AFSCME Iowa Council 61 v. State of Iowa and Iowa Public Employment Relations Board

Opinion

IN THE SUPREME COURT OF IOWA No. 17–1841

Filed May 17, 2019

AFSCME IOWA COUNCIL 61, JOHNATHAN GOOD, RYAN De VRIES, TERRA KINNEY, and SUSAN BAKER,

Appellants, vs.

STATE OF IOWA and IOWA PUBLIC EMPLOYMENT RELATIONS BOARD,

Appellees.

Appeal from the Iowa District Court for Polk County, Arthur E.

Gamble, Judge.

Public employee union and several members appeal summary judgment dismissing constitutional challenges to 2017 amendments to Iowa Code chapter 20, the Public Employment Relations Act. AFFIRMED.

Mark T. Hedberg and Sarah M. Baumgartner of Hedberg & Boulton, P.C., Des Moines, for appellants.

Matthew C. McDermott, Michael R. Reck, Kelsey J. Knowles, and Espnola F. Cartmill of Belin McCormick, P.C., Des Moines, for appellees.

WATERMAN, Justice.

This appeal, submitted with Iowa State Education Ass’n v. State, ___ N.W.2d ___ (Iowa 2019), also filed today, presents constitutional challenges to the 2017 amendments to the Public Employment Relations Act, Iowa Code chapter 20. The amendments ended payroll deductions for union dues and narrowed the scope of mandatory collective bargaining topics for bargaining units comprised of less than thirty percent “public safety employees,” defined to include most police officers

and firefighters. The new classifications result in many public employees losing significant statutory bargaining rights compared to other public

employees with arguably similar jobs. A public employee union and several of its members filed this action against the State of Iowa and the Public Employment Relations Board (PERB) seeking injunctive and declaratory relief. The plaintiffs allege the amendments violate the equal protection clause of the Iowa Constitution and violate their right to freedom of association. The district court granted the defendants’ motion for summary judgment dismissing the action, and we retained the plaintiffs’ appeal.

Our role is to decide whether constitutional lines were crossed, not to sit as a superlegislature rethinking policy choices of the elected branches. We conclude the 2017 amendments withstand the constitutional challenges. The plaintiffs concede there is no constitutional right to public-sector collective bargaining or payroll deductions. The parties agree the equal protection claims are reviewed under the rational basis test. The legislature could reasonably conclude

that the goal of keeping labor peace with unions comprised of at least thirty percent public safety employees, and the greater risks faced by

emergency first responders, justified the classification. We hold the

3

legislative classifications are not so overinclusive or underinclusive as to be unconstitutional under our highly deferential standard of review. We further hold the amendments do not violate constitutional rights of freedom of association. Public employees remain free to belong to the same unions. Accordingly, we affirm the district court’s summary judgment.

I. Background Facts and Proceedings.

We begin by reviewing the statute in place before the 2017

amendments to put the constitutional challenges in context.1 In 1974, after public employees engaged in multiple strikes, the Iowa legislature

enacted the Public Employment Relations Act (PERA), codified at Iowa Code chapter 20. See generally Waterloo Educ. Ass’n v. Iowa Pub. Emp’t Relations Bd., 740 N.W.2d 418 (Iowa 2007) (detailing the history of public sector collective bargaining). PERA sought to create an orderly system of collective bargaining for public employees by establishing rules and procedures and by prohibiting strikes.2 Iowa Code §§ 20.6, .9, .10 (2017). PERA permitted, but did not require, public employees to join a public employee organization (union).3 Id. § 20.8. Employees could vote to select a union to represent them. Id. An employee who joined a union had the option to pay dues through automatic payroll deductions. Id. § 20.9; id. §§ 70A.17A, .19.

1The plaintiffs do not challenge the payroll deduction prohibition, a provision we hold withstands constitutional scrutiny in Iowa State Education Ass’n, ___ N.W.2d at ___.

2As of 2010, only one-half of the states had a comprehensive collective

bargaining statute. See Marilyn Raskin-Ortiz & Emily Martin, Bargaining in States Without Public Sector Collective Bargaining Legislation, ABA Labor & Emp’t Law Section Subcommittee Report, at 1 (2010) [hereinafter Raskin-Ortiz & Martin]. http://apps.americanbar.org/labor/slgbcomm/mw/papers/2010/home.shtml.

3PERA defines unions as “employee organizations.” Iowa Code § 20.3(4).

4

Once employees selected a union, PERA required the union and public employer to bargain in good faith on these topics:

wages, hours, vacations, insurance, holidays, leaves of absence, shift differentials, overtime compensation, supplemental pay, seniority, transfer procedures, job classifications, health and safety matters, evaluation procedures, procedures for staff reduction, in-service training and other matters mutually agreed upon.

Id. § 20.9.

If a public employer and union were unable to reach an agreement

on these mandatory topics, PERA established a procedure for resolving the impasse through mediation and binding arbitration. Id. §§ 20.20,

.22. If an impasse reached arbitration, each party submitted a final offer to an arbitrator. Id. § 20.22(3). The arbitrator was required to consider the following factors:

a. Past collective bargaining contracts between the parties including the bargaining that led up to such contracts.

b. Comparison of wages, hours and conditions of employment of the involved public employees with those of other public employees doing comparable work, giving consideration to factors peculiar to the area and the classifications involved.

c. The interests and welfare of the public, the ability of the public employer to finance economic adjustments and the effect of such adjustments on the normal standard of services.

d. The power of the public employer to levy taxes and appropriate funds for the conduct of its operations.

Id. § 20.22(7). After considering the proposals and the relevant factors, the arbitrator “select[ed] . . . the most reasonable offer, in the arbitrator’s judgment, of the final offers on each impasse item submitted by the

parties.” Id. § 20.22(9).

PERA imposed harsh penalties for engaging in strikes. Id.

§§ 20.10(3)(h), .12. PERA authorized courts to issue injunctions to

5

restrain any actual or imminently threatened strike. Id. § 20.12(3). Anyone who failed to comply with an injunction faced contempt sanctions and punishment including up to six months in jail, fines, and automatic discharge from employment for an employee, or immediate decertification as a union. Id. § 20.12(3)–(6). See generally Iowa Code ch. 665 (contempt). There have been no strikes by public employees in Iowa since PERA’s enactment in 1974. The University of Iowa Labor Center, “To Promote Harmonious and Cooperative Relationships”: A Brief

History of Public Sector Collective Bargaining in Iowa, 1966 to 2016, 7 (2016), https://www.iowaaflcio.org/system/files/history_of_ia_public_

sector_bargaining.pdf.

In February 2017, the Iowa legislature enacted House File 291, amending PERA. 2017 Iowa Acts ch. 2 (codified in part at Iowa Code ch. 20 (2018)). On February 17, the Governor signed House File 291 into law. The amendments altered the scope of mandatory collective bargaining and arbitration and eliminated payroll deductions for all union dues. See generally Iowa Code ch. 20.

Collective bargaining laws for public employees vary by state, with some states allowing collective bargaining rights for police and firefighters not shared by other public employees.4 House File 291 gave

4As of 2018, twenty-eight states require collective bargaining. Eric J. Brunner & Andrew Ju, State Collective Bargaining Laws and Public-Sector Pay, 72 ILR Rev. 480, 487 (2019) [hereinafter Brunner & Ju]. Fifteen states allow state employers to decide whether or not to collectively bargain. Id. The range of topics public employees are able to bargain over varies from state to state, as does the employees’ ability to compel arbitration in the event of an impasse. Raskin-Ortiz & Martin at 4–10.

Of the states that require or permit collective bargaining, Alabama, Delaware, Idaho, Kentucky, Oklahoma, Rhode Island, and Wyoming have separate bargaining rights for police officers and/or firefighters. Id. Three states—North Carolina, South Carolina, and Virginia—prohibit collective bargaining for any public employees. Brunner & Ju at 487. Arizona and Texas limit collective bargaining to police officers and firefighters, while Georgia limits collective bargaining rights to firefighters alone. Id.

6

public employees different bargaining rights depending on whether they are part of a bargaining unit with at least thirty percent “public safety employees.” Public safety employees are defined to include

a. A sheriff’s regular deputy.

b. A marshal or police officer of a city, township, or special-purpose district or authority who is a member of a paid police department.

c. A member, except a non-peace officer member, of the division of state patrol, narcotics enforcement, state fire marshal, or criminal investigation, including but not limited to a gaming enforcement officer, who has been duly appointed by the department of public safety in accordance with section 80.15.

d. A conservation officer or park ranger as authorized by section 456A.13.

e. A permanent or full-time fire fighter of a city, township, or special-purpose district or authority who is a member of a paid fire department.

f. A peace officer designated by the department of transportation under section 321.477 who is subject to mandated law enforcement training.

Iowa Code § 20.3(11). Not included in the statutory definition of public safety employees are university police, probation or parole officers, fraud bureau investigation officers, airport firefighters, corrections officers, and emergency medical service providers.

If a union represents a bargaining unit with at least thirty percent public safety employees, it may exercise broad bargaining rights on behalf of all of its members, including those who are not public safety employees. Id. § 20.9(1). The union continues to have the right to bargain and, in the event of an impasse, the right to mediate and arbitrate with public employers on the following mandatory topics:

wages, hours, vacations, insurance, holidays, leaves of absence, shift differentials, overtime compensation, supplemental pay, seniority, transfer procedures, job classifications, health and safety matters, evaluation procedures, procedures for staff reduction, in-service

7

training, grievance procedures for resolving any questions arising under the agreement, and other matters mutually agreed upon.

Id.

