"di h Commodity Futures Exchanges"
di h Commodity Futures Exchanges Historical Evolution and New Realities John Baffes John Baffes The World Bank Forum for Agricultural Risk Management in Development Forum for Agricultural Risk Management in Development Annual Conference 2011 http://www.agriskmanagementforum.org/farmd/ Price Volatility and Climate Change Implications for the Agricultural Risk Management Agenda June 9‐10, 2011 Zurich, Switzerland 1 OUTLINE o Milestones in the evolution of Commodity Futures Exchanges (slides 3‐9) g ( ) The New Reality (slides 10‐15) o The “New Reality” (slides 10 15) Does the new reality matter? (slides 16 22) o Does the new reality matter? (slides 16‐22) 2 y Milestones in the evolution of Commodity Futures Exchanges o Early history (17th and 18th century): Amsterdam (1695), Japan (1730, Dōjima rice market in Osaka). o Information travels faster than commodities (1840‐1865): Following the introduction of the steamship and the telegraph, information traveled faster than commodities. commodities o Globalization (1865‐1940): After the installation of the transatlantic cable, futures exchanges become global institutions. The post WWII downfall (1940 1970): Policy interventions at national and o The post‐WWII downfall (1940‐1970): Policy interventions at national and international levels and, often, outright prohibition of futures contracts. o The rebirth (1970‐2000): Following the collapse of Breton Woods numerous financial futures are launched. Options trading is improved (Black‐Scholes formula); policy reforms in commodity markets remove an important impediment to futures trading; CFTC is created; a crude oil market is established; international commodity agreements collapse. 3 E l hi t ff t t di Early history of futures trading g gg o Most historians agree that futures trading goes as far back as 1695 where sporadic futures trading was taking place in Amsterdam (sources cited in Goss and Yamey 1978). The Dōjima rice market in Osaka is credited with the birth of o Th Dōji i k ti O k i dit d ith th bi th f future trading c. 1730 (Markham 1987). o The contracts at the Dōjima market were centrally cleared. o They were derived from an underlying asset with standard characteristics in terms of size and quality. o They were margins and mark‐to‐market accounting rules. y g g o At the end of the trading session, a settlement price was achieved, called the “fuse cord price”. It was used as the next session’s opening p price. 4 Mid 19th t F t h t k h Mid‐19th century: Futures exchanges take shape Two inventions (telegraph and steamship) changed the price formation process in futures exchanges by taking into account, for the first time, non‐local demand and supply conditions. For example: o The telegraph. Buffalo and New York City were connected by telegraph in late 1846 Then a flurry of activity took place in 1847 in various commodities including 1846. Then, a flurry of activity took place in 1847 in various commodities including flour, corn, wheat, and eggs (Williams 1982). o The steamship. With the expanded use of the steamship during the 1840s, the time to cross the Atlantic was reduced from two months to two weeks. This allowed demand and supply conditions of cotton in the US, the major supplier to the UK at the time, along with cotton samples to arrive much earlier than cotton itself. Based on this information, merchants in Liverpool were trading “to arrive” or in transit contracts more than one month prior to the physical transaction “in transit” contracts, more than one month prior to the physical transaction (Dumbell 1927). 5 g The transatlantic cable renders futures exchanges global institutions o The first attempt (unsuccessful) to connect old and new world by cable was made in 1854 with a proposed land link to pass through Alaska, Bering Straits, Siberia, Eastern Europe, and England. o A transatlantic cable connecting US and England was installed in 1857 (4,200 kilometers). It never transmitted a single message. A second cable as installed a kilometers) It never transmitted a single message A second cable as installed a year later. It only transmitted around 500 messages at the cost of $10 per 5‐letter word (equivalent to 82 kilograms of cotton lint at the time). y o The first successful cable was installed in 1864. It was financed by John Pender, a , textile merchant from Manchester. The funds of this and subsequent cables were financed by the British textile industry which was the first industry to realize the benefits of instantaneous communication (they were interested in the supply conditions of cotton in the US). The cost of a 5‐letter word fell to 25 cents by 1888. conditions of cotton in the US) The cost of a 5 letter word fell to 25 cents by 1888 o By the end of the 1880s, five cotton futures exchanges were connected by cable (New York, New Orleans, Liverpool, Havre, and Alexandria) were trading cotton g g futures contracts. Futures exchanges become global institutions. Source: Hugill (1999), Garside (1925). 