP spending, however, went into representation. Between the two CAP funds, more than two-thirds went into political and lobbying activities (68.8 percent) while less than a third (30.3 percent) went into contributions, gifts and grants. What little remained (less than one percent of CAP spending) was attributed to overhead. In summary, the UAW CAP programs are modest in size, but by their own reckoning contribute nothing to workplace representation.[44]

The UAW's national headquarters in Detroit oversees nearly three quarters (74.1 percent) of per-member expenditures. According to its LM-2 form, UAW headquarters appear to be reasonably focused on the representation function: by our calculations 63.0 percent of UAW spending falls into this category. This figure does not seem to be padded by overhead costs; our review of itemized representation expenditures indicates that overhead items make up only 5.8 percent of the UAW's itemized representation expenditures.

Some caution is in order before the UAW's national headquarters is lauded for its exemplary focus on representation. Much of the reason for the UAW having such a large percentage of spending dedicated to representation lies in the time allocations claimed for its staff. According to the UAW, 79.2 percent of its employee compensation went to officers and staff working on representation, an unusually large portion.[45] But officers and staff are not required to itemize projects that take up significant portions of their working hours, while with cash disbursements any item over $5,000 must be itemized, which serves as a check on the misallocation of spending. The LM-2 does not give us any other information we can use to verify how employees made use of their time.

Compared to the NEA, we found relatively few representation expenses that might be political rather than representational in nature, but there were some transfers to CAP funds listed under representation and, as we described earlier, CAP funds are primarily political in nature.

With these caveats in mind, it appears that 55.2 percent of UAW dues ultimately go toward representation, with 21.2 percent going into administration and 9.6 percent spent on overhead. Our summary of the UAW follows:

Service Employees International Union

The Service Employees International Union is notable for being at the center of a split in organized labor. SEIU and six other unions, including the Teamsters, broke away from the AFL-CIO in 2005 to form the Change to Win Federation.[46] SEIU Joint Council 35, its statewide Michigan affiliate, claimed more than 34,000 members in 2006, making SEIU one of the larger unions in the state.[47]

SEIU has two locals in Michigan that filed LM-2 forms in 2006: Local 79 in Detroit and Local 517, also known as the Michigan Public Employees, located in Lansing. Our estimate of SEIU local spending is based on an average of the two.

According to their LM-2 reports, the two locals spent an average of $338.59 per member, which is nearly two-thirds of the overall per-member spending for SEIU as a whole. Representation made up 43.0 percent of local spending, while overhead takes up 29.8 percent of local spending and union administration is responsible for another 15.1 percent. Michigan Public Employees had lower representation costs and higher overhead than Local 79.[48] The fact that 2006 was not a contract negotiation year for most state government employees may have been a factor.

SEIU Joint Council 35, the SEIU's intermediate organization for Michigan, appears to be a fairly modest organization, according to its LM-2 form. Total spending by JC 35 was less than $400,000, far smaller than either Local 79 or the MPE. Per-member spending was $11.42, or 2.2 percent of the per-member total.

Like the UAW's CAP programs, JC 35 was largely a political operation. Only 8.1 percent of JC 35's expenditures were related to representation, while more than half the budget - 53.6 percent - went into politics and lobbying. Internal union business (administration) took up more than a quarter (25.3 percent) of JC 35 spending.[49]

SEIU National Headquarters in Washington, D.C., spent $162 per member, a bit less than a third of per-member spending. According to its LM-2 figures, representation made up a respectable 58.7 percent of SEIU national spending, but this figure may be particularly inflated, as will be shown by our further breakdown of representation spending (see Graphic 19).[50]

First, the number of entries that we found to be likely to be political in nature (listed in our "suspect" category) was significantly higher than for other union organizations. These included items whose political nature was fairly obvious, such as a $20,000 payment to ACORN International, payments totaling $19,500 to Jobs with Justice affiliates in Massachusetts and Philadelphia, $29,360 to the Chicago Coalition for the Homeless, $68,182 for sponsorship of Air America, $83,400 to the Voter Activation Network and $225,000 to the Campaign for Our Children's Future. Lest the reader be left with the impression that labor union organizing is strictly a leftist cause, SEIU also classified a $5,000 payment to the Michigan Republican Party as a representation expense.[51]

