This policy brief discusses several foundational legal concepts and sets forth model language for a legally sound right-to-work amendment to the Michigan Constitution.
Twenty-two states have statutes or state constitutional amendments that are meant to ensure that joining or paying fees to a labor union is not a requirement of lawful employment. Such provisions are known as "right-to-work" laws, and if Michigan chooses to adopt one, the state should enact language that is likely to achieve the provision’s purposes while minimizing the opportunities for legal challenges that seek to narrow or nullify the law.
The right-to-work language presented here — see Page 15 or the back cover — is written as a model state constitutional amendment, not as a statute. One value of a right-to-work provision is the message it sends about the state’s business climate, and given Michigan’s labor history, this message would be stronger if it were placed in the state constitution. Further, work is central to human existence, and key rights related to this basic concern are proper subjects of the state’s social contract.
In preparing model language for a right-to-work amendment in Michigan, the case law from all 22 right-to-work states was reviewed. This review suggested a number of elements that should be part of a right-to-work law in Michigan.
A clause preventing discrimination against nonunion workers in employment is fundamental to the nature of right-to-work laws, but a clause preventing discrimination against union members is also recommended here. This union nondiscrimination clause would ensure both that the state does not run afoul of the U.S. Constitution’s "equal protection" clause and that the amendment provides basic fairness given the requirements of federal labor law. This union nondiscrimination clause could conceivably be pre-empted by federal law, thereby providing some opportunity for a legal challenge to the amendment, but other clauses in the model language — specifically a clause regarding "severability" — would discourage such a challenge.
Some state courts have ruled that public employment must be explicitly included in a right-to-work law in order for right-to-work protection to be extended to government workers. This amendment therefore specifically includes public-sector employment.
In the absence of a right-to-work provision explicitly banning the payment of any union fees as a condition of employment, one state court held that in filing a grievance, a nonunion worker would have to pay a significant portion of the union’s attorney fees and the costs related to the hearing. The model language includes a clause expressly prohibiting payment of any fees or expenses to a union as a condition of employment.
The most recent pertinent challenge to a state right-to-work law contended that enough of the law was pre-empted by existing federal labor statutes that the law should be nullified in its entirety. The model language safeguards against such a challenge by explicitly stating that any portion of the amendment not pre-empted remains in effect — i.e., that the amendment is "severable."
There has been recent public discussion about reviving Michigan’s economy by having the state adopt “right-to-work” provisions — laws that prevent joining or paying fees to a union from becoming a requirement of lawful employment.* In July 2007, Michigan Senate Majority Leader Mike Bishop indicated that he would support a right-to-work ballot initiative to amend the Michigan Constitution.[1] Right-to-work bills have been submitted in both the Michigan House and Senate.† In April 2007, Peter Karmanos Jr., the CEO of Detroit’s Compuware Corp., publicly suggested that Michigan should consider becoming a right-to-work state, saying, “The state unions are very powerful, the (teachers union) is very powerful and they are, as far as I’m concerned, fiscally irresponsible.”[2]
There are currently 22 states with right-to-work laws,** and many of these laws have been subject to union legal challenges. The subsequent legal record indicates that some right-to-work provisions are more likely to survive these challenges than others are. This policy brief therefore discusses several foundational legal concepts and sets forth model language for a legally sound right-to-work amendment to the Michigan Constitution.
* For a discussion of the economics of right-to-work laws, see William Wilson, “The Effect of Right-to-Work Laws on Economic Development” (Mackinac Center for Public Policy, 2002), and Paul Kersey, “The Economic Effects of Right-to-Work Laws: 2007” (Mackinac Center for Public Policy, 2007).
† House Bill 4454 and Senate Bill 607, which are identical substantively.
** The 22 states are (1) Alabama – Ala. Code §§ 25-7-30 to 36; (2) Arizona – Ariz. Const. art. XXV; Ariz. Rev. Stat. §§ 23-1301 to 1307; (3) Arkansas – Ark. Const. amend. 34 §§ 1-2; Ark. Code Ann. §§ 11-301 to 304; (4) Florida – Fla. Const. art. I, § 6; Fla. Stat. § 447.17; (5) Georgia – Ga. Code Ann. §§ 34-6-20 to 28; (6) Idaho – Idaho Code Ann. §§ 44-2001 to 2009; (7) Iowa – Iowa Code §§731.1-8; (8) Kansas – Kan. Const. art. 15, § 12; Kan Stat. Ann. § 44-831; (9) Louisiana – La. Rev. Stat. Ann. §§ 23:981-87; (10) Mississippi – Miss. Const. art. 7 § 198-A; Miss. Code. Ann. § 71-1-47; (11) Nebraska – Neb. Const. art. XV, §§ 13-15; Neb. Rev. Stat. §§ 48-217 to 219; (12) Nevada – Nev. Rev. Stat. §§ 613.230, 613.250 to 300; (13) North Carolina – N.C. Gen. Stat. §§ 95-78 to 84; (14) North Dakota – N.D. Cent. Code §§ 34.01.14 to 14.1; (15) Oklahoma – Okla. Const. art. XXIII, § 1A; (16) South Carolina – S.C. Code. Ann. §§ 41-7-10 to 90; (17) South Dakota – S.D. Const. art. VI, § 2; S.D. Codified Laws § 60-8-3 to 8; (18) Tennessee – Tenn. Code Ann. §§ 50-1-201 to 204; (19) Texas – Tex. Lab. Code Ann. §§ 101.003 to 004, 101.051 to 053, 101.101 to 102, 101.111, 101.121 to 124; (20) Utah – Utah Code Ann. §§ 34-34-1 to 17; (21) Virginia – Va. Code Ann. §§ 40.1-58 to 69; and (22) Wyoming – Wyo. Stat. Ann. §§ 22-7-108 to 115.
To understand key issues related to a right-to-work amendment, it is necessary to have a general understanding of labor policy, labor law and the manner in which state and federal laws interact. These topics are discussed in the two subsections that follow.*
* Those seeking a comprehensive legal understanding of the relationship between compulsory unionism, mandatory union fees and right-to-work law in Michigan should consult Robert P. Hunter, “Compulsory Union Dues in Michigan” (Mackinac Center for Public Policy, 1997); and Robert P. Hunter, “Michigan Labor Law: What Every Citizen Should Know” (Mackinac Center for Public Policy, 1999). Readers can also consult Thomas W. Washburne and Paul Kersey, “Right-to-Work FAQ,” Mackinac Center for Public Policy, (accessed August 12, 2007).
