The first point to be made in talking about structural adjustment, stabilization, and growth in almost all developing countries is that it is absolutely essential for these countries to shift from the inwardly oriented policies that have protected and insulated their domestic markets. If adjustment is to succeed, their policies must be outwardly oriented. The question is not whether this change should be made, but how, and most of the real issues lie in the particulars of accomplishing the reforms, not in the broad outline.

Anne O. Krueger

The Need for an Outward Orientation

The first point to be made in talking about structural adjustment, stabilization, and growth in almost all developing countries is that it is absolutely essential for these countries to shift from the inwardly oriented policies that have protected and insulated their domestic markets. If adjustment is to succeed, their policies must be outwardly oriented. The question is not whether this change should be made, but how, and most of the real issues lie in the particulars of accomplishing the reforms, not in the broad outline.

This shift in trade strategies is necessary in part because of the failure of the inwardly oriented policies that have been so important in influencing the structure of domestic production and incentives and in part because of the potential the international market offers. For a variety of reasons, inwardly oriented strategies have led to increasing inefficiencies over time. For example, the shortage of foreign exchange has had a number of consequences that have adversely affected the efficiency of production. Factories can remain idle until spare parts arrive. In more than one country, harvests have been lost because there has been no fuel to transport the crops.

The absence of domestic competition has also contributed to the growing inefficiencies. I do not mean perfect competition, in the classic textbook sense, but simply competition. In most countries that rely on domestic markets to create new industries and that have consequently encountered shortages of foreign exchange, the domestic markets are too small to provide competition for domestic firms. When this absence of competition is combined with an import control system that evolves because of foreign exchange shortages, the outcome is often a lack of incentives for producers to reduce costs. In fact, producers are distracted; their attention is diverted from reducing costs to activities such as obtaining import licenses, locating the necessary new materials and inputs, or simply ensuring that production takes place at all. The same factors also result in a lack of the market pressures that might force inefficient firms to close down or encourage efficient firms to expand.

In an outwardly oriented economy, by contrast, access to international markets can help spur necessary competition and, in turn, improve resource allocation. Low-cost profitable firms can expand despite the size of the domestic market. But in inwardly oriented economies, once production matches domestic demand—usually at monopoly prices—producers simply look for something else on the import list they can produce. These producers are more willing to go after monopoly profits in another market rather than they are to try expanding production, even of their own reasonably cost-effective products.

In the past several years, economists have increasingly been looking at other factors that create inefficiencies, including the problems that arise when producers cannot interact easily with international markets. Not being able to interact globally means losing access to the latest production techniques, quality control standards, and technological advances. Some economists are now arguing that this inability to interact globally has cost producers at least as much as other factors have.

Finally, developing countries that rely exclusively on a domestic market have basically trapped themselves, again because these domestic markets are so small. Not only do developing countries tend to have small populations, but consumer purchasing power is also limited by low incomes. India is a very large country, but the Indian market for industrial production—other than for essential consumer goods such as clothing—is no larger than Sweden’s. Although the Swedes know very well that they cannot rely only on their domestic market, leaders of many developing countries have not been as wise, with disastrous consequences. Even firms with excellent managers and modern production techniques will not be especially cost-effective or efficient if their production runs are 1 percent (or less) of those in other countries. The unit costs will be too high, and too many resources will be used in the process of changing from one specification to another.

For all these reasons, many developing countries have begun to recognize some of the disadvantages of the inwardly oriented strategy. But there is a second part to the story that needs to be taken into account in considering the need for structural adjustment: the fact that the same incentives used to encourage import substitution discourage exports. First, tariffs and quantitative restrictions on raw materials, intermediate goods, and capital goods—not to mention spare parts and replacements—all raise the costs of doing business within the domestic economy. Either producers experience delays in receiving imports, or the imports become much more expensive.

Second, tariffs and quantitative restrictions make producing for the domestic market very profitable. A producer of bicycles can sell the finished products on the domestic market at 60 percent above the world price at a cost that is only 80 percent of the world price. This producer has a choice: to expand bicycle production and try to export at a much lower price, or to produce something else on the import list, using the 60 percent mark-up in another monopoly position. Anyone but the most loyal of patriots will choose another domestic activity rather than expanding bicycle production for export. So there are a pool of resources on the import substitution side and a high cost on the export side.

A third and often overlooked factor is that countries following import substitution policies usually restrict imports, in part because they want to use their foreign exchange to import the capital goods that will increase capacity in import substitution industries. The result is an exchange rate that is not consistent with a fully open trade and payments regime. Or, to put it another way, the real exchange rate is appreciated relative to what it would be in an outwardly oriented trade regime that actively encourages exports.

All three of these factors make exporting less attractive. A recent World Bank study on agricultural exports examined the situation of farmers producing export crops.1 The study looked at the factors that typically penalize these farmers and weighed the effects of each—the exchange rate, tariffs, and the resultant higher prices on both consumer goods and inputs such as fertilizers. I was surprised to discover that the high costs farmers incur as a result of policies designed to protect domestic industry are at least as important as those caused by exchange rate misalignment. In addition, an interaction term makes the combination of the two even worse than either one alone would be.

In most developing countries, analysts have been puzzled by slow export growth. Why do these countries restrict imports? As it turns out, not only do they want to encourage domestic import substitution, but they also feel that they cannot generate sufficient earnings to make producing for export worthwhile. This argument fails to recognize that these countries’ inability to profit from exporting is built into their incentive systems. In terms of structural adjustment, it is necessary to shift the structure of protection and challenge the system of incentives so that the impediments to open trade are removed.

