Frenkel, Roberto. MACROECONOMIC SUSTAINABILITY AND DEVELOPMENT PROSPECTS:
THE LATIN AMERICAN PERFORMANCE IN THE NINETIES. CEDES, Buenos Aires, Argentina. 1995. p. 92. (Economía: Nº 111) Disponible en la World Wide Web: http://bibliotecavirtual.clacso.org.ar/ar/libros/argentina/cedes/frenkel3.rtf
RED DE BIBLIOTECAS VIRTUALES DE CIENCIAS SOCIALES DE AMERICA LATINA Y EL CARIBE, DE LA RED DE CENTROS MIEMBROS DE CLACSO
MACROECONOMIC SUSTAINABILITY AND DEVELOPMENT PROSPECTS:
THE LATIN AMERICAN PERFORMANCE IN THE NINETIES
El Area de Economía del CEDES agradece el apoyo a la investigación de UNCTAD, del IDRC (Canadá), la SAREC (Suecia), la Mellon Foundation (Estados Unidos) y la Universidad de La Plata (Argentina).
Buenos Aires, 1995
Macroeconomic Sustainability and Development Prospects:
the Latin American Performance in the Nineties1
1. The nineties Latin American performance and the Mexican crisis. Since the beginning of the nineties Latin America has shown indications of recovery from the generalized crisis of the previous years. There has been a marked decline in the rate of inflation and a recovery of positive rates of growth. Until mid-1994, the megainflationary process in Brazil placed this country outside the above-mentioned trends and so, given the significant relative weight of the country in the region, made the general characterization dubious. But the "Plano Real", the stabilization program launched in July 1994 and still successful in both the inflation and activity level fronts, has brought Brazil's macroeconomic performance up to the average of the rest of the main countries.
Paradoxically, a few months after Brazil began to perform like its well-behaved neighbors, some of these showed renewed symptoms of instability and the need for another round of external adjustment. The Mexican crisis marks the clear end of the early nineties, a period characterized by the above-mentioned trends. This seems to be so not only because of the significant regional weight of Mexico but also because other factors enhance its importance beyond its demographic and economic relative weight.
Despite the fact that both Brazil and Mexico are the largest countries in the region, there is a substantial difference between Brazilian instability, recently overcome, and Mexican instability, recently reopened. While the Brazilian instability was seen as an enduring symptom of the eighties, awaiting a definitive solution, the Mexican crisis represents the emergence of instability and uncertainty in a country that was supposed to have left these symptoms definitively behind. The Mexican crisis symbolizes, in this sense, a turning point for Latin America as a whole because Mexico actually led the regional process of stabilization and structural reform and was held up as an example for other countries to follow. The country was not only the benchmark with regard to fiscal discipline, privatization and deregulation, but its experience was also supposed to lead - through the NAFTA agreement - a regional path of beneficial growing trade integration with the USA.
Other factors also contribute to spreading the meaning and effects of the Mexican events. Mexico's performance plays a leading role in the formation of the international investors' expectations about Latin America as a whole. Then, independently of the actual similarities between the individual situations in the region, bandwagon expectations exert a generalized effect on capital flows. In countries like Argentina, where macroeconomic configuration resembles in many aspects the Mexican pre-crisis situation, reduced capital inflows have a self-fulfilled impact on the country's performance. But, to a different extent and with different consequences, the "herd behavior" of international capital flows will surely effect every country in the region.
Beyond any speculation about future prospects, the Mexican crisis has already had a significant impact on the region, as can be seen through the immediate reaction of the Latin American international bonds and domestic stock prices. Moreover, the broad character of the acknowledgment of the generalized effects of the Mexican crisis can be stressed by mentioning that they constituted the main foundation of the rescue package promoted by the American government.