In sharp contrast, for unions representing a bargaining unit with less than thirty percent public safety employees, House File 291 limited mandatory bargaining and, in the event of an impasse, mediation and arbitration, to the subject of “base wages and other matters mutually agreed upon.” Id.5 The amendment specifies that these subjects “shall

be interpreted narrowly and restrictively.” Id. The amendments allow public employers to voluntarily bargain over formerly mandatory topics.

Longevity pay, shift differentials, and overtime compensation are still permissive subjects of bargaining. See Iowa Code § 20.9(1), (3). This leaves it up to the state or local government or school board whether to negotiate on these matters. See Waterloo Educ. Ass’n, 740 N.W.2d at 421. Public employees, like all citizens in our state, have the ability to affect those decisions. A unit of state government, a municipality, or a school board that wishes to negotiate on these matters with the employee organization is free to do so. But the union may not bargain over “insurance, leaves of absence for political activities, supplemental pay, transfer procedures, evaluation procedures, procedures for staff reduction, and subcontracting public services.” Iowa Code § 20.9(3).

During arbitration with a bargaining unit consisting of at least thirty percent public safety employees, the arbitrator considers most of the same factors as before the 2017 amendments. Compare id.

5A 2010 report found that a number of states limit mandatory binding arbitration to certain classes of employees, including police and firefighters. See Raskin-Ortiz & Martin at 11–12 (noting Alaska, California, Illinois, Michigan, Montana, Nevada, New Jersey, New York, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, Washington, Wisconsin, and Wyoming limited compulsory arbitration to certain classes of employees including firefighters and/or police officers).

8

§ 20.22(7) (2018), with id. § 20.22(7) (2017). The only change House File 291 made is that the arbitrator may no longer consider “[t]he power of the public employer to levy taxes and appropriate funds for the conduct of its operations.” Id. § 20.22(7)(d) (2017).

For all other public employee units, the arbitrator, in reaching a final decision, must consider

(1) Comparison of base wages, hours, and conditions of employment of the involved public employees with those of other public employees doing comparable work, giving consideration to factors peculiar to the area and the classifications involved. To the extent adequate, applicable data is available, the arbitrator shall also compare base wages, hours, and conditions of employment of the involved public employees with those of private sector employees doing comparable work, giving consideration to factors peculiar to the area and the classifications involved.

(2) The interests and welfare of the public.

(3) The financial ability of the employer to meet the cost of an offer in light of the current economic conditions of the public employer. The arbitrator shall give substantial weight to evidence that the public employer’s authority to utilize funds is restricted to special purposes or circumstances by state or federal law, rules, regulations, or grant requirements.

Iowa Code § 20.22(8)(a) (2018). The arbitrator shall not consider,

(1) Past collective bargaining agreements between the parties or bargaining that led to such agreements.

(2) The public employer’s ability to fund an award through the increase or imposition of new taxes, fees, or charges, or to develop other sources of revenues.

Id. § 20.22(8)(b).

Regardless of the makeup of the bargaining unit, the arbitrator must still determine the most reasonable offer. Id. § 20.22(10)(a).

However, if the bargaining unit is made up of less than thirty percent public safety employees and there is an impasse on base wages, the

arbitrator is prohibited from selecting an offer, even if it is reasonable,

9

that provides for an increase in base wages that would exceed in any year the increase in a specified consumer price index or three percent, whichever is less. Id. § 20.22(10)(b)(1).

House File 291 also eliminated the right of all public employees, including public safety employees, to bargain over union dues checkoffs and to pay union dues through payroll deductions. Id. § 20.9(3); id. § 70A.19. Public employees may still make other payments through payroll deductions, such as insurance premiums, charitable

contributions, and dues in professional associations. Id. §§ 70A.15A, .17, .17A.

The plaintiffs in this case are a public employee union and four of its members. Iowa Council 61 of the American Federation of State, County and Municipal Employees (AFSCME) represents public employees throughout Iowa. The individual plaintiffs, Johnathan Good, a corrections officer; Ryan De Vries, a police officer; Terra Kinney, a motor vehicle enforcement officer; and Susan Baker, a drafter, are public employees and members of AFSCME. All of AFSCME’s bargaining units in Iowa are comprised of less than thirty percent public safety employees. House File 291 restricted collective bargaining rights for every AFSCME bargaining unit, including those with public safety employees.

In February 2017, the plaintiffs filed this civil action for declaratory and injunctive relief. The defendants, the State of Iowa and PERB, answered, and the parties filed cross-motions for summary judgment. The plaintiffs’ motion for summary judgment argued House File 291 violates article I, section 6, the equal protection clause of the Iowa

Constitution because it unconstitutionally deprives some public employees of rights guaranteed to other, similarly situated public

employees. The plaintiffs also argued that House File 291 deprives all

10

AFSCME-represented state public safety employees of the right to meaningful collective bargaining, violating their fundamental right to freedom of association, and the court should therefore evaluate the law under a strict scrutiny standard.

The district court denied the plaintiffs’ motion for summary judgment and granted the defendants’ motion for summary judgment. The court rejected the plaintiffs’ freedom of association argument. With regard to the equal protection challenge, the court applied the rational

basis test and ruled that House File 291 is constitutional. The court concluded that while the amendments distinguish between similarly

situated people, the State’s desire to avoid public safety employee strikes was a realistically conceivable purpose and was based in fact, and the relationship between the classification and the purpose was not so weak as to be viewed as arbitrary.

The plaintiffs appealed, and we retained their appeal.

II. Scope of Review.

“We review summary judgment rulings for correction of errors at law.” Baker v. City of Iowa City, 867 N.W.2d 44, 51 (Iowa 2015). “We view the entire record in the light most favorable to the nonmoving party, making every legitimate inference that the evidence in the record will support in favor of the nonmoving party.” Bass v. J.C. Penney Co., 880 N.W.2d 751, 755 (Iowa 2016).

We review constitutional claims de novo. State v. Groves, 742 N.W.2d 90, 92 (Iowa 2007). Our standard of review with regard to constitutional challenges to statutes is well established.

We review constitutional challenges to a statute de novo. In doing so, we must remember that statutes are cloaked with a presumption of constitutionality. The challenger bears a heavy burden, because it must prove the unconstitutionality

11

beyond a reasonable doubt. Moreover, “the challenger must refute every reasonable basis upon which the statute could be found to be constitutional.” Furthermore, if the statute is capable of being construed in more than one manner, one of which is constitutional, we must adopt that construction.

State v. Seering, 701 N.W.2d 655, 661 (Iowa 2005) (quoting State v. Hernandez-Lopez, 639 N.W.2d 226, 233 (Iowa 2002)), superseded by statute on other grounds, 2009 Iowa Acts ch. 119, § 3 (codified at Iowa Code § 692A.103 (Supp. 2009)), as recognized in In re T.H., 913 N.W.2d 578, 587–88 (Iowa 2018).

III. Analysis.

The plaintiffs argue that House File 291 amendments to Iowa Code chapter 20 fail rational basis scrutiny under article I, section 6 of the equal protection clause of the Iowa Constitution. Additionally, the plaintiffs argue that the amendments violate their right to freedom of association. We address each challenge in turn.

A. Iowa’s Equal Protection Analysis. The plaintiffs argue that the amendments to Iowa Code chapter 20 violate their right to equal protection under the Iowa Constitution because the defendants’ asserted rationale is unsupported by the legislative facts and further because House File 291’s extreme degrees of overinclusiveness and

underinclusiveness render the amendments arbitrary. We conclude the plaintiffs’ equal protection challenge fails because the plaintiffs cannot

meet their burden of refuting every reasonable basis upon which the classification could be sustained.

Article I, section 6 of the Iowa Constitution is referred to as the equal protection clause and provides, “All laws of a general nature shall

have a uniform operation; the general assembly shall not grant to any citizen, or class of citizens, privileges or immunities, which, upon the

12

same terms shall not equally belong to all citizens.” Iowa Const. art. I, § 6.

Iowa’s equal protection clause “is essentially a direction that all persons similarly situated should be treated alike.” Varnum v. Brien, 763 N.W.2d 862, 878–79 (Iowa 2009) (quoting Racing Ass’n of Cent. Iowa v. Fitzgerald (RACI), 675 N.W.2d 1, 7 (Iowa 2004)). In Varnum, we noted,

Even in the zealous protection of the constitution’s mandate of equal protection, courts must give respect to the legislative process and presume its enactments are constitutional. We understand that Iowa’s tripartite system of government requires the legislature to make difficult policy choices, including distributing benefits and burdens amongst the citizens of Iowa. In this process, some classifications and barriers are inevitable. As a result, courts pay deference to legislative decisions when called upon to determine whether the Iowa Constitution’s mandate of equality has been violated by legislative action. More specifically, when evaluating challenges based on the equal protection clause, our deference to legislative policy-making is primarily manifested in the level of scrutiny we apply to review legislative action.

Id. at 879.

To prove an equal protection violation, the plaintiffs must first establish that the statute treats similarly situated individuals differently. McQuistion v. City of Clinton, 872 N.W.2d 817, 830 (Iowa 2015). Generally, however, determining whether classifications involve similarly situated individuals is intertwined with whether the identified classification has any rational basis. State v. Dudley, 766 N.W.2d 606, 616 (Iowa 2009).

Here, House File 291 distinguishes first between public safety employees and all other public employees, and second between

bargaining units comprised of at least thirty percent public safety employees and all other bargaining units. The parties agree that rational

basis review applies to the plaintiffs’ equal protection challenge.

13

“The rational basis test is a ‘very deferential standard.’ ” NextEra Energy Res. LLC v. Iowa Utils. Bd., 815 N.W.2d 30, 46 (Iowa 2012) (quoting Varnum, 763 N.W.2d at 879). Plaintiffs bear “the heavy burden of showing the statute unconstitutional and must negate every reasonable basis upon which the classification may be sustained.” Id. (quoting Bierkamp v. Rogers, 293 N.W.2d 577, 579–80 (Iowa 1980)).

