The Misconceived Modern Attack on Right to Work Laws

After a long period of relative quiescence, labor and employment law have once again become hot topics on both the legislative and judicial fronts. One area of deep contention concerns state right to work (RTW) laws in the private sector, controversy over which, along with two recent judicial opinions on the topic, will be the focus of this paper.

Two decisions, Judge Diane Wood’s dissent from the Seventh Circuit’s decision in Sweeney v. Pence,1 and the opinion of Wisconsin Circuit Court Judge C. William Foust in International Ass’n of Machinists District 10 v. Wisconsin2 have both insisted that the state implementation of right to work laws (RTW), which allow nonunion members in a bargaining unit not to pay any dues or fees to the union representative, count as uncompensated takings of union’s right to collect some dues from these nonunion members. Judge Wood also insisted that under the National Labor Relations Act the entire governance of labor-management relationships is governed by federal law, which preempts the ability of the states to insulate those workers who opted out from the union from paying their “fair share” of the union’s expense in order to overcome the freerider problem that otherwise would arise.3 Judge Wood’s opinion was a dissent, and Judge Foust’s decision was recently overturned by an appellate court.4

Nonetheless, the issues remain sufficiently salient that a more thorough examination of this problem seems appropriate, especially since of these RTW laws are now in effect in twenty-eight states.5 The distribution is far from random. There are no RTW states in the northeast or along the west coast, but every southern state, as well as many Midwestern states, furnish RTW laws. In the last five years the ranks of RTW states has increased as Indiana (2012), Kentucky (2017), Michigan (2013), Missouri (2017), West Virginia (2016) and Wisconsin (2015) have adopted such laws. As the RTW movement has gained ground, so too has the resistance to its expansion. Union stalwarts see RTW laws as a mortal threat to their continued existence. The AFL-CIO states, “The real purpose of right to work laws is to tilt the balance toward big corporations and further rig the system at the expense of working families. These laws make it harder for working people to form unions and collectively bargain for better wages, benefits and working conditions.”6 In contrast, the National Right to Work Foundation puts the issue in terms of individual autonomy, “A Right to Work law secures the right of employees to decide for themselves whether or not to join or financially support a union.”7

This article is organized as follows. The Section I outlines the key feature of RTW laws. Section II then develops the two sides of the takings argument, the first of which involves the free-rider arguments raised by the unions and the second the conflicts of interest claims raised by dissenting workers. Section III then explains why the duty of fair representation (DFR) is not sufficient to deal with these conflict claims. Section IV then analyzes and rejects the claim that federal law preempts any effort by the state to enact a RTW law that prevents the union from claiming a “fair share” payment for the services that it renders dissident workers within the bargaining unit. Section V then returns to the takings issue by looking at how strict scrutiny and rational basis tests apply to both the NLRA and to its RTW provision.

I.     The Structure of RTW Laws

In order to get to the merits of this dispute it is important to understand the legal environment in which these rules are adopted. As traditionally understood, RTW laws do not guarantee that every person gets a job. Instead these laws have long been understood as a counterweight to the risk of compulsory unionization otherwise created by statute under the National Labor Relations Act (NLRA),8 whose 1935 incarnation in the Wagner Act was one of the decisive moments in the New Deal.9 That statute authorized a system of collective bargaining, in which any union that won a representation election became the exclusive bargaining agent for all the workers inside that bargaining unit, whether or not they voted for the union.10

Three manifestations of compulsory union-employee relationships subsequently emerged from this system. Under the original 1935 arrangement, the union could insist on a closed shop, under which all covered employees had to be union members in order to keep their jobs.11 A variation on the closed shop is the union shop, which allows for the hiring of non-union workers, but only on condition that they agree to join the union within a certain period of time after employment, usually thirty or sixty days.12 A third variation on the workplace arrangement, which holds for public as well as for private unions, is the so-called agency shop. Under this arrangement, employees need not join the union, and hence are not subject to its dues and other membership requirements, but are required to pay the union an “in lieu” fee equal to union dues.13 Collectively, these three kinds of relationships are commonly called “union security” arrangements for the protection that they afford labor unions that have secured the right to represent workers in a particular bargaining unit. It is now settled that any payments under these bargaining units can only be directed toward core union operations dealing with the collective bargaining agreement, but do not cover political expenditures undertaken by the union, for which workers have the right to opt out.14 However, RTW laws have established a fourth union-employee relationship whereby it is possible to have an open shop arrangement, under which individual workers are neither members of the union nor required to pay dues. Yet nonetheless these nonmembers both benefit from and are bound by all the terms of the union agreement. It is the nature of this arrangement that has become the focus of contemporary controversy.

The 1947 Taft-Hartley Act profoundly altered the New Deal-era situation with the passage of Section 14(b), which created a large exception to the three originally recognized union security arrangements (closed shop, union shop, and agency shop) that were otherwise allowed for under section 8(a)(3). Section 14(b) reads: “Nothing in this Act shall be construed as authorizing the execution or application of agreements requiring membership in a labor organization as a condition of employment in any State or Territory in which such execution is prohibited by State or Territorial law.”15 The effect of this provision, as stated in the standard hornbook on the subject by Professors Gorman and Finkin, is:

[T]o retreat from the all-embracing federal principles which characterize the Labor Act and to permit the states to enact right-to-work laws to invalidate union security provisions that are otherwise lawful under section 8(a)(3) provisos. The term “union security” embraces a number of different kinds of arrangements designed to bolster the membership and finances of [the] union.16

And, as Gorman and Finkin note, these union security arrangements include the closed, union, and agency shops, without differentiation.17 It is Taft-Hartley’s Section 14(b) that opened the door to modern RTW laws.

II.    The Takings Law Dispute: Free Riders and Conflicts of Interest

The logic of this position was erroneously rejected, as noted above, in Sweeney v. Pence,18 and Machinists.19 The purpose of this paper is to offer a critique of both opinions. Both of these decisions held that the application of the RTW provisions violated the Takings Clause of Fifth Amendment to the United States Constitution, which provides, “nor shall private property be taken for public use, without just compensation.”20 One basic claim made by the plaintiff unions was that RTW laws deprived them of the “in lieu” dues they were entitled to from nonmembers, while the unions were still required to provide services to these nonmembers. Thus, a taking allegedly had occurred. However, litigation under the Takings Clause is complicated because of the difficulty in choosing the level of scrutiny that courts bring to takings challenges. The basic distinction is that a per se takings rule, with strict scrutiny, is applied to the physical occupation by the government of property owned by any private property.21 But the far-lower threshold rational basis test is applied to laws that only impose restrictions on the use or disposition of private property.22 These two tests, both of which were originally developed in connection with real estate, have been carried over to other areas of the law, including labor relations, with at best mixed results.