The Process and Sequencing of Trade Reform

The first step in serious trade reform—as contrasted with another export promotion campaign or some other superficial attack on the problem—must involve granting exporters unfettered, ready access to international markets for intermediate goods and raw materials. This idea is not as simple as it sounds, because it already supposes that the exchange rate is appropriate and serves as an incentive to induce exporters to act in an economically efficient manner while, at the same time, pursuing profits. If the exchange rate, differentiated or otherwise, is overvalued, giving exporters easy access to imports is difficult and normally results in widespread cheating or evasion.

For similar reasons, if exporters are to have easy access to international markets, tariffs cannot be too high or quantitative restrictions too stringent. High tariffs or binding quantitative restrictions tempt exporters to import items that are valuable in the domestic market, leading the authorities to inspect all imports to see whether they are really needed. And once inspections and other controls are in place, there is no easy way for exporters to gain access to the inputs they need.

In a sense, then, just saying that exporters will be freed from the constraints of existing exchange and trade controls is a very strong statement, even in the short run. There cannot be tremendously high rewards for those who would import certain items under the guise that these items are the widgets needed for production. No country I know of has really succeeded in structural adjustment without moving very quickly to allow exporters virtually effortless access, especially administratively, to international markets. This situation is not an end in itself, only a minimum precondition.

Second, exporters must have, and must be able to anticipate, a reasonable rate of return—that is, there must be not only a reasonable exchange rate in the short run but the expectation that rewards will remain adequate. In Chile in the early 1980s, for instance, some exporters felt that it would not really matter if the exchange rate depreciated, because copper was so important to the country that the government let the exchange rate appreciate every time the price of copper rose. These businessmen did not expect the government’s policy to make very much difference in the long term. Given the behavior of Chile’s exchange rate over a long period, the authorities found it difficult to convince potential exporters to pay attention to the prevailing relative prices and to persuade these exporters that the prices would remain attractive long enough to merit investment.

The policy problem is one not only of making prices rewarding today but also of finding ways to make producers believe that returns will remain rewarding in the longer term. This “credibility gap” is perhaps the greatest difficulty policymakers confront when they really want to change the structure of an economy. The issue is more complex in some countries than in others and depends in part on a country’s history and previous exchange rate policy. In Chile, there had already been three or four preannounced reforms. As in many countries where there have been unsuccessful preannounced reform efforts, people in Chile found it difficult to believe that another announcement could be any more than a pious statement of intent. This distrust made the task of reformers in the 1970s and 1980s much harder than it would have been if the earlier efforts had not been made. Unfortunately, when credibility is lost, the difficulties for policymakers increase.

Third, along with allowing exporters easy access to international markets and ensuring a reasonable and sustainable rate of return, governments must reduce incentives for import substitution in order to prevent resources from being pulled in that direction. Penetrating the export market is not easy, and people will not make the effort if other, less stressful money-making activities are available. In particular, certain mechanisms cannot be maintained, such as the automatic prohibition of imports once domestic production has started; and some policies must be discontinued, including those that guarantee monopoly positions in the domestic market even after incentives have changed. At a minimum, quantitative restrictions that protect new undertakings and allow expansion of existing import substitution in the domestic market need to be removed.

Finally, in order for an outwardly oriented trade strategy to succeed, incentives must be uniform. There cannot be a genuine structural adjustment program if the export subsidy is 50 percent for steel, 25 percent for tables, and 35 percent for copper products. Uniform incentives for potential exports need not be achieved only through the exchange rate. Other instruments can be used, but the resulting incentives must be created for exports, not for the iron and steel industries, the automobile sector, or other domestic businesses policymakers feel deserve more subsidies.

Problems Affecting Reform

If the fundamentals just discussed are in place for structural adjustment, what comes next, and what are the problems? First, most firms engaged in import substitution do not understand what they will gain from a reform program and may in fact be quite convinced that they will lose. In the mid-1970s, with the help of the Korea Development Institute, I located prominent Korean businessmen who had been working in the country in the late 1950s. My goal was to discuss with them the situation before and during Korea’s trade reforms. Every one of these businessmen stated that the reforms had been good for the economy and observed that the business climate had improved as a result. Yet all those I spoke with also said they had opposed the reforms in the beginning because of the uncertainty involved, which many described as “overwhelming.” (In general, trade reforms do pose a real risk, from the viewpoint of policymaking, of creating a kind of “contagious hysteria” as people talk each other into a state of high anxiety about potentially dire outcomes.)

Some high-cost firms or industries will suffer under genuine structural adjustment. Some activities are simply not economically feasible, and since structural adjustment is intended to raise productivity and living standards, these activities must adjust dramatically or fail altogether. In most instances, however, businesses are remarkably adept at changing their cost structures in response to new incentives, altering product mixes and taking other actions to increase productivity. In addition, the costs of some intermediate goods drop, offsetting price pressures somewhat, so that in the end, the number of firms suffering large losses is much smaller than people expect. In addition, economic growth tends to speed up within two or three years, so that everyone begins to benefit, even though a few firms may feel they are suffering because of competitive pressures.

Because true reform inevitably brings with it a degree of genuine hardship, my acid test of whether a reform program is working is to ask whether policymakers have been willing to allow any large plants to close. If the answer is “no,” then I conclude that the reform is not serious. My test is a hard one, especially since such closures are politically troublesome. But without this evidence of a government’s commitment to reform, there is room for doubt. And as long as that doubt remains, there is every reason for businesses not to try to change their behavior.