To understand the nature and effect of a right-to-work amendment, labor relations can be viewed from three perspectives. First is a free-market approach, which maximizes government neutrality. As Robert Hunter, a former member of both the National Labor Relations Board and the Michigan Civil Service Commission, puts it:
"The free-market (or government-neutral) approach requires that the government neither encourage nor discourage the formation of labor unions. Workers who choose to form a union are free to do so. Government does not prohibit union membership or union activity, provided existing laws against fraud, violence, and property damage are not violated. Individual workers may join or not join a union, and union leaders must earn each worker’s voluntary support by providing desired benefits. Under this approach, employers may choose to deal or not deal with the labor union and workers are free to strike. ..."[3] (Emphasis in original.)
A competing vision is "compulsory unionism," which uses governmental power to ensure that unions are given a monopoly status as the bargaining representative for a group of workers whenever a majority of those workers have voted to allow the union to perform this task.* These workers, who are typically referred to as a "collective bargaining unit," are a set of employees who have a common employer and are held to have similar labor interests in relation to that employer. As Hunter describes compulsory unionism:
"[In this approach, the] government plays an active role in encouraging labor unions. The government forces employers to recognize labor unions and negotiate with them in a process called ‘mandatory collective bargaining.’ Unions are recognized by law as ‘exclusive bargaining representatives’ who may prohibit individual workers in their bargaining units from negotiating individual working arrangements with their employer, even if they would be better off doing so and their employer is willing. ...
"Likewise, the privilege of exclusive representation prevents workers from being represented by a union other than the one approved by a majority of their co-workers. Everyone in the bargaining unit is in turn required to work under the terms of any contract the union approves."[4]
A labor contract that allows a union to compel fees from all workers within the union’s collective bargaining unit (even if some of those workers have not actually joined the union) is generally called an "agency-shop agreement."[5] An agreement whereby an employee must join a union and pay dues or fees within a certain time after being hired is called a "union-shop agreement."
In cases of compulsory unionism, labor contracts typically include "union secu-rity clauses," which require employers to fire workers who fail to support the union either by becoming a union member or by paying union dues or fees.
A third, intermediate approach to labor policy involves right-to-work laws, which are a means of lessening the impact of compulsory unionism. Generally, right-to-work laws prevent an employee from being hired or fired for being a union member or a nonmember, and they prohibit unions from charging fees to nonmembers in a collective bargaining unit. In effect, right-to-work laws nullify union security clauses, meaning that a worker no longer has to join a union or pay union bargaining fees as a condition of employment.
Note that right-to-work laws do not end a union’s status as the government-mandated exclusive bargaining agent for a collective bargaining unit. Individual workers in the collective bargaining unit cannot negotiate independently with their employer or bargain through a second union. Right-to-work laws therefore do not produce a free-market labor environment. Nevertheless, given the current state of federal labor law, right-to-work laws provide workers with a degree of freedom despite the monopoly bargaining power the unions retain.
* Unions are generally empowered to represent a bargaining unit by a majority vote of the unit’s workers. Once created, however, the union often enjoys a presumption of continued majority support. Thus, in some longstanding bargaining units, many employees may not have had an opportunity to explicitly accept or reject union representation.
In regard to labor-management relations, the federal government has generally adopted a compulsory unionism policy. This approach is prescribed for most private-sector employers through the National Labor Relations Act,[6] though there are other federal laws that involve labor relations and allow for compulsory collective bargaining.*
The manner in which federal labor laws interact with state laws is important to understanding the role of right-to-work laws. The "supremacy clause"[7] of the U.S. Constitution means that federal laws supersede any state law that might conflict with them. The U.S. Supreme Court has ruled that federal legislation, through a pre-emption doctrine based on the supremacy clause, may render a state law inoperative:
"A fundamental principle of the Constitution is that Congress has the power to preempt state law. Even without an express provision for preemption, we have found that state law must yield to a congressional Act in at least two circumstances. When Congress intends federal law to ‘occupy the field,’ state law in that area is preempted. And even if Congress has not occupied the field, state law is naturally preempted to the extent of any conflict with a federal statute. We will find preemption where it is impossible for a private party to comply with both state and federal law and where ‘under the circumstances of [a] particular case, [the challenged state law] stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress.’"[8]
A federal district court in Oklahoma has also determined that where Congress is silent, "Rules of statutory construction dictate that the court should find the federal government exempt from the application of state statutes which attempt to regulate any collective bargaining agreement to which the federal government is a party."[9]
Further, the U.S. Constitution’s "commerce clause" gives Congress the power to prevent state right-to-work laws from becoming legally effective. For example, in Ry. Employes’ Dep’t. v. Hanson,[10] the Supreme Court in 1956 held that a railroad union could enforce a security clause related to a union-shop agreement.† While recognizing, "Powerful arguments have been made … that the longrun [sic] interests of labor would be better served by the development of democratic traditions in trade unionism without the coercive element of the union … shop,"[11] the court held that the Railway Labor Act, which explicitly allowed union shops,[12] trumped the right-to-work amendment in the Nebraska Constitution.
Although this pre-emptive power exists, Congress has not used it in every instance. Importantly, section 14(b) of the NLRA explicitly allows state right-to-work laws:
"Nothing in this subchapter 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 or application is prohibited by State or Territorial law."[13]
Because state right-to-work laws are permitted under the NLRA, the overwhelming majority of employees within a state will receive the benefits of right-to-work provisions, even though the coverage of some employees will be pre-empted by other federal labor statutes. In effect, state right-to-work laws establish about as much labor-relations freedom as is possible under current federal labor law.**
* For example, The Railway Labor Act governs labor relations in the railroad and airline industries. (See 45 U.S.C. § 151 et. seq.) The Civil Service Reform Act applies to most federal agency employees. (See 5 U.S.C. § 7101 et. seq.) The Postal Reorganization Act concerns postal employees. (See 39 U.S.C. § 1203.)
† The Railway Labor Act prohibited union-shop agreements from 1934 until 1951. (See Ry. Employes’ Dep’t, 231.)
