Abstract
THE MOVE TOWARD the elimination of restrictions and artificial impediments to foreign trade and capital flows, or what has come to be called the “liberalization” or “opening up” of economies, by some developing countries in the mid- to late 1970s created a great deal of interest on the part of academics and policymakers alike. The countries of the Southern Cone of Latin America—Argentina, Chile, and Uruguay—where the liberalization strategy was pursued most aggressively, and one might say with considerable fanfare, were being touted as the economic success stories of the past decade in both the international financial press and banking circles.1 Developing countries were being continually apprised of the benefits of “outward-looking” policies and encouraged to emulate the examples offered by the experiences of the Southern Cone countries. By now, however, the earlier enthusiasm has dissipated quite sharply as these very same countries have found themselves in serious economic difficulties that have precipitated the adoption of stabilization programs and, more important, even a reversal of direction in their liberalization policies. The sharp declines in growth rates in all three countries, and the increases in current account deficits to levels that were no longer sustainable, exerted considerable pressures on the authorities to retreat from the policies they had earlier announced. As these difficulties persist, the resolve of even the national authorities most committed to an open economy model is being severely tested. Needless to say, not many developing countries are being persuaded that the benefits of liberalization clearly outweigh the costs in the present set of circumstances. It seems that the euphoria that accompanied the initially successful attempts at liberalization has been replaced by serious misgivings, and perhaps even by a degree of pessimism, about opening up.
THE MOVE TOWARD the elimination of restrictions and artificial impediments to foreign trade and capital flows, or what has come to be called the “liberalization” or “opening up” of economies, by some developing countries in the mid- to late 1970s created a great deal of interest on the part of academics and policymakers alike. The countries of the Southern Cone of Latin America—Argentina, Chile, and Uruguay—where the liberalization strategy was pursued most aggressively, and one might say with considerable fanfare, were being touted as the economic success stories of the past decade in both the international financial press and banking circles.1 Developing countries were being continually apprised of the benefits of “outward-looking” policies and encouraged to emulate the examples offered by the experiences of the Southern Cone countries. By now, however, the earlier enthusiasm has dissipated quite sharply as these very same countries have found themselves in serious economic difficulties that have precipitated the adoption of stabilization programs and, more important, even a reversal of direction in their liberalization policies. The sharp declines in growth rates in all three countries, and the increases in current account deficits to levels that were no longer sustainable, exerted considerable pressures on the authorities to retreat from the policies they had earlier announced. As these difficulties persist, the resolve of even the national authorities most committed to an open economy model is being severely tested. Needless to say, not many developing countries are being persuaded that the benefits of liberalization clearly outweigh the costs in the present set of circumstances. It seems that the euphoria that accompanied the initially successful attempts at liberalization has been replaced by serious misgivings, and perhaps even by a degree of pessimism, about opening up.
What went wrong in such a short space of time is a question that is being repeatedly raised by interested parties both within and outside the countries that undertook the trade and financial reforms. Broadly speaking, it is possible to identify four main lines of thought on this issue. First, there is the view that the liberalization policies were themselves misconceived and were not really relevant for developing countries at even the theoretical level. For various reasons, given the institutional and structural characteristics of developing countries, it is argued that opening up is destined to fail, and therefore it is of no great surprise to find the countries that pursued such policies in their present straits. A second view, which is related to the first, does not condemn liberalization policies per se but holds that it was the implementation of these policies that was at fault. In other words, the proponents of this view, while tending to accept the neoclassical premise that opening up has long-run advantages, are nevertheless critical of how the policies were actually executed. Third, it is possible that the countries were the victims of exogenous—specifically external—shocks that coincided with their attempts to liberalize, and that the deteriorating international environment bears some part of the responsibility for the problems encountered by these countries. It is generally recognized that liberalization is fraught with difficulties, even under the most ideal circumstances, and that the task for the policymaker is made doubly demanding when the country is faced with external shocks while it is in the process of liberalizing. Finally, there is the argument developed by Edwards (1982), Dagnino Pastore (1983), Sjaastad (1983), and Dornbusch (1984), among others, that lays the principal blame for the difficulties on what are referred to as “domestic policy inconsistencies.” In essence their position is that fiscal, wage, credit, and exchange rate policies were not sufficiently coordinated and, in the end, proved to be in fundamental conflict with the overall strategy of opening up.
The true picture most likely combines elements of all the arguments listed above to some degree, and it would be difficult to pick only one as being the predominant cause of the “failure” of the liberalization experiments. For example, there is little dispute, even among the most ardent proponents of liberalization, that opening up does involve costs in the short term and the medium term. Whether these outweigh the potential benefits is a matter that has not yet been examined in detail, and it is also a difficult issue having strong welfare implications.2 In addition, it must be noted that several developing countries, principally in Southeast Asia, have been relatively successful in pursuing outward-oriented policies (although of a selective nature and with active government support) for many years now. The experience of these countries does provide a counterexample against any broad indictment of liberalization policies. How the policies ought to be implemented (in particular, whether they should be undertaken gradually or suddenly), and whether they should be simultaneously or sequentially applied to the trade and capital accounts, also are difficult questions. The short-run outcome for the economy is not independent of the way in which barriers to trade and capital flows are removed.3 Although different strategies may yield similar results in the long run, during the transition the behavior of the main macroeconomic variables can be quite different, and the choice between strategies would naturally have to depend on the government’s objective function.
From a less normative perspective, developing countries involved in the process of liberalization benefited at the beginning from quite favorable external economic conditions: bouyant export markets, improving terms of trade, and very low (even negative) foreign real interest rates and abundant capital inflows. In the late 1970s and early 1980s, however, these same countries were facing a worsening of the international economic climate marked by declining terms of trade, falling growth rates in industrial countries, sharp changes in the availability of foreign financing that were accompanied by a dramatic increase in real interest rates on external borrowing, and, finally, the growth of protectionist pressures in the developing countries’ principal export markets. Although these external changes affected most developing countries in some degree, the countries that were opening up found that their liberalization policies had perhaps made them relatively more vulnerable to shocks emanating from abroad than other developing countries that had continued to maintain restrictions on trade and capital flows. Furthermore, serious domestic policy inconsistencies arose in a few of the countries that were liberalizing, possibly because of, to quote Sjaastad (1983), the absence of any “master plan” of reforms. The level of fiscal deficits and the rapid growth in private expenditures financed by foreign borrowing turned out to be incompatible with the desired expansion in aggregate demand in certain cases; institutional wage indexation schemes continued to be the rule; domestic financial and other structural reforms, together with stabilization programs, led to excessively high real interest rates that had a negative effect on both investment and the cost structure of production; and, with the benefit of hindsight, it is now clear that the exchange rate policies adopted often led to steady real overvaluation of the respective currencies (see Dornbusch (1984)).
In a previous paper (Khan and Zahler (1983)) we analyzed the macroeconomic effects of opening up and dealt with the issues of the timing and sequencing of reforms in some detail. Because the focus of that paper was exclusively on the subject of liberalization, however, the analysis was conducted under two critical assumptions. First, we assumed that the international environment was unchanged; second, we assumed that for the most part there was no change in domestic fiscal and monetary policies, and that the nominal exchange rate was fixed. Although these two assumptions were obviously necessary to isolate the direct effects of liberalization, they are of course unrealistic from a historical perspective. The main purpose of the present paper is to relax these assumptions and to conduct some further simulation experiments with the model that was developed in our previous paper, comparing the outcomes with and without external and internal shocks. This type of exercise allows us to ascertain whether such shocks could indeed significantly alter the paths taken by the main macroeconomic variables after the opening up of the foreign sector. We purposely have excluded from the discussion the broader question of whether liberalization is in some sense beneficial (or not), and whether a particular type of liberalization strategy may be “optimal” for developing countries. Issues of this nature, although extremely important, are outside the scope of the essentially quantitative approach adopted here. We should also stress that, whereas the types of shocks we study here have indeed occurred in the 1970s, the exercise is still in large part hypothetical, and we do not pretend to reproduce the experience of any specific country.
