Statistical issues

by Dr Ashfaque H Khan

Statistics speak all languages, affect all policies and touch all aspects of people’s lives. The significance of critical statistics for policymakers cannot be overemphasised. In recent years, significant changes have taken place in Pakistan’s statistics which have not only distorted some key macroeconomic statistics but made researchers’ lives miserable. Such distortions need to be understood and addressed urgently.

National income accounts, employment/unemployment and inflation are among the most vital statistics for any economy. Pakistan’s national income accounts have witnessed structural changes in the year 2012-13. It has seen its national account (GDP) rebased from 1999-2000 to 2005-06 while the measurement of national accounts changed from the System of National Accounts (SNA) 1993 to SNA 2008. While the rebasing of GDP was highly desirable, the movement towards SNA 2008 was premature and inconsistent with the level of expertise available at the Pakistan Bureau of Statistics (PBS).

The rebasing of GDP was required because every country witnesses structural changes in production, consumption and investment over the years on account of continuous development and innovations. Practice around the world suggests that some countries rebase their GDP after five years and others after ten years. Pakistan rebased its GDP from 1999-2000 to 2005-06, the results of which were released in 2012-13. The SNA 2008 is highly sophisticated and measures GDP by complicated techniques. As such, hardly any developing country like Pakistan has adopted SNA 2008. Furthermore, our statistical information gathering system is not strong enough to gather all the information required for SNA 2008.

The whole exercise of rebasing of national account from 1999-2000 to 2005-2006 and moving away from SNA 1993 to SNA 2008 was led by Mr. Bernd Struck, a German National Account expert who spent seven years in PBS. Little or no consultation was made while undertaking this exercise with those who use, or have any understanding, of the national accounts. Even the staff of the finance ministry, Planning Commission, and the Central Bank were out of the picture during the whole exercise. They only came to know about the change when the results of this exercise were presented to them.

Such an exercise has created enormous difficulties for researchers and has at the same time posed challenges for policymakers. The reliability of the data has become a serious issue. There exist no consistent time series data for national account prior to 2005-06; hence no meaningful economic research could be undertaken. Every researcher generates his/her own consistent series of national account using different methodologies. In particular, investment numbers are totally distorted. In the words of a renowned economist, Dr Hafiz Pasha, the investment numbers are “rubbish”.

This exercise has produced some strange numbers. For example, while real GDP grew at an average rate of 2.9 percent per annum over the last five years, a relatively unknown component of the GDP with the name of ‘other private services’ has grown at an average rate of nine percent per annum and accordingly emerged as the fifth largest component of GDP after agriculture, manufacturing, wholesale and retail trade and transport and communication. Similarly, investment in this sector was only one-third of the investment in manufacturing in 2007-08; whereas, it has now surpassed the investment in manufacturing during 2012-13. Does that make sense?

The PBS has released the data of the large-scale manufacturing for the first quarter (July-September) of the current fiscal year. Large-scale manufacturing has grown by 12.8 percent in September and by 8.4 percent in July-September this year and over the corresponding period of the last year. Such an impressive recovery in large-scale manufacturing is based on fictitious growth in the production of several components of this sector – for example, vegetable ghee registered a growth of 131 percent, footwear (46.1 percent), paints and varnishes (36.2 percent), air-conditioners (106.7 percent), electric bulbs (43.4 percent), TV sets (86.4 percent), diesel (132.7 percent), billets (32.5 percent) and so on. Does that make sense? In particular, these statistics were distorted in the month of September 2013.

Inflation is another important statistic for any country. This statistic has also undergone considerable changes since 2011-12. First, the base year measuring CPI-based inflation changed from 2000-01 to 2007-08; its results are being reported since August 2011. As a result of change in the base year, the weight of food items declined from 40.3 percent to 34.8 percent. Since inflation in developing countries like Pakistan is, to a large extent, driven by food and fuel prices, it has been understated to that extent and also rendered non-comparable with previous years.

Second, natural gas price distorted CPI since July 2012, when reduction of the number of slabs for domestic consumers was interpreted by the PBS as 49 percent reduction in tariff. With 1.5756 weight of gas in CPI, such a large ‘reduction’ understated inflation for the year 2012-13.

Third, the PBS data for electricity prices apparently account for the increase in base tariff only with no adjustment for monthly fuel adjustment charges paid by the consumers. Fourth, house rent inflation has been averaging 6.5 percent for the past 15 months prior to March 2013 and is measured 6.9 percent in March 2013 and 7.9 percent in November 2013. With property prices rising steeply in major urban cities, such low house rent inflation is not reflecting the ground reality.

Fifth, money supply and domestic credit have been growing at an average rate of 15 percent and 20.5 percent respectively over the last two years. Furthermore, borrowing directly from the central bank to finance fiscal deficit has been on the rise for the last two years and has accelerated with unprecedented pace in the past few months. Surprisingly, inflation exhibited a declining trend in the midst of rising credit expansion. Inflation has surged in recent months owing to food prices. There appears to be a disconnect between rising money supply and CPI-based inflation. Even if we double the supply of money, inflation may rise only marginally.

Unemployment is yet another key macroeconomic variable. Pakistan’s unemployment rate does not reflect the ground reality. There appears to be a total disconnect between economic growth and unemployment. The Pakistan Economic Survey 2012-13 says: “unemployment rate decreased marginally from 7.7 percent in 2003-04 to 5.9 percent in 2010-11” (PP. 162). Does it make sense?

Based on the above facts, it is abundantly clear that something has gone fundamentally wrong with Pakistan’s key macroeconomic statistics. Any policy measures taken on the basis of these statistics are flawed. I would urge the prime minister and the finance minister to look into the matter and take urgent measures. These measures must include formation of a high level committee of professional economists, statisticians, academicians and the users of these statistics to thoroughly examine these statistics, the way they are measured and compiled and the technical capabilities of the PBS.

The finance minister must request the IMF to send experts from its Statistics Department to assist the high level committee in its work. The sooner, the better.

The author is principal and dean at NUST School of Social Sciences & Humanities, Islamabad Email: ahkhan@nbs.edu.pk

http://www.thenews.com.pk/Todays-New...istical-issues