Keynote Address: Disrupted Laws in the Modern Workplace

How many people here are Monty Python fans? Who remembers the opening scene of The Meaning of Life—the scene that depicts a roomful of very old men toiling away as office clerks for The Permanent Assurance Company, hunched over dark desks, abused by their bosses? In the movie, they suddenly join together, rebel, and set “sail” in their building to attack The Very Big Corporation of America’s skyscraper. (Eventually they fall off the world, but we don’t need to follow them that far.) Think of what that workplace looked like, and you’ll have a good picture of what employment law evolved to address. As the many excellent papers that have been prepared for this Symposium make clear, that picture has changed considerably, but the law has not kept up. The “gig” economy doesn’t look much like nineteenth century England and America, nor does it resemble the business environment of the pre-Information Age. Then, the idea of “going” to work had only a literal interpretation—people left their homes, went to a different place, worked, and returned home again.

Although that model has not disappeared, it is now just one of many. Other important changes have also taken place, including most notably the rise over the twentieth century of a network of laws governing workplace interactions—labor laws, antidiscrimination laws, safety laws, and environmental laws, to name a few. These changes have placed enormous stresses on the legal constructs that traditionally have applied to the workplace. Law being the conservative institution it is, we lawyers spend a great deal of time hammering ever-larger square pegs into ever-smaller round holes. Sometimes, though, you just have to buy a different peg—one that fits the hole. This afternoon, I’d like to suggest a number of rules that may be ready for the trash bin, if we are serious in our desire to craft a legal system for the 21st-century workplace.

Let’s start with the basics: who occupies the workplace? Unless you are talking about a sole proprietorship in which the owner does all the work (and to be fair, there are such businesses), there will be a person in charge and another person (or persons) who are working to get the job done—the product assembled, the service performed, or some combination of the two. Sometimes we call the first person an “employer,” and the other people “employees.” Sometimes the first person is called a “principal,” and the others are “agents.” And sometimes the first person is simply called the owner or manager, and some or all of the others are “independent contractors.” Perhaps these different labels meant something at some point, and maybe they even have some residual practical meaning today. Yet aside from allocating responsibility for complying with various parts of the tax laws, for instance, or for determining who has to buy what insurance policies, the categories have lost much of their meaning.

Take, for instance, the standard definition of the word “employee,” which appears in the Tenth Edition of Black’s Law Dictionary, edited by Bryan Garner. An “employee” is “someone who works in the service of another person (the employer) under an express or implied contract of hire, under which the employer has the right to control the details of work performance. . . . Cf. agent; independent contractor.”1 If you go to the definition of “agent,” you find that a human agent is “someone who is authorized to act for or in place of another; a representative. . . .”2 An “independent contractor” is defined as “someone who is entrusted to undertake a specific project but who is left free to do the assigned work and to choose the method for accomplishing it.”3 That definition adds the conclusory comment that “unlike an employee, an independent contractor who commits a wrong while carrying out the work usually does not create liability for the one who did the hiring.”4

Let’s look at this more closely. Suppose I hire a person to do a comprehensive rehab of my condo—we’re going to rip everything out down to the “studs” and modernize, redecorate, and generally transform the place. We spend hours poring over plans, picking out tile, granite, cabinets, bathroom fixtures, hardwood, etc. (ad nauseam). Each day I stop by to see how the work is going. Is the person doing the work an employee or an independent contractor? Most of you would say automatically that he is the latter. But why? It is not at all accurate to say that we have no contract, or that I am not controlling the details of how he does the work. To the contrary, I am micromanaging probably to an annoying degree. Nor is it accurate to say that I have left up to him the “method” of doing the work, at least no more than I leave up to my law clerks the “method” they must use as they research the cases that are coming before me. And yet the clerks are certainly employees of the Judiciary. They are just professional employees who enjoy a great deal of discretion over the precise way in which they accomplish the assigned task.