As we noted in Varnum,

The rational basis test defers to the legislature’s prerogative to make policy decisions by requiring only a plausible policy justification, mere rationality of the facts underlying the decision and, again, a merely rational relationship between the classification and the policy justification.

763 N.W.2d at 879.

“We will not declare something unconstitutional under the rational-basis test unless it ‘clearly, palpably, and without doubt infringe[s] upon the constitution.’ ” Residential & Agric. Advisory Comm., LLC, v. Dyersville City Council, 888 N.W.2d 24, 50 (Iowa 2016) (alteration in original) (quoting RACI, 675 N.W.2d at 8). Nevertheless, the rational basis standard, while deferential, “ ‘is not a toothless one’ in Iowa.” Varnum, 763 N.W.2d at 879 (quoting RACI, 675 N.W.2d at 9). “[T]his

court engages in a meaningful review of all legislation challenged on equal protection grounds by applying the rational basis test to the facts

of each case.” Id.

We use a three-part analysis when reviewing challenges to a statute under article I, section 6. “First, we must determine whether there was a valid, ‘realistically conceivable’ purpose that served a legitimate government interest.” Residential & Agric. Advisory Comm.,

LLC, 888 N.W.2d at 50 (quoting McQuistion, 872 N.W.2d at 831). “To be realistically conceivable, the [statute] cannot be ‘so overinclusive and

14

underinclusive as to be irrational.’ ” Id. (quoting Horsfield Materials, Inc. v. City of Dyersville, 834 N.W.2d 444, 459 (Iowa 2013)). “Next, the court must evaluate whether the ‘reason has a basis in fact.’ ” McQuistion, 872 N.W.2d at 831 (quoting RACI, 675 N.W.2d at 7–8). “[A]lthough ‘actual proof of an asserted justification [i]s not necessary, . . . the court w[ill] not simply accept it at face value and w[ill] examine it to determine whether it [i]s credible as opposed to specious.” LSCP, LLLP, v. Kay- Decker, 861 N.W.2d 846, 860 (Iowa 2015) (alteration in original) (quoting

Qwest Corp. v. Iowa State Bd. of Tax Review, 829 N.W.2d 550, 560 (Iowa 2013)); see also King v. State, 818 N.W.2d 1, 30 (Iowa 2012) (“[W]e have

continued to uphold legislative classifications based on judgments the legislature could have made, without requiring evidence or ‘proof’ in either a traditional or a nontraditional sense.” (Emphasis added.)).

“Legislative facts are relevant in deciding these constitutional issues because courts must normally analyze ‘whether there exist circumstances which constitutionally either legitimate the exercise of legislative power or substantiate the rationality of the legislative product.’ ” Varnum, 763 N.W.2d at 881 (quoting 2 John W. Strong, McCormick on Evidence § 328, at 370 (5th ed. 1999)). Legislative facts “may be presented either formally or informally” and consist of “social, economic, political, or scientific facts.” Id. (first quoting Welsh v. Branstad, 470 N.W.2d 644, 648 (Iowa 1991)).

The plaintiffs ask that if we find House File 291 to be constitutional, we reevaluate our rational basis standard. The plaintiffs argue that courts should not be able to rely on unstated rationales in

upholding a statute. We disagree. As the foregoing authorities make clear, we are not limited to considering only the facts stated on the

record during a legislative debate.

15

Finally, “we evaluate whether the relationship between the classification and the purpose for the classification ‘is so weak that the classification must be viewed as arbitrary.’ ” Residential & Agric. Advisory Comm., LLC, 888 N.W.2d at 50 (quoting McQuistion, 872 N.W.2d at 831).

As we recently reiterated in unanimously rejecting a federal equal protection challenge, courts have only a limited role in rational basis review,

We many times have said, and but weeks ago repeated, that rational-basis review in equal protection analysis “is not a license for courts to judge the wisdom, fairness, or logic of legislative choices.” Nor does it authorize “the judiciary [to] sit as a superlegislature to judge the wisdom or desirability of legislative policy determinations made in areas that neither affect fundamental rights nor proceed along suspect lines.” For these reasons, a classification neither involving fundamental rights nor proceeding along suspect lines is accorded a strong presumption of validity. Such a classification cannot run afoul of the Equal Protection Clause if there is a rational relationship between the disparity of treatment and some legitimate governmental purpose. Further, a legislature that creates these categories need not “actually articulate at any time the purpose or rationale supporting its classification.”

Instead, a classification “must be upheld against equal protection challenge if there is any reasonably conceivable state of facts that could provide a rational basis for the classification.”

Baker, 867 N.W.2d at 57 (alteration in original) (quoting Heller v. Doe ex rel. Doe, 509 U.S. 312, 319–21, 113 S. Ct. 2637, 2642 (1993)).

Our role is similarly limited under the Iowa Constitution. See Qwest Corp., 829 N.W.2d at 560 (“[In RACI,] we made clear that actual proof of an asserted justification was not necessary, but the court would not simply accept it at face value and would examine it to determine

whether it was credible as opposed to specious.”); King, 818 N.W.2d at 30 (“RACI has not been the death knell for traditional rational basis review.

16

Since RACI was decided, we have continued to uphold legislative classifications based on judgments the legislature could have made, without requiring evidence or ‘proof’ in either a traditional or nontraditional sense.”).

The district court found that the valid, realistically conceivable purpose for House File 291 was a concern for labor peace, especially among public safety employees. The State also asserts that another purpose was the unique health and safety concerns public safety

employees face. We consider each justification.

1. Labor peace rationale. The plaintiffs argue that House File

291’s legislative history belies the labor peace justification because no one mentioned this justification during the recorded legislative debates as a reason for amending PERA. The plaintiffs also argue that House File 291’s definition of public safety employees includes employees who would not be crucial to maintaining labor peace, such as park rangers, DOT motor vehicle enforcement officers, fire marshals, and gaming enforcement officers, while excluding employees who may be necessary to maintain peace during a strike, including university police officers and other emergency medical service providers. The plaintiffs note that police officers already routinely enforce laws against union members, neighbors, friends, and even other police officers. Finally, the plaintiffs argue that the labor peace rationale is belied by the fact that there has not been a strike since PERA was enacted in 1974.

The plaintiffs also argue that even if a labor peace rationale could support House File 291, the law is so overinclusive and underinclusive

“it cannot [reasonably] be said to . . . further that goal.” LSCP, LLLP, 861 N.W.2d at 861 (alterations in original) (quoting Bierkamp, 293 N.W.2d at

584). The plaintiffs contend that the thirty percent threshold ignores the

17

bargaining unit’s size, and some cities could have entire police forces that do not have expanded bargaining rights. These public safety employees would not have the same incentive to avoid strikes. The plaintiffs give examples of the effect House File 291 has on public safety employees. For example, plaintiffs identify a number of populous counties including Tama (population 17,337), Fayette (population 20,257), Delaware (population 17,403), Dubuque (population 97,125), Harrison (population 14,265), and Black Hawk (population 133,455) in

which sheriff’s deputies are unable to exercise the broad collective bargaining rights guaranteed to public safety employees in House File

291 because they are in bargaining units made up of less than thirty percent public safety employees. Yet deputies from comparably populated counties such as Floyd (population 15,960), Woodbury (population 102,782), Cedar (population 18,340), Webster (population 37,071), and Washington (population 22,247) are able to exercise broad collective bargaining rights under House File 291. Plaintiffs argue this extreme arbitrariness is not justified by any of the purported rationales of House File 291.

The defendants argue that the thirty percent threshold is rational because the risk from labor unrest is materially greater in a unit with a larger percentage of public safety employees. The defendants argue this thirty percent threshold had another rationale, protecting the public fisc. The thirty percent threshold also provides greater assurance that in the event of labor unrest there would be a critical mass of public safety employees available to enforce the law and preserve public safety.

The defendants rely on Wisconsin Education Ass’n Council v.

Walker, in which the United States Court of Appeals for the Seventh

Circuit rejected a public employee challenge to recent amendments to the

18

Wisconsin public collective bargaining statute. 705 F.3d 640 (7th Cir. 2013). The Seventh Circuit held there was a rational basis for the amendments, stating,

[E]xperience has borne out the state’s fears: in the wake of Act 10’s proposal and passage, thousands descended on the state capital in protest and numerous teachers organized a sick-out through their unions, forcing schools to close, while the state avoided the large societal cost of immediate labor unrest among public safety employees. Wisconsin was free to determine that the costs of potential labor unrest exceeded the benefits of restricting the public safety unions.

Id. at 655.

The plaintiffs argue Walker is unpersuasive because Wisconsin’s collective bargaining law is fundamentally different than Iowa’s. The Wisconsin statute treats all safety employees alike, regardless of their unit placement, while restricting the bargaining rights of all general public employees. The Wisconsin statute also has less onerous antistrike penalties. Additionally, Wisconsin’s statute was evaluated under a federal equal protection framework, while our court applies a more stringent rational basis “with teeth” standard of review.

We hold that maintaining labor peace is a valid, realistically conceivable purpose and has a basis in fact. The legislature could

reasonably have found that giving public safety employees expanded bargaining rights would discourage them from engaging in strikes or

sick-outs. It is true that there have been no strikes of public employees in Iowa since PERA was enacted in 1974. But it is also true that until 2017 there had never been legislation substantially curtailing the collective bargaining rights of Iowa public employees. Iowa legislators in

2017 could consider what happened several years earlier in Wisconsin to see that labor unrests and strikes may result when legislative

amendments curtail public union bargaining rights. Wisconsin public

19

employees staged mass protests in 2011, occupying the rotunda of the state capitol with great media fanfare. See Walker, 705 F.3d at 642–43, 655; Wis. Right to Life State Political Action Comm. v. Barland, 664 F.3d 139, 144–45 (7th Cir. 2011) (discussing the political unrest occurring in Wisconsin leading up to and after the collective bargaining amendments). See generally Madison Teachers, Inc. v. Walker, 851 N.W.2d 337 (Wis. 2014) (rejecting union challenge to Wisconsin’s amended collective bargaining statute).

