COMMENTARY

and there is no reason for the price of credit to increase. This has been known since Kahn and Keynes (Patnaik 2001).

Another way of seeing why it is wrong to link higher fiscal deficits to high interest rates is to note that the short-term nominal interest rate is not a marketdetermined price at all – it is an administered price of which RBI itself is the administrator. The interest rate at which RBI lends and borrows determines the interest rates which prevail in the economy. If the interest rate for private sector loans is significantly lower than what the RBI offers to those who lend to it, people would borrow from the private sector and lend to the RBI. If the interest rate for private sector loans is significantly higher than what the RBI charges from those who borrow from it, people would borrow from the RBI and lend to the private sector. Thus whatever interest rate the RBI charges on its loans becomes the interest rate which prevails throughout the economy.

Of course, loans made to the private sector would always carry a higher interest rate than loans to the RBI since private loans are risky whereas there is no risk in lending to the RBI. But since there is no reason why this risk premium demanded from the private sector would depend in any way on the level of the government’s fiscal deficit, we ignore variations in it in our discussion below and assume that a change in RBI’s interest rates translates into a one-is-to-one change in the interest rates for the private sector. We can also ignore for simplicity the lags that exist between changes in interest rates by the RBI and changes in retail interest rates.

So if the RBI wants interest rates to be low, it just needs to announce itself ready to lend and borrow at low interest rates. How much the government does or does not borrow should not have any impact on the effectiveness of this announcement. If the quantity of government bonds that are outstanding exceeds the amount of bonds people are willing to hold at the interest rate fixed by the RBI, they will offset this excess holding of bonds by borrowing an equivalent sum from the RBI.

One may object that borrowing from the RBI (which creates high powered money) is not a perfect substitute for government bonds. In our response, we distinguish between short-term and long-term government bonds. The only difference between short-term bonds and money is that the latter is slightly more liquid, i e, slightly more acceptable in exchange. However, in an economy like ours with developed financial markets this liquidity advantage is likely to be small and therefore substitution between money and short-term government bonds is unlikely to have large consequences. Moreover, this “monetisation” is a consequence of the low interest rate policy, something which is not being questioned by those like the RBI governor who want lower fiscal deficits. The case of long-term bonds is different, since holding long-term bonds involves taking interest rate risk while holding money does not. So in case the government borrows by issuing long-term bonds the substitution argument made earlier in this paragraph may not hold. But what is happening in this case is a change in maturity structure of the net stock of fi nancial assets in the economy, a policy instrument different from that of low interest rates that we have been discussing.

Sometimes the case for a high fiscal deficit raising interest rates is presented in terms of increased government borrowing pre-empting a larger proportion of the total credit available in the economy. First, we have argued that the supply of credit by the private sector is not fixed, it increases when private incomes increase as a r esult of higher government expenditure. Further, so far in our argument we have assumed that the RBI is willing to lend and borrow as much as the private sector demands at the interest rate it has fixed. The amount of credit available in the economy can therefore be restricted only if either RBI actually restricts the amount it is willing to lend or if not all agents in the private sector have access to the RBI’s borrowing window and those who do have access are not willing to borrow on behalf of those who do not.

In the first case, where the RBI is lending too little, the remedy is simple – the RBI needs to lend more rather than asking the government to borrow less. The second case is more realistic and interesting. As things stand, only banks can borrow from the RBI. It may be the case that banks refuse to lend to the rest of the private s ector, instead deciding to hold government bonds. But in this case the excess holding of government bonds is just the symptom, the real problem is banks refusing to lend to businesses. Trying to treat just the symptom by restricting the stock of government bonds will be counteracted by banks running down their reserves or by their turning around and lending to the RBI. The true cure would involve either making businesses creditworthy again, in which demand injection through fiscal stimulus can only help, or in more extreme cases by the central bank opening up its borrowing window to a larger class of p rivate agents, as the US Fed seems to be doing.

Thus, there is no reason why larger borrowing by the government need raise the short-term nominal rate of interest. It may once again be argued that what matters for the investment expenditure is not the short-term nominal rate but the long-term real rate of interest. However, this poses no new difficulty. By arbitrage, the return on long-term bonds is a product of a sequence of returns on short-term bonds,