In my view, the decisions of both the Seventh Circuit and Wisconsin Circuit Court are wrongheaded because the consistent application of either standard of judicial review can both uphold the constitutionality of the basic NLRA and strike down the RTW. Either both stand together under a rational basis test, or both fall under a per se taking rule. A high standard of judicial review knocks out (correctly in my view) the NLRA in its entirety including the subsidiary RTW provisions. Conversely, any lower standard of review that upholds the NLRA necessarily validates any permutation on the basic scheme, including the RTW accommodations to the NLRA that have been allowable since the enactment of Section 14(b).

As mentioned briefly before, the basic union claim is that RTW laws should be rejected because they create an intolerable free-rider situation under which workers who opt out of the union get all the benefits of union membership without having to pay for those benefits. In perhaps the most well-known articulation of this proposition, at least outside the context of union speech, Justice Scalia observed:

What is distinctive, however, about the “free riders” who are nonunion members of the union’s own bargaining unit is that in some respects they are free riders whom the law requires the union to carry—indeed, requires the union to go out of its way to benefit, even at the expense of its other interests. In the context of bargaining, a union must seek to further the interests of its nonmembers; it cannot, for example, negotiate particularly high wage increases for its members in exchange for accepting no increases for others. Thus, the free ridership (if it were left to be that) would be not incidental but calculated, not imposed by circumstances but mandated by government decree.23

Judge Foust in Machinists cites Lehnert v. Ferris Faculty Association24 as consistent with its own conclusion that by virtue of RTW, “[a] free-rider problem is born—the ability of non-members to refuse to pay for services unions are compelled to provide by law.”25 Judge Wood in Sweeney takes the same position: “If there is no way to compel the nonmember employee to pay the actual cost of the services the union is obligated to provide for him, a classic ‘free-rider’ problem arises.”26 The argument is then strengthened by the claim that unions are bound by a duty of fair representation (DFR) to advocate for, and bear the costs of, their nonmembers, as well as themselves, in the collective bargaining negotiations and also to represent the nonmembers in grievance procedures. Given their role as the exclusive representative, unions are bound 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.”27 In effect, this DFR protects the nonmembers against harmful union conduct and obligates, on Judge Wood’s reading, the union to provide certain services.28 This is the false foundation on which the violation of the Takings Clause is subsequently presumed.

The attacks on the RTW laws offered in Sweeney and Machinists are profoundly wrong in three relevant dimensions. The first is that neither decision correctly understands the economics of freeriding within the union context. The second is that, as Gorman and Finkin noted, Section 14(b) of the NLRA insulates dissident workers from all forms of union security agreements, including agency shops.29 The third is that the effort of both courts to apply a takings analysis to this case is badly misguided, regardless of whether the case is characterized as a taking of services, cash, or other forms of property. The brief explanation for this conclusion is that a takings analysis of RTW laws simply cannot presume the constitutionality of the NLRA, which in its current form could not survive a takings analysis under any heightened level of judicial scrutiny. The purpose of this critique is to explain why RTW laws are socially desirable, and why they should escape constitutional condemnation.

III.  The Duty of Fair Representation Does Not Eliminate Serious Conflicts of Interest within the Union Context

The critique of RTW starts with the proposition that the DFR offers sufficient protection to dissident workers, whether or not union members. Such a presumption is the only manner in which to establish the relationship between unions and employees necessary to sustain the logic of the takings claim. A closer examination of the case law shows that this argument fails in all relevant circumstances, including the case of Vaca v. Sipes,30 which I critiqued many years ago in my student note in the Yale Law Journal.31 The key objections to this reliance on the DFR are two-fold. First, the argument wrongly presumes that a union under the DFR is bound to represent union members during the grievance process. In fact, the statutory scheme only requires impartiality in the basic negotiation of the collective bargaining agreement, while leaving it to the workers themselves to handle their own grievances at their own expense. Second, at the negotiation stage, the DFR is unable to overcome the acute conflicts of interest among the many workers to whom it owes that duty.

A.    Control of Individual Grievance under Vaca v. Sipes

In Vaca, the union refused to process the union member’s grievance beyond the first four stages of the internal arbitration procedure.32 The plaintiff then sought to maintain a legal action on his own account, but the Court held that the union had acted fairly and impartially when it decided in good faith that it would not go forward with that grievance.33

The evident conflict of interest in such a case arises because the individual worker has all his eggs in a single basket, while the union has to deal with a broad class of cases and cannot devote all its resources to individual claims, especially ones it regards as suspect. Justice White acknowledged the conflict of interest when he wrote: “The collective bargaining system as encouraged by Congress and administered by the NLRB of necessity subordinates the interests of an individual employee to the collective interests of all employees in a bargaining unit.”34 It is surely correct to say, therefore, that the union need not back this particular claim, or any other individual claim, in light of the evident conflict. It is, however, a very different proposition to insist that its good faith decision can block the individual’s pursuit of his own claim, where he has a far greater interest. Indeed, the point was well recognized under the NLRA in Section 9(a),35 nowhere mentioned in Vaca. That key section sets out an explicit arrangement whereby the individual worker is entitled to press his own grievance consistent with the collective bargaining agreement. As an offset, the union is allowed to attend the grievance session to protect its own interest by seeing that the settlement does not go outside the boundaries of the underlying collective bargaining agreement, given the possible implications for other cases.36 It is for this reason why the union is allowed “to be present” at the proceeding, so it can present its views of the contract provisions.37 But this last caveat does not wrest the control of the grievance from the worker, because to read it in this fashion nullifies the initial allocation of control in its entirety. The statutory process set out in Section 9(a) offers a far superior accommodation of the relevant interests than a simple rule that lets the union take over, subject to an implicit obligation nowhere stated in Section 9(a). The fact that unions typically negotiate for such control is not consistent with the underlying structural safeguard. This statutory right is insulated from reversal by the collective bargaining agreement.

One consequence of following the statutory rule is that it undercuts the claim made by Judge Wood in Sweeney that the union is somehow entitled to receive compensation because the DFR sometimes requires the union to bear the costs of presenting the grievance to the employer for which it receives no compensation at all. But if the correct arrangement of Section 9(a) were followed, the worker would have to bear that cost by himself unless he made some arrangement to purchase insurance, whether from the union or a third party, to cover those expenses. Simply stated, union compensation is not a right in every circumstance, and as such cannot possibly provide the basis for the validation of the takings claim. Indeed, the statutory rule here is no different from one that requires each party to bear his own cost in litigation. It is also the case that the worker can, of course, enter individually into any agreement that cedes the union total control over the case. But what cannot be done is for the union to force this result as part of the collective bargaining agreement. Section 9(a), passed as part of Taft-Hartley, was intended to provide worker protection against the union, not to render the employee subservient to it. Hence once the correct rule is used in Vaca, any case for requiring the state to compensate the union for its loss of dues under the Takings Clause falls apart, given that the union has no statutory duty to process and pay for the grievance.

