Is Pakistan in a manufacturing recession?

If the latest release of Large-Scale Manufacturing (LSM) index is any guide, the country is surely facing a manufacturing contraction.

Although, the first year-on-year contraction appeared in 3QFY19, the number was too small (negative 0.02%) and perhaps not too convincing of a data given the usual revisions in provisional data releases. But the two quarters hence – i.e. 4QFY19 and now 1QFY20 – have witnessed relatively bigger quantum of back to back year-on-year production decline of 1.01 percent and 5.62 percent respectively.

By one account, a fall in output in two consecutive quarters is construed as a recession, and therefore one could argue that Pakistan is facing a manufacturing recession, or at least was facing one until September this year.

But that’s precisely the problem, for there is no official definition of a recession, according to economists Stijn Claessens and M. Ayhan Kose who wrote a piece titled ‘What is a recession’, in IMF’s Finance and Development magazine in March 2009. Ten years on there is still no globally accepted standard definition.

However, one definition that is somewhat better accepted is the one by America’s National Bureau of Economic Research. The NBER defines a recession more broadly: as "a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in production, employment, real income, and other indicators. A recession begins when the economy reaches a peak of activity and ends when the economy reaches its trough."

Of all these indicators Pakistan produces only LSM data on a monthly basis. It is rather strange that despite all the noise about unemployment and joblessness in this country by business, politicians and citizens alike, unemployment datasets aren’t estimated on monthly basis. But that’s another subject discussed earlier in this space (See BR Research’s Is unemployment really Pakistan’s big issue?).

Based on LSM dataset alone, one could argue that production contraction is underway, which ‘may’ lead to a recession in manufacturing GDP by the end of ongoing fiscal year. But let’s wait is out for at least two more months before pronouncing a judgment.

If the 2MFY20 LSM numbers were the worst large-scale manufacturing performance since FY09, the 1QFY20 numbers have shown a bit of improvement. This is somewhat visible in the table published alongside this piece which shows percentage change in production numbers in key LSM sectors. Of course, this may well be only be catching on a straw of hope.

But there is another straw of hope: the central bank’s Purchasing Managers Index (PMI), which the SBP started tabulating since FY18. It’s a part of SBP’s Business Confidence Survey, conducted at a bi-monthly frequency to obtain the views of senior managers of large companies in the industry and services sector.

The central bank’s PMI had been falling consistently since it began and is still in the negative zone; below 50 shows poor confidence. But its October 2019 wave has improved over the last wave conducted in August 2019 and has in fact shown improvement for the first time after eight waves. On year-on-year basis, if the PMI was down 22 percent in August 2019, its down 16 percent in October 2019.

Does this mean a recovery in LSM contraction will soon be witnessed? And what shape might it take – we know contractions come in various alphabets, V-shaped, W-shaped, U-L-or J-shaped – these are the kind of questions that can only be answered once the tide is over, for in absence of data it is difficult to make such conjectures – perhaps even harmful. Such is the nature of Pakistan’s patchy-data economy.

Note: The ongoing LSM data series has its base year as 2005-06. Data for monthly LSM index prior to FY10 is not readily available on the website of Pakistan Bureau of Statistics, whereas the LSM of 2005-06 base was first published by the SBP in Jan-2012 Statistical bulletin. Ergo, monthly and quarterly data for FY10 and FY09 used for the purpose of this research is from BR Research’s own archive. However, there are sufficient reasons to believe that revisions to provisional LSM data release have not tampered broader trends.