Abstract
The study estimates multidimensional poverty in Pakistan following Alkire-Foster methodology. The analysis is based on Pakistan Social and Living Standard Measurement Survey 2004-05 and 2014-15. The study adopts expert opinion weights, frequency-based weights and equal weights for the provision of estimates of MPI at national and provincial levels. The results show that the estimates of MPI range from 14% to 20% at national level and these estimates are quite sensitive to the choice of weights. Whereas, equal weights always underestimate the magnitude of poverty. Moreover, the inter-temporal analysis of poverty reveals that the intensity of poverty has lower contribution in the reduction of multidimensional poverty in Pakistan. Therefore, the deprived regions of the country should be focused separately (especially the deprived districts of Balochistan) to target poverty. Besides, the regional allocation of resources can be made according to the intensity of poverty. The study concludes that the measurement of poverty is a complex phenomenon and it is quite sensitive to the choice of weights. So, the researcher should be careful about the choice of weighting scheme while providing estimates of multidimensional poverty.