• Print
  • Email

Chicago Fed Letter, No. 47, July 1991
Unemployment Insurance: Countercyclical or Counterproductive?

According to estimates by the Congressional Budget Office (CBO), nearly 8.5 million people are expected to be unemployed in 1991, up almost 25% from 1990. The Unemployment Insurance (UI) system’s goal of providing income maintenance, and thus economic stability, will surely be put to a test this year. Policy issues relating to the program’s effectiveness as a countercyclical tool will also become increasingly important.

In this Chicago Fed Letter, we review the UI system’s role as a countercyclical tool and address concerns about the system that have developed over time, including how changes in the U.S. labor force over the last 55 years have affected the UI system and how underfunding of Employment Services has compounded the system’s problems. Where relevant, we present a regional perspective of these issues as they relate to the five states within the Seventh Federal Reserve District: Illinois, Indiana, Iowa, Michigan, and Wisconsin.

Does it work as intended?

The UI system was first authorized by the Social Security Act of 1935 to provide income to those workers who, through no fault of their own, are temporarily out of work. As originally envisioned, the system was supposed to come into play during economic downturns by providing a countercyclical balance, supporting the incomes of unemployed workers and thereby allowing them sufficient funds to seek new employment while preventing further deterioration in the nation’s aggregate demand for goods and services. But many critics of the system feel that it no longer serves its primary purpose.

According to the original design, the UI system accumulates sufficient funds during good times to pay benefits during bad times. This worked fairly well during the early years and through the early 1970s. The first big draw on the system occurred in the 1970s when unemployment rose sharply and high benefit payouts depleted individual state funds. Several states, including Illinois and Michigan in the Seventh District, found themselves in need of federal loan assistance. Critics of UI believe that the need for federal loans, coupled with the availability of federal loans, erodes the cyclical nature of the UI system.

Rather than accumulating adequate funds during good times, states came to rely on federal loans as a regular source of funds during downturns, according to this view.1

Until 1981, federal loans for UI were interest free. But increased loan activity in the late 1970s and early 1980s prompted the federal government to change the rules regarding federal loans, by requiring that loans be paid in the same fiscal year as borrowed to avoid interest.

States of the Seventh District were among the heaviest borrowers. Throughout the 1980s, unemployment in most of the District states was higher than the national average, forcing District states to borrow as their individual trust funds dwindled. The five District states accounted for roughly 37% to 46% of total federal loans outstanding from 1980 through 1987. As shown in figure 1, trust fund balances and loan activity at both the national and Seventh District level displayed a pattern of increased reliance on federal assistance.

1. District trust fund reserves and end-of-year loan balances ($000), 1980-90

*December 31 balance.
Source: Unemployment Insurance Financial Data, DOL/ETA.
Unemployment trust fund reserves* Outstanding loan balance year-end
Year IL IN IA MI WI % of U.S. IL IN IA MI WI % of U.S.
1980 66 231 115 209 271 7.7 984 0 0 842 0 36.6
1981 25 176 101 9 54 3.0 1,405 0 0 1,075 0 39.5
1982 0 63 0 0 0 0.8 2,069 0 63 2,186 413 44.5
1983 0 100 0.7 0 0.8 1.3 2,418 0 127 2,322 628 41.1
1984 0 245 0 0 124 3.2 1,707 0 38 1,666 534 41.6
1985 45 392 51 454 88 6.4 1,124 0 0 1,289 256 43.7
1986 474 436 145 880 68 9.9 889 0 0 1,121 0 41.8
1987 314 509 238 978 404 9.9 0 0 0 953 0 46.4
1988 824 634 427 982 756 11.4 0 0 0 782 0 1,00.0
1989 1,268 770 518 972 1,032 12.2 0 0 0 603 0 1,00.0
1990 1,459 879 575 740 1,210 11.6 0 0 0 418 0 1,00.0

Work disincentives

Some critics believe that the UI system fosters work disincentives. Economic theory suggests that any benefit system such as UI will lead to some degree of work disincentive. In essence, UI recipients can, in practice, collect benefits even if they could find a job. In other words, some unemployed workers will choose to delay re-entry into the labor force because they are collecting UI benefits. The extent to which this actually occurs remains a topic of heated debate.

