The business economist at work: an unemployment insurance actuary

Business Economics, April, 1994 by John R. Palumbi

SINCE BECOMING responsible for a national Unemployment Insurance (UI) revenue forecasting model in September 1992, I have been involved in shaping a new attitude and perspective toward UI.

As an actuary, I use economic analysis and econometric models to forecast the effects of changing economic conditions on the UI system. Policymakers are particularly interested in the impact of legislation of perceived interest groups and the inevitable winners and losers associated with any new laws.

St. Louis & Associates is under contract to the Unemployment Insurance Service within the U.S. Department of Labor to provide actuarial support to their Division of Actuarial Services. Three contract actuaries round out the Division's three federal actuaries, with both federal and contract staff comprising what is called the Benefit Financing Group (BFG). The entire group reports to a career federal actuary who, as the Chief of the BFG, coordinates everyone's activities and the group's interaction with the rest of the Unemployment Insurance Service.

We are all general experts on UI, performing a myriad of overlapping tasks, although each actuary does tend to specialize in a particular aspect of the system. I specialize in revenue estimation, another handles overall benefit payments, two provide technical assistance to states, the fifth researches issues pertaining to our annual workload forecasts and the sixth does a lot of work with the weekly UI initial claims release. Each actuary's activities are broadly defined, and special project teams form or dissipate as needed, encouraging cross-training and the development of a broad range of knowledge. Technical issues and new ideas are placed before the group in ways both formal and informal, and appropriate action is generally determined through consensus. By providing a forum, the BFG encourages innovation while minimizing risk.

UI BACKGROUND

To portray the nature of the actuarial work within UI adequately, it is necessary to discuss briefly the general framework of the system.

The UI system is financed through an employer payroll tax upon the first $7,000 to $24,000 of a worker's wages, depending upon the state where the worker is employed. (Washington, DC, Puerto Rico and the U.S. Virgin Islands are also part of the UI system: the word "state" as used here refers to all of the jurisdictions subject to UI.) Benefit payments are financed through this payroll tax, the level of the tax varying from firm to firm within a state depending upon each firm's history of layoffs (called experience rating of UI taxes).

The market mechanism of experience rating is one aspect of UI that makes it so intriguing to economists. Although not a perfectly efficient measure, each firm's tax rate can be viewed as the price of unemployment. Experience rating also ensures the vigilance of employer groups in maintaining an efficiently managed UI system and keeping moral hazard to a minimum.

Among states, the revenues collected from employers are paid into each state's interest-bearing state trust fund account. Benefit payments to qualified claimants are then drawn from those accounts, while administrative costs, emergency benefits and loan reserves are financed through a federal payroll tax. As a result, the UI system is self-financing and self-administered.

An adverse change in the economy, e.g., due to the onset of a recession, would cause benefit payments to increase and aggregate state trust fund balances to decline. Because employer tax rates are experience rated, tax rates for the next year would then adjust upward as a consequence of the increased benefit payments. Tax receipts would begin to rise and net trust fund outflows would ultimately be reversed, eventually restoring state trust funds to their previous levels. Once trust funds have been replenished in the recovery period, tax rates would ease until the cycle began again.

The speed with which the system adjusts is vitally important; if taxes respond too quickly to increased benefit payments, the tax burden on business could increase at a point in the recession when employers are least able to absorb higher labor costs. This might pressure employers to reduce their work force and have the counterproductive effect of fueling the recession. If a state's tax system adjusts too slowly, either the system becomes insolvent as benefit payments exhaust the trust fund, or the trust fund becomes unnecessarily large, draining funds from the economy that could otherwise be utilized more efficiently. Each state balances these two concerns differently, depending upon particular economic conditions and the current political climate regarding taxation.

I have two goals as a UI actuary, each posing unique obstacles: (1) to model UI as it is; and (2) to model UI as it might be. While some relationships are easy to explain and model in isolation, it has been a real problem getting variables to cooperate realistically as part of a system. With revenue forecasting, I began with a simple framework and added sophistication as time and resources allowed.