The Making of Pro‐Poor Growth
| Published date | 01 November 2011 |
| Author | Fabrizio Carmignani |
| Date | 01 November 2011 |
| DOI | http://doi.org/10.1111/j.1467-9485.2011.00563.x |
THE MAKING OF PRO-POOR GROWTH
Fabrizio Carmignani*
ABSTRACT
A system of three endogenous equations is used to estimate the determinants of
poverty dynamics. The system incorporates: (i) the direct effect of growth and
income inequality on poverty, (ii) the feedback effect of poverty on inequality
and growth, and (iii) different channels through which economic policies can
contribute to poverty reduction. Results suggest that countries tend to move
towards one of two possible equilibria. The positive (virtuous) equilibrium is
characterized by fast growth, decreasing inequality, and rapid poverty reduction.
The negative (vicious) equilibrium involves slow (or even negative) growth,
sharpening inequalities, and resilient poverty. The policy mix is critical in deter-
mining to which of the two equilibria a country converges.
II
NTRODUCTION
On average, economic growth reduces poverty, and there is a voluminous
body of empirical research to support this statement. However, the dispersion
around the average is wide and the elasticity of poverty with respect to eco-
nomic growth varies considerably across countries and over time. The chal-
lenge for development economists and policymakers is to understand to what
extent economic policies contribute to making growth more or less pro-poor
1
This paper takes up the challenge from a macroeconometric perspective: a sys-
tem of three endogenous equations is estimated to disentangle the different
channels through which economic policy variables affect poverty dynamics.
The specific variables that are investigated capture some of the key dimensions
of the economic policy space in developing countries: openness to interna-
tional trade, macroeconomic stabilization, financial development, government
size, and investment in human capital. The effect of institutional quality is
also studied.
*The University of Queensland
1
For the purpose of this paper, saying that growth is more or less pro-poor is equivalent
to saying that poverty is more or less responsive to economic growth. Responsiveness is in
turn associated with a notion of elasticity. Letting Dyand Dp, respectively, denote the per-
centage change in per-capita income (or consumption) and poverty headcount, then the mar-
ginal effect of Dyon Dpgives a measure of the responsiveness of poverty to growth.
Scottish Journal of Political Economy, Vol. 58, No. 5, November2011
©2011 The Author. Scottish Journal of Political Economy ©2011 Scottish Economic Society. Published by Blackwell
Publishing Ltd, 9600 Garsington Road, Oxford, OX4 2DQ, UK and 350 Main St, Malden, MA, 02148, USA
656
There is a fast-growing literature on the relationship between growth and
poverty. Evidence of a significant poverty-reducing effect of economic growth
goes back to Fields (1989) and Squire (1993), who estimate income elastici-
ties of poverty reduction well in excess of two. Similar findings have been
reported by Bruno et al. (1998) and Adams (2004). In two companion con-
tributions, Dollar and Kraay (2002, 2004) show that the elasticity of mean
income of the poor with respect to average mean income is not significantly
different from one, meaning that growth is unambiguously good for the
poor. Kraay (2006) provides additional evidence that most of the observed
cross-country variation in the dynamics of poverty is due to the variation in
the rate of average per-capita income growth. Ravallion (1997, 2001, 2004),
Bourguignon (2003), Epaulard (2003), Kakwani et al. (2004), Mosley et al.
(2004), and Kalwij and Verschoor (2007) find that the growth elasticity of
poverty significantly differs across countries and that these differences are
largely explained by the interaction between growth and income inequality.
In a recent contribution, Loayza and Raddatz (2010) take a complementary
perspective and show that the sectoral composition of growth affects its
capacity to reduce poverty.
This paper extends the existing literature by exploring the role of economic
policies in shaping the growth–inequality–poverty relationship. The theoretical
prior of the paper is that economic policies can affect poverty in three ways:
(i) by determining the rate of economic growth, (ii) by determining the
dynamics of income inequality, and (iii) by determining the responsiveness of
poverty to growth for any given rate of growth and change in inequality.
These three channels are jointly estimated using a system of three endogenous
equations. The paper is therefore related to the literature on the simultaneous
evolution of growth and inequality, although the addition of a poverty equa-
tion constitutes a significant innovation relative to the previous studies of
Lundberg and Squire (2003) and Huang et al. (2009). The use of a system of
equations also makes the paper quite different from Chhibber and Nayyar
(2008), who explore the impact of economic variables on the growth elasticity
of poverty within a single-equation framework.
The empirical analysis of this paper draws on the statistical information
available from POVCALNET. This is the on-line tool for poverty measurement
developed by the Development Research Group of the World Bank. POVCAL
NET reports measures of poverty and inequality based on a collection of
nearly 700 randomly selected households in 116 countries.
2
Even though pov-
erty also exists in other countries, including high-income economies, which are
not featured in this collection, POVCALNET covers a relatively large and repre-
sentative sample of the household population in developing countries. It is
however important to stress that in POVCALNET consumption-based surveys
are mixed with income-based surveys. Therefore, poverty (and inequality)
measures are not always determined from the same welfare concept. To deal
2
The POVCALNET software, the complete database, and the background documentation are
freely accessible on-line at: http://econ.worldbank.org/povcalnet.
THE MAKING OF PRO-POOR GROWTH 657
Scottish Journal of Political Economy
©2011 The Author. Scottish Journal of Political Economy ©2011 Scottish Economic Society
To continue reading
Request your trial