Household Poverty and Welfare in Pakistan: Evidence from Pooled HIES Data 2005-2014

Abstract

Aim of the Study: This study examines household poverty and welfare in Pakistan using five pooled repeated cross-sectional rounds of the Household Integrated Economic Survey for 2005-06, 2007-08, 2010-11, 2011-12, and 2013-14. The household is treated as the primary unit of analysis, and expenditure per adult equivalent is used as the principal monetary welfare indicator. Poverty status is defined by comparing household expenditure per adult equivalent with the relevant poverty line for each survey year. The study first estimates poverty headcount ratios to describe changes in poverty incidence across survey rounds and then examines the socioeconomic correlates of poverty status and welfare among poor households.

Methodology: Since the welfare equation is estimated only for households classified as poor, the empirical strategy uses a logit-based two-stage selection-correction approach to account for possible non-random selection into this subsample. The first stage models the probability that a household is classified as poor, while the second stage estimates log expenditure per adult equivalent among poor households after including the generated selection-correction term.

Results: The results show a declining poverty headcount under the comparable poverty-line approach. The second-stage results indicate that education, asset value, access to safe water, sanitation, and electricity are positively associated with log expenditure per adult equivalent among poor households. Household size, age of the household head, food insecurity, rural residence, and later-survey-year indicators are negatively associated with welfare among poor households.

Conclusion: The findings highlight the importance of distinguishing poverty incidence from welfare variation within the poor-household subsample.