Some Issues In The National Income Accounts Of Pakistan : rebasing, Quarterly And Provincial Accounts And Growth Accounting
Author: Muhammad Farooq Arby

This dissertation takes on some important issues related with national accounts of Pakistan including (a) re-estimation of past series of national accounts (prior to the year 1999-00) to make it consistent with the new official series for years 1999-00 onward at new base year prices; (b) quarterisation of annual series of national accounts to remove one of the major stumbling block in research by making available high frequency data, (c) estimation of provincial accounts with new base year prices, and (d) estimation the contribution of total factor productivity (TFP) to economic growth with the new series of gross domestic product and its sub-sectors. Thus by embarking upon the above mentioned issues, the dissertation contributes to the economic literature in the following respects: i) It provides a new set of national accounts at 1999-00 prices as well as at current prices for a period from 1970-71 to date consistent with the new official estimates for recent years. ii) It gives quarterly data of GDP and all its sectors/sub-sectors both at constant prices of 1999-00 and at current prices.1 A by-product of this exercise is quarterly GDP deflator (with 1999-00=100) which was earlier not available in Pakistan. iii) It presents estimates of provincial GDP and all its sectors/sub-sectors covering all the provinces in a consistent framework.2 iv) It provides series of gross fixed capital formation and capital stock estimated at disaggregated level and at prices of 1999-00. v) It estimates contribution of TFP, capital and labour to the growth of GDP and its sectors/sub-sectors (growth rates of new series at 1999-00 prices). In its attempt to re-estimate the previous series at new base of 1999-00, the dissertation followed, to the extent possible, the same methodology as of the Federal 1 Earlier work on quarterisation include Bengaliwala (1995) and Kemal and Arby (2004), however, both are at old base year prices of 1980-81 and available only at constant prices. 2 Earlier work on provincialisation is Bengaliwala (1995) and Bengali and Sadaqat (2005), however, both are old base year prices of 1980-81. Bureau of Statistics (FBS). Rather, it has improved upon FBS methodology in case of some sub-sectors including livestock and slaughtering by re-estimating population of different animals and the number of animals slaughtered with more logical techniques (detail discussion in chapter 4). For quarterization of national accounts, maximum information available in official sources or in different studies have been used which include: harvest calendars of all major crops and most of the minor crops (with province level detail); seasonal patterns of milk production, fish catch, and timber; quarterly production of different minerals, manufacturing goods, and cement (for construction activities); seasonal pattern of utilities consumption; quarterly imports and financial indicators like M2; etc. Moreover, mechanical technique of quarterisation as proposed by Lisman and Sandee (1964) has also been used in case of some sub-sectors like Transport, storage & communication, ownership of dwellings and other services. The provincial distribution of national accounts have been done by using some related indicators; however, provincial value added for sectors like crops, fishing, and mining & quarrying have been estimated directly by product approach just like national accounts. In order to estimate the contribution of total factor productivity to growth, the dissertation has undertaken a detailed exercise of estimating the capital stock at constant prices of 1999-00 and skill-adjusted labour force – both the inputs for all sub-sectors of GDP. The non-parametric approach as suggested by Solow has been used in factorization of overall GDP growth as well as growth in all sub-sectors. The results show that the series of national accounts estimated by this dissertation are fairly close to the official series for years 1999-00 to 2004-05, which gives a confidence to estimates of this dissertation for series prior to 1999-00. It has been found that on average 21.8 percent of the annual GDP is produced in the first quarter (Jul-Sep) followed by the third quarter (Jan-Mar) with 25.2 percent of annual GDP. In the second quarter (Oct-Dec) the production of goods and services is vi the highest at 26.9 percent. In the last quarter (Apr-Jun) production is also high with 26.1 percent of the annual. The provincial distribution of gross domestic product shows that the Punjab holds the highest share in gross domestic product (52.3 percent); it is followed by Sindh (30.6 percent), NWFP (11.5 percent) and Balochistan (5.5 percent). However, over the years the Punjab’s share has declined: during 1970s, about 54 percent of the country’s GDP was being generated in the Punjab that declined to 51.8 percent in 2000s. On the other hand, shares of NWFP and Balochistan in total GDP have increased during this period; there is no significant change in the share of Sindh in total GDP during the period of 1970-2005. The results of growth accounting exercise shows that average contribution of total factor productivity to GDP growth during the period 1970-2005 had been 1 percentage point. It was higher in 1970s and early eighties and remained below 1 percent in subsequent years with negative growth during the periods of late 1980s and late 1990s. However, resurgence in total factor productivity growth has been witnessed in recent years. Comparing the relative contribution of capital and labour, the results show that labour remained the biggest contributor to economic growth during 1970-2005; however, in recent years, capital contribution has surpassed the labour contribution. Supervisor:- Dr. A. R. Kemal

Meta Data

Keywords : Annual National Accounts, Factorization of GDP, National Income Accounts, National Income Accounts-Pakistan, Pakistan, Provincialisation of National Accounts, Rebasing of National Accounts
Supervisor: A. R. Kemal
Cosupervisor: Musleh ud Din

Related Thesis​