** Note that a state law or constitutional amendment that attempted to establish a statewide free market in labor relations policy would conflict with the NLRA’s provision establishing private-sector unions’ monopoly bargaining power and would be overturned in court. In fact, such a law would be pre-empted not only by the NLRA, but also by other federal labor laws, such as the Railway Labor Act.
The model language presented later in this study is based on a review of the case law for right-to-work provisions in the 22 states that have such laws. The discussion that follows focuses on the right-to-work litigation surrounding Oklahoma’s state constitutional amendment, because the Oklahoma case is the most recent and instructive of the legal challenges. Nevertheless, pertinent litigation from Arizona, Nebraska, Nevada, North Carolina, South Carolina and Tennessee will be discussed as well.
In September 2001, the voters of Oklahoma passed a right-to-work amendment to their state constitution by a margin of 54 percent to 46 percent,[14] making Oklahoma the most recent state to pass a right-to-work amendment. Almost immediately after passage, several unions filed a federal lawsuit seeking to prevent the amendment’s implementation. Since a similar challenge would likely be filed in Michigan if a right-to-work amendment were ratified, the legal arguments presented in Oklahoma are discussed below. The Oklahoma amendment is reproduced in Graphic 1.
Graphic 1: Oklahoma’s Right-to-Work Amendment (Oklahoma Constitution Article XXIII, Section 1A) § 1A. Participation in labor organization as condition of employment prohibited. A. As used in this section, “labor organization” means any organization of any kind, or agency or employee representation committee or union, that exists for the purpose, in whole or in part, of dealing with employers concerning wages, rates of pay, hours of work, other conditions of employment, or other forms of compensation. B. No person shall be required, as a condition of employment or continuation of employment, to:
C. It shall be unlawful to deduct from the wages, earnings, or compensation of an employee any union dues, fees, assessments, or other charges to be held for, transferred to, or paid over to a labor organization unless the employee has first authorized such deduction. D. The provisions of this section shall apply to all employment contracts entered into after the effective date of this section and shall apply to any renewal or extension of any existing contract. E. Any person who directly or indirectly violates any provision of this section shall be guilty of a misdemeanor. |
The unions’ basic argument against the amendment was that as written, the amendment would be pre-empted in so many places by federal law that the entire constitutional provision would be invalid because it was not "severable"[15] (a "severable" provision is one whose valid or unchallenged parts remain in force even when a court invalidates other portions of it). In particular, the unions began by contending that the amendment applied to employees that were "governed by the RLA [Railway Labor Act], the CSRA [Civil Service Reform Act], the PRA [Postal Reorganization Act], as well as those individuals employed at facilities within federal enclaves[*] in the State of Oklahoma."[16] The unions claimed that none of the employees subject to those federal laws was protected by the right-to-work amendment, and they concluded that this pre-emption so radically diminished the scope of the right-to-work amendment that the entire amendment should be struck because it was nonseverable.
The trial court rejected the foundation of the argument, holding that even though Oklahoma’s right-to-work amendment did not contain an explicit provision exempting employees pre-empted by federal law, the amendment did not seek to challenge federal authority:
"... [I]t is simply not a reasonable construction to extend the scope of Oklahoma’s right-to-work law to include those individuals subjected to regulation under the RLA [Railway Labor Act], the CSRA [Civil Service Reform Act], the PRA [Postal Reorganization Act], and federal enclave jurisprudence. Consequently, the court interprets Oklahoma’s right-to-work law as excluding from its coverage those individuals subject to the RLA, the CSRA, the PRA, as well as those individuals subject to federal enclave jurisdiction. From this conclusion, it follows that the preemption suggested by Plaintiffs with respect to these individuals has no application to any portion of Oklahoma’s right-to-work law."[17]
The unions’ efforts to invalidate large portions of the amendment — and thereby the entire amendment — even led union lawyers to do something counterintuitive: They challenged the amendment provision that stated no person should be required to quit a union, refuse to join a union, or refuse to support a union in order to become employed or remain employed. This argument was also rejected by the court, which stated:
"Plaintiffs argue that under [Section 14(b),] states only have the authority to prevent employers from requiring membership in labor organizations as a condition of employment — they do not have authority to prevent employers from prohibiting union membership as a condition of employment. The court disagrees. The United States Supreme Court has upheld state right-to-work laws which prohibit discrimination in employment based on both union membership and non-membership alike. [Lincoln Federal Labor Union 19129 v. Northwestern Iron & Metal Co., 335 U.S. 525 (1949)]."[18]
But the trial court did find some provisions of the right-to-work law were pre-empted by federal law, specifically a provision outlawing "hiring halls" and a paycheck-protection provision.†[19]
Having found a portion of the right-to-work amendment to be ineffective, the trial court then performed a severability analysis, which was guided by an Oklahoma statute that generally indicated that Oklahoma legislative acts were severable.[20] The court therefore held that the pre-empted provisions were severable:
"With respect to Oklahoma’s right-to-work law, it is clear that the overriding purpose of the law was to ensure that employment was not conditioned upon one’s membership in, voluntary affiliation with, or financial support of a labor organization or on a refusal to join, affiliate, or financially support a labor organization. Enforcement of the core provisions of the law which carry out this undeniable purpose is in no way hindered by the court’s invalidation of the subsidiary provisions of subsections (B)(5) [the hiring-hall provision] and (C) [the paycheck-protection provision]. Consequently, the invalid provisions of [the amendment] are severable from the core provisions and the remainder of Oklahoma’s right-to-work law is upheld."[21]
The unions appealed to the 10th U.S. Circuit Court of Appeals. That court found another provision of the right-to-work amendment was pre-empted: the
nondiscrimination provision related to those who wished to join or support a union.[22] To properly analyze the 10th Circuit’s conclusion regarding this provision, we first need to review two seminal Supreme Court decisions that upheld right-to-work laws: Lincoln Federal Labor Union No. 19129, American Federation of Labor v. Northwestern Iron & Metal Co.;[23] and American Federation of Labor, Arizona State Federation of Labor v. American Sash & Door Co.[24]
* Very generally, federal enclaves are federally owned land obtained with the consent of the state in which the land is located. The question of whether state right-to-work laws apply within a federal enclave depends on whether the federal government has obtained exclusive jurisdiction over the enclave. Where jurisdiction in an enclave is concurrent between the state and federal government, state right-to-work laws are effective. (See Int’l Ass’n of Machinists and Aerospace Workers, AFL-CIO and Local Lodge 2771 v. Dyncorp, Aerospace Operations, Sheppard ENJJPT Div., 796 F. Supp. 976 (N.D. Tex., 1991).)