The difficulty, then, is that uncertainty about a government’s commitment to reform feeds into the uncertainty felt by potential exporters. They may decide that policies will return to what they were before, since the politicians are under fire; that the exchange rate will stay fixed in nominal terms; that inflation will take off again, leading to balance of payments difficulties; that exchange controls will be reimposed; or that new factories built to produce for export will be unprofitable. Businesses will then take a “wait and sec” attitude, as they often have even in countries such as Korea, where reforms and structural adjustment ultimately succeeded.

It has never been true that exporters will start investing heavily the day after reforms are announced. In general, two or three years of uncertainty follow, during which businesses produce for export out of their existing capacity and perhaps learn about the international market. But primarily they are waiting to see what will happen. During this period, the government needs to do whatever it can do to increase the reform program’s credibility. The sooner credibility is established, the more likely it is that investment will occur. Of course, if there are widespread protests and vociferous opposition to the reforms, the lag is likely to be longer.

In short, businesses must be able to turn a profit soon enough after reform to overcome uncertainty, but this profitability must be sustainable. There is no point in taking measures that are likely to be reversed if some people object to them. The first steps are usually fairly easy and include adjusting the exchange rate and providing exporters with ready access to needed imports. These two steps typically encourage some additional exports, even in the short run. However, what normally happens first is something I mentioned before, which I call “excess capacity exporting.” New, productive capacity is not immediately brought on line to produce exports in response to the improved incentives. Rather, producers use their existing plants and equipment to take advantage of the new opportunities.

In any of the countries that have been major success stories in terms of external sector reforms—such as Korea, Chile, or Turkey—the commodity composition of exports in the year or two after the reforms has become more diverse. Some activities have later dropped off the export list, while others have expanded rapidly. This development shows that the new incentives are effective, that people are now beginning to invest, and that exports can continue to grow because of this investment, ensuring that enough is being produced to supply both the export and the domestic markets.

Other Aspects of Reform

Several issues will occupy policymakers in the first three or four years after trade policy reform is initiated on imports: eliminating quantitative restrictions; reducing the maximum tariff rate (with the expectation that it will fall further); maintaining a realistic exchange rate; moving to a positive minimum real interest rate; controlling inflation; and liberalizing the capital account.

Eliminating quantitative restrictions

Removing quantitative restrictions is not difficult, but it should be done in steps. The first step, which I have already mentioned, is to ensure that exporters are not subject to controls, either on the imports they need for production or on their exports. The second step usually is to switch from a negative list to a positive list—that is, to make sure that only the items listed are prohibited and that all other imports are allowed. This one simple step cuts down delays enormously. Third, exporters and importers can be allowed to receive licenses automatically for extended periods. Restrictive licensing mechanisms can be terminated in favor of less restrictive ones, and at the same time the process of eliminating restrictive lists can be initiated. Turkey started with three lists: a bilateral list, a quota list, and a so-called liberalized list, which were abandoned one by one over a four-year period. In Mexico, 87 percent of imports were once covered by quantitative restrictions that have gradually been eliminated; they now cover less than 20 percent of imports.

Reducing tariffs

In addition to eliminating quantitative restrictions, most countries fare better if they begin reducing the highest import tariffs and announce an intended schedule of reductions. Policymakers disagree on how soon the announcement should be made and how quickly the changes should be implemented, but the arguments for moving quickly become more important when credibility is at stake. Typically, the tariff cutting is initiated with the announcement of a maximum tariff rate and a schedule for lowering both that rate and tariffs in general. An alternative is to set a maximum tariff rate and then to cut all tariffs by a certain percent across the board. In some countries, the situation is complicated by the fact that tariffs may be so high that cutting them does not affect the domestic price. That is, while a tariff rate may be 600–700 percent, the domestic price is only around 500 percent more than world prices. The first round of tariff cuts should be sufficient to eliminate any redundancy in the tariff. More than one country’s policymakers have learned too late that although they intended to reduce tariffs, all they had really done was to reduce the excess protection in the rates.

It is important to reduce the highest tariffs first. Then, as structural adjustment begins to take effect, it is important to make further cuts. High tariffs are economically inefficient, and leaving them in place diverts resources and imposes costs not only on producers but also on consumers, who must pay higher prices. More important, if a government does not take action to reduce high tariffs immediately, people will suspect that the authorities are not serious about the new trade strategy.

Exchange rate policy

Thus far, I have discussed trade policy as if it could be changed without being accompanied by other reforms, and of course it cannot. For example, liberalizing trade requires an exchange rate policy that assures producers the rate will not become overvalued in real terms. This issue, which is key to exporters, can be approached in a variety of ways. In Korea, subsidies were announced (along with a number of other incentives) for all exports in won-per-dollar terms, and the number of won per dollar changed as the exchange rate fluctuated. The incentives were uniform—that is, there was no differentiation by commodity—and as time went on, the exchange rate was unified. As usually happens, the export subsidies could be supported in the early stages of structural adjustment and were in fact effective in maintaining the real exchange rate for the first three or four years. However, as exports grew in importance, budgetary costs made the subsidies unfeasible.

Excess domestic demand

Another general policy matter that must be considered in liberalizing trade is domestic demand. If there is excess demand and a rising rate of inflation, once again, by definition, the domestic market is highly profitable. And if the domestic market is very attractive, producers are not going to risk trying to expand into the export market.

