Commodity Supercycles & Metal Prices
Reviewing 100 years of the research on commodity supercycles
The concept of prices rising and falling in cyclical patterns has been around since at least the late nineteenth century.
Writing in the 1920s, Russian economist Nikolai Kondratiev examined commodity prices, along with other factors, such as industrial production, interest rates, and foreign trade, to conclude that prices ebbed and flowed over regular periods of about 40 to 60 years. He believed the stimulus behind these long-term price cycles was technological change and innovation.
His explanation for these long-term fluctuations was technological change.
Influenced by Kondratiev, the Austrian economist Joseph Schumpeter wrote in the 1930s of overlapping cycles of differing duration in. Kondratiev cycles came and went every five decades or so, whereas, Juglar cycles averaged about nine years and Kitchin cycles came and went, on average, every three to five years.
Different factors, Schumpeter suggested in Business Cycles (1939), drove the various cycles, but he agreed with Kondratiev's assessment that technological innovation was the main driver behind long-term price cycles.
More recent empirical analyses have looked specifically at commodity price cycles and long-term price trends for specific metals. While many economists disagree with the existence of regular economic cycles, there is much research supporting the idea of cyclical commodity price trends.
In 1950, Hans Singer and Raul Prebisch explained Britain's improving terms of trade vis-à-vis developing countries as due to a long-term drop in the real value of primary commodities relative to manufactures.
After 2000, the discussion of commodity price cycles was pushed to the forefront due to steadily advancing metal, energy and agricultural product prices, with many arguing that China's industrialization was driving a new commodity supercycle. Indeed, it was difficult to pick up the business section of a newspaper or magazine in the years leading up to the 2008 financial crisis without seeing some sort of reference to the impact of China's development on commodity prices.
Alan Heaps, a commodities analyst at Citigroup, wrote an insightful piece in 2005 arguing that China's ongoing industrialization and urbanization was the stimulus for a new commodity supercycle.
Heaps analysis identified two commodity supercycles over a period of 150 years prior to 2005. The first was driven by economic growth in the US and ran from the late 1800s into the early 1900s. The second was spurred by post-war reconstruction and Japanese economic expansion and ran from 1945 to 1975.
The analyst also saw the beginnings of a third cycle, brought about by materials intensive economic growth in China.
Material intensive economic growth is the driver behind commodity supercycles according to Heaps. As a maturing economy shifts from one oriented towards infrastructure and manufacturing to more of a service-based economy, this cycle comes to an end. This suggests that demand - not supply or technological change - is responsible for the rise and fall of prices.
Looking more specifically at copper prices, the Citigroup analyst recognized the importance of China's urbanization, industrialization and fixed capital formation in overall demand, demand growth and intensity of use for copper.
Contrary to Prebisch and Singer, Heaps did not believe that innovation would result in long-term real prices trending downward.
Post-financial crisis writing has put the recent commodity price rally in more context. Bilge Erten and José Antonio Ocampo's empirical analysis of about 30 non-oil commodities made use of prices from between 1865 and 2010.
Looking specifically at the metals, which included aluminum, copper, iron ore, lead, nickel, silver, tin and zinc, the researchers identified four supercycles ranging in length from 30 to 40 years. These cycles ran from 1885 to 1921, 1921 to 1945, 1945 to 1999 and from 1999 onwards to the time of writing (2013).
Erten and Ocampo also found a relationship between metal prices and world GDP, suggesting that global output acceleration (growing demand) drives commodity price growth, with metal prices being particularly sensitive to changes in growth.
In support of Prebisch and Singer's work, Erten and Ocampo's research showed that real mean prices at the end of each cycle were lower than the mean price at the end of the previous cycle. This trend was particularly pronounced for metals and agricultural products.
One final piece of research worth looking at is David S. Jacks' 2013 paper for the National Bureau of Economic Research (NBER), which also looked at commodity price supercycles as well as short-term boom and bust cycles, akin to Kitchin cycles first theorized nearly a century earlier.
Jacks' research examined 30 commodity prices between 1850 and 2010, including seven metals (aluminum, copper, lead, nickel, steel, tin and zinc), five minerals (bauxite, iron ore, chromium, manganese, and potash) and two precious metals (gold and silver).
Like others, he found trends supporting the existence of a number of 10- and 35-year price supercycles that were driven by demand brought about by mass industrialization and urbanization. However, Jacks saw these as long-term cycles are comprised of numerous short boom-bust cycles that range in length from one to five years and define price volatility.
Price booms could be associated with real price spikes of 50 to 100 percent above the long-run trend prices, while commodity price busts reflected real prices drops of 30 to 50 percent below the long-run trend.
Jacks suggests that these boom-bust cycles have become longer and larger since 1950 as a result of floating nominal exchange rates.
This look at boom-bust cycles helps to explain short-term commodity price volatility more effectively than discussions of decades-long supercycles.
In order to examine the relationship between commodity cycles and economic growth, Jacks looked at Australia, whose economy is highly dependent upon natural resources. Between 1900 and 2012, 14 commodities made up 43 percent of the country's exports. The statistical analysis from this period shows that price booms, on average, boost Australia's GDP growth by over six percent, while the busts cut GDP growth by over eight percent from the long-term real trend.
In sum, what we can take from the research is:
- There is strong evidence of regular multi-decade cycles of rising and falling commodity prices.
- That these commodity price supercycles are primarily demand driven and correspond with global output growth, as well as large-scale industrialization and infrastructure spending.
- A new supercycle, driven by China's infrastructure-focused economic growth, began in the mid- to late- 1990s and likely peaked between 2007 and 2013.
- Metal prices are particularly reactive to the increase in demand brought about by greater intensity of use, as reflected by the price volatility of copper in the post-2000 period.
- Short-term price volatility, characterized by boom-bust cycles, may be increasingly pronounced and longer than in previous eras.
While there may be strong evidence for the existence of commodity price supercycles, predicting how the current cycle will run its course and when the next may begin will forever be a topic of conversation among those in the metals business.
Erten, Bilge and José Antonio Ocampo.."Super-cycles of commodity prices since the mid-nineteenth century". DESA Working Paper No. 110. February 2012
Jacks, David S. "From Boom to Bust: A Typology of Real Commodity Prices in the Long Run". National Bureau of Economic Research Working Paper 18874. March 2013.
Heap. Alan. China - The Engine of a Commodities Super Cycle. Citigroup. March 31, 2005.
Follow Terence on Google+