Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Macro
27 June 2008
Commodities Quarterly
Commodities As An Asset Class: Total returns on the DBLCI-Optimum Yield have risen 40.5% year to date. We believe the appearance of contango in the crude oil curve and increasing spot price risks are increasing the attractiveness of non-directional alpha strategies, such as a long exposure in the DB Commodity Harvest Index. Oil: We believe the factors driving oil prices higher are fundamental and not speculative. Oil needs to rise to USD150/bbl for oil as a share of global GDP to reach the levels that occurred in the early 1980s. We believe at that point we will start to see more signs of demand destruction and an eventual tipping point in oil markets. US Natural Gas: We believe US natural gas prices are still trading cheap relative to crude oil. We are therefore positioning for further price gains. The main risk, in our view, is that the recent price advances may require a hot US summer to sustain the current momentum. Precious Metals: The complex has been neglected during the second quarter as investors contemplate a turn in the US dollar. We believe any rally in the US dollar is likely to be based on shaky foundations and consequently we are maintaining our long recommendation. Industrial Metals: We are maintaining our bullish outlook for aluminium and copper. We also believe the price outlook will start to improve for zinc from the middle of next year. Agriculture: We believe fundamentals still warrant higher prices, but, seasonal factors may conspire to push corn prices lower over the summer. We expect wheat to outperform corn particularly if the 2008-09 Australian wheat harvest disappoints. Carbon Emissions: We target Cal10 carbon prices rising to EUR40/tonne. However, we place a high probability of price spikes during the 2008-2012 Phase 2 period because borrowing from futures periods is not allowed under ETS rules. Trade Recommendations: We outline directional and relative value plays across the commodities complex on page 8.
Dark Markets & Monsters
Table of Contents Commodity Views ..................................... 2 #1 Executive Summary .............................. 3 #2 The Role Of Speculators ...................... 4 #3 Trade Recommendations...................... 8 #4 Trade Review ...................................... 12 #5 Commdity Indices............................... 16 #6 Commodity Alpha ............................... 18 #7 Long Term Pricing............................... 21 #8 Crude Oil ............................................. 26 #9 Crude Oil & Fuel Subsidies ................. 31 #10 Crude Oil Contango: Here To Stay?.. 33 #11 Commodity Volatility......................... 38 #12 Crude Oil: Analysts vs. Markets........ 42 #13 US Natural Gas.................................. 44 #14 Carbon Emissions ............................. 47 #15 Precious Metals ................................ 51 #16 Exchange Traded Funds ................... 55 #17 PGMs ................................................ 57 #18 Industrial Metals ............................... 60 #19 Agriculture ........................................ 70 #20 Uranium ............................................ 74 #21 Bulk Commodities............................. 78 Commodities Chartbook ......................... 81 Price Forecasts ........................................ 91 Correlation Matrix .................................... 94
Research Team London Michael Lewis 44 20 7545 2166 Jude Brhanavan 44 20 7547 1558 Hong Kong Amanda Lee, CFA 852 2203 8376 Paris Mark Lewis 33 1 4495 6761 Isabelle Curien 33 1 4495 6616 New York Joel Crane 1 212 250 5253 Washington Adam Sieminski, CFA 1 202 662 1624
Commodities
Deutsche Bank AG/London All prices are those current at the end of the previous trading session unless otherwise indicated. Prices are sourced from local exchanges via Reuters, Bloomberg and other vendors. Data is sourced from Deutsche Bank and subject companies. Deutsche Bank does and seeks to do business with companies covered in its research reports. Thus, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision. Independent, third-party research (IR) on certain companies covered by DBSI's research is available to customers of DBSI in the United States at no cost. Customers can access this IR at http://gm.db.com, or call 1-877-208-6300 to request that a copy of the IR be sent to them. DISCLOSURES AND ANALYST CERTIFICATIONS ARE LOCATED IN APPENDIX 1
27 June 2008
Commodities Quarterly
Commodity Views
Energy USD WTI Brent Heating oil Gasoline (RBOB/gallon) US natural gas (/mmBtu) Coal (API#2/tonne) Uranium (/lb) EUR Emissions Cal'10 Precious Metals Spot (USD/oz) Gold Silver Platinum Palladium Industrial Metals 3M Fwd (USD/tonne) Aluminium Copper Lead Nickel Tin Zinc Agriculture 1 nearby (USD) Corn Cotton Live cattle Soybeans Sugar Wheat
st
12M Low 67.77 68.69 1.94 1.86 5.43 77.30 57.00 19.20
12M High 138.54 137.69 3.97 3.55 13.21 199.50 138.00 28.73
View
View
View
View
Source: DB Global Markets Research, Bloomberg (Prices as of close of business June 25, 2008)
Page 2
27 June 2008
Commodities Quarterly
#1 Executive Summary
We are maintaining our bullish outlook for the agricultural sector and specifically corn and soybeans. However, we recognise that corn is not only trading rich relative to wheat, but, since 1989 the summer months have typically been a hazardous period for corn returns. We are positioning for wheat to outperform corn in the short term and for the wheat curve to move back into backwardation. We believe oil markets are in a state of confusion unwilling to believe that the forces that drove oil prices higher over the past year, namely lower OPEC production, disappointing non-OPEC supply, strong economic growth in the developing world and a weaker US dollar may now be working in reverse or at least not working in favour of even higher prices. We are reluctant to take a short position in crude oil at the moment particularly given reports of increased Chinese oil demand after the recent dismantling of fuel subsidies and the risks of US dollar overshooting. However, we believe we are close to a tipping point and if and when crude oil price correct lower this could help lift global equity markets and alleviate fears towards stagflation in our view. We believe one possible beneficiary of such a scenario could be industrial metals. Elsewhere in the energy complex we believe US natural gas prices have a further 15% to rise to bring them into line with crude oil. Although sentiment remains bullish we fear that a hot summer in the US is require to sustain this trend. We believe the 60% decline in uranium prices since its peak in 2007 is overdone. We believe demand fundamentals suggest a bullish price outlook over the next few years. Given the appearance of contango in the crude oil curve and heightened spot price risks in the energy and agricultural sectors, we recommend non-directional alpha strategies, such as going long the DB Commodity Harvest Index We are maintaining our bullish view towards aluminium and copper and believe from 2009 the outlook for zinc price will improve considerably. Investors appear to be contemplating a turn in the US dollar. While we believe any rally in the US dollar will be based on shaky foundations, we expect this will sustain headwinds for the precious metals complex in the near term. However, we have stated for some time that every US dollar cycle since 1978 has ended with central bank FX intervention either to rescue the US dollar or the euro (DEM before 1999). We believe it would therefore be unwise to abandon a bullish call on the precious metals complex, but, we recognise that given the already extreme undervaluation of the US dollar particularly against the euro, we believe the precious complex will require additional factors to drive prices higher. We believe platinum will outperform within the group. We also recommend selling volatility across the commodities complex and specifically for crude oil and gold.
In next section of this report we examine the role of speculators and specifically commodity index investors on driving commodity prices. We estimate that the role and power of speculators has been exaggerated.
Palladium Aluminium Soybeans Platinum RBOB gasoline Nickel WTI Heating oil Coal (API#2) Corn US natural gas -60 -40 -20 0 20 40 60 80
Page 3
27 June 2008
Commodities Quarterly
2)
Markets where index trading is greatest as a percentage of total open interest have actually suffered from falling prices during the past year. The level of speculation in the agriculture commodity and the crude oil markets has remained relatively constant in percentage terms at a time when prices have risen. Speculators such as managed money traders are both buyers and sellers in these markets. For example, data shows that there are almost as many bearish funds as bullish funds in wheat and crude oil. CFTC studies in agriculture and crude oil markets have found that speculators tend to follow trends in prices rather than set them.
4)
Figure 1: The performance of financially and nonfinancially traded commodities since 2001
1400 1200 1000 800 600 400 200 0
M ol yb de nu C ad m m i R um ho Fe d rro ium ch ro m e C o Tu bal t n M gst an en ga ne se R ic Iro e n or e R Ste ut he el ni um W T C I op pe r N at Tin ur al ga s Le ad N ic ke l Si lv er G ol d C or n Al Zin um c in iu m
Source: DB Global Markets Research, Bloomberg, Reuters, data as of end May 2008
We believe the US Senate and the CFTC among others are concerned with three broad areas of regulatory oversight: (1) OTC markets: Over-the-counter electronic energy trading has been largely exempt from detailed federal record-keeping requirements, the so called Enron loophole. (2) Swaps trading: Investment banks undertaking swaps for clients, which do not have the same reporting requirements and limits on trading positions that would apply to the customer directly, the Swaps loophole. (3) Non-US commodity exchanges: Trading of USbased WTI crude oil and other products on Londons InterContinental Exchange (ICE) is largely exempt from US Federal oversight and record-keeping requirements, the London loophole. The CFTC is therefore working on achieving the right balance of enhanced market oversight and transparency while promoting market innovation and competition.
The view that higher crude oil and agriculture commodity futures prices are being driven by speculators in the financial markets has captured significant market attention during this year. Along with many other public institutions and regulators, the US Commodity Futures Trading Commission (CFTC) has analysed the role of investors, both hedgers and speculators and concluded that there is little economic evidence to demonstrate that prices are being systematically driven by speculators. The CFTC data show: 1) Prices have risen sharply for many commodities that have neither developed futures markets nor significant institutional fund flows, Figure 1.
Page 4
27 June 2008
Commodities Quarterly
We agree with the argument that there is little evidence to suggest that the rise in commodity pries has become a speculative phenomenon. We believe prices are responding to tightening fundamentals which are either trying to resolve acute shortages in agricultural commodities, where inventory to consumption ratios have fallen to levels not seen since the early 1970s, or prices are attempting to ration demand and increase supply to avert a significant imbalance in market conditions that are set to appear at the end of the decade, for example in the oil market. In this article we present further analysis to support this view. The role in institutional investors To assess the importance of commodity index investors we first attempt to track the flows of funds into commodity indices during this decade. Our aim is to gauge to what extent these investor flows could distort the market away from underlying physical supply and demand fundamentals. Most industry estimates suggest money tracking commodity indices has grown from approximately USS10bn at the beginning of the decade to around USD200bn. However, we are now presented with two challenges: First, the number of investible commodity indices has increased significantly since the beginning of the decade, which means commodity index inflows are scattered across various index products. Before 2003, there were only three investible indices, the S&PGSCI, the Dow Jones-AIG and the Rogers International Commodity Index. In February 2003, Deutsche Bank launched a family of commodity indices to challenge these incumbent index providers and since then the index universe has been joined by a new wave of commodity indices from other investment banks. Thankfully for our purposes, the distribution of funds allocated to commodity indices has tended to be concentrated in just a handful of fund providers. Last month Reuters estimated that the S&PGSCI and Dow Jones-AIG commodity indices represented around 75% of the share of the commodity index market. In the same article, Reuters reported that the family of Deutsche Bank commodity indices was the next largest index provider, although representing no more than 10% of the market. Based on our estimates, the amount of money tracking the DB indices is split between the DBLCI-Mean Reversion, USD10bn, and the DBLCI-Optimum Yield, USD8bn. The second challenge to assess how index money could be affecting commodity markets is to recognise that the weights to the various commodity components and sectors varies according to a particularly commodity index. For example, the energy weight on the S&PGSCI is currently over 75% compared to less than 40% on the Dow Jones-AIG. Consequently, tracking investor inflows into commodities via indices one needs to make assumptions on the market share of individual
commodity indices and then in which sectors and individual commodities those indices are investing. Given the degree of market attention on global energy markets and the role of speculation, we examined the potential size of commodity index inflows that have entered into the crude oil market since the beginning of the decade. For simplicity sake, we assume that the commodity index market can be represented by the S&PGSCI, the Dow Jones-AIG, the DBLCI-Mean Reversion and the DBLCI-Optimum Yield. We believe this is realistic since we believe this captures more than 80% of the commodity index market. Of the USD200bn tracking commodity indices, we assume USD128bn is indexed to the S&PGSCI and USD55bn to the Dow Jones-AIG. For the DBLCI-MR and DBLCI-OY we use our own estimates of USD10bn and USD8bn respectively. Figure 2 presents the various underlying components, weights and derived US dollar notional amounts that correspond to these four commodity indices. We find that of the USD200bn invested in commodity indices in 2008 approximately USD66.2bn is invested in the WTI sweet light curve oil future and USD18.8bn in the Brent crude oil future, or USD85bn in total. Consequently between 40% and 45% of US dollar notional is in crude oil (or 85/200 x 100). If true this would imply that crude oil length attributable to commodity index investors has grown from USD4bn in 2000 to around USD90bn today (45% of USD200bn). If we then divide our estimate for crude oil length by the average oil price for that year we can derive the total number of barrels that would need to be held back to satisfy index demand. On our calculations, this would be equivalent to 750 million barrels of oil in 2008. However, we need to capture incremental barrels of oil needed to be held back each year to assess the marginal effect of index investors on the oil market. We estimate this not only as an absolute number, but, also as a fraction of the annual physical usage. The International Energy Agency estimates oil consumption in 2008 will total 86.8 million barrels per day, equivalent to annual crude oil usage of 31.7 trillion barrels. We then can compare the scale of index money demand for crude oil compared to actual physical demand. We find that even if every paper barrel held by an index investor had been held back by the oil producers the net impact on crude oil supply would have been insignificant. For example, in 2007, index investors added 80 million barrels of crude oil length into the market via indices. Since oil consumption in that year was almost 31 trillion barrels, index investment represented just 0.26% of annual usage, Figure 3.
Page 5
27 June 2008
Commodities Quarterly
Estimating index flows into crude oil We believe the role of speculators and their impact on commodity markets has been distorted by many commentators not recognising that of the USD200bn invested in commodity indices only a fraction of this is invested in crude oil. Moreover that sum has not been built up in just one year, but, gradually over the course of this decade. Consequently we believe a significant amount of the value of commodity index positions is price appreciation. Moreover we estimate that marginal inflows are relatively small particularly compared to annual average oil consumption. Another way of presenting this analysis is that of the USD200bn of commodity index length, USD85-90bn is linked to crude oil index positions. Assuming an average oil price of USD120/bbl in 2008, this amounts to crude oil length of 750 million barrels of oil. This accumulation in crude oil length was built up over eight years during which time world consumption of crude oil amounted to a cumulative 240 billion barrels.
Consequently we estimate that the build up of index length represents approximately 0.31% of global crude oil usage. Or put another way, equivalent to 260k bbl/day. As a result, even if producers held back oil to satisfy index investors, we believe 260k bbl/day can not explain the move in the oil price from USD20 to almost USD140/barrel today. To put this into perspective, 260k bb/day is the equivalent to total oil production in the Republic of Congo or 0.5% of world crude oil exports in 2007. We believe the eventual price of any commodity futures is therefore conditioned by the physical market, with futures buyers unable to directly impact that market. However, we recognise that one way speculators could affect prices indirectly would be if physical players sell the futures to the indexers and then hoard physical inventory. Such hoarding would raise prices, but even here there is no evidence of an inventory build in the OECD in oil or for that matter in other commodity markets such as grains.
Page 6
27 June 2008
Commodities Quarterly
Another possibility is that oil producers are holding back inventories in the ground, possibly reflecting resource nationalism or due to government development policy decisions. In effect that would be equivalent to hoarding, but, it would not show up in inventories. A falling US dollar could also have the effect of postponing the development and exploration of oil fields. Conclusion We believe speculators have been an easy scapegoat for commentators to blame for the rise in commodity prices. In our view, events today have resonance with the turbulence in foreign exchange markets during the collapse of the Exchange Rate Mechanism in 1992 and 1993. At that time speculators realised the inconsistency of government macroeconomic policies across Europe, such as the UK, which were attempting to operate a pegged exchange rate regime, in an environment of free capital mobility, but, preserving their own monetary independence. What became known as the inconsistent trio was clearly not sustainable over the long run and exchange rate regimes collapsed. Interestingly, speculators did not push foreign exchange rates away from equilibrium rather they pushed them closer to fair value since many of them were significantly overvalued versus the Deutschmark. If it had succeed attempts to rein in speculative activity, via a Tobin tax on short-run foreign exchange transactions, would have merely slowed the process of exchange rate adjustment. Consequently we view speculators as a force for good in FX markets in the early 1990s although they discredited many European governments macroeconomic policies in their wake.
Today the shoe is on the other foot. US, not European, regulators are proposing action on speculative activity but this time in commodity markets. However, we believe regulators may find themselves disappointed by steps they may try to take to curb speculative activity and specifically investment via commodity indices. If we are correct, commodity indices have had a negligible effect on commodity prices. Rather we believe it is fuel subsidies in many parts of the non-OECD that are one of the greatest distortions in oil markets currently. The rise in the oil price to levels of valuation we last saw in the 1980s is likely to accelerate the process of dismantling fuel subsidies, which we believe will help to shift the oil demand curve over the medium term.
Michael Lewis, (44) 20 7545 2166 michael.lewis@db.com Adam Sieminski, (1) 202 662 1624 adam.sieminski@db.com
Figure 3: Measuring annual crude oil length as a percentage of crude oil consumption
2000 Index length (USD bn) Crude oil length (USD bn) Average crude price (USD) Crude oil length (bbl bn) Added crude oil length (bbl bn) Annual crude oil consumption (bbl tn) Added crude oil length as a % of annual crude usage 0.41% 0.20% 0.38% 0.07% 0.24% 0.39% 0.26% 0.39%*
* data for 2008 annualised; Source: DB Global Markets Research
8 4 30 0.12
Page 7
27 June 2008
Commodities Quarterly
#3 Trade Recommendations
#1 Short Dec09 WTI variance swap
The run up in oil prices to USD140/bbl has also sent oil implied vol significantly higher. Indeed, back-end vol has jumped to the highest level in 5 years and the entire term structure has flattened substantially. Consequently vol has shifted from the lower quartile of the past eight year distribution at the beginning of this year to stand currently significantly above the 95-percentile mark. We believe the current level of elevated crude oil vol is not sustainable because although realised vol has climbed as well, it has failed to catch up with implied vol and is yet to break the previous high set by the Katrina related shutdown in 2005. Summer tends to see volatilities declines across asset classes and commodity is no exception. Indeed, actual volatility of crude oil has tended to fall in July and August using data since July 1993. Short Dec09 WTI variance swap
WTI Dec09 implied vol is trading substantially higher than the previous peak reached by the historic dated futures contracts
WTI 260 day realised volatility
37.5 35.0 32.5 30.0 27.5 25.0 22.5 20.0 17.5 15.0 12.5 10.0 7.5 Oct-03
Oct-04
Oct-07
3.00
2.00
1.00
0.00
-1.00
-2.00
-3.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Page 8
27 June 2008
Commodities Quarterly
Implied vol of platinum is also trading much higher than the realized vol
- SUVs sales in the US are collapsing; - global growth is moderating and at risk of slowing further on rising inflation 3. Refining margins remain weak.
Aug-84 Aug-85 Aug-86 Aug-87 Aug-88 Aug-89 Aug-90 Aug-91 Aug-92 Aug-93 Aug-94 Aug-95 Aug-96 Aug-97 Aug-98 Aug-99 Aug-00 Aug-01 Aug-02 Aug-03 Aug-04 Aug-05 Aug-06 Aug-07 Aug-08 Aug-09 Aug-10 Aug-11 Aug-12 Aug-13 Aug-14 Aug-15
Long DBLCI-OY-CL vs. short DBLCI-CL Long DBLCI-OY vs. short DBLCI Short M02 WTI vs. Long M03 WTI and roll the positions every month The maximum theoretical loss to this trade is unlimited
Source: DB Global Markets Research
Page 9
27 June 2008
Commodities Quarterly
#5 Bullish Nickel
Since the collapse of the nickel price from USD51,600/tonne in the middle of 2007 nickel has been trading in a very tight yet low range between USD25,000 and USD35,000 over the past one year. We believe the current nickel price at USD21,800 is very close to if not below the marginal cost of production and indeed we believe the substitution threat from nickel pig iron, which was the main driver of the nickel crash, is diminishing this year. We believe the cost of producing nickel pig iron could serve as a floor to the nickel price, and the estimated cost is USD24,180 in 2008 and USD24,260 in 2009, according to Brook Hunt data. Although we have a cautiously bullish view on nickel demand, we believe the risk is to the upside such that the weak demand was exaggerated by producers de-stocking which will not and cannot continue forever. Indeed, the latest trade data from China provides strong evidence of robust demand and Chinas net import of nickel has gone up despite the threat of nickel pig iron substitution. On an annualized basis, for 2008 we believe China is set to record nearly double the 2007 net import balance. We believe the downside risk is limited by the production cost and the nickel price could surprise to the upside when the global economy picks up and therefore we would recommend a long position in nickel. Long Dec08 Nickel
40000
35000
35000
30000
30000
25000
25000
20000
20000
M03
M06
M09
M12
M15
M18
M21
M24
M27
#6 Bullish Zinc
The surface of the zinc story is undeniably (overwhelmingly?) bearish: China has ramped up zinc production, there are excessive capacities, and the zinc market is expected to stay in surplus over the next few years. However, similar to nickel, we believe the zinc price is close to the marginal cost of production which could serve as a floor to the zinc price. Indeed, we feel more comfortable in buying zinc at below USD2,000/tonne when compared to the USD4,000 plus level traded in late 2005 (or USD138/bbl WTI, or USD8,300 copper). In addition, we believe zinc has another edge over nickel, which is their relative inventory position. While nickel inventory level stands at its highest level since 1999, zinc inventory level stays close to the bottom of the 5-year range. We also believe the zinc market might turn when the Chinese government imposes an export tax on refined zinc in order to conserve energy and contain pollution (similar to todays primary aluminium market), or alternatively, when global growth picks up again. We believe the downside risk of zinc is limited by the marginal cost of production and at the current level it offers good risk adjusted returns potential. Long Dec08 Zinc
Source: DB Global Markets Research
800 700 600 500 400 300 200 100 0 Jan Feb Mar
A pr May
Jun
Jul
A ug Sep
Oct
Nov Dec
Source: LME
Page 10
27 June 2008
Commodities Quarterly
140
120
100
80 Aug-97
Dec-98
Apr-00
Aug-01
Dec-02
Apr-04
Aug-05
Dec-06
Apr-08
Page 11
27 June 2008
Commodities Quarterly
#4 Trade Review
#1 Bullish Aluminium: China To Become A Net Importer
We believe the global aluminium market is undergoing a fundamental shift in which supply is becoming increasing fragile and uncertain. The latest supply shock resulting from the power crisis in South Africa best illustrates this trend. With electricity comprising at least one third of the total production cost of aluminium, we continue to believe longer dated aluminium is extremely undervalued given the substantial rise in crude oil, natural gas as well as coal prices. We believe aluminium production growth is now at risk as a result of Chinas attempt to curb production in energy intensive industries by closing down inefficient smelters and restrict the level of aluminium exports. These measures have contributed to Chinese net aluminium exports this year falling to their lowest levels since 2001 and China moving to become a net importer of aluminium over the past few months. The recent power tariff hikes in China also echos our view. We continue to stand by our core investment strategy Buy what the Chinese are short of and sell what the Chinese export. With China importing more and more aluminium we have no choice but to remain bullish long dated aluminium. Maintain long Dec09 aluminium.
700
500
Net exporter
300
100
-100
Net importer
-300 1996 1998 2000 2002 2004 2006 2008?
Source: USDA
Page 12
27 June 2008
Commodities Quarterly
Peak demand
Operational capacity
Installed capacity
Source: Eskom
Page 13
27 June 2008
Commodities Quarterly
DBLCI-MR performance
250
backwardation
200
150
100
50
-50
contango
-100 Apr-06
Jul-06
Oct-06
Jan-07
Apr-07
Jul-07
Oct-07
Jan-08
Apr-08
Jul-08
In addition we believe the soybeans market would be no exception to this curve steepening environment given the substitution properties among these three crops. Maintain Long Dec08 vs. short Dec09 Corn. Maintain Long Dec08 vs. short Dec09 Wheat. Long Nov08 vs. short Nov09 Soybean The maximum theoretical loss to this trade is unlimited
Page 14
27 June 2008
Commodities Quarterly
Maintain Long Dec08 vs. short Dec09 Corn. Maintain Long Dec08 vs. short Dec09 Wheat. Long Nov08 vs. short Nov09 Soybean
#7 Bullish Uranium
We believe that the world is on the verge of a uranium renaissance, and in our opinion, the financial markets continue to underestimate the potential for a rapid increase in uranium demand going forward. Based on the IEA's new Global Energy Mix target, uranium demand could increase by around 300% by 2050 with the construction of over 1000 new reactors. Uranium has been one of the worst performing commodities in spot price terms, falling by more than 35% year to date. We find that the performance of uranium prices is significantly underperforming that of other energy commodities such as crude oil, thermal coal and natural gas this year. We believe the current spot price of USD57/lb is not high enough to encourage new uranium supply and expect that continual supply issues and the likelihood of increased demand from the utilities sector will drive price higher going forward. Maintain Long Uranium
Source DB Global Markets Research Source: IEA
Average Annual Power Generation Capacity Additions in the 50% Cut Scenario 2010-2050
Page 15
27 June 2008
Commodities Quarterly
#5 Commodity Indices
DBLCI-OY Up 40.5% Year To Date
Energy and agriculture have been the engine room of performance in the second quarter of this year. In comparison, returns in the metals sector have been uninspiring. We believe the industrial metals sector has been polluted by concerns towards the global economic outlook as well as falling equity markets. Meanwhile the benefit of a rapidly falling US dollar in the first quarter of the year on the precious metals complex has given way to a possible turn in the US dollar. We expect the metals sector could spring to life during the second half of the year if and when oil prices start to decline. If there were a strong correction in energy prices, we would expect this to lift global equity markets, alleviate fears of stagflation and boost industrial metal prices higher. However, optimism towards an uplift in global growth from lower oil prices could benefit the US dollar in our view and therefore sustain strong headwinds for the precious metals complex.
