Journal Issue

The Magnitudes of Exchange Devaluation

International Monetary Fund. External Relations Dept.
Published Date:
June 1968
  • ShareShare
Show Summary Details

Margaret G. de Vries

IN RECENT YEARS a number of economists, especially in the universities, have lamented the relatively little use that countries have made of exchange rates to solve their balance of payments difficulties. For example, many of these economists have been advocating flexible rates to overcome the infrequency of change of fixed rates or have been considering how fiscal or monetary policies can be made more flexible to compensate for the rigidity of exchange rates. For the most part, these economists have focused their observations on the large industrial countries. They have been noting that, until the devaluation of the pound sterling in November 1967, few changes had been made since 1949 in the exchange rates of the major industrial powers.

On the other hand, the business community and commercial traders have had a keen awareness of all exchange rate changes. With an eye on their competitive positions, their fears of other countries’ devaluations may have been exaggerated.

With so much current interest in exchange rates, it is worthwhile to take a brief look at the magnitudes and frequencies of exchange devaluation that have occurred in a broad spectrum of countries since World War II. It is evident that use of the exchange rate mechanism has been rather common.

Measuring the Magnitudes

Percentage changes in the exchange rates for 109 countries—by major geographic region—for the 19-year period from the end of 1948 to the end of 1967 are presented in Table 1; by using long-term data numerous small devaluations can be compressed into a single figure. The end of 1948 as a starting date has been preferred to the end of 1946 (when initial par values were set for most of the Fund’s original members) because the exchange rates of several countries were more settled by that year (or by early 1949) than they were immediately after the war, and because data for 1948 are readily available.



No Devaluation1321271
Less than 30%1200732
More than 75%2464572

Because the results depend on the particular exchange rates used, and because in some instances (such as countries with multiple currency practices) a choice of rates existed, an endeavor has been made to select for these calculations currently used exchange rates as against, say, legal parities. For most countries, par values, or at least fixed official rates, could be used; but for some countries, free market rates have been used, and for a few a choice among multiple rates has been made. The percentage changes in exchange rates have been measured relative to gold, or to the U.S. dollar with gold content expressed in the weight and fineness in effect on July 1, 1944, with 100 per cent being the maximum possible devaluation.

How Much Devaluation?

Table 1 also shows the distribution of 109 countries over five major geographic regions according to the net percentages by which their currencies have been devalued. The first group shows 13 countries which have not devalued at all: the United States, 7 Central American countries, Ethiopia and Liberia, and Switzerland and Japan.1 In Lebanon a slight appreciation occurred. The second group comprises 12 countries that have devalued less than 30 per cent—that is, approximately the amount by which sterling was devalued in September 1949; in this group are not only several major industrial nations (Belgium, Canada, Germany, Italy, Luxembourg, and the Netherlands) but also 3 Latin American countries (Costa Rica, Nicaragua, and Venezuela) and Portugal, Saudi Arabia, and the Syrian Arab Republic. Germany and the Netherlands, after devaluing by about 30 per cent in September 1949, both revalued their currencies upward by 5 per cent in 1961.

The third group comprises 22 countries which have devalued a total of 30-39 per cent. This includes mainly countries which devalued along with sterling in 1949 but which did not go along in the devaluations of 1967. It also includes Denmark, which devalued again in November 1967, by 7.9 per cent—less than the 14.3 per cent of the 1967 sterling change. Apart from Australia, Denmark, Norway, and Sweden, the group includes developing countries in Asia, the Middle East, and Africa—for example, Burma, Iraq, Kenya, Malaysia, Nigeria, Pakistan, and Singapore.

What is often not fully realized is that for 62 countries (groups four and five) the magnitude of postwar devaluation has exceeded 40 per cent. Some 38 countries have devalued between 40 and 75 per cent; this includes not only the United Kingdom and Ireland, which have devalued just over 40 per cent in total, but also Austria, Finland, India, Mexico, New Zealand, Peru, the Philippines, and Turkey, which have devalued by larger amounts.

For another 24 countries devaluation has gone beyond 75 per cent. Among European countries in this last group are France, Greece, Iceland, Spain, and Yugoslavia. Other countries devaluing by these large amounts have been 7 Latin American countries (Argentina, Bolivia, Brazil, Colombia, Chile, Paraguay, and Uruguay), as well as Ghana, Indonesia, Israel, Korea, and Viet-Nam.

In sum, devaluations, even by the major powers, fall in all five categories. Among less developed countries two patterns of devaluation are apparent. One pattern follows that of the major currencies. Countries in Central America have conformed generally to the U.S. dollar and have devalued either not at all or very little. Many others in Asia and Africa have followed the lead of either sterling or the French franc.

The second pattern of devaluation among developing countries is related to the price and balance of payments experiences of the individual country. Many of them in all geographic regions have thus devalued independently of the major currencies, by amounts from 40 to nearly 100 per cent.