The district court correctly concluded that Iowa legislators could take note of that Wisconsin experience. The district court also carefully

considered Iowa’s history of labor peace in the broader context of national developments.

Even assuming a strike is improbable, reasonable legislators could also be rationally concerned that public employees who experience a reduction in collective bargaining rights will be more likely to experience low morale and labor unrest. Labor unrest short of a strike could reasonably be considered by legislators to contribute to instability in the public sector workforce. Other jurisdictions have experienced incidents of civil disobedience through sickouts by public employees and “Blue Flu” by law enforcement in response to less desirable terms and conditions of employment. . . . The State cites numerous news articles about police officers in New York City, Memphis, Tennessee, Selma, Alabama and East Orange, New Jersey calling in sick in large numbers in order to protest issues such as unsafe conditions, low pay, and lack of benefits.

We agree with the district court that legislative facts readily available to Iowa lawmakers support concerns that labor unrest among police could undermine public safety, if not through strikes, then through reduced initiative or “blue flu.” Historically, police officers in

other states have used strategies such as the blue flu to protest labor conditions and policy changes. See generally Baker v. City of Detroit, 483

20

F. Supp. 930, 943 (E.D. Mich. 1979) (discussing white officers’ use of a blue flu and ticket strike to oppose the police department’s desegregation efforts in 1959–60); 30 Richard A. Lord, Williston on Contracts § 77:92, at 578 (4th ed. 2004) (“The length of a work stoppage may be temporary, as in the case of work slow downs, increasing call-ins by ‘sick’ workers or ‘blue flu,’ or a strike may persist for years.”); Illya Lichtenberg, Police Discretion and Traffic Enforcement: A Government of Men?, 50 Clev. St. L. Rev. 425, 444–45 (2003) (discussing how police officers have used

the blue flu as a bargaining strategy). The district court aptly observed that in Iowa “[t]he potential for ‘Blue Flu’ or some other exhibition of

labor unrest short of a strike is realistically conceivable.”

Against that backdrop, Iowa legislators in 2017 could rationally decide to extend more beneficial negotiating rights to bargaining units comprised of at least thirty percent public safety employees. The public safety rationale need not be voiced during the floor debates over House File 291 or proven with evidence. See Qwest Corp., 829 N.W.2d at 560. Iowa legislators individually and collectively can have multiple or mixed motives. Courts applying rational basis review do not take testimony from senators or representatives. See Rhoades v. State, 880 N.W.2d 431, 447 (Iowa 2016) (“In considering the statute in its full context, we do not give weight to the affidavit submitted by Rhoades from a former state legislator. On occasion, we have stated that a court may consider affidavits from legislators describing the factual background of legislation. We have consistently, however, held that affidavits from legislators or former legislators are inadmissible on the subject of

legislative intent. We do not depart from our established precedent in this case.” (Citations omitted.)).

21

The Fourth Circuit, in rejecting a constitutional challenge to legislation ending payroll deductions for union dues, aptly quoted Justice Scalia’s warning against efforts to ascertain the subjective intent of a group of legislators.

[D]iscerning the subjective motivation of those enacting the statute is, to be honest, almost always an impossible task.

The number of possible motivations, to begin with, is not binary, or indeed finite. . . . [The legislator] may have thought the bill would provide jobs for his district, or may have wanted to make amends with a faction of his party he had alienated on another vote, or he may have been a close friend of the bill’s sponsor, or he may have been repaying a favor he owed the Majority Leader, or he may have hoped the Governor would appreciate his vote and make a fundraising appearance for him, or he may have been pressured to vote for a bill he disliked by a wealthy contributor or by a flood of constituent mail, or he may have been seeking favorable publicity, or he may have been reluctant to hurt the feelings of a loyal staff member who worked on the bill, or he may have been settling an old score with a legislator who opposed the bill, or he may have been mad at his wife who opposed the bill, or he may have been intoxicated and utterly unmotivated when the vote was called, or he may have accidentally voted “yes” instead of “no,” or, of course, he may have had (and very likely did have) a combination of some of the above and many other motivations. To look for the sole purpose of even a single legislator is probably to look for something that does not exist.

S.C. Educ. Ass’n v. Campbell, 883 F.2d 1251, 1261 (4th Cir. 1989) (alteration in original) (quoting Edwards v. Aguillard, 482 U.S. 578, 636– 37, 107 S. Ct. 2573, 2605 (1987) (Scalia, J., dissenting)).

Our rational basis review is purposefully limited and does not include evidentiary fact-finding on the motives of individual legislators or validity of the labor peace rationale. Qwest Corp., 829 N.W.2d at 560.

A State, moreover, has no obligation to produce evidence to sustain the rationality of a statutory classification. “[A] legislative choice is not subject to courtroom factfinding and may be based on rational speculation unsupported by evidence or empirical data.” A statute is presumed constitutional and “[t]he burden is on the one attacking the legislative arrangement to negative

22

every conceivable basis which might support it,” whether or not the basis has a foundation in the record. Finally, courts are compelled under rational-basis review to accept a legislature’s generalizations even when there is an imperfect fit between means and ends. A classification does not fail rational-basis review because it “is not made with mathematical nicety or because in practice it results in some inequality.” “The problems of government are practical ones and may justify, if they do not require, rough accommodations—illogical, it may be, and unscientific.”

Baker, 867 N.W.2d at 57 (alterations in original) (quoting Heller, 509 U.S. at 319–21, 113 S. Ct. at 2643); see also FCC v. Beach Commc’ns, Inc., 508 U.S. 307, 315, 113 S. Ct. 2096, 2102 (1993) ([B]ecause we never require a legislature to articulate its reasons for enacting a statute, it is entirely irrelevant for constitutional purposes whether the conceived reason for the challenged distinction actually motivated the legislature. . . . ‘Only by faithful adherence to this guiding principle of judicial review of legislation is it possible to preserve to the legislative branch its rightful independence and its ability to function.’ ” (quoting Lehnhausen v. Lake Shore Auto Parts Co., 410 U.S. 356, 365, 93 S. Ct. 1001, 1006 (1973)); see also RACI, 675 N.W.2d at 7 n.3 (discussing what makes a justification plausible and credible as opposed to merely

specious).

We agree with the district court’s well-reasoned ruling applying our

court’s three-part test.

Potential liability resulting from a reduction of public sector collective bargaining rights could reasonably create a rationally credible concern regarding the effectiveness of law enforcement in the event of an emergency. A reasonable legislature could rationally conclude it is necessary to preserve the rights of collective bargaining units of at least thirty percent Public Safety Employees [(PSE)] in order to preserve a reliable corps of law enforcement authorities to deal with emergencies. The legislature could rationally establish as a priority the preservation of a satisfied, welltrained and experienced corps of [PSE]. The purpose of the classification is realistically conceivable. It has a credible

23

basis in fact. The relationship between the classification and the purpose of retaining a stable public safety force is not so weak as to be arbitrary.

Applying the rigorous rational basis test of [article I, section 6] of the Iowa Constitution, the Court concludes AFSCME failed to negate every reasonable basis for the classification that might support disparate treatment between units thirty percent or more PSEs and units of less than thirty percent PSEs. There is a rational basis for this legislative classification. The presumption of constitutionality prevails. H.F. 291 does not violate the Equal Protection Clause of the Iowa Constitution.

(Citations omitted.) We affirm on that basis. We turn next to the health

and safety rationale.

2. Health and safety rationale. The main rationale advanced

during the legislative debates on House File 291 centered on the health and safety risks that public safety employees face on the job. Because of these risks, legislators determined that public safety employees should retain broader bargaining rights, including on topics directly relating to their health and safety, such as insurance. This rationale was not reached in the district court ruling, but provides another ground for upholding the classifications in House File 291.

We note that on November 2, 2016, just over three months before the enactment of House File 291, two police officers were fatally shot in their squad cars in Des Moines and Urbandale, respectively.6 And the preceding summer, five police officers were gunned down in Dallas, Texas,7 and another three officers were shot dead two weeks later in

6KathyA. Bolten, Police ‘Heartbroken’ After Ambush Leaves 2 Des Moines-Area Officers Dead, Des Moines Reg. (last updated Nov. 3, 2016, 1:23 PM), https://www.desmoinesregister.com/story/news/2016/11/02/2-police-officers-killedambush -attacks/93155012/ [https://perma.cc/GC2R-ESTM].

7Manny Fernandez et al., Five Dallas Officers Were Killed as Payback, Police Chief Says, N.Y. Times (July 8, 2016), https://www.nytimes.com/2016/07/09/ us/dallas-police-shooting.html [https://perma.cc/45T9-2FEF].

24

Baton Rouge, Louisiana.8 These legislative facts provided vivid reminders to the Iowa General Assembly of the dangers police face on the job. While this appeal was under submission, a firefighter died in the line of duty in Clinton, Iowa, with another firefighter seriously injured.9 It is inarguable that the legislature could rationally conclude public safety employees face significantly greater risks to their health and safety than other public employees.

The plaintiffs argue House File 291 is impermissibly overinclusive

and underinclusive and denies the expanded collective bargaining rights to many public safety employees who belong to a bargaining unit

comprised of less than thirty percent public safety employees. The plaintiffs further argue that some public employees with public safety functions, such as corrections officers, university police, DOT road safety workers, and psychiatric aids, lack expanded bargaining rights under House File 291. The plaintiffs also argue that House File 291 arbitrarily grants certain public employees expanded bargaining rights because they are part of a public safety bargaining unit, even if they are not public safety employees. Finally, the plaintiffs argue that the topics over which public safety employee units are able to bargain is not limited solely to health and safety issues.

