Energy and Financial Markets Overview: Crude Oil Price Formation
Richard Newell, Administrator May 5, 2011
www.eia.gov EIA’s Energy and Financial Markets Initiative • Collection of critical energy information to improve market transparency – improved petroleum storage capacity data – other improvements to data quality and coverage
• Analysis of energy and financial market dynamics to improve understanding of what drives energy prices – internal analysis and sponsorship of external research
• Outreach with other Federal agencies, experts, and the public – expert workshops – public sessions at EIA’s energy conferences – solicitation of public comment on EIA’s data collections
2 Richard Newell, May 5, 2011 Synthesis of current perspectives on oil price formation
3 Richard Newell, May 5, 2011 Who’s who in global oil markets • Producers – international oil companies (private) – national oil companies (government-controlled) • Consumers – individuals (autos and space-heating) – transportation fleets (trucking, shipping, air) – industrials (chemicals and plastics manufacturers) and power generation • Traders, hedgers, speculators, and investors – merchants: firms facilitating physical trade – financial market participants • index investment by pensions, endowments, and others • commodity trading advisors, hedge funds, and individuals • swap dealers: typically banks and merchants serving customers • Policymakers – demand-side (subsidies); supply-side (OPEC quotas); market regulation (position limits)
4 Richard Newell, May 5, 2011 Many factors influence the formation of oil prices and other energy prices
Supply Physical balancing Demand Affected by current • Inventories Affected by current conditions and future conditions and future expectations for: Markets & market behavior expectations for: • energy prices • Energy prices • energy prices • OPEC supply capacity ‐ spot • economic growth • usable spare capacity ‐ futures • industrial production • non-OPEC capacity ‐ options • goods transport • geopolitics ‐ spreads • personal transport • weather ‐ swaps • weather • E&P costs • Other financial markets • innovation in energy- • E&P investments using equipment ‐ other commodity prices • E&P innovations ‐ commodity investment ‐ currency exchange rates ‐ stocks and other assets ‐ interest rates
5 Richard Newell, May 5, 2011 EIA is actively examining the role of both supply-demand fundamentals and other factors in oil price formation
• EIA has asked several academic partners to conduct thorough reviews of economic literature regarding supply-demand fundamentals and the role of financial market speculation and investment in the oil-price formation process – the work is ongoing, but still at an early stage • Some researchers are finding evidence that factors including unexpectedly strong economic growth in China and stagnant supply were at least associated with, and may have contributed to, the sharp oil price run-up and subsequent decline during the 2007-2008 period • The researchers are also finding some evidence suggesting that the price run-up and decline may have been exacerbated by the formation and collapse of an oil price bubble, perhaps triggered by fundamental factors in both the oil market and the broader global economy • As discussed later in the presentation, both internal EIA and academic research is also addressing the major increase in oil derivatives trading, significant change in the composition of derivatives traders (such as the growth of swap dealers, hedge funds, and commodity index funds), and increased correlation of oil and other markets over the past several years
6 Richard Newell, May 5, 2011 Crude oil prices react to a variety of economic and geopolitical events
price per barrel (real 2009 dollars, quarterly average)
140 Global financial collapse Imported refiner acquisition cost WTI crude oil 120
Iran-Iraq War 100 Low spare capacity 80 Saudis abandon 9-11 attacks swing producer role 60 U.S. spare capacity Asian financial crisis exhausted 40 OPEC cuts quotas Iranian revolution 4.2 mmbpd 20 Iraq invades Kuwait OPEC cuts quotas Arab Oil Embargo 1.7 mmbpd 0 1970 1975 1980 1985 1990 1995 2000 2005 2010 Sources: EIA, Thomson Reuters
7 Richard Newell, May 5, 2011 World oil prices move together due to arbitrage price per barrel (real 2009 dollars, monthly average) 160
WTI crude oil 140 Brent crude oil MARS crude oil 120 Tapis crude oil Dubai crude oil 100
80
60
40
20
0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Sources: Bloomberg, Thomson Reuters
8 Richard Newell, May 5, 2011 Crude oil prices are the primary driver of petroleum product prices price per gallon (real 2009 dollars, monthly average) price per barrel (real 2009 dollars) 4.00 Forecast 160 Unplanned refinery 3.50 down-time 140 Post-hurricane 3.00 refinery repairs 120 2.50 100
2.00 Hurricane Katrina shuts down refineries & pipelines 80
1.50 60
1.00 40 gasoline wholesale (left axis) 0.50 diesel wholesale (left axis) 20 WTI crude oil (right axis) 0.00 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Sources: EIA, Thomson Reuters
9 Richard Newell, May 5, 2011 Oil demand
10 Richard Newell, May 5, 2011 Oil demand and consumption: What matters? • Oil consumption is determined by demand drivers and price
