Case against Rushing intoFull Capital AccountConvertibility
A strong financial sector is required if a nation is to reap the potential gains from trade in assets. In the financial markets, the collapse of a few institutions could lead to a collapse of the entire system. The experience of emerging economies suggests that one should approach the issue of throwing open the capital account with extreme caution.
PARTHA SEN
I
India is a latecomer to the debate involving capital account convertibility (indeed it is a latecomer to the whole set of macroeconomic management issues) and there are benefits from learning about the advantages and pitfalls of having a more open capital account. The Asian crisis was a setback to those who believed that the markets should be given a free rein. These people are now back pointing to the mountain of foreign exchange that India is sitting on – a mere hillock compared to China and others in Asia – to make a case for immediate liberalisation of the capital account.
Our experience with unfettered capital mobility is recent. In the last quarter of a century there have been two peaks in net yearly private flows to developing or emerging countries. (I will use these expressions interchangeably.) The first peak of about $ 50 billion occurred in the early 1980s. The second peak of about $ 250 billion occurred just prior to the Asian crisis. Since then net flows to the emerging markets have trickled to zero or even turned negative – net transfers to Latin America including interest and profit repatriation are about minus $ 40 billion (India and
Economic and Political Weekly May 13, 2006 China are exceptions).1 A worrying aspect of this drought is that it includes trade credit. One implication of this is that the positive aspect of a currency crisis, viz, a depreciation that used to stimulate exports, does not do so anymore because of the paucity of trade credits.2
The period is then too short for theorising on a grand scale – not that this has prevented anyone from doing so if they so wished. But there are some empirical regularities that one should bear in mind.
Broner and Rigobon (2004) analyse capital flows to 23 industrialised countries and 35 “emerging” countries – they looked at the capital account surplus for each country – using annual data over the period 1965-2003.3 They found that the latter group has a standard deviation that is 1.8 times the former, and (left) skewness that is 1.5 times as much. They try to provide three sets of explanations: (a) the fundamentals may be different for the two groups;
(b) the response to fundamentals may be different; and (c) non-fundamental shocks. The last set of factors includes “crises” or the role of outliers, “persistence” or the role of lags, and “contagion” or comovement. They found that the third set of factors explain both the high variance and skewness, although country characteristics also matter.
Developing countries are told by mainstream economists to liberalise their trading regimes, and with some justification. Outward-looking economies (barring primary goods exporters) have done well from the GATT regime. Trade in goods at its crudest may be thought of as being timeless. If the quality of goods is not an issue, then exchange can take place at any instant.
Should this enthusiasm for free trade carry over to trade in assets? At the level of an individual it is unfair that while a supplier may export a commodity and get a higher price abroad, a borrower should be prevented from accessing a supplier of loans abroad. Proponents of liberalisation of the capital account point to the opportunities for a small economy that the international market provides for diversification of idiosyncratic risk. Presumably, by holding a diversified portfolio one can buy insurance against country specific (temporary) shocks. Trade in asset markets inevitably involves time. Borrowing is to be repaid at a later date, etc. Contracts have to be written, issues of liquidity worked out. For the contracts to be enforceable, there has to be a legal authority, whose role is much more complex than one which might be checking on the quality of goods. Bankruptcy and the so-called inalienability of human capital only add to the complexity of the issues at hand. Underdeveloped countries with ill-functioning (or worse, non-existent) legal systems and thin (or worse, non-existent) markets need more care. Crony capitalism and asset price collapses are indicators of these – although on both counts Texas is also a candidate!
Historically, capital inflows to the colonies took the form of foreign direct investment (FDI) in mining and plantations and trade credits (where collateral is readily available). Also historically, trade credits (which are very short term) and FDI (which is long term) were both very stable components of capital flows. In the next historical episode, inflows took the
INDIRA GANDHI INSTITUTE OF DEVELOPMENT RESEARCH
Applications are invited for the post of Assistant Librarian, in the pay scale of Rs. 8000-275-13500. The post will be filled either on deputation basis or on contract basis for a period of one year. In case a regular vacancy occurs at the end of the one-year period, the selected person may be considered for regular appointment subject to satisfactory report on his/her performance, etc.
Minimum essential qualifications, experience and age: Master’s degree in Library science/ Information science/Documentation with at least 55% of the marks or its equivalent grade with good academic record. The candidate should have cleared the National Eligibility Test (NET) conducted for the purpose by the UGC or similar test accredited by the UGC. Minimum experience of 5 years in a computerized library environment in a reputed institution. Age below 35 years.
Desirable: Working knowledge of HTML, any one Operating System, RDBMS and Digital Library Software.
Applications, on plain paper, with copies of all relevant certificates and names of two referees, should reach the Registrar, Indira Gandhi Institute of Development Research, Gen. A. K. Vaidya Marg, Goregaon (E), Mumbai-400 065, on or before May 31, 2006. Those who are employed in Govt. organizations/Autonomous bodies should route their applications through proper channel.