The average (that is, the arithmetic mean) devaluation of all 109 countries for these years is 48.2 per cent. The arithmetic mean for those 96 countries that did engage in some devaluation (eliminating the 13 countries of group one) is, of course, higher—54.7 per cent. The global picture is seen in better perspective, however, when account is taken of the differing importance in world trade of various countries. Arithmetic means have, therefore, been calculated (Table 2) which have been weighted by the share in 1966 of the country’s exports in total world exports.




Australia, Canada, New Zealand, and the United States45.2
Middle East1238.4
Asia (excluding Japan)1446.1
Latin America2362.2

Australia, Canada, New Zealand, and the United States, as a group, have devalued by the least amounts. The weighted average for Europe is 23.5 per cent. The Latin American countries have devalued by the greatest amounts—a weighted average of 62.2 per cent—with the Middle East, Asia, and Africa in between. As would be expected (because the larger trading nations devalued less than most smaller ones) the over-all world weighted average (calculated for 109 countries) is 22.8 per cent, much less than the unweighted average.

The Frequency of Devaluation

The frequency with which devaluation occurs is also often underestimated. Admittedly some 12 countries have not devalued at all in the postwar period—1 has even slightly appreciated its currency—and another 27 have devalued only once, most of them in the general realignment of 1949. However, an impressive total of 48 countries has devalued at least once more after 1949, and another 21 have devalued quite frequently. Table 3 shows these frequencies.


No Devaluation since 194812
One Devaluation27
Two Devaluations24
Three Devaluations24
Frequent Devaluations, but no change since 19625
Frequent Devaluations, including changes since 196216

Among the 24 countries that have devalued once again after 1949, for a total of two devaluations, are not only those which devalued in November 1967 along with the United Kingdom (such as Ceylon, Denmark, Ireland, Jamaica, and New Zealand) but several which found it necessary earlier to devalue once more the rates they had set in 1949: Austria, Ghana, Greece, India, and the United Arab Republic. France and the French franc area have devalued twice since 1949—once in 1954 and again in 1958—making a total of three devaluations.

Devaluations for countries with multiple or fluctuating rates have been even more numerous. Some five countries—Bolivia, China, Paraguay, Thailand, and Yugoslavia—had frequent devaluations until 1962, but have not changed their rates subsequently. Another 16 countries have continued to devalue fairly often, even in the last five years. These include not only several Latin American countries (Argentina, Brazil, Chile, Colombia, and Uruguay) but also a few countries in Europe and several in the Middle East and Asia.

Comparison With Prices

Interesting as these figures are, measurements of the changes in exchange rates for various countries are more revealing when juxtaposed with movements in their domestic prices. While the Brazilian cruzeiro and the Chilean peso have, for example, been devalued by 99.9 per cent betwen 1948 and the end of 1967, their domestic price levels have, in the meantime, gone up 100-fold. The question, therefore, arises of how much the exchange rate has changed in comparison with rises in internal prices?

Accordingly, ratios have been calculated between actual exchange rates at December 1967, and notional rates which are those which have just kept pace with subsequent inflation. A notional rate for each of 65 countries has been obtained by taking the exchange rate in 1948 and adjusting it to take account of the rise in the cost of living index in that country from 1948 to mid-1967.2 This rate shows what exchange rate would have prevailed in mid-1967 had the rate been devalued since 1948 merely to the same extent as the country’s consumer prices have increased.

It is thus possible to see what changes in exchange rates have occurred in something like “real terms”—that is, abstracting from differences in internal price changes. The actual exchange rate for each country in December 1967 was expressed as a percentage of the notional rate, that is, as a ratio. A ratio of 100 indicates that the exchange rate and prices are in the same relationship in 1967 as they were in 1948. Where the ratio exceeds 100, devaluation has been less than the increase of the country’s prices of consumer goods. Where the ratio is below 100, the exchange rate has been devalued more than the country’s prices have increased.

The 65 countries have been grouped into five categories, and an average actual-notional exchange ratio obtained for each group. These average ratios are presented in Chart 1. Nineteen countries—the United States, Canada, the major powers of Western Europe, Japan, and others—have been placed in group one as “industrial countries.” Most of these have had what is here labeled as “moderate inflation”—that is, the level of consumer prices in mid-1967 was about 1.4 to 2.0 times its 1948 level.


The remaining 46 countries—comprising what are usually called the developing countries—have been distributed over four groups according to the degree of inflation. Twelve—some in Central America, a few in the Middle East, and 2 in Asia—have had what can be regarded as “little inflation,” using as a benchmark less inflation than that of the United States (group two). Another 10—the rest of Ceneral America and a few countries in Asia and Africa—are in group three, that of “moderate inflation,” as defined above.

