7/21/2019 Barclay's Commodity http://slidepdf.com/reader/full/barclays-commodity 1/216 BARCAP_RESEARCH_TAG_FONDMI2NBUR7SWED Please read carefully the important disclosures at the end of this publication.The Commodity Refiner Commodities Research Q1 2006 Seismic shift New forces are shaping the commodity markets. After a long period of retrenchment, very few natural resource companies possess either the opportunities or capabilities to swiftly raise their output to keep abreast of the sustained move up in demand growth that China has brought. The global economy is proving highly resilient, there is a lack of low-cost competing materials to provide substitutes, and the overhang of excess capacity and inventory seen in most energy and metals markets during the 1980s and 1990s is unlikely to return. Commodities will not continue moving up indefinitely, but we believe average price levels will stay high for a long time to come. In 2006, we expect a slowing of the broad upward trend in commodity prices, a more diverse performance across different markets and the potential for much greater levels of volatility. Crude oil is expected to hit fresh historical highs, with fundamental balances implying no imminent reduction in upstream tightness and refinery capacity shortages getting worse. The dynamics of base metals are strong moving into 2006, with further severe supply- side tightness likely to become manifest. We expect the upward price trends to persist, with zinc still our top pick. We are unconvinced regarding the sustainability of the rise in precious metals prices, however, and prefer to reduce exposure to them. Prices in the agricultural sector have the potential to continue moving up modestly from current low levels, though performance will vary markedly across the sector, with sugar likely to outperform, in our opinion. George Hopley[email protected]+1 212 412 2079 Paul Horsnell[email protected]+44 (0)20 7773 1145 Kevin Norrish[email protected]+44 (0)20 7773 0369 Ingrid Sternby[email protected]+44 (0)20 7773 7034 Sudakshina Unnikrishnansudakshina.unnikrishnan @barcap.com +44 (0)20 7773 3797 Yingxi Yu [email protected]+44 (0)20 7773 1336 www.barcap.com
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
We remain positive on the commodity sector as a whole, but have varying degrees of
enthusiasm across the sub-sectors. We are unconvinced of the sustainability of the
rise in precious metals prices, and prefer to reduce exposure to them. Prices in theagricultural sector possess the potential to continue moving up modestly from their
current low levels, though performance will vary markedly across the sector, with
sugar likely to be the outperformer. The dynamics of base metals are strong moving
into 2006, with further severe supply-side tightness becoming manifest. We expect
the upward price trends to persist, with zinc still our top pick. It is our belief that oil
prices will remain well supported, with fundamental balances implying no imminent
reduction in upstream tightness. We also expect an increase in downstream
dislocations, particularly in deep conversion capacity. Natural gas markets in the US
have eased and could weaken further in the short term, but are likely to tighten as
winter demand kicks in during Q4 06.
There are three very distinct commodities stories in operation currently. The first is the
continued capacity constraints within industries that exhibit long supply lags, and
where investment has been distinctly tardy. This story covers most of the base metals
and also energy, together indirectly with a few agricultural commodities that have links
with the energy market. It results in price dynamics that combine long-lasting grinding
upward trends with the possibility of occasional episodes of violent upwards spikes. The
key factor across these commodities is that of friction, with demand pushing up against
capacity that has proved to be too low relative to the demands of non-OECD growth, in
particular.
The second commodities story is evidenced in some industries where there are no
binding supply-side constraints, and where inventories are often a significant multipleof output flows. In some of these sectors, price dynamics are being primarily
determined by the flow of funds into the market, and in particular by speculative
money moving into a commodity on the basis of what can sometimes be fairly flimsy
fundamental stories. That story is at the moment being reflected in the behaviour of
precious metals prices. The third, and final, story is that of those commodities with
relatively fast supply-side responses, where the fundamental balances can often be
weak, and where upward price moves tend to be corrected within months, rather than
grinding on for years, as in the first story. This is the case for most agricultural
commodities.
Our general bias is that we like exposure to the first story, where we expect somefurther upside in some commodities, and rough stability with upward price risks in
others. We are, by contrast, keen to avoid heavy exposure to the other stories, which
can, in our view, produce price dynamics of extreme fragility. With some exceptions
within categories, that leads us to prefer to have exposure to base metals and energy,
and to prefer to reduce exposure to precious metals and most agricultural
commodities. That bias towards the market going forward has also been the one that
has been the most successful over the past year, as is shown in Figure 5. In terms of
absolute price changes, energy and base metals have been the strongest sectors, in that
order. In total return terms, it has been the same sectors that have been strongest, but
in the reverse order, with energy held back by a time structure of prices that has been
leave us enthusiastic about the prospects for commodities in general, there are some
differences in our approach to the individual sub-sectors and in our relative degree of
enthusiasm across those sub-sectors, as is explained in further detail below.
Agricultural commodities
Agricultural commodity prices have, in general, failed to make much headway over the
past year. Most, but not all, agricultural commodities lack the underlying structure that
has been so positive for base metals and energy, namely the appearance of capacity
constraints in coexistence with long lead times for incremental investment. As was
shown in Figure 5, agricultural commodities (including livestock) have made up all, bar
one, of the select group of commodities which have experienced a Y/Y fall, and they
make up 18 of the 20 commodities with the worst Y/Y price performance.
Figure 6: Strong fundamentals continue to support
sugar prices (US cents/lb)
0
2
4
6
8
10
12
14
1618
20
01 02 03 04 05 06
Source: EcoWin.
Figure 7: Corn prices have weakened with negative
sentiment (US cents/56 lbs bushel)
180
200
220
240
260
280
300
320
340
01 02 03 04 05 06
Source: EcoWin.
Within the sector, price increases of 25% or more have come in three groups. Firstly,
there are the Asian rubber contracts with their underlying industrial demand base. As
oil prices have risen, so has the cost of producing synthetic rubber, and this has helped
to pull up the price of natural rubber. The second group is comprised of two smaller US
futures contracts, both exhibiting relatively limited volumes, namely oats and orange
juice (prices for the latter have been pushed up by hurricane damage to the citrus
industry in Florida). Finally, there is Robusta (but not Arabica) coffee and sugar, the
best performer in dollar terms among agricultural commodities.
Despite the already strong rise in sugar prices, we believe it still has the best and mostpersuasive underlying fundamental story among the agricultural commodities. Demand
remains strong; the supply response thus far has been fairly lacklustre in comparison. On US
Department of Agriculture (USDA) projections, global inventory cover for the end of the
current season is placed at 11.5 weeks of demand, three weeks fewer than at the end of the
2003/4 season. Sugar prices are also being supported by the link, via ethanol, to the energy
market, and thus, like rubber, sugar has been partly pulled up by the path of oil prices. In
particular, ethanol production out of sugar for vehicle fuels has risen very sharply in Brazil.
This has had a strong enough effect on the physical market, and also on market sentiment,
as to create a partial but continuing link to oil prices. With both the underlying balances of
the market and investor sentiment remaining strong, we expect sugar prices to achieve
further highs.
While sugar may be benefiting from positive sentiment, a broad swathe of cereal crops
has been affected by a raft of negative sentiment, brought on, most specifically, by
investor unease about the potential impact of avian flu on the feedstock sector. Like
sugar, the US corn market should be positively supported by the link to ethanol and bio-
fuels, but any such effect has been swamped by weak short-term fundamental balances
and the negative turn in sentiment, bringing prices close to their lows for the decade.
Similar concerns have also affected wheat, the soybean complex and other oilseeds.
While livestock prices have recovered from major reverses earlier in 2005, neither the
physical market balances nor the recent price dynamics suggest that any significant andsustainable move forward is likely. Within the agricultural sub-sector we would be
overweight those contracts with a strong link to the energy market (namely sugar and
natural rubber), and we would continue to be underweight oilseeds and cereals.
Precious metals
Precious metals prices have continued to push upwards, in what is perhaps best described
as a sustained bout of market exuberance. Gold has pushed through $550/oz, platinum
through $1000/oz and silver through $9/oz. These rises have come despite a lack of
concerted dollar weakness. Given the breaching of key psychological levels and the
associated triggering of a series of positive technical conditions, it is perhaps hard to standin the way of the strong move up. Such has been the scale of the flow of money into the
sector, that there does of course remain some scope for further short-term gains. However,
looked at over a quarterly horizon and beyond, we have doubts as to the sustainability of
the upwards move. We believe that the rally in prices is now looking stretched and has a
distinct fragility; as a result, we are underweight precious metals within the commodities
sector as a whole.
There are two main negative dynamics for precious metals: the first is the
concentration of speculative interest in futures markets, which appears to us to now be
becoming overextended. The strength of the flow of speculative funds into precious
metals has been the primary trigger of the recent decoupling of prices from the path of
the dollar. While upwards price momentum remains, those funds are likely to stay
heavily invested. However, once the rally stalls, the sheer size of speculative length
suggests that the downwards correction might be swift and brutal. The second main
negative is the generally rather weak fundamental picture. In gold, for example, higher
prices have caused Asian demand to slow markedly and increased the level of recycling.
We remain unconvinced about stories implying imminent buying of gold from major
central banks. In practice, the ability of central banks to purchase significant quantities
of gold is severely constrained by limited availability in relation to those reserves, and
we do not see any significant role for gold as an instrument of central bank reserve
Figure 8: Gold has continued to push sharply higher
despite recent dollar strength (US$/oz)
250
300
350
400
450
500
550
600
01 02 03 04 05 06
Source: EcoWin.
Figure 9: Platinum has tracked on to $1000/oz, and
even palladium has turned upwards (US$/oz)
0
200
400
600
800
1,000
1,200
01 02 03 04 05 06
Platinum
Palladium
Source: EcoWin.
We suspect that the path of the Japanese economy may also become a critical variablein the medium term. Much of the heaviest buying of precious metals has been seen on
the Tokyo exchange, and seems to be reflective of some aversion from Japanese
investors to domestic yen-based assets. If the path of the Japanese recovery follows our
base case, we would expect that aversion to fall, and reduce a source of buying pressure
in the market.
Given that the overall balance of risks is, in our view, becoming skewed to the downside,
we would wish to reduce exposure to precious metals, except as part of a very short-
term trading position. However, given the positive technical price trends in place and
the evident enthusiasm of many investors to purchase gold even at current elevated
price levels we certainly would not advocate an overt short position in this market.
Base metals
The rise in base metals prices, and most particularly copper, has been attributed to a
variety of factors in recent months. These have included speculative funds and the
consequences of failed short positions taken by key market participants. We see no
reason to ascribe the main source of strength to forces from the shadows, when the
fundamental basis of the three-year long move up remains so strong. It is our view that
price behaviour is being driven primarily by binding supply-side constraints across the
metals sector, which have resulted in falling inventory cover in the face of continued
strength from the demand side. Costs are rising, and the supply-side response has been
made more lacklustre by shortages of equipment and, in some cases, by a lack of
sufficient incremental resource prospects.
Price rises in base metals over the past year have been led by copper and zinc, both of
which have risen by more than 50% Y/Y in London, as was shown in Figure 5. We
expect the move up in both metals to continue into H1 06, setting fresh multi-year
highs, with some extreme supply tightness emerging in the face of robust demand.
Given our base macroeconomic forecasts, particularly for global manufacturing and for
China, we expect the demand side to remain vibrant across the sector. In the case of
copper, while there are some price-related drawbacks in demand in process, the rapid
expansion of the Chinese power supply network is helping to buttress demand. Further,
despite the price impact, copper demand is ending the year in stronger shape than it
did the first half of the year, partly due to the timing of consumer de-stocking. In thecase of both copper and zinc, we continue to recommend buying into any short-term
price weakness. While we expect the cycle to show signs of turning into H2 06, there is
still a final leg up to be ridden over the coming quarter.
Figure 11: After bottoming out, nickel prices appear
poised for a recovery (US$/tonne)
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
20,000
02 03 04 05
Source: Reuters, EMC, Barclays Capital.
While copper and zinc have been the swans of the base metals world over the past year,we believe that there are now reasons for interest in some of the other base metals. In
particular, we recommend exposure to nickel, where prices now appear to have
bottomed out after weakening across the first three quarters of the year. In our view,
nickel prices are showing a significant degree of upside potential. Producer sentiment
has turned less bearish, and availability has started to show some tightness at European
steelmakers. We believe that there is already a considerable degree of bad news
contained in the nickel price, and the tightening of balances we expect in early 2006
should start to chip away at the current surplus.
Aluminium market fundamentals are also strong, the result of a lack of availability of
raw materials, notably alumina, and escalating power costs that have slowed primary
capacity growth and resulted in a number of recent smelter closures. While we see the
potential for temporary pull-backs in prices, the extremely positive long-term
fundamentals suggest further significant upside potential in 2006.
Given the current tightness across most of the base metals, and the increasing
sensitivity of prices to supply-side disruptions, we are maintaining an overweight
position on base metals relative to the commodities sector as a whole. In addition, we
also believe that further forward portions of the base metals time curve currently
represent better value than the equivalent positions in other commodities. In terms of
prompt prices, the overweight position is compelling for Q1 and into Q2. However, our
current view of the timing for the top of cycle is such that we would be more neutral,
and indeed tend towards an underweight view, when it comes to positioning in a
commodities portfolio for H2 06.
Energy
There are currently two striking dynamics within energy markets. The first is the sharp
rise in the volatility of natural gas prices relative to oil prices. On both sides of the
Atlantic, natural gas markets have demonstrated a seasonal propensity to extreme
sensitivity to weather conditions. We would see this as further evidence of extremely
stressed logistical systems that sometimes can only find short-term physical or
psychological balance by creating a sudden swathe of demand destruction through
extreme price behaviour. The UK and US natural gas markets have already movedthrough one such extreme short-term cycle.
The second key dynamic is within the oil sector, and concerns the extent to which price
signals are having to do work that used to be achieved by physical capital. The drying
up of spare capacity throughout the supply-chain, (including the input chains), and
most profoundly the tightness in deep-conversion capacity within refining, means that
most price differentials have become significantly more volatile, be they differentials
across time, place or quality.
Figure 12: US natural gas prices have moved higher and
become more volatile (US$/mmbtu)
0
24
6
8
10
12
14
16
01 02 03 04 05 06
Source: EcoWin.
Figure 13: UK natural gas prices have been through an
extreme price spike (UK pence/therm)
0
20
40
60
80
100
120
01 02 03 04 05 06
Source: EcoWin.
US natural gas markets have fallen back significantly from their peaks of early
December 2005 but short-term dynamics and sensitivity to weather patterns mean that
there is still some potential short-term upside. One of the reasons we gave for
exercising caution on precious metals prices was the massing of speculative positions
on the long side of the market. By contrast, speculative funds have been overall net
short of natural gas since April 2005 and although the position has been trimmedrecently, the balance of speculative interest is still heavily short. We would expect to see
Q2 and Q3 prices ease back further from current levels, but there is the potential for
another big move up to the $10 per mmbtu region as winter 2006 draws near.
UK natural gas prices are currently settling back after an extreme spike due to the
supply tightness revealed by an earlier cold snap. The spike has demonstrated the
practical and political difficulty of moving incremental supply from the rest of Europe
into the UK market, the infrastructural constraints within the logistical system, and have
highlighted the consequences of the UK’s sharp move towards net natural gas importer
status given the rapid decline of domestic output.
Figure 14: WTI prices have moved back up close to their
hurricane-related peak (US$/barrel)
10
20
30
40
50
60
70
01 02 03 04 05 06
Source: EcoWin.
Figure 15: US retail gasoline prices have moved well
below $3 (regular unleaded average, US cents/gallon)
140
160
180
200
220
240
260
280
300
320
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
200620052004
2003
Source: US Energy Information Administration.
We expect that the key characteristic of oil prices next year will be an enhanced
volatility of the main benchmarks and of price differentials across place, time and
quality. In terms of absolute price levels, we project a modest Y/Y increase. The 2006
average for West Texas Intermediate crude oil is forecast to be $68 per barrel, an
increase of about $11 from the 2005 average, and a deceleration in the pace of increase
after the $15 average rise seen in 2005. In our view, the causes of price volatility in
2005 should continue into 2006, with spare capacity remaining close to current levels
in crude oil, while we expect tightness in oil products to intensify.
The implication of our projected oil market balances for 2006 is that the market is
entering a period of holding station in terms of the degree of the underlying upstream
tightness. In particular, we expect that the increased market call on OPEC output will
roughly match the increase in OPEC’s production capacity, leaving the degree of global
spare capacity little changed over the next year. It would require a period of significant
demand weakness or strength in non-OPEC supply to start to rebuild the buffer of
global spare capacity, and we do not expect either of these dynamics to occur in the
coming year. On the demand side, our strong outlook for global growth detailed
elsewhere in this report is accompanied by a belief that demand will also remain robust
based on strong income effects. Such growth would be supported by a continued
deceleration in the Y/Y increase in prices thereby continuing a process we have
witnessed in recent years, in which the short- and medium-term importance of price
elasticities has been side-lined relative to the role of income elasticities.We currently expect that global oil demand will average 85.4 mb/d in 2006, which
would represent an annual growth of just over 1.5 mb/d (1.8%). That would make 2006
the fourth year in a row of strong global demand growth, and bring the cumulative
increase in demand over those four years up to 7.7 mb. The supply side has not even
come close to matching that pace, which has resulted in a very sharp reduction in spare
capacity, making the sustained nature of the strong demand increase the key driver of
the market, together with the relative torpidity on the supply-side.
The strong increase in global demand growth has caught non-OPEC supplies ratherflat-footed and demonstrated that it is prone to structural weaknesses. We expect 2006
to be a slightly better year for non-OPEC growth than 2005, and estimate this to be just
over 0.5 mb/d (1%). Non-OPEC supply growth is then expected to fall some 1 mb/d shy
of the increase in global demand. That would bring the total increase in non-OPEC
supply over four years up to 2.2 mb/d, the same four years over which we expect
demand to have grown by 7.7 mb/d.
The onus for filling that 5.5 mb/d gap has fallen on OPEC, and that has been the brutal
mathematics which has reduced supply cover to such wafer thin levels. We expect all of
the net increase in non-OPEC supplies over the four years to 2006 to come from the
Former Soviet Union (FSU), and Russia in particular. The growth of 0.46 mb/d we arecurrently projecting for non-OPEC outside the FSU simply cancels out the fall in 2005,
and would bring the total growth from 2003 to 2006 to just below zero. There has been
growth in several key areas of non-OPEC, and in particular Canada, Brazil, Angola and
Chad. However, the net increments from the areas of growth have been cancelled out
by declines from some mature non-OPEC producers. The UK has been a source of
particular weakness, with output having dropped there by about 1 mb/d this decade. In
2005, the weakness in the UK was compounded by steep falls in Norway and in the US,
with the fall in the latter compounded by the temporary effects of the hurricanes. In all,
2005 proved to be the worst year this decade for non-OPEC supply growth.
The implications of the poor supply performance in 2005 outside of the FSU were
heightened by a sharp slowing in the rate of growth of Russian output. Throughout this
decade, Russia has been the major engine of growth within non-OPEC, and at points grew
at Y/Y rates that were only a little shy of 1 mb/d. However, in 2005 Russian growth slowed
to just above 0.2 mb/d, with the drag brought about by a combination of fiscal changes,
technical issues in the mature fields, lack of investment in new producing areas, and the fall-
out from the Yukos affair. We expect the balance of growth within the FSU to shift away
from Russia and towards the Caspian states in coming years, but the extent to which
Russian growth stabilises in 2006 is likely to be one of the key factors on the supply side of
the market.
We expect the degree of upstream tightness to remain unchanged through the next
year. However, a relative dearth of new refinery projects is likely to increase the degreeof tightness in downstream. While there is currently no binding shortage of short-run
distillation capacity, there is a shortage of cracking and other deep conversion capacity.
The market has reacted to that shortage by increasing the absolute size and the
volatility of the key differentials between crude oil and products and between oil
products. Adjustments that used to be accomplished by relatively small price changes
and by the use of physical capital within the refining sector are now requiring far
greater price responses due to the tightness in conversion. Indeed, we expect that the
general theme of a transfer from output response to price response due to deficiencies
in the flexibility of the supply system is likely to be a key one for the market in coming
years. Indeed, short of a period of unexpected demand-side weakness, the long supply
side lags and the impediments to the acceleration of investment profiles, mean that the
sub-optimality of the capital base within the oil industry is unlikely to be corrected
before the end of the decade.
With both producers and consumers having become more comfortable with prices in
the rough range of $50-60, we expect that the coming year will help cement the
perception that there has been a structural shift in oil pricing over the course of the
current decade. For quite some time, the conventional wisdom was that oil prices were
unsustainable above $20; this period finally appears to be over, with the acceptance
growing that long-term prices at these levels generated too little investment relative to
the needs of longer-term demand. While there are no floors to the oil price in
operation, we believe that longer-term averages above $50 per barrel should now be
considered as the base case, particularly given the demonstration in recent years that$50 oil can coexist with strong global economic growth.
A long period of underinvestment in natural resource exploration and development
combined with a lack of skilled manpower with which to quickly rectify the situation (the
result of drastic cost-cutting during the previous decade), plus a severe shortage of
materials and equipment means that the medium-term picture for supply growth in many
markets is tightly constrained. This has left the supply side of the industry struggling to
catch up with accelerating demand (mainly due to China and India), a situation that given
the extremely long lead times involved in the highly capital intensive mining and energysectors, is unlikely to be remedied for several years to come.
In theory, what should set the upper limits for commodity prices in such an environment is
the consumers’ ability, or willingness, to pay. In many commodity markets high prices are
playing a balancing role by rising to levels where consumption growth slows or contracts.
However one of the lessons of the past few years is that the global consumer has proved a
lot more “willing to pay” than any economist previously thought possible. The global
economy continues to prove highly resilient to rising commodity prices and at the same
time the price of alternative materials (eg, plastics for copper tubing) has also accelerated
dramatically, limiting the ability of consumers to find substitutes.
Meanwhile, so long as consumers remain prepared to meet even small downward pricecorrections with substantial levels of buying interest in both physical and futures
markets as continues to be the case in early 2006, it is difficult to characterise recent
price developments as simply a speculative bubble. In this type of environment,
commodity price volatility is likely to persist for some time to come, providing trading
opportunities for many different types of investor.
Eventually some combination of rising supply and reduction in demand is likely to restore
better balance to many commodity markets. However we expect fundamentals to remain
tight for a very long time to come and the kind of excess capacity and inventory levels that
characterised many energy and metals markets during the 1980s and 1990s, providing a
buffer against sudden supply shortages or surges in demand, are extremely unlikely to re-emerge.
Consequently, the characterisation of commodity markets as being at the high point of a
boom and bust cycle is wrong in our view. Fundamentals have changed and when prices do
fall back in metals and energy markets, it will be to much higher levels than in the past.
Outlook for investment flows in 2006
After another very strong year for investing in commodities in 2005, significant fresh
inflows are likely in 2006, in our view. We estimate that 2005 saw fresh investments
into commodity indices and structured products of around $25bn-30bn, taking total
commodity investments worldwide to around $75bn-80bn, with the bulk of this in
commodity index tracking investments, but with the structured commodity products
sector also growing rapidly. The continued strong performance of commodity
investments, allied to the desire of many institutional investors to diversify their equity
and fixed income exposures, suggests that commodity investments are likely to
continue growing strongly in 2006. In general, institutions remain under-invested in
commodities and the anecdotal evidence suggests a significant net inflow of fresh funds
has already occurred during the early part of H1 06.
Figure 22: Annual flows into commodity investments
0
10
20
30
40
50
60
70
80
90
1990 1995 2000 2005
Commodity medium term notes
Commodity index swaps
Notional value of funds under manageme nt ($bn)
Source: Barclays Capital estimates.
Figure 23: Monthly flows into commodity mutual funds
-200
0
200
400
600
800
1,000
1,200
1,400
1,600
Jan 03 Aug 03 Mar 04 Oct 04 May 05 Dec 05
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000Monthly real inflows into
commodity linked mutual funds
(US$m, LHS)
Total US commodity linked mutual
fund assets (US$m, RHS)
Source: Bloomberg, Barclays Capital. Note that December 2005 fund
inflow figure is adjusted for dividend payments.
Estimates of the flow of funds into US commodity-linked mutual funds dipped in late
2005 primarily due to adjustments made for dividend payments in the December data.
The amount of new cash flowing into US commodity-linked mutual funds is estimated
at $42m in December 2005, bringing the total inflows in 2005 to almost $5.37bn,
compared with $5.4bn in 2004. Total assets under management almost doubled over
the course of the year rising to a new all-time high of $13.96bn.
Moreover, anecdotal and survey evidence supports the view that investors are likely to
continue raising their exposure to commodities over the next three years, viewingdirect investment in commodities as a long-term strategy designed to diversify their
portfolios. This was one of the key messages that emerged from a survey of US
investors carried out at Barclays Capital’s first annual US Commodity Investor
Conference in NY in late December, attended by more than 70 institutional investors.
The survey showed that roughly two-thirds of the respondents already had some
commodity exposure, but for most this was 5% or less of their total portfolio. However, over
the next three years, the survey results showed that almost 70% of the respondents
expected to increase commodity exposure to 5% or more of their portfolio.
The result was consistent with the audience’s view that large fresh inflows of institutional
investor flows into commodities are likely over the next three years, with 54% expectingfunds under management in the sector to increase from $75bn currently to between $90bn
and $120bn by 2008, and 32% expecting a figure even greater than $120bn by then.
Almost half of the audience is expected to hold their commodity exposure for three years or
longer with most (63%) citing portfolio diversification as the most important factor in the
decision to add commodities to their mix of assets. The major concern of investors was
current high commodity prices (58% of the audience cited this as their number one
concern). For this reason, only 11% of the audience expected to invest in commodity
indices over the next three years, while 68% expected their investments to take the form of
a combination of different strategies, including active management and structured
These same themes of strong performance in price levels across a number of different
commodity sectors were a key element in the impressive returns to those investors
taking out their commodity exposure in the form of commodity baskets in 2005. The
rationale for this type of investment differs significantly to that of an open-ended
financial swap, with returns linked to a commodity index investment. A commodity
basket is often customised to an investor’s requirements with a relatively short
investment horizon (typically three to five years), while the commodities included andtheir weightings are flexible and a trade-off between potential return and capital at risk
is usually part of the deal.
The ability to provide highly leveraged returns in such structures is possible due to the
traditionally downward sloping curves for both price and volatility that many
commodity markets typically display over the two- to five-year time horizon and
further forward. This feature enables relatively cheap far forward options to be
purchased and a return to be generated as those options positions roll up the volatility
curve to maturity. Meanwhile, a portion of the total capital invested is placed in an
interest bearing account, providing a capital guarantee that varies with the risk-reward
trade-off of the specific structure in question.
The leverage contributed by the use of options enables the issuers of such notes to offer
returns based on multiples of the performance of the reference prices of the underlying
commodities themselves (usually a price point on the forward curve), and thus it is not
surprising that many of these types of products have outperformed simple commodity
index investments over the past 12 months. Another feature boosting the strong
performance of structured products relative to commodity index products is that the
upward sloping price curves that exist in most agricultural product futures markets
(plus their relatively lower levels of liquidity further out along the futures curve), mean
that agricultural commodities (which have performed poorly relative to other
commodity markets) are very rarely included in such structures.