Economic and Political Weekly May 13, 2006
form of syndicated bank loans to the government. It was much later that other forms of capital inflows, e g, investment in equity, bank lending to the private sector, occurred. (This is the first peak in private flows mentioned earlier in the introduction.) The surge of capital inflows to the emerging economies which began in 1989 was supposed to be different from the one in the late 1970s in that in the previous episode almost all the lending was syndicated bank loans, whereas the later inflows were more diversified. One feature of a financial market that is not fully mature is that banking is the main form of financial intermediation. And since banks engage in maturity transformation, they are fragile (i e, prone to crises). This is true of the rich countries as well. But they are able to cope with idiosyncratic shocks better since their production structure tends to be more diversified and banks constitute a smaller proportion of the financial sector.
Given their underdeveloped financial markets, developing economies typically have (not unrealistically) a fear of outflow of capital if flows are liberalised. The two major types of outflows are (i) capital flight: these may take place due to arbitrage opportunities presented by different tax rates for domestic residents and foreigners and in response to economic and political uncertainty; and (ii) outflows for portfolio diversification. Both kinds of outflows take place even in the presence of controls through a porous current account, through underinvoicing of exports and overinvoicing of imports – for example, the errors and omissions in China’s balance of payments statistics grew by $ 19.3 billion between 2002 and 2004 in anticipation of a revaluation of the RMB.4
Inflows and Outflows and Macroeconomic Performance
Capital account openness is supposed to provide a country with access to the international capital market and provide it with insurance. By way of an appetiser, I note that Caballero (2000) finds that for Chile – which is regarded as a resounding success story in terms of financial liberalisation – all the major macroeconomic series move, more or less, in tandem with the price of copper, which is Chile’s main export.
Capital inflows of all kinds, following liberalisation, generate an appreciation of the real exchange rate. These squeeze the domestic tradeables sector, which would put the whole liberalisation process at risk. Also, if the financial system is not developed this gives rise to a lending boom (through, e g, implicit guarantees to foreign capital).5This over time leads to losses on loans. If a stock market boom, as in India in recent times, led to investment and real growth, it would be welcome. What happens, more often than not, is that a consumption boom occurs with the party ending with the inevitable outflow. So, some form of capital controls helps provide time for the domestic financial sector to prepare for the challenge of opening up.
Outflows leave individuals and banks with foreign currency denominated liabilities – hedging is not well developed and is simply unable to cope with a systemwide crisis. In the wake of the outflow, interest rates rise and the currency depreciates. These put further strain on the financial sector. The positive aspects of a currency depreciation in raising exports needs, as mentioned earlier, trade credits.
Among the capital account items, FDI is more stable and it provides new technology and creates jobs – not all of FDI does this, but the bulk of these flows do. It is portfolio and bank lending that constitute a problem. FDI and portfolio investment flows are not correlated. Among the developing counties, China is the singlelargest recipient of FDI – its capital account was quite closed till recently– while Brazil was the largest recipient of portfolio investment.6
In east Asia, bank lending, which constituted an inflow of $ 114 billion in 1996 turned into an outflow of $ 54 billion in 1998, the figure for 2000 being an outflow of $ 13 billion. Portfolio investment declined from $ 17 billion to $ 5 billion. Similarly, in Latin America, between 1996 and 1999, bank loans fell from $ 49 billion to -$7 billion. For all emerging markets between 1996 and 2000, bank loans fell from being about 56 per cent of all inflows to about a fifth of all inflows.
Countries which have restricted the flow of capital have used two broad types of capital control mechanisms: (a) one which distinguishes between transactions
– implemented, e g, in Chile and Argentina (where there are some items that remain closed to all market participants) and
(b) one which differentiates between domestic residents and others, with domestic residents being forbidden from participating in certain transactions – almost inevitably this translates into liberalising inflows first, and then outflows.
This is the strategy followed in India and South Africa.
When Is Full ConvertibilityPossible?
There are three prerequisites for liberalising the capital account highlighted in the literature:(a) financial sector reforms,
(b) fiscal balance; and (c) properly designed monetary (and exchange rate) policy. But as pointed out by Griffith-Jones, Williamson and Gottschalk (2005), it is the absence of these factors but with the presence of capital flows that an external crisis occurs. For instance, India had a weak banking system and so did Thailand. But India was not affected by the Asian crisis.
