Methodological issues of Poverty Measurement

Issues relating to defining poverty as lack of resources like income and wealth to a household or individual to meet the basic needs to maintain a reasonable standard of living in a given society have been addressed along with various approaches to measure poverty. Limitations of Multidimensional Poverty Index (MPI) including aggregation of the chosen indicators have been discussed. The paper suggests multiplicative aggregation of ratios of chosen indicators at current period and the base period to find an index of overall poverty status of a country at the C-th time period over the based period (OPIc0). The index OPIc0 involves no selection of weights, satisfies desirable properties with significantly reduced substitutability among the indicators and producing no bias for advantaged groups. Replacing the base period data by the national targets to remove all forms of poverty, the modified index will indicate how far the country is at the C-th time period from the goals set at national level. The plot of growth/decline of multidimensional poverty in terms of changes in OPIct for a country reflects effectiveness of policy measures adopted and can be used in projection of poverty for the country. The index OPIc0 facilitates to form two sub-indices: (i) New Multidimensional poverty index (INMPI) based on suitable indicators, in line with MPI without any cut-off marks leading to rejection of households/individuals and (ii) Transient poverty index for a period (ITPt ) based on chosen indicators like poverty spell, chronic and transient poverty and associated factors. Using longitudinal data, probability of movement into poverty and out of poverty can be estimated over time (pNP-P and pP-NP) and temporal poverty of a country during the period (t,t_p) can be better understand by P - Escape(t,t_p) and P - Stic.(t,t_p). Estimation of such probabilities along with economic trajectory of a country will help the planners to achieve the national goals.