Poverty being a multidimensional phenomenon invites the attention of polic makers and researchers in Pakistan to visualize it in non-conventional ways. There has been felt a dire need to shift the research endeavour from mere uni-dimensional estimation to multidimensional diagnostics so as to draw appropriate & workable policy inputs. As against the money metric approach, the latter accommodates all such dimensions as education, health, housing, public services, employment, thereby offers broader picture of economic development. This study aimed at identifying various socio-economic dimensions in the mapping of multidimensional poverty along with establishing cut-off points according to Millennium Development Goals (MDGs). The study also addressed the estimation of incidence, depth & severity of multidimensional poverty across regions and time by employing modified and adjusted FGT class of poverty. Similarly, poverty correlates have also been explored with the aid of probabilistic empirical specifications. On an overall basis, incidence of multidimensional poverty was estimated as 43.34 percent in 1998-99. With minor fluctuation, the value for 2007-08 was 38.31 percent. The decline in urban area was more pronounced than that of urban area across the time and space. The regional situation exposes that the poverty index was higher in the rural area, but it was interesting to note that the percentage decrease was substantial in rural areas as compared to the urban areas. This indicates that growth has been effectively translated to the rural poor during that specific decade. The provincial trends also followed the overall shifting over the time. It revealed from the regional scenario that Sindh urban was the only
region where incidence of multidimensional poverty increased overtime. The declining trend in Punjab province was higher than that of Sindh. Balochistan remained the worst of all in terms of reduction in multidimensional poverty. In terms of percentage contribution of incidence of multidimensional poverty to the overall country level poverty, Balochistan ranked number one followed by KPK, Sindh and Punjab during 1998-99. The urban contribution of Sindh in overall poverty was found to be the lowest followed by Punjab, KPK and Balochistan. But in 2007-08, the proportional contribution of Balochistan in the overall multidimensional poverty was the highest both in rural and urban areas. Similarly in each province there has been made further estimations at divisional level and thus identified the percentage contribution of each division to the overall poverty statistics in the respective provinces. For all the five years, separate multilogistic analysis indicate that probability of occurrence of poverty in a specific area was attributed to household size, household head age, education, child index, unemployment, dependency ratio, rural urban and provincial dummies.
The research necessitates the priority intervention in the poorest segments as well as highly deprived regions which are contributing largely in the poverty index, particularly in the rural sector of respective province and division. The overall findings are offering the government some policy lessons for constructing a comprehensive Poverty Reduction
Strategy Papers (PRSPs) while meeting World Bank standards. Any poverty alleviation policy should be designed keeping multidimensional poverty statistics in view rather than merely focusing of uni-dimensional results. Finally, region specific policies would be more effective rather than targeting the poor on an overall basis which create regional
inequalities – one of the major causes of violent conflict in the country.