Useful but ineffectual

Financial Express, 26 June 2008

RBI's recent moves are, to some extent, in the right direction. However, the operating framework of monetary policy is under stress and the difficulties could worsen. More importantly, this set of moves is ineffectual and will do little to impact inflation. The only real lever that the government has in fighting inflation is the exchange rate: rupee appreciation would reduce prices of traded goods and counteract inflation.

In a mature market economy, there is a clear policy rate of the central bank. Pinning down the short-term interest rate for riskless government bonds is integral to the creation of fiat money. The policy rate unambiguously pins down the short term interest rate, and expectations about the future of the policy rate trace out the entire yield curve. This is the foundation for the bond market and the banking system.

In India, the central bank does not pin down the short term policy rate. Instead, we now have an enormous band between the `repo rate' (the price at which RBI lends) and the `reverse repo rate' (the price at which RBI borrows). The reverse repo rate is now 6% while the repo rate is 8.5%. There is a gap of 2.5 percentage points between the two. In other words, RBI is leaving the short term rate to fluctuate between 6% and 8.5%. No sound central bank anywhere in the world leaves the bond market and banking in this kind of confusion.

Put together, the recent announcements add up to a 20 basis point hardening of interest rates. The rate on 90 day treasury bills has probably gone up from 8.25% to 8.45%. With inflation at roughly 10%, this is still a negative real rate. A negative real rate is a very expansionary stance of monetary policy. Much worse is the situation with bank deposits. With inflation at 10%, current accounts (which earn nothing) are getting -10% in real terms, and savings accounts (which earn 3.5%) are getting -6.5% in real terms. Every rational person will ensure that the balance held in current accounts and savings bank accounts is minimised. In the process, this will generate dislocations in banking.

A properly structured central bank forecasts inflation at future dates, and adjusts the policy rate in response to such forecasts. When anticipated inflation goes up, the central bank must always mount a greater than one-on-one response. E.g. if expected inflation goes up by 100 bps, then the policy rate must go up by more than 100 bps. In India, monetary policy has failed to react ahead of time, and has failed to show this greater-than-one-on-one property.

With inflation at 10%, the 90-day treasury bill must go above 12% for monetary policy to combat inflation. While this needs to be done, it will yield remarkably little traction in the fight against inflation. This is because India lacks the Bond-Currency-Derivatives (BCD) Nexus, the interlinked set of markets which translate changes in the short-term policy rate into changes in all interest rates in the economy. RBI's sustained efforts at preventing the development of financial markets have resulted in the absence of the BCD Nexus. As a consequence, monetary policy is rendered ineffectual. While changes in the policy rate need to be made, they do not reach out and shake the economy.

In countries like the US and the UK, central banks have not prevented the BCD Nexus from coming about. This has given them a properly functioning `monetary policy transmission'. If the US Fed or the Bank of England want to combat inflation, raising interest rates by 25 to 100 basis points over a few months generally has quite an impact. In India, if interest rates have to be used to combat inflation, massive increases are required - as was done in the mid-1990s - because monetary policy is so ineffectual.

By implementing the Mistry and Rajan reports, the BCD Nexus can be created in roughly two years time. In the short term, the lack of the monetary policy transmission is a given. While the real rate must get back to positive territory, this will not combat inflation. In other words, RBI's announcements are not going to make a dent on inflation.

With the existing policy mix, in coming weeks, elections will be uncomfortably close and inflation will not improve. What could make a difference to inflation? There is only one tool which has a demonstrable impact on inflation: the exchange rate. By reversing the capital controls of 2007 and selling reserves on a large scale, it is easy for the government to obtain a 10% rupee appreciation. Selling reserves would support the modification of liquidity conditions which would combat inflation in the medium term. We are holding excessive reserves in India; this is an ideal opportunity to undertake corrective action by selling reserves.

A 10% rupee appreciation would yield a roughly 2% impact on inflation rapidly. Rapid action of this nature would break the self-reinforcing spiral of inflationary expectations.

Back up to Ajay Shah's 2008 media page
Back up to Ajay Shah's home page