Exchange Rate Policies in Economies in Transition 10
- Richard Bart, and Chorng-Huey Wong
- Published Date:
- June 1994
Exchange rates, which are central to good economic performance, have much to do with credibility and are vastly underrated as economic instruments that can have a pervasive influence on macroeconomic stability, static competitiveness, and the dynamics of an economy’s evolution. Inappropriate exchange rates will come under severe pressure and, ultimately, will have to change, possibly resulting in a political and economic crisis. Because of their importance, exchange rates are a central topic in economic management. Unfortunately, there is a solid body of prejudice concerning them, even within the IMF; everybody has a different idea about what they should be. Many of these prejudices have been tested and eliminated, so that the only way to make additional headway on the subject is through discussion, not by using a textbook.
The first point I want to make is that no country’s situation is unique. Many economists maintain that the same rules apply to all countries and that therefore broad generalizations are the only way to proceed. (Of course, another school of thought, to which Tip O’Neil would have belonged, maintains that everything is in the details.) With exchange rate issues, it is necessary to determine whether a country’s situation is so unique that no other experience is relevant, or whether one country’s experience resembles that of any other in the principal points and therefore can be subject to the same conclusions.
In the context of transition, which is my more narrow topic, this question is particularly important, because it is tempting to suggest that command economies, with their 99 percent state ownership, are unique and that what characterizes Mexico, for example, cannot have any relevance to Belarus. Such thinking is unproductive, because it overlooks the many lessons that can be learned from years of exchange rate failures and successes.
I. A Historical Comparison
In making a case for using other countries’ experiences as a starting point in examining specific exchange rate issues, I would like to spend a moment discussing the end of the Austro-Hungarian empire. The collapse of the empire is more than distantly related to the collapse of the former Soviet Union—in fact, I would argue that the two events were very similar. When the last Austro-Hungarian emperor, Charles I, fell from power in 1918, many independent republics rapidly emerged from the previous empire, very much under the influence of U.S. President Woodrow Wilson’s declaration in support of nationalities. From one day to the next, new countries were created: Poland was one, Czechoslovakia another. Many territories were ceded to Italy, Poland, Yugoslavia, and Romania, until finally nothing remained of the former Austrian empire except Vienna, which was left with the public debt and the bureaucrats. In everybody’s pocket, on the day of separation, were Austrian crowns, because they were the money of yesterday and a replacement currency could not be created overnight.
Of course, one of the first priorities of an independent country is to print a lot of its own money to pay the bills. In this case, the result was that Austria and Hungary had hyperinflation, and Czechoslovakia did not. It is interesting to learn why not all of the new countries experienced hyperinflation and to follow the Austrian crowns through the transition from a single empire to nation-states. Czechoslovakia tried for a short time to run a joint central bank with Austria. It was agreed that the Vienna-located central bank would not print money to finance the Austrian Government and that a bank commissioner would supervise all activities. Predictably, the bank did nothing other than finance the Government by printing money 24 hours a day, and the bank’s board met when the Czech commissioner was out of town. Obviously, this situation did not last very long: after three months, the strict Czech Finance Minister, Alois Rasin, closed the Czech borders for a week, had all the currency the residents held stamped, and required residents to surrender 50 percent as a “wealth tax” that was expected, by reducing the amount of money in circulation, to push prices back to their prewar level. Of course, not everyone was totally honest; despite heavy penalties, unstamped notes found their way to Vienna and helped fuel the hyperinflation there. The value of the stamped Austrian crowns, which had become Czech money, immediately rose, and the crowns were soon commanding a high premium in Zurich. For months, stamping occurred in every region, but the efforts were not coordinated: new monies emerged in parallel with the stamped crowns, and hyperinflation developed that would last for three years.
It took around four years for hyperinflation to take hold in Austria; Russia has had only two years. So while Russia is well positioned to achieve hyperinflation, it is not yet there. In the end, how was stabilization achieved in Austria? It had been politically impossible to balance the budget, because food subsidies could not be removed, but keeping them in order to maintain social peace also helped fuel hyperinflation. Finally, the League of Nations intervened with a substantial stabilization loan—5 percent of GNP—with some conditions. The Parliament had to pass a law giving the Government the right to do anything necessary to balance the budget over the next two years. In addition, the League of Nations required an independent central bank, a fixed exchange rate, and a League of Nations commissioner in Vienna. The commissioner would submit monthly reports on the situation, and, if things did not go well, the loan would be suspended. Even before these arrangements were made public, the exchange rate stopped deteriorating. Price stability came within a week, and prosperity within two.
Here is a story with all the earmarks of something that could be happening today between the Ukraine and Russia: competitive money creation, the stamping of monies, the groping for a central bank arrangement, the attempt to balance the budget—all of these things are very much the same, because of a number of common factors. The first is that the transition from an empire to a democracy invariably involves very unstable politics. Everybody is in agreement about throwing out those in charge, be they emperors or Communists. The moment those in power are toppled, the broad coalition breaks up, and suddenly everybody has a separate agenda. Because there is a very active democracy, balancing the budget is the one thing that is impossible, and so huge budget deficits develop. Without external financing and a domestic capital market, huge budget deficits mean money creation—and that soon means inflation and, ultimately, hyperinflation.
The second common factor is nationalism. The Austro-Hungarian empire had fiercely suppressed nationalism. The entire senior civil service was Austrian; Austrian or Hungarian had to be learned in school, even in areas where Czech was the mother tongue. The moment the republics became independent, their single preoccupation was to turn their backs on everything in the past, imposing tariffs and quotas and creating their own money. Of course, the integrated trade arrangements that had existed before became obsolete immediately, and that loss had a terrible effect on output and trade patterns.
Another factor involves the diminishing interest of the West. As it became accustomed to famines in Austria, the West lost interest. The Allied Supreme Council as much as blamed the problem solely on Austria’s budget deficits. The Austrians reacted in a somewhat extreme manner: the Jesuit Chancellor of Austria traveled to Italy to ask the Italians to annex his country. The neighboring countries were deeply disturbed by this development, fearing that if the Italians annexed Austria, Hungary and Czechoslovakia would be next, so a group of countries asked the Allied Supreme Council for assistance for Austria. The equivalent of this episode today is the Ukraine’s reluctance to give up its nuclear missiles, provoking the West to react with greater involvement.
This example supports my case that in terms of exchange rates, parallels can be drawn from many earlier experiences and brought to bear on current situations. At some stage, however, it is necessary to decide which of the details that could influence potential outcomes ought to be given special consideration.
II. Why Are Exchange Rates Important?
One of the striking features of transition economies is that all relative prices seem too low. How can this problem be remedied? Perhaps the best method is to use free trade and an exchange rate system to create, in a stable, predictable, and transparent fashion, a pricing system that will accurately reflect world prices. A country very badly needs the right set of prices; otherwise, whatever is positive and useful in the country is underpriced and underused. It is important to utilize assets and resources constructively. The fact that the structure of prices can be so wrong makes a strong case for paying attention to the exchange rate as one of the primary ingredients in improving a country’s economic structure. I think the exchange rate is even more relevant in a transition economy than it would be in basically capitalistic economies that are closed, very closely managed, or mismanaged and in need of modernizing.
For example, Mexico before Presidents Miguel De La Madrid and Carlos Salinas was a closed economy with high tariffs and quotas on top of the tariffs. Licenses were often required, and they were difficult to get. As a result, Mexican prices differed greatly from U.S. and world prices. In addition, a very rich Government was living off oil, keeping low-income people happy through a system of subsidies. In that economy, there was still a connection to the rest of the world, because although domestic capital goods were extremely expensive, import licenses were available, and profits were an issue, even in some state enterprises. In this description, does Mexico differ qualitatively from, for example, Ukraine today? True, Mexico had far more connection with the rest of the world in its pricing system than Ukraine currently has in its: Mexico may have been closed, but its people could see enough of the rest of the world to understand which types of goods should be smuggled. But both Mexico and Ukraine demonstrate that the only way to have the right price structure is through a relatively open trade system and a well-focused exchange rate system. This is one of the arguments for a special role for the exchange rate system.
A second argument is economic organization. In a command economy, or an economy that is somewhere between command and the market, it is important to have trade with and direct investment from the rest of the world. Luring such investment requires a well-focused exchange rate system. Economic modernization proceeds best if, rather than being legislated, it is nurtured by cross-border fertilization and allowed to spread to the rest of the economy. Mexico again provides an example. After six years of a relatively open economy, modernization is spreading rapidly, in large part because of external support and contacts with the rest of the world. If a country’s exchange rate system functions poorly, such interaction will eventually stop. This idea, incidentally, is nineteenth-century thinking about the advantages of trade for economic development, and I highlight here that a sensible exchange rate system is an essential ingredient to enjoying this modernizing influence free of charge.
A third argument involves linkages. The former Soviet Union was heavily integrated across borders with other command economies, but it has now defined a new trading pattern. We are seeing that the exchange rate systems in this region are too poorly managed to accommodate the new trade. Either there is exchange control, no exchange rate system, or a very unstable exchange rate. These problems do not help the new trading system, which recognizes the advantages of trading with the West but does not see maintaining trade with the former empire as taboo. I take that to be a strong case for a European Payments Union-style arrangement, something I will discuss later.
Lastly, there is the issue of corruption. In certain countries, it is extremely easy to manipulate a foreign exchange system in order to win a special rate. This behavior has an extraordinarily negative influence on the economy. It is also true that because the exchange rate is so visible, a single price can become more important than the price of bread. It can become a highly politicized price, and if there is no professional competence to act as a counterweight, the exchange rate—and, ultimately, the economy—may become a victim of politics. Ignorance is also often very powerfully at work when it comes to exchange rates. For instance, some may believe that imports have to be cheap so people can buy them. But the truth is that if people cannot afford imports, they ought not to buy imports—and that truth is not popular.
As I have tried to show, there are a number of considerations that make exchange rates a particularly important topic for the transition economies, especially in light of the new and relatively widespread interest in these economies. There has always been an exchange rate, but it was an unrealistic one; now it is necessary to move to some level of realism. Of course, there are black market rates, which may be two, three, or even four times the official rate. In the former Soviet Union, for instance, one ruble to the dollar was the price of foreign exchange for state enterprises, which bought foreign exchange and foreign goods at this rate despite any economic considerations. The absence of a link between the exchange rate and economic considerations in the minds of economic agents highlights an important factor in thinking about exchange rates: economic transactors must recognize the exchange rate’s effects on economic decisions.
III. Choosing the Right Regime
What are the primary considerations in searching for an exchange rate regime? There are really three. The first is the degree of convertibility. Is there to be strict exchange control, current account convertibility, or full convertibility? The second is the relationship between the central bank and the foreign exchange market. At one extreme, we might have the equivalent of a gold standard; at the other, a complete separation between the two, although the central bank would have a powerful influence. The third consideration is the relationship between the trade and exchange rate regimes, which must be considered together. A trade regime may be liberal, for instance, but its potentially positive effects may be suppressed by the exchange rate regime. It does not make any difference where a tariff is collected, for instance. An explicit 50 percent tariff on certain imports and a special exchange rate for certain imports are equivalent tariffs in a protective structure. And multiple exchange rates simply do in the foreign exchange market what is not done for one reason or another—perhaps administrative—in the port.
