Faced with public pension crises, many states have recently enacted pension-reform laws-increasing the rates at which their employees must contribute to their pension funds, reducing or eliminating cost-of-living adjustments, increasing the retirement age, or even converting to an entirely different type of pension system. Public employees and retirees have aggressively challenged these reforms, arguing that the makeup of their pension plans was part of the employment contracts they agreed to years ago. One of their main weapons is a relatively forgotten part of the constitution: the Contract Clause.
The text is simple: “No State shall . . . pass any . . . Law impairing the Obligation of Contracts.” The Contract Clause is one of the few restrictions against states to be contained in the original constitution, rather than in the Bill of Rights or the post-Civil War amendments, and is also one of the few economic-rights provisions. In the early Republic, courts used this clause aggressively to protect the obligation of private and public contracts alike: states were restricted in how much they could relieve debtors of the obligation to pay their creditors, just as they were restricted in repudiating their own bonds or reneging on their promises of tax exemption. James Ely writes that the Contract Clause was “the most litigated provision in the Constitution and was the chief restriction on state authority.”
This changed gradually throughout the late 19th and early 20th centuries, as the Supreme Court increasingly deferred to state economic regulation, especially during “emergencies.” The protection of public contracts was the quickest to go, but private contracts lost protection as well. Since the New Deal, the Contract Clause, much like other economic rights, has been afforded a relatively low level of protection.
But examining the current crop of cases by public employees and retirees challenging pension-reform laws-there are at least a dozen such cases just in 2013 so far-shows that the Contract Clause (as well as similar state constitutional Contract Clauses) nonetheless remains a vibrant area of litigation.
These are important challenges for many reasons. At the most general level, presumptively everyone, even governments, should keep their promises, since contractual security is an important feature of a free-market economy. Still, figuring out what’s part of the contract and what’s merely an alterable state policy can be tricky. Public-sector contracts have the added twist that they’re politically determined and can be used to transfer resources from taxpayers to politically preferred groups, and-especially in tough budgetary times-keeping them may be beyond a state’s means. These challenges are also important for the future of privatization, since part of the appeal of privatization is the high cost of paying public employees; while reforming pensions at public employees’ expense may improve a state’s finances, it might also hide the true cost of in-house provision of government services.
* * *
Under the modern Contract Clause, public contracts get more protection than private ones (perhaps surprisingly, given the historical evolution of the doctrine); the theory now is that government has a greater conflict of interest when it’s a party to the contracts that it abridges. A modern Contract Clause claim, applied to public contracts, has several components: (1) Was there a contract? (2) Did a state law impair the obligation of the contract? At this point, one would think the inquiry would be over, since the constitutional text straightforwardly prohibits states from passing laws impairing the obligation of contracts. But now, given states’ greater leeway to regulate economic matters, we go on to ask: (3) Was the impairment substantial? and (4) Did the impairment serve a legitimate public purpose, and was it reasonable and necessary?
Let’s start with question (1), whether there’s a contract. It turns out that this can be a difficult question in the context of public employment: as the Supreme Court stated in 1985, the presumption is against finding a contractual obligation. “Policies, unlike contracts, are inherently subject to revision and repeal, and to construe laws as contracts when the obligation is not clearly and unequivocally expressed would be to limit drastically the essential powers of a legislative body. . . . Thus, the party asserting the creation of a contract must overcome this well-founded presumption, and we proceed cautiously both in identifying a contract within the language of a regulatory statute and in defining the contours of any contractual obligation.” This presumption seems hard to maintain in the case of pensions, which are a form of deferred compensation and thus akin to current wage promises. Most states thus don’t take the presumption that seriously. As Stuart Buck writes, “the vast majority of states treat pension laws as creating a contractual entitlement to something“; but “the real question is what that something is.”
Suppose you’re retired from county employment and have always been covered under the county’s health plan. Every month, you pay a premium, which is calculated on a group-wide basis, based on the expected costs of the group’s medical care for that month. Clearly, what group you’re bundled with makes all the difference: retirees incur greater medical costs than active workers, so if retirees and active workers are in separate groups, retirees will pay higher premiums and active workers will pay lower ones; conversely, if retirees and active workers are bundled in the same group, everyone pays some intermediate premium, so the active workers cross-subsidize the retirees. The county now proposes to segregate the two groups, though they had been pooled together for over 20 years. You worked for the county in the days of pooling; you paid premiums that cross-subsidized retirees, hoping to benefit from the same subsidy after you retired; but now this new segregation threatens to end the subsidy. Of course you’d rather not pay the higher retiree rates. Are you contractually entitled to a unified pool for your lifetime?