Components of the DBLCI* Crude Oil* Heating Oil* Aluminium* Gold* Wheat* 7.43 4.24 5.79 0.24 14.26 19.47 36.17 33.18 3.65 -2.96 -4.64 29.15 46.35 51.25 25.44 5.01 16.78 59.29 3.49 2.97 0.53 1.47 1.67 2.60
Total returns on the DBLCI-Optimum Yield and DBLCIMean Reversion are up 40.5% and 36.2% respectively since the end of last year. This compares with returns of 25.0% and 38.9% on the Dow Jones-AIG and S&PGSCI respectively. Crude oil, heating oil and corn have been the engine room of performance within the DBLCI during the second quarter of this year, with total returns up approximately 30% during the quarter. We believe corn returns have benefited from downgrades to the US corn harvest for 2008-09. The US constituted around 60% of global corn exports in 2007-08 and according to the USDA it was attributable to 80% of global corn production growth last year. We expect the progress of the US corn harvest over the summer months will remain critical to the short-term outlook for the performance of corn returns. Given the extremely low level of the corn inventory to consumption ratio we believe it is difficult to adopt a short position in this market. However, we are aware that the summer months have historically been a hazardous time for corn returns. We estimate that since 1989 corn returns have declined 70% of the time and by approximately 2% in June, July and September.
Corn*
Performance of other benchmark indices S&PGSCI DJ-AIG S&P 500 EFFAS US Bond 7.30 7.19 -6.01 0.12 26.39 14.04 -0.16 -2.96 38.94 25.00 -9.59 1.40 2.95 1.91 -0.61 1.33
* Returns defined by DBLCI sub-index contract and rolls For quotes and compositions see Reuters: DBLCI and Bloomberg: DBCM Source: DB Global Markets Research, Bloomberg (As of cob Tuesday)
We believe one possibly beneficiary of the rapid rise in corn prices this year could be wheat, which is now trading cheap relative to corn. Although total returns on the DB wheat index are down 4.6% during the second quarter, this masks a recent recovery. We expect this recovery to continue. Since 1989 we find that wheat returns have tended to perform strongly in August and September. We would also expect wheat prices to benefit from any downgrades to Australian wheat production estimates due to poor rainfall conditions.
Page 16
27 June 2008
Commodities Quarterly
Total returns on the DB aluminium index have been trading broadly sideways during the second quarter. We continue to expect further price gains based on Chinese aluminium imports. In addition, we expect the recent dismantling of fuel subsidies in China will sustain upward pressure on production costs and with it the threat of further smelter closures.
Conclusion Total returns on commodity indices have posted a strong performance in the first half of this year. However, we believe the strong grains in energy and agricultural returns may be leading to a loss in appetite among investors to adopt strong directional trading strategies. In the next article we examine alpha strategies in the commodity index space which we believe offer attractive benefits in todays markets. For example, going long the DB Commodity Harvest Index, a rule based market neutral strategy that uses the concept of carry in the commodity complex to generate stable returns.
DBLCI-OY Energy
7.52%
51.61%
3.24
60.14%
-6.35%
41.04%
-0.10%
6.19%
1.28
20.74%
27.14%
26.15%
2.55%
12.02%
-0.21
51.32%
89.85% -10.97%
DBLCI-OY Agriculture
11.61%
23.82%
2.24
14.10%
15.78%
31.28%
Page 17
27 June 2008
Commodities Quarterly
#6 Commodity Alpha
DB Commodity Harvest Index
We believe the surge in energy and agricultural prices this year has increased spot price risks associated with a long only commodity index exposure. In our view, the appearance of contango in the crude oil forward curve has also introduced new challenges for commodity index products that roll their energy futures contracts on a pre-defined (monthly) schedule. We believe the DB Commodity Harvest Index offers attractive properties particularly in the current environment since it is spot neutral and attempts to extract carry out of commodity markets via a more dynamic rolling programme.
sectors. We believe it benefits from the larger number of components since at any one time we would expect a number of the components to have forward curves either in contango or backwardation. For each commodity in the index, the long leg chooses the futures contract which maximises the implied carry, or roll return by employing the Optimum Yield technology while a short position is established in the nearby futures contract using the pre-defined rolling schedule employed by the S&PGSCI Light Energy Index, typically monthly. The Optimum Yield (OY) technology attempts to overcome contango by avoiding a pre-defined monthly rolling schedule and instead selecting that futures contract which generates the maximum implied roll returns from the list of tradeable futures which expire in the next 13 months. The technology therefore aims to maximise the potential roll benefits in backwardated markets and minimise the losses from rolling up the curve in contangoed markets. Figure 2 details not only the 24 components and weights of the S&PGSCI Light Energy Index, but also the long versus short contracts currently employed by the two legs of the trade. The long side will be rolling into that futures via the OY technology while the short side employs the pre-defined monthly rolling procedure governed by the S&PGSCI Light Energy. When markets are in backwardation the long and short legs will most often be rolling into the same contract. However, contango allows the OY technology to extract carry or the risk premium via its ability to roll further along the curve.
At the end of last year we introduced the DB Commodity Harvest Index. The index is a rule based market neutral strategy that uses the concept of carry to generate stable returns. This strategy can be viewed as a way of extracting alpha in the commodity space without taking a directional view on commodity markets. Given the surge in energy and agricultural commodity prices so far this year, we believe this non-directional property of the index is an appealing feature given the increasing spot price risks now attached to directional commodity trading strategies. So far this year excess returns on the DB Commodity Harvest have risen 3.8%, Figure 1.
Figure 2: Commodity weights and futures contracts used in the DB commodity Harvest Index
No. Commodity Current weight
as on 9-June-08
Ticker
Commodity Exchange
Long Contract
as on 9-June-08
Short Contract
as on 9-June-08
140
1 2 3 4 5 6
Energy WTI crude oil Brent crude oil Heating oil RBOB gasoline Gas oil Natural gas Base metals Aluminium Copper Zinc Nickel Lead Precious metals Gold Silver Agriculture Wheat Kansas red wheat Corn Soybeans Cotton Sugar Coffee Cocoa Livestock Live cattle Feeder cattle Lean hogs
48.54% 25.28% 9.14% 3.39% 2.88% 3.33% 4.52% 14.06% 5.05% 6.13% 0.93% 1.30% 0.65% 4.13% 3.61% 0.52% 26.32% 7.06% 1.69% 8.06% 4.82% 1.51% 1.58% 1.11% 0.49% 6.97% 3.98% 0.76% 2.23%
CL CO HO XB QS NG
120
100
7 8 9 10 11
LA LP LX LN LL
80 Aug-97
Dec-98
Apr-00
Aug-01
Dec-02
Apr-04
Aug-05
Dec-06
Apr-08
12 13
GC SI
COMEX COMEX
Aug-08 Jan-09
Aug-08 Sep-08
The DB Commodity Harvest Index has been constructed to strip out stable excess returns without direct exposure to the underlying asset class. For the purposes of the construction the DB Commodity Harvest Index, we have employed the S&PGSCI Light Energy Index as the base index. This index consists of 24 commodities representing the energy, precious metals, industrial metals, agriculture and livestock
14 15 16 17 18 19 20 21
W KW C S CT SB KC CC
22 23 24
LC FC LH
Page 18
27 June 2008
Commodities Quarterly
Since the DB Commodity Harvest Index goes long and short the same index, with identical components and weights and the only difference being the rolling procedure, the strategy is market neutral. Reweighting takes place annually to bring index weights in line with the benchmark. In addition the index is rebalanced every month according to the differences of the performance of the long and short legs to maintain spot neutrality. We believe the main hazard to the performance of the DB Commodity Harvest Index is when a commodity forward curve moves from contango to backwardation. When this occurs the strategy of going long the back end of the commodity future whose curve is in contango and short the front end leg will suddenly trigger a drawdown event. However, we find that losses during drawdown events since 1997 have tended to be recouped quickly. The two most recent drawdown events were in September 2006 and June 2007. On the first occasion the wheat curve moved rapidly from contango to backwardation on account of downgrades to the Australian wheat harvest. The next event was during the summer of last year when the oil forward curve moved from contango to backwardation in response to OPEC production cuts. However, in both instances drawdowns were relatively shortlived. In both events, the DB Commodity Harvest Index had recouped its losses within a seven week period.
Conclusion Commodity index investing has become more challenging this year. We believe the surge in energy and agricultural prices may deter investors from adopting aggressive directional trading strategies given the associated higher spot price risks. Moreover, we believe the appearance of contango in the crude oil forward curve once again exposes the dangers of commodity index investing when employing a predefined monthly rolling schedule. We also expect that, if sustained, contango will enhance the benefits of the OY technology employed on the suite of Deutsche Bank commodity indices. Given spot price risks and the dynamic nature of commodity forward curves we believe the DB Commodity Harvest Index offers an attractive route to extract alpha from commodity markets. Indeed we find that the DB Commodity Harvest Index has generated steady year-on-year returns with limited drawdowns. The index had two negative calendar years over the past 10 years with its worst drawdown of -5.2% in 2000. Consequently we expect the index will be attractive option given its property of spot neutrality. Figure 3 highlights that even during a cyclical downturn in commodity markets, the DB commodity Harvest Index has been able to generate positive returns.
Figure 3: The comparative performance of various alpha and beta allocation strategies
Total Return* Beta allocation strategies
DBLCI TM S&P GSCI SM DJ-AIGCI
TM
Volatility**
Excess Return*
Sharpe Ratio #
Monthly drawdown ##
maximum
no. of months < -5%
19 23 14
15 4 5 9 23 3 14
SM
5.45%
7.09% 4.22%
0 1 1
* Annualised returns based on total return and excess return; ** Annualised vol of the daily lognormal returns; # calculated as a quotient if excess returns and the volatility; ## based on total return. Data from September 1997 t 25 June 2008 Source: DB Global Markets Research, Bloomberg
Page 19
27 June 2008
Commodities Quarterly
Trade recommendation:
Long DB Commodity Harvest. In our view, the risk to this trade would be a sudden move in one of the commodity forward curves from contango to backwardation.
Page 20
27 June 2008
Commodities Quarterly
Real crude oil price (PPI deflated, 2005 US dollars) Real copper price (PPI deflated, 2005 US dollars)
10000
8000
6000
4000
It will not come as a surprise that our evaluation of the market has produced significant increases to our longterm price assumptions. This conclusion implies cost structures and market values have permanently changed, for a number of reasons: We believe the shift in long-term oil and gas futures prices is structural not cyclical. Higher capital development. costs for new project
Higher costs associated with infrastructure development. Increased royalty rates. Rising labour and consumables costs and New but often more expensive technological developments.
Introduction We have traditionally reviewed our long-term price assumptions for commodities at the beginning of each year. However, the environment over the last two years has prompted us to review these assumptions more regularly, last occurring in September 2007 and now again. Among the number of material shifts that have occurred throughout this commodities bull-cycle, one key theme has been the rapid change in long-term equilibrium prices. Indeed it is a widely accepted presumption that commodity prices are mean reverting over the medium to long-term. However, this cycle is promising to significantly raise those averages. Part of this shift is a direct consequence of an increase to the cost of production. Prior to 2003, operating costs were generally declining in response to the downward trend in real commodity prices.
Energy is precious In our view, energy supply and security has become one of the most pressing issues facing commodity consumers this decade. Already in the first six months of the year, there have been we have witnessed several events in China, southern Africa, Australia, India and Brazil that have had considerable ramifications across the commodities complex. High energy prices are also forcing companies to re-evaluate projects, as exemplified by a major diversified miners decision to indefinitely delay its large scale aluminium smelter in South Africa. For example, we believe the cancellation of fuel and power subsidies, particularly in countries that are major commodities producers such as China, Indonesia and India, will be viewed as a watershed event in this cycle since it will have a substantial impact on prices. According to Brook Hunt, global power tariffs increased by 15% in 2007, the largest ever annual increase since the independent consultant began costing the industry. We expect this years annual increase will be even higher.
Page 21
27 June 2008
Commodities Quarterly
Methodology In this article we introduce five routes by which to determine the long-term fair value for various commodities: 1) Commodity prices relative to G7 per capita income Commodity prices in real terms Finding and development costs The marginal cost of production The five-year forward price
2) 3) 4)
Source: Brook Hunt, Reuters Note: The Metallgesellschaft Base Metals Index (MGMI) comprises the six LME base metals, weighted by contract size
Given international experience in power sector reform, we can only assume that greater capital market influence among utilities in non-OECD countries will lead to rationalisation of supply and perhaps even constrain new capacity build. Consequently this could lead to even tighter power markets. In addition, given energy price increases over the last few years, we believe it would not be surprising if power suppliers become increasingly reluctant to enter into new longterm power contracts. Labour, consumables and other factors The global mining boom has stoked competition for skilled workers and raised prices for equipment and contractors at a time of record fuel costs. As demand for bio-fuels like ethanol grows, sulphuric acid has become scarce and expensive. Roughly 60% of sulphuric acid produced goes into agriculture, primarily in the manufacture of phosphate fertilizers which feeds into corn, soybean and sugar production. Metal processing also relies on sulphuric acid and bumped up against growing demand in the agricultural sector, producers in our view will be scrambling to secure long-term supply. Exploration expenses are another source of increased expenditure as drill-rigs are in short supply, primarily as a result of a lack of operators. Exchange rate trends The depreciation of the US dollar during this decade has also been a significant factor affecting costs. According to Brook Hunt, locally denominated costs (ie producer countries) in US dollar terms rose by 56% between 2002 and 2007. Although a falling dollar help to limit input costs in local currency terms, since commodities are priced in USD, producer revenues outside the US feel the pinch.
5)
1) Commodity prices relative to G7 per capita income In April 2008, the 3M copper forward price hit a new all time nominal high of USD8730/tonne. However, relative to per capita income the price of copper and other industrial metals are simply moving back to the levels that prevailed during most of the 1970s and 1980s. We track how many tonnes of copper could be purchased with respect to G7 per capita income, that is how much copper could an average G7s consume purchase over time. We have repeated this exercise for other commodities and the full results are presented in the Commodities Chartbook at the end of this report. We find that during the 1960s and 1970s copper was even more expensive relative to per capita income than it is today. Indeed at the beginning of the 1970s G7 per capita income was only able to buy 2.6 tonnes of copper. To bring purchasing power back to these levels would require the copper price rising to USD17,000/tonne. If we assume copper prices move back towards fair value over the next five years it would only imply the copper price declining towards USD5,550/toone (USc2.52/lb). Moreover, the lesson of the past 45 years would suggest that copper prices can trade at excessively cheap or expensive levels relative to per capita income for extended periods of time. This could signal that copper prices could stay at or above current extreme levels well into the next decade. At the moment, oil and industrial metals are trading rich relative to per capita income. This is not the case for precious metals and agricultural prices, which are trading close to fair value on this measure.
Page 22
27 June 2008
Commodities Quarterly
35 30 25 20
15
10
15 10
0 1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
2) Commodity prices in real terms The second route to establish the long-term fair value of commodity prices is to examine the price of individual commodities in real terms. The equilibrium oil price is not necessarily a constant, it can change up and down because of currency fluctuations, GDP changes, technology, geopolitics, taxation, environmental rules, etc. On the supply side there are the competitive costs of tar sands, heavy oil, gas-to-liquids, and biofuels among others. On the consumption side, technology may foster efficiency gains or offer up alternate fuels that could put downward pressure on prices, which in turn could force costs down. However, if we assume prices in real terms move back to their long run historical averages it would imply copper pries moving back to USD3,850/tonne (USD1.75/lb) and oil prices moving back to USD40/barrel. 3) Oil: Finding and development costs One key factor that offers compelling evidence for longer-term energy price forecasts is the continuing escalation in finding & development costs, Figure 4. The rise in per barrel costs reflects a combination of higher total finding and development costs which include bonuses, royalties, labour, materials and poor reserves replacement.
Prior down-cycles in oil prices were generated by a combination of lower oil products demand, dramatic improvements in seismic and drilling technology, and greater access to reserves (aperatura & perestroika). A repeat of this confluence of such events is certainly possible, but seems improbable over the course of the next few years. Oil demand growth is increasingly being driven out of Asia and the Middle-East, regions not immediately susceptible to economic downturns in the US or Europe. Technology is obviously improving, but does not appear to be on the verge of the enormous breakthroughs (3-D seismic, horizontal drilling, and subsea completions) of the last two decades. Access to the areas of the highest geologic potential seems to be decreasing as resource nationalism moves steadily into the mainstream of geopolitical discourse in the producing nations. Per barrel costs come from the interaction of absolute costs and reserve changes. A decline in reserve additions over the past few years has been an important part of the overall increase in finding costs. BP just reported in its June 2008 Statistical Review of World Energy that global oil reserve actually fell from 2006 to 2007. Rising development costs, particularly in the offshore sector where rising global demand have driven drilling, sub-sea and production facilities costs steeply upwards, continue to plague the industry. In addition, the price of exploration licences for new acreage in prospective basins has surged. High oil prices are encouraging host governments to tighten fiscal terms. These factors are being reflected in higher total F&D costs. Our projections for real 2007-2015 F&D costs assume a 20% rise in 2007, 15% in 2008, 10% in 2009, and trailing off toward 5% in 2015. This would push F&D costs overUSD35 in 2010 and toward USD45-50/bbl by 2015.
Page 23
27 June 2008
Commodities Quarterly
2007E - 2015E
of the cost curve meaning their costs are higher than most of the rest of the world. Thus an increasing share of production will come from the upper end of the cost curve. Given the recent increase in Chinese power tariffs, even higher costs are inevitable in our view. Our own estimates are close to the ones Chalco published for tis own production costs in FY06 of USD2,126/t (DB estimate was USD2,161/t) and FY07 of USD2502. We estimate these have risen to USD2,700/tonne in 2008.
Figure 3 illustrates the relationship between oil prices and finding & development costs. Using EIA data on F&D costs over the period to 1980-2006, the regression equation is: Oil price = 2.6x F&D costs + 7.5 From a cost perspective, this suggests our nominal oil price deck is supported at USD100/bbl in 2010 and USD125/bbl in 2015. . 4) Metals: Marginal cost of production The marginal cost assumption is that price will be driven by the production cost of the marginal producer. Our goal is to discover the threshold rates of return required to continue to attract capital into commodity production and the hurdle rate required for the marginal producers at the upper end of the cost curve to successfully exist. Compared to oil, we have found deriving up-to-date and detailed production costs for metals producers more challenging. Although we have the ability to employ detailed third party cost models to obtain data from a sample of producers, we find this method limiting due to the models inability to capture more recent trends, particularly energy price inflation and associated cost pressures. Although, we have endeavoured to flex these models to take into account such trends, we have also attempted to estimate total costs of producers located at the upper end of the cost curve by taking a sample of some of these producers. The results of this exercise are not surprising with all examinations showing substantial increases to operating and capital costs primarily as a result of energy price inflation. Trends in the aluminium market provide the clearest evidence. The aluminium production process is the most energy intensive among the industrial metals complex and as such is the most exposed to energy price inflation. China is by far the largest primary aluminium producer, currently supplying around 35% of the worlds metal, a share forecast to increase to 42% by 2012. At the same time, Chinese producers operate at the upper end
Source: Brook Hunt, Company data, DB Global Markets Research *Other costs include freight, petroleum coke, labour and consumables
Conducting this exercise in other metals produced similar results, namely a strong rise in production costs. Given the decline in a number of industrial metal prices over the past year for example in nickel and zinc, we believe current prices in these metals are close to or below marginal cost of production. We therefore assume the threshold rates of return required to continue to attract capital to new projects in order to sustain production in line with long-term demand trends has significantly increased. At the same time, we are also taking the opportunity to acknowledge the changing trends in long-term demand growth which in the past have been projected, in our opinion, too conservatively. Generally speaking, most global economic forecasters (including Deutsche Bank) have throughout most of this commodities cycle predicted a gradual easing of global demand growth including Chinese and other emerging market growth. With these assumptions baked into producers forecasts, the incentive for investment in new projects had been reduced. We believe this mind-set is changing rapidly and we are now seeing evidence of miners investing in regions in the world with high sovereign risk that would not have been considered in the past. In addition, many producers are starting to reevaluate mines as long-term average prices increase. 5) Forward markets Another route we employ to assess long-term commodity prices is the futures market and specifically the current five year forward price. For crude oil we find that long-dated crude oil prices are trading at just over USD130/barrel. However, we need to recognise that using the forward curve as a future predictor of price has tended to work well for the year ahead but its accuracy starts to fade over a multi-year horizon. For example, the gold forward curve is simply a function of the current spot
Page 24
27 June 2008
Commodities Quarterly
price and interest rate and warehousing costs going forward and can therefore move substantially on a multi-year basis solely as a result of spot price changes.
Conclusion We remain convinced the rally in commodity prices that began around five years ago is fundamentally based. We do not ascribe to the view that his is a speculative bubble rather the forces that depressed commodity prices in the 1990s have all been moving in reverse during this decade. This does not preclude the risk of short-term corrections in these markets, but, in our view the world is undergoing a sector-wide re-rating in the fundamental valuation of commodities.
Joel Crane, (1) 212 250 6263 joel.crane@db.com Adam Sieminski (1) 202 662-1624 adam.sieminski@db.com
Like crude oil, the copper forward curve is heavily backwardated. This, in our view, reflects tight market fundamentals and in particular the low level of global inventories particularly across the industrial metals complex.
Page 25
27 June 2008
Commodities Quarterly
#8 Crude Oil
Moving Towards A Tipping Point?
We believe energy demand is responding to rising income levels and changing lifestyles of a growing population across the globe. The latest DB forecast for the global economy puts world GDP growth for 2008 at 3.5%. This is significantly less than the 4.6% forecast in December 2007, but not a recession. World oil markets are starting to loosen up fundamentally, but the degree of tempering does not appear to be acute and corrective forces are taking time. Even with an economic slowdown, we expect oil demand will rise by 0.5-0.8mmb/d in 2008 and by 1.0-1.2mmb/d in 2009, mostly in the subsidised nations of Asia and the Middle East. We expect non-OPEC supply growth forecasts look for a rise of 1.0mmb/d for 2008 and 2009. In our view, this suggests that the call on OPEC for 2008 and 2009 will not be under much, if any, downward pressure. We estimate that oil prices at USD138/bbl are currently is now approaching the top of the USD100-150/bbl extreme level of valuation implied by a number of measures.
Furthermore, we believe the oil producers are becoming more accustomed to higher prices and our review of the extremes in oil valuation suggests that prices might have to remain at elevated levels in order to curb demand growth
1.8 1.7
We believe projected energy demand is increasingly coming from developing and emerging economies, while oil and natural gas production in the US and Europe is struggling. We believe the endowment of global energy resources is enormous, but it is subject to above-ground risks such as access, resource nationalism, security concerns, political shifts, and environmental considerations. Al of which can significantly affect the productibility, conversion, and deliverability of energy, and the timing of needed investments in our view. These factors have weighed heavily, in our opinion, on what investors have been willing to pay for oil and natural gas futures. We believe the supply and demand fundamentals in the oil market continue to be relatively tight. Even with a significantly slower global economy, we find that world oil demand is still growing, driven largely by growth in Asia and the Middle East where strong economies, growing populations and subsidised consumers are prevalent. Since we expect non-OPEC supply to rise by the same level as demand, on our estimates OPECs market share is not expected to erode in 2008. Although we expect a rebound in non-OPEC supply growth in 2009, the global economy should be doing better then too, and that implies higher oil demand and yet another year with minimal pressure on OPECs quota system.
Page 26
World demand We believe that world oil demand is almost entirely a function of GDP growth at 1:1 and an annual offset of approximately 2% for technological progress (aided by higher prices). As shown if Figure 2, the IMF now estimates a baseline world GDP forecast of 3.7% (down from a 4.8% estimate late last year), and now sees global GDP growth potentially under 2% in 2009, but views this outcome as a very low probability. With inflation emerging as an issue in both China and India, we believe downside growth risks are emerging in these economies, but so far, the non-OECD Asia countries are experiencing strong oil demand growth. In 2008, we expect total world demand to rise to 86.7mmb/d against an upwardly revised 86.0mmb/d in 2007, for a gain of 0.7mmb/d. We believe that OECD demand will fall by 0.4mmb/d and that the rest of the world will see a gain of approximately 1.1mmb/d. We forecast demand in China to grow by 0.4mmb/d, while the Middle East increases by 0.3mmb/d. We expect US consumption to fall by 0.4mmb/d and Europen oil demand growth to be flat.
1 2005 10% 30% 2006 50% 2007 base 2008E 50% 2009E 70%
Source: IMF
27 June 2008
Commodities Quarterly
Non-OPEC oil supply We find that Non-OPEC supply remains weak despite six years of rising prices. EIA analysts who correctly predicted low non-OPEC supply performance in 2008, remain convinced that the pace and timing of new nonOPEC supplies will be subject to project delays and impaired by accelerating production decline in older fields. Despite annual average growth in non-OPEC output of only 0.5mmb/d in our forecast for 2008, growth from Q2 to Q4 in 2008 could average close to 1.5mmb/d based on estimates from OPEC, DOE/EIA, and the IEA. We expect 1.2mmb/d of non-OPEC growth in 2009, almost exactly matching the total increase in global demand and thus suggesting that OPECs market share in 2009 is not threatened.
2002
2003
2004
2005
2006
2007
2008
OECD inventories The IEA reported in on 10 June that OECD industry stocks fell 8mmb in April to 2562mmb, in stark contrast to the typical build. Stocks should build over the summer, but, we believe they are unlikely to exceed 53-54 days of forward cover by Q4. Although this is substantially higher than the 49-51 day lows seen over the past few years as shown in Figure 5, 53 days coverage is, in our view, still relatively skimpy compared to the 55-56 day levels seen in 2002. It is this relatively low level of cover that has been a continuing source of worry for IEA officials over the last year. In some respects, the Saudi offer made in June to increase production regardless of their view that markets are well supplied is a reflection that the IEA arguments have some merit.