Equally important, when inflation rates rise, relative price signals lose their ability to communicate information about relative scarcities to decision makers; it is not clear if a price is rising because of general inflation or if a sustainable relative price change is taking place. During structural adjustment, there must be visible relative price changes. In particular, exportable prices must rise relative to the prices of import-competing and home goods. Furthermore, these relative price changes must be seen as signaling something important about relative scarcities. In that sense, macroeconomic policy during structural adjustment must eliminate enough excess demand not only to ensure that relative price signals are clear but to keep the domestic market from becoming so profitable that producers simply ignore the international market.

Usually, the primary source of excess domestic demand is the government’s own budget, and the impact is most often on home goods prices. In turn, the public sector, with its excess of expenditures over receipts, often bids away the very goods and services the domestic economy should be providing more efficiently for such purposes as export, transport, and communications. Macroeconomic fiscal and monetary policy must also be looked at from this standpoint.

Controlling inflation

There is no point in raising the nominal price of exports if the only result is that the price of everything else also increases, forcing export prices up still further. Maintaining control of aggregate expenditures, so that goods and services are freed up for the international market, is a key issue in trade reform. Continuing that strategy, if it is successful, will mean moving further toward a more stable price level and a lower inflation rate over time. Thus, there is both good news and bad news: while the inflation rate must be reduced, it need not be reduced to zero all at once. In reality, most countries have enough rigidities built into their economies to make the idea of zero inflation unfeasible, except when the starting point is hyperinflation and the situation is desperate.

The financial market

The financial market must be liberalized enough to allow profitable firms access to credit to finance expansion of their exporting activities. Most analysts agree that, at a minimum, the real interest rate must be positive—not necessarily a market-clearing rate in the early stages, but positive nonetheless. Without such a rate, loans are worth less when they are paid back than when they were first extended, creating a resource misallocation of enormous proportions and diverting credit to activities that are uneconomic but that do not appear so to private producers with access to subsidized loans. So I would argue that very early in the macroeconomic reform effort, interest rate policy must be changed so that borrowers are, at the least, paying a positive real price for credit.

The capital account

The capital account is a problematic issue, because a fully liberalized capital account requires even more stringent exchange rate, interest rate, monetary, and fiscal policies than are generally required in the move toward an outwardly oriented trade strategy. The real interest rate must be at least equal to the world interest rate at both domestic and at international price levels (as perceived through the exchange rate). Thus, there must be an anticipated path for the exchange and interest rates that is consistent with price level adjustments. Investors must not be tempted to move funds to, for example, Germany or the United Kingdom in order to make a profit when the exchange rate depreciates. The experience of most countries has been that it is difficult, if not impossible, to achieve full capital account convertibility in the early stages of structural adjustment.

Capital account liberalization must follow the relevant macroeconomic magnitudes, aligned and constrained in such a way that prices will maintain capital account equilibrium. Capital flows are sensitive, and policymakers often unintentionally create the conditions for private citizens to profit (with little or no downside risk) by speculating on the exchange rate. For that reason, the relevant argument is that policymakers want to open up and gradually remove the restrictiveness of controls on the capital account carefully and slowly, making sure that current account conditions are appropriate for capital account liberalization.

However, with a liberalized current account, there cannot be too many restrictions on the capital account. If, there are, too many profitable opportunities are created for activities that, in effect, use the current account to speculate against anticipated currency devaluations. It makes sense to think in terms of postponing the removal of the controls on large capital movements or large purchases of foreign currency by domestic residents until fairly late in the process of structural adjustment. Controls can be eliminated once the exchange and domestic financial markets are working fairly well, usually after a period of time has elapsed during which people learn that they can buy and sell foreign exchange with ease. Opening up the capital account, especially for domestic residents’ transactions abroad, as the first step in external sector liberalization is not generally recommended. Rather, the capital account should be opened when there is proof that most other reforms are succeeding and people believe they will continue.

The Experiences of the Successful Reformers

In light of what I have just said, it is interesting to see exactly what characteristics are shared by countries that have undergone successful structural adjustment. First, all successful reformers have started with uniform export incentives and a realistic exchange rate. In addition, exporters are often offered special privileges at a unified rate—for example, export credits at two percentage points below the nominal interest rate. Although the rate is subsidized, the subsidy itself is uniform, as are tax incentives and access to credit and imports. I know of no case of successful structural adjustment that did not include the removal of disincentives for exporting soon after the beginning of reforms. While the exchange rate and other policies may not have been exactly right, the successful countries moved quickly to make sure that producers knew exporting would be profitable in both the short and the long run.

Second, all countries that have moved from an inwardly to an outwardly oriented trade strategy as part of successful structural adjustment have seen their inflation rates fall in the first two years after reform began. This fact demonstrates the crucial link between adjustment policies and the macroeconomic situation. In some countries, however, external sector reform has been more successful than macroeconomic stabilization over the longer term. Turkey is one such country. The Turks freed their exchange rate, eliminated quantitative restrictions, and abolished most tariffs in the early 1980s, but they could not bring inflation under control. While they reduced the annual rate from 100 percent in 1980 to 35 percent in 1986, it has subsequently risen, and most observers today would say that without better macroeconomic controls, Turkey’s dramatic success on the trade side will be jeopardized.