† For the hiring-hall provision, see Okla. Const. art. XXIII, § 1A (B)(5); for the paycheck- protection provision, see Okla. Const. art. XXIII, § 1A (C).
In H.A. Artists & Assoc. v. Actor’s Equity Ass’n, the purpose of union hiring halls was described by the U.S. Supreme Court as follows: “[U]nions maintain hiring halls and other job referral systems, particularly where work is typically temporary and performed on separate project sites rather than fixed locations. By maintaining halls, unions attempt to eliminate abuses such as kickbacks, and to insure fairness and regularity in the system of access to employment.” (See H.A. Artists & Assoc. v. Actor’s Equity Ass’n, 451 U.S. 704, 721 n. 28 (2001).)
In a separate case, the Supreme Court held that the National Labor Relations Board was not allowed to prohibit hiring halls. The court reached this conclusion despite recognizing the potential for hiring-hall discrimination against nonunion workers. (See Local 357, Int’l Bhd. of Teamsters, Chauffeurs, Warehousemen and Helpers of America v. NLRB, 365 U.S. 667, 672 (1961).)
In Lincoln Federal, the U.S. Supreme Court considered challenges to North Carolina’s right-to-work law and Nebraska’s right-to-work amendment. As the court described these right-to-work provisions, "A North Carolina statute and a Nebraska constitutional amendment provide that no person in those states shall be denied an opportunity to obtain or retain employment because he is or is not a member of a labor organization."[25] The provisions were challenged, "[O]n the ground that insofar as they attempt to protect non-union members from discrimination, the laws are in violation of rights guaranteed employers, unions, and their members by the United States Constitution."[26]
The unions’ most relevant constitutional claim for our purposes was that these provisions violated the 14th Amendment’s "equal protection" clause,* which basically requires state laws to treat people equally. The Supreme Court rejected the unions’ argument: "[I]n identical language these state laws forbid employers to discriminate against union and non-union members. Nebraska and North Carolina thus command equal employment opportunities for both groups of workers."[27] Thus, right-to-work laws containing both a union- and nonunion-nondiscrimination clause clearly do not violate the federal equal protection clause.
In the second seminal case, American Sash, the Supreme Court stated, "A difference between the Arizona amendment and the amendment and statute considered in the Nebraska and North Carolina cases has made it necessary for us to give separate consideration to the contention in this case that the Arizona amendment denies appellants equal protection of the laws."[28] The difference was, "The language of the Arizona amendment prohibits employment discrimination against non-union workers, but it does not prohibit discrimination against union workers."[29] The unions claimed, according to the court, "[A] failure to provide the same protection for union workers as that provided for non-union workers places the union workers at a disadvantage, thus denying unions and their members the equal protection of Arizona’s laws."[30]
This argument was rejected, but only after the court discussed other Arizona laws that provided protection to unions:
"[W]e are unable to find any indication that Arizona’s amendment and statutes are weighted on the side of non-union as against union workers. We are satisfied that Arizona has attempted both in the anti-yellow-dog-contract law and in the anti-discrimination constitutional amendment to strike at what were considered evils, to strike where those evils were most felt, and to strike in a manner that would effectively suppress the evils."[31]
The court’s additional analysis of Arizona’s other laws implies that without additional state laws that protect union members, a right-to-work law without a union nondiscrimination clause might be declared a violation of the federal equal protection clause.
* Two other constitutional arguments of the unions deserve mention — one because it led to a holding that directly supports the "agreement, understanding or practice" provision of the model language provided later in this study, and a second because it was so ambitious. The Supreme Court rejected an argument that employees and employers could enter into a contract "obligating [the employer] to refuse to hire or retain union workers." The court stated, "If the states have constitutional power to ban such discrimination by law, they also have the power to ban contracts which if performed would bring about the prohibited discrimination." (See Lincoln Federal, 335 U.S. at 533.)
The unions’ more ambitious claim was described by the Supreme Court as follows:
"The right of unions and union members to demand that no non-union members work along with union members is ‘indispensable to the right of self organization and the association of workers into unions’; without a right of union members to refuse to work with non-union members, there are ‘no means of eliminating the competition of the non-union worker’; since, the reasoning continues, a ‘closed shop’ is indispensable to achievement of sufficient union membership to put unions and employers on a full equality for collective bargaining, a closed shop is consequently ‘an indispensable concomitant’ of ‘the right of employees to assemble into and associate together through labor organizations. * * *’ Justification for such an expansive construction of the right to speak, assemble and petition is then rested in part on appellants’ assertion ‘that the right to work as a non-unionist is in no way equivalent to or the parallel of the right to work as a union member; that there exists no constitutional right to work as a non-unionist on the one hand while the right to maintain employment free from discrimination because of union membership is constitutionally protected.’" (See Lincoln Federal, 335 U.S. at 530-31.)
The Supreme Court gave short shrift to this argument:
"We deem it unnecessary to elaborate the numerous reasons for our rejection of this contention of appellants. Nor need we appraise or analyze with particularity the rather startling ideas suggested to support some of the premises on which appellants’ conclusions rest. There cannot be wrung from a constitutional right of workers to assemble to discuss improvement of their own working standards, a further constitutional right to drive from remunerative employment all other persons who will not or can not, participate in union assemblies. The constitutional right of workers to assemble, to discuss and formulate plans for furthering their own self interest in jobs cannot be construed as a constitutional guarantee that none shall get and hold jobs except those who will join in the assembly or will agree to abide by the assembly’s plans." (Ibid., 531.)
