In this article, we look at why commodity prices have a tendency to go through extended periods of boom and bust (a.k.a. super-cycles). Commodity price movements are important for Canada, since they help determine our terms-of-trade, employment, income, and inflation.
The asymmetric band pass filter is one, but by no means the only, method for identifying super-cycles in commodity price data. The results of this research support the view that there have been four broad-based commodity price super-cycles since the early 1900s.
One potential driver of commodity price super-cycles is through the interaction of large, unexpected shocks to demand and slow-moving supply responses. This is widely seen to be the case with the most recent super-cycle, which was driven by rapid growth in China and other emerging market economies (EMEs). We have now entered the downswing phase of the current super-cycle. However, depending on how events unfold, we could see a new upswing phase within the current cycle (or a cycle-within-a-cycle).
In the second half of 2014, oil prices experienced a sharp decline, falling more than 50 per cent between June 2014 and January 2015. A cursory glance at this oil price crash suggests similarities to developments in 1986, when the price of oil declined by more than 50 per cent, initiating an episode of relatively low oil prices that lasted for more than a decade. This analytical note compares the 1986 price decline with the current episode more closely, and its key findings suggest important differences. While oil demand had been falling in the beginning of the 1980s, demand growth currently is being sustained by emerging economies and is projected to be more stable. Also, spare production capacity is significantly smaller today. Due to higher decline rates and shorter investment cycles of unconventional production, current supply is expected to adjust faster to low prices and reductions in investment spending. As long as oil demand from emerging economies remains robust, increases in production will require additional investment in high-cost production. The cost of this incremental production points to higher prices in the medium term than were observed in 2015, although the potential size of a price increase is limited because of ongoing cost-cutting initiatives and technological advances. Due to the fundamental changes in the oil market, it is unlikely that a decade of low oil prices—similar to the experience following the 1986 oil price crash—will repeat itself.
Commodity-equity return co-movements rose dramatically during the Great Recession. This development took place following what has been dubbed the “financialization” of commodity markets. We first document changes since 1995 in the relative importance of financial institutions’ activity in agricultural futures markets. We then use a structural VAR model to ascertain the role of that activity in explaining correlations between weekly grain, livestock, and equity returns in 1995-2015. We provide robust evidence that, accounting for shocks which are idiosyncratic to agricultural markets, world business cycle shocks have a substantial and long-lasting impact on the latter’s co-movements with financial markets. In contrast, changes in the intensity of financial speculation have an impact on cross-market return linkages that is shorter-lived and not statistically significant in all model specifications.
We examine whether herding among speculators in U.S. crude oil futures markets affects market prices and volatility. Using detailed data on the positions of hedge funds and swap dealers from 2005-2009, we find little evidence that herding destabilizes the crude oil futures market. To the contrary, herding among speculative traders is negatively correlated with contemporaneous volatility and does not lead next-day volatility. Our impulse-response analysis shows that market regulators should monitor herding since a shock to herding among all groups may lead to price changes, and, in the case of hedge funds, may lead to increased volatility. Interestingly, however, increased swap dealer herding actually dampens crude oil price volatility.
We document that, starting in the Fall of 2008, the benchmark West Texas Intermediate (WTI) crude oil has periodically traded at unheard-of discounts to the corresponding Brent benchmark. We further document that this discount is not reflected in spreads between Brent and other benchmarks that are directly comparable to WTI. Drawing on extant models linking oil inventory conditions to the futures term structure, we test empirically several conjectures about how calendar and commodity spreads (nearby vs. first-deferred WTI; nearby Brent vs. WTI) should move over time and be related to storage conditions at Cushing. We then investigate whether, after controlling for macroeconomic and physical market fundamentals, spread behavior is partly predicted by the aggregate oil futures positions of commodity index traders.
We analyze the role of hedge fund, swap dealer and arbitrageur activity in a Markov regime-switching model between high volatility bear markets and low volatility bull markets for crude oil, corn and Mini-S & P 500 index futures. We find that these institutional positions reflect fundamental economic factors within each market. More importantly, institutional positions also contribute incrementally to the probability of regime changes displaying the synchronization patterns modeled in Abreu and Brunnermeier (2002; 2003). Conditioning on hedge fund activity and arbitrageur activity significantly improves our probability estimates, demonstrating that institutional positions can be useful in determining whether price trends resembling bubble patterns will continue or reverse.
