Working Paper: NBER ID: w15433
Authors: Maxim Pinkovskiy; Xavier Sala-i-Martin
Abstract: We use a parametric method to estimate the income distribution for 191 countries between 1970 and 2006. We estimate the World Distribution of Income and estimate poverty rates, poverty counts and various measures of income inequality and welfare. Using the official $1/day line, we estimate that world poverty rates have fallen by 80% from 0.268 in 1970 to 0.054 in 2006. The corresponding total number of poor has fallen from 403 million in 1970 to 152 million in 2006. Our estimates of the global poverty count in 2006 are much smaller than found by other researchers. We also find similar reductions in poverty if we use other poverty lines. We find that various measures of global inequality have declined substantially and measures of global welfare increased by somewhere between 128% and 145%. We analyze poverty in various regions. Finally, we show that our results are robust to a battery of sensitivity tests involving functional forms, data sources for the largest countries, methods of interpolating and extrapolating missing data, and dealing with survey misreporting.
Keywords: income distribution; poverty; inequality; welfare
JEL Codes: F01; O1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| parametric estimation method (C51) | understanding the world distribution of income (D31) |
| parametric estimation method (C51) | world poverty rates fell by 80% (F63) |
| economic growth (O49) | total number of poor decreased from 403 million to 152 million (F63) |
| changes in income distribution (D31) | decline in various measures of global inequality (F62) |
| lognormal distribution (C46) | validity of findings (C90) |
| sensitivity analyses (D79) | reinforce causal relationships (C32) |