Supporting a relatively bullish view of near-term supply and demand balances, the EIA still holds the lowest non-OPEC supply forecast for 2008, a gain of a mere 0.3mmb/d estimate. The EIA also has the lowest Q2 to Q4 growth estimate for non-OPEC production at 1.1mmb/d. This contrasts with the IEAs Q2 to Q4 rise of 1.9mmb/d. We believe the growth in Canadian oil sands, Russia and Brazil remains uncertain. We expect the sizeable increase in non-OPEC growth in 2009 shown in Figure 3 will be subject to the same delays that have resulted in disappointing production growth over the last few years, but higher prices, with normal lags, are likely to boost prospects. OPEC spare capacity Spare oil production capacity is almost solely held within OPEC. The IEA estimates that OPEC effective spare capacity (excluding Iraq, Indonesia, Nigeria and Venezuela which face ongoing security, operational or investment issues) has now slipped below 2mmb/d. Both the IEA and US DOE/EIA see OPEC crude capacity increasing by the end of 2008 and into 2009, especially in Saudi Arabia with streaming of the delayed AFK (Khursaniyah) project. Nevertheless, we expect there is little hope of achieving levels similar to 2002 absent unexpectedly poor global GDP performance.
Jan-00
Jan-02
Jan-04
Jan-06
Jan-08
Oil price outlook In our view, there is a tug of war occurring in the oil markets based on two distinct views of how the marginal price of oil is set. One view is that marginal cost of supply should dominate, and we see this price being driven by changing cost and access issues; for now it might be near USD75-100/barrel. The other view is that prices are rising toward the level required to destroy demand, or to get it to slow dramatically. We
Page 27
27 June 2008
Commodities Quarterly
believe this could require oil prices in excess of USD150/barrel that would presumably bring demand growth under control over the next 5 years and thus eliminate the market worry (expectation) that demand growth trends are running well in excess of supply growth trends. Deutsche banks oil and gas price forecasts shown in Figure 6 are substantially higher than our forecast made in March of this year. Our prior forecast for Brent and WTI crude oil prices in 2009 called for an USD102.50/barrel average, and long-term prices settling toward USD85/barrel. We have based the new deck on a view that incorporates the continuing lags being seen in global supply/demand elasticities, and the persistent escalation in finding and development costs. In terms of what is likely to arrest the rise in the oil price in near term we would view a further dismantling of fuel subsidies across the non-OECD, world GDP growth falling to 2.5% or below and/or a stabilisation or even strengthening in the US dollar. Aside form the US dollar, these would carry a strong chance of depressing world oil demand. A third possibility is the apparent decision by Saudi Arabia to sharply increase production over the course of the next few months. Saudi Arabias King Abdullah has reportedly made the decision that oil prices are too high based on an alarming wave of protests and strikes around the world over crippling prices, the growing food crisis that is seen as a direct consequence of the oil price rise, and the rising danger to producers from keen interest in alternative technologies and energy sources. This view was apparently the driving force behind the Kings desire to convene the Producer-Consumer conference held in Jeddah on June 22. In our view the best long-run outcome from the meeting would involve Saudi Arabia pumping more oil and holding fast to its schedule for increasing production at Khursaniyah along with the mid-2009 streaming of the 1.2mmb/d Khurais light crude project. US consumers could further reduce their gasoline use, and Fed attempts to stabilise the dollar might also help in our view. Further steps to cut Chinese fuel subsidies could also help to reduce oil demand over the medium term in our view, and there is a chance that Russia will be able to stem its Q1 oil production decline in our view. Meanwhile the international integrated oil companies could, in our view, boost their capital spending budgets and keep their refineries running.
Since we believe that demand is showing signs of slowing, our forecast for next year is that oil prices should average USD120/barrel, but, we still see the risk to the upside from geopolitical events. Our forecast for the longer-term is predicated on the belief that prices will eventually settle toward the cost of marginal supply discussed at length in the Long-Term Prices article of this report, which would imply close to USD100/barrel in 2010 and adjusted for inflation rising by USD5/bbl annually. Macroeconomic impact of high oil prices Work done by both the IMF and IEA suggests that (1) the vulnerability of oil-importing countries to higher oil prices varies markedly depending on the degree to which they are net importers and the oil intensity of their economies, and (2) the economic impact of higher oil prices on oil-importing developing countries is generally even more severe than for OECD because their economies are more energy-intensive. The models built by the IEA and IMF in 2004 demonstrate that a sustained USD10 per barrel increase in oil prices from USD25 to USD35 was estimated to result in the world as a whole losing 0.5% of GDP in the first and second years of higher prices. With oil prices up by nearly an order of magnitude beyond the 2004 studys USD10/bbl rise (an increase of USD100/bbl from USD25 to USD125), the rule is obviously flawed since the GDP impact has been considerably less. As a rough revision to the rule, we
Page 28
27 June 2008
Commodities Quarterly
might attribute the IMFs downgrading of global growth from 4.9% to 3.7% to higher energy prices (excludes possible impacts from housing and financial issues), when the consensus amongst analysts shifted from clustering on USD50/bbl oil prices to targeting USD100/bbl. This would suggest that every USD25/bbl rise might cause a loss of 0.5% GDP using a base near USD75/bbl for 2007. Other issues to consider in the macroeconomic arena that could amplify negative economic impacts include: (1) The enormous transfer of wealth from oil importers to oil exporters (with 55mmb/d in global trade, each USD25/bbl is the equivalent of 500 billion dollars) Loss of business and consumer confidence, and Policy mistakes made as pressure rises to insulate commodity users from higher prices through greater subsidies and trade restrictions.
(2)
(3)
Figure 7: The level of oil prices required to represent extreme levels of valuation versus a variety of indicators
Indicator In real terms (PPI) Analyst forecasting error In real terms (CPI) Versus the US dollar Futures market forecasting error Relative to per capita income As a percent of US disposable income As a share of the S&P500 As a percent of global GDP Average
Source: DB Global Markets Research
Oil price level USD94 USD116 USD118 USD120 USD130 USD134 USD145 USD145 USD150 USD128
Conclusion We believe the oil market is in a state of confusion unable to believe that the forces that have driven prices higher over the past year, namely OPEC production cuts, non-OPEC supply problems, strong economic growth in the emerging markets and a falling US dollar may be moving in reverse or at least not moving in the direction of even higher oil prices. We believe the dismantling of fuel subsidies could be an important catalyst for even lower oil demand growth going forward. Adam Sieminski (202) 662-1624 adam.sieminski@db.com
Page 29
27 June 2008
Commodities Quarterly
China's grow th slow er in 2010-15 but is still the fastest global rate
7.5 5.0 7.5 19.9 12.8 0.1 3.8 2.7 3.9 1.6 2.6 14.6 2.1 0.3 49.7 0.3 4.8 0.1 31.1
7.5 4.6 7.4 19.5 13.1 0.1 3.9 2.8 4.1 1.6 2.6 15.1 2.1 0.5 50.3 -0.1 5.2 -0.2 31.5
7.7 4.5 7.3 19.5 13.9 0.1 3.7 2.7 4.3 1.6 2.8 15.3 2.1 0.7 51.5 -0.1 5.4 -0.4 31.4
7.6 4.3 7.2 19.1 14.4 0.1 3.7 2.6 4.4 1.6 2.8 15.3 2.2 0.9 51.9 -0.1 5.6 -0.4 32.1
7.5 4.1 7.4 19.0 14.6 0.1 3.7 2.5 4.5 1.6 2.7 15.2 2.2 0.9 51.9 -0.1 5.8 -0.4 33.0
7.4 3.9 7.8 19.1 14.7 0.1 3.7 2.4 4.5 1.6 2.7 15.0 2.2 0.9 52.0 -0.1 6.0 -0.4 33.8
7.2 3.7 8.1 19.1 14.8 0.1 3.7 2.3 4.4 1.5 2.6 14.7 2.3 1.0 51.8 -0.1 6.3 -0.4 35.0
7.1 3.5 8.4 19.1 14.9 0.1 3.6 2.2 4.4 1.5 2.5 14.3 2.3 1.0 51.7 -0.1 6.5 -0.4 36.1
7.0 3.4 8.7 19.0 15.0 0.1 3.6 2.1 4.3 1.4 2.5 14.1 2.3 1.1 51.6 -0.2 6.6 0.0 37.1
-1.8 -3.8 1.0 -1.4 8.3 -2.3 2.2 3.0 2.6 -3.3 3.7 1.8 1.6
0.4 -5.1 -0.4 -1.3 4.1 -3.5 0.7 -0.8 3.2 -2.4 2.7 1.1 3.1 53.4 1.3 4.5 0.4
-1.5 -4.8 3.6 -0.1 0.9 0.0 -0.5 -3.7 -0.7 -3.3 -2.4 -1.7 1.5 5.0 -0.1 3.5 3.0
Non-OPEC production could peak som etime before 2015 unless NIM BY restrictions are lifted
10.1 13.6 11.9 6.8 90.2 1.3 36.6 -0.5 1.8 0.0 1.1 0.3 1.3 -0.5
72.66 119.30 120.00 100.00 105.00 110.00 115.00 120.00 125.00 72.36 120.00 120.00 100.00 105.00 110.00 115.00 120.00 125.00 66.66 114.30 115.00 95.00 100.00 105.00 110.00 115.00 120.00
Source: OPEC Secretariat, International Energy Agency, US DOE/EIA, DB Global Markets Research
Page 30
27 June 2008
Commodities Quarterly
In terms of crude oil we examined the rates of oil demand growth in a different sub-sets of countries: (1) (2) (3) The OECD where, with the exception of Mexico, fuel subsidies do not operate Nations that subsidise prices paid by their oil users, primarily in the non-OECD non-OECD countries that do not significantly subsidise domestic oil consumption.
We find that of the roughly 86mmb/d of oil use in 2008, 45mmb/d is in the unsubsidised OECD. Not surprisingly this region has the lowest growth rate in oil demand. 25mmb/d is consumption that is occurring in countries where fuel is subsidised, such as India and China. The remaining 15mmb/d is occurring in nonOECD countries where oil consumption is mostly unsubsidised. Not surprisingly oil demand growth rates between 2000 and 2007 are highest in those non-OECD countries which operate fuel subsidies. However, the characteristics of these countries where fuel subsidies exist can be classified according to oil exporters and oil importers. We find that of the 25mmb/d in oil consumption occurring in subsidised countries, two-thirds are oil importing nations such as Indonesia and the remaining one third are oil exporting countries such as Iran and Venezuela. Not surprisingly, oil importers rather than exporters have been under greater pressure to dismantle fuel subsidies in a period of high oil prices. This has been highlighted by recent government action in cutting subsidies in Indonesia, Malaysia, India and more recently China.
The rise in food and energy prices over the past few years has led to a significant policy headache for central bankers around the world. These problems have been most acute in the developing world where food and energy constitute more than 40% of their CPI baskets. Not surprisingly many central banks around the world are failing to meet their inflation targets. One solution to this problem has been the proliferation of fuel and food subsidies around the world. However, such steps are leading to distortions in global commodity markets. In the case of crude oil, it is sustaining the vigour in crude oil demand where subsidies exist and delaying the point at which oil prices should fall in response to higher prices. In agricultural markets, governments are trying to restrict the export of food via export tariffs given the risk of shortages at home. This is leading in some circumstance to domestic farmers, unable to access higher international prices for their products, to cut plantings due to the rise in input costs such as fertilizers, land and water. Consequently government interference is leading to lower agricultural production as a result of these distortions.
2.5% 1.8%
0.4%
The remaining administered fuel price regimes that the IEA would like to see move toward market-set petroleum product prices include most of the oil producing countries in the Middle East, Nigeria, Venezuela, and several large non-OPEC Latin American countries, especially Argentina, Mexico, Colombia and Chile.
Page 31
27 June 2008
Commodities Quarterly
In 2008, the IEA forecasts global oil demand rising 800mmb/d. If we categorise oil demand growth into the key subsidised and non-subsidised nations we find that .the entire growth in oil demand in 2008 is originating from the group of nations that operate fuel subsidies. It is still clear to us that below-market fuel prices are a major factor in promoting higher oil demand growth rates.
1008 802
240 -447
Non-OECD Subsidised
Non-OECD Unsubsidized
OECD
Total Demand
Admittedly, removing subsidies will not completely slow oil demand growth, in our view. The IEA points out that as long as economic activity, the primary driver of oil demand growth, remains strong, oil demand will shrug off moderate price hikes and will continue to be buoyant. In some circumstances, removing price controls could even accelerate oil demand growth in the short run if retail price hikes eliminated artificial shortages caused by controls in the first place. Nevertheless, we believe that ending subsidies sets the correct tone for shifting the demand curve over the long-term.
Page 32
27 June 2008
Commodities Quarterly
Short term we believe the oil curve is likely to stay in a modest contango because we expect crude oil fundamentals to lossen and inventory conditions to become less tight:
1. 2. Saudi Arabia is pumping more oil. Demand destruction happening: is finally
Asian governments oil subsides are being removed or at least reduced; many airlines are closing down or at least grounding planes and shrinking fleets; mileage driven & gasoline consumption have declined sharply in the US; global growth is moderating and at risk of slowing further on rising inflation Refining margins remain weak.
3.
Trade recommendations:
Long DBLCI-OY-CL vs. short DBLCI-CL Long DBLCI-OY vs. short DBLCI Short M02 WTI vs. long M03 WTI and roll the positions every month
Risks
Hurricane attacks, refineries ramping up productivities, heightened geopolitical risks, oil production problems or accidents
Commodity index investment has often been cited as the culprit behind the appearance of contango in the crude oil forward curve when its first appeared in an environment of record high crude oil prices in 2005. However, such claims were seriously flawed in our view. Taken to its extreme, it would imply the crude oil curve would be perpetually in contango not least given the surge in funds tracking commodity indices since 2005. However, this has not been the case. Crude oil term structure: what has happened In this section we analyse what we believe are the drivers of the crude oil term structure to determine whether the crude oil price, speculators or the crude oil inventories are most connected to the oil curve. To start with we review the latest move in the crude oil term structure. In Figure 1 we overlay the crude oil forward curve with historic WTI prices. Prior to 2002, backwardation was normally associated with high crude oil prices (red curves), and contango at low prices (green curves).
Page 33
27 June 2008
Commodities Quarterly
Crude oil term structure vs. crude oil price The crude oil price and the term structure are plotted against each other in Figure 3. We find that there is little correlation between the two. The uninterrupted rally in frontline prices from the beginning of 2007 has seen the second-to-fourth month curve flipping from steep contango to backwardation and back to steep contango.
160
WTI frontline prices (USD/bbl, lhs) M2-M4 Nymex WTI spread ($/bbl, inverted scale, rhs)
-4.0
140
-3.0
However this relationship collapsed in 2005 when contango made its first appearance in a high oil price environment. Contango persisted for more than two years and it was not until the middle of 2007 that the crude oil curve flipped back into backwardation only to reverse and move back into contango in May 2008. We find the first assault to USD130/bbl moved the entire crude oil curve into steep contango, the top (blue) line in Figure 2, where we plot the price of the futures contracts as a percentage of the frontline crude oil price. The subsequent rally has seen the degree of contango moderating and in fact, the USD10/bbl jump in frontline prices to reach an all time high of USD138.54 briefly sent the curve into steep backwardation, the bottom (green) line. The current curve is shown in orange (middle line). Although it is relatively flat, it has a small contango at the front end, backwardation in the middle of the curve and contango from three years out.
3.0
In addition, we also examined whether there is any leading, lagging or coincidental relationship between crude oil price and the crude term structure. In the following analysis we usie data from 3-Jan-05 onwards for three different scenarios to see: 1. Whether the crude oil curve is leading the crude oil price. Whether the crude curve is moving in sympathy with the crude price. Whether the crude curve is lagging the crude price.
2.
3.
98 current 96
In Figure 4 we applied a 3-month-lead to the curve (the time spread between M02 and M04), this means the crude oil price on 3-Jan-2005 is plotted against the time spread on 1-Oct-04. In Figure 5 there is no time difference, and in Figure 6 a 3-month-lag. In all cases the correlation is very weak and therefore we believe the crude oil price does not have any leading, coincidental or lagging relationship with the crude oil term structure.
94
92 Forward month
Page 34
27 June 2008
Commodities Quarterly
100
80
60
Crude oil term structure vs. speculative positioning The relationship between the crude oil price and speculative positioning has been well researched. There are many academic studies concluding speculators are price followers rather than price setters, for example research from both the IMF and the CFTC. In Figure 7 we examine any possible connection between the net speculative position in WTI future contracts against the crude oil term structure. Not surprisingly we could not find any compelling evidence linking the two.
40
Figure 7: WTI net speculative positioning vs. crude oil term structure
150 WTI net spec positions ('000, lhs) M02-M04 (USD/bbl, rhs) speculators net long 6 5 4 3 50 2 1
3-months lead: e.g. crude @ 3-Jan-2005, spread @ 1-Oct-2004 Source: DB Global Markets Research
100
0 -1
-50
100
-2 -3 -4 Sep-08
contango
80
-100 Jan-02
Sep-02
May-03
Jan-04
Sep-04
May-05
Jan-06
Sep-06
May-07
Jan-08
60
40
No lag: e.g. crude @ 3-Jan-2005, spread @ 3-Jan-2005 Source: DB Global Markets Research
Crude oil term structure vs. open interest & volume In this section we examine the movement of the crude oil price along each part of the term structure and compare this against the open interest and trading volumes at that respective part of the curve. In Figure 8, the blue diamonds represent the price movement of each WTI contract along the forward curve out to 72M between 21-May-08 and 3-Jun-08. The frontline contract lost USD8.26/bbl, and this compared to the decline of USD14.81 suffered by the M72 contract (Dec2016 contract). The open interest and total trading volume for each contract is highlighted in grey and dark blue bars respectively.
100
80
60
40
3-months lag: e.g. crude @ 3-Jan-2005, spread @ 4-Apr-2005 Source: DB Global Markets Research
Page 35
27 June 2008
Commodities Quarterly
Figure 8: WTI price movement along different part of the curve, vs. open interest and volume (WTI on the way down, frontline from USD133/bbl to USD124/bbl)
350000 open interest for each contract (3-Jun-08, lhs) total volume for each contract (3-Jun-08, lhs) 300000 16
Figure 9: WTI price movement along different part of the curve, vs. open interest and volume (WTI on the way up, frontline from USD101/bbl to USD133/bbl)
350,000 open interest for each contract (3-Jun-08, lhs) total volume for each contract (3-Jun-08, lhs) 50 48 46
Jul'08
15
300,000
Dec'16
100,000
10
50000
50,000 32 0 spot 5M 10M 15M 20M 25M 30M 35M 40M 45M 50M 55M 60M 65M 70M 30
0 spot 5M 10M 15M 20M 25M 30M 35M 40M 45M 50M 55M 60M 65M 70M
We find that along the crude oil futures curve, the further away from the prompt month, the bigger the price movement. This also coincides with decreasing liquidity along the curve, as measured by open interest and trading volumes. We do not believe this is surprising since the liquidity at the back-end of the curve is extremely thin. For example, open interest in the frontline contract is roughly 100 times that compared to the Dec16 contract. The trading volume in fact sees a much bigger disparity. While the open interest and trading volume of the frontline contract is roughly equal, this is not the case for the Dec16 contract. The open interest is roughly 20 times more than the trading volume on 21May-08, and the ratio is almost 2000 times on 3-Jun-08. We also found this relationship held both when crude oil rallied from USD101/bbl to USD133/bbl, and on the way down from USD133/bbl to USD124 as shown in Figure 9.
Therefore we believe if speculators are to blame for the fluctuation in the oil price, they assume a much bigger role on very long dated oil contracts, as opposed to the prompt contracts. However, we would not generalise it to the entire crude oil market in such a way that speculators are solely responsible for the run up to USD140/bbl. Crude oil term structure vs. crude oil inventory We have always tended to believe that the crude oil term structure (and to a certain extent, crude oil prices) are primarily driven by physical market fundamentals, such that an oversupplied market with excess crude oil inventories and weak demand would push the oil curve into contango and on the other hand, shortages would drive up the storage risk premium and therefore encourage crude to leaving storage and be delivered to those who are willing to pay a premium for immediate access. In fact, the sudden flip of the oil curve from backwardation to contango also coincides with a rise in inventory levels at Cushing, the delivery point of the WTI contract as shown in Figure 10.
Page 36
27 June 2008
Commodities Quarterly
Figure 10: WTI term structure vs. inventory level at Cushing (the exact delivery point of the WTI future contracts, data since April 2004)
Cushing Crude Stocks (million bbls, lhs) 30 M2-M4 Nymex WTI spread ($/bbl, inverted scale, rhs) -3.0 -4.0
Figure 12: De-trended & de-seasoned Padd 2 crude oil stocks vs. WTI term structure
20 Padd 2 Crude Stocks deviation from 5-year avg ('000 bbl, lhs) M2-M4 Nymex WTI spread ($/bbl, inverted scale, rhs) -4 contango
15
-3
-2 10
-2.0 25 -1.0 contango 0.0
-1 5 0
20
0 1
1.0 15 2.0
-5
-10
10 Jan-04 3.0 Jul-04 Jan-05 Jul-05 Jan-06 Jul-06 Jan-07 Jul-07 Jan-08
Jan-04
Since Cushing storage data are only available from April 2004 onwards, we examined a longer term relationship by using Padd 2 inventory. The relationship has been relatively strong since January 1995, Figure 11.
Conclusion: short term we believe the oil curve is likely to stay in a modest contango because we expect crude oil market fundamentals to loosen and inventory conditions to become less tight: 1. 2. Saudi Arabia is pumping more oil. Demand destruction happening: is finally
Figure 11: WTI term structure vs. inventory level at Padd 2 (the delivery point of the WTI future contracts, data since January 1995)
90 85 80 75 -1 70 0 65 1 60 55 50 45 Jan-95 Jan-97 Jan-99 Jan-01 Jan-03 Jan-05 Jan-07 2 Padd 2 Crude Stocks ('000 bbl, lhs) M2-M4 Nymex WTI spread ($/bbl, inverted scale, rhs) contango -4 -3 -2
Asian governments oil subsides are being removed or at least reduced; many airlines are closing down or at least grounding planes and shrinking fleets; S mileage driven & gasoline consumption have declined sharply in the US; global growth is moderating and at risk of slowing further on rising inflation Refining margins remain weak.
3.
Trade recommendations:
Long DBLCI-OY-CL vs. short DBLCI-CL Long DBLCI-OY vs. short DBLCI Short M02 WTI vs. Long M03 WTI and roll the positions every month
In fact, if we refine the data further, by taking out the trend and seasonality (by subtracting the 5-year average value for that reporting week), we found an even closer relationship between the WTI term structure and the deviation from the 5-year average storage level, Figure 12. However, the small divergence between the two means either the storage level has to rise further in respect to the average level or the current contango is not sustainable in our view. In addition, we believe this also helps explain the sudden, but, brief move back into backwardation and the moderating degree of contango of the crude oil curve.
Risks
Hurricane attacks, refineries ramping up production, heightened geopolitical risk, oil production problems or accidents
Page 37
27 June 2008
Commodities Quarterly
Conclusion: Volatilities has tended to subside over the summer months across asset classes and commodities volatilities is no exception. Given the current elevated level of implied vol we believe shorting vol via variance swap offers good value. Trade recommendations: Risks Continuous disruption in the supply of crude oil, gold, or platinum, in the form of geopolitical or weather events, as well as production problems or accidents. Short Dec09 WTI variance swap Short Dec09 Gold variance swap Short Dec09 Platinum variance swap
Commodities vol is trading at elevated level In the crude oil market, it is not only price that has hit a new record high, so has implied volatility and in particular longer dated implied vol. WTI long dated implied vol has climbed to its highest level in five years, Figure 1. The 24-month implied vol has climbed almost 12 vols to 32 vols since beginning of this year (data as of June 24, 2008).
Figure 1: WTI long dated implied vol has climbed to its highest level in five years
Page 38
27 June 2008
Commodities Quarterly
Indeed, the entire term structure of WTI implied vol has flattened since the beginning of the year, Figure 2. Figure 2 compares the implied volatility today (red circles) with the vol curves at the beginning of the year (yellow diamonds). The blue box represents the 25-75 percentile range and the bottom and top end of the whiskers mark the 5 and 95 percentile range of the last six year distribution.
Figure 3: Crude oil realized vol has failed to catch up with the implied vol
Oil implied vol cone: current curve vs. past 6-year distribution The blue boxplot for each month shows: i) the interquartile range (blue box) ii) the 5-95 percentile range (whisker) iii) the median of the 6-year history (dashed line)
Median
45%
45%
03-Jan-08
40% 35% 30% 25% 20% 15% 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Forward month
23-Jun-08
Indeed the entire forward curve has shifted from the lower quartile at the beginning of this year (yellow diamonds curve) to currently standing significantly above the 95-percentile mark (orange circles curve). From the fourth month onwards WTI vol is trading above the 95-percentile mark indicating how extreme the level of implied vol is trading currently. We believe such an elevated level of implied vol is not sustainable and we believe it is a good sell. If the WTI implied vol is trading rich with respect to history, how about versus realised vol? In Figure 4 we compare the WTI implied vol vs. realised vol. The M12 contract realised vol has climbed higher in the current rally, however, it has failed to catch up with the significant rally in implied vol. In fact realised vol has yet to break the previous high set by the Katrina hurricane related shutdown in 2005.
In order to assess how extreme the current level of realised vol is, we compare realised vol of a wide selection of dated WTI contracts as far back as the Dec03 contract as shown in Figure 5. Each dated contract had tended to start trading at a lower realised vol level and which has climbed steadily as the contract approaches maturity. That was one of the factors why we recommended going long the WTI variance swap at the beginning of this year and highlighted by the vertical line in Figure 5. The sharp rally in vol over such a short period of time motivated us to take profit of our long vol position two weeks ago. We believe the time is ripe to reverse the trade and go short this time.
