A V Rajwade: Cause, effect and correlation

WORLD MONEY

The other day I came across a study by some economists (unfortunately I have misplaced the document) arguing that intervention in the exchange market is not only ineffective but sometimes produces changes in prices opposite to what the central bank intends: in other words, for example, the dollar may fall against the rupee on the day the RBI buys dollars in the market. The conclusion was, as usual, supported by statistical correlations. What puzzled me was one question: the analysis should really compare where the level of the exchange rate would have been in the absence of intervention, with where it actually was after the intervention, and not merely the actual price movement. The problem, of course, is that the level of the rate in the absence of intervention, is a "known unknown", but drawing conclusions only on the basis of what is known can lead to illogical conclusions. To elaborate, if the exchange rate started at, say, 40 and ended the day at 39.50 after the RBI had bought a few hundred million dollars, does it prove the counter-productivity of the intervention effort? To my mind, no. The actual comparison should be between the admittedly unknowable level in the absence of intervention, and the actual level. For instance, if there were a way to know that the rate would have gone to 35 in the absence of intervention, then clearly intervention has been successful.

In a broader context, the point is that we often confuse correlations with cause and effect, the messenger with the cause of the message, and then come to often counter-productive administrative actions. To quote two recent instances, consider the ban on export of certain products, cement and rice for example, and the ban on trading in futures contracts on some commodities, both imposed in recent weeks by the authorities in a bid to substantiate their anti-inflation credentials. In fact, as politically sensitive food prices remain high, more than 40 countries have banned the export of various grains and other eatables. To my mind, such actions can be counter-productive in relation to the price level, and also more generally:

It is widely accepted that the purpose of anti-inflationary intervention, for example, the tightening of monetary policy, is more to moderate inflation expectations rather than influence prices directly. [As Nobel Laureate Joseph Stiglitz argued recently (ET, May 12), "Unless taken to an intolerable level, these measures by themselves cannot bring inflation down to the targeted levels."] Is the ban on exports actually fuelling global inflation expectations and hence counter-productive? Ngozi Okonjo-Iweala, former finance minister of Nigeria, recently wrote that restrictions on food exports "lead to even higher world prices, greater smuggling across borders and, once again, depressed prices for domestic farmers" (Time, May 12).

The last point is very important in our case. Even as our political masters continue to voice their concern about rural poverty, they are fighting shy of accepting that high commodity prices have perhaps come to stay, and provide an excellent opportunity to reduce the disparity in the per capita incomes between rural Bharat and urban India.

In a broader context, such bans strengthen the anti-globalisation lobbies in western countries, which is surely not in our long term interest: every poor Asian country which has grown fast during the last four decades, has done so on the basis of embracing globalisation.

Exports of cement were banned recently. There are news reports that cement companies in Gujarat are cutting production because of over-supply in the market! Again, as for steel, isn't the correct solution to remove bottlenecks in the way of creation of additional capacity rather than trying to control prices?

Whose interests are really being served by keeping the administered prices of four major petro-products at unrealistically low levels? The subsidy could cost something like Rs 200,000 crore in the current year, and oil companies have already stopped fresh connections of LPG because of the cash crunch. This apart, do the artificially low prices not discourage conservation? Even the profligate, spendthrift Americans have reduced the consumption of oil, for the first time since 1977, because of the high petrol pump prices. Surely price signals are not necessarily negative but influence behaviour positively? The macroeconomic consequences of the subsidised prices are also large. For one thing, the government has lost a huge amount of money because of the fall in the share prices of the three marketing companies: surely this cannot be ignored even if the government does not follow mark-to-market accounting? Again, many analysts are getting increasingly concerned about the true level of fiscal deficit. Inclusive of the oil bonds, fertiliser bonds, food bonds, loan write offs, the Pay Commission recommendations, and so on, it is approaching 10 per cent of GDP (centre and states together), and has started frightening foreign investors.

Perhaps the most interesting example of confusing correlation for cause and effect is the ban on trading of several commodity derivatives. But more on that next week.

avrajwade@gmail.com