Self-insurance—The Debate India Must Have
by Arvind Subramanian, Peterson Institute for International Economics
Op-ed in the Business Standard, New Delhi
November 26, 2008
© Business Standard
This crisis must leave Indian policymakers shaken, not just stirred. Unlike the Asian crisis of ten years ago, the financial impact of this crisis has been severe (we are yet to see the full effects on the trade and real sides). The stock market has lost about half its value, the rupee has depreciated by 25 percent, reserves have declined by about $60 billion, and the successive waves of liquidity scrambles have been unprecedented.
This financial contagion should not have come as a surprise because, unlike in the 1990s, India has become highly integrated in financial terms. But a surprise it has been, in part because everyone underestimated the extent of financial integration. Large inflows from the outside into the stock market were visible. The less-visible aspect of integration was the foreign funding of Indian financial institutions and firms, especially those that had borrowed abroad to finance their mergers and acquisitions in the last few years.
So, lesson number one is: The greater the financial integration, the greater the susceptibility to financial crises, and the larger their final cost.
Lesson number two is: Self-insurance (the accumulation of foreign exchange reserves) helps and absolute self-insurance helps absolutely. Call it the Powell doctrine of financial crises.
It is undeniable that India's foreign exchange reserves have helped in limiting the impact of the crisis. If India had gone into the crisis with $100 or even $150 billion of reserves instead of $300 billion, the rupee would have declined more steeply and more quickly. China, with its nearly $2 trillion of reserves, suffered very little effect on the currency.
In the aftermath of the Asian financial crisis, India did not consciously embark on self-insurance. Its sizable precrisis reserve levels reflected less a deliberate policy choice than favorable circumstances, including the software export boom.
In thinking about and preparing for the next crisis, two questions arise: Should self-insurance become an explicit policy objective and, if so, how much self-insurance should India take out? What does self-insurance imply for other policy choices such as the exchange rate and capital account opening? Consider each question in turn.
A lot, of course, will depend on how events unfold over the next few months. This crisis is far from over. As the financial crisis morphs into a real sector crisis, we will see growth declines, corporate sector difficulties, and further asset price declines. Will residents retain faith in the economy and the rupee? If they do, we will emerge from the crisis with about $200 billion in reserves and a currency no worse than say Rs. 50–52 to the dollar. Policymakers will feel vindicated in their unconscious neglect of the self-insurance objective, and will target some modest build-up of reserves for the future, hopefully calibrated to the level of future capital inflows.
If, on the other hand, the crisis turns ugly in terms of real sector performance, and residents' confidence in the rupee is affected, we will see a very different outcome. When the dust settles on this crisis, the rupee could touch levels of Rs. 55–60 to the dollar, and reserves could decline to uncomfortable (double-digit) levels. If that happens, policymakers will, and should, embrace self-insurance with a vengeance.
What future reserve levels should India target? Clearly, these will have to be related to potential capital outflows during a crisis. The standard Guidotti rule—that reserves should be as much as debt falling due within a year—no longer provides an adequate basis for determining self-insurance. The recent experience suggests that any funding by Indian firms in foreign capital markets (such as those of TATA), which may not be included in domestic external debt, should also be included in thinking of the size of potential outflows.
But one cannot stop there. In the recent crisis, nonresident investments in the equity market were also prone to sudden withdrawal. Withdrawals of portfolio investments have a self-limiting aspect to them: Prices adjust to moderate outflows. But even if they are self-limiting, the process can easily play itself out enough to create significant pressures on the currency, especially if there have been sizable cumulative inflows.
One could go even further. If there are no capital controls on residents, crises could also lead to pressure on foreign exchange reserves from residents fleeing rupees directly or indirectly through trade channels. If governments want to cushion against all these sources of outflows, prudential levels of reserves could be very high: A $1 trillion figure for India would not seem excessive.
What does self-insurance imply for policy, especially exchange rate policy and capital account opening? It is preferable to build up reserves by running current account surpluses, which in turn requires a competitive exchange rate. Maintaining a competitive exchange rate requires policy and structural reforms, but it is rendered difficult, if not impossible, when there are large inflows of capital. Such inflows may have long-run benefits, but from a self-insurance perspective, they are a double whammy: They increase the vulnerability to crises but also make self-insurance against crises more difficult because competitive exchange rates are less easy to maintain. Self-insurance therefore requires a mercantilist slant to exchange rate policy and caution about capital account opening.
But self-insurance is not without costs even as insurance. This strategy will increase reliance on exports and foreign markets for sustaining growth. Export-to-GDP ratios will increase, but the flip side is that India's exports and growth will become more vulnerable to growth slowdowns in foreign markets. So, what a country gains from self-insurance on the swings (protection against financial contagion) it loses partly on the roundabouts (vulnerability to trade contagion). Policymakers will have to trade-off these benefits and costs.
Some time in the not-too-distant future, when the storm clouds recede, when the rupee is at, say, Rs. 50 to the dollar, Indian exports will be hypercompetitive, and Indian growth prospects will be restored to precrisis levels of 8–9 per cent. At that stage, capital, attracted by the higher returns, will once again come pouring into India. That is almost certain. If India were to welcome all that capital, the rupee will very quickly find itself at Rs. 40 or even Rs. 35 to the dollar, leading to current account deficits and undermining India's ability to become self-insured consistent with the Powell doctrine. What quantum of solace will that provide?