What are the possible exchange rate arrangements? Exchange rate regimes are generally classified according to whether or not the rate moves. But I would suggest that the very best monetary arrangement for a transition economy is the adoption of another currency, such as the deutsche mark. The deutsche mark is a solid currency—probably the most solid—followed by the Dutch guilder, which would be an equally good choice. It is a travesty to believe that an underdeveloped country without any management experience, financial expertise, or real background in the world ought to create money. Yet the countries least able to manage money are those that most want their own. Even during hyperinflation, these countries are proud of their money and will not admit that it is a disgrace. But, by printing it, they may be financing a deficit—another good reason not to have a national currency.
In Western Europe, the race is on to abolish national monies, because some countries, including Italy and France, have had trouble with theirs. Again, the deutsche mark is the most stable currency, and other countries would like to emulate its stability. The experience of a number of countries suggests that a country with no appropriate institutional or political setting should not suddenly adopt its own variant of the ruble and pay a lot of money to a firm in Canada to get it printed. As I have said, hyperinflation will be the result. To those people tempted to put forward a national currency as a way to symbolically honor national heroes, I would say that the national heroes can be put elsewhere—on monuments, for instance. In reality, in these situations a national money is just an extravagance.
What about currency boards? Currency boards are a way to demonstrate that the national currency is sound, through 100 percent backing for the existing stock or at least at the margin. There is a big difference between these two situations, because starting at the margin with a large existing stock means that the slightest run can wipe out available reserves. This type of backing is really nothing more than a marginal gold standard. But 100 percent backing for the entire money supply is the equivalent of being pegged to the deutsche mark or the dollar until further notice. “Further notice” is that time when the need for 100 percent backing is called into question—and, starting from 100 percent, the need can be questioned all the way down to zero. Somewhere on the way, it will be convenient to start printing money.
Argentina, in the aftermath of many hyperinflationary episodes, has a monetary arrangement of 100 percent marginal and average backing. But even with that arrangement—one to one against the dollar—they are having a run on their currency. How is that possible? It is possible because people do not believe that Argentina really has a currency board. The country has had 45 percent inflation in the last 18 months, despite the dollar backing and fixed rate. Nobody believes the currency will hold, because for 100 years Argentina has had a currency crisis just before Christmas, and the crisis is occurring again. But currency boards are nonetheless attractive, because they foster credibility in the exchange rate system.
With a currency board, it is important to choose the right currency for the peg. If a country closely tied to countries in the European monetary system (EMS) pegs to the U.S. dollar, for instance, the fluctuations of the dollar relative to Europe—up to 50 percent—could result in extraordinary real instability. So the choice for a country like Estonia, with its newly founded currency board, really is between the deutsche mark or a basket. The argument for a basket is that the currencies are weighted against each other—dollars, lira, and so on—but again, my view is that one currency is preferable. With a single peg, people can say their currency is as good as (for example) the deutsche mark; further, when something happens in the EMS, an extraneous problem will not suddenly develop. Credibility is also an issue here: people want to be able to bite into a coin and feel that it is real. In essence, then, a currency board ought to be a very thin national veneer above an underlying reputable currency. Anything else will become a problem for financial markets—including a basket, which can become inconvenient, because the standard is discretionary. In this regard, Estonia was very right to go to a deutsche mark peg, and if inflation develops and the currency board attempts to move to a basket, people should criticize the currency board.
On my shopping list for transition economies, I go further, to fixed exchange rates. If a currency board has 100 percent marginal reserves, foreign exchange that flows in through balance of payments surpluses is bought up by the central bank, which then issues one-for-one local currency. But when foreign exchange flows out, it is as if there were a gold standard: the money supply is reduced, there is a credit squeeze, and interest rates go through the roof. Funneling foreign exchange through a central bank at a fixed rate is an automatic mechanism for stabilization. No discretion is involved; one person can run the central bank and oversee foreign exchange operations. A fixed exchange rate provides the critical link that gives the rate credibility—namely, the fact that the money supply is geared to foreign exchange movements. It is clear that the currency is being managed according to a strict criterion. If the fixed exchange rate becomes a problem, interest rates can be raised to defend it; if it is not defended, a crisis and devaluation ensue. Fixed exchange rates are noninstitutionalized currency arrangements, useful at the beginning, but they soon wear out if the central bank is not following appropriate policies and the institutional setting is weak. The bank will continue to hand out money, and the fixed exchange rate will collapse because of the lack of institutional support. Currency boards, on the other hand, have a lot of institutional support.
Multiple exchange rates
A country may feel that it needs a number of economic policy instruments to meet its objectives. For instance, a poor country may want to keep food prices low and make luxuries very expensive. On the export side, because primary commodity industries do not respond to exchange rates on account of the long gestation lags for their investments, a country may choose to give these industries a bad rate. At the same time, it may want to give nontraditional industries, which have no lags at all, much better exchange rates, because these industries are viewed as the entrepreneurs that build a modern economy. This kind of thinking can result in up to 50 different exchange rates.
Multiple rates encourage bureaucracy and corruption: the civil service administers the multiple exchange rate system, opening itself to bribes. It is essential to come out against this sort of system immediately, because even two exchange rates may be too many. Multiple rate systems result from the conjunction of people with a feeling for the structure of industry and politicians who have a feeling for what is necessary for the next election. When these two types of people meet, the result is a really awful exchange rate system. So a system of multiple exchange rates is to be avoided, even during a transition. One exchange rate should suffice, because its essence is that, at the margin, the costs and benefits of foreign exchange are equalized.
Subsidies and the tax system can be used to offer industries special treatment, but a country going this route will soon find out just how much it costs. The foreign exchange market should not be used to obscure priorities, and governments should not make the mistake of thinking that just because something has a gestation lag, it ought to have a bad exchange rate. In five years, an industry offered a poor rate will disappear, because, with such an unfavorable rate, businesses have no choice but to take their equipment out of the country and sell it.
Dual exchange rates
If multiple exchange rates are out, how does a country gain some leeway in its exchange system? One way is to have two rates: a fixed rate for current account transactions and a flexible rate for the capital account. The fixed rate for current account transactions provides some macroeconomic stability for countries wishing to avoid a wildly fluctuating rate that upsets the price level, then wages, social peace, the budget, the money supply, and again the exchange rate. An anchor is necessary, and countries that are comfortable intellectually or politically with controlling the money supply can use the exchange rate on the current account.
Of course, such an anchor can create a problem with capital flows. Capital flows begin the very day the system begins to open. Someone has a cousin in Chicago, for instance, who wants to send a thousand dollars into the country—a transaction that is certainly not going to be made through the central bank. Capital flows exist, and the only question is how to manage them. One way is by channeling them all through the central bank at a fixed exchange rate on the way out, but this method is unpopular. The alternative is to create a market system for them, divorcing the central bank from capital account transactions and allowing the banking system and people on the street corners to deal freely in foreign exchange. At the same time, the central bank can maintain a fixed rate for approved commercial transactions. Any volatility in policies that leads to volatility in capital flows will then affect the flexible capital account rate as if it were a black market rate, leaving the price level unaffected and reducing the risk of macroeconomic instability.
While there is a lot to this argument, there are two problems with it. The first is that some sort of exchange control is clearly necessary in order to license the people receiving foreign exchange for import transactions. However, the moment a good rate is established for import transactions, people will start inventing transactions, importing used equipment described as “new,” for example, or rare drugs that turn out to be aspirin. There is also the business of import inspection, which requires that people be employed to check boxes to ascertain whether the imports are actually there. So dual rates are an attractive concept, but they require a bureaucracy that inspects effectively, and if a country is institutionally weak or democracy has reduced the bureaucracy too much, they are probably the wrong answer.
The second argument against dual rates is that they require isolating the price level from the capital account. The following example illustrates the problem: if the capital account rate has a 100 percent premium over the fixed rate for current account transactions, every exporter will want the capital account rate. The exporters then “under-invoice” their exports and pay what they save into a Swiss bank account. The value of Latin American exports to the United States in Latin American statistics is 30 percent less than the value of U.S. imports from Latin America in U.S. statistics—a discrepancy that reflects export invoicing. But the moment exports are under-invoiced to take advantage of the depreciated rate for capital account transactions, a feedback is created to the domestic price level, and the prices of domestic exportables at home rise. Then, because people have the alternative of getting these goods from outside the country, imports increase and pressure on prices is reduced.
One rule for dual exchange rate systems, then, is that the rate for capital account transactions must diverge greatly from the current account rate. It can fluctuate, but there has to be a limit. What would the limit be? Above 20 percent, it is impossible to stop commercial transactions from getting the capital account rate, but 15 percent is probably an even better rule. Either the current account rate or monetary policy needs to be flexible enough to keep the rates in line. Mexico had a dual exchange rate system from 1982 until around 1987 and very carefully watched the divergences. As a result, there are few horror stories about that experience with a dual system. If there is little divergence, however, there may not be enough leeway. A rate that can move a bit may serve as a shock absorber; if a 10 percent depreciation of the capital account rate will not affect the price level, this flexibility is an advantage. But again, a whole bureaucracy is required to run such a system, and in an institutionally weak setting, this need for oversight is a very strong reason not to choose this path.
The alternatives are to have one fixed rate, to allow total flexibility, or to “forget about” the capital account. This last is an attempt to avoid the convertibility question by wishing away capital flows. The prejudice here is the dichotomy “current account transactions are good, capital account transactions are bad.” I think this distinction between current and capital accounts is overdone, however. The capital account does not have to do only with speculation and making money, and the current account does not have to do only with food, exports, and other “good” things. Everybody recognizes that unless foreign companies can remit their profits, investment will be low. If businesses have to keep extensive records and account for every penny, an unfriendly environment develops. And if firms have to argue with bureaucrats about whether certain remittances of profits are capital outflows or not, the environment becomes hostile. My suggestion would be, in the absence of an effective bureaucracy, to have unified rates and convertibility on both current and capital accounts.
A country may argue that by adding extra personnel, it can manage a multiple rate system. This argument does not hold up, however, because the costs are too high, as the country following this route will soon discover. The bureaucrat who controls foreign exchange has absolute power over it, and in the end, the businesspeople needing licenses will have to bribe someone. It is more efficient to forget about distinguishing between the transactions, to establish full convertibility, and to learn to live with it. In a market-oriented economy with a democratic, accountable government, many things are possible. But for a country just coming out of a repressive experience, the worst thing has to be perpetuating that repression at the foreign exchange desk, where people know only too well how to stop others from doing good business.
Fixed and flexible rates; crawling pegs
Once multiple or dual rates are eliminated, the choice is between fixed and flexible rates. Fixed rates are a problem, because inflation makes them unsustainable. Interest rates go up to defend them, balance sheet problems arise, and recessions develop, because these rates are often not defensible. In the end, there may be a collapse that is invariably blamed on speculators and “foreign professors.” On the other hand, flexible rates fluctuate, creating macroeconomic and political problems. A better option is a crawling peg, which can be very powerful. The crawling peg is designed to adjust the exchange rate when domestic inflation exceeds international inflation by depreciating domestic price levels in a way that does not invite speculation. The depreciation should take place at a steady pace, in increments so small that nobody is willing to speculate and everybody understands what is going on. Interest rates will then reflect the anticipated pace of depreciation, and competitiveness will not be lost. Brazil has used a crawling peg with extraordinary success, and since that experience the crawl has become routine in countries that do not want to fight inflation at the expense of competitiveness. A crawling peg is the right answer when a country wants to avoid having its exchange rate misused in the fight against inflation. When the exchange rate is misused, a barrier of protection often develops, and the rate becomes misaligned.
The crawling peg is not popular with central bankers. Central bankers—the new generation—are all hard money men. They want a firm anchor to fight inflation and therefore advocate a completely fixed exchange rate. Their type of thinking has influenced Europe for the last two years, and what has happened since October 1992 is the complete collapse of the arrangement they advocated. Exchange rates have become overvalued either because countries have too much inflation—as in Spain, Italy, and Great Britain—or because their competitiveness has suffered on account of a loss of markets, a terms of trade shock, or a domestic financial crisis. Instead of allowing a substantial depreciation, these countries held on to their fixed rates, maintaining that the markets would ultimately come to believe in these rates. A crawling peg could be a strong answer to inflation problems of this type, which are not going to be resolved soon and which can affect a country’s competitiveness. However, the crawl does not actually solve the inflation problem; rather, it validates the inflation and tells the foreign exchange market that the exchange rate cannot bring inflation down. Incomes policy, monetary policy, or both must be used to reduce inflation, but not the foreign exchange market. Countries with strong and sustained economic performance and significant inflation always have a crawling peg. Chile has used one with extraordinary success in the last ten years, ever since an episode of brutal overvaluation with a terrible collapse in 1982. Chile, incidentally, is currently the only Latin American country with Asian-style growth.
If fixed rates are a problem in inflationary economies, it is possible to use totally flexible rates. The primary concern with totally flexible rates is the absence of a nominal anchor. Monetary policy is a potential nominal anchor, but in an economy that is just developing a financial structure, the demand for money is unknown, and the money supply cannot be closely controlled. As a result, the exchange rate can fluctuate widely, especially if it is affected by a capital account driven by expectations about monetary policy, and if the central bank does not really know where things are headed. For these reasons, flexible exchange rates are difficult to use properly in a rapidly changing economy. For example, during the financial innovations of the 1980s in the United States, the dollar went up by 50 percent because people were watching the various measures of monetary aggregates, which were reflecting primarily those financial innovations. This kind of movement is often an issue in a transition economy, where the entire economic structure is evolving and predicting money demand is almost impossible. Even in industrialized countries, money demand equations are often difficult to use well, although econometricians have tried for 20 years to make them work. Finding a money demand equation in a transition economy is a heroic undertaking, because the data do not exist. For this reason, I am skeptical of flexible exchange rates in these situations.
Again, a crawling peg is probably the right answer, because it sets up the exchange rate as the single priority to preserve competitiveness. I see competitiveness, rather than financial stability, as the most important function of an exchange rate, more so in a transition economy than in an industrial one, because competitiveness is an essential ingredient of modernization, reform, and an open economy.
A crawling peg exchange rate system has two variants: a preannounced crawl (perhaps tied to the daily rate) and a crawl geared to the difference between the domestic and external inflation rates. The preannounced crawl contains inflation at the margin, because the preset exchange rate path allows for some inflation but does not automatically accommodate full inflation. I think this system is a good one for a country that has a predictable inflation rate and that wants some nonaccommodation. After first using a crawling peg, Mexico moved to preannounced mini-devaluations, which at 2 percent turned out to be too small, given the 15 percent inflation rate. The rate dropped to 7 percent, which was tolerable for a while, but such a rate ultimately has serious consequences, as Mexico found out. Over the last few years, Mexico’s wholesale prices have increased 40 percent in dollar terms, compared with the United States, where wholesale prices have increased by 5-6 percent. The result has been a sharp drop in competitiveness, a gigantic trade deficit, a loss of confidence, and soaring interest rates. What will happen next? There will be a speculative attack—maybe not tomorrow, but someday. Mexico needs to accelerate its crawl in order to improve competitiveness.
I mention Mexico because it is an economy that has undergone extensive modernization and runs a budget surplus. In fact, Mexico has managed to get everything right except the exchange rate. And the faulty exchange rate now means that interest rates are 16-17 percent in real terms, that the budget is being further tightened to cut the trade deficit and defend the exchange rate, and that economic growth is 2 percent less than population growth. The next two years—1993 and 1994—will be the same, if the crisis has not occurred by then. Even in exceptionally well-managed economies, then, the exchange rate can go wrong in the end, as it did in Chile in 1979, at the tail end of modernization and an “economic miracle.” There was a large overvaluation, a terrible collapse, and then five years of up to 30 percent unemployment. Since that episode, Chile has never again misused its exchange rate the way Mexico is misusing its. It is simply not enough to do everything right if the exchange rate is not in order, because the exchange rate is a central link once capital mobility has been established—and capital mobility always happens, particularly if the exchange rate is in the wrong place. If it is, it can eliminate accumulated successes and cause macroeconomic instability with high interest rates and low growth. And in the end, external events will be the determining factor. The vote on Maastricht broke Italy, and the U.S. Congress’s vote on the North American Free Trade Agreement could adversely affect Mexico. Certainly, Mexico does not want an overvalued rate as that vote takes place.
Again, if a country lacks stability and cannot predict its money supply, it cannot manage a flexible rate. It can try a fixed rate, but a fixed rate will not adjust for inflation. In such a case, a country should let the rate crawl, although without preannouncements, which tend to encourage mismanagement. Brazil used to depreciate by the difference between the local consumer price index and foreign wholesale prices, a differential to which a margin was added to provide an extra 1–2 percent real depreciation every year. This system was extraordinarily good for competitiveness, allowing the country to develop a substantial trade sector as well as a rule that was followed long enough to become virtually tamper-proof. The rest of the system bore the entire burden of living with indexation. However, a complete indexation in the foreign exchange market puts pressure on the budget, the central bank, and incomes policy to manage inflation. As currency prices in Europe show, managing inflation from the foreign exchange market is very successful for three years. But by the fourth year, it results in huge bills.
IV. A Payments Union?
Whatever Estonia does will have relatively little effect on the world. But what Russia does has implications for all the states of the former Soviet Union and for the East European countries. If each of these economies adopts a different system of exchange control, there is a real possibility that trade will contract. Each country will be defensive about allocating foreign exchange for imports, because there are so few exports. The current trade collapse in the former Soviet Union is in part the result of foreign exchange controls. Of course, there is also a strong dose of nationalism involved. The outside observer—the IMF, for example—must realize that full convertibility with a crawling peg may not happen. Countries undergoing difficult restructuring are unlikely to have clean foreign exchange markets, and sooner or later foreign exchange is going to have to be administered. Most exchange rates have a tendency to be overvalued, so foreign exchange is always scarce, trade is repressed, and production falls. This scenario developed in the former Soviet Union and in East European countries that belonged to the Council of Mutual Economic Assistance (CMEA). Trade with Russia has fallen by a factor of 20–30 percent or more, and all the former CMEA members are paying for it. The region’s disorganized exchange rate systems are not the only reason for this collapse, but they are one significant factor.
What is to be done? The answer lies in something like the European Payments Union (EPU), which was organized by the United States in 1950 as part of the Marshall Plan in an effort to create an economically prosperous region that would act as an antidote to communism. In 1947, trade had come to a virtual standstill among the European countries. Everyone wanted dollars, because dollars would buy food, raw materials, machines—things countries could not buy with other currencies. Countries defended the few exports they had in order to turn them into dollars. The EPU was developed as a type of a clearing union that would allow the European countries to trade with one another without using dollars. Bilateral clearing was the first round, but this system proved too primitive, and trade surpluses developed. A surplus with one country could be used to pay a third, so multilateral clearing was the next logical step. The U.S. authorities organized the clearing system so that it could function without forcing countries to offset exports with imports daily: accounts were cleared monthly, and each country could draw on an interest-bearing balance in the EPU. Even countries like Switzerland, which had a fully convertible currency, were members, because they wanted to trade with countries that lacked hard currencies. Within ten years, Europe had moved to full convertibility and full liberalization of trade, and the EPU ended in 1958.
CMEA countries need this type of system to encourage the kind of trade among themselves that is not now taking place, because these countries are so set on earning dollars and, as a result, are ignoring some other potentially useful trade. Because there are no hard monies, a payments union is essential as a way of settling balances without using rubles, coupons, or the like. The issue is not just Poland or the Czech Republic—certainly not the Czech Republic, which is financially stable, has a convertible currency and substantially free trade, and is virtually merging with southern Germany. The problem is the countries that are not making it—and will not be for ten years—but that do have interesting trade that could be beneficial to all parties. I conclude that an important part of thinking about exchange rates is considering a payments system that can function even when currencies are mismanaged (as some of those in the transition economies are likely to be) and that will not destroy trade. The EPU is a perfect economic model and a good political model, because an EPU-style arrangement for the countries of the former Soviet Union could dampen many of the enthusiasms that are slowly pushing the countries of that region closer to civil war.
Thomas A. Wolf
In mid-1992, many observers were advocating a fixed exchange rate for Russia, in line with the other elements of the stabilization program that the Government and Central Bank had committed themselves to with the support of the first credit tranche arrangement with the IMF. Since there is so much support for either a fixed exchange rate or a crawling peg for most transitional economies, why did the Russian authorities ultimately not select one of these two regimes?
The first—and perhaps most salient—reason is that the country had only negligible reserves, an important consideration in this kind of situation. Second, limits existed on the extent to which other countries felt themselves able, due to their own budgetary problems, to provide direct balance of payments financing to help support a fixed exchange rate. The reluctance of these governments to provide balance of payments support above and beyond what was already being provided may also have been due to perceptions that Russia was highly unstable, both politically and economically. The lack of a coordinated monetary policy within the ruble area meant, simply, that the Russian authorities did not have full control over their money supply or inflation rate. And there were certainly doubts about the authorities’ ability to carry out appropriately tight financial policies and coordinate them adequately between the Government and Central Bank. These perceptions have been borne out somewhat by the experience since August 1992. The idea now is to try to defend the liberalization of the foreign trade and exchange rate systems and not to compromise the reform effort by trying to defend a fixed exchange rate that, as I will try to explain, is at best arbitrary.
It would have been difficult to determine which exchange rate could have been defended in Russia in the second half of 1992. Even with additional resources from abroad, the Russian authorities had very little information or experience to go by, especially in comparison with some of the East European countries. Russia’s interbank market for foreign exchange was quite thin at that time, and the authorities lacked very basic data on the extent of implicit subsidization of foreign trade. Some of the East European countries, on the other hand, had been able to draw on a great deal of information in this area, so they had some sense of the exchange rate equivalent of various explicit or implicit export subsidies, for example. Finally, the Russians had no track record of maintaining solid financial policies, and, as I said before, there were real questions about the Government’s ability to carry out coordinated, tight fiscal and monetary policies.
Russia’s situation differed somewhat from Poland’s at the end of 1989 or the former Czechoslovakia’s at the end of 1990, because major price liberalization had already occurred in Russia at the beginning of 1992. It was generally agreed that Russia no longer had an immense overhang of money, but some key prices were still controlled, especially for energy, and the impact of increases in those prices on the overall price level had not been determined. The impact would have depended on the demand for money and the downward flexibility of other prices, among other things, and there was very little solid evidence on these variables in Russia at that time.
What would have been the appropriate range within which to fix the exchange rate? As others have said, the precise level at which the exchange rate is set is probably not terribly important in the transitional economies. The objective should be to get the rate in roughly the right range, and it is key that the currency not start out overvalued. In Poland in late 1989, for example, despite a substantial track record of currency auctions and a data series that permitted calculations of the exchange rate equivalent of export subsidies, it was extremely difficult to determine an appropriate exchange rate. In other words, there was little agreement on the precise rate that would provide a trade balance strong enough to help build up reserves and, at the same time, prevent unnecessary inflation and a decline in real wages (which might not have been socially acceptable). Expert estimates for the correct zloty rate varied anywhere from about zl 6,500 per dollar to about zl 13,000 per dollar—a considerable range. The rate that was finally selected was around zl 9,500 per dollar. Even in this situation, therefore, the uncertainties were tremendous, but they pale in comparison with the circumstances in Russia.
The purchasing power parity-based exchange rate mentioned for the Russian ruble in the second half of 1992 was in the range of rub 20 to rub 30 per dollar. The current rate—rub 418 per dollar (appreciated somewhat from the 450 that Rüdiger Dornbusch mentioned)—is far from that rate. In mid-1992, not only were there numerous purchasing power parity calculations, there was a very recent record of strong Central Bank intervention in the interbank market. As a result, the exchange rate in that market had appreciated in nominal terms to around rub 130 to the dollar—five to six times the purchasing power parity rate. There were also calculations based on an assumed dollar wage in Russia—once stabilization had more or less succeeded—similar to the average dollar wage in Eastern Europe in the year or two following macrostabilization. Estimating inflation and nominal wage growth in Russia under a stabilization program, it was possible effectively to solve for the exchange rate that would provide the dollar-equivalent wage (the calculations suggested an exchange rate even more depreciated than rub 130 to the dollar). These examples illustrate the very broad range within which the monetary authorities had to work.
It is important not to focus on purchasing power parity, however, especially during a period of transition when relative prices, production, and consumption are undergoing tremendous changes and inflation is high. Purchasing power parity calculations in such circumstances do not mean very much, except perhaps for tourists or businesspeople making everyday expenditures, and do not tell us which short- or even medium-term exchange rate would clear the foreign exchange market. Why might this be particularly true in Russia, and why is it important to be so careful about assuming that the appropriate exchange rate should be close to some purchasing power parity rate? A classic reason is that a fairly predictable relationship exists between official exchange rates and purchasing power parity rates, and for a country at a relatively low level of development, the differential between these two rates will be considerable.1 This logic suggests that if Russia had already achieved some degree of stabilization, the market exchange rate would have been considerably more depreciated than the purchasing power parity rate.
With a few exceptions, Russia had almost completely liberalized imports in early 1992. In theory, exports had also been liberalized, but the major exportables (oil, gas, many precious metals, other metals, and timber, among others) were still subject to export quotas. It was an unusual situation with liberalized imports but not liberalized exports. Holding everything else constant, then, a much more depreciated market clearing exchange rate could have been expected than what a purchasing power parity calculation suggested, simply because exports were artificially constrained. Furthermore, it is generally recognized that at least with respect to most manufactured products, especially finished manufactures, many Russian producers are simply not competitive in world markets. This fact suggests that a purchasing power parity calculation—which typically would be based primarily on food or a household consumption basket—may not contain many of these potentially tradable manufactured products and may actually understate the true purchasing power parity rate (or at least the rate relevant to foreign trade). The same kind of problem exists with imports. As many Russians know, many of their imports are really complementary imports, or nonsubstitutes. There is a tremendous pent-up demand for many products that simply are not produced in Russia or, if they are, do not meet the standards of quality found on world markets. Once again, the implication is that an adjusted purchasing power parity calculation would provide a rate somewhat more depreciated than one based on a market basket heavily loaded with fresh vegetables and various consumer staples.
Finally, those exportables that make up the lion’s share of Russian exports to the world—namely energy products—must be taken into account. Here the Russians have been faced with a continuing decline in production not entirely unrelated to economic policy. In the very short run, however, the drop in petroleum output and the stagnation—even slight decline—in natural gas production is, most experts agree, relatively unaffected by policies. Over the medium run, with the right investments, pricing policies, and other factors, a tremendous scope exists for expansion of output and exports. But in the short run, production and exports—because of inadequate policies to curb consumption—have been falling.
Taken together, these factors make clear the difficulty of using anything close to a purchasing power parity rate for Russia as a fixed exchange rate that could be defended with very scarce financial resources (a low level of reserves or money borrowed, or perhaps given, by the rest of the world). And what should that exchange rate have been? No one knows, and in this case there were so many uncertainties and questions about the stance of economic policy and the ability of the authorities—regardless of their commitments—to carry through these policies that it would have been a mistake to try to fix the exchange rate in mid-1992. Indeed, experience with economic policy in the second half of 1992 suggests that the correct decision was not to try to peg the rate at that time.
Since we really do not have very solid evidence that one exchange rate system is better than another, a number of circumstances need to be taken into account in deciding which system is preferable for a given country. Thus, I would like to organize my comments around four sets of issues: one dealing with convertibility, the second with the role of the exchange rate in planned versus transition economies, the third with the experience of countries in Central and Eastern Europe, and the last with the interpretation of various indexes of real effective exchange rates.
I very much agree with the approach to convertibility taken in the paper suggested for background reading.1 Convertibility has become a slogan for many in Central and Eastern Europe. However, it is more important to determine exactly what external liberalization contributes to the process of reform than simply to repeat the slogan. First, external liberalization of trade and trade-related services contributes to internal reform: it corrects relative prices, for which world prices provide the best guide, and promotes competition, especially in countries where extensive domestic monopolies prevail. Second, liberalization of foreign direct investment and related transactions, such as the repatriation of capital and profits, is important to the promotion of growth. Although opinions here differ, to me these are the areas in which early liberalization is important; it is less urgent in other areas.
The Role of the Exchange Rate in Planned and Transition Economies
The exchange rate played a very small role in planned economies. The reason is simple: the exchange rate is a price, and prices did not matter, or, if they did, their effect in influencing behavior was typically blunted by the absence of hard budget constraints. It is not possible to affect the consumption of a particular item by raising its price tenfold and then compensating consumers after the fact for the increased cost. Various studies of the influence of the exchange rate on Hungarian exports make this fact clear. No statistical evidence that the exchange rate had had a positive influence on exports was found until the availability of “soft money” (subsidies, credit, and ruble exports) was considered concurrently. After this factor was taken into account, the exchange rate was found to have had the expected positive influence on exports. Exchange rate policy, therefore, plays a crucial role in the reforming or transition economies, because in these economies, prices matter.
Exchange Regimes in Central and Eastern Europe
An important difference among the stabilization and reform programs adopted by the Central and Eastern European countries lay in exchange rate policy. In Czechoslovakia, as in Poland, the exchange rate was fixed after large initial devaluation. Bulgaria and Romania chose floating regimes. After a relatively small initial devaluation, Hungary followed a policy of adjusting the rate flexibly in light of balance of payments developments.
This eclectic collection of pragmatic approaches reflects the considerations particular to each case. From a macroeconomic perspective, a fixed rate can be particularly helpful as a nominal anchor when other policy tools are hard to use or interpret, as monetary policy is in the economies in transition. A fixed exchange rate can provide a simple yardstick for measuring the stance of financial policies and setting interest rates. Most of the economies in transition are still some way from being able to conduct monetary policy efficiently with indirect instruments and continue to rely on direct credit limits and interest rate guidelines. Even with these, however, monetary policy remains a much less satisfactory tool than it is in more developed market economies. Changes in velocity as these economies undergo transformation make reliance on monetary targeting as the sole or primary policy tool difficult. The financial sector reforms under way mean that money demand will likely remain unstable for some time. In these circumstances, policy must be judged in light of developments in the variables that it is seeking to control—inflation and the balance of payments or pressures on the exchange rate—as well as by the intermediate tools of credit and money growth.
From a microeconomic perspective, the fixed exchange rates in Poland and Czechoslovakia provided a more stable bridge for the importation of world prices, the realignment of relative prices, and the stabilization of inflationary expectations. But the commitment to the fixed rate had to be credible, requiring an adequate level of reserves—or at least a level of reserves perceived as adequate. While reserves were quite low in Czechoslovakia, the early commitment of financial assistance from the IMF, the European Community, and the Group of 24 probably bolstered perceptions of the authorities’ ability to defend the rate.
However, fixing the exchange rate was not considered a feasible option in Bulgaria and Romania, given the lack of reserves, or in Hungary, given the vulnerability of the balance of payments to fluctuations in the capital account. In both Bulgaria and Romania, the acute shortage of foreign exchange contributed to sharper-than-expected depreciations and increased inflation. Although exchange rate fluctuations have often been held responsible for exacerbating economic instability, I hold that these variations were less causes than symptoms of continuing perceptions of political and economic instability and of prolonged delays in mobilizing the targeted financial assistance. In these circumstances, flexible rates made possible a program of external liberalization—which might have had to be abandoned or which might have occasioned a more severe drop in output had other objectives been subjugated to a firm commitment to a fixed rate.
Finally, the adjustable peg regime worked well in Hungary, which chose to manage the exchange rate flexibly during the initial stages of its more gradual reform in order to safeguard an external position vulnerable to shifts in private capital flows. With limited endogenous responses of capital to movements in interest rates, the country would have had to offset a large negative external shock primarily through the current account. In such a case, adjusting the exchange rate is likely to elicit a faster response than tightening financial policies would. As its balance of payments position has strengthened and prices have stabilized, Hungary has gradually moved to harden its exchange rate policy in order to reinforce the gains against inflation.
The Hungarian policy of gradually adjusting the exchange rate has not, however, been without cost. Gradual price liberalization and exchange rate adjustment have contributed to the entrenchment of inflationary expectations. None of the economic agents—including those that watched economic developments closely—believed that the country’s accelerating inflation would be reversed. Although the inflation rate fell to about 15 percent annually during the second half of 1991, inflationary expectations remained much higher for an extended period, together with corresponding expectations of further exchange rate depreciations. The result was continuing high nominal interest rates and rapidly rising real interest rates that depressed economic activity for at least a year after inflation rates started falling. Interest rates are only now dropping—and rather slowly, especially on the credit side.
Thus, initial conditions were an important consideration in the choice of exchange regimes in the reform and stabilization programs introduced by the various countries of Central and Eastern Europe. Subsequently—as is the case everywhere else—the sustainability of fixed rate regimes and the stability of flexible rates have reflected, to a significant extent, the degree to which underlying financial policies are consistent with maintaining a stable exchange rate. In Poland, slippages of financial policies that eroded the country’s competitiveness eventually led the authorities to abandon their fixed exchange rate policy and implement a crawling peg system. In contrast, despite its initial and surprising volatility, the Bulgarian lev subsequently achieved a significant measure of stability for an extended period.
Interpreting Real Exchange Rate Movements
Real exchange rate indexes are well known. They measure relative inflation performance adjusted for exchange rate changes and are used to indicate whether a country’s competitiveness is improving or lagging. However, these indexes must be very carefully interpreted for economies in transition.
As subsidies are reduced during the reform process and the share of labor remuneration paid in the form of wages increases, we can expect a significant appreciation of the usual indexes of real effective exchange rates that is consistent with maintaining or improving both competitiveness and profitability. This change will perhaps be most immediately obvious in indexes of the real effective exchange rate based on the consumer price index, because eliminating consumer subsidies (or, as part of tax reforms, introducing a value-added tax) will significantly raise inflation measured on the basis of consumer prices, without affecting external competitiveness or profitability. Producer price-based or unit labor cost-based indexes will also be affected, although to a lesser extent, as the reform process can be expected to lead to an increase in the share of wages in total labor remuneration. Simultaneously, reducing taxes on enterprises or other costs that had previously financed the nonwage component of labor remuneration, such as subsidies, services, and vacation houses, would maintain competitiveness or after-tax profitability despite rising wage costs. These shifts simply reflect a narrowing of the well-known difference between “market” and purchasing power parity-based exchange rates.
This process of reform and stabilization is far from over in the transition economies of Central and Eastern Europe, as illustrated by data comparing prices for various consumer goods in individual countries.2 While price and external liberalization, introduced by all countries at the outset of their reform programs, resulted in the expected broad convergence of the prices of tradable goods, substantial differences still remain in the prices of nontradable goods, in particular housing and utilities. As these prices are also adjusted over time, higher inflation rates and wage increases than in partner countries will cause real appreciation of the usual indexes of real exchange rates—without, however, implying losses of competitiveness or profitability, provided that other enterprise costs, including taxes, are reduced accordingly.
Also, over a longer period, gains in productivity (probably larger for tradables due to the central role of external liberalization in the reform process) should permit a further real appreciation without impairing the performance of the balance of payments. Judging when this should happen requires going well beyond the usual in dicators of real effective exchange rates and looking carefully, in particular, at after-tax profitability and the balance of payments.
Martin J. Fetherston
Considerable attention is being paid at the moment to the transition in Eastern Europe and the states of the former Soviet Union. Those of us working in the Asian region need to speak up from time to time and remind our colleagues that Asian transition economies also offer useful and interesting economic lessons. I would like to touch on the exchange rate experiences of four of those economies: China, Viet Nam, Laos, and Mongolia.
What exactly is the meaning of the phrase “transition economies”? This question is important because there are several dimensions to the transition issue. Each of these dimensions may apply in some countries more than in others, and considering them should help to counteract the tendency to have a particular kind of situation in mind when addressing the policy issues that arise in a transition economy. For example, Rüdiger Dornbusch emphasized the problems faced by countries that are establishing new currencies, have not had the experience of managing their own monetary policies, and therefore must start from scratch. When making that point, Mr. Dornbusch is obviously thinking of some of the former Soviet republics. But there are plenty of economies undergoing transition that do not find themselves in this same situation—that, for example, already have their own currencies.
In trying to define a transition economy, I would identify five dimensions that vary considerably from one country to another.
The first is the severity of the initial macroeconomic imbalances the economy faces, including high inflation rates and balance of payments problems. Of the four Asian economies mentioned, initial imbalances were particularly severe in Mongolia, Viet Nam, and Laos, but considerably less so in China.
The second dimension is the length of time that the economy has been centrally planned—or, to put it the other way around, how long it has been since the country’s last experience as a market economy. And, again, in Asia, there is a range of experience: obviously, substantial parts of Viet Nam and Laos were market economies not that long ago—up until the mid-1970s, or about the same length of time as the states of the former Soviet Union.
The third is the extent to which the economy was integrated into the Council of Mutual Economic Assistance (CMEA) trading system, because the extent of that integration determined the severity of the problems the economy faced when the CMEA trading arrangements broke down. And, once again, there is a spectrum among the four Asian economies. Mongolia was the most highly integrated into the CMEA system; Viet Nam and Laos were considerably less so; and China was not a member of the CMEA system at all.
The fourth dimension involves the political conditions and, more specifically, whether the country is undergoing a political as well as an economic transition. In China, there have been considerable economic reforms, but the political system has remained basically unchanged. At the other end of the spectrum, Mongolia underwent substantial political change beginning in 1990 and adopted a new Constitution incorporating a parliamentary democracy.
The fifth and final dimension of the transition process involves the country’s prospects for attracting external financial support for its transition efforts. Once again, there is a broad range of experience among these countries. China has been able to attract significant support, primarily in the form of project assistance (rather than as balance of payments financing). Mongolia had considerable success in mobilizing support from the international community, including the IMF, which has also supported policies in Laos under the structural adjustment facility. An exception is Viet Nam, which has been making its transition without external financial support.
What are the exchange rate policy issues that have come up in these four economies during their transition processes? Like Gérard Bélanger, I will emphasize eclecticism and pragmatism. Despite the strong arguments put forward by Rüdiger Dornbusch and others for some kind of rules-based arrangement, I think that in very broad terms the experience of these four Asian economies supports the desirability of having some degree of flexibility. It certainly does not indicate that a nominal anchor approach is the only one that can be a success.
Both Laos and Viet Nam had rather complicated exchange rate systems until the mid- to late 1980s, with a proliferation of official exchange rates for various purposes. Of course, this type of setup has characterized many of the centrally planned economies. Flexible, unified exchange rate systems were established in Laos in 1987 and in Viet Nam a year or two later. In both countries, the official exchange rate was adjusted frequently, based on movements in the parallel market. A key issue in choosing a flexible approach—particularly for Viet Nam—was the absence of reserves that could be used to defend a fixed rate. In addition, as Thomas Wolf just emphasized, there was great uncertainty about the level at which such a fixed rate should be set.
Some interesting contrasts have emerged between the experiences of Laos and Viet Nam since the new systems were adopted, mainly in the area of inflation. In Laos, the inflation rate was brought down to a fairly low level, and the market exchange rate over the two years or so has been stable, largely because the authorities have adopted a policy of using exchange rate developments to calibrate their financial policies. Thus, if the market exchange rate begins to show signs of depreciation, the decrease is taken as a signal that monetary policies need to be tightened. Until recently, Viet Nam’s experience with inflation was less favorable, and the inflation rate—which had been reduced in 1989—jumped back to around 70 percent in 1991 before it was brought under control. The key factor in these fluctuations was that Viet Nam’s monetary authorities were less successful in maintaining the tight financial policies needed to preserve a reasonable degree of stability in their exchange rate while keeping inflation low. But in terms of economic growth, in fairness it must be said that both countries—particularly Viet Nam—have enjoyed some success. Viet Nam, for example, has recently been growing at a rate close to 8 percent. A very interesting issue for further research would be the reasons why transition economies in some parts of the world—particularly Europe, the former Soviet Union, and Mongolia—have suffered very large declines in output at the start of their transition process, whereas some of the Asian economies—such as Laos, Viet Nam, and China—experienced no output declines at all and in fact have enjoyed rather rapid growth rates.
Mongolia was one country that initially established a fixed exchange rate. The rate was devalued substantially in mid-1991 but almost immediately became unrealistic because of the very low level of international reserves and lack of supporting domestic financial policies. The Mongolian authorities have recently announced that they intend to establish a floating exchange rate system in the next few months.1
Of these four countries, China has the exchange rate system that conforms most closely to the crawling peg approach. During the 1980s, China implemented a number of exchange rate adjustments, making some moderately large discrete adjustments at some points but also using periods of successive mini-devaluations. Since 1986, China has had a dual exchange market that differs from the kind of dual exchange market Mr. Dornbusch was describing, with different rates for current and capital account transactions. China’s second market is for enterprises wanting to trade their quotas for retained foreign exchange, while other transactions—both current and capital—still take place at the official rate.2 One important aspect of the Chinese experience is that the overall results have been impressive. Chinese exports, for example, have increased sevenfold since China initiated its reforms in the late 1970s. Exchange rate policy has played a role in that, but many other factors were also involved.
One concluding observation about the role of supporting policies is in order. If there is one common thread among the diverse experiences in Asia and other transition economies, it is that even if a new exchange rate regime is put in place, it will not work in the end if it is not supported by good financial policies. These policies will always be essential.
Summary of Discussion
Most of the discussion took the form of questions to Rüdiger Dornbusch on his presentation, in particular on his recommendation that the transition economies adopt a payments system similar to the European Payments Union (EPU), which combined multilateral clearing and a credit mechanism with well-defined rules. Mr. Dornbusch indicated that the parallel between the current situation in the transition economies was strikingly similar to the situation in Western Europe after World War II, when countries attempted to engage in U.S. dollar trade at the expense of inter-European trade. He urged the transition economies to consider trade among themselves instead of focusing almost exclusively on trade with the West. And he strongly recommended a payments union that would allow the transition economies to cultivate the interregional trade that he felt ought to be taking place, not only among the very complementary republics of the former Soviet Union (FSU), but also among the East European countries. Moreover, a system similar to the EPU would force countries to settle their growing deficits with other members of the system in U.S. dollars, forcing hard currency discipline on countries that habitually run deficits. In addition to encouraging trade, such an arrangement would create a strong incentive to adjust.
On the question of whether FSU countries should set up currency boards, Mr. Dornbusch stated that although currency boards would work in principle, they would not be long lived. Because they would eventually be asked to raise the ratio of domestic currency issue to foreign exchange receipts, inflation would accelerate, and soon a multiple rate system would develop. Only if the fiscal situation was resolved and the government no longer had to print money to cover budget deficits could a currency board system work. Finally, on the issue of national currencies for each of the countries of the FSU, Mr. Dornbusch was highly skeptical about the possibility of moving to a system of convertible national monies, since an orderly transition did not seem possible.