Third, in all cases of successful adjustment, exporters have been allowed free access to imports, which has translated quickly into further liberalization of the entire import regime. Turkey allowed its exporters to retain 15 percent of their export earnings in order to purchase any imports, even those on the control list. Immediately, exporters were bringing in commodities with high domestic prices, implicitly reducing the highest tariff. In Korea, exporters were given a so-called wastage allowance—the cost of the material inputs wasted in the production process. Those materials, however, were not wasted; rather, they were used for domestic market production in place of inputs subject to import restrictions.

In all cases there was an initial excess capacity response and additional foreign exchange earnings, which enabled the authorities to allocate more foreign exchange and gradually cut down the number of restrictive import lists. Import liberalization soon followed, and current account controls were greatly relaxed. Restrictions on business travel were removed in almost all successful exporting countries shortly after the initial opening up, and foreign services became much more accessible.

While there has not been full capital account liberalization, the controls over people’s purchases of foreign exchange for current account transactions have been reduced, if not eliminated. In addition, restrictions on capital account transactions have gradually been relaxed, and the remaining restrictions are really only precautions. In all cases, the exchange rate stayed realistic for exporters. In Turkey, as I said, the inflation rate increased, but the currency was devalued sharply in 1980 and continued to depreciate slightly in real terms through 1985. At no point was the problem of domestic inflation allowed to cut off the incentives for exports.

Quantitative restrictions on imports have also diminished greatly in importance in the reforming countries. In counties where an export-oriented strategy has been in place more than ten years, the restrictions have generally disappeared completely. In general, quantitative restrictions give way to tariffs, because giving exporters quick rebates is much easier in a tariff system than in a system that uses quantitative restrictions. In all cases, it is interesting to note that the exchange rate, the interest rate, and other indirect instruments of control eventually became the primary levers policymakers relied on to support economic activity. Discretionary instruments, such as export subsidies, credit subsidies, and tax exemptions, became less important and finally disappeared. Indexed measures are uniform for all economic activities, and the lack of discrimination is one reason for their success.

Korea, as mentioned earlier, initially offered export subsidies and tax inducements. But by 1974, the tax and credit subsidies had been phased out, and by 1980 the remaining export exemptions were gone. The exchange rate now supports exporters and that carries the entire weight of the incentive structure, as it should. It took Korea more than 15 years to reach this point, and the process of liberalizing financial markets, which is not complete, continues. But as the Korean economy grows, new bottlenecks develop; the Koreans have learned that policymakers tend to view successful structural adjustment as an ongoing process.

Further, I do not think that any country with a long history of promoting an outwardly oriented trade strategy has not seen its inflation rate drop markedly. Korea had a virtually stable price level by the mid-1980s, and Taiwan’s price level actually fell for two or three years in the late 1980s. Other countries have had similar experiences. In all cases, low inflation has not necessarily been achieved at the outset but has been realized over time.

What happens when countries embark on a trade policy that is successful in shifting the structure of their economies? In some cases, the effects are remarkable. In Turkey, the share of exports in GNP rose from 5 percent in 1980 to 22 percent in 1987. In Korea, the share rose from 3 percent in 1960 to 45 percent in 1980. The changes may not be too pronounced in the first two or three years, but within five or six years they can become dramatic. Policymakers have increasingly discovered that it is impossible to maintain the direction of economic activity with hands-on, sector-specific controls once such changes have begun. As the domestic economy becomes more intertwined with the world economy, efforts to affect any one industry have unanticipated ramifications for the rest of the economy. For this reason, the tendency has been to use more general instruments to guide the economy—such as a unified exchange rate or positive real interest rates—rather than to rely on sectorial intervention.

As the economy becomes more complex, the costs both of sectorial interventions and of inflation rise. Over time, therefore, reducing the fiscal deficit becomes increasingly important. While a deficit of 3–4 percent of GNP may be acceptable when reform begins, it can soon become a burden. Even 1–2 percent of GNP becomes a drain on the budget and a huge expense in terms of the objectives of countries undertaking reforms.

My talk has focused on the kind of macroeconomic policy needed for successful trade policy reform. What kind of trade policy is needed for macroeconomic stabilization? Can a macroeconomic stabilization and structural adjustment program attain satisfactory growth without trade sector reforms? The answer to this last question is no, for a variety of reasons, the most important having to do with global economic interdependence.

In today’s world, the benefits of this interdependence are great—so great that no economy can afford to be cut off from the rest of the world. Given this fact, a structural adjustment program cannot be credible unless trade issues are addressed. If they are not, people will realize that the adjustment program will fail, or they may think that trade reform will come later. However, if they think trade reform will come later, people will simply wait for it, and the same kinds of uncertainties exporters often have about the sustainability of reform will develop.

Occasionally macroeconomic policy uses the exchange rate as a tool, and more than one country has made a conscious effort to bring down the inflation rate by fixing the exchange rate so that it keeps foreign prices constant and puts pressure on the domestic price level. But while these efforts may bring some pressure to bear on prices, they do so too slowly. The result is an exchange rate that becomes increasingly unrealistic. Any gains that have been made on the trade side diminish, balance of payments difficulties multiply, and soon everyone realizes that the exchange rate will change. Expectations are skewed, and, if anything, neither trade nor macroeconomic targets are met. Under no circumstances is it a wise policy to use the exchange rate as an anchor for the price level when domestic, fiscal, and monetary policies are not appropriate.

