Let me begin by expressing my appreciation of Sudipto Mundle (Finding A Goldilocks Policy Interest Rate Mint, 19 October) for bringing to public attention the extremely important, but neglected, issue of the appropriate level of interest rates in India. In recent years, the discourse has focused on whether the current levels of interest rates should go up or down without reference to any benchmark. The general attitude is that interest rates are like any other price, determined in the market by supply and demand, with the government’s role limited to manipulating it at the margin to sub-serve some policy objective, usually inflation rate.
This view is incorrect, and perhaps even dangerous. The level of interest rate should be a key public policy objective in itself since it is central to determining savings and investment, choice of technologies and the structure and time-path of the economy. Moreover, it is naïve to believe that interest rates are “pure” prices determined entirely by the market. Interest rates on public debt instruments of different maturities set the floor for all interest rates in the economy.
Mundle makes a strong case for the repo rate to be only marginally above the expected inflation rate. He works backwards from a normative estimate of the long-term interest rate, i.e. on 15-year government bonds. This is necessary since there is no real theory of short-term interest rates, but there are two established theories to guide the choice of the long-term rate: (a) social rate of time preference (Fisher-Ramsey tradition); and (b) natural rate of interest (Wicksellian tradition). Mundle mainly relies on the former by invoking the 1982 report of the Sukhamoy Chakravarty committee on the working of the monetary policy system.
The committee recommended that the interest rate on 15-year government bonds should be 3% above expected inflation rate. However, as Mundle rightly points out: “This social rate of time preference should be determined in principle by elected governments in representative democracies”. The elected union government of that time disregarded the recommendation and fixed the 15-year bond rate at 6% above the target inflation rate (12% nominal). This rate persisted until the late-1990s, post which it was brought down to 5% (9th Plan) and then to 4% (11th Plan).
Why did Chakravarty recommend a rate so much lower than what was politically acceptable? He was possibly influenced by Frank Ramsey’s ethical preference for a zero rate of time preference, which means that consumption of future generations should be valued on a par with that of the present generation. In a country with high poverty levels, such a position seems hard to justify, at least politically.
Be that as it may, the real yield on 15-year government bonds at 2.75% is already below Chakravarty’s 3% and the 11th Plan’s 4%. I believe further reduction is not justified; indeed, it should rise. More importantly, moving from a normative estimate of the long-term interest rate to the short-term requires some sense of the desired slope of the “yield curve”—which can crudely be expressed as the difference between the interest rates on the longest and shortest maturity government securities. Chakravarty’s recommendation was 3 percentage points, which I concur with, but the actual today is a mere 0.75 percentage point, and it has been so for several years. This is a much more serious issue than the quibbles about the long-term rate, and one which has received scant mention in recent discourse, whether in the public domain or government or even the monetary policy committee.
The long-term rate and the yield curve together determine the minimum that the repo rate can be reduced to. A repo rate significantly lower than the T-bill yield will create an arbitrage opportunity for banks whereby they borrow from the Reserve Bank of India (RBI) only to re-invest in T-bills, and the process goes on until interest rates equalize at the new repo rate, with unearned profits for banks and significant increase in money supply. Seen in this light, the present 6% repo rate appears to be about as low as the RBI can go given the yield curve.
But such a flat yield curve raises more fundamental problems. Short-maturity debt is usually used for working capital and short-gestation investment, while longer-maturity debt for capital-intensive, long-gestation investment. The present yield structure effectively penalizes current production and labour-intensive investments in favour of capital-intensive products and processes. Thus, the current problems of low employment growth and increasing automation can, at least partly, be attributed to an undesirably flat yield curve.
How has this state of affairs come to pass? I believe it is the outcome of excessive focus on the repo rate as the principal instrument of monetary policy. Consequently, management of the yield curve has dropped out of consideration as an objective of monetary policy, since it cannot be done merely by playing around with the repo rate. It requires active open-market operations in which the maturity structure of public debt instruments is as important as the volume.
As things stand, there are too few long-term government bonds and too many shorter-term ones in the market. This may not entirely be RBI’s fault since the finance ministry could have insisted on issuing more short-term debt as a method of reducing the government’s interest burden. At present, however, with just a 0.75 percentage point spread, the gain to the government simply does not justify the damage to the economy. The process of correction needs to begin now, with gradual reduction in the volume of short-term government paper and corresponding increase in longer maturities. If done right, there is no reason why the repo rate cannot be reduced by another 1 percentage point, provided that the top rate is raised to at least 8%.