It is all too easy to multiply examples in which the distinction between an employee and an independent contractor is gossamer-thin, or worse, artificial. That might not bother you if the consequences of placing people into the proper box were equally trivial. But they are not. Consider, for instance, Title VII of the Civil Rights Act of 1964,5 which prohibits employment discrimination on various bases (race, color, religion, sex, or national origin). Title VII addresses employer practices, not practices of those who engage independent contractors. Its operative language, under the heading “Unlawful employment practices; (a) Employer practices” says that “[i]t shall be an unlawful employment practice for an employer” to discriminate with respect to compensation or other terms or conditions of employment on any of the prohibited bases.6 The term “employer” is defined in the statute as “a person engaged in an industry affecting commerce who has fifteen or more employees” during a described period. In order to decide whether the statute even applies, it is therefore necessary to decide whether the person complaining of discrimination is an employee or an independent contractor.7

One of the many cases addressing that topic arose shortly after I joined the Seventh Circuit: Ost v. West Suburban Travelers Limousine, Inc.8 The question was whether a female limousine driver, who had incorporated her own company but who performed services for West Suburban, was an employee (thus protected by Title VII) or an independent contractor (not protected). Ost sued West Suburban, asserting that it had subjected her to discrimination in the driving assignments she received, it had discharged her based on her sex, and it had denied a promotion based on sex. The district court ruled in West Suburban’s favor, finding both that West Suburban did not have enough “employees” to qualify as an “employer” for purposes of Title VII, and that its drivers were independent contractors and thus not protected by the statute.

The Seventh Circuit affirmed. What interests me here about that decision is the way in which it approached the task of deciding whether Ost and her fellow drivers were employees or independent contractors. The court examined five “factors” to guide its analysis:

(1) the extent of the employer’s control and supervision over the worker, including directions on scheduling and performance of work,

(2) the kind of occupation and nature of skill required, including whether skills are obtained in the workplace,

(3) responsibility for the costs of operation, such as equipment, supplies, fees, licenses, workplace, and maintenance of operations,

(4) method and form of payment and benefits, and

(5) length of job commitment and/or expectations.9

Of those factors, it added, the employer’s right to control is the most important.

I have already commented on how slippery that critical point can be, in my condo-remodeling story. At a broader level, it is hard to see what policy is served by making workplace protection against invidious discrimination depend on an ancient distinction between employees and independent contractors. Interestingly, the Supreme Court has jettisoned just this distinction in a closely-related area: political patronage. Back in 1976, the Court decided in Elrod v. Burns10 that government officials may not discharge public employees for refusing to support a political party or its candidates, unless political affiliation is a reasonably appropriate requirement for the job in question.11 In O’Hare Truck Service, Inc. v. City of Northlake12 (maybe not coincidentally, from the Chicago area), it faced the question whether the Elrod principle extends to an independent contractor whose contract is terminated or who is removed from an official list of authorized contractors, in retaliation for refusing to comply with demands for political support.13

Yes, the Court answered, independent contractors do enjoy the same protection, found in the First Amendment, against retaliation based on political affiliation. (Justice Scalia dissented in both O’Hare Truck and in another case decided that day, Board of County Commissioners, Wabaunsee County, Kansas v. Umbehr,14 which considered the question whether the First Amendment protects independent contractors from the termination or prevention of automatic renewal of at-will government contracts in retaliation for their exercise of their freedom of speech.15 Briefly put, his objection was to the Court’s conclusion that the First Amendment prohibits political patronage (or at least some forms of it), in the face of the reality that patronage has existed since the birth of the Republic. He was particularly disturbed by what he described as the expansion of the Elrod principle (rooted in employment) to the “massive field of all government contracting.”16 )

The Elrod-O’Hare line of cases naturally raises the question whether the distinction between employees and independent contractors is doing any useful work, at least for purposes of anti-discrimination principles. That’s a hard case to make, especially in light of the fact that there are many other rules prohibiting discrimination in contracting. For the federal government, such discrimination has been banned since 1969, when President Richard Nixon signed Executive Order 1147817 banning discrimination in the federal civilian workforce on the basis of race, color, religion, sex, national origin, handicap, or age. Later Presidents have expanded those prohibitions to government contractors and have added sexual orientation and gender identity to the list. State laws also prohibit discrimination, and typically reach private conduct that cannot be addressed through measures that depend on state action. Section 1981 of the Civil Rights Act of 186618 prohibits even private race discrimination in many contexts. A patchwork of laws thus seem to add up to a fairly comprehensive ban on discrimination with respect to contractors. Nothing but the drag of old classifications, however, forces us to accomplish the task this way. A fresh, functional look at the responsibilities assumed by different actors in the workplace, free of these labels, would be more efficient as well as more equitable.

