February 17, 2014

Reckless Unions

Pension reform could save cities on the brink of bankruptcy.

Public employee pension reform is a hot topic these days. This past week Rhode Island settled its dispute with the unions over its sweeping pension reform statute of 2011. Detroit and San Jose are wrangling over their own deals with their respective unions. Other cities and counties will follow. None of these jurisdictions can ignore the huge holes in their pension programs for both retired and active employees. One estimate puts the total amount of unfunded pensions in the United States at between $730 billion and $4.4 trillion, with the smart money betting on the higher estimates. Systematic deficits of this sort do not just happen. Fierce union pressure, bad economic circumstances, and unsound pension design all contribute to the bottom line.

Defined Benefits v. Defined Contributions

Start with some pension basics. Most public pension dollars are invested in defined benefit plans, under which the state, county, or municipality promises its workers a pension benefit based on a complex formula that combines years worked with salary earned. The risk of variation in the stock market, or changes in longevity, is therefore borne in the first instance by the government entity, which has lately faced heavier obligations on both counts.

epstein
Illustration by Barbara Kelley

The alternative system, a defined contribution plan, puts the risk on the employee, whose payments are fully vested at the time they are made. The individual worker is then left to adopt an investment strategy to deal with the portfolio, subject to whatever distributional and diversification constraints are incorporated in the plan. Defined contribution plans leave no uncertain residual liabilities in the hands of the government. Benefits vest at once, and workers are free to take their pensions (like their individual retirement accounts) with them if they change jobs. Private pension plans have moved sharply in the direction of defined contribution plans.

The stresses on defined benefit plans are compounded enormously by the powerful role that unions play in negotiations. In boom times, government officials and unions both found it easier to use deferred compensation in the form of defined benefit plans to protect themselves from public criticism of generous compensation packages. Now, everyone knows that there is a problem, but no one can agree on a common cure. Any structural pension reform that requires union consent will lead to concessions that in practice will prove both too little and too late. Only unilateral modifications by the government employer can save the day. The question is this: How can the government best roll back pensions in ways that satisfy key economic requirements without running afoul of serious constitutional concerns?

The Economic Issues

The first problem with pensions is figuring out what the government employer’s liabilities are. The six-fold difference in estimates of total pension liability reflects the difficulties of that task. Pension obligations must be sustainable for the long run, requiring projections spanning generations. Figuring out rates of return is no picnic. Indeed, multiple market gyrations have left the Dow about 11 percent below 2000 levels in real terms, adjusted for inflation. The needed revenue is just not there.

Pension valuation also heavily depends on the precise formulas used to calculate the payoffs. Defenders of the current system point to the moderate annual pensions for retired workers. Detroit pensioners receive an average pension of $19,000 per year, hardly a princely sum. Their union defenders point to this low figure as evidence that these pensions should remain intact throughout the bankruptcy process.

But the correct conclusions depend not only on the annual payments, but also on the total burden any individual pension places on the overall system. Take a worker who retires after 25 years of service, whose pension equals, say, 75 percent of his last salary. That obligation could kick in at age 50, at which point the prudent pension plan administrator must shift its investment from equity to bonds to fund current obligations. But debt instruments unfortunately lower the overall rate of return. That same pension payable at age 65 is a pale shadow of the first plan, because the pension fund has an extra 15 years over which to accumulate the wealth needed to service the pension, and 15 fewer years over which the money has to be paid out. The crushing county or municipal debt comes from the total size of the obligation, not the yearly payments.

Similarly, this annual figure does not capture the worker’s position. Workers who retire early often take second jobs, which provide social security, whose own bizarre compensation formula favors parties who enter the workforce close to retirement.

The Legal Issues

The great legal challenge in all these cases is when the pension rights negotiated in the original employment contracts should “vest” so as to receive protection against unilateral variation by local governments. The Constitution in Article I, Section 10 commands with deceptive simplicity that “No state shall. . . pass any . . . law impairing the obligation of contracts.” Many state constitutions contain similar provisions. The issue came to a head in recent litigation in San Jose, where the estimated unfunded liability for the pension plan was put at $3.7 billion. Under the leadership of San Jose Mayor Chuck Reed, San Jose adopted “Measure B” by referendum in 2012, which let the City trim, but not eliminate, pension benefits in order to release funds to restore key city services.

The unions promptly challenged Measure B in court, where they won a major victory before Judge Patricia Lucas, who held that Measure B violated the vested rights of union members. By way of full disclosure, I have given some informal advice to Mayor Chuck Reed about how to attack that decision on appeal. The gist of this dispute lies in the interaction between the key provisions of the San Jose City Charter and California case law on vested pension rights. San Jose has two defined benefit plans, one for uniformed services and one for its “civilian” employees. The Charter explicitly reserves to the San Jose legislature the right to “alter or amend” each plan unilaterally. However, this authority does not explicitly include the power to revoke the pension benefits. On its face, it appears that the Charter lets the City cut down pension benefits, without setting out a formula defining how to do so.

