70 the following table presents information on the

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: the crisis. As a crisis itself lasts typically for one or two years, a country may actually suffer through a period of 4‐5 years before it is able to get back to its position before the fall. The last row presents the March 2009 projections for per capita GDP for 2009. The GDP growth forecasts are taken from the Consensus Forecasts and these were converted to GDP per capita projections using population data from the World Development Indicators. As is evident from the table, the projected losses of GDP per capita in 2009 for Mexico and Ecuador could already be considered large by historical standards. In these cases, it may take three years to recover the losses in poverty. Estimating the Impact of the Economic Crisis on Poverty The previous section simply presented the information on past relations between declines in per capita GDP and past changes in poverty. This information can help frame the expectations for the likely change in poverty. An alternative approach to considering what might be the effect of a given change in per capita GDP is to estimate the elasticity of changes in poverty to changes in per capita GDP and then use the estimated elasticities to predict the change in poverty for a specific projected growth rate for per capita GDP. Poverty elasticities were estimated using a specification similar to Ravallion (1995)105 and Adams (2004)106. It should be noted that these elasticities make use of the periods of positive increase in per capita GDP growth and declines of poverty. We did estimate the elasticities with a spline to allow the coefficient on GDP growth to take on a different value depending on whether growth was positive or negative. However, the difference was not statistically significant. These results are presented in Annex 3. We also estimated elasticities using the POVCALNET data instead of the SEDLAC data and the differences were not very great. Table 4 reports the different elasticities and the significance level of the estimated elasticities using the SEDLAC data. Table 4 Elasticities of Changes in Poverty Measure with changes in per capita GDP Poverty Line National Extreme* National Moderate* Elasticity ‐2.63 ‐1.62 P value 0.00 0.00 Note: (*) SEDLAC data; std errors clustered at country level; population weighted point estimates; controls: Log(Gini) and time trend. The estimated elasticities presented in Table 4 can be used to simulate the impact of the economic crisis on poverty in LAC. This is done by taking forecasted rates of growth of GDP, converting them into rates of growth of GDP per capita and using the elasticities to generate an estimated effect on poverty. The forecast economic growth rates are taken from the March 16th LAC Consensus Forecasts, which compile 105 Ravallion (1995) used the private consumption component from national accounts. Since our main objective here was to use these elasticities to simulate the impact of changes on GDP per capita, we chose to follow the approach of Adams (2004) and oth...
View Full Document

This document was uploaded on 11/14/2013.

Ask a homework question - tutors are online