The remainder of the paper covers the following. In Section I we discuss the changes in the international picture and how it affected developing countries as a group, as well as some of the domestic policies implemented by developing countries that embarked on the liberalization process. Section II presents the basic framework of the analysis, including a brief description of the model used. The results from the various simulations are presented in Section III. The concluding section brings together the main results and attempts to provide a judgment on the significance of external and internal factors in the liberalization experiments, and whether these factors were sufficiently important to unravel the whole opening-up process itself.
I. The International Environment and Domestic Policies
The late 1970s and early 1980s have been characterized as a period of considerable strain for non-oil developing countries facing an international environment increasingly inimical to their growth and current account prospects. Recent papers by Khan and Knight (1983) and Massad and Zahler (1984) have identified three external factors as being mainly responsible for the serious current account difficulties of this group of countries: the deterioration in the terms of trade, the slowdown in economic activity in the industrial world, and, toward the end of the decade, the sharp rise in real interest rates in international capital markets.4 At the same time, domestic developments—as evidenced by rising fiscal deficits and consequent inflationary pressures or by increased private expenditures financed by excessive foreign borrowing (see Massad (1983) and Zahler (1983))—combined with rigid exchange rate policies to compound the external payments difficulties that resulted from the deterioration in the international economic climate.
Insofar as the external factors are concerned, the terms of trade of the oil importing developing countries deteriorated at an average annual rate of 2 percent during the period 1973-83 (Table 1). 5 After falling sharply in 1974-75 in the wake of the fourfold jump in the world price of energy products in 1973-74, the terms of trade improved somewhat in the following two years as primary commodity prices in the world markets registered substantial increases. From 1978 to 1982 there was a steady worsening of the terms of trade as commodity prices continued to fall at the same time that these countries were again faced with another significant increase in the price of imported oil during 1979-80. Exacerbating the problem was the decline in the growth rates in industrial countries.6 After increasing at about 3 percent during 1973-77, the average growth rate in the industrial world declined in the following six-year period to around 2 percent a year. During 1980-83 the real gross national product (GNP) of the industrial countries grew only at an average rate of a little over 1 percent a year (Table 1). The increase in petroleum prices, the decline in primary product prices, and the fall in growth rates in industrial countries combined to worsen the terms of trade of oil importing developing countries at an annual average rate of over 3 percent during 1978-83. By contrast, the terms of trade had been relatively constant on average during the earlier period 1973-77.
Selected Macroeconomic Variables of Net Oil Importing Developing Countries, 1973-83
(In percent)
Three-month Eurodollar rate, adjusted for percentage changes in export unit values.
Consumer prices.
As percentage of exports of goods and services.
Selected Macroeconomic Variables of Net Oil Importing Developing Countries, 1973-83
(In percent)
| Variable | 1973 | 1974 | 1975 | 1976 | 1977 | 1978 | 1979 | 1980 | 1981 | 1982 | 1983 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Terms of trade | 4.0 | -8.7 | -7.5 | 5.6 | 6.8 | -4.5 | -2.1 | -7.1 | -4.8 | -2.9 | 1.8 |
| Foreign real interest rates1 | -16.2 | -17.3 | 6.4 | -0.6 | -7.7 | 3.4 | -3.2 | 1.7 | 19.9 | 20.3 | 11.6 |
| Growth of real gross domestic product (GDP) | 4.1 | 5.5 | 3.3 | 5.2 | 5.7 | 6.4 | 4.7 | 4.7 | 2.2 | 1.6 | 2.2 |
| Inflation2 | 23.8 | 29.9 | 30.1 | 29.7 | 23.8 | 21.4 | 25.9 | 33.2 | 32.3 | 31.1 | 39.1 |
| Current accounts3 | -9.0 | -24.5 | -27.6 | -16.2 | -12.7 | -15.7 | -19.4 | -22.0 | -22.5 | -18.8 | -13.4 |
| Memorandum item | |||||||||||
| Growth of real gross national product (GNP) of industrial | |||||||||||
| countries | 6.1 | 0.5 | -0.6 | 5.0 | 3.9 | 4.1 | 3.5 | 1.3 | 1.6 | -0.1 | 2.3 |
Three-month Eurodollar rate, adjusted for percentage changes in export unit values.
Consumer prices.
As percentage of exports of goods and services.
Selected Macroeconomic Variables of Net Oil Importing Developing Countries, 1973-83
(In percent)
| Variable | 1973 | 1974 | 1975 | 1976 | 1977 | 1978 | 1979 | 1980 | 1981 | 1982 | 1983 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Terms of trade | 4.0 | -8.7 | -7.5 | 5.6 | 6.8 | -4.5 | -2.1 | -7.1 | -4.8 | -2.9 | 1.8 |
| Foreign real interest rates1 | -16.2 | -17.3 | 6.4 | -0.6 | -7.7 | 3.4 | -3.2 | 1.7 | 19.9 | 20.3 | 11.6 |
| Growth of real gross domestic product (GDP) | 4.1 | 5.5 | 3.3 | 5.2 | 5.7 | 6.4 | 4.7 | 4.7 | 2.2 | 1.6 | 2.2 |
| Inflation2 | 23.8 | 29.9 | 30.1 | 29.7 | 23.8 | 21.4 | 25.9 | 33.2 | 32.3 | 31.1 | 39.1 |
| Current accounts3 | -9.0 | -24.5 | -27.6 | -16.2 | -12.7 | -15.7 | -19.4 | -22.0 | -22.5 | -18.8 | -13.4 |
| Memorandum item | |||||||||||
| Growth of real gross national product (GNP) of industrial | |||||||||||
| countries | 6.1 | 0.5 | -0.6 | 5.0 | 3.9 | 4.1 | 3.5 | 1.3 | 1.6 | -0.1 | 2.3 |
Three-month Eurodollar rate, adjusted for percentage changes in export unit values.
Consumer prices.
As percentage of exports of goods and services.
The other major external factor affecting the payments positions of oil importing developing countries during the late 1970s was the increased level of service payments on foreign debt. In the years before about 1977, debt service had not posed a serious problem for many oil importing developing countries because conditions in the international credit markets were in general favorable; even though foreign real interest rates rose in 1975,7 for the period 1973-77 the average was still strongly negative (–7.1 percent; see Table 1). Beginning in 1978, however, the picture changed quite dramatically as developing countries began to acquire increasing amounts of external debt at floating rates while interest rates in the international capital markets were climbing to postwar highs and export prices were beginning to weaken.8 Foreign real interest rates rose sharply and became positive—averaging about 9.5 percent a year during 1978-83—and in the last three years (1981-83) had reached the astronomical average level of over 17 percent. This turnaround of more than 16 percentage points between the periods 1973-77 and 1978-83 forced a number of oil importing countries to undertake strong adjustment efforts once the limits on their recourse to financing had been reached (see Massad and Zahler (1984)).
These external events were obviously beyond the control of the developing countries, but inappropriate domestic policies also contributed in no small measure to the overall difficulties faced by these countries. Fiscal policies in the non-oil developing countries continued to be expansionary as a rule, and budget deficits, expressed as a proportion of gross domestic product (GDP), more than doubled—from about 2 percent in 1973 to 4.6 percent in 1983 (see International Monetary Fund (1984, p. 51)). The pressure on available resources created by rising government expenditures, as well as the rapid increase in liquidity caused by the financing of these deficits and of private sector expenditures through the banking system, led to a significant upsurge in inflation. Because exchange rate changes typically tended to lag behind domestic price increases that were in excess of those experienced by trading partners, upward pressure was put on the real exchange rate, which was allowed to appreciate substantially in several developing countries during the decade (Khan and Knight (1982, 1983)).
The cumulative effect of the external shocks and inappropriate domestic policies on the main macroeconomic variables in the oil importing developing countries is evident from Table 1. With 1978 taken as a convenient breakpoint, the annual average rate of GDP growth fell from about 5 percent in 1973-77 to an average of a little over 3.5 percent during 1978-83. The steady decline in the growth of real GDP from 1978 is particularly noticeable, and in the last two years (1982-83) real GDP grew by an average of less than 2 percent. After averaging about 10 percent in the period 1968-72, inflation rose to an annual average rate of nearly 30 percent during 1973-83; for the most recent years 1980-83, inflation was running at an average annual rate of about 34 percent.
Finally, the current account position of the oil importing developing countries, expressed as a proportion of exports of goods and services, deteriorated sharply following the first major oil price increase in 1973-74, with the deficit reaching a peak of nearly 28 percent of exports of goods and services in 1975. Favorable movements in the world prices of primary commodities led to a marked improvement in the current account balances in 1976-77, but from then on the deficit rose steadily through 1981. Because the possibilities of increasing exports significantly were small in view of the relatively flat foreign demand situation, the adjustment policies initiated by some countries in late 1981 consequently put considerable emphasis on reducing aggregate demand, which caused a decline in imports and economic activity. The current account picture did improve as a result of these efforts, and the average ratio of the current account deficit to exports of goods and services during 1982-83 was reduced to about 16 percent. Yet the overall economic picture in 1983 was one of striking deterioration on all fronts. The decade of the 1970s, except for the years of abundant foreign financing and low real foreign interest rates, contained some of the worst periods from an economic standpoint for countries that were perhaps the least equipped to handle such adverse circumstances.
The relative influence of external and domestic factors on the current account positions during 1973-80 was tested empirically by Khan and Knight (1983) for a sample of 32 non-oil developing countries. The results of that study showed that the most important determinant of the current account balances was the terms of trade, followed by foreign real interest rates, fiscal deficits, and real effective exchange rates, which were all roughly equal in importance. The growth of real GNP in industrial countries played a relatively minor role, but it can be argued that the effect of this variable is already captured to some extent in the terms of trade and foreign real interest rate variables.
Given the empirical results obtained by Khan and Knight (1983), it would be fair to hypothesize that the experiences of the countries engaged in opening up their economies—in particular the Southern Cone countries—would have been similar to those of the other non-oil developing countries. The economic situation of the Southern Cone countries during the past two to three years, however, has been far worse than the average outcomes for the group reported in Table 1. Growth rates in Argentina, Chile, and Uruguay had turned negative by the early 1980s; inflation in Argentina was far in excess of the average rate for net oil importing countries;9 and the ratio of the current account deficits to exports of goods and services was markedly higher in all three countries.
There would seem, therefore, to be at least some prima facie evidence that external shocks and domestic policy inconsistencies had a more severe effect on the countries that were liberalizing their foreign sectors. Why this was the case is the central question we focus on, and by analyzing it we should be able to take a somewhat firmer stance in trying to explain the failures of the liberalization experiments.
II. Framework of Analysis
The analysis of the effects of shocks that occur while the foreign sector is being liberalized is conducted within the framework of the dynamic general equilibrium model that was developed by Khan and Zahler (1983) to examine the transitional macro-economic effects of changes in barriers to trade and capital flows. The model has its roots in the general equilibrium econometric models developed by Clements (1980); the computational general equilibrium models such as the ones designed by Feltenstein (1980), among others; and the more monetary models typically specified to analyze short-term stabilization policies.10
A detailed description of the basic model is given in Khan and Zahler (1983); here we present only a brief outline of some of its main features. The model contains three composite goods—exportables, importables, and nontradables—for which supply and demand equations are separately defined. The supply equations are derived in a manner outlined by Clements (1980) in the framework of a multiproduct supply model. The supply of each good depends exclusively on the relative prices of the three goods, the technical conditions of transformation of one good into another, and the initial resource endowment.11 Total output of the economy is simply the aggregate of the outputs of the three goods, and the unemployment of resources is modeled as a linear function of the difference between potential output (equal to the resource endowment) and total output.
The demand system incorporated in the model represents a fairly straightforward application of standard demand theory. The private component of total aggregate expenditures is related to disposable income, the excess supply of money, and the domestic interest rate; government expenditures, as is customary, are assumed to be exogenous.12 By invoking separability we argue that, once total expenditures are determined, the distribution between importable, exportable, and nontradable goods is determined by a process of maximization that is subject to a budget constraint represented by the (given) level of aggregate expenditures. This yields demand equations for each good, as a proportion of aggregate expenditures, that depend solely on relative prices, satisfying the properties of symmetry and additivity.
The domestic price of exportables (Px) is taken to be equal to the product of the international price of tradables (Pf) and the exchange rate (ε), and the domestic price of importable goods, allowing for tariffs, is defined as
where Pi is the domestic price of importable goods, τ is the tariff, ∊ is the exchange rate, and Pf is the international price of tradable goods. Because the prices of importable and exportable goods are essentially given from abroad (assuming that the exchange rate and the tariff level remain unchanged), disequilibrium in the tradable goods markets results in changes in imports and exports. Imports are defined as the difference between domestic demand and domestic supply of importables; similarly, exports are equal to the domestic excess supply of exportables.
The price of nontradable goods (Pn) is, however, endogenously determined and is assumed to respond positively to excess demand for nontradable goods (and variations in foreign prices). The general price index—constructed as a Divisia index of the three composite goods, with the (endogenous) weights corresponding to the expenditure shares of each of the three goods—is therefore endogenous as well. Expectations of inflation are also incorporated into the model, although in a fairly simple fashion using an adaptive expectations formulation.
The monetary sector of the model contains three basic relations: a money demand function, a money supply identity, and an equation that links changes in the domestic interest rate to the excess demand for money. The demand for money is specified in the customary way; that is, relating money holdings to income, inflationary expectations, and the domestic interest rate. The supply of money is made up of net international reserves, credit to the private sector, and credit to the government. It is assumed that all fiscal deficits are financed by government borrowing from the banking system, so that there is a one-to-one correspondence between the budget deficit and variations in the money supply brought about by changes in credit to the public sector. For the case of the interest rate, a standard LM mechanism is assumed, so that an excess demand for (supply of) money leads to an increase (decrease) in the domestic interest rate. In the model, monetary disequilibrium affects aggregate demand both directly, through the spillover into private expenditures, as well as indirectly, through changes in the interest rate.
Capital flows, aside from an autonomous component, are assumed to be determined by the differential between domestic and foreign interest rates, adjusted for expected exchange rate changes and a country risk premium. The presence or degree of controls on capital movements is represented by a parameter, β, that scales the explanatory variables in the following way:
where DK is the flow of capital (with
To allow for the possibility of an upward-sloping supply curve of foreign credits, the risk premium is made a function of the ratio of external debt to income:
where ρ0 is a constant, Bf is the stock of external debt, and Y is the level of income. The parameter ρ1 is assumed to be positive, so that as the ratio (Bf/Y) rises the risk premium will also increase. This will reduce net capital inflows to the country even though domestic and foreign interest rates, and the expected exchange rate, remain unchanged.13
Despite its high level of aggregation compared with the computational general equilibrium models, the model is sufficiently detailed to be able to provide meaningful answers relating to the short-run consequences of opening up. The model explicitly incorporates the links between the balance of payments and the fiscal and monetary sectors, as well as their relations to expenditures and output. Moreover, considerable attention is paid to the role of relative prices in the demand and supply functions for the three composite goods. Finally, because it is formulated in dynamic form, the model is able to provide the path of adjustment of the main macroeconomic variables from one equilibrium to another.14 The analysis of the transition path, which is essential in devising operational liberalization strategies, clearly requires the introduction of some type of dynamics into the system.
The main theoretical characteristics of this model can be shown through some simple experiments relating to trade and financial liberalization. Consider first the case in which a country has a 100 percent tariff on imports, which it then reduces to zero. The relative price effect of this measure, following Dornbusch (1974), can be analyzed through the aid of Figure 1 (panel A). If income and expenditures are assumed to be equal, along the HH schedule there is no excess demand for tradable goods, and by Walras’s Law excess demand for nontradable goods is zero as well. Northeast of the HH schedule the relative price of nontradable goods is too low, and there is an excess supply of tradable goods (trade balance surplus) and an excess demand for nontradable goods. Similarly, southwest of HH there would be trade balance deficit and an excess supply of nontradable goods.
Real and Monetary Effects of Liberalization
Citation: IMF Staff Papers 1985, 001; 10.5089/9781451956696.024.A002
If τ equals 100 percent, the nominal exchange rate is fixed (and for simplicity set equal to unity), and the economy is closed to capital movements, the initial equilibrium is at point A, where the ray 0T (the slope of which measures the domestic price of importables in terms of the price of exportables) intersects HH. At A the relative prices of importable and exportable goods in terms of nontradable goods are Pi°/Pn° and Px°/Pn°, respectively, and there is equilibrium in both the trade balance and the nontradable goods market.
If τ is reduced to zero, the domestic price of importables falls (to Pi1) and rotates the ray to 0T”. Assuming that Pn is unchanged, the initial effect of the tariff reduction is represented by a movement from A to B,15 which involves an appreciation of the real exchange rate (defined as the ratio of the price of non-tradables to the price of tradables, including tariffs). Obviously this is not an equilibrium position, since at B there is an excess demand for tradable goods and an excess supply of nontradable goods, requiring a fall in Pn along 0T’ so as to restore general equilibrium at point C, with Pn1 < Pn°. This movement from B to C has been identified in the literature as the real exchange rate depreciation associated with trade liberalization. Although at point C the trade account is in balance, with both imports and exports above their respective values in the original equilibrium (A), the initial effect of opening up (point B) does generate a trade balance deficit. In other words, the depreciation of the real exchange rate associated with the movement from B to C represents a transitory equilibrating movement necessary to close the foreign exchange gap created by the trade deficit that occurs in the process of moving from A to C (through B).
The changes in relative prices and their effects on demands and supplies that result from tariff removal correspond to a sort of production and expenditure “switching” effect. Note, however, that opening up also creates an expenditure “augmenting” effect. Assuming that inflation is zero initially,16 the fall in the prices of importable and nontradable goods causes a reduction in the general price level, which in turn creates an excess supply of money and a fall in the domestic interest rate. This stimulates expenditures that reinforce the trade balance effect and, in the short run, dampen the fall in the relative price of nontradable goods (Blejer (1978)).
The effect of trade liberalization on aggregate supply and output can be seen in panel B of Figure 1, which relates the production possibilities between importables and exportables, with the resources used by the nontradables sector assumed to remain constant.17 At the initial relative price DD the economy would be at point A, producing XA of exportables and IA of importables. When the tariff on imports is reduced to zero, the country will face the new (domestic) terms of trade FF, and the new equilibrium will be at B. If adjustment were instantaneous, the economy would simply move along the transformation curve from A to B, and output of tradables would be unchanged. If the reduction in the production of importables is faster than the expansion of exportables, however, then the path of tradables output would be pushed inside the transformation curve (indicated by the dashed line). In such a case, during the transition period as the economy moves from A to B, it will be operating below its productive potential, creating greater resource unemployment and a larger output gap in comparison with the respective long-run equilibrium levels of these variables.
In sum, as demonstrated in panels A and B of Figure 1, the main theoretical results of a tariff reduction in the short run are a trade balance deficit and consequent loss of international reserves, an increase in both imports and exports, a lowering of the price level, a fall (rise) in the nominal (real) interest rate, and, if the production of importables is assumed to adjust faster than the production of exportables, a temporary decline in output and increase in resource unemployment.
To analyze financial opening up, as in Khan and Zahler (1983), we start from an initial equilibrium in which the domestic interest rate is above the foreign rate plus the risk premium and in which capital movements are completely restricted (β = 0). Financial liberalization takes the form of increasing the value of β, and capital movements then take place (with the expected—and actual—nominal exchange rate assumed to be constant) as long as
In the traditional IS-LM framework (panel C of Figure 1) the initial equilibrium point would be A, with real income at y0, the domestic interest rate equal to rd°, and zero foreign debt (Bf = 0).18
With a constant risk premium and foreign interest rate, the (small) country faces an infinitely elastic supply of international financial capital that, when monetized, makes the effective LM curve horizontal.19 The short-run effect of financial opening up is therefore represented by shifting LM to KK. At point B expenditure (yB) exceeds income (Y°) and induces a current account deficit.20 Whether international reserves rise or fall obviously depends on the size of the capital inflows relative to the current account deficit. As a consequence of the capital inflow, the stock of foreign debt would naturally rise.
In Khan and Zahler (1983) it was assumed that the resource endowment (potential output) was fixed, an assumption that implies zero net savings and investment.21 Consequently, as output remains constant and foreign debt increases, the risk premium rises and increases the total cost of financing faced by the country. This shifts KK in panel C of Figure 1 upward to K1 K1, reducing the difference between expenditures (yc) and income (y°) and, therefore, reducing the current account deficit. At the point C the inflow of capital is smaller because of the lower interest differential, and the foreign debt rises at a smaller rate. The process continues until a new overall equilibrium is reached at the original values of income and the domestic interest rate (point A), with expenditure equal to income and the current account in balance. At A, however, there now is a larger stock of foreign debt and higher risk premium, and a lower level of real expenditures on goods and nonfinancial services, compared with the initial equilibrium.
The main results of financial opening up are that the domestic nominal interest rate initially declines and then rises back to its original level. The current account deficit is financed by increases in foreign debt rather than by a fall in international reserves, as was the case in trade liberalization. During the transition period real expenditures on goods and nonfinancial services increase, but then are lower in the final equilibrium because of the need to service the now larger stock of foreign debt.22
The model embodying the characteristics described was simulated in our earlier paper (1983) for a variety of opening-up strategies—including, among others, the gradual and sudden removal of barriers to trade and capital flows (both simultaneously as well as in different sequences)—and was found to yield generally sensible results. The way in which this model is structured makes it quite capable of handling a large variety of shocks aside from those directly related to opening up. The only change we made in the original model was to introduce a distinction between the price of importables and exportables, which had previously been assumed to be equal to a single international price level. This change had to be made to enable us to discuss terms of trade variations, and therefore the current version of the model contains two separate foreign prices—one for importables and the other for exportables.
Formal models of any type are clearly not able to analyze all of the interesting questions arising from foreign sector reforms, nor for that matter can they capture the complex nature of some of the other structural changes implemented that are less amenable to quantification. But the advantages of using a model, such as the one here, to determine the effects of liberalization, external shocks, and autonomous domestic policy changes are obvious in comparison with the approaches taken by, for example, Ffrench-Davis and Arellano (1981), Ffrench-Davis (1982), Dagnino Pastore (1983), Edwards (1982), Harberger (1982), Hanson and de Melo (1983), Sjaastad (1983), Wogart (1983), Zahler (1983), and Ramos (1984). Several things tend to be occurring simultaneously during the period of interest—opening up itself, stabilization efforts and their various effects, structural and institutional changes, and so forth—and it is only with a model that one can realistically hope to identify and isolate the effects of different sets of factors. In other words, by using a model one is able to make suitable assumptions for other things being equal, something that is not really possible in the type of historical and descriptive studies that have hitherto addressed the issue.
The simulation experiments start with the case of a gradual reduction in trade barriers and restrictions on capital movements. This particular case was studied by Khan and Zahler (1983) as well, and here it is taken as the “control” or baseline simulation with which the other simulations are compared. The specific foreign shocks we consider are a simultaneous temporary increase in the nominal foreign interest rate and a temporary deterioration in the terms of trade. The change in the terms of trade is taken, for purposes of this particular exercise, as a decline in the price of exportables relative to the price of importables.23 The domestic policy inconsistency analyzed here is represented by a simulation in which there is a temporary increase in the fiscal deficit.24 Furthermore, since it is assumed in all the simulations that the nominal exchange rate is fixed, this policy implicitly yields a second inconsistency that has been stressed in the recent literature. Keeping the exchange rate unchanged while opening up in the presence of certain external shocks or an expansionary fiscal policy will in general lead to a real appreciation, defined here as an increase in the price of nontradables relative to the price of tradables, a weakened external payments position, and increased vulnerability to speculative attacks on the currency. In each of the simulations undertaken, we trace the response of the following macro-economic variables: the general price level, the domestic interest rate, the current account balance, international reserves and foreign debt, the real exchange rate, and real expenditure on goods and nonfinancial services.
These various simulations obviously do not cover all the shocks that occurred during the 1970s. For example, we do not explicitly consider the effects of a slowdown in the growth rates in industrial countries. Because this effect was not found to be particularly significant in the results reported by Khan and Knight (1983), we felt that we could exclude it from consideration here. Furthermore, we do not attempt to determine the effects of growing protectionist pressures in industrial countries on the exports of developing countries. Neither of these simulations is particularly difficult to perform, but each would require some respecification of the basic model to incorporate a foreign demand function for exports. As the model is currently formulated, it utilizes a small country assumption and implicitly assumes that foreign demand for exports is infinitely price elastic. Finally, we do not go into the question of wage indexation because the model does not include an explicit wage determination equation, although it is possible, as discussed in Khan and Zahler (1983), to handle this indirectly. Nevertheless, we feel that the simulations here provide sufficient information to enable one to form a reasonable judgment about the principal effects that some specific external shocks and certain policy changes are likely to have in the course of liberalization.
III. Results
In the initial equilibrium the economy is assumed to have a uniform tariff of 100 percent on imports, and capital flows are completely restricted.25 In other words, the economy is not completely closed to trade because imports are allowed (although at a domestic price substantially higher than the world price) and the country does engage in export activities. Yet neither capital inflows nor outflows are permitted. The balance of payments, the current account, and the government budget are all in balance; prices are constant; the economy is assumed to be on its aggregate transformation curve (with a “normal” rate of unemployment of 5 percent assumed); and the exchange rate is fixed. In specific terms, liberalization involves lowering the tariff rate gradually to zero over four periods and simultaneously eliminating restrictions on capital flows, also over four periods.26 In the control simulation the foreign interest rate, the terms of trade, and the government budget deficit are kept unchanged (Chart 1). For the external shock scenario, the nominal foreign interest rate is raised to 15 percent in period 3 and is then lowered back to its original level of 5 percent after four periods (panel A). Concurrently, the terms of trade are assumed to deteriorate by 5 percent per period for four periods, and then to improve progressively so that by the seventh period they are at their original level (panel B). The domestic shock is represented by the emergence of a fiscal deficit (approximately equal to 7-8 percent of national income) for four periods (panel C), after which the budget is once again assumed to be balanced.
A critical point to note in analyzing the simulations reported here is that the outcomes for the variables under consideration are conditional on the numerical values of the parameters of the underlying model (see Khan and Zahler (1983, pp. 276-79)). Clearly, alternative scenarios could be created by changing the parameter values employed. Moreover, the values chosen for the shocks, and the periods over which they extend, are only illustrative and are not intended necessarily to be realistic. Even though these specific shocks are arbitrary, however, they nevertheless should give a reasonable idea of what can be expected to happen if shocks of certain types are superimposed on the opening-up process.
Control Simulation: Simultaneous Liberalization of Trade and Capital Flows
Liberalization of the trade and capital accounts directly lowers the price of importables (by the amount of the reduction in the tariff rate) and thus initially raises the relative prices of both exportables and nontradable goods in terms of importable goods. The change in the pattern of demand and production caused by the change in relative prices tends to exert downward pressure on the price of nontradables as well. As a consequence, the general price level falls quite rapidly in the beginning and then, once the effects of the tariff reduction have worked themselves out, stays permanently at the new lower level (Chart 2, panel A).
Combined Effects of External and Fiscal Shocks
Citation: IMF Staff Papers 1985, 001; 10.5089/9781451956696.024.A002
As expected, the fall in the overall price level lowers the nominal demand for money; furthermore, because the domestic interest rate is assumed initially to be above the corresponding foreign rate, with the removal of capital controls there is a large inflow of capital from abroad that augments domestic liquidity. The resultant excess supply of money causes the domestic interest rate to decline (panel B), and both these factors have an expansionary effect on aggregate demand. The combination of the change in relative prices and the rise in domestic absorption yields a pronounced deficit in the current account (panel C) that persists for several periods.27 The volume of trade (imports plus exports), however, is larger than in the initial equilibrium, which is the desired result of the liberalization policy. Given the parameters of the model, the capital flows generated by the interest rate differential are not adequate to cover the deficits in the current account, so that the country will continue to lose international reserves until monetary equilibrium is re-established (panel D). By the end of the transition the stock of international reserves falls to less than one half of the original level. In the context of our model this result points to an important precondition for liberalization policies: that, when starting off the process of opening up, the policymakers should ensure that the country has a comfortable cushion of reserves. The foreign debt of the country rises in a somewhat cyclical fashion, reflecting closely the path taken by domestic interest rates and the resultant capital inflows. Until the risk premium rises by enough to close the differential between domestic and foreign interest rates, the stock of foreign debt will continue to increase. In this particular simulation, equilibrium is reached when the final stock of foreign debt is about 25 percent of national income (panel D).
Two additional results, which were not stressed in Khan and Zahler (1983), are worth mentioning. First, real expenditures on goods and nonfinancial services, which can be treated as a proxy measuring the welfare effects of liberalization policies, increase substantially when the domestic price of importable goods falls (Chart 2, panel E).28 This tendency is then reversed as interest payments on foreign debt absorb an increasing proportion of the income of residents, although, given the model structure and specific parameter values, in the long run equilibrium real expenditures on goods and nonfinancial services are still higher than their level before the liberalization.
Second, as was analyzed in Figure 1, during the course of liberalization the domestic relative price of importable goods decreases with respect to the other goods, and the relative price of exportables tends to rise. With the assumed parameter values and the initial shares of the three goods in total output, the real exchange rate (defined as the ratio of the price of nontradable goods to the price of tradables) will appreciate (panel F). This real appreciation is a natural consequence of the removal of tariffs on importable goods, and the economy has to move to a new equilibrium real exchange rate. Other things being equal, this appreciation will result in a loss of international competitiveness and a worsening of the current account for some time. Although this movement represents an equilibrium change, the authorities could reduce its effect on the current account through appropriate exchange rate policy.29 What is more important, however, is the prevention of a real appreciation beyond the new equilibrium real exchange rate that is consistent with the elimination of restrictions on trade and capital movements.30
Effects of External and Domestic Shocks
The effects of a combination of external shocks and the emergence of a budget deficit while the foreign sector is being liberalized are also shown in Chart 2. It is quite evident from panel A that the various shocks appear to have little effect on the path of the general price level that is the result of opening up alone. We would expect the deterioration in the terms of trade to exert additional downward pressure on prices because the decline initiated by the fall in the domestic price of importables is amplified by the reduction in export prices. At the same time, however, the fiscal deficit would tend to push up the price of nontradable goods by increasing aggregate spending. The values of the parameters of the underlying model are such that these effects tend to offset each other, and the net effect on the general price level turns out to be negligible.
Although the domestic interest rate does fall when the shocks are superimposed, the decline is somewhat smaller than in the control simulation (panel B). Because the foreign interest rate is increased, there is a net capital outflow during the first few periods and a smaller excess supply of money, even though the financing of the fiscal deficit expands the nominal money supply.31
A more striking difference between the two sets of simulations can be observed in the case of the current account position (Chart 2, panel C). Although there is less excess liquidity in the economy during the initial periods, the combined effect of the deterioration in the terms of trade and of the expansionary fiscal policy causes the current account balance to be significantly worse than it would be in the absence of such shocks. Starting from an equilibrium position, the current account deficit, as a proportion of nominal income, reaches around 18 percent by period 6 (compared with less than 7 percent in the same period in the control simulation). Once the shocks have worked themselves out, the paths of the current account balance from the two simulations become quite close.
Accompanying this larger current account deficit is an initial outflow of capital because of the increase in the foreign interest rate and the decline in the domestic interest rate brought about as a result of the removal of capital controls. As a consequence, international reserves decline much more rapidly in this scenario (panel D); in the final equilibrium the stock of international reserves actually becomes negative. In marked contrast to the control simulation, the stock of foreign debt falls for the first eight periods or so 32 and rises steadily thereafter, although its level remains permanently smaller than in the control simulation (panel D). This scenario would necessarily mean that debt service payments would be smaller than in the scenario without shocks, despite the temporary increase in the foreign interest rate.
From periods 2 to 6 real expenditures on goods and non-financial services increase significantly more than in the control simulation (Chart 2, panel E). This increase is primarily due to the expansionary effect of the budget deficit, moderated somewhat by the effect of the terms of trade deterioration on domestic spending. As the terms of trade and the foreign interest rate return to their respective original levels, and the fiscal deficit is eliminated, real expenditures end up being slightly higher than in the control simulation because of the smaller debt service payments.
The appreciation of the real exchange rate also turns out to be more pronounced when there are external and domestic shocks (Chart 2, panel F). This marked appreciation occurs for two reasons. First, the price of tradable goods falls relatively more, with the decline in import prices caused by the tariff reduction now being accompanied by a fall in the price of exportables. Second, the expansion in aggregate demand caused by the fiscal deficit increases the price of nontradable goods. Eventually, as the foreign price of exportables returns to its original level and the fiscal balance is re-established, the real exchange rate first depreciates (relative to the control simulation path) and then moves to a slightly higher equilibrium level. This long-run result occurs because fewer resources from the tradable goods sector—that is, smaller trade balance surpluses—are required to service the now lower stock of foreign debt.
The movements in the real exchange rate clearly highlight the importance of adopting an appropriate exchange rate policy during the liberalization process. Maintaining a fixed nominal exchange rate—or, for that matter, simply operating a policy that does not permit the exchange rate to move in line with the relative price of tradables to nontradables when there are external shocks or domestic fiscal imbalances—would not appear to be suitable. In particular, budget deficits, and excessive private expenditures financed by foreign borrowing, cause an expansion in aggregate demand that is basically inconsistent with the reduction in the price of nontradable goods needed to keep the current account deficit (and consequent loss of international reserves) within reasonable limits. In such circumstances fixing the nominal exchange rate would exacerbate the situation. This issue of inconsistency between excess domestic expenditures and exchange rate policies has been discussed by several authors engaged in analyzing the experience of the Southern Cone countries during the 1970s (for example, Edwards (1982), Dagnino Pastore (1983), Sjaastad (1983), Zahler (1983), and Dornbusch (1984)).
In connection with the exchange rate issue, an interesting question arises concerning the likely effects if the country actually did adopt a more flexible exchange rate policy while it engaged in opening up. We analyzed this case by repeating the experiment of reducing tariffs and eliminating capital account restrictions, but now instead of maintaining a fixed exchange rate we allowed for a gradual depreciation of about 50 percent between periods 3 and 6. This policy led to a much smoother evolution of the real exchange rate over time, although, as expected, in the final equilibrium there was still a real appreciation. The decline in the general price level and domestic interest rate was markedly smaller, as was the increase in real expenditures, relative to the control (fixed exchange rate) simulation. There was also an improvement in the current account position, even though restrictions on trade and capital flows were removed, and the stock of international reserves rose initially before settling down to a value quite close to the original equilibrium level. Although the policy of steady depreciation is not totally without costs—there is an increase in the foreign debt above that observed in the control simulation, and the price level falls by a smaller amount—nevertheless it can be argued that at least some of the negative aspects of the transition period following liberalization can be moderated if the authorities pursue a more flexible exchange rate policy.33
IV. Conclusions
The widespread interest generated by countries that embarked in the direction of opening up their economies to allow for the freer flow of goods and capital across borders still continues. There has, however, been a radical shift in the kind of questions being raised about the relative economic performance of these economies. Initially the questions focused on the success these countries achieved in some macroeconomic areas, but now, in the light of their current situation, equally relevant questions are being asked about the supposed failure of outward-oriented policies. The burgeoning literature analyzing the experience of the countries in the Southern Cone of Latin America continues to attest to the considerable puzzlement that remains about how the situation there could have changed so dramatically in the space of only a few years.
Certainly there has been no shortage of reasons presented for this turn of events. In this paper we have focused on two specific reasons: first, the role of certain external shocks; and second, various domestic policy actions that proved in the end to be inconsistent with the overall strategy of opening up. These particular factors are ones that are relatively more amenable to quantitative analysis. It is well recognized that external events have contributed to the difficulties experienced by developing countries in the recent past, and that these events have obviously been outside the control of these countries. Nevertheless, it can be argued that opening up so as to increase economic efficiency and improve resource allocation has made the countries in question more vulnerable to foreign shocks. Furthermore, in some cases the inability to control excessive domestic spending, whether public or private, and the general inflexibility of exchange rate policies during the liberalization process compounded the problems. These failings thus can be held partially responsible for the negative consequences that emerged in the wake of the liberalization attempts.
In this paper we have specifically examined the short-run and medium-run effects that certain types of external shocks and an expansionary fiscal policy can have on key macroeconomic variables while the economy is being opened up. The simulation experiments, conducted using a general equilibrium model, were not intended to be completely realistic, in the sense of either being applicable to any particular country or reproducing the actual shocks that buffeted developing countries, but rather should be treated as providing essentially hypothetical scenarios. In particular it was shown that the removal of barriers to trade and capital flows produces some costs in the short run. Although the rate of inflation and the domestic interest rate tend to approach their respective international values, this is accompanied by a rise in the real rate of interest, a decline in output and employment, a worsening of the current account, a loss of international reserves, and a significant buildup of foreign debt. There is, furthermore, an appreciation of the real exchange rate to a new equilibrium level. In the context of this particular exercise, the negative effects of opening up become magnified if one then also allows for the possibility of domestic policy inconsistencies and an adverse international climate during the liberalization process.
Whether the liberalization experiments undertaken in the Southern Cone would have been successful in the absence of external shocks, and if the right macroeconomic policies had been in place, is a question we have purposely refrained from addressing. What we can say, however, is that the possibilities of success were greatly diminished once the international picture worsened and countries pursued other domestic policies that apparently worked at cross purposes with the policy of liberalization. On the basis of the results reported here it is possible that, if external shocks are large and persist for long enough and if inconsistent domestic policies are maintained, the economy may go into a recession if compensatory action is not taken. In reality, limits on external borrowing prevented the developing countries in question from financing their way out of the difficulties brought about by the various shocks and inappropriate policies, and these limits finally forced them to undertake painful adjustment—and even to reverse the overall strategy and reimpose barriers to trade and capital flows.
What then is the policy lesson that can be drawn from the analysis conducted here? The conclusion that comes out quite forcefully is that opening-up policies have to be supported by active domestic macroeconomic management. This becomes even more imperative if the country is subjected to external shocks while it is in the process of liberalizing the foreign sector. Whereas the individual developing country can do little about changes in the international environment and has to take those as given, it would seem that a judicious combination of external financing, use of international reserves, and domestic adjustment would be called for at an early stage to offset or to minimize the effects of any external shocks that occur. Because there are quantitative limits to the amount of international reserves and foreign financing, it is obvious that eventual adjustment of the basic supply-demand balance in the economy is necessary. Fiscal and monetary restraint, to control both public and private spending, and a more flexible exchange rate policy would seem to be the relevant instruments of adjustment under the circumstances in which several developing countries found themselves in the late 1970s and early 1980s. Although there were attempts in this general direction, they can perhaps be characterized as having been too little, too late. As a consequence, many developing countries, and particularly those engaged in the process of opening up, found that they eventually had to undergo more painful adjustment than would have been necessary if appropriate actions had been taken more promptly.
REFERENCES
Blejer, Mario I., “The Short-Run Dynamics of Prices and the Balance of Payments,” American Economic Review (Nashville, Tennessee), Vol. 67 (June 1977), pp. 419–28.
Blejer, Mario I., “Nontraded Goods and the Monetary Approach to the Balance of Payments: A Graphical Note,” Quarterly Review of Economics and Business (Urbana, Illinois), Vol. 18 (Summer 1978), pp. 102–06.
Blejer, Mario I.,, and Roque B. Fernandez, “The Effects of Unanticipated Money Growth on Prices and on Output and Its Composition in a Fixed-Exchange-Rate Open Economy,” Canadian Journal of Economics (Toronto), Vol. 13 (February 1980), pp. 82–95.
Clements, Kenneth W., “A General Equilibrium Econometric Model of the Open Economy,” International Economic Review (Osaka, Japan), Vol. 21 (June 1980), pp. 469–88.
Dagnino Pastore, José Maria, “Assessment of an Anti-Inflationary Experiment: Argentina in 1979-81,” in Economic Liberalization and Stabilization Policies in Argentina, Chile, and Uruguay: Applications of the Monetary Approach to the Balance of Payments, ed. by Nicolás Ardito Barletta, Mario I. Blejer, and Luis Landau (Washington: World Bank, 1983), pp. 21–38.
Diaz-Alejandro, Carlos, “Goodbye Financial Repression, Hello Financial Crash,” Journal of Development Economics (Amsterdam), forthcoming (1985).
Dornbusch, Rudiger, “Tariffs and Nontraded Goods,” Journal of International Economics (Amsterdam), Vol. 4 (August 1974), pp. 177–85.
Dornbusch,Rudiger, “Stabilization Policies in Developing Countries: What Have We Learned?” World Development (Oxford, England), Vol. 10 (September 1982), pp. 701–08.
Dornbusch,Rudiger, External Debt, Budget Deficits and Disequilibrium Exchange Rates, NBER Working Paper No. 1336 (Cambridge, Massachusetts: National Bureau of Economic Research, April 1984).
Edwards, Sebastián, Economic Policy and the Record of Economic Growth in Chile in the 1970s and 1980s, UCLA Working Paper No. 283 (Los Angeles: University of California, November 1982).
Edwards, Sebastián, and Sweder van Wijnbergen, Welfare Effects of Trade and Capital Market Liberalization: Consequences of Different Sequencing Scenarios, NBER Working Paper No. 1245 (Cambridge, Massachusetts: National Bureau of Economic Research, December 1983).
Edwards, Sebastián, and Sweder van Wijnbergen, The Order of Liberalization of the External Sector in Developing Countries, Essays in International Finance No. 156 (Princeton, New Jersey: Princeton University, December 1984).
Feltenstein, Andrew, “A General Equilibrium Approach to the Analysis of Trade Restrictions, with an Application to Argentina,” Staff Papers, International Monetary Fund (Washington), Vol. 27 (December 1980), pp. 749–84.
Ffrench-Davis, Ricardo, El experimento monetarista en Chile: Una síntesis crítica, Colección Estudios CIEPLAN 9, (Santiago, Chile: Corporation de Investigaciones Económicas para Latinoamerica, December 1982), pp. 5–40.
Ffrench-Davis, Ricardo, and José Pablo Arellano, Apertura financier a externa: La experiencia Chilena en 1973-80, Colección Estudios CIEPLAN 5, (Santiago, Chile: Corporación de Investigaciones Económicas para Latinoamerica, July 1981), pp. 5–52.
Frenkel, Jacob A., “The Order of Economic Liberalization: Lessons from Chile and Argentina—A Comment,” in Economic Policy in a World of Change, ed. by Karl Brunner and Allan H. Meltzer, Carnegie-Rochester Conference Series on Public Policy, Vol. 17 (Amsterdam: North-Holland, 1982; New York: Elsevier, 1982), pp. 199–201.
Goldstein, Morris, and Mohsin S. Khan, Effects of Slowdown in Industrial Countries on Growth in Non-Oil Developing Countries, Occasional Paper No. 12 (Washington: International Monetary Fund, August 1982).
Goreux, Louis M., Compensatory Financing Facility, Pamphlet Series No. 34 (Washington: International Monetary Fund, 1980).
Gotur, Padma, “Interest Rates and the Developing World,” Finance & Development (Washington), Vol. 20 (December 1983), pp. 33–36.
Harberger, Arnold C, “The Chilean Economy in the 1970s: Crisis, Stabilization, Liberalization, Reform,” in Economic Policy in a World of Change, ed. by Karl Brunner and Allan H. Meltzer, Carnegie-Rochester Conference Series on Public Policy, Vol. 17 (Amsterdam: North-Holland, 1982; New York: Elsevier, 1982), pp. 115–52.
Hanson, James A., and Jaime de Melo, “The Uruguayan Experience with Liberalization and Stabilization, 1974-1981,” Journal of Interamerican Studies and World Affairs (Beverly Hills, California), Vol. 25 (November 1983), pp. 477–508.
International Monetary Fund, World Economic Outlook: A Survey by the Staff of the International Monetary Fund, Occasional Paper No. 27 (Washington, April 1984).
Khan, Mohsin S., and Malcolm D. Knight, “Stabilization Programs in Developing Countries: A Formal Framework,” Staff Papers, International Monetary Fund (Washington), Vol. 28 (March 1981), pp. 1–53.
Khan, Mohsin S., and Malcolm D. Knight, “Some Theoretical and Empirical Issues Relating to Economic Stabilization in Developing Countries,” World Development (Oxford, England), Vol. 10 (September 1982), pp. 709–30.
