On the nature of micro-entrepreneurship: evidence from Argentina

Abstract

We analyse the nature of micro-entrepreneurship in Argentina. We focus on whether the sector resembles its counterpart in industrialized countries, characterized by the risk-taking nature of the entrepreneurial activity, or if it is the result of labour market distortions and disguised unemployment, as in the dual economy hypothesis. Our results suggest a segmentation of the micro-entrepreneur sector. Both young uneducated and middle aged highly educated salaried workers have the highest likelihood of becoming entrepreneurs. However, the first segment has a high probability of becoming own-account workers, while the probability of becoming micro-entrepreneurs with employees is strictly increasing in both age and education. Moreover, the probability of entrepreneur failure (as measured by the transition to the salaried sector) has an inverted U shape, implying that both high and low skill individuals are more likely to remain entrepreneurs.

Acknowledgement

Financial support for this article was provided by The Tinker Foundation. We are grateful to Werner Baer and William Maloney for their guidance and to an anonymous referee for his constructive comments and suggestions.

Notes

¹ We define the sector as own-account workers and owners of firms with less than 16 employees.

² The literature refers to entry when individuals enter the entrepreneurial activity from other labour status. Conversely, it refers to exit when individuals leave the sector for other labour status.

³ Different terminologies have been used for this portion of the labour force. Todaro (1969) called it ‘urban traditional’, Santos (1979) the ‘lower circuit’, McGee (1971) ‘protoproletariat’, Cole and Sanders (1985) ‘urban subsistence’.

⁴ Some exceptions are Moore and Mueller (2002) and Tervo (2006) who find that men are more likely to be self-employed than women.

⁵ We will use the terms self-employed and micro-entrepreneur interchangeably, to refer to those individuals who are own-account workers or owners of a small firm (less than 16 employees). Own-account workers can be considered as the most basic firm structure.

⁶ We do not pursue identification of the parameters in the earnings equation.

⁷ We are grateful to an anonymous referee for this suggestion.

⁸ Although longer series are available, there are serious problems identifying individuals over time for regions other than Gran Buenos Aires.

⁹ However, we also use information from the wave in between (August) to observe individuals who experience unemployment. See below on the construction of the variable Lost Job.

¹⁰ Logit and probit models gave identical results.

¹¹ Wage effects are directly captured by the variable Log of hourly wage.

¹² We are also grateful to an anonymous referee for pointed this specification.

¹³ This group is sufficiently small to be accurately captured by a typical household survey.