Are Dues Check-Off and Agency Shop in the Public Interest?

In 2011, Republican-majority legislatures in Wisconsin and Ohio passed, and governors in both states signed, legislation reforming government labor relations. Among the specific provisions, the measures sought to restrict collective bargaining for most public workers and eliminate government’s collection of union dues. A pitched battle has ensued, most notably in Wisconsin where Governor Scott Walker faces a recall election that some say will set the tone for November’s presidential contest. This debate will examine the arguments for and against “dues check-off” and “agency shop” legal provisions that benefit public-sector unions. These provisions, under which the government automatically withholds union dues from the salaries of both union and non-union public employees, have been the focal point of debate in Wisconsin and across the country.

Joseph Slater

Joseph Slater

Joseph Slater is the Eugene N. Balk Professor of Law and Values at the University of Toledo College of Law. He holds a B.A. from Oberlin College, a J.D. from the University of Michigan Law School, and a Ph.D. in history from Georgetown University. Be...Read More »

Daniel DiSalvo

Daniel DiSalvo

Daniel DiSalvo is Assistant Professor of Political Science at the City College of New York—CUNY. He is the author of Engines of Change: Party Factions in American Politics, 1868-2010 (Oxford) and co-editor of Building Coalitions, Making Policy (Joh...Read More »

Day 1 – Opening Remarks


Union security agreements and dues check-off provisions exist to avoid free-rider problems.  Unions have a duty to fairly represent all members of a bargaining unit, whether or not an employee pays any dues or not. Further, no employee can be obligated to pay dues that go to activities not related to collective bargaining (e.g., politics).

American law at most allows the “agency shop.” In an “agency shop,” employees are required to pay that portion of their dues that goes to activities related to collective bargaining—mostly contract negotiating and contract enforcement (e.g., arbitration). Union political expenditures are considered not related to collective bargaining. Thus, if a union spends 95% of its dues income on contract negotiations and enforcement and 5% on politics, an employee could object to, and not be charged for, 5% of her dues. 

Banning the “agency shop” creates free rider problems because unions have a “duty of fair representation” (DFR) to all employees in the bargaining unit. A union violates its DFR (and can be liable for significant damages) if it provides inferior representation because an employee does not pay dues.  Thus, in “right to work” jurisdictions (which bar the “agency shop”), unions often must pay for representing employees who don’t pay dues, e.g., in arbitrations. This is a classic free rider problem. I like my cable company, but if it had to give me HBO whether I paid or not, I might not pay.

Making dues check-off agreements illegal is another way to burden unions financially: making it more difficult for them to collect money lawfully owed them. 

Like the anti-union laws of 2011, support for these rules is clearly and sometimes admittedly motivated by the desire to cripple unions financially for partisan gain. These rules lack even the pretense of a relationship to public budgets.


In government labor relations, the “dues check-off” and “agency shop” excessively tilt the political playing field in favor of public employee unions.

According to the Supreme Court, American workers cannot be forced to join unions but they can be required to pay “agency fees” to unions for representational purposes. Today, half of the states allow mandatory contributions to unions. The collection of “agency fees” and union dues by government guarantees public employee unions an abundant and reliable source of money, sparing them the need to spend resources on recruitment, retention, and fundraising. Consequently, government unions have more members and money than they would if government didn’t solve their collective action problems for them and they were forced to operate like other interest groups. Public employees unions have 7.6 million members and annual revenues of more than $1 billion. Few other interest groups can match such an arsenal. In some states and localities, they are nearly unrivaled.

And when it comes to getting their way in American politics, public employee unions have other advantages. First, they have automatic access to policymakers through collective bargaining—something for which other groups must fight. Second, given their common workplace and highly educated and experienced workforce, they can more easily mobilize for political activity than other mass membership groups. Third, government workers unions care intensely about government policy, since it affects them directly, while the attention span of other groups is often short and episodic. Fourth, they are closely aligned with the Democratic Party and do not split their contributions between the two parties like many business groups.

Insofar as groups representing government workers would be powerful political players without laws that smooth organizing and fundraising, these provisions are unwarranted. This is doubly the case when the interests of public employees and the public interest are hardly synonymous and sometimes antithetical.

Day 2 – Rebuttals


Agency fee provisions do not give unions an unfair advantage compared to other “interest groups.” First, unlike other groups (corporations, issue advocacy groups, etc.) unions are obligated to represent all members of the union bargaining unit without regard to whether the worker pays dues. Again, without agency fee agreements, unions face a financially crippling free rider problem. Thus, unions deserve a “reliable source of money” to fulfill their legal obligation to represent all workers in the bargaining unit.

Second, the vast majority of union expenditures from dues income do not go to politics.  Typically, over 90% of such union expenditures go to the prosaic task of representing employees in contract matters that have nothing to do with politics (e.g., arbitrations over discipline). In the leading case, Chicago Teachers Union v. Hudson (1986) (reaffirming the right to object to political expenditures), the union spent 95% of its dues income on matters related to collective bargaining, not politics.

