How Should We Measure Poverty in a Changing World? Methodological Issues and Chinese Case Study


  • We thank Nan Geng, Zhouran Zhou, Lynn Lethbridge, and Cheryl Stewart for their excellent assistance, SSHRC for its financial support (Grant 410-2001-0747), and anonymous referees, Patricio Aroca, Mark Brenner, Björn Gustafsson, Min-Dong Lee, Shi Li, Zhicheng Liang, Xin Meng, Tony Shorrocks, Terry Sicular, Guanghua Wan, Meiyan Wang, Xiaolu Wang, Xiaobo Zhang, Yin Zhang, and other participants at the UN's conferences (2004 in Manila and 2005 in Helsinki) for their constructive comments.

* Xu: Department of Economics, Dalhousie University, Halifax, NS, Canada B3H 3J5. Tel: (902) 494-2026; Fax: (902) 494-6917; E-mail: Osberg:


This study asks whether, in a rapidly changing world, the estimated proportion of the world's population with income below US$1 (adjusted according to purchasing power parity) per day is still a good measure of trends in poverty. It argues that strong economic growth in nations such as China implies that the commonly accepted international poverty line definition of one half median national equivalent income is increasingly relevant and that poverty intensity (the normalized deficit or Foster–Greer–Thorbecke (FGT) index of order one) is a better summary index. This index has a convenient graphical representation—the “poverty box”. Using the proposed poverty line and the example of ranking the level of rural poverty in Chinese provinces, the study demonstrates how poverty intensity replicates the poverty rankings of the Sen family of poverty indices and captures most of the information content of higher-order FGT indices.