Commodity baskets have a big advantage over commodity index investments in that
they are not vulnerable to the negative yield currently attributable to rolling of
positions forward at the front end of many commodity price curves. However, their
performance is still dependent on positive correlations between the specific
commodities included in each basket. More specifically, a commodity basket’s
performance is enhanced by including a selection of commodities whose price
performance has historically not been very closely correlated, but where those
commodities experience a phase of positively correlated price performance after the
basket has been issued. This is because low levels of historical correlation among a
group of commodities usually reduce the overall implied volatility of the basket, making
the basket options cheaper and thus enabling a greater degree of leverage to the actualprice movement of the basket.
levels of correlation between the underlying commodities have been reduced. In part,
this reflects the perception held by many market participants that both long-term
average commodity price levels and potential levels of price volatility have risen
compared to the recent past. Part of the mechanism by which this has taken place has
been via the issuance of commodity baskets themselves and the resulting bidding up of
far forward price levels and implied volatility in the face of a lack of opposing selling
pressure on the part of commodity producers or tactical investors.
In conclusion to this section, we would emphasise the following key points:
Both commodity index investments and commodity baskets provided strong
returns to investors in 2005, with some commodity baskets performing particularly
strongly due to the specific commodities included and the ability to leverage
returns via the use of options.
The positive returns provided by both types of commodity investment strategy
were enhanced by an unusually high level of correlation between many disparate
commodity sectors in 2005.
The existence of negative roll yield so far in early 2006 (only copper out of the 24commodities included in the GSCI is currently trading consistently in
backwardation) suggests that another very strong performance will be required in
order for the commodity index investments to perform as strongly again in 2006.
A move up at the back end of commodity futures price and implied volatility curves
have reduced the advantageous trade-off between principal protection and
leveraged returns offered by commodity basket structures in 2005. Again, for the
type of returns generated by some commodity baskets in 2005 to be repeated big
moves up in reference price levels are required.
From the above, it is clear that a significant driver of commodity returns in 2006 will be
derived from price direction across the broad range of different sectors that investorsare now targeting. In the next section, we review the price outlook for 2006, focusing
specifically on the implications for commodity investors.
Outlook for the main commodity sectors in
2006 and implications for investors
In historical terms, we expect another year of exceptionally high commodity prices in
2006, with prices in most energy, agriculture, industrial and precious metals markets
staying a long way above their long-term average levels. The key themes that we seedriving commodity markets in early 2006 are as follows:
Strong global economic activity: We project global economic growth at 4.3%, in
2006, unchanged from the level currently estimated for 2005 and see little sign of
any significant slowing of growth in the medium term with growth in 2007
currently forecast at 4.2%. The composition of that growth is likely to change
slightly, with the US and China slowing a little and with Europe and Japan
accelerating.
Accelerating commodity demand: In 2005, oil demand growth fell sharply due to
slower growth in China, plus post-hurricane price-related demand losses in the US.
In industrial metals, consumer destocking cut primary metals demand growth. Witha rebound in US and Chinese oil demand likely in 2006 and an end to metals de-
stocking we expect stronger demand growth rates for metals and energy in 2006
and steady demand growth for most agricultural markets.
Favourable movements in currency markets: Market consensus is for a weakening
of the US$ as growth differentials narrow between the US and Europe and Japan.
Historically, there is a negative correlation between the value of the US dollar and
commodity prices. Dollar weakness in 2004 saw additional buying of commodity
futures by US hedge funds as a currency hedge. Within the commodities universe
gold is likely to be the main beneficiary of US dollar weakness.
Rising geopolitical tensions: Early 2005 has seen a cluster of geopolitical events in
key oil exporting countries including Iran, Nigeria and Russia. Any escalation of the
threat to oil supplies that these represent would at the very least provide support to
energy prices and could possibly send them sharply higher. Gold could also benefit
under such a scenario.
Slow growth in supply: The high price of many commodities is failing to stimulate
the sharp increases in supply that many market participants had expected to see in
2005. We see 2006 as another year of relatively slow growth in metals and energy
sectors with supply constrained by a general lack of new opportunities, rising costs
for commodity producers and shortages of equipment and technical expertise.
Agricultural markets do not face the same constraints and provided harvests are
not hampered by bad weather, another year of steady growth looks likely.
Continued vulnerability to shock: Without a significant acceleration in supply
growth, the cushion of spare capacity and inventory will remain very low in energy
and metals sectors. Supply problems or unforeseen surges in demand are likely to
result in large price reactions.
While these factors suggest another very strong year for commodities their interaction
is likely to play out to varying degrees across different commodity markets. While
strong global growth will be positive for commodity demand in general, the importance
of supply constraints varies quite widely across the different commodity sectors.
Energy sector : It is within the energy sector that this particular factor is of most
significance and we see 2006 as a year where growth in crude oil production capacity
will at best only just match demand growth, leaving little growth in flexibility in the
upstream part of the business. In downstream, the refinery part of industry supply
growth is set to fall even further behind demand and the balance between the supply
and demand for products, such as gasoline and heating oil, will get even tighter.
Industrial metals sector : In industrial metals the supply picture is likely to remain one
of extremely slow growth. The major diversified metals producers continue to targetprecious metals and bulk commodities such as coal and iron ore for expansion rather
than the industrial metals where opportunities are more scarce. Although many
markets are expected to move back into better balance between supply and demand,
inventory levels are likely to stay extremely low and demand firm.
Agricultural sector : Agricultural markets are quite different to energy and metals from
a fundamental point of view and we struggle to find the same compelling supply side
constraints. As is usually the case, any supply shortfalls within the agricultural sector
are likely to result from the impact of weather on harvests in different parts of the
world rather than a long-term structural mis-match between supply and demand
growth rates.
Precious metals sector : In precious metals, supply is likely to prove a relatively minor
issue in determining price direction. In the gold and silver markets, fluctuations in
Figure 36: Historical and forecast price changes by
commodity sector
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
2004 2005 2006
Energy
AgricultureIndustrial Metals
Precious Metals
Average
Change in average s pot prices by commodity sector
(Q4 vs. Q4, unweighted)
Source: EcoWin, Barclays Capital.
Figure 37: Historical and forecast price trends in
selected commodity markets
50
100
150
200
250
300
350
400
Q1 2000 Q1 2002 Q1 2004 Q1 2006
Oil
Gold
Industrial metals index
Wheat
Sugar
Selected quarter average energy, a griculture, precious and
industrial metals prices (indexed Q1 2000=100)
F'cast
Changes are derived from average price levels in Q4 of each year.
Figures for 2006 are BarCap forecasts. Source: EcoWin, Barclays
Capital.
We see further upside for metals prices in H1, at a time when a seasonal easing in
fundamentals could see energy prices moving lower. However, in H2 we see the
potential for energy prices to outperform metals as oil moves to fresh all-time highs.
These forecasts are made mainly on the basis of supply/demand fundamentals and as
such carry a degree of risk, with geopolitical factors possessing the potential to push oil
prices significantly higher than our base case forecasts currently project.
This of course highlights the ever-present risk that a spike up in oil prices as the resultof any one of a number of different geopolitical flashpoints around the world could
result in a dramatic slowdown in global growth with important implications for other
commodity assets. Broadly speaking, we would expect the performance of industrial
metals to be hardest hit, with agricultural markets reaction generally neutral and for
gold prices to be buoyed as the result of the metal’s status as a safe haven asset.
Commodity investment strategy in 2006
Commodity index investments have performed extremely well over the past few years.
However, the persistence of a substantial negative roll yield means that positive totalreturns are now heavily reliant on spot price appreciation. Should the negative roll yield
continue at the level evident in 2005, then the GSCI spot index would need to achieve
an appreciation of a further 14% across 2006, simply to maintain a neutral
performance for the GSCI excess return. This does not look likely given our current
forecasts for the individual commodity sub-sectors.
Regarding commodity index investments, we make the following recommendations:
For investors such as pension funds that are seeking long-term diversification of
their portfolios via an investment in commodities, exposure to an index is still a
very good way of achieving this. We would expect that over the long term, overall
returns should remain at close to their historical average levels and that negative
correlations with equities and fixed income investments will persist.
Figure 50: Relative performance of commodity indices
– past 12 months
Average weekly returns annualised
0%
5%
10%
15%
20%
25%
GSCI RICI DBLCI CRB DJ-AIG DBLCI-
MR
Source: EcoWin, Barclays Capital.
Figure 51: Relative performance of commodity indices
– past five years (indexed to Dec 2000)
60
80
100
120
140
160
180
200
220
Dec 00 Oct 01 Aug 02 Jun 03 Apr 04 Feb 05 Dec 05
Rogers International Commodity Index
DB Liquid Commod Index
DB Liquid Commodity Index (MR)
DJ-AIG
GSCI TR
Reuters/Jefferies C RB
Source: EcoWin, Barclays Capital.
Figure 52: Comparative performance of commodity indices over various time periods
Weekly returns (annualised):
Last 12 months Last 5 years Last 10 years Start of data
GSCI TR 23.4% 12.8% 12.1% 13.2%
GSCI Energy TR 26.9% 18.6% 21.1% 17.4%
GSCI Industrial Metals TR 41.9% 16.3% 8.2% 11.1%
GSCI Precious Metals TR 27.3% 15.0% 5.7% 8.6%
GSCI Agriculture TR 7.7% -4.0% -6.1% 6.3%
GSCI Livestock TR -2.6% 2.7% 1.9% 11.5%
DJ-AIG Composite Index TR 19.4% 11.1% 9.7% 8.3%
DJ-AIG Energy TR 29.2% 20.4% 24.3% 17.7%
DJ-AIG Industrial Metals TR 39.7% 16.2% 9.3% 7.5%
DJ-AIG Precious Metals TR 28.0% 15.2% 5.5% 4.7%
DJ-AIG Agriculture TR 6.6% 1.2% -2.0% 1.5%
DJ-AIG Livestock TR -5.7% 1.6% 1.4% 2.6%
Reuters/Jefferies CRB TR 20.2% 8.4% 5.3% 5.0%
Deutsche Bank DBLCI TR 20.5% 16.8% 16.1% 14.7%
Deutsche Bank DBLCI-MR TR 8.3% 13.8% 12.8% 13.9%
Rogers International Index 21.8% 16.5% na 17.8%
JPM Govt Bond Index TR 3.1% 5.6% 6.1% 7.5%
S&P 500 Composite Index TR 8.5% 2.1% 10.5% 12.2%
Note: Weekly changes based on Friday 20 Jan close, except for the Rogers International Index where the latest data available is as of Wednesday 19
Jan. The respective dates for the start of our data are: GSCI TR, GSCI Agriculture, GSCI Livestock from Jan 1970; GSCI Energy from Dec 1982; GSCI
Industrial metals from Jan 1977; GSCI Precious metals from Jan 1973; DJ-AIG Composite Index and all sub-indices from Jan 1991; Reuters CRB Index
from Jan 1982; DBLCI and DBLCI-MR from Dec 1988; Rogers International Index from Jul 1998; JPM Govt Bond Index from Jan 1986; S&P 500Composite Index from Jan 1989.
Oil market overview In 2006, we expect the annual average of oil prices to increase for the fifth successive
year. We currently project that WTI will average $68 per barrel for the year, with
upside risk provided by the potential for geopolitical risks to impact on physical
supplies. We believe Iran, Nigeria and Iraq provide the three main geopolitical risks,
and we expect these risks to be the dominant driver of price behaviour. We expect to
see both demand growth and non-OPEC supply rebound from their relative weakness
at the end of 2005. Demand growth is expected to be driven primarily by the Middle
East, Latin America, China and the US. Non-OPEC supply growth is expected to
resume after a flat 2005, but our projected increase is only about one half of current
consensus forecasts. Should non-OPEC supply perform significantly better relative to
demand, then we would expect OPEC to cut output and to defend price levels close to
the mid $50s for WTI. The current misalignment between crude oil and oil products is
expected to be corrected by a period of lower refinery margins and a heavier-than-
usual maintenance season.
The oil market has entered 2006 with its price dynamics continuing to be driven by thetriangle of interaction between supply, demand and supply capacity. On the one hand,
supply, at the level of oil products, has recently managed to more than keep pace with
demand on a flow basis. However, on the other hand, demand remains uncomfortably
close to the limits of sustainable capacity. For several quarters the pricing dilemma for
the market has been how to balance the sometimes price-negative implications of
short-term flows and their associated inventory situation, against the risks implicit in
running the system at highly sub-optimal levels of spare sustainable capacity. The price
signal is then having to assimilate a considerable amount of information. It has to adjust
to the short-term flow of data, as well as variations in the perceived probability of
politically induced supply-side risks, and also to send the correct long-term investment
signal along the price curve.
The oil price must currently reflect all that, and do it from a starting point at levels
where there is only very limited experience as to the precise behaviour of supply and
demand elasticities. Hence, the system is also going through a process of feeling out or
recalibration in judging which price range has become the new norm. Indeed, changes
in the perception of the necessary baseline perhaps represent as big a price risk as
current politics or short-term data flows. For example, should the period in which prices
have stayed above $60 become prolonged without any obvious significant sustained or
fast easing of capacity tightness, then a move to a $10 higher range becomes distinctly
possible as part of the markets’ process of feeling out supply and demand dynamics in
an unknown price environment. Likewise, should credible signs of a fast enough easingemerge, the market could just as well take another look at a lower range to retest and
reconsider the dynamics at those levels.
At this point, we believe that a recalibration upwards is more likely than a downwards
shift. Indeed, our current price forecast deck is intended to signal that we are prepared
to take that into our base case by Q4 of this year at the latest. However, given the
normal fog in the data, no one should be overconfident about the certainty of any
change which would be primarily driven by market perceptions of the underlying
There is one key asymmetry in this potential process of recalibration and trying different
ranges. Other than the ultimate market response, there is no necessary bar to the number
of upward transitions the market could potentially make. However, the number of
downward moves is, we believe, strictly limited by producer response. In particular, we
believe that producer response would become a significant factor were prices to move to
about $55 for WTI or roughly $50 for the value of the OPEC basket. That is not to say that
there is an effective short-term floor, but it is to imply that under the current alignment ofpolicies and response, a push below those levels is unlikely to prove to be sustainable.
We believe that the immediate balance of importance has started to shift somewhat
towards the demand and capacity interaction and away from shorter-term supply and
demand balances. This shift is due to a marked increase in the degree of political risk in key
producing areas, heightening the fundamental concern that the margin between demand
and effective sustainable capacity could disappear due to the possibility of constrictions of
supply. Were capacity to be less constrained, the presence of political risk at the margin
would be expected to have a considerably diminished effect. However, with a reduced level
of that shock absorber available, the magnitude and probability of shock required to have a
given impact has become significantly less.
This is not a matter of trying to price fundamentals against geopolitical risk, in that the
price cannot be sliced into an element that is “justified by fundamentals” and a so-called
premium that is down to political developments. The reason that prices are showing an
enhanced sensitivity to politics is about as fundamental as it could be, ie, the cumulative
effect of a three-year period in which demand growth has outweighed capacity growth,
thus reducing spare upstream capacity to less than 2 mb/d. Indeed, in many ways that is a
far more fundamental impact on prices than the effect of the short-term flows. Before
evaluating the key features of the fundamental balances for 2006, we first run through what
we see as the three main geopolitical risks that are likely to add to market volatility and
price averages in the short and medium terms. Given the changed sensitivity of the market,
it is perhaps no surprise that the bias of coverage in this commentary reflects the bias in themarket towards greater responses to geopolitical developments.
The first of the key geopolitical drivers of the market is Iran. In our view, the importance of
the Iranian situation for the oil market comes from three main factors. Firstly, and most
obviously, Iran is a key producer whose exports are some 1 mb/d greater in magnitude than
the remaining level of global upstream capacity. Secondly, the nuclear issue is likely to
remain interlinked with oil, in that it is difficult in policy terms for the two to ever be
credibly separated. Thirdly, the possibility of a policy mistake on both sides of the dispute
appears to us to be distinctly non-trivial. Taking these three factors together produces a
non-trivial possibility of some extreme market outcomes, enough to make the maintenance
of a significant short position more difficult, and hence enough in itself to support prices at
higher levels.
The extreme case of a withdrawal of Iranian exports from the market would most likely
require the use of strategic reserves and create pressure for the last remaining tranche of
OPEC spare capacity to be brought onto the market. Whether it would be politically
acceptable to bring that last tranche into play depends somewhat on the circumstances
under which the withdrawal of oil occurred. Indeed, there are some not implausible cases
under which the entire adjustment would have to met though strategic reserves. The
implicit linking of oil with the nuclear issue is perhaps the most effective diplomatic tool
that Iran has at its disposal, particularly in terms of causing enough concern to create some
decoupling of key Security Council members from an overly proactive stance against Iran.
Given that, and also given that oil remains the most instant form of retaliation available to
Iran, the Iranian nuclear issue is likely to remain a concern for the oil market for a while to
come yet, and changes in the perceived probability of the more extreme outcomes are likely
We believe that the chances of a major policy mistake surrounding Iran’s external
relations are being increased by several factors. Firstly, over a period of years there has
been a general degradation of the understanding of Iran’s political dynamics in several
key foreign ministries. Secondly, the underlying policy towards Iran in some key
governments remains one that involves the possibility of subversion more than
engagement, a feature which does not seem likely to set the stage for a sustainable
agreement. Thirdly, some key red lines have already been drawn. It is becoming harder tosee how far the Iranian government can step back without losing support on what has
become a major nationalistic issue, and it is also difficult to see other governments
wishing to accept some potential minimal outcomes against Iran on the nuclear issue.
With political expediency, nationalism, lack of mutual trust and understanding, and
potentially poor political analysis all involved, we believe that there is a non-trivial
possibility of a failure in attempts to find an optimal and sustainable solution to the issue.
The second geopolitical issue that is capable of adding to volatility is Nigeria. It appears
to us that there has been a significant escalation in the nature of what has been a
prolonged period of attacks against oil infrastructure in the Niger delta. Until recently,
those attacks could best be characterised as having their roots in very local issues and
local demands for shares in the oil rent. However, more recently the attacks have come
from groups that are of a different nature. They appear to be better organised, capable
of mounting significant offshore as well as onshore attacks, better financed, better
armed, and to be driven by wider regional and ethnic divisions. In short, while the
intensity of violence in the delta regions has tended to wax and wane over many years,
the most recent attacks seem to be the result of a more credible medium- and longer-
term threat to the integrity of Nigerian exports. In current market circumstances, the
impact of this political risk is further enhanced by the nature of Nigerian crude, ie,
predominantly light and medium low sulphur crude with high gasoline and distillate
yields which plays a key swing role between regional crude import markets.
Until now, Iran and Nigeria have received the greatest market attention. However, we
believe there is a third major risk: the evolution of the situation in Iraq. Within the Iraqi
oil industry, there are no obvious signs of sustainable improvement, given the
continued degradation of the system, and the lack of sufficient investment to even
cover depreciation, let alone to improve the system from its current low base. Export
levels are likely to be volatile, but in recent months they have continued to set new
post-war lows. That background of poor performance is likely to continue, particularly
given the lack of security and the contentious nature of oil within the Iraqi political
situation. However, it appears to us that larger risks are emerging under which
centrifugal political forces could lead to either the de facto or de jure break up of Iraq.
Under such circumstances the entire regional political balance is likely to be affected,
with a southern zone that is at least sympathetic to Iran, a contentious northern zone,
and an unstable central Sunni area that could be subject to various competing claims. In
short, while it is not yet a base case, we believe that there is a non-trivial probability of a
weakening of links and then political disintegration process in Iraq. That would not only
potentially impinge on Iraqi exports. It could also have further implications for Iran and
international policy towards Iran, as well as changing the dynamic of a series of broader
regional issues.
There are other geopolitical factors that could affect the market. However, at least at
this stage we do not expect any of them to play an immediate role. This remains the
case in our view for US and Venezuelan relations. While relations have continued to
deteriorate in what has become something of a diplomatic comedy of errors, which
began with external reactions to the failed coup against President Chavez, we would not
expect that degradation as it stands at the moment to affect prices in the short to
medium term. There are also a number of other potential but more minor flashpoints,
particularly in some of the emerging non-OPEC producers, eg, Sudan and Chad, whose
output growth has been a factor that has helped to compensate in recent years for
declines in the North Sea and the US. Even within the context of the enhanced
sensitivity to geopolitics brought on by low levels of spare capacity, it is hard to escape
the conclusion that the political risk of oil supplies at the margin has been increasing in
a significant manner.
In terms of supply and demand flows, the key question for 2006 is to what extent, if
any, the dynamics of 2005 will be reversed. On the demand side, growth slowed in 2005
until by Q4 global demand was rising Y/Y by only about 1%. Non-OPEC supply growth
first slowed in 2005, and it then became negative by Q4. Across the year as a whole,
non-OPEC supply was flat, with a fall outside the Former Soviet Union (FSU)
compensated for by growth within the FSU. Some structural weaknesses in non-OPEC
were assisted by the significant impact of the US Gulf hurricanes to produce a Y/Y fall of
about 1.2% in Q4. Given these dynamics, the question for 2006 is which side of the
market will rebound the most convincingly.
In our base case, we expect both demand and non-OPEC supply to rebound, with thenet effect being an increase in the call on OPEC crude oil that is roughly the same as the
expected growth in OPEC capacity. We therefore expect the cushion of spare capacity
by year-end to remain roughly the same as it was at the end of 2005, with geopolitical
risks threatening to reduce that cushion relative to the base case.
Figure 98: Y/Y change in call on OPEC crude by quarter
(mb/d)
-2.0
-1.0
0.0
1.0
2.0
3.0
01 02 03 04 05
Source: Barclays Capital.
Figure 99: Cumulative increase in call on OPEC over
previous three years (mb/d)
-2
-1
0
1
2
3
4
5
01 02 03 04 05
Source: Barclays Capital.
The most relevant quantity in terms of price formation is the change in the call on OPEC,
and the associated implications for spare capacity. In particular, prices are being driven,
through spare capacity changes, by the cumulative changes in the call on OPEC. Hence,
the key characteristic of 2005 was not so much that demand growth slowed or non-OPEC
supply fell, it was that the call on OPEC increased again. Indeed, in terms of Y/Y changes
by quarter, the call on OPEC has not decelerated noticeably, and the Y/Y change has been
positive for 13 straight quarters. When we aggregate those changes to show the
cumulative pattern over the past three years, the cumulative nature of the grounds for
current price strength become clearer, and it also becomes clearer as to why prices could
continue to rise in a quarter where demand growth was relatively mild.
There are three main grounds for projecting a rebound in the rate of demand growth
from the Q4 level, a rebound which on our projections would leave annual demand
growth for 2006 roughly the same as in 2005. The first element is China, where demand
growth is expected to resume in official data. We suspect that official data probably
underestimated growth in 2005, although the data is of such poor quality that it is
probably not rational to over-analyse it. However, with Chinese economic growth
expected to improve in 2006, we feel confident that oil demand growth will resume at arate of at least 6%, or probably a couple of points higher. The second element is the US,
where demand growth is likely to be boosted by robust economic conditions, and some
positive Y/Y effects stemming from the hurricane-induced abnormalities in the second
half of 2005. Elsewhere in the OECD, a rebound in the Japanese economy is expected to
keep oil demand level or slightly up, reversing the falls in some recent years, while
European oil demand is also expected to be flat to slightly up. The third main element
driving demand is the robustness of economies in the Middle East and Latin America.
Between them, these two regions have produced some 0.5 mb/d of demand growth in
each of the past two years, and we expect a similar performance in 2006, given their
current economic trajectories.
While we are expecting non-OPEC supply growth to improve in 2006 after its flat
performance in 2005, there is an issue as to the scale of that improvement. We are
expecting growth of about 0.6 mb/d, while consensus estimates are about twice that
level, and some projections are as much as three times that level. The qualitative case
for a rebound rests primarily on the easier Y/Y comparisons in the second half of the
year, particularly in the US Gulf. The wildcard in that comparison is of course the
severity of this year’s hurricane season, and it would perhaps be unwise to assume that
all the Y/Y losses in the US will necessarily be recovered. The quantitative difference
arises largely from our assumptions on decline rates in mature areas; a factor which we
think has systematically been underestimated in recent years, and on our relatively
downbeat view of prospects in Russia. While Y/Y growth in Russia has improvedrecently, given that recent comparisons have involved a period of M/M production falls
in late-2004, we are not convinced that capital is particularly incentivised at the
moment within the industry. With few new projects due on stream, the question
appears to us to be when rather than if output will flatten out and stall.
Beyond the playing out of geopolitical developments, the evolution of the call on OPEC
crude will be key to the market in 2006. Over the past three years, market conditions
have been favourable enough to allow the OPEC 10, (ie, OPEC excluding Iraq), to
increase crude output by 5 mb/d in addition to an increase of over 1 mb/d in NGLs.
They are producing more than 20% more liquids at more than twice the price. There is,
therefore, some ammunition left in reserve in order to defend prices from falling back
towards levels where longer-term economic growth and political robustness might be
jeopardised. If we prove to be wrong, in that demand growth disappoints and non-
OPEC growth is significantly better than we project, then we would expect to see cuts in
OPEC output before the year is over. However, if we are roughly correct, then OPEC will
need to produce more and not less as the year progresses, and the probability of
another recalibration of prices upwards is likely to increase.
For Q1 06, the US natural gas market reverses the dilemma encountered at the start ofthe winter season. With more than enough gas in storage to allay any fears of running
low on supply at the end of March 2006, the focus has shifted to how the market will
reconcile expected flush balances with still lingering high market prices. We believe that
this condition will be contained to short-term markets, extending well into Q2 06.
However, the winter 2006/07 forward curve and beyond will continue to receive pricing
guidance from an elevated crude complex, and will incorporate the perception that
marginal demand will pick up during the first half of the year and may be able to absorb
higher prices later. We are reducing our expectations for Q1 06 down to $8.60 and Q2
even further, but maintain a contango shape for the balance of the year. We are also
keeping forward years near levels that were issued in October 2005. The risks leading to
even lower cash and short-term prices stem from current winter weather remaining in awarm phase throughout. Countering that, longer-dated prices have potential to move
higher on the strength of normal or warm summer 2006, another active hurricane
season and higher crude prices.
Having withstood the onslaught of a record hurricane season and scope of damage now
readily recognized as historic, the Q4 05 price response was immediate and lofty. With
a challenged supply network, the perspective shifted to watching the response of
demand. The initial evidence of price elasticity, however, was muddied by the broad
warming pattern that began in the last week of December and has swept over North
America since. The reduction in core heating demand by residential and commercial
sectors is so substantial that now the outlook for the balance of withdrawal season ishow much gas remains in storage, above prior years’ metrics. The dampening effect on
gas prices has largely reduced the once-ebullient levels in the front months. Beyond the
2006 injection season, prices have remained firm, above the early October 2005
forward view, and have even tilted higher, thanks in part to the higher petroleum prices.
The supply-side has generally performed well, post-Katrina and Rita, growing onshore
volumes both in the US and in Canada. Unconventional production from shale and coal
bed methane basins continues to witness a brisk drilling pace, and the permitting
process is proceeding to develop adequate infrastructure to bring this supply to market.
With the turn in onshore markets to better-supplied, the LNG sector has been given a
pass to proving its delivery capability. While no absolute growth of landed LNG was
witnessed in 2005, the development activity on regasification moves forward, but also
offshore liquefaction plants have brought more supply to world markets. Robust
demand conditions persist in global markets, leaving the US NYMEX under other major
market indices, and unable to attract marginal spot volumes.
For 2006, domestic production and imports have been granted temporary relief from
showing error-free performance. Through the first half of the year, the storage cushion
will keep prices from spiking, and reveal wider production area basis discounts. During
the summer months the two keys will be day-time temperatures in major metropolitan
centres, and tropical disturbance activity in the Atlantic and Gulf of Mexico. Given last
year’s record on these two items, the markets will remain sceptical of avoiding
complicating circumstances, and subsequent price weakness could only continue toemerge beyond the start of third quarter.