B.    DFR in Negotiation

Even if the DFR has no proper place at the grievance stage, it does, however, have a role to play at the contract negotiation stage where no single worker has a disproportionate stake in the outcome. But it is nonetheless fair to ask whether the DFR offers adequate protection for dissenting workers at the negotiation stage. Unfortunately, the DFR is not up to the task. It is at best a blunt instrument that cannot, even with the most conscientious union, paper over internal disputes within a firm. Nor when unions act in bad faith is it strong enough to protect the local minority from the majority faction.

In 1944, the DFR was created judicially in Steele v. Louisville & Nashville Railroad Co.,38 which arose under the Railway Labor Act of 192639 to deal with a manifest case of union bad faith. Prior to the 1926 Act, black and white railway workers were in separate unions. The law required the appointment of a single union to represent all workers, which meant in practice that white workers dominated the union and systematically relegated the minority workers to inferior positions—an arrangement present on all major railroads.40 The evident injustice prompted the Supreme Court to invent the DFR to redress the balance. But in truth, the situation for the minority workers grew far worse, because they always had to battle a hostile leadership. Indeed, thirteen years later, the famous civil procedure case Conley v. Gibson41 showed how little progress minority workers had made in disputes of this sort since Steele. A separate union with its own leadership would have offered these workers far greater protection.

The situation does not resolve itself when matters of racial prejudice are put to one side. Even under ideal conditions, conscientious unions have to make all sorts of decisions that impact all members, and which can disproportionately affect differently situated individuals. Do they make aggressive demands in collective bargaining negotiations, or do they settle for less in order to achieve greater certainty? Or what should be done when two firms merge so that it becomes necessary to integrate the seniority lists for the two plants: is the advantage given to the incumbent workers at the surviving plant, or are the two lists dovetailed? In Humphrey v. Moore,42 the Court rightly held that any integration of the two lists would necessarily prejudice some workers no matter which way it was decided.43 A union cannot be subject to a damned-if-you-do, damned-if-you-don’t legal regime in every case where it has to make hard choices, and hence the Court insulated the union from the charge that it breached its duty.

The DFR is similar to the norm that governs in corporate contexts under the business judgment rule. That rule typically requires directors and officers of the corporation to act reasonably and in good faith for their shareholders.44 The raison d’être of the rule is that no one will take on these duties if they can be held liable when their responsible choices go wrong, but must then stand aside when the shareholders reap the benefits of all successful decisions. The insulation from liability for decisions made through fair processes and with honest judgment is a pillar of the corporate law, and it is only displaced by the “fair value” rule in those instances where a conflict of interest puts these officials on both sides of a given transaction. At that point, they are required to show that their deal is entirely fair to the shareholders as a class.

A variation on this rule is at stake in cases like Vaca, but with two key differences.45 In the business context, shareholders in any public corporation can diversify their holdings in a way in which individual employees cannot.46 In addition, shareholders of public corporations who dislike the corporate position can sell their shares in a ready market.47 But individual workers, often with accumulated seniority and entrenched interests, must incur huge losses if they decide to resign from the firm, given that they cannot sell their union position to anyone else. The point here is not to say that Humphrey was wrongly decided under the law. It was not. It is only to insist that, even when correctly applied, the DFR offers far weaker protection to minority workers than the conventional accounting presupposes. Even if these workers are allowed to not pay dues, there is no way that they can repudiate the union representation in order to strike their own deal, or to reject individually deals that they do not prefer. Still others who support the union are leery about taking a passive stance on negotiations lest the terms negotiated for the group be less suitable for them than those that they might have able to obtain when they participate inside the union. Thus, even though some workers surely think of themselves as free-riders, many, even if only a minority, of these dissident workers will not regard themselves as free-riders, given that they would prefer to remain outside the union even if they had to contribute zero dollars to its support.

Judge Wood in Sweeney shows her awareness of many of these difficulties when she writes that there is nothing “inevitable” about the current structure of American labor law, which could be reconfigured to give minority workers additional rights.48 But she adds: “to repeat, that is not the system that the United States has adopted.”49 She also makes a political analogy when she notes: “This is hardly an unfamiliar arrangement in a democracy. Even after the most hotly contested presidential election, the person who is declared the winner becomes the President for all citizens, not just those who voted for him or her.”50

This analogy to political governance breaks down at the most fundamental level. The reason we have elections for presidents is that government must possess a monopoly of force within the nation in order to provide peace and order and other collective goods. There is no individual exit option, as there is in voluntary markets, where the repeal of the NLRA results in clear movement to competition, which provides stabilized state protection of private property and contractual relations. It is, however, a mistake to trivialize the collective action problem by assuming the sole function of government is to overcome free-rider and hold-out problems that pose obstacles to collective life, which could be overcome by a tax on all that leaves each person better off than before.51

That collective action model only works where goods are homogenous and equally valued by all alike—an assumption that does not even hold with respect to the provision of street lights: any given light may disturb the sleep of nearby residents even as it provides needed illumination to passers-by. It certainly does not hold for the great questions of war and peace, where, after the deliberation, the dissenters are bound, even if they are by no stretch of the imagination free-riders on the expenditure of others. But given that exit is possible in labor markets, why tolerate this easily avoidable mass dissatisfaction? Union supporters will say it is because they need economic strength to secure their interests. But that claim is no better than the claim of industry moguls who need protection of legally enforceable cartels in order to secure maximum benefit from their operations. The early labor law refused to recognize any such special privilege. Loewe v. Lawlor52 subjected unions to the full force of the antitrust law, only to see that decision overruled by Section 6 of the Clayton Act,53 which exempted unions from its provisions, and thus set in motion a set of special privileges ultimately embodied in the NLRA. Thus, any justification for binding dissenters in governmental contexts does not carry over to dissenting employees in labor markets.

The RTW laws do not create complete exit rights for dissenting workers because they cannot escape union representation even if they need not pay dues. But the competitive pressure created under RTW laws has a generally positive effect on union operations. Two amicus curiae briefs written in Friedrichs v. California Teachers Ass’n,54 which dealt with First Amendment issues outside the scope of this essay, illustrate the point.55 Once unions are faced with the possibility of defection, they will take steps to counter those losses by becoming more responsive to the interests of actual and potential members. The salaries for union officials go down, as do dues, on the order of ten percent, which means a better deal for union members, leading to a reduction in membership losses. Union members are still worse off than before the adoption of RTW laws, but the conflict of interest between unions and workers, which is hidden by the current rhetoric, is more readily resolved by market effects. No union membership is perfectly homogenous. The DFR thus cannot simultaneously protect two classes of workers: those who would rather negotiate without the union, and those who lose from a collective bargaining interest that prefers some other faction to their own.

IV.  Membership Requirements under Section 14(b)

In Sweeney, the union made the claim that the RTW law in the case fell outside the scope of Section 14(b) because it called for a total elimination of all dues, and therefore did not allow the union to recover from these workers their “fair share” of the costs of collective bargaining, a point Judge Wood accepts in her dissent.56 Initially, the point seems odd because there is no reference in either the statute or the legislative history to a fair share regime, nor any indication of just how that fair share might be calculated. Indeed, the key cases on which Judge Wood relies to deal with this matter point quite clearly in the opposite direction.

The initial statement of the problem is found in Retail Clerks v. Schermerhorn,57 as follows:

At the very least, the agreements requiring “membership” in a labor union which are expressly permitted by the proviso are the same “membership” agreements expressly placed within the reach of state law by § 14(b). It follows that the General Motors case58 rules this one, for we there held that the “agency shop” arrangement involved here—which imposes on employees the only membership obligation enforceable under § 8(a)(3) by discharge, namely, the obligation to pay initiation fees and regular dues—is the “practical equivalent” of an “agreement requiring membership in a labor organization as a condition of employment.” Whatever may be the status of less stringent union-security arrangements, the agency shop is within § 14(b).59

The position was reiterated in Retail Clerks v. Schermerhorn,60 where Justice Douglas was, if anything, even more emphatic, when noting that “even if the union-security agreement clears all federal hurdles, the States by reason of § 14(b) have the final say and may outlaw it.”61 In the face of this stern language, Judge Wood claims the only impact of these decisions was to allow states to outlaw a requirement of full payment of union dues, but to leave open the payment of lesser amounts under the fair-share obligation.62 But that point is surely wrong. In Retail Clerks I, the RTW clause in question exempted from dues contributions issued toward political activities that were covered in the earlier General Motors case.63 The case law takes the view that all types of agency clauses are governed by the same rule. Thus, in International Union v. NLRB,64 the D.C. Circuit wrote as follows:

The Union’s “representation fee” is a “less stringent union-security arrangement” than the fee in [Retail Clerks I] because it is not set to equal union dues. The representation fee thus escapes [Retail Clerks I’s] holding, but not its rationale as restated in recent dicta: “Section 14(b) simply mirrors that part of § 8(a)(3) which focuses on post-hiring conditions of employment.”65

Judge Wood tries to deflect this uniform line of case law by accusing the majority of double think for “its decision to assign the status of statutory ‘members’ to nonmembers of the union,”66 which sounds like a contradiction until it is understood against the paramount need to ensure that judicial interpretation does not negate the statute by allowing a union to insist on its practical equivalent, namely by collecting dues from nonmembers. The effort to use this form of anti-circumvention norm is par for the course in all areas of interpretation, in order to make sure that close substitutes do not have disjointed treatment from those activities that fall within the statute.67 The standard definition of a union security agreement is intended to preserve that functional equivalence.

To be sure, it does not follow that the reach of Section 14(b) is unlimited. In NLRB v. Houston Chapter, Associated General Contractors of America, Inc.,68 the Fifth Circuit held that nothing in Section 14(b) prohibited a clause that required a nondiscriminatory hiring hall in the construction industry.69 “[T]he long and the short of this matter is that § 14(b) contemplates only those forms of union security which are the practical equivalent of compulsory unionism.”70 There is not the slightest suggestion that this case even questioned the broad reach of Section 14(b). Nor is there any reason to think that some dues reduction destroys the “practical equivalen[ce]” and the various fees. The initial statement of Gorman and Finkin has to be correct because Section 14(b) allows the state to knock out all forms of agency shop arrangement in their entirety.71 But that section does not allow the individual workers to repudiate union representation so long as they remain in the bargaining unit.

V.    A Multi-layered Takings Analysis: Strict Scrutiny versus Rational Basis

Once it is held that Section 14(b) does not preempt the ability of the state to outlaw all forms of the agency shop, the question arises whether that Section can be challenged under the Takings Clause of the Fifth Amendment. As the matter was teed up, the only question before the court in Sweeney and Machinists was whether the states can use their RTW laws to extricate nonmembers from dues without paying the union a sum equal to the dues lost from their membership. The implicit theory is that the union has either a property right or some investment-backed expectation in the continued receipt of the dues that were “taken” by the state once the RTW statute is put into play. In both Judge Wood’s Sweeney dissent, and Judge Foust’s Machinists opinion that follows it on most particulars, the key cases are federal constitutional cases in support of the proposition that just compensation is needed before the union loses any advantage conferred upon it by the NLRA.72 On this view, it would seem that even Congress could not amend the NLRA to impose, at the federal level, a RTW law similar to that adopted in Indiana and Wisconsin.

Analytically, there is no reason why the rational basis analysis that sustained the basic NLRA statutory scheme would require a different result if Congress decided to move the balance a bit closer to management’s side of the ledger rather than labor’s side. Once the federalism overtones are removed from the debate, adding the RTW principle into the NLRA is just one of a number of permissible statutory options open to Congress. It is therefore a mistake to argue that the adoption of RTW necessarily creates a one-way ratchet, so that once that option is put into place, it cannot be taken out. It cannot be the case that even though Congress has no constitutional duty to pass a collective bargaining statute, it must give a union just compensation from the federal government for the cost of running its operations for the benefit of dissident workers, when it has already conferred upon them enormous monopoly power.

Indeed, to take this position raises issues never broached on the passage of Taft-Hartley, for, if adopted, a constitutional gloss on RTW laws strengthens the union position relative to a statute that contains no such RTW option. Under Judge Wood’s proposed regime, the union should now be quite happy to encourage defections by disgruntled workers because it recoups all the lost revenue from the state or federal government, as the case may be. Hence it gets to rule with an iron hand because the dissidents now have no say in the governance structure of the union. It is inconceivable that either the federal government or the states would write this kind of blank check, so that the proposal in practice spells the end of RTW laws in all forms.

In light of these considerations, one should pause long and hard before reaching the conclusion that the elaborate political compromise which led to the passage of the NLRA is now constitutionally mandated by the Takings Clause. Nonetheless, ironically, just that view has been attributed to me by Heather Whitney, who in writing about just compensation issues set up the problem as follows:

If you take the position that “all regulations, all taxes, and all modifications of liability rules [should be understood as] takings of private property prima facie compensable by the state,” the answer is easy: yes. Conversely, if you are an originalist and believe that the Takings Clause was narrowly meant to require compensation when the government exercises its power of eminent domain—that is, when the government physically seized property—the answer is also easy: no.73

The quoted material in the first part of the passage is from my book Takings: Private Property and the Power of Eminent Domain,74 and the second is from William Treanor’s influential article on the original understanding of the Takings Clause, which takes the view that the Takings Clause only applies to physical takings and not to regulations that simply limit use and disposition of property.75 In my own view, it is not possible to defend an originalist, or indeed any other, view of the Takings Clause that limits its scope solely to physical seizure or occupation. Does one want to insist that there is no taking when the government prevents by force an owner from entering his own property that the state in turn does not occupy, when the action is surely the practical equivalent of taking a restrictive covenant over the entire property? But for these purposes, Whitney does capture my position on the comprehensive reach of the Takings Clause: that the Takings Clause covers, prima facie, all partial takings by government that include the loss of air rights, mineral rights, liens, or reversionary interests.76

