After the dollar deluge, costly choices

The Foreign Currency Non-Resident (Bank) or the FCNR(B) scheme has largely been celebrated as a successful defence of the rupee. It attracted dollars and prevented disorderly depreciation. But stabilising the rupee was only the first step. By absorbing these dollars, the Reserve Bank of India (RBI) has created an immediate domestic-liquidity problem. The scheme has already imposed costs, even before any future exchange-rate loss or repayment pressure arises. 

The special facility mobilised $136.4 billion, including $127.2 billion through FCNR(B) deposits. As the RBI absorbed much of the foreign exchange and released rupees in return, surplus liquidity in the banking system rose beyond ₹10 trillion. What should the RBI now do with this money? Every available choice is costly. 

The first option is to allow the dollars into the market and let the rupee appreciate. This was appropriate in 2013, when India’s current-account deficit was 4.7 per cent of gross domestic product (GDP), inflation was high and the economy was in serious trouble. The rupee faced a genuine run. The FCNR(B) and related windows attracted about $34 billion and helped arrest the collapse. 

The present episode is different. The rupee’s real depreciation against a broad basket of trading partners was around 10 per cent, less than half the comparable decline in 2013. India has substantial reserves, healthy growth, strong remittances and a current-account deficit close to zero. This was not a run. 

Allowing the entire inflow into the market could nevertheless have pushed the rupee below ₹90. That would have hurt exports, cheapened imports and weakened domestic manufacturing. It would also have subsidised foreign portfolio investors’ exit by giving them more dollars for the same rupee proceeds. Absorbing much of the inflow was therefore justified, but it has converted an exchange-rate problem into a liquidity problem. 

The second option is to leave the reserves with banks to support lending. But reserves are high-powered money. When banks lend, they create deposits. Credit and broad money could therefore expand by several times the original injection, producing inflation after a lag. 

There is a deeper danger. Policymakers should carefully read Viral Acharya, Rahul Chauhan, Raghuram Rajan and Sascha Steffen’s paper, “Liquidity Dependence: Why Shrinking Central Bank Balance Sheets Is an Uphill Task”. 

The authors show that banks reorganise their balance sheets around abundant reserves by creating deposits, credit lines and other liquidity commitments. These obligations remain when the central  bank removes the reserves. Banks may then hoard liquidity, curtail lending, sell securities or scramble for short-term funding. The central bank may be forced to inject reserves again. 

This is the reserve ratchet. Creating reserves is easy. Removing them after banks become dependent on them can be disruptive. India could first experience inflation and then discover that controlling it threatens financial stability. 

The third option is to increase the cash reserve ratio (CRR). This would immobilise the reserves by requiring banks to hold additional funds with the RBI without interest. 

The burden would be large and uneven. Banks that raised FCNR(B) deposits must pay depositors as much as 6 to 7 per cent. If the corresponding rupees are locked in CRR, they earn nothing. A system-wide increase would also penalise banks that did not mobilise these deposits. 

A fourth option is for the RBI to sell government securities already on its balance sheet. No new debt would be created. Existing debt would move from the RBI to banks. But sales on this scale could raise bond yields, reduce the RBI’s income and exhaust securities needed for normal monetary operations. 

Fresh ordinary government bonds would not permanently absorb liquidity if their proceeds were spent, because the money would return to the banking system. Banks would effectively have borrowed abroad, swapped the dollars into rupees and financed the government, while the RBI guaranteed the future exchange rate. This begins to resemble indirect sovereign foreign-currency borrowing. 

The final option is the market stabilisation scheme. Under MSS, the government issues special bonds and impounds the proceeds. The money cannot be spent and therefore remains outside the banking system for the life of the bonds. 

Sterilising ₹7 trillion to ₹8 trillion  at 6 per cent for three years would require gross interest payments of approximately ₹1.25 trillion to ₹1.45 trillion. Taxpayers would ultimately bear this cost through higher interest expenditure or reduced fiscal space.

In my view, MSS is the least damaging option. It makes the cost explicit, prevents a large monetary expansion and does not impose the entire burden on banks. Most importantly, it prevents the banking system from becoming dependent on reserves that the RBI may later struggle to withdraw. 

The scheme may have stabilised the rupee, but it did not eliminate costs. It merely shifted them. India must now choose among currency appreciation, inflation, bank losses, indirect public exposure and an expensive sterilisation programme. Having reached this position, paying the visible cost of MSS is preferable to gambling with inflation and the reserve ratchet.

 

The writer teaches at Indian School of Business