The Economic Survey: Data Vs Reality

The Economic Survey for 1996-97 will be out tomorrow. It will give two sets of figures. First, what happened in 1995-96. Dr Manmohan Singhs Survey for 1995-96 quoted an estimated GDP growth rate of 6.2 per cent over the previous year. This was revised in Mr Chidambarams Survey, in July 1996, to around 7 per cent. After a few months, the CSO reduced it to 6.5 per cent, making many people worried about the veracity of the CSO data. To compound matters, the CSO a few months ago increased it back to 7 per cent! Many people have been quite bewildered with the fluctuating growth rate in a year that has already happened!

 

In the forthcoming Survey, the matter will hopefully be put to rest. Unfortunately, going by experience, it will still be a provisional quick estimate, late by a year! We will know only next year, that is two years after 1995-96, how the economy behaved in that year. Are we really bothered with what happened two years ago?

The second matter of interest in the upcoming Survey is, what has happened in 1996-97, given that we have data for only part of the current fiscal year. The answer to this question is one that may make or break the current coalition government. The Congress has publicly voiced its concerns about the slow pace of reforms while the Left has been trying to ensure that the reforms have a human face, whatever that means. The net result has been that the finance minister, in spite of his very best wishes, has been unable to carry out major reforms that he had promised in the last budget.

The dilly-dallying over implementing the reform agenda, coupled with the CBI investigations of prominent industrialists, have resulted in some degree of uncertainty on the economic front. Although some observers maintain that this has had its effect on industry, agriculture seems to be doing well. As for the services sector, over the last few years it has grown at a stable rate of 5-6 per cent and, is expected to do so this year too.

There is a quick and easy method of figuring out the growth rate in the GDP. The economy can be broken up into three sectors: industry, agriculture and services. One can take their growth rates, multiply them by their respective shares in the total GDP, add these up and arrive at a rough (and ready) estimate of the GDP growth rate. The shares of each of the sectors remain more or less stable in two consecutive years and, hence, are easily available from the last years estimates.

The problem is with getting the data on the performance of each of the individual sectors. Data is now available for the index of industrial production (IIP) upto October, 1996, i.e for only the first seven months of fiscal 1996. In agriculture, we can resort to the governments estimate. As for services, very little information is available and services account for the largest share in the GDP - slightly over 41 per cent. Of course, by information we mean data that is available in the public domain. Nonetheless, we can still construct a forecast by this method.

From the last years data, the proportion of agriculture, industry and services in GDP are 31, 28 and 41 per cent respectively. The newspaper estimates of agricultural growth rate are about 3.5 per cent. Data for the overall IIP indicate a growth rate of 9.8 per cent over the same months in the previous year. For services, one can stick to its historical growth rate in the nineties of about 6 per cent. Putting these together, we get an estimated growth rate of GDP, in 1996-97, of 6.3 per cent.

There must, however, be something wrong with this method. This is because, if a similar calculation is performed for 1995-96 for which most of the data is now available, we get a corresponding GDP growth rate figure for 1995-96 at 5.7 per cent, well below the latest estimates of 7 per cent. This can only mean that the so-called provisional data are exactly what they claim to be !

So, there has to be another method of going from the provisional to the actual figures. This one can do if one can identify a pattern in the co-movements of the different variables in the economy. Specifically, one has to be able to figure out the relationship between variables whose observability is easy, reliable and more frequent with those that make up the GDP. Monthly data on several important variables are quite easily available in the public domain. Our model utilises this monthly information to generate forecasts for the GDP.

Reliable data, however, is available now for only the first half of the current fiscal year. Based on this data, our model forecasts a growth rate of 8.5 per cent in this year. This is dependent on the assumption that the economy will continue to perform at the same pace as during the first half of the year. For instance, industry registered a growth rate of almost 10 per cent in the first half, over the same period in the previous year. Agriculture, this year, has grown faster than the previousyear, when it was nearly stagnant. All of this may change in the second half of the year.

Recently, there has been growing pessimism regarding the growth of the economy. Export growth is down, stock markets are sluggish to say the least, industry is unhappy with a wavering government, credit costs are too high, etc. Since the data for the first half of the year does not reflect any of this, these fears must pertain to what is happening in the remaining part of the year. Unfortunately, we do not have reliable data for this period.

This brings us to our second estimate of GDP growth. If we hold the economy to levels achieved till the end of October 1996, i.e. then the growth rate will be no lower than 4 per cent. But this is too extreme an assumption and is highly unlikely to hold true. One has to remember that there is some inertia in the economy investment decisions are taken well ahead of time and are often difficult to reverse quickly. Also, the production of certain commodities cannot be halted easily, with producers making inventory adjustments in the face of changing demand. Taking these into account, the growth rate should be no lower than 6 per cent.

All of this may suggest that we are being typical two-handed economists, unwilling to commit ourselves. Anticipating this criticism, let us hasten to add that the above discussion is to give a flavour of the thoughts that went behind our calculations. Since the data available is less than complete, it is only fair to consider possible scenarios which give the upper and lower bounds to the forecast. The number we will stand by is an 8 per cent growth in real GDP in 1996-97. For those of you who want to check this out, unfortunately, the wait will be two years! In the meantime, let us both wait for the provisional estimates of the Economic Survey, 1996-97.

(Shubhashis Gangopadhyay is President and Wilima Wadhwa, the Executive Director, of the Society for Economic Research and Financial Analysis (SERFA), New Delhi)

The dilly-dallying over implementing the reform agenda, coupled with the CBI investigations of prominent industrialists, have resulted in some degree of uncertainty on the economic front.