The twin events of the strongest hurricane season on record, and a very cold December
left NYMEX prices pushing into new territory. But the about-face in northern US
temperature patterns left prices moving back to pre-hurricane levels. As if on-cue, the
day after the official start to the winter season, temperatures warmed across the
Midwest and Northeast, beginning the price decline for cash and 2006 futures. FromEIA statistics, the month of December historically shows the greatest probability for
core demand variability, which was borne out by the temperature variation in Midwest
and Northeast cities, during the month. January verified completely warm, with no daily
major core demand that reached levels considered normal, base on 30-year average
weather.
Despite the weakness shown by the front of the gas curve early this year, crude and
products have held to relatively steady value throughout, although the lack of home
heating demand has allowed No.2 fuel oil to build stocks in key Gulf and Eastern
markets, weighting similarly on product prices. Competition for marginal British
thermal unit (btu) demand has historically dictated that when gas prices turn higherthan No.2 switch-able demand will move to the cheaper fuel, and prices will correct
back. The theory holds under the most recent evidence, although many other factors
came into play during the recent changing of price leadership.
Across key regions, the accumulation of heating degree days for the winter has faltered well
below normal counts, or even below last January, which was itself below average in eastern
markets. With winter weather stuck in such a mild weather pattern, the impact on regional
basis development has been dramatic, narrowing the spread of Midwestern points over
Henry Hub, and keeping the Northeast spreads at transport values, since deliverability issues
are far from being a concern at present.
Figure 202: January regional heating degree days
200
400
600
800
1,000
New
England
Middle
Atlantic
E N
Central
W N
Central
South
Atlantic
Jan-06 Normal Jan-05
Source: NWS, Barclays Capital.
Figure 203: Market area basis, $/mmbtu
-3
-2
-1
0
1
2
3
Aug 05 Sep 05 Oct 05 Nov 05 Dec 05 Jan 06
Northeast
Midwest
Source: NYMEX, Barclays Capital.
Current price perspective
The most notable feature about the forward price curve evolution has been transition
to contango market over the next year, with the backwardation previously more
common through 2005, gradually returning after 2007. The shape reconciles the
weakness implied by high storage, which is deemed to be only temporary in the current
storage cycle. In the last two years we have seen how the perception of scarcity orabundance can emerge within the year or even within the season. Without actually
reaching shortage scenario, the extreme events of 2005 (warmer summer, colder
winter, active tropics) forced responses by the market well before March scarcity could
develop. Volatile weather patterns, then, can work for or against balanced markets.
However, once one cycle of storage has been consumed, the onus then is to prove that
supplies are ample enough to meet future winters. Implicit in this view are long range
assumptions that on-shore production is more getting more difficult to increase
consistently, and is matched with an inherently growing demand base. Given the recent
weather patterns, the uncertainty about resolving this condition now resides further out
in winter 2006/07 and beyond rather than the current year.
This shape is also influenced by the crude oil curve, which has attained a similar
appearance and has incorporated the perception of critical points over the next two
years, but then retains a flat profile. The key to the gas market weakening further out is
the arrival of liquefied supplies, no secret there. However, given the increased concern
about geo-political risk, and the seemingly global reliance on LNG to clear many local
markets, the risks have increased that the rational allocation of gas supply by ship may
not occur in an adequate or quick enough fashion. The more full impact of increased
LNG may not come before 2008, over which timeframe the domestic market could turn
tighter again.
Separating out the individual months becomes a contingent process with weather
playing the key role through the injection season. The nearly foregone conclusion nowis that an even cooler-than-normal February and March cannot undo the work of a mild
January lets the debate begin well into the second quarter. A flush supply profile made
even more bearish by operational and contractual requirements to drawdown storage
capacity by the end of the second quarter, will let weaker cash markets linger though
the April/May period, as demand again sinks to a seasonal spring trough. Maintenance
season for base-load electric power plants could provide lower price level support,
especially if planned work is deeper than usual in anticipation of meeting tighter
environmental standards. Summer season will see a tendency of gas prices to converge
with if not stay above residual fuel values, given the active switching capacity in Easternand Southern regional power markets. Recent forward curves have indicated this
support resides near $8.00, with each unit generating or power pool having unique
characteristics to account for sulphur and fuel transport differentials.
Through the summer months, the verification of heat in key markets and the advent of
another robust hurricane season will keep value from slipping to pre-2005 levels.
Absent those warmer conditions, cash prices could shift lower again, as demand again
would appear absent from the short-term markets. Moving through August and
September, the primary concern for energy and non-energy interests, will be how
active tropical disturbances become. Given the last two years of storm activity, market
observers discount the unfettered performance by Gulf of Mexico area gas supply –
prudence suggests factoring in at least 100 bcf of shut-in from the threat of storms,
with no indication that damage would result. Absent extreme circumstances, an
inventory build to 3.3 tcf at least would be the minimum expected, a level which has
evolved into the norm from market observations. The question would then revert to
weather during the winter season, as the storage capacity levels would be inferred as
capped some level near 3.5 tcf, given time constraints at end of season. Thus the
current flush supply appearance could lose its impact by the end of injection season.
2006 begins with discordant themes, none of which gets resolution quickly and easily.
During the year, the market must decide which trends have greater import – but
underlying will be persistent volatility in the near term, and persistent strength in
longer-dated values. 2005 demonstrated that the question of growing LNG supplies for
the US would not be answered yet. The flow of ships globally behaved in a rational
manner, moving to higher priced markets. Other factors affecting the valuation of gas
during 2006 will take time to resolve and could even persist into 2007.
Figure 204: Recent and current 2 year NYMEX strip
6
7
8
9
10
11
12
13
14
15
Nov 05 Mar 06 Jul 06 Nov 06 Mar 07 Jul 07 Nov 07
early Feb
early Oct
Source: NYMEX, Barclays Capital.
Figure 205: Storage spreads prior to injection season,
The perennial theme about the importance of working gas in storage is that the
absolute levels have significance only for the current year, and cannot imply much
about future price levels beyond the current storage cycle. In addition, the withdrawal
season 2005/2006 has proven that trends may change abruptly from appearing well-
supplied to something less than, and back again – all within a month’s timeframe. So far
this year, weekly storage reports estimate very high inventory. For consecutive months
in 2006, it appears working gas inventories have the potential to log their highest levels
in the NYMEX Henry Hub contract era, or since 1993. The end-January level should
reach 2,340 billion cubic feet, rivalling the 2002 level. End of March should reach 1,550
bcf, also slightly higher than the 1,517 hit in 2002. Before then, such high levels in
storage had not been common or even close. The years 1992 and 1993 ended thedecade-long trend of high first-quarter storage when end-January levels were at least
2,200 bcf and March ended with at least 1,600 bcf or even 2,000 bcf. To be so full this
year, each region has had to outperform prior year levels, and each has according to
specific fundamental reasons.
The bulk of US working gas storage is conventional, with mostly depleted reservoirs and
aquifers totalling almost 8,000 bcf, working and base combined. Salt capacity (caverns
and domes) has grown more quickly since the late 1990s, yet still only amounts to 300
bcf of working capacity. Storage operations are still dominated by traditional users
(local distribution companies) who plan for conventional use of little or no intra-
seasonal cycling. The cycling issue will be highlighted this year because the end ofMarch is likely to have such a high inventory level. Whether the working gas remaining
in place must be pulled to allow storage operators to perform maintenance or contract
terms allow for rolling inventory into the injection season is a key issue for capacity
holders with higher valued inventory anticipating a weak spring cash market. If there
are forced inventory reductions by traditional users, cash markets will react
accordingly, and non-traditional merchants will have opportunities to acquire cheaper
gas. In either case, after March, the injection season could commence at a rapid pace, if
spring weather remains mild and industrial loads are sluggishly responding to
weakened gas prices. The cash spread to winter months will likely remain robust and
well above historical averages, which were less than $1.00 until last year. The
propensity of next winter’s prices to converge with near-term levels will be delayedthrough the summer, but weaker winter prices could develop if a historically high
Figure 210: Daily storage operations at AECO, mmcf
-2,000
-1,500
-1,000
-500
0
500
1,000
Jul 05 Aug 05 Oct 05 Dec 05 Jan 06
Source: Encana, Barclays Capital.
Figure 211: End of Sept Canada storage, 2005 & 2004, bcf
100
110
120
130
140
150
160
170
180
East West
Jan 05 Jan 06
Source: Enerdata, Barclays Capital.
Core sector demand greatly reduced
The warm trends during last summer extended well into the fall, with first half of Novemberverifying very warm in principal cities along the northern tier of US states. While December
showed evidence of cooler temperatures during the first half of the month, such pattern
was completely shut down prior to the Christmas holiday break. Even normal temperatures
have been absent from the Midwest and Northeast since late December forcing core
demands to historically lowered levels. Occasional tele-connection forecasts have indicated
potentially cooler air, but prior to February all such indications failed to verify. The variable
weather in December had left some meteorologists calling for at least some warming in
January but none had foreseen the extent of warm recent trends. February and March may
still reveal some cooler temperature patterns, but two facts have emerged. The almost
complete lack of snow-cover in northern states, compared to normal conditions has left the
ambient conditions warmer, with less energy reflected back from the surface. This couldtranslate into lower potential for sustained cooler conditions in the balance of this winter.
Second, typical core demand levels in February and March are lower than January levels,
meaning even a cold February will not provide the same level of space-heating gas burn as
mid-January seasonal peaks do. Short-term price elasticity would dictate the return of more
demand at the tail end of the withdrawal season, if prices continue to weaken, leaving a
small positive impact, but perhaps inconsequential to the broad warming patterns
observed.
Figure 212: Q1 core gas demand, bcf
20,000
25,000
30,000
35,000
40,000
45,000
50,000
2000 2001 2002 2003 2004 2005
Jan
Feb
Mar
Source: EIA, Barclays Capital.
Figure 213: January average temperatures in New York
US domestic production still struggling with hurricanes,
growing elsewhere
In considering the state of US Gas production during 2005, it is helpful to separate the
hurricane-related events from on-shore trends. Prior to the two major hurricanes at the
end of the summer, on-shore production was doing better than just breaking even on a
net basis. Increased production has been noted mostly in the Texas districts 5 and 9,covering the Barnett shale area. Additional shale development activity has been
accelerating in the west Texas Woodford play. Behind Texas, the Rockies have been
increasing steadily, adding more than 300 mmcf per day for the first eight months of
2005, compared to the January-August period in 2004. Countering this growth has
been the drop in Oklahoma and Kansas production, over 200 mmcf/day, during the
eight month periods. For the Federal off shore system, Y/Y declines surged to 1250
mmcf/d through the end of August, about the time when shut-ins began in anticipation
of the first major storm to hit the central Gulf of Mexico. Without the Gulf declines, US
production increased by almost 250 mmcf/d – with Alaska adding another 50
mmcf/day. Total lower 48 production through August dropped 1.0 bcf per day.
The history of Gulf gas production up to 2004 has perennially anticipated the impact of
hurricanes on daily production. With the landfall of Ivan in late August, the extent of
foregone production was significant by comparison, but importantly, the pace of
recovery was slow enough so that Q1 05 still logged 300 mmcf/day offline, due to the
storm impact. This was reduced to a rate below 100 mmcf/day by the middle of the
summer. But still the event showed the difficulty in recovery from strong storms. Of
course the impact of Katrina followed by Rita was the worst by far for the Gulf coast
energy industry. Through the end of 2005, over 560 bcf was reported by MMS as lost
production, a figure shy of combined state impacts as well. Significantly, the daily shut-
in rates through January averaged above 1.6 bcf per day, and show little improvement.
Indeed, government assessments, with guidance from commercial operators indicatethat recovery will slowly progress through H1 06, and significant production will remain
offline, or will ultimately be abandoned, in favour or Greenfield development.
The industry is struggling with resource allocation and the shift to higher prices
throughout the curve, which provides incentive for every producer to engage in E&P
activity equipped with a growing list of economic prospects. Key infrastructure has
gradually been brought back to service, especially the larger gas processing facilities
that sustained damage during the storm. Drilling activity in the Gulf of Mexico had
averaged 80 rotary rigs during the summer, comparable to the 2004 summer. However,
by November that rate dropped into the 30s, and is showing no signs of a turnaround
yet. Some larger production projects anticipated for start-up during early 2006 have
run into delays, attributable to storm-related complications, eg, the Atlantis joint
venture project slated to produce as much as 180 mmcf/day. Recent news that the
timing was under review indicates the uncertainty still around completion. Also, the
existing Mars production platform, which has been a target for hurricanes in the last
two seasons, is now expected to start up in the middle of 2006, with full production by
the second half of the year. This Mississippi Canyon facility has typically produced 400-
500 mmcf/d in the past. The US MMS has estimated that ultimately 1-2% of Gulf gas
production has been lost permanently, as the damage sustained by smaller and older
platforms preclude re-investment. To offset such losses, the service has indicated it is
making plans to offer financial incentives to the industry for projects that would
otherwise be uneconomic. We expect progress to be made on recovery, improving toless than 1 .0 bcf per day still shut-in during Q3.
Figure 222: Texas district gas production, 4, 5, 8 & 9,
2000 - 2005, mmcf/d
0
2,000
4,000
6,000
8,000
10,000
Jan 00 Dec 00 Nov 01 Oct 02 Sep 03 Aug 04 Jul 05
RR 4 RR 8 RR 5 RR 9
Source: TX RRC, Barclays Capital.
Figure 223: Hurricane count by decade, since 1851
0
5
10
15
20
25
1 8 5 1 -
1 8 6 0
1 8 7 1 -
1 8 8 0
1 8 9 1 -
1 9 0 0
1 9 1 1 -
1 9 2 0
1 9 3 1 -
1 9 4 0
1 9 5 1 -
1 9 6 0
1 9 7 1 -
1 9 8 0
1 9 9 1 -
2 0 0 0
All
Major
Source: NHC, Barclays Capital.
Figure 224: Federal Gulf of Mexico gas productiontrends, bcf/d
8,000
10,000
12,000
14,000
16,000
Jan 00 Apr 01 Jul 02 Oct 03 Jan 05
0
40
80
120
Gas Production
Rigs
Source: EIA, Baker Hughes, Barclays Capital.
Figure 225: Alberta gas production, mmcf/day
15,000
15,500
16,000
16,500
17,000
17,500
18,000
J a n
F e b
M a r
A p r
M a y
J u n
J u l
A u g
S e p
O c t
N o v
D e c
200320042005
Source: Alberta EUB, Barclays Capital.
Canadian supply continues to expand in West
Total gas output in Alberta during 2005 showed few clear trends higher or lower than the
prior two years, although ultimately cumulative production through November increased
sequentially from 2003 to 2005. During H1 05, production started stronger but finished
weaker than the prior year. Summertime trends surged higher again, while October total
went back to flat. For November the output was 200 mmcf per day higher than November.Since October, drilling activity has surpassed year prior levels, through January 2006,
leading expectations that Western Canadian sedimentary basin output will remain strong
through Q1 06. Further west, British Columbia January-October output has averaged 200
mmcf per day higher than the first 10 months of 2004, but rig counts show signs of slow
growth, and slipping behind the pace in BC during the end of 2004. Future gains from the
recent level may be limited until more drilling activity is supported by the data. Eastern
Canadian output from offshore Nova Scotia remains limited to 450 mmcf during good
months and can slip below 300 quickly, a disappointing trend but nothing new. The rate of
decreasing output has levelled off but thoughts of a turnaround must await further offshore
developments.
With the weak demand profile in the US during the fall time, post Katrina and Rita, the
imports from Canada remained reduced from 2004 levels, but the cold spell in mid
December did result in stronger flow south across the Midwestern and Eastern corridors,
than seen in the last year. The surge was limited, and recent export levels have been
flagging. With reasonable full storage in the US for January and lack of apparent demand,
the slow export is just another bearish factor that will force local prices to slump further,
relative to Henry Hub values. Stronger exports are possible but would just be competing
with cheap gas in US markets also.
Figure 226: British Columbia gas production, mmcf/d
1,500
2,000
2,500
3,000
3,500
4,000
Jan 00 Dec 00 Nov 01 Oct 02 Sep 03 Aug 04 Jul 05
Source: BC Ministry of Energy & Mines, Barclays Capital.
Figure 227: Western Canadian rig counts
0
100
200
300
400
500
600
Sep 02 Apr 03 Nov 03 Jun 04 Feb 05 Sep 050
20
40
60
80
100
120
140
160
180Albe rta (LHS) Britis h C olum bia
Source: CAODC, Barclays Capital.
Figure 228: Nova Scotia Gas Production, mmcf/d
0
100
200
300
400
500
600
700
Jan 00 Dec 00 Nov 01 Oct 02 Sep 03 Aug 04 Jul 05
Source: CNSOBP, Barclays Capital.
Figure 229: Canadian Gas Pipeline Exports, mmcf/d
0
2,000
4,000
6,000
8,000
10,000
12,000
Nov 04 Jan 05 Mar 05 May 05 Jul 05 Oct 05 De c 05
West Midwest Northeast
Source: Facilities, Barclays Capital.
LNG supply dropped during 2005, changes profileFor the reported months through November, 2005 is set to log less import of LNG than
2004, a significant change from the common expectation that more ship traffic would pick
up during the second half of the year and lift daily average regasification up to 2.0 bcf/d orhigher. Indeed, the most recent months (December and January) have been the weakest
with Lake Charles significantly lacking its usual flow rate. More recently the Cove Point
facility has shown slower receipts, although we anticipate that this trend is temporary and
will not mimic the Gulf terminals’ complete lack of traffic. Excelerate Energy has not had
success in landing new spot or contract cargoes but recently has announced plans to source
new supply ex-Pakistan, from a planned liquefaction terminal in the port of Karachi using
their patented on-board technology. Presumably this is still a few years away. Elba Island
and Distrigas have shown more consistency but the trend is clear for now that spot supplies
will continue to flow to the highest priced market and the US market values trail European
prices which are cheaper still than Asia. In Q4 05, global demand for spare LNG supplies
staged a strong turnaround, from the summer time. Firm demand from Japan and Korea for
winter utility service pushed market prices for spot cargoes up to $20/mmbtu for
January/February delivery into Korea. Only recently have values returned to $15.00, for spot
volumes. Long-term contracts notionally are quoted around $8/mmbtu delivered.
The profile of supply to US markets shows Trinidad volumes declining during 2005, and
Egypt essentially taking up the slack, as Algeria still represents a variable source of
supply. Looking forward to 2006, spot cargoes will still be an unsteady source, but more
liquefaction capacity allows better chance of marginal cargoes to land in the US, if only
at a higher price. Elba Island has recently completed the expansion of its terminal
raising total throughput capability to 1.2 bcf/day. Extra storage capacity was added,
almost doubling space to 7.3 bcf. Incremental supply is scheduled to load from Qatar,eventually. Within the next two years, at least four new projects are set to bring
commercial quantities of LNG to the North American continent. This could total over 5
bcf of incremental regasification capacity. However, most of this capacity is due to
start-up in 2008 if not 2009. Until then, better utilization of the five continental import
terminals, and subsequent expansions therein, will be the sole source of increased
capacity.
Figure 230: Source of LNG imports, bcf/d
0.0
0.5
1.0
1.5
2.0
2.5
Jan 05 Mar 05 May 05 Jul 05 Sep 05 Nov 05
Trinidad Algeria
Egypt Malaysia
Nigeria other
Source: terminals, Barclays Capital.
Figure 231: LNG infrastructure, bcf
0.0
0.5
1.0
1.5
2.0
2.5
3.0
E n e r g i a
C o s t a
A z u l
C a m e r o n
F r e e p o r t
S a b i n e
P a s s
Source: EIA, Barclays Capital.
Electric power demand key for summertime
Electric power demand for gas in the summer of 2005 showed the biggest consumption
rate since 2002, due to much warmer weather than verified during summers 2003 and
2004. Since Q3, ambient temperatures have been milder than normal on average,
lowering the call on marginal gas fired capacity. Higher gas prices have forced capacity
owners of nuclear and coal based units to optimize output given the btu value well under
the market rates for natural gas. Demand profiles of electric power shift lower during end
of Q1 and Q2, but return in importance during the June-September period.
For 2006, the key themes to electric power from a natural gas demand perspective are
summer weather in eastern states, operational dynamics of other base-load fuels, and
relative prices for oil products, residual fuel oil and distillate heating oil. The climate
regionally in the US was warmer than normal in most locations. But interestingly, for
Texas and generally the south central states, mean daily temperatures in the July/August
period were below the 30-year normal. This is significant given that Texas is the largest
consumer of natural gas, averaging 5.6 bcf/day last August, which is almost double the
next largest consuming state, California. In addition, the Midwest region, with Illinois at its
centre, registered a summer climate just over normal. Lifting the overall profile of the
summer to warmer than normal were the West Coast and East Coast, also key gas
consumption regions. In comparison to just normal weather across all regions, the year
2005 was warmer than normal, leaving expectations for this year to be cooler than 2005.However, showing just normal or warmer than normal in the South Central region would
prevent gas demand from declining all the way back to levels of summer 2005. When
considering the electric power load growth which has generally been averaging 2% per
year during the last decade, this implies that the underlying growth expected for this
summer (June-September) calls for 9,500 average megawatts of incremental energy,
needed from any energy resource. If more than half of that energy comes from available
gas capacity then an additional 1.2-1.5 bcf per day or load growth can be imputed,
outside of weather variations.
Of course, the behaviour of coal, nuclear and hydro resources will help dictate the portion
of load growth available to gas, or any incremental changes that re-order the capacity
stack. New nuclear units are not near construction, although new designs are getting
ready for the permitting phase. Upgrades from existing capacity have been commonly
made, but these are limited to 1000-200 MW totals at a time. The spring of 2005 saw
deep maintenance and refueling activity, which is not expected to be repeated this year.
With more nuclear available through the spring season, less demand is inferred for the
gas capacity stack. Coal-fired generation hit new highs during 2005, with new monthly
record set in August, with little new capacity online. The outlook for new capacity this
year is less than 500MW and averages only 1400MW until 2008. Only by 2009, does the
list of planned new units deliver substantial capacity, over 8,000MW in that year.
Meanwhile new gas-fired capacity should average 14,000MW until 2009, then reducing
to just 5,000MW of new resources. Given tightening environmental restrictions, the
prospect for coal-fired energy to capture all incremental power growth is not great,
although existing capacity has shown the ability to increase efficiencies.
Hydro presents another alternative to natural gas, especially in western markets. The bulk
of Northwest hydro is located in the Columbia River system, including the Snake River.
The water supply outlook is presented regularly by the Northwest River Forecast Centre,
which forecasts total water supply for the basin during multi-month blocks. The January-
July amount this year is currently expected to be greater than the 30-year average. This is
significant because every year since 2000, the outlook has been for less water supply than
average. Typically, early season forecasts have less accuracy, and the trend in recent years
has been to expect less water than what actually flows through downstream points. This
year the early call is for more than normal water, indicating a strong likelihood of
abundant water available for hydro-power generation, and thus less gas demand from
western power markets.
Residual and distillate fuel oils are the remaining competitive fuel for generating electric
power. Since units capable of burning either fuels are common in eastern and southern
markets, the price sensitivity to btu-based comparison is relatively transparent. However,
Independent System Operators (ISO) are fully capable of re-dispatching the grid, so all
units compete on a security constrained unit cost and availability basis. Over the last year
and before, the two fuel oils have represented a band within which gas prices have tended
to fluctuate. Residual values have represented a more definite bound, while higher pricesfor natural gas have made excursions well beyond distillate prices. With forward prices for
residual fuel oil near $8.00 in New York Harbor for benchmark 1%, lower gas prices would
imply greater gas consumption. While these amounts may be nominal during light
demands in the spring, the more full effect would come during May-September. Various
estimates abound for installed switching capacity, but evidence suggests that over 1.0 bcf
per day would be sensitive to gas prices under benchmark fuel oil prices. Additional but
less capacity exists for industrial consumers switching from residual fuel (closer to 0.5
bcf/day). Of course, residual fuel oil prices are held higher by the crude curve, and
offshore demand. Weakening in either would depress US residual values further, and
lower the price signals to natural gas – this influence remains strongest in the
summertime, but this remains just a signal, ie, gas prices are free to move lower, with an
Industrial gas demand weakened by storms and prices
Through 2005, the creeping value in gas prices began to register an effect on industrial
demand as monthly average demand began to slip away from prior years’ levels, even
before August and the hurricane season. Of course the impact of hurricanes took out a
large swath of industrial capacity in the Gulf coast, physically. And the price spikes up
to $15.00 left many smaller gas-buying concerns out of the market. The near shutdown
in industrial processes left open the prospect for importation of feed-stocks (eg,
ethylene) and end-products (polymers). Such a rundown in inventory of these products
left local prices soaring over European and Asian values, temporarily. Producers of
fertilizer witnesses record high pricing, but even those levels meant some facilities
could not make margin with higher gas prices. The steep decline to pre-hurricane gas
price levels during January has provided incentive for industrials to return back to the
US market. Strong market prices for end-products bolstered by a healthy economy,
have kept margin available for industrial consumers still in the market. Indeed, strong
quarterly results from larger chemical companies confirm the limited impact of the
hurricane, although real and significant damage did occur, with life-changing
consequences for locally based personnel. With the most recent statistics available fromEIA covering October 2005, industrial gas demand was still short of 2004 levels by 100
bcf or over 3 bcf per day. Return of more demand during Q1 06 is expected, and should
show the strongest representation after the spring outage of crackers and refiners.
Some fertilizer production has been noted returning to the market, which only came
offline during November 2005. This trend should be even stronger as cash prices
weakened during the spring period, providing incentive to those industrials still looking
for sub $8.00 values.
Figure 238: Industrial demand for gas, mmcf/d
15,000
17,000
19,000
21,000
23,000
25,000
Jan Mar May Jul Sep Nov
2002 2003
2004 2005
Source: EIA, Barclays Capital.
Figure 239: US Fertilizer prices, $/ton
$250
$280
$310
$340
$370
$400
Dec -04 May 05 Oct 05 Jan 06
US Gulf
Tampa
Source: Green Markets, Barclays Capital.
With the recent gas price still attractive to power producers using coal for fuel, the
incentive to maximize output still exists. Transportation problems encountered moving
coal from the Powder River Basin on the Joint Line (BNSF and UP) have been addressed,
but the chronic condition of impaired transport out of the PRB suggests future
problems may arise. With such strong emissions credit prices impacting eastern coal
prices, for all units needing to compensate for higher sulphur consumption, demand is
set to increase for western coal over the near term. Eastern coal prices could even
weaken in order to compete accordingly. The forward view on NYMEX spec coal is
relatively flat at just below $60/ton, or inconclusive about future dynamics. Relatively
lower stocks and potentially strong demand during the summer season will preventmuch of the decline in prices, but a mild summer and weaker gas prices could put
pressure on the summer months, which then would extend to end-of-year values, at the
conclusion of August and September peak potential demand period. Recent tragedies in
West Virginia have forced scrutiny over mining practices and even given pause to
production. If regulatory developments emerge from these events, the prospect for
future production growth could be lowered, also providing notional support to Eastern
coal prices.
Figure 240: Eastern coal price, $/short ton
$30
$40
$50
$60
$70
1st
Q 04
2nd
Q
3rd
Q
4th
Q
1st
Q 05
2nd
Q
3rd
Q
4th
Q
Bal
06
Cal
07
Spot Jan 06
Source: NYMEX, Barclays Capital.
Figure 241: Emissions credit prices, $/ton
$300
$500
$700
$900
$1,100
$1,300
$1,500
$1,700
Sep 04 Dec 04 Mar 05 Jun 05 Sep 05 Dec 05
Source: Canto Fitzgerald, Barclays Capital.
A complete change in the profile of short-term fundamentals has occurred, with strong
supply rebound from the hurricane season, and very weak demand during the current
winter. A healthy storage profile will persist throughout the 2006 injection season,
leaving end of October levels at or above at 3.5 tcf. Such abundant supply will tend to
depress in-season prices, but leave forward values less affected, but not completely so.
Going through the first and second quarters 2006, there are more bearish signals than
bullish ones, but most of those conditions are well identified by the market. Further
weakness could emerge during the first half of the year, as these events are realized inthe cash markets. We are lowering each of our quarterly prices for 2006, especially for
Q2 to below $8.00. But the forward calendar years remain unchanged, and indeed the
current market has moved up to recognize value in those years, and the implied
uncertainty made more substantial by geopolitical risks associated with accessing LNG
supplies.
Figure 242: Barclays Capital natural gas price forecast
Thermal CoalThe following article has been contributed in good faith by coal portal ( www.coalportal.com ,
a segment of coal industry specialists, Barlow Jonker) for Barclays Capital. Barclays Capital
accepts no responsibility for the content.
Spot price bounce driven by northern winter,but market moving to balanced supply
The sharp decline in thermal coal prices in both the Atlantic and Pacific markets has
been halted and prices are again on the rise.
Very cold weather in Europe and Japan has led to increased electricity demand that in
turn has increased coal consumption and demand.
Uncertainty over European gas supplies centred on Russia’s ability to deliver requisite
volumes under contract augurs well for coal.
Production in the main coal exporting countries has risen to record levels pushed by
high demand. Expansion of mines and projects is under way in Australia, Indonesia,
South Africa, Colombia and Russia.
Indonesia has become the world’s largest thermal coal supplier, displacing Australia.
Indonesian coal is a cheaper alternative, especially when freight rates are included.
China has stated that it will be able to satisfy internal demand with domestic coal
through production of more than 2.1Gt of coal in 2006. A major reason for the
current state of the global thermal coal market is Chinese demand for energy and its
contraction as an export supplier.
Early semi-soft coal prices have declined 30% from Japanese fiscal year 2005 (JFY 05)
and this could lead to increased supply in the Pacific market as unwanted coal is sold
into the thermal market (typical buyers in recent months have been Israel and Spain).
Some evidence of oversupply is emerging in Indonesia. More than any other country,
Indonesia has pushed new coal mine developments and production and stockpiles
continue to grow.
Growing domestic demand will place pressure on Indonesian coal export growth in
the medium and long term. Three new coal-fired power plants are due to come online
in 2006 that will consume low energy (low CV) coal.
Russia is a marginal supplier on a cost basis and higher Atlantic prices are drawingthem further into the European market. More Russian supply into Europe should keep
a lid on prices as they reach the high US$40s/t and low US$50s/t.
US thermal coal inventories are low and will add to US demand as restocking takes
place. US demand for coal is strong and electricity demand is growing at about 2% pa
and low stockpile levels could add up to 20Mt of demand in 2006. The majority of
Powder River Basin coal has been settled at between US$14.74/t and US$17.58/t and
is well up on 2005 levels.
Korean buyers are re-entering the spot market after an absence of more than 18
months. Korean Power Utilities (KPUs) drastically reduced the volume of coal procured
on the spot market as prices rose, but probably sense that the market has bottomed.
Term contract negotiations with Japanese Power Utilities are under way and we
expect a settlement of between US$42/t and US$45/t.
Crucially, Japanese Power Utilities (JPUs), the world’s largest consumers of exportthermal coal, are meeting with major coal suppliers in the current quarter (Q1 06) to
determine the contract price for coal for JFY 06. More than any other event, the annual
contract negotiations influence the forward price of coal as large tonnages are agreed
(with and without options) at tiered prices below a “reference” or “benchmark” price.
These negotiated prices reached an all time high last year when the “benchmark” was
set at US$53.00/t FOB. Barlow Jonker forecasts the current round of negotiations will
agree a price between US$42.00/t FOB and US$45.00/t FOB, or a reduction of between
16.6% and 22.2%, as the severe supply restrictions that materialised in 2003 and 2004
have eased. China and India remain the growth engines for thermal coal demand, but
there is some evidence that these two consumers are coming to terms with supplyshortages. For example, recent reports said critical stockpile shortages at Indian utilities
had been overcome, largely because heavy rain in the south took helped hydroelectric
power plants and relieved the pressure on thermoelectric generators. Restocking has
been so successful that Coal India, the government-run coal mining company, has said
that it might scale back production as mine stocks are around 17Mt and power plant
stocks are at 15Mt, compared to normal levels of about 10Mt.
Figure 244: Barlow Jonker Index (BJI) v ACR Asia Index v Japan
Benchmark* (Jan 86-Aug 05) (US$/t)
20
25
30
35
40
45
50
55
60
65
Jan 86 Oct 88 Jul 91 Apr 94 Jan 97 Oct 99 Jul 02 Apr 05
Reference*
ACR Asia Index
BJI Spot Price
Note: *Barlow Jonker estimate from 2002. Source: www.coalportal.com.
Traditionally, the “benchmark” has acted as a cap to the spot price of coal (Figure 244).
In the past, the vast majority of coal was settled against the benchmark system, with a
relatively small tonnage agreed at the top price and larger tonnages settled against a
lower, tiered price structure. Consumers, realising the opportunity to procure additional
coal on the spot market below benchmark prices, began to move towards spot marketsand reduced their reliance on long-term purchases. This flexibility worked well when
coal supply outstripped demand and prices were low, such as from the mid-1990s to
the early 2000s.
Figure 245: Korean spot purchases of thermal coal have declined with
As in parts of Europe, the Japanese winter has been particularly severe with heavy
snow in the north of the country and bitter cold in the west. The initial reaction of
the JPUs was to burn more oil because of its ready transport. However, a high
consumption rate by the utilities has led to a drawdown of coal stockpiles and some
rebuild is now necessary (Figure 249 and Figure 250). Traditionally lower than the
summer peak, Japanese electricity demand in winter is matching that trend.
Consumption at the 10 major Japanese utilities in the last two quarters of 2005 wasabove the stock level and buyers entering the market have been putting upward
pressure on prices. Twenty-two days of stock on hand is considered a critical
marker because it represents the minimum safe time needed to deliver coal into the
utility stockpile.
Figure 252: Coal stocks & consumption – 10 Major JPUs
(Mt)
0
2
4
6
8
10
12
14
16
1Q04 2Q04 3Q04 4Q04 1Q05 2Q05 3Q05 4Q05
0
2
4
6
8
10
12
14
16Consumption Stocks
Source: www.coalportal.com.
Figure 253: Days of stock on hand
0
5
10
15
20
25
30
35
40
45
1Q04 2Q04 3Q04 4Q04 1Q05 2Q05 3Q05 4Q05
Source: www.coalportal.com.
Figure 254: Monthly consumption (Mt) vs BJI (US$/t
FOB)
0
1
2
3
4
5
6
Jan 04 Aug 04 Mar 05 Oct 05
0
10
20
30
40
50
60
70
80Consumption BJI (monthly)
Source: www.coalportal.com.
Figure 255: Stock on hand (no. of days) vs BJI (US$/t
FOB)
0
510
15
20
25
30
35
40
45
Jan 04 Aug 04 Mar 05 Oct 05
0
10
20
30
40
50
60
70
80Stock on Hand BJI (monthly)
Source: www.coalportal.com.
Reports suggest that thermal coal stocks in the Hunter Valley (NSW, Australia)
were comparatively low and that pressure was on some suppliers to meet
shipments. It is understood that stocks in the Hunter Valley will remain under
pressure in Q1 06 because of heavy delivery schedules.
Japan halted its imports of Russian coal because of contamination by rocks in
shipments. Imports of Russian coal are expected to resume once a team of Japanese
experts visits Far East Russia to determine the cause.
Semi-soft coking coal prices appear to have declined by around 30% to US$55/t
FOB from last year’s settlement price of US$79.50/t FOB. The demand for these
types of semi-soft coals has fallen away sharply and anecdotal evidence from Japan
suggests that some consumers would like to arrange a cancellation of unfilled semi-
soft cargoes from JFY 05. The upshot is that unwanted semi-soft coal often finds its
way into thermal coal market and adds to supply. Felix Resources, a producer of
semi-soft and thermal coal in Australia, recently foreshadowed a fall in profitbecause Asian steel mills were delaying delivery of semi-soft coal. It has been forced
to place the coal with thermal buyers and thus is realising a lower return.
Evidence from Indonesia suggests that the country may be moving to an
oversupply situation. Some producers are already reported to be scaling back 2006
production targets as a result. A balancing factor is the growth of power demand in
the country, which is currently running at about 8% pa. Government policy to
replace oil consumption in power plants with gas and coal is taking hold and
considerable fuel switching is under way, especially at the smaller end of industry
and among the general population. There are three coal-fired power plants coming
on line in 2006 that will lead to an increase in domestic coal consumption of
approximately 7Mtpa and lift domestic sales to more than 40Mtpa. More coal-fired
stations are planned with the stated intention of using Indonesia’s massive low and
medium CV coal resource for domestic power generation.
Most switching from oil to gas and coal in other parts of Asia has already taken
place and coal consumption growth in this area is limited.
Chinese exports are still constrained by the high internal demand for energy.
However, there were reports of cheap Chinese coal being offered to Japanese
buyers late in Q4 05 and this may have been an attempt to win back the market
share that has been lost over the past two years. Moreover, the Chinese coal export
quota has been maintained at last year’s volume of 80Mt.
Figure 256: Actual and forecast Australian Thermal Coal price to Asia
Figure 259: Annual thermal coal exports for the major supply countries
0
20
40
60
80
100
120
140
A u s t r a l i a
C h i n a
C o l o m b i a
I n d o n e s i a
P o l a n d
R u s s i a
S . A f r i c a
U S A
V e n e z u e l a
2004
2005 (p)
2006 (f)
Source: www.coalportal.com.
Exports from Pacific suppliers have continued apace, with Indonesia posting record volumes
and leap-frogging Australia into the top spot for thermal coal suppliers. Australia, too,
exported a record volume of thermal coal but growth has slowed.
Both countries have significant production growth capacity but unlike Indonesia, Australia
is constrained by fixed infrastructure, ie, rail and port facilities. Indonesia benefits from a
large number of floating port facilities that provide flexibility in times of high demand. All of
the country’s main coal producers have embarked on expansion programmes, including
infrastructure support, underpinning the record result. Our data suggests that Indonesia
exported around 117.5Mt in 2005, a year-on-year rate of growth of 13.6%.
The seasonal rains in Kalimantan have arrived, but at this stage appear to be less severethan last year, when production was severely hampered by two bouts of torrential weather.
To date, our information suggests that only the smaller producers on the Mahakam River in
East Kalimantan have suffered significant production losses due to the rain, although we
caution that all operations in Kalimantan are at risk of disruption throughout the first
quarter of the year. In any case, further expansion by the Indonesian majors has been
flagged for 2006 giving the country an export target of around 124Mt this year.
Of some concern to local coal-producing companies is the introduction of a new export
coal levy of 5% to be based on a reference price, mooted to be US$30/t in the first instance.
The workings of the levy remain unclear as Generation I Coal Contracts of Work (the main
Indonesian companies) claim exemption under the agreements they signed with thegovernment in the 1990s. Moreover, the Government wants the right to change the
reference price on a monthly basis, a move being resisted by producers. Presumably, coal
under the threshold is taxed at the threshold, while cargoes above the reference pay the
rate based on their FOB price. To date, the levy has been applied only on a single, 8,000t
barge shipment of coal from the state-run coal miner, PT Tambang Batubara Bukit Asam to
Malaysia.
Australian production is continuing to grow with several new coal projects slated for
operation in 2006 and infrastructure improvements continue to gather pace. In addition,
freight rates have fallen and Australian coal is regaining some of the competitiveness it lost
when transport costs soared in 2003 and 2004. Thermal coal exports are forecast to grow
2.0% between 2004 and 2005 to be 109.6Mt. More significant growth is likely in 2006 and
we forecast thermal coal exports of up to 123.0Mt.
In November 2005, the RBCT in South Africa announced an expansion from its existing
72Mtpa to 92Mtpa. This includes the original Phase 5 expansion (86Mtpa), which has been
under discussion for some time. The total cost will be approximately ZAR1bn (US$150mn),
and the expansion is scheduled for completion by July 2008.
The precise phasing of the incremental export tonnage is subject to colliery and railinfrastructure developments and will be determined once the projects that will use the
increase in terminal capacity have been identified. Initial estimates, however, indicate that
the full rail and mine capacity will only be available after 2009.
The RBCT, which accounted for 97% of total South African coal exports in 2005, exported
19.64Mt during Q4 05 compared with 17.14Mt in Q3 05. The increase in exports was
largely due to the rail parastatal Spoornet's continued improved performance railing
18.04Mt to the RBCT during the quarter. Coal stocks at the RBCT decreased from 3.6Mt to
3.0Mt during the quarter as the terminal increased its tempo considerably.
Total South African coal exports increased by 5% from 67.9Mt in 2004 to 71.1Mt in 2005.
This export total was 5% lower than the target rate of 74.8Mt. Nevertheless, it reflects aremarkable recovery in exports during the second half of 2005 (37.8Mt), considering that
only 33.2Mt were exported during the first half of the year.
The Richards Bay Coal Terminal (RBCT) achieved its highest annual level of coal exports
during a calendar year, exporting 69.2Mt in 2005 compared with 65.9Mt the previous year
(up 5%). The RBCT's previous export record was attained in 2003 (68.3Mt). The port
authority was only expected to export in the region of 68.4Mt, however, the terminal did
not close for Christmas day (normally a 32-hour shut down) as it has done in the past, and
was able to increase loadings substantially, exporting 7.24Mt in December 2005. Ships that
had arrived in port by 31 December were included in the month's loadings.
The terminal increased its rate dramatically during the second half of the year due toincreased railings to the RBCT. Railings to the terminal were hampered severely by two
major derailments, which occurred in March and May 2005, and as a result only 32.4Mt
Figure 260: Russian seaborne Coal shipments by region
Region/Port 2005 (Mt) 2004 (Mt) %
Far East Russia
Vostochny 14.072 14.364 -2.0
Other 0.350 0.819 -0.6
Sub-total 14.429 15.183 -5.0
Northern/Western Russia
Murmansk 10.995 8.850 24.2
Vysotsk 3.499 3.100 12.9
St Petersburg 2.500 2.517 -0.7
Ust-Luga 0.500 0.575 -13.0
Other 0.438 0.638 -31.3
Sub-total 17.932 15.680 14.4
Other Russia 3.098 3.145 -1.5
Baltic Ports
Riga 10.714 9.400 14.0
Ventspils 4.598 3.900 17.9
Tallin 4.086 2.274 79.7
Sub-total 19.398 15.574 24.6
Ukraine
Yuzhny 4.889 4.950 -1.2
Other 5.388 5.388 -
Sub-total 10.277 10.338 -0.6
Total 65.134 59.920 8.7
Source: www.coalportal.com.
Venezuela’s coal exporters experienced a difficult year with lower-than-expected
production and an increase in costs as a result of exchange rate fluctuations, the late
arrival of new mining equipment and the worst rain in five years. Production and
exports at Venezuela’s largest coal mine, Carbones de Guasare’s Paso Diablo was
around 5.9Mt and well below the 6.8Mt target for 2005. Carbones de la Guajira also
experienced problems at Mina Norte. The initial production target for 2005 was 1.4Mt
but was reduced to 1.2Mt then further to 0.85Mt as higher stripping ratios affected
production.
In November, Venezuelan President, Hugo Chavez, announced the approval of funding
and resources for the construction of a deep-water port in the Venezuelan Gulf near the
entrance to Lake Maracaibo. The port will include coal terminals and rail infrastructure
to transport coal from the mines in Venezuela’s southwest to the port. Basicengineering work for the port has been completed and construction is to commence in
early 2006 taking around four years to complete. The project has been around for a
number of years, however, it has only now received the financial backing to allow it to
go ahead.
Colombia’s coal exports in 2005 totalled 58.2 Mt, preliminary data indicate. The results
were lower than expected in part because heavy rain last year disrupted production and
shipments from the Cerrejon and Drummond mines, which make up 80% of output.
Exports from Cerrejon totalled 25.5Mt compared with 25Mt last year, while output at
Drummond this year rose 7% to 22Mt.
A number of new projects commenced in Q4 05, including the San Luis Mine, locatedon a 30-year concession in Santander Department. Coal production began on 4 October
After the onslaught of the major hurricanes at the end of summer 2005, global petrochemical markets went through gyrations to address the acute western shortages
of feedstock and polymers. The reduction of available cracking capacity limited demand
of hydrocarbons, whose production from the Gulf of Mexico had been shuttered
simultaneously. Relative steady overall plastics demand reflect a steady economy, and
meant that trade flows made up for shortfalls, but ultimately price spikes rationalized
demand accordingly. Since then, individual markets have continued in their dynamic
paths, but have reduced the price excursion seen in Q4 05. In North America, polymer
demand and production have both gone through some recovery, and lower feedstock
prices means plastics pricing has moderated. A positive outlook remains for the rest of
first quarter, in our view. In Europe, stronger demand and rising feedstock values have
left polymer prices rising. Asian polymer prices have also been rising but demand hasmoderated, coincident with annual holiday season. Energy costs have diverged, allowing
natural gas based production to benefit more than naphtha-based production. The
larger debate emerging is whether the strong trends currently underpinning the
petrochemical sector will persist through 2006 or if the peak is in fact being seen
currently, and the rest of the year is downward. We believe that strong economics and
firm energy prices will be supportive to polymer pricing through the first half of the year
at least.
The key issue on the supply side for 2006 in plastics will be the direction of oil and gas
prices. The hurricanes highlighted the impact and problems associated with these fuel
prices on the petrochemical sector. This was further complicated by the temporary lossof capacity. Since Q4 05, oil prices have diverged from natural gas prices, especially in
North America, where a very weak winter has brought natural gas price back to levels
last seen in mid-summer 2005, 50% lower than October and December highs.
Meanwhile, crude prices have stayed above $60, close to their strongest nominal levels.
These volatile and uncorrelated price movements have a significant impact on the
industry, which regionally creates polymers from different feedstock sources. More
specifically, for natural gas, 2005 highlighted the regional differences in gas values and
the acute issues that could confront each individual market. While gas prices could soar
in the US, they did not match values seen in Europe during the cold spell, nor Asia,
seeking spare LNG cargoes. This only heightens the eventual trend to site new cracking
capacity in Middle East, closer to the source of cheaper hydrocarbons. Relatively stronglight naphtha values have shown little variation in values among the continents, but the
price direction has been steadily higher, leaving all markets over $500/mt since August
2005.
This price action which has moved higher on all fronts did impact demand, temporarily,
but end-2005 company results indicate that end-use customer demand has remained
firm for most products with strong economic conditions boosting consumer spending.
The first half 2006 will likely continue this strong demand trend, in our view, requiring
better capacity utilization, and absorbing green-field start-ups in Asian markets.
Inventories can thus be rebuilt.
US industrial production rose in October, November and December at a gradual pace,but still achieved a 3.8% annualized growth rate for the quarter. Durable goods
manufacturing rose at a slower rate, with the decline in motor vehicle and parts output
of 2.8% weighing down the average. This is significant for the plastics industry, given
the growing content of polymers in auto manufacture. Cost cutting by the
petrochemical producers – a skill further honed during energy price spikes – should
continue to pay off during the first half of the year, allowing margins to stay wide. Price
signals from relatively tight feed-stock markets should provide support to polymer
prices during the first quarter and second quarters, in our view.
Monomers
In the US, the rotating position of which feed-stock provides the greatest cost
advantage for polyolefin production has been the theme since the summer 2005. The
integrated nature of petrochemical industry along the Gulf Coast dictated that oil and
gas market trends are immediately translated into monomer market trends. The
shortage of hydrocarbons also forced the relative shortage of ethylene – which then
sent spot values soaring – and compressed margins of ethylene buyers. The additional
response from the market was to increase imports of ethylene or ethylene-based
products to avoid having to pay higher prices, whether by contract or spot terms. Spot
ethylene values in the US were advantaged by almost 10 cents during Q1 05. Thisspread disappeared in July and then reversed to almost 20 cents in favour of western
European supply by November. This spread held through December but has since
reverted to just a small US premium. Of course, during this time natural gas prices
surged to record levels – above $15.00 – in prompt futures month pricing. Relative to
light naphtha based manufacturing, US natural gas values were disadvantaged by at
least 10 cents at 30 cts/lb; a lack of industrial demand due to price elasticity or
otherwise, helped clear the market. Now light naphtha is almost 10 cents over natural
gas which is the more economic feedstock.
Figure 267: US ethylene supply and demand, mn lbs
12,000
12,400
12,800
13,200
13,600
14,000
14,400
14,800
15,200
Q1 '04 Q3 Q1 '05 Q3 Q1 '06
Supply
Demand
Source: CMAI, Barclays Capital.
Figure 268: US propylene supply and demand, mn lbs
7,500
8,000
8,500
9,000
9,500
10,000
Q1 '04 Q3 Q1 '05 Q3 Q1 '06
Supply
Demand
Source: CMAI, Barclays Capital.
For the macro balances, the H2 04 left US ethylene fairly balanced between supply and
demand, but tilted in favour of production for H2 05. Higher energy prices lowered
incentive for production during Q3 05, even before the hurricane events. Both aggregates
fell precipitously but demand remained above supply through the end of the year. Firm
demand is set to lead supply higher during the first half of the year, as output regains Gulflosses. However, a rash of other acute outages has prevented a swift return of multiple
ethylene units. Notably the Nova facility in Ontario, Canada has experienced various issues
with restart, down originally in November. Compressor seal leaks kept the restart from full
realization, and the unit has only just come back online, allowing the company to recall
force majeure. Still idled units in Mexico have staggered restart plans, with only one certain
for H1 05. Along the Gulf Coast, at least 8 units were recently offline, half due to
maintenance issues, according to the CMAI. In our opinion, February is showing signs that
continued maintenance issues may keep substantial capacity offline.
The start of 2006 revealed that Europe would have its share of capacity problems,
overshadowing the challenges in the US. No less than seven feedstock units went down
across the continent for unrelated and diverse reasons. While two of the outages were
planned for maintenance, even those developed complications which made the outages last
longer. The resulting impact was to kick spot ethylene values higher, squeezing cash
margins for the spot buyers. The trend gathered from uneven production performance has
now provided support for feedstock prices well into the first quarter of the year, even
though stocks were being rebuilt right up till H2 05. Coincidently, the firm or rising
petroleum price environment had similarly impacted naphtha values and thus now have
preserved margin for feedstock suppliers as well.
Asian price trends also reflected firm demand, as spot quotes went higher than US andEuropean markets. Some large crackers had gone down for unscheduled reasons, but more
generally the planned turnarounds had the impact of accelerating buying of prompt
supplies by producers to cover the reduction of operating capacity. The impact on market
prices should be limited due to the already stated reasons, as long as all goes well through
the scheduled downtime. The other factor that provides resistance to Asian markets
moving higher is the additional production capacity coming to market from the
CNOOC/Shell Petrochemcials joint venture at Daya Bay. This plant was completed in
December, and will feature 800,000 tonnes per annum of ethylene and 430,000 tpa of
propylene, from condensate or naphtha cracking. Mid-January start-up led to early
February operations reaching 50% of nameplate capacity
The Middle East has been the region that has aggressively expanded ethylene capacity
during the past decade, doubling its share of world capacity to 10%. The pace of expansion
should continue through the rest of this decade, putting aggregate capacity on a more
competitive size with the large North American markets. Importantly, the marginal supply
of product will be emerging from this region more frequently.
Figure 282: US olefin refinery production, ’000 bbls
-
50
100
150
200
250
300
350
Jan 04 May 04 Sep 04 Jan 05 May 05 Sep 05
-
1,000
2,000
3,000
4,000
5,000
6,000
7,000
8,000
9,000
10,000Ethylene
Propylene (RHS)
Source: EIA, Barclays Capital.
Polymers
Like the monomer market trends, the current market conditions that polymer products face
stem from the hurricane recovery. During Q4 05 prices spiked, reflecting damage to the
supply chain but continued demand for products. The price-elasticity effects followed, and
demand levels for products sank through to the end of the year. January has witnessed price
recovery firstly in polyethylene products and more recently in polypropylene. Elevated
energy prices, especially in the petroleum-based feed-stocks, will provide support to prices
during the first and second quarters of 2006. Natural gas prices could see further weakness
during H1 06, but this will help provide margin to ethane-based producers, while higher
priced naphtha production stays online, encouraging higher market prices. A strongrefinery turn-around season implies to us that the supply side is tightly balanced.
Momentum in the US economy and globally should provide adequate demand to at least
match supply growth. Trade flows will likely revert to more typical patterns as the premium
market in the US rationalizes, then signalling for imports decline from the strong Q4 05
levels, and allowing US exports to expand again.
Important new capacity comes online in parallel to the monomer production at the
CNOOC/Shell j/v with 240,000 tpa of polypropylene, 250,000 tpa of LDPE and 200,000
capacity of either LLDPE or HDPE. This follows two major polyolefin capacity additions
during the past year: 400,000 at Yangzi and 200,000 at Daqing. Further capacity
development is anticipated in both the Middle East and Asian markets, leaving mature
North American capacity to see incremental growth primarily through upgrades.
Polyethylene
The lower prices rolled over into the first month in 2006 – trending down from the heights
of Q4 05 – have stabilized, and prices going into February have responded to more demand.
The higher prices over much of the petrochemical sector during the end of 2005 no doubt
held back the price-elastic demand, and the emergence of lower energy prices may have
spurred demand into action. In advance of the refinery turnaround season, there may be
some effort to secure supply, making up for the relatively light inventories on hand. Priceincreases are intimated for March contracts, even though energy values may point to spot
price weakness.
For European markets, there has been a strong start after the weak end-of-year markets.
Demand has been steady and warranted price increases from major producers. Helping to
tighten the balances are the plant outages on the continent, noted previously.
Asian markets are typically quiet during the lead-up to the Lunar holiday break. Some
concern about supply lingered as feed-stock production problems surfaced, but more
common was product on-offer from multiple sources helping to hold prices in check. The
eventual start-up of new capacity could further alleviate any tendency of prices to bound
higher during the quarter. For Chinese import data through November 2005, strong
increases were noted for the LL and LDPE product, especially gains made by Saudi Arabia
and the US. The surge in Middle East imports will likely be a recurring theme as new capacity
comes online in the medium term, advantaged by cheaper feedstock than found elsewhere.
Figure 283: US polypropylene supply/demand, mn bls
4,100
4,200
4,300
4,400
4,500
4,600
4,700
4,800
4,900
Q1 '04 Q3 Q1 '05 Q3 Q1 '06
Supply
Demand
Source: CMAI, Barclays Capital.