The defendants rely on the Seventh Circuit decision that rejected an equivalent challenge.

8Alan Blinder, The 3 Officers Killed in Baton Rouge, N.Y. Times (July 18, 2016), https://www.nytimes.com/2016/07/19/us/the-3-officers-killed-in-baton-rouge.html [https://perma.cc/8DXG-F8JJ].

9Thomas Geyer & Amanda Hancock, Clinton Firefighter Killed, Another Seriously Injured While Battling Fire Saturday, Quad City Times (Jan. 5, 2019), https://qctimes.com/news/local/clinton-firefighter-killed-another-seriously-injuredwhile -battling-fire-saturday/article_9816ef7d-415c-531b-b258-497f14a09a8b.html [https://perma.cc/K2HY-XFFL].

25

[W]e cannot, as the Unions request, determine precisely which occupations would jeopardize public safety with a strike. Even if we accept that Wisconsin imprudently characterized motor vehicle inspectors as public safety employees or the Capitol Police as general employees, invalidating the legislation on that ground would elevate the judiciary to the impermissible role of supra-legislature. . . .

Distinguishing between public safety unions and general employee unions may have been a poor choice, but it is not unconstitutional.

Walker, 705 F.3d at 656. The Seventh Circuit specifically considered the omission of correctional officers, explaining,

Even if we agree with the Unions that Act 10 should have placed prison guards in the public safety category, “a legislature need not run the risk of losing an entire remedial scheme simply because it failed, through inadvertence or otherwise, to cover every evil that might conceivably have been attacked.”

Id. at n.11 (quoting McDonald v. Bd. of Election Comm’rs, 394 U.S. 802, 809, 89 S. Ct. 1404, 1409 (1969)).

3. The thirty percent threshold. In our view, the foregoing authorities make clear the Iowa Constitution permits the State to treat public safety employees differently from other public employees and to treat bargaining units comprised of at least thirty percent public safety

employees better than bargaining units with a smaller percentage. The plaintiffs nevertheless argue that the thirty percent threshold itself is

unconstitutional, even if the labor peace and public safety rationales would permit preferential treatment of a bargaining unit comprised solely of public safety employees. The State responds that the legislative classifications reflect the current reality that local government bargaining units in Iowa happen to be comprised of a mix of public safety employees

and other employees. The State notes it would be impractical to segregate for collective bargaining purposes public safety employees and

other employees with different employers and unions. The plaintiffs fail

26

to persuasively rebut that State’s showing of the practical problems with interunit collective bargaining. Nor do the plaintiffs suggest a different, higher threshold that concededly passes constitutional muster. Ten percent? Forty percent? Ninety percent? Perhaps in plaintiffs’ view preferential treatment can be allowed only for public safety employees isolated in their own bargaining unit with no one else, as in Wisconsin.10 It is not the court’s role under our separation of powers to redraw the legislature’s chosen thirty percent threshold. The Seventh Circuit

recognized that “[d]efining the class of persons subject to a regulatory requirement . . . requires that some persons who have an almost equally

strong claim to favored treatment be placed on different sides of the line . . . [and this] is a matter for legislative, rather than judicial, consideration.” Walker, 705 F.3d at 655 (alterations in original) (quoting Beach Commc’ns, Inc., 508 U.S. at 315–16, 113 S. Ct. at 2102). We reiterate that “[f]or legislation to be violative of the Iowa Constitution under the rational basis test, the classification must involve ‘extreme degrees of overinclusion and underinclusion in relation to any particular goal.’ ” Ames Rental Prop. Ass’n v. City of Ames, 736 N.W.2d 255, 260 (Iowa 2007) (quoting RACI, 675 N.W.2d at 10). We hold that the thirty percent threshold is not so extremely overinclusive or underinclusive as to flunk our deferential rational basis review.

10Iowa public safety employees now in bargaining units below the thirty percent threshold may be able to vote with their feet and reorganize into a new bargaining unit to attain the preferential bargaining rights, as AFSCME’s counsel acknowledged at oral argument.

Unlike Iowa, the Wisconsin statutory scheme separates public safety employees by occupation into their own statewide bargaining unit. See Wis. Stat. Ann. § 111.825 (West, Westlaw current through 2017 Act 370). The Wisconsin legislature was not confronted with bargaining units comprised of public safety employees and other employees.

27

We decline to second-guess the legislature’s constitutional policy choices. We conclude that the plaintiffs have failed to meet their burden of negating every conceivable basis upon which House File 291 could be upheld. For that reason, plaintiffs’ equal protection challenge fails.

B. Plaintiffs’ Freedom of Association Claim. The plaintiffs claim that House File 291 unconstitutionally infringes on their right to associate with AFSCME. The plaintiffs argue that associating with the union of their choice is a fundamental right under the First Amendment

and House File 291 is therefore subject to strict scrutiny. The district court correctly rejected this claim.

“The First Amendment embodies the freedom of association, the right to ‘enter into and maintain certain intimate human relationships [without] undue intrusion by the State.’ ” Baker, 867 N.W.2d at 52 (alteration in original) (quoting Roberts v. U.S. Jaycees, 468 U.S. 609, 617–18, 104 S. Ct. 3244, 3249 (1984)); see also Sioux City Police Officers’ Ass’n v. City of Sioux City, 495 N.W.2d 687, 697 (Iowa 1993) (discussing freedom of association). Strict scrutiny applies when a suspect classification or fundamental right is involved. King, 818 N.W.2d at 31. As the United States Supreme Court noted in Roberts,

[T]he constitutional shelter afforded such relationships reflects the realization that individuals draw much of their emotional enrichment from close ties with others. Protecting these relationships from unwarranted state interference therefore safeguards the ability independently to define one’s identity that is central to any concept of liberty.

Roberts, 468 U.S. at 619, 104 S. Ct. at 3250.

Plaintiffs argue that public employees have a right to organize and

join labor unions. See State v. Keul, 233 Iowa 852, 855, 5 N.W.2d 849, 852 (1942) (“The right to form labor unions and by lawful means to act in

furtherance of their legitimate purposes is not open to question.”). The

28

plaintiffs note that “[w]e have traditionally followed the U.S. Supreme Court’s guidance in determining which rights are deemed fundamental.” King, 818 N.W.2d at 26. The plaintiffs also note that the United States Supreme Court has found that the right to join a union is a protected associational right. Bhd. of R.R. Trainmen v. Virginia ex rel. Va. State Bar, 377 U.S. 1, 8, 84 S. Ct. 1113, 1118 (1964) (“[T]he Constitution protects the associational rights of the members of the union precisely as it does those of the NAACP.”); see also Am. Fed’n of State, Cty., & Mun. Emps. v.

Woodward, 406 F.2d 137, 139 (8th Cir. 1969) (“Union membership is protected by the right of association under the First and Fourteenth

Amendments.”).

The plaintiffs argue that House File 291 both intentionally and incidentally infringes on their fundamental right to associate with a union of their choice. AFSCME represents a number of state peace officers and firefighters excluded from the definition of public safety employee. Further, all AFSCME-represented bargaining units are comprised of less than thirty percent public safety employees and thus are excluded from bargaining over any matter other than base wage. For that reason, plaintiffs argue House File 291 infringes on a fundamental right and is subject to strict scrutiny review. Under strict scrutiny, “the statute will survive a constitutional challenge only if it is shown that the statute is narrowly drawn to serve a compelling state interest.” City of Maquoketa v. Russell, 484 N.W.2d 179, 184 (Iowa 1992) (quoting City of Panora v. Simmons, 445 N.W.2d 363, 367 (Iowa 1989)).

We reiterate that the scope of collective bargaining rights of public

employees “is a matter for the legislature, not the courts.” State Bd. of Regents v. United Packing House Food & Allied Workers, Local No. 1258,

175 N.W.2d 110, 113 (Iowa 1970); see also Bennett v. City of Redfield,

29

446 N.W.2d 467, 473 (Iowa 1989) (“The right to public employment is not a fundamental right.”). House File 291 does not prohibit or restrict unions from soliciting members, disseminating materials, engaging in political activities, or expressing their views. As the State argues, “There is a fundamental distinction between the right to associate and whether someone must listen when you do. Declining to collectively bargain over certain topics does not inhibit the ability to associate.” We agree and apply rational basis review to this challenge. Nothing in House File 291

prohibits public employees from joining AFSCME or any other union.

The Fourth Circuit rejected a claim that legislation ending payroll

deductions for union dues violated associational rights. S.C. Educ. Ass’n, 883 F.2d at 1263–64. That court noted that the

legislation does not prohibit, regulate, or restrict the right of the [union] or any other organization to associate, to solicit members, to express its views, to publish or disseminate material, to engage in political activities, or to affiliate or cooperate with other groups.

Id. at 1256. The same is true of House File 291.

The plaintiffs charge that House File 291 “red circles” AFSCME-

represented bargaining units.11 The plaintiffs contend this targeting was

artful and capricious and performed with “scalpel-like precision” to specifically target AFSCME. The district court expressly “reject[ed]

Plaintiffs’ contention that [House File 291] ‘red circles’ AFSCME bargaining units or impinges on freedom of association with AFSCME.”

