RETIREMENT BENEFITS, ECONOMICS AND ACCOUNTING: MORAL HAZARD AND FRAIL BENEFIT DESIGNS

North American Actuarial Journal, Jan 2005 by Gold, Jeremy

ABSTRACT

This paper uses economic principles to analyze alternative recognition schemes for end-of-period retirement plan liabilities; the candidates, using U.S. nomenclature, are the vested benefit obligation (VBO), the accumulated benefit obligation (ABO) and the projected benefit obligation (PBO).

In competitive employment markets with rational contracting we are unable to justify projected costing (PBO-based) for typical pay-related defined benefit plans. Projected costing misrepresents the economic obligations incurred by shareholders and invites moral hazard.

Employee exposure to moral hazard may be minimized by exit costing (VBO-based) which recognizes only those benefits to which an exiting employee is entitled under the explicit benefit contract. But exit costing may not fully inform shareholders about the obligations that they have incurred under implicit contracts that extend beyond the plan document. Accrued costing (represented in the United States by the ABO) may better measure shareholders' economic commitments.

Small differences between the ABO and the VBO may measure a human capital asset incented by delayed vesting and benefit eligibility. Large differences are a marker for frail benefit design and potential moral hazard.

Moral hazard options exercised by employers disappoint employees and may lead to unwelcome ex-post results-oriented repairs imposed by legislators, regulators and courts.

1. INTRODUCTION

This paper uses economic principles to analyze alternative recognition schemes for end-of-period retirement plan liabilities; the candidates, using U.S. nomenclature, are the vested benefit obligation (VBO), the accumulated benefit obligation (ABO) and the projected benefit obligation (PBO). We are concerned only with economics and financial reporting and do not address funding.

Each end-of-period liability recognition scheme implies an allocation of costs over periods of employment. Because the VBO adheres to the explicit "contract" that defines the benefit entitlement of an exiting employee, we may call the VBO-based allocation exit costing. Accrued costing is based on the ABO, which exceeds the VBO because it assigns probabilities to future service that may meet eligibility requirements for vesting and for subsidized benefits. The ABO recognizes an implicit contract to continue employment beyond the current reporting date. Projected costing is based on the PBO, which adds a second layer of implicit contract, recognizing the benefit impact of estimated future pay increases.

The end-of-period liabilities must properly inform shareholders of economic obligations that they have incurred. The cost allocation method should provide reliable information to support economically rational decisions by managers, employees, investors and regulators.

Fully vested defined contribution (DC) plans provide an example of exit costing. The cost recognized by the employer in each period is identical to the cash it contributes. Defined benefit (DB) and other postemployment benefit plans accrue over an employee's career in ways that make the cost attribution less certain. Difficulties can arise with provisions for vesting and benefit eligibility, and with the treatment of future salary increases under pay-related DB plans. When do benefits that are subject to vesting and eligibility rules accrue, over the full crediting period or only when they vest? Do final-pay benefits accrue on the basis of current pay or expected final pay? What about benefits that may be revoked at the company's discretion, such as retiree medical coverage or ancillary pension benefits that are not protected by statute?

We assume competitive markets for labor and capital, and rational contracts. These contracts may be explicit, implicit or both and may include risks. Under these conditions, we argue that there cannot be a rational implicit contract for future pay increases that induces noncompetitive total compensation. This means we can rule out projected costing for common pay-related DB plans. But U.S. and international accounting standards prescribe exactly that approach. This accounting for benefit costs over employees' careers is flawed because it misrepresents the underlying contracts and invites moral hazard.

Some plan designs deliberately delay vesting and eligibility, leading to an ABO that is greater than the VBO. Such designs put employees at risk, raise expected compensation, reduce turnover, enable training investments and enhance productivity. Small differences may mark tradeoffs where productivity gains exceed expected compensation increases; thus the difference represents a human capital asset. Large differences, however, indicate frail contracts where the increased compensation required for employee risks is very likely to exceed any productivity gains; the difference exceeds any human capital asset and reflects an opportunity for moral hazard.

In Section 2, we review principles of employment economics and provide a literature review.