Case Study of Czechoslovakia
Leslie J. Lipschitz
Recently, Czechoslovakia has embarked on a pegged, fixed-rate exchange rate policy. My talk concerns the choice of this regime and the results of that choice. I will start with some basic, relevant facts, touch on the fears many felt when the Czechoslovak authorities decided on a nominal exchange rate peg, and finish up with how the scenario has played out. It is a story with what looks to be a happy ending.
In Czechoslovakia, 1990 was a year of preparing for the transition to a market economy. For exchange rate policy, preparation involved unifying a number of different rates and then devaluing the rate several times. In January, certain rates were unified and the koruna was devalued; it was devalued again (to Kčs 24 per $1) in mid-October. On December 28, the tourist rate was unified with the principal rate (at about Kčs 28 to the dollar), and the exchange rate was fixed in terms of a basket of five trading partner currencies. Three days later, 85 percent of prices were freed, and most quantitative restrictions on trade were abolished.
Why did the Government choose a peg rather than a float or crawl? Besides the obvious aspect—that a fixed rate involves less foreign exchange risk in trade—there was quite clearly a desire for a transparent nominal anchor in Czechoslovakia. The authorities accepted the idea that a liberal trade regime with a fixed exchange rate would stabilize the prices of traded goods, providing a reference point around which wages and the prices of nontraded goods would organize themselves and price expectations would cohere.
How was the rate chosen? Probably the choice was less an intellectual than a practical decision. In September, with reserves falling, the Government was considering two scenarios. One envisaged an exchange rate of Kčs 24 to the dollar, the other a rate of Kčs 28 to the dollar. Shortly thereafter, the rate was set at Kčs 24 per dollar. After a very short respite, the koruna weakened further in the parallel market. But reserves continued to fall (from a level that was already problematically low), so the rate was adjusted in December. The December devaluation was part of a broader package involving tight credit, a restrictive incomes policy, an austere fiscal stance, and a substantial commitment of funds from the IMF and the Group of 24 (G-24). That package was seen as substantive and credible; the exchange rate held, and it has held ever since.
II. The Risks Involved
There were four major risks involved in setting this rate: (1) getting the rate wrong; (2) finding that domestic imbalances were inconsistent with maintaining the rate; (3) putting an excessive financial squeeze on the economy; and (4) being unable to defend the rate against short-term pressures because of low reserves and, consequently, suffering a loss of credibility.
Getting the rate wrong
It is not necessary to know the long-run equilibrium rate to embark on a fixed-rate policy. With some price flexibility, there is a reasonable continuum of nominal rates that can be considered consistent with an appropriate real rate.
Conventional purchasing power parity (PPP) calculations are irrelevant here, because it takes some time for prices to reflect flow costs (rather than inventories and other past influences). Price indexes also are not terribly meaningful; internationally competitive, cheap final goods are no comfort to unemployed workers priced out of the market by an exchange rate that leaves their wage costs uncompetitive.
Also, the short-run exchange rate consistent with stabilization objectives bears no relationship not only to the PPP but to the long-run equilibrium exchange rate. This, I think, is a fundamental theoretical issue that warrants elaboration. If the long-run (equilibrium) real exchange rate is known with certainty, should an immediate move to this rate be proposed? The answer, obviously, is no. In all the transition economies, there is a huge stock of capital, a whole industrial structure, that has been accumulated under arbitrary relative prices. Knowing that comparative advantage is different from what exists and moving immediately to a new real exchange rate that would be in line with comparative advantage in five years’ time would almost certainly render a large proportion of the existing capital stock and the industrial structure obsolete. The objective of stabilization policy has to be a more gradual structural change.
Thus, in East Germany, where real exchange rates were set to be consistent with a rapid move toward a West German capital structure, it has been necessary effectively to write off much of the starting capital stock and related jobs. This situation is not ideal. The transitional exchange rate cannot be impervious to its influence in the labor market. Rather, what policymakers should aim at is a set of relative prices that put the most egregiously uneconomic enterprises (or branches of enterprises) out of business and encourage new businesses without obliterating much of the industrial base.
In Czechoslovakia, the real exchange rate has been consistent with a relatively gradual transformation. Large enterprises are closing down their weakest operations and shedding labor, while new, small private enterprises are springing up, encouraged by a good competitive position. Vacancies and new private sector jobs, at least in the Czech lands,1 have kept pace with the elimination of old positions.
If the initial devaluation had been excessive, wage settlements in the ensuing year would probably have strained the limits set by incomes policy; in fact, as wages easily stayed below these limits, the prima facie evidence does not appear to support the idea that the devaluation was too large. Of course, real wage increases can be expected over time, consistent with a changing and more productive capital stock. And an endogenous and equilibrating real appreciation over time to an exchange rate more in line with the long-term structure of the economy would not be at all problematic and should not be resisted.
Maintaining the rate
Obviously, a large budget deficit, especially if financed by money creation, can undermine a peg. In the case of Czechoslovakia, however, conditions were propitious. There were no large initial imbalances, and both domestic and foreign debt levels were low. The Government, moreover, was committed to a budget that was close to balance. (In fact the deficit amounted to 2 percent of GDP, excluding proceeds from privatization; including these proceeds, it was balanced.)
“Squeezing” the economy
There was real concern that the combination of a fixed exchange rate and inflation inertia would put too much pressure on the economy. Inflation during 1990 had been 18½ percent, and the devaluations of late 1990 would feed into 1991 prices. The simultaneous terms of trade losses would also raise prices. In addition, there was also some fear that, with price liberalization, monopolistic enterprises would raise their prices sharply (and perhaps repeatedly) to increase profit margins.
Three aspects of policy were seen as guarding against continued high inflation. First, relatively tight credit policies were set. Second, the opening up of trade was seen as curtailing monopoly power. Third, a strict tax-based incomes policy was put in place; it envisaged a fall of some 10 percent in real wages from the end-1990 level.
Keeping up reserves
The danger that reserves would be insufficient to sustain the exchange rate and that a forced devaluation would undermine the credibility of the whole program was of particular concern to then Finance Minister Vaclav Klaus. Reserves were indeed low—below $1 billion, or equivalent to only three weeks of imports. IMF and G-24 financing was seen as a way to alleviate this problem. The IMF provided about $1.5 billion under its stand-by arrangement and compensatory and contingency financing facility; the G-24 pledged an additional $1 billion. The plan was to build reserves to some $2.5 billion by the end of 1991.
III. The Strategy’s Success
The strategy was successful in terms of its objectives. Prices jumped by 50 percent in the first half of 1991 and then stabilized.2 There was no inflation in the third quarter and very little in the fourth. There is evidence that some large enterprises raised their prices disproportionately in the first few months of the year, perhaps basing their behavior on the monopolistic price determination and automatic credit of the old system. But owing to nonaccommodating macroeconomic policies, these prices produced a buildup in inventories of finished goods that was difficult to finance. Prices therefore had to be lowered quickly. The balance of payments proved stronger than projected, and by the end of 1991 reserves had risen to $3.3 billion.
Real GDP dropped, but the reduction was unavoidable, given the huge terms of trade loss and the demise of the Council for Mutual Economic Assistance (CMEA) market. Output bottomed out at the turn of the year. There is likely to be some real overall growth in 1992, although the uncertain political circumstances related to the dissolution of the country may well influence the outcome. There is a burgeoning private sector made up mainly of small companies, especially in the Czech lands. At present, both unemployment and job vacancies in the Czech lands stand at 2.5 percent. Inflation in 1992 should be about 10 percent, and the current account will probably show a surplus of about $500 million.
The future is somewhat obscured by the division of the Czech Republic and Slovakia, which is scheduled for January 1, 1993. In the Czech lands, there is no question about the course of exchange rate policy. I would expect a gradual real appreciation as wage increases are facilitated by a shift in the structure of industry. The appreciation will, of course, hasten the demise of the old industries. In Slovakia, the whole macroeconomic strategy is being reconsidered. Most likely, there will not be a fundamental shift in strategy there, but it is too early to say exactly how the authorities will use the various instruments under their control.
Czechoslovakia’s economy has successfully undergone the first phase of the transformation to a market economy with a stable exchange rate. The country’s external position has improved and is now broadly in equilibrium. The 1991 current account ended in a surplus of $360 million—much better than anticipated—and 1992 is also likely to close with a surplus. Despite the breakup of the Council for Mutual Economic Assistance (CMEA) and an overall decline in foreign trade in 1991, Czechoslovak firms have proved to be very flexible, and exports to convertible currency markets have continued to expand in 1992. A significant decline in imports in 1991 contributed to an acceptable trade balance deficit of $450 million, although the deficit worsened in 1992.
These results were facilitated by several devaluations of the koruna in 1990, which totaled more than 100 percent and allowed the economy to begin benefiting from a stable exchange rate. The last of these devaluations, carried out shortly before the liberalization of most domestic prices on January 1, 1991, contributed to a significant jump in prices that was quickly arrested by strict monetary and fiscal policies. By midyear, monthly inflation was less than 1 percent. The money supply declined overall in real terms in 1991, and the state budget recorded an acceptable deficit of 1.8 percent of GDP at year’s end.
As prices stabilized, costs as a share of wages fell because real average wages declined markedly in 1991. GDP showed a sharp drop of 16 percent for 1991, but only a small part of this decline may have been caused by the exchange rate devaluation; most of it was due to the breakup of Czechoslovakia’s traditional export market, the CMEA. The decline in GDP has slowed, and a recovery could begin in 1993.
Czechoslovakia’s experience shows that a single strong devaluation, followed by medium-term exchange rate stabilization, can be successful. But our experience raises a question: for which economies and situations is our technique suitable? Rüdiger Dornbusch pointed out that Czechoslovakia was the only country among the successor states of Austria-Hungary that was able to avoid hyperinflation after World War I—something it accomplished with a very restrictive monetary and fiscal policy. The tradition of price stability continued in the following years, so that there has always been a tendency in Czechoslovakia to favor external and internal stability and a readiness to make sacrifices to this end.
Because of this tradition, Czechoslovakia was in a better position than most other countries to launch economic reform in 1991, in part because our external and internal debts were relatively small. External debt in convertible currencies stood at $9.4 billion and the Government’s domestic debt at 5.2 percent of GDP in 1991 (domestic debt now stands at 9 percent). A tradition of price stability, a smaller financial burden, a less severe dislocation of capital, and internal political stabilization have worked together in Czechoslovakia to support exchange rate stability. The massive initial devaluation and strict monetary and fiscal policy, supplemented by wage regulation, were the immediate means to this end. Now, we are able to see some benefits—although, since we have had only two years of experience with economic reform, it can be argued that a positive assessment is somewhat premature.