Third, especially after the past year, it is hard to see unions as the 800-pound gorilla of politics. Overall union density is lower than at any time since the National Labor Relations Act (NLRA) was passed. Corporate contributions to the Democratic Party greatly exceed union contributions (and the ratio is much higher for Republicans). Further, unions have well-funded and well-organized opponents. The American Legislative Exchange Council (ALEC) and its allies have successfully pushed many state laws that slash the benefits and workplace rights of public employees. Beyond that anti-tax, pro-privatization, and other interest groups that oppose unions in general and/or on specific topics wield significant influence. 

Finally, many well-publicized claims that public sector unions have exercised their political power harmfully are false or misleading. For example, no significant correlation exists between bargaining rights and budget deficits, and most studies conclude public workers are notoverpaid relative to comparable private workers.


Professor Slater is right that agency shops make free riding on union representation illegal. Workers who refuse to join a public employee union can be forced to pay for representation the union provides. This sounds like (because it is) a form of coercion. Professor Slater nicely advances the positive side of the matter. However, agency shops are controversial and forbidden in half the states precisely because they restrict individual freedom.

Professor Slater is also correct that government workers can’t be forced to pay for political activity. However, in many jurisdictions agency fees are nearly identical to union dues. This provides a huge incentive for workers to join the union, especially when it offers select incentives such as insurance or legal protection. In jurisdictions with agency shops, over 90% of teachers belong to unions; in jurisdictions without them, the percentage of teachers belonging to unions falls about 30 points.

Even workers who exercise their “Hudson rights” and demand a refund of monies spent on political activity find that the amounts returned are often small and determined by the union. Union finances are thus hardly affected.

Professor Slater describes the idea of unions collecting their own dues as a “burden.” Yet that burden is borne by every other interest group in America that raises money for political activity. If dues and agency fee monies were strictly confined to collective bargaining expenses that would be one thing. But they aren’t.

As for crippling the unions financially, this is a hardly a concern at present. For the past twenty-five years, public employee unions have been among the biggest spenders in American politics. The connection to public budgets is also clear: unions’ purpose is to win better pay, benefits, and job protections for their members – all of which drive up the costs of government.

Day 3 – Question from Ellen, a reader in Wisconsin

Thank you both for participating in this debate. As a Wisconsin resident who will vote in the June 5 election but who is still undecided, I am struggling with whether this entire system is in the public interest. I consider the public interest as getting the best service for the least amount of money. This has always been important in Wisconsin. I think that public workers should be allowed to belong to unions and to bargain collectively, but it seems like the whole process costs way too much money. If, as you both cite, between 90-95% of union dues go towards collective bargaining representation and arbitration, it seems like we are talking about billions of dollars a year in taxpayer money going towards a process. What the heck costs so much? So my question: do you agree that too much money is being spent on the process of collective bargaining? If so, how would you improve the system to save my tax dollars? If not, how would you defend this expense. Thank you.


Thanks for your thoughtful question. First, we’ve been debating the union security and “dues check-off” clauses. They cost taxpayers nothing. They only involve whether employees may be required to pay some dues to the union, and whether employers can agree to automatically deduct dues from paychecks (essentially costless). Indeed, the fact that these rules don’t cost employers or taxpayers anything shows that their real purpose is not helping public budgets, but rather crippling unions financially. Again, unions must represent employees whether they pay dues or not.

Second, you ask about the broader costs of collective bargaining. The “90-95%” figure is not taxpayer money going to bargaining, but rather the percentage of dues income unions spend on bargaining. This amount is nowhere near billions of dollars: union dues average a few hundred dollars a year.

Public employers do spend money on collective bargaining mainly negotiating contracts every few years, and in arbitrations over employee discipline (again, nowhere near billions). But without collective bargaining, employers would still have to spend money on various human resources issues, developing and enforcing rules, litigating discipline cases in the civil service process, and etc.

Unions often make human resources more efficient, by providing a unified employee voice for the employer to deal with, and rules forged with the input of employees – who know about their workplace. I’ll list more benefits of collective bargaining on Thursday.

What about the costs of contract terms, mainly compensation? As I mentioned yesterday, most studies find that public employees are not overpaid compared to similar private employees. Further, at some point, paying workers less increases taxpayer costs by producing an unqualified (and demoralized) workforce. Finally, again, no significant correlation exists between bargaining rights and state budget deficits.

If you want to follow up, feel free to e-mail me directly.


Labor-management relations in government are complicated, which translates into greater costs. Contract negotiations require spaces – such as hotels – for contract negotiations, as well as lawyers, administrators, and secretaries to carry out key tasks. The unions must pay a full time staff that includes many lawyers, many of whom are handsomely compensated.