Apr-04
Oct-05
Apr-07
Oct-07
Page 39
27 June 2008
Commodities Quarterly
Dec09 WTI implied vol is currently trading at around 35% which is substantially higher than the previous peak reached by the historic dated futures contracts as shown in Figure 6. The thick orange line near the top of the chart is the current level of Dec09 implied vol, which is roughly 7 vols higher than peak reached by the most volatile Sep05 contract.
Figure 7: Implied vol of platinum is also trading much higher than the realized vol
Figure 5: WTI Dec09 implied vol is trading substantially higher than the previous peak reached by the historic dated futures contracts
WTI 260 day realised volatility
37.5 35.0 32.5 30.0 27.5 25.0 22.5 20.0 17.5 15.0 12.5 10.0 7.5 Oct-03
Oct-04
Oct-07
Seasonality of volatilities In an article we published in July 2004, we found that examining data from the 1990s and into this decade gold volatility tended to subside over the summer moths only to spring higher in September. Crude oil vol also displayed a similar seasonal pattern. We have updated the analysis and we find that gold has continued to display this seasonal pattern, such that gold implied vol tends to trade lower in July and August and spike higher in September.
Crude oil vol is not alone in trading at extreme levels, gold implied vol is also trading rich with respect to historical averages, Figure 7. While front-end gold vol has traded lower over the past month, back-end gold vol has climbed to be trading at the 95-percentile mark.
Figure 8: The seasonal pattern of gold vol: quiet summer and the September Effect
Average change in gold implie vol (Jan 1998 - May 2008) 5
28% 26% 24%
1M
3M
6M
12M
4 3 Change (Vols) 2 1 0 -1 -2
Aside from gold, in the precious metals complex platinum vol is also trading rich in our view. The 12M implied vol is now trading at 5-year high despite the fact that the realised vol remains lower than the previous high as shown in Figure 8.
To avoid any distortion coming from a big event risk (such as the CGBA in 1999, and the terrorist-related events in the US in 2001), in Figure 10 we focus on the month of July and plot the changes in gold implied vol in the month of July against June over the past ten years. Indeed we found a very consistent pattern including the previous summer characterised by the onset of the subprime debacle.
Page 40
27 June 2008
Commodities Quarterly
Figure 9: The changes in gold implied vol in July (vs. June) since 1998: a very consistent pattern
2
6.00
4.00
2.00
0.00
M06
-3
M12
-2.00
-4.00
-4
-6.00
-5 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
This seasonal effect also exists in other asset classes, most notably equities and foreign exchange. July has tended to see a drop in EURUSD implied vol with last year being the exception, Figure 11.
Figure 10: The changes in EURUSD implied vol in July: 2007 being a big exception
1.0
3.00
2.00
Changes in implied vol (EURUSD, July, vols)
1.00
0.5 M01 0.0
0.00
M02
-1.00
-0.5
M03
-2.00
-3.00 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Conclusion: Volatilities has tended to subside over the summer months across asset classes and commodities volatilities is no exception. Given the current elevated level of implied vol we believe shorting vol via variance swap offers good value.
Regarding our short variance swap trade, we believe a closer scrutiny of the seasonal pattern of realised vol is more important than that of implied vol. Therefore the seasonal pattern of WTI and gold actual realised vol is charted in Figure 12 and 13. In both cases, the realised vol declines in July and August only to spring back hard in September.
Trade recommendations:
Short Dec09 WTI variance swap Short Dec09 Gold variance swap Short Dec09 Platinum variance swap
Risks
Continuous disruption in the supply of crude oil, gold, or platinum, in the form of geopolitical or weather events, production problems and/or accidents
Page 41
27 June 2008
Commodities Quarterly
Figure 1: The distribution of oil prices in the 1990s and during this decade
450 400 350 Number of days 300 250 200 150 100 50 0 10 20 30 40 50 60 70 80 90 100 WTI crude oil price (USD) 110 120 130 Average WTI oil price 2000-2008=USD47.5/barrel Average WTI oil price 1990-1999=USD19.7/barrel 2000-2008 1990-1999
However, the rules of price forecasting do not appear to have adapted to this new world order in the oil market. Instead forecasting crude oil prices has tended to be solely based on where the current spot price is trading. We find that since 1990 the consensus 12 month oil price forecast has been equal to the current spot price times 0.85 plus 2 dollars and 28 cents, Figure 2. Since the oil price is approximately USD138/barrel using this correlation it implies the consensus price forecast for crude oil at the end of June 2009 is USD119.6/barrel.
Figure 2: Consensus oil prices forecasts have tended to be a function of the current spot price
120 100 Oil price 12M forecast 80 60 40 20 0 January 1990 - June 2008 12M consensus oil price forecast = 0.85 x (Current oil price) + 2.28 R square = 0.98
Forecasting crude oil prices during this decade has been fraught with difficulties. In the 1990s, when the crude oil price averaged USD20/barrel analysts typically employed mean reversion to set oil prices forecasts. This technique worked well since over 75% of the time the crude oil price traded within a range of USD16/bbl to USD24/bbl. In this decade, oil price forecasting has become more challenging.
20
40
60
80
100
120
140
Not surprisingly this technique in forecasting oil prices has not been a successful strategy during this decade. Indeed in January 2004 we published research examining the consensus prices for crude oil at the start of each year since 1999 and the eventual outturn. We found that the analyst community had consistently under-estimated the oil prices by an average of 30% between 1999 and 2003. At that time we advised oil market participants to add 30% to the consensus Reuters oil price forecast to reach a more realistic assessment for the oil price outlook. At the start of this year the consensus oil price forecast stood at USD72.9/bbl. Like previous years, we applied the average forecasting error between 1999 and 2007 and it
Page 42
27 June 2008
Commodities Quarterly
implied Brent crude oil prices averaging USD98/bbl in 2008. However, on our estimates, the analyst community will have underestimated the average crude oil price in 2008 by 72%, Figure 3. Moreover according to the last Reuters oil price survey published at the end of May, the consensus price forecast for crude oil in 2009 stands at USD104.3/barrel. If we apply the average forecasting error since 1999, then it would imply the crude oil price averaging USD140.9/bbl next year (USD104.3 x 1.35).
Calendar Brent swap price at the start of the year Outturn Average absolute forecasting error 1999-2008 = 19% 33%
19%
0% 18%
24%
Figure 3: Tracking the crude oil forecasting error among the analyst community
150 Brent forecast at the start of the year Outturn Average forecasting error 1999-2008 = 35% 110 USD/barrel 72% 35% 130
70
30
10 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008f 2009f
Another route to assess the likely oil price for the coming year is to examine the calendar Brent swap price at the start of the year and compare it to the final outturn. We find that like analysts the futures market has also consistently under-estimated the strength in crude oil prices. However, the track record of the futures market is superior to analysts since the forecasting error drops from +35% to +19% between 1999 and 2008. This superior track record was demonstrated at the beginning of this year when we applied the average forecasting error between 1999 and 2007 to the Cal08 Brent swap price. It implied the crude oil prices would average USD110/barrel in 2008. If we apply the average forecasting error of 19% to the current Cal09 Brent swap price of USD136.5/bbl, it yields an expected average oil price for 2009 of USD162.4/bbl, Figure 4.
Options market outlook Another route to assess the oil price outlook is via the options markets. We examine the risk neutral probabilities of various WTI contracts expiring above or below different crude oil price levels. Not surprisingly the surge in oil prices during 2008 has led to a significant increase in market participants seeking protection against a spike in oil prices. Inevitably this has led to the probability the options market attaches to the Dec-08 WTI contract expiring above USD200/bbl to rise to 12% currently while sub 90 dollar oil is considered a 5% probability event. During this month, the options market was also pricing an equal probability of the Dec12 WTI contract expiring below USD105/bbl and above USD245/bbl highlighting the markets ongoing fears towards even higher oil prices.
Figure 5: Probabilities of the Dec08 WTI contract expiring above/below different price levels
WTI crude oil price (USD/bl) 50 60 70 80 90 100 125 150 200 Below 0% 0% 1% 2% 5% 10% 31% 59% 88% Probability Above 100% 100% 99% 98% 95% 90% 69% 41% 12%
Page 43
27 June 2008
Commodities Quarterly
Growth in the industrial sector, which increased by 4.8% in the first quarter of 2008 compared with the corresponding period last year, seems to be tied to export strength and some resurgence in gas-intensive industries, such as fertilizers. Production Data published May 30 by the US DOE/EIA indicates that US lower-48 states production rose 0.5% in March from February 2008, and shows wet gas output up in all areas except the Gulf of Mexico federal offshore. EIA stated that GOM declines were due to platform issues and shut-in wells. Since January 2007, we calculate the average m-o-m increase at 0.6%. Even assuming this trend continues into April, May and June (April data to be released by the EIA on 30-June), we expect that the total production figure to be reported by the EIA for April will be down circa 0.7% from March due to the April 9 unexpected shutdown at Independence Hub that took 0.9bcf/d off the market. With the hub repairs now completed, according to owner Enterprise Products Partners, production should jump substantially in June after a relatively stagnant May.
Consumption Total natural gas consumption is expected to increase by just over 2% in 2008 and by about 1% in 2009. Year-over-year increases in the residential, commercial, and electric power sectors have been largely weatherdriven. In 2009, residential and commercial sector consumption is expected to decline slightly while natural gas consumption for electricity generation is expected to increase by 2.5%.
Figure 1: EIAs gas demand forecasts (consumption in bcf/d and % change y-o-y)
Source: US DOE/EIA
The most recent weekly US gas rig count is running about even against the year-ago period, in stark contrast to June 2007 when rig activity was running up 10% and compared to June 2006 when the y-o-y pace was 15%. The US Gulf of Mexico and onshore Texas areas that are currently providing the gains may not be able to sustain their momentum. Gas volumes out of the US Gulf of Mexico region were in pronounced decline prior to the startup of the Independence Hub platform during the second half of last year, with an underlying decline rate larger than the new volumes from Independence Hub (circa 0.9 bcf/d assuming 90% utilization of nameplate capacity). We estimate that deterioration in other GOM fields will offset output from the Independence Hub by Q4 2008, meaning that the Gulf as a whole could resume declining at a double-digit pace by late this year.
Page 44
27 June 2008
Commodities Quarterly
Liquefied natural gas The outlook for the global LNG market is tight. The US market's size, liquidity, diversity of supply and storage capacity make it the "sink" for LNG cargoes but a shortage of global natural gas has left the sink without a sufficient tap of supply. The US "call on LNG" exceeds available supply. Long term it gets worse. The expected 2007 FID of nearly 90Mta of LNG capacity saw only 10Mta receive project sanction i.e., a nearterm supply growth breather in 2009 will give way to shortages post-2010. While supply struggles, demand strength continues to surprise, led by earthquakes in Japan, droughts in Spain, and the emergence of new demand centers in Kuwait, Singapore, Chile and Argentina the southern hemisphere now competes for US summer imports. Short supply has polarized returns, driving shipping & regas often below cost of capital, yet upstream returns to > 50% IRR.
US natural gas price outlook Uncertainty over natural gas demand by the electric power sector during the summer and the possibility of hurricane-related supply disruptions as well could impact spot prices in the coming months. The inability of electric utilities to get coal plants permits is increasing the potential for power shortages after 2010. Nuclear power has growing potential (but not until well after 2013, and maybe not then, depending on US energy policies. Wind and renewables projects are moving as fast as possible. In our view, this suggests that any shortfall in coal-fired electricity will have to be made up by natural gas.
2000
2005
2010E
2015E
Strong international demand plus a security of supply and environmental premium in Asia and Europe has left international prices set by the marginal price of demand, pushing long-term contracts to straight line oil parity, with implied pricing of USD17/mmBtu at USD100/bbl oil. We expect global tightness will leave the US exposed to gas price spikes most likely weather induced. We believe the key point is that rival gas markets in Asia and Europe are prepared to pay more for LNG supply than US buyers.
Page 45
27 June 2008
Commodities Quarterly
Figure 5: World natural gas supply & demand fundamentals (billion cubic feet/day)
Annual Avg Rate % 2007 CONSUMPTION Residential Yr/Yr % Change Com m ercial Yr/Yr % Change Industrial Yr/Yr % Change Electric Power Yr/Yr % Change Other Yr/Yr % Change Total Demand Yr/Yr % Change DOMESTIC SUPPLY Alaska Gulf of M exico Offshore Other U.S. Total Marketed Production Total Dry Gas Production Yr/Yr % Change Net Storage Withdraw Other & Balance Total Dom estic Supply Exports IMPORTS Pipeline LNG 2008E 2009E 2010E 2011 2012 2013 2014 2015E 00-05 05-10 10-15
12.9 8.1 8.2 6.0 18.2 2.1 18.8 10.4 5.0 4.8 63.2 6.5
13.3 2.9 8.5 2.6 18.5 1.7 19.1 1.7 5.1 2.6 64.5 2.1
13.5 1.3 8.6 1.3 18.6 0.5 19.5 2.0 5.0 (1.5) 65.2 1.1
13.7 1.3 8.7 1.3 18.7 0.5 19.9 2.0 5.0 (1.5) 65.9 1.1
13.8 1.3 8.8 1.3 18.8 0.5 20.5 3.0 4.9 (1.5) 66.8 1.4
14.0 1.3 8.9 1.3 18.8 0.5 21.1 3.0 4.8 (1.5) 67.7 1.4
14.2 1.3 9.0 1.3 18.9 0.5 21.8 3.0 4.7 (1.5) 68.7 1.4
14.4 1.3 9.1 1.3 19.0 0.5 22.4 3.0 4.7 (1.5) 69.7 1.4
14.6 1.3 9.3 1.3 19.1 0.5 23.1 3.0 4.6 (1.5) 70.7 1.4
-0.6
0.7
1.3
-1.1
1.1
1.3
Modest growth as industrial production grows, but pressured by high prices
-4.1
0.6
0.5
2.5
4.4
3.0
Electric power growth use of gas slows due to price increases
-0.9
1.0
-1.5
-1.1
1.8
1.4
Overall demand growth slightly below the 2005-2010 period; residential and industrial use stabilises
1.2 7.6 46.4 55.2 52.8 4.3 0.5 0.4 53.7 2.2 12.6 10.5 2.1
1.2 7.8 47.6 56.6 54.2 2.6 -0.3 -0.4 53.5 2.2 13.2 11.6 1.7
1.2 7.6 48.1 56.9 54.7 0.9 0.0 0.5 55.2 2.2 12.2 10.1 2.1
0.9 6.8 51.1 58.7 -1.2 2.1 0.3 -8.6 0.7 -2.7 3.0 -2.0 1.0
Gulf of Mexico OCS slows; onshore production shows modest response to higher prices
54.9 55.2 54.9 55.2 55.5 55.8 0.5 0.5 -0.5 0.5 0.5 0.6 0.0 0.5 55.4 2.2 12.6 10.1 2.5 0.0 0.5 55.7 2.1 13.2 9.9 3.4 0.0 0.5 55.4 2.1 14.5 9.6 4.8 0.0 0.5 55.7 2.1 15.1 9.4 5.7 0.0 0.5 56.0 2.1 15.8 9.2 6.6 0.0 0.5 56.3 2.0 16.4 8.9 7.5
Exports to Mexico border areas expected to slow as Baja and Gulf Coast LNG penetrates
Conventional imports from Canada decline and the gap between US supply and higher demand must be filled by LNG
Memo Items: Im ports % of domestic use End Prd Wrkg Storage (tcf) Baker Hughes Rig Count Oil Gas % Gas Nymex Price ($/mmBtu) Yr/Yr % Change
23 3.89 2039 410 61.9 8.5 -0.3 10.5 6.1 7.0 15.0 5.5 0.0 1.0 1.0 1.0
Gas prices expected to normalise at $12/mmBtu in 2010, and then begin to rise at circa 50 USc/mmBtu per year
7.12 11.31 12.25 12.00 12.50 13.00 13.50 14.00 14.50 2.0 58.8 8.4 -2.0 4.2 4.0 3.8 3.7 3.6
15.9
5.9
3.9
Source: OPEC Secretariat, International Energy Agency, US DOE/EIA, DB Global Markets Research
Page 46
27 June 2008
Commodities Quarterly
signal to effect CCS as the new entrant of choice by 2020 will always be reflected in the EUA price today. However, by proposing far in advance both a specific level for the ETS cap by 2020 and that the cap be reduced thereafter at a fixed rate of 1.74% (with a review of this reduction rate in 2025), the Commission has surrendered discretionary power to modify the 2020 cap in response to changes in the key variables that drive EUA prices (in particular, commodity prices, weather conditions, and rates of economic growth). In practice, this means that whether or not CCS becomes the new entrant of choice by 2020 will depend to a large extent on the evolution of gas prices in the next few years. Based on our revised forecast for long-term oil, gas, and coal prices, our revised base-case scenario for the residual abatement required via fuel switching over 2008-20 leads us to conclude that the Phase-3 cap proposed by the Commission under the existing EU commitment to reduce GHG emission by 20% by 2020 against 1990 levels will indeed be tight enough to make CCS the new entrant of choice by 2020. Our revised forecasts imply a 2008 EUA price of EUR40/t, a price consistent, in our view, with CCS convergence in 2020. We now see 2008 EUAs at EUR40/t
The EUA forward curve should ordinarily be in contango With EUAs bankable into future trading periods, the EUA forward curve should ordinarily be in contango, slanting upwards at the cost of carry to the most valuable point in the future, i.e. the point where the EUA price hits the level required to fulfil the ETS policy objective. The EUs policy objective is to restrict global warming to no more than 2C against pre-industrial levels, and Stern and other authorities are agreed that this can only happen with an accelerated deployment of CCS. This means that if the ETS is to serve its ultimate purpose, the EUA price will have to reach and then stabilize at the level required to make CCS viable on a large-scale commercial basis in the next 10-15 years. Indeed, the Commission has specified that it wants to see CCS become the new entrant of choice by 2020. With Phase-2 EUAs bankable into Phase 3 and beyond, the Commission is in theory able to drive the price to the level required to make CCS the new entrant of choice by 2020 by ensuring that the cap is at all times perceived by the market to be as tight as necessary to achieve this end. This is because if the cap is at all times perceived to be tight enough, the necessary price
In an efficient market the fact that EUAs are bankable puts a floor on their price but not a ceiling at the level of the long-term supply curve. Our revised commodity-price estimates assume a longterm gas price of EUR8.9/GJ in real terms today (based on a long-term oil price of USD85/boe in real terms today), and a long-term all-in coal price of EUR2.32/GJ (based on USD90/t in real terms today). On these assumptions, we calculate that the EUA price required today to make CCGT plant the new entrant of choice over conventional coal is EUR40/tonne, and that this implies a long-term power price of EUR82/MWh in real terms today. At the same time, our model indicates that assuming equal annual use of all of the carbon credits available to the ETS over 2008-20 under the Commissions proposals (i.e. EUAs plus CERs/ERUs), there is currently not enough spare gas-fired generation capacity in the EU for fuel switching to achieve the average annual residual abatement requirement over 2008-20 of 125Mt at prices up to EUR40/tonne. As a result, with CCS probably not technically viable on a large-scale commercial basis before 2020 and nuclear disadvantaged by its long lead-time and its political
Page 47
complications in many Member States, our model projects a need for 60GW of new CCGT capacity by 2020 simply in order to provide the switching capacity necessary to enable the Phase-2 and Phase-3 caps to be met at lowest cost. This being the case, we think the most rational compliance strategy over 2008-20 is for ETS installations to use a disproportionately large share of their total carbon credits (EUAs plus CERs/ERUs) over 2008-12, and thereby give themselves time to build the new CCGT capacity necessary to enable greater fuel switching over 2013-20. If we averaged the total amount of all carbon credits available to ETS installations over 2008-20 (EUAs plus CERs/EUAs), then the equal annualized amount would be 2,045Mt. However, we think installations will have to use 2,163Mt per year over Phase 2 in order to restrict the cost of compliance to EUR40/t, as on our estimates there is only enough fuel-switching capacity available at the moment to reduce emissions by 65Mt per year at prices up to EUR40/tonne.
By definition, then, the optimal cost curve for compliance over 2008-20 will only be constructed if the 2008 EUA price rises to EUR40/t in the relatively near future to enable the required CCGT capacity to be built. On our estimates, a price of EUR40/t today is consistent with making CCS the new entrant of choice by 2020. In other words, the EUA price required today to make CCGT the new entrant of choice over conventional coal is consistent with a long-term supply curve that would see CCS become the new entrant of choice in 2020. The more efficient the market, the more quickly the EUA price will move to the level of the long-term supply curve, and the more likely it will be that the carbon price will not have to deviate materially from this trajectory thereafter. By contrast, the less efficient the market, the longer it will take for the EUA price to reach the level implied by the long-term supply curve, and the greater the likelihood of the price spiking above and perhaps significantly above the long-term supply curve thereafter. Our new base-case scenario for the EUA price and the optimal compliance strategy for the ETS over 2008-20 is set out in Figure 1 below.
Figure 1: DB base-case scenario for the EUA price: 40/t in 2008 rising to EUR67/t by 2020E
Rise in temperatures < 2C
EU Cap (MtCO2)
2,083 1,720
2008
2012
banking
2020
banking
Phase 2
Phase 3
Borrowing not allowed Borrowing not allowed
CO2 price(/t)
67
67/t
Risk of backwardation
40
40/t
* * * * * ** * * ** *
Mt p.a.)
* *
* * * *
*
CCS plant
In power-plants
** * *
Page 48
27 June 2008
Commodities Quarterly
but EUA prices could spike very sharply during Phase 2 Because borrowing from future periods is not allowed under ETS rules, backwardation in the EUA forward curve may occur at times. Indeed, we think there is a non-trivial risk of a significant spike in the EUA price towards the end of Phase 2, and that prices of EUR100/t or more are possible under certain scenarios. In particular, we think that if there were insufficient CERs/ERUs available over 2008-12 this would lead to a spike in the Phase-2 EUA price, as in the absence of sufficient switching capacity at prices of up to EUR40/t the marginal price would be set on the short-term supply curve rather than the long-term supply curve. Given the implied cost of fuel switching on our commodity-price estimates for 2008-12, we think that under extreme circumstances this could lead to Phase-2 prices of EUR100/t or higher. Valuation and risks In our view, the European Commissions proposed cap for Phase 3 of the ETS points to an EUA price of EUR40/t in 2008, and to a price of EUR67/t by 2020 assuming a cost of carry of 4.5%. Other things being equal, higher gas prices than we are currently assuming in our model would imply a higher 2008 EUA price than EUR40/t (and hence an earlier date for CCS convergence). As indicated in Figure 2, a gas price based on a longterm oil price of $100/boe (in real terms in 2008) would imply a supply curve starting at EUR65/t in 2008, and rising to EUR110/t by 2020. In turn, this would imply a long-term power price of EUR100/MWh. Correspondingly, this would then imply an even higher level of carbon-credit usage in Phase 2, and hence a greater amount of new CCGT capacity to be built over 2013-20 for abatement purposes. By contrast, lower gas prices than we are currently assuming in our model would imply a lower EUA price than EUR40/t (and hence a later date for CCS convergence). As indicated in Figure 3, a gas price based on a longterm oil price of $70/boe (in real terms in 2008) would imply a supply curve starting at EUR20/t in 2008, and rising to EUR34t by 2020. In turn, this would imply a long-term power price of EUR70MWh. Correspondingly, this would then imply a much lower level of carbon-credit usage in Phase 2, and hence a lower amount of new CCGT capacity to be built over 2013-20 for abatement purposes.
More generally, if enthusiasm for the Kyoto methodology of setting limits on emissions were to wane owing to the high price of carbon credits, then this could lead to a re-negotiation and/or abrogation of the Kyoto Protocol, although we think that EU policymakers are committed to the ETS, and that there is therefore less of a risk that this scheme would collapse even under a high-price scenario.
Figure 2: DB sensitivity: Long-term oil price of USD100/boe and long-term coal price of USD90/t in real terms 2008 implies a carbon price of EUR65/t in 2008
Rise in temperatures < 2C
EU Cap (MtCO2)
2008
2012
2,083 1,720
banking
2020
banking
Phase 2
Phase 3
Borrowing not allowed Borrowing not allowed
CO2 price(/t)
110/t
65/t
Figure 3: DB sensitivity: Long-term oil price of USD70/boe and long-term coal price of USD90/t in real terms 2008 implies a carbon price of EUR20/t in 2008
Rise in temperatures < 2C
EU Cap (MtCO2)
2008
2012
2,083 1,720
banking
2020
banking
Phase 2
Phase 3
Borrowing not allowed Borrowing not allowed
CO2 price(/t)
34/t
20/t
Page 49
Figure 4: EUA Dec08 and CER Dec08 price since June 2007 (EUR/t)
30 EU Em ission Certificate (Dec' 08) CER sw ap (Dec' 08) 25
20
15
Source: Datastream
Implications for power prices Figure 4 shows the trend in EU one-year forward baseload power prices over the last 12 months. Although UK power prices of 86/MWh, and German prices of EUR82/MWh both represent record levels, our carbon-price forecasts point to higher power prices still in both markets. With gas the marginal fuel in the UK at the moment, then other things being equal an increase in EUA prices to EUR40/t from the current levels of EUR27/t would imply an increase in UK power prices of approx. 4/MWh. With coal the marginal fuel in Germany at the moment, then other things being equal an increase in EUA prices to EUR40/t from the current levels of EUR27/t would imply an increase in German power prices of approx. EUR11/MWh.
