Much ado about growth forecasts

The statistical arm of the government, the Bureau of Statistics (BBS), published national income accounts data for the current fiscal year at the beginning of the fourth quarter. The most discussed datum of the national accounts in the media is the GDP growth rate. It so happens that at about this time several other organisations, among which are the World Bank, the International Monetary Fund and the Asian Development Bank, also make public their forecasts about the country's GDP growth rate.

The government has put forth its estimate of the GDP growth rate for the fiscal year 2015-16 at 7.05 percent. However, the forecasts of all of the aforementioned organisations were significantly lower than this rate, with the World Bank giving the most pessimistic outlook for the economy (6.3 percent). It also expressed doubts about the plausibility of the government estimate. Not unexpectedly, some government leaders were riled by the lower GDP growth forecasts and pointedly advised the public to ignore these, since these were based on speculations. They maintained that the BBS estimate was the accurate figure for the GDP growth rate.

Since the national accounts data were published at the beginning of April, employees of the Bureau of Statistics working out these data could not have had the relevant information for months much beyond last December. The principal agricultural crop, boro, would be a few weeks from harvesting such that only aus and aman harvests, done by December, would be known with certainty. Industrial production data are published with a lag of 3-4 months. In this event, the national accounts estimates of BBS are most likely six-month forward forecasts. It is extraordinary that these forecasts later become the final estimates with no or very little revisions.

There are few developing countries that can provide their national accounts data even after three months of the end of the year, but BBS does so three months before the end of the year! Not many developed countries or organisations have as yet developed a forecasting technique that can provide national accounts estimates three months before the end of the year as accurately as BBS does. This is what the government should be highlighting as its achievement rather than the mundane 7.05 percent GDP growth rate, which many developing countries, including all South Asian countries except Nepal, have already exceeded in the recent past. Bangladesh also had a higher growth rate in 2006-07.

Forecasts are basically educated guesses or speculations about the future on the basis of the knowledge of past outcomes and the current state of affairs. Although statisticians provide certain methods of making guesses, these hardly ever yield accurate predictions, as would be apparent from a cursory check of the forecasts of, say, world or country GDP growth made by the reputed organisations mentioned above. Since each organisation has its own method of forecasting, the numbers churned up by their forecasters are seldom the same although they do not usually differ much. Forecasts are often revised periodically as more recent information comes to hand. The forecasts of GDP or any other variable made at the beginning of the year is most unlikely to be the same as that done toward the end of the year.

Hence the difference in the forecasts of the GDP growth rate made by the government on the one hand and these organisations on the other is neither unusual nor surprising. Indeed if they were the same, one should be suspicious of some unhealthy collusive practice. What is surprising, though, about these forecasts is the pattern of the difference between them.

If all forecasts were done diligently and competently, they would be randomly distributed. However, during the last several years the estimates of the government were always significantly higher than the forecasts of the multilateral organisations. This does not fail in giving rise to some controversy, with the government defending its forecast even as others suspect some upward bias.

Every year when the forecasts of these organisations (in particular that of the World Bank) are made public — and these are invariably much lower than the government estimates — ministers and  government leaders angrily whip up a storm alleging bias and an unholy attempt to belittle the government's achievements. It is not very clear why the government should be so sensitive to the guesswork of some people in these organisations.

Unless it has a few skeletons in the cupboard the government should take a benign view of what they may be saying. After all, the only organisation that has the statutory authority to actually estimate national accounts is the government. Such estimation involves the collection of sensitive information on revenue, expenditure and other variables from households and firms. They are unlikely to divulge all such information unless required by law. Only the government can force them to reveal the necessary information such that only it can provide credible data. Hence, the government is the sole provider of national accounts data in about every country of the world.

However, detailed information is not necessary for forecasts (although these could be helpful if available). Forecasts of GDP are done by utilising already available data and information on a small range of macroeconomic variables that are usually available without a long lag and are indicative of the current state of the economy. The utility of forecasts is that they are useful for formulating current policies of both business enterprises and the government, which require some idea about the future. The forecasts are not claimed to be actual estimates. When the final estimates are published by the appropriate authorities, these will be accepted as the true data. The GDP growth rates of Bangladesh on the data website of the World Bank or the Asian Development Bank are the same as that on the BBS website.

It is disconcerting that a large number of citizens seem to have more trust in the predictions of the multilateral organisations than in those of their own government and believe that the GDP growth numbers of BBS are inflated. This can happen when people do not have much faith in their government keeping its promises or telling time-invariant truth. Such problems may arise when the optimal policy of the government at one point in time is no longer optimal at a future point in time such that the government has an incentive to renege on its earlier promise. Economists have shown that there are significant welfare costs when this happens.

The only way of avoiding these costs would be to build up a reputation of never reneging on promises or saying things that are not true. Admittedly this is a tough task.