This scenario actually happened in 2007 in Orange County, California. The Retired Employees Association of Orange County sued the county in federal court. The district court ruled in favor of the county, holding that there had never been a promise to continue providing the pooling benefit: as a matter of California law, the court held, the county couldn’t be liable for any obligation it didn’t undertake explicitly through a board resolution. The retirees appealed to the Ninth Circuit, which certified the question to the California Supreme Court. In 2011, the California Supreme Court answered that generally, public employment is held by statute, not by contract; but the result is different when the hiring units (here, counties) are authorized to enter into contracts, collective bargaining agreements, memoranda of understanding, and the like. In such cases, contractual terms can be implied from legislative enactments (like board resolutions or ordinances) just as they can from private contracts, based on the parties’ expectations, their course of conduct, and so on.
In light of the California Supreme Court decision, the Ninth Circuit sent the case back to the district court, which ruled in favor of the county again, and the retirees appealed again. That case is still pending, so we don’t know the final answer yet as to the Orange County retirees. But in the meantime, a similar case arose in Sonoma County, California. Concerned about rising healthcare costs, Sonoma County decided to limit its healthcare benefit contributions to $500 per month for retirees. The Sonoma retirees sued in federal court. The district court dismissed the complaint, holding that there could be no implied contract, but on appeal, in Sonoma County Ass’n of Retired Employees v. Sonoma County, the Ninth Circuit-in light of the California Supreme Court’s intervening Orange County decision-held that this was mistaken. Nonetheless, the retirees still lost. It’s true that county board resolutions or ordinances ratifying memoranda of understanding can in principle create binding contracts if that’s the intent that emerges from their text; but in this case, the retirees failed to allege that there was any such resolution or ordinance. The Ninth Circuit thus affirmed the district court’s dismissal of their complaint.
That was on February 25, 2013, and the retirees may yet come back with an amended complaint, so this story, too, is not yet over. But what these cases illustrate is that, for a Contract Clause violation, at a minimum, one has to show that there was some contract, and in the context of public employment this is far from trivial.
* * *
Now let’s go on to question (2): did a state law impair the obligation of the contract? Note that, to violate the Contract Clause, it’s also not enough for a state to simply breach a contract. A breach of contract merely gives rise to a state-law suit for breach of contract. What the Contract Clause prevents is the states’ changing their law so as to impair the obligation of contracts. Thus, Anne Warkentine, a teacher with the Salina, Kansas public school district, was denied a $35,000 early-retirement incentive. The school district argued that the incentive was available only for teachers with at least 15 consecutive years of full-time employment with the school district immediately before retirement; because Warkentine had had a brief spell of part-time employment within the last 15 years, she wasn’t eligible. Warkentine disputed the school district’s interpretation of the contractual language and argued that she was eligible because of her 28 years of full-time employment (though not all in one consecutive chunk) before retirement. She brought a number of claims against the district, including a breach of contract claim and a Contract Clause claim (which is what allowed her to sue in federal court). But because she couldn’t point to any change of law, a federal district court in Kansas, in Warkentine v. Salina Public Schools, denied the Contract Clause claim on February 1, 2013, allowing the breach of contract claim to go forward.
But even suppose there’s a clear contract-a longstanding state “preservation of rights” statute has clearly declared that state employees’ pension rights are contractual in nature. And suppose there’s a clear law altering that contract. So far, your pension has been noncontributory-the state paid all the premiums-but now the state legislature changes the law so you must henceforth pay 3% of your salary to the retirement system. And suppose, too, that cost-of-living adjustments (COLAs) are eliminated for service after the effective date of the law. Does the change in the law impair the state’s obligations under its contract? This happened in Florida in 2011 with the passage of Senate Bill 2100. The general rule, established by the Florida Sheriffs Ass’n case from 1981, was that the preservation of rights statute “vests all rights and benefits already earned under the present retirement plan so that the legislature may now only alter retirement benefits prospectively.” The SB 2100 challengers contended that the Florida Sheriffs rule applied to changes in benefits but didn’t apply to the change of the plan from noncontributory to contributory. But on January 17, 2013, the Florida Supreme Court ruled in Scott v. Williams that the Florida Sheriffs rule applied to any purely prospective change in the state pension system. Because the 3% contribution rule only applied for future service, and because COLAs were likewise eliminated only for future service, SB 2100 didn’t “impair any statutorily created contract rights.” (The Florida Supreme Court’s analysis proceeded under the Florida constitution, but Florida’s Contract Clause functions similarly to its federal counterpart.)
