MUSHTAQ GHUMMAN

ISLAMABAD: State Bank of Pakistan (SBP) has sought feeder-wise data of electricity consumers to be used to project growth statistics and other research on its models based on theft data.

In a letter to Secretary Power Division, Deputy Governor SBP Murtaza Syed has explained that the primary mandate of SBP is to ensure price stability besides ensuring financial sector stability and supporting Government’s general economic policies in fostering economic growth. Data on aggregate demand or GDP are of paramount importance for the formulation and conduct of monetary policy. However, the frequency of GDP estimates is annual which is too low.

To make high-frequency estimation of economic growth on regular basis, SBP needs to access power consumption data at consumer and feeder-level in machine readable format from PITC on regular basis.

For this purpose, SBP has requested Power Division to facilitate it in getting the desired information which will remain confidential.

According to the Bank, PBS compiles data on GDP for Pakistan on annual basis. This frequency is too low for policy conduct. Further, the said GDP number is not disaggregated at a sub-national level which renders the observance and analysis of regional dynamics in the economy almost impossible. In this context, it is important to have a data series of GDP which is of at least quarterly frequency and which is disaggregated at district/ tehsil level. It would be an extremely rich dataset that can be used in all areas of economic research.

SBP team worked on the project of compilation of GNI series through various machine learning models using datasets such as inter alia, nightlights (from NASA), atmospheric observation (from NASA), prices and last but not least, electricity consumption data. This regression allowed SBP to predict GNI for a given tehsil every month from 2012 onwards.

As part of GNI computation, SBP would be able to use this data in other research projects pertaining to the overall economy in general and power sector in particular. This involves, among other things, estimation of price elasticity of consumption, computation of an economic activity index, credit situation in the power sector and other research on electricity theft using feeders and subsequent adjustment of SBP models based on theft data.