Source: Datastream
Mark Lewis, (33) 1 4495 6761 mark-c.lewis@db.com Isabelle Curien, (33) 1 4495 6616 isabelle.curien@db.com
Page 50 Deutsche Bank AG/London
27 June 2008
Commodities Quarterly
Turning points in the US dollar have occurred when the greenback has hit extreme levels of over/undervaluation against the euro, or synthetic euro pre-1999. In this cycle, the US dollar peaked against the euro in October 2000, but, it was not until January 2002 that the US dollar embarked on a new long run downtrend. At the last turning point in a dollar cycle, we estimate the US dollar was approximately 25% overvalued against the euro, Figure 2.
1.40
1.20
1.20
1.00
1.00
-25%
0.80
We believe the outlook for the gold price remains closely tied to the course of the US dollar. Since 1972, we find that the US dollar has displayed long-run cycles of rising and falling for extended periods of time. These cycles have an average duration of seven years with bear runs historically longer (8.5 years) than bull runs (5 years), Figure 1.
It should come as no surprise that the dollar started to overshoot this year because 2008 is not only the seventh year in the current US dollar cycle, but, also because the US dollar had lost significant interest rate support following aggressive Fed easing. We estimate the US dollar is currently 35% undervalued against the euro. Since this cycle is long in the tooth and the US dollar is close to its largest misalignment against the euro (DEM) in recorded history investors are beginning to contemplate the possibility that the US dollar may start to embark on a new long term uptrend. This could have serious implications for precious metal prices given the strong positive correlation of gold and silver prices to EURUSD during this decade, Figure 3.
3.00
0.25 0.00 -0.25 -0.50 -0.75 In 2005 EURUSD falls from 1.35 to under 1.17. However, gold prices rise USD100 to USD520/oz.
2.00
1.50
1.00 1972 1976 1980 1984 1988 1992 1996 2000 2004 2008
-1.00 Dec-99
Apr-01
Aug-02
Dec-03
Apr-05
Aug-06
Dec-07
Source: DB Global Markets Research, Bloomberg * 30-day log daily changes correlation Source: DB Global Markets Research, Bloomberg
Page 51
Indeed we find that when a new long term uptrend in the US dollar eventually commences, the move in the greenback has tended to be ferocious. We examined US dollar cycles since 1980 and find that in the first year of a dollar turn the greenback moves by an average of 27% in the first year and a further 14% in the same direction in the second year, Figure 4.
the current US dollar cycle will continue to track the decline in the dollar that followed the signing of the Plaza Accord in September 1985, then it implies the US dollar only hitting rock bottom in September 2011, Figure 6.
Figure 4: The scale of the dollars reversal in the two years after the dollar cycle has turned
USDDEM 1980 USD bull cycle 1985 USD bear cycle 1995 USD bull cycle 2002 USD bear cycle Average* Year 1 40.8% -35.0% 11.3% -19.9% 26.7% Year 2 0.37% -26.3% 13.5% -13.9% 13.5%
* Absolute average change in the US dollar in Year 1 & Year 2 Source: DB Global Markets Research
Such a large correction in the US dollar would, in our view, be sufficiently severe to end what has been the longest gold price rally in history, Figure 5.
2)
Bull cycles have typically been preceded by a significant narrowing in the US trade deficit. Although the US current account deficit has fallen from 5.8% in 2006 to an estimated 3.7% of GDP in 2008, the deficit still remains large. We expect the rise in oil prices may also limit further improvement in the deficit and constrain a turnaround in the US dollars fortunes for the time being. The start of a new uptrend in the US dollar has tended to coincide with a turn in interest rate differentials in favour of the US dollar. While the dollar may have benefited from Fed tightening being priced into interest rate markets, we are doubtful that the US economy is robust enough to support such tightening. Rather we expect the environment of negative real interest rates is expected to continue, which we believe will sustain a supportive environment for precious metals complex.
3)
41 8 34 84
Although the gold price had been rising before August 1971, we take this as the start point for this rally since it marks the date the US government informed the IMF that the US dollar would no longer be convertible into gold. This consequently led to the collapse of one of the main pillars of the 1944 Bretton Woods system. Highs and lows in the gold price relate to closing prices Source: DB Global Markets Research, Bloomberg
We believe there are several factors to suggest that a turn in the US dollar may still be a distant prospect: 1) Bear cycle tends to be more durable than bull cycles and that uptrends from down cycles have tended to be preceded by an extended bottoming-out period. If this cycle conforms to historical average then it would imply the US dollar will not embark on a new long term uptrend until August 2010, or 8.5 years after the dollar turned. Alternatively, if one assumes
We examined year on year returns for gold and silver since 1971 in various real interest rate environments. We find that gold and silver tend to perform strongly in low or negative real interest rate environments. It has only tended to be when real interest rates in the US have moved above 3.5% that the dangers for precious metal returns has become hazardous. Given the fragility of the US banking and housing sectors, we believe US real interest rates will remain low for the foreseeable future and consequently limit the degree of US dollar strength that is likely to unfold over the coming year.
Page 52
27 June 2008
Commodities Quarterly
Figure 7: Gold & silver returns in different real interest arte environments
100 80 60 yoy returns (%) 40 20 0 -20 -40 -60 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 Real short-term Fed funds (%) Gold and silver returns perform strongly in low or negative real interest rate environments Year-on-year returns 1971-2008 Gold Silver
5)
Every turn in the US dollar since 1978 has coincided with aggressive rounds of G3 foreign exchange intervention, Figure 9. For the time being dollar weakness is not a policy headache for public sector authorities. It assists in dampening inflationary pressures in Europe, in rising commodity price environment, while it helps cushion the slowdown in the US economy. As a result, we believe investors should only position for dollar strength or at the very least stability in the US dollar exchange rate when central banks start intervening and buying the US dollar.
4)
A trend reversal in the US dollar has historically been associated with a longer-term turn in capital flow trends. However, US credit markets have been the primary channel of capital inflows into the US in recent years consequently ongoing stress in the US credit and housing sectors are expected to hinder a turnaround in US capital flows in the near term.
1.50
G3 action to stop the dollar falling Louvre Accord (1987) to stabilise the dollar
1.70
1.50
1.30
1.30
1.10
1.10
0.90
0.90
0.70
We monitor a modified version of the USs basic balance. This tracks net foreign purchases of US equities, corporate and agency bonds less the monthly trade deficit. While the US dollar did not benefit from an improvement in the basic balance during 2006, we find that since the middle of last year there has been a significant deterioration in the USs basic balance, which has coincided with extreme US dollar weakness, Figure 8.
0.70
0.50
0.50
Before 1999, FX intervention rounds are conducted in USDDEM Source: DB Global Markets Research, Bundesbank, Bloomberg
To assess the markets outlook for EURUSD, we examined the cumulative Black-Scholes knockout probabilities for various levels of EURUSD over various time horizons. We find that the implied probability of EURUSD trading the 1.35 level stands at 19.6% in one year and 33% in two years. Conversely the probability of the dollar remaining weak at 1.50 is considered a 77% and 80% probability event over the next two years. Moreover, it shows an equal probability of EURUSD trading at 1.45 and 1.65 over the next three years. Although there is no predictive power in these probabilities it provides an indication that the market does not envisage a significant uptrend in the US dollar commencing in the near term.
0.90
* The modified basic balance is defined as net foreign purchases of US stocks and bonds (excluding US Treasuries) less the US trade deficit Source: DB Global Markets Research, US Treasury
Page 53
Source: DB Global Markets Research; Data as of June 25 2008 EURUSD spot 1.555
Investment demand for precious metals has been another factor aside from the US dollar that has helped to drive gold and silver prices during this decade. Indeed in certain circumstances the launch of several physically backed Exchange Traded Funds has meant investors have had a direct impact on the physical market. The next section outlines the growing importance of ETFs, which have become particularly relevant for the precious metals sector.
Page 54
27 June 2008
Commodities Quarterly
of assets under management (AUM) in the US and Europe followed by index based commodity ETFs.
Across
The
ETF investments have become increasingly popular among private and institutional investors during this decade. In our view, this reflects their accessibility, transparency and low cost structure. A surge in investor interest in commodities has led a sharp rise in commodity listed ETFs as well as a substantial growth in asset under management. We estimate total asset under management in commodity ETFs has risen from USD4.2 billion in 2005 to approximately USD53.7 billion so far this year. We believe investor focus in commodity ETFs is gradually shifting from the precious metals sector towards the energy and agriculture sectors. Despite being one of the strong performers so far this year, investor interest in the industrial metals sector remains subdued perhaps a reflection of the fewer products offered in this sector.
We believe the dominance of precious metals in commodity ETFs is mainly due to two main reasons. Firstly precious metals such as gold and silver ETFs were amongst the first wave of commodity related ETFs to be launched. Secondly physically backed precious metals ETFs avert the carry costs associated with commodities that have upward sloping forward curves.
Commodity ETFs have attracted a substantial amount of investor interest since the launch of the first gold based ETF in 2004. This is highlighted by the rapid growth in asset under management (AUM), which according to our estimates has grown from USD4.2 billion in 2005 to approximately USD53.7 billion so far this year, Figure 1.
One common feature of the gold and silver market is that the forward curves are almost always in contango. We find that the degree of contango is considerably higher than the storage, insurance and management fees charged by ETFs. The relative out performance of GLD, a physically back gold ETF over S&P GSCI gold total returns index is illustrated in Figure 3.
Figure 3: Relative performance of GLD vs S&P GSCI gold total return index
8 Relative performance of the GLD vs. S&PGSCI gold total returns index (%)
There are some structural differences between European and US exchange traded products. While exchange traded certificates account for the largest slice of the European market, the US market is dominated by ETFs. However, we find that the mix of assets in the US and Europe is broadly similar, Figure 2. The precious metals sector accounts for the lions share
2005
2006
2007
2008
Source: Bloomberg
Page 55
However, recent trends suggest investors are gradually shifting their focus to other sectors such as energy and agriculture. We find that AUM in agriculture has risen from USD0.5 billion to USD3.42billion since August last year while AUM in energy has increased from USD0.7billion to USD3.9 billion over the same period. One could argue that price appreciation, not investor flows is the main culprit behind the most of the growth in AUM. We disagree since total returns on the DBLCIOY energy and DBLCI-OY agriculture have risen approximately 90% and 60% respectively over the same period. Despite being one of the strong performers so far this year, investor interest in industrial metals sector has remained subdued with AUM gaining a relatively small 60% since the beginning of 2008. In terms of commodity ETFs in the US, turnover in US energy ETFs has recently overtaken precious metals as the most actively traded sector among the ETF universe. By the end of May, the energy sector accounted for over 50% of all commodity ETF trades. This compared to turnover in precious metal ETFs declining from almost 90% of all turnover at the start of this year to currently just under 37%, Figure 4.
2006
2007
2008
This note is based on the Portfolio & Index Strategy report Exchange Traded Funds: Europe & US. Commodity ETF Liquidity Trends published by Nizam Hamid and Yvonne Sandford, Deutsche Bank Index Research
Page 56
27 June 2008
Commodities Quarterly
#17 PGMs
Rhodium Market Deficit Set To Tighten
We have made minor changes to our platinum and palladium price forecasts for 2008 (both up by around 2.9%), although we have raised our rhodium forecast by 5.5%; all of these changes represent a mark to market of 2Q08 pricing. For 2009 and 2010 we have left our pricing profiles unchanged. We are forecasting 2008 average PGM prices of USD2,000 per ounce for platinum, USD466 per ounce for palladium and USD9,323 per ounce for rhodium. The key risks to our forecasts include a continued depreciation of the rand against the US dollar, which would likely result in our forecasts being too high. When considering upside risks, we cannot ignore the possibility of further short-term supply issues and renewed investment interest.
conservative. With these pressures increasing further, in conjunction with power and consumables also experiencing significant levels of inflation, we have been forced to reassess our long term prices again. Ultimately, the majority of the 13.5% increase in our long term basket price (to $1027 per 3PGM oz) has been a result of higher than inflation wage increases (which can be assumed to be structural in a South African labour context) and a doubling in real terms of unit power costs over the next five years (power represents around 8% of the average SA PGM producers working costs). Even with these fairly sizeable upgrades to our long term prices, we believe that the risk remains to the upside for long term prices. Platinum We believe the platinum market will continue to be in deficit for the foreseeable future. We are forecasting a 711 000 ounce deficit for 2008 (Johnson Matthey recently confirmed a deficit of 480koz for 2007), and a 458 000 deficit in 2009, following massive production downgrades from AngloPlat and Lonmin (and smaller ones from Northam and Aquarius) in the last year. While our forecast deficit declines in 2010, it remains substantial at around 293 000 ounces. Although higher platinum prices may well accelerate palladium substitution in diesel autocats (largely complete in gasoline autocats) from around 10% of the PGMs in current autocats to around 25% in future autocats, we believe the risks to our platinum forecasts are balanced, in the absence of a substantial shift in the USD. With little above ground stocks (that we know of at any rate), we would expect the deficit to be filled, in time, with lower jewellery sales and a return of jewellery accumulated by Japanese buyers (around 20moz) over the last two decades. In the current set of revisions we have made a mark-tomarket adjustment for 2Q08. We have also made substantial changes to our long term assumptions on the back of higher structural costs in the industry (due to skills shortages, high wage increases and significant real increases unit power costs). In 2008, we have lifted our forecast marginally to USD1 999 per ounce (previously USD1 942 per ounce) but left our 2009 and 2010 forecasts unchanged at USD2 150 per ounce and USD1 925 per ounce respectively. The long-term forecast has been raised to USD1 110 per ounce in 2006 real terms (USD1 136 per ounce in 2007 real terms). We expect continued penetration of diesel vehicles into Europe, but more importantly, it appears that US authorities have been showing increasing support for diesel-powered vehicles of late. While the market typically expects a penetration rate of around 3% in the US over the next few years, AngloPlat has cautioned that it believes this figure could reach as high as 5 8% within the next five years. This could result in additional demand of around 300 000-500 000 ounces pa. Continued high oil prices also have the potential to
Price forecasts We believe the case for robust PGM (platinum group metals) demand and prices remains firmly intact. Although there has been elevated speculative involvement in the PGM markets in the last few years, we think it is with just cause. The fundamentals of these markets are strong; thus we believe markets are likely to continue attracting speculative involvement. While the speculative activity cannot be ignored, we believe the basis for the strength in the PGM prices continues to be founded on accelerating consumer demand (most significantly in autocatalysis) as well as long-only fund purchases and more recently, we have seen an element of ETF purchases as well. We establish our long-term PGM prices on the basis of the marginal cost of future production. Southern Africa will ultimately account for more than 80% of world production, as well as world reserves for platinum, rhodium and palladium. Southern Africa currently accounts for c.50% of newly-mined metal (that is set to grow to c.55% by 2009) and c.80% of global resources. Due to the dominance of Southern Africa in the current and long-term production of platinum, palladium and rhodium, any change in the long-term real rand/USD exchange rate has significant implications for the global marginal cost of production (in USD terms). As the changes to our rand forecasts in this quarter are limited to the short-term, their implications for the long-term marginal cost of production are negligible. In 3Q07, we made some adjustments to our marginal cost of production based on the industrys apparent inability to contain costs (skills and materials shortages, etc.) but unfortunately, these changes have proven too
Page 57
accelerate research into fuel cell technology, although our assumptions include minimal demand from this sector for years to come.
-600
Palladium We believe that in time, the palladium market fundamentals will improve, with symptomatic evidence of reverse substitution in autocats (largely complete in gasoline autocats but increasing in diesel autocats from the current 10% to around 25% of the PGMs in the catalyst over the next few years) as well as improved jewellery off-take (although the last two years have seen declines in off-take, we believe consumers have absorbed a fair amount of the high levels of manufacturing purchases seen in 2005, which bodes well for the future). Unfortunately, we expect the overhang of above-ground stocks (around 8m ounces in Zurich and anywhere between 2m ounces and 10m ounces of Russian stock) to cap the palladium price for the foreseeable future. Aside from the above-ground stocks, we are also concerned that autocat scrap will increase considerably over the next few years (vehicle manufacturers switched aggressively from platinum into palladium in the mid- to late-1990s and with the average vehicles lifespan being around 12 years, we could see autocat recycling increase to as much as 1.5m ounces pa over the next few years). Our final concern is that the Russians (both Norilsk and the government) may attempt to manage the palladium market if prices move aggressively higher to provide peace of mind to vehicle manufacturers (the price spikes at the turn of the century that resulted in substantial write-downs of stock will not be repeated). Our short and medium forecasts have been left largely unchanged but we have raised our long term forecasts on the back of higher structural costs in the industry (due to skills shortages, high wage increases and significant real increases unit power costs). Our long-term forecast has been raised to USD392 per ounce in 2006 real terms (USD402 per ounce in 2007 real terms)
Rhodium The rhodium market remains difficult to forecast; we expect it to remain extremely volatile given the relative illiquidity of the market. Rhodium is heavily exposed to autocats (85% of demand) and we expect demand from this sector to remain strong given tightening regulatory environment for NOxs (the various forms of nitrous oxides). Rhodium is particularly important in reducing NOx emissions that occur predominantly in diesel vehicles and in a lean-burn environment (as manufacturers continue to strive for efficiency in gasoline engines, we would expect an increase in the population of lean-burn engines). Although small in the context of the overall market, consumption by the glass manufacturing industries (particularly LCD glass) should remain supportive from the current base. Ultimately, however, we expect the long-term price to settle at levels substantially below spot as mining operations in South Africa continue to migrate to the UG2 and Eastern Limb reef horizons (which carry significantly higher rhodium content than the traditional Merensky reef horizon). Although one cannot with certainty predict the timing of an inevitable price decline, we believe it would be imprudent to be anything but conservative with our medium and long term rhodium price forecasts. In the short to mediumterm, we have recognised that the impact of continued supply shocks (production issues at AngloPlat and Lonmin in particular) in a tight, illiquid market will likely result in high prices (relative to our long term forecast) being sustained for at least the next few years but if production problems were resolved, we could see a sharp price decline over a short period. We have raised our 2008 forecast by 5.5% to $9 293 per ounce (essentially a mark to market) and raised our long-term forecast by 13.5% to USD2 660 per ounce in 2006 real terms (USD2 730 per ounce in 2007 real terms), to reflect higher structural costs in the industry (due to skills shortages, high wage increases and significant real increases unit power costs).
Page 58
27 June 2008
Commodities Quarterly
short to medium-term. Other lesser-known PGMs such as ruthenium, which dramatically increases storage capacity in hard drives, should in our view also remain at levels that are significantly higher than their long-term averages, despite having moderated significantly from their recent highs. We believe the biggest industry risk is significant price spikes (perhaps worsened by the growth in ETFs) which permanently damage certain PGM demand categories. We expect this would create more secular problems for the business as a whole. While it seems that the days of continued delays and project cancellations (historically a result of a sharp strengthening in the rand) have passed, we have experienced persistent smelting accidents in the industry over the last few years that, if repeated, could keep PGM prices high (or higher) for longer. We believe two of the three major producers have also found it increasingly difficult to reach their design capacity or even target production levels. More recently, the South African government and the South African PGM industrys tougher stance against mine accidents has further contributed to lower than anticipated production levels, a situation that has been exacerbated by Eskoms (SA power utility) power rationing, in our view. Gary Pearson (27) 11 775-7247 gary.pearson-sa@db.com
Conclusion In our view, the case for PGMs is easily explained. The metals have highly unique and desirable physical characteristics, making them useable in a number of industrial applications. PGMs have the unique ability to catalyse reactions, many of which have desirable environmental consequences. Many of these applications happen to be in areas that face increasingly strenuous environmental legislation, requiring the use of more catalytic material. Also, in the case of autocats (that account for more than 50% of the consumption of all PGMs), the legislation has tightened to such an extent that it has become a strategic and critical path item for automotive manufacturers. On the supply-side, reserves are highly concentrated and barriers to entry very high. This enables producers to essentially design their own returns by controlling the rate of new supply. That said, PGM markets remain at a crucial junction. In our view, historical delays in the South African and Zimbabwean expansions will create a continued supply squeeze in platinum and rhodium. This is could result in prices (in particular for platinum and rhodium) remaining high (or even spike significantly) over the next few years despite residual substitution back into palladium for gasoline-based autocatalysis, increased substitution into palladium in diesel-based autocatalysis and despite weaker long-term fundamentals for rhodium. The most likely buffer (for platinum) appears to be the platinum jewellery market but even here, it seems the market has become accustomed to buying at prices of around USD2 000 per ounce. We expect this situation should improve market conditions for palladium, which should be the biggest beneficiary of continued substitution. We therefore expect to see prices remain firm until autocat recycling ultimately caps the upside (as long as above-ground stocks are managed prudently). We believe rhodium continues to enjoy the effects of regular supply shocks (smelter rebuilds, production cutbacks, declining grades, etc.) in an already tight, illiquid market and, in our view, is likely to remain at elevated levels in the
Page 59
Introduction The time for questioning whether or not this commodities bull-market is a typical price cycle is over. Unlike previous cycles that typically featured a supply shock, the world is undergoing a demand shock as a result of the unexpected potency of the structural shift in global consumption trends resulting from the social and economic transformation in the emerging markets. Today most base metals markets are left with a legacy in which producers are struggling to keep up with unrelenting demand.
5.0%
-15.0% -16.2% -35.0% Tin Alum inium Copper Nickel Zinc -15.4% -28.9% Lead
We think regional power availability and security issues will dominate supply-side dynamics and heighten the risk of supply disruptions in the coming quarters. With stock-to-consumption ratios in all metals still at historic lows, supply shocks will continue to have powerful price implications. Similar to the past two months of May, the LME metals complex suffered vigorous selloffs. However, unlike 2006 and 2007, the downtrend in
Page 60
27 June 2008
Commodities Quarterly
2008 was mitigated by a number of interruptions to output usually related to power supply. In the past, miners could afford to be picky about where they chose to develop projects, but are now increasingly being forced to look to more politically sensitive and remote areas for new production. Furthermore, mining companies may find it harder to invest in new projects because of accelerating inflation and higher borrowing costs. At the consumption end of this equation, China and the rest of the emerging market economies are still driving the demand engine. Urbanisation and the development of the middle class demanding an increasing amount of consumer items require a level of industrial raw materials that the world has yet to properly forecast. At the same time, the OECD economies are entering a period in which the economic outlook is weak, inflation is rising and business and consumer confidence is low not a great combination for industrial metals. Nevertheless, recent industrial data out of the Eurozone has proved encouraging. Both industrial production growth rate and industrial orders for came in higher than expected, demonstrating that while economic activity may be slower than recent years, they are certainly not at critical levels.
delay in natural gas supply to many commodities producers in Western Australia. We list a number of other events occurring just in the past six weeks that serve to illustrate why we think energy supply is the most important factor in the aluminium market which in our view will lead to higher prices through 2009: China: The 4.7% increase in power prices is, in our view, the beginning of a trend of the removal of many global energy-sector subsidies. United Kingdom: Anglesey Aluminium was forced to reduce its output by two-thirds at its 148Ktpy smelter for at least three months following a fire that knocked out power. USA: Alcoa shut down annual capacity of around 120Ktpy at its Rockdale smelter in Texas because they had been recently forced to buy power on the open market where energy costs have escalated to an unsustainable level. New Zealand: The 350Ktpy Tiwai Point aluminium smelter cut production by 10% in response to soaring spot electricity prices because of low water levels in the countrys hydro system, with influential lobbyists calling for more cuts in order to conserve the countrys energy. Brazil: Monthly aluminum production (4.4% of global production) in April and May was lower than the previous years after smelters voluntarily lowered output levels in response to higher spot power prices. South Africa: National power provider Eskom won regulatory approval for a 25% rise in power rates for the current financial year and a strong indication the regulator will support tariff increase of 20-25% for each of the next three years. India: Aluminium producer Nalco was one day away from shutting down a 345Ktpy facility after its coal stocks nearly ran out following a trucker strike. Abu Dhabi: Plans for a 375Ktpy smelter were shelved after the government decided to save its energy supplies for higher priority uses. Saudi Arabia: The start-up of a 750Ktpy smelter was delayed until 2012 with costs rising 50% more than previously estimated because of a delay in the power plant construction.
Jun-98
Jun-00
Jun-02
Jun-04
Jun-06
Jun-08
Source: Reuters
Aluminium
Other than tin, aluminium has been the most resilient metal this year, rising over 30% since January. With growth in OCED uncertain, supply side events have dominated the aluminium market in 2008. Starting with the snowstorm in China in February, there have been a number of events that have placed extraordinary pressure on the global market balance and we are expecting supply deficits through 2010. Our view is underpinned by the recent decision in China to increase power prices for industrial users.
Supply
Security of reliable and affordable power sources is the most significant issue affecting the aluminium market, observed most recently with the minium three month
The other major factor dominating market sentiment is recent indications pointing to much higher than expected Chinese smelting capacity. We have based our bullish call on aluminium prices against the premise Chinese production growth would decrease as a result of the governments concerted effort to slow the sector and conserve energy. However, evidence has recently surfaced suggesting capacity growth may not be
Page 61
slowing as rapidly as thought. In the first quarter of the year, we had forecast smelting capacity increases of 11% and 7.5% in 2009 and 2010 to 16.5Mt and 17.7Mt, respectively. We have now revised the 2009 and 2010 forecasts to 18.0Mt and 20.0Mt. Many forecasts in the market suggest much higher capacity growth. Indeed if capacity and resulting production growth eventuate much higher than our current forecasts, the outlook for aluminium is much bleaker that we currently are predicting. So far in 2008, data has shown production is in fact decreasing. Between January and May, Chinas ytd production growth rate was 15.7% compared to 32.6% in the first five months of 2007. That statistic obviously includes a slower rate in February because of the drastic reduction in output following the snowstorms, but the five-month total remains remarkably lower than the previous year.
