Self-employment rates and project size vary greatly across countries. The main message of this paper is that these broad regularities
are consistent with an environment in which a common self-employment technology is available worldwide, but where (a) financial
intermediation costs and (b) alternatives in “paid” work differ greatly. Our model indicates that alternatives in paid work
are crucial for explaining self-employment
rates, whereas high financial intermediation costs primarily affect the
scale of projects. We also show that credit use is not informative for predicting either rates of self-employment or the scale
of self-employment projects.
Keywords Self-employment - Project scale - Development
JEL Classification J23 - E21 - D31
We thank Anne Villamil and Cristina DeNardi for helpful discussions. We also thank seminar participants at the 2007 SAET Meetings
(Kos, Greece), and Midwest Macro Meetings 2008 (Philadelphia), especially Juan Sanchez, Leo Martinez, and two anonymous referees
for thoughtful comments. Brian Gaines and Nadezhda Malysheva provided excellent research assistance. This was written when
Akyol was visiting the Federal Reserve Bank of Richmond. The views expressed are those of the authors, and do not necessarily
represent those of the Federal Reserve Bank of Richmond, or the Federal Reserve System.