Explaining Poverty and Income Inequality in Pakistan: The Long-Run Role of Education, Employment and Macroeconomic Factors

Abstract

This study examines the long-run and short-run determinants of poverty and income inequality in Pakistan during the period 2000–2026 with special reference to the role of secondary education Enrollment (EDUSEC) and employment. Two equations are estimated in OLS Error Correction Model (ECM) framework using annual time-series data of World Bank World Development Indicators. The first one is a poverty equation (Model 1) and the second one is an income-inequality equation (Model 2, measured by GINI coefficient). The Augmented Dickey-Fuller (ADF) unit root tests confirm the presence of a mixture of I(0) and I(1) variables which satisfy the condition for ARDL bounds testing. Although the ARDL bounds tests resulted inconclusive F-statistics, cointegration is confirmed by the negative and significant error correction term (ECT) in both ECMs. The four estimated equations pass a battery of diagnostic tests for the absence of serial correlation, heteroskedasticity and non-normality in the residuals. In the poverty equation, secondary school Enrollment has a large and highly significant negative effect on poverty (coefficient = −1.320, p < 0.001), confirming that expanding access to education is the single most powerful long-run lever for poverty reduction in Pakistan. The GINI coefficient enters negatively and significantly (coefficient = -3.413, p < 0.001), trade openness has a small negative effect on poverty at the 10% significance level, while access to credit has a positive effect on poverty, contrary to expectation. Employment is retained as an important variable in the model but is statistically insignificant. This is a reflection of the near complete stagnation in Pakistan’s aggregate employment share over this period rather than the irrelevance of employment per se. In the inequality equation, education again has a significant reducing effect on GINI (coefficient = −0.325, p < 0.001), as does falling poverty (coefficient = −0.132, p < 0.001), confirming a bidirectional poverty-inequality nexus. Inflation decreases inequality (−0.055, p < 0.05), FDI increases it significantly (0.984, p < 0.001) and a higher population growth decreases it (−1.459, p < 0.05). The ECT coefficients are −0.530 (p < 0.01) and −0.681 (p < 0.05) for Models 1 and 2, respectively, implying adjustment speeds of 53 and 68 % per year towards the long-run equilibrium. The policy recommendations include emphasis on universal secondary education, reform of targeted credit delivery, FDI regulation with emphasis on quality and investment in labor-intensive employment.