11The plaintiffs have not alleged that House File 291 is an unconstitutional bill of attainder. See U.S. Const. art. I, § 9, cl. 3 (“No Bill of Attainder . . . shall be passed”); Iowa Const. art. I, § 21 (“No bill of attainder . . . shall ever be passed.”); Atwood v. Vilsack, 725 N.W.2d 641, 651 (Iowa 2006) (“A bill of attainder ‘is a legislative determination that metes out punishment to a particular individual or a designated group of persons without a judicial trial.’ ” (quoting State v. Phillips, 610 N.W.2d 840, 843 (Iowa 2000))).

30

The text of House File 291 is facially neutral. The plaintiffs offered no evidence that the thirty percent threshold was chosen to target AFSCME. As we have explained above, House File 291 survives rational basis review. House File 291 was enacted within the power of the general assembly. Accordingly, we will not inquire into the subjective motives of individual legislators, regardless of whether political payback inspired some of them.

The Sixth Circuit, in rejecting a constitutional challenge to a

Michigan statute prohibiting public school employee payroll deductions for union dues, expressly declined to “ ‘peer[] past’ the [statutory] text . . .

‘to infer some invidious legislative intention.’ ” Bailey v. Callaghan, 715 F.3d 956, 960 (6th Cir. 2013) (quoting Walker, 705 F.3d at 649–50)). The Sixth Circuit concluded it was bound by the “familiar principle of constitutional law that this Court will not strike down an otherwise constitutional statute on the basis of an alleged illicit legislative motive.” Id. (quoting United States v. O’Brien, 391 U.S. 367, 383, 88 S. Ct. 1673, 1682 (1968)); see also In re Hubbard, 803 F.3d 1298, 1312 (11th Cir. 2015) (same). We too apply the O’Brien principle and decline to consider alleged motives to red circle AFSCME.

In upholding the Wisconsin enactment more broadly curtailing public employee collective bargaining rights, the Seventh Circuit squarely addressed the claim legislators were motivated by politic payback.

As unfortunate as it may be, political favoritism is a frequent aspect of legislative action. We said as much in Hearne v. Board of Education, 185 F.3d 770, 775 (7th Cir.

1999). There, members of the Chicago Teachers Union challenged on various constitutional grounds, including the Equal Protection Clause, an act of the Republican-dominated legislature that severely curtailed Chicago teachers’ job security relative to teachers in other parts of the state. Id. at 773. The unions argued, in part, that the Republican legislature retaliated against them for opposing Republicans

31

in the previous election. Id. We candidly remarked, “there is no rule whereby legislation that otherwise passes the proper level of scrutiny . . . becomes constitutionally defective because one of the reasons the legislators voted for it was to punish those who opposed them during an election campaign.” Id. at 775. We went further stating, “[i]ndeed one might think that this is what election campaigns are all about: candidates run a certain platform, political promises made in the campaign are kept (sometimes), and the winners get to write the laws.” Id. These sorts of decisions are left for the next election. Accordingly, we must resist the temptation to search for the legislature’s motivation for the Act’s classifications.

Walker, 705 F.3d at 654. We likewise decline to weigh the subjective motivations of legislators in our rational basis review under the Iowa Constitution. See Qwest Corp., 829 N.W.2d at 559–60; see also Rhoades, 880 N.W.2d at 447 (declining to give weight to legislator’s affidavit).

The 2017 amendments do not infringe on a fundamental right of association. The plaintiffs “come to us with a problem suitable only for political solution.” See Brown v. City of Lake Geneva, 919 F.2d 1299, 1304 (7th Cir. 1990). The plaintiffs are free to attempt to persuade public employers, such as the State and local governments and school boards, to voluntarily bargain over formerly mandatory terms. The

plaintiffs otherwise must look to the ballot box and the elected branches to change this lawfully enacted statute.

IV. Conclusion.

For these reasons, we affirm the district court’s summary judgment in favor of the State and PERB.

AFFIRMED.

Mansfield, Christensen, and McDonald, JJ., join this opinion.

Cady, C.J., files a dissenting opinion in which Wiggins, J., joins.

Appel, J., files a separate dissenting opinion in which Cady, C.J., and Wiggins, J., join.

32

#17–1841, AFSCME v. State CADY, Chief Justice (dissenting).

I respectfully dissent and join the dissent by Justice Appel. I write separately to emphasize the important role of courts and how a statute that treats people differently must not only have a rational basis, but one that fits the statute. See Varnum v. Brien, 763 N.W.2d 862, 879 (Iowa 2009) (“[T]he deference built into the rational basis test is not dispositive because this court engages in a meaningful review of all legislation challenged on equal protection grounds by applying the rational basis test to the facts of each case.”).

The legislation at issue gives expansive collective bargaining rights to public safety employees, but very limited collective bargaining rights to other public employees. Our constitutional doctrine of equal protection recognizes that most all laws tend to discriminate in some way, so the focus of the analysis turns on whether the unequal treatment is properly justified. See NextEra Energy Res. LLC v. Iowa Utils. Bd., 815 N.W.2d 30, 46 (Iowa 2012) (explaining a classification resulting in some inequality does not necessarily violate equal protection). If it is, the constitutional mandate of equal protection is not violated. Id.

The majority opinion finds a rational basis to justify the disparate treatment in this case from the special need to protect the public against

the potential harm of labor unrest by public safety employees and to give special protection to public safety employees from the health and safety risks they face on the job. The premise is that public safety is a vital concern in Iowa and this concern supports special laws that give greater

bargaining rights to public safety employees than other public employees to help keep them on the job, instead of engaging in strikes or becoming

injured and unable to perform their jobs.

33

I agree with the conclusion in the majority opinion that the rational reasons identified are an adequate justification for disparate collective bargaining treatment among public employees. I also agree it is not the role of courts to find criticism of public policy based on disagreement over policy. Any such form of criticism, even implicit, has no place in the analysis by courts. Instead, the only role of the courts in the process is to decide if the discrimination is justified under the facts and circumstances.

In this case, the legislation offends our constitution. The problem with the law is not its purpose or justification to discriminate, but how

the general assembly failed to apply this purpose in articulating the law. Instead of treating public employees differently by dividing them into one group of public safety employees and another group of other public employees, the general assembly passed a law giving different rights to public employees based on their membership in a collective bargaining agreement. The problem is that bargaining units in Iowa contain both public safety employees and other public employees. Thus, while the law purported to put public safety employees in a separate class based on a valid purpose, it created classifications by using bargaining units and permitted the bargaining units to contain up to seventy percent of persons who are not public safety employees. This means the statute enacted ended up giving many public employees rights of public safety employees and denied many public safety employees those rights.

This type of line drawing falls far too short of our constitution’s demands. While line drawing can never be clean and can present a

variety of obstacles, this case is not even close to a fair delineation. Moreover, there is simply no reason why the general assembly could not

have drawn the lines to eliminate the unconstitutional distinctions. The

34

law cannot purport to give needed special protection to one group of people and then allow that group to be populated by up to seventy percent of other people not included within the purpose. This approach is a bad fit and destroys the justification for the law.

If the line drawing needed to accomplish the stated purpose in this case were difficult to do, as it can be in some cases, leeway would exist. But, in truth, there is no reason it cannot be done in this case. Constitutional lines are clearly available. A collective bargaining statute

in Wisconsin is one such example. See Wis. Educ. Ass’n Council v. Walker, 705 F.3d 640, 642–43 (7th Cir. 2013). Here, the line required by

the constitution simply was not drawn. Id.

The majority opinion relies on Walker as its authority to support the constitutionality of the Iowa statute. That reliance is misplaced. Walker involved a Wisconsin statute that gave broader collective bargaining agreements to those public employees designated as public safety employees based on the same rationale used in this case. Id. at 655. Yet, unlike the Iowa statute, the Wisconsin statute divided public safety employees and other public employees into two separate groups. Id. at 642–43. Unlike the Iowa statute, the statute did not use percentages of public safety employees within a bargaining unit to allow for such dramatic overlap. Thus, the purpose of the statute fit the categories drawn, and this fit allowed the legislation to pass constitutional scrutiny. Id. at 654–57.

Accordingly, the Walker decision does not support the constitutionality of the statute in this case, and the majority opinion

offers little more analysis. It only relies on the justification to discriminate and ignores the vast overinclusiveness and

underinclusiveness of classifying employees based on membership in

35

bargaining units. As a result, the Iowa statute ends up treating many similarly situated public employees in Iowa differently based solely on the bargaining unit they belong to and not for the reason the constitution would justify different treatment of public employees. Our constitution requires laws to treat similarly situated people equally unless there is an adequate reason otherwise. In this case, the overinclusiveness and underinclusiveness written into the statute drowned this reason out. Our constitutional form of government depends on courts to see it and

demand better.

Wiggins, J., joins this dissent.

36

#17–1841, AFSCME v. State APPEL, Justice (dissenting).

I. Introduction.

House File 291 is an odd statute. See 2017 Iowa Acts ch. 2, §§ 1, 6 (codified at Iowa Code §§ 20.3(11), .9 (2018)). It slices and dices the universe of public employees entitled to collective bargaining by various categories in multiple novel ways that are overinclusive and underinclusive. One of the two ostensible purposes—labor peace—is

advanced even though severe sanctions for striking have been in place for forty years, and during that period, there has never been a strike by

any public employees. Further, the means chosen by the legislature—an arbitrary grouping and shuffling of public employees that is overinclusive and underinclusive—bears no rational relation to either labor peace or promoting the health and safety of public employees exposed to danger in their jobs.

In my view, therefore, the law does not survive rational basis review under article I, section 6 of the Iowa Constitution. Accordingly, I respectfully dissent.

II. Overview of House File 291.

I begin with a discussion of the remarkable classification system created by the law. It identifies an oddball group of public employees and throws them into the burlap grab bag labeled “public safety employee[s].” Id. § 1 (codified at Iowa Code § 20.3(11)). Then, some of those within the grab bag are denied privileges that others receive. Id. § 6. And some public employees not within the grab bag receive the

benefits denied to a portion of public safety employees, while others do not. Id. Perplexing, I know. The classification system is illogical.