• Demand is closely related to the state of the global economy – oil is an input into industrial production and the transport of goods – household income and employment drive personal transport demand
• Demand in recent years has come increasingly from non- OECD countries, led by China – growth expectations for Asia ex-Japan approached 10% by 2010
• Price increases tamp-down fundamental demand drivers – the net effect on consumption depends on the relative strength of price impacts and economic growth
11 Richard Newell, May 5, 2011 In non-OECD countries, economic growth has a strong impact on oil consumption percent change (year-on-year)
12 Forecast 10
8
6
4
2
0
-2 non-OECD oil consumption non-OECD GDP -4 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Sources: EIA, IHS Global Insight
12 Richard Newell, May 5, 2011 Economic growth in Asian economies surprised to the upside in 2007 and 2009 and to the downside in 2008 percent GDP growth in Asia-ex Japan (annual expectations)
10
9
8
7
6 2007 5 2008 4 Starting in January of each year, each line shows the expected forecast of GDP growth for 2009 3 the specified calendar year, which tends to move toward the actual realized growth outcome 2010 2 as the year progresses. Expectations continue to evolve into the next calendar year as revised 1 GDP data become available (e.g., 2007 GDP expectations are revised even during 2008). 0 2007 2008 2009 2010 2011
Source: IHS Global Insight
13 Richard Newell, May 5, 2011 In OECD countries, price increases have coincided with lower consumption
percent change (year-on-year) price per barrel (real 2009 dollars)
6 135 Forecast
4 90
2 45
0 0
-2 -45 OECD consumption (left axis) -4 WTI crude oil (right axis) -90
-6 -135 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Sources: EIA, Thomson Reuters
14 Richard Newell, May 5, 2011 Rising oil prices held down global oil consumption growth from 2005-2008, despite high economic growth
percent change (year-on-year) price per barrel (real 2009 dollars) 6 135 Forecast 5
4 90
3
2 45
1
0 0
-1
-2 World oil consumption (left axis) -45 World GDP (left axis) -3 WTI crude oil (right axis)
-4 -90 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Sources: EIA, Thomson Reuters
15 Richard Newell, May 5, 2011 Oil supply
16 Richard Newell, May 5, 2011 Oil supply and production: What matters? • OPEC – actively sets production quotas for members (compliance varies) – can influence prices with announced price targets and quotas – maintenance of spare capacity by Saudi Arabia gives it the option to pursue price objectives – in times of low spare capacity, OPEC’s ability to respond to supply shocks or rising demand is reduced – this reduced ability coincided with 2004-2008 oil price increases
• Non-OPEC – production is typically at or close to full capacity – unlike OPEC producers, almost no near-term price-driven variation in production levels – even with rising prices, there was almost no net annual production increase during 2005-2008
17 Richard Newell, May 5, 2011 OPEC production often acts to balance the oil market; OPEC quota cuts tend to lead to price increases
million barrels per day change (year-on-year) percent change (year-on-year) 5 100
4 80
3 60
2 40
1 20
0 0
-1 -20
-2 -40
-3 OPEC quotas (left axis) -60 WTI crude oil (right axis) -4 -80
-5 -100 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Sources: EIA , Thomson Reuters
18 Richard Newell, May 5, 2011 Non-OPEC production saw strong growth in 2009-2010. EIA expects these increases to slow in 2011-2012
million barrels per day change (year-on-year) price per barrel (real 2009 dollars) 2.0 140 Forecast
1.5 105
1.0 70
0.5 35
0.0 0
non-OPEC production (left -0.5 axis) -35 WTI crude oil (right axis)
-1.0 -70 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Sources: EIA, Thomson Reuters
19 Richard Newell, May 5, 2011 Non-OPEC supply expectations were adjusted upward in 2009-2010 after production decreases during downturn million barrels per day (annual average expectations) 52.0 2007 2008 2009 2010 2011 51.5
51.0
50.5
50.0
Starting in January of each year, each line shows the 49.5 expected forecast of non-OPEC supply growth for the specified calendar year, which tends to move toward the actual realized growth outcome as the year progresses. 49.0
48.5 2007 2008 2009 2010 2011
Source: EIA, Short Term Energy Outlook
20 Richard Newell, May 5, 2011 During 2003-2008, OPEC’s spare production levels were low, limiting its ability to respond to demand and price increases spare capacity (million barrels) price per barrel (real 2009 dollars) 7 140 spare capacity < 2.5 million barrels Forecast
6 120 OPEC spare capacity (left axis) 5 100 WTI crude oil (right axis)
4 80
3 60
2 40
1 20
0 0 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Source: EIA, Thomson Reuters
21 Richard Newell, May 5, 2011 In summary, 2003-2008 saw periods of very strong economic and oil demand growth, slow supply growth and tight spare capacity percent change (year-on-year) price per barrel (real 2009 dollars) 6 135 spare capacity < 2.5 million barrels Forecast 5
4 90
3
2 45
1
0 0
-1 World capacity* (left axis) -2 -45 World GDP (left axis) WTI crude oil (right axis) -3
* World capacity = OPEC capacity plus non-OPEC production -4 -90 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Sources: EIA, Thomson Reuters