6 P t WWII d f ll Post‐WWII downfall The period between 1880 and WWII was the “golden age” of futures markets. But the importance diminished after WWII: Some exchanges were shut down while others diminished in importance. Attempts to “resurrect” them were mostly unsuccessful because: Farm programs in the US (where most of the exchanges were flourishing) made o Farm programs in the US (where most of the exchanges were flourishing) made most futures contracts redundant. o The introduction of the Common Agricultural Policies in Europe, especially its g , stock‐holding mechanisms, made futures markets redundant. o International Commodity Agreements often worked against the creation of futures markets. o Central planning policies in many countries did not allow futures trading (Eastern Europe, USSR) o Strong state control in in many countries prohibited futures trading . Examples include China, India, and Egypt. 7 When governments intervene commodity futures exchanges suffer 10 100% CCC‐held stocks as a percent of total stocks (right axis) NYCE volume, million tons (left axis) 8 80% 6 60% 4 40% 2 20% 0 0% 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 Source: Hieronymus (1977) and US Department of Agriculture Th bi th f th 1970 The rebirth of the 1970s o Breton Woods arrangement collapses. In turn numerous markets are creating the need to price discovery and managing risk. Exchanges rates fluctuate: exchange and interest rate markets are created. Gold price is delinked from the dollar: a gold market emerges. The oil crisis of 1972: a crude oil market emerges. o Pricing of options becomes easier (Black‐Sholes option pricing formula) o With the introduction of financial futures, the stakes become high and the , g need for regulation emerges. The US creates the Commodity Futures Trading Commission, the regulatory body that oversees the functioning of futures markets. o Throughout the 1980s and the 1990s, agricultural policies become more “market friendly" first by replacing stock‐holding mechanisms by price supports and latter by replacing price supports with decoupled mechanisms. 9 Wh t h d th “ lit ”? What shaped the “new reality”? o Literature: Numerous influential authors have argued that commodities may be a profitable asset class. o New players and the financialization of commodities: Many funds (investment, hedge, pension, and sovereign wealth) began including commodities in their portfolios in order to diversify their holdings and receive higher returns (not portfolios in order to diversify their holdings and receive higher returns (not unique to commodities). o The financialization of futures exchanges: They changed from member‐holding ( q ) g p ( institutions (the equivalent of a Credit Union) to exchange‐listed companies (the equivalent of an investment bank). o Deregulation: The “2000 Commodity Modernization Act” (not unique to commodities, e.g., repeal of Glass‐Steagall Act). o Liquidity: Low policy rates, stimulus packages, government spending, and quantitative easing (not unique to commodities). o Technology: Electronic trading, ETFs, index funds, and information technology made commodity investment accessible (not unique to commodities). made commodity investment accessible (not unique to commodities) 10 Th lit t The literature o Rogers (2004): Argued that commodities should be part in every investors portfolio. Creator of the “Rogers commodity index” with high credibility within the investment community, especially because he anticipated both the early 1970s and mid‐2000s commodity price booms. Heap (2004): Argued that commodities (mostly extractive) have entered a super‐ o Heap (2004): Argued that commodities (mostly extractive) have entered a super cycle, that is a prolonged period of high prices primarily demand‐driven by emerging economies (especially China). ( ) p yp o Gordon and Rouwenhorst (2006): Compared a hypothetical commodity‐based y with bond and equity portfolios and concluded that the risk premium on commodity futures is essentially the same as equities, commodity future returns are negatively correlated with equity and bond returns. Cuddington and Jerrett (2008): The confirmed the presence of a super‐cycle in o C ddi d (2008) h fi d h f l i various metals. 11 p y New players and the financialization of commodities o Precious metals: Historically, gold and silver have been the only commodities which have been used as hedging instruments by agents who have nothing to do with the “precious metals industry”. Thus, gold prices depend more on inflationary expectations, armed conflicts, crude oil supply disruptions, and central bank intentions and less so on production and consumption of gold. intentions and less so on production and consumption of gold. o Other commodities: During the early 2000s, non‐precious metal commodities were added in the portfolios of various investors (investment, hedge, pension, and sovereign wealth funds). Such inclusion was primarily driven by diversification and higher return objectives. o New investments: As of end of 2010, about US$ 380 were invested in commodities, two thirds in energy. Some characteristics of the new investment: o S h i i f h i Funds have, to date flowed mostly in, not out. They invest on the basis of fixed weights or past performance criteria. Most sovereign wealth funds (key source of index funds) are commodity‐based Most sovereign wealth funds (key source of index funds) are commodity‐based. These funds are small relative to their holding but large relative to commodity markets. 