It can be argued that the CRR should remain, and RBI has been fair to banks as the cost of the CRR is permitted to be included in the calculation of the base rate and the MCLR.
With the credit policy coming up next week and the cash reserve ratio (CRR) increase already in force by 50 basis points (bps), it may be time to reconsider the value of having a CRR in place. This also rhymes with what the late Deputy Governor of RBI, KC Chakrabarty, argued for—doing away with the CRR. Today, the net demand and time liability (NDTL) for the banking system is around Rs 160 lakh crore, and 4% CRR means around Rs 6.4 lakh crore is impounded by the Reserve Bank of India (RBI) under this stipulation on which no interest is earned. Further, the markets were quite sensitive to the RBI announcement in the last policy of increasing the CRR in two steps. G-Sec yields were nudged up as it was assumed that the signal was one of tightening rates even though it was pointed out by the central bank that this was not the case. There is compelling reason to revisit this issue.
A CRR has been defined by most central banks because it is meant for solvency of the banking system. If a bank goes bust and there are issues in liquidity, the central bank can use the cash that has been impounded to make the necessary payments to begin with, before using other measures to save the deposit holders. In the Indian case, this has never been used and all bank failures have been caught early by the central bank and action taken. Where the central bank has been caught off guard, the CRR money has not been deployed to pay the deposit holders. In fact, there have been restrictions put on withdrawals and the CRR was never used for this purpose. The deposit insurance scheme is already there to address issues of safety of bank deposits. Therefore, the solvency explanation is not too strong.
The other justification for the CRR is that it is part of monetary policy toolkit, and by increasing or decreasing this ratio, RBI can manage liquidity. Hence, unlike repo rate where changes can only nudge the banks to follow suit, a CRR change deals with the supply of liquidity directly, which, in turn, affects interest rates. Intuitively, if the CRR is increased, the supply of lendable funds falls and rates would go up. Unlike repos/reverse repos which are of a short tenure or OMOs which are of smaller quantities, the CRR is a permanent deal with liquidity. Therefore, the impact is sharper here. Quantitative measures like the CRR have an advantage of being large and direct, and hence effective. However, it is a one-time shot. And once the CRR change is absorbed by the system, a reversion to earlier equilibrium is possible. OMOs, on the other hand, are smaller doses which can be used periodically to steer the market.
Globally, the CRR exists and is as high as 17% in Brazil, 11% in China and 8% in Russia. These rates are much higher than in India (which will soon be 4%). It is nil in the US, 1% in the UK and 2.5% in South Africa. Therefore, there are different ratios in countries depending on local conditions. But the concept is not alien.
Now, bankers would always argue that the CRR should not be there. The idea of collecting deposits is to lend the money to the productive sectors. By impounding funds through the CRR, there is loss of liquidity. Here it can be counter-argued that banks anyway are never lending all their funds and are investing in government securities far in excess of what is required, either out of regulatory pressures of Basel III or preference for the same. Therefore, even if they had these funds, it would not have been necessarily used for credit deployment. In fact, in FY21, RBI had lowered the CRR by 100 bps, which would be roughly Rs 1.5 lakh crore, which could be matched with large reverse repo deployments through the year (of the order of above Rs 5 lakh crore on a daily basis). Hence, the original objective of lending to industry was never achieved as the money went back to RBI and earned 3.35% interest. But the market was happy that more funds were permanently made available to the system.
Then there is the question of interest payment on the CRR. Earlier there was an interest paid on CRR balances till 2007. This made sense when the CRR was in the double-digit range. Now, while demanding an interest payment on CRR payment is legitimate, it should be realised that banks actually have around 9-10% of their deposits that are rolled over continuously as demand deposits. No interest is paid on these deposits, but the funds are deployed for lending as there is stability in these deposits, just like savings deposits which are around 25% and cost not more than 3%. It can be argued, therefore, that as money is fungible, the 9% demand deposits that come free of cost have a part withdrawn through the CRR by RBI which is interest free. Hence, banks are not really losers given the inherent structure of banking in India.
On balance, it can be argued that the CRR should remain, and RBI has been fair to banks as the cost of the CRR is permitted to be included in the calculation of the base rate and the MCLR (Marginal Cost of Funds Based Landing Rate). Therefore, the cost is loaded finally to the borrower and not really borne by the bank exclusively.
The purpose so far has mainly been to use the CRR as a monetary policy tool rather than a final recourse for failing banks. In this situation, a pertinent question to ask is, can these funds be used for some productive purpose? It is important because even forex reserves which are accumulated by the central bank are deployed in federal bonds or other investments. Here, the existing balance of, say, Rs 6 lakh crore is idle, and it is possible for these funds to be deployed by RBI. These funds can be partly used for zero-cost emergency lending to the government for short-term periods including WMA (ways and means advances) which have the advantage of not leading to creation of new reserve money. Probably RBI can decide on what part of the CRR can be used for these purposes while the base CRR money which can be defined as being 50% is used for monetary policy purposes. How about lending to the new development finance institution (DFI)? It cannot be for the long-term and has to be for the short-term only because otherwise the conduct of CRR tool for monetary policy will be impeded. A discussion needs to start on this subject for sure.
The author is Chief economist, CARE Ratings, and the author of ‘Hits & Misses: The Indian Banking Story’. Views are personal