37

The identification of public safety employees is made not on the basis of an employee’s duties or functions, but rather by the title an employee holds. Id. § 1 (codified at Iowa Code § 20.3(11). In some respects, the “public safety” grab bag is astonishingly inclusive. The grab bag was stretched astoundingly wide. It accommodates park rangers, gaming enforcement officers, and peace officers designated by the department of transportation. Id.

But then, it excludes employees with obvious public safety

responsibilities. The grab bag has no room for university police who, just like other police officers, are law enforcement officers pursuant to Iowa

Code chapter 80B, are trained and certified by the Iowa Law Enforcement Academy, and engage in law enforcement and emergency response alongside other city police officers. Id. Airport firefighters are excluded even though they too work alongside the firefighters designated as public safety employees by House File 291. Id. The law also excludes others, like parole officers and fraud bureau investigators, who work in unpredictable environments with broad arrest powers and the obligation to respond to emergencies. Id. And none of our state’s corrections officers, jailers, and emergency medical service providers are considered public safety employees. Id. Yet all of those public employees work in “protection occupations.” Iowa Code § 97B.49B(1)(e).

As is evident, the statutory classification of public safety employees is obviously remarkably overinclusive and underinclusive. No one questions that. And no one questions that the overinclusiveness and underinclusiveness are among the features that make the classifications

in House File 291 suspect.

I think it very doubtful that the classification of public safety

employees makes much sense, but we are not done with the

38

irrationalities of the statute. This is because classification as a public safety employee does not even determine whether a public employee gets the benefits of collective bargaining granted to some and denied others.

Under House File 291, full collective bargaining rights are provided to all state employees in bargaining units with at least thirty percent public safety employees. Id. § 6 (codified at Iowa Code § 20.9). In such a unit, the thirty percent public safety employees get the benefit of representatives who have the ability to collectively bargain fully over

wages and a list of other terms and conditions of employment. Id. But so do the remaining seventy percent of non-“public safety employees” in

the bargaining unit. Id. Thus, in a unit with thirty percent public safety employees, a supermajority of the beneficiaries are not public safety employees! If the law is designed to target public safety employees for preferential treatment, it is way, way overbroad.

And yet, once again, it is also underinclusive. What about a bargaining unit with twenty-nine percent public safety employees? Those public safety employees, along with their colleagues in the bargaining unit, are out in the cold. See id. For instance, we are told that the police officers serving the City of Guttenberg, along with the police and fire departments of the City of Decorah, are all relegated to disfavored status under House File 291. So too are the deputy sheriffs of Humboldt County, even though the county sheriffs and Humboldt police officers serve side-by-side and share equipment.

Thus, the set of public safety employees benefiting from the statute is doubly underinclusive. The definition of “public safety employees” is

underinclusive, and then only some public safety employees are doled out benefits based solely on whether the employee happens to fall within

39

a given type of bargaining unit. Id. §§ 1, 6 (codified at Iowa Code §§ 20.3(11), .9).

What kind of statute is this? Notably, the parties have failed to identify a similar statute anywhere at any time. House File 291 is unlike the recent legislation passed in Wisconsin because, under the Wisconsin law, all those designated as public safety employees receive broader collective bargaining rights and all those who are not so designated do not receive those rights. See Wis. Educ. Ass’n Council v. Walker, 705

F.3d 640, 642–43 (7th Cir. 2013). Make no mistake, House File 291 is really odd.

III. Framework for Rational Basis Review.

Our approach to rational basis review is well-established. See LSCP, LLLP v. Kay-Decker, 861 N.W.2d 846, 859 (Iowa 2015); Racing Ass’n of Cent. Iowa v. Fitzgerald (RACI), 675 N.W.2d 1, 7–8 (Iowa 2004). In general terms, “to pass the rational basis test, the statute must be ‘rationally related to a legitimate state interest.’ ” LSCP, 861 N.W.2d at 858 (quoting Qwest Corp. v. Iowa State Bd. of Tax Review, 829 N.W.2d 550, 558 (Iowa 2013)). In undertaking our analysis, we employ a three- part test.

First, we identify the classes of similarly situated persons treated differently. LSCP, 861 N.W.2d at 859. This is a threshold determination. Id. We do not make intricate distinctions between purported classes of similarly situated individuals, as “[n]o two groups are identical in every way.” Qwest, 829 N.W.2d at 561.

Next, we “examine the legitimacy of the end to be achieved.” LSCP,

861 N.W.2d at 860 (quoting Fed. Land Bank of Omaha v. Arnold, 426 N.W.2d 153, 156 (Iowa 1988)). We consider “whether there was a valid,

‘realistically conceivable’ purpose that served a legitimate government

40

interest.” Residential & Agric. Advisory Comm., LLC v. Dyersville City Council, 888 N.W.2d 24, 50 (Iowa 2016) (quoting McQuistion v. City of Clinton, 872 N.W.2d 817, 831 (Iowa 2015)). “A legitimate interest can be any reasonable justification, not just the one the legislature actually chose.”12 LSCP, 861 N.W.2d at 858.

Our review of the legitimacy of the end to be achieved is not toothless. Id. at 860. We consider whether the claimed state interest is “realistically conceivable” and “decide whether this reason has a basis in

fact.” Id. (quoting RACI, 675 N.W.2d at 7–8). “The ‘realistically conceivable’ standard requires more than ‘a purely superficial analysis

and implies that the court is permitted to “probe to determine if the constitutional requirement of some rationality in the nature of the class singled out has been met.” ’ ” Id. (quoting RACI, 675 N.W.2d at 7 n.3). “ ‘Basis in fact’ means ‘the court will undertake some examination of the credibility of the asserted factual basis for the challenged classification rather than simply accepting it at face value.’ ” Id. (quoting RACI, 675 N.W.2d at 8 n.4). “In other words, although ‘actual proof of an asserted justification [i]s not necessary, . . . the court w[ill] not simply accept it at face value and w[ill] examine it to determine whether it [i]s credible as opposed to specious.” Id. (alterations in original) (quoting Qwest, 829 N.W.2d at 560).

Third, we consider the relationship between the classification and the purpose of the classification. The fit between the means chosen by the legislature and its objective need not be perfect, but it must be rational. Id. at 859. “[W]e must consider whether the relationship 12Some states decline to consider hypothetical justifications in considering equal

protection claims brought under state constitutions. See State v. Russell, 477 N.W.2d 886, 889 (Minn. 1991). We need not consider the question here because the statute is infirm even under the hypothetical justifications advanced by the State.

41

between the classification . . . and the purpose of the classification is so weak that the classification must be viewed as arbitrary.” Id. at 860.

“Under the Iowa Constitution, we determine whether a classification rationally furthers a legitimate state interest by evaluating whether the classification features ‘extreme degrees of overinclusion and underinclusion in relation to any particular goal.’ ” Id. at 861 (quoting Bierkamp v. Rogers, 293 N.W.2d 577, 584 (Iowa 1980)). “If a classification involves extreme overinclusion or underinclusion ‘in

relation to any particular goal, it cannot [reasonably] be said to . . . further that goal.’ ” Id. (quoting Bierkamp, 293 N.W.2d at 584).

Of course, we have explained that rational basis review is a “very deferential standard.” NextEra Energy Res. LLC v. Iowa Utils. Bd., 815 N.W.2d 30, 46 (Iowa 2012) (quoting Varnum v. Brien, 763 N.W.2d 862, 878 (Iowa 2009)). “The plaintiff has the heavy burden of showing the statute unconstitutional and must negate every reasonable basis upon which the classification may be sustained.” Id. (quoting Bierkamp, 293 N.W.2d at 579–80). Further, “[a] classification ‘does not deny equal protection simply because in practice it results in some inequality; practical problems of government permit rough accommodations.’ ” Id. (alteration in original) (quoting In re Det. of Morrow, 616 N.W.2d 544, 548 (Iowa 2000)).

But “[t]he deference we afford the legislature’s classifications ‘is not, in and of itself, necessarily dispositive’ under article I, section 6.” LSCP, 861 N.W.2d at 859 (quoting Bierkamp, 293 N.W.2d at 581). Our “rigorous standards have not . . . prevented this court from finding

economic . . . legislation in violation of equal protection provisions.” RACI, 675 N.W.2d at 8–9. “[E]ven in the economic sphere, a citizen’s

guarantee of equal protection is violated if desirable legislative goals are

42

achieved . . . through wholly arbitrary classifications or otherwise invidious discrimination.” Fed. Land Bank, 426 N.W.2d at 156. “It is for the judicial department to determine whether any department has exceeded its constitutional functions . . . .” Luse v. Wray, 254 N.W.2d 324, 327 (Iowa 1977) (quoting 16 C.J.S. Constitutional Law § 144, at 688).

IV. Applying Iowa’s Rational Basis Review.

A. The Statutory Classifications. As described above, House

File 291 classifies public employees in multiple unusual ways. Among the public employees with safety responsibilities, it identifies some as

public safety employees and omits others. 2017 Iowa Acts ch. 2, § 1 (codified at Iowa Code § 20.3(11)). Then, it goes on to even omit some of the public safety employees from the benefits of broader collective bargaining while allowing large numbers of those not branded as public safety employees to benefit from broader collective bargaining. Id. § 6 (codified at Iowa Code § 20.9).

Therefore, House File 291 treats many similarly situated persons differently. First, some public employees with safety responsibilities— like university police, airport firefighters, corrections officers, jailers, and emergency medical service providers—are similarly situated to other public employees with safety responsibilities yet treated differently. See id. § 1 (codified at Iowa Code § 20.3(11)). Second, only some of the public employees that House File 291 itself considers similarly situated—public safety employees—are able to benefit from broader collective bargaining. Id. § 6 (codified at Iowa Code § 20.9). Third, among the public employees

that are not considered public safety employees by the law, some are able to engage in broader collective bargaining and others are not. Id.