Kt
Figure 4: Chinese aluminium production is still growing, but at a much lower rate
14.00 Chinese prim ary alum inium production, Kt (rhs) Rolling 12-m onth production grow th rate, Kt (lhs) 1200 1100 1000 900 10.00 800 700 8.00 600 500 6.00 M ay-04 Nov-04 M ay-05 Nov-05 M ay-06 Nov-06 M ay-07 Nov-07 400 M ay-08
12.00
The other swing factor in the market is the situation surrounding Chinese aluminium trade. Since Q2 2007, we have forecast China would become a net importer of primary aluminium which is now the in fact the case. However, we do acknowledge that while exports of primary have sharply decreased, exports of aluminium alloy have increased as Chinese producers devise clever ways to label primary metal otherwise as alloy does not carry an export tax yet. We suspect this loophole will be closed sometime over the next year, applying the true test of the efficacy in the governments coolingdown policy.
Source: Reuters
Until we see evidence otherwise, we will continue to forecast a profile of lower than average capacity growth, particularly given the June 19 power tariff increase. In our view, this partial removal of a subsidy is a watershed event in the aluminium market. Although the effect has not been immediate, we think that by the time we update this report next September, we will be reporting lower than forecast production growth and higher costs of production.
200
150
100
50
Source: Reuters
Demand We remain confident the drivers of demand growth for this industrial metal are still in place and macroeconomic indicators across the developing economy spectrum have been persistently strong throughout the credit crisis. It is the less constructive outlook among the mature economies that will keep aluminium consumption growth in check. Indeed the continuation of the US housing slowdown and recent cutbacks in the US auto sector is damaging for
Page 62
27 June 2008
Commodities Quarterly
aluminium demand and perhaps illustrative of consumption growth trends across the rest of the OECD economies. Regardless, we stick to our thesis emerging market growth will offset weakness elsewhere, meaning any surprises to the upside among the mature economies will only benefit industrial metals. Market outlook Despite the already stellar performance of aluminium this year, prices are not that far from the marginal cost of production which is equally appreciating. Higher power costs in China will eventually feed through to the international market.
Current, 2.5
2005 31.97
World primary consumption 31.91 Market balance Stk-to-Consptn ratio (wks) Average cash price (USD/t) Average cash price (USc/lb)
Supply According to the latest reports from the International Copper Study Group and the World Bureau of Metal Statistics, the global copper market so far this year has been in deficit (67Kt and 42Kt, respectively), primarily as a result of shortage on the mined side. This is a similar phenomenon witnessed in 2007, with Brook Hunt estimating 950Kt of mine production lost through supply disruptions. We have factored in 520Kt production loss for 2008. In fact, extreme tightness in concentrate supply has remained a constant over the last three years. Although there is plenty of smelting and refining capacity to supply global demand for copper metal, the miners are simply not able to feed the industry. For instance, Jiangxi Copper announced in late June it is operating well below its smelting capacity of 700Kt due to a lack of concentrate (as well as worries over power supplies over the summer period). Given the cost pressure we detailed in the long-term price chapter of this volume, we expect the deficit in the concentrate market could last well into next year.
Copper
Although copper prices spent most of April in May in a slow but steady decline, in nominal price terms, copper delivered it highest ever quarterly average in Q2 2008. Since the beginning of June, prices have shown resilience on stronger than expected demand indicators as well as a clear down-trend in global exchange inventories including Shanghai Futures Exchange stocks. In fact, global stocks when mapped against consumption levels (ie stock-to-consumption ratio) remain historically low. This has been evident by the widening of the cash to three-month spread as a steeper backwardation reduces the incentive to sell at the front end of the curve.
Figure 10: Weakness in the treatment and refining charge market indicates low concentrate supply
50.0 Realised treatment charge USD/t conc. (lhs) Month end cash price USD/t (rhs) 8800
40.0
6800
10.0
800 May-08
Source: Brook Hunt Note: Treatment and Refining charges (TC/RCs) are the fees miners pay to smelters and refineries to turn their concentrate into copper metal. In times of low mined supply, smelters must either lower their fees to attract metal of reduce production.
Source: Reuters
Page 63
For this reason, the future fortunes of the copper market rely on new production growth. In the coming years, the share of new production growth will increasingly come from frontier countries, or regions of non-traditional output. Whereas these areas of the world has previously been by-passed because of sovereign risks and lack of infrastructure and other development, miners have recently indicated no region is out of their realm of potential investment. Most projects in frontier countries are quality assets, but the ability of this material to be delivered to the market on time are questionable.
Around USD8,000/t on the LME, Chinese imports tend to dissipate as consumers turn to domestic supply, or in other words, undergo a de-stocking phase as evident by SFE inventory movements. At the same time, buying tends to reappear as the London premium dissipates and/or Shanghai inventories touch the critical point of 20Kt something that has occurred only eight times since the exchange began holding inventories at the end of 1996.
Shanghai Premium
100.0
80.0
-500
60.0
-1000
40.0
-1500
20.0
London Premium
-2000 Jun-06 Oct-06 Feb-07 Jun-07 Oct-07 Feb-08 Jun-08 0.0
40% 6% 3% 4% 49%
Kt
The other supply-side theme that we anticipate will also constrain supply is the increasingly more frequent and more violent labour unrest particularly in the Latin American region. With the Cananea mine in Mexico shut indefinitely, non-company specific unrest in Peru, frostier receptions in Argentina and Bolivia and outright expropriations in Venezuela there are few "safe havens" for metals production in Latin America aside from stalwarts Brazil and Chile. Demand Similar to aluminium, refined copper consumption in the OECD economies has been constrained by the downturn in residential and commercial construction particularly in the USA, Germany and Italy. Also analogous to aluminium, growth elsewhere has more than offset that weakness. With Chinese GDP and industrial production still growing at double digits, it is hard to argue that Chinese demand for copper can slow. Around 40% of Chinese copper consumption is accounted for by power infrastructure and given the country remains short energy, we are bullish power sector build-out. Many read declines in Chinese refined copper imports as a slowdown in demand. However, monthly trade patterns are more reflective of the arbitrage between SFE and LME prices than consumption patterns.
Long-term price
Estimating a global market balance is not an exact science because of the many variables in the market (supply disruptions, scrap generation etc). However, we remain bullish in our price assumptions because we think the risks of a greater than expected deficit are greater than a surplus. Moreover, given what we discovered in our cost study, we do not think prices are headed back to 1990s averages. Capital costs have been rising at a rate of 25%
per annum for the past three years and show no sign of abating. We examined one producer operating toward the upper end of the cost curve. Antofagasta (the worlds 10th largest copper producer and a company that is unusually frank in their reporting) primarily produces in Chile said it faces a ``major challenge to maintain margins because the worst drought in 50 years reduced the amount of hydropower available. The company recently reported its cash costs per pound of copper produced for two operations in
Page 64
27 June 2008
Commodities Quarterly
Chile. In its latest reporting period costs at its El Tesoro operation rose from USD1,927/t (87.4USc/lb) to USD2,853/t (129.4USc/lb) while its Michilla operation rose from USD2,714/t (123.1USc/lb) to USD4,220/t (191.4USc/lb) an average of around 50% cost inflation. Given these two operations are around the 60th and 80th percentiles on the cost curve, we think these statistics provide useful insight to cost inflation across the industry.
steel production. We have adjusted our global supply/demand model to account for the lower levels of primary nickel in stainless which has resulted in a market surplus in 2008 instead of a deficit we had originally forecast. While these findings are consistent with the findings of the International Nickel Study Group which has estimated the global refined nickel market balance to be in a 13.7Kt surplus in the first four months of 2008, the World Bureau of Metal Statistics recently calculated the global balance to be in deficit of 7.2Kt over the same period. The discrepancy highlights the delicate balance between surplus and deficit, and is also likely a result of inconsistency in nickel pig iron data (discussed below). The most significant supply-side event affecting the nickel market this quarter occurred in the beginning of June. An interruption of natural gas supply to a major commodity producing region of Western Australia led many producers to either seek alternate sources of power or shut down operations. The owner of the large-scale Kalgoorie smelter decided to push forward its scheduled furnace rebuild by one year, affecting around 28Kt, or 1.8%, of global production this year. Lower LME nickel price means less of a threat from Chinas nickel pig iron sector Just as in 2007, the key to understanding the global nickel market is grasping the trend in Chinese nickel pig iron production. China began importing laterite ores in 2005 in order to produce a low nickel bearing product called pig iron, a substitute for primary nickel. Although cheaper to produce than purchasing primary nickel at LME prices, pig iron is also characterised as labour intensive, energy inefficient and pollutive process. However, the level of production is highly sensitive to the primary (LME) nickel price the lower the price, the less incentive to produce the substitute. Data on Chinas nickel pig iron sector is difficult to ascertain, but despite increased levels of imports of ore from Indonesia, it appears output in 2008 will be lower than last years. Substantially lower nickel prices and subsequent reduced demand from the stainless sector have influenced the decline, but the sector has also been squeezed by government clampdowns on environmental concern as well as rising power and coke costs Thus we find that the range in nickel prices during this period lends to discovering the producer price-pain threshold in nickel pig iron. Based on the observations and data provided by Brook Hunt, Figure 15 details the total costs for producing nickel pig iron.
World refined consumption Market balance Stk-to-Consptn ratio (wks) Average cash price (USD/t) Average cash price (USc/lb)
Nickel
Nickel delivered perhaps the most surprising performance in the LME complex in the second quarter as we expected prices to stay relatively stable, but instead average around 8.5% lower than in Q1. Market weakness derived from a slower than anticipated recovery in the global stainless steel sector especially in Asia. Nevertheless, we think the pull-back was perhaps excessive given LME inventories slowly but steadily drifted lower from their cyclical highs throughout most of the period.
Source: Reuters
Supply Following 20-year lows in global inventories in 2007, refined nickel availability in 2008 is at the highest since 1999. The build occurred while stainless steel producers de-stocked of material purchased at much higher prices and stayed relatively clear of the nickel market. The slow and steady decline inventory declines reflect the equally sluggish rebound in global stainless
Page 65
2008 1542
2009 1674
1429
We also see evidence of robust demand from China from trade data showing the country is delving deeper into net importer territory in spite of market concern over nickel pig iron production. On an annualized basis, China is set to record nearly double the 2007 net import balance.
50000
Demand Converse to other metals in the complex where supplyside dynamics are more persuasive price movers, demand factors are dominating the nickel market. Sentiment toward the Asian stainless sector remains weak, with the regions largest producer, Posco, announcing output cuts in July because customers are delaying orders on expectations that falling nickel prices would lead to lower stainless-steel prices. Some other East Asian stainless steel producers said they may switch a proportion of capacity to produce carbon steel against the slackness in the stainless market. We find this an interesting development because if lower stainless output ends up tightening the market, consumers could very quickly wake up to a market in short supply. Although most evidence suggests the Chinese stainless steel sector is weak, we would have a hard time subscribing to the theory current demand factors are signalling a more permanent downtrend. Recent economic indicators, such as IP growth suggest anything but slower industrial demand. According to independent consultants CRU, it appears that Chinese consumers are satisfying much of their stainless requirements from the countrys high stock levels, with adverse implications for production.
Source: Reuters
World refined consumption Market balance Stk-to-Consptn ratio (wks) Average cash price (USD/t) Average cash price (USD/lb)
14,751 24,237 37,060 26,539 25,105 22,597 6.69 10.99 16.81 12.04 11.39 10.25
Page 66
27 June 2008
Commodities Quarterly
Zinc
Fundamentals in the zinc market have changed significantly in the past 18 months. During the period 2001-2006, there was insufficient concentrate supply to meet demand from smelters resulting in a considerable mine-side deficit in those years. This situation forced many inefficient smelters out of the market or compelled others to cut utilization rates. There was a point in late 2006 when questions were raised over the competence of the global zinc market to adequately supply the world. It was at this point the Chinese emerged as the saviour of the market, scheduling the construction of a vast fleet of new or expanded smelters, increasing capacity nearly 75% between 2006 and 2010. Supply After recording global market deficits every year since 2004, the balance has shifted to surplus in 2008. Both the International Lead Zinc Study Group and the World Bureau of Metal Statistics have calculated a respective 78Kt and 64Kt surplus in the first four months of the year. Despite initial worries, we now know the earthquake in Chinas Sichuan province has little impact on Chinese production. Chinas refined capacity is expected to rise by 15.5% in 2008 and 18% in 2009, accounting for 84% of global new refined growth.
Regardless of what happens with Chinese smelting capacity and production growth, the country will permanently rely on imports of concentrates. Since expanding 54% between 2001 and 2006, Chinese mine capacity has since levelled off while average ore grades have dropped from 5% to 2.5%. Relying on imports brings into question the logistical capability of achieving the sheer volumes needed to supply such a large smelting fleet, particularly given freight availability and cost. We also maintain a belief that similar to refined aluminium, the Chinese government will impose higher tariffs on refined zinc exports they have already removed all import restrictions. We expect any such move to occur from late 2008 or beyond and it would have significant implications across the market. Since the market deficits between 2001 and 2007, the only refined capacity developed has been in China. If Chinese producers lose incentive to export and therefore smelt, the world would again be short of required metal. We have reflected this view in our model and are therefore forecasting a recovery in zinc prices from 2010. Demand Around half of the worlds zinc is used in galvanizing steel, most of which is headed for the infrastructure and construction sectors the majority of which is consumed by developing countries. It is against this backdrop we are forecasting global demand growth to grow in 2008 despite weakness in the OECD economies. We think the spill-over effect from high energy and food price inflation will negatively affect consumer demand in the mature economies, particularly in the auto sector. High fuel prices are having an immediate impact on galvanised steel consumption as smaller cars need less of the material. According to Brook Hunt, yoy US sales in April of pick-up trucks fell by 20% and SUVs dropped by 33% while sales of small cars rose by 19.2%.
Kt
At the same time, the impressive capacity growth may not necessarily lead to metal entering the market with immediate effect as Chinese capacity utilizations rates are reportedly still averaging around 80%. In fact, one surprising development in the market has been the lack of Chinese exports so far this year. In the first five months of the year China has been a net importer or refined zinc. We conclude this phenomenon is likely a function of Chinese players taking advantage of a favourable arbitrage as LME prices dip to their lowest since the end of 2005 and expect exports to emerge later in the year.
Page 67
Coupled with a downturn in most mature economies residential and commercial construction sectors, we are forecasting negative demand growth in OECD countries as an aggregate in 2008. It is the sharp growth in China that is offsetting weakness elsewhere. We remain encouraged by persistently strong Chinese industrial production growth and our Chinese economic team recently increased annual GDP growth forecasts in 2008 and 2009 as a consequence of increased construction as the country reconstructs from the devastating earthquake in the highly populated southwest.
World refined consumption Market balance Stk-to-Consptn ratio (wks) Average cash price (USD/t) Average cash price (USc/lb)
Kt
Lead
The lead outlook looks heavy. Since peaking around USD3,500/t (USD1.59/lb) last October, prices have since fallen 50%. Interestingly, this occurred amid supply constraints from the worlds largest refined producer China. Although prices fell sharply, we must be cognizant that drop came from a historically high level. Supply The run-up to an all-time real and nominal price high was underpinned by two factors. First, shipments from Ivernia's Magellan mine in Australia which accounts for about 3% of global production were halted in March 2007 after the Australian Department of Environment and Conservation suspended the lead export license from the Western Australian port of Esperance following bird deaths. It also found elevated lead levels in the blood of local residents and investigated excessive levels of nickel dust during loading. Second, the introduction in China of an export tax has limited global supply availability since January 2007. In the first five months of this year, Chinese exports are down over 77% from the same period last year and the country is set to record it lowest ever net export trade balance.
Long-term price In order to determine the long-term price for zinc, we essentially have to discover the trigger point of incentive pricing. According to Brook Hunt, that trigger point in 2002 was USD800/t (35.3USc/lb), a level marginally above that years average annual price. Chinese zinc smelters struggled to generate decent profits last year when the price averaged USD3,262/t (USD1.48/lb). Treatment charges (TC) were too low, and there are quite a few examples of smaller plants closing last year because of a shortage of working capital. While there has been an improvement in the TC market this year, the zinc price has not fared as well, averaging USD2,308/t (USD1.05/lb) so far in 2008. Given this scenario coupled with a number of capacity downgrades and efforts to reduce costs announced by producers throughout the year, we feel comfortable nominating USD2,000/t (90.7USc/lb) as the current trigger point.
Page 68
27 June 2008
Commodities Quarterly
Prices have come under pressure recently against market speculation one of these factors is about to change. Although not officially announced, it is expected Ivernia will re-open the Magellan mine very soon. Coupled with recent sharp increases to LME inventories, prices have rapidly declined.
Kt
2009 3156 11.0% 35.2% 1508 -2.0% 16.8% 1610 -0.2% 18.0% 2688 3.9% 30.0% 8,961 4.4%
2010 3393 7.5% 36.5% 1517 0.6% 16.3% 1620 0.7% 17.4% 2774 3.2% 29.8% 9,304 3.8%
Source: Reuters
Demand The second quarter is generally a slow period for lead demand as it falls between the winter and summer auto battery replacement seasons. Consumption indications have also been lower during this period as a result of lower vehicle sales in the economically weaker OECD. According to Brook Hunt, car output in Western Europe fell 4.4% in Q1. Combined with vehicle production cuts from the top US manufacturers, growth auto sector in those two regions is certainly at risk. Regardless of new car growth, the US an Europe still possess massive vehicle fleets that require replacement batteries, one of the largest end-uses for lead. With prices falling back to more sustainable levels, the battery industry is reporting solid sales. The fortunes of the lead market in the coming quarter will rely on temperatures in the US and Europe as extreme levels cause more battery replacements. Among the emerging markets, auto, motorcycle and electric bicycle demand remain robust. Also, a newer segment of demand has arisen. Demand for small power generators in various regions of the world has emerged as a result of power supply disruptions which is bullish of lead demand. Many disaster stuck areas require these generators which typically run on diesel or oil, but also require lead-acid batteries for start-up power.
Tin
Tin has consistently been the best performing metal since 2005, with prices appreciating nearly unabated. The largest factor in this impressive run has been falling production in Indonesia and China. In May of this year however, prices dropped sharply on little more than technical factors. The contract had been supported by a few long-term bulls and we believe the sell-off did not reflect sales to customers. The contract is illiquid enough that once selling began, it cascaded until buyers re-appeared. Prices have since settled where we expect them remain fairly steady for at least the next two quarters. Market outlook Trends in Indonesia and China dominate the market. Industry consultant CRU recently downgraded production from both countries and is forecasting an 18Kt supply deficit. Indonesia recently indicated it will control production in order to maintain price stability, nominating USD15,000/t as an ideal price of tin. Meanwhile, Chinese exports have become negligible as a result of lower feed from Indonesia. Therefore we are forecasting prices to sustain near current levels over the next year. Joel Crane, (1) 212 250 5253 joel.crane@db.com
Page 69
#19 Agriculture
Breaking Records
This month corn prices hit a new all time nominal high of USD7.46/bushel. Consequently corn prices have rallied 241% since the August 2006 low. This has consequently proved to be the most powerful rally in corn prices in recorded history surpassing the gains that occurred in the 1970s. However, prices would need to hit USD12.80/bushel or rise an additional 79% for prices in real terms to touch the highs that occurred in July 1974. For this to mark the most durable rally in corn, prices would need to rise beyond January 2010. In our view, seasonal factors would caution against further gains in corn returns until the summer has past. We expect during the hazardous summer period that wheat is likely to outperform corn.
After spending many years in the doldrums, prices in many parts of the agricultural complex have sprung to life during the last two years. Indeed since the end of 2005, corn, soybeans and wheat have risen by more than 150%, Figure 1.
Last year, US farmers were responsible for a more than 80% of the increase in global corn production as US domestic corn production rose 30%. This strong supply side response was, in our view, partly responsible for corn prices declining by around 50% between February and June 2007. However, this strong increase in production was not sufficient to prevent a decline in the corn inventory-to-consumption ratio during 2007-08. In our view, this tightening in fundamentals was due to strong demand not only from the US ethanol industry, which accounted for 60% of global corn consumption growth last year, but, also from China.
US corn production
2007-08
Bushels (billions)
2008-09
2004-05
2005-06
2006-07
10.2 M J J A S O N D J F M M J J
Source: USDA
Source: Bloomberg
In our view, these price gains reflect a significant tightening in fundamentals across the complex. This has been most marked in corn, where available inventories are now able to cover just 47 days of consumption. The last time inventories were at such critically low levels was in 1974, Figure 2.
Agricultural imports into China have surged over the last few years. This has been most noticeable for soybeans in response to increasing cattle feed requirements. In addition, like thermal coal and aluminium, Chinese net exports of corn have also collapsed with the possibility of the country becoming a net importer of corn going forward, Figure 4.
Page 70
27 June 2008
Commodities Quarterly
Earlier this year, wheat prices had rallied 337% from their December 2005 lows and so the price appreciation in percentage terms fell just shy of the trough to peak rise that occurred in the early 1970s, Figure 6.
-10000
-20000
-30000 Tonnes (000s) -40000 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Since 1972, the average magnitude and duration of rallies in grain and soybean prices has been 130% and 20 months respectively. Rallies in wheat have tended to be stronger and persist for longer while price gains in soybeans have been more modest and unravelled sooner. However, the current rally in agricultural prices has proved to be significantly more powerful than historical averages. Indeed the rise in corn prices since August 2006 has been the most powerful in recorded history with prices appreciating 241% from tough to peak, Figure 5. Between June 1972 and October 1974 corn prices rose 237%.
Highs and lows in the wheat price relate to closing nominal prices. Source: DB Global Markets Research, Bloomberg
The problems that have beset the wheat market since February have been signs of a rebound in global wheat production. This follows extreme droughts in major wheat exporting nations during 2007. However, we expect Australian wheat production trends may be supportive for further gains in the wheat price during the second half of this year. Figure 7 examines the accuracy of official Australian wheat production forecasts. We find that weather events have been responsible for significant revisions to the wheat production outlook during recent years. Indeed severe droughts have cut wheat production in half when compared to the official estimate published during the June quarter of that year. Although ABARE expects Australian wheat acreage to rise 13% to 14 million acres in 2008-09, and which is expected to lead to an 82% rebound in wheat production compared to last year, this is contingent on favourable weather conditions persisting into the second half of this year. We expect wheat prices may also be supported by the strong advance in corn prices which we believe will increase the attractiveness of wheat as an animal feedstock.
Highs and lows in the corn price relate to closing nominal prices. Source: DB Global Markets Research, Bloomberg
Page 71
and which was on a par with the price performance at the beginning of the 1970s, is also running out of steam. However, prices in real terms still remain relatively low on an historical basis, Figure 9. In real terms, we would view soybeans and wheat as trading cheap and corn prices now trading rich. This is in stark contrast to earlier in year when wheat was trading rich relative to corn.
25
Tonnes (million)
20
15
10
Figure 9: Agricultural prices in real terms still remain low compared to the 1970s
40 Deflated by US PPI Wheat price in real terms (2005 US dollars) Corn price in real terms (2005 US dollars) Soybean price in real terms (2005 US dollars)
35 30 USD bushel
25 20 15 10 5
For soybeans, prices have rallied 196% since their September 2006 lows. The rally is consequently in its nd 22 month. As a result, soybean prices only need to rally beyond October this year and it would represent the most durable rally in soybean prices since the beginning of the 1970s, Figure 8.
0 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008
Source: Bloomberg
We have extended this analysis of examining commodity prices in real terms across the commodities complex. We examined monthly commodity price data since 1960 and deflated by the US producer price index. We then calculated how far prices (in real terms) today are away from their all time high hit over the past fifty years. We find the sugar and coffee prices would need to rise approximately 1400% and 790% respectively to hit their all time highs reached in November 1974 and April 1977 respectively, Figure 10. One of the most dramatic moves this year has been US natural gas prices in real terms, which are now only 18% shy of their all time high touched in December 2000.
Figure 10: How far are prices today away from their all time highs in real terms since 1960
1600 1417 1400 1200 1000 800 600 787
Difference between the current price and the all time high in real terms (%)
Cheap
Expensive
Highs and lows in the wheat price relate to closing nominal prices. Source: DB Global Markets Research, Bloomberg
400 200
370 343
270
222
79
68
60
We find that the rally in corn, wheat and soybean prices has proved significantly more powerful than historical norms. Wheats rally has proved interesting since its trough to peak rise was unable to surpass the gains in the 1970s. The recent correction in corn prices could be a sign that the rapid price appreciation since 2006,
24
18
0
Z S o in c yb ea ns W he at N ic ke l Le Al um ad in iu m C or n C of fe e Si lv er C ot to n C oc oa Su ga r
Source: DB Global Markets Research, IMF, Bloomberg (data as of June 23, 2008)
Page 72
Ti n G ol C d op U pe S na r tg a C ru s de oi l
27 June 2008
Commodities Quarterly
Conclusion The performance of agricultural markets over the past few years has been one of winners and losers. For example, in 2005 sugar was the worlds best performing commodity but a rapid supply response a year later led to all the price gains in 2005 being surrendered by the beginning of 2007. The performance of grain prices has also been highly divergent, with wheat treading expensive to corn in 2007 to be now trading cheap relative to corn, Figure 11.
2.2
1.8
1.4
Wheat trades cheap relative to corn Apr-06 Aug-06 Dec-06 Apr-07 Aug-07 Dec-07 Apr-08
1 Dec-05
Source: Bloomberg
We expect further price gains across the agricultural sector. However, the game of winners and losers suggest wheat prices may start to outperform relative to corn in our view given the risk of further downgrades to the Australian wheat harvest and seasonal factors since 1989 that have tended to favour wheat over corn during the summer months.
Average monthly wheat returns on the DBLCI Average monthly corn returns on the DBLCI
Page 73
#20 Uranium
Atomic Potential
Given that the world is becoming increasingly short of affordable and reliable energy, we maintain our belief that the planet is on the verge of a nuclear renaissance. We note that uranium is significantly underperforming other energy commodities (oil, thermal coal, natural gas) in 2008. At the same time, if it were to occur, we think the uranium industry could initially struggle to respond to such a spike in demand. As we have seen repeatedly throughout this commodities cycle, lower supply leads to higher prices. We believe the long-term price indicator is a good harbinger of where market players see fair value prices. At the moment, the longterm price is nearly precisely between last Junes peak and this Junes trough. Although the shorter-term market appears to be well supplied, we suspect utilities may aggressively re-enter the spot market this year.
