Originally posted on LinkedIn by Erico Matias Tavares
Dr. David Jacks is the Chair and Professor of Economics as well as an Associate Member of the School for International Studies at Simon Fraser University. He is also a Faculty Research Fellow at the National Bureau of Economic Research (“NBER”) in Cambridge, MA. He has also served as a consultant for Asia Pacific Economic Cooperation, the Richard Chandler Corporation and Tsao Family Holdings.
Dr. Jacks’ research spans the areas of economic and financial history, as well as international trade and finance. He has been invited to give presentations at over 75 conferences, organizations and universities in 17 countries and has been a visitor at 9 universities. His research has been funded by the Economic and Social Research Council of the UK, the French National Research Agency, the National Science Foundation of the US and the Social Science and Humanities Research Council of Canada.
He received a Bachelor in Economics and History (summa cum laude) from the University of Memphis, followed by a Masters in Economics from the University of Memphis and a Masters in Economic History (with distinction) from the London School of Economics. He received his PhD in Economics from the University of California, Davis.
Erico Tavares: Dr. Jacks, thanks for being with us today. You have recently published “From Boom to Bust: A Typology of Real Commodity Prices in the Long Run”, a paper which looks at price performance for forty commodities over 160 years. Contrary to popular belief, your research showed that the commodities sector as a whole has been in an uptrend for the better part of the 20th century, despite some big swings in the interim. Can you talk about your approach and key findings?
David Jacks: I like motivating this topic by first considering a famous bet which emerged from an academic debate in the 1970s. It begins with Paul Ehrlich, a biologist, winning acclamation and fame for his book, The Population Bomb. As the name would imply, Ehrlich saw the prodigious growth in human population over the past two centuries as a big problem in a world of finite resources. In his mind, population growth in time would hit a hard constraint, and humankind would be consigned to a fate of conflict, disease, and famine.
One of his biggest intellectual opponents was an economist by the name of Julian Simon. They sparred for the better part of a decade with Simon finally responding with his book, The Ultimate Resource. As the name would imply, this turned Ehrlich’s reasoning on its head: yes, we live in a world of finite resources but humanity in all of its creativity and ingenuity represented the ultimate resource, one which would figure out better means for both extracting and substituting away from scarce resources.
This gave rise to the Simon-Ehrlich wager, a bet on the direction of commodity prices from 1980 up to 1990. The terms were such that they would essentially construct an equally-weighted portfolio of chromium, copper, nickel, tin, and tungsten and track its performance over time. For every percentage point decline in the portfolio, Ehrlich would pay $10. And for every percentage point increase in the portfolio, Simon would pay $10. At the end of the day, Ehrlich lost decisively by writing Simon a check for $600, representing a nominal decline in the value of that equally weighted portfolio of 60%.
In the past few years, large increases in commodity prices once again lead to the view that the world was quickly running out of key materials. The necessary consequence of this scarcity is that economic growth must grind to a halt. But for others, this view is misguided as history suggests otherwise: that is, going long commodity prices is equivalent to shorting human ingenuity. Thus, we were basically watching the re-emergence of the Simon-Ehrlich debate 30 years later.
My contribution to this debate is that we have to start with the idea that commodity prices are cyclical. That is, commodity prices have both trends and cycles which may be long in duration. Because of this, long-run patterns can be easy to miss because we have a natural tendency to confuse cycles for trends.
As cycles and trends can span decades, we need very long-run commodity price data to accurately detangle the two. And as cycles and trends can differ across goods, we need a wide range of commodity price data. To these ends, I have collected annual price data for 40 commodities back to 1900, representing 8.72 trillion USD of production in 2011.
The bottomline of my research comes in two parts: 1.) real commodity prices increased by roughly 0.75% per year in real terms from 1950; 2.) 2012 marked the peak of above-trend pricing, that is, the peak of the currently-evolving commodity price cycle which began in 1998.
ET: One important distinction that you make is that the uptrend has been much more pronounced in “commodities in the ground”, like petroleum, natural gas and metals, relative to “commodities to be grown”, like grains, which have actually been in a long term downtrend. What accounts for this difference in your opinion?
DJ: For “commodities in the ground”, most of the price increase has been from 1950. This upward trajectory has been driven by deregulation of key markets (in particular, petroleum but also to a lesser extent, gold). We also have to contend with much higher capital costs in the mining sector from having to go into more remote areas and deeper into the ground.
