By contrast, non-food primary commodities, which were more directly affected by tendencies in international trade, show much sharper deceleration. In fact, Chart 6 shows that prices of non-food primary commodities actually fell after 1998, and have remained low thereafter.

Chart 6 >> Click to Enlarge
 
Most categories of manufactured goods also show substantial slowdown in terms of rates of price change. This is indicated in Chart 7, which suggests that sharp declines in inflation rates occurred for textiles, food products, machinery and other manufacturing categories. In fact, for textiles and food products, prices have declined in absolute terms by the end of the period, from their earlier peaks.

Chart 7 >> Click to Enlarge
 
Chart 8 brings all this together in the form of average rates of inflation over two different sub-periods, 1990-91 to 1995-96 and 1995-96 to 2001-02. The annual rate of increase of the general WPI index, which was 10 per cent over the first sub-period, fell to less than half that, at an average of 4.8 per cent over the next sub-period. Food items – and particularly foodgrain – did not slow down in price so substantially. But textile products actually fell in price in the second period, on average. And non-food primary products also decelerated very sharply in terms of inflation. This clearly reflected the effect of low world prices for most such commodities. Next in terms of significance in this regard was machinery, which was probably adversely affected by the falling import prices of similar goods.

Chart 8 >> Click to Enlarge
 
It is worth noting that fuels, light and lubricants actually showed a higher average rate of inflation in the second sub-period compared to the first. This category constitutes an important element of costs for all producers in the economy. Thus, a significant part of costs did not decelerate, and even accelerated in terms of price increase, for most producers, even as prices of final products appeared to decelerate or decline. This implies a squeeze on domestic producers, which would be reflected in falling profit margins or falling wage shares or both.
 
So in the aggregate, what has caused this very evident decline in inflation? The current mainstream explanations, which are heavily influenced by the monetarist approach, tend to put a lot of emphasis on money supply and the control of central bank lending to the government. According to this position, tighter control over broad money (M3) and limiting the ability of the government to finance its deficits by borrowing cheaply from the Reserve bank of India which would print money in consequence, are the main causes of the overall decline in inflation.
 
There are of course several problems with this argument at a theoretical level. The monetarist argument is based on the twin assumptions of full employment (or exogenously given aggregate supply conditions) and aggregate money supply determined exogenously by macro policy. Neither of these assumptions is valid. In fact there is a strong case for arguing that in a world of financial innovation where quasi-moneys can be created, the overall liquidity in the system cannot be rigidly controlled by the monetary authorities. Rather, the actual liquidity in the system is endogenously determined.
 
Therefore the real monetary variable in the hands of the government is the interest rate and thus attempts to control money supply typically end up as forms of interest rate policy instead.  Further, the notion of a stable "real demand for money" function (where the demand for money is determined by the level of real economic activity) is one which gets demolished by the possibility of speculative demand for money, a feature which if anything is enhanced by financial sophistication and the greater uncertainties of operating in today's economies.
 
In any case, the empirical justification for such an argument is also worse than negligible. In fact, there is no clearly discernible relationship between the rates of growth of money supply and of inflation on the one hand, and real output growth on the other. This is evident not only in India, but indeed in all instances elsewhere in the world where monetarist prescriptions have been pursued. The more complex definitions of money that are being developed elsewhere show this very starkly for other countries.
 
In the recent Indian case, a simple look at the relationship between M3 and inflation rates would indicate immediately that there is no relationship at all. This is shown in Chart 9. While the stock of money (M3) continued to grow at an average rate of around 17 per cent per annum throughout this period, the aggregate inflation rate, as we have already seen, declined quite sharply to less than half of its earlier rate, progressively over the decade.

Chart 9 >> Click to Enlarge