Friday, April 25, 2003
Ice World  |  Smart Investor  |  The Strategist  |  BS Motoring  |  BS Weekend
Regional News
BS Headlines
BS Opinion
Special Features
Business Law
Money Manager
Business & Values
Creative Business
Personal Business
Q & A
  Lunch with BS
  Book Review
  Today's Main Column
BS Magazine
Catch Up With News
Partner with Us
Advertise with Us
Contact Webmaster
To worry or not to worry
Ila Patnaik
Business Standard, May 23, 2001

Official data are flights of statistical fancy compiled in the stream of consciousness genre, says Ila Patnaik

“I’d rather be vaguely right than precisely wrong,” said Keynes. Most of India’s economists, Sudhir Mulji should be pleased to note, seem to be following Keynes’s lead. They are forecasting a 6 per cent GDP growth for this year.

Hard to believe, of course, considering what is happening to industry. But not to worry. The forecast simply has to be right because in recent years — quite regardless of whether people’s incomes and spending have moved up or down — the government has always come up with a number around six. Even when this is revised later, it usually remains between 6 and 6.5 per cent. Is that, in the fitness of things, the new Hindu rate of growth?

So stop worrying about whether the current crop of forecasts will match the government numbers. It decidedly will. Also, by the time you get the estimates, the revised estimates, the actuals, the revised actuals, not to mention the actually revised, as opposed to unrevised revised, the numbers would have changed considerably. By then it is unlikely to make a difference anyway.

You have to feel a bit sorry for government economists, though. They have the unenviable job of keeping confidence high, which means they can’t allow bad forecasts to become self-fulfilling prophecies.

After all, when confidence in the government depends on the numbers put out by it, who can blame its economists for trying to keep sentiment up by coming up with, shall we say, “encouraging” data.

But all this data juggling avoids the real issue, which is whether official statistics reflect what is happening in the economy. While, in general, there is agreement that a very large part of the economy is left uncovered by official data, there is little agreement on what this share actually is. This probably lies at the heart of the problem.

Of course, even in the the US where data are far better than ours, it is sometimes argued that about 10 per cent of GDP could well be unaccounted for.

For India, this component is much higher. Some say that the hidden economy constitutes about 25 per cent of GDP. Others put the figure at 40 per cent. So, for all we know, GDP may be growing at 10 per cent if there is high growth in the unreported segments.

I am, in fact, quite willing to take a small bet that the economy is actually growing at 7-8 per cent and that the government is clueless about it. The fact is that during the 1990s high growth has taken place in new sectors like entertainment, courier services, software, etc., while the old sectors have not fallen behind much. Add them together, and what I say makes at least intuitive, if not immediately verifiable, empirical sense.

I find it difficult to comprehend government statistics. For instance, in 2000-01 when most industries felt a significant slowdown and when many parts of the country faced drought and agriculture fared badly, government estimates said that GDP grew by 6 per cent. Good, you say; how wonderful, what a resilient economy we have built.

But then, there is this other fact: in 1999-2000, which was a good year at least for industry, GDP growth was 6.4 per cent. How can it be that in a bad year growth is still 6 per cent and in a good year it goes up only to 6.4 per cent?

But how did the government make this happen? By making industrial growth figures for both years identical! In its GDP estimates, CSO estimated industrial growth in both years to be 6.2 per cent. This despite the fact that the Index of Industrial Production (IIP) released by the CSO indicates that production grew in 1999-2000 by 6.7 per cent and in 2000-01 by 4.9 per cent. If you say this to the CSO, you will get a standard response: both numbers are going to be revised, and re-revised. Until, of course, one has forgotten all about them.

But then why should one believe CSO on its IIP numbers? Would these not be overstated as well? Could it be that these might actually be even lower than published figures? A little investigation reveals that might indeed be the case. The coverage of IIP was changed early this year to leave out some of the slow-growing sectors. But even that didn’t seem to be doing the trick. So what was done was as follows.

On January 12, 2001, when the government provided industrial production data, it noted that the CSO had revised the components of the IIP. For instance, from IIP manufacturing the CSO had deleted data for the production of four items: radio receivers, photosensitised paper, chassis for heavy commercial vehicles (buses and trucks) and engines.

Of course, as we can guess, these items were showing low growth in the year 2000-01. Indeed, for instance, the production of commercial vehicles is currently so slack that growth numbers are negative. So if chassis produced for commercial vehicles and engines are deleted from the series, it makes the series look better. This move pushed up the growth rate in IIP in 2000-01.

The reason for this change — as given — was that the production in these items was reported by single producers, so it was subject to significant variation. One would have thought that the thing to do was to improve coverage but, surprisingly, the CSO chose instead to delete these from the index altogether.

But now even that doesn’t seem to be working. Other sectors have also started showing much slower growth. In fact, growth in most capital and basic goods has either been negative or very low and these constitute nearly half of IIP. So unless you go on a dropping spree and delete all these goodies that are doing badly,i.e., nearly half the items from the production series, you can’t avoid showing lower growth in production.

Perhaps it is better, even at the risk of some loss in business confidence, to expand the coverage and reveal the true numbers even if, at times, this hurts in the short run. In the long run, it will add to the credibility of the government and its numbers.

Of course, for economic forecasters this means that they will be wrong more often than right. But that is fine because, by definition, an economist is one who will know tomorrow why the things s/he predicted yesterday didn’t happen today. So we’ll still be in business.

Search Archives
Market Indices
Sensex : 2924 -13
Nifty : 924 -6
Re-$ : 47.3575
Nasdaq : 1457 -9
Nikkei : 7699 -155
BS Services
Free Newsletter

BS Opinion Poll
Can the amended exim policy boost exports?
Can't say
Previous Polls
Top ^   

Ice World  |  Smart Investor  |  The Strategist  |  BS Motoring  |  BS Weekend 
Business Standard Ltd.
Nehru House, 4 Bahadur Shah Zafar Marg, New Delhi - 110002. INDIA
Ph: +91-11-23720202-10. Fax: 011 - 23720201
Copyright &  Disclaimer
Ila Patnaik