For many analysts, both the Mexican crisis and its present and future probable regional consequences were not unexpected. This is not to say that the December 20 devaluation and the following episodes were predicted in detail, but that the increasing dependence on short-run capital inflows to finance rising trade and current-account deficits made the medium-run unsustainability of that process clear. For instance, in the 1993 UNCTAD Trade and Development Report it was asserted that in the cases of Mexico and Argentina "even though autonomous capital flows have so far been more than enough to finance external deficits, these countries need to undertake an external adjustment through higher investment and exports. If the opportunities for an expansionary adjustment are not exploited, deflationary adjustment may eventually become unavoidable." The rationale behind this expectation was the low probability attributed to a permanently increasing amount of capital inflows, in conjunction with the trends exhibited by the external accounts. If, as was expected, net capital inflow falls at some point below the financial needs of the current account, reserves will start to decline. The reserves reduction, through its direct effects on money and credit and by pushing the domestic rate of interest up - automatically or as a policy reaction intended to attract foreign capital - will induce recession and might ultimately lead to a run, forcing a devaluation. Actually, trade-and current-account deficits continued growing in Mexico and Argentina in 1994. In Argentina, the reserves stagnated in the first part of the year and declined afterwards, moderately before the Mexican crisis and more rapidly afterwards. In Mexico, the reserves stood at $29 billion in February 1994, when American monetary policy started to raise the interest rates. From then on, Mexico lost reserves throughout the year and at a faster rate just before and after the devaluation. By December 22, reserves were down to $6 billion and the authorities left the currency to float. In all of 1994 reserve losses amounted to $20 billion, while the current account showed a $30.6 billion deficit. This means that net capital inflows in 1994 were about $10 billion, one-third of the almost $30 billion that Mexico received in 1993.
In 1993 Mexico and Argentina showed the worst regional indicators of external fragility and their external trends pointed to a future turning point, some changes in policy and some kind of reversion of the early nineties performances. But afterwards, in 1994 and before the Mexican devaluation, some new elements emerged in the regional performance that represented, in fact, the first symptoms of a turning point. One of them was the above-mentioned new trend of the international reserves in Mexico and Argentina. The starting point of the process was February 1994, when the American Federal Reserve began to increase the short-term rate of interest.
From February on, against conventional expectations, the prices of long-term bonds fell and long-term interest rates went up together with the short-term rates. Both movements had a more than proportional impact on Latin American bond prices and lending interest rates. Simultaneously with the rise in short and long-term American interest rates, there was a generalized increment in the Latin American country risk spreads charged to the region. This can be seen in Tables 5 and 6. The tables show that in 1994, before the Mexican crisis, the spreads of long-term bonds on US Treasury bills augmented very significantly. These facts should be stressed because they represent the first negative2 international markets herd behavior with regard to the region in the nineties.
Under what circumstances should country risks rise when American interest rates go up? Only if international investors get a perception of the regional external fragility and of the negative impact that higher international rates would have on it. But, by asking for higher compensatory risk spreads, the financial markets' behavior accentuates that negative effect. So, it was not necessary to wait until Mexico devalued to see evidence of self-fulfilling prophecy behavior regarding Latin America. While the Mexican crisis coordinated that behavior at the end of the year, a similar coordination was provided throughout 1994 by the rising American rates of interest. In this sense, the Mexican crisis can be interpreted as a specific episode following a period of financial distress that started at the beginning of the year. As we argued above, the episode itself is showing a significant feed-back effect on the whole region. The sequence is far from new. The analogy with the late seventies-early eighties is apparent, but, more generally, the sequence of a financial boom, followed by a period of distress and a final abrupt contraction, reproduces accurately the phases of a Minskyan financial crisis.
There are also other elements in the 1994 financial markets that complement the above-mentioned evidence. During the year and before the Mexican crisis, individual country risk spreads seem to have risen in correlation with the relative external fragility indicators, less in the cases of Chile and Colombia and more in the cases of Mexico and Argentina. This "revealed attitude" of the financial investors regarding the relative external fragility of individual countries is consistent with the mentioned attitude towards Latin America as a whole, and also with our hypothesis about the main sources of the perceived risk.
