Finance and Development

About F&D

Back Issues

Write Us

Free Email Notification

Receive emails when we post new items of interest to you.

Subscribe or Modify your profile

A quarterly magazine of the IMF

March 2003, Volume 40, Number 1

Commodity Currencies
Paul Cashin, Luis C�spedes, and Ratna Sahay

Developing countries reliant on commodity exports see the fate of their exchange rates tied to fickle commodity markets

For decades, economists have tried with little success to model long-run movements in real—that is, adjusted for inflation—exchange rates. Almost all of the studies have focused on industrial countries, trying to pinpoint whether fundamentals such as government spending, current account imbalances, and differences in productivity and interest rates hold the key to explaining exchange rate movements. But the results have been disappointing, with many models that are based on fundamentals failing to provide a convincing explanation of the behavior of real exchange rates in industrial countries.

In contrast, studies of the behavior of developing country real exchange rates are scarce. The few studies that have examined the determinants of these rates have focused largely on Latin America and have emphasized the role of terms of trade movements in driving the real exchange rate. However, a natural assumption for developing countries is that fluctuations in real commodity prices have the potential to explain a large share of changes in real exchange rates, given that so many of these countries are highly dependent on commodities—in some cases, a single commodity—for the bulk of their export revenues. Indeed, several studies have explored this relationship for a handful of commodity-exporting industrial countries, such as Australia, Canada, and New Zealand. But the biggest hurdle in extending these studies to developing countries has been the lack of country-specific data on commodity export prices.

That is why we undertook a study of the relationship between the real exchange rate and real commodity prices for all commodity-dependent economies. We asked two questions: Do real commodity prices and real exchange rates move together? And does the exchange regime affect a country's ability to cope with commodity price swings? The findings are surprising and raise serious questions for policymakers, given that the real exchange rate holds the key to a country's competitiveness in global trading markets.

Identifying commodity currencies

Our study began with the construction of new monthly indices of national real commodity export prices and the gathering of monthly real exchange rate data for 58 commodity-exporting countries for the period January 1980 to March 2002. Each country's (nominal) commodity export price index is a geometric weighted average of world prices for 44 individual nonfuel commodities (taken from IMF commodity price data), using country-specific export shares (averaged over 1990–99) as weights. The national indices of the nominal (U.S. dollar) price of nonfuel commodity exports are then deflated by the IMF's measure of the nominal (U.S. dollar) price of exports of manufactured goods to form the real price of commodity exports for each country. The real price of commodity exports can also be described as a measure of the terms of trade of each country, expressed in world prices. The real exchange rate is the IMF's real effective exchange rate, which is defined for each country as the trade-weighted average of bilateral exchange rates vis-�-vis trading partners' currencies, adjusted for price differentials between the home country and trading partners.

The 58 commodity-exporting countries include 53 developing countries (classified by the IMF's World Economic Outlook as exporters of nonfuel primary products and those with diversified export earnings) and 5 industrial countries—all of which rely on commodity exports for a major share of their export income. Indeed, during the 1990s, the cross-country mean share of total export receipts derived from primary commodity exports was about 48 percent. Commodity exports typically exceeded 50 percent of the total exports of several sub-Saharan African countries, especially Burundi (97 percent), Madagascar (90 percent), and Zambia (88 percent). The share of primary commodity exports in total exports was quite high even for the industrial countries (Australia, 54 percent; Iceland, 56 percent). In addition, many countries remain overwhelmingly dependent on export receipts from their dominant exportable commodity—the dominant exportable exceeded 90 percent of commodity export receipts in Dominica (bananas), Ethiopia (coffee), Mauritius (sugar), Niger (uranium), and Zambia (copper).

Armed with the real exchange rate and real commodity export prices for each country, we then checked to see whether these two series displayed a close relationship. We found that, for many countries, such as Australia and Burundi (both of which have flexible nominal exchange rates), this was indeed the case (see chart), while others appear to display a relationship once a onetime movement in the real exchange rate (such as the 1994 devaluation for the CFA franc zone countries of Mali and Togo) is accounted for.

