Previous Page Table of Contents Next Page

Commodity Prices, Exchange Rates and the International Monetary System[2]

Presented by
Dr Robert Mundell
University Professor of Economics
Columbia University
(1999 Nobel Prize in Economic Science)

In discussing commodity prices, one is dealing with commodity prices in currencies; so it may be expected that monetary variables are among the explanations of real change. International commodity prices are mostly expressed in dollars, for example, in the IMF International Financial Statistics, or in terms of indices based on dollar prices. Such commodity prices are obviously affected by inflation as well as real developments, and also by the value of the dollar exchange rate. There is a definite link between monetary policies, exchange rates and commodity prices, and this is the subject which I wish to discuss today.

Pricing in gold and dollars under fixed rates

Prices are relationships between two quantities, a quantity of the object for sale, and a quantity of a quid pro quo-usually money--offered for it. It may therefore be expected that changes in prices could reflect not only market-specific trends but also monetary development. In a world of inflation, for example, commodity prices would be rising, and in a world of deflation they would be falling. Both would be clear manifestations of monetary rather than real disturbances. There would not be a problem of "commodity prices," there would be a problem of monetary stability. To analyze significant trends in "commodity prices," therefore it is important first to isolate the monetary disturbances (if they are present) from the real disturbances.

Superimposed on general movements of world-wide inflation or deflation are influences of exchange rates. In our world of multiple currencies and flexible exchange rates, commodity prices might rise in one currency but fall in another. The statement of commodity prices in dollars could reveal either a problem concerning commodity prices or a problem of the dollar. This bring sup the question: in what currency or currencies should commodity prices be quoted?

In the post-war world, the dollar was by far the most important currency in the world and had been since World War I. It was natural to use it as the basic unit of account and the convertible dollar --the "1944 gold dollar"-- was the anchor for exchange rates. Parities for currencies were expressed in weights of gold (the dollar was 1/35 of an ounce or .888671 grams of gold), but currency units and exchange rates were more normally expressed in terms of the more familiar gold dollar. As long as the dollar was exchangeable into gold at $35 an ounce, the dollar had the legal role and legitimate status as the international unit of account. It was natural also to use dollar quotations as the basis for the index of commodity prices.

All that changed with the international monetary system broke down in the early 1970s. The dollar was no longer convertible into gold, and foreign currencies were no longer convertible into the dollar. The dollar lost its judicial status as both monetary anchor and unit of account. Exchange rates became flexible. The IMF Board of Governors then officially scrapped the IMF constitution based on fixed exchange rates and officially accepted the a new regime of market-based "managed" flexible exchange rates. The idea was to let markets determine exchange rates. At the same time it was decided to rid gold of its "mystique," and to auction off at least part of IMF gold stocks as well as US Treasury holdings, and to introduce in its place as a numeraire the index of the value of a basket of a few major currencies that the Special Drawing Rights (SDR) had become.

What numeraire under flexible exchange rates?

Unfortunately, there was not at the time much understanding of how the new regime would work or of what would fulfil the functions formerly filled by gold and the dollar. Unlike the previous system, which had been built upon the experience of hundreds of years of monetary history, there was no precedent for the new regime of paper currencies connected by fluctuating exchange rates. In addition there had been little theoretical analysis of the problems likely to be encountered.

One of the problems related to the use of a unit of account. With all currencies on the same footing, international payments would be in chaos. At the most rudimentary level, how would exchange rates be quoted? With n currencies in the world there are ½ n (n-1) exchange rates. If n = 200 there are 9900 exchange rates! Flexible exchange rates in the absence of a numeraire in which to express currency prices would involve enormous confusion.

Fortunately, the market found the solution. Under flexible exchange rates the dollar was more rather than less important than before. Exchange rates were quoted mainly in dollars, the currency most frequently used in exchange markets and the main reserve asset (apart from gold) of central banks. If all currencies were quoted in dollars and perfect arbitrage could be relied on there would be "only" 199 independently flexible exchange rates.

The same solution was adopted in the statistical monthly, IMFIFS. There was no longer any legal basis for using the dollar as the numeraire for expressing exchange rates but it was the expedient solution. Dollar exchange rates gave some coherence to international monetary transactions. But it was far from a solution. The usefulness of a currency as numeraire depends partly on its stability. But was the dollar stable?

Impact of the dollar cycle

There would have been no problem if the dollar had been stable vis-à-vis other currencies. But in fact that has not been the case. Since the 1970s, the two most important currencies, besides the dollar, have been the German DM and the Japanese yen. The dollar has gyrated against other major currencies. Against the DM, for example, the dollar was DM 3.5 in 1975 and fell in half to DM 1.7 five years later, in 1980. Then the dollar doubled to DM 3.4 by early 1985, and then fell below DM 1.35 in August 1992, at the peak of the ERM crisis in Europe. Since that time the dollar has risen far above DM 2.0 This instability of the dollar/DM rate means that commodity prices in dollars and DM would be for much of the period moving in opposite directions. In a period when the commodities prices were rising in dollars, they might have been falling in DMs, and vice versa.

Yen-dollar fluctuations have been just as extreme. In the hey-day of Bretton Woods, the yen-dollar rate was fixed at 360 yen to the dollar. After the 1970s this rate became flexible. By 1985 the dollar was 250 yen. Ten year later, by April 1995, the dollar had fallen to 78 yen. In other words the yen had tripled in value against the dollar. This was the period in which the balance sheets of Japanese companies were undermined, and Japanese banks ended up with the non-performing loans that persist in trillions of dollars to this day. But from April 1995 to June 1998, the dollar soared from 78 yen to 148 yen, creating the setting for the Asian crisis. After 1948 the dollar fell to 105 yen but then rose again. During these fluctuations, dollar and yen prices frequently moved in opposite directions.

The instability of dollar-mark and dollar-yen exchange rates does not prove by itself that the dollar is unstable. The instability could equally be due to events in Germany or Japan or elsewhere. Yet there has been a distinct cycle of the dollar measured against other major currencies, and this means that quotations of commodity prices in dollars may give rise to grave misinterpretation.

The movement of the dollar with respect to the SDR was one way of measuring the stability of the dollar. Initially, the SDR was equivalent to 1/35 of an ounce of gold, i.e., a 1944 gold dollar, but its gold basis was stripped away in the later 1970s and it eventually became a basket of five major currencies including, besides the dollar, the yen, the mark, the pound and the franc. The weights were changed as seemed appropriate with changing circumstances. The value of this SDR basket in terms of the dollar was of course unity in 1970, when the first issue of SDRs were made. Then it rose to $1.32 in 1979 with the dollar's depreciation, fell to $0.98 in 1984 as the dollar soared in the early 1980s, then rose to $1.49 in 1995 with the weakening of the dollar and then fell to $1.24 at the end of 2001 as the dollar strengthened in the late 1990s. These fluctuations, it should be noted, have been extremely large, especially in proportion to differences in price levels and inflation rates.

Commodity price cycles

The question needs to be asked whether the cycle of the dollar against major currencies is related to the cycle of the dollar commodity prices. A casual reading of the statistics suggests that this relationship is quite close. Thus the index of non-oil dollar commodities tripled in the 1970s when the dollar was depreciating sharply relative to the SDR; it then fell by more than 20 per cent from 1980 to 1986 when the dollar was soaring; then it rose by 50 per cent from 1986 to 1995 when the dollar was again depreciating; and it has fallen by 30 per cent since 1995 when the dollar has been appreciating. There is therefore a very pronounced association of the cycle of the dollar against other major currencies (as measured by the SDR) with the cycle of dollar commodity prices.