As noted earlier, the unions opposing the Oklahoma right-to-work amendment challenged the amendment’s provision barring employers from discriminating against union members — a counterintuitive objection, since it questioned a provision meant to protect union members. Still, this challenge was meant to persuade the court to annul the entire amendment on grounds that the amendment’s parts were not severable. Specifically, the unions challenged the nondiscrimination provision on grounds that it was pre-empted by the federal NLRA.*
But neither of the U.S. Supreme Court rulings in Lincoln Federal and American Sash explicitly discussed the issue of whether right-to-work provisions protecting union members were pre-empted by the NLRA. This silence was seen as key by the 10th U.S. Circuit Court of Appeals in its analysis of Oklahoma’s right-to-work amendment:
"It is readily apparent that the Court in Lincoln Federal and American Sash was focused on the very narrow question of whether the state provisions at issue violated the Equal Protection Clause of the Fourteenth Amendment. There is absolutely no discussion of the question of preemption and, hence, no indication that provisions [that prohibit discrimination against union members] are not preempted by the NLRA."[32]
The 10th Circuit also contended that equal protection concerns do not mandate that state right-to-work laws contain provisions protecting union members from employer discrimination:
"Nor can Lincoln Federal and American Sash be read to stand for the proposition that the Equal Protection Clause of the Fourteenth Amendment requires the states to adopt [clauses that prevent discrimination against union members] if they choose to enact right-to-work laws. The Court was not required to reach this ultimate question because the state schemes at issue in both cases provided mutuality of protection [i.e., protection from discrimination for both union and nonunion workers]. Lincoln Federal, 335 U.S. at 532-33; American Sash, 335 U.S. at 541."[33]
The court also provided a second rationale for its view that a union nondiscrimination clause was not necessary on equal protection grounds by discussing the level of scrutiny to which the right-to-work amendment’s language would be subject in the event of an equal protection challenge. The court claimed that such a challenge almost certainly would fail because it would be subject to the lowest level of judicial scrutiny (known as the "rational basis" test):†
"Moreover, we note that neither union nor non-union status implicates a fundamental right or constitutes a protected class, so that a statute which addresses or favors one group over another need only reflect a rational basis. Accordingly, for defendants to prevail on their claim that the Fourteenth Amendment mandates mutuality in the treatment of union and non-union workers they must demonstrate that it would be irrational for a state to only provide protection against employment discrimination to non-union workers."[34]
In rejecting the argument that the nondiscrimination clause was necessary to satisfy the equal protection clause, and in noting that the Supreme Court had not made a ruling regarding pre-emption, the 10th Circuit felt justified in pre-empting the Oklahoma nonunion discrimination clause. This holding increased the number of pre-empted provisions in the amendment to three (the hiring-hall provision, the paycheck-protection provision and the union-nondiscrimination provision). The court then asked the Oklahoma Supreme Court to decide the question of severability.**[35] In addition, the 10th Circuit sought guidance from the Oklahoma Supreme Court on the technical question of whether the federal trial court should have applied the Oklahoma severability statute in this case, given that the law being challenged was a constitutional amendment, not a statute.[36]
* The pre-emption in this case would occur in the technical sense that the federal law would supersede the state amendment. Practically speaking, the “pre-emption” would have had no effect on the legality of discrimination against union members, since the NLRA prohibits such discrimination, just as the Oklahoma amendment attempted to do.
† When a law is challenged on grounds that it fails to satisfy the federal Constitution’s equal protection clause, it is subject to one of three levels of scrutiny. Strict scrutiny typically applies whenever the claim is based on an allegation of racial discrimination or a denial of a court-recognized fundamental right. Intermediate scrutiny generally applies to allegations of sex discrimination. The lowest level of scrutiny — “rational basis” review — applies to any other challenges that are based on the equal protection clause, and this review requires only that a governing body have some “rational basis” for its legislation. In practice, this rational basis test rarely leads to a law being overturned.
** The legal process in which one court asks another to decide a question is known as “certifying” the question.
The Oklahoma Supreme Court held that a severability analysis was not necessary, since the drafters of the amendment contemplated that parts of the amendment would be pre-empted by federal law:
"It is important to keep in mind that the federal courts in this matter have not declared any provision of the right to work law unconstitutional. Instead, the federal courts have merely held that the right to work law does not apply in certain circumstances due to the primacy of federal law, not that preemption lead to invalidation of any of the right to work law’s provisions. ...
"Just as whether some of the right to work amendment’s provisions were preempted by federal law was a question of federal law, whether the finding of the federal court’s [sic] requires us to engage in severability analysis is a question of state law. We hold that severability analysis is not necessary here for the reason that the right to work law contemplated that some of its provisions might be preempted by federal law and because plaintiffs failed to overcome the presumption that the right to work law is valid and enforceable. Thus, we decline to address plaintiffs’ various legal arguments in support of their claim that the rulings of the federal courts in this matter establish that the voters were somehow mislead [sic]."[37]
Justice Marian P. Opala concurred, noting that state law in the labor field needed to be flexible: "Because federal labor law is neither stagnant nor mummified in its present form, the drafters understood the outer boundaries of [a] right-to-work amendment must be flexible to remain in conformity with present as well as future federal re-definitions."[38]
After receiving the Oklahoma Supreme Court’s answer to the certified question, the 10th Circuit affirmed the trial court’s decision. The unions did not seek an appeal to the U.S. Supreme Court.
In Lincoln Federal, American Sash and the recent Oklahoma litigation, unions sought to have the entirety of the relevant right-to-work laws or amendments overturned. But in other challenges, the goal was not to have the amendments or laws overturned; rather, it was to limit their scope. These challenges have often related to whether all union fees are prohibited and whether public employees are covered by these amendments and laws.
In Cone v. Nevada Service Employees Union/SEIU Local 1107, the Nevada Supreme Court held that despite Nevada’s right-to-work law, unions could charge nonunion workers a fee for handling the nonmembers’ individual grievances.[39] The fee was substantial: Nonunion workers filing a grievance had to cover half the cost of the hearing officers and arbitrators for the grievance hearing and all of the union’s attorney fees up to $200 an hour. The fee schedule was imposed after a number of employees opted not to join the union.[40]
Oddly, the court relied on the same "free-rider" policy arguments that unions often make to voters and legislators when seeking to prevent enactment of a right-to-work law:
"We see no discrimination or coercion, however, in requiring nonunion members to pay reasonable costs associated with individual grievance representation, and … we are convinced that the exclusive bargaining relationship establishes a ‘mutuality of obligation’: a union has the obligation to represent all employees in the bargaining unit without regard to union membership, and the employee has a corresponding obligation, if permissible under the CBA [collective bargaining agreement] and required by the union policy, to share in defraying the costs of collective bargaining services from which he or she directly benefits. Our recognition of this mutuality of obligation will, in part, serve to discourage ‘free riders’ — employees who receive the benefits of union representation but are unwilling to contribute to its financial support.