In addition, SIEU contributions, gifts and grants included a number of disbursements to political groups, including another $100,000 to Jobs with Justice, $40,000 to La Raza, $100,000 to Illinois ACORN, $100,000 to Rock the Vote and $100,000 to the Rainbow/PUSH Coalition.[52]

There were also an unusually high number of transfers to SEIU locals or intermediate councils. Transfers made up 38.6 percent of SEIU representation costs; this is second only to the national NEA. Most other functional categories in the SEIU report had high numbers of similar transfers, except overhead, where SEIU made only a handful of transfers to its locals.[53]

There was also a curious transfer of $32,237 for representation purposes to a "Michigan State Council, SEIU," located in Lansing, which to our knowledge did not file any financial disclosure form with the Department of Labor in 2006 or any other year. It does, however, have its own Web page.[54] The same mysterious organization received $179,981 in political funds and was the source of a $26,056 transfer to the SEIU national organization.[*]

More than any other labor organization, the LM-2 forms for SEIU are a confused tangle of misallocated spending and transfers to local organizations that are likely to lead to double-counting of union funds. In light of this, we recommend that our figures for SEIU be taken with an extra pinch of salt. Nonetheless, it appears that less than half of SEIU membership dues are ultimately directed to representation. Our summary is shown in Graphic 21.

American Federation of State, County, and Municipal Employees

As its name would indicate, the American Federation of State, County, and Municipal Employees is primarily focused on government employees and therefore the vast majority of locals are exempt from the federal financial reporting requirement. Nonetheless, four AFSCME locals in Michigan filed LM-2 forms for 2006.

Of these four, however, one claimed that all paid staff time was spent on representation: Local 1640 (Community Social Agency Employees) in Detroit.[55] While such diligence would be commendable, we found it unlikely that staff had no duties relating to internal union administration.

Two of the locals claimed no money or significant staff and officer time was spent on representation in 2006: Local 1820 (The St. Joseph Mercy Hospital Employees Union) based in Bloomfield Township, and Local 2568 (Oakwood Employees) based in Lincoln Park.[56]

Local 1820 would file a similar report, indicating no representation activity, in 2007 and reported some spending on representation, but no staff or officer time dedicated to representation, in 2005.[57] (None of the representational spending is itemized.) Such a pattern is conceivable, but barely. One explanation would be that the local brought in outside counsel for collective bargaining in 2005, allowed them to run the negotiations with minimal oversight, and has had no workplace issues, such as employee grievances, arise since then. Local 2568 filed a report indicating no representation activities in 2007.[58] In 2005 the local filed an LM-3 form, and as a consequence we cannot tell what their representation activity might have been. If taken at face value, the reports filed by locals 1820 and 2568 indicate that the local organizations are almost entirely superfluous in most years and perhaps irrelevant in contract negotiations as well.

Ultimately, we concluded that LM-2 reports from all three locals were not credible. The remaining AFSCME affiliate in Michigan was Local 875 (McLaren Registered Nurses) based in Flint. In 2006, the local spent $126.43 per member, which amounts to 27.1 percent of overall dues. Of that, 40.2 percent went into representation, with union administration taking nearly a third of local spending (33.1 percent) and overhead slightly more than a quarter.[59]

AFSCME's statewide organization for Michigan is Council 25, headquartered in Lansing. The LM-2 for Council 25 indicates that the council's responsibilities are substantial, much more than is typical for intermediate union bodies; Council 25 spending amounts to $228.90 per member, nearly half of total per-member spending by AFSCME and its affiliates.

Council 25's spending profile is not particularly lean: overhead (38.4 percent) is nearly as large as representation (39.9 percent). Politics and lobbying make up a significant portion of Council 25's expenditures, 13.8 percent.[60]

AFSCME's National headquarters take up the remaining quarter of AFSCME spending, with disbursements of $111.49 per member. AFSCME's national office is even less involved with representation, and more with politics, than Council 25: Representation takes up less than a third of expenditures by the national office, and is nearly matched by political and lobbying expenses.[61]

Breaking down representation expenses, we found that spending by AFSCME's national office included a larger-than-usual amount of overhead costs and an unusually large number of transfers to locals. To its credit, AFSCME had no significant entertainment costs, and there was little evidence that political spending was mischaracterized as representation, although a number of contributions, gifts and grants items (such as $73,000 in payments to Rainbow/PUSH) were to political groups.[62]

Overall, though, AFSCME comes across as a particularly inefficient workplace representative. We estimate that only 38.0 percent of AFSCME member dues are dedicated to workplace representation. Overhead (29.8 percent) and administrative (17.1 percent) expenses combined exceed worker representation by a sizable margin.[63] Our summary is shown in Graphic 26.