On another topic, Mr. Dornbusch suggested in his talk that the former CMEA countries might want to imitate the system of payments used in Western Europe after World War II. I agree with Gérard Bélanger that substantial differences exist between the present situation in Central Europe and that of postwar Western Europe. While the West European payments model would be very attractive to us, I am not sure that it would be effective. To date, our trade with a number of countries, particularly in the former Soviet Union, has led to growth in nonusable assets, even with the system of government clearing that was in place in 1991. To some extent, then, the resources that we paid for in convertible currencies have been transferred to the former Soviet Union. The current system of trade means not only a loss of resources, but also support for the old structure of production, which is not a viable one in the long run.
Why not provide direct assistance to the states of the former Soviet Union, supporting the development of trade among former CMEA countries? In fact, some banks in the region are already making efforts to establish a small clearing system based on new principles that could (at least in part) help renew mutually advantageous trade. However, returning to a government-led clearing system would be very difficult. Czechoslovakia, a small, open economy, is ready to trade with any business partner able to pay. Recent developments point to the potential for expanding Czechoslovakia’s trade, first with industrial countries and then with some developing economies.
Despite significant success in containing inflation, our consumer price index increased by about 56 percent in 1991, and a 12 percent increase is anticipated in 1992. Reform of the tax system means that relatively high inflation is expected again in 1993. Accordingly, the koruna has appreciated substantially in real effective terms since 1990, and the issue of Czechoslovakia’s export competitiveness is coming to the fore.
While some firms may see profits from exports decline in 1992 compared with 1991, exports are still more profitable overall than domestic sales.
Wage costs are now growing in real terms, but this growth is not likely to offset the wage decrease recorded in 1991. The 1991 devaluation stimulated exports of products with low value added and of commodities that trading partners view as “sensitive.” A further attempt to stimulate exports by devaluing would only exacerbate the problem of protectionism all economies undergoing transformation currently face.
In any case, a devaluation does not seem to be necessary, since the adjustment of the pattern of production and trade to the 1991 devaluation is still far from complete. In addition, there is room for cuts in production costs, as illustrated by the low overall rate of unemployment in Czechoslovakia.
An expected increase in imports of investment goods will likely lead to a slight deficit in the trade balance, but this kind of disequilibrium means rapid export growth in the near future. In any event, it seems that a growing surplus in the services account (i.e., transport and tourism) will finance most of the trade deficit and that the remainder will be more than sufficiently covered by direct inflows of foreign investment, so that some increase in reserves can be expected. This optimistic view has been tempered somewhat in the second half of 1992 by expanding imports and signs of capital flight, which are causing a decline in foreign exchange reserves. Unfavorable expectations have arisen because of the imminent division of Czechoslovakia into two independent states and the upcoming tax reform (including the implementation of a value-added tax). However, when this difficult period ends, we will have good prospects for maintaining our stable nominal exchange rate and benefiting from the advantages it provides.
Case Study of Poland (1990–91)
In some important respects, Poland’s experience differed from Czechoslovakia’s; nonetheless, there were a number of common elements.
I. Reforms before 1990
It is important to remember that Poland had made several attempts at reforming its economy before 1990. In fact, as far back as 1957, there had been attempts to introduce some reforms into the Polish economy. (One of the most famous postwar economists, Oscar Lange, was a Pole celebrated for showing that, in principle, it was possible to find a set of prices within a planned economy that would function like the price system in a competitive market environment.) These early reform efforts, however, were if anything counterproductive and in part explain why Poland faced some particularly serious problems in the years leading up to 1990.
The aim of the reforms—particularly in the 1980s, when things started to go wrong—was to give enterprises greater autonomy. Because it was believed that decentralizing decision making would enable them to react more flexibly to economic changes, enterprises were allowed a certain degree of independence in their own management and finances. But the attitude toward the degree of decentralization that should take place was ambiguous: the authorities really saw the market as facilitating a more efficient realization of the planners’ dreams and did not intend to unleash unchecked free-market forces. As the authorities decentralized, giving enterprises a greater degree of autonomy, one of the side effects was a weakening of existing control mechanisms, because it was no longer possible to order state enterprises to do whatever the Government wished. In other words, corporate governance started to break down.
It is important to emphasize that the enterprises were not being irrational in the way they behaved but were operating according to the rules then in force. They were not facing appropriate market incentives—specifically, no one was looking after the interests of capital, and the Workers’ Councils were exercising considerable influence.
The year leading up to the “big bang,” 1989, was a year of economic crisis in Poland. The authorities lost control of the budget and banking system, and the balance of payments started to deteriorate dramatically. The authorities continued to take steps toward reform. Private citizens were granted access to the parallel foreign exchange market, and as a result, the “black” market became in essence a “white” market. One of the side effects of this development was a gradually accelerating shift out of zlotys into dollar-denominated assets—hardly surprising in the circumstances. In April 1989, as a result of round-table talks between Solidarity and the Government, formal wage indexation was introduced. By August, real wages were up 73 percent from their January level. In August 1989, food prices were liberalized, and the impact was immediate: the monthly inflation rate climbed to 39 percent.
At this point, the Government was having increasing difficulty controlling the fiscal deficit. State enterprises faced soft budget constraints, to which powerful labor interests reacted by increasing wage demands. Against the background of these developments, the ratio of the parallel to the official exchange rate rose in August to 7.4, a very large difference. In late August, the Government took additional steps, introducing a measure of fiscal restraint in an attempt to regain control of the budget. The indexation coefficient on wages was cut in order to break the wage-inflation cycle, and measures were introduced to tighten credit. While inflation remained high, the measures had some beneficial impact: even before the reform program came into effect on January 1, 1990, the ratio of the parallel to the official exchange rate had declined significantly, so that the flight out of zloty-denominated deposits into dollar-denominated deposits slowed. Real wages were adjusted down because of the change in the indexation formulae, but they rose again in December. It was clear something further had to be done.
II. The 1990 Reforms
It is important to evaluate the 1990 reforms as a package—that is, as a broad range of measures covering a variety of issues—and not to focus solely on the exchange rate. An exchange rate rule on its own would not have achieved much; what had to be addressed were the underlying macroeconomic imbalances and structural flaws in the system. The 1990 program had that objective. It was meant to curtail sharply the fiscal deficit and, by introducing a tax-based incomes policy, to control the behavior of wages. The issue of corporate governance was addressed by initiatives concerning ownership transformation. Of these initiatives, mass privatization received the greatest public attention, but it must be emphasized (particularly since mass privatization has yet to occur in Poland) that many Polish enterprises followed more individual routes to privatization—and that these enterprises have been responsible for many of the changes since 1990.
When the 1990 program came into effect, it had immediate effects. What happened in that first year? Instead of the planned fiscal deficit of about zero in 1990, Poland ended up with a fiscal surplus equivalent to 4 percent of GDP. This fiscal adjustment was far greater than anticipated. Moreover, the balance of payments in 1990 was far stronger than anyone had expected: net international reserves increased by about $4.4 billion. Real wages plunged initially, because the tax-based incomes policy allowed for a predetermined increase in nominal wages in January and February, based on an inflation projection that turned out to be a serious underestimate. Accordingly, the wage increases that were granted in no way compensated for the large burst of inflation, which reached 78.6 percent in January alone. The change in the price level occurred primarily in the first two weeks of January and took many people by surprise.
The strength of the balance of payments, the surprising fiscal surplus, and the decline in real wages were, of course, interrelated. Maintaining a fixed exchange rate and damping down demand relative to supply—even if output is falling—is likely to result in a strong balance of payments. One of the reasons for the fiscal surplus that has continued to cause trouble was the enormous increase in the nominal value of inventories, which were taxed for the first time in 1990. As a result, revenue from the enterprise sector showed a large transitory increase that did not reflect real values. Rather, it reflected the impact of inflation and contributed to the long-term weakening of the enterprises’s balance sheets. One of Poland’s problems even now is that the economy is doing well in many areas, but the tax system continues to rely to a significant degree on the shrinking financial base of the state enterprise system. It has been very difficult to effectively extend the tax base, although important steps have been taken, including approval of legislation mandating a value-added tax (VAT).
On the other hand, there has been added pressure on expenditures, albeit for very legitimate reasons—to support the social safety net, for example, because unemployment has increased rapidly. The result has been something like a dual economy, in that the public sector is struggling to make ends meet as it adjusts (although it is adjusting rapidly) and the private sector is growing rapidly but managing to escape most of the tax net. In Poland’s economy, an estimated 58 percent of total employment is now in the private sector—a remarkable transformation, despite the fact that agriculture, which was always considered to be in the private sector, employed a large proportion of the population. Even in the industrial sector, where state enterprises were once pre-eminent, close to 30 percent of enterprises are now considered to be private.
As is well known, there was a sharp decline in output in 1990 and another drop in 1991, when Poland felt the impact of the breakup of the Council for Mutual Economic Assistance (CMEA). But since about November 1991, output has begun to turn around, and the Polish economy has had a fairly positive growth rate for a number of months—one of the best pieces of news from any of the Central European economies. In year-on-year terms, Poland is now looking at positive growth rates.
III. The Exchange Rate
Even before pegging occurred on January 1, 1990, the authorities had already started devaluing the official rate. Nonetheless, the January zloty rate of 9,500 per dollar represented a large devaluation. Some felt the rate was overly depreciated, but others argued for a strong initial devaluation. The comments Leslie Lipschitz was making earlier apply here, too: there certainly was the sense that if the authorities didn’t go far enough, there would be trouble. If the authorities had introduced the peg—which was intended to be a cornerstone of the new program—without devaluing sufficiently and had run into balance of payments problems shortly afterward, the credibility of both the peg and the program would have been seriously damaged. It was important to give the peg a chance to work, and it would do no harm to give the export industries an initial stimulus to get them going, considering the changes to which the economy was being subjected.
Prices did increase sharply in January 1990, and discussion continues on what might have been the cause. One explanation that has been suggested in the context of other transforming economies is that a monetary overhang existed at the outset. According to this argument, such overhangs existed because of repressed inflation under the former centrally planned regimes, with individuals accumulating large liquid balances in the absence of anything to spend money on. In Poland’s case, this theory may not hold, for two reasons. First, Poland had significant inflation before January 1, 1990, so that any large monetary overhang would have been reduced in value in the months leading up to that date. Second, “dollarization” had occurred before 1990—that is, money was moving out of zlotys into dollar-denominated assets. Accordingly, to the extent that there was an overhang, it may not have been in zloty-denominated assets. Although people could not purchase all the goods and services they wanted in the marketplace, dollar-denominated assets provided an alternative outlet that was used quite extensively.
Now, with liberalization, people with money to spend have tended to switch from dollar-denominated assets into goods. In fact there is some evidence to suggest that this shift occurred in Poland in 1990, strengthening the zloty. In the first three months of 1990, deposits denominated in foreign currency began declining again. This behavior is a fairly accurate bellwether of the sort of confidence the population at large had in the program the Polish Government was implementing. In fact, the steady drift out of dollar-denominated assets into zloty-denominated assets has continued to this day, so that the Polish economy is now really a zloty economy.