As long as people know that an exchange rate adjustment is imminent or that the trade regime must be opened up, they will wait. The bottom line on trade adjustment and macroeconomic policy reform is that they must go hand in hand. Without trade reform, macroeconomic policy will not be credible, and without macroeconomic policy, trade reform is not credible.

There are a range of issues I have not touched on, including communications and transport. However, the essentials on the trade side are as I described: initially, the emphasis must be on expanding exports, then on cutting the incentives for import substitution, and then on gradually unifying the entire exchange rate regime.


David Burton

My talk will deal with external liberalization and macroeconomic policy in the context of Eastern Europe. I hope it will illustrate some of the points that Anne Krueger discussed and perhaps also raise one or two new angles on those issues.

External liberalization in Eastern Europe is particularly interesting because it was a very dramatic reform carried out almost overnight. Several of the phases that Ms. Krueger talked about were basically collapsed into one round of liberalization as part of the more general transition to a market economy.

The starting point for the East European countries was central planning, and the coordination of imports and exports was part of the planning process. Firms generally did not trade directly with the outside world; rather, they traded through foreign trade organizations at heavily distorted prices. Trade with other centrally planned economies took place through the Council for Mutual Economic Assistance (CMEA), which was essentially a system of bilateral barter. Within the CMEA, relative prices diverged substantially from world levels. In particular, the East European countries could import energy and other raw materials at low relative prices. CMEA trade resulted in strong but rather arbitrary dependence among these economies and the Soviet Union and certainly was not based on comparative advantage at world prices.

In lumping the East European countries together, I am running the risk of suggesting that their initial conditions and policy strategies were all the same, when in fact they were not. Their initial conditions differed in a number of respects. Some of these countries—particularly Hungary and Poland—had undertaken partial reforms prior to the recent bold reforms of the early 1990s. In addition, the East European economies entered into the 1990 reforms with macroeconomic imbalances of different sizes. Poland, for example, was close to hyperinflation in late 1989, and the actual or incipient imbalances were also large in Bulgaria and Romania. By contrast, the macroeconomic imbalances were fairly small in Hungary and Czechoslovakia.

Before turning to external liberalization, I want to mention the overall policy strategy in many East European countries, the so-called “big bang” approach. “Big bang” is something of a misnomer. What this approach really involved was implementing some reforms immediately, notably external sector and price liberalization, together with financial policy tightening. Other reforms—tax reform, privatization, and the restructuring of banking systems, for example—by their very nature could be completed only gradually, over a much longer period.

Some economists favored a more gradual approach to the liberalization of prices and the external sector, arguing that markets in these economies were characterized by powerful domestic monopolies and therefore would not allocate resources effectively. But even though some markets may not have worked well initially, the general feeling among reformers in Eastern Europe was that the existing markets were preferable to the old administrative mechanisms. Also, as Ms. Krueger suggested, a country that adopts a gradualist approach to reform is likely to run into sequencing problems. Liberalization in just a few economic areas can create serious problems in others; these problems are difficult to anticipate and can easily make the situation worse.

Another reason gradualism was rejected was purely political. Reformers saw that the window of opportunity that would allow them to sweep away all the mechanisms of central planning might soon close and were anxious to act while there was time.

What about external sector liberalization? As I mentioned, these reforms were bold in all the East European economies, although there were some differences. Importantly, nearly all the quantitative restrictions on imports and exports were eliminated immediately, and state monopolies on trade were abolished. The tariff regimes put in place as part of the reform programs generally established fairly low average tariff rates and reasonably uniform structures. In some countries, import surcharges on goods were introduced (often on consumer goods only), partly in order to raise revenue. I will come back to this issue when I talk about the coordination of structural reforms.

Trade policy within the CMEA continued in its old form during 1990, so that in Hungary and Poland, liberalized trade with the West took place alongside regulated trade with CMEA countries. CMEA trade was reformed radically only in 1991.

To get an accurate picture of the extent to which a country’s external current account has been liberalized, it is necessary to look at the exchange system as well as at trade restrictions. In Eastern Europe, exchange systems were liberalized and a high degree of current account convertibility was introduced at the outset. Further, all the arguments Ms. Krueger made in favor of liberalizing the current account are valid for these countries, and, in my view, two are especially pertinent. The first is that in economies undergoing radical transformations, it is particularly important to get the structure of relative prices right in the beginning in order to guide the process of restructuring industry. The second is that liberalizing the external current account is an effective way of curbing the power of domestic monopolies in the tradable goods sector.

Not all economists agreed that a rapid move to international relative prices was advisable. Some argued for the introduction of a temporary tariff structure that would give industries threatened by the changes in the terms of trade time to adjust. A preannounced schedule for removing these temporary tariffs would provide an appropriate relative price structure to guide investment decisions. However, as Ms. Krueger said, the success of this sort of preannounced tariff reduction strategy depends very much on its credibility. In Eastern Europe, the new governments had little or no experience with the kind of economic challenges they faced. Given the circumstances, it seems unlikely that economic agents would have had faith in such a schedule. They were more likely to simply postpone making investment decisions (or perhaps to make the wrong ones during the transition).

The East European governments faced difficult issues in deciding how much to liberalize their capital accounts. As Ms. Krueger mentioned, capital account liberalization has a number of advantages. One is that it allows individuals and firms to hold any quantity and mix of domestic and foreign assets. Paradoxically, removing capital account restrictions can also encourage inflows of foreign capital. A liberal capital account regime, by providing evidence of a government’s commitment to reform and convincing economic agents that capital brought into the country can also be taken out, may encourage direct foreign investment and repatriation of capital by domestic residents. In this connection, I would note that imposing capital account controls when they cannot be enforced may provide a country with the worst of both worlds. Outflows will not be prevented, but inflows will be discouraged. Finally, capital account convertibility also imposes a discipline on financial policies that has some advantages.