Mined uranium ore is a mildly radioactive element with two principle isotopes, U-235 and U-238, with the former suitable for use as nuclear fuel. This material must be processed before it can be used as a fuel for a nuclear reactor, and the first step, milling, is usually done at the mine site. The mixture of uranium oxides produced after milling is called uranium oxide concentrate (U3O8) and often referred to as yellowcake because of its khaki colour. It is in this form that the primary producers sell to the power producing industry. It is then up to the nuclear utilities to organise the contracts for the next three steps in the nuclear fuel cycle, conversion, enrichment and fuel fabrication. All uranium transactions are over the counter with the market split between long term supply contracting and a spot market, roughly a 90/10 split in favor of contracting. The spot market generally consists of contracts for delivery of material within a six-month period and is what is commonly quoted as current price. Term contracts average delivery term and volume vary by transaction, but as an example, a typical contract does not usually begin before 2-3 years and lasts roughly 3-5 years with an average volume per delivery of around 115tU. Uranium the fuel Currently, around 16% of the worlds electricity is produced from nuclear energy. The rate of expansion in the industry was high from the early 1970s until 1980s, but has since slowed as fewer plants have come into operation. Today there are 448 nuclear reactors with a total output capacity of more than 375GWe operating in 31 countries. According to our estimates, a further 84 reactors are under construction or permitted to begin power generation between now and 2015. Dozens more around the world are proposed or in the application process.
Uranium the commodity Uranium is a relatively common element that is found throughout the world and is mined in several different countries. It is as about as common as zinc and about 40 times as abundant as silver. Canada and Australia mine by far the largest amount of material and although Canada currently produces more on a recurrent basis, Australia has the largest known economically recoverable reserves. Kazakhstan has rapidly become a global player and is set to become the world largest producer.
Canada Australia Kazakhstan Niger Russia Namibia Uzbekistan USA Ukraine South Africa China Czech Repub. Total world Supply growth
Source: Ux
2005 11.6 9.5 4.4 8.0 3.3 3.1 2.3 1.0 1.0 0.7 0.7 0.4 41.7 5.4%
2006 9.9 7.6 5.3 3.4 3.5 3.1 2.3 1.6 0.8 0.5 0.7 0.4 39.5 -5.2%
2007 9.5 8.7 6.9 3.5 3.6 3.1 2.3 1.9 0.8 0.6 0.7 0.3 42.7 8.0%
2008 10.1 8.8 12.9 3.5 3.8 3.0 2.4 2.6 0.8 0.7 0.7 0.3 51.4 20.5%
2009 11.1 9.0 13.7 3.5 4.2 3.2 2.5 2.8 0.8 0.7 0.7 0.3 55.5 7.8%
2010 9.5 9.1 16.2 3.5 4.2 3.4 2.5 3.7 0.8 0.7 0.7 0.3 58.9 6.1%
Source: EIA
Page 74
27 June 2008
Commodities Quarterly
Asia China India Japan Korea Pakistan Taiwan Total South America Argentina Brazil South Africa Total Eastern Europe Armenia Czech Rep Bulgaria Hungary Lithuania Romania Russia Slovakia Slovenia Ukraine Total Western Europe Belgium Finland France Germany Netherlands Spain Sweden Switzerland United Kingdom Total North America Canada Mexico USA Total
11 19 55 20 2 6 113
18 5 10 7 1 2 44
A nuclear renaissance In our view, the world is on the verge of a nuclear renaissance as many countries energy policy makers seek to reduce their great dependence on oil, gas and imported power. While we are projecting a robust demand profile for existing nuclear power programs, a number of countries around the world without existing programs have signalled their interest in starting one, from Latin America to the Middle East. Nevertheless, the real growth story in nuclear new build is found in the regions with the most amicable regulatory environment, North America and Eastern Europe/Russia and Asia. China has stated its goal to expand the installed capacity of nuclear power to at least 40 gigawatts (GWe) to account for more than 5% of the national total power installed capacity by 2020 (the share is currently around 2%, or ~9GWe). As an incentive, the Ministry of Finance is offering nuclear power companies a valueadded tax rebate of 75% each year during their first five years of operation, 70% in the following five years and 55% for the third five year period. In terms of the upcoming new US administration, both candidates are bullish for nuclear power new build, with John McCain much more vocal than his competitor Barack Obama. McCain claims he will push to add 100 new nuclear reactors, almost doubling the current fleet of 104 as part of his plan to address the nation's energy woes. Obama has said he supports exploring greater use of nuclear power as part of the U.S. energy mix, but did not similarly commit with numbers of reactors. The Russian government has confirmed its very ambitious plans which were first announced last September for a substantial increase in nuclear generation capacity by 2020. As well as the 4.8GWe capacity now under construction, a further 12GWe is planned for completion mostly by 2016, and then another 16 to 22GWe proposed by 2020.
2 2 2 6
1 1 1 3
0 0 0 0
1 6 4 4 1 2 31 6 1 15 71
376 3472 1906 1755 1185 1361 21743 2034 686 13330 47848
0 2 2 0 1 0 17 2 0 3 27
1 0 2 0 1 0 2 2 0 0 8
0 3524 3906 1755 1000 1361 37486 2466 686 16330 68514 43%
7 4 59 17 1 8 10 5 23 134
5761 2656 63363 20384 452 7430 8933 3220 11192 123391
0 1 1 0 1 0 0 0 0 3
3 0 0 7 0 1 1 0 16 28
4014 4256 64963 12018 452 6984 8766 3220 3838 108511 -12%
448 374046 85 Global Total Source: DOE, EIA, Reuters, Bloomberg, Ux, WNA
Page 75
market will require all new projects to successfully come to market without delays.
80%
70%
60%
50% 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Source: Ux
We recently reviewed the International Energy Agencys (IEA) 2050 energy targets presented to Energy Ministers from the G8, China, India and Korea. The group outlined its target of significantly reducing carbon emissions that will require an energy mix that has a considerably higher contribution from nuclear energy. The IEA estimates that between 24 and 32 new 1,000Mw nuclear reactors will be required per annum between 2010 and 2050. This equates to between 960 and 1,280 new reactors, on top of the 448 reactors currently in operation. Although a scenario as ambitious as the IEAs would require a sea-change in global legislation and energy policy shift, we are believers in such a shift. At the same time, if it were to occur, we think the uranium industry could initially struggle to respond to such a spike in demand. As we have seen repeatedly throughout this commodities cycle, lower supply leads to higher prices. Supply availability: Will there be enough to go around? The shorter-term market appears to be relatively wellsupplied. This has been achieved through lower than expected spot procurement, the success of Paladin and other new entrants to the market and adequate secondary material (HEU feed, recycled fuel, US government stocks). However, we think the market will retighten relatively quickly for a number of reasons. First, secondary supplies are finite and rapidly being depleted. Since 2006, the percentage of mined supply needed in the global market has increased and will approach 90% by 2014. Second, as mined supply increases in importance, so does the significance of the success in new projects entering the market. Severe under investment in the uranium sector has resulted in few if any significant new uranium discoveries since 1980. Two of the worlds largest projects - Cigar Lake and the Olympic Dam expansion have been delayed by several years with construction costs ballooning into the stratosphere. As a consequence, by 2014, the
Figure 7: Percentage of new projects that need the go-ahead to meet new demand
100%
75%
50%
25%
Kazakhstan hopes to be the answer to that dilemma. With the world's second largest uranium reserves, the country wants to become the planet's largest uranium supplier by 2010. Although there are very ambitious mined capacity increases, given recent experiences we remain sceptical of those targets being met. Earlier in the year, guidance for output in 2008 was reduced as a result of sulphuric acid shortages and questions remain over the countrys ability to provide adequate power supply, transportation and skilled labour. Market outlook Between 2003 and 2007 uranium prices rose just over 1,200%. Since its peak it has declined almost 60% and during this period the uranium market has undergone significant change, in our view. For the first time in two decades, market demand has became sensitive to price after spending years in the doldrums. Now both buyers and sellers have become more cautious in bids and offers. We find that the utilities are happy to buy only when necessary, forcing sellers of uranium to come up with increasingly competitive offers which has pressured prices lower.
Page 76
27 June 2008
Commodities Quarterly
So far in 2008, although the spot market activity has been slow compared to normal, the actual volumes are at standard levels meaning spot transactions have been few but large. This tells us there are plenty of sellers patiently waiting on the sidelines, fully aware of the bullish scenario outlined above. Movement in the long-term price indicator or lack there of is also a good harbinger of where market players see fair value prices. Since reaching USD95/lb in May 2007, the long-term price has fallen to just USD90/lb today. To us, the price movements in this market are typical of lower-liquidity markets and take time to correct themselves. The long-term market appears to represent where fair value is and at the moment, is nearly precisely between last Junes peak and this Junes trough. Thus we feel confident in forecasting higher prices from here. Going forward, we believe the combination of production problems and dwindling secondary supply will lead to a greater propensity for utilities to scramble to secure supply. We believe that utilities will re-enter the market strongly from the third quarter of this year and lead to strong advances in price over the coming year.
Source: Ux
In our opinion, the current spot price of USD57/lb is not high enough to encourage new uranium supply and believe that continual supply issues and the likelihood of increased demand from utilities should drive the price higher in the coming months. We think this level represents the bottom of the market as that is the estimated level of new, higher cost production in Africa, Kazakhstan and even in the US. In fact, at USD60/lb, some projects begin to look uneconomic, thus eroding the supply outlook at the marginal cost of production primarily smaller producers. As such we have nominated USD58/lb as our long-term average price from the year 2013.
Page 77
80
60
20
The rise in bulk commodity prices during 2008 has been substantial compared to other parts of the commodities complex and specifically those commodities where a futures market exists for example in crude oil or copper. Within the bulk commodities, the doubling in iron ore lump prices looks trivial compared to the price gains that have occurred in coking coals. Figure 1 illustrates the expected price appreciation for a variety of commodities this year compared to 2007. Price progression has clearly favoured the bulk commodities.
While the thermal coal market remains tight in its own right, the extreme tightness in the coking coal market has impacted the thermal coal market to the extent that some high quality thermal coals have been moved into the semi-soft coking coal market. This added pressure on the thermal coal market has continued to drive prices up with recent spot coal prices from Newcastle being struck north of US$160/t FOB. We are not expecting a repeat of the Queensland rains next year an to that extent believe that some of the semi-soft coking coal will again make its way back into the thermal coal market, but not enough to lower benchmark prices and we are now expecting another rise in the thermal coal price next year. The continued tightness in the coal supply chain was highlighted over the last fortnight with adverse weather conditions impacting the loading operations at the Australian coal port of Newcastle resulting in the ship queue increasing back to 40 ships.
Bulks Other
Coal In the near-term we expect the thermal coal market to remain tight in 2009 resulting in another price rise to the contracted coal before moving back into balance during 2010 as the semi-soft coking coals move back into the steaming coal market.
Page 78
27 June 2008
Commodities Quarterly
Long term looking much stronger Energy prices have been rising with increasing global demand and the underlying energy commodity prices have followed in a broadly similar manner as seen in the chart below. Differences in the various commodity price responses are due to the level of historic overinvestment (hence the timing of demand catch-up) and regional supply response differences between the commodities. The chart below shows the spot price performance of four key energy commodities.
2011/2012E (for thermal coal). Given that the exports are such a small part of the entire coal equation for China, there is a risk that this could change much more quickly. 2) South Africa: A mismatch between power generation capacity growth and demand in South Africa has reduced the energy reserve margin from 25% in 2002 to a mere 8-10% now. We expect it will fall even further in the coming years. Eskoms initial presentations to stake-holders on the state of play has revealed a stark outlook with demand reduction the main solution until additional base load coal-fired power generating capacity is brought on stream in 2012. More recently Eskom has announced that it is ceasing load shedding (in line with its plan) as energy saving has eased the power shortage. However, a number of producers remain on reduced load with resultant reduced production. While the supply appears to be pulling back from crisis point, the growth outlook in the medium term remains muted, in our view. 3) Australia: Australia continues to face major supply issues, compounded by heavy rains which have forced a number of key producers to declare force majeure in the first quarter of this year. We expect the port and rail issues in Australia will continue to restrict export capability until 2010, and hence limit tonnage growth. Earlier this year the Australian Corporate Regulator (ACCC) approved a continuation of the existing export quota system (CBS) for 2008, rejecting the proposal by the terminal operators for a new system based on rail contracts. The continued use of the export quota system and the lack of immediate capacity growth will, in our view, limit thermal coal exports from Australia for the next 12 months or more. Iron ore tight despite expected producer response We forecast the iron ore supply demand market to remain tight through 2009 despite the delivery of a number of projects on the supply side. The most recent IISI release for May showed global steel production grew +6.7% yoy with the third consecutive month of annualized global steel production above 1.4bt. In the last 12 months, China and India have continued to dominate the production growth accounting for 60% and 11% of the worlds steel production growth respectively. This represents an annual steel production growth rate of 11% for China and 21% for India, Figure 5.
Figure 4: Relative spot performance of oil, gas, uranium and thermal coal
1600 1400 1200 1000 800 600 400 200 0 May-90 May-92 May-94 May-96 Uranium May-98 May-00 May-02 May-04 May-06 May-08
Crude Oil
Natural Gas
Thermal Coal
For setting our long-term thermal coal price expectation, we have been particularly interested in the relationship between oil price and coal price. We are expecting that this general relationship between oil and coal prices a will be maintained, and assuming a long run oil price of USD100/bbl this would tend to imply a thermal coal price of around USD80/tonne by the middle of the next decade. Rising costs, including energy, have had a significant impact on coal production costs in our view. We believe there are three key reasons for tightness in the thermal coal markets to continue: 1) China: Chinese net exports of thermal coal have more than halved from a peak of around 65Mt in 2003 to just over 30Mt in 2007. On our estimates, domestic production of thermal coal in China in 2007 was around 2.1bn tonnes. Consequently net exports represent just 1.5% of total production, and are a minor part of the equation for China, but Chinese exports have a significant impact on the regional seaborne market. Despite being a small part of the Chinese coal production, the Chinese government has been restricting exports in light of the power crisis. As a result, we view the uptick in April net export levels as a catch up from the low levels in March. We now await the May figures, which we expect will be significantly impacted by the recent earthquake and the ensuing Government response to preferentially transport food and coal to regions in need. In the medium term we expect the net export position to dwindle even further with China moving to a balanced position by
Page 79
Despite the continuing run up in steel production, demand remains strong with producers continuing to be able to achieve price rises at a surprisingly rapid rate. The continuing price rises that we are seeing in steel give us confidence that benchmark iron ore prices will likely rise again next year as has been seen in previous price cycles. We find that iron ore demand in China remains voracious with April recording records in iron ore production, which is running at an annualised level of 830Mt up 26%yoy) as well as imports which have risen to an annualised 521 Mt and up 28% yoy. This is the first time iron ore imports have risen over the 500Mt mark and we believe is a reflection of the declining iron ore content of domestic iron ore supplies. We expect the market to remain in apparent deficit until 2011 despite some significant step ups in supply. In 2008 we expect iron ore production from Australia to increase by 24% and then by 22% in 2009 as FMGs production ramps up and the large Australian producers continue to deliver expansions (this is after a more modest 8% increase in 2007). The real push over the line to apparent surplus comes in 2011 when we see Vale delivering on its growth projects. As can be seen on the chart below, the market has been in apparent deficit most of this decade, which has been reflected in the price rises seen, we believe that the deficit may be over-estimated as we expect that not all Chinese production has been accounted for in historic figures.
Conclusion We view the rise in bulk commodity prices as fundamentally based this year. We expect tightness in thermal coal and global iron ore markets to continue.
Page 80
27 June 2008
Commodities Quarterly
Commodities Chartbook
Commodity consumption around the world relative to per capita income Figure 1: Oil consumption intensity
3.0 Oil consumption per capita (gallons per day) Canada 2.5 South Korea 2.0 Japan Australia 1.5 Germany Sweden France UK Italy Venezuela MexicoRussia Thailand 0.5 Indonesia India 0 5 Brazil China Taiwan US
3 South Korea Thailand India Indonesia Mexico Ven Russia China Brazil 5 10 15 Taiwan Japan Canada UK Sweden 40 US
1.0
0.0
0
10 15 20 25 30 GDP per capita ('000 USD) 35 40 45 50
35
45
50
US
Sweden Germany
US
Japan
10 8 Germany 6 4 2 0 0 China Mexico Russia Thailand Brazil India Venezuela Indonesia 5 10 15 20 25 30 GDP per capita ('000) Italy France Canada Japan Sweden UK US
Italy
US
35
40
45
50
Page 81
27 June 2008
Commodities Quarterly
Commodities Chartbook
Commodity consumption around the world relative to per capita income Figure 7: Iron ore consumption intensity
1200 Japan 1000 Iron Ore consumption (Kg per capita)
Australia 800
South Korea
10
600
Venezuela
Russia
Taiwan Germany
Sweden
8 Germany 6 France Sweden Russia India Brazil China 5 10 15 20 25 30 GDP per capita ('000) 35 40 45 50 Canada Japan US
400
China
Brazil
Italy France
Canada UK US
200 India 0 0 5 10 15
2
20 25 30 GDP per capita ('000) 35 40 45 50
UK
0 0
40 Venezuela
30
20
Taiwan
10
40
50
Page 82
27 June 2008
Commodities Quarterly
Commodities Chartbook
Commodities relative to G7 per capita income Figure 1: Crude oil prices relative to per capita income
3000 2500 Number of barrels of oil 2000 1500 1000 500 0 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 G7 per capita income divided by the price of oil Oil price decline helps to boost the purchasing power of a G7 consumer
Ounces 100
80
60 ` 40
20
Tonnes
4 3 2 1 0
40 30 20 10 0
1970-2007 average
1972
1977
1982
1987
1992
1997
2002
2007
Page 83
Ounces
27 June 2008
Commodities Quarterly
Commodities Chartbook
Commodity inventory-to-use ratios Figure 1: OECD oil inventory-to-consumption ratio
65 Total OECD crude oil and product inventories relative to demand
60
Days
55
3000
2000
50
45 1995
1997
1999
2001
2003
2005
2007
0 1Q88
1Q90
1Q92
1Q94
1Q96
1Q98
1Q00
1Q02
1Q04
1Q06
Source: EIA
250
16
50
0 1Q88
1Q90
1Q92
1Q94
1Q96
1Q98
1Q00
1Q02
1Q04
1Q06
0 1Q08
0 1Q88
1Q90
1Q92
1Q94
1Q96
1Q98
1Q00
1Q02
1Q04
1Q06
0 1Q08
0 1Q88
1Q90
1Q92
1Q94
1Q96
1Q98
1Q00
1Q02
1Q04
1Q06
0 1Q08
Source: USDA
Page 84
27 June 2008
Commodities Quarterly
Commodities Chartbook
Commodities prices in real terms Figure 1: Crude oil prices in real terms
140 120 100 80 60 40
200
Real crude oil price (PPI deflated, 2005 US dollars) Real crude oil price (CPI deflated, 2005 US dollars)
Real gold price (USD/oz, lhs) Real silver price (USD/oz, rhs)
80 70 60 50 40 30 20 10 0
20 0 1972
0 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006
1976
1980
1984
1988
1992
1996
2000
2004
2008
Real aluminium price (USD/tonne, lhs) Real copper price (USD/tonnes, rhs)
10000
Real zinc price (USD/tonne, lhs) Real nickel price (USD/tonnes, rhs)
50000
5000
8000
40000
4000
6000
30000 3000
3000
4000
20000 2000
2000
2000
1000 0 1960
10000
1000 1960
Real lead price (USD/tonne, lhs) Real tin price (USD/tonne, rhs)
3000
2000 ` 1000 15000 10000 5000 0 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 0
10
0 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 2008
Page 85
27 June 2008
Commodities Quarterly
Commodities Chartbook
Commodity Forward Curves Figure 1: WTI crude oil forward curve
150
140
Price (USD/tonne)
Price (USD/tonne)
130
June 2008
120
100
90 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 Tenor (month)
8000 7800 7600 7400 7200 7000 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 Tenor (month) June 2008
11
13
15
17
19
21
23
25
27
Tenor (month)
June 2008
2300
Price (USD/tonne)
2200
2100
850 Mar08
Jun08
Sep08
Dec08
Mar09
JunSepMonth 09 09
Dec09
Mar10
Jun10
Sep10
Dec10
Page 86
27 June 2008
Commodities Quarterly
Commodities Chartbook
BRIC & OECD commodity demand Figure 1: Aluminium demand
55,000 45,000 35,000 25,000 30% 15,000 5,000 -5,000 BRIC 2003 2004 OECD 2005 2006 2007 2008 2009 2010 2011 2012 20% 10%
0% 2003 BRIC 2004 2005 2006 2007 2008 2009 2010 2011 2012
Tonnes (000s)
20,000
40%
0%
Others
BRIC as a % of w orld
OECD as a % of w orld
OECD
OTHERS
BRIC As a % of World
OECD As a % of World
OTHERS
BRIC As a % of World
OECD As a % of World
OECD
OTHERS
BRIC As a % of World
OECD As a % of World
OTHERS
BRIC As a % of World
OECD As a % of World
OTHERS
BRIC As a % of World
OECD As a % of World
Page 87
27 June 2008
Commodities Quarterly
Figure 1: Occidental Petroleum CDS spreads and the WTI crude oil price
80 70 60 50 40 30 20 10 0 Jan03 5Y CDS spread (bp, lhs) Oil prices (WTI USD/bbl, inverted rhs) 15 25 35 45 55 65 75 85 95 105 115 125 135 145 Jul03 Jan04 Jul04 Jan05 Jul05 Jan06 Jul06 Jan07 Jul07 Jan08
Figure 3: Burlington Resources, Inc. CDS spreads and the US natural gas price
80 70 60 50 8 40 10 30 20 10 0 Jul-03 Jan-04 Jul-04 Jan-05 Jul-05 Jan-06 Jul-06 Jan-07 Jul-07 Jan-08 12 14 16 5Y CDS spread (bp, lhs) US natural gas price (USD/mmBtu, inverted rhs) 4 6 2
Figure 2: Conoco Phillips CDS spreads and the WTI crude oil price
80 70 60 50 40 30 20 10 0 Jan03 5Y CDS spread (bp, lhs) Oil prices (WTI USD/bbl, inverted rhs) 15 25 35 45 55 65 75 85 95 105 115 125 135 145 Jul03 Jan04 Jul04 Jan05 Jul05 Jan06 Jul06 Jan07 Jul07 Jan08
Figure 4: Devon Energy Corporation-CDS spreads and the US natural gas price
110 100 90 80 70 60 8 50 40 30 20 10 0 Jan-03 Jul-03 Jan-04 Jul-04 Jan-05 Jul-05 Jan-06 Jul-06 Jan-07 Jul-07 Jan-08 14 16 10 12 4 6 5Y CDS spread (bp, lhs) US natural gas price (USD/mmBtu, inverted rhs) 0 2
* A credit default swap allows one party to buy protection from another party for losses that might be incurred as a result of default by a specified reference credit (or credits). The buyer of protection pays a premium for the protection, and the seller of protection agrees to make a payment to compensate the buyer for losses incurred upon the occurrence of any one of several specified credit events. Source: Moodys
Page 88
27 June 2008
Commodities Quarterly
Figure 1: Alcoa CDS spreads and the LME aluminium spot price
170 150 130 110 90 70 50 30 10 Jun-05 Alcoa 5Y CDS spread (bp, lhs) Aluminium spot price (USD/metric ton, inverted rhs) 1,050 1,300 1,550 1,800 2,050 2,300 2,550 2,800 3,050 3,300 Dec-05 Jun-06 Dec-06 Jun-07 Dec-07
Figure 3: Phelps Dodge CDS spreads and the LME copper spot price
286 256 226 196 166 136 6,000 106 76 46 16 Jun-05 7,000 8,000 9,000 Dec-05 Jun-06 Dec-06 Jun-07 Dec-07 Phelps Dodge 5Y CDS spread (bp, lhs) Copper spot price (USD/metric ton, inverted rhs) 1,000 2,000 3,000 4,000 5,000
Figure 2: Inco CDS spreads and the LME nickel spot price
140 120 100 80 34,000 60 42,000 40 20 Jun-05 50,000 Inco 5Y CDS spread (bp, lhs) Nickel spot price (USD/metric ton, inverted rhs) 2,000 10,000 18,000 26,000
Figure 4: Barrick Gold CDS spreads and the gold spot price
110 100 90 80 70 60 50 40 30 20 10 Jun-05 920 1,020 1,120 Dec-05 Jun-06 Dec-06 Jun-07 Dec-07 520 620 720 820 Barrick Gold 5Y CDS spread (bp, lhs) Gold spot price (US$/troy ounce, inverted rhs) 420 320
Dec-05
Jun-06
Dec-06
Jun-07
Dec-07
* A credit default swap allows one party to buy protection from another party for losses that might be incurred as a result of default by a specified reference credit (or credits). The buyer of protection pays a premium for the protection, and the seller of protection agrees to make a payment to compensate the buyer for losses incurred upon the occurrence of any one of several specified credit events. Source: Moodys
Page 89
27 June 2008
Commodities Quarterly
Figure 1: Alcoa stock price and the LME aluminium spot price
55 50 45 40 35 30 25 Jun-06 2600 2300 2000 Dec-06 Jun-07 Dec-07
Alcoa share price (USD, lhs) Aluminium spot price (USD/tonne, rhs)
Figure 4: Freeport McMoRan stock price and the LME copper spot price
140 130 120 110 100 90 80 70 60 50 40 30 Jun-06
Freeport McMoran share price (USD, lhs) Copper spot price (USD/tonne, rhs)
Dec-06
Jun-07
Dec-07
Figure 2: Norilsk Nickel stock price and the LME nickel spot price
355 305 255 205 155 105 55 5 Jun-06 Dec-06 Jun-07 Dec-07
Norilsk Nickel share price (USD, lhs) Nickel spot price (USD/tonne, rhs) 70000 60000 50000 40000 30000 20000 10000
Figure 5: Newmont Mining stock price and the gold spot price
80 70 60 50 40 30 Jun-06
Newmont share price (lhs) Gold price (USD/oz, rhs)
Dec-06
Jun-07
Dec-07
Figure 6: Bumi Resources stock price and spot thermal coal prices (Asia & Europe)
90 80 70 60 50 40 30 20 10 0 Jun-06 Dec-06 Jun-07 Dec-07
Bumi Resources share price (IDR, lhs) API#2 Coal price (Cal 08, USD/tonne, rhs) API#4 Coal price (Cal 08, USD/tonne, rhs)
Page 90
27 June 2008
Commodities Quarterly
Price Forecasts
Energy Price Forecasts
USD/barrel WTI Brent RBOB gasoline Heating oil IPE gasoil Q108 97.8 96.31 2.48 2.74 871 Q208 122.6 121.5 3.00 3.31 1084 Q308 135.0 135.0 3.52 3.63 1210 Q408 125.0 125.0 2.90 3.46 1120 2007 72.36 72.66 2.07 2.04 638 2008 120.0 119.5 2.91 3.28 1071 2009 120.0 120.0 2.88 3.29 1075 2010 100.0 100.0 2.48 2.79 895 2011 105.0 105.0 2.50 2.96 940 2012 110.0 110.0 2.65 3.08 985 2013 115.0 115.0 2.80 3.21 1030
Figures are period average and relate to the spot price; Source: DB Global Markets Research forecasts
Page 91
27 June 2008
Commodities Quarterly
Price Forecasts
Precious Metals Price Forecasts
USD/oz Gold Silver Platinum Palladium Q108 925 17.6 1866 447 Q208 898 17.2 2030 443 Q308 925 16.9 2000 480 Q408 960 17.5 21001 495 2007 698 13.4 1306 356 2008 927 17.3 1999 466 2009 1000 18.2 2150 505 2010 925 16.8 1925 530 2011 800 14.6 1700 545 2012 650 11.8 1250 450 2013 600 10.9 1225 440
Page 92
27 June 2008
Commodities Quarterly
Glossary
API: American Petroleum Institute sets standards for specific gravity of crude oil. API#2 TFS API#2 average price index for coal delivered CIF ARA. API#4 TFS API#4 average price index for coal loading FOB Richards Bay, South Africa. ARA: Amsterdam-Rotterdam-Antwerp major delivery hub for cargo entering Northwest Europe. Backwardation Market condition in which forward prices decline as tenor increases. Barrel (bbl): liquid measure of 42 US gallons. Bcf: Billion cubic feet macro measure of natural gas volume. Bpd: Barrels per day measure for oil production or use. Bunkers Fuel oil used to power ships. CAT: Cumulative average temperature. The sum of daily high + low)/2, usually over a month or season. CDD: Cooling degree day excess of daily average temperature over 65F; usually cumulated over time. CIF: Cost, Insurance, and Freight denotes commodity price delivered to destination, e.g. fuel oil CIF Rotterdam. Clean Spread: The spark spread minus the cost of emissions. Contango Market condition in which forward prices increase as tenor increases. Crack Price spread between crude oil and refined product (after the refining process of cracking large molecules to make smaller). DBLCI: Deutsche Bank Liquid Commodities Index tracks six commodities, rolling positions in crude oil and heating oil monthly, and in gold, aluminum, corn and wheat once per year. Reuters: DBLCI. Bloomberg: DBCM. DBLCI-MR: DBLCI-Mean-reverting rule-based variant of the above; under-weights those commodities amongst the six which are expensive relative to their long-term average, and overweights those which are relatively cheap. Distillate: Class of refined oil products including heating oil (aka gasoil) and diesel, and usually jet fuel and burning kerosene.
