UI Effects on Unemployment: Some Data on Competing Theories

• Published date: 01 September 1983
• Date: 01 September 1983
• Author: HARRY C. BENHAM
• DOI: http://doi.org/10.1111/j.1468-232X.1983.tb00944.x

RESEARCH

NOTES

HARRY

C.

BENHAM*

UI

Effects on Unemployment:

Some

Data

on Competing Theories

RECENT

RESEARCH

HAS

DISCLOSED

many empirically strong rela-

tionships between various parameters of the unemployment insurance (UI) system

and certain unemployment characteristics. However, these observed relationships

frequently are consistent with different behaviorial theories. The purpose of this note

is to utilize a unique data source to distinguish between search based and contracting

based theories of UI’s effect on unemployment.’ The results show a positive effect

for

the

benefit replacement ratio which is consistent

with

both theories. The contracting

based hypothesis receives additional support, however, in

the

existence of a significant

experience rating effect.

Theory

The

first of

the

two

predominant theories of UI’s effect on unemployment

is

based on

the

well established inverse relationship between search

costs

and reservation

wages in job search models.

As

UI benefits reduce search costs, theory would predict

higher reservation wages and longer duration. Kiefer and Neumann

(1979)

have

empirically verified

the

reservation wage effect, while Ehrenberg and Oaxaca

(1976)

and others have found evidence that UI benefits result in longer durations of joblessness.

Hamermesh

(1979, 1980)

and Burgess and Kingston

(1981)

have presented evidence

that

the

duration of benefit eligibility

also

influences unemployment duration. Indeed,

the evidence indicating that UI benefits tend to increase unemployment duration is

so

extensive that Tope1 and Welch

(1980)

consider the issue resolved.

The

second theory is rooted in Baily’s

(1974)

and Azariadis’

(1975)

work on contracting.

Firms and employees are viewed as entering into long-term agreements on wages,

hours, laypff, probabilities, and layoff durations, with wage premiums required to

compensate for higher layoff probabilities or longer layoff durations. UI benefits can

affect the terms of these agreements by reducing the necessary wage premiums. In

imperfectly experience-rated UI systems, the expected value of agreements with

layoffs increases, since firms can offer lower wage agreements and are not

fully

taxed

for the benefits paid.

As

the degree of experience rating is increased, agreements

with layoffs become less desirable and consequently fewer layoffs are optimal. In an

imperfectly experience-rated system, Baily

(1977)

and Feldstein

(1976)

argue that

*Assistant Professor

of

Economics, University

of

Oklahoma.

‘I

would like to thank an anonymous referee for helpful comments and suggestions.

~

INUUSTHIAL

RELATIONS,

Vol.

22,

No.

3

(Fall

1983).

01983

by

the Regents

of

the University of California.

0019/8fi7~83/1015/403/$10.00

403