Growth Elasticity of Poverty

• Definitions and Concepts 
• Limitations 
• Notes / Extensions 
• Quick Results (with ADePT and STATA) 
• Data Requirements 
• Helpful Tips 
• Annotated Examples 
• References / Related Papers 

Definitions and Concepts

The growth elasticity of poverty indicates how effectively growth has translated into poverty reduction. The total growth elasticity of poverty is the percent change in poverty with respect to a one percent change in per capita GDP (or mean income or expenditure per capita):

where P is any of the Foster-Greer-Thorbecke (FGT) poverty measures but usually the headcount index, and Y is per capita GDP, income, or expenditure.

Since the total growth elasticity of poverty reflects many different factors such as changes in inequality, initial level of inequality, initial level of development, growth rate, and data issues, this seemingly straightforward indicator should be interpreted carefully (see p. 81 of Pro-Poor Growth in the 1990s).

To see how growth elasticities vary across sectors (states), one can estimate sectoral (state-specific) growth elasticities of poverty (e.g. for rural and urban sectors), as in Ravallion and Datt (2002).

Limitations

Notes / Extensions

Semi-elasticity of growth on poverty: This is an alternative measure of the sensitivity of poverty reduction to economic growth. Klasen and Misselhorn (2006) argue that percentage point changes in poverty reduction may be of more interest to policymakers and advocate for the use of the semi-elasticity measure. They point out that percent changes in poverty reduction (and the growth elasticity of poverty) can be easily misinterpreted. If the initial poverty headcount is relatively low, then a small absolute reduction in poverty can constitute a large percent change in poverty. For example, compare two countries with different initial poverty rates, one with an initial poverty rate of 6% and the other 60%. If they both reduce poverty by half, then one will have reduced poverty from 6% to 3% while the other from 60% to 30%. If they experienced the same GDP growth rates, the growth elasticity of poverty would be identical. They also point out that with growing countries, the growth elasticity of poverty tends to increase over time and may give the impression that growth is becoming more pro-poor.

Discrepancies between survey and national accounts data: It is not uncommon for the level of private consumption expenditures per capita derived from national accounts data to differ from the mean household expenditure per capita derived from household survey data. Growth rates estimated from these two sources can differ as well. When estimates differ substantially, one cannot presume that one is better than the other since both data sources are susceptible to errors, Thus, it is good practice to use both data sources when available and determine the extent of discrepancies. See Ravallion (2001) or Deaton and Kozel (2005) for a more detailed discussion of the potential sources of error and tests for systematic differences.

Quick Results with Stata

Stata: Elasticities can be estimated quickly by regressing the log of poverty on the log of the welfare. See the annotated examples for details using Stata; other econometrics or statistics packages can also be used.

Data Requirements

• Per capita GDP (from national accounts data) or mean per capita income or mean per capita expenditure (from household data) for at least two years.
• Poverty measures for the corresponding years.

Helpful Tips

• Despite the name "growth elasticity of poverty," remember that the elasticity is the percent change in poverty with respect to the percent change in GDP rather than the change in the growth rate of GDP.

Annotated Examples

Here are examples of the growth elasticity of poverty estimation using the Uganda sample data with notes explaining the command line specification and the output. The data sets and variables are defined in the Sample Data section.

Example 1: National Accounts Data

We can use national accounts data (i.e., per capita GDP in constant local currency) and any one of the poverty measures.  By dividing the percent changes in poverty by the percent changes in per capita GDP, we arrive at an elasticity of -0.8 and -3.1 for, respectively, the 1993-2003 and 2003-2006 periods.  So for every 1% increase in per capita GDP, poverty dropped by 0.8%.

Example 2: Household Data

We can also use household data to see how poverty responded to growth in mean per capita consumption.

First, estimate poverty measures and mean per capita consumption for each year.

. use ugahh92, clear
. sepov welfare [pw=iwe], povline(spline) strat(stratum) psu(ea)

Poverty measures for the variable welfare: Consumption per AE (constant UG shillings)

Survey mean estimation

. mean welfare [pw=iwe]

Mean estimation

. use ugahh05, clear
. sepov welfare [pw=iwe], povline(spline) strat(stratum) psu(ea)

Poverty measures for the variable welfare: Consumption per AE (constant UG shillings)

Survey mean estimation

. mean welfare [pw=iwe]

Mean estimation

Then, divide the percent change in poverty by the percent change in per capita consumption to calculate the elasticity.

We can also estimate the semi-elasticity of growth on poverty. For this example, we use the national accounts data, but the household survey data would be fine too. Instead of using percent changes in poverty, we use the absolute changes in poverty to estimate the growth semi-elasticity of poverty. The results indicate a semi-elasticity of -0.5 and -1.2 for, respectively, the periods 1993-2003 and 2003-2006. So for every 1% growth in per capita GDP per year, poverty dropped by 0.5 percentage points per year on average during 1993-2003 and by 1.2 percentage points during 2003-2006.

References / Related Papers

Ravallion, M. (1997). "Can High Inequality Development Countries Escape Absolute Poverty?" Economics Letters, 56(1).

(Link above is for Science Direct subscribers. Other versions are available: World Bank Policy Research Working Paper – No. 1775)

Ravallion, M. (2001). "Growth, Inequality and Poverty: Looking Beyond Averages." World Development 29(11): 1803-1815.

(Link above is for Science Direct subscribers. Other versions are available: World Bank Policy Research Working Paper – No. 2558)

Ravallion, M. (2004). "Pro-Poor Growth: A Primer." The World Bank, Policy Research Working Paper No. 3242.

Growth semi-elasticity of poverty

Klasen, S. and M. Misselhorn (2006). "Determinants of the Growth Semi-Elasticity of Poverty Reduction," Proceedings of the German Development Economics Conference, Berlin 2006 No. 15, Verein für Socialpolitik, Research Committee Development Economics.

Discrepancies between survey and national accounts data

Deaton, A. and V. Kozel (2005). "Data and Dogma: The Great Indian Poverty Debate," The World Bank Research Observer, 20(2), pp. 177-199.

(Link above is for World Bank Research Observer subscribers. Other versions are available: September 2004 paper) 

Ravallion, M (2003). "Measuring Aggregate Welfare in Developing Countries: How Well Do National Accounts and Surveys Agree?" Review of Economics and Statistics, 85(3), pp. 645-652.

(Link above is for subscribers. Other versions are available: World Bank Policy Research Working Paper # 2665)

Sectoral elasticities of poverty

Ravallion, M. and G. Datt (2002). “Why has Economic Growth been more Pro-Poor in Some States of India than Others?” Journal of Development Economics, 68(2), pp. 381-400.

(Link above is for Science Direct subscribers. Other versions are available: World Bank Policy Research Working Paper # 2663) 

Growth-Inequality Decomposition | Sectoral Poverty Decomposition | Growth Incidence Curve |
Rate of Pro-Poor Growth | Growth Elasticity of Poverty |
Installation of Stata ado files | Sample data

Back to Measuring the Growth-Poverty Link