What is common about both Wood’s Sweeney dissent and Foust’s Machinists decision is that they take for granted the current intellectual framework of takings law, without thinking through its structure as a matter of first principle. The gap between my position in Takings and the current law, which gives scant protection to regulatory takings, is indeed enormous. My essential argument in Takings was that all takings are prima facie covered in the same fashion, whether they are for a total or a partial interest in property. The range of partial interests in property is huge, for it includes all the various life estates and remainders, leases, mortgages, covenants, and easements. To be sure, the comprehensive theory includes extensive set-offs to the prima facie that include the police power justifications on the one hand, and, on the other, rules dealing with implicit-in-kind compensation that come into play when a broad regulation simultaneously benefits and burdens the same group of individuals.77 There is no reason to cut down on the broad definition of private property that faithfully tracks the common law definition.

Unfortunately, the Treanor approach requires drawing a categorical distinction between “mere” restrictions on land use and physical occupations that are built into the law in Penn Central Transportation Co. v. City of New York,78 and on which the Machinists analysis is said to rest. But the conceptual line is untenable: Is keeping someone from using his own air rights, a recognized interest at issue in Penn Central,79 a physical or a regulatory taking? Should no compensation be provided if the government excludes the owner from property rights even if it chooses not to occupy those rights itself ? The case looks far closer to those, most notably Loretto v. Teleprompter Manhattan CATV Corp.,80 in which there is a permanent or temporary occupation of property by or authorized by the government that counts as a per se taking, i.e. one that is subject to a few narrow exceptions.

In order to see how these alternative frameworks apply to RTW laws, it is necessary to first ask how they apply to the NLRA as originally enacted before Section 14(b) and RTW. As I have long argued, the NLRA is dead on arrival under the classical liberal view.81 At the very least it forces employers to do business with workers in competitive industries when there is no common carrier arrangement involved. The loss of the ability to choose one’s trading partners is a limitation on the rights of disposition of their property, including their capital. Yet there is no return benefit of any sort to the employer, nor plausible argument that an ordinary business is a public or private nuisance that the state can regulate under the police power. At this point the entire NLRA falls. It was just that conclusion that was rightly reached in two key pre-1937 cases, Adair v. United States82 and Coppage v. Kansas,83 which were written in the classical liberal tradition. Sadly, both cases have been overruled, but if the higher level of scrutiny that they embody still applies, there is no reason to worry about the RTW exception to the basic law. It falls when the NLRA falls under the combined weight of these two distinctions. It is not likely that any defender of the modern labor law would want to embrace this analysis.

The next question is how the NLRA fares under the more modern analysis that does not give substantial protection to economic liberties. Again, the question of characterization makes this issue harder than it seems. The statute, like the rent control statutes,84 was upheld before the modern distinction between physical and regulatory takings was adopted. But even if one takes the Treanor view that limits the Takings Clause to physical invasions, both rent control and labor statutes are unconstitutional under Loretto v. Teleprompter Manhattan CATV Corp.85 because the government has authorized the CATV company to physically enter a landlord’s property, or a firm’s plant, without requiring payment of just compensation.86 Judge Wood accepts a version of this proposition when she writes that what the Interest on Lawyers Trust Account (IOLTA)87 cases establish “for us is that a state law compelling one private party to give property to another private party must be assessed under the Takings Clause. The fact that those two cases involved money [IOLTA], while our case involve the compulsory provision of services, is of no moment.”88

Although she does not discuss it, Judge Wood’s same logic carries over to rent control laws where the landlord renews the lease of a tenant for rates below the fair market value required under the Just Compensation Clause, but there is no liability under current law. Her logic also applies to Republic Aviation Corp. v. NLRB,89 which recognizes that a union has long-term rights to enter into limited portions of the employer’s property in order to organize labor workers against the employer’s interest.90 The decisions in both the rent control and the union case did not accept any version of a takings argument, even though the private party was allowed to enter or stay in possession of property under government authorization. Neither of these cases accepted the need for just compensation under the Takings Clause, even though modern law treats the right to exclude as lying at the core of private property.91 But under Judge Wood’s view the law has to reverse itself on both counts, revealing a seemingly insurmountable logical contradiction. She cannot have it both ways: she cannot ignore the possessory rules in dealing with the validity of the NLRA and then treat it as decisive with respect to the state option to enact a RTW law. The account that she gives to require just compensation from Indiana under the RTW is the account that invalidates the NLRA at its inception. And once the NLRA is struck down on takings grounds, any issue of the constitutionality of the RTW law never arises.

To make matters even worse, there is a credible argument that neither the rent control laws nor the labor laws satisfy the public use prong of the Takings Clause. Both of these statutory frameworks are solely for the direct benefit of private parties. At no point do members of the public have the right to enter a rent controlled apartment or a union workplace. Hence these statutes should be suspect even under the narrow reading of public use found in Kelo v. City of New London,92 which says categorically: “it has long been accepted that the sovereign may not take the property of A for the sole purpose of transferring it to another private party B, even though A is paid just compensation.”93 That conclusion is doubly true in these two cases, since in the aggregate, both statutes, by reducing the size of the social pie, have negative effects on the public welfare as a whole. Under the conventional readings of Kelo, these arguments will of course be rejected out of hand, with indignation and anger, given that public use tends to be equated with any conceivable public purpose. After all, the highly relaxed rational basis test that governs the inquiry lets the legislature make its own assessment of public convenience, without any serious judicial oversight.94 But if the Public Use Clause bans takings for private use, then even under the current law, the NLRA falls as a whole, so that the question of the constitutionality of section 14(b) does not arise.

Judge Wood also writes as if the union was forced to donate its services to its nonmembers,95 but in so doing she overlooks the quid pro quo that the union receives. The first point is that the dissenting workers cannot turn down a gift, which is always possible in ordinary donative situations. She also undervalues the enormous leverage that the NLRA gives to unions, with or without Section 14(b), in the form of the exclusive bargaining rights for all workers against management that is now bound to bargain with the unions in good faith. At one point she chides the majority by denying that the union’s “seat at the bargaining table” somehow compensates it for the myriad real costs it incurs on behalf of nonmembers.96 She then observes two points. First, she insists that the union only claims its seat at the bargaining table by winning a representation election: “it does not win that seat either through the grace of the employer or in exchange for some kind of quid pro quo from either the employer or the bargaining-unit employees.”97 Second, she notes that the collective bargaining process also confers “such features as no-strike clauses, management rights clauses, and a grievance procedure, all of which are a win-win for both labor and management.”98

These arguments fail. Under her account, employers should be eager to welcome unions in order to obtain the benefits of a collective bargaining agreement. But this rosy account is false. The real quid pro quo given to the union is the list of unfair labor practices that forces management to bargain with the union at all. It is the statutory grant of the power to bargain, coupled with the employer’s duty to bargain in good faith, which offer unions enormous chips that they did not have at common law, where any employer could walk away from any union if it chose without having to divide its control of the firm through long, bitter, and potentially disruptive negotiations. These enormous and undeserved benefits remain with or without the RTW option at either the state or federal level, and should be duly considered as a counterweight in this context. It is foolish to say that the union receives no compensation when it retains these advantages, even if it cannot force dissident workers to support their mission under Section 14(b).