Carpenters (United Brotherhood of Carpenters and Joiners)

The United Brotherhood of Carpenters and Joiners of America and its affiliates are among the more numerous of the construction industry unions in Michigan, and their Michigan regional council has advertised extensively on cable television in the state. By our estimates, the Carpenters union is easily among the biggest spenders in the union movement, at least on a per-member basis. The total per-member spending for the local, intermediate and national organization is $1,509.82.[64] While per-member spending for all unions is substantial, a total of $400 to $700 is more typical. None of the other unions we looked at had total per-member expenditures in excess of $1,000.

Finding a representative Carpenters local in Michigan is greatly simplified by the fact that only Local 1102 in Warren, filed an LM-2 form in 2006. None of the other Carpenters locals in Michigan had sufficient disbursements to warrant filing an LM-2.

Local 1102 spent $300.90 per member in 2006, 19.9 percent of the total. Its members might wonder why any of this was needed since more than three quarters of that $300 went into overhead, while only 6.6 percent - one fifteenth - of the local's spending was dedicated to representation.[65] This would appear to be an extremely inefficient operation.

These spending figures were greatly influenced by the manner in which personnel allocated their time. According to its LM-2 report, the local's staff is more or less evenly divided between officers and employees. While most of the local's officers (president, vice president, treasurer, etc.) reported that a majority of their time was spent on representation tasks, their compensation was relatively low.[66] The local's employees (secretaries, bookkeepers and one business agent) were more highly compensated, while they reported no representational activity. The overwhelming majority of employee time was dedicated to general overhead, with some time also allocated to administration.[67]

We decided to use this LM-2 because, unlike the AFSCME locals discussed earlier, there is some indication that Carpenters Local 1102 did make an effort to differentiate between uses of staff time.[68] This peculiar time allocation, while not as extreme as that of the AFSCME locals, should be borne in mind as the Carpenters spending is evaluated.

The Michigan Regional Council of Carpenters and Millwrights, headquartered in Detroit, had the highest per-member spending of any union organization (local, intermediate or national) that we reviewed. With disbursements of over $18.6 million for a membership of less than 18,000, the Michigan Regional Council spent $1,054.19 per member, more than the total (including local, intermediate and national) for any other union. The regional council is responsible for nearly 70 percent of per-member spending by the Carpenters union.

Compared to Local 1102, the Regional Council is much more focused on representation, with 48.9 percent of spending going into that function. Overhead costs making up 34.4 percent of spending are higher than most union organizations, but closer to normal.[69]

The National Headquarters of the Brotherhood of Carpenters and Joiners spends $154.92 for each of its 523,000 members, making up around 10 percent of overall spending. As is the case with the local, the national headquarters has little to do with representation; only 11.1 percent of their spending falls into that category. By comparison, contributions, gifts and grants make up a somewhat larger portion (14.3 percent) of the national union's expenditures. Unlike other unions, relatively few of the donations made by the Carpenters union went to obviously political groups. Most of the itemized contributions were given to various construction trade unions.

But the largest portion of the Carpenters spending was overhead, which took up more than a third of the national headquarters spending, while administration accounted for another 12 percent. The Carpenters national headquarters is also notable for having more than a quarter of its spending going into purchases of fixed assets and investments, the highest rate of any LM-2 report we examined.[70] With overhead and administrative costs exceeding representation by a factor of four-to-one, the national headquarters of the United Brotherhood of Carpenters and Joiners appears to be afflicted by extreme waste.

Fortunately for its members, the more severely bloated levels of the Carpenters union appear to be responsible for less than half of the Carpenters Union spending. Still, representation makes up less than half of the spending for every level of the Carpenters union, and less than 40 percent of spending for the Carpenters overall. In fact, we estimate that more of a typical Michigan Carpenter's dues will go into overhead than into representation.[71] Of the unions that we reviewed, only MEA/NEA makes less efficient use of their members' dues. Our summary appears in Graphic 30.