We construct a uniquely detailed, comprehensive dataset of trader positions in U.S. energy futures markets. We find considerable changes in the make-up of the open interest between 2000 and 2010 and show that these changes impact asset pricing. Specifically, dynamic conditional correlations between the rates of return on investable energy and stock market indices increase significantly amid greater activity by speculators in general and hedge funds in particular (especially funds active in both equity and energy markets). The impact of hedge fund activity is markedly lower in periods of financial market stress. Our results support the notion that the composition of trading activity in futures markets helps explain an important aspect of the distribution of energy returns, and have ramifications in the debate on the financialization of energy markets.
We employ data over 2005-2009 which uniquely identify categories of traders to test whether speculators like hedge funds and swap dealers cause price changes or volatility. We find little evidence that speculators destabilize financial markets. To the contrary, speculative trading activity largely reacts to market conditions and reduces volatility levels, consistent with the hypothesis that speculators provide valuable liquidity to the market. These results hold across a variety of products and suggest that hedge funds (with approximately constant risk tolerance as in Deuskar and Johnson ) improve overall market quality.
We provide direct empirical evidence that who trades helps explain the joint distribution of commodity and equity returns. Using a unique, comprehensive dataset of individual trader positions in 17 U.S. commodity and equity futures markets from 2000 to 2010, we document major changes in the composition of the open interest in commodity-futures markets during that period. We then show that the correlations between the returns on commodity and on equity indices increase significantly amid greater activity by speculators in general and one type of traders in particular – hedge funds. We find no such effect for other kinds of traders, including swap dealers and index traders. The impact of hedge fund activity is complex. In particular, it is lower during periods of financial market stress. Our results support the notion that the composition of trading activity matters for asset pricing, and have implications for the debate on the “financialization” of commodity markets.
OPEC producers, individually or collectively, often make statements regarding the “fair price” of crude oil. In some cases, the officials commenting are merely affirming the price prevailing in the crude oil market at the time. In many cases, however, we document that they explicitly disagree with the contemporaneous futures price. A natural question is whether these “fair price” pronouncements contain information not already reflected in market prices. To find the answer, we collect the “fair price” statements made from 2000 to 2009 by officials from OPEC or OPEC member countries. Visually, the “fair price” series looks like a sampling discretely drawn (with a lag) from the daily futures market price series. Formally, we use several methodologies to establish that “fair price” pronouncements have little influence on the market price of crude oil and that they supply little or no new news to oil futures market participants.
The coincident rise in crude oil prices and increased number of financial participants in the crude oil futures market from 2000-2008 has led to allegations that “speculators” drive crude oil prices. As crude oil futures peaked at $147/bbl in July 2008, the role of speculators came under heated debate. In this paper, we employ unique data from the U.S. Commodity Futures Trading Commission (CFTC) to test the relation between crude oil prices and the trading positions of various types of traders in the crude oil futures market. We employ Granger Causality tests to analyze lead and lag relations between price and position data at daily and multiple day intervals. We find little evidence that hedge funds and other non-commercial (speculator) position changes Granger-cause price changes; the results instead suggest that price changes precede their position changes.
The possibility that speculative trading destabilizes or creates a volatile market is frequently debated. To test the hypothesis that speculative trading is destabilizing we employ a unique dataset from the U.S. Commodity Futures Trading Commission (CFTC) on individual positions of speculators. While others have used a more aggregated version of our data, here we test, for the first known time, whether speculators cause, in a forecasting sense, price movements and volatility in futures markets and, therefore, destabilize markets. Our findings provide evidence that speculative trading in futures markets is not destabilizing. In particular, speculative trading activity reduces volatility levels.