Figure 284: US polypropylene exports, ’000 metric tons
Base metals: Yet higher The outlook for base metals prices remains extremely constructive. After very strong
performances again in the first month of 2006, we see the price up-trends continuing,
with particular upside pressure in H1 06. We remain most positive on zinc for 2006,
followed by aluminium and copper and lead, respectively (compared with 2005
average prices). We also see nickel and tin prices reversing their falling price trends of
H2 05 (see Figure 317 for details). We have made some adjustments to our forecasts
since our last review on 6 January. Far forward contracts also continue to offer value,
in our view, especially in zinc. In nickel, we expect a nearby backwardation to return as
the market starts to tighten from its current surplus.
Base metal prices have risen for four consecutive years. Copper, zinc and lead
continue to hit new all-time highs, while aluminium is around its highest levels
since the late 1980s. Against general expectations, high prices have failed to rapidly
attract new production. At the same time, the global demand environment has
remained strong. This powerful combination, as well as extremely low global metals
inventories, will generate even higher prices in 2006, in our view.
The broader macroeconomic backdrop remains extremely positive for the
industrial metals, with a low interest rate environment, robust global growth and
prospects of a weaker dollar. The sceptics continue to cite rising interest rates,
downside risk to the US economy and subsequent risk to the China economy.
However, we see these fears as overdone; we believe China’s metals demand will
continue to grow strongly even in the event of an external economic slowdown, due
to domestic infrastructure-led investment. Strong demand from other emerging
markets, for example India, is also becoming increasingly important.
In a strong demand and price environment, production should grow. But serious
short- and long-term supply constraints persist, and production costs are on therise. We believe the risk of a sudden near-term flow of new production causing a
correction in prices is therefore small.
Fund involvement is also widely regarded as a threat to the sustainability of high
base metals prices. Indeed, long-term investment money continues to flow into
commodity markets, including the base metals, and this trend shows no signs
of abating.
Shorter-term fund activity, however, is generally not excessive on the long side. As
a result, we see little risk of substantial fund-driven corrections in prices. In the
event of downside price pressure, it is likely to be shallow and brief once again,
given the amount of buying interest from consumers and funds on lower numbers.
Figure 317: Barclays Capital base metal price forecasts (US$/t)
Our visit to China at the end of November 2005 was timely. It came at the height of
widespread interest in the copper market on rumoured large short positions held by
Chinese traders, possibly involved with the Strategic Reserve Bureau (SRB). While SRB
activities remain secretive, albeit arguably less so now that they claim they have up to
1.3Mt in stock, they are likely to remain a big talking point going forward. However,
here we leave speculation aside and focus on underlying supply and demand
fundamentals which, after all, are the reasons for the strength in metals prices.
In 2005, there was a ferocious battle between the bull and the bear in the metals
market (particularly in copper). The bears pointed at slowing demand and the likely
return to surplus markets in 2006, while in contrast, the bulls were pointing at low
inventories, as well as highlighting the potential for supply to disappoint and demand to
surprise on the upside – a scenario that materialised in H2 05.
Figure 324: The metals cycle – reached the peak?
20
30
40
50
60
70
80
90
100
110
120
130
140
150
Jul 69 Oct 76 Jan 84 Apr 91 Jul 98 Oct 05
Nominal base m etal
price index
Source: Barclays Capital.
Figure 325: Adjustment to a higher price environment
necessary
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
5,000
2000 2001 2002 2003 2004 2005 2006
High
Low
Mean
Actual
US$/t
Copper forecast survey
Source: Reuters, Barclays Capital. Note: Survey conducted in Jan 2006.
Metals prices are high in a historical context; nominal metals prices are trading around
all-time highs again at the start of 2006 (see Figure 324 above). In fact, the current
uptrend in base metal prices has now lasted the same length of time as that seen during
the late 1970s (about 48 months), and has been just as sharp as the one in the late1980s. The current metals cycle is breaking away from the traditional pattern seen over
the past 20 years, when cycles were becoming shorter. However, high prices in
themselves are not a good reason to adopt a negative outlook.
There is no strong consensus on future metals prices. In fact, the spread between the
highest and lowest analyst price forecasts is very large again for 2006. With global
metal inventory levels being so low, only small differences in one’s assumptions on
supply and demand can have large consequences on the price conclusion. Figure 325
above shows the latest half-yearly Reuters forecast survey for copper, conducted in
early January, and the situation is similar for other base metals. In 2005, even the most
bullish forecaster was too cautious on prices, and judging by the start of 2006, thiscould be repeated again in 2006. Upward revisions to price estimates, and adjustment
to a higher price environment, are still necessary.
Adjusting for inflation, and looking at base metal prices in real terms, prices are not
that high (see Figure 326 below). As an example, they are still some 40% below the
peak in the late 1980s. This points to the potential for substantial upside. Figure 327
shows in broad terms how we see the metals cycle developing over the longer term.
Metals prices should maintain a cyclical pattern, but fluctuate around a higher average
than in the past, raising the floor for future metals prices due to structural changes
related to underinvestment in supply and a higher trend rate in demand growth. Wealso believe there is more room on the upside in this cycle, before positive cyclical
factors ease back in strength and more supply comes on stream.
Figure 326: Real metals prices bottoming out
50
75
100
125
150
175
200
225
250
275
300
325
350
Jul 69 Oct 76 Jan 84 Apr 91 Jul 98 Oct 05
Real base metal
price index (current US$)
Source: LME, Barclays Capital.
Figure 327: Where are we in the metals price cycle?
Here?
Source: Barclays Capital.
Over the past year, the strong divergence in base metals prices and key leading demand
indicators, as seen in Figure 328, has been a disturbing phenomenon for many. The
slowdown in the demand environment helped fuel the bears. So why did metal prices
not follow the lead of these indicators? We believe there are several reasons, which all
form the basis of our conclusion on the outlook for metals prices in 2006:
Economic growth stays positive; base metals prices corrected temporarily only
when, for example, the OECD leading indicator moved into negative territory
during Q2 05.
Strong demand from China; while consumers refrain from buying as much aspossible due to high prices, the backwardation and difficulties in passing higher prices
on to their customers, the underlying trend of demand remains strongly positive,
partly due to infrastructure expansion driven by the vast urbanisation process.
Supply constraints; production estimates keep being revised lower due to
important short- and long-term cost-related issues.
Low inventories; total reported base metal stocks are around all-time lows, and stock-
In the euro area, economic activity is finally starting to improve, allowing the European
Central Bank to hike interest rates for the first time in five years at the end of 2005.
European consumer confidence has improved markedly as a result, fears over
unemployment are falling, and the German industrial activity is picking up pace. In
Japan, our economists expect the economic recovery to stay firmly intact in FY 06, with
real GDP growth of 2.3%. In fact, they see Japanese demand pushing the economy
toward its most prolonged expansion in the post-war period. We feel industrial metalsprices are unlikely to fall back in this positive economic environment.
Still, metal-specific demand statistics were discouraging for most of 2005. No wonder
some (read: most) market participants took bearish views on prices. Copper
consumption in the US, for example, slowed sharply during 2005, much more so than
economic activity itself (see Figure 331 below). The picture is similar when looking at
other metals, also in other regions. It is really only “emerging markets” such as China
and India that have seen metals demand growing over the previous year. Several factors
are behind these poor readings for metals demand:
Slowing orders: In line with a slowdown in the economy, metal orders also slowed
in 2005.
Substitution: Consumers are moving away from metals to other materials wherever
possible (away from copper in piping and roofing for example). While substitution
opportunities are still relatively limited over the near term because of quality issues
and high prices of other commodities, it could become a more important issue over
the medium term (two to three years) as manufacturers develop products
containing less metal if they can.
Larger scrap usage: The scrap ratio at semis manufacturers is increasing. In China,
scrap availability is set to increase as the government is stipulating policies to
encourage recycling. However, scrap availability and stocks are low today, after a
large surge in scrap supply to the market at the initial stage of the price rally.
De-stocking: Importantly, this was the biggest contributing factor to poor demand
statistics in 2005. Consumers have been using their inventories rather than buying
new material. Consumer stocks have fallen to very low levels as a result, and
consumers will be forced to buy at higher-than-desired levels in 2006 if economic
Given a strengthening of economic activity, heavy inventory drawdowns (de-stocking)
by consumers in 2005 is likely to have important implications for availability and prices
going forward. While consumers are unlikely to aggressively replenish their stocks in
the near future due to steep backwardations (implying high financing costs), they are
likely to support prices on any weakness, or take advantage of favourable currency
moves (eg, dollar weakness against key consuming currencies). There is already
evidence of an advance in copper-specific demand statistics. The US Copper & BrassServicenter Association reported that shipments of copper products in the US rose a
hefty 5.3% Y/Y in October 2005. In Japan, copper wire and cable shipments rose by
5.8% Y/Y in December.
Furthermore, China’s metals demand growth has been underestimated in recent years –
not only by metals analysts, but more importantly by producers. We believe China’s
copper demand potential in coming years is still being underestimated. We remain
long-term bulls on China metal demand, even though temporary periods of slowdown
might occur on the way. Figure 332 below shows the low level of metal consumption
per capita in China still, compared with its neighbouring Asian countries. This is despite
the fact that China is already the world’s largest consumer of base metals.
Figure 332: China – Strong demand potential
0
5
10
15
20
25
30
4 8 12 16 20 24 28
$ '000 GDP per Capita (PPP)
C o p p e r C o n s u m p t i o n ( k g / p e r s o n )
Japan
South Korea
Taiwan
China
Source: Brook Hunt, University of Pennsylvania, Barclays Capital.
Figure 333: China to drive above-trend growth in demand
-800
-600
-400
-200
0
200
400
600
800
1,000
1990 1993 1996 1999 2002 2005 2008
China US
Europe Japan
Growth
Copper, Kt
Source: CRU, Brook Hunt, Barclays Capital.
As in the western world, many Chinese market participants are puzzled over ongoingprice strength, as they see consumers being reluctant to buy at high prices, and see
their difficulty in passing on higher prices to customers. Large price differentials
between the LME and SHFE prices have also caused scepticism, encouraging short LME
and long SHFE positions in copper. And there are large concerns over potential negative
effects on the Chinese economy from an eventual slowdown in international economies.
However, we believe the strength in China’s domestic metals demand, which is driven
by the infrastructure spending needed to support the vast urbanisation process, is still
being underestimated. In addition, rising incomes are supporting consumer demand for
goods containing metal, including cars, air-conditioners and electronic products.
As a result, once the western world economies slow, we expect China to remain a key
driver of global metals consumption (see Figure 333 above). China’s economic growth is
powering ahead robustly. 2005 saw another outstanding year of economic performance
in China, with GDP growing by 9.9%. Our economists point out that China has now
surpassed the UK and France to become the world’s fourth largest economy. In
addition, China’s economy has been moving towards a more balanced growth path, and
is gaining new resilience. Thus, our economists believe China will be able to maintain a
high level of growth rates, although they may moderate somewhat this year.
But it is not only metal consumption in China that has a positive outlook. Indian metals
consumption is also growing much faster than anticipated, albeit from a lower base and it
will take some time before it has the same impact on global demand as China has today.
Nonetheless, the Indian Government has plans to spend some US$350mn on domestic
infrastructure over the next few years. At the same time, earnings growth of the Indian
middle class is rising fast, while the quality of earnings has also improved. The rapid
Indian growth is also reflective in a shift in the trade of steel; previous export volumes are
now being absorbed internally. European steelmakers have indicated that they see growth
in the Indian construction sector of at least 10% per year going forward.
Supply fails to grow sufficiently fast
To meet these strong demand conditions, a response from metal producers is necessary,
and high prices are required to attract investment in new output capacity. The basemetals markets are still feeling the effects of underinvestment in output capacity in recent
years at a time when prices were low. Mining exploration spending has picked up. The
Minerals Economics Group estimates total exploration budgets have increased by a
further 40% in 2005, to US$5.1bn, after an already large rise in the previous year (see
Figure 334 below). While the vast majority of total spending is still in gold, copper and
zinc are two markets where exploration spending has risen especially fast (by some 60%
and 90% Y/Y, respectively).
Figure 334: Bullish supply prospects to be realised?
0
1,000
2,000
3,000
4,000
5,000
6,000
78 81 84 87 90 93 96 99 02 05
100
200
300
400
500
600
700
800Mining Exploration Spending , US$M (LHS)
CRB Metals Index, Current US$ (RHS)
Source: Mineral Economics Group, Barclays Capital.
Figure 335: Downward revisions to supply
14,600
14,800
15,000
15,200
15,400
15,600
15,800
16,000
2005 2006
Sep 04Mar 05Today
'000
tonnes
ICSG mine output
forecast m ade in:
Source: ICSG, Barclays Capital.
Supply growth continues to disappoint, and we would argue it is now taking longer than
ever to bring capacity on stream. In copper specifically, it was the shortfall in output
against very low global inventories that was a key reason for the ongoing strength in pricesin H2 05. Output growth forecasts, by the International Copper Study Group, for
example, keep being downwardly revised (see Figure 335 above).
Remarkably, at a time when copper prices keep pushing fresh all-time highs, output from
some of the world’s largest copper producers has been declining. The Chilean Copper
Commission has also been forced to revise lower its forecast for total Chilean copper
production, and sees no growth in Chilean copper output this year. Elsewhere, major
producer Freeport-McMoRan is reporting copper output from its vast Grasberg mine
(second largest in the world) in Indonesia to be running below expectations because of poor
ore-grade quality – its ore grades fell for three consecutive quarters during 2005. The slowramp-up in smelter capacity expansions (eg, India) is further enhancing tightness in refined
copper supply.
The strong relationship between metals prices and the Caterpillar share price is no
coincidence (see Figure 336 below). Reporting a 54% rise in Q4 profits, and raising its
2006 outlook, Caterpillar, the world’s largest maker of earthmoving equipment, say it is
experiencing “unprecedented customer demand”. While Caterpillar has raised prices of
its equipment (+5% last year), its sales are up 60% over the last couple of years. Its
mining equipment products are sold out through 2007, though pointing out that it
could have responded faster to increased global demand if components were more
readily available. Caterpillar said the current cycle looks stronger than upturns in the
late 1980s and 1990s. To us, this provides further firm evidence of the severe supply
constraints in the metal markets.
Figure 336: Surging demand for construction and mining equipment
1,500
2,000
2,500
3,000
3,500
4,000
4,500
5,000
Dec 02 Jun 03 Dec 03 Jun 04 Dec 04 Jun 05 Dec 05
20
25
30
35
40
45
50
55
60
65
70
Copper price, US$/t (LHS)
Caterpillar share price, US$ (RHS)
Source: Reuters, Barclays Capital.
At the same time, budgets for new projects are being revised up. The increase in
operating and capital costs of commodity producers is becoming an increasinglyimportant supply side constraint (see Figure 337 and Figure 338 below). Almost every
component of costs, from labour through to energy and materials, has escalated
dramatically over the past two years. In addition, copper output has suffered from
higher mining of ore-grades rich in by-products fetching even more attractive prices
Estimated increase in copper mine construction costs
(2005 vs. 2004)
Source: Brook Hunt, Barclays Capital.
The reality today is that while the trend rate of demand growth is on the rise, there are
few large metal deposits being developed for the remainder of this decade. The average
size of planned projects is smaller, and higher cost. The trend of supply constraints is not
isolated to copper. Aluminium smelters are struggling from high input costs, due both to
energy and alumina. Zinc smelters are also extremely short of feed, likely to last into
2007, while nickel mine output growth is also particularly low in relation to price levels.
With no signs of the infrastructure-led expansions in emerging markets slowing
significantly over the next few years, producers’ focus on cost control rather thaninvestment is pivotal. Indeed, Rio Tinto reported record capital expenditure of US$2.5bn
for 2005 (and record revenues), and said this figure is expected to growth further in
2006-07. It has also expressed a long-term commitment to exploration by a joint-venture
with Norilsk Nickel. But, Rio Tinto’s capex was up by less than 20% from last year, and its
share of revenues has hit the lowest over the past 12-year period (see Figure 339),
highlighting the relatively low level of investment in new projects in relation to demand
and price conditions. The lack of sufficient output growth is set to continue to underpin
record high commodity prices for some time to come, in our view.
The combined effect of these supply and demand conditions is low inventories. In
copper, total stocks measured as weeks of consumption have fallen below a “critical
low” of four weeks (see Figure 340 below). We do not see stocks returning to more
comfortable levels until 2007. Sheer physical tightness has the power to keep prices
strongly supported even when demand slows and production grows. To quote producer
Xstrata: “prices are being underpinned by chronically low stock levels around the
world”. We expect this situation to persist for some time to come.
Global zinc stocks are also moving rapidly toward “critical lows”. In aluminium, we see
relatively better availability overall, with some material entering LME warehouses
(although partly delivered for coverage of short positions). But with the alumina market
set to stay tight for a large part of this year (H1 at least), we expect renewed drawdowns
in aluminium stocks. Moreover, against poor demand from the stainless steel sector, LME
nickel stocks have been rising during H2 05, but we see this trend reversing again in H106 when we expect improved demand from the stainless steel sector.
Modest fund length is evident not only in the relatively low level of LME open interest
for most base metals, but also from a small net long non-commercial (or speculative)
position in copper on Comex. Indeed, both the gross long and short has risen, but in
today’s marketplace we would still regard the risk for short-covering as much larger
than for long liquidation, given the resilience in prices and a steep backwardation (see
Figure 341 above). Arguably, LME metals prices have received an extra push higher
lately from forced short-covering (eg, copper and lead).
Remarkably, consumers were also seen entering short positions in mid-2005, believing
the ultimate cyclical peak had been reached when prices were correcting during Q2 05.
And producers, never believed prices would be even higher in 2005-06, compared with
an already strong 2004, hence they produced and sold as much as they could, and have
been unable to supply any extra material into renewed price strength.
Investment in commodities as a long-term strategy designed to diversify portfolios, has
become very popular (see Figure 342 below), however. And there are signs that
investors plan to significantly raise their exposure to commodities over the next three
years. Indeed, investors are concerned about current high price levels and plan to
reduce exposure to index-based products, substituting less directionally sensitivestructured products such as commodity baskets and becoming more active managers
of their commodity portfolios. These are the key results of a survey of US investors
carried out at Barclays Capital’s first annual US Commodity Investor Conference in New
York on 12 December 2005, attended by more than 70 institutional investors.
Following on from an earlier event in Barcelona, the opportunity was taken to repeat an
audience survey carried out at the earlier conference. On average, at the New York
conference, roughly two-thirds of the respondents already had some commodity
exposure, but for most this was 5% or less of their total portfolio. However, over the
next three years the survey results showed that almost 70% of the respondents
expected to increase commodity exposure to 5% or more of their portfolio. This patternof increasing commodity exposure was also evident in the survey carried out at our
Barcelona conference (see Figure 343).
Figure 342: Sharp rise in commodity-linked mutual
funds
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
01 02 03 04 05
Credit Suisse Commodity Return Strategy Fund
Merrill Lynch Real Investment Fund
Oppenheimer Real Asset Fund
PIMCO CommodityRealReturn Strategy Fund
Assets under management ($mn)
Source: Barclays Capital.
Figure 343: Investors to continue raising exposure to
commodities
0%
10%
20%
30%
40%
50%
60%
70%
Zero 1 -5% 5 - 10% Above 10%
Barclays Capital New
York Commodity
Conference -
December 2005
Barclays Capital
Barcelona
Commodity
Conference -
Februar 2005
Audience response to the question: "How much of your
portfolio will be in commodities over the next three years?"
The result was consistent with the audience’s view that large fresh inflows of
institutional investor flows into commodities are likely over the next three years, with
54% expecting funds under management in the sector to increase from $70bn
currently to between $90bn-120bn by 2008 and 32% expecting a figure even greater
than $120bn by then. Almost half the audience expected to hold their commodity
exposure for three years or longer with most (63%) citing portfolio diversification as
the most important factor in the decision to add commodities to their mix of assets.The major concern of investors was current high commodity prices (58% of the
audience cited this as their number one concern). For this reason, only 11% of the
audience expected to invest in commodity indices over the next three years, while 68%
expected their investments to take the form of a combination of different strategies
including active management and structured commodity products.
Conclusion
2006 is set to be another strong year for base metals prices. The serious supply
constraints evident through 2005 will not be swiftly resolved. We believe any
meaningful slowdown in China's metal consumption is unlikely, given expansion of the
power network and infrastructure in response to the vast urbanisation process. At the
same time, the outlook for western world metals demand is positive – we believe that
continued robustness in US industrial production, and strengthening activity in Europe
and Japan, will force renewed consumer buying and attract fresh fund involvement on
the long side.
In addition, we would regard the prospects for a weaker dollar this year as a big positive
for metals. First, US dollar depreciation is likely to attract investment money into assets
with traditional negative correlation to the currency. Indeed, dollar strength (against
the euro) last year failed to pressure metals prices, but only because other metal-
specific fundamental factors were sufficiently forceful, such as supply disruptions.
However, at a time when consumers are “under-hedged” and when producers arealready facing upside pressures on costs, we believe dollar weakness is likely to have
significant positive implications for metals prices.
AluminiumEven though aluminium prices at the start of 2006 are testing their highest levels since 1989, we maintain a positive
view on the price outlook. Despite the strong price environment, output curtailments are occurring at aluminium
smelters across different geographical regions. At the same time, demand conditions remain healthy. This should
push the global aluminium market deeper into deficit this year, and cause renewed inventory drawdowns. Recent
LME inflows, we believe, are largely in response to short position holdings, hence not a reflection of an easing of thesupply deficit.
Ongoing rises in production costs are making aluminium smelting unprofitable in many traditional producing
regions. As a result of high energy prices, Alcoa has already announced the closure of its 195Kt/y Eastalco smelter in
the US. And in Europe, three smelters will face closure in the period 2006-07 (112Kt/y Hamburg, 70Kt/y Stade and
50Kt/y Lannemezan). We estimate another approximate 800-900Kt/y of capacity could be at risk for closure over
the next couple of years (with the majority in Europe) as long-term power supply deals come up for renewal.
In China, now the world’s largest primary aluminium producer, it is the high alumina spot price (about US$600/t cif)
that is of particular concern. To combat high raw material costs, Chinese smelters have, in a joint effort, reduced
output by 10%, or about 350Kt/y. This has enabled previously strong growth in China’s smelting output to slow, to
about +9% compared with growth of about 15-20% in H1 05. A direct consequence of this, together with a rise in
export tariffs and strong domestic demand, is that China’s exports of primary aluminium are sharply reduced, and
this is helping to keep the western world market tight.
While new smelting production is starting up in regions with readily available, and more affordable, power supplies,
future alumina availability will be key to the outlook for aluminium, in our view. Alumina refineries have (like with
other commodity production) been hit by several negative factors such as technical problems, strikes and high
natural gas prices. With maximum capacity utilisation rates at refineries and now very low alumina inventories at
Chinese smelters, we expect alumina prices will remain well underpinned for some time to come. Medium-term
contract prices are currently settled at a much higher percentage of the LME aluminium price (16-18%), compared
with in the past (typically around 12-14%), while one-year contracts have been reported at around 21-23%.
Demand conditions (usually the prime driver of the metals price) are generally positive for aluminium. US orders are
not spectacular, but robust, and physical spot premiums have firmed. While there are strong signs of improvingdemand in both Europe and Japan, aluminium consumption in China continues to grow strongly, driven by the
construction and transport sectors. While partly satisfied with secondary material, China’s aluminium semis output
registered a new record high in December (609Kt), and was up 29% in 2005 as a whole, to 6.0Mt.
Figure 356: Barclays Capital global supply and demand summary for aluminium
Copper Just because prices are already trading at record highs, does not mean they cannot go higher still. Low inventory levels
form the basis of our continued positive outlook on copper prices; however, the combination of low interest rates, robust
global growth and a weak US currency creates an ideal macroeconomic environment for rising metals prices. Furthermore,
producers remain constrained and are still unable to respond fast enough to rising prices. Consumers are no longer
protected with inventories, and are showing good buying interest on any price weakness. Many blame high prices oninvestors. Indeed, in this environment, funds are forced to cover any outstanding short positions, while the longer-term
investment flow remains positive.
Copper consumption was weak overall in 2005 as consumers de-stocked, substituted copper for other materials if they
could (mainly in plumbing applications) as well as used more secondary material. In a strong global economy, we expect
low consumer stocks to help boost demand numbers this year. However, the backwardation and high prices are likely to
prevent any aggressive restocking activity. We expect substitution to become an increasingly important issue over the
medium term (next two to three years), but for now, we believe demand destruction is limited.
We are especially upbeat on demand prospects in Europe and Japan this year, while we expect to see manufacturing
activity in the US continue progressing solidly also. Even if growth in the western world was to fall back below our
expectations, we expect total demand conditions for copper to remain very healthy, driven by the growth in emergingmarkets. China plans to spend Rmb800bn (almost US$100bn) on upgrading its electricity network over the next five years.
And in India, construction activity is growing by some 11% pa. While India still only represents less than 5% of total copper
demand, it is becoming an increasingly important force in the structural bull trend for metals prices.
This positive demand picture is against a trend of ongoing downward revisions to supply forecasts, driven by falling ore-
grades, technical problems, labour disputes etc. In addition, expansions are taking longer due to rising costs related to
energy, steel, labour, water etc, while new mine deposits are of much smaller scale than in the past. And critically, over the
near term, smelting capacity remains constrained. Continued drawdowns of refined inventories are the result, and we do
not expect total reported inventories to return to more comfortable levels (above four weeks of consumption) this year.
Everyone wants to know about the impact of investor activity in the metals market. First, tactical hedge funds hold a near-
neutral net position in copper, we estimate, hence the risk for a sharp price correction on a turn in sentiment is limited.Second, long-term investor interest in commodity markets shows no signs of slowing. While copper accounts for a
relatively small share in the broad-based commodity indices (e.g. 2.74% of the GSCI), it tends to attract a relatively larger
weighting in commodity baskets, due to its attractive forward curve (backwardation). These positions are often based on
strategic, long-term, decisions to diversify portfolios. As a result, once the positive cyclical factors ease back in strength, we
expect structural supply and demand issues, as well as this investor money, to continue to provide solid support to
especially the back end of the forward price curve.
Figure 375: Barclays Capital global supply and demand summary for copper
LeadThe lead market provides a prime example of how prices can rally sharply higher in the event of supply disruptions
due to critically low inventories. Being one of the smaller contracts on the LME, this is also a market that has been
largely ignored by the large flows of strategic fund money. Indeed, in reality, the lead market has been the subject of
short fund selling in response to record high prices. This sparked a sharp short-covering rally at the start of the year
as supply concerns remerged at a time when demand has remained very strong. The return to a surplus market willtake longer than previously expected, in our view.
Environmental issues have once more moved to the forefront in the lead industry after the Shaoguan smelter in
China closed down in mid-December following a toxic spill into the Beijing River. It is unclear how soon the 100Kt/y
lead operation will restart commercial production, but unlikely before June. The Zhuzhou Smelter (95Kt/y of lead)
also halted output temporarily at the beginning of the year, also due pollution fears. In an attempt to protect the
environment amid a booming economy, the China State Environmental Protection Administration said it will demand
officials to report toxic spills within an hour, or they will face criminal prosecution.
Labour disputes can also be expected to reoccur in the metals markets this year, as workers’ unions become more
demanding partly in response to record high prices. Workers at Peruvian lead and zinc miner Volcan were protesting
at the start of the year against the company’s plans to outsource labour, and are threatening more strike actions.Despite recent disruptions to smelting output, spot concentrate treatment charges have fallen back below US$100/t,
highlighting a tight global lead-in-concentrates market. Of key interest will be the pending government approval for
the MacArthur River mine to move from underground mining to open pit. The approval has been delayed and the
mine has stockpiled concentrates only until April. Given the large size of the mine, about 90Kt/y of contained lead,
any further delay could have dramatic market implications. The US market is particularly tight of refined supply,
evident from low exchange stocks and rising physical spot premia.