Chart: Moving together

Next, we used regression analysis to formally examine whether there was a stable, long-run relationship between a country's real exchange rate and the real price of its commodity exports—in other words, which of the commodity-exporting countries qualified as having "commodity currencies"? In the analysis, we allowed for a structural shift in the relationship between the two series to account for onetime movements in either series (which typically involve rapid movements in the nominal (and real) exchange rate). We found just such a long-run relationship for 22 of the 58 commodity-exporting countries (Table 1). It is not unexpected that sub-Saharan African countries, given their dependence on commodity exports, account for half the commodity-currency countries. Moreover, for these 22 countries, over 80 percent of the variation in the real exchange rate can, on average, be accounted for by movements in real commodity prices alone—a surprisingly strong result. As for those commodity-exporting countries where such a long-run relationship could not be found, it is likely that factors other than real commodity prices played a key role in real exchange rate movements.

Table 1
Identifying the commodity-currency countries

Two-fifths of commodity-exporting countries have commodity currencies: there is a long-run relationship between their real effective exchange rate and their real commodity export prices.

Sub- Saharan Africa	Asia- Pacific	Middle East and North Africa	Western Hemisphere	Europe

Burundi Cameroon Central African Republic C�te d'Ivoire Ethiopia Ghana Kenya Madagascar Malawi Mali Mauritania Mauritius Mozambique Niger Senegal South Africa Sudan Tanzania Togo Uganda Zambia Zimbabwe	Australia Bangladesh India Indonesia Malaysia Myanmar New Zealand Pakistan Papua New Guinea Philippines Sri Lanka Thailand	Morocco Syrian Arab Republic Tunisia Turkey	Argentina Bolivia Brazil Canada Chile Colombia Costa Rica Dominica Ecuador Guatemala Honduras Mexico Nicaragua Paraguay Peru St. Vincent and Grenadines Suriname Uruguay	Iceland Norway

Note: Countries found to have commodity currencies are in bold.
Source: Cashin, C�spedes, and Sahay, 2002.

How large an impact do real commodity price movements have on the real exchange rates of commodity-currency countries? We found that the elasticity typically ranged between 0.2 and 0.4, with a median of 0.38. Thus, a 10 percent drop in the real price of the commodity exports of countries with commodity currencies is typically associated with a 3.8 percent depreciation of their real exchange rate. Further analysis also indicated that, when deviations from the relationship between exchange rates and commodity prices occurred in countries with commodity currencies, they were caused primarily by changes in real commodity prices. Following a movement in commodity prices, it is typically the real exchange rate that then adjusts to restore its long-run relationship with real commodity prices.

Does the exchange regime matter?

Theory tells us that, for economies prone to frequent real external shocks, flexible nominal exchange rates facilitate the smoothing of real output to such shocks, especially when domestic wages and prices are slow to change. Following a negative real shock (such as a decline in the world price of a key export), a nominal depreciation raises the domestic price of exported goods (to partially offset the decline in the international price) and reduces real wages in line with reduced labor demand. In contrast, in countries with inflexible nominal exchange rates experiencing such negative real shocks, prices and wages need to fall to ensure that employment and output do not decline. Accordingly, we examined whether nominal exchange rate flexibility or relative price flexibility was more important in driving the flexible real exchange rates of commodity currencies.

We began by classifying our commodity currencies by exchange rate regime. As Table 2 shows, we do this first in terms of the IMF's de jure classification (which is based on the publicly stated commitment of the national authorities) and then in terms of a de facto classification proposed in a recent study by Carmen Reinhart and Kenneth Rogoff (which is based on the observed behavior of market-determined exchange rates, including that of active parallel exchange rate markets). After all, through exchange market intervention and monetary policy, policymakers can transform a de jure flexible exchange rate regime into a de facto pegged regime. Similarly, active parallel markets can transform de jure pegged official exchange rates into de facto flexible regimes. Under either classification, our commodity currencies contain a mix of currencies with flexible and fixed exchange rate regimes. We then further divide our commodity currencies into three groups using the Reinhart-Rogoff classification of exchange rate regimes: pegs (single currency, SDR, and official basket pegs); limited flexibility or managed floats (including crawling-peg bands and cooperative arrangements); and flexible regimes.

