ERISA & uncertainty.

• Author: Maher, Brendan S.
• Position: In employment and health insurance reform

TABLE OF CONTENTS INTRODUCTION I. UNCERTAINTY IN THE BENEFIT PROMISE A. Performance Uncertainty B. Expectation Uncertainty C. Collective Uncertainty II. THE ERISA BENEFIT PROMISE A. The Defined Benefit Pension Promise B. The Defined Contribution Pension Promise C. The Welfare Benefit Promise III. UNCERTAINTY IN THE ERISA BENEFIT PROMISE A. Defined Benefit Pension Promise Uncertainty B. Defined Contribution Pension Promise Uncertainty C. Welfare Benefit Promise Uncertainty IV. UNCERTAINTY IN THE SUPREME COURT A. Civil Enforcement B. Limiting Remedies C. Limiting Review CONCLUSION INTRODUCTION

More than other developed nations, the United States relies on private promises to assure health and retirement security. (1) These promises involve "employee benefits." They are subsidized by the first- and third-largest tax expenditures in the federal budget. (2) And they are heavily regulated by a landmark statute known as ERISA. (3) In this Article, we develop a theory of uncertainty designed to evaluate ERISA and its regulation of the benefit promise. (4)

To set the stage, employee benefits come in varying forms. They include traditional monthly pensions, (5) 401(k) contributions, (6) and the payment of health insurance premiums. (7) As economists have long noted, these benefits are wage substitutes. (8) In other words, the promise of benefits entails a corresponding reduction in salary. No one disputes, therefore, that these promises should be secure, understood by both parties, and not too costly to make or administer. But legal rules that promote security and clarity may render benefit promises more costly; in other words, rules may have both desirable and undesirable consequences, and a given rule's ultimate desirability will virtually always require the assessment and balancing of competing concerns regarding security, clarity, and cost.

We argue that the most useful way to compare alternatives--both in terms of prospective rule selection and retrospective rule evaluation--is to frame the inquiry in terms of context-specific uncertainty. Economists have long recognized the power of such framing in making difficult choices between competing legal rules. (9) More recently, this mode of analysis has been employed by legal scholars to explain and evaluate specific areas such as commercial contracts, (10) property rights, (11) and criminal plea negotiations. (12) In our view, a similar approach is sorely needed in the ERISA context, where existing scholarly and judicial debates often suffer from a profoundly undertheorized conception of the benefit promise and its regulation.

To be sure, there will always be disagreement regarding whether, and how, government should provide, subsidize, or regulate pension and health-care benefits. (13) At the same time, thoughtful examination of existing policy often reveals areas in which some intervention is necessary. (14) Broad thinking is essential because, as ERISA scholars have long observed, the stakes are high. (15)

The Article proceeds as follows: In Part I, we argue that benefit promises necessarily implicate three species of uncertainty--(i) performance uncertainty (i.e., the likelihood that an agreed-upon benefit promise will not be performed); (ii) expectation uncertainty (i.e., the likelihood that a benefit promise does not reflect a mutual understanding of promise terms); and (iii) collective uncertainty (i.e., the likelihood that a proposed rule will undesirably reduce, overall, the number or generosity of future benefit promises). In Part II, we briefly rehearse the most common benefit arrangements regulated by ERISA. In Part III, we evaluate how the different categories and aspects of ERISA benefit promises implicate different mixes of uncertainty. In Part Iv, we apply our model to several important Supreme Court decisions, explaining, in part, why the Court has written opinions that appear indefensible on purely doctrinal grounds. We conclude by criticizing both the Court and Congress for failing to candidly acknowledge that central questions inadvertently left open by ERISA cannot be resolved without a comprehensive legislative response. (16)

1. UNCERTAINTY IN THE BENEFIT PROMISE

   The phrase "employee fringe benefit" is commonly used to describe any nonwage item of value provided by an employer to an employee. Yet employer-provided health insurance, pensions, and other perquisites now constitute a significant percentage of total compensation for working Americans. (17) Consequently, most scholars refer to these items only as "employee benefits."