At this point, it becomes clear that the only way to salvage the NLRA is to revert to the highly deferential rational basis analysis that was invoked to sustain it against various constitutional challenges in the first place. When it comes to complex economic arrangements, the argument goes, Congress is boss. The oft-cited proposition from Usery v. Turner Elkhorn Mining Co.99 states: “It is by now well established that legislative Acts adjusting the burdens and benefits of economic life come to the Court with a presumption of constitutionality, and that the burden is on one complaining of a due process violation to establish that the legislature has acted in an arbitrary and irrational way.”100 That decision justified retroactive legislation by which one party was forced to pay into a compensation fund for harms for which it bore no responsibility. In effect, legislation required a donation from one group to another. If, however, the rational basis test is used to sustain the basic NLRA from constitutional attack, it also sustains the Indiana RTW law as part of the elaborate ad hoc compromise in management labor relationships. If Congress could retroactively impose taxes on particular parties, it is not bound to keep the original form of the NLRA for all time. Put otherwise, if the rest of the Taft-Hartley amendments were constitutional, so too is Section 14(b).

A similar analysis applies to the confused reasoning offered by Judge Foust in Machinists, which on many points tracks the arguments of Judge Wood on two key but erroneous propositions. First, that the DFR protects non-union members against discrimination;101 and second, that the union receives nothing of value in exchange for the services that it is required under law to supply to these workers.102 He also insists that services rendered are property protected by the takings law.103 His distinctive contribution is to analyze this case under the low-level rational basis test in Penn Central, only to strike it down on the grounds that the government cannot put forward any reason to justify the adoption of the RTW law. He thus looks to the famous three-part test in that case to ask about “(1) the economic impact of the regulation on the claimant, (2) its interference with distinct investment backed expectations, and (3) the character of the governmental action.”104 He then recites the loss of dues to the union that follows from the adoption of the RTW and assumes it creates the prima facie case. Then he makes the further case, without argument or evidence, that the adoption of the RTW “is threatening to the unions’ very economic viability,”105 while failing to explain why unions have continued to operate unimpeded for years in RTW states.

He next pushes the extreme claim that the plaintiffs have the “distinct, investment-backed expectation that they would always have a right to collect fair-share payments from non-members as long as they were compelled to provide them [with] services.”106 It is also worth asking how unions possibly have formed any expectation that they could receive dues from all workers when RTW laws have been on the books long before the Sweeney litigation. But in all that time no one had ever advanced the fair-share argument. Nor is it explained how unions can have formed that view in the teeth of a statute that gives states the explicit power to pass RTW laws that blocked the agency shop. Under current law, even in those cases where strong contractual rights were at issue, it has been repeatedly held that the government can strip individuals of their rights without offering any compensation at all. In Pension Benefit Guaranty Corp. v. R.A. Gray & Co.,107 the Court rejected a challenge under the Due Process Clause in a unanimous decision that placed extensive reliance on the Court’s earlier decision in Turner Elkhorn. In Connolly v. Pension Benefit Guaranty Corp.,108 the Supreme Court again unanimously relied on Penn Central and Turner Elkhorn to undermine vested rights when it said: “This interference with the property rights of an employer arises from a public program that adjusts the benefits and burdens of economic life to promote the common good and, under our cases, does not constitute a taking requiring Government compensation.”109 The Court then concluded that the applicable legislation did not interfere with reasonable investment-backed expectations: “Prudent employers then had more than sufficient notice not only that pension plans were currently regulated, but also that withdrawal itself might trigger additional financial obligations.”110 It is worth saying that the claimant’s case in both Gray and Connolly was far stronger than any claim that might be raised here, because the private parties relied on explicit contractual guarantees that were nowhere to be found in the RTW law. The decision of Judge Foust in Mechanists is so far off the reservation as to defy belief.

Conclusion

The RTW laws are a long-accepted feature of American labor law. In this essay, I have argued that they are an acceptable response to the federal system of collective bargaining, and, in fact, improve overall social welfare by cutting down on the monopoly power that unions can acquire under the NLRA. It is also the case that their constitutional status was widely accepted on all sides of the debate until the recent flurry of activity which has led to the adoption of RTW laws in the Midwest. There is little doubt that these laws further compromised the already-troubled status of organized labor leading unions to bring forth a set of initiatives, including litigation, to strengthen their hands. In dealing with these issues, the current legal system adopts the low-level of scrutiny associated with Turner Elkhorn and Penn Central to justify a wide set of privileges to union power that were never accepted anywhere at common law. Long before the current controversies developed, I have passionately insisted that the rejection of the common law rules has set in motion a train of events that have wreaked genuine dislocation on labor markets.111 In another time and another place, I would be happy to defend this view against all comers. But for these purposes it is only necessary to take a more modest position in defense of the status quo ante, which has now come under attack. The simple proposition is Section 14(b) of the NLRA means what it has always been understood to mean: States are free to pass RTW laws that remove any and all union security arrangements, including the agency shop, and this position does not offend the Constitution by taking private property from the union, including their union services, without just compensation. To my mind, these propositions are so utterly conventional that it is hard to see why or how they could be rejected by anyone on any side of the political spectrum. But once those arguments have been advanced in a forceful way, it becomes incumbent to say why they are against the statutory text, the legislative history, the uniform practice, and the constitutional principles that govern labor relations.