Another relic of the past is the use of the contract model to describe the relationship between employer and employee. (I would say the same thing about the relationship between businesses and consumers in the marketplace, but that is not the subject of this Symposium.) Contracts are quintessentially “agreements” between two or more parties. That implies that each side has some ability to take it or leave it; back in some sepia-tinged Utopia, it may even have meant that true bargaining either did take place or could have done so.

Employment markets, however, do not typically work that way, and so it is misleading to use the rubric of contract to describe them. (That fact may well underlie some of the passionate support that existed for Donald Trump’s candidacy: workers who once had jobs that they regarded as good cannot find employment at all, or they must settle for positions that are far inferior to what they once had.) Some fields may indeed be buyers’ markets: information technology, engineering, and health-care administration come to mind. But many remain firmly on the seller’s market side. Reductions in federal funding for the National Institutes of Health make it difficult for newly minted PhD’s to obtain post-doc positions; at the other end of the spectrum, jobs for people with no more than a high school diploma become more difficult to find every year. (And that is primarily not because the jobs have fled to places outside the United States; it is because of automation.) I do not have a long list of recommendations here, but I do suggest that if we understood most employment relationships to be the one-sided creatures that they are, we might re-think such diverse matters as minimum wage laws, protections against arbitrary exercise of the rights that attend “employment at will,” and rights under welfare-benefit and pension plans that confer discretion on the plan administrator.

In this connection, I’ll focus on just one issue that has restricted employee efforts to enforce the protections that they do enjoy, whether under a contract or pursuant to law: mandatory arbitration. Until the Supreme Court’s 2001 decision in Circuit City Stores, Inc. v. Adams,19 it was not clear what the Federal Arbitration Act (FAA) meant when, in section 1, it excluded from the Act’s coverage “contracts of employment of seamen, railroad employees, or any other class of workers engaged in foreign or interstate commerce.”20 The first two groups were clear enough: seamen and railroad employees. But what about others “engaged in foreign or interstate commerce”? An enormous number of statutes use those terms, or their equivalent, when Congress inserts a jurisdictional “hook” into a statute. If all it takes is to work in an industry that affects foreign or interstate commerce, then the rule excluding employer/employee arbitration agreements from the scope of the FAA would be quite broad. (Note that this would not necessarily forbid arbitration agreements, either in general or under other statutes. They just would not benefit from the FAA’s provisions.) If, on the other hand, the last phrase were understood (using the canon noscitur a sociis—a word is known by the company it keeps, perhaps supplemented by ejusdem generis, which calls for general words to be implicitly restricted to items similar to others in a list) as referring only to transportation industries, then nearly all employment arbitration would fall under the FAA.

The Supreme Court, in a 5-4 decision, opted for the latter interpretation. In reaching that result, the majority used a rather formal analysis, rather than a policy-based one. It characterized the final phrase as a “residual clause,” and commented that there would have been no need for Congress to use the phrases “seamen” and “railroad employees” if those classes of worker were already encompassed within the residual clause. The Court also rejected the argument that the 1925 Congress that passed the FAA had no idea that it had the power to legislate with respect to private employment arrangements in every industry that affected interstate commerce, even though those developments were yet to come. Federal-state comity did not bother the Court either: it brushed aside the objections from twenty-one state attorneys general, who were concerned that at a stroke, the Court had wiped out large swaths of state law that restricted the ability of employees and employers to enter into arbitration agreements.