It should be stressed that the markets' "revealed perceptions" were in conflict with the judgments made public by some international financial institutions, such as the IMF and the World Bank. According to these official points of view, risks could only emerge from an undisciplined fiscal performance or from threats to the persistence of the structural reforms. Because none of these conditions applied in the cases of Mexico and Argentina, both countries ranked in the lower risk category.
The Mexican crisis and its immediate consequences should have a significant impact on the theoretical views about stabilization and development and on their policy implications. From the official point of view of the IMF and the World Bank, the Mexican crisis should never have happened. Although we can take for granted that some ad hoc explanations will be found in order to save political positions and theoretical principles from the effects of empirical evidence, this time it will be a very difficult task.
As we mentioned above, international financial crises are not new and the essential features of the early nineties cycle resembles that of the late seventies-early eighties. In particular, there is a marked analogy between the recent and present Mexican evolution and the Chilean early eighties case. Nevertheless, after the debt crisis, theoretical principles were saved by ascribing the debt crisis to fiscal deficits and ISI. Brazil, Mexico and Argentina were exemplary cases in this regard. The Chilean case did not fit into that diagnosis well, and that is why its early eighties external and financial crisis was rarely mentioned or analyzed.
Regarding the possibilities of developing an analogous ideological operation, the present situation differs in some important aspects. Firstly, the crisis directly involves the most important, exemplary and leading country regarding orthodoxy and free market structural reform. Secondly, many countries in the region have strictly followed the same policy prescriptions, particularly Argentina, which is the second most affected economy. Thirdly, there has been considerable direct public involvement of the American administration in the evolution of the Mexican economy since well before the crisis, not only through the NAFTA agreement and the inclusion of Mexico in OECD, but also through the swaps agreements signed in 1993 and 1994. Fourthly, although the IMF expressed the same opinion and optimistic judgments about the Chilean economy in 1981 that the institution expressed recently about Mexico, the IMF and the World Bank seem never to have been so committed to an individual country's performance as they are now to Mexico's.
All of these circumstances help to explain the extraordinary characteristics of the rescue package promoted by the American administration and the IMF. But the operation itself represents a severe shock to the prevailing ideas. The theoretical basis of current policy orientations, particularly regarding financial markets deregulation and opening, assumes rational expectations agents behavior and market with self-stabilizing properties. After years of advocating against public intervention and regulation on that basis, the calls for a massive public rescue package seem rather pathetic. So, even if the rescue package serves its intended purpose to avoid the worst immediate consequences of the Mexican crisis and allows the financial markets to stabilize for some time, the lack of consistency of the ideas that founded the model have been irremediably exposed.
2. The macroeconomic sustainability and the long-term development aspects of the early nineties performances. The main purpose of this paper is to evaluate the trends in the nineties as they evolved up until the Mexican crisis, and to discuss its prospects. The discussion of the economic prospects should focus both on the observed trends and on the probable effects of the Mexican crisis, not only on Mexico itself but also on other countries.
It is possible to talk about the regional performance because the countries' individual performances have some common aspects and share some common external influences. Some have been pointed out above. But the countries differ in other aspects that are crucial for our purposes. To pay the necessary attention to this diversity we have chosen five countries we think represent three different stylized patterns: Argentina and Mexico, Chile and Colombia, and Brazil. It should be mentioned that beyond their capacity to illustrate these patterns, these countries also make up a representative sample of the region given their economic and demographic weight.
Our evaluation of these cases intends to focus on two issues. The first is the sustainability of the recent macroeconomic trends from a macroeconomic point of view. The second is the long-term development aspects of the recent growth performance. The first issue relates mainly to questions regarding the fragility of macroeconomic equilibria vis-à-vis plausible changes in external and domestic conditions. The second intends to tackle a more difficult question. To what extent is it possible to assume that the positive rates of growth that most of the countries have been experiencing - some of them only recently, like Argentina, but others, like Chile and Colombia, since the mid-eighties - represent the first stage of a self-sustaining and lasting development process?
