Arizona Voters Approve Major Overhaul of Public Safety Officers’ Pensions

On May 17, 2016, Arizona voters approved Proposition 124, a legislatively referred constitutional amendment affecting post-retirement benefit increases for participants in the Arizona Public Safety Personnel Retirement System (PSPRS). The voter initiative had little opposition—it passed by a 70-30% margin—and consisted of a minor-looking change to Article 29, § 1 of the Arizona Constitution, the so-called “Pension Clause.” The purpose of this constitutional change was to ensure the constitutionality of S.B. 1428, a piece of legislation signed into law in February 2016 that made significant changes to PSPRS, which covers state and local law enforcement personnel and firefighters.

But to understand Arizona’s Pension Clause and how it stymies pension reform—and why some constitutional change was necessary—we have to go back two years to February 2014 and examine an Arizona Supreme Court case called Fields v. Elected Officials’ Retirement Plan.

*     *     *

In 1985, Arizona created an Elected Officials’ Retirement Plan (EORP). At the time, there was no provision for officials’ pension benefits to increase over time—the legislature just passed ad hoc increases from time to time. In 1995, the legislature enacted a statute, § 38-818, which provided for automatic annual post-retirement benefit increases. But this statute was unlike the “cost-of-living adjustment” formulas that you can find in many pension plans: instead, the benefit increase each year was tied to the Plan’s return on its investments. In good years (i.e., if the Plan’s total investment return exceeded 9%), retired officials’ benefits increased according to a formula, up to a 4% cap. If there were still “excess investment earnings” left over after funding those benefit increases, the remainder was to be set aside for benefit increases in future (leaner) years.

Like many public-sector pensions, Arizona’s plan ran into mounting underfunding problems. As the Plan’s funding ratio declined between 2000 and 2010—from 142% to 67% (and as of 2015, it stood at 48% funded)—the legislature decided to be more careful about funding benefits. In 2011, it passed a new law, S.B. 1609, making several changes to the § 38-818 system. The leftover excess investment earnings were no longer to be set aside for future benefit increases, but were instead to be applied to fund the basic retirement plan. The 9% threshold was increased to 10.5% (which reduced the amount of funding available for automatic benefit increases). The amount of benefit increases in the future was tied to how well the Plan was funded: 4% increases were to be available if the Plan was over 80% funded, no increases if the Plan was under 60% funded.

Naturally, some people’s benefits were lower under the new S.B. 1609 scheme than they would otherwise have been. Kenneth Fields and Jefferson Lankford, retired Arizona judges, challenged S.B. 1609 on several grounds, on behalf of themselves and others similarly situated. They argued that the law violated the federal constitution’s Contract Clause, which provides that “[n]o state shall . . . pass any . . . Law impairing the Obligation of Contracts.” They also argued that it violated the Arizona constitution’s similar Contract Clause, which provides that “[n]o . . . law impairing the obligation of a contract[] shall ever be enacted.” And, finally, they argued that it violated the Arizona constitution’s Pension Clause—added in 1998—which provides that:

“[m]embership in a public retirement system is a contractual relationship that is subject to [Arizona’s Contract Clause], and public retirement system benefits shall not be diminished or impaired.”

The state and federal Contract Clauses look similar, and many states—including Arizona—interpret their own Contract Clauses similarly to the federal Clause. (I have discussed this issue in a previous post.) Though the Clauses seem fairly absolute, in reality courts apply a three-part test:

1. Does the challenged law substantially impair a contractual relationship?
2. If so, does the law have a significant and legitimate public purpose?
3. If so, is the impairment reasonable and appropriate to the public purpose?

The state of Arizona argued that S.B. 1609 passed all these steps.

But in ruling on this case in 2014, the Arizona Supreme Court sidestepped this whole three-part Contract Clause test. Instead, it relied exclusively on the Pension Clause. Take a look at the text of the Pension Clause above, and you’ll see that it explicitly references the Contract Clause, defines public pensions as contracts, and provides that they won’t be diminished or impaired—strikingly similar to the Contract Clause itself. One could easily take the position that the Pension Clause provides no additional protections.

Indeed, in Detroit’s recent bankruptcy procedure (which I have discussed in a previous post), the federal bankruptcy judge took exactly that position with respect to Michigan’s Pension and Contract Clauses. The Michigan Contract Clause states that “No . . . law impairing the obligation of contract shall be enacted.” The Michigan Pension Clause states that “The accrued financial benefits of each pension plan and retirement system of the state and its political subdivisions shall be a contractual obligation thereof which shall not be diminished or impaired thereby.” The judge held that the Pension Clause was no stronger than the Contract Clause, and went straight to the Contract Clause.