Other developments in arbitration law make it clear that nothing prevents employees (or anyone else) from agreeing to arbitrate public-law issues. Securities-law issues, antitrust issues, and many others that were once thought to be beyond the reach of private dispute-resolution mechanisms, are now routinely included within the scope of enforceable arbitration agreements. It is interesting, however, to see that some push-back has started to develop, and that not all of the push-back comes from the proverbial little guys. That may be for several reasons: first, the increasingly documented fact that arbitration is not as cheap and fast as it is advertised to be; second, the finality of arbitral awards, reflected in the extremely narrow grounds for vacatur or refusal to enforce in the FAA and comparable laws (all well and good if you win, but what if you lose?); and third (as I will discuss in a moment), businesses may not yet have fully put a stake in the heart of collective arbitration proceedings.

The first reason should, in time, lead to a lower incidence of arbitration clauses—even those that are imposed, for all practical purposes, unilaterally—because they would not be paying for themselves. I would like to pause for a moment on the second reason, because the Supreme Court went against the grain in this area and issued a decision that tends to make arbitration a bit less popular. That decision was Hall Street Associates, L.L.C. v. Mattel, Inc.21

In Hall Street, the Court considered the question whether the parties to a contract (there, a lease) have the option of crafting a procedure that is a hybrid of arbitration and litigation. Normally, the choice is binary: either one is litigating in court, using court procedures, and taking advantage of whatever appellate review opportunities exist; or one is engaged in arbitration, in which court involvement is severely limited by the grounds for confirmation, vacatur, or modification listed in FAA sections 10 and 11. The parties in Hall Street wanted the best of both worlds, and so they drafted an arbitration agreement with the following clause:

[t]he United States District Court for the District of Oregon may enter judgment upon any award, either by confirming the award or by vacating, modifying or correcting the award. The Court shall vacate, modify or correct any award: (i) where the arbitrator’s findings of facts are not supported by substantial evidence, or (ii) where the arbitrator’s conclusions of law are erroneous.22

In typical FAA review, courts definitely do not review the arbitrator’s findings of fact for correctness, whether under a substantial-evidence test or any other. Indeed, in the Seventh Circuit we have even eschewed review for “manifest disregard of the law,” on the theory that this does not appear in the FAA and that it is hard to give any clear meaning to it.23

Hall Street argued before the Supreme Court that the FAA’s procedural and substantive grounds for judicial confirmation or review of an arbitral award were nonexclusive, but the Court was not persuaded. Hall Street also argued that, if arbitration is a creature of contract (as all agree it is), then there is no reason why the parties should not be able to specify how far arbitration will go, and at what point the regular courts should re-enter the picture. Again, the Court wasn’t buying it. The central question was whether “the FAA has textual features at odds with enforcing a contract to expand judicial review following the arbitration,”24 and the Court concluded that it does. Everything listed in section 10 and 11 reflects extreme malfunctions in the arbitral process, and so once again the rule of ejusdem generis compelled a finding that “softer” flaws could not be corrected.

By erecting such a high wall between arbitral and judicial procedures, the Court may well have been respecting the language of the FAA. But it also appears to have caused some parties to re-think the desirability of arbitration, given the attractiveness of the option of a second bite at the apple. Some have turned to mediation in lieu of arbitration; others may be litigating. Many, of course, are still arbitrating. For that reason, courts have also been called on to consider restrictions on the arbitral process itself. The restriction that has received the most attention is the one to which I now turn: class or collective arbitration. I mention this as a lead-in to a recent decision of the Seventh Circuit—one that I happened to author.

In AT&T Mobility LLC v. Concepcion,25 the Supreme Court considered “whether the FAA prohibits States from conditioning the enforceability of certain arbitration agreements on the availability of classwide arbitration procedures.”26 The case involved a typical arbitration agreement in an individual cellphone contract. The contract called for arbitration of all disputes between the parties and stipulated that claims had to be brought exclusively in the parties’ individual capacities, not in any group format. The plaintiffs in the litigation, the Concepcions, did not follow those procedures when they wished to complain about AT&T’s failure to provide them with a free phone (as they believed it had promised to do). Instead, they filed a lawsuit in the U.S. District Court for the Southern District of California, and their suit was later consolidated with a putative class action. AT&T moved to compel arbitration; the district court denied the motion based on a California law under which the arbitration clause was unconscionable; and the case made its way to the Supreme Court.