22 Richard Newell, May 5, 2011 Inventories: The balancing point
23 Richard Newell, May 5, 2011 Inventory builds go hand-in-hand with increases in future oil prices relative to current prices (and vice versa) million barrels change (year-on-year) dollars per barrel change (year-on-year)
200 20
150 15
100 10
50 5
0 0
-50 -5
-100 -10 OECD inventory (left axis) -150 -15 WTI crude futures spread (12-month-ahead minus next-month's futures price; right axis) -200 -20 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Sources: EIA, Thomson Reuters
24 Richard Newell, May 5, 2011 Financial market activity
25 Richard Newell, May 5, 2011 EIA is reviewing information that bears on several proposed linkages between physical and financial markets
• Our academic partners are looking at the growth of swap dealers, hedge funds, and commodity index funds in oil derivatives trading – to date, they have not identified any rigorous research that either proves or disproves that speculation caused or contributed to 2007-2008 oil price volatility – some evidence suggests that the price run-up and decline may have been exacerbated by the formation and collapse of an oil price bubble, perhaps triggered by fundamental factors in the oil market and the global economy • Internal work is looking at several relationships – the relationship between levels of index investment, both for crude and commodities more generally, and oil price changes – movement in the correlation of daily returns between energy and other assets, as one indicator of changing linkages among markets – the relationship between changing information in the marketplace and changing expectations, building on our recent use of futures and options prices to measure the uncertainty surrounding future price expectations
26 Richard Newell, May 5, 2011 Commodity derivatives have multiple roles in the market • Successful futures and derivatives markets can help improve market efficiency – provide transparent price discovery – tools for managing and hedging price risk – serve as a vehicle for investors to diversify their portfolios
• Derivative trading can also influence the degree of price fluctuations – moderation (arbitrage can reduce price discrepancies and smooth them over time) – amplification (if herding behavior and/or bubbles occur)
• Prices could diverge from what physical factors alone may warrant, particularly if current conditions and expectations about the future are uncertain – identifying the magnitude of such episodes is extremely challenging
27 Richard Newell, May 5, 2011 The attraction of commodity diversification brought increased institutional investment during the oil price rise • Crude oil and other commodities were an attractive investment – expectations for crude oil demand were high in 2007 and the first half of 2008 – crude oil price had historically been slightly negatively correlated or uncorrelated with other investments up until 2008
• There were increased risks to traditional asset classes, as evidenced by unfavorable trends in indicators linked to these assets over the period July 2007-July 2008
– equities: unemployment rose from 4.6% to 5.8% (Bureau of Labor Statistics) – bonds: annual inflation expectations rose from 2.6% to 3.1% (Cleveland Federal Reserve Bank) – real estate: S&P/Case Shiller Home Price Index fell 14.9%
28 Richard Newell, May 5, 2011 Open interest in crude oil futures grew over the last decade as more participants entered the market average daily open interest in crude oil futures (thousands of contracts)
1,800 1,600 1,400 1,200 1,000 800 600 400 200 0
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Source: Bloomberg
29 Richard Newell, May 5, 2011 Daily return correlations between oil prices and other asset classes have generally strengthened in recent years • The correlation between daily returns on crude oil and other globally-traded commodities has strengthened over the past few years
• Crude oil also began showing stronger absolute correlations in 2008 with the S&P 500 and U.S. Treasury bonds
• The correlation between crude oil and the dollar has varied over the past decade, with a strong negative correlation in recent years
• In contrast, crude oil and natural gas returns have been correlated over the past decade, but the relationship has recently weakened
30 Richard Newell, April 21, 2011 Correlations between daily price changes of crude oil and other commodities generally rose in recent years 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Energy Natural Gas
Gold
Metals Copper
Silver
Soybeans
Agriculture Corn
Wheat
WTI crude oil price peak
< -0.65 -0.65 to -0.4 -0.4 to -0.25 -0.25 to 0.25 0.25 to 0.4 0.4 to 0.65 > 0.65 Negative correlation Positive correlation Note: Correlations computed quarterly
31 Richard Newell, May 5, 2011 Correlations (+ or -) between daily returns on crude oil and financial investments have also strengthened
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
S&P 500
U.S. Dollar
U.S. Treasury
WTI crude oil price peak < -0.65 -0.65 to -0.4 -0.4 to -0.25 -0.25 to 0.25 0.25 to 0.4 0.4 to 0.65 > 0.65 Negative correlation Positive correlation
* U.S. Dollar Index (DXY), which is a weighted index of a basket of currencies, per U.S. dollar. As the dollar strengthens against other currencies, the value of the index rises.