Financial Sector Reforms
Financial sector reforms can be painful even in the absence of capital flows. To absorb capital inflows, most of which tend to be short-term, what is needed is a strong domestic financial system resilient enough to cope with inflows and outflows. A movement away from quantitative restrictions to more market determined interest rates, prudential norms, development of a money market, markets for government securities, and foreign exchange, the existence of a yield curve for pricing floating rate instruments, etc, are required to be in place. The Basel prudential norms, which have been criticised (correctly) for other reasons, have helped set some standards in the banking sector. In addition to currency mismatch there could also be maturity mismatch. It is only after financial liberalisation that the banking sector’s nonperforming assets (NPAs) have been recognised as a problem – in the past they were just swept under the carpet. The definition, provisioning, etc, are norms that are still evolving in developing economies.
As mentioned above, bank credit tends to grow quickly following an inflow of capital. This leads to funding of high-risk projects and because of the inadequate monitoring and information the banks’ balance sheets become strained. Excessive risk taking is more likely in a regime where there are implicit or explicit government guarantees. In these situations it is important that banks should not have access to foreign borrowing, because excessive lending takes place without adequate monitoring or hedging of foreign currency. Corrective action takes place only after a crisis
Economic and Political Weekly May 13, 2006 erupts, as in Thailand. The Thai banking crisis, which heralded the Asian crisis in 1997, is widely believed to be due to poor monitoring by the Thai central bank. About half of the (unhedged) foreign exchange loans were made to firms located in the non-traded goods sector.
Monetary (and Exchange Rate) Policy
Fiscal and monetary policies have to ensure that volatility of capital flows is minimised. Low inflation, budget balance and an independent central bank are prerequisites for such a policy. Foreign capital inflows may cause additional headaches for the monetary authorities. First, they may want to sterilise the effects of the flows on money supply to prevent changes in the money stock, which are pro-cyclical. This may be done through the usual channels of monetary policy, though for an economy, which is financially integrated with the rest of the world, the cash reserve ratio (CRR) cannot be too much out of line with the levels abroad. Second, the authorities may conduct an open market operation. This requires some depth in financial markets, which may be missing. Sterilisation through this route may have budgetary implications because now the government, by engaging in it, places in the hands of the public an interest-bearing liability in place of cash. A third alternative is to target a segment of the flows directly. Chile (also Colombia and Thailand) has had some success in transforming the maturity of capital flows through an ‘encaje’ or unremunerated reserve requirement (URR) but whether there was any effect of this on total flows is doubtful.7
Exchange rate flexibility is also necessary. Much has been written on the impossibility of maintaining free mobility of capital, fixed exchange rates and an independent monetary policy – the impossible trinity. A regime of flexible exchange rates has its drawbacks but is clearly preferable to a fixed exchange rate regime in a situation where capital account convertibility is being contemplated. Fixed exchange rates do not allow for inflation differentials, convert returns into foreign currencies one-for-one and are prone to one-way bets against the central bank. In the recent past most developing countries have switched to a floating exchange rate system. Malaysia is still on a fixed exchange rate regime – it had also imposed temporary capital controls – while Argentina was on a currency board till recently.
There is an even bigger argument against a fixed exchange rate regime in developing economies, especially with implicit guarantees: it causes inflows with the borrowing denominated in the foreign currency, which may then be lent to the non-traded goods sector. If it looks as if an exchange rate crisis could occur, market participants move out of the domestic currency (a oneway bet causes a run on the foreign exchange reserves of the central bank), which may cause the fixed exchange rate regime to collapse. The collapse makes the domestic banks vulnerable to a crisis because their balance sheets deteriorate in terms of the foreign currency (domestic currency loans especially those to the non-traded goods sector fall in value). This is what happened in Thailand and Indonesia. The central bank may try to protect the peg by raising interest rates. This may cause fragile businesses to go under and thus worsen the banks’ balance sheets further (see Mishkin (1999) for a discussion).8
One justification, often overlooked, for some form of a pegged rate is that capital flows are prone to reversals – there was the debate mentioned above about whether capital flows are caused by pull factors (ie, country-specific performance) or push factors (OECD business cycle, interest rates, etc). If indeed a reversal of a capital flow is likely, then with some nominal wage-price inertia it may be optimal not to have a floating exchange rate regime. A floating exchange rate regime would cause an immediate real appreciation and a current account deficit, the output and employment costs of these could be substantial. With a pegged exchange rate system, on the other hand, the overvaluation occurs gradually over time. It is very difficult for the authorities, given the few episodes of international capital flows, to decide whether the shock (i e, the inflow of capital) is permanent or temporary and therefore what the appropriate policy response should be.
A hard-peg (e g, a currency board), which for many analysts was the favourite form of a nominal anchor, has become less fashionable. It involved surrendering discretionary monetary policy altogether. Its fall from grace is due to Argentina tying its fortunes to the US dollar. Having tied its own hand it suffered the consequences of a strong dollar and an inflexible labour market.9 To see why just compare the recent experience of Brazil, which has a floating rate and Argentina, which had a currency board (this is equally true of Hong Kong versus the rest of east Asia). Some form of nominal exchange rate fixing
– with the currency board as an extreme form – pays rich dividends in bringing down inflation in countries prone to high (or hyper) inflation. It is a good idea, however, to move to a floating regime once this has been achieved. But what is the correct time to dismount the tiger of “credibility”?
