Behind the scenes of CPI blooper

The recent CPI blooper leading to a much-more-than-expected inflation number for October 2018 has caught the attention of business and economic stakeholders across the country. The Pakistan Bureau of Statistics (PBS) has since found itself in a round of fire. And rightly so. But is it entirely the fault of the PBS staff? Not quite entirely!

The PBS maintains that it has accounted for gas revisions as per its methodology being followed since the last base revision in 2007-2008. So why the ruckus now! One of the key reasons that caused higher-than-expected inflation in October numbers was the difference in the number of slabs for gas prices the PBS tracks for CPI measurement, and the number of gas slabs announced by Ogra. The PBS tracks five slabs, Ogra has seven slabs. Adjusting the two requires some statistical tinkering, that is not reported publicly – so there is a transparency gap. Hopefully, the PBS will share the details of its statistical adjudgment, regardless of whether it will make corrections in October numbers.

Second, the methodology itself is flawed. The PBS does not calculate utilities’ price inflation on weighted average basis – weighted in terms of giving higher weight to the slab where there are more consumers and lesser weight to the slab that has less number of consumers. This is a methodological issue, which according to PBS sources, will likely be resolved in the new series of CPI that has been constructed and whose time series data is being developed every month. The new CPI series is expected to fix these and other methodological issues such as the absence of population weighted house rent inflation in various cities.

Herein lie two problems. For one, the government does not fully appreciate the importance of inclusiveness. It has some private sector representation on the panel that oversees CPI rebasing; it also has a fairly respectable (on paper) representation in its governing council. But the organisation simply does not effectively engage with media and the civil society at large.

The other end of this problem is the weak importance that media places on geeky aspects of economic journalism, even though economics affects all and sundry across class, ethnicity, gender and so forth. When the CPI rebasing was initiated by the PBS in 2015, BR Research was the only section of press that carried back to back analysis and opinions on its methodological issues. The rest of the media hardly talked about it, even though measurement of CPI (done right or wrong) affects rich and poor alike; it even affects the salaried class that so-called well educated, urban, important voter block called the chattering class.

Now if the new CPI also has methodological problems, then the same class and the media will raise hue and cry. But it may be too late because rebasing exercise does not happen overnight; it takes at least about two years to put together a new base.

Another reason why the PBS staff is not entirely responsible for slow reforms? And perhaps even for the latest CPI blooper if it so transpires, is the failure of governance?

It had come as quite a shock to many when Dr Sajjad Akhtar, former acting chief statistician wrote an opinion piece in this newspaper flagging that new hiring at the PBS was stalled between 2011 and 2017 “on the pretext that it will be undertaken after the approval of the new rules of the merged department. As a result nearly 1/3rd of the workforce of PBS was lost to regular attrition, with no corresponding replacement”. (See also his op-ed ‘Data revolution’ published October 25, 2018. Other reports put that number to employees being halved due to periodic retirement.)
https://fp.brecorder.com/2018/10/20181028419367/

The failure to replace the staff, Dr Sajjad wrote, is attributed to the fact that attempts at devising new service rules under the new statistics act of 2011 was mired in bureaucratic red-tape on one context or the other. As a result of this poor governance, for instance, the staff that had been producing inflation and other statistics for 30-40 years gone, with no people coming in.

Those who joined the PBS starting from 2017 when the civil services act was invoked to prevent the organisation from coming to a grinding halt know little. Why? Well because the new staff has not had the opportunity to be trained by their predecessors. Classic failure of succession planning! Consequently, the institutional memory of the now retired PBS staff who knew, for instance, how to deal with data outliers or jumps through third party validation, is now gone.

The poor state of governance can also be measured by the fact that at present, the PBS has no chief statistician; the statistics division has no secretary; and no minister; its governing council is incomplete and hasn’t met for at least nine months; it has only about two-thirds of its employees (if not half as some sources say) with another 1000 employees expected to retire in next 12-18 months; half of its senior management are on contractual or consultative posts, whereas the last two appointments at the office of chief statistician were short term when in fact the post, being an administrative one, demands a strong anchor.

Whether the PTI will continue with the legacy left by PML-N and PPP at the PBS or whether it will truly make the organisation a competent and independent body of international repute remains to be seen.
So far they not have even paid lip service to improving statistics, considering they have not made the right and reputable appointments, nor is the fixing of statistics bodies on their 100-day agenda. Will Naya Pakistan be built on the foundation provided by weak socio-economic statistics?