As the editors of Let's Talk Development, we want to respond to questions raised recently in social media channels about use of 2011 International Comparison Program (ICP) as well as during events and discussions about poverty and measurements during the Annual Meetings of the World Bank and IMF last week.
The World Bank currently uses an international poverty line of $1.25 (per person per day) in 2005 prices to monitor global poverty. The process draws on several data sources, including the ICP. The most recent global and regional poverty estimates cover the period 1981-2011 and are available from the recently updated Povcalnet database; they are based on data from well over 1,000 household surveys, covering nearly all developing countries. The latest estimates have been published and explained in both the recent Policy Research Report and the Global Monitoring Report, published last week.
The release this past summer of the results from the most recent round of the ICP in 2011 highlighted the potential role for price data to alter the overall profile of global poverty. Whereas the 2005 ICP data indicated a significant increase in price levels in several low income countries compared to previous estimates (resulting in a corresponding increase in the estimates of poverty incidence for those countries), the 2011 PPP data suggest a drop in price levels for some countries.
While new rounds of the ICP improve our understanding of the differences in cost of living across countries, there are important questions regarding their application for poverty measurement. For example, how representative are the data collected of prices paid by the poor? How well do they represent rural and urban areas? Scholars have been debating these and other points; see, for example Deaton, Angus, and Aten, Bettina. 2014 “Trying to Understand the PPPs in ICP 2011: Why Are the Results so Different?” NBER Working Paper no. 20244; and Ravallion, Martin. 2014. “An Exploration of the International Comparison Program’s New Global Economic Landscape.” NBER Working Paper no. 20338.
It’s also important to keep in mind that the ICP is a extensive statistical operation spanning almost 200 countries, so answering these questions and understanding the application of the data for the purpose of poverty monitoring is by no means a trivial or simple exercise.
The challenge of when and how to adopt new estimates of PPPs for poverty measurement purposes is not new; careful review has occurred with previous releases of new PPP data. For example, the initial results of the 2005 round were released in December 2007 but were not used in the global poverty counts until August 2008. Similar prudence is warranted this time, too.
The initial 2011 ICP results were released in April 2014: The Bank’s poverty analysts will continue their intensive research and data validation, and anticipate making a decision about the use of the 2011 PPPs early next year.
Join the Conversation