CPI Bias and its Implications for Poverty Reduction in Africa.

International poverty estimates for countries in Africa commonly rely on national consumer price indexes to adjust trends in nominal consumption over time for changes in the cost of living. However, the consumer price index is subject to various types of measurement bias. This paper uses Engel curve estimations to assess bias in the consumer price index and its implications for estimated poverty trends. The results suggest that in 13 of 16 Sub-Saharan African countries in this study, poverty reduction may be understated because of consumer price index bias. With correction of consumer price index bias, poverty in these countries could fall between 0.4 and 5.2 percentage points per year faster than currently thought. For two countries, however, the paper finds the opposite trend. There is no statistically significant change in poverty patterns after adjusting for consumer price index bias for only one country. [ABSTRACT FROM AUTHOR]