DOE: US Department of Energy. Often used as synonym for its EIA arm. EIA: Energy Information Administration. Statistical arm of the US DOE, which releases weekly and monthly data. FOB: Free on Board denotes commodity price loaded and cleared for export at load port, e.g. coal FOB Richards Bay, South Africa. Fuel oil (FO) Dense refined oil product used to fuel ships and generating stations. German Dark Spread: The spread between German power and coal -- Dark Spread = German power coal/(2.65*EURUSD) HDD: Heating degree day deficit of daily average temperature below 65F in US, 18C elsewhere. Henry Hub: Louisiana delivery point for NYMEX natural gas. HSFO: High sulphur fuel oil. LNG: Liquefied natural gas can be shipped on specialpurpose tankers. mmBtu: million British Thermal Units Natural gas heatcontent measure, approx. 1000 cubic feet. PADD: Petroleum Area of Defense District US regions for petroleum market data, defined approximately as: PADD1 East coast PADD2 Midwest PADD3 Gulf coast PADD4 Inter-mountain west PADD5 West coast Spark Spread: Price spread between electricity and the fuel (see also UK Spark Spread and German Dark Spread). UK Spark Spread: The spark spread represents the marginal value of selling UK electricity and buying UK natural gas for a gas fired power station. Market standard UK Spark Spread = UK power UK Natural Gas * 0.6944 Therm: 100,000 British Thermal Units, or 0.1 mmBtu WTI: West Texas Intermediate benchmark US crude oil; for NYMEX futures, delivered Cushing, Oklahoma.
Contacts
Name
Richard Jefferson Louise Kitchen
Title
Head of Commodity Sales, Europe, Asia Global Head of Commodities Structuring & Sales
Telephone
44 20 7547-7689 44 20 7547-5395
Email
richard.jefferson@db.com louise.kitchen@db.com
Location
London London
Page 93
Page 94
LGO MR TR EUR GBP NOK CAD AUD JPY ED ECU 3m AUD 3m SPX NG MAL MCU HG MNI MZN GC PL SI W C DBLCI DBLCI- GSCIiBOX
Correlation Matrix
27 June 2008
FALSE
CL
LCO
HU
HO
Commodities Quarterly
Light Crude Brent Unleaded Petrol Heating Oil Gas Oil Natural Gas
0.19 0.19 0.13 0.24 0.22 0.21 0.54 0.57 0.37 0.32 -
0.51 0.52 0.41 0.44 0.39 0.28 0.33 0.32 0.64 - 0.01 0.22
LME Al LME Cu High Grade Copper LME Nickel LME Zinc Comex Gold Future NYMEX Platinum Comex Silver
0.55 0.56 0.50 0.51 0.40 0.21 0.30 0.33 0.49 0.12 0.24 0.70
0.48 0.48 0.37 0.38 0.32 0.35 0.11 0.20 0.66 0.06 0.14 0.84 0.67
0.37 0.34 0.34 0.24 0.12 0.06 0.03 0.06 0.29 0.30 0.04 0.35 0.36 0.34
0.31 0.29 0.24 0.25 0.07 0.19 0.16 0.09 0.15 0.21 0.09 0.20 0.34 0.32 0.67
0.95 0.94 0.86 0.90 0.45 0.47 0.23 0.15 0.43 0.26 0.19 0.59 0.64 0.56 0.55 0.51
0.92 0.91 0.83 0.87 0.48 0.52 0.34 0.23 0.54 0.24 0.29 0.64 0.67 0.60 0.43 0.41 0.96
0.98 0.98 0.92 0.90 0.51 0.56 0.22 0.14 0.41 0.25 0.21 0.56 0.60 0.54 0.40 0.38 0.97 0.95
0.23 0.25 0.24 0.26 0.32 0.21 0.51 0.47 0.24 0.00 0.18 0.45 0.37 0.33 0.06 - 0.09 0.28 0.35 0.27 -
- 0.01 0.03 - 0.00 - 0.09 0.04 - 0.14 - 0.13 0.11 0.20 - 0.27 - 0.19 0.18 0.13 0.33 0.05 - 0.04 0.01 0.04 0.00 0.43 -
0.24 0.25 0.22 0.27 0.36 0.10 0.44 0.49 0.19 0.05 0.21 0.43 0.34 0.30 0.09 0.11 0.27 0.31 0.26 0.90 0.36 -
- 0.01 0.00 - 0.08 0.02 0.02 0.18 - 0.20 - 0.18 - 0.14 - 0.22 - 0.18 - 0.17 - 0.23 - 0.06 - 0.15 0.02 - 0.05 - 0.04 - 0.02 - 0.20 - 0.05 0.20
0.13 0.17 0.13 0.11 0.26 0.06 0.48 0.56 0.30 0.10 0.30 0.49 0.43 0.35 0.00 0.22 0.14 0.23 0.16 0.64 0.32 0.71 0.33
0.12 0.08 0.09 0.02 0.27 0.02 0.32 0.25 0.14 0.04 0.04 0.17 0.17 0.05 0.11 0.11 0.06 0.04 0.07 0.54 0.12 0.43 0.04 0.30
0.13 0.13 0.12 0.11 0.12 0.10 0.01 0.01 0.16 0.01 0.07 0.15 0.14 0.12 0.18 0.01 0.15 0.13 0.11 0.04 0.03 0.02 0.09 0.07 0.13
ED ECU 3m AUD 3m
0.43 0.40 0.35 0.36 0.50 0.03 0.15 0.17 0.24 0.04 0.01 0.23 0.21 0.08 0.09 0.01 0.38 0.36 0.41 0.24 0.14 - 0.26 - 0.01 0.09 0.01 - 0.02
0.13 0.12 0.08 0.15 0.15 0.05 0.12 0.03 0.12 0.16 0.01 0.02 0.09 0.03 0.19 0.20 0.15 0.10 0.10 0.06 0.01 0.15 0.02 0.10 0.04 0.04 0.03 -
0.98 0.91 0.88 0.48 0.98 0.92 0.88 0.47 0.91 0.92 0.81 0.41 0.88 0.88 0.81 0.58 0.48 0.47 0.41 0.58 0.51 0.50 0.47 0.50 0.25 0.19 0.21 0.16 0.28 0.23 0.11 0.13 0.08 0.13 0.27 0.38 0.37 0.28 0.25 0.18 0.20 0.19 0.28 0.17 0.00 0.19 0.19 0.13 0.24 0.22 0.51 0.52 0.41 0.44 0.39 0.55 0.56 0.50 0.51 0.40 0.48 0.48 0.37 0.38 0.32 0.37 0.34 0.34 0.24 - 0.12 0.31 0.29 0.24 0.25 - 0.07 0.95 0.94 0.86 0.90 0.45 0.92 0.91 0.83 0.87 0.48 0.98 0.98 0.92 0.90 0.51 0.23 0.25 0.24 0.26 0.32 0.01 0.03 - 0.00 - 0.09 0.04 0.24 - 0.25 - 0.22 - 0.27 - 0.36 0.01 0.00 - 0.08 0.02 0.02 0.13 0.17 0.13 0.11 0.26 0.12 0.08 0.09 - 0.02 - 0.27 0.13 0.13 0.12 0.11 - 0.12 0.43 0.40 0.35 0.36 0.50 0.13 - 0.12 - 0.08 - 0.15 0.15 0.29 - 0.28 - 0.24 - 0.35 - 0.23 0.14 0.16 0.09 0.17 0.12 0.24 - 0.03 0.17 0.27 0.21 0.28 0.21 0.35 - 0.06 0.19 0.47 0.52 0.56 0.21 - 0.14 - 0.10 0.18 0.06 - 0.02 - 0.10 0.03 0.05 - 0.17 0.06 0.72 0.29 0.24 0.54 0.33 0.30 0.11 0.03 0.16 0.23 0.34 0.22 0.51 0.13 0.44 0.20 0.48 0.32 0.01 0.15 0.12 0.30 0.08 0.43 0.22 0.57 0.32 0.33 0.20 0.06 0.09 0.15 0.23 0.14 0.47 0.11 0.49 0.18 0.56 0.25 0.01 0.17 0.03 0.18 0.04 0.24 0.37 0.64 0.49 0.66 0.29 0.15 0.43 0.54 0.41 0.24 0.20 - 0.19 - 0.14 0.30 0.14 0.16 0.24 0.12 - 0.14 0.11 0.32 0.01 0.12 0.06 0.30 0.21 0.26 0.24 0.25 0.00 0.27 0.05 0.22 0.10 0.04 0.01 0.04 0.16 0.09 0.11 0.22 0.24 0.14 0.04 0.09 0.19 0.29 0.21 0.18 0.19 0.21 0.18 0.30 0.04 0.07 0.01 0.01 0.03 0.05 0.70 0.84 0.35 0.20 0.59 0.64 0.56 0.45 0.18 0.43 0.17 0.49 0.17 0.15 0.23 0.02 0.27 0.27 0.67 0.36 0.34 0.64 0.67 0.60 0.37 0.13 0.34 0.23 0.43 0.17 0.14 0.21 0.09 0.14 0.23 0.34 0.32 0.56 0.60 0.54 0.33 0.33 0.30 0.06 0.35 0.05 0.12 0.08 0.03 0.10 0.12 0.67 0.55 0.43 0.40 0.06 0.05 - 0.09 - 0.15 0.00 0.11 0.18 0.09 - 0.19 0.01 0.09 0.51 0.41 0.38 0.09 0.04 0.11 0.02 0.22 0.11 0.01 0.01 0.20 0.09 0.10 0.96 0.97 0.28 0.01 0.27 0.05 0.14 0.06 0.15 0.38 0.15 0.30 0.15 0.95 0.35 0.04 0.31 0.04 0.23 0.04 0.13 0.36 0.10 0.29 0.14 0.27 0.00 0.26 0.02 0.16 0.07 0.11 0.41 0.10 0.30 0.15 0.43 0.90 0.20 0.64 0.54 0.04 0.24 0.06 0.23 0.06 - 0.36 - 0.05 0.20 0.32 - 0.71 - 0.12 0.43 - 0.03 0.02 0.14 - 0.26 - 0.01 0.15 0.19 0.18 - 0.19 - 0.08 0.33 0.04 0.09 0.01 0.02 0.04 0.02 - 0.30 0.07 0.09 0.10 - 0.08 0.16 - 0.13 0.01 - 0.04 0.21 - 0.29
- 0.02 0.04 - 0.03 - 0.26 - 0.29 - 0.08 0.43 - 0.01 0.05 - 0.35
- 0.29 - 0.28 - 0.24 - 0.35 - 0.23 - 0.17 - 0.30 - 0.18 - 0.14 0.09 - 0.03 - 0.27 - 0.14 - 0.10 0.01 - 0.09 - 0.30 - 0.29 - 0.30 - 0.23 0.19 0.18 0.04 - 0.08 0.21 - 0.26 - 0.29 - 0.08
0.14 0.16 0.09 0.17 0.12 0.06 0.08 0.04 0.11 0.11 0.05 0.27 0.23 0.12 0.09 0.10 0.15 0.14 0.15 0.06 0.19 0.08 0.02 0.16 0.29 0.43 0.01 0.05 0.35
This Pearson moment correlation matrix is calculated from the daily returns of the 60 most recent business days data. For most, the first nearby futures contract is used. A roll adjustment is made by back-creating the price series according to the daily return of the prompt contract on the roll date. This avoids severe consequences for those commodities with significant term structure, where rolls introduce spurious jumps, which lower correlations. The shading scheme is as follows: numbers in interval [-0.4, 0.4] are unshaded; numbers in [-0.85, -0.4] and [0.4,0.85] have a light grey background; the highly correlated pairs (with a magnitude >=0.85) are shown with the darker grey.
27 June 2008
Commodities Quarterly
Appendix 1
Important Disclosures Additional information available upon request Special Disclosures
Deutsche Bank AG and/or an affiliate(s) are acting as Corporate Broker to Xstrata Plc Deutsche Bank AG and/or an affiliate(s) are acting as defense advisor to Rio Tinto PLC in relation to a potential offer by BHP Billiton PLC. For disclosures pertaining to recommendations or estimates made on a security mentioned in this report, please see the most recently published company report or visit our global disclosure look-up page on our website at http://gm.db.com.
Analyst Certification
The views expressed in this report accurately reflect the personal views of the undersigned lead analyst(s). In addition, the undersigned lead analyst(s) has not and will not receive any compensation for providing a specific recommendation or view in this report. Michael Lewis
Page 95
27 June 2008
Commodities Quarterly
Additional disclosures required under the laws and regulations of jurisdictions other than the United States
The following disclosures are those required by the jurisdiction indicated, in addition to those already made pursuant to United States laws and regulations. Analyst compensation: Analysts are paid in part based on the profitability of Deutsche Bank AG and its affiliates, which includes investment banking revenues Australia: This research, and any access to it, is intended only for "wholesale clients" within the meaning of the Australian Corporations Act. EU: A general description of how Deutsche Bank AG identifies and manages conflicts of interest in Europe is contained in our public facing policy for managing conflicts of interest in connection with investment research.Disclosures relating to the firm's obligations under MiFiD can be found at http://globalmarkets.db.com/riskdisclosures. Germany: See company-specific disclosures above for holdings of five percent or more of the share capital. In order to prevent or deal with conflicts of interests Deutsche Bank AG has implemented the necessary organisational procedures to comply with legal requirements and regulatory decrees. Adherence to these procedures is monitored by the ComplianceDepartment. Hong Kong: See http://gm.db.com for company-specific disclosures required under Hong Kong regulations in connection with this research report. Disclosure #5 includes an associate of the research analyst. Disclosure #6, satisfies the disclosure of financial interests for the purposes of paragraph 16.5(a) of the SFC's Code of Conduct (the "Code"). The 1% or more interests is calculated as of the previous month end. Disclosures #7 and #8 combined satisfy the SFC requirement under paragraph 16.5(d) of the Code to disclose an investment banking relationship. Japan: See company-specific disclosures as to any applicable disclosures required by Japanese stock exchanges, the Japanese Securities Dealers Association or the Japanese Securities Finance Company. Disclosures under the Financial Instruments and Exchange Law Company name: Deutsche Securities Inc. Registration number: Registered as a financial instruments dealer by the Head of the Kanto Local Finance Bureau (Kinsho) No.117 Member of associations: Japan Securities Dealers Association (JSDA), The Financial Futures Association of Japan This report is not meant to solicit the purchase of specific financial instruments or related services. We may charge commissions and fees for certain categories of investment advice, products and services. Recommended investment strategies, products and services carry the risk of losses to principal and other losses as a result of changes in market and/or economic trends, and/or fluctuations in market value. Before deciding on the purchase of financial products and/or services, customers should carefully read the relevant disclosures, prospectuses and other documentation. New Zealand: This research is not intended for, and should not be given to, "members of the public" within the meaning of the New Zealand Securities Markets Act 1988.
Page 96 Deutsche Bank AG/London
27 June 2008
Commodities Quarterly
Russia: The information, interpretation and opinions submitted herein are not in the context of, and do not constitute, any appraisal or evaluation activity requiring a licence in the Russian Federation. South Africa: Publisher: Deutsche Securities (Pty) Ltd, 3 Exchange Square, 87 Maude Street, Sandton, 2196, South Africa. Author: As referred to on the front cover. All rights reserved. When quoting, please cite Deutsche Securities Research as the source. Turkey: The information, interpretation and advice submitted herein are not in the context of an investment consultancy service. Investment consultancy services are provided by brokerage firms, portfolio management companies and banks that are not authorized to accept deposits through an investment consultancy agreement to be entered into such corporations and their clients. The interpretation and advices herein are submitted on the basis of personal opinion of the relevant interpreters and consultants. Such opinion may not fit your financial situation and your profit/risk preferences. Accordingly, investment decisions solely based on the information herein may not result in expected outcomes. United Kingdom: Persons who would be categorized as private customers in the United Kingdom, as such term is defined in the rules of the Financial Services Authority, should read this research in conjunction with prior Deutsche Bank AG research on the companies which are the subject of this research. Disclosures relating to the firm's obligations under MiFiD can be found at http://globalmarkets.db.com/riskdisclosures.
Page 97
David Folkerts-Landau
Managing Director Global Head of Research Global Company Research Stuart Parkinson Chief Operating Officer Europe Pascal Costantini Regional Head Principal Locations
Deutsche Bank AG London 1 Great Winchester Street London EC2N 2EQ Tel: (44) 20 7545 8000 Deutsche Bank AG New York 60 Wall Street New York, NY 10005 United States of America Tel: (1) 212 250-2500 Deutsche Bank AG Boston 225 Franklin Street Boston, MA 02110 United States of America Tel: (1) 617 988 8600 Deutsche Bank AG Hong Kong Cheung Kong Center, 2 Queens Road Central Hong Kong Tel: (52) 2203 8888 Deutsche Bank AG Singapore One Raffles Quay South Tower Singapore 048583 Tel: (65) 6423 8001 Deutsche Bank AG Japan 2-11-1 Nagatacho Sanno Park Tower Chiyoda-ku, Tokyo 100-6171 Tel: (81) 3 5156 6701 Deutsche Bank AG Australia Deutsche Bank Place, Level 16 Corner of Hunter & Philip Streets Sydney NSW 2000 Tel: (61) 2 8258 1234
Global Fixed Income Strategies & Economics Marcel Cassard Global Head
Deutsche Bank AG Frankfurt Grosse Gallusstrasse 10-14 60272 Frankfurt am Main Germany Tel: (49) 69 910 0 Subscribers to research via email receive their electronic publication on average 1-2 working days earlier than the printed version. If you would like to receive this or any other product via email please contact your usual Deutsche Bank representative. Publication Address: Deutsche Bank AG London 1 Great Winchester Street London EC2N 2EQ United Kingdom (44) 20 7545 8000 Internet: http://gmr.db.com Ask your usual contact for a username and password.
Global Disclaimer
The information and opinions in this report were prepared by Deutsche Bank AG or one of its affiliates (collectively Deutsche Bank)for its clients. The information herein is believed by Deutsche Bank to be reliable and has been obtained from public sources believed to be reliable. With the exception of information about Deutsche Bank, Deutsche Bank makes no representation as to the accuracy or completeness of such information. This published research report may be considered by Deutsche Bank when Deutsche Bank is deciding to buy or sell proprietary positions in the securities mentioned in this report. For select companies, Deutsche Bank equity research analysts may identify shorter-term opportunities that are consistent or inconsistent with Deutsche Bank's existing, longer-term Buy or Sell recommendations. This information is made available on the SOLAR stock list, which can be found at http://gm.db.com. Deutsche Bank may trade for its own account as a result of the short term trading suggestions of analysts and may also engage in securities transactions in a manner inconsistent with this research report and with respect to securities covered by this report, will sell to or buy from customers on a principal basis. Disclosures of conflicts of interest, if any, are discussed at the end of the text of this report or on the Deutsche Bank website at http://gm.db.com. Opinions, estimates and projections in this report constitute the current judgement of the author as of the date of this report. They do not necessarily reflect the opinions of Deutsche Bank and are subject to change without notice. Deutsche Bank has no obligation to update, modify or amend this report or to otherwise notify a reader thereof in the event that any matter stated herein, or any opinion, projection, forecast or estimate set forth herein, changes or subsequently becomes inaccurate, except if research on the subject company is withdrawn. Prices and availability of financial instruments also are subject to change without notice. This report is provided for informational purposes only. It is not to be construed as an offer to buy or sell or a solicitation of an offer to buy or sell any financial instruments or to participate in any particular trading strategy in any jurisdiction or as an advertisement of any financial instruments. The financial instruments discussed in this report may not be suitable for all investors and investors must make their own investment decisions using their own independent advisors as they believe necessary and based upon their specific financial situations and investment objectives. If a financial instrument is denominated in a currency other than an investors currency, a change in exchange rates may adversely affect the price or value of, or the income derived from, the financial instrument, and such investor effectively assumes currency risk. In addition, income from an investment may fluctuate and the price or value of financial instruments described in this report, either directly or indirectly, may rise or fall. Furthermore, past performance is not necessarily indicative of future results. Derivative transactions involve numerous risks including, among others, market, counterparty default and illiquidity risk. The appropriateness or otherwise of these products for use by investors is dependent on the investors' own circumstances including their tax position, their regulatory environment and the nature of their other assets and liabilities and as such investors should take expert legal and financial advice before entering into any transaction similar to or inspired by the contents of this publication. Trading in options involves risk and is not suitable for all investors. Prior to buying or selling an option investors must review the "Characteristics and Risks of Standardized Options," at http://www.optionsclearing.com/publications/risks/riskchap1.jsp. If you are unable to access the website please contact Deutsche Bank AG at +1 (212) 250-7994, for a copy of this important document. Furthermore, past performance is not necessarily indicative of future results. Please note that multi-leg options strategies will incur multiple commissions. Unless governing law provides otherwise, all transactions should be executed through the Deutsche Bank entity in the investors home jurisdiction . In the U.S. this report is approved and/or distributed by Deutsche Bank Securities Inc., a member of the NYSE, the NASD, NFA and SIPC. In Germany this report is approved and/or communicated by Deutsche Bank AG Frankfurt authorised by Bundesanstalt fr Finanzdienstleistungsaufsicht. In the United Kingdom this report is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange and regulated by the Financial Services Authority for the conduct of investment business in the UK and authorised by Bundesanstalt fr Finanzdienstleistungsaufsicht (BaFin). This report is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. and in Singapore by Deutsche Bank AG, Singapore Branch. In Japan this report is approved and/or distributed by Deutsche Securities Inc. The information contained in this report does not constitute the provision of investment advice. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS) relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product. Deutsche Bank AG Johannesburg is incorporated in the Federal Republic of Germany (Branch Register Number in South Africa: 1998/003298/10) Additional information relative to securities, other financial products or issuers discussed in this report is available upon request. This report may not be reproduced, distributed or published by any person for any purpose without Deutsche Bank's prior written consent. Please cite source when quoting. Copyright 2008 Deutsche Bank AG