Commodity terms of trade: The history of booms and busts(1)

Nikola Spatafora (Senior Economist) and Irina Tytell (Economist), Research Department, The International Monetary Fund

(solamente en inglés)

How do commodity-price booms affect the economic performance of commodity exporters? This column presents comprehensive new data on country-specific commodity terms of trade. It finds that, on average, countries grow nearly 2 percentage points faster during booms than during busts. But policy plays an important role — sharp currency appreciations and large government deficits are associated with lower growth.

How commodity prices affect the macroeconomic performance of commodity-exporting countries — particularly developing countries — has long been the subject of heated debate. Some studies find that commodity booms raise growth, while others suggest a “resource curse” that undercuts sustainable growth.

In recent years, this debate has regained momentum. As shown in Figures 1 and 2, the rapid growth observed in many developing countries from the start of the 2000s until the beginning of the current financial crisis coincided with a sharp increase in commodity prices.

Until the turn of the 21st century, commodity prices had been in general decline for some decades. During mid-2008, commodity prices rose by about 75% in real terms on average. By historical standards, this latest commodity boom was broad-based and sustained, with the prices of most commodities rising sharply (especially energy and industrial inputs, including agricultural raw materials and metals, but also major food crops and some beverages). After the financial crisis erupted, commodity prices fell rapidly, and they have since remained very volatile. It remains to be seen to what extent strong commodity price pressures re-emerge as the global economy recovers.


Figure 1. GDP growth: Emerging and developing economies

Figure 2. Commodity prices indices (2005=100)


New evidence on the effect of commodity-price booms

Figures 1 and 2 raise two crucial questions. First, looking back, to what extent was the growth acceleration in developing countries a by-product of the commodity boom?

Second, and related, how will developing countries’ future growth trajectories depend on the outlook for commodity prices?

Unfortunately, existing analyses of commodity booms generally rely on unsatisfactory price data, either a country’s aggregate terms of trade, or else simply the price of one or two key commodity exports.

In a recent IMF Working Paper (Spatafora and Tytell 2009), we use the first comprehensive dataset on country-specific commodity terms of trade. While Deaton and Miller (2006) and Cashin et al (2004) construct country-specific commodity export prices in a similar way, the terms-of-trade measure used in our study takes into account both commodity export and import prices. Moreover, our study adjusts for the importance of commodities in overall trade of each country.

Our study identifies country-specific commodity-price cycles. It dates and characterises commodity booms and busts for a sample of more than 150 countries over nearly 40 years starting in the early 1970s. Additionally, we begin the task of harvesting this dataset to analyse the link between commodity-price cycles and macroeconomic performance, how this link depends on various policy variables, and whether the experience of the past few years suggests a significant break with previous historical patterns.

Frequent booms and small busts

Some of our key initial findings are as follows. Regarding the nature of commodity-price movements, commodity-price booms tend to be larger than busts. On a cautionary note, however, around 1/3 of all booms are followed by busts (and, likewise, around 1/3 of all busts are followed by booms). Further, the larger the boom, the larger the subsequent bust.
Turning to the impact of commodity-price booms, we find that median output growth is nearly 2 percentage points per annum higher during booms than during busts. That being said, policy plays an important role in determining the response to booms. In particular, during both booms and busts, large real currency appreciations are associated with significantly lower growth. Also, the larger the pre-boom government deficit, the smaller the growth during the subsequent boom. One interpretation is that larger initial deficits increase the potential for crowding-out of private spending.

Finally, it appears that higher growth during the latest commodity-price cycle was at least partly due to non-global factors. Rather, it reflected factors specific to those countries that experienced a commodity terms-of-trade boom. Such factors included smaller real currency appreciations than in the past, as well as better initial conditions –including stronger initial fiscal positions in particular.

Put differently, improved policy frameworks and policy responses helped countries benefit from improved terms of trade to a greater extent than in the past. This gives grounds for optimism looking ahead. Still, the commodity booms of the 2000s were also, on average, quite long and, in particular, lasted longer than in the past — a pattern that may not be repeated in the future.


Cashin, Paul, Luis F Cespedes, and Ratna Sahay (2004), “Commodity Currencies and the Real Exchange Rate”, Journal of Development Economics, 75:239-268.
Deaton, Angus, and Ronald Miller (1996), “International Commodity Prices, Macroeconomic Performance, and Politics in Sub-Saharan Africa”, Journal of African Economies,5:99-191.
Spatafora, Nikola and Irina Tytell (2009), “Commodity terms of trade: The History of booms and busts”, IMF Working Paper 09205, September.