Money & Finance

Should India hold US$400 billion of foreign exchange reserves?

  • Blog Post Date 03 October, 2017
  • Perspectives
  • Print Page

RBI’s foreign exchange reserves have now crossed the US$400 billion mark. In this article, Dr Gurbachan Singh discusses why India’s Central Bank should not hold such large reserves.

Foreign exchange (FE) reserves with the Reserve Bank of India (RBI) have now crossed the US$400 billion mark. While this is being celebrated by many people, I take a very different view in this article.

Back to the classroom

Suppose a country follows a fixed exchange rate regime. If the demand for FE is high and the exchange rate cannot be increased to induce a reduction in demand, then the central bank needs to have adequate reserves so that it can supply FE and meet the excess demand. In contrast, under a flexible exchange rate regime, the price of FE can adjust to bring about a balance between demand and supply. In this context, FE reserves are not required.

A rough but useful rule of thumb used is that in the long run the exchange rate between currencies of two countries needs to change at a rate equal to the difference in the inflation rates in the two countries. So, there is a glaring imbalance if a country follows fixed exchange rates despite persistent higher inflation (Krugman 1979). The imbalance can be sustained with large FE reserves but eventually there is a crisis. This can be avoided if flexible exchange rate regime is adopted, obviating the need for large FE reserves.

It is true that under a flexible exchange rate regime, the flexibility can give way to considerable volatility in the FE market. In this case, it helps to have FE reserves with the central bank so that it can intervene and reduce the volatility. However, this need arises a lot more if the central bank has multiple objectives and it follows discretionary policy than in the case where the central bank has adopted ‘inflation targeting’1. In the latter case, there is less uncertainty for currency markets. The price mechanism works better in the currency market and the need for central bank intervention is less. Accordingly, the central bank does not need to hold large FE reserves, if it has adopted inflation targeting.

The story of India

The RBI has been operating under flexible exchange rate regime since the early 1990s; prior to that it followed a fixed exchange rate regime (which is, in fact, an important reason why there was a balance of payments crisis in 1990-91). Furthermore, it has formally adopted inflation targeting since June 2016 (previously the RBI followed ‘multiple indicator approach’2, which arguably played a role in the events culminating in, what may be viewed as, near-crisis in the currency market in India in 2013).

Given the two important changes in policy regimes, there is no compelling need for the RBI to maintain very large reserves now. This is true not just of India but most other emerging economies. They are mostly going through a hangover of the currency crisis that they had experienced. It is usually forgotten that the context then was different; most emerging economies now have flexible exchange rates and (formal or informal) inflation targeting.

The proposal here to cut down FE reserves is not new. It had been put forward earlier by the erstwhile Planning Commission in 2004-05 but it was criticised. However, the criticism of the proposed policy was not always valid. For example, the critique in Panagariya (2008) was actually about the short-run adjustment process involved in shifting from FE reserves to investments in infrastructure projects rather than about the issue of what is an optimal level of reserves in the long run. These arguments, to a great extent, missed the wood for the trees. It is worth revisiting the basic proposal again.

Often there is a tendency to keep reserves equal to the value of six months of imports. The rationale for this is not clear. But if at all we must think on these lines, then FE reserves are, as Professor Kaushik Basu had emphasised, required to finance only the gap between imports and exports (current account deficit) and not imports as a whole. By this yardstick, the FE reserves with the RBI at present are huge. If we take 7.5% of GDP (gross domestic product) as a figure for current account deficit (which is extremely high), then there is a need of about US$85 billion of FE reserves.3 The actual reserves are US$400 billion.

At the end of March 2017, FE reserves were equal to 78.4% of India’s total external debt. This is a very large proportion. This is particularly true when short-term debt is only 23.8% of the total debt. There are some difficulties with the definitions and concepts used but the essential story that FE reserves are large relative to the flow of hot money4remains unchanged – more so if other policies are also adopted.

Foreign exchange reserves and other policies

Conventional wisdom is that if there are sudden and large outflows of capital, the RBI must intervene and sell FE in the currency market; this would stabilise the rupee. However, there can be another form of intervention. The Ministry of Finance (MoF) can impose a tax on capital outflows, if these are sudden and large (Jeanne and Korinek 2010). Such a tax can discourage large capital outflows; the rupee is then not under pressure in the currency market.