Not so in Arizona. The Arizona Supreme Court reasoned that if this were so, the Contract Clause (prohibition of “impair[ment]”) and the second half of the Pension Clause (prohibiting “diminish[ment]” or “impair[ment]”) would be redundant—and we should read legal language to avoid redundancy.

The “no redundancy” principle is often a useful rule of thumb, though it shouldn’t always be decisive. Sometimes constitutions say the same thing in two different ways, out of an excess of caution or because constitutions are also rhetorical documents. Not all repetitions should be read to have different and independent effect.

But that’s how the Arizona Supreme Court read the two Clauses. Therefore, it said, “the Pension Clause confers additional, independent protection for public retirement benefits separate and distinct from the protection afforded by the Contract Clause.” The effect of the ruling was that the old post-retirement benefit increase formula for retirees was restored.

*     *     *

That’s not all the Arizona Supreme Court did, though.

One could certainly sympathize (as I do) with a strong rule against abrogating contracts, and believe (as I do) that the current three-part Contract Clause test is too forgiving of governmental abridgments of contracts. (The Takings Clause requires fair compensation whenever property is taken, always, without exception, even when there’s a really good reason and exigent circumstances. Why not the same rule when governments want to abrogate their contracts?) But even if one holds these views, one could still argue that prospective changes in pensions, which retain all the benefits accrued so far, can never be an impairment of a pension benefit.

For instance, consider a hypothetical employee hired 10 years ago qualifying for the now-closed EORP defined benefit pension plan. If he retires today, his benefits will be equal to 40% of his salary (assuming the 4% benefit multiplier in effect for the EORP plan at the time, times years served). But suppose he doesn’t retire today, and instead works for many more years (accumulating, roughly, 4% of final salary for each additional year he works, up to a maximum of 80% of his average yearly salary). If the formula were changed after the tenth year, but only applied to his later earnings—so the first ten years’ 40% continued to be calculated under the original rules—it seems difficult to say that any benefits have been impaired. Some states, for instance Maryland, follow this sensible approach, under which any pension changes are constitutional as long as they operate only on benefits to be earned in the future. (Under this interpretation, retired judges like Fields and Lankford would be entitled to all the old rules, but current employees wouldn’t be entitled to the old rules on their new earnings.)

Consider the consequences of the opposite rule—one that would say you’re entitled to pension formulas that existed on the day you were hired. Everyone agrees that public employees don’t have a constitutional right to the continuation of their starting salary; the legislature can reduce public salaries down to minimum wage if they so choose. They don’t have a constitutional right to the quality of the working conditions that they experienced on day one; the legislature can steadily make their jobs more and more miserable, within the constraints of federal workplace regulation. They don’t have a constitutional right to the continued existence of their own job positions; the legislature can abolish those without notice. They don’t even have a constitutional right to continue working in their existing positions; any existing civil service protections can be abolished by the legislature. But God forbid the legislature should make any unfavorable prospective change in public employees’ pension calculation formulas: those are vested rights.

It thus seems to make little sense to interpret the “benefits” protected under the Pension Clause to include not only benefits based on work already done, but also the prospect of future benefits based on the rules that existed when one was hired. And yet, this is exactly what the Arizona Supreme Court held—similarly to many other states that use the so-called “California Rule.” (I have written about the California Rule in this recent blog post and this recent policy study.) The court held that an employee has “a right in the existing formula by which his benefits are calculated as of the time he began employment and any beneficial modifications made during the course of his employment.” (In doing so, it explicitly noted that the Arizona Pension Clause is worded more broadly than parallel clauses in other state constitutions—including in Michigan, whose Pension Clause, quoted above, states merely that “accrued financial benefits” are a contractual obligation.)

*     *     *

The Fields case has already had repercussions beyond Arizona. In May 2015, in In re Pension Reform Litigation, the Illinois Supreme Court struck down a major Illinois public-pension reform package on essentially the same grounds (Illinois’ Pension Clause provides greater protection than its Contract Clause, and formulas are protected benefits from day one), citing Fields favorably. And the Illinois Supreme Court reaffirmed that holding in March 2016, in Jones v. Municipal Employees’ Annuity and Benefit Fund of Chicago.

Having examined the Fields case in detail, we’re now in a position to understand the recent Arizona pension reform and constitutional amendment.