Khan, Mohsin S., and Malcolm D. Knight, “Determinants of Current Account Balances of Non-Oil Developing Countries in the 1970s: An Empirical Analysis,” Staff Papers, International Monetary Fund (Washington), Vol. 30 (December 1983), pp. 819–42.
Khan, Mohsin S., and Roberto Zahler, “The Macroeconomic Effects of Changes in Barriers to Trade and Capital Flows: A Simulation Analysis,” Staff Papers, International Monetary Fund (Washington), Vol. 30 (June 1983), pp. 223–82.
McKinnon, Ronald I., “The Order of Economic Liberalization: Lessons from Chile and Argentina,” in Economic Policy in a World of Change, ed. by Karl Brunner and Allan H. Meltzer, Carnegie-Rochester Conference Series on Public Policy, Vol. 17 (Amsterdam: North-Holland, 1982; New York: Elsevier, 1982), pp. 159–86.
Massad, Carlos, “La deuda externa y los problemas financieros de America Latina,” Revista de CEPAL, Comisión Económica para America Latina de las Naciones Unidas (Santiago, Chile), No. 20 (August 1983), pp. 149–63.
Massad, Carlos and Roberto Zahler, “Adjustment Policies in the Eighties: The Need for a Global Approach,” in Adjustment with Growth: A Search for an Equitable Solution, ed. by Khadija Haq and Carlos Massad (New York: North-South Roundtable, 1984).
Ramos, Joseph, “Estabilización y liberalización ecóndmica en el Cono Sur,” Estudios e informes de la CEPAL 38 (Santiago, Chile: Comisión Económica para America Latina de las Naciones Unidas, 1984).
Sjaastad, Larry A., “Failure of Economic Liberalism in the Cone of Latin America,” World Economy (Oxford, England), Vol. 6 (March 1983), pp. 5–26.
Wogart, Jan Peter, “Combining Price Stabilization with Trade and Financial Liberalization Policies: The Argentina Experience, 1976-1981,” Journal of Interamerican Studies and World Affairs (Beverly Hills, California), Vol. 25 (November 1983), pp. 445–76.
Zahler, Roberto, “Estrategias alternativas de apertura: Un modelo de simulation,” Monetaria (Mexico City), Vol. 5 (July-September 1982), pp. 303–58.
Zahler, Roberto, “Recent Southern Cone Liberalization Reforms and Stabilization Policies: The Chilean Case, 1974-1982,” Journal of Interamerican Studies and World Affairs (Beverly Hills, California), Vol. 25 (November 1983), pp. 509–62.
Mr. Khan, Advisor in the Research Department, is a graduate of Columbia University and the London School of Economics and Political Science.
Mr. Zahler, Economist with the United Nations Economic Commission for Latin America (ECLA), is a graduate of the University of Chile and the University of Chicago. He was formerly a member of the Research Department of the Central Bank of Chile and Professor of the Graduate Program for Latin American Economic Studies (ESCOLATINA).
The participation of ECLA in this study was conducted under United Nations Development Programme project RLA/77/021, “Implications for Latin America of the Situation of the International Monetary System,” directed by Carlos Massad. We are grateful to Ken Clements, Sebastian Edwards, Jacob Frenkel, Carlos Massad, and colleagues in our respective institutions for helpful comments on an earlier draft.
See Harberger (1982). Recent papers by Edwards (1982) and Diaz-Alejandro (1985) contain an interesting sampling of quotations from the financial press on this issue.
An interesting recent attempt to analyze these types of welfare-related questions is contained in Edwards and van Wijnbergen (1983).
See McKinnon (1982), Frenkel (1982), Edwards (1984), and Edwards and van Wijnbergen (1983) for a discussion of the issues involved in the sequencing of reforms.
See also Khan and Knight (1982). Other external factors would include the growth in protectionist pressures in the principal export markets of the developing countries and the fairly drastic contraction in capital flows to the non-oil developing countries in 1982 and 1983 that sharply increased the costs of adjustment of debtor countries. It is, however, difficult to deal with either of these factors in a quantitative fashion.
This group basically includes all non-oil developing countries except those referred to as “net oil exporters.” For a precise classification, see International Monetary Fund (1984, pp. 167-68).
There is some empirical evidence now on the positive relation between growth in industrial countries and the international prices of non-oil primary commodities; see Goreux (1980) and Goldstein and Khan (1982).
The empirical evidence on the relation between high real international interest rates and prices of primary commodities is discussed by Gotur (1983).
The rate of inflation in Uruguay was close to the average rates experienced by developing countries as a group, whereas that of Chile was well below the average.
We can include in this group the models of Blejer (1977), Blejer and Fernandez (1980), and Khan and Knight (1981).
Because the resource endowment is fixed, we do not allow for any net investment or savings.
Private expenditures on goods alone require that interest payments on foreign debt be subtracted out. Note also that we assume that the pattern of government spending on the three goods is identical to that of the private sector.
The relevance of this particular formulation to the analysis will be made clear later in this section.
In general this is not possible in the larger computational general equilibrium models.
Following Khan and Zahler (1983), we assume here that nontradable goods substitute with each of the tradable goods, but for simplicity we rule out cross-price effects between importables and exportables. This assumption does not change the conclusions reached by Khan and Zahler (1983) in any important way.
Beginning with a positive rate of inflation would not alter the analysis.
In general, a change in relative prices would change production of nontradable goods, and the transformation curve between importable and exportable goods would have to shift. We make this restrictive assumption in panel B of Figure 1 only for expositional purposes.
Starting with a positive stock of foreign debt would not change any of the basic results.
In the Khan and Zahler (1983) model it was assumed that the response of foreign capital to interest rate differentials was high, although not instantaneous. The approach adopted here for the graphical analysis—that is, shifting the LM curve—yields qualitatively similar results.
The current account deficit is also marginally increased by the increase in the relative price of nontradable goods that results from the excess supply of money created by the inflow of capital.
This would hold, for example, if domestic and foreign savings were perfect substitutes.
It should be stressed that, by assuming that net savings are zero, the potential benefits of financial opening up are in a sense being minimized. For an alternative approach that allows productive capacity to grow with foreign savings, see Zahler (1982).
Obviously one could also obtain the same decline in the terms of trade by increasing the price of importables relative to the price of exportables. The outcomes, however, are not symmetrical, so that one has to be careful to note that our results are conditional on how the terms of trade change is specified.
A similar set of results is obtained if private sector expenditures are increased through an expansion in domestic credit.
For the reasons for choosing an equilibrium position to begin from, see Khan and Zahler (1983, p. 245).
All changes in the exogenous variables are assumed to occur in the third period.
In equilibrium the current account is set equal to 100; values below 100, therefore, imply a current account deficit.
Real expenditures, as defined here, are closely related to the concept of national income adjusted for changes in the terms of trade.
Of course nominal depreciation of the currency would entail certain costs, particularly with respect to inflation.
For a discussion of the reasons that the real exchange rate may appreciate in the course of a stabilization program, see Dornbusch (1982, 1984).
One would normally expect an increase in the fiscal deficit to result in a greater excess supply of money, but in this framework there is a larger increase in the demand for money (owing to the increase in nominal income); furthermore, the additional monetary expansion created by the deficit leaks out very rapidly through the balance of payments.
In the model only foreign residents are allowed to acquire domestic debt, and domestic residents (including the government) are restricted from holding foreign debt. Under these conditions an increase in foreign interest rates, by reducing the incentive for foreigners to invest, leads to a smaller (or even negative) inflow of capital and a lower stock of foreign debt.
This result confirms the argument, put forward by Dornbusch (1984), that the exchange rate policies in the Southern Cone countries led to steady overvaluation of their respective currencies, and that this outcome could have been avoided through more flexible exchange rate management.