12 Th fi i li ti ff t h The financialization of futures exchanges o For most of their history, futures exchanges have been non‐profit organizations serving the needs of their members. o The New York Cotton Exchange (NYCE) was established in 1870 and operated as an independent entity until 1998. It merged with the Coffee, Sugar, and Cocoa Exchange (CSCE), which was founded in 1882. They formed the New York Board of Exchange (CSCE) which was founded in 1882 They formed the New York Board of Trade (NYBOT). In 2007, NYBOT became a unit of the Intercontinental Exchange (ICE). ICE has been listed at NYSE since 2005. g ( ), , o The Chicago Board of Trade (CBOT), established in 1848, is the world’s oldest commodity futures exchange. It merged with the Chicago Mercantile Exchange in 2007 to form the CME Group. CME has been listed at NASDAQ since 2003. o Interestingly, Carlton (1984)‐‐a Chicago law professor‐‐noted: “One can view a futures exchange as a type of firm. Although exchanges are non‐profit and have f h f fi lh h h fi dh many distinctive institutional features, their objective is to succeed by generating volume.” Now, futures exchanges are firms. o Now futures exchanges are firms 13 ICE has been listed at NYSE since 2005 ICE has been listed at NYSE since 2005 CME has been listed at NASDAQ since 2003 CME has been listed at NASDAQ since 2003 D th “ lit ” tt ft ll? Does the “new reality” matter after all? q There are two questions of interest Question # 1: Has the “new reality” contributed to better facilitation of risk or more efficient hedging instruments? Answer: Yes Answer: Yes. Question # 2: Has the “new reality” improved the price discovery process? Answer: Not clear. Th i diff The views differ, the opinions are very strong, the empirical evidence is, at best, mixed, and h i h b li i i d b h fi i l d l h “ the issue has been politicized by the financial and popular press; the “new reality” is often discussed in the context of “speculation” without explicitly stating what it means. 16 Th i t The views are strong ON THE ONE HAND ON THE OTHER HAND o g Krugman: “… a futures contract is a bet o Soros: commodity index trading is Soros: “… commodity index trading is about the future price. It has no, zero, nada intellectually unsound, potentially direct effect on the spot price” (NY Times, destabilizing, and distinctly harmful in its June 23, 2008). economic consequences” (US Congress o Wolf :“if speculation were raising prices testimony , June 3, 2008). b h dl l ld above the warranted level, one would expect o Calvo: “Increases in commodity prices during to see inventories piling up rapidly, as supply 2007/08 were a result of portfolio shift exceeds the rate at which oil is burned. Yet against liquid assets by sovereign investors there is no evidence of such a spike in and sovereign wealth funds, partly triggered inventories” (FT, May 13, 2008). ( , y , ) by lax monetary policy (weblog 2008) by lax monetary policy” (weblog, 2008). o Frankel: “The evidence does not support the o Rubini: “Improving fundamentals … justify oil claim that speculation has been the source going from $30 to maybe $50. The other $30 of, or has exacerbated the price increases” is all speculative demand feeding on it— (weblog, July 25, 2008). speculation and herding behavior.” (FT, p g ( , o Verleger: “… the 2007/08 crude oil price November 1, 2009). spike was caused by the incompatibility of o Khan: “While market fundamentals played a environmental regulations with the global key role in the run‐up of the oil prices after crude supply. Speculation, he argued, had 2003, the price increase of 2008 was a price nothing to do with the price increases” (CFTC nothing to do with the price increases (CFTC bubble.” (2008). b bbl ” ( ) testimony, August 5, 2009). 17 , p But, the empirical evidence is weak: Some find little or no impact o IMF (2006): “Little evidence that speculative activity affects either price levels over the long run or i i i th h t price swings in the short run”” o Haigh, Hranaiova, and Overdahl (2007): ”Failed to observe a link between price volatility and changes in hedge find positions” o CFTC (2008): “Failed to find that changes of traders’ positions caused changes in crude oil prices” o IMF (2008): “No apparent systematic connection between financialization and price changes” o Büyükşahin, and Harris (2009): “Concluded changes of traders’ positions did not cause changes in oil prices” o Irwin, Sanders, and Merrin (2009): ”Positions of long‐only index funds do not lead futures prices” o Verleger (2009): “Money flows into oil contracts have not affected oil prices” o Till (2009): “The balance of outright speculators in the US oil futures and options markets was not excessive relative to hedging” o Kilian and Murphy (2010): “The results eliminate speculation as an explanation of the 2003‐08 Kilian and Murphy (2010): The results eliminate speculation as an explanation of the 2003 08 surge in oil prices” o Sanders and Irwin (2010): “The evidence that index fund positions impact returns across commodities is scant” o Stoll and Whaley (2010): Inflows and outflows from commodity index investment do not cause Stoll and Whaley (2010): “Inflows and outflows from commodity index investment do not cause futures prices” 18 Whil th fi d d t