It is true that the recent situation in India has been the opposite one. There have been large inflows of capital. The rupee appreciated and adversely affected exports. The rupee would have appreciated much more if the RBI had not intervened and bought a large amount of FE. In fact, this is basically how the FE reserves went up in recent months. So, it appears that the RBI does need to intervene often. However, again there is an alternative policy. Instead of the RBI, the MoF could have intervened. The MoF could have imposed a tax on inflows of capital (just as it can impose a tax on sudden and large outflows of capital).

If instead of maintaining large FE reserves, the funds are used to finance, say, useful infrastructure projects, the returns will be much higher. So, the opportunity cost of FE reserves is very high. In contrast, if the MoF uses the tax policy, the revenues of the Government of India rise (and the costs of large FE reserves are not incurred). So, the proposed tax policy by the MoF is superior to the RBI’s policy of using FE reserves to stabilise the rupee.

The tax considered here is not the same as ‘Tobin tax’ (this is a small tax imposed on an ongoing basis for the purpose of curbing speculation). In contrast, the tax considered here is used only in times of sudden and large flows of capital; also, it can be increased in a calibrated manner if the capital flows warrant the same.

Besides the tax policy, there is yet another safeguard available for stability of exchange rate. The public authorities (RBI or the MoF) can buy a credit line from the IMF (International Monetary Fund) or elsewhere (Singh 2014). Such a credit line is an option that gives India the right (but not obligation) to borrow if a crisis situation were to arise in future. This instrument reduces the need for large FE reserves. India already has a credit line from Japan to the tune of US$50 billion. Additional credit lines can be bought (though not necessarily from Japan). The advantage of a credit line is that its cost is much less than the opportunity cost of holding FE reserves. So, FE reserves can be much less.

India and China

It is true that China’s reserves are much larger than those of India but the Chinese story is an outlier. Many China specialists have observed that the Yuan had remained undervalued for long, which pushed exports. The large reserves are then a cumulative effect of that policy. However, that story is over and not really replicable now. Also, it is important to note that though China’s reserves were as high as US$3.84 trillion in 2014, they have now come down to US$3.1 trillion. It appears that the reserves are being brought down further. Also, US$800 billion out of the FE reserves is China’s ‘sovereign wealth fund’ (the rest are invested primarily in US dollar-denominated treasury bills on which the nominal yield at present is around 1%; earlier it was close to zero).

If the RBI must have large foreign assets (though there is really no need), then it can be divided into two parts. One part can be the standard FE reserves which are liquid and give a low return. The other part can be a sovereign wealth fund, which is invested in a variety of long-term foreign assets and gives a high return.


  1. In January 2014, the Urjit Patel Committee Report, which was appointed to ‘Revise and Strengthen the Monetary Policy Framework’, proposed a new framework for monetary policy – flexible inflation –targeting (FIT). The new framework was then formalised as an agreement between the government and the Reserve Bank of India (RBI), making price stability the primary goal of RBI, while keeping in mind the growth objective. The inflation target in the year starting April 2016 and beyond was kept at 4% (+/-2%).
  2. Under this approach, interest rates or rates of return in different markets along with movements in currency, credit, fiscal position, trade, capital flows, inflation rate, exchange rate, refinancing and transactions in foreign exchange – available on a high frequency basis – are juxtaposed with output data for drawing policy perspectives (Mohanty, 2010).
  3. The data used is that India’s GDP for 2016 was Rs. 2,263.52 billion (
  4. 4. These are funds that are controlled by investors who seek high short-term yields when the funds are likely to be reinvested somewhere else at any time.

Further Reading

Jeanne, Olivier, and Anton Korinek (2010), “Excessive volatility in capital flows: A Pigouvian taxation approach”, American Economic Review: Papers & Proceedings, 100(2):403-407.

Krugman, Paul (1979), “A model of balance of payments crises”, Journal of Money, Credit, and Banking, 11(3):311-325. Available here.

Panagariya, A (2008), India: The Emerging Giant, Oxford University Press, New York.

Singh, G (2014), ‘Systemic Sudden Stop, Credit Lines, and Funding Liquidity’, in M Callaghan, C Ghate, S Pickford and F Rathinam (eds.), Global Cooperation Among G20 Countries: Responding to the Crisis and Restoring Growth, Springer India, p. 185-198.

No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter