Latin American countries (LACs) have been strongly affected by the changes that have occurred in capital flows over the last quarter of a century. During the 1970s, a large supply of funds was made available to the region; then, during the 1980s, there was a generalized severe shortage of financing, and the region became a net exporter of funds. Between 1991 and 1994, it received a large capital surge again, only to experience another sharp reduction focused in Mexico and Argentina, and a generalized drop of portfolio flows in late 1994 and early 1995. The so-called tequila crisis was followed by a renewed access in 1996-97. In 1998-99, Latin America has been experiencing a new shortage of external financing, aggravated by a worsening of the terms of trade. The crisis centered in Asian countries has now given rise to of a new recessive macroeconomic adjustment in the region. It appears impossible to hide the significant instability of financial markets.

On all those occasions, the changes that were first expansive and then contractionary, began in the international markets and had a strong impact on national economies. Up to 1996, the successful emerging economies of Asia appeared to be immune to the instability associated with capital surges. The recent events have shown that is no longer so. Some of the causes are common with those of Latin America. The new crisis provides a renewed opportunity to significantly improve the architecture of the international financial system (ECLAC, 1999), and to reform the domestic approach so as to achieve sustainable macroeconomic equilibria and to enhance its contribution to growth.

Net capital inflows to Latin America averaged nearly 5% of GDP in 1991-94 and 1996-97. During both periods the current account deficit rose sharply and exchange rates increased (see ECLAC, 1995); naturally, imports grew more rapidly than exports, and external liabilities rose steadily. Indeed, soon all these variables reflected a growing macroeconomic imbalance.¹ In many cases they were anchored to one dominant "balance", that of a falling CPI tied to an increase in real exchange rates and climbing external deficits. Those recipient countries which had larger current account deficits heavily financed with short-term liabilities and exhibiting stronger exchange rates, tended to become increasingly more vulnerable to external creditors. Creditors, given the high exposure of financial assets placed in the region, subsequently became more sensitive to any "bad news".

The dramatic increase in the flow of international financial resources in recent years has been more diversified during the current decade than it was during the 1970s. But the situation is potentially more unstable, inasmuch as the trend has been to move from medium-term bank credits to investments in liquid stocks, bonds and deposits. A very high percentage of this supply of financing is of a short-term and liquid nature. Paradoxically, there has been a diversification toward volatility in the 1990s; the relative improvement after the "tequila crisis", with a rising share of Foreign Direct Investment (FDI), still included a significant proportion of volatile flows. Latin America was a receptive destination, and offered the expectation of high rates of return.

Why the high rates of return? Naturally the rate should tend to be higher in the "productive" sectors of capital-scarce regions, like Latin America. Several factors contributed to this situation. Initially, prices of equity stocks and real estate were highly depressed. That allowed a 300% average rate of return (in current US dollars) in the stock markets of Latin America between late 1990 and September 1994. After a sharp drop around the tequila crisis which affected all Latin American stock markets, between March 1995 and June 1997 average prices nearly duplicated, due directly to portfolio inflows. Domestic interest rates were high as well, reflecting the external restrictions prevailing in 1990 and the tight monetary policy in place (plus the nature of financial reforms implemented; see Ffrench-Davis, 1999). Finally, in a non-exhaustive list, the renewed supply of external financing generated a gradual exchange-rate appreciation that also encouraged short-term or liquid inflows (see tables in Ffrench-Davis, 1999, ch. 10).

The yearly current account deficits rose persistently. By 1994 Mexico and the region had accumulated 4 years of large flows of external liabilities. While in 1991 the actual stock of assets of the new investors in Latin America was evidently below the "desired" stock, in 1994 it was notably large. We had entered a vulnerability zone, with the economy growingly sensitive to bad political or economic news. The longer and deeper the permanence in that zone, the harder to get away without a crisis.

During the 1990s, capital inflows contributed to a recovery of economic activity, after the recession that had prevailed throught 1990 in most LACs. Argentina and Perú were two outstanding cases of underutilization of capacity; Chile and Mexico were the exceptions in the other corner.

Annual GDP growth rose from 1.2% in the 1980s, to 4.1% between 1991 and 1994 and to 2.9% in 1995-98 (table 1). This growth was meager, however. On the one hand, the comparison with the previous golden age is shocking. In 1950-80, Latin America had averaged a GDP growth of 5.5% per annum; and domestic investment had risen fast. Subsequently, in the 1980s the investment ratio fell sharply by 7 points of GDP, with only a mild recovery in the 1990s. In fact, investment grew much less during this decade than did capital inflows. Thus, a significant part of the external flows financed increased consumption and consequently crowded-out domestic savings (Ffrench-Davis and Reisen, 1998).

Actually, the new supply of financial flows initially had a positive effect on Latin America. Thanks to a larger utilization of installed capacity, GDP increased beyond the expansion of production: about one third of the 4% annual GDP growth rate in 1991-94 reflected an increased use of installed capacity. The phenomenon was particularly intense in countries such as Argentina and Peru. Subsequently, the strong recovery of Argentina and Mexico in 1996-97 rested to a significant degree on the underutilization created in 1995. And, again, recovery was short-lived, giving way to slower growth since mid-1998, and open recession in Argentina in 1999.

The renewed access to external capital challenged the stability and sustainability of macroeconomic equilibria. Indeed, the affluence of capital had an adverse effect on the evolution of exchange rates,² the money supply and domestic credit, the accumulation of external liabilities (mostly with short-term maturities), and thus made the economy more vulnerable to future negative external shocks.

What happens? Why debtors don’t stop capital inflows before a crisis?

In the majority of cases, authorities allowed the capital surges to move into their domestic economies. Most of them thought they should do nothing or could do nothing, or preferred to "benefit a little longer" from the anchoring of the domestic CPI to international inflation.

In general, observers tended to agree that disequilibria had been created. For voluntary flows to take place both debtors and creditors must agree on it. Then, why did creditors not act in due time to curb flows? The specific nature of the main agents on the creditor side is crucial. The conflictive segment of flows, that was predominant in both surges and busts, naturally works over the short term.

When creditors discover an emerging market, they start out with non-existent exposure. Then they generate a series of consecutive flows which accumulate in rapidly increasing stocks. As said, the creditors' sensitivity with regard to bad news increases remarkably with the level of stocks placed in a country (or region), and with the degree of debtors' dependence on additional flows (current account deficit plus refinancing of maturing liabilities). Additionally, the most relevant feature is that after a significant increase in asset prices and exchange-rates, accompanied with rising stocks of external liabilities, the probability of reversal of expectations grows steeply.

The accumulation of stocks and a subsequent reversal of flows can be considered to be a "rational" behavior, given the nature of predominant agents. Investors with short horizons are not concerned whether the foundations (over a long term) are weakened by capital surges while they continue to bring inflows. What is relevant for them is that the crucial indicators –real estate prices, stocks, exchange rates– can continue providing profits in the near term; thus, they will continue pouring money until expectations of near reversal start to built.

The most relevant issue is that myopic agents specialized in microfinance have begun to determine the macroeconomic situation and to design policies. The outcome, unsurprisingly, turns out to be unsustainable macroeconomic imbalances, "wrong" prices, and an environment not conducive to productive investment. Then what appears to be "irrational" is that the domestic macroeconomy has become increasingly influenced by experts in microfinance. Macroeconomic authorities must face their responsibility and insure that macroeconomic balances both sustainable and suitable for growth are achieved. That requires that they avoid entering vulnerability zones during economic booms-cum-capital surges.

It is important to design economic policies that will attract resources and also ensure that they flow in a fashion and quantities that are sustainable and directed more towards long-term investment than to consumption.

The Mexican crisis which exploded in 1994 is a good example of the harm that can be caused when a country absorbs an excessive volume of capital flows, giving way to a large stock of external liabilities, especially when the composition of such financing makes it volatile. Producers and consumers adjusted to a level of overall expenditures that was much higher than the potential GDP, and after a while the amounts involved became unsustainable. Recessive adjustment inevitably followed. In 1995 Mexico experienced a 6.6% drop in GDP and a nearly 30% drop in capital formation. These were closely associated, first, to a persistent revaluation of the exchange rate and a growing current account deficit and, later, to a sharp cutback in financing on the part of creditors. The country was forced into a highly recessive adjustment and a huge devaluation, despite the large package of international support it received in 1995 (Lustig, 1997; Ros, 1999).

As said, the seeds of the crisis date back to the period 1992-94, when there was a massive capital inflow, mostly short term. Aggregate demand grew rapidly, leaving far behind the potential GDP; it leaned increasingly towards tradable goods, especially encouraged by the strengthening of the exchange rate. Thus, in those years, there was a maladjustment that would most likely have to be reversed in the future. What is extremely important is that such disequilibrium was encouraged by capital inflows. Since the public sector was balanced, the disequilibrium occurred in the private sector.

In 1995, the Argentine economy was seriously affected by the contagion. Although this did not lead to a currency crisis with a sharp exchange rate devaluation, as many had feared in 1995, Argentinean GDP fell by 5% and investment diminished by 16%. The overall growth rate of Latin America went down sharply, to a figure below the population increase, while the regional investment ratio also fell substantially. During 1995, in diverse countries, negative flows were observed in several segments of the supply of funds (especially bonds, deposits and stock markets). Average growth in Latin America was negative in the four quarters included between March 1995 and 1996. Subsequently, the flow of funds was reactivated once again, exceeding US$80 billion in 1997. The resulting economic reactivation was particularly significant that year. However, some of the same problems apparent in the 1991-94 recovery reappeared in 1996-97, and actually claimed their due in 1998-99. Nevertheless, the consequences were moderated by the following variables. First, the new financial boom period was short-lived; it lasted only a couple of years. Second, there was a tendency to increase the share of FDI flows, which are most steadfast than flows to the stock market or short-term credits. Third, owing to a sharp reduction in 1995 of the current account deficits (and a significant exchange rate depreciation, particularly in the case of Mexico), a high deficit was recorded only in one year (1997). Fourth, taking into consideration the banking crisis of Mexico and Argentina, following the "tequilazo", these and other countries introduced reforms to their financial systems which strengthened the regulation and supervision of their banking systems.

Consequently, the 1998 adjustment took place in economies with a more moderate stock of volatile liabilities than in 1995, with healthier banking portfolios, and with less overheated economies. However, in all, the Asian crisis caught Latin America after most of the region's countries had strengthened their exchange-rates and their current account deficit had risen sharply. The same old story was being repeated, at a time when the international scenario was hit by a recession, worse terms of trade, declining trade flows and access to finance, and rising spreads.

Chile's performance was opposite that of Mexico in 1995-96, in spite of numerous similarities during the years prior to 1994. Both the Mexican crisis and Chile’s strength were built up over time. Towards the end of the 1980s, both countries had already opened up their trade considerably, their budgets had improved substantially, privatization was well under way, annual inflation was around 20%, and the two countries had similar domestic savings rates. The reason why Chile performed better in 1995 is that, faced with an abundance of external funds in 1990-94, it deliberately followed a cautious policy (Agosin and Ffrench-Davis, 1999; Stiglitz, 1998). Instead of taking and spending all the large supply of external resources available, which would have led to a significant appreciation of the peso and to a rising current account deficit, it chose to discourage short-term capital inflows. In 1991 a tax was imposed and substantial non-interest-bearing reserves for external credit were required; the reserves requirement was subsequently extended to deposits in foreign currencies and investment in second hand stocks, while primary issues and venture FDI capital were exempted;³ FDI had to be held in Chile for one year at least; the financial system was subject to relatively strict regulations, including a selective supervision of assets and required provisioning, as well as restrictions and drastic penalties on operations with related parties. The set of measures adopted effectively discouraged speculative capital inflows (Agosin, 1998; Agosin and Ffrench-Davis, 1999).

As a consequence, by late 1994, Chile had a moderate external deficit, high international reserves, a modest and manageable short-term debt, a domestic savings rate that was rising instead of falling (the latter being the case in Mexico and Argentina), a level of domestic investment that since 1993 has been the highest recorded in history and the exchange rate in 1990-94 was comparatively closer to equilibrium than that of most of the countries of the continent, as reflected by a moderate current account deficit.

The measures adopted were effective in achieving their targets for most of the 1990s. However, in 1996-97 this policy mix and the intensity with which it was applied remained unchanged, in spite of a new vigorous surge in capital flows to most countries in the region, but particularly to Chile, a country immune to the tequila contagion. This surge should have been met with increased restrictions on rising inflows. Since these are a market-based mechanism that alters the relative prices of capital flows, what happened was that capital inflows ended up paying the cost of the reserve requirements, with no evidence of significant evasion.

The fact is that, in general, the policies were rather unchanged in face of a stronger capital surge during 1996-97. As a consequence, despite heavy intervention in the foreign exchange markets, the Central Bank was unable to prevent a sharp rise in the exchange-rate and the current account deficit. Nonetheless, the benefits of the active regulations implemented in previous years had left large international reserves, a low stock of foreign liabilities and a small share of volatile flows. Unfortunately, those strengths were partially undermined by the excessive rise in the exchange-rate and the high external deficit recorded in 1997.

Optimism regarding Latin America returned to the international financial markets in 1996-97. The current net capital inflow climbed to pre-crisis levels. Composition improved, with a larger share of FDI. GDP decline in various LACs was reversed. In fact, a dynamic growth for the region as a whole was observed since mid-1996 until mid 1998.

GDP recovery in Argentina and Mexico was particularly vigorous, although after the sharp decline in both countries due to the Tequila Effect in 1995, there was a large gap between effective GDP and productive capacity. This allowed for a significant reactivation which led to a complacent view of the effects of crises and of countries' capacity to recover. Nevertheless, in both countries only by 1997 did per capita GDP reach the levels achieved in 1994, while average wages were still lower in 1998, with Mexico 22% lower than in 1994. Costs for the real economy and equity are large and long-lasting.

It should be noted that since the GDP increase comprised a larger recovery share, effective GDP was once again close to the production level. However, such level moved upward slowly, because productive investment was still low, while real exchange rates were once again on the rise. Consequently, as long as productive investment did not increase substantially, that rate of growth was not sustainable. In effect, at the beginning of 1998 it was estimated that the 5.3% growth of 1997 would be moderate, around 4%. With the intensification of the Asian crisis and its contagious effects on finance and trade, effective growth contracted to 2.3% in 1998 and is expected to close near below zero in 1999. Thus, the region is experiencing a new significant adjustment. The following lessons can be derived from recent experiences.

It is important to ensure that the volume of inflows is consistent with the absorptive capacity of the host country. Absorption must, of course, refer to use in productive investment. The composition of flows has relevant incidence in three dimensions. First, FDI (excluded purchases of existing assets) goes directly to capital formation, as well as long term loans to importers of capital goods. Second, volatile flows tend to impact more directly the foreign exchange and stock markets. Third, the capital surges or deviations from the trend tend to leak toward consumption, because of the faster capacity of consumers to respond as compared to irreversible productive investment. There is growing evidence that the greater the instability of flows (or deviation from the trend), the lesser the share directed to productive investment.

Opening up the capital account indiscriminately can be very detrimental to productive development and to the welfare of the majority of people and firms, inasmuch as externalities and other imperfections of international capital markets give rise to frequent cycles of abundance and shortage of external financing (Rodrik, 1998; Wyplosz, 1998). Effective, efficient regulation is possible. Chile proved this from 1991 onwards.

The recent experience of emerging economies has shown dramatically that allowing the market, dominated by short-sighted agents, to determine the volume and composition of capital flows can have a very high cost for the recipient country. This is why the use of regulations on capital inflows should not be neglected a priori. On the contrary, the microeconomic costs associated with the use of such instruments should be balanced against the social benefits in terms of macroeconomic stability, investment and growth (Williamson, 1993; Zahler, 1998).

Governments must ensure that capital flows do not generate atypical (outlier) prices or significant distortions of basic macroeconomic indicators, such as interest and exchange rates, aggregate demand, the composition of expenditures in terms of consumption and investment, and the production of tradables.

Capital surges should not be used for achieving an extreme objective related to a single domestic economic variable, such as to anchor inflation, by appreciating the real exchange rate. This tends to throw other major variables off balance. It is risky to remain bound to a fixed nominal rate, and worst to fully dollarize except if the economy is in an optimum currency area with the US. Is there any country in the region that fits this description? The methods for regulating the exchange rate can be extremely diverse, several of them involve some form of a crawling-band, with some type of intra-marginal intervention.

It is generally agreed that across-the–board opening-up of the capital account has been premature and should have been postponed, moving only in a selective way, until a long term process in which other major reforms had been consolidated and new equilibrium prices had been established. The lesson to be learned from this experience is that during structural adjustment, with open capital accounts (especially when international financing is abundant) capital flows can increase too fast and have destabilizing macroeconomic and sectoral effects (Williamson, 1993; Wyplosz, 1998).

In the first place, in Latin America many countries adopted deep trade reforms in the 1990s together with increases in the exchange-rate. Second, if productive investment capacity reacts slowly and/or with a lag and domestic financial markets remain incomplete and poorly supervised, additional external resources cannot be absorbed efficiently in the domestic economy, and thus they threaten the future stability of the flows themselves. In the third place, fiscal parameters need to be consolidated, since in the absence of a sound tax base and flexible fiscal mechanisms the authorities will have to depend excessively on monetary policy to regulate aggregate demand. Finally, since part of the aggregate demand generated by capital flows is inevitably spent on non-tradable goods, when actual demand comes close to the production frontier, the relative price of non-tradables tends to rise. This in turn is reflected in a higher current-account deficit. A real revaluation of the currency can obviously distort the allocation of resources and investment, seriously weakening the structural mid-term objective of penetrating external markets with new exports (ECLAC, 1995; Ffrench-Davis, 1999, ch. III).

It is not wise to make an inflexible commitment to indiscriminately keeping the capital account open, particularly in light of the crucial importance of macroeconomic stability, along with the disproportionate volume of the international capital markets compared with the small size of LACs markets, and the serious shortcomings of both markets. As long as market movements depend to a significant extent on short-term transactions and domestic securities markets remain shallow, there will be a risk of great instability in this new modality of linkages with the international economy. In fact, Mexico’s, Korea’s and Thailand’s recent critical experiences attest to the wisdom of discouraging both large financial inflows and the accumulation of short-term external liabilities.

Understanding better the working of domestic and international financial markets is at the core of the future of the world economy. More pragmatism and systematic efforts should be undertaken.

References

• Agosin, M. (1998), "Capital Inflows and Investment Performance: Chile in the 1990s", in Ffrench-Davis and Reisen (1998).
• Agosin, M. and R. Ffrench-Davis (1999), "Managing Capital Inflows in Chile", S. Griffith-Jones, M.F. Montes and A. Nasution (eds.), Short-Term Capital Flows and Economic Crises, Oxford University Press and United Nations University/ WIDER, forthcoming.
• ECLAC (1995), Policies to Improve Linkages with the Global Economy, United Nations, Santiago. Second revised Spanish edition published by Fondo de Cultura Económica, Santiago, 1998.
• ECLAC (1999), "Towards a New International Financial Architecture", United Nations, Santiago, January.
• Ffrench-Davis, R. (1999), Reforming the Reforms in Latin America: Macroeconomics, Trade, Finance, Macmillan, London, forthcoming.
• Ffrench-Davis, R. and H. Reisen (1998) (eds.), Capital Flows and Investment Performance: Lessons from Latin America, OECD Development Centre/ECLAC, Paris.
• Lustig, N. (1997), "The United States to the Rescue: Financial Assistance to Mexico in 1982 and 1995", CEPAL Review, N° 61, Santiago, April.
• Rodrik, D. (1998), "Who Needs Capital Account Convertibility?", in P. Kenen (ed.), Should the IMF Pursue Capital Account Convertibility?, Princeton Essays in International Finance, Nº 207.
• Ros, J. (1999), "From the capital surge to the financial crisis and beyond: Mexico in the 1990s", ECLAC project on Preventing financial crises: lessons from successful emerging economies, draft, June.
• Stiglitz, J. (1998), "The Role of the Financial System in Development", Presentation at the Fourth Annual Bank Conference on Development in Latin America and the Caribbean (LAC ABCDE), San Salvador, El Salvador, June 29, 1998.
• Williamson, J. (1993), "A Cost-Benefit Analysis of Capital Account Liberalization", in H. Reisen and B. Fischer (eds.), Financial Opening, OECD, Paris.
• Wyplosz, Ch. (1998), "International Financial Instability", Graduate Institute of International Studies, Geneva and CEPR, London, July.
• Zahler, R. (1998), "El Banco Central y la Política Macroeconómica de Chile en los Años Noventa", Revista de la CEPAL N° 64, April.

2. It should be recalled that several LACs were implementing sharp liberalization of import regimes pari passu with exchange-rate appreciation. See Ffrench-Davis (1999, ch. III) and ECLAC (1995, ch. V).

http://www.sela.org
sela@sela.org
   SELA,  Permanent Secretariat
Torre Europa, Fourth floor, Urb. Campo Alegre, Av Francisco de Miranda,
Caracas 1060- Venezuela
Tlf: (58) (212) 955.71.11 Fax: (58) (212) 951.52.92