This is occurring amid reports of very strong demand conditions at the start of 2006, driven by industrial battery
manufacturing in the US, while there are signs of consumer re-stocking in Europe. Compared with other base metals,
lead demand is typically more dependent on the season than the economic cycle, given its large exposure to
replacement batteries and consequently variations in temperatures. The two defined peak periods of demand tend to
occur during late autumn/early winter and toward the end of Q2 in the northern hemisphere summer period.
But at present, the lead market is not only facing strong seasonal and cyclical demand, but demand is strong also
from a structural perspective, driven by the rapidly expanding transport sectors primarily in China and India. In line
with this, China is set to lead the way in consumption again this year, helped by strong growth in battery demand.
Figure 394: Barclays Capital global supply and demand summary for lead
At the start of 2006, nickel has continued to underperform the other LME metals even though it has regained most of its
losses seen during H2 05. In contrast to all other base metals (bar tin), the nickel market is in surplus, and LME inventories
are up substantially over the past six months. However, with a lot of bad news in the price, we regard downside price risk
as limited. We also note that even with the terrible demand conditions during H2 05, the lowest level the price dropped to
was US$11,500/t, which is still high in a historical context. There are now early signs of fundamental marketimprovements, in our view, and we are positive on the price outlook as a result.
Western world nickel demand fell into negative territory last year, driven by cut-backs in stainless steel production and
consumer inventory reductions. Europe’s stainless steel output was down some 12% in 2005. Arcelor cut output due to
falling prices and overcapacity in the industry, and Outokumpu (the world’s third largest stainless steel producer) reported
its stainless steel deliveries were 8% lower in 2005. Developments among Asian producers, eg, South Korea, followed a
similar trend.
The new year has started on a more positive note, however. Stainless steel mills in Europe have been able to raise prices for
the first time since May 2004 helped by strong demand, while US and Asian steelmakers are following suit. While volumes
are likely to remain subdued until market conditions have improved further, there are already reports of increased
production levels at Asian plants. Chinese stainless steel production is set to rise by over 20% this year. Furthermore,consolidation within the steel industry should be an overall positive development for nickel, potentially improving
producer discipline and ease price volatility.
In line with this, the rising trend in Chinese refined nickel imports has resumed, rising to their highest level for a year in
December (at 8.9Kt net). For full-year 2005, imports were 40% higher compared with the previous year. Despite the
expansion at China’s largest nickel producer Jinchuan (to 130Kt/y), China’s requirements for imported nickel will stay
large; local research bureau Antaike estimates 107Kt of refined metal (net), compared with 75Kt in 2005.
A key factor limiting the downside in nickel prices has been various disruptions to supply. Output losses amounted to some
45Kt last year. This was partly due to technical problems and labour disputes, but most importantly, feed shortages. And
this is something which will not be rapidly resolved. Like in other commodity markets, a number of factors are keeping
output growth constrained. Growth in global nickel mine output eased to about 3% Y/Y in the January-November period,
according to the latest statistics from the International Study Group. Major new projects are not planned until 2008, when
Goro and Rvensthorpe are due to start production. In this environment, any threats to current supply will keep a solid floor
under the price, with upcoming labour negotiations at Inco’s Sudbury operations in May a focal point. In line with these
supply and demand developments, inflows to LME warehouse have become smaller, perhaps signalling that the rising
inventory trend is about to reverse.
Figure 413: Barclays Capital global supply and demand summary for nickel
TinMetals prices are not strong just because of the sheer weight of investment money, irrespective of underlying market
fundamentals. The tin market is a good example of this. In contrast to all other base metals, tin prices fell during
2005 (-13%), despite ongoing large increases in strategic commodity investment, although admittedly less geared
toward a small market like tin. Also in sharp contrast to other base metals, the tin market saw large supply increases
last year. However, the recent lows of US$6000/t registered in mid-Q4 05 caused a response among producers – alsosuffering from rising production costs – while demand conditions have remained solid. We see the declining price
trend of last year reversing, and see prices gaining strength through the first half of this year.
The strike at the 40Kt/y Minsur smelter in Peru (in December) coincided largely with the turning point in prices,
even though it reportedly had little impact on availability amid a seasonal slowdown in demand over the New Year.
Nonetheless, the smelter is the world’s largest individual producer of refined tin, and after steady rises in LME tin
stocks during H2 05, inventories have now started to draw again.
More significant, however, has been recent production cutbacks at Indonesian mines since the start of the monsoon
season (which usually runs from late November through February) – operations that are also suffering from declining
profit margins due to rising fuel prices. The CRU estimates total reported Indonesian tin production fell to about 14Kt
in Q4, below levels of the same quarter a year earlier, and compared with a quarterly average production figure ofabout 16.5Kt earlier in 2005.
Elsewhere, a slowdown in tin production has also emerged. Even though Bluestone Tin has now commissioned its
3.7Kt/y Collingwood tin mine in Australia, problems related to installing ventilation equipment will keep output
below targets for some 2-3 months. The project is already about six weeks behind schedule and about 15% above
budget. Furthermore, the company is awaiting more strength in the tin price (to above US$7500/t on a sustainable
basis) before restarting its 11Kt/y Renison Bell tin mine in Tasmania, which it closed in October. Elsewhere, Brazilian
tin output has fallen sharply due partly to exhausting reserves, while in China output growth rates have also slowed.
In conjunction with strong domestic demand, Chinese net exports of refined tin has slowed markedly over the past
year. Statistics from China Customs show that while China was a marginal net exporter of refined tin in December (of
89t), its refined tin imports were up a hefty 63% (to 29.8Kt) in 2005 as a whole.
Further on the demand side, we estimate that consumer stocks are running low, and expect a boost to prices as
consumers replenish their inventories in line with positive demand indicators and continued structural changes in
consumption trends. The EU is introducing lead-free regulations from the middle of this year, which will require more
tin (at the expense of lead) in electronic and solder applications.
Figure 432: Barclays Capital global supply and demand summary for tin
ZincThe zinc price has not looked back since the US$1200/t-level traded in July. The most actively traded three-month
contract has gained almost 100% since then, making zinc the best performer on the LME in H2 05. Far forward prices
are also starting to catch up with large rises in the equivalent contracts of other base metals. We see more upside
potential in prices due to this market’s particular severe supply constraints over the coming couple of years.
Together with aluminium, zinc has been the favourite metal among investors for some time. Because of the relentlessrising price trend, systematic CTA fund length has built up. A change in sentiment could therefore cause a sharp
correction in prices. In that event, however, we see buying interest emerging fast. Continuous drawdowns in LME
inventories reflect the strong supply and demand situation, and total reported inventories are approaching critical
lows fast.
Production failures and closures are a key factor behind rising zinc prices. Korea Zinc’s decision to close the 110Kt/y Big
River smelter in the US (now potentially being bought by UK-based ZincOx) aided sentiment on the LME and spurred
physical premia. However, in theory it should have a limited impact on the global refined balance due to ample smelting
capacity and the ability for released concentrates to be processed elsewhere. The technical failure due to a power
overload at Grupo Mexico’s 100Kt/y San Luis Potosi refinery in Mexico and a shutdown of the 140Kt/y Shaoguan
smelter in China (on pollution concerns) should have similar market implications. A labour dispute at Volcan’s mine
operations in Peru has had a more direct market impact, with workers threatening renewed strike actions.
These supply disruptions are accelerating inventory drawdowns on the LME. Even more critical for the longer term
outlook, however, is the lack of new mine projects. Although exploration spending in the zinc industry has picked up
especially fast (an estimated 90% last year) in response to higher prices, there are few major new projects coming on
stream until H2 2007, when the 200Kt/y San Cristobal mine is due to start production. In the past, Chinese mine
supply has been fast to react to rising prices. However, China’s zinc mines are generally small and reserves are
depleting. We estimate the possibility of a sudden flow of new production as small, and China is now becoming an
increasingly important net importer of zinc, of approximately 340Kt in 2005, compared with about 25Kt in 2004.
The lack of spare output capacity is a major factor supporting the current bull-run in commodity prices, however, the
zinc industry actually has some idled mine capacity. Teck Cominco said it would consider re-starting its 150Kt/y
Lennard Shelf mine in Australia once LME zinc stocks fall to 200Kt. We estimate this could happen toward the end ofMay, with stocks currently about 150Kt above that level.
In the largest end-use sector for zinc consumption – galvanised sheet – demand has remained strong, with negative
effects from rising zinc prices partly offset by easier steel prices. In addition, substitution opportunities are limited, as
zinc is the only metal that provides the required corrosion resistance in sheet products for the automotive,
construction and white good sectors.
Figure 451: Barclays Capital global supply and demand summary for zinc2002 2003 2004 2005 2006E 2007E 2008E
bull-run. As the bullish sentiment gradually but steadily spread across the marketplace,
some negative signals for gold from the macroeconomic environment (healthy
economic growth and contained inflationary pressures) were largely shrugged off,
while positive indications (such as speculations of future central bank buying,
geopolitical tensions, and high oil prices) were warmly embraced.
Our precious metals price outlook for 2006
2006 began with strong investor interest in gold, and there remains room on the upside for
prices even though speculative length among tactical investors is already large. One of the
key positive factors we see for gold in the coming year is the prospect of a dollar
depreciation, which our FX team expects. Despite the recent divergence of gold prices from
the dollar, in a bull market where good news outweighs bad, a move in a direction positive
for gold (dollar weakness) would be strongly supportive, although the relationship would
possibly not be as tight as before, given that investors now base their enthusiasm on many
other factors.
The main negative dynamic for gold is the amount of speculative froth currently in theprice. As long as the upward momentum continues, these short-term speculators will
remain committed, but once the rally stalls, the substantial speculative long positions in the
Comex gold market suggest that the downward correction could be swift and brutal.
A wide variety of reasons have been put forward to explain gold’s rally and to argue for its
long-term sustainability, including macroeconomic risks like economic slowdown and
inflationary concerns, hopes of large scale central bank buying, strong physical demand,
declining supply, and long-term investors gaining interest in gold. We are sceptical about
many of them. While we can see the rationale behind some, they are more often than not
overdone or still unrealised. However, we recognise that there are early signs of a changing
attitude towards gold among longer-term investors, aided by the exchange-traded funds
(ETFs), which were created to ease some of the difficulties in investing in physical gold and
to allow more retail investors access to gold (See the Gold section of this report for more).
At the start of the year, we adjusted our gold price forecast higher by $60 to $525/oz for
2006, mainly to take into account the current elevated price levels. With prices at multi-year
highs and short-term investors behind much of the action, volatility can be expected to
remain high, with the potential to spike above $600/oz and fall towards the $470 area,
where physical markets should provide good support. That said, we do see upside risks to
this forecast, if the pick-up in interest among retail and institutional investors is sustained.
We will continue to monitor this area closely going forward.
Accordingly, we have made upward revisions to other precious metals, which have recently
been following on gold’s coattails. We see no compelling fundamental reason for silver togain much independence from gold, and as such expect it to continue to track gold closely,
and to a lesser extent the base metals. That said, the gold/silver ratio trading chart could
provide a reason to be long silver, given that it is a widely followed trading tool. We revised
our 2006 silver price forecast up by $1.80 to $8.90/oz, and expect it to trade in a range of
$7.80/oz – $10.00/oz in 2006.
Platinum remains our preferred precious metal on a commodity fundamental basis. We
expect a combination of resilient fund interest and robust consumer demand to support
prices at historically high levels, and look for an average of $995/oz in 2006 (from our
previous forecast of $910). Palladium has benefited from the renewed wave of fund interest
this time round, after having lagged the complex in previous rallies. We expect its robust
price performance to continue in 2006, led by investor interest across the sector andimproved fundamentals. Thus, we have adjusted our 2006 price forecast higher by $55 to
We have raised our 2006 annual average price forecast from $465/oz to $525/oz
since our last Commodity Refiner . Gold prices moved up sharply at the end of 2005
and again in early 2006. In light of the strong upward momentum, we see further room on the upside.
However, prices have been driven largely by speculators, not fundamentals. Thus,
they are vulnerable to a change in sentiment.
A variety of reasons have been given for gold’s rally and why it is sustainable. We are
sceptical about many of them. While we recognise the rationale behind some, we
think their importance is often overstated, or that they are still at a premature stage.
We believe that the truth is that investors have been attracted to gold’s solid uptrend.
There are early signs of a change in attitude among longer-term investors. Gold
inflows into the exchange-traded funds (ETFs) have accelerated lately, indicating
increased interest among retail investors.
Talk of large-scale Asian central bank purchases will remain an important
influence, whether or not they materialise. We view the ability of central banks to
buy in large quantities as constrained, largely due to the small size of the market.
Physical demand is extremely weak due to the high prices, while scrap selling has
emerged. Output has declined in the traditional producing countries, but we expect
growth in other regions like Latin America, China, Russia and other parts of Africa
to continue expanding.
In other gold-specific factors, positive signs are seen from the gold equities, often
seen as an indicator of market sentiment. However, producer de-hedging is likely tocontinue at a slower pace, while the European central banks should continue to sell.
Figure 473: Gold supply and demand
Tonnes 2001 2002 2003 2004 2005 2006E 2007E
Mine production 2,621 2,589 2,593 2,463 2,494 2,532 2,560
% change 1.2% -1.2% 0.1% -5.0% 1.3% 1.5% 1.1%
Old gold scrap 708 835 939 834 840 860 880
Official sector sales 527 545 617 478 600 600 600
Total physical supply 3,856 3,969 4,149 3,768 3,997 3,992 4,040
Key market themesAs gold continues to scale fresh multi-year peaks, the search for reasons to explain its
powerful uptrend and attempts to make a convincing argument as to why the rally is
sustainable are becoming more intense. The key reasons are as follows:
1) Macroeconomic and geopolitical concerns will attract more investors
2) Central banks will become large buyers of gold
3) Soaring demand and falling supply will drive prices higher
4) Large inflows of money from long-term investors
We believe the reality is that investors looking for a healthy uptrend have been
attracted to gold, and many of these are simply “post-rationalisation” to justify holding
a long position in gold. Foundations for the steady rising trend in gold prices were set
back in 1999 when the European central bank gold agreement (EcbGA 1999) was
signed to limit their gold sales, removing the fears of a “flood” of central bank gold
sales. Then came a trend of gold producers being forced (later voluntary) to cut back
their hedge positions, a term widely known as “de-hedging”. The weak US dollar against
the euro was also an important factor, while the downturn of US equity markets over
the same time period encouraged investors to look at alternative assets.
We would not agree with some of the justifications given for the recent bull-run. While
we recognise the rationale behind some, we think their importance is often overstated,
or that they are still at a premature stage. Nevertheless, we have sensed early signs of a
change in attitude among longer-term investors, a group which will provide much of
the futures hopes for the sustainability of this rally.
1) Macroeconomic and geopolitical concerns will attract more investors
Uncertainty over the health of the global economy and rampant inflation are popularreasons put forward in making a case for gold, though we think this scenario is an
unlikely one. Barclays Capital remains upbeat about the outlook of both the global and
US economic growth. In our latest Global Outlook released in December 2005, our
economists have revised their forecasts for global GDP growth to 4.3% for both 2005
and 2006. One factor that has bolstered their confidence is the resilience of the global
economy to the surge in prices of energy and other commodities, as well as to natural
disasters. Another reason is the recent improvement in Japan and Europe, two of the
laggards in the global economic recovery thus far. That said, our analysis of gold prices
and global growth relative to trend yields no clear relationship between the two, with a
negative correlation of just -0.04 (see Figure 474).
Figure 476: Dollar depreciation against the euro likely
to boost gold price
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
Jan 00 Jan 01 Jan 02 Jan 03 Jan 04 Jan 05 Jan 06
1 month rolling correlation of gold with USD/EUR
Source: EcoWin, Barclays Capital.
Figure 477: A pause in Fed tightening could be positive
for gold though recent rate hikes had little negative
impact
240
280
320
360
400
440
480
520
560
600
Nov 99 May 01 Nov 02 May 04 Nov 05
0
1
2
3
4
5
6
7Gold spot price (US$/oz, lhs)
US Fed fund rate (%, rhs)
Monetary accomodation
Rate
tightening
Source: EcoWin, Barclays Capital.
After what will have been almost two years of tightening, the Fed seems about to go on
hold. Minutes for the FOMC meeting on 13 December indicated that the Committee
thought the number of additional rate hikes needed would “not be large”, and that “policy
decisions would depend to an increased extent on the implications of incoming economic
data for future growth and inflation”. Our economists are expecting two more rate hikes in
2006, while noting that the FOMC minutes suggests upside risk to this forecast in the near
term is limited. The successive rate cuts from 2001 to 2004 have helped support gold’s
price uptrend. Even though recent rate hikes did not have a discernible negative impact on
gold, a pause in the tightening cycle should provide an additional boost to sentiment (see
Figure 477).
Importantly, the ongoing geopolitical concerns – tensions over Iran’s nuclear programme,
potential for a leadership vacuum in Israel, threats of more terror attacks and spread of bird
flu – are likely to make investors nervous about sustaining an aggressive short position, and
thus should provide support for prices.
2)
Central banks will become large buyers of gold
The argument for Asian central bank buying of gold has been based on, firstly, Asian central
banks being overly exposed to the US dollar (which we expect to weaken), and secondly,Asian central banks still have a low proportion of their reserves in gold compared to their
European and US counterparts (see Figure 478). However, it should be noted that Japan
(765.2 tonnes) and Mainland China (600 tonnes) are already among the top-ten largest
holders of gold in terms of absolute quantity.
These hopes were given a boost in early 2006 by comments from the Chinese State
Administration of Foreign Exchange (SAFE) that one of its targets for 2006 was to “improve
the operation and management of foreign exchange reserves and to actively explore more
effective ways to utilise reserve assets… to improve the currency structure and asset
structure of our foreign exchange reserves, and to continue to expand the investment area
of reserves.” This could mean that an increasing pool of reserves would be actively managedand invested in a broader array of financial assets, including commodities. It should be
noted, however, that a senior central banker at the People’s Bank of China (PBoC) has
clarified that it is unlikely for China to sell off its USD-denominated reserves to achieve
diversification. Our economists for China believe that China would be more interested in
gradually increasing the portion of non-USD assets in its new purchases rather than selling
off existing USD reserve assets.
We view the ability of Asian central banks to purchase gold in any meaningful amounts for
FX diversification as constrained, primarily due to the small size of the gold market. China
currently holds roughly 600 tonnes of gold (already more than the 500-tonne annual sales
limit under the EcbGA), accounting for 1.1% of total reserves. If this level were to be raised
to 10%, some 4107.5 tonnes, or around two years of global mine production, would be
absorbed. Investing in other commodities like oil and base metals would be more possible,
or the outcome could well be a combination of elements from different asset classes.
Meanwhile, if we assume that the central banks of Japan, China and India (the sort of
central banks that should be buying more gold, under the popular thesis) each raise their
gold holdings to 10% (a reasonable level for FX diversification purposes), this would
require roughly 290 Moz (9,017 tonnes) of gold, calculated at $500/oz (see Figure 479
for more calculations). Given that this represents around 3.7 years of global mine
production, the most plausible way of transacting is through a large off-markettransaction.
Figure 478: Asian CBs proportion of gold holdings
small compared to European and US CBs
0%
10%
20%
30%
40%
50%
60%
70%
80%
0 2,000 4,000 6,000 8,000 10,000
Gold reserve (tonnes)
G o l d ' s s h a r e o f r e s e r v e s
US
Germany
France
ECB
Switzerland
Asian central banks
Source: World Gold Council, Barclays Capital.
Figure 479: Amount of gold needed by each Asian
central bank to increase gold holdings to 10%
Tonnes of gold
in reserves
% of reserves Tonnes of gold
needed to
reach 10% Japan 765.2 2 4,380.8
China 600.0 1.1 4,107.5
Taiwan 423.3 2.4 1,170.8
India 357.7 3.6 528.5
Indonesia 96.5 4.1 105.3 Thailand 84.0 2.4 220.2
Total 2,327 10,513
Source: World Gold Council, Barclays Capital. Note: Calculations are
made based on a gold price of $500/oz.
However, the renewal of the European central bank gold agreement in 2004 (EcbGA 2004)
that limits gold sales to 500 tonnes a year effectively dashes realistic hopes of large-scale
buying by Asian central banks (it would take them over 20 years to accumulate at the speed
the European central banks are selling). In addition, the existence of the EcbGA highlights
the difficulty of exiting a large position in gold, which could be a major deterrent for other
countries to increase their gold by a significant amount. Further, low gold lease rates mean
that yields from holding gold are extremely unattractive.
Another key factor driving gold prices in the first few weeks of 2006 was speculation that
the Iran central bank has been buying gold as part of their preparation for any UN action
over their nuclear research. However, such rumours were dealt a blow by comments fromthe Iran central bank that they are not repatriating their foreign exchange assets, and they
“have no intention of buying gold at the high price at the moment… are not in the market
for the time being and are not going to be in the market”, though it failed to have any
noticeable negative impact on prices.
3) Soaring physical demand and falling supply will drive prices higher
Physical demand from India was surprisingly resilient in the high price environment during
H1 05, but had weakened considerably during H2 and so far in 2006 on persistent price
volatility and strength (Figure 480). Physical purchases were also slow in other parts of Asia,with physical premiums turning into discounts or zero on several occasions, as the locals
engaged in scrap selling. Jewellery demand in European countries continued to decline, but
there is evidence of rising demand from the Middle East and China. That said, physical
demand rarely assumes the role of a price driver in the gold market given its massive
inventory overhang (conservatively, over 300 weeks of demand, see Figure 481), and could
only affect gold prices indirectly via sentiment.
However, it is argued that as consumers become accustomed to higher prices, they would
return to the market and help lift the price floor to higher levels. We agree with the
rationale behind this, although to put this into perspective, physical demand would not be
sufficient to prevent a freefall should all the funds turn together (though we are not in any
way suggesting that this would happen).
Figure 480: High and volatile prices discouraging
Indian imports
0
20
40
60
80
100
120
140
160
Jan 00 Jun 01 Nov 02 Apr 04 Sep 05
300
350
400
450
500
550
600
650
700Indian imports (tonnes, lhs)
Gold price (Rupee/g, rhs)
Source: World Gold Council, CRU, Barclays Capital.
Figure 481: No lack of gold above the ground
0
50
100
150
200
250
300
350
400
Gold Silver Pd Pt Zn Al Cu
Inventory as weeks of demand
Source: GFMS, Johnson Matthey, Brook Hunt, CRU, Barclays Capital.
Despite rising sales from the official sector, expectations of a large fall in mine productionhas been one of the major themes supporting the thesis of a rising trend in gold prices.
There is a basis for this argument – mine output growth has been modest in recent years
and turned negative in 2004. With the decline in exploration spending during the 1990s,
expected depletion at several major ore bodies in coming years, and rising costs in the
traditional producing countries of South Africa, the US, Canada and Australia (see Figure
482) it is not unreasonable to come to such a conclusion.
However, despite declining output from the four traditional producing countries in recent
years, we have seen growth in other regions like South America, Asia, Russia and other
parts of Africa and the former Eastern bloc countries ( Figure 483). Mine output in China
rose by 6% in 2004, and there is potential for further growth due to market liberalisation in
recent years. The expectations for a significant fall in mine output are unlikely to be realised,
in our view, and we expect mine output to continuing growing over the coming years. In
line with our view, GFMS noted that mine output growth in 2005, concentrated towards H2
05, was mainly a result of new mines in Latin American countries, including Peru, Mexico
and Brazil. Meanwhile, an improvement in the giant Grasberg mine contributed to a
significant increase in Indonesian output, though output in South Africa continued its
downtrend. That said, the large above-ground stocks in gold also imply that constrained
supply is unlikely to be an important driver of prices.
Figure 482: Cash costs rising for most traditional
producing countries
150
200
250
300
350
400
1998 1999 2000 2001 2002 2003 2004
South Africa
Australia
United States
Canada
(US$/oz)
Source: GFMS, Barclays Capital.
Figure 483: Mine output by major region (tonnes)
0
300
600
900
1,200
1,500
1,800
2,100
2,400
2,700
93 94 95 96 97 98 99 00 01 02 03 04
Former East bloc
Other Western
Sth A frica,USA,Canada,Aus)
Source: GFMS, Barclays Capital.
4) Large inflows of money from long-term investors
The arguments for holding gold as an investment – portfolio diversification, an
inflationary and/or US dollar hedge, safe haven asset, etc – are far from new. The rise ofalternative assets and the uptrend in gold prices have returned gold to many investors’
radar screen. Other than traditional means of investing in gold, including bar hoarding,
coins, futures and options on exchanges and over-the-counter (OTC) markets, it
appears that new hopes for the long-term sustainability of gold’s rally are increasingly
placed on retail investments into gold exchange-traded funds (ETFs) and institutional
investors like pension funds gaining interest into gold.
We do see encouraging signs in the gold ETFs, but suspect that institutional investment in
gold through commodity indices and structured products is still quite small at present.
However, there are hopes that this will rise as institutional investors increase their
commodity exposure, with gold as one of the beneficiaries. The bulk of money drivinggold in the recent rally has come from speculators on major exchanges and OTC markets
worldwide, which usually adopt a more tactical or short-term approach to investing.
Fund activity on major exchanges and OTC markets
It is difficult to obtain an accurate measure of fund positions on various exchanges and
the OTC markets. Data released by the Commodity Futures Trading Commission (CFTC)
every week provide a gauge for fund activity on COMEX – under their classification,
“non-commercial” accounts are generally seen as tactical investors acting on short-
term views of price direction. In the week ending 17 January 2006, non-commercial net
length stood at 148.5K lots (461.9 tonnes, see Figure 484). Although this is some way
below the all-time high of 177.4K lots (551.8 tonnes) in mid-October, it is still quitesignificant and accounts for 41% of the total open interest in gold (see Figure 485),
third-highest among all commodities covered in the CFTC data series. Given the
sizeable gross long positions, the threat of large long liquidations if the sentiment turns
is indeed very real.
According to data from the LBMA, COMEX and TOCOM, the daily market size of the
London OTC market is the largest among the three, with a daily market size of around
$7.1bn, compared with $2.8bn on COMEX and $2.0bn on TOCOM. Unfortunately, there
is no way of getting reliable estimates for the OTC market due to its nature, neither is
there good data for TOCOM, but we believe that the huge amount of buying in these
two markets have resulted in massive long positions. Therefore, concerns of massive
profit-taking will remain a key issue for the high prices, though it is still unclear what
could jolt the market out of its current mood.
Figure 484: Speculative length on COMEX remains
large…
-100
-75
-50
-25
0
25
50
75
100
125
150
175
200
Mar 02 Dec 02 Sep 03 Jun 04 Mar 05 Dec 05
250
275
300
325
350
375
400
425
450
475500
525
550
575Net Position ('000 contracts, LHS)
Price (US$/oz, RHS)
Source: CFTC, Barclays Capital.
Figure 485: … and its proportion of total open
interest near record high
-50%
-40%
-30%
-20%
-10%
0%
10%
20%
30%
40%
50%
95 96 97 98 99 00 01 02 03 04 05 06
Source: CFTC, Barclays Capital.
