On PMI data,look before you leap

The RBI,and analysts,are just as likely to be right by tossing a coin as by using the much quoted PMI data on output and inflation

Written by Surjit S Bhalla | Published: May 12, 2012 3:10 am

The RBI,and analysts,are just as likely to be right by tossing a coin as by using the much quoted PMI data on output and inflation

Much has happened to the financial markets over the last two weeks. Each of the three Indian markets — bonds,currency and stock — has depreciated sharply. There are several determinants of each of these markets,but a common factor is inflation. If inflation bounces back,then there is some reason for the RBI to be on hold,and perhaps even to hike rates,as some analysts have begun to argue,and even plead.

Inflation data for April will be out next week,and the purpose of this article is not to forecast what that level will be. Rather,it is to assess the prospects for inflation in India over the next several months; inflation patterns that should influence and form the basis of RBI’s policy regarding interest rates. Unfortunately for the RBI hawks and the followers,these patterns include a distinct slowing down of the world economy,and a distinct decline in commodity international prices,including food. There are noises about surplus of wheat from none other than Ashok Gulati,chairman of the Commission on Agricultural Costs and Prices. My previous article (‘Populism of Paddy Prices’,IE,May 9),emphasised that Indian procurement prices for paddy in 2011 were close to the highest in the world. International food prices are declining,so how much further up can domestic food prices go? Crude oil prices are down more than 10 per cent in the last two weeks,to below $100 a barrel (although the RBI,and the finance minister,are still talking about heightened inflationary expectations and $150 oil!). And growth is trending below the lower end of RBI’s forecast for 2012-13.

So it was this present,and possible future,that RBI Deputy Governor Subir Gokarn was looking at when he casually opined on May 8 at a FICCI seminar in Hyderabad: “if you look at our inflation projections,in relation to what we consider as long-term or medium objective,there are inflation pressures. That,in a sense,limits the room that we have to reduce rates” (emphasis added). It is the job of the RBI to make projections about the future and make policy accordingly. It will be tragic if the RBI made policy on the basis of the past,or what some people have called rear-window economics.

However,the RBI’s record on both growth and inflation projections over the last two years does not inspire much confidence. Nor does it inspire confidence that the RBI’s favourite policy prescription is to change the goal posts,or the rules of the game,according to its predisposed (pre-ordained?) game plan.

Consider the following three goal post changes that the RBI has engineered in the last six months. (Parenthetically,this is a new record for goal post changes for India,and beats the legislature and Supreme Court who induce a constitutional amendment every six months.) First,the RBI talked of bringing WPI inflation below 7 per cent. Just as that target was about to be breached,the RBI talked of pricing power in the hands of Indian corporations keeping core inflation upwards of 7.5 per cent. And just as this target was decisively breached (last reading was of core inflation at 4.7 per cent,year-on-year),the RBI said fiscal deficit needs to be reduced. Just as the government acknowledged that concern via the budget for 2012-13,the RBI now talks about limited room because of heightened inflationary expectations. Since this is an animal much like the invisible hand and therefore cannot be found,the RBI is finally on safe ground — while it can’t be proved right,at least it can’t be proved wrong!

At best,forecasts are a mugs game,whether these forecasts are about love,inflation,bonds or stock markets. So the RBI and Gokarn,and the FM may still be right about inflation and much else. The only point I wish to make is that the forecasts should have a logical base,an empirical reality.

It is possible,if not likely,that the RBI’s certitude about future inflation was influenced by the PMI data distributed by a leading investment bank,HSBC. These data have a long history in the US (since 1948) and a shorter history in emerging markets (in India since 2005). It is de rigueur for every analyst and the RBI,to cite these data in support of growth and inflation trends. PMI data for both output and prices suggest robust growth and robust inflation in 2012. So much so that the marketers of PMI data have scornfully suggested that the RBI made a huge mistake in cutting rates because their data showed no reason for doing so. Further,the RBI was losing credibility as an institution,and/or that it was submitting to pressure from Delhi. Many investment banks (both domestic and foreign) have reinforced the PMI data by buying its trends in toto — and to the neglect of all other data.

Are the PMI data so good that analysis is no longer worthwhile? These figures come out on the first of every month,so traders,investors and the RBI,do not have to look at stale data anymore. The table documents the relative worth of the PMI data for both India (available since April 2005) and the US where it is present since 1948. For comparison,the US results are presented since April 2005.

The test conducted for the “accuracy” of PMI is very simple — it is whether it gets the direction of monthly changes right in the corresponding data for industrial production and WPI for manufacturing (neither series is seasonally adjusted so month-on-month changes is the right comparison; use of year-on-year changes yields broadly similar results). For example,data released on May 11 showed that the IIP index improved in March to 186 from 174.5 the month before. As reported more than a month-and-a-half earlier,the PMI for March had declined from 56.6 to 54.7. For this match up,the PMI scores a zero. When the directions match,the score is 1. The scores are summed for 84 matches,from April 2005 to March 2012. If the cumulative score is 42,then the PMI is no better than the toss of a coin.

The average score for India is 42 out of 84 possibilities,or a strike rate equal to a random toss. For the US,the scores are better,but still a rate only marginally better than random,52 per cent. And these data form the basis of RBI forecasts for output and inflation. So what should the RBI do when making policy on interest rates? Toss a coin,or quote the PMI data? It just doesn’t matter.

The writer is chairman of Oxus Investments,an emerging market advisory firm

For all the latest Opinion News, download Indian Express App

    Live Cricket Scores & Results