Research on the disincentive effects of UI investigates the relationships between the UI system and variables such as unemployment duration, reservation wages, and the job-search behavior of benefit recipients. In a comprehensive analysis of the work disincentive effects of UI, the Upjohn Institute for Employment Research presented various work disincentive evidence, including analysis of similar or related programs, as well as direct evidence—empirical analysis of the UI system itself.2

The Upjohn study concluded that reviews of labor supply studies and social experiments, and comparison of the UI system to related transfer programs such as Aid for Dependent Children, were inconclusive. Examination of direct evidence of work disincentives was conducted through analysis of the ratio of benefits to prior wages (also called an earnings-replacement ratio) with duration of unemployment; demographic characteristics of exhaustees (people who have used up or exhausted their UI benefits) as compared to other claimants; patterns of post-exhaustion experience; and partial-benefits analysis (people who collect partial benefits because they chose part-time or reduced wages while seeking full-time employment).

Analysis of direct evidence tended to provide more support for the work disincentive theory. In summary, the Upjohn authors noted that the weight of the evidence suggests work disincentive effects do exist, but that there may be some economic and social benefits resulting from the work disincentive effects (such as better or more stable jobs arising from recipients’ ability to be more selective due to UI benefits), and that there are also some work incentive effects of the UI system.3 For example, marginal unemployed workers may continue to look for employment as a result of receiving UI benefits. In the absence of UI benefits, these workers may withdraw from the labor force.

The changing mix of the labor force

The changing industry mix in the U.S. may have also affected the UI system. Some economists believe that the evolution to a service economy has changed the demands on the UI system funds over the course of the business cycle. Service jobs tend to pay less, and, because of the larger number of service jobs, unemployment spells tend to be shorter. As a result, a shift toward a service economy means reduced weekly unemployment benefits for workers, because benefits are based on past earnings, and a shorter period of unemployment for each individual.

Changes in family demographics have also affected the UI system. In the 1930s, most families relied on one “breadwinner” as their sole source of income. Accordingly, the UI system as originally devised was supposed to provide income maintenance when the family’s sole earner is unemployed. A 1990 study by the Congressional Budget Office on family income of UI recipients looked at how changes in the number of wage earners in a family has compromised this aspect of the UI system.4,5 In particular, this study looked at family income of long-term UI recipients (defined as recipients who received at least four consecutive months of benefits). The CBO study showed that nearly 60% of long-term recipients were in families with at least one other person working and that family income averaged just under 80% of its level prior to UI benefits.

The implication is that while the UI system was designed to provide assistance to the family’s sole earner, it now supplements family income. However, this must be weighed against the fact that, according to the study, although few family incomes were below poverty levels prior to unemployment, 20% were below poverty levels while receiving UI benefits and 45% would have been below poverty levels in the absence of UI.

Declining coverage ratios

Those who believe that the UI system works well as an income-support program are concerned with the declines in both coverage, or the number of people eligible for benefits, as well as the share of wages replaced by UI benefits, or the earnings-replacement ratio. The heart of this issue is that states, through stricter eligibility requirements, are reducing the number of people eligible for benefits. Most analysts believe that this is a direct consequence of enhanced state effort to keep individual UI trust funds solvent. In addition, the actual level of benefits paid to recipients as a percent of former earnings has also declined. These actions, whether intentional or not, erode the countercyclical nature of UI by reducing the level of income replacement and thus the purchasing power of the unemployed.

One measure of declining coverage is the ratio of insured unemployment (people receiving UI benefits) to total unemployment. This ratio, called the IU/TU ratio, is presented for the U.S. and District states over the 1978-1989 period in figure 2. The IU/TU ratio for the entire U.S. has declined from 44.4% in 1980 to a low of 28.5% in 1984. Since then, the ratio showed a slight rebound to 30.8% in 1985, with a leveling off thereafter.

2. Insured as a proportion of total unemployment

Figure 2 is a line graph showing the IU/TU ratio for the U.S. and the Seventh District states from 1978 to 1989. In all locations, the ratio reached its highest peak around 1980 and declined until around 1984-85.
Source: Unemployment Insurance Data, DOL/ETA and Geographic Profile of Employment and Unemployment, DOL/BLS, various years.

District states have followed the general U.S. pattern of declining IU/TU ratios, with Michigan and Wisconsin showing higher levels of volatility than the rest. For example, in 1980 when the U.S. IU/TU ratio peaked at 44.4%, Michigan’s peaked at 52.7% and Wisconsin’s at 58%. Both states had substantial increases in total unemployment that year relative to the total U.S. increase. Another considerable deviance occurred in 1988 when Wisconsin’s IU/TU ratio jumped to 40% (from 33.6% in 1987) while the U.S. rate remained flat. Again, this variance can be explained by the fact that the state’s unemployment rate fell sharply from 6.1% to 4.3%, or nearly 30%, while the weekly insured level fell only 14%. What might be gleaned from this very small sample is that the U.S. figures contain a high degree of state variability and geographic dispersion. Nonetheless, the downward trend in coverage has been pervasive.