At the same time, there may be potential costs to introducing full capital account convertibility early on. It can certainly make a new government vulnerable to speculative attacks, the danger of which may be considerable if the government’s credibility is not yet fully established. A related argument is that when foreign exchange is in short supply, as it certainly was in these economies at the outset, capital outflows resulting from risk hedging or portfolio diversification can in turn cause high real interest rates, with potentially adverse implications for growth. In Eastern Europe, there was also concern that the domestic financial systems were not well enough developed to withstand foreign competition.

The policymakers in these countries decided that, on balance, the risks of full capital account liberalization outweighed the benefits, and some capital account restrictions were maintained. These restrictions were supported by repatriation or surrender requirements for foreign exchange earnings. I would note, however, that in at least one case—Bulgaria—the capital account restrictions were not strictly enforced, and something close to full capital account liberalization took place. In other countries, such as Poland, where financial policies were restrained, the restrictions appear not to have been particularly binding, judging from the behavior of the parallel market exchange rate, which stayed very close to the official rate.

The points Ms. Krueger made about the coordination of macroeconomic stabilization with external liberalization also hold for Eastern Europe. The stabilization programs initially targeted the approximate fiscal balance, along with credit and monetary policies consistent with a rapid reduction in inflation. The experience has been mixed thus far. Several countries have made considerable progress in controlling inflation (the Czech Republic, the Slovak Republic, Hungary, and Poland), but in others inflation has remained quite high (Bulgaria and Romania). Most countries, however, have encountered difficulties in containing the fiscal deficit. Pressures on expenditures have mounted as the costs of reconstruction grow—including payments to a growing number of unemployed—and tax revenues from state enterprises are declining faster than other sources of revenue can be developed. These problems have highlighted the importance of moving quickly with fiscal reforms such as the introduction of broad-based tax systems and the rationalization of social security systems to ensure a targeted social safety net.

When a country undertakes trade liberalization, it is important to get the exchange rate reasonably right at the outset. There were particular problems in setting rates in Eastern Europe, for two reasons. First, most of the countries were believed to have a liquidity overhang (repressed inflation), because in the past adjustments to administered prices had not kept pace with monetary expansion. However, the demand for money and the precise size of the overhangs were hard to judge, creating uncertainty about what would happen to the price level when prices were liberalized. Second, with the structural changes taking place, including external liberalization, it was difficult to predict the level at which the real exchange rate was likely to emerge.

The East European economies approached these problems in different ways. Poland and Czechoslovakia decided to peg their nominal exchange rates, Poland against the dollar, Czechoslovakia against a basket of currencies. But because of prevailing uncertainties, both governments decided to err on the side of depreciation in order to avoid overvaluing their currencies. An overvalued currency would have resulted in a stagnating export sector, and probably in a balance of payments crisis that, in turn, could have forced the governments to abandon the peg, undermining the credibility of the entire stabilization and reform process. The result of this overdepreciation was that while the external sector strengthened dramatically in both countries, inflation was initially higher than had originally been hoped. Both Bulgaria and Romania, by contrast, initially adopted floating exchange rate systems, in part because the level of foreign reserves in these countries was considered too low to support a pegged exchange rate.

In the end, however, it is important to remember that trade is a two-way street. While the East European countries have undertaken bold liberalization measures, the West must reciprocate if these countries are to realize the full benefits of trade liberalization. Some progress has been made through Association Agreements with the European Union, but barriers to exports from Eastern Europe still remain. It is encouraging that a decision has been reached to accelerate the removal of these barriers. Of course, completion of the Uruguay Round would help all countries, including Eastern Europe.1

It is difficult to make an overall evaluation of the rapid external liberalization that has taken place in Eastern Europe because of the difficulty of separating its effects from those of other far-reaching reforms undertaken at the same time. There have certainly been much larger output losses from the reform programs as a whole than were originally expected, but they are due in part to the collapse in CMEA trade. (The growth of private sector activity is probably also not fully captured by available statistics.) Some of these countries are now showing signs of recovery, particularly Poland, where the private sector is leading the way. Most East European countries have been successful under their new, more liberal trade regimes in generating exports to Western markets to replace exports lost as a result of the collapse of the CMEA. The challenge now is to stay the course, persevering with supporting reforms in other areas and maintaining financial stability.

Ichiro Otani

About 20 years ago, when I was still a young economist at the IMF, I went to Sri Lanka as a mission member. At that time, Sri Lanka was a model of the inwardly oriented economy that Anne Krueger mentioned. It was very closed; controls were everywhere; growth was anemic; inflation was high; and corruption was endemic. The IMF staff tried to negotiate a program for a stand-by arrangement, but we failed because the kind of policies the authorities were pursuing then were not compatible with the IMF’s view of what was needed.

Over the intervening years, I thought little about Sri Lanka until I visited the country again in May 1993. What I found was amazing: many modern hotels, new office buildings, heavy traffic, well-stocked shops, and busy restaurants. I would like to explain what has been done there, because it is so close to what Ms. Krueger has described as the best way of setting about liberalizing external markets. Since the focus of this session is on liberalization and the macroeconomic policy mix, I will also focus on these issues.