The new pension reform, S.B. 1428, was signed into law in February 2016 and applies to the Arizona PSPRS, which covers law enforcement personnel and firefighters. [Note: Reason Foundation was a key player in providing analysis and facilitating the negotiations that led to this compromise; for more details, see this analysis by Reason analysts Leonard Gilroy, Pete Constant, and Anthony Randazzo.]

S.B. 1428 has a number of interesting features. First, for current public safety employees and retirees, the existing formula for benefit increases (similar to the one described above for elected officials, based on investment returns) is replaced by an actual cost-of-living adjustment (COLA) tied to inflation, with a 2% cap—and, unlike in the previous system, this COLA will be explicitly funded in advance like the rest of the pension benefits.

New hires on and after July 1, 2017 will have the choice of taking a defined contribution plan (which is automatically fully funded, because you contribute whatever you contribute and it earns whatever it earns) or a defined benefit hybrid plan. Like for existing employees and retirees, the defined benefit plan’s COLA is tied to inflation, with a 2% cap; however, for new hires, COLAs will be restricted or eliminated if the plan is less than 90% funded. And employers and employees will equally share all retirement costs (current employee contributions are capped).

The new system has a number of additional features, and opining on its merits is beyond the scope of this post. But from a legal perspective, the most vulnerable part seems to be the part that applies to current employees and retirees. New hires have no vested interest in formulas that existed before they were hired, but under the Arizona Supreme Court’s interpretation in Fields, existing employees and retirees are entitled to the indefinite existence of whatever rules applied the day they were hired (plus any upward adjustments since then), and the Pension Clause categorically bars any diminution of these benefits.

This is where Proposition 124 comes in—which was approved by the voters on May 17. Prop. 124 takes care of S.B. 1428’s potential unconstitutionality in a direct and ultra-specific way. The new Pension Clause now reads, in relevant part (additions in italics): “Public retirement systems shall not be diminished or impaired, except that certain adjustments to the public safety personnel retirement system may be made as provided in Senate Bill 1428, as enacted by the fifty-second legislature, second regular session.”

The constitutional amendment thus guarantees that the Pension Clause won’t be a problem for S.B. 1428 (the parties behind the bill presumably feel confident enough that they can pass the Contract Clause’s three-part test), while simultaneously avoiding the long-term problem of the Arizona Supreme Court’s excessively sticky interpretation of what counts as a constitutionally protected pension benefit.

Prop. 124 makes one more change. It adds constitutional text to the Pension Clause clarifying that the legislature may still “modify public retirement system benefits for prospective members of public retirement systems.” This change was arguably unnecessary, but it’s hard to see how it could hurt.

Looking forward, what does this whole experience portend for the future? This particular problem (the funding of public-safety employee pensions) has been solved for the moment in this particular state, but of course this one-time constitutional carveout will have no legal effect on other pension reforms or outside of Arizona.

Politically, though, this could prove to be meaningful, especially if S.B. 1428 is also challenged and upheld under Arizona’s Contract Clause. Now we know that there’s a constituency for meaningful pension reform, and that reforms can be politically feasible even if they include some retroactive changes.

In states that only have a Contract Clause similar to Arizona’s—and that don’t have a Pension Clause, at least not one that has been interpreted Arizona-and-Illinois-style—the way forward is procedurally easy. Just reach the proper political compromise, and the legislature can simply enact the corresponding statute.

But in states with a Pension Clause like Arizona’s and Illinois’, any future reform along these lines will have to involve constitutional change, either in the text of the state constitution or in the state Supreme Court’s interpretation. Could Arizona’s experience make state Supreme Courts more likely to rethink their Pension Clause doctrine? Maybe—Supreme Court justices don’t like to see their work overruled by constitutional amendment; courts, in the long run, follow election returns; and union-friendly justices might realize that their strict Pension Clause interpretations aren’t always pro-worker. But that’s all very hypothetical: maybe justices could instead retain their strict Pension Clause doctrine, on the theory that the people can easily make one-time exceptions as needed.

Fortunately, state constitutions tend to be much easier to amend than the federal constitution. So a series of one-time carveouts in different states for different reforms, while moderately cumbersome, isn’t unimaginable. My hope is that, though the political will for a broader change didn’t exist this time, voters will eventually approve a broader constitutional reform—one that would explicitly fold Pension Clause protections into Contract Clause doctrine, treating the state’s pension promises on a par with its other contractual obligations.

Alexander “Sasha” Volokh is an associate professor of law at Emory Law School. An archive of his previous Reason.org articles is available here.