"Although appellants cite much precedent, including NLRB opinions, in support of their position … we disagree with this authority because it leads to an inequitable result that we cannot condone, by essentially requiring union members to shoulder the burden of costs associated with nonunion members’ individual grievance representation."[41]
Nevada’s right-to-work law does not have a specific provision prohibiting collections of fees and charges, which likely aided the Nevada Supreme Court in reaching its decision. Thus, while silence regarding the charging of fees does not necessarily mean that a union would be allowed to charge them, this silence could allow a court to arrive at that conclusion.*
Another area in which courts have found ambiguity in right-to-work laws is public-sector employment. In Branch v. City of Myrtle Beach, the South Carolina Supreme Court held that public employees were exempt from the state’s right-to-work laws.[42] The lower court had noted that the right-to-work law explicitly applied to "any employer" and held that a public employer fit within the plain meaning of that term. But the South Carolina Supreme Court held the plain meaning analysis was improper, writing, "At the time the legislature enacted the right-to-work statute, labor relations statutes did not apply to public employment unless coverage was specifically required by the statute’s language."†[43]
A similar result was reached by the Tennessee Supreme Court in Keeble v. City of Alcoa.[44] The court indicated that where a law is to apply to a sovereign power — the state of Tennessee in Keeble v. Alcoa — the law must do so in "clear and specific language."[45] As in South Carolina, the court rejected an analysis based on the "plain meaning" of the statute.[46] Because the Tennessee right-to-work statute did not specifically include public employers, the court concluded that the right-to-work law did not apply to state government or its political subdivisions.
* A different result was reached by the Arizona Court of Appeals in American Federation of State, County and Municipal Employees, AFL-CIO, Local 2384 v. Phoenix. The court noted, "[N]othing in Arizona’s constitution or statutes specifically prohibited requiring the payment of a pro rata share of a union’s expenses, or similar fees, as a term or condition of employment." (See American Federation of State, County and Municipal Employees, AFL-CIO, Local 2384 v. Phoenix, 142 P.3d 234, 236 (Ariz. Ct. App. 2006).) The unions therefore sought a "fair share" fee, basically (in the words of the court) "a pro rata share of expenses incurred by the union for negotiation, administration, and enforcement of collective bargaining agreements benefiting all members of the bargaining unit without regard to union membership." (Ibid., 235 n. 2.)
The Arizona Court held such fees were improper in light of Arizona’s right-to-work amendment and accompanying right-to-work laws:
"The clear intent of the electorate of Arizona in enacting [Arizona’s constitutional right-to-work provision] and Arizona’s ‘right to work’ laws was to ensure the freedom of workers to choose whether to join and participate in a union. Allowing the proposed ‘fair share’ fee would be contrary to the intent voiced by Arizona citizens because it would essentially render meaningless the distinction between union membership and non-membership. Non-members would be forced to contribute to, and thus support, the Union, albeit in an amount slightly less than full union dues. Consequently, the proposed ‘fair share’ fee would, in its practical effect to non-union employees, be little different than mandatory membership dues. Such a ‘fair share’ fee is no less onerous to freedom of employment than a compulsory arrangement requiring the payment of full union dues." (Ibid., 242-43 (internal citations and footnote omitted).)
The court correctly summarized the fundamental point in this case, writing, "[I]t is the compulsion and not the amount which is determinative." (Ibid., 242 (emphasis in original).) Ultimately, in other words, a union provides collective bargaining and grievance services precisely because the union itself sought and obtained an exclusive power to bargain on behalf of all workers in the collective bargaining unit, including those workers who do not wish to be members of the union. With this monopoly power, the union has accepted the duty to represent all workers in that bargaining unit. Compelling a worker who has chosen not to join the union to subsidize the union’s choice is questionable on its face and a clear contradiction of the intent of a right-to-work law.
† At first blush, exempting public-sector employees from South Carolina’s right-to-work statute would seem to be significant. But as the South Carolina Supreme Court wrote, “Unlike private employees, public employees in South Carolina do not have the right to collective bargaining.” (See Branch v. Myrtle Beach, 532 S.E.2d 289, 292 (South Carolina 2000).) Thus, “[E]ven if the right-to-work statute applied to public employment, significant aspects of the statute would be totally irrelevant.” (Ibid.)
Model language for a right-to-work amendment to the Michigan Constitution appears in Graphic 2. If implemented, this language presumably would be added at the end of Article 1.
This model language is presented as a constitutional amendment, rather than as a statute, because part of the value of a right-to-work provision is the message the provision sends to businesses about the state’s business climate. This message might be attenuated if the right-to-work provision were simply placed in statute, which is easier to manipulate, change or repeal — a natural concern in Michigan, a state with a long history of a strong union political presence. Further, a right-to-work provision is more than just an economic measure: It involves a basic employment right that is arguably more appropriate as a constitutional, rather than statutory, requirement. Nevertheless, if the Legislature and the governor wished to pass this model right-to-work language as a statute, the model language would serve that purpose.*
Graphic 2: Model Right-to-Work Amendment for Michigan A. As used in this section, "labor organization" means any agency, union, employee representation committee, or organization of any kind that exists for the purpose, in whole or in part, of dealing with employers concerning wages, rates of pay, hours of work, other conditions of employment, or other forms of compensation. B. No person shall be required as a condition of obtaining or continuing public-sector or private-sector employment to:
C. An agreement, contract, understanding, or practice between a labor organization and an employer that violates this section is unlawful and unenforceable. This section will apply only to those agreements, contracts, understandings or practices that take force or are extended or renewed after this section takes effect. D. Any person who suffers an injury or a threatened injury under this section may bring a civil action for damages, injunctive relief, or both. In addition, the court shall award a prevailing plaintiff costs and reasonable attorney fees. E. This section shall be self-executing. If any part or parts of this section are found to be in conflict with the United States Constitution or federal law, the section shall be implemented to the maximum extent that the United States Constitution and federal law permit. Any provision held invalid or inoperative shall be severable from the remaining portions of this section. |
* Of course, minor modifications would be necessary to adapt the language to the requirements of statutory law. For example, legislators would need to strike the sentence in clause E indicating that the section was self-executing, and they would likely need to separately amend Michigan’s Public Employment Relations Act (see MCL § 423.201 et seq.). As will be discussed later, the Legislature would also need to decide whether violation of the right-to-work provisions should constitute a misdemeanor offense.