Where Union Money Goes

As mentioned earlier, our estimates are based on LM-2 reports filed by individual union organizations at the national, intermediate and local level. Much of the data is unverified and the rules give individual unions considerable leeway as to how individual items might be classified. The union's own word, without supporting documentation, is the basis of the allocation of staff and employee time among the various functional categories of representation; politics and lobbying; contributions, gifts and grants; general overhead; and administrative (union governance), as well as the resulting distribution of wages and benefits among those functions.

The various unions — and sometimes different units within the same union hierarchy — treated LM-2 reports very differently. The Carpenters union generally avoided attempting to pass off payments to political organizations as representation expenses, while SEIU had a number of entries attributed to representation that were likely political in nature. The National Education Association also had a sizeable number of dubious entries attributed to representation; its Michigan subsidiary had somewhat fewer. AFSCME locals in Michigan varied wildly in the extent to which staff time was attributed to representation or administration.

In assembling these estimates, we have taken what we believe to be the best approach to analyzing union spending given the limitations of the LM-2 reports. We have generally taken the reports filed by the various unions at face value, unless those reports strained credulity. Most of the reports filed by the largest locals in Michigan were used, as were all of those filed by intermediate and national organizations. Where inappropriate expenditures were found, they were noted, but as a general rule we did not attempt to adjust the unions' own figures except, as described earlier, to apportion the costs of benefits for the unions' own employees and to include strike benefits as a cost of representation.

The vast majority of union-represented workers in Michigan are required to pay union dues or agency fees as a term of employment. This examination of LM-2 reports was undertaken so that Michigan workers, lawmakers and anyone else interested in the prosperity of Michigan's working men and women could have a better understanding of how unions in Michigan use these funds.

Are the unions that receive all these mandatory dues and fees using this money wisely? Are these funds being used for their proper purposes — ensuring that these workers have effective representation in the workplace — or are they being used for something else? The answers that we found are not encouraging.

Among the six unions that we reviewed, representation made up as much as 55.2 percent, and as little as 29.5 percent, of total spending. On average, the unions we examined spent less than half of their funds on representation. General overhead took up more than a quarter; overhead and administrative costs combined made up nearly as large a portion of union expenditures as workplace representation.

We conclude with a high level of confidence that the 6.1 percent figure for political and lobbying expenditures underestimates union political activity; the only question is by how much. We have documented that unions have attributed payments to political advocacy groups under both representation and contributions. It is entirely possible that non-itemized expenditures have been misallocated in the same manner. And itemized disbursements make up a minority of the expenses for most unions. Employee time is allocated almost entirely on the honor system. A union that is willing to present a contribution to the Campaign for America's Future or ACORN as a representation expense is certainly capable of treating time spent on political activism as representation time as well. We cannot prove this has happened and will not speculate as to how prevalent it is, but this is a likelihood that workers and policymakers should bear in mind as they examine how union staff are spending their time.

What we can say with more confidence is that LM-2 forms suggest the union movement is a fairly bloated one, with overhead and administrative costs far out of line with what is needed to support the core union mission of worker representation.

Comparing Unions to Nonprofits

To illustrate the problem, it might be useful to compare union overhead costs with those of nonprofits. There has been considerable research on charities and their spending on overhead and administration costs versus program costs.

The needs of charities for overhead and administrative support is a subject of some controversy, but a review of the literature indicates that a reasonably well-run nonprofit should be able to dedicate close to two-thirds of its spending to its core programs. Under the "Standards for Charity Accountability" published by the Better Business Bureau Wise Giving Alliance, program service expenses should make up 65 percent of total expenses.[72] The Combined Federal Campaign, the federal government's workplace charitable donation program, deems a charity ineligible if its administration, fundraising and overhead costs are persistently above 25 percent of revenue.[73] In an August 2005 statement, in which it urged donors to make allowance for charities' administrative and overhead costs, the Center for Philanthropy at Indiana University asserted, "[O]verhead costs can reasonably be 10 percent, or 20 percent or sometimes even 35 percent of an organization's budget, depending on the organization's work and its size."[74] With representation making up less than half of expenditures, it is not clear that any of the unions we reviewed would have met these standards.

