2001年-世界发展银行全球_Is_Inequality_Bad_for_Business___A_Nonlinear_Microeconomic_Model_of_Wealth_Effects_on_Self-Employment_34页_1mb

Summary of "Is Inequality Bad for Business?"

A Nonlinear Microeconomic Model of Wealth Effects on Self-Employment

Core Content

This working paper investigates the relationship between wealth inequality and self-employment, focusing on return migrants in Tunisia. The study challenges the common assumption that wealth inequality is inherently detrimental to business activity and explores the nonlinear effects of wealth on entrepreneurship.

Main Findings

• Wealth and Entrepreneurship:

  • Higher wealth inequality reduces the level of new business activity, but the effect is relatively small.
  • The aggregate number of business start-ups is an increasing function of aggregate wealth but a decreasing function of wealth inequality.
  • At any given mean wealth level, greater initial inequality leads to a lower rate of new business start-ups.

• Liquidity Constraints and Diminishing Returns:

  • The relationship between wealth and self-employment is influenced by liquidity constraints and diminishing returns to capital.
  • At low wealth levels, the relationship is convex (increasing returns), but becomes concave (diminishing returns) as wealth increases.
  • There is a unique point of inflection where the curvature changes, implying that wealth distribution has a nonlinear effect on entrepreneurship.

• Empirical Ambiguity:

  • Theoretical models suggest that wealth distribution can have both positive and negative effects on entrepreneurship, depending on the wealth level.
  • Redistributing wealth from those above the inflection point to those below may have ambiguous effects on the total number of business start-ups.

Key Insights

• Credit Market Failures:

  • Credit market failures are assumed to play a role in limiting entrepreneurship, especially for the poor.
  • Wealth is critical for accessing credit, and the ability to borrow depends on an individual's wealth level.

• Nonlinear Wealth Effects:

  • The wealth effect on self-employment is nonlinear, with increasing returns at low wealth and diminishing returns at higher levels.
  • The paper uses nonparametric regression to capture this nonlinearity, as a linear form is unlikely to be appropriate.

• Empirical Methodology:

  • The authors use partial linear models and nonparametric regression to estimate the wealth effect on entrepreneurship.
  • They introduce a specification test to assess whether individual characteristics interact with wealth in a non-separable way.
  • The test involves first-order Taylor series expansion and interaction terms to determine if the wealth effect is independent of other variables.

• Data and Sample:

  • The study uses micro data from a 1989 survey of return migrants in Tunisia.
  • Return migrants are considered a more homogeneous group than the general population, making them a suitable sample for analyzing wealth effects.
  • The paper focuses on male return migrants who intend to stay in Tunisia indefinitely.

• Descriptive Statistics:

  • Self-employed return migrants had significantly higher savings (1086 dinars) compared to salaried return migrants (442 dinars) and the full sample (466 dinars).
  • Most self-employment projects are funded by personal savings, with only a small percentage (12.4%) receiving extra funds from special programs.
  • No one relied on formal bank credit, and many cited difficulty in accessing credit markets as a major obstacle.

Implications

• Redistribution and Business Activity:

  • The findings do not support the idea that wealth redistribution is an effective way to stimulate entrepreneurship.
  • Instead, the authors suggest that reducing liquidity constraints might be a more promising intervention.

• Policy Recommendations:

  • More research is needed on interventions that reduce liquidity constraints for potential entrepreneurs.
  • The paper contributes to the broader understanding of how wealth distribution influences macroeconomic activity and occupational structure.

Conclusion

• The effect of wealth inequality on business activity is small and nonlinear, depending on the level of wealth.
• The model suggests that inequality can be good or bad for entrepreneurship, depending on whether the affected individuals are above or below the inflection point.
• The empirical evidence supports the theoretical framework, but the size of the effect is not large enough to justify widespread redistribution as a policy tool.

Methodology Summary

• The authors use nonparametric regression to estimate the wealth effect on self-employment.
• They apply higher-order differencing to improve efficiency.
• A specification test is introduced to check for non-separable effects of wealth and individual characteristics.
• The paper also discusses potential selection bias and how it is addressed through the inclusion of inverse Mills ratio.

References and Contact

• Authors: Alice Mesnard, Martin Ravallion
• Affiliation: World Bank, Development Research Group, Poverty and Human Resources
• Contact:
  • Patricia Sader: psader@worldbank.org
  • Martin Ravallion: mgravallion@worldbank.org
• Date: January 2001
• Pages: 28 pages