Inequality and Growth: What Can the Data Say?

The attached PDF contains a summary of the article

Banerjee, A., Duflo, E. (2003), Journal of Economic Growth, Vol , pp. 267-299.

Abstract
This paper describes the correlations between inequality and the growth rates in cross-country data. Using non-parametric methods, we show that the growth rate is an inverted U-shaped function of net changes in inequality: changes in inequality (in any direction) are associated with reduced growth in the next period. The estimated relationship is robust to variations in control variables and estimation methods. This inverted U-curve is consistent with a simple political economy model but it could also reflect the nature of measurement errors, and, in general, efforts to interpret this evidence causally run into difficult identification problems. We show that this non-linearity is sufficient to explain why previous estimates of the relationship between the level of inequality and growth are so different from one another.

Data on inequality from Deininger and Squire (1996) high quality, cross-country and panel
structure; allows using advanced techniques on analysis. Results from previous OLS work
typically found negative relationship between growth and inequality,
Benhabib and Spiegel (1998), Forbes (2000): fixed effects estimates arguing that omitted
country specific effects bias OLS. Fixed effect approach yields positive relationship: increases
in inequality within same country promote growth.
Barro (2000): a 3SLS approach treating country-specific error terms finds no relationship;
after breaking up sample into more homogeneous, rich and poor subsamples, he finds
negative relationship for poor and positive relationship in rich country sample.
 causal interpretation of evidence not clear, as variations of inequality correlated with a
range of unobservable factors associated with growth; observation of data without imposition
of linear relationship eye opening; changes in inequality in either direction associated with
lower growth rates; same for relationship between growth rates and inequality lagged by one
period
non-linearity may explain why different variants of linear model (OLS; fixed effects,
random effects) have generated very different conclusions.

The relationship from Macroeconomic Data between inequality and growth is disputed. Relatively good data on inequality comesfrom Deininger and Squire (1996). They have a cross-country panel on inequality measures.  Results from previous OLS  cross-sectional work typically found negative relationship between growth and inequality. Benhabib and Spiegel (1998), Forbes (2000) work with the panel and use fixed effects estimatation. They argue that previously omitted country specific effects biases OLS. Their fixed effect approach yields positive relationship between inequality and growth.

Barro (2000) on the other hand uses a 3SLS approach treating country-specific error terms finds no relationship. However, after breaking up sample into more homogeneous, rich and poor subsamples, he finds negative relationship for poor and positive relationship in rich country sample. Clearly a causal interpretation of evidence not clear, as variations of inequality correlated with a range of unobservable factors associated with growth; observation of data without imposition of linear relationship eye opening; changes in inequality in either direction associated with lower growth rates; same for relationship between growth rates and inequality lagged by one period non-linearity may explain why different variants of linear model (OLS; fixed effects, random effects) have generated very different conclusions.

Read the rest…Inequality and Growth: What Can the Data Say?

Share and Enjoy: These icons link to social bookmarking sites where readers can share and discover new web pages.
  • MisterWong
  • Y!GG
  • Webnews
  • Digg
  • del.icio.us
  • StumbleUpon
  • Reddit

One thought on “Inequality and Growth: What Can the Data Say?”

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong> <pre lang="" line="" escaped="">

Login with Facebook: