The business economist at work: an unemployment insurance actuary

Business Economics, April, 1994 by John R. Palumbi

STATE REVENUE ESTIMATION

Because of the federal government's role as a partner in the UI system, one of my duties is to create and maintain a realistic model of overall trust fund balances and revenue flows, and to estimate their impact on the solvency of the trust fund arrangement.

Anticipated revenues, benefits, and accumulated state trust fund balances are also important because state UI trust funds and federal UI surpluses, as part of the federal government's budget, help offset any deficit.

With the goal of eventually moving toward an increasing hazard system, I have built a partial adjustment model for aggregate state trust funds theorizing that, over time, the average aggregate state payroll tax rate will adjust toward some constantly evolving equilibrium. Another way of conceptualizing the situation is to regard it as a stock-flow problem, with the overall trust fund balance as the stock of interest.

I was asked to use this model to anticipate the revenue impact of a proposal to finance temporary emergency unemployment compensation through an increase (from one week to two weeks) in the amount of time one must wait to receive UI after becoming unemployed. This increase in the waiting period would have the effect of immediately lowering estimates of overall benefit payments from the regular program. The difference would create a window whereby the federal government could provide additional emergency benefit weeks to those having exhausted their regular benefits, while still satisfying laws stipulating that new spending be offset by additional taxes or equivalent spending cuts from elsewhere.

The reduced benefit payments would result in higher aggregate state trust fund balances, as well as a consequent downward pressure on firm's tax rates, meaning that only a fraction of the reduced benefit payments would be translated into higher aggregate state trust fund balances to offset the costs of emergency benefits. I used my trust fund model to derive the magnitude of such a multiplier. Final results indicated that, because of the dynamics of the UI system, only 90 percent of the benefit savings from increasing the waiting period would be available to offset the increased spending associated with the additional weeks of emergency benefits. The rest would be a windfall to employers through lower tax rates.

Although the proposal to increase the waiting week was ultimately rejected, it is nonetheless a good example of the kind of UI dynamics that I try to capture in my models.

WORKFORCE SECURITY

My work during the past year has increasingly involved the continued refinement of the revenue model and the provision of estimates supporting possible legislation, particularly a proposal called the Workforce Security Act (WSA). The concept underlying the WSA is that the nature of unemployment has changed. The UI system should be aiding not only its traditional constituency -- those affected by the cyclical downturns of the economy -- but also the increasing numbers of workers who become dislocated as a result of structural shifts in employment demand. By dislocated, I mean workers who are permanently separated from their jobs and have characteristics, such as little schooling or long job tenure, that make them more likely to suffer extended jobless spells.