Let RBI make money from forex reserves

As foreign currency comes in through exports, invisibles, foreign investment, NRI deposits and external commercial borrowings (ECBs), banks hold on to what is required to support other forex outflows and the balance gets accumulated as reserves with RBI. These are exogenous inflows as far as the central bank is concerned. RBI accounts show that, for FY20 the earnings on forex assets were 2.65%. The question really is whether the central bank can earn more? It is in this context that the debate has erupted.

Related posts

EXCLUSIVE – Crypto giant Binance controlled ‘independent’ US affiliate’s bank accounts

June 5, 2023

Why investing just to save tax may be bad for you

June 5, 2023

At one level, the issue can be brushed aside on the grounds that the central bank is not a commercial entity and should not be looking to earn money. The job is to ensure that the reserves are safe, and deployment in sovereign bonds of some countries offers a solution. They can be converted to dollars whenever required, and were another Lehman-like crisis to take place, RBI can get the dollars back. In FY09, forex reserves fell by $58 billion, which is the highest decline in any single year. The argument, hence, is that it is not the job of the central bank to earn money on what is in safe custody merely because the need for the same may never really arise. A crude analogy: Quite in the manner a commercial bank cannot use what is kept by customers in the lockers to earn money, the central bank can’t earn off the nation’s forex reserves.

In the last couple of years, however, RBI has become a critical part of the Union Budget as the surpluses are transferred to the Centre, and these transfers can go up to as much as Rs 1 lakh crore and add to non-tax revenue. So much so that there were strong arguments made and executed over transferring of reserves to the government. A special committee was set up to advice on the same, as the central bank did not seem too inclined to do so as the balance sheet is really a notional one for any such authority where there are no limits on issuing currency. But, ultimately, there was a recommendation for the same as fiscal compulsions have prevailed.

Hence, from the point of view of fiscal argument, there is a case for working on the feasibility of this proposition. In fact, with the ushering in of LTROs where funds are being provided to banks at the repo rate, disbursal of Rs 1 lakh crore for a year will earn Rs 4,000 crore to RBI, which under ceteris paribus conditions will get transferred to the government at the end of the year. Now, with reserves of $515 billion, we are talking of deploying around Rs 38 lakh crore of foreign currency assets, which is a big amount. Intuitively, if the earnings can be enhanced by 1% on this sum, there could be around Rs 38,000 crore earned by the central bank. Can this theoretically happen?

To begin with, any exercise of investing the dollars outside the present circuit of sovereign bonds of other governments would be only partial, as a core component of the reserves has to be maintained for all times in this ring. This can be kept at six or eight months of imports while deploying the balance in higher earning avenues.

The first thought that comes to mind is investments in AAA-rated international corporate bonds. At present, the yields are not very different from what is being earned on sovereign bonds, which means that there would have to be a step down in terms of quality of the bonds. A call also must be taken on whether RBI would be a ‘buy and hold’ player or a trader in this segment? The additional consideration would be to mark to market (MTM) the portfolio and the option to sell in the secondary market, which would be a major liquidity consideration. Forex reserves can, however, become volatile in case the MTM results in lower valuation. Are we prepared for this? Currently, it has been observed that the gold reserves carry the same volatility factor as change in the price of gold will involve valuation adjustments.

The issue really is that if there is a default, then the loss must be borne by the central bank, which is not acceptable. How about lending to banks? RBI can consider lending to foreign banks in foreign territories, but these banks anyway get money at a lower cost and hence will not find a cost of, say, 3.65% (2.65 plus 100 bps) attractive. Hence, practically speaking, finding a safe zone for investment outside the country will be difficult and any inroads outside the current domain will require a modicum of risk-taking, which involves treasury activity. The central bank probably may not like to get into this line as in general such authorities do not take on the role of traders.

If one were to think out of the box on the issue, the question to be asked is whether we can mimic other markets used by Indian borrowers for forex loans? The ECB market is attractive as it offers companies with a certain stature access to foreign markets. Companies borrow from this market as the cost is lower. Therefore, RBI can consider lending the same to Indian commercial banks, which can offer forex loans to those companies that currently borrow in the ECB market. With the rate of benchmark of LIBOR plus 450 bps being the ceiling, banks can be asked to lend to these companies in an analogous manner. There would, however, be a risk of default, but that would have to be taken by the bank and not RBI. This way, RBI can lend to banks at, say, more than 265 bps depending on what yield is being targeted, and banks can add their spread on the same. The final rates must be made comparable with those in the ECB market or else companies will not find these attractive.

If we are not averse to the concept of RBI deploying excess forex reserves for commercial returns, there are strong grounds for exploring such options. This will mean breaking away from the conventional central bank and turning partly commercial. There has been a discussion even earlier on RBI funding infra finance from these reserves and hence the idea is not new, though motivation is different.