The model amendment benefited from the Oklahoma experience. As Justice Opala noted, federal labor law is not static; big changes sometimes occur. One example is the 1951 Railway Labor Act amendment that allowed union shops, which had previously been prohibited. Another is the 1947 Taft-Hartley amendments to the NLRA, which included section 14(b), the provision that permits state right-to-work laws. In clause E, the model amendment explicitly seeks to regulate to the full extent permissible by federal law at all times.*
Clause B(1) of the model amendment also contains a union-nondiscrimination clause, despite the 10th Circuit’s contention that such a provision would be pre-empted by federal law. It is by no means certain that the 10th Circuit properly interpreted Lincoln Federal and American Sash. The Supreme Court held that the two cases must be considered separately, a fact not properly accounted for in the 10th Circuit’s analysis. The Supreme Court issued separate opinions on the two cases because the Arizona amendment, unlike the Nebraska amendment and North Carolina law, did not contain a union nondiscrimination clause. The Supreme Court’s decision to issue separate opinions leads to the conclusion that a law that failed to include such a union nondiscrimination provision could risk an equal protection violation, despite the 10th Circuit’s assertion that such a law would likely pass muster under the "rational basis" test.
Of course, the presence of this nondiscrimination clause could theoretically lead to a court challenge to the model language similar to the lawsuit in Oklahoma. But the unions’ strategy in Oklahoma was entirely dependent on an argument that the language of the amendment was not severable. The model amendment proposed here contains an explicit severability clause, rendering a union challenge to the nondiscrimination clause all but fruitless. In clause E, the model amendment explicitly recognizes that federal law may (or may not) pre-empt some portion of its language. This explicit recognition obviates any argument that voters were unaware of a potential overlap or conflict with federal law. The clause also ensures that voters are directly informed that the remainder of the law would be enforced in the event of a pre-emption of any part of the amendment.
Note, too, that the amendment does not include either the hiring-hall provision or the paycheck-protection provision that did not pass muster in the Oklahoma case. The absence of these elements means that there is less language in the model that could be susceptible to a pre-emption claim, and therefore that there is less chance a court would use pre-emption and severability to annul the amendment in its entirety. It should also be noted that Michigan already has a paycheck protection statute (though it is somewhat narrow).[47]
And ultimately, there is much to be said for a labor relations policy that maximizes government neutrality. Given that federal law provides unions with a collective bargaining monopoly regardless of the wishes of some of the individual workers and employers involved, and given that the U.S. Constitution (and simple fairness) requires equal treatment under the law, a neutral approach would support providing antidiscrimination protection to both union members and nonmembers.†
The Oklahoma experience is not the only one that provided guidance. The beginning of Clause B of the model amendment discusses both private- and public-sector employment because the state supreme courts of South Carolina and Tennessee have held that their right-to-work laws did not apply to public employees in the absence of language explicitly saying so. The model language would recognize the same right-to-work freedoms for public employees as it would for private workers.
Clause B(3) is informed by Nevada’s Cone decision, in which the court ruled that a union could assess nonunion workers a fee when those nonmembers filed a grievance. While other courts have reached a different conclusion where the right-to-work provisions did not explicitly prohibit such fees, the model amendment in B(3) eliminates any ambiguity by explicitly prohibiting the collection of such an assessment (or any other type of fee).
Clause B(4) is included because people with religious objections to union membership are typically required to pay their union dues or agency fees to a charity instead of the union. Note that the clause would also exempt non-religious-objectors from such payments if the unions ever attempted to require the payments.
Clause D provides civil remedies for violations of the amendment, a stipulation common in other states’ right-to-work provisions. The remedies are meant to fully compensate a worker and to minimize obstacles to workers’ enforcing their rights. In particular, the provision granting an employee who prevails in court under this amendment both costs and reasonable attorney fees makes it easier for employees with legitimate claims to access the courts.
Many states include misdemeanor criminal sanctions in their right-to-work statutes, but only one state — Oklahoma — does so in a right-to-work amendment. The model language here does not include such a sanction. No other provision in the Michigan Constitution creates criminal penalties for violations; the state’s constitution is not generally considered the place to insert minor criminal infractions. Nothing in the model language prevents the Legislature from creating misdemeanor criminal penalties for violations of the amendment or from providing further civil remedies.** In other words, when it comes to enforcement of the section’s provisions, the amendment acts as a floor, not as a ceiling.
* This clause differs from the approach adopted in Michigan Senate Bill 607 and House Bill 4454. Those bills specifically list the federal laws that might pre-empt parts of the proposed right-to-work legislation and exempt the employees subject to these federal laws from the bills’ right-to-work guarantees. Such an approach was not adopted in the model amendment because these federal laws might change, and the model embraces an elastic approach that does not require a new state amendment for each change in federal law. For example, if Congress were to pass a new provision of the Railway Labor Act paralleling the NLRA’s section 14(b), then state right-to-work laws would be permitted to cover railway and airline employees. If the Senate and House bills were passed as written, however, the Legislature would have to amend the resulting state right-to-work law to include those transportation employees. The model amendment, in contrast, would automatically include them, since the amendment states, “[T]he section shall be implemented to the maximum extent that the United States Constitution and federal law permit.” For a discussion of the other differences between the Michigan Senate and House bills and the model amendment, see the two footnotes that follow.
† Thus, the model amendment differs from Michigan Senate Bill 607 and House Bill 4454, which do not include a union nondiscrimination clause.
** Indeed, both Senate Bill 607 and House Bill 4454 include misdemeanor criminal penalties for violation of the proposed statutes. Note that the bills also state that prosecutors or the attorney general “shall investigate each complaint of a violation of this act and shall prosecute the criminal case if credible evidence of a violation exists.” This provision of the proposed statutes would likely prompt a constitutional challenge on grounds that the “separation of powers” doctrine prevents the Legislature from restricting the discretion of prosecutors, who are part of the executive branch of government.
Case law shows that unions and other supporters of compulsory unionism will attempt to exploit any flaw in a right-to-work law or amendment. Nevertheless, the case law also indicates that with careful drafting, such laws or amendments can operate as intended.
A right-to-work amendment is likely to be a catalyst for economic improvement in Michigan. But more fundamentally, such an amendment would move Michigan closer to government neutrality in the regulation of labor relations. No doubt some unions would challenge the legality of such an amendment. But reviewing so many previous legal challenges makes it easier to foresee probable lines of attack. The model language provided here should produce an amendment that clearly articulates the will of the people and effectuates it, regardless of subsequent legal challenge.
