The Reserve Bank is not ‘overcapitalised’

Over the last few years, there have been demands for the use of the RBI’s internal reserves for fiscal purposes, like recapitalisation of PSBs and higher dividend payments to the government. The previous RBI Governor was fairly dismissive of this idea. It appears that the present leadership of the RBI is again facing some pressure in this regard, if the unprecedented interim dividend payment of ₹100 billion in 2017-18 is anything to go by. The basic premise of the government is that the RBI holds capital (share capital, free reserves and valuation reserves) much in excess of what is needed, and hence the government is justified in laying a claim on the free reserves. This claim is, however, contestable.

Complex issue

What quantum of economic capital is adequate for a central bank (CB) in the present era of fiat money is not easy to determine. These issues have been the subject of a good deal of debate and research over the last two decades, especially after it became known in the late 1990s that the Bank of Japan hesitated to take appropriate and timely monetary policy measures to address the country’s acute deflationary situation obtaining then, because of its concern about their likely consequences on its balance sheet.

At the heart of the issue is: what constitutes appropriate back up against the monetary liabilities from the point of view of macro-financial stability? Two distinct back-up systems exist: One, the tax-based system, in which the central bank’s balance sheet does not have much relevance and the present and future tax revenues of the government provide the ultimate support to the monetary liabilities of the RBI.

Two, the reserve-based system, in which the central bank’s balance sheet strength and internal reserves provide support for the monetary liabilities. One important operational difference between the two is that central banks of the latter category do not undertake quasi-fiscal activities and have aversion toward credit risk. Most central banks occupy intermediate positions between the two types, but the RBI is closer to the second, particularly in its evolution in the post-reform period.

The RBI’s statute does not permit buying of securities both in India and and abroad which are not issued/guaranteed/supported by the sovereign. Beginning in 1993-94, when the RBI began building its forex reserves after the BoP crisis of 1990-91, its composition of assets has undergone a major structural shift in that foreign assets (including gold) now are significantly higher than domestic assets. The former’s share in total assets, which was about 50 per cent in 2000, rose to about 80 per cent in 2018. Among other contributory factors, this transformation of the backing for RBI’s monetary liability has certainly been a reason for the decline in structural inflation in India from close to double digits in 1990s to 4-5 per cent now.

But the preponderance of foreign assets comes with a price: higher risk with lower return vis-à-vis domestic assets, necessitating higher capital. Further, there exists an extra layer of risk associated with foreign assets since their composition is not entirely a result of the RBI’s monetary policy operations: RBI buys US dollars through domestic forex market interventions, but maintains a roughly 50:50 currency and asset composition in US dollars and in other major currencies in its foreign currency assets (FCA). The RBI does this not as a monetary authority but as a financial institution with its particular risk-return strategy for portfolio management of foreign currency assets.

Nevertheless, the strategy entails higher risk, the chief manifestation of which is the valuation change of foreign currency assets from time to time. The conclusion here is the RBI needs to maintain significant capital to cover all the currency, interest and credit risks that it faces, not counting operational risk and the risks due to its domestic liquidity operations, lender-of-the-last-resort function, expected support to DICGC, various types of legal risks and any quasi-fiscal operation that it could be asked to undertake.

Augmenting internal reserves

The RBI’s free reserves comprise its reserve fund, asset development fund and contingency fund (CF), the last one providing the overwhelming bulk and being the first cushion for absorbing general loss. It was almost entirely exhausted in 1993-94 as a result of the exchange rate guarantee that the RBI provided to FCNRA deposits introduced in early 1980s. Largely as a consequence of this experience, the RBI felt the need to adopt a policy in 1997 to build its CF, in terms of which a target 12 per cent for the ratio of CF to its total assets was set. Subsequently in 2004, a detailed review exercise in this regard was undertaken but its recommendations were not accepted and the status quo ante was maintained. The CF was subsequently built up, reaching a high of 10.9 per cent of total assets in 2008-09.

Currently, this ratio is much lower. It stands at a little over 6 per cent and shows a declining trend. The CF in relation to total assets at little over 6 per cent is much lower than the target and is declining. This is worrying and can have adverse implications for macro-financial stability. The large balance in the currency and gold valuation account of ₹6,916.41 billion as on June 30, 2018 provides little comfort, since this balance is available only for absorption of currency and gold valuation losses. The importance of the CF can be gauged from the fact that in 2017-18 alone, valuation loss of ₹168.74 billion in foreign securities portfolio was charged to the CF.

Way forward

The RBI should adopt a policy framework to determine dividend payable to the government each year, after assessing CF adequacy vis-à-vis an identified set of risks, following a rich methodology. This should be supplemented by an analysis of the RBI’s earnings under the following categories:

Seigniorage: Earnings on investing the counterparts of monetary liability (reserve money), which as on June 30, 2018 was about 69 per cent of total assets. This is the expected core dividend, net of proportionate share of operational expenses and valuation deficits, if any.

Counterpart of capital: Earnings on investing capital.

Counterpart of other liabilities: Earnings on other liability items.

Currency risk: Valuation change of FCA, which needs to be recognised separately in the books of accounts.

The earnings of the last three categories should be used to consolidate the CF and also to smooth dividend payments. Among other benefits, this approach will incentivise the RBI to follow appropriate policies for portfolio management of FCAs.

(The writer is a former central banker and consultant to the IMF)