Table 2
The type of exchange rate regime does not matter

The volatility of the real exchange rates of commodity-currency countries is similar, whether these countries have fixed or flexible nominal exchange rate regimes.


	Reinhart- Rogoff (2002) classification of exchange rate regime (de facto)	IMF classification of exchange rate regime (de jure)	Volatility of real effective exchange rate	Volatil- ity of nominal effective exchange rate	Volatil- ity of relative price levels	Ratio of volatil- ity
Peg
Bangladesh Cameroon Central African Rep. C�te d'Ivoire Mali Niger Papua New Guinea Togo Mean	Peg Peg Peg Peg Peg Peg Peg Peg	Managed float Peg Peg Peg Peg Peg Peg Peg	0.020 0.045 0.046 0.043 0.044 0.051 0.045 0.047 0.043	0.016 0.044 0.039 0.044 0.042 0.043 0.026 0.043 0.037	0.011 0.019 0.019 0.014 0.027 0.025 0.008 0.018 0.018	1.6 2.3 2.1 3.1 1.6 1.7 3.4 2.4 2.3
Limited flexibility/managed float
Iceland Morocco Bolivia Indonesia Tunisia Mauritania Burundi Kenya Paraguay Syrian Arab Rep. Mean	Limited flexibility Limited flexibility Limited flexibility Limited flexibility Limited flexibility Limited flexibility Managed float Managed float Managed float Managed float	Peg Peg Managed float Managed float Managed float Managed float Managed float Managed float Independent float Peg	0.021 0.012 0.197 0.066 0.013 0.028 0.036 0.035 0.044 0.047 0.050	0.025 0.011 0.235 0.067 0.012 0.026 0.029 0.034 0.053 0.046 0.054	0.016 0.007 0.118 0.013 0.004 0.018 0.022 0.015 0.034 0.022 0.027	1.6 1.7 2.0 5.1 3.3 1.5 1.3 2.3 1.5 2.1 2.2
Freely floating/falling
Australia Ecuador Ghana Malawi Mean	Freely floating Freely falling Freely falling Freely floating	Independent float Managed float Managed float Managed float	0.023 0.054 0.114 0.046 0.059	0.023 0.058 0.122 0.049 0.063	0.003 0.019 0.026 0.019 0.017	7.1 3.1 4.7 2.6 4.4

Note: Exchange rates are classified according to the regime in place for the majority of years during 1980–2002. The nominal effective exchange rate (the trade-weighted average of bilateral exchange rates vis-�-vis trading partners' currencies); relative price level (the domestic price level relative to those of trading partners, which is the differential between the domestic price level and the trade-weighted foreign price level); and real effective exchange rate indices are in logarithms (base 1995=100) and are monthly. Volatility is measured as the standard deviation of the (monthly) rate of change of the series. Ratio of volatility is the ratio of the standard deviation of the (monthly) rate of change of the nominal effective exchange rate to the standard deviation of the (monthly) rate of change of the relative price level.
Source: Cashin, C�spedes, and Sahay, 2002.

Do exchange rate regimes influence the behavior of commodity currencies? Our study shows—surprisingly—that the variability of the real exchange rate is similar across the various nominal exchange rate regimes. This result is contrary to a 1986 study by Michael Mussa, which found that industrial country real exchange rates were substantially more variable during flexible nominal exchange rate regimes than during fixed nominal exchange rate regimes. The volatility (as measured by the standard deviation) of each country's real effective exchange rate can be broken down into the volatility of the nominal effective exchange rate and the volatility of the price differentials between the home country and its trading partners (see Table 2). As expected, the volatility of the nominal effective exchange rate relative to the volatility of the relative price levels is smallest for the countries with pegged exchange rate regimes and largest for the countries with flexible regimes, with the average ratio of volatilities almost doubling from 2.3 (pegged) to 4.4 (flexible).

While the variability of the real effective exchange rate is similar across the various nominal exchange rate regimes, for countries with pegged nominal regimes, a larger relative share of real exchange rate variability is driven by the variability of relative prices—which is consistent with the study by Reinhart and Rogoff. They find that African CFA franc zone countries experienced by far the most frequent bouts of deflation (as measured by declines in the 12-month percent change in consumer prices)—about 28 percent of the time—during 1970–2001. This indicates that commodity currencies with pegged exchange rate regimes (three-fourths of which are CFA franc zone countries) experience deflation (or, at least, inflation rates that are lower than their trading partners') when they adjust their real exchange rates downward in response to the typically adverse movements in their real commodity prices.

The bottom line is that it is the nature of real shocks to the economy that determines the behavior of real exchange rates, not the type of nominal exchange rate regime. When real shocks (such as a change in real commodity prices) are the dominant influence on real exchange rates, to restore the efficient relative price between traded and nontraded goods, commodity-dependent countries need to have either flexible nominal exchange rate regimes (which facilitate the slow change of relative inflation rates) or flexible wages and prices (which facilitate the maintenance of nominal exchange rate pegs).

A handy crystal ball
Our study found evidence in support of the comovement of national real exchange rates and real commodity prices in a group of commodity-exporting countries. For these commodity-currency countries, the world price of their commodity exports has a stable and important effect on their real exchange rate. This empirical regularity is surprisingly robust, given the repeated failure of previous attempts to use models that are based on fundamentals to explain exchange rate movements.

For policymakers in commodity-exporting developing countries, understanding the effects of commodity price movements on exchange rates should be of great interest in guiding the conduct of monetary and exchange rate policies, particularly as such countries liberalize their capital markets and increase the flexibility of their exchange rate regimes. For countries with commodity currencies, commodity prices are the key determinant of their real exchange rate; they can be used as a benchmark in determining when exchange rates have deviated excessively from their equilibrium value. Policymakers in countries with commodity currencies can also use spot or futures prices on world commodity markets as important information in guiding monetary policy. Finally, it is likely that commodity prices will play a greater role in the design of inflation-targeting arrangements, particularly as such arrangements are adopted by commodity-exporting developing countries.

References:
Paul Cashin, C. John McDermott, and Alasdair Scott, 2002, "Booms and Slumps in World Commodity Prices," Journal of Development Economics, Vol. 69, pp. 277–96.
Paul Cashin and C. John McDermott, 2002, "The Long-Run Behavior of Commodity Prices: Small Trends and Big Variability," IMF Staff Papers, Vol. 49, pp. 175–99; http://www.imf.org/External/Pubs/FT/staffp/2002/02/cashin.htm
Yu-chin Chen and Kenneth Rogoff, 2002, "Commodity Currencies and Empirical Exchange Rate Puzzles," IMF Working Paper 02/27 (Washington); http://www.imf.org/external/pubs/cat/longres.cfm?sk=15638.0
Michael Mussa, 1986, "Nominal Exchange Rate Regimes and the Behavior of Real Exchange Rates: Evidence and Implications," Carnegie-Rochester Series on Public Policy, Vol. 25, pp. 117–213.
Carmen Reinhart and Kenneth Rogoff, 2002, "The Modern History of Exchange Rate Arrangements: A Reinterpretation," NBER Working Paper 8963 (Cambridge, Massachusetts: National Bureau of Economic Research).

Paul Cashin is a Senior Economist and Ratna Sahay is an Assistant Director in the IMF's Research Department.

Luis C�spedes is an Economist in the IMF's European I Department. This article is based on the authors' "Keynes, Cocoa, and Copper: In Search of Commodity Currencies," IMF Working Paper 02/223 (Washington, 2002), which is available at http://www.imf.org/external/pubs/cat/longres.cfm?sk=16194.0. IMF data on world commodity prices can be found at: http://www.imf.org/external/np/res/commod/index.asp