The following is the text of Michigan Senate Bill 607, a proposed right-to-work statute. As noted earlier in this study, the language of this statute is essentially identical to that of Michigan House Bill 4454. Page breaks in the bill are indicated by a solid line.
SENATE BILL No. 607
A bill to prohibit employers from placing certain conditions
on employment; to grant rights to employees; to impose duties and
responsibilities on certain state and local officers; to make
certain agreements unlawful; and to provide remedies and penalties.
THE PEOPLE OF THE STATE OF MICHIGAN ENACT:
1 Sec. 1. This act shall be known and may be cited as the "right
2 to work law".
3 Sec. 3. As used in this act:
4 (a) "Employer" means a person or entity that pays 1 or more
5 individuals under an express or implied contract of hire.
6 (b) "Labor organization" means an organization of any kind, an
7 agency or employee representation committee, group, association, or
8 plan in which employees participate and which exists for the
9 purpose, in whole or in part, of dealing with employers concerning
—————————————
1 grievances, labor disputes, wages, rates of pay, hours, or
other
2 terms or conditions of employment.
3 Sec. 5. Except as provided in section 13, a person shall not
4 require an employee to do any of the following as a condition of
5 employment or continued employment:
6 (a) Become or remain a member of a labor organization.
7 (b) Pay dues, fees, assessments, or other similar charges to a
8 labor organization.
9 (c) Pay to a charity or other third party an amount equivalent to
10 or pro rata portion of dues, fees, assessments, or other charges
11 required of members of a labor organization.
12 Sec. 7. Except as provided in section 13, an agreement,
13 understanding, or practice between a labor organization and employer
14 that violates employee rights granted under this act is unlawful and
15 unenforceable.
16 Sec. 9. A person who suffers an injury or a threatened injury
17 from a violation of this act may bring a civil action for damages,
18 injunctive relief, or both. The court may award a prevailing plaintiff
19 costs and reasonable attorney fees. The civil remedy is independent
20 of, and in addition to, any criminal proceeding or sanction prescribed
21 for a violation of this act.
22 Sec. 11. A person who violates this act is guilty of a
23 misdemeanor. The prosecuting attorney of the county or the attorney
24 general shall investigate each complaint of a violation of this act
25 and shall prosecute the criminal case if credible evidence of a
26 violation exists.
27 Sec. 13. This act does not apply to any of the following:
—————————————
1 (a) An employer or employee covered by the federal railway
labor
2 act, 45 USC 151 to 188.
3 (b) A federal employer or employee.
4 (c) An employer or employee at an exclusively federal enclave.
5 (d) An employment contract entered into before the effective date
6 of this act, except that this act applies to a contract renewal or
7 extension that takes effect after the effective date of this act.
8 (e) A situation in which it would conflict with, or be
9 preempted by, federal law.
10 Enacting section 1. This act does not take effect unless
11 Senate Bill No. 608
12 of the 94th Legislature is enacted into law.
1 "Bishop Wants RTW on Ballot," MIRS Capitol Capsule, July 17, 2007, https://mirsnews.com/capsule.php?gid=785 (accessed August 19, 2007).
2 Eric Morath, "Karmanos rails against unions," The Detroit News, April 20, 2007, https://www.detroitnews.com/apps/pbcs.dll/article?AID=/20070420/BIZ/704200365/1001 (accessed August 19, 2007). Parenthetical text appears in the original.
3 Hunter, "Michigan Labor Law: What Every Citizen Should Know," 5.
4 Ibid.
5 Davenport v. Washington Educ. Ass’n, __ U.S. __, __; 127 S. Ct. 2372, 2376 (2007).
6 Hunter, "Michigan Labor Law: What Every Citizen Should Know," 7.
7 U.S. Const. art. VI, § 2.
8 Crosby v. Nat’l Foreign Trade Council, 530 US 363, 372-73 (2000) (citations and footnote omitted).
9 Local 514, Transport Workers Union of America v. Keating, 212 F.Supp.2d 1319, 1325 (E.D. Ok. 2002).
10 Ry. Employes’ Dep’t. v. Hanson, 351 U.S. 225 (1956).
11 Ibid., 233.
12 See 45 U.S.C. § 152 Eleventh (a).
13 29 U.S.C. § 164(b).
14 Author’s calculations based on http://www.ok.gov/~elections/02sq.pdf (accessed August 17, 2007).
15 Local 514, Transport Workers Union of America v. Keating, 212 F.Supp.2d 1319, 1323 (E.D. Ok. 2002).
16 Ibid., 1324.
17 Ibid., 1326.
18 Ibid., 1327 n. 6.
19 Ibid., 1326-27.
20 Ibid., 1328 (citing Okla. Stat. Ann. tit. 75, § 11a).
21 Ibid., 1331.
22 Okla. Const. art. XXIII, § 1A (B)(1).
23 Lincoln Federal Labor Union No. 19129, American Federation of Labor v. Northwestern Iron & Metal Co., 335 U.S. 525 (1949).
24 American Federation of Labor, Arizona State Federation of Labor v. American Sash & Door Co., 335 U.S. 538 (1949).
25 Lincoln Federal, 335 U.S. at 527-28 (footnote omitted).
26 Ibid., 528.
27 Ibid., 532.
28 American Sash, 335 U.S. at 540.
29 Ibid.
30 Ibid.
31 Ibid., 541.
32 Local 514, Transport Workers Union of America v. Keating, 358 F.3d 743, 753 (2004).
33 Ibid., 754.
34 Ibid.
35 Ibid., 744.
36 Ibid.
37 Local 514, Transport Workers Union of America v. Keating, 83 P.3d 835, 840 (Oklahoma 2003).
38 Ibid., 845-46 (Opala, J., concurring).
39 Cone v. Nevada Service Employees Union/SEIU Local 1107, 998 P.2d 1178 (Nevada 2000).
40 Ibid., 1180.
41 Ibid., 1182-83 (internal citations and footnote omitted).
42 Branch v. Myrtle Beach, 532 S.E.2d 289 (South Carolina 2000).
43 Ibid., 291.
44 Keeble v. City of Alcoa, 319 S.W.2d 249 (Tennessee 1958).
45 Ibid., 251.
46 Ibid., 250.
47 MCL § 169.255(6).