Themes > News Analysis       Print           
Print this article
Revisiting Capital Flows
C.P. Chandrasekhar & Jayati Ghosh

A striking feature of the recent global financial crisis and its aftermath is the behaviour of private international capital flows, especially to emerging markets. Prior to the crisis, in the years after 2003, a number of analysts had noted that the world was witnessing a surge in capital flows to emerging markets. These flows, relative to GDP, were comparable in magnitude to levels recorded in the period immediately preceding the financial crisis in Southeast Asia in 1997. They were also focused on a few developing countries, which were facing difficulties managing these flows so as to stabilise exchange rates and retain control over monetary policy. They also included a significant volume of debt-creating flows, besides other forms of portfolio flows.

Interestingly, these developments did not, as in 1997, lead up to widespread financial and currency crisis originating in emerging markets, as happened in 1997. However, the risks involved in attracting these kinds of flows were reflected in the way the financial crisis of 2008 in the developed countries affected emerging markets. Financial firms from the developed world, incurring huge losses during the crisis in their countries of origin, chose to book profits and exit from the emerging markets, in order to cover losses and/or meet commitments at home. In the event, the crisis led to a transition from a situation of large inflows to emerging markets to one of large outflows, reducing reserves, adversely affecting currency values and creating in some contexts a liquidity crunch.

Given the legacy of inflows and the consequent reserve accumulation, this, however, was to be expected. What has been surprising is the speed with which this scenario once again transformed itself, with developing countries very quickly finding themselves the target of capital inflows of magnitudes that are quickly approaching those observed during the capital surge. As the IMF noted in the latest (April 2011) edition of its World Economic Outlook: ''For many EMEs, net flows in the first three quarters of 2010 had already outstripped the averages reached during 2004–07,'' though they were still below their pre-crisis highs.

One implication of the quick restoration of the capital inflow surge is the fact that, in the medium-term, net capital inflows into developing countries in general, and emerging markets in particular, has become much more volatile. As Chart 1 shows, net capital flows which were small though the 1980s, rose significantly during 1991-96, only to decline after the 1997 crisis to touch close to early-1990s levels by the end of the decade. But the amplitude of these fluctuations in capital inflows was small when compared with what has followed since, with the surge between 2002 and 2007 being substantially greater, the collapse in 2008 much sharper and the recovery in 2010 much quicker and stronger.

When we examine the composition of flows we find that volatility is substantial in two kinds of capital flows: ''private portfolio flows'' and ''other private'' flows, with the latter including debt (Chart 2). There has been much less volatility in the case of direct investment flows. However, in recent years the size of non-direct investment flows has been substantial enough to provide much cause for concern. Further, besides the fact that direct investment flows are differentially distributed across countries (with China taking a large share), the definition of direct investment is such that the figure includes a large chunk of portfolio flows. The magnitude of the problem is, therefore, still large.

Does this increase in volatility during the decade of the 2000s speak of changes in the factors driving and motivating capital flows to emerging markets? The IMF in its World Economic Outlook does seem to think so, though the argument is not formulated explicitly. In its analysis of long-term trends in capital flows the IMF does link the volatility in flows to the role of monetary conditions (and by implication monetary policy) in the developed countries, especially the US, in influencing those flows.

As the WEO puts it, ''Historically, net flows to EMEs have tended to be higher under low global interest rates, (and) low global risk aversion,'' though this assessment is tempered with references to the importance of domestic factors. Shorn of jargon, there appears to be two arguments being advanced here. The first is that capital flows to emerging markets are largely influenced by factors from the supply-side, facilitated no doubt by easy entry conditions into these economies resulting from financial liberalisation. The second is that easy monetary policies in the developed countries has encouraged and driven capital flows to developing countries. This is because easy and larger access to liquidity encourages investment abroad, while lower interest rates promote the ''carry-trade'', where investors borrow in dollars to invest in emerging markets and earn higher financial returns, based on the expectation that exchange rate changes would not reduce or neutralise the differential in returns. Needless to say, when monetary policy in the developed countries is tightened, the differential falls and capital flows can slow down and even reverse themselves.

The evidence clearly supports such a view. The period of the capital surge prior to 2007 was one where the Federal Reserve in the US, for example, adopted an easy money policy, involving large infusion of liquidity and low interest rates. While this was aimed at spurring credit-financed domestic demand, especially for housing, so as to sustain growth, it also encouraged financial firms to invest in lucrative markets abroad. Flows reversed themselves when the losses and the uncertainty resulting from the sub-prime crisis and its aftermath resulted in a credit crunch. Finally, flows resumed and rose sharply when the US government responded to the crisis with huge infusions of cheap liquidity into the system, aimed at relaxing the liquidity crunch. A substantial part of the so-called stimulus consisted of periodic resort to ''quantitative easing'' or the loosening of monetary controls.

This close link between monetary policy in the developed countries and capital flows to emerging markets is of particular significance because, with the turn to fiscal conservatism, the monetary lever has become the principal instrument for macroeconomic management. Since that lever can be moved in either direction (monetary easing or stringency), net flows can move either into or out of emerging markets. As a corollary, the consequence of monetary policy being in ascendance is a high degree of volatility and lowered persistence of capital inflows to these countries.

From the point of view of developing countries the implications are indeed grave. When global conditions are favourable for an inflow of capital to the developing countries, these countries experience a capital surge. This creates problems for the simultaneous management of the exchange rate and monetary policy in these countries, and leads to the costly accumulation of excess of foreign exchange reserves. Costly because the return earned from investing accumulated reserves is a fraction of that earned by investors who bring this capital to the developing economy. Moreover, when global conditions turn unfavourable for capital flows, capital flows out, reserves are quickly depleted and there is much uncertainty in currency and financial markets.

The problem is particularly acute for countries that are more integrated with US financial markets, since dependence on the monetary level is far greater in that country, partly because of the advantages derived from the dollar being the world's reserve currency. The IMF's WEO, therefore, predicts: ''economies with greater direct financial exposure to the United States will experience greater additional declines in net flows because of U.S. monetary tightening, compared with economies with lesser U.S. financial exposure.'' This tallies with the evidence. Overall, ''event studies demonstrate an inverted V-shaped pattern of net capital flows to EMEs around events outside the policymakers' control, underscoring the fickle nature of capital flows from the perspective of the recipient economy.''

This increase in externally driven vulnerability explains the IMF's recent rethink on the use of capital controls by developing countries. Having strongly dissuaded countries from opting for such controls in the past, the IMF now seems to have veered around to the view that they may not be all bad. However, its endorsement of such measures has been grudging and partial. In a report prepared in the run up to this year's spring meetings of the Fund and the World Bank, the IMF makes a case for what it terms capital flow management measures, but recommends them as a last resort and as temporary measures, to be adopted only when a country has accumulated sufficient reserves and experienced currency appreciation, despite having experimented with interest rate policies. This may be too little, too late. But, fortunately, many developing countries have gone much further. Only a few like India, which is also the target of a capital surge, seem still ideologically disinclined.

May 05, 2011.


© International Development Economics Associates 2011