The Court reversed, finding that the California rule had been used in a way that disfavored arbitration, and that such a use conflicted with the FAA. As it introduced that discussion, it said:

the overarching purpose of the FAA, evident in the text of sections 2, 3, and 4, is to ensure the enforcement of arbitration agreements according to their terms so as to facilitate streamlined proceedings. Requiring the availability of classwide arbitration interferes with fundamental attributes of arbitration and thus creates a scheme inconsistent with the FAA.27

It bears noting, however, that it made this comment against the backdrop of an agreement in which the party had promised not to use classwide arbitration (which is available through many providers, such as the American Arbitration Association). Later in the opinion the Court also commented that “the switch from bilateral to class arbitration sacrifices the principal advantage of arbitration—its informality—and makes the process slower, more costly, and more likely to general procedural morass than final judgment.”28 In the end, however, all the Court did was to hold that California’s unconscionability rule is preempted by the FAA.

Let me now turn to a case that lies on the axes between dispute-resolution mechanisms within the workplace and collective rights: Lewis v. Epic Systems Corporation,29 decided by the Seventh Circuit in May 2016 (and in the interest of full disclosure, written by me). This was not a union workplace. The employer, Epic (a health-care software firm) had agreements with each of its employees under which they were required to bring any wage-and-hour claims against the company only through individual arbitration. The agreement then provided that the employee waived “the right to participate in or receive money or any other relief from any class, collective, or representative proceeding.”30 It went on to say, however, that if the “Waiver of Class and Collective Claims” was unenforceable, then “any claim brought on a class, collective, or representative action basis must be filed in a court of competent jurisdiction.”31 Finally, it said that employees were “deemed to have accepted this Agreement” if they “continue[d] to work at Epic.”32 (More of the contract fiction to which I referred earlier.)

At some later point, Jacob Lewis had a dispute with Epic. Rather than proceeding under the arbitration clause, he filed suit against Epic in federal court. From a substantive standpoint, he alleged that Epic had violated the Fair Labor Standards Act and its Wisconsin equivalent. When Epic moved to dismiss and compel individual arbitration, Lewis responded with the argument that the arbitration clause violated the National Labor Relations Act (NLRA), section 7, because it cut off all possibility of joint or collected activities. Because the arbitration clause was unenforceable, he continued, the court action could proceed. The district court agreed and denied Epic’s motion, and we affirmed.

Our reasoning proceeded in two stages. First, we looked at Lewis’s NLRA section 7 argument; then we turned to the issues concerning arbitration. Section 7 provides that “[e]mployees shall have the right to self-organization, to form, join, or assist labor organizations, to bargain collectively through representatives of their own choosing, and to engage in other concerted activities for the purpose of collective bargaining or other mutual aid or protection.”33 The first important point, often overlooked, is that section 7 is not limited to collective action in the form of unionization. It covers any kind of concerted activity. Second, the reference to “other concerted activities” has for years been understood to include (as the Supreme Court put it) “resort to administrative and judicial forums.”34 The Eighth Circuit recognized that a lawsuit filed in good faith by a group of employees to achieve more favorable terms and conditions of employment “counted” as concerted activity for section 7 purposes.35 The Fifth and First Circuits have come to the same conclusion. And “joinder” is not a synonym for “class action.” The more common form of joinder is covered by Federal Rule of Civil Procedure 20, which sets out the (liberal) circumstances under which joinder of plaintiffs or joinder of defendants is permitted. Rule 22 interpleader creates another type of joinder; Rule 23.1 derivative actions another; and Rule 24 intervention another.

We held in Lewis that “the NLRA’s history and purpose confirm that the phrase ‘concerted activities’ in Section 7 should be read broadly to include resort to representative, joint, collective, or class legal remedies.”36 In so concluding, we rejected Epic’s argument that because the Rule 23 class action did not exist in 1935, when the NLRA was passed, the Act could not have been meant to protect an employee’s right to resort to class remedies. First, as I just noted, there are many forms of joint action other than class procedures. Second, Epic was mistaken if it thought that class actions were born with the 1938 Federal Rules of Civil Procedure, or perhaps with the 1966 amendments to Rule 23. As a number of excellent scholarly works have demonstrated, the equity courts had been entertaining class actions for years before the Federal Rules came into effect. Professor Stephen Yeazell has documented group litigation in England as far back as 1199 C.E.37

Importantly, nothing in this part of our analysis said anything about the use of courts versus the use of arbitral procedures. The point was only that section 7 prevents an employer from banning an employee from acting collectively with one or more of her fellow employees.

We then concluded that Epic’s “contract” (using the word loosely) impinged on Lewis’s section 7 rights. The contract flatly prohibits any collective, representative, or class legal proceeding. Assent to this agreement was a condition of Lewis’s continued employment with the firm.

Finally, we concluded that Epic’s agreement was not saved by the FAA. Indeed, we found no conflict between the NLRA and the FAA, because the FAA provides that agreements to arbitrate are enforceable “save upon such grounds as exist at law or in equity for the revocation of any contract.”38 Illegality is one such ground, and so to the extent that the NLRA makes the waiver of collective action unenforceable, it also takes that part of the agreement out of the FAA.

In addition, it is important to note that the Epic contract chose to avoid arbitration if, for some reason, the collective-action waiver was held to be unenforceable. In that event, the contract expressly elected the court system for dispute resolution. We thus did not have before us a contract that opted squarely for arbitration in all circumstances. Had Epic instead said that in the event the collective-action waiver were found to be unenforceable, then any collective proceeding would have to proceed in arbitration, there is no apparent reason why that would not have been enforceable. The fact that the Supreme Court, in dicta, expressed the view that class arbitration is clumsy and expensive, does not require a contrary conclusion. No facts bear out the Court’s pessimistic assessment of class arbitration. It would have been possible, in short, to have respected the employees’ collective action rights and at the same time allowed Epic to opt for an arbitral tribunal rather than the courts.

I freely acknowledge that there is a conflict in the circuits on this point. Should the Supreme Court decide to take the case, it will be interesting to watch from the sidelines. And it is worth recalling the final footnote of Lewis, which (following Seventh Circuit Local Rule 40(e)) says that the opinion was circulated to all active judges in the circuit, and no one wanted to hear the case en banc.

Moving one step further from Lewis takes me to the issue of unions—a dying breed, to be sure, but still important in many workplaces. (USA Today recently reported that in 1979, 34% of male private-sector workers were union members, and in 2013 that number had shrunk to 10%; for women, the corresponding numbers are 16% and 6%.)39 There are many reasons for this decline. Wages for nonunion workers tend to be lower, and some states seeking to attract businesses have taken the view that a good way to do so is to enact strict right-to-work laws—laws that have been permissible ever since the Taft-Hartley Act of 1947.40 These laws provide the backdrop for the last issue I would like to raise this afternoon: the question whether those who choose to stay out of the union in a right-to-work state have any obligation to pay the union for any of the services it is compelled by law to give them. In labor parlance, these have been called “fair-share” dues, but I’ve stayed a little bit away from that phrase in recognition of the fact that some objectors believe that the amount fails to reflect only the benefits conferred and some money slips over into political activities that they are entitled not to subsidize.

This came up in the Seventh Circuit in 2014 in the case of Sweeney v. Pence.41 The majority (Judges Tinder and Manion) found that Indiana’s Right to Work Act did not violate any of the Union’s rights under the Constitution, and that it was not preempted by federal legislation. I disagreed with that outcome and wrote a dissent. I’ll just outline, in a point-counterpoint format, the arguments that the majority found persuasive, and my responses to them.

The Indiana law states as follows:

A person may not require an individual to:

(1)  Become or remain a member of a labor organization;

(2) Pay dues, fees, assessments, or other charges of any kind or amount to a labor organization; or

(3) Pay to a charity or third party an amount that is equivalent to or a prorate part of dues, fees, assessments, or other charges required of members of a labor organization as a condition of employment or continuation of employment.42

Indiana argued, and the majority accepted, that this meant that a worker who chooses not to be a union member cannot be compelled to pay as much as one cent to the union. Section 14(b) of the NLRA, they noted, says that the Act cannot be “construed as authorizing the execution or application of agreements requiring membership in a labor organization as a condition of employment in any State” with a right-to-work law.43

The majority construed the term “membership” to mean not just actual membership, but also any relationship that entailed the obligation to pay any amount in dues. They relied on the Supreme Court’s comment in the General Motors case to the effect that the term “membership” in section 8(a)(3) has been “whittled down to its financial core.” That “financial core” includes “those fees and dues necessary to performing the duties of an exclusive representative of the employees in dealing with the employer on labor-management issues.”

I did not read the term “membership” in the same way. The very case, Communication Workers of America v. Beck,44 that coined the “whittled down to its financial core” language, also made it clear later in the opinion that the Beck objectors were not union “members.” They were nonmembers and thus entitled not to pay “dues.” The only payment that could be compelled was payment for services rendered.

Does the right-to-work law effectuate an unconstitutional “taking”? The majority said no, it does not. It first found this argument forfeited and questioned whether the Union had sued the right people. And I agree with them that the issue was not briefed as well as it should have been. But from an academic standpoint and for today’s purposes, that does not matter. Here are the key justifications the majority offered:

First, it said that “because it is federal law that provides a duty of fair representation, Indiana’s right-to-work statute does not ‘take’ property from the Union—it merely precludes the Union from collecting fees designed to cover the costs of performing the duty.”45 I found this to be an admission that the Union was indeed performing services, as a result of the combined operation of the two laws, for which it was receiving no compensation. Federal law standing alone did not bar the Union from charging fees for services. It was the language in the Indiana statute that did so.

Second, the majority said “we believe the union is justly compensated by federal law’s grant to the Union [sic] the right to bargain exclusively with the employer. The reason the Union must represent all employees is that the Union alone gets a seat at the negotiation table.”46 The duty of fair representation, it added, is a corresponding duty for this exclusivity.

I had a number of responses to that point. First, the duty of fair representation extends well beyond the bargaining table; it covers all union representational activity, including the grievance and arbitration processes that are often established in exchange for a no-strike clause.47 I am willing to assume that the marginal costs of representing the objectors during the bargaining process are low enough that they can be disregarded (though even this is not clear, if a substantial faction of non-union members have a point that they believe should be reflected in the collective bargaining agreement). The majority brushed aside, however, the fact that a major part of the work assigned to most unions under a CBA is not the periodic need to negotiate a new agreement (a task that normally comes up only once every three or four years). It is the administration of the grievance procedure. As I pointed out, assigning a union representative to an employee who wishes to file a grievance, and shepherding that grievance through the various required stages—possibly through arbitration—is expensive. I also pointed out that the Labor Arbitration Rules of the American Arbitration Association outline a comprehensive process that costs real money.48 Under Indiana’s (and comparable) regimes, the Union is forced to perform all of those services for the objectors for free.

If the Union were free to turn away requests for these services when non-union members make them, the case might be different. But it is not. As the Supreme Court stated in Vaca v. Sipes,49 the duty of fair representation requires the exclusive bargaining representative to “serve the interests of all members [of the bargaining unit] without hostility or discrimination toward any, to exercise its discretion with complete good faith and honesty, and to avoid arbitrary conduct.”

No one has argued that the non-members of the Union are indigent, such that one consequence of the privilege of serving as exclusive representative is the duty to perform a certain amount of pro bono work. The Union’s position is thus fundamentally different from the position of the organized bar, for example, whose members have the exclusive right to practice law, but who in return must donate a certain amount of service in the public interest. Indeed, I am aware of no profession or line of business that is required by law to give away its products or services for free, in the absence of any such justification.50

I do not agree that the Union’s status as exclusive representative is valuable enough to justify the free-riding on the efforts of the Union, and the dues of its member, with respect to the bargaining process. I recognize that the Union probably has little interest in carving out the non-members and permitting them to work for cheaper wages. But it would not be difficult to charge the non-members on a piece basis for any post-bargaining services, such as the time of a representative who assisted in the grievance process, or in obtaining proper welfare benefits, or in sorting out the pension plan, or in any other normal post-bargaining activity.