• 1767 F.3d 654 (7th Cir. 2014).
• 2No. 2015CV000628, 2016 WL 8504732 (Wis. Cir. Ct. Apr. 8, 2016) (Machinists).
• 3Sweeney, 767 F.3d at 680. For earlier cases on preemption, see Local 174, Teamsters v. Lucas Flour Co., 369 U.S. 95 (1962); San Diego Bldg. Trades Council v. Garmon, 359 U.S. 236 (1959).
• 4No. 2016AP820, 2017 WL 4158729 (Wis. App. Ct. Sept. 19, 2017).
• 5For the tally, see Right-to-Work States, Wikipedia, https://en.wikipedia.org/wiki/Right-to-work_law#/media/File:Right_to_Work_states.svg, https://perma.cc/VXW3-MLQS.
• 6Right to Work, AFL-CIO, https://aflcio.org/issues/right-work, https://perma.cc/7FVC-23YW.
• 7Right to Work States, Nat’l Right to Work Legal Def. Found., http://www.nrtw.org/right-to-work-states/, https://perma.cc/Q2JK-GWY4.
• 829 U.S.C. §§ 151–69 (2012). Section 157, as modified under the Taft-Hartley Act, touches all the bases when it allows for workers to join collectively or to opt of unions, except insofar as there are union security clauses. 29 U.S.C. § 157.
• 9For a straightforward account, see Robert A. Gorman & Matthew W. Finkin, Basic Text on Labor Law: Unionization and Collective Bargaining 897–902 (2d ed. 2004), written before the current controversies arose.
• 1029 U.S.C. § 157.
• 1129 U.S.C. § 8(a)(3) (1947), amended by Labor Management Relations (Taft-Hartley) Act of 1947, 29 USC §§ 141–97 (2012).
• 12See Gorman & Finkin, supra note 8, at 900.
• 13See id.
• 14Commc’ns Workers of Am. v. Beck, 487 U.S. 735, 762–63 (1988) (quoting Ellis v. Railway Clerks, 466 U.S. 435, 448 (1984)) (The current rule covers “only 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.’”).
• 1529 U.S.C. § 164(b).
• 16Gorman & Finkin, supra note 8, at 900 (emphasis added).
• 17Id.
• 18767 F.3d 654 (7th Cir. 2014).
• 19Machinists, No. 2015CV000628, 2016 WL 8504732 (Wis. Cir. Ct. Apr. 8, 2016).
• 20U.S. Const. V.
• 21Loretto v. Teleprompter Manhattan CATV Corp., 458 U.S. 419 (1982).
• 22Penn Central Transp. Co. v. City of New York, 438 U.S. 104 (1978).
• 23Lehnert v. Ferris Faculty Ass’n, 500 U.S. 507, 556 (1991).
• 24500 U.S. 507 (1991).
• 25Machinists, No. 2015CV000628, 2016 WL 8504732, at *6 (Wis. Cir. Ct. Apr. 8, 2016) (citing with approval Scalia’s free-rider argument in Lehnert).
• 26Sweeney v. Pence, 767 F.3d 654, 673 (7th Cir. 2014) (Wood, J., dissenting).
• 27Vaca v. Sipes, 386 U.S. 171, 177 (1967) (cited in Machinists, 2016 WL 8504732, at *4; Sweeney, 767 F.3d at 672).
• 28For critique of the argument, generally see the discussion of Vaca infra, Section I; see also Richard A. Epstein, Individual Control over Personal Grievances Under Vaca v. Sipes, 77 Yale L.J. 559 (1968).
• 29Gorman & Finkin, supra note 8, at 900–01.
• 30386 U.S. 171 (1967).
• 31Epstein, supra note 27.
• 32Vaca, 386 U.S. at 174–76.
• 33Id. at 195.
• 34Id. at 182 (citing J.I. Case Co. v. NLRB, 312 U.S. 332 (1944)).
• 3529 U.S.C. § 159.
• 3629 U.S.C. § 159(a) (“Exclusive representatives; employees’ adjustment of grievances directly with employer—Representatives designated or selected for the purposes of collective bargaining by the majority of the employees in a unit appropriate for such purposes, shall be the exclusive representatives of all the employees in such unit for the purposes of collective bargaining in respect to rates of pay, wages, hours of employment, or other conditions of employment: Provided, That any individual employee or a group of employees shall have the right at any time to present grievances to their employer and to have such grievances adjusted, without the intervention of the bargaining representative, as long as the adjustment is not inconsistent with the terms of a collective-bargaining contract or agreement then in effect: Provided further, That the bargaining representative has been given opportunity to be present at such adjustment.”).
• 37Id.
• 38323 U.S. 192 (1944).
• 3945 U.S.C. §§151–65 (2012).
• 40Steele v. Louisville & Nashville Railroad Co., 323 U.S. 192, 195–96 (1944).
• 41355 U.S. 41 (1957).
• 42375 U.S. 335 (1964).
• 43Id. at 348–50 (1964).
• 44For an early case, see Otis & Co. v. Pennsylvania R. Co., 61 F. Supp. 905, 911 (E.D. Pa. 1945) (“[M]istakes or errors in the exercise of honest business judgment do not subject the officers and directors to liability for negligence in the discharge of their appointed duties.”).
• 45See Richard A. Epstein, Redistribution Within Collective Organizations: What Corporations, Condominiums and Unions Tell Us about the Proper Use of Government Power, 8 N.Y.U. J.L. & Liberty 280, 308–10 (2014), for relevant discussion.
• 46Epstein, supra note 44, at 284–85, 287, 297.
• 47See generally, Epstein, supra note 44, § I.
• 48Sweeney v. Pence, 767 F.3d 654, 672 (7th Cir. 2014) (Wood, J., dissenting).
• 49Id.
• 50Id. at 671–72.
• 51For this model, see Mancur Olson, The Logic of Collective Action (1965).
• 52208 U.S. 274 (1908).
• 53Clayton Antitrust Act of 1914, 15 U.S.C. § 17 (2012) (“The labor of a human being is not a commodity or article of commerce. Nothing contained in the antitrust laws shall be construed to forbid the existence and operation of labor, agricultural, or horticultural organizations, instituted for the purposes of mutual help, and not having capital stock or conducted for profit, or to forbid or restrain individual members of such organizations from lawfully carrying out the legitimate objects thereof; nor shall such organizations, or the members thereof, be held or construed to be illegal combinations or conspiracies in restraint of trade, under the antitrust laws.”).
• 54136 S.Ct. 1083 (2016).
• 55Brief for the Buckeye Institute for Public Policy Solutions as Amicus Curiae in Support of Petitioners, Friedrichs v. CA Teachers Assoc., 136 S.Ct. 1083 (2016) (No. 14-915), 2015 WL 5317008; Brief for the Mackinac Center for Public Policy as Amicus Curiae in Support of Petitioners, Friedrichs v. CA Teachers Assoc., 136 S.Ct. 1083 (2016) (No. 14-915), 2015 WL 5461532.