Fiscal Policy
The government’s fiscal stance is also important for a successful liberalisation of the capital account. If there is a large budget deficit (or more correctly debt) then the real interest rate will tend to be high, attracting inflows. This will cause a real appreciation, which with a floating rate could be achieved through a nominal appreciation. In the presence of fixed exchange rates this inflow could cause an increase in the nominal money supply and generate inflationary pressures. To sterilise the money supply the central bank could conduct an open market operation, which, as noted above, requires some depth in these markets. But even a sterilisation operation is only a temporary fix because, as mentioned in the previous sub-section, this increases the government debt held by residents and hence adds to the interest cost.10 In any case the tradeables sector, which was being squeezed before capital mobility was allowed, is hurt even more. This could make the current account deficit unsustainable. In anticipation of such an eventuality, a reversal of capital flows could take place. Brazil was able to attract a lot of foreign portfolio capital in the early 1990s. But the government’s (both federal and provincial) deficits were sizeable.
It should be noted that prudent fiscal behaviour by itself is not enough to keep as crisis at bay. After all, most of the southeast Asian countries had “reasonable” deficits and contagion tends not to discriminate between those who are well behaved and those who are not. But the experience of Argentina, Chile and Uganda shows that with the government deficit under control the financial system can handle inflows better, and as mentioned above, in Brazil fiscal profligacy did lead to a crisis.
In conclusion, one can say that a strong financial sector is required if a nation is to reap the potential gains from trade in assets. Even to benefit from trade in goods, some sectors are expected to contract, while others expand. In the financial markets, the
Economic and Political Weekly May 13, 2006
collapse of a few institutions could lead to a collapse of the entire financial system. The experience of emerging economies suggests that one should approach the issue of throwing open the capital account with extreme caution.
m Email: partha@econdse.org
Notes
1 China, like India, has opened up its capital account gradually over the last two decades. Initially, to prevent capital flight, inflows were first liberalised. Among the inflows, the more illiquid long-term flows were liberalised earlier. Anticipating an appreciation of the currency, the RMB, the capital account surplus grew by $ 78 billion between 2002 and 2004. Of this growth, portfolio investment accounts for $ 30 billion and “other investments”– these are mainly loans raised abroad by China-based firms – $ 42 billion.
2 See Griffith-Jones, Williamson and Gottschalk (2005) for a discussion of how this may require government intervention and the development of a secondary market.
3 Note the period precedes the opening up of capital accounts worldwide and thus possibly understates the possible problems that one may expect with capital mobility to emerging markets.
4 Up to 2001 this component was negative, indicating capital flight from China.
5 An Indian example was how recently the RBI was stopped in its tracks when it tried to discipline some brokers.
6 This may have been due to the high real interest rates in Brazil due to the government’s budget problems. It is not surprising that Brazil attracted very little FDI in this period.
7 Encaje is the Spanish word for URR. Edwards (1999) casts doubt on whether even the composition of the capital flows was altered. Griffith-Jones, Williamson and Gottschalk (2005) view them as being successful.
8 Empirically, Kaminsky and Reinhart (1999) find that the banking crisis occurs before the exchange rate crisis.
9 Frankel (1999) points out that the dollar for Argentina does not constitute an optimal currency area in terms of the usual criteria.
10 This imposes a quasi-fiscal cost.
References
Broner, F A and R Rigobon (2004): Why Are Capital Flows So Much More Volatile in Emerging Markets than in Developed Countries? University of Pompeu Fabra.
Caballero, R (2000): ‘Macroeconomic Volatility in Latin America: A View and Three Case Studies’, National Bureau of Economic Research Working Paper No 77782.
Edwards, S (1999): ‘How Effective Are Capital Controls?’, National Bureau of Economic Research Working Paper No 7413.
Frankel, J A (1999): ‘No Single Currency Regime Is Right for All Countries or At All Times’, National Bureau of Economic Research Working Paper No 7338.
Griffith-Jones, S , J Williamson and R Gottschalk (2005): Should Capital Controls Have a Placein the Future International Monetary System? Institute of Development Studies.
Kaminsky, G, S Lizondo and C Reinhart (1998):‘Leading Indicators of Currency Crises’, International Monetary Fund Staff Papers, 45, 1-48.
Kaminsky, G and C Reinhart (1999): ‘The Twin Crises: The Causes of Banking and Balance of Payments Problems’, American Economic Review, 89, 473-507.
Mishkin, F S (1999): ‘Lessons from the Asian Crisis’, National Bureau of Economic Research Working Paper No 7102.
Economic and Political Weekly May 13, 2006
