The Reserve Bank of India's somewhat drastic decision last Friday to hike both its refinance (repo) rate and the cash reserve requirement for banks seems to have taken the markets by surprise.

However, few can claim that it was entirely unanticipated. Headline inflation has remained stubbornly close to the 6.5 per cent mark. Despite a raft of lending rate revisions by banks over the last four months, overall credit growth still seems uncomfortably high at 30 per cent or so.

Money supply growth printed at 22 per cent for the fortnight ended March 16, way ahead of the nominal GDP growth, which at best would come in at 16 per cent for the January-March quarter. Thus, there seems to be enough indication that there is too much money chasing a smaller supply of goods.

The central bank could have other concerns. Much of the upward push on money supply has been driven by foreign capital inflow. While a fair bit of this money has come in as foreign direct investment (traditionally considered "superior" to the more fickle portfolio flows), a significant proportion has come in as FDI in real estate, a sector that has stayed on the RBI's "overheating watch" for a while now.

There is also anecdotal evidence of escalating bad loans in retail portfolios of banks. To add to its woes, strong capital inflow has driven up the real nominal exchange rate of the rupee. This comes at a time when export growth, with the exception of oil products, shows signs of flagging.

The RBI's tack in managing the recent crisis in the money market also perhaps pointed to the fact that some stern measures were likely to follow. Its stubborn refusal, for instance, to pump in liquidity into the market (a dose of un-sterilised intervention in the currency market could have done the trick) when call money rates touched stratospheric levels could have been construed as a signal to banks that tougher times lay ahead if they did not calibrate their assets and liabilities.

Over the last four months or so, a couple of things have happened in both the political economy of inflation control and the choice of policy instrument and they need to be borne in mind.

First, the public sparring between the finance ministry and the RBI on the issue of interest rate hikes seems to have died down. While this could imply a convergence of views between the two institutions, it also means that the exchequer has pushed the inflation ball squarely in the central bank's court.

The task of inflation management is now almost exclusively in the RBI's domain and this appears to have given monetary policy an increased sense of urgency.

In this period the distinction between and the debate over supply-side inflation and demand-driven price rises have finally been jettisoned. I can think of a number of reasons for this. For one, as calculations done by colleagues at my bank show, the shares of demand-driven components of inflation have indeed gone up.

Besides supply shocks, an imbalance in the demand and supply of a farm product, for example, is meant to be transient. If the "shocks" linger long enough (as they have in our case), they tend to embed themselves as inflationary expectations.

Perhaps the only way to excise these expectations is through monetary tightening, signalling that the central bank will not tolerate enhanced levels of inflation going forward.

Finally, in an economy like India where supply shortages are exacerbated through things like hoarding, a possible policy recourse is to hike the cost of hoarding or "carry". That too warrants monetary tightening.

The third theme that I want to emphasise relates to the choice of instruments. The current regime at the RBI has quite categorically deviated from the orthodoxy that only market-based instruments - be it the repo rate or the monetary stabilisation bonds - ought to be used in liquidity management. The central bank has chosen to resurrect the somewhat discredited cash reserve ratio in mopping up excess liquidity.

In fact, this is the third hike of its kind in the last four months. The rationale for deviating from classical monetary management principles is not that difficult to fathom. The CRR hikes have been far more successful in driving increases in borrowing and lending rates by banks than increases in signal rates like the repo rate.

The cost of using the CRR versus the cost of market instruments like the MSS bonds is also an issue. While on the MSS bonds, the government has to bear an interest charge of anywhere between 7.5 and 8 per cent, depending on the tenor, the RBI shells out a meagre 0.5 per cent on CRR balances.

This cost should not be seen in isolation. Both the CRR and MSS bonds are used essentially to impound liquidity, which gets created when the RBI intervenes in the currency market to hold the value of the rupee down.

Forex reserves earn a return of about 4-4.5 per cent. Thus on every dollar that it adds to reserves and then "sterilises" through an MSS bond issuance, it incurs a carry-cost of 3-3.5 per cent.

If it were to use the CRR, the "carry cost" is negative. The consolidated entity consisting of the government and the RBI actually makes a profit of about 3.5 to 4 per cent. As domestic rates rise, this becomes an increasingly critical issue.

In short, by hiking the CRR, the central bank creates an easy and cost-effective way of intervening in the forex market and impounding the liquidity that gets created.

Ultimately, central banking is all about striking a balance between the conflicting objectives of supporting growth and aiming at inflation control. Going by its recent moves, the RBI seems to have made a clear choice. It is hell-bent on getting headline inflation down and is willing to accept slower economic growth as its unavoidable corollary.