We identify and explain a structural change in the relation between crude oil futures prices across contract maturities. As recently as 2001, near- and long-dated futures were priced as though traded in segmented markets. In 2002, however, the prices of one-year futures started to move more in sync with the price of the nearby contract. Since mid-2004, the prices of both the one-year-out and the two-year-out futures have been cointegrated with the nearby price. We link this transformation to changes in fundamentals, as well as to sea changes in the maturity structure and trader composition of futures market activity. In particular, we utilize a unique dataset of individual trader positions in exchange-traded crude oil options and futures to show that increased market activity by commodity swap dealers, and by hedge funds and other financial traders, has helped link crude oil futures prices at different maturities.
Amid the rise in commodity investing that started in 2003, many have asked whether commodities now move more in sync with traditional financial assets. Using daily, weekly and monthly data from January 1991 through November 2008, we provide evidence largely to the contrary. First, we apply dynamic conditional correlation and recursive cointegration techniques to the prices of, and the returns on, key investable commodity and U.S. equity indices. Compared to the 1991-2002 period, both short- and long-term relationships between passive commodity and equity investments are generally weaker after 2003. Even though the correlations between equity and commodity returns increased sharply in Fall 2008, during extraordinary economic and financial turbulences, they remained lower than their peaks in the previous decade. Second, we analyze the co-movements between equity and commodity returns in periods of extreme returns. We find little evidence of a secular increase in spillovers from equity to commodity markets during extreme events. Overall, our results suggest that commodities provide substantial diversification benefits to passive equity investors, but also that those benefits are weaker precisely when they are most needed.
We test the prevalence, sources and effects of herding among large speculative traders in thirty U.S. futures markets over 2004-2009. Using unique U.S. Commodity Futures Trading Commission (CFTC) data identifying daily trader positions we compare herding among hedge funds and floor market participants and find similar levels of herding across groups at slightly higher levels than in equity markets. We analyze the sources of herding and find that the number of traders, trading volume and floor-based markets are positively associated with herding. Notably, we find that the moderate levels of herding by hedge funds serve to stabilize, rather than destabilize, prices in futures markets.
The impact of news, events or volatility on the underlying market microstructure has been studied extensively in finance literature. Common to many of these studies is the arrival of "true" news. In this paper we use unique data on error trades that have occurred on the CBOT and CME and assess their impact on the underlying market microstructure. Critically, our research can be discerned from other research in that an error trade is, by definition, an event that was triggered by accident and is not "true" news. As such, an evaluation of "false news" stemming from human errors is unique. We find that despite the fact that participants are informed well after the event has occurred that the "news" turned out to be noise, market efficiency prevails in the sense that trading activity drives market prices back toward their pre error levels long before the exchange announces to the traders that an error has occurred. In line with the market microstructure (volatility) literature we also find that the composition of traders changes around the error event and some types of traders limit their trading. We also show that the types of orders placed by traders changes (more market orders relative to limit orders). While the increased dominance of electronic trading over the open-outcry form of trading has called regulatory attention to error trading policies, we identify no area of concern regarding these policies from a price discovery standpoint. Further, we find that market microstructure reactions to human errors seem to be similar to those found when "true news" is released.
Some of the world’s largest futures exchanges impose daily limits on the price movements of individual contracts. Using data from three of the most active US commodity futures contracts, we show that these price restrictions are largely ineffective because traders are able to take similar positions using other contracts. When price limits become binding on the futures market, the associated (but unrestricted) options market becomes the price discovery market: much of the trading that would have occurred on the futures market migrates to the options market, and options prices accurately predict the (unconstrained) futures price the next day. We also show that the presence of options mitigates the effect of price limits on information revelation by documenting that futures markets reflect more accurate information on days following limit hits when the associated options were trading on the previous day. Overall, our evidence suggests that price limits in US futures markets have little effect on prices when options markets exist.
In this paper an Information–Theoretic method for reconstructing noisy and blurry images is developed. Basically, the inverse problem is transformed into a generalized moment problem, which is then solved by an information theoretic method. This estimation approach is robust for a whole class of distributions and allows the use of prior information. The resulting method builds on the foundations of information-theoretic methods, uses minimal distributional assumptions, performs well and uses efficiently all the available information (hard and soft data). This method is computationally efficient. A number of empirical examples are presented.