   Before delivery, an employee benefit is simply a wage substitute expressed as a promise of future consideration in whatever form the benefit takes (e.g., a monthly pension check, employer-paid health insurance premiums). In a well-known ERISA opinion written over twenty years ago, Judge Richard A. Posner expressed the point with characteristic elegance: "the less an employee's pension rights are worth, the higher are the wages that he will demand." (18) Although the existence of this wage-benefit tradeoff is widely accepted by economists, (19) it was not recognized by most American courts until the middle of the twentieth century. (20)

   Any benefit promise will necessarily present "risks" (i.e., undesirable outcomes that could materialize). (21) And any rational decision maker (22) will attempt to quantify such risks before selecting a particular course of action. (23) If a decision maker cannot quantify a material risk, she faces what is often referred to by economists as "uncertainty." (24) As scholars regularly observe, "[i]gnoring major problems because of uncertainty is an invitation to disaster." (25) In this Part, we present our theory of uncertainty. (26) Part I.A addresses what we refer to as "performance uncertainty." Part I.B addresses what we refer to as "expectation uncertainty." Part I.C addresses what we refer to as "collective uncertainty." (27)

   1. Performance Uncertainty

      Imagine the following promise made by Promisor A to Promisee B: "If you today relinquish to me your seat on this crowded bus, exactly five weeks from today I will pay you five hundred American dollars in cash." The danger in relinquishing the seat is not that the promised benefit is unclear and that one might be entitled to something less than five hundred dollars; the danger is that the promise will not be performed. Among the many reasons the promise may not be performed is that the promisor may not have five hundred dollars available to give in five weeks' time. A meeting of the minds does not ensure performance, and in the benefit setting--where the beneficiaries are often elderly or ill when the promise ripens--performance is paramount. (28)

      In the benefit context, performance uncertainty describes the likelihood that the promisor will not perform in a way consistent with the shared expectations of the parties at promise inception. That is, it describes the possibility that the promisor will fail to deliver a benefit when there was, in fact, an original meeting of the minds regarding its amount and triggering conditions. The most obvious reason for such uncertainty is the possibility that the promisor will lack the assets needed to confer the promised benefit when the entitlement matures. (29)

      Financial inability, however, is not the only threat to rightful benefit conferral. Performance uncertainty also describes the possibility that dishonesty, strategic play, or carelessness by the promisor or its agents will result in a wrongful refusal to confer a benefit, even where the promisor has sufficient assets and there was an original meeting of the minds regarding the terms of the benefit promise. In practice, it is often difficult to distinguish such an occurrence from what we call "expectation uncertainty." (30) This difficulty can pose challenging problems for policy makers and near-insurmountable problems for the judiciary. (31)

   2. Expectation Uncertainty

      Expectation uncertainty describes the likelihood that, at promise inception, the parties do not share a material expectation regarding the meaning (usually, the applied meaning) of the promise. (32) The absence of a shared expectation can occur because (i) the parties have firm but differing initial expectations regarding the meaning of the promise in a particular circumstance (a "circumstance-specific expectation"), (ii) one party has an initial circumstance-specific expectation and the other party does not, or (iii) neither party has a circumstance-specific expectation.

      Although the last two variants describe a total or partial absence of a specific expectation, they are nonetheless instances of expectation uncertainty because, in virtually all cases, a broad standard of conduct encompassed by the promise (or imposed by law) supplies a general expectation of promise content (e.g., the promisor would follow "fiduciary" standards of conduct in performing the promise). Missing, however, is an expectation regarding the application of that standard in a particular circumstance (e.g., that a fiduciary in situation A would do X and not Y). (33)

      Indeed, a useful generalization regarding expectation uncertainty tracks the time honored rule-standard continuum. (34) If a one-sentence description of law is that it attaches consequences to conduct or circumstance (i.e., if Conduct A occurs, Consequence B follows), then, to oversimplify, classic "rules" are legal directives that, in objectively discernible circumstances, impose determinate results. (35) Classic "standards" are legal directives that, in circumstances possessing a certain character, authorize a range of consequences sensitive to situational facts. (36) Generally, a benefit promise that is contractually or statutorily "rule-based" will contain less expectation uncertainty than a "standard-based" promise. The more objective and discretely conditioned the promise, the more modest the expectation uncertainty. (37)...