One can see the same dynamic at work in Sweeney v. Daniels, where union members challenged Indiana’s 2012 “right to work” law, which prevents forced union membership as a condition of employment. An Indiana district court held on January 17, 2013 that there was no violation of the Contract Clause because the law only applied prospectively. But not all states follow the same prospective-retrospective dichotomy: California and about a dozen other states hold that there’s a contractual right in pension promises from the very beginning of employment; thus, a state can fire you, lower your salary, or reduce your other benefits, but can’t decrease your rate of pension accrual while you’re employed. This seems strange, though what California gives with one hand, it might take away with the other, as will become apparent in question (4) below.
* * *
As mentioned above, if questions (1) and (2)-was there a contract, and did a law impair the obligation of the contract?-are answered affirmatively, the constitutional violation seems to be complete. But modern cases look, in addition, to (3) whether the impairment was “substantial” and (4) whether the impairment serves a legitimate public purpose and is reasonable and necessary.
As to question (3), consider Taylor v. City of Gadsden, where Alabama firefighters challenged increases to their required pension contribution rates from 6% to 8.5%. On July 29, 2013, a district court in Alabama ruled, first, that there was no contractual right to a perpetual contribution rate of 6%; statutes can create contract rights-the Florida preservation of rights statute discussed in Scott v. Williams did just that-but there was no such statute in this case. But the district court went further and held that, even if there had been a contract, its impairment in this case wasn’t substantial. Factors affecting whether an impairment is substantial include “whether the impaired term was central to the contract, whether settled expectations have been disrupted, and whether the impaired right was reasonably relied on.” And these factors cut against substantiality in this case: the Retirement Systems of Alabama handbook, after all, warned employees that “[t]he member’s contribution rate is determined by statute and subject to change by the Alabama legislature,” which undercuts any claims of centrality, settled expectations, or reasonable reliance. But this inquiry is highly fact-specific: in New York State Correctional Officers & Police Benevolent Ass’n v. New York, on December 3, 2012, a New York district court refused to dismiss a complaint related to changes in required contribution rates, holding that plaintiffs had a “plausible claim” of a substantial impairment-which they would have to go on to prove at trial.
Once questions (1) through (3) are satisfied, the state will often lose under prong (4). In the New York case mentioned above, the district court noted that the public purpose couldn’t simply be the “financial benefit of the sovereign,” so a garden-variety funding crisis presumably wouldn’t justify substantially impairing public employees’ contract rights. Since the state is a party to these contracts, courts tend to afford less deference to its claims that the impairment of the contract has a legitimate public purpose and is reasonable and necessary. California, on the other hand, is more willing to allow substantial pension reform; its case law holds that employees are entitled to a “substantial or reasonable pension” but not a specific amount of benefits. Perhaps this has to do with its (overly?) generous approach to finding a contractual right to specific levels of pension benefits on the front end.
* * *
The future of pension reform thus depends heavily on this constitutional issue. On the one hand, it seems right that a state shouldn’t be able to violate its contracts whenever they seem too expensive, even if the contracts were ill-advised to begin with, and even if this will cost the taxpayers. After all, the state can’t take private property without paying for it (obviously, at taxpayers’ expense); why should it be able to “take” a contract right for free? If public employment is really so expensive, perhaps state officials will learn next time not to make promises that are too generous. On the other hand, in most states, a great many pension-related promises aren’t contractual, especially ones that only relate to future periods of work. This seems to give states substantial leeway in crafting pension-reform bills that will withstand constitutional scrutiny.
Alexander “Sasha” Volokh is an associate professor of law at Emory Law School. An archive of his previous Reason.org articles is available here.