There may be no totally adequate explanation for the burst of inflation at the beginning of 1990. Probably it was spurred in part by money coming back into the country from dollar-denominated assets, and in part by the initial overdevaluation of the exchange rate. As for the fixed exchange rate itself, the choice was relatively clear. The earlier discussion concerned Czechoslovakia’s potential to break inflationary expectations—something that was, if anything, a higher priority in Poland. The view was that a fixed exchange rate would help achieve this and, in the absence of better ways to implement corporate governance, would also discipline state enterprises. And the exchange rate was not the only nominal anchor in the system: the wage policy, carried out through tax-based incomes policy, served the same function. This second anchor worked, by some accounts, extremely well at the outset of the program, as real wages declined. In fact, it worked too well, and a correction took place in the middle of 1990 to compensate for the unexpected plunge in real wages earlier in the year.
The basic rationale for the fixed exchange rate has not changed. Initially, the fixed rate was expected to last about three months, but it survived from January 1, 1990 until May 1991. By May 1991, the real effective exchange rate had more or less converged with the real effective rate that had existed prior to the January 1990 program. Subsequently, the zloty was devalued twice. A preannounced crawling peg of about 1.8 percent per month has replaced the fixed exchange rate (this crawl is also fixed, with one time derivative). For many months in 1990, inflation remained high, averaging about 5 percent per month. It has been coming down since but has proved to be relatively obstinate, so the tactic that has been selected to combat it is the crawling peg. Presumably the rate of crawl will be adjusted in the future as the anti-inflation program improves, and eventually the economy will tend toward a fixed exchange rate. But the crawling peg still acts as an anchor for the system. Some may claim that it provides a floor for inflation, but at least it encourages some inflationary discipline.
The task for monetary and exchange rate policy in Poland’s stabilization program was to create a stable monetary environment for decision making and to reduce deep-seated inflationary expectations. Meeting these objectives was not easy. In 1988 and 1989, the zloty began falling into disuse as foreign currency became the favored means of transaction and people sought to hold their wealth in either goods or foreign exchange. By the end of 1989, the inflation rate had soared to 260 percent. After several years of high and accelerating inflation, inflationary expectations were deeply ingrained, and the credibility of both official institutions and government commitments was low.
Under these circumstances, the authorities decided that the short-term goal for monetary policy would be to establish convertibility for the zloty at a fixed nominal exchange rate in order to anchor price expectations. Compensation for phasing out the remaining indirect and direct measures for export promotion and the margin necessary to maintain export competitiveness were the basic considerations in determining the new rate. On January 1, 1990, the official and parallel exchange rates were unified and the new rate fixed at Zl 9,500 per dollar (at the end of 1989, it had been Zl 6,500). Inflation remained high in the first quarter of 1990.
The fixed exchange rate was regarded as a significant factor in Poland’s macroeconomic stabilization. Initially, it was decided to fix the zloty against the dollar on a weekly basis according to movements in cross rates, allowing for fluctuations against other currencies (the exchange rate has been determined daily since the beginning of 1991). The peg against the dollar was considered the best way to discount the psychological effects of the stable exchange rate, since about 50 percent of Polish exports and imports in convertible currencies were denominated in dollars. The U.S. currency was also the one most commonly used in convertible currency settlements between individuals.
The Government originally intended to maintain the stable exchange rate only during the early months of 1990, but actual market pressure on the zloty was substantially weaker than had been feared. The $1 billion Polish stabilization fund that Western countries—notably the United States and Germany—had established to support convertibility was not needed, but it did allow for the possibility of stabilizing the center rate within a narrow band of the official rate, and official reserves grew throughout 1990. Under these conditions, it was possible to keep the zloty fixed much longer than had been planned—17 months, in fact. The initial exchange rate, which appeared to be undervalued, helped to soften the effects on enterprises of the decreased domestic demand that resulted from the stabilization measures. The foreign exchange reserves accumulated in 1990 (gross official reserves increased by $2.4 billion) helped to absorb the shock of the collapse of the Council for Mutual Economic Assistance (CMEA) and the transition to convertible currency settlements among the former members. The pressure on foreign exchange reserves associated with the real appreciation of the zloty became visible in the last quarter of 1990, increasing in the first months of 1991 as convertible currency settlements became more common among the former CMEA countries and the U.S. dollar appreciated.
On May 17,1991, the zloty was devalued to ZI 11,199 per dollar. In order to prevent the fluctuation characteristic of currencies pegged to a single currency when international foreign exchange markets are unstable, it was decided to return to the pre-1990 policy of setting the value of the zloty against a weighted basket of currencies. The five currencies used most widely in Polish foreign trade are represented in this basket: the dollar (45 percent), the deutsche mark (35 percent), the pound sterling (10 percent), the French franc (5 percent), and the Swiss franc (5 percent).
The inflation differentials among Poland and its main trading partners, which were still significant in 1991–92, weakened the competitiveness of Polish products. These differentials were the main reason for the change of exchange rate regime that took place on October 14, 1991, when the weighted basket was introduced and the crawling peg adopted. The value of the basket was increased by 9 zlotys per business day, or the equivalent of a monthly depreciation of around 1.8 percent. Because this pace of devaluation did not compensate for the growth in domestic prices, the exchange rate still functioned as an instrument of stabilization.
A higher-than-expected real appreciation of the zloty between October 1991 and February 1992 resulted in a drop in the competitiveness of Polish goods, both at home and abroad, causing foreign exchange reserves to fall. Against this background, a step devaluation was made on February 26, 1992, amounting to 12 percent against the basket. Since then, the “crawling” devaluation has continued at the monthly rate of about 1.8 percent (now 12 zlotys per day).1
The growth in exports in 1992 was in part the result of a more active exchange rate policy. In the first six months of 1992, Poland accumulated a trade surplus of $946 million. Foreign exchange reserves reached $4.1 billion at the end of June, covering more than four months of average monthly imports, and reserves increased further in the summer months. In addition, the inflation rate did not exceed the expected level.
Polish residents must channel foreign transactions of more than the equivalent of $10,000 through licensed foreign exchange banks. All export proceeds in foreign exchange must be immediately surrendered to these banks in exchange for zlotys. Initially, when the so-called internal convertibility of the zloty was introduced, foreign exchange banks had to resell foreign exchange to the National Bank of Poland (NBP), which manages external reserves. The NBP provided the necessary foreign currency for the settlement of liabilities due foreigners for all imports, property rights, and attendant services, as well as other services and capital transactions under general or individual permits.
The Foreign Exchange Act of February 15, 1989, which introduced the internal convertibility of the zloty on January 1, 1990, directly involved the NBP in this system of obligatory purchases and sales of foreign exchange. In order to improve the efficiency of interbank settlements in foreign currencies (as defined by the President of the NBP), foreign exchange banks have been authorized since May 1992 to effect within one business day bilateral transactions (including cash transactions) of amounts derived from obligatory resales of foreign currencies by domestic enterprises. Such transactions are conditional, however; the NBP sets limits on how much foreign exchange these banks can hold, and at the end of each business day the banks must sell everything above those limits back to the NBP. Banks that are allowed to handle foreign exchange for domestic and foreign entities may also conclude domestic transactions in foreign currencies among themselves, using resources available from other of their activities—for instance, the foreign currency accounts of individuals and other banks, and banks’ foreign exchange deposits.
Foreign exchange banks have also been given the right to establish the exchange rate they use, except in transactions with the NBP, within a range of ± 2 percent of the average daily rate set by the NBP. Resident native persons may make transactions freely on the parallel foreign exchange market. In 1991–92, the parallel exchange rate did not differ significantly from the NBP’s average rate (in 1992, the differences were usually not more than ± 1 percent).
In the near future, the NBP plans to work out new rules that will authorize foreign exchange banks to establish foreign exchange open position limits. These limits should, on the one hand, reduce the level of exchange rate risk and, on the other, accelerate the operations of the interbank foreign exchange market. According to the NBP’s scheme, foreign exchange banks will be obligated to report whether their foreign exchange positions are short or long and to offset positions that exceed the limits on the interbank foreign exchange market. The limits should not be more than 15 percent of the banks’ own funds.
Summary of Discussion
Participants discussed the exchange rate component of Czechoslovakia’s 1990 economic program. In particular, there were questions as to why a country with such a low level of reserves would adopt a fixed exchange rate system. The point was made that while the exchange rate was the most worrisome element of the program, the Czechoslovak authorities had believed the overall package would be strong and credible enough to deflect speculative runs on the currency, particularly in the event of an unforeseen shock. Moreover, the country’s reserves nearly doubled as soon as the agreement with the IMF was signed, as IMF and other resources quickly became available. In the end, the current account balance was much better than expected.
Although Poland’s official exchange rate was pegged at a somewhat less depreciated level than the parallel market rate, the new rate was low enough to represent a clear break with the previous policy of incremental devaluations (around six in a few months). While an initial burst of inflation just after the new devaluation of the zloty caused a sharp decline in real wages, labor was relatively restrained in view of the rising unemployment rate and a 12 percent decline in output for 1990. In the first few months of Poland’s program, labor adopted a wait-and-see attitude, despite the growing economic insecurity. Interest rates increased dramatically under the program, moving from negative to positive in real terms.
Finally, participants were in broad agreement that the success of the Polish and Czechoslovak programs was attributable largely to the comprehensive package of reforms undertaken by both governments. But to some extent, these successes masked other urgent needs, especially for banking system reforms and accelerated privatization.
See, for example, Irving B. Kravis, Alan Heston, and Robert Summers, World Product and Income (Baltimore: The Johns Hopkins University Press, 1982).
Joshua E. Greene and Peter Isard, Currency Convertibility and the Transformation of Centrally Planned Economies, IMF Occasional Paper 81 (Washington, D.C.: International Monetary Fund, 1991).
See “Cost of Living in Eastern Europe” (table), Business Eastern Europe, April 13, 1992, pp. 172–73.
A floating rate was established on May 27, 1993.
China unified its dual exchange rate system in January 1994.
Note: This presentation was made before the division of the Czech Republic and Slovakia and reflects arrangements made before the formal separation.
The Czech Republic of the Czech and Slovak Federal Republic before separation.
Enterprises used to a monopolistic structure quickly learned to respect the constraints of the marketplace, especially in terms of tradables. The first thing that happens when trade is opened is that new price relativities are established. A bicycle produced in Germany, for example, will be made of lightweight aluminum and titanium, with many high-tech features. The equilibrium relative price with a Czechoslovak bicycle may be 2:1, but it might take some time to establish this relative price in the marketplace. Very few German bicycles will be imported at double the price of a local one at current Czechoslovak salaries. But it is quite likely that local producers will raise prices closer to German levels for a time. They will probably not be able to sell much at that price—people prefer to save longer for the German product or just to withhold consumption. In Czechoslovakia, this is what happened: people postponed consumption of those kinds of durables, inventories built up, and prices fell back to something closer to a sustainable level.
On August 27, 1993, the authorities devalued the zloty by 8 percent against the basket (7.4 percent in foreign currency terms) and reduced the rate of crawl from 1.8 percent to 1.6 percent per month.