43

B. Examining Ends. The State has suggested two purposes for House File 291—labor peace and the health and safety of public safety employees. Neither provides a basis for sustaining this statute.

I begin with the purported purpose of labor peace. First, the historical record is striking. No one claims that there has ever been a strike of any public employees, let alone public safety employees, since the enactment of the Public Employment Relations Act over forty years ago. Further, no one claims that such a strike has been seriously

threatened. The lack of any facts to support the asserted rationale is troubling.

Second, as plaintiffs point out, labor peace was not a rationale for the law asserted by any Iowa legislators during the floor debate. That is striking. If there was truly a risk to public safety that a strike by public safety employees would create, surely the legislators would have said so. The fact that avoiding strikes was not even mentioned in the debates further suggests a lack of basis in fact.

Third, for forty years, draconian sanctions have been in place in the event any public employee contemplated striking. The sanctions can include imprisonment for six months; daily individual fines of $500; daily union fines of $10,000; termination from employment and ineligibility for public employment for one year; decertification of union and one-year waiting period for recertification; injunctions; contempt; and “any other legal or equitable remedy or penalty.” Iowa Code § 20.12(3)–(6). The existence and effectiveness of the sanctions undermine the State’s argument that labor peace is a purpose of the discriminatory treatment

in House File 291. See, e.g., U.S. Dep’t of Agric. v. Moreno, 413 U.S. 528, 536–37, 93 S. Ct. 2821, 2827 (1973) (explaining that pre-existing

provisions addressing fraud “necessarily casts considerable doubt upon

44

the proposition that the 1971 amendment could rationally have been intended to prevent those very same abuses”).

In Iowa, we “examin[e] . . . the credibility of the asserted factual basis for the challenged classification rather than simply accepting it at face value.” LSCP, 861 N.W.2d at 860 (quoting RACI, 675 N.W.2d at 8 & n.4). Based on the history, the lack of justification in legislative debates, and the existence of strong sanctions already addressing the problem, I conclude that the labor peace rational fails that test.

As an alternative, the State generally claims that the health and safety of certain endangered public employees could be a legitimate end

for the law. No one questions the general proposition that promoting the health and safety of employees is a legitimate state interest. Of course, that generalization may be declared as supporting every statute. But to the extent House File 291 provides nebulous health and safety benefits apparently arising from robust collective bargaining rights, what is the rationale for denying those benefits to other public safety employees under the statute? It seems odd to suggest that some public safety employees are entitled to the health and safety benefits afforded by robust collective bargaining and benefits and others are not. Why, say, are park rankers entitled to the health and safety benefits of robust collective bargaining while corrections officers are not? While health and safety benefits may justify robust collective bargaining rights, that benefit is equally applicable to the excluded public safety employees.

C. Examining Means. I now turn to the question of whether there is a rational relationship between the purported goals of the statute

and the means chosen by the legislature. For the reasons expressed below, I find it hard to see a rational relationship between the means

chosen and the ends asserted.

45

If labor peace were the goal, why aren’t corrections staff, or parole officers, or university police officers, or healthcare workers, provided the benefits of the statute? Other states deal with the potential of strikes in inclusive ways. See, e.g., Cty. Sanitation Dist. No. 2 v. L.A. Cty. Emps. Ass’n, 699 P.2d 835, 846 & n.26 (Cal. 1985) (en banc) (noting that ten states permit public employees to strike “unless such strikes endanger the public health, safety, or welfare” and explaining that “[t]he statutes generally prohibit strikes by police and fire-protection employees,

employees in correctional facilities, and those in health-care institutions”). If House File 291 is designed to prevent strikes that would

jeopardize public safety, it is remarkably underinclusive.

Conversely, is there anything in the record suggesting that public safety employees included in the House File 291 grab bag have threatened to strike? And if they have, have they threatened to strike more frequently or more intensively than the corrections officials and university police? Is there anything in the record suggesting that a strike by gaming enforcement officers would be a threat to public safety? And could it be of the same magnitude as a strike by the many police and fire departments left out in the cold by the thirty percent threshold? Moreover, is the danger of a strike by non-“public safety employees” in a favored bargaining unit somehow of such concern that they, too, need special bargaining rights? These questions, of course, must be answered in the negative, and reveal the arbitrariness and extreme overinclusion of the classifications if they are designed to ensure labor peace.

Most importantly, perhaps, is the absence of a rational connection

between doling benefits and preventing strikes. Does the record, or any legislative facts, show that only some public safety employees—i.e., those

in unions in which they comprise more than thirty percent of members—

46

need special benefits to convince them not to break the law and strike? Or that, unless they are doled out special benefits, police officers in those units (and I guess gaming enforcement officers, park rangers, and DOT officers) will refuse to do their duty in the face of others breaking the law and striking? Are public safety employees in units in which they comprise less than thirty percent of members somehow better able to resist lawbreaking? Or are those public safety employees less important to public safety?

The classifications in House File 291 are arbitrary if the goal was labor peace. There is no relationship between the classifications, which

feature extreme degrees of overinclusion and underinclusion, and labor peace. And the linchpin of the argument advanced in favor of the law— public safety employees might strike in the face of criminal penalty if they are not granted special status—is specious. In the face of such irrationality, I would not uphold the oddball classifications in House File 291.

I now turn to the question of whether the classifications in House File 291 may be supported as health and safety measures. In my view, the slicing and dicing in House File 291 bears no rational connection to protecting health and safety of public employees exposed to greater risks. In short, and as detailed below, the law confers privileges on some public employees and withholds them from others without regard to whether the persons actually face greater danger. It also does not consider whether the privileges are rationally related to protecting health and safety.

The legislature’s choice of who may be allowed greater collective bargaining rights is grievously underinclusive towards achieving a goal of

protecting health and safety of public employees exposed to danger. Why

47

omit university police officers, corrections officers, jailers, emergency medical service providers, airport firefighters, and others from the category of public safety employees? 2017 Iowa Acts ch. 2, § 1 (codified at Iowa Code § 20.3(11)). The record shows that many employees in these jobs, especially corrections officers and university police officers, face similar or greater risks than those classified as public safety employees. Psychiatric aides and medical technicians, according to the record, are approximately four times as likely to be injured on the job as

are police officers and approximately 150 times as likely to be injured on the job as firefighters. Meanwhile, the number of road safety workers

killed in Iowa exceeds the number of police killed in the line of duty.

Moreover, why prevent even some of the public safety employees from being able to attain the collective bargaining benefits that, purportedly, would protect their health and safety? Id. § 6 (codified at Iowa Code § 20.9). If the purpose was to protect the health and safety of public employees exposed to danger, or even just the health and safety of an anointed group of those public employees, the statute is fatally underinclusive.

The statute, of course, is also overinclusive if the goal is to protect health and safety of those exposed to greater danger. Why allow supermajorities of non-“public safety employees” to access those benefits? See id. There is no rational explanation.

Further, the limitations on bargaining applicable to the two groups are numerous, and in almost every respect, divorced from health and safety. For instance, House File 291 gives units with thirty percent

public safety employees greater rights than other units in arbitrating over wages. Id. What rational connection is there between a cap on the

wages that an arbitrator may award to some bargaining units and the

48

health and safety of a portion of the members of other bargaining units? There is none.

Similarly, House File 291 makes seven bargaining subjects mandatory in the case of units with thirty percent public safety employees and prohibited in the case of other units. These subjects are “insurance, leaves of absence for political activities, supplemental pay, transfer procedures, evaluation procedures, procedures for staff reduction, and subcontracting public services.” Id. Again, protecting the

health and safety of some members of some bargaining units does not rationally connect to giving all of the members of those units the

exclusive right to bargain over those subjects.

Indeed, House File 291 requires mandatory negotiation over seventeen specified subjects in collective bargaining with units comprised of at least thirty percent public safety employees. Id. Only one of those subjects is “health and safety matters.” Id. And for other unions, none of the seventeen subjects is a mandatory topic of negotiations. Id. The reason for giving the favored units such significantly greater collective bargaining rights is not rationally explainable by a desire to protect the health and safety of a portion of the benefited units.

The situation we face is not like that in Wisconsin. Unlike Wisconsin, as noted above, the Iowa law shuffles the public safety employees into some groups that are entitled to greater benefits and others which are not. Id. In Wisconsin, all public safety employees were entitled to greater collective bargaining rights. Wis. Educ. Ass’n Council, 705 F.3d at 642–43. Moreover, in Iowa, many non-“public safety

employees” are granted greater collective bargaining rights while others are not. 2017 Iowa Acts ch. 2, § 6 (codified at Iowa Code § 20.9). In

Wisconsin, no nonsafety employees were granted greater bargaining

49

rights. Wis. Educ. Ass’n Council, 705 F.3d at 642–43. Further, the federal court addressing the Wisconsin law did not consider whether the gamut of differential collective bargaining rights present in House File 291 bear any rational relation to health and safety. See id. at 656–57 (discussing recertification requirements and elimination of payroll deductions for all public employees). Finally, reliance on federal precedents ignores that, in Iowa, extreme overinclusion and underinclusion can render a statute so arbitrary as to fail the rational

basis test. LSCP, 861 N.W.2d at 859; RACI, 675 N.W.2d at 7–8.

It seems to me that the extreme overinclusiveness and

underinclusiveness of this statute is so striking that it does not pass constitutional muster under RACI principles.

V. Conclusion.

For the above reasons, I would reverse the judgment of the district court.

Cady, C.J., and Wiggins, J., join this dissent.