Signs of improvements in the gold ETFs
After a period of inconsistent inflows, investments into the gold ETFs gathered pace
during the last few months of 2005, not only in the largest US StreetTRACKs fund, but
also in products that had pretty much stagnated soon after inception, like the Gold
Bullion Securities product listed on the London Stock Exchange (LSE) and ABSA
NewGold in Johannesburg Securities Exchange (see Figure 486). In 2005, a total of
189.1 tonnes of gold entered the gold-held-in-trust at gold ETFs worldwide, a 52%
increase on 2004. Out of this, 79 tonnes occurred in Q4 05 alone. Meanwhile, around46 tonnes have already entered the ETFs in less than a month into 2006, of which 30.3
tonnes is attributable to the US StreetTRACKs fund. Developments at the ABSA fund
are also noteworthy – after stagnating around 3 tonnes since its launch in November
2004, gold-held-in-trust has grown steadily since mid-December to around 9.42
tonnes. Roughly 406.12 tonnes of gold is now held at the five ETFs globally, amounting
to a net asset value of $7.2bn, a spectacular 735% increase from the $0.87bn at the
start of last year.
So exactly how significant will the gold ETFs be? We believe that this is an important
area to watch going forward, firstly because they are arguably an indication of the
amount of retail interest in gold, and secondly, an additional buying force is required tomake up for the expected slowdown in producer de-hedging. Thirdly, increased inflows
of funds into the gold ETFs could potentially encourage short-term speculators to stay
One of the major themes surrounding commodity markets as we step into 2006 is the
potential for commodities, including gold, to gain more traction among long-term
investors like pension funds. Or according to some, flow of money from institutional
investors is already one of the main driving forces behind gold prices.
However, we suspect that investment from this group of investors in the gold market is
relatively small at present. Institutional investment into commodities is estimated to be
around $80bn. Assuming that 5% of that has been in gold, then institutional investment
represents around 1.5% of gold’s monthly combined turnover on COMEX, TOCOM andthe London OTC market; and less than one quarter of its notional value of open interest
on COMEX (see Figure 488). Thus, gold’s fortunes currently still hinges critically on
speculative activity of funds with a shorter time horizon. Nonetheless, it is hoped that
this will grow as investors increase their exposure to commodities in portfolios over the
next few years, with gold one of the beneficiaries.
Institutional investors in commodities often look for opportunities to benefit from a
feature unique to commodities – backwardation. Backwardation occurs when a physical
market is in shortage of material, and therefore consumers have little choice but to pay
a premium for prompt delivery of raw materials. This will give rise to a downward
sloping forward price curve, and hence the investor is paid to roll his position forward,
generating a positive roll yield. However, gold is virtually in a permanent state of
contango (upward sloping forward price curve) due to its massive above-ground stocks
available at extremely low borrowing costs. As a result, there is no yield from rolling
Figure 489: No roll in gold means index investors can
only gain from price appreciations
0
100
200
300
400
500
600
700
800
900
1000
Jan 87 Oct 91 Jul 96 Apr 01 Jan 06
Oil GSCI (Total Return/Spot Return)
Copper GSCI (Total Return/Spot Return)
Gold GSCI (Total Return/Spot Return)
Source: EcoWin, Barclays Capital.
Other main themes
Producer de-hedging is expected to be less supportive for gold – De-hedging is
estimated to have declined by 54% to 195 tonnes in 2005 (though this large fall
was partly a result of the extremely high level in 2004). We expect de-hedging to
continue at the new slower pace, given the reduced likelihood of large buybacks
and a smaller global hedgebook.
Gold Equities have started to outperform gold bullion – The HUI Gold Bugs Index
has often been viewed as an indicator for sentiment towards gold. On that basis, we
have noticed some positive signals recently. The HUI Gold Bugs Index repeatedly
reached record highs at the beginning of 2006 together with the strong prices in
gold bullion. In addition, after lagging behind gold for most part of 2005,
performance of the equity index has picked up significantly in late December, and is
currently outperforming gold prices (see Figure 489).
European central banks continue to sell – Up to the week ending 24 January 2006,
or about four months into Year 2 of the European central bank gold agreement
(EcbGA 2004), the signatories have sold roughly 149.8 tonnes of gold. According to
figures released by the IMF International Financial Statistics in December, 33 tonnes
have been sold by France, 21 tonnes by Spain, 10 tonnes by Portugal, 6 tonnes by
the Netherlands and 1 tonne by Sweden. Meanwhile, Germany will decide this year
whether it would sell its gold (it has an option to sell 120 tonnes a year under the
EcbGA 2004, but it only sold 8 tonnes in Year 1). The latest development on this
was comments from the finance minister that it is possible to find a solution “which
represents the independence of the Bundesbank on the one hand, and on the other,to find a solution which… may allow us to mobilise money from these gold reserves
Resilient fund interest has been the predominant factor behind silver’s steady rise since
2003 and the more pronounced rally recently. Fund buying has been based largely on
silver’s traditional role as an investment alternative to gold, with silver seen as
possessing “better relative value”. Investors looking for attractive risk-reward
propositions have used the gold/silver ratio as a trading tool on which to base their
strategies. Silver’s robust performance in December moved the ratio below 58
(compared with its 2005 average of 61). While this ratio rose to the 61-62 range duringthe first few weeks of 2006, the recent strength in silver prices has brought it back
towards 58. On this technical note, our technical analysts have noted in their recent
report (see: Silver plays catch-up with gold , 12 December 2005) that a move lower
than $58.57 could set up the longer-term line towards $56 and lower.
As perceptions of silver swing between it being a precious or an industrial metal,
investors have also increasingly traded silver in line with the base metals, particularly
copper (see Figure 501). Overall, we remain upbeat about base metal prices, due to
their strong fundamentals and expectations for robust global industrial activity. Thus,
silver can be expected to gain some benefit from this.
Figure 500: Gold/silver ratio moved up in early 2006
35
45
55
65
75
85
95
Jan 95 Jul 97 Jan 00 Jul 02 Jan 05
Gold/Silver Ratio
average over period
Source: EcoWin, Barclays Capital.
Figure 501: Silver also gaining some strength from thebase metals
Index of Performance since 2003
90
110
130
150
170
190
210
230
250
270
290
310
Jan 03 Sep 03 Jun 04 Mar 05 Dec 05
Copper
Silver
Source: EcoWin, Barclays Capital.
In the first few weeks of 2006, speculators on COMEX have reduced their net exposure
to the silver market to levels below the hefty level seen in December 2005 (Figure 502).However, fund length remains significant and accounts for more than 45% of open
interest (Figure 503); pointing to risks of liquidations should the bullish mood turn. The
fact that prices rose despite the decline in fund length highlights the resilience of fund
interest in the over-the-counter (OTC) market, where we expect speculative length to
Figure 503: … and accounts for a significant portion
of total open interest
-10%
0%
10%
20%
30%
40%
50%
60%
95 96 97 98 99 00 01 02 03 04 05 06
Source: CFTC, Barclays Capital.
The steep rally in prices over the latter part of 2005 cannot be attributed to any
significant improvements in silver’s commodity fundamentals. After falling to the
lowest level in our records in 2004, Indian demand recovered during H1 05. However,
Indian imports fell significantly in H2 05 (Figure 504) and traders report to us that local
demand remains extremely weak at present, due to high and volatile prices.
We expect fabrication demand from the photographic sector to continue falling at an
accelerated pace. Over the past few years, photographic demand has declined sharply
in North America, Europe and Japan due to the increasing popularity of digital cameras,though it has remained largely stable in other parts of the world. In fact, Chinese
photographic demand has been rising since 1999, due mainly to an expansion of the
total photographic market, the higher cost of digital cameras and the lower level of
technology involved in film cameras. However, this could change quickly together with
China’s robust economic growth, rising incomes and increased access to information
technology. Furthermore, reports suggest that digital camera makers may start to
concentrate more on emerging markets such as China, India and Korea, as the US
market could be nearing its peak and the Japan market has already peaked, in 2004.
On the supply side, we expect mine output to continue expanding at a modest pace.
Mexico, the world’s largest silver producer, has said recently that it expects silveroutput to increase by 9% to 3,080 tonnes in 2006, adding that silver mines are
increasing production to take advantage of the high prices. Meanwhile, Peru, the
second-largest producer, reported an 8.8% Y/Y rise in silver output to 274,546kg (275
tonnes) in November. Nonetheless, with two-thirds of silver being generated as a by-
product of other metals, the constrained situation in the lead and zinc concentrate
markets will help to restrict the pace of silver mine output growth. In addition, we
expect scrap supply to continue on a downtrend, due to the lower recovery from
photographic sources.
Short-term rallies are the hallmark of the silver market. Such rallies are usually
associated with increases in silver lease rates (Figure 505). Lease rates have picked up
recently together with the surge in prices, though they still remain relatively low. We
believe the modest lease rates suggest that there are no imminent shortages in the
Figure 515: Platinum remains the most “speculative”
commodity…
-20% 0% 20% 40% 60% 80%
PlatinumSilver Gold
Palladium
Feeder CattleWheat
GasolineCocoa
Live CattleOrange Juice
Sugar CoffeeCotton
SoybeanCorn
Heating OilCopper
Crude OilLean HogsSoybeans
Soybean OilWheat
Nat Gas
Futures Net Position
as % of Open Interest
Source: CFTC, Barclays Capital.
Figure 516: … even though speculative length cut
from record levels
-2
0
2
4
6
8
10
12
Mar 02 Dec 02 Sep 03 Jun 04 Mar 05 Dec 05
400
500
600
700
800
900
1,000
1,100Net Position ('000 Contracts, LHS)
Price ($/oz)
Source: CFTC, Barclays Capital.
Platinum expected to remain in deficit for the eighth straight year
A catalyst for the surge in platinum prices during Q4 05 was the release of Johnson
Matthey’s (JM) Interim Review on 15 November 2005, in which it forecast another deficit in
2005 (previously it had estimated the market to be more or less in balance), mainly on the
back of strong consumer demand. Further, demand growth is expected to remain robust in
2006, while several supply side challenges remain. Therefore, another market deficit is likely
over the coming year, in our view.
Figure 517: Another year of deficit for platinum in2005…
-800
-600
-400
-200
0
200
400
1976 1980 1984 1988 1992 1996 2000 2004
Balance ('000oz)
Source: Johnson Matthey, Barclays Capital.
Figure 518: … as autocatalyst demand soars in Europe
0
200
400
600
800
1,000
1,200
1,400
1,600
1,800
2,000
1976 1980 1984 1988 1992 1996 2000 2004
Europe
North America
Japan
Rest of World
Platinum autocatalyst demand ('000 oz)
Source: Johnson Matthey, Barclays Capital.
Demand growth to remain strong
We expect platinum autocatalyst demand growth to remain robust, driven primarily by
European demand (see Figure 518) as diesel engines continue to gain market share (around50% of new car registrations in Europe in 2005). Tighter emissions standards for both light
and heavy duty diesel vehicles should lead to increased loadings, while the high fuel prices
could serve to further boost the sales of diesel vehicles, which are more fuel-efficient. In
addition, rising vehicle sales globally and growth in light vehicle production in China are also
expected to contribute to higher autocatalyst demand. New emissions legislations in the US
will prompt greater demand for platinum from heavy duty diesel vehicles, which could
offset the impact of the ongoing shift towards palladium-based gasoline engines.
Meanwhile, recent comments from a variety of carmakers and research agencies that they
expect diesel vehicles to make inroads into the US markets should continue to boostinvestor interest, whether or not these expectations are realised in the short term.
A key factor that can potentially pose downside risks for platinum is a pick-up in the
substitution towards cheaper palladium in diesel autocatalysts. While the technology is
available and can be expected to be popular given the price spread between platinum and
palladium, we believe that a truly significant impact on the physical markets remains some
time away. For instance, it takes around 1-2 years for the use of palladium for diesel engines
to be adopted on a large range of models. Furthermore, it is also important to note that
25%-30%, and not 100%, of platinum would be replaced. Nonetheless, news on
developments in this area could well keep a lid on prices.
Figure 519: Chinese jewellery demand likely tocontinue falling…
0
300
600
900
1,200
1,500
1993 1995 1997 1999 2001 2003 2005
350
450
550
650
750
850
950Chinese Platinum Demand
('000oz,lhs)
Platinum Price
(US$/oz,rhs)
Source: Johnson Matthey, Barclays Capital.
Figure 520: … but Chinese buying still re-emerging onprice dips
The situation in platinum jewellery demand is not as rosy, however. High and volatile prices
have continued to hurt manufacturers’ margins, and jewellery demand is expected toremain on a downtrend this year. Chinese jewellery demand is likely to have recorded the
third successive year of decline in 2005 (see Figure 519). Nevertheless, as we have
highlighted a few times, Chinese demand on the Shanghai Gold Exchange (mainly physical)
has re-emerged on price dips even at prices over $900, though they have remained modest
recently, with prices above $1,000.
Supply to grow but challenges remain
On the supply side, South African output is expected to have increased in 2005, even after
accounting for the temporary closure of Angloplat’s Polokwane smelter. In 2006, South
African output is expected to increase, while the material built up in Angloplat – due to the
Polokwane incident in 2005 – would boost output further. The producers’ rand basket pricehas improved slightly in 2005, but recent strength in the rand (currently around 6.00 vs
USD, well below the 7.00 level deemed the lowest many of these projects could withstand)
could pose a significant threat to output if sustained.
Elsewhere, Russian sales are expected to have increased marginally in 2005 and sales from
state stocks are likely to remain small this year. North American supply is estimated to
decline in 2005, due to lower output from Inco and North American Palladium, but should
improve in 2006, according to JM.
Figure 521: Producers’ rand basket price improved in
2005…
50,000
75,000
100,000
125,000
150,000
175,000
200,000
98 99 00 01 02 03 04 05
Nominal South African PGM Basket Price (R/kg)Real South African PGM Basket Price (R/kg)
Source: EcoWin, Barclays Capital.
Figure 522: … but more strength in rand is a threat to
expansions
2
4
6
8
10
12
14
Jan 95 Mar 97 May 99 Aug 01 Oct 03 Jan 06
USD/ZAR
Source: EcoWin, Barclays Capital.
Another important factor influencing prices is the inventory of the metal. On that basis,
platinum stocks relative to demand have declined quite significantly over the years, and weestimate that they stand at around seven weeks of demand in 2005. However, the one-year
lease rates remain quite modest, implying that there is currently little concern over the near
term availability of platinum (see Figure 524).
Figure 523: Platinum’s stocks have declined
significantly after years of deficits…
0
5
10
15
20
25
30
35
40
45
50
1995 1997 1999 2001 2003 2005
Estimated platinum inventory as weeks of
consumption
Source: Johnson Matthey, Barclays Capital.
Figure 524: … but the modest lease rates imply little
Figure 538: Palladium jewellery demand in China soars
0
300
600
900
1,200
1,500
1995 1997 1999 2001 2003 2005
Chinese platinum jewellery demandChinese palladium jewellery demand
'000 oz
Source: Johnson Matthey, Barclays Capital.
Figure 539: Strong Chinese car sales positive for
palladium
0
100,000
200,000
300,000
400,000
500,000
600,000
700,000
Jan 97 Jan 99 Jan 01 Jan 03 Jan 05
Chinese motor vehicles sales (units)
Source: EcoWin, Barclays Capital.
In contrast, palladium autocatalyst demand is estimated to have fallen slightly in 2005,
negatively affected by the declining trend of gasoline vehicle production in Europe.
While demand in the US is benefiting from the ongoing shift towards palladium-based
gasoline autocatalysts, negative effects of weak auto sales and reductions in average
loading (due to thrifting) dominated last year, resulting in a fall in autocatalyst demand
for palladium in the US.
In other parts of the world, we expected autocatalyst demand to remain robust, aided
by higher vehicle sales and rising emissions controls. In China particularly, demand for
palladium in autocatalysts is estimated (by JM) to have increased by 43% in 2005, andthere is potential for further growth as rising incomes and strong economic growth
encourage more demand for cars, which predominantly use palladium-based
autocatalysts at present (palladium offtake roughly four times that of platinum).
Supply declines in 2005 but expected to grow in 2006
Importantly for the supply side, ever since some substantial sales from Russian state
inventories in Q4 04 were carried over to Q1 05, there has been little evidence of
further stock selling, resulting in a fall in Russian exports last year. In addition,
production from North America also declined last year due to lower output at Inco and
North American Palladium. In 2006, Stillwater Mining would have sold its entire
inventory transferred by Norilsk Nickel by the first quarter, while Norilsk Nickel expects
production to decline to the range of 2.90-2.95Moz (from 3.133Moz in 2005). North
American output should improve, and South African output is expected to increase
significantly, according to JM, though the strength of the rand (around 6.00 versus USD
currently) will pose a threat to expansions if sustained.
Agriculture market overviewIn addition to the usual weather and crop-disease related price movements, key themes
to watch out for in the agricultural commodity markets in our view are Chinese demand
levels, biofuel production and speculative interest. It is well known that the vast and far-
reaching economic changes occurring in China, including industrialisation, rural-urban
migration and higher living standards, have spurred commodity demand. What is
known to a lesser degree is the impact these changes have had on agricultural
commodity demand. Chinese demand has risen strongly over the past few years for
most agricultural commodities, apart from wheat, albeit in some markets from low
levels. Higher per-capita income and changed diets have led to robust soft commodity
consumption, while rising protein and meat demand has galvanised feed markets. China
is already the biggest consumer of cotton, wheat and the biggest importer of soybeans,
cotton and corn. Increased Chinese demand and imports therefore in our view will
materially impact the agricultural markets.
High and volatile oil prices in conjunction with geopolitical tensions have highlighted
the need for alternative fuel production. Key among these alternative fuels is ethanol,made from sugar in Brazil and from corn in the US. While there has been a palpable
shift in sugar prices, corn prices have not received any sustained lift from their link to
ethanol. This we believe is due to lower US ethanol demand compared to Brazilian
ethanol demand, the percentage of corn made into ethanol is lower than the amount of
sugarcane diverted to ethanol in Brazil, and because supplies are tighter in the sugar
market than in the corn market.
The other key theme is increasing fund interest in the agricultural commodity markets.
At the start of 2006, funds stated targeting the agricultural commodity sector as the
one with most upside price potential. Aggregate CFTC net long positions in the main US
agricultural futures markets have switched from broadly neutral at the start ofDecember to a long position of over 257,000 lots in early February. We expect this fund
interest to add both buoyancy and volatility to prices.
Figure 548: Spot agricultural commodity returns set to
rise in 2006
0%
5%
10%
15%
20%
25%
30%
35%
40%
45%
50%
2004 2005 2006
Energy
Agriculture
Industrial Metals
Precious Meta ls
Average
Change in average spot prices by commodity s ector
(Q4 vs. Q4, unweighted)
Source: EcoWin, Barclays Capital.
Figure 549: The high cost of carry erodes returns for
Agriculture spot indices were one of the few commodity sectors to register larger
percentage gains in 2005 than in 2004 (see Figure 548). We expect another step up in
the rate of spot price appreciation in 2006, in contrast with most other commodity
sectors where the increase in average spot prices will continue to slow. Despite the rise
in spot prices, returns for investors in the agriculture sector in 2005 were disappointing
due to the high negative roll yield (see Figure 549). For investors targeting the
agriculture sector in 2006, a crucial issue is how to benefit from rising spot priceswithout seeing potential gains eroded by the high costs of carry. In order to achieve
this, markets should be targeted where there is the potential for spot price gain, but
also for long periods of backwardation in futures curves.
We remain bullish on sugar’s price performance through 2006, in light of the strong
market fundamentals – the deficit market, strong demand levels for both sugar and
ethanol and tight stocks coupled with speculative interest and positive investor
sentiment. In light of cotton’s positive market fundamentals and supportive market
sentiment, we view cotton prices with an upside bias as well. Strong Chinese cotton
import demand for its burgeoning textile industry, adverse climate in cotton-growing
areas of the US and projected record levels of global use all paint a positive scenario for
prices in our view. For cocoa, due to robust production levels coupled with strong
global stocks, we do not expect a significant rise in prices from current levels. However,
short-term price spikes in connection to the volatile political situation in the Ivory Coast
remain a strong possibility. For coffee, we expect prices to be firmly underpinned in Q1
amid volatility, on the back of modestly positive short-term fundamentals and
speculative interest. However, we do not anticipate sustained appreciation in prices
through 2006.
Price appreciation prospects for the wheat market also look good, in our view,
especially for KCBT’s Hard Red Winter Wheat (HRW) which has been boosted by strong
export demand and by adverse climatic conditions. However, for the feed grains, the
outbreak of bird flu dampened sentiment in 2005, and if spreads in 2006, it has the
potential to act as a dampener on feed demand and prices.
Figure 550: Key agricultural product information
Commodity Main Exchange Quotation Contract Size Contract expiration
Softs
Cocoa NYBOT (CSCE) US$/t 10 tonnes Mar, May, Jul, Sept, Dec
Coffee NYBOT (CSCE) USc/lb 37500 lbs Mar, May, Jul, Sept, Dec
Sugar NYBOT (CSCE) USc/lb 50 long tonnes Mar, May, Jul, Oct
Biofuel
Ethanol CBOT US$/gallon 29K US gallons 12 consecutive months
Ethanol CME US$/gallon 30K US gallons 12 consecutive months
Front month NYBOT cocoa prices in Q4 ranged between a low of $1315/t to a high of
$1525/t, averaging $1434/t through the quarter. Prices temporarily spiked at the start
of 2006 on the back of violence in the world’s largest cocoa producer, the Ivory Coast,
taking prices to a high of $1600/t, a level not seen since April 2005. Funds were
shorting the NYBOT cocoa market from the start of Q4 till end-November, when net
fund position turned long, and it has been in positive territory since. In light of the
strong production levels coupled with large stocks, we do not expect cocoa prices to risesignificantly higher from present levels. However, short-term price spikes continue to be a
strong possibility, in light of the volatile political environment in the Ivory Coast. Global
production for 2005/06 is estimated by the EIU at 3,398kt, a 7.5% Y/Y increase from the
previous year’s 3,162kt figure, but still below 2003/04’s record harvest of 3,482kt.
Production from the world’s two largest cocoa producers, the Ivory Coast and Ghana, is
expected to be robust, with the two countries accounting for 60% of total production. In
the longer term, the EIU forecasts oversupply in the cocoa market, expecting a 219Kt
surplus in five years, with governments of many producer countries supporting higher
production.
With the Ivory Coast dominating global cocoa production (40%), it is by far the single
most important dynamic on the production side and therefore, any potential supply
disruption from the country is bound to push prices higher. The latest spate of violence
in the Ivory Coast in January took place in Abidjan, in the southern region controlled by
the Ivory Coast Government (the north is controlled by the rebels). Abidjan, a main
shipping port for cocoa exports was hit by violent protests carried out by the pro-
Gbagbo “Young Patriot” supporters of the government. Protestors targeted their ire on
the UN, as foreign mediators recommended that the national parliament be dissolved,
which is dominated by the President’s party. In early February, the UN imposed
restrictions on three political figures within the Ivory Coast, which led to cocoa prices
rising 2.6% on the day.
In line with recent trends, China’s cocoa bean imports were robust through 2005,
amounting to 42,659 tonnes and posting a 115% Y/Y, albeit from low levels. The rising
imports reflect stronger demand on the back of urbanisation, economic change and
higher per capita income. Grindings data from mature markets was mixed, with US Q4
grindings falling 2.2% Y/Y to 97,673 tonnes according to the Chocolate Manufacturers
Association. However, Europe’s Q4 cocoa grindings posted a 4.7% Y/Y increase to
301,742 tonnes, according to the European Cocoa Association.
Figure 553 China’s cocoa bean imports rise by 115% Y/Y in 2005
0
1000
2000
3000
4000
5000
6000
7000
8000
Mar 04 Jun 04 Sep 04 Dec 04 Mar 05 Jun 05 Sep 05 Dec 05
Front month NYBOT coffee prices in Q4 ranged between a low of 91¢/lb and a high
of 108.5¢/lb, averaging 105.4¢/lb through Q4. Prices at the start of 2006 have been
firmer, rising to a high of 125.9¢/lb at the end of January.
We expect NYBOT coffee prices to be volatile but firmly underpinned in Q1, helped
by modestly positive short-term fundamentals and speculative interest. However,
we do not anticipate sustained appreciation in prices through 2006.
The supply scenario for the coffee market looks stronger than it did in Q4, with no frost
damage to the Brazilian crop and less than expected damage from hurricanes in the US
Gulf and in Central America.
While 2005/06 global production is estimated by the EIU to rise by a strong 9.7% Y/Y
at 89mn bags, global consumption is estimated at 86.3mn bags (+1.4% Y/Y), taking the
coffee market into a surplus following two successive deficits.
Dynamic coffee demand comes from emerging market economies, such as Russia,
Eastern Europe and parts of Asia. However, coffee demand in these regions is price-
sensitive and contingent upon affordability.
Figure 560: NYBOT coffee prices rise in 2005/06
40
90
140
190
240
290
340
Oct 93 Jul 95 Apr 97 Jan 99 Oct 00 Jul 02 Apr 04 Jan 06
NYBOT Arabica prices (cents/lb)
Source: EcoWin.
Figure 561: Global supply and demand summary for coffee
01/02 02/03 03/04 04/05 05/06E 06/07EExportable Production (mn 60kg bags)
Global 89.5 85.9 83.7 81.1 89.0 94.4
Y/Y change 1.0% -4.0% -2.6% -3.1% 9.7% 6.1%
Brazil 28.3 26.9 21.9 23.0 27.2 29.2
Share of Global 31.6% 31.3% 26.2% 28.4% 30.6% 30.9%
Y/Y change 33.5% -4.9% -18.6% 5.0% 18.3% 7.4%
Consumption
Global 81.6 85.2 84.2 85.1 86.3 87.2
Y/Y change -1.0% 4.4% -1.2% 1.1% 1.4% 1.0%
Balance 7.9 0.7 -0.5 -4 2.7 7.2
Total Stocks 44.4 45.1 44.6 40.6 43.3 50.5
Weeks of consumption 28.3 27.5 27.5 24.8 26.1 30.1
Annual Average Price (cents/lb) 48.9 62.4 69.8 104.1 104.2 Note: The marketing year for coffee runs from October to September. Average prices for 2005/06 are to the
31 st of January 2006. Source: Economist Intelligence Unit, EcoWin, Barclays Capital.
Front month NYBOT coffee prices ranged between a low of 91¢/lb and a high of 108.5¢/lb,
averaging 105.4¢/lb through Q4 2005. Prices were firmer at the start of 2006, rising to a
high of 125.9¢/lb. Net fund length in the NYBOT coffee market was in positive territory
from middle October to early December, following which for three weeks in December, net
fund length in the market fell temporarily into negative territory. Since end-December net
fund length has been steadily increasing, from 1.1K lots to 23.7K lots at end-January. NYBOT
coffee prices have been supported by strong speculative interest; however, this also makesthe market prone to extreme volatility.
In Q3 05, the supply outlook for coffee was bleak with potential damage caused by
hurricanes in the US Gulf and in Central America, drought in Vietnam and fears of frost in
Brazil. However, at the onset of 2006, the supply outlook is more positive, with limited
hurricane-induced damage, no Brazilian frost and the end of drought in Vietnam’s key
coffee regions. The EIU estimates 2005/06 global production at 89mn bags, and global
consumption at 86.3mn bags, taking the coffee market into a surplus following two
successive deficits. While this would indicate potentially lower prices; there are certain
short-term factors supporting NYBOT coffee prices. These include strong seasonal demand,
lowered stocks in the world’s biggest producer, Brazil, coupled with strong Brazilian coffee
consumption, which is projected at a record high. Brazil’s stocks have been projected by its
crop supply agency Conab to fall to 5.6mn bags in March, the lowest in decades. On the
back of these modestly positive short-term fundamentals as well as strong fund interest in
the NYBOT coffee market, we view coffee positively in Q1.
