Review and Empirical

From the early eighties to the nineties, the Standard Rate increased modestly, but is still below its mid-seventies level. While researchers have identified many reasons for the low UI recipiency rates over the past twenty years, many questions remain as to the causes behind the low rate and steps that policy and program officials might take to increase it. While the Standard Rate is the most commonly used measure to evaluate the effectiveness of the UI program, researchers have developed alternative UI recipiency rates to address some of the limitations of the standard measure.

The standard measure is expressed as the ratio of the insured unemployed (i. e. , the number of regular UI claimants) to the total number unemployed. Alternative measures have been designed to better capture the effectiveness of the UI program by including the full range of UI programs available to the unemployed (beyond the regular program) and by more accurately defining the UI target population (a subset of unemployed workers).

Purpose and Methodology The purpose of this report is to examine why the Standard Rate, as well as alternative recipiency rates, declined sharply in the early eighties and continued to remain well below their midseventies level in the early nineties. We critically reviewed the findings from the research literature to explore the factors others have identified to explain the drop in the UI recipiency rate. The literature review enabled us to identify factors for inclusion in our empirical analysis and to assess the effects of factors that could not be included in our own analysis.

Our empirical analysis is based primarily on the methodology used by Burtless and Saks (1984) and focuses only on changes in the UI recipiency rate over recessionary periods. It is important to compare similar economic periods because the UI recipiency rate is higher during recessionary periods and lower during periods of economic expansion. We first replicated the analysis from Burtless and Saks, estimating the effects of various factors that influenced the rate used in their original analysis from the seventies recession (1975-76) to the eighties recession (1981-83).

We then extended their earlier analysis by testing the effects of additional factors during that period. Next, we updated the analysis to include data from the most recent recessionary period in the nineties (1991-92). We chose the period in the nineties to be consistent with the periods of rising unemployment rates selected by Burtless and Saks. Finally, we extended their analysis by using the Standard Rate and two additional measures of UI recipiency selected to measure the performance of the UI programs during recessionary periods.

Our conclusions about the effects of various factors on the UI recipiency rate are based on the findings from both the critical review of the literature and our empirical analysis. We also present evaluation design options to address some of the limitations of current knowledge. The Lewin Group, Inc. E-1 156059 Executive Summary C. UI Recipiency Rate Measures Four UI recipiency rate measures were selected for the empirical analysis.

Standard Rate: number of weekly claims for regular program unemployment insurance benefits, as a proportion of all unemployed workers;1 All Programs Rate: number of weekly claims for all program (regular, extended and federal) unemployment insurance benefits, as a proportion of all unemployed workers; Standard Short-term Rate: number of weekly claims for regular program unemployment insurance benefits, as a proportion of job losers unemployed less than 27 weeks; and All Programs Job Loser Rate: number of weekly claims for all program (regular, extended and federal) unemployment insurance benefits, as a proportion of all job losers.

The final three UI recipiency rates deviate from the Standard Rate by changing the definition of UI claimants, unemployed workers, or both. Because the All Programs Rate and the All Programs Job Loser Rate include all UI program claimants, Wandner and Stengle (1996) argue that they are generally better measures of UI coverage during recessionary periods when extended benefit programs are provided. The All Programs Job Loser Rate differs from the All Programs Rate because it targets a subset of unemployed workers (i. e. , job losers) who would be most likely to qualify for UI benefits. The Standard Short-term Rate only includes regular program claimants and the general “target population” for the regular state program, job losers unemployed less than 27 weeks.

This final measure was used in the original Burtless and Saks analysis. All three alternative rates are larger than the Standard Rate because they use either a more expansive definition of UI claimants and/or a more restrictive definition of unemployed workers. From the seventies to the eighties, all four recipiency rates declined sharply (Exhibit 1). The largest reductions are for the All Programs Rate and the All Programs Job Loser Rate. These rates declined by more than the Standard Rate because of the large cutbacks in the extended benefit programs that were implemented in the early eighties. From the eighties to the nineties, the Standard Rate increased slightly.

There is not, however, a large change in either the All Programs or All Programs Job Loser rates over this period, due to the small number of extended claimants. If, however, the analysis were extended to periods following March 1992, there would be an increase in both of these rates because of the extension of benefits through the Emergency Unemployment Compensation (EU3) program. 2 The Standard Short-term Rate follows the same general pattern as the Standard Rate, though there is a much sharper drop-off in the Standard Short-term rate in the early eighties that corresponds with fewer short term job losers receiving regular program benefits.