California pension law cases, without careful analysis, have evolved to freeze minimum pension levels for all employees the moment that they take their jobs. Workers may be dismissed or demoted from their jobs, but the pension structure is said to remain inviolate as a result of two key decisions. In Kern v. City of Long Beach (1947), the California Supreme Court held that a pension could not be repealed in its entirety just before a Long Beach fireman was about to retire, which would have stranded covered workers high and dry. Fair enough. However in 1955, and without serious reflection, that rule morphed in Allen v. City of Long Beach to require that any “changes in a pension plan which result in disadvantage to employees should be accompanied by comparable new advantages."

The gap between Kern and Allen is enormous, for while the former prevents the City from taking advantage of its workers, the second neuters the power of local governments to alter and amend, by wiping out all government flexibility to correct prior errors in pension program design or funding. That rigidity is exceedingly costly for two reasons: first, because of the obvious difficulties in making any once-and-for-all estimate of future pension benefits, and, second, because it creates a built-in ratchet whereby any future increase in pension benefits is permanently added to the base, without the possibility of further reduction. The upshot is a financial death spiral.

Clearly, the Charter language can only be made effective by finding some middle ground between total flexibility and total rigidity. Indeed, that middle ground is constitutionally necessary in light on the well-established proposition in Stone v. Mississippi (1880): “All agree that the legislature cannot bargain away the police power of a State.” The legislature cannot permanently contract away its right to govern. This search for the middle ground ultimately rests on the notion of “good faith modification” that plays a central role in the ordinary law of contract by requiring anyone with the unilateral power to modify existing arrangements to take due regard of the interests of the other claimants on public funds. San Jose Measure B sought to do this by limiting the amount of pension reductions to those needed to deal with its estimated $3.7 billion shortfall, without cutting off anyone entirely as in Kern.

To be sure, the determinations of good faith are difficult to make, but it is surely better to tolerate some ambiguity at the margin than to force the City, which has multiple fiduciary obligations, to make wholesale cuts in its current programs, cuts that might include the size of its police and fire forces, the maintenance of its parks and schools, and various welfare programs. In her opinion, Judge Lucas heroically sought to confine her inquiry to “one of law, not of policy,” and thus missed the key point that the pension straight jacket has political as well as legal consequences.

 All of these consequences can be avoided ex ante by holding that pension rights “vest” only for work completed. Just that position, for example, is taken in the Michigan constitution, which in Article IX, Section 24 states: “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.” That definition is consistent with San Jose’s Measure B, and leaves it to another day to answer the question of whether federal bankruptcy law can force reduction in all unsecured claims, including those that are fully vested in retirees under state law (as Judge Steven Rhodes held could be done in the Detroit bankruptcy).

This correct definition of vesting is consistent with federal contracts clause jurisprudence that denies the government the unilateral power to alter and amend. Thus in United States Trust v. New Jersey (1977), the Supreme Court held that the state could not strip a secured lender of his right to the property pledged to the loan. Likewise in United States v. Winstar (1996), the Court applied ordinary principles of contractual interpretation to bind the government to its promise to count good will against the capital requirements of the bank when it assumed liabilities on which the federal government was guarantor.

The key ground of distinction is this: in both United States Trust and Winstar, the private party had performed in full, so that they had no option to leave the transaction if the government changed its terms. With the future pension, the workers do have the option to leave the job—an option they will exercise if the cutbacks are too extreme. There is in effect a self-help mechanism available to San Jose’s workers that was not available to either the United States Trust or Winstar. For good reason, the federal case had the open-ended “alter and amend” language found in the San Jose Charter. The faithful adherence to contractual language makes it equally imperative not to read into an agreement an “alter and amend” clause that is not there. But by the same token it makes it indefensible to take that same clause out of a charter.

San Jose and other local governments could be on the brink of bankruptcy. The unions are playing a reckless game. If they don’t back off, they may dig the bankruptcy hole deep enough that their fully accrued benefits could easily be put in jeopardy as well. California’s appellate court should affirm the constitutionality of Measure B, and avoid a far worse fate.


Richard A. Epstein, Peter and Kirsten Bedford Senior Fellow at the Hoover Institution, Laurence A. Tisch Professor of Law at New York University, and senior lecturer at the University of Chicago, researches and writes on a broad range of constitutional, economic, historical, and philosophical subjects. He has taught administrative law, antitrust law, communications law, constitutional law, corporate law, criminal law, employment discrimination law, environmental law, food and drug law, health law, labor law, Roman law, real estate development and finance, and individual and corporate taxation. His publications cover an equally broad range of topics. His most recent book, published in 2013, is The Classical Liberal Constitution: The Uncertain Quest for Limited Government (2013). He is a past editor of the Journal of Legal Studies (1981–91) and the Journal of Law and Economics (1991–2001).


Letters to the editor may be sent to definingideas@stanford.edu. Editors reserve the right to reject or publish (and edit) letters.