The measurement of transient poverty: Theory and application to Pakistan

The present paper investigates the measurement of transient poverty when each person's welfare level fluctuates due to exogenous risk. The paper namely characterizes the sensitivity of transient/chronic poverty decomposition with respect to the poverty line and to the expected welfare level so that the decomposition analysis will be based on solid theoretical foundations and be robust empirically. Theoretical results show that poverty measures associated with prudent risk preferences perform better than other measures in assuring that the value of transient poverty increases with the depth of chronic poverty and that the decomposition is not highly sensitive to the poverty line. Poverty measures such as those associated with constant relative risk aversion are thus superior to popular Foster–Greer–Thorbecke (FGT) measures such as headcount, poverty gap, and squared poverty gap indices. These theoretical arguments are confirmed empirically by the application of the decomposition to a two-period household panel dataset from rural Pakistan. The relative magnitudes of transient versus chronic poverty are more robust to changes in the poverty line when poverty measures associated with constant relative risk aversion are used than when FGT poverty measures are used.