This is of course not unexpected. It is natural that there would be a correlation of the prices of commodities in dollars with the price of currencies in dollar. Whenever US monetary policy is easy, as it was in the late 1970s and the late 1990s and early 1990s, the dollar depreciates against foreign currencies and commodities; and when it is tight, as in the early 1980s and the late 1990s, the dollar appreciates and dollar commodities prices decline.

It makes a great difference if commodity prices are quoted in dollars, euros or SDRs. The IMF world index which covers about 50-70 non-fuel commodities quoted every month in dollar terms, indicates that prices have fallen in the last 3 years. Starting in 1970, the index, with 1995 as one hundred, was 32.8 in 1970, and 57.0 in 1975, and 90.7 in 1980. From 1980 to 1986 it dropped from 90.7 to 67.8, and then rose to the peak of one hundred in 1995. Subsequently it fell to 70.2 at the end of 2001, a very precipitous 30 percent drop.

Broken down into major commodity groups, the index in 2001 for food was 76.5, beverages 47.2, agricultural raw materials 68.7, and metals 71.2, while that for fertilizers was quite different at 102.2. Another index, that of the World Bank for lower middle income countries' commodity exports stood at 62.4, with 1995 equal 100. All these numbers show that except for fertilizers which can be considered more of a manufactured product, all other commodity prices showed a very steep decline during the 1995-2001 period.

Let us look again at the dollar-SDR rate. From a parity of 100 in 1970 in the dollar-SDR rate, by 1975 the SDR price of the Dollar had fallen to 82.4 cents, and by 1980 had dropped further to 77 cents. Then it soared to 98 cents in 1986. Subsequently it fell to 66 cents and then it rose again, to 89 cents. With one exception in the 1970s, that cycle mirrors that of commodity prices fixed in dollars. When the dollar is weak then commodity prices rise.

From 1967 to 1981 commodity prices in dollar terms tripled, while the value of the dollar fell to one-third over that inflationary period. These were the years of the very strong oil prices that brought euro-dollars into the system, a big expansion of the euro-dollar market and inflationary prices in the United States, with two-digit inflation rates during 1979 to 1981.

Then from 1980 to 1985 a big fall occurred in commodity prices, with the index dropping from 90.7 to 67.8. That coincided almost exactly with the soaring value of the Dollar. The reversal of the policy mix under Ronald Regan included sweeping tax cuts. Marginal tax rates were cut from 70 percent at the federal level to 28 percent for the highest income tax brackets. Corporate taxes were cut from 48 percent to 34 percent and capital gains taxes were also reduced over that period. Big increases in government spending were combined with a tightening of Federal Reserve monetary policy. There was a sharp fall of the price in gold, which dropped from 850 Dollars/ounce in February 1980 to 300 Dollars/ounce within something like two years. This was a period of big deflation or disinflation. The inflation rate in the United States fell from 13 percent in 1980 to 4 percent in 1986. And that period of disinflation was a period of very sharply falling commodity prices.

After 1985, there was another shift in United States policy, aimed at depreciating the US Dollar. The Dollar started to go down slowly against the Yen first, and subsequently the Yen soared following the drop in oil prices in 1985/86. Over that period, Dollar/SDR rate fell from 98 cents in 1986 to 66 cents in 1995, and commodity prices soared from an index value of 67.8 in 1986 to 100 in 1995. However, from 1995 onwards index of commodity prices fell to 70, while during the same period, the Dollar soared against the SDR, rising from 66 to 89.

Future of the euro and the dollar

What are the indications for the future? Many believe that the Dollar has reached its peak and that the future will see a much weaker Dollar and a stronger Euro as a result of many positive developments in the European economy. However one long-run force which might contribute to a weakness of the Euro would be the accession of countries, such as Poland, Romania and other countries of Central and Eastern Europe, which could raise the level of debt in the Euro zone.