Let us try to determine the meaning and limits of these analytical perspectives. In principle, it seems difficult to distinguish clearly one from the other. For example, macroeconomic sustainability refers to the conditions that should be fulfilled in order to give a lasting character to low inflation and positive rates of growth. But, at the same time, these conditions seem to be necessary for long-term development, as was illustrated by the impressive evidence provided by the "lost decade." On the other hand, lasting positive growth rates, as an essential ingredient of macroeconomic sustainability, will require, aside from other conditions, rates of investment in physical and human capital which suffice to assure a compatible growth of productive capacity. Then, in thinking of sustainable rates of growth, the analysis enters the core of the long-term development process. So, without going any further it is easy to see that both perspectives, in a broad sense, overlap substantially. However, we think that the distinction can be analytically fruitful if the issues are narrowed to focus on the main questions posed by the actual situation of the countries.
The advantage of this distinction is that it takes into account the special way in which economists view the country experiences. Certainly this is not the best, but it is a fact that there are two differentiated bodies of analysis and specific questions, regarding stabilization and macroeconomic equilibrium, on the one hand, and development on the other. So, differentiating the analyses makes it easier to anchor the discussion in the current debates on stabilization and development.
3. The early nineties regional macroeconomic performance and the international financial conditions. Regarding the macroeconomic sustainability of recent trends, our analysis focuses on the paths followed by the countries to "close" the external and fiscal gaps and on the fragility of those "closures" vis-à-vis domestic and international factors. Because of the leading role of the early nineties capital inflows and because their potential volatility seems to be the most relevant source of renewed disequilibria, we look at these factors with special attention.
We assume that lasting closures of both the public sector and the external accounts are necessary conditions for stability. The assumption is derived from both stabilization theory and previous Latin American experiences3. To make stabilization feasible, both gaps have to be closed and to make it last, both closures have to be credibly lasting4.
Some stabilization efforts in the eighties were initially well founded and showed good outcomes for some time, but, in the end, most of them were frustrated by the inability of economic policies to give long-term sustainability to the closures of either the external or the fiscal gap, or both. Although few stabilization programs implemented in the eighties after the debt crisis succeeded and lasted, like Chile's and Bolivia's, these cases were the exception. In general, in most of the countries in the region, shocks stemming from one or both of the mentioned disequilibria were the main sources of the recurrent instability.
The characterization of the regional situation of the external and fiscal accounts was just the opposite in the nineties. Almost all countries could manage their economic polices to reach closures of the fiscal and the external gaps that have lasted so far5. Behind the indicators that show positive rates of growth and lower inflation, this has been the most remarkable difference in the regional macroeconomic picture between the eighties and the nineties.
The most notable factor explaining the difference between the two periods is the evolution of the external sector and the international financial conditions the countries were confronted with. While the countries faced rationed financial markets and experienced capital outflows in the eighties, there has been a significant fall in the international interest rate followed by a sudden and substantial inflow of capital in the nineties.
With the relaxation of the external constraint, many of the mechanisms that contributed to the macroeconomic instability during the eighties were deactivated. The availability of external credit allowed for an expansion in domestic absorption. The reversal of capital inflows was so abrupt and significant that many countries faced an excess supply of foreign exchange, even though the trade and current-account deficit have been increasing fast. Consequently, the region as a whole accumulated international reserves during the period and most countries appreciated their currencies. The expansion of economic activity and the real appreciation had beneficial effects on stability. The lagging exchange rate played an important role in the observed process of disinflation. It also contributed to fiscal balance by lowering the real value of interest payments on the public debt. Fiscal revenues were buoyed by the recovery. The lower inflation rates contributed directly to improving the real value of collected taxes and easing the introduction of administrative and tax reforms aimed at augmenting the average rate of collected taxes. In some countries fiscal equilibrium resulted substantially from the proceeds of privatization financed by capital inflows.