Another big cost is arbitrations over government manager’s attempts to discipline or fire public workers. These processes are set up by collectively bargained contracts or by state laws. The unions’ duty is to protect all workers they represent, so they push for maximal job protections. In California, disciplinary and dismissal decisions by public managers can be reviewed by the Personnel Board, a $26-million public agency with a staff of 170. Over the last ten years, the Los Angeles school district spent $3.5 million trying to fire seven teachers (out of 33,000). It managed to get rid of four. 
For years New York City paid as many as 700 teachers a year not to teach, at a cost of $35 to $65 million a year, while their cases moved through the city’s labyrinthine disciplinary process. In the Empire State as a whole, the average teacher hearing takes 502 days and costs $216,588. As the UC-Davis pepper spray incident revealed last year, California police officers have such robust due process rights that they are unlikely to be disciplined or fired even when there has been clear misconduct.

Private sector workers can only dream of such job protections. These costly procedures result from restrictive collective bargaining contracts, state laws, and public sector unions, who are the power behind the contracts and the laws. Reducing and rationalizing some of these protections would save taxpayers money and make government more efficient.

Day 4 – Closing Remarks


First, respectfully, Prof. DiSalvo’s objections really go to the principle of majority, exclusive representation, a rule central to American labor law. The rule is based on democracy: a majority of employees can vote in a union, vote for different union leaders, vote to change unions, and vote for no union.  If a majority selects a union, the union must incur the costs of representing all employees. Just as citizens must pay taxes even when their preferred candidate loses, it should be legal to charge employees for the costs of representing them when a majority of their co-workers choose representation.  Some European labor laws lack the exclusive, majority principle. But, tellingly, none of the anti-union laws of 2011-12, including those that attempted to illegalize the agency shop, even tried to eliminate the union’s duty to represent employees who don’t pay dues. That’s because the goal was not to re-imagine labor law, but rather to cripple unions.

Second, politicians who would make the agency shop illegal are not concerned about employees being “coerced” or employee “liberty.” Such politicians regularly oppose legal rules that would increase protections to individual employees: anti-discrimination, wage and hour, workplace safety, and similar laws. The push for “right-to-work” is an anti-union move, not a pro-worker move.

Finally, unions do considerable good in society. They help maintain a middle class under siege in an era when income inequality is at unprecedented heights. They bring democratic values and practices to the workplace. Many studies show that unions increase workplace efficiency. I represented unions for over a decade, and I know they aren’t always “right” about everything. But I also know mid-level managers aren’t either. Unions allow an effective method for the voices of workers to be heard in the workplace. Crippling them by legally mandating massive free-rider problems is wrong.


My reading of the evidence on a number of points raised differs from Professor Slater’s. First, in my view, many unions often spend more that 5-10% of their revenues on political activity. Some people estimate that the unions spend as much as 30% or more. An exact percentage is hard to divine but a safe estimate is about 20%. But that translates into huge sums in dollar terms. Consider California’s public-sector unions: they represent about 1 million workers, with estimated annual average dues of $500 per worker. If they have 20 percent of those funds available to spend on politics, that’s $120 million. And in 2010, they spent nearly $100 million on donations to candidates and parties, independent expenditures, and ballot measures. Over the last decade, the California Teacher’s Association alone spent more on politics in the Golden State than the pharmaceutical, oil, and tobacco industries combined.

Second, the available evidence also belies the claim that “corporate contributions to the Democratic Party greatly exceed union contributions.” According to the Center for Responsive Politics, nine of the top 15 spenders on federal elections from 1989-2012 were labor unions (six of them primarily public employee unions). They gave 85% or more of their money to Democrats. The total sum is much more than what Democrats received from AT&T, Goldman Sachs, and Citigroup.

Third, the balance of the scholarly (not think tank) studies of comparable worth point to a premium in public sector compensation. 
Fourth, economists have found correlations between public employee union strength and higher interest rates on state bonds and greater state debt per capita. Neither of which is good for states’ finances.

It is a curious fact of our times that the biggest spenders on politics in many parts of the country are government employees unions. Agency shops and dues check-offs are what make this state of affairs possible.

Joseph Slater

Joseph Slater is the Eugene N. Balk Professor of Law and Values at the University of Toledo College of Law. He holds a B.A. from Oberlin College, a J.D. from the University of Michigan Law School, and a Ph.D. in history from Georgetown University. Before coming to Toledo in 1999, he practiced labor and employment law in Washington, D.C. for over a decade. Selected recent publications include: “The Rise and Fall of SB-5 in Political and Historical Context,” University of Toledo Law Review (forthcoming 2012); “Public Sector Labor in the Age of Obama,” Indiana Law Journal (2011); and “Employee Voice: Lessons from the Public Sector,” Marquette Law Review (2011).

Daniel DiSalvo

Daniel DiSalvo is Assistant Professor of Political Science at the City College of New York—CUNY. He is the author of Engines of Change: Party Factions in American Politics, 1868-2010 (Oxford) and co-editor of Building Coalitions, Making Policy (Johns Hopkins). His scholarly work focuses on public sector unions, political parties, elections, and public policy. He has written on these topics for scholarly and popular publications, including Congress & the Presidency, The American Interest, The Weekly Standard, The Forum, National Affairs, The New York Times, The New York Daily News, and The Journal of Policy History.

Add Comment Add yours ↓

Your Comment