For “commodities to be grown”, most of the decline has been from 1970. This downward trajectory has been driven by radical improvements in crop science resulting in greater resistance and higher yields for grains and soft commodities.
ET: You divide historical price performance between a long term trend, and shorter term cycles, which oscillate around that trend. This is quite interesting because we sometimes get lost on short time movements without taking into consideration the long term perspective. What has driven these cycles, in some cases quite pronounced, over the last 100 years or so?
DJ: Talk of cycles may seem a little mystical to some of my economist colleagues, but it really is a story rooted in ECON 101. Generally, we can think of these cycles emerging from the interaction of: 1) surging demand related to episodes of mass industrialization and urbanization; and 2) acute capacity constraints in energy, metals, and minerals. The result is that we can see above-trend commodity prices for years, if not decades, on end because it takes substantial time to build additional capacity in some of these sectors.
ET: How stable are these cycles going forward? What is often found in capital markets is that the duration and magnitude of such cycles can vary a great deal going forward, and therefore using them as a predictor or forecast can produce quite inaccurate results.
DJ: I would agree with that assessment and would say that any predictions must be taken with a grain of salt. Having said that, the empirical distribution of past cycles suggested a typical length of 14 years. This coupled with an unambiguous start-date of the current cycle in 1998 made 2012 the most likely year for the peak. This has proven to be a relatively accurate hunch, given the performance in commodity markets over the past two years.
ET: The graphs above show the real price index since 1900 for the commodities sector as a whole, highlighting the trend (graph on the top left) and the cycles (graph on the top right). It seems that the cycles shown here are peaking. Is this saying that the commodities sector may correct even further from here?
DJ: I have incorporated the recent crash in petroleum prices into some preliminary (unreported) analysis. With a WTI price of $50 per barrel, this analysis suggests we are around two-thirds of the way there in terms of establishing a definitive floor for commodity prices. Again, this must be taken with a grain of salt since commodity markets are so unpredictable...and therefore, so much fun to study.
ET: OK, so let’s just focus on crude oil. The cycle shown in the graph on the top right seems to be peaking, but we are in a long term bull market correct?
DJ: That is correct. The world is hungry for energy in all of its forms. And a few years out from now, I have a hard time believing we will be looking at such low prices for petroleum. Indeed, my work in decomposing trends and cycles in petroleum suggests a long-run price of $75 per barrel for WTI.
ET: In contrast, the price of rice, a food staple particularly in Asia, has been in a big downtrend since the 1970s, as per the graph above. Your thoughts here?
DJ: This is actually a crop and a sector for which I am mildly bullish at the moment. The Chinese transition from fixed capital accumulation to a consumption-based economy —and suburbanization—is tentatively beginning. If the Communist Party of China is successful in this effort, we are likely to see an increase in demand for goods “to be grown” and a potential inflection in its long-run trend. Here, I emphasize that quality matters—both in terms of the form of consumption (e.g., the switch in caloric consumption away from grains and into animal products) but also in terms of the safety of sources (e.g. the switch in Chinese consumption away from domestically produced grains and into those sourced abroad).
ET: On the whole, what does your research suggest of where we are in the overall commodities cycle? And what implications does this have for major producers, such as Australia, Canada and many emerging markets?
DJ: It suggests that the boom years of the past decade and a half were the exception and not the rule. Australia and Canada will have a bit of rough patch in the years to come, but will manage through as they always do. The much touted growth prospects of many of the BRICs will prove to be nothing more than a commodity-boom-fuelled mirage.
ET: That’s quite concerning, and we may be seeing some of that already playing out right now. Final question, what else are you working on right now?
DJ: I have some work addressing the question of what drives commodity prices in the long run, particularly in the context structural VAR models. These allow us to decompose fluctuations in commodity prices into global demand shocks, commodity supply shocks, and an inventory/speculative demand shock. The first results are interesting in that they suggest supply shocks just are not that big of a deal in the long run and that demand in its various form seems to be running the show.
Finally, I am very interested in the theme of cyclical resource investment and see obvious implications for investment decisions on the ground. At the same time, I am lacking sufficient data and time to have made much progress on this front.
ET: This has been great, thank you very much for sharing your thoughts with us today. Keep up the great work!
DJ: Thanks for giving me the opportunity to speak to a topic which is near and dear to my heart!