A synthetic quantitative picture of the early nineties trends, for Latin America as a whole and for each of the selected countries, is presented in Table 1. We have disaggregated the information corresponding to 1994 to underline the more recent experiences and changes.
The average regional rate of growth was 3.3% in 1991-93 and somewhat higher in 1994. Mexico's growth was about 3% in 1991 and 1992, almost zero in 1993 and was pushed again to 3% in 1994, more by expansionary short-run polices than by development impulses. Argentina's GDP augmented almost 9% per year in 1991-92, after having experienced a deep recession in 1989-90. GDP growth was pulled up by an abrupt expansion of domestic demand that followed the implementation of the Convertibility Plan - the stabilization program launched in April 1991 - and which lasted for about two years. In 1994, although the rate of GDP growth was about 7%, it was mainly explained by services and construction, because the industrial sector's activity level has shown the exhaustion of the shock's demand effects since 1993. Also Chile grew faster in 1991-93 and decelerated in 1994, but in this case the slowdown was managed by policy, aimed at "cooling" the expansion of domestic demand. Colombia's growth performance has been essentially stable. The 1994 improvement with regard to the previous years' average is due to a transitory slowdown in 1991. Lastly, in mid-1993 Brazil initiated a recovery from a recession that had begun in 1990. In the second half of 1994, after the implementation of the "Plano Real", domestic demand experienced the same kind of sudden expansion that was seen in Argentina, under similar circumstances, in 1991.
Regarding the inflation figures, Table 1 reflects deceleration in all cases. In Argentina, "residual" inflation immediately following the Convertibility Plan was about 3% per month, a rate similar to those experienced in analogous circumstances by Mexico in 1988, and Brazil in the second semester of 1994. Owing essentially to the fixed exchange rate, inflation decelerated persistently in Argentina from 1991 on. Mexico's figures show a similar performance, also explained mainly by the role of the pre-fixed exchange rate. Chile is an indexed economy and so the current exchange rate pace and the demand pressures operate marginally on a "core" inflation determined by past rates. In this country, the inflation rate has been gradually falling from a high of 27.3% in 1990. The deceleration process is due to both the real exchange rate appreciation and demand management policies. Colombia is also an indexed economy where inflation has been fluctuating between 20% and 30% for several years. The rate reached 32.4% in 1990 and has since been falling slowly. The deceleration is explained, as in the other cases, by the exchange rate appreciation process. Only very recently, in late 1994, did the government attempt a comprehensive deindexation program - instrumented by a social pact with trade unions, private business organizations and local authorities - aimed at reducing the 1995 inflation rate to 18%.
The rest of the information included in Table 1 describes the capital inflows and the aggregate evolution of the external sector. In 1991-93 net capital flows to Latin America amounted to $166 billion while current-account deficits totalled about $98 billion. In all cases capital inflows were higher than current-account deficits, causing accumulation of reserves. From the mentioned total inflow, $75 billion corresponded to Mexico, $29.3 billion to Argentina, $19.5 billion to Brazil and $7.7 billion to Chile. So, in 1991-93 these four countries accounted for 80% of the global regional inflow. Mexico alone accounted for about 45%. It should be mentioned that, out of our sample of countries, capital flows were also significant in Peru and Venezuela. In 1993, net regional capital inflow reached an annual maximum of $70 billion, from which $29.5 billion corresponded to Mexico and $14.8 billion to Argentina. In both cases those figures also represented the maximum annual inflows in their individual records.
In 1994 the regional trend changed sign because total inflows fell to $47 billion6. It was not the case of Brazil. This country's 1994 inflows shifted upwards to $13 billion. Colombia also showed also an upward shift - less important than Brazil's - and in the rest of the region, with the exception of Argentina and Mexico, capital inflows were similar to the previous year. So, the regional decrease is fully explained by Mexico and Argentina and most of it by the former, whose net capital inflow fell to $10.5 billion, about one third of the previous year's. But the reduction was also significant in the Argentine case, where inflows totalled in 1994 $10.2 billion, 30% less than in the previous year.
