The jury seems to be still out on whether interest rates have peaked in India. Those who claim that there is still some upside left point to the fact that there is enough momentum left in the economy to stop worrying about inflation rates. They are likely to use the following data to support their case.

The March index of industrial production showed a year-on-year growth rate of around 13 per cent, way higher than the consensus forecast of 10 per cent. Despite a strong "base effect", which kicked in from April, and a continuously appreciating rupee, headline inflation remains stubbornly over 5.5 per cent.

Proponents of this view also argue that real interest rates are still lower than the highs reached over past cycles. For monetary policy to effectively curb the "overheating", nominal or money rates will have to climb a little more to push real interest rates up further.

I must confess that I am a little confused about how to compute and interpret real interest rates in the Indian context. Going by the widely accepted formula that the real interest rate equals the nominal rate minus inflation, the real interest rates could vary widely, depending on which nominal rate is used.

If the ten-year government bond rate is used to compute inflation, then the real rate is currently close to 2-2.5 per cent, going by an inflation rate of 5.5-6 per cent. If, instead, the average prime lending rates of banks (roughly 14 per cent) are used, the computed real rate is 8-8.5 per cent.

This is perhaps a trivial issue. It is somewhat futile to get bogged down in the nitty-gritty of what the absolute real rates should be. What needs to be done is to use comparable rates at different points in time and see how they compare.

If the same interest rate metric is used to compute the real rates, then it is perhaps true that real interest rates today are lower than previous peaks.

I am a little more concerned about the appropriate measure of inflation. Economists acknowledge the fact that using the current or past levels of inflation is somewhat pointless unless of course they are accurate predictors of future rates. The logic is simple. From the borrowers' perspective, the interest rate represents the cost of servicing a loan over its tenure in the future. The "real" value of this debt service cost is contingent on the rate of inflation that would prevail in the future over the span of the loan, not on the current rate of inflation.

Thus, to reiterate the point I made earlier, the current rates of inflation hardly make sense unless they represent future levels of inflation. I claim that they don't.

This is not just an esoteric academic point. It is critical to the question of whether higher rates will begin to impinge on investments by companies and whether with a further rise in rates, the possibility of a hard landing for the economy escalates.

One way to interpret the real interest rate is to see it as a consolidated index comparing the return on capital to the cost. The anticipated rate of inflation indirectly measures the returns that are linked to pricing power. I would think that most companies planning their projects would do well to factor in much lower pricing power going forward.

With falling import tariffs and rising domestic competition, the average manufactured product inflation could settle at 2.5-3 per cent.

That works out to a real interest rate of 11-11.5 per cent if it is set against the average PLR. If this is the true real interest level, companies could start doing a reality check on the viability of their expansion plans. Instead of chugging along despite high borrowing costs, the investment engine might just start losing steam.

My prediction of manufacturing inflation need not be as far-fetched as it seems at first glance. From a level of around 9 per cent in the decade spanning the second half of the eighties to the first half of the nineties, the average rate of manufactured product inflation moderated to less than 5 per cent in the following decade.

Even today, if the index is shorn of things like food products and a couple of commodities, manufactured product inflation prints below 5 per cent.

Manufacturers have been able to hold on to their margins on the back of drastic cost reduction that have followed productivity gains over the last decade. There is, unfortunately, a limit to improvement in productivity and companies cannot bank on infinite reduction in operating costs.

Domestic economic variables are, with increasing integration, converging to international averages. In that context, a 2.5-3 per cent rate of manufactured product inflation seems high enough. Commodity producers are likely to see wild swings in their pricing power. Phases of high product price inflation and profitability are likely to be followed by sharp cyclical dips. On average, they are unlikely to beat this 2.5-3 per cent benchmark that I've taken as the trend growth in pricing power.

The policy implications of this are obvious. Interest rate policy has to be forward-looking in gauging the implications for investment spending. In a situation where structural change that erodes the pricing power of companies further is inevitable, it is critical to choose the right measure of real interest rates.

If an appropriately long-term forecast for inflation is chosen, we have perhaps reached precariously high levels of real interest rates. If nominal rates were to climb up much further the result could be a sharp slowdown in corporate spending.