• 56Sweeney v. Pence, 767 F.3d 654, 660 (7th Cir. 2014) (Wood, J., dissenting).
• 57373 U.S. 746 (1963) (Retail Clerks I).
• 58NLRB v. General Motors, 373 U.S. 743 (1963).
• 59Retail Clerks I, 373 U.S. at 751–52.
• 60375 U.S. 96 (1963) (Retail Clerks II).
• 61Id. at 102–03.
• 62Sweeney, 766 F.3d at 676.
• 63Retail Clerks I, 373 U.S. at 751–52.
• 64675 F.2d 1257 (D.C. Cir. 1982).
• 65Int’l Union v. NLRB, 675 F.2d 1257, 1261–62 (citing Oil, Chem. & Atomic Workers Int’l Union v. Mobil Oil Corp., 426 U.S. 407, 417 (1976)).
• 66Sweeney, 767 F.3d at 675 (citing NLRB v. General Motors, 373 U.S. 734, 742 (1963)).
• 67For a relatively long discussion of the point, see Richard A. Epstein, The Classical Liberal Constitution: The Uncertain Quest for Limited Government Ch. 3 (2014).
• 68349 F.2d 449 (5th Cir. 1965).
• 69A hiring hall is organized by unions to place their members in different jobs. These are needed in industries like construction and stevedoring where work is often done under a daily and short-term basis, so that no given worker is paired with a single employer throughout the term of the master agreement that usually covers multiple employers. See, e.g., NLRB Hiring Halls, Nat’l Lab. Rel. Board, https://www.nlrb.gov/rights-we-protect/whats-law/employees/i-am-represented-union/hiring-halls, https://perma.cc/Z3A2-64V8.
• 70NLRB. v. Houston Chap. Ass’n Gen’l Con., 349 F.2d 449, 453 (5th Cir. 1965), cert. denied, 382 U.S. 1026 (1966) (citing NLRB. v. General Motors Corp., 373 U.S. 734 (1963); Retail Clerks I, 373 U.S. 746 (1963)).
• 71See Gorman & Finkin, supra note 8, and accompanying text.
• 72Sweeney v. Pence, 767 F.3d 654, 671–85 (7th Cir. 2014) (Wood, J., dissenting); Machinists, No. 2015CV000628, 2016 WL 8504732 (Wis. Cir. Ct. Apr. 8, 2016).
• 73Heather M. Whitney, Friedrichs: An Unexpected Tool for Labor, 10 N.Y.U. J.L. & Liberty 191, 194 (2016) (emphasis added). Ms. Whitney was a clerk for Judge Wood.
• 74Richard A. Epstein, Takings: Private Property and the Power of Eminent Domain (1985).
• 75William Michael Treanor, The Original Understanding of the Takings Clause and the Political Process, 95 Colum. L. Rev. 782, 840 (1995).
• 76Epstein, supra note 73, at 35–106.
• 77See Epstein, supra note 73, at 107–60, 195.
• 78438 U.S. 104 (1978).
• 79Penn Central Transp. Co. v. City of New York, 438 U.S. 104, 136 (1978).
• 80458 U.S. 419 (1982); see also San Diego Gas & Elec. Co. v. City of San Diego, 450 U.S. 621, 657 (1981) (“[T]he fact that a regulatory ‘taking’ may be temporary, by virtue of the government’s power to rescind or amend the regulation, does not make it any less of a constitutional ‘taking.’ Nothing in the Just Compensation Clause suggests that ‘takings’ must be permanent and irrevocable. Nor does the temporary reversible quality of a regulatory ‘taking’ render compensation for the time of the ‘taking’ any less obligatory.”). For a more recent attempt to resuscitate the distinction, see Horne v. Department of Agriculture, 135 S. Ct. 2419 (2015); Richard A. Epstein, The Unfinished Business of Horne v. Department of Agriculture, 10 N.Y.U. J. L. & Lib. 734 (2017).
• 81Epstein, supra note 73, at 280–81.
• 82208 U.S. 161 (1908), overruled by Phelps Dodge Corp. v. NLRB, 313 U.S. 177 (1941).
• 83236 U.S. 1 (1915) (invalidating the Kansas law), overruled by Phelps, 313 U.S. 177 (1941).
• 84See, e.g., Block v. Hirsh, 256 U.S. 135 (1921) (validating two-year rent control statute).
• 85458 U.S. 419 (1982).
• 86Id. at 426.
• 87Phillips v. Wash. Legal Found., 524 U.S. 156 (1998).
• 88Sweeney v. Pence, 767 F.3d 654, 674 (7th Cir. 2014) (Wood, J., dissenting), discussing Phillips v. Wash. Legal Found., 524 U.S. 156 (1998), amongst other authorities.
• 89324 U.S. 793 (1945).
• 90Republic Aviation Corp. v. NLRB, 324 U.S. 793, 803–05, n. 10 (1945) (citing 49 N.L.R.B. at 843–44).
• 91See, e.g., Kaiser Aetna v. United States, 444 U.S. 164, 179–80 (1979) (accepting the right to exclude as a private property right: “we hold that the ‘right to exclude,’ so universally held to be a fundamental element of the property right . . .”).
• 92545 U.S. 469 (2005).
• 93Id. at 477. Note if indirect benefits are counted, then this rule is vapid.
• 94See Berman v. Parker, 348 U.S. 26, 33 (1954) (“We do not sit to determine whether a particular housing project is or is not desirable. The concept of the public welfare is broad and inclusive.”); see Day-Brite Lighting, Inc. v. Missouri, 342 U.S. 421, 424 (1952).
• 95Sweeney v. Pence, 767 F.3d 654, 674 (7th Cir. 2014).
• 96Id. at 684 (quoting majority opinion at 667).
• 97Id.
• 98Id.
• 99428 U.S. 1 (1976).
• 100Id. at 15 (citing Ferguson v. Skrupa, 372 U.S. 726 (1963); then citing Williamson v. Lee Optical Co., 348 U.S. 483, 487–88 (1955)).
• 101Machinists, No. 2015CV000628, 2016 WL 8504732, at *4 (Wis. Cir. Ct. Apr. 8, 2016).
• 102Id. at *4.
• 103Id. at *8–11.
• 104Id. at *11 (citing Penn Central Transp. Co. v. City of New York., 438 U.S. 104, 124 (1978).
• 105Id. at *13.
• 106Id.
• 107467 U.S. 717 (1984).
• 108475 U.S. 211 (1986).
• 109Connolly v. Pension Benefit Guaranty Corp., 475 U.S. 221, 225 (1986) (citing Penn Central Transp. Co. v. City of New York., 438 U.S. 104, 124 (1978); then citing Usery v. Turner Elkhorn Mining Co., 428 U.S. 1, 15–16 (1976)).
• 110Id. at 227; see Pension Benefit Guaranty Corp. v. R.A. Gray & Co., 467 U.S. 717, 732 (1984); see also Fed. Housing Admin. v. Darlington, Inc., 358 U.S. 84, 91 (1958) (“Those who do business in the regulated field cannot object if the legislative scheme is buttressed by subsequent amendments to achieve the legislative end.”).
• 111Richard A. Epstein, A Common Law for Labor Relations: A Critique of the New Deal Legislation, 92 Yale L.J. 1357 (1983); Richard A. Epstein, Common Law, Labor Law, and Reality: A Rejoinder to Professors Getman and Kohler, 92 Yale L.J. 1435 (1983); Richard A. Epstein, Labor Unions: Saviors or Scourges?, 41 Cap. U. L. Rev. 1 (2013).