First, as regards exchange rate policy, the Sri Lankan authorities devalued the currency significantly about five years ago in order to correct the distorted exchange rate. Since then, the authorities have tried to maintain a rate that will protect external competitiveness. Thus, they have attempted to control monetary expansion in order to keep domestic inflation from accelerating. Such an exchange rate devaluation, followed by tight monetary policy, is an important element of policy coordination, as Ms. Krueger pointed out.

Second, the authorities have liberalized the trade system. Tariff rates have been lowered to reduce distortions so that more industries can compete effectively. Of course, this liberalization has had implications for other parts of the economy as well. When tariff rates are reduced in countries like Sri Lanka, which rely on customs duties for tax revenue, the governments often suffer revenue losses. Unless something is done to counteract these losses, the authorities will have a budgetary crisis on their hands that could hinder attempts to establish macroeconomic stability.

Against this background, the Sri Lankan authorities have introduced various tax measures to offset possible revenue losses from trade reform. In practice, there have been many constraints on introducing new tax measures and collecting additional revenues. As a result, the new measures have been introduced gradually. Tariff reform accompanied by new revenue measures has become a major factor in policy coordination.

Third, quantitative restrictions on imports have gradually been reduced, increasing potential demand for imports. Subsequently, the authorities have tried to tighten monetary and fiscal policy by stressing the need for coordination between structural reform in the external sector and financial policy.

I hope these experiences in Sri Lanka have demonstrated the importance of coordinating structural reform and macroeco-nomic policy.

Summary of Discussion

Most of those participating in the discussion tended to endorse Anne Krueger’s analysis and conclusions, although a number of participants supported some form of protection. Some argued that continued agricultural protection may be necessary in certain African countries. Because productivity levels for key crops such as rice are often many times lower in African countries than they are in more competitive economies, removing import barriers is likely to lead to a flood of rice imports and could cause serious damage to local economies. Ms. Krueger disagreed, maintaining that the low producer prices set by monopolistic marketing boards are more frequently the cause of African farmers’ problems. While the farmers’ low incomes can be attributed in part to low productivity, price incentives are poor, and there is no evidence to show that protectionist approaches alone will improve agricultural productivity. Sri Lanka was offered as an example of a country where successful trade reform helped the country achieve self-sufficiency in the production of rice.

Supporters of continued protection for consumer goods pointed to China, which, it was claimed, had retreated from its earlier policy of liberalizing prices on certain consumer goods. Ms. Krueger argued that when a country—for example, Korea—is concerned about excessive imports of luxury goods, the best approach is to impose high excises or similar taxes on both imports and domestic production of such goods. There is no merit in merely taxing or restricting imports, since such taxes only encourage local production of the same goods.

Some participants expressed concern about the problem of replacing the fiscal revenues lost when import tariffs are lowered or removed. Ms. Krueger did not see the lost revenues as a major problem, pointing to the many new and improved tax systems that have been put in place as part of reform efforts. These systems have raised more revenues than import tariffs ever did.

On a related point, participants expressed concern about fiscal revenue losses associated with enterprise reform in Eastern Europe. Some participants wondered whether this problem argues in favor of slowing the reforms in order to defend revenue levels. David Burton maintained that the reforms should not be slowed, noting that tax reform and cutbacks in unnecessary government expenditures are important accompaniments to structural reform and can compensate for some of the losses.

Participants expressed differing opinions on the question of trade reform in Eastern Europe. Some argued that the situation of the East European countries is unique, since these economies started their reforms with no functioning trade system and therefore have not had much choice about the pace of liberalization. Other participants disputed this analysis, observing that because trade reform had begun in some East European countries before the dismantling of the Council for Mutual Economic Assistance, gradualist options had been available, at least in principle. David Burton pointed out that these options had not been particularly attractive, largely because of the complex sequencing questions that would have emerged had countries such as Poland opted for only partial reforms.

The problems of using the exchange rate as an anchor for inflation were brought up, with participants expressing concern about the difficulty of determining an appropriate level for the rate. The alternative—aggressive exchange rate depreciation to defend competitiveness—often means that exchange rate changes provide an impetus for inflation. In response, it was noted that reforming countries need to achieve a reasonable amount of control over their domestic monetary and fiscal policies. Economies that succeed in gaining a modicum of control over financial aggregates will not need large depreciations, which could lead to large current account surpluses. Countries with large surpluses run the risk of seeing their limited savings used to finance investment in other countries. If moderate inflation continues despite tightened monetary and fiscal policies, a carefully managed crawling peg can be used to stymie speculators.

Finally, Egypt was mentioned as a country that seems to have followed all the rules for successful trade liberalization but that has seen disappointing results in both the real economy and the export sector. Ms. Krueger suggested that the Egyptian authorities need to examine any remaining restrictions that might be discouraging exporters. In addition, there may have been water in the tariffs originally, in which case the original reductions may not have lowered the rates as much as the authorities had hoped. Possibly, some controls also remain that are difficult to discern and that provide too much incentive for domestic production, lowering the relative returns for exporting.


The Political Economy of Agricultural Pricing Policy, edited by Anne O. Krueger, Maurice W. Schiff, and Alberto Valdes (Baltimore: Johns Hopkins University Press for the World Bank, 1991), 5 vols.


The Uruguay Round Agreement was signed in Marrakesh in March 1994.

Proceedings of the Seminar Coordination of Structural Reform and Macroeconomic Stabilization, Washington, D.C., June 17-26, 1993