i t While others find moderate impact o Gilbert (2007): “Some evidence that index investment may have been responsible for raising some dit i d i th tb commodity prices during the recent boom” ” o Plastina (2008): ”Investment fund activity might have pushed cotton prices 14 percent higher” o Medlock and Jaffee (2009): “In order to diversify against falling $US, oil‐based index funds added oil into their portfolios, thus pushing crude oil prices up” o Robles, Torero, and Braun (2009): “Speculative activity might have been influential” o Gilbert (2010): “Index‐based investors appear to have inflated food commodity prices” o Silvennoinen and Thorp (2010): “Higher commodity returns volatility is predicted, among others, by financial traders open positions” o Tang and Xiong (2010): “Commodity price volatility of recent years was related to the presence of index investors” o Singleton (2011): “There is an economically and statistically significant effect of investor flows on p futures prices” o UNCTAD (2011): “Due to their financial strength, financial investors can move prices in the short term; this leads to increased volatility, which can harm markets” 19 The inconclusive nature of the findings is not surprising given the numerous factors that affected commodity prices during the boom factors that affected commodity prices during the boom 2001‐05 2006‐10 change Crude oil prices (US$/barrel) 33 75 +130% Exchange rates (US$ against a broad index of currencies) 119 104 ‐13% Interest rates (10‐year US Treasury bill) 4.7 4.1 ‐14% Funds invested in commodities ($ billion) 30 250 +730% GDP growth, low and middle income countries (annual %) 5.0 5.8 +16% Industrial production, low and middle income countries (annual %) Industrial production low and middle income countries (annual %) 63 6.3 7.1 71 +13% Stock‐to‐use ratio of maize, wheat, and rice (months of consumption) 3.2 2.5 ‐20% Biofuels production (million of barrels per day equivalent) 0.4 1.3 +200% Average yields of wheat, maize, and rice (tons/hectare) 3.8 4.0 +7% Growth in yields (% change per annum) 1.4 1.0 ‐32% Natural disasters (droughts, floods, and extreme temperatures) 374 441 +18% 20 Source: World Bank, IMF, FRED, USDA, CRED, IEA Y t th i it i Yet, there are some inconsistencies o Verleger (2009): “The 2007/08 oil price spike was caused by the incompatibility of environmental l ti ith th l b l d il l S l ti h d thi t d ith i i ” regulations with the global crude oil supply. Speculation had nothing to do with price increases” Later, however, he acknowledged that the possible liquidation of future positions was one of the key reasons tied to the collapse of oil price from July to December 2008. QUESTION: If the decline in oil prices was caused by the liquidation of futures positions, shouldn’t the undertaking of such position have caused a proportional increase in oil prices? position have caused a proportional increase in oil prices? o Verleger (2010): “The stability of energy prices during December 2009 and January 2010 partly to the increase of passive investors that allocated a portion of their portfolios to commodities.” COMMENT: Therefore, passive investors affect prices! o IOSCO Task Force of Futures Markets (2009): It concluded that economic fundamentals rather than IOSCO Task Force of Futures Markets (2009): It concluded that economic fundamentals rather than speculative activity explains explained commodity prices during 2007/08. It recommended, among others, “to understand to role of speculation and gain a more comprehensive view of trading activities in the underlying markets.” QUESTION: If IOSCO does not understand the role of p g , y p p speculation and the nature of trading activities, why did the report conclude that speculation played no role? o Many authors have argued that the investments brought into futures markets by various funds added liquidity in these markets, thus enhancing the price discovery process. QUESTION: Some of q ( g g ) y g y the contracts under question (e.g., grains, cotton) have been traded for 120 years. Such longevity means that the contracts were liquid. Was there any need for additional liquidity during those 120 years and we did not know it? 21 ,p , Some final, personal, and non‐technical observations on the “new reality” o Historically, blaming speculation has originated from politicians and populists, a reflection of ideology or search for scapegoats to assign blame for unpopular price increases. Most economists and market participants, however, have been strong defenders of futures markets which are viewed (rightly) as the most important price discovery mechanisms. o The current debate has divided economists and market participants. o Most empirical studies follow a similar approach in the sense that they “correlate” a quantity variable (open interest, volume traded, or various ratios of commercial/non‐commercial positions) with a price variable (futures contract) positions) with a price variable (futures contract). o Most studies that find impact come from financial economists. o Although the empirical evidence is weak (2/3 of the studies find no impact), recent studies tend find more impact compared to earlier studies (perhaps because of more data and better i h i econometric techniques). ) 22 Thank you 23