** U.S. Treasury is based on the negative of the change in yield on 30-year U.S. government bonds because as yields rise, bond prices fall.
Note: Correlations computed quarterly
32 Richard Newell, May 5, 2011 Non-commercial participation in commodity derivative markets has been increasingly on the buy-side (net long) number of contracts (thousands)
250 Non-commercial net open interest 200 150 Commercial net open interest 100 50 0 -50 -100 -150 -200 -250 -300
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011
Note: Non-commercial consists of managed money and other market participants. Commercial consists of producers, merchants, processors, end users and swap dealers. Source: CFTC Commitment of Traders
33 Richard Newell, May 5, 2011 Did the level of commodity index investment move with or counter to movement in oil prices? More evidence is needed • CFTC “special call data” on commodity investment is an important source of information, but it is only available post- December 2007 – suggests that investment positions in oil futures and derivatives moved counter to oil prices during 2008 and early 2009 – the use of supplemental CFTC commitment-of-traders data to look back further for oil (by backing out other commodities) is methodologically difficult
• We are also looking at other measures of index investment in all commodities – constructing a measure of index investment that aggregates across commodities is even more challenging, even if we limit attention to the post- December 2007 period – preliminary evidence suggests that a broader commodity measure may have a different relationship to oil price movements than one that considers only the level of oil index investment
34 Richard Newell, May 5, 2011 Crude oil plays a major role in commodity investment 2011 Target Weights of the Dow Jones - UBS Commodity Index Nickel Live Cattle Lean Hogs Zinc 2.2% 3.4% 2.0% Crude Oil 2.8% 15% Silver 3.3%
Aluminum 5.2% Natural Gas 12% Copper 7.5%
Heating Oil 3.6% Gold 10.4% Gasoline 3.5%
Cotton 2.0% Soybeans Coffee 7.9% 2.4% Soybean Oil Corn 2.9% Sugar Wheat 7.0% 3.3% 4.6% Source: Dow Jones Indexes, CME Group
35 Richard Newell, May 5, 2011 Index investors' CFTC futures-equivalent positions may have actually run counter to movements in oil prices index investor future-equivalent positions (thousands of contracts) price per barrel (real 2009 dollars) from the end of 2007 through early 2009, the size of CFTC- reported index investor positions 700 fell as prices rose, then rose as 135 prices fell 600 125 500 115 400 105 300 95 200 85 100 75 0 65 -100 55 -200 45 Long (left axis) Short (left axis) WTI 1st nearby (right axis) -300 35 2007 2008 2009 2010
Sources: CFTC Special Call Report, Bloomberg
36 Richard Newell, May 5, 2011 Assets under management by largest U.S. commodity funds compared to CFTC’s total index notional investment asset value ratios relative to December 2007 (normalized dollars)
2.4 2.2 2 1.8 1.6 1.4 1.2 1 0.8 0.6
0.4 Assets under management (five largest public U.S. commodity funds) 0.2 Commodity index investment reported to CFTC under "special call" 0 2005 2006 2007 2008 2009 2010 2011 Source: CFTC, Bloomberg
37 Richard Newell, May 5, 2011 Oil price formation: both physical and financial markets matter • Supply and demand fundamentals remain a key underlying driver of oil prices – the low near-term responsiveness of both oil consumption and production to price changes is an essential factor underpinning oil price volatility
• Financial market conditions in the past few years have, however, increased interest in oil as an asset class
• Activity in markets for other commodities, equities, bonds, and foreign exchange has become more highly correlated with oil, and should be considered in analysis of oil price formation
• Additional research is needed to better understand the linkages among these markets
38 Richard Newell, May 5, 2011 For more information U.S. Energy Information Administration home page | www.eia.gov
Short-Term Energy Outlook | www.eia.gov/steo
Annual Energy Outlook | www.eia.gov/aeo
International Energy Outlook | www.eia.gov/ieo
Monthly Energy Review | www.eia.gov/mer
EIA Information Center (202) 586-8800 | email: [email protected] Live expert from 9:00 a.m. – 5:00 p.m. EST Monday – Friday (excluding Federal holidays)
39 Richard Newell, May 5, 2011