Funding Employment Services

Another concern about the UI system involves the funding of Employment Services. In concept and practice, UI programs are also charged with the responsibilities for speeding the temporarily unemployed back into the labor force. Funded by federal taxes on employers, activities of the Employment Services, which include job listings and counseling of unemployed workers, have been hampered over the 1980s by serious underfunding. For example, staff levels have declined from about 30,000 in 1980, to 17,000 today.6

In this instance, actions by the federal government may be directly responsible for diminished activities of the Employment Service. An estimated $2.5 billion in funds will have been accumulated but left unspent in the Employment Security Administrative Account by the end of this fiscal year.7 As part of the federal unified budget, unspent funds reduce the size of the federal deficit.


Changing demographics, industry mix, and state UI statutes have all contributed over time to create a UI system that is no longer countercyclical and may be counterproductive. Suggestions to change the system range from eliminating the system and replacing it with a needs-based system, to pumping more money into the system by raising the employee tax base, perhaps from $7,000 per covered employee up to the present Social Security base of $53,400.

The issues presented here are by no means new—they have been, in most cases, well documented and discussed in a variety of research by concerns such as the W.E. Upjohn Institute for Employment Research, the Congressional Budget Office, and the Department of Labor. In addition, the UI system’s problems are well-known to the public and legislators. The time has come for a rethinking of and resolution to the problems of the UI system. If the UI system no longer serves as a countercyclical tool or provides adequate income maintenance during unemployment, then we need to determine whether, and if so how, these goals can be accomplished.

MMI-Midwest Manufacturing Index: Current expansion

The Midwest Manufacturing Index is a line graph showing manufacturing activity in the Midwest and the U.S., indexed to 100 in January 1983. The MMI value in April of 1991 was 133.4 (compared with 129.6 the previous month and 132.9 the previous year). The USMI value was 131.3 (compared with 129.8 the previous month and 132.2 the previous year).
Note: The MMI and the USMI are composite indexes of 17 manufacturing industries and are derived from econometric models that estimate output from monthly hours worked and kilowatt hours data. For a discussion of the methodology, see “Reconsidering the Regional Manufacturing Indexes,” Economic Perspectives, Federal Reserve Bank of Chicago, Vol. XIII, No. 4, July/August 1989.

Manufacturing in the Midwest appeared to reach a low point this March, rebounding 2.9% from that trough in April (current data reflect the 1990 rebenchmarking). Industry performance was mixed, however, with roughly one-third of the 17 industries continuing to decline. Solid gains were posted by primary metals, machinery, and especially transportation equipment, coinciding with a pickup in auto production.

The region has closely followed the national pattern during the recession. The Midwest decline since mid-1990 was greater than the nation’s, largely because of the transportation equipment industry. But that industry should continue to be a major source of strength to the Midwest in the coming months.


1 On the other hand, money managers might argue that the accumulation of large balances is not prudent budget management because the dollars could be used for more current, pressing issues.

2 Munts, Raymond and Irwin Garfield, The work disincentive effects of unemployment insurance, The W.E. Upjohn Institute for Employment Research, September 1974.

3 Ibid., p. 48.

4 Family incomes of unemployment insurance recipients and the implications for extending benefits, Congressional Budget Office, February 1990.

5 While the CBO study did provide these results, its focus was to assess the economic condition of long-term unemployed workers both while receiving benefits and thereafter.

6 See Paula Duggan’s article entitled “Failing net,” Northeast Midwest Economic Review, May 6, 1991, for a more complete review of this issue.

7 Ibid., p.7.

Opinions expressed in this article are those of the author(s) and do not necessarily reflect the views of the Federal Reserve Bank of Chicago or the Federal Reserve System.

Having trouble accessing something on this page? Please send us an email and we will get back to you as quickly as we can.

Federal Reserve Bank of Chicago, 230 South LaSalle Street, Chicago, Illinois 60604-1413, USA. Tel. (312) 322-5322

Copyright © 2022. All rights reserved.

Please review our Privacy Policy | Legal Notices