Another 9 (group four)—some from each region—have had what is here called “strong inflation,” that is, their price levels rose more than two, but less than three, times from 1948 to 1967. Finally, in group five, are 15 countries characterized as those with “chronic inflation”; their consumer price levels have gone up more than three times, and many by much more; half of these countries are in South America, but a few in southern Europe, the Middle East, and Asia are also included.

This arrangement of countries enables one to interpret the actual-notional exchange rate ratios with reference to the degree of inflation. Low ratios in countries with little inflation, for example, reflect their having had very little increase in prices; they may not have adjusted their exchanges rates at all since 1948. On the other hand, low ratios for countries with very large inflation reflect degrees of exchange devaluation which are even greater than the rises in their internal prices.

Realignment of Postwar Exchange Rates

In Chart 1, low ratios appear on the left side of the 100 mark and high ratios on the right. It is revealed that the major industrial powers have an average ratio well above 100—indeed 141. In other words, the general level of postwar consumer prices in these countries has risen more than their exchange rates have been devalued.

The ratios for the four groups of developing countries, on the other hand, were considerably lower, and two of the four groups of developing countries were even below 100. For the developing countries in group two, the explanation for the average ratio of only 95 is simple: their price increases have been virtually negligible. Several of them have also devalued in line with major currencies, although only two or three of them have devalued in the last few years. The average ratio of 121 for group three, developing countries with moderate inflation, is also lower than that of the industrial countries, partly because the rates of inflation in the industrial countries have been somewhat lower and also because some of these countries have devalued by larger amounts than most of the developing countries.

What is more surprising are the relatively low ratios for groups four and five, developing countries with relatively strong and even chronic inflation. This clearly reflects extensive exchange depreciation, even in real terms (apart from price rises). And the greater the inflation, the larger, on average, is the depreciation, even in real terms.

Several factors help to explain this extensive devaluation by the developing countries even relative to their greater inflation. First, the exchange rates of many developing countries may well, in 1948, have been more out of line with prices in the rest of the world, compared with prices at home, than the exchange rates of the major countries. Hence, more “correction” may have been necessary to realign their rates. Second, developing countries have used, more than industrial countries, exchange rate adjustment as a policy tool. Industrial countries have relied more heavily on monetary and fiscal policies and on other domestic measures. Third, demand for the exports of the industrial countries has, on the whole, been more dynamic than that for the exports of the developing countries. Moreover, the productivity advances in industrial countries in export industries have also been more marked than in developing countries; these have helped to keep their export prices relatively lower than their general consumer prices. Hence, the general level of prices in industrial countries could often run ahead of exchange rates without damaging their export positions; this has been much less true for developing countries.

Fourth, the mechanism by which exchange rates were adjusted also seems to have been a factor in the extent of depreciation. Countries which have used multiple rates—and these were mainly developing countries—have devalued much more in relation to their general domestic prices than have countries with unitary exchange rates. As countries have eliminated multiple rates, they have usually devalued down to the prevailing free market rate; the free market rate, at least at the time of devaluation, has usually been below that determined by general prices. Finally, it has been suggested in a Fund study that inflation itself, because of its price distorting effects, may induce a degree of exchange depreciation in excess of the rise in domestic prices.3

Not Equilibrium Rates

These comparisons between actual and notional exchange rates based on internal prices cannot, of course, be taken as suggesting equilibrium rates. For one thing, the base period used here, 1948, is widely known to be anything but an equilibrium year to be used as a starting point. And the 1967 pattern is undoubtedly a more realistic structure. More importantly, there has been increasing recognition in the last 20 years that equilibrium exchange rates are a function not only of relative prices but also of changes in productivity, capital flows, and levels of tariffs and trade barriers considered appropriate, as well as the domestic monetary, fiscal, and employment policies a country wishes to pursue.

However, it is evident that over the long haul, from 1948 to the end of 1967, important alterations had occurred in the structure of international exchange rates. Many countries—industrial and developing alike—have devalued their exchange rates extensively since the postwar international monetary system commenced in 1946. Furthermore, compared with that starting point, the relative realignment of rates between countries that has occurred has been striking. Countries with chronic inflation have engaged in severe and frequent devaluation which has more than offset their subsequent aggregate domestic price rises. Countries with lesser amounts of inflation have also offset considerable amounts of their domestic price increases via adjustments in their exchange rates. And a substantial realignment of the exchange rates of the developing countries vis-à-vis the major industrial countries has occurred, even allowing for the price advances of the developing countries. Where their prices have risen, they have usually devalued a good deal more than the price rise. Where they have not devalued—or devalued very little—generally their price rise has been very small.

Japan’s official rate was set in April 1949 and has since remained unaltered.

Those 65 countries have been selected for which price data back to 1948 are available.

Graeme S. Dorrance, “The Effect of Inflation on Economic Development,” Staff Papers, International Monetary Fund (Washington, D.C.), Vol. X (March 1963), pp. 1-47.

Other Resources Citing This Publication