On the other hand, the factors which might contribute to a weaker Dollar include a 400 billion Dollar current account deficit, or four percent of the country's 10 trillion Dollar GDP. And with the United States recovery, the trade deficit could rise to 600 billion Dollars, or 6 percent of GDP. Of course, there were deficits of 3.5 percent in the 1980s, but this was a time when the United States was still a net creditor. Now the United States has become the biggest debtor in the world to the extent of 25 percent of GDP; in other words, its international liabilities exceed by 2.5 trillion Dollars its international assets, and that amount is rising by 4-6 percent of GDP every year. At that high rate, the ratio will increase to 29 percent at the end of this year and 35 percent in the year following. Such a rate of increase in indebtedness cannot be sustained for long without giving rise to strong pessimistic views about the future of the Dollar. The only way in which this situation can be corrected is to reduce the United States current account deficit, and to do so would require the depreciation of the Dollar. A halting of lending to the United States would correct a good part of the deficit but probably not all of it, but when market sentiment starts to turn, there could be pressure for a very rapid downward trend in the value of the Dollar. The resultant portfolio shifts could be expected to cause a correction of commodity prices.

Productivity, the dollar cycle and commodity prices

The link between the commodity price cycle and the dollar cycle is apparent, but the underlying causes are not clear. Obviously, arbitrary exchange rate changes can lead to commodity price changes, and as I have said before dollar prices may not reflect truly trends in real commodity prices. Prices in SDR terms would be better, as would in some cases an index of gold prices. Using some other types of measures, the swings in commodity prices are much attenuated.

It should be pointed out that despite the weight of the United States economy in the world economy, there is not necessarily a direct causal relationship between the strength of the dollar in currency markets and commodity prices. It could be that the same factors that cause the dollar cycle also cause the commodity cycle. One of the factors, for example, which has caused the dollar cycle has been the IT revolution and the resultant very rapid increase in productivity in tradable goods, which meant that the real Dollar had to appreciate. With the United States inflation rate around 2.5-3 percent throughout the period, the appreciation of the real Dollar needed because of the "internet economy" was something like 5 percent, which was accompanied by deflation in all the countries that kept their currencies fixed to the Dollar such as Argentina, China, Hong Kong SAR, the Gulf States and Panama. Every one of those countries had deflation in this period, prime evidence that productivity effects were behind this strong Dollar.

The United States economy accounts for about 25 percent of the world economy measured at current exchange rates. So anything that affects the Dollar, the currency of that big economy, is certainly going to affect real events; and those factors that led to the Dollar weakening or strengthening can also lead to fluctuations of commodity prices. Thus, very strong productivity growth and the change in the real exchange rate, coupled with tightness on the part of the Federal Reserve to keep the consumer price index below 3 percent contributed to the slowdown in the United States and the global economy, and that would certainly be an important factor in depressing commodity prices.

However, looking for a single cause, is simplistic. For example, there are two kinds of mistakes that one can make in relating exchange rates to basic real commodity prices.

One is to say that exchange rates do not matter, while the other is to consider exchange rates as responsible for a whole series of different problems. In fact, in the short run they matter, while in the long run they do not matter very much. Therefore, it would be a good idea to have a reform of the international monetary system in order to avoid any possible link between exchange rates and commodity prices. Moreover, a restoration of a fixed exchange rate system would provide countries with a new rudder for monetary policy and would be a great step in the improvement of economic policy.

I want to conclude by emphasizing that the current international monetary arrangements are far from optimal. They do not constitute a system. If the Balkanized world were suddenly transformed into a centralized empire, its first act would be to create a common currency that would be acceptable everywhere, with a great improvement in potential welfare. In the absence of a hegemonic empire, monetary efficiency depends on cooperation which in turn requires a world at peace that can be enforced. The end of the Cold War opened up a new era of globalization and the emergence of a global economy. As Paul Volcker has said, a global economy needs a global currency.

[2] Summary of the verbatim record presented at the Consultation.

Previous Page Top of Page Next Page