If anything, that standard should probably be higher for unions than for other charities, because unions, unlike most other nonprofits, do not have to solicit funds voluntarily. Fundraising is a major factor in most nonprofits' overhead costs, but because of the availability of agency fees in most states, including Michigan, unions can draw revenue without the difficult work of identifying potential donors or making appeals to them — at least not in the manner required of most nonprofits. Because they are largely relieved of the burdens and uncertainties of fundraising, unions should be free to put more of their resources into representation.

In fact, to the extent that unions engage in "fundraising," the costs are inevitably treated as representation or "program." Unions secure dues revenues by organizing workers in new bargaining units and then negotiating collective bargaining agreements that include agency fee clauses. Union organizing drives are considered part of representation under the current LM-2 rules, and negotiating collective bargaining agreements is an essential part of the representation process.

Implications for Beck Rights

Our review of union LM-2 reports also has important implications for those workers who elect to pay an agency fee and exercise their rights under Beck to have their union membership dues limited to the pro-rata portion of the costs of representation. Under the typical Beck analysis, unions are given an allowance for overhead and administrative costs. We have shown that the LM-2 rules allow unions to apportion costs that are typically considered "overhead" to representation so long as they support representation work, and we have also demonstrated the tendency of some unions to categorize some likely political costs as representation.

If anything, our estimates of union spending on representation are likely to be high. As described earlier in this report, courts that have examined union spending have found significantly lower spending on representation. This is most likely due to the limitations of the LM-2 form and the absence of any independent verification of many items. A diligent court will have far greater access to union records than is provided for by the LM-2 form.

The amounts we have calculated based on LM-2 reports are probably a more than adequate estimate of union representation costs, including an appropriate amount for overhead and administrative support. If so, dues collected from Beck objectors should be limited to the percentage indicated on their LM-2 form unless a union can demonstrate that its actual needs for staff and materials to support representation are greater.

Following such a rule, the typical reduction for Beck objectors represented by affiliates of the National Education Association in Michigan would increase from the current 20 to 33 percent to around 70 percent. Those unions most focused on representation only spent 55 percent of their total expenditures on representation. Based on that, the minimum dues reduction to which Beck objectors are generally entitled would be 45 percent. Public-interest law foundations with an interest in promoting worker rights should consider litigation that builds on the Beck decision and establishes that in the absence of other evidence, Beck objectors should be entitled to a dues refund based on the percentage of union spending devoted to representation as indicated on its LM-2 forms.

Needed Improvements to LM-2 Forms

While the Department of Labor deserves considerable credit for updating LM-2 forms and rules, our review revealed numerous weaknesses in the current system.[*] First, the definitions given for each of the functional categories are too general. As was earlier noted, a certain amount of vagueness is to be expected when the functional categories are new, but with nearly three years of LM-2 reports to review, the Department of Labor should be in a position to know where the gray areas are and how they could be replaced with bright lines.

The primary purpose of a union is representation of workers. Unions have a direct interest in making this function appear as large as possible. Consequently, the greatest need for sharp but fair distinctions is here.

As the rules currently stand, representation is one of several functional categories. But worker representation is not just one of several functions that a union performs; it is the raison d'etre of the entire union movement. Having "representation" as a single expense category on form LM-2 is the equivalent of General Motors treating "cars" as a single category on its expense statements. Shareholders would be likely to expect more details. Future LM-2 forms should break representation into its constituent parts:

• Collective Bargaining — disbursements and staff time spent preparing for and conducting collective negotiations with employers for the purposes of establishing, modifying or extending collective bargaining agreements.


• Contract Administration and Enforcement — relating to the implementation of an existing or newly ratified collective bargaining agreement, including related litigation and the prosecution of any grievances that might arise.


• Organizing — relating to attempts to become the representative for bargaining units or to maintain the union's position as representative of an existing bargaining unit. This category should include communications directed primarily at bargaining unit members, any necessary communications with employers and legal costs.


• Overhead and administrative support — disbursements and support staff time reserved for the use of union employees who are primarily engaged in collective bargaining, contract administration and enforcement, and organizing.


• Finally, the Department of Labor should require unions to separate all transfers to affiliates from regular spending in order to resolve the double-counting problem and simplify the task of assessing union priorities.

While we combined strike benefits with general representation costs for our analysis, these items should be kept separate in future LM-2 revisions. Breaking representation down into these constituent parts will make it easier for DOL to develop clear rules regarding what is and is not a representation expense, and for union staff to follow those rules. The creation of these subcategories will also give union members a more complete understanding of the services that the union provides for them and where union priorities lie.