The debate in Sweeney and the general question of the sweep of right-to-work laws raises another aspect of the question whether the legal structure we are using for the workplace still works. De facto, in right-to-work states we have a bifurcated workplace, but we are pretending that the Union speaks for everyone. It is unclear what is wrong with democracy, when it comes to unions: no workplace would be unionized if a majority of the bargaining unit had not voted to take that step; and unions can be de-certified as well if that is the workers’ wish. Moreover, the workplace itself in the “gig” economy does not resemble the one that gave birth to the NLRA and its amendments. Interesting examinations of the way in which unions can or do operate for home-based workers of many types are taking place in the courts and in the law reviews. These are obviously not places like the floor of a huge steel mill or a trucking business. Legislative change to bring the laws closer to the world to which they ostensibly apply will probably be necessary: my own sense is that these problems far exceed the competence of courts.

In the meantime, we judges will continue to pound square pegs into round holes, and we will continue to do our best to apply the doctrines that are baked into the legislation that exists. I look forward to hearing the new ideas that the speakers at this Symposium have sketched out in their abstracts, and I commend the Legal Forum for its choice of topic. Thank you all very much.

• 1Employee, Black’s Law Dictionary (10th ed. 2014).
• 2Agent, Black’s Law Dictionary (10th ed. 2014).
• 3Independent Contractor, Black’s Law Dictionary (10th ed. 2014).
• 4Id.
• 542 U.S.C. §§ 2000e–2000e17 (2012).
• 6Id. § 2000e-2 (emphasis added).
• 7Id. § 2000e(b).
• 888 F.3d 435 (7th Cir. 1996).
• 9Id. at 439.
• 10427 U.S. 347 (1976).
• 11Id.; see also Branti v. Finkel, 445 U.S. 507 (1980).
• 12518 U.S. 712 (1996).
• 13Id.
• 14518 U.S. 668 (1996).
• 15See 518 U.S. at 686 (Scalia, J., dissenting).
• 16Id. at 687.
• 1734 Fed. Reg. 12,985 (Aug. 8, 1969).
• 1842 U.S.C. § 1981 (2012).
• 19532 U.S. 105 (2001).
• 209 U.S.C. § 1 (2012) (emphasis added).
• 21552 U.S. 576 (2008).
• 22Id. at 1400–01.
• 23George Watts & Son, Inc. v. Tiffany and Co., 248 F.3d 577 (7th Cir. 2001).
• 24Id. at 586.
• 25563 U.S. 333 (2011).
• 26Id. at 336.
• 27Id. at 344.
• 28Id. at 348.
• 29823 F.3d 1147 (7th Cir. 2016).
• 30Id. at 1151.
• 31Id.
• 32Id.
• 3329 U.S.C. § 157 (2012) (emphasis added).
• 34Eastex, Inc. v. NLRB, 437 U.S. 556, 566 (1978) (collecting cases).
• 35Brady v. Nat’l Football League, 644 F.3d 661, 673 (8th Cir. 2011).
• 36823 F.3d at 1153.
• 37See Stephen C. Yeazell, From Medieval Group Litigation to the Modern Class Action 38 (1987).
• 389 U.S.C. § 2 (2012).
• 39Paul Davidson, Decline of Unions Has Hurt All Workers: Study, USA Today (Sept. 5, 2016), http://www.usatoday.com/story/money/2016/08/30/decline-unions-has-hurt-all-workers-study/89557266/ [https://perma.cc/DF5B-3DCP].
• 40See Labor Management Relations Act of 1947, 29 U.S.C. §§ 149–197 (2012).
• 41767 F.3d 654 (7th Cir. 2014).
• 42Ind. Code § 22-6-6-8 (2012).
• 4329 U.S.C. § 164(b).
• 44487 U.S. 735 (1988).
• 45Sweeney, 767 F.3d at 666.
• 46Id.
• 47See Vaca v. Sipes, 386 U.S. 171, 190–91 (1967).
• 48Sweeney, 767 F.3d at 672.
• 49Vaca, 386 U.S. 171 (1967).
• 50Id. at 177.