The fall in Mexico's and Argentina's 1994 net capital inflows occurred while current-account deficits continued to expand. In 1993 they amounted to $23.5 billion in Mexico and $7.5 billion in Argentina, and shifted to $30.6 billion and $11.1 billion respectively in 1994. So, as a result of both capital inflow and current-account deficit trends, reserves declined in both countries in 1994 for the first time in the early nineties.
Table 1 includes information about the evolution of the regional trade that lies behind the mentioned current-account performances. The Latin American trade deficit showed a rising trend, attaining $15.3 billion in 1993 and accumulating $16 billion in the 1991-93 period. But these aggregate figures comprise contrasting country performances. Between the years 1991-93 Brazil showed an accumulated trade surplus of $39.8 billion and another $11.3 billion surplus in 1994 despite the imports boom of the second semester. Mexico and Argentina, on the other side of the regional spectrum, recorded an accumulated deficit of $71.3 billion in 1991-93, explained in both cases by a fast growth of imports. This trend persisted in 1994, when the deficit of the two countries added up to $29.6 billion. The Chilean trade account was in surplus in the nineties - except for 1993 - in spite of a moderate rising trend in imports. In Colombia, lastly, imports augmented fast throughout the nineties. The trade account passed from a $2.3 billion surplus in 1991 to deficits of $1.6 billion in 1993, and $2.1 billion in 1994.
Let us close this quantitative description of the early nineties performance by looking at the evolution of real exchange rates. Data are shown in Tables 2a and 2b. The figures of Table 2a correspond to the dollar exchange rate, measured in units of domestic currency per US dollar. The figures in Table 2b correspond to real effective exchange rates for exports and so, in addition to nominal exchange behavior and domestic consumer prices, they also account for each country trade partners real exchange rate performances.
As can be seen, appreciation throughout the nineties has been a generalized process. However, the countries differ with regard to its intensity. To assess these differences from a longer term point of view, the last column in the table shows the real exchange indexes as proportions of the 1986-90 averages. As can be seen, Mexico and Argentina show the most appreciated currencies. In Mexico, where the stabilization program was inaugurated in late 1987, appreciation was still important in 1988, in the first stage of the program. It continued at a slower pace until 1990 and then accelerated from 1991. In Argentina, there was a significant appreciation in 1990 and so the real value of the nominal exchange rate fixed in early 1991 had already appreciated. Chile and Colombia entered the nineties with relatively depreciated exchange rates and ranked on the opposite side of the regional spectrum in 1994. Chile appreciated less than the other countries. In Colombia, the appreciation process accelerated in 1994. In Brazil, there was a huge appreciation in 1990, as a consequence of the Collor Plan stabilization attempt. After its failure, the exchange rate was continuously depreciated until 1993 when the exchange policy again followed an appreciation path. In the second semester of 1994, after the "Plano Real" was launched, the real exchange rate appreciated about 30%.
In concluding the above analysis of the early nineties evolution of the external sector and the 1994 turning-point, two points deserve to be stressed. Firstly, the inertial character of the current-account trend and its short run autonomy with respect to the reduction of both capital inflows and reserves. In both Mexico and Argentina, current-account deficits rose persistently in 1991-93 while capital inflows were higher than them and reserves accumulated. In both countries nominal exchange rates were fixed or quasi-fixed and the real exchange rate gradually appreciated. The external sector situation changed in 1994, as was mentioned earlier, while the real exchange rate remained stable. Under these circumstances, the performance of the current account disregards the hypothesis of a fast and smooth adjustment vis-à-vis capital flows and stresses the role of the real exchange rate and the aggregate demand.
