Frontiers of Entrepreneurship Research
1996 Edition
SUMMARIES

Back to Index96
Order hard copy editions of Frontiers of Entrepreneurship Research by mail


WHY DOES THE SELF-EMPLOYMENT RATE VARY ACROSS COUNTRIES AND OVER TIME?


Zoltan J. Acs 

University of Baltimore
Robert G. Merrick School of Business 1420 North Charles Street
Baltimore, MD 21201

Telephone:(410) 837-5722

Fax :(410) 837-5012

Principal Topics

Several lines of research have found that something happened to the centuries-old trend towards larger businesses: depending upon the measure of business size examined, the trend decelerated, ceased or reversed itself sometime between the late 1960's and late 1970's . An important aspect of the trend away from larger businesses, at least in the United States, is the percent of the labor force that is self-employed.

Despite a burgeoning literature on the determinants of self-employment and entrepreneurship, there has been little research into whether the increase in the self-employment rate is unique to the United States. It turns out that it is not. However, there is a tremendous diversity in the level and time-series pattern of self-employment. That diversity is the subject of this paper. It raises two questions. (1) Why is there such diversity of the self-employment rates across countries? Does this diversity result from cultural, institutional, or economic factors? (2) Why has self-employment declined in some countries and risen in others?

Methods

We define the self-employment rate as the percent of the nonagricultural labor force that is working for themselves. We calculate the self-employment rate for 1966-1990 from data obtained from the Organization for Economic Cooperation and Development Labor Force Statistics and from the International Labour Organization's Yearbook of Labour Statistics. Statistical techniques are used to explore six possible sources of inter- and intra-country variations: (1) the stage of economic development; (2) the bias of technological change; (3) changes in industry composition; (4) demographic characteristics; (5) unemployment; and (6) cultural factors.

Data were available for most of the countries and most of the years on self-employment, per-capita gross national product, manufacturing value added, service value added, and the female labor-force participation rate. We have also included variables on technological change and unemployment.

Major findings

We find that the self-employment rate decreases with increases in per capita gross national product, female labor-force participation, and the relative importance of manufacturing increases with the relative importance of services. One problem with looking only at regressions among the OECD countries is that none of them represent the earlier stages of development that are present among the LDC data set. For the LDC countries self-employment varies negatively with per--capita gross national product, negatively with manufacturing value added as a percent of gross national product and positively with service value added as a percent of gross national product. Female labor force participation was insignificant.

The variables for high technology and unemployment are negative for the OECD countries. That is after controlling for cross national differences, the self-employment rate is lower when unemployment is higher, and when high technology industries are relatively more important. As for the cultural factors, we find that self-employment is positively correlated with the uncertainty avoidance index, and negatively correlated with the individualism index.

Implications

A major explanation for the diversity of self-employment is the stage of economic development. The implications of the study is that economic development is an extremely powerful force behind the secular decline in self-employment. The convergence of several factors -notable the decline of heavy manufacturing, the growth of services, and possible, the increase in unemployment - in the 1970's tended to stem the decline in the self-employment rate. Self-employment will most likely continue its downward trend as per-capita wealth increases.

Back to Index96