Figure 562: The coffee market moves into a surplus in
2005/06
-4
-2
0
2
4
6
8
10
12
1996/97 1998/99 2000/01 2002/03 2004/0520
40
60
80
100
120
140
160
180
200Coffee market balance (mn 60kg bags,LHS)
NYBOT coffee prices (cents/lb, RHS)
Source: EIU, Barclays Capital.
Figure 563: Brazilian stocks trend lower, while its coffee
consumption is at a record high
0
10
20
30
40
50
60
70
80
60/61 71/72 82/83 93/94 04/05
6
8
10
12
14
16
18Brazil stocks (LHS)
Brazil cnsumption (RHS)
mn bags mn bags
Source: USDA, Barclays Capital.
Akin to the majority of soft commodities, dynamic growth in coffee demand continues to
come from emerging market economies such as Russia, Eastern Europe, Brazil and parts of
Asia. However, the majority of this coffee demand is price-sensitive, and therefore is
strongly influenced by any rise in international prices. Brazilian demand continues to rise
steadily, galvanised by marketing by domestic coffee producers, and is estimated by the
International Coffee Organisation to have risen by 6.5% Y/Y to 16.5mn bags in 2005,
well above the 1.4% Y/Y rise in global consumption projected by the EIU. Coffee
demand from China similarly remains robust, reflected in China’s 2005 coffee importsposting a 13.3% Y/Y rise at 16.3Kt, albeit from low levels. Urbanisation, westernisation
and rising per capita income suggest that Chinese coffee demand will remain buoyant.
NYBOT sugar prices continued to post gains through the last quarter of 2005 before
surging into 2006. Prices started 2006 on a firm note, in the upper ranges of 14¢/lb
before leaping past the18¢/lb and then 19¢/lb in early February.
Despite their impressive gains, we continue to remain bullish on sugar’s priceperformance due to the strong market fundamentals: the market falling into a deficit,
strong demand levels for both sugar and ethanol and depleting ending stocks.
However, this rally has also been strongly fuelled by speculative interest and the
positive investor sentiment has provided both support and buoyancy to prices.
A key factor driving prices higher has been ethanol demand for flex-fuel cars and
the increasing amount of sugarcane diverted into ethanol production by the world’s
largest producer and exporter of both sugar and ethanol: Brazil. This also
establishes Brazil as the most important supply-side dynamic for both markets.
Demand from emerging markets like China, Russia and India continues to be
robust. US sugar imports have also received a fillip due to a fall in production in thewake of damage caused by Hurricane Katrina last year.
Figure 570: Sugar prices propel ahead through Q4 05 and into 2006
0
2
4
6
810
12
14
16
18
20
Jan 97 Apr 99 Jul 01 Oct 03 Jan 06
NYBOT Sugar Prices (Usc/lb)
Source: EcoWin.
Figure 571: Global supply and demand summary for sugar
02/03 03/04 04/05E 05/06E 06/07E
Production (mn tonnes)
Global 148.2 141.9 141.3 145.7 148.7
Y/Y change -4.3% -0.4% 3.1% 2.1%
Brazil 24.4 27.1 29.0 30.3 31.3
Share of global 16.5% 19.1% 20.5% 20.8% 21.0%
Y/Y change 11.1% 7.0% 4.5% 3.3%
Consumption
Global 139.5 141.7 144.8 147.9 151.2
Balance 8.7 0.2 -3.5 -2.2 -2.5
Total Stocks 68.2 68.4 64.9 62.7 60.2
Weeks of consumption 25.4 25.1 23.3 22.0 20.7
Annual Average Price (cents/lb) 7.3 6.3 8.9 12.8
Note: The marketing year for sugar runs from September to August. Average prices for 2005/06 are to the 31 st
of January 2006. Source: Economist Intelligence Unit, EcoWin, Barclays Capital.
The rally in front-month NYBOT sugar prices through Q4 2005 was consistent and
steady in comparison with the sharp upward swings witnessed at the start of 2006.
NYBOT sugar prices in Q4 rose steadily from trading within ranges of 11¢/lb to 12¢/lb
in mid-November, breaching 13 ¢/lb in early December and 14¢/lb mid-December.
Sugar prices opened January 2006 by trading around 14¢/lb before moving up rapidly
breaching 18¢/lb and then 19¢/lb in early February. While sugar market fundamentals
are strong, prices have also been buoyed by positive investor investment. While fundnet length in sugar has scaled down from the peaks reached in end-September last
year, placed in a historical context it is still considerable (see Figure 580).
Although the production outlook for 2005/06 is stronger in comparison to the previous
three years, estimated by the EIU at 145.7mn tonnes, the strong rise in sugar
consumption levels taking it to 147.9mn tonnes would imply that the sugar market
once again falls into a deficit in 2005/06. Lowered sugar production due to adverse
climatic conditions is expected from Thailand, which in the past has been an important
sugar exporter. Further, lowered supply can be expected going forward due to reforms
in the EU’s sugar subsidy system. The key supply side dynamic however is the diversion
of sugarcane into the production of ethanol. An estimated 52.5% of 2005/06’s Brazilian
cane was used in ethanol production and the remainder for sugar, a ratio that is expected to
rise to 55% in ethanol’s favour in 2006/07compared with 51.1% in 2004/05.
Figure 572: Strong sugar demand from most key
emerging economies...
2
7
12
17
22
92/93 95/96 98/99 01/02 04/05
Brazil India China Russian Federation
mn tonnes
Source: USDA, Barclays Capital.
Figure 573: …which are also experiencing declining
stocks
0
2
4
6
8
10
12
14
92/93 94/95 96/97 98/99 00/01 02/03 04/05
Brazil China India Russian Federation
mn tonnes
Source: USDA, Barclays Capital.
Sugar demand levels have been robust, especially in key emerging economies such as Brazil,
China, India and Russia. India, the world’s biggest sugar consumer is projected by the EIU to
use 20.9mn tonnes (up from 20.3mn) in 2005/06. Similarly, Chinese demand is forecast at
12.2mn tonnes by the EIU (up from 11.7mn in the previous year). Rising domestic sugar
prices in China prompted the government to make a series of sales from its reserves over
the course of 2005 due to a production fall arising from drought in key sugar producing
areas. In early February 2006 it was reported that China was planning to release an
additional million tonnes from its state reserves and could remove import tariffs to ease
tightness in the domestic market and contain prices. Russia, the world’s largest sugar
importer, posted a 12% Y/Y increase in sugar imports in 2005. Russia also reduced its raw
sugar cane import tariff to $140/t in October last year (from $164/t). In addition to this
strong emerging market demand, US sugar imports have risen due to the production
shortfall in the wake of Hurricane Katrina. The USDA in February 2006 said that it would be
allowing an extra 500,000 short tons of sugar imports and also permit sale of all US
produced sugar in 2006 to tackle the tight supply situation in the US market.
High and volatile oil prices in conjunction with geopolitical tensions have highlighted
the need for alternative fuel production. Key among these alternative fuels is ethanol,
made from sugar in Brazil and from corn in the US. While there has been a palpable
shift in sugar prices, corn prices have not received any sustained lift from their link to
ethanol. This we believe is due to lower US ethanol demand compared to Brazilian
ethanol demand, the percentage of corn made into ethanol is lower than the amount of
sugarcane in Brazil, and because supplies are tighter in the sugar market than in the
corn market. With increased demand for ethanol as a cheaper supply of fuel in the context
of the current geopolitical environment, sugar’s exposure to oil via ethanol has added both
volatility and increased liquidity to the sugar market with increased speculative interest.
Strong ethanol demand has been reported from Brazil fuelled by the popularity of flex-fuel
cars (cars using a combination of gasoline and ethanol). Sales of flex-fuel cars in January
2006 came in at 73% of all new car sales, compared to 27.5% of cars in January 2005. Steepdomestic ethanol prices rises in Brazil led the government in January 2006 to discuss with
mills to stop the spike in prices. The outcome of the discussions is that the cane mills have
promised to contain ethanol prices. Brazil’s Agriculture Minister said that Brazil would have
to plant an additional 2.5mn hectares of sugarcane by 2012 in order to be able to meet a
projected rise in ethanol demand, implying a 42% expansion of Brazil’s officially estimated
sugarcane area of 5.88mn hectares. Strong profits were reflected in Brazil’s cane mills in
2005, with millers reporting record revenue of $4.7bn in sugar and ethanol exports, up
sharply from $3.1bn in 2004. A survey by the Sao Paulo Cane Agroindustry Union (Unica)
reported that 82 projects have been announced, with cane millers estimated to invest
approximately $6bn to increase ethanol production capacity and renovate factories
However, in light of the vast amount of arable land in Brazil, if sugar and ethanol prices
continue to trade at firm levels, increased cane plantings can be expected from Brazil which
continues to hold the best prospects of servicing the global sugar and ethanol market. The
downside of the world’s dependence on Brazil for supplying sugar and ethanol is that it
makes the market vulnerable to supply shock.
Figure 574: Brazil: the key supplier of world sugar
In light of cotton’s positive market fundamentals and supportive market sentiment,
we view cotton prices as having an upside bias.
According to the USDA, the cotton market is expected to fall into a 2.9mn bale
deficit due to lower global production in 2005/06 in conjunction with robust globalcotton use, which is expected to rise to a record 115.2mn bales.
If the less-than-ideal climatic conditions in key cotton-growing areas of the world’s
second-largest cotton producer, the US, continue into the planting season, lowered
US supply should buoy prices higher.
China’s cotton import figures also paint a positive picture of demand fuelled by its
textile industry, with cotton imports in 2005 posting a 35% Y/Y rise. China’s cotton
imports in 2005/06 are projected by the USDA to more than double from
2004/05’s 6.4mn bales to 16.5mn bales.
In addition to textile exports by China, cotton demand for domestic use in both
China and India is expected to rise.
Figure 581: Cotton futures prices (cents/lb)
25
35
45
55
65
75
85
Jan 00 Jul 01 Jan 03 Jul 04 Jan 06
NYBOT Cotton Prices (cents/lb)
Source: EcoWin.
Figure 582: Global supply and demand summary for cotton
00/01 01/02 02/03 03/04 04/05E 05/06E
Production (mn 480lb bales)Global 88.9 98.8 88.3 95.3 120.4 112.4
Y/Y change 1.3% 11.1% -10.6% 7.9% 26.4% -6.7%
China 20.3 24.4 22.6 22.3 29.0 24.5
Share of global 22.9% 24.7% 25.6% 23.4% 24.1% 21.8%
Y/Y change 15.3% 20.2% -7.4% -1.3% 30.0% -15.5%
Use
Global 92.2 94.3 98.3 98.1 108.6 115.2
Y/Y change 1.4% 2.3% 4.2% -0.2% 10.8% 6.1%
Balance -3.3 4.4 -10.1 -2.8 11.7 -2.9
Total Stocks 46.8 52.1 42.3 40.7 51.6 50.8
Weeks of consumption 26.4 28.7 22.4 21.6 24.7 22.9
Annual Average Price (cents/lb) 54.9 37.1 50.8 64.0 48.7 51.7 Note: The cotton marketing year runs from August to July. Average prices for 2005/06 are to the 31st of
January 2006. Source: USDA, EcoWin, Barclays Capital.
Front-month NYBOT cotton prices through Q4 05 held over the 50 cents/lb level until
mid-November, when they dipped to a low of 48.3 cents/lb before staging a recovery in
early December. Cotton prices in Q4 averaged 52.3 cents/lb, marking a rise from Q3’s
average of 49.5 cents/lb and traded in a range between 48.3 cents/lb and 57.8
cents/lb. Prices at the start of Q1 06 have held steady over 50 cents/lb and in our view,
look poised to rise higher on the back of a supportive demand and supply scenario. In
line with this, speculative positive exposure in the NYBOT cotton market has alreadystarted to rise, with fund net length turning positive, after being negative since
November.
Global production in 2005/06 is forecast by the USDA at 112.4mn bales, which
although falling well short of the previous year’s record harvest of 120.4mn bales, is the
second highest harvest on record. However, in marked contrast to the ideal climatic
conditions that prevailed in the US in 2004/05, extreme dryness, strong winds and a
lack of rainfall in key cotton producing regions of the US – Texas and the Southwest
Plains – represent a potential setback to supply. If the adverse climatic conditions
continue into the planting season, cotton prices should be supported higher, given that
the US is the world’s largest cotton exporter and second-largest producer.
On the demand side, global cotton use is projected by the USDA to come in at a record
115.2mn bales, taking the cotton market into a 2.9mn deficit and thereby again
providing a rather supportive backdrop to cotton prices. The key demand side dynamic
in the market is Chinese cotton demand and imports, which in turn feed into China’s
rapidly expanding textile exporting industry. With the end of the Multi-Fibre
Agreement (MFA) at the start of 2005, global textile trade has received a fillip,
especially with regards to textile exports from China and India. Chinese textile exports
to the US and Europe rose by 62.7% for the first 10 months of 2005, and total Chinese
textile exports posted a 20% increase, to amount to US$116bn. In January 2006, China
issued an additional 1.5mn tonnes of cotton import quotas in order to supply its textile
industry. The US continues to be the biggest cotton exporter to China, but Indian cotton
exports to China have also risen, now totalling over half of India’s total cotton exports
(which have more than doubled from the previous year to a projected 1.8mn bales).
Figure 583: Chinese cotton imports more than double…
Front month CBOT wheat prices in Q4 05 traded in a range between $2.93/bushel
and $3.52/bushel, averaging $3.19/bushel through the quarter. Prices at the start
of 2006 have been firm, especially towards the end of January.
According to the USDA, the global wheat market is set to fall into a 5.4mn tonnedeficit in 2005/06, compared to the previous year’s 17.9mn tonne surplus.
A deficit in the wheat market combined with declining global stocks, dry
unfavourable conditions in parts of the US and excess cold in Eastern Europe reflect
the upside potential for wheat prices. However, bird flu has been a dampener on
feed demand and its spread could exert strong downward pressure on prices.
China’s wheat production in 2005/06 is forecast by the USDA at 97mn tonnes,
posting a 5mn tonne increase over the previous year. China’s wheat consumption
on the other hand, is forecast to decline by a million tonnes, as are imports, which
are expected to fall steeply from 6.8mn tonnes in 2004/05 to 2mn tonnes in
2005/06, owing to changing consumption patterns and lessened feed demand dueto the outbreak of bird flu.
Figure 591: CBOT front-month wheat prices
2
3
4
5
6
7
8
Jan 95 Oct 97 Jul 00 Apr 03 Jan 06
CBOT Wheat Prices ($/bushel)
Source: EcoWin.
Figure 592: Global supply and demand summary for wheat
00/01 01/02 02/03 03/04 04/05 05/06E
Production (mn tonnes)
Global 581.4 581.1 567.7 554.6 626.6 616.4
Y/Y change 0.0% -2.3% -2.3% 13.0% -1.6%
China 99.6 93.9 90.3 86.5 92.0 97.0
Share of global 17.1% 16.2% 15.9% 15.6% 14.7% 15.7%
Y/Y change -5.8% -3.8% -4.2% 6.3% 5.5%
Consumption (mn tonnes)
Global 583.8 585.2 604.0 588.5 608.7 621.8
Y/Y change 0.2% 3.2% -2.6% 3.4% 2.2%
Balance -2.4 -4.1 -36.3 -33.9 17.9 -5.4
Total Stocks 206.5 202.5 166.1 132.2 150.1 144.7
Weeks of consumption 18.4 18.0 14.3 11.7 12.8 12.1
Annual Average Price ($/bushel) 2.61 2.79 3.38 3.69 3.14 3.23 Note: The wheat marketing year runs from July to June. Average prices for 2005/06 are to the 31 st of January
CBOT wheat prices in Q4 05 traded between $2.93/bushel and $3.52/bushel, averaging
$3.19/bushel through the quarter, as compared to Q3’s average price of $3.24/bushel.
Prices started Q4 05 on a firm note, but began to fall from mid-October, reaching their
lowest levels through November and early December, falling below $3/bushel. Prices
staged a recovery by end-December and have traded over $3.30/bushel through most
of January, peaking at month end. The outlook for wheat prices in Q1 is positive in our
view, in light of declining stocks, the market deficit coupled with dry arid conditions inthe US and fear of damage to wheat crops by the excess cold in Eastern Europe.
However, significant price appreciation also depends on bird flu-related developments,
which in the past have negatively affected feed demand. Further, investor sentiment
towards wheat also continues to be negative, with net fund position in the CBOT wheat
market being short since April 2005 (see Figure 600), but this could spur short covering
on the back of weather related supply disruption.
Global production for 2005/06 is estimated at 616.4mn tonnes by the USDA, posting a
1.6% Y/Y fall from the previous year’s record crop of 626.6mn tonnes. As compared to
the previous year, higher production is forecast for China, Australia, Canada and Russia,
while lowered production is forecast for Argentina, the EU and the US. If the dry arid
weather in wheat-producing parts of the US continues, and/or if the frost in Eastern
Europe affects wheat crops, wheat supply could suffer and prices could rise in response
to both regions being major wheat producers. Global wheat consumption is projected
at a record 621.8mn tonnes, taking the wheat market into a 5.4mn tonne deficit, with
higher consumption anticipated from the EU, Russia, the US and Ukraine.
According to the USDA, China’s wheat stocks are expected to fall in 2005/06, marking
the sixth year of stock drawdowns, while domestic wheat production is expected to rise
for the third consecutive year and consumption is forecast to fall for the sixth
successive year. This does not bode very well for the wheat market in 2005/06, given
that China, a major grain consumer is expected to make limited imports. Customs data
reveals that for 2005 as whole, China’s wheat imports came in at 3510kt, a 51.4% Y/Y
decline. China’s wheat consumption has been on a declining trend due to urbanisation,
changed diets containing less grain, and rural-urban migration.
Figure 593: Chinese use of wheat as feed falls sharply,
while world wheat feed use rebounds
0
20
40
60
80
100
120
140
60/61 69/70 78/79 87/88 96/97 05/06
0
2
4
6
8
10
12World use of wheat as feed (mn tonnes)
China use of wheat as feed (mn tonnes)
Source: USDA, Barclays Capital.
Figure 594: China’s wheat stocks have posted successive
• Front month CBOT corn prices traded in a 186-216¢/bushel range through Q4 05,
averaging 199¢/bushel. Owing to strong projected demand levels, ethanol demand and
positive fund sentiment, we expect CBOT corn prices to rise in 2006.
•
The USDA estimates production in 2005/06 at a strong 683.5mn tonnes, but below2004/05’s record 708.3mn tonne crop. High levels of production coupled with
large stocks can be expected to cap the upside.
• World corn consumption is projected at a historical high of 686.2mn tonnes. This
takes the corn market into a modest 2.7mn tonne deficit after the previous year’s
large surplus. Ethanol production is slated to be a supportive end-sector for corn
use going forward.
• China’s corn crop in 2005/06 is projected at a record 134mn tonnes, while demand
is robust, galvanised by feed demand for China’s livestock industry. Viewed in
tandem with China’s declining stocks, falling exports and a strong rise in imports,
Chinese demand is a key dynamic in the corn market. However, the outbreak of birdflu dampened sentiment towards corn in 2005, and if it spreads in 2006, it holds the
potential to act as a dampener on corn feed demand.
Figure 601: Prices fall on the back of strong corn production forecasts
150
200
250
300
350
400
450
500
550
Jan 95 Nov 96 Sep 98 Jul 00 May 02 Mar 04 Jan 06
CBOT Corn Prices (¢/bushel)
Source: EcoWin.
Figure 602: Global supply and demand summary for corn00/01 01/02 02/03 03/04 04/05 05/06E
Production (mn tonnes)
Global 590.0 598.9 601.7 623.0 708.3 683.5
Y/Y change 1.5% 0.5% 3.5% 13.7% -3.5%
United States 251.9 241.4 227.8 256.3 299.9 282.3
Share of global 42.7% 40.3% 37.9% 41.1% 42.3% 41.3%
Y/Y change -4.2% -5.6% 12.5% 17.0% -5.9%
Consumption (mn tonnes)
Global 609.8 622.5 625.5 644.4 681.1 686.2
Y/Y change 2.1% 0.5% 3.0% 5.7% 0.7%
Balance -19.8 -23.6 -23.8 -21.3 27.2 -2.7
Total Stocks 172 149 125 104 131 128
Weeks of consumption 14.6 12.4 10.4 8.4 10.0 9.7Annual Average Price (cents/bushel) 208.3 216.9 235.7 261.4 209.4 202.7 Note: The corn marketing year runs from October to September. Average prices for 2005/06 are to the 31 st of
January 2006. Source: USDA, EcoWin, Barclays Capital.
Front month CBOT corn prices through the fourth quarter of 2005 were weak,
averaging 199¢/bushel and trading in a range between 186¢ and 216¢/bushel. Prices
started the quarter, holding above 200¢/bushel until the end of October after which
they rose over 200¢/bushel in mid-December. Prices have been firmer at the start of
this year, rising to a high of 222¢/bushel at the end of January. Much of the rise in
prices from December onwards can be attributed to speculative activity, with fund
length in CBOT corn turning positive from end-December and all through January. Thisis quite in contrast to the period from end-August to December 2005 when investors
were shorting the CBOT corn market (see Figure 611).
Global corn production in 2005/06 is projected by the USDA at 683.5mn tonnes, the
second highest corn crop ever, following the previous year’s record harvest of 708.3mn
tonnes. Much of this strong level of production can be attributed to US corn production,
which although posting a fall from 2004/05’s record crop, is estimated to have its
second highest crop, at 299.9mn tonnes. The other country with projected strong
production levels in 2005/06 is China, which is expected to have record production of
134mn tonnes. While global production is slated to be high, global corn consumption
levels projected by the USDA, at 686.2mn tonnes are a record as well. Strong demand
levels are expected from China (+1.9%Y/Y), the US (+1.3%Y/Y) as well as Brazil
(+3.9%Y/Y), taking the corn market into a 2.7mn tonne deficit. However, corn stocks
are high, a result of large carryover stocks resulting from the previous year’s record
harvest, at 128.3mn tonnes.
Figure 603: World corn consumption keeps pace with robust levels of
Front-month CBOT soybean prices were steady through Q4 05, trading between a
low of $5.4/bushel and a high of $6.2/bushel and averaging $5.8/bushel. In light of
the strong production figures, the market surplus and the record level of global
soybean stocks, we do not expect any appreciable rise in soybean prices.
The USDA projects 2005/06 global soybean production at a record high of 223mn
tonnes, with robust production from key exporters such as Brazil, Argentina and
the US. Soybean crushings are also expected to rise to a record 184.8mn tonnes in
2005/06, as is total soybean use, at 214.8mn tonnes.
Global 2005/06 soybean ending stocks are estimated by the USDA at a record high
of 53.2mn tonnes, buoyed by an almost doubling of US soybean stocks.
The use of soy oil in bio-diesel production is a supportive factor for soybean prices
going forward. For soymeal, demand is strongly tied to protein demand for feed,
which was hit in 2005 due to the outbreak of bird flu. If it were to spread in 2006,
prices would be weighed down.
Figure 612: CBOT Soybean prices ($/bushel)
4
5
6
7
8
9
10
11
Jul 00 Jun 01 May 02 Apr 03 Mar 04 Feb 05 Jan 06
CBOT Soybean Prices ($/bushel)
Source: EcoWin.
Figure 613: Global supply and demand summary for soybean
00/01 01/02 02/03 03/04 04/05 05/06E
Production (mn tonnes)
Global 176.0 185.1 197.0 186.8 215.3 223.0
Y/Y change 5.2% 6.5% -5.2% 15.3% 3.6%
United States 75.1 78.7 75.0 66.8 85.0 84.0
Share of global 42.6% 42.5% 38.1% 35.8% 39% 38%
Y/Y change 4.8% -4.7% -11.0% 27.3% -1.2%
Use (mn tonnes)
Crushings 146.7 158.2 165.6 163.6 176.0 184.8
Total use 173.1 184.0 190.8 190.0 205.8 214.8
Y/Y change 7.8% 6.3% 4.7% -1.2% 7.6% 5.0%
Balance 2.9 1.1 6.2 -3.2 9.6 8.3
Total Stocks 31.9 33.3 40.5 35.7 44.9 53.1
Weeks of consumption 9.6 9.4 11.0 9.8 11.3 12.9
Average Annual Price ($/bushel) 4.7 4.8 5.8 8.1 6.0 5.8 Notes: The soybean marketing year runs from September to August. Average prices for 2005/06 are to the 31 st
of January 2006. Source: USDA, EcoWin, Barclays Capital.
Front-month CBOT soybean prices through Q4 were steady, ranging from a low of
$5.4/bushel to a high of $6.2/bushel and averaging $5.8/bushel through the quarter. Net
fund length in the CBOT soybean market was positive from the start of Q4 to mid-
November, after which net fund position was short until mid-December. In light of the
strong production figures, the market surplus and the record level of global soybean
stocks, we do not expect any appreciable rise in soybean prices. Funds once again started
shorting the CBOT soybean market during the final three weeks of January (see Figure 621)
Global soybean production for 2005/06 has been estimated at a record level of 223mn
tonnes by the USDA. Robust production has been forecasted for the world’s biggest
soybean producer, the US, at 84mn tonnes, although this is below last year’s record harvest
of 85mn tonnes. Higher production has also been estimated for key South American
soybean exporters, Brazil (at 58.5mn tonnes) and Argentina (at 40.5mn tonnes).
Global crushings are also projected by the USDA to be a record 184.8mn tonnes (+5%Y/Y)
and global use to have risen to 214.8mn tonnes reflecting strong soymeal demand for feed
as well as soyoil demand. Global soybean stocks are also projected at a record high of
53.2mn tonnes and much of this rise in stocks can be attributed to US stocks almost
doubling from 7mn tonnes in 2004/05 to 13.7mn tonnes in 2005/06. This strong level ofcarryover coupled with strong production and the uncertain demand outlook due to the
bird flu can be expected to weigh down on market sentiment.
Figure 614: World soybean production sets a record at
223mn tonnes in 2005/06
0
50
100
150
200
250
64/65 70/71 76/77 82/83 88/89 94/95 00/01
World productionmn tonnes
Source: USDA, Barclays Capital.
Figure 615: Chinese soybean imports continue to post
strong rises while production steadies
0
5
10
15
20
25
30
64/65 72/73 80/81 88/89 96/97 04/05
Production Importsmn tonnes
Source: USDA, Barclays Capital.
Chinese soybean production in 2005/06 is estimated by the USDA to be 17mn tonnes, akin
to the previous year’s production level of 17.4mn tonnes, while China’s soybean crushings
are forecast to post a 14% Y/Y rise, to a record 34.6mn tonnes, also highlighting the
country’s overcapacity in crushing. Strong levels of Chinese soybean demand are reflected
in trade data released by Chinese customs, which reveals that China imported 26.6mn
tonnes of soybean in 2005, posting a 31.4% Y/Y increase. For 2005/06, the USDA estimated
China’s imports at 27.5mn tonnes. The strong level of Chinese soybean imports has been
propelled primarily by the use of soymeal as a protein additive to feed for China’s expanding
livestock and poultry industry, demand for soy oil and to a lesser degree for use in bio-diesel
production. In 2005, however, demand received a setback due to the outbreak of bird flu
and into 2006, this continues to be a key variable determining Chinese soybean demand.