The outcome should not be surprising. The interests on the external debt stock and the trade-account results are the main elements of the current account. The first has an obvious inertial component. The second depends both on the exchange rate and on the domestic demand. Without devaluation, current account adjustment can only take place via recession. But the income-elasticity of imports is not very high and the adjustment process will take some time. So, under these circumstances, the current account will not adjust quickly - i.e., quick enough to fully compensate for the contraction of the capital account. Recession should be deep and lasting and while it develops, the current-account deficit should be continuously financed. Moreover, even if the exchange rate is devalued, its effects on imports and exports also involve time lags. This should be particularly true in Mexico and Argentina, where real appreciation took place together with trade liberalization and the productive structure has adapted considerably to cheap intermediate and final imported goods by specializing its production lines and operating with a high proportion of imported inputs - the so-called de-industrialization and de-substitution processes. So, a fast reduction of the current-account deficits should not have been expected in 1994, in either Argentina or Mexico and should not be expected to take place in the near future.
If capital inflows continue falling, both in the Argentine case - where the exchange rate has not yet been devalued and external adjustment depends exclusively on income effects for now - and in the Mexican case - because of the reasons mentioned above - current-account deficits will show a decreasing trend, but the adjustment will be relatively slow and there will be, then, continuous pressures on the reserves and permanent threats of speculative attacks.
It is precisely the inertial character of the current-account deficit, in conjunction with capital flows volatility, that ultimately brings on external crises. This difference between the velocities of adjustment of current and capital account is the main argument for regulations intended to smooth the volatility of the latter. Rapid shifts in financial portfolios are usually the triggering episode of the crisis - Mexico was not an exception in this regard - but they almost always follow a more or less extended period of reserve losses that would not occur if the current account adjusted quickly7.
The second point that seems worth stressing refers precisely to the reversion of the capital inflows trend observed in 1994. This should also have been expected. In the first stages of the capital inflows boom it was argued both in Argentina and Mexico that the fast imports trend was only a transitory adjustment effect of the trade and financial liberalization. The authorities recurrently announced throughout the early nineties that the trade account would show an improving trend in the near future. This was expected to take place because of the positive effects on competitiveness owing to free trade itself, higher capital goods imports, privatizations, and other structural reforms with supposedly positive effects on domestic productivity. In the Mexican case, particularly, the NAFTA agreement played an important role in the announced official expectations.
In fact, beyond the official expectations, when faced with the persistently rising current-account deficits, the Mexican and Argentine authorities, together with the Washington institutions, thought and acted as if the early nineties capital inflows trend could be permanent, or, at least, lasting enough to allow the countries to finance an extended period of very high external deficits.
Why should this have happened? In most of the countries in the region, particularly in Argentina and Mexico, direct foreign investment represented only a small fraction of capital inflows. A proportion of them were attracted by privatizations that can take place only once. Another proportion was portfolio diversification adjustments that stop at some point. Lastly, there were also short- term capital inflows attracted by very high financial yields that could not be sustained in the long run. So, the changing trend experienced in 1994 should have been expected. Clearly, this was not the case in Argentina and Mexico. But it was in some countries of the region, as we will discuss below, who did not allow their domestic economies to fully adapt to the early nineties massive inflows of capital to the extent that Mexico and Argentina did.
The description presented above intends to reflect the global regional evolution and serves to identify the changes in the international financial environment as the main differentiating factor of the early nineties with respect to the eighties. It should be mentioned that the emphasis we placed on the capital inflows phenomenon and on the external sector evolution does not mean to disregard the positive role some structural reforms played in the reactivation process. Neither do we disregard the efforts developed by some countries to improve the efficiency of public expenses and tax regimes. But it should also be stressed that most of the reforms could not have been developed and most of the efficiency gains could not have been realized if the countries had continued to be in the position of the eighties, i.e. financially rationed and obliged to heavy external transfers.
Even if it is true that capital inflows have effected almost every country in the region and that macro-performances are due everywhere to some extent to those inflows, national situations differ with regard to fragility, as was already mentioned. The differences between countries stem from both the characteristics of the closures of the external and fiscal gaps and from the policy regimes adopted in each country. As we will see, both are correlated. Before looking more closely at these issues, let us succinctly present some alternative views about what has happened in the region as a whole.