Businesses and even nonprofits are regularly expected to provide independent verification, including an independent audit, as part of their financial disclosure requirements under federal law. As we discussed above, the fact that unions claim much of their income through agency fee clauses, requiring workers to pay union dues or agency fees as a condition of employment, makes the need for thorough financial disclosure by unions as compelling as it is for publicly traded companies — perhaps more so. Independent verification of the information contained in LM-2 forms would give union members greater confidence in the information they receive from their union and do much to reduce the number of questionable entries.

Under current law, there is no penalty if a union fails to file an LM-2 form or refuses to allow its members to review the underlying records. Instead, the Department of Labor or individual union member must ask a court for an injunction to compel compliance, a cumbersome enforcement mechanism that among other things is likely to encourage unions to take liberties with their financial reporting duties. When DOL must litigate to compel the filing of a report, or documentation is difficult to find, the department is more likely to accept an LM-2 report with incorrect or questionable entries. Legislation has been introduced to allow DOL to levy fines on unions that fail to meet their obligations under the Labor-Management Reporting and Disclosure Act.[75] Legislation of this sort is long overdue.

An even greater legislative oversight, however, is the failure of the state of Michigan to establish any financial reporting requirements for unions representing state or local government employees. While state labor federations generally do file federal financial disclosure forms, government employee union locals across the state are largely exempt from making any financial disclosures other than tax forms available to their members or the general public. Some of these locals represent thousands of members and handle hundreds of thousands of dollars in mandatory union dues and agency fees. For all the weaknesses of the current LM-2 forms, Michigan government employees would benefit greatly if the state were to simply adopt federal standards and require all unions in Michigan with annual disbursements of more than $250,000 per year to file form LM-2 with the Department of Labor or with the state.

Conclusion: Funds Wasted and Privileges Abused

For decades, the policy of Michigan and other states without right-to-work laws has been to allow unions to negotiate for and collect mandatory union membership dues or agency fees from the workers they represent. This significant curtailment of workers' rights of free association has been tolerated because it is believed that unions must have a reliable stream of funds to represent workers effectively, and that it is unfair to allow individual workers to be "free-riders" who avoid bearing the costs of that representation. These assumptions are essential to understanding union objections to right-to-work; if unions do not need these funds, the moral case against free-riders loses whatever force it might otherwise have.

A review of LM-2 forms filed by various unions leaves much of that rationale in serious doubt. Based on the information in these reports, it is very likely that the typical Michigan union could function effectively as a worker representative on little more than half its current revenue by reducing political activism and prudently trimming excessive overhead and administrative costs.

Assuming Michigan were to enact a right-to-work law, empowering workers to withhold financial support from a workplace union, the effect on union revenue would probably be substantial. But unions would be unlikely to see a cut in revenues that approaches 50 percent. A revenue reduction of 10 to 25 percent would be far more likely. In right-to-work states, the vast majority of workers who are covered by collective bargaining agreements opt to join the union and pay regular dues. That percentage ranges from a high of 88.6 percent in Utah to a low of 77.4 percent in Mississippi.[76] Adjusting to this sort of drop in income would be challenging for a union, but should not entail any loss in the quality of worker representation.

The case for continuing to force workers to financially support unions might have some weight if the typical union were a reasonably economical organization focused on worker representation, but the picture that emerges after examining union financial reports suggests that unions are anything but lean and focused. The typical union structure is weighed down by high overhead and administrative costs, and distracted by political activism.

We have generously granted unions privileges that we rarely give to other private organizations. Michigan unions appear to have been negligent about how they use these privileges and in some cases may have abused them. There appears to be no compelling reason for those privileges to be continued: Michigan unions do not really need all the money they have been given.

Author Acknowledgements

I would like to express my gratitude to Mackinac Center summer intern James Vote, who pored through the hundreds of entries on LM-2 reports — a difficult chore but essential to the project. On this and other projects Vote proved to be a valuable member of the Mackinac Center team. I would also like to thank my peer reviewers: Diana Furchtgott-Roth of the Hudson Institute, Prof. Leo Troy of Rutgers University and Mike Reitz of the Evergreen Freedom Foundation, for their valuable feedback that served to make this a better, clearer paper.