Transcript Slide 1
Some Empirical Aspects of Entrepreneurship By David S. Evans and Linda S. Leighton Agata Narożnik Levan Bzhalava Introduction • 4.2 million people operate business on a full time basis. • They employ tenth of all wage workers • Self-employed has increase since mid-1970s Small Firms • Major source of industrial changes • Creating a disproportionate share of new jobs • Many states have programs to stimulate smallfirm formation Great Britain, France, Belgium, Netherlands Helping unemployed workers start business Determinants of selfemployment? Contents • I - Data description. • II - Statistics on self-employment entry and exit over the life cycle and report estimates. • III - Determinants of self-employments earnings • IV - Conclusion Data description • The National Longitudinal Survey (NLS) – White men between 14-24 in 1966 – Surveyed yearly between 1966 and 1971 and 1973, 1975, 1976, 1978, 1980 and 1981 – 1966 – 3918 respondents – 1981 – 2731 respondents Data description • The March Supplement to the Current Population Survey (CPS) for 1968-1987 – 150000 white men between age 18-65 – Information on the employment status for the survey year and for the previous Entry and Exit over the life cycle • Simple time-homognous Markov model – The probability that an individual will operate business T years after entering the labour force is: • e – probability of entering the self-employment • x – probability of exiting self-employmet Entry and Exit over the life cycle Entry and Exit over the life cycle Entry and Exit over the life cycle Entry and Exit over the life cycle Entering self-employment Expected utility of self employmet > Expected utility of wage work THAN Individual switchs to selfemployment Entering Self-employmet • The probability of switching into self-employment increases with net worth – assets (NLS), liquidity (CPS) • Wage experience is statisticaly insignificent • Individuals with low wages are more likely to switch • Individuals with longer job tenure are less likely to switch into self-employment • Individuals who have changed job frequently are more likely to switch Entering self-employmet • The individuals who have had prior selfemployment experience are more likely to switch • The effect of previuos unemployment is not consistent – 1980-1981 positive and significant – 1976-1978 negative and insignificant – 1978-1980 negative and insignificant Entry and Exit - Conclusion • Entry is time homogenous • Exit decreases sharply with time • Workers with low wages and a history of instability are more likely to switch to selfemployment holding assets and education constraints. III. Self Employment Selection and Earnings • Previous Works • Borjas and Bronars (1987) • Evans (1985) • Brock and Evans (1986) 1980 Census data • Rees and Shah (1985) U.K. data on small cross section • Blau (1985) Malaysian farmers Data Data problems of Previous works 1. No information on self-employment versus wage experiences 2. Sparse information on personal characteristics. NLS Data • Gives possibility to estimate a much more refined model and to investigate the effect of wage and self-employment experience on wage and self-employment earnings. Data • Consist 2,405 white man who were in the 1981 NLS survey. • A total of 272 individuals were deleted either because they held both wage and self-employment jobs or because information was inconsistent. Result of Probit Estimation Probability of being self-employed increases with • labor-market experience • Changing job frequently • Relatively more unemployment experience • More educated individuals in professional occupations • More internal locus of control • Having father manager Differences in Return • wage experience in self-employment (2.1%) • wage experience in wage work (5.6%) • Self-employment experience in selfemployment (4.6%) Differences in Return • For professional workers the returns to education are higher in self-employment than in wage work • Unemployment experience carries larger penalty in self-employment than in wage work Interpretation of the Differences • Human-capital accumulated through wage work is less valuable in self-employment than wage work. • Individual who switch into self-employment later in their careers (and who have thereby accumulated more wage experience) are relatively poorer wage worker IV. Conclusions • Probability of entering self-employment is independent of age or experience for the first 20 years of employment But • Individuals face liquidity constraints and have to accumulate assets to start business. • Discovering a business opportunities takes time. Older people might be more likely to have identified an opportunity but less likely to choose to exploit it. IV. Conclusions Determinants of switching from wage work to self-employment • Relatively low wage. • Frequently changing jobs. • Frequently experiencing unemployment as wage workers. • More internal locus of control Questions, Comments….