Outlook for metals and minerals - Investor seminar

Outlook for metals and minerals – Investor seminar

Outlook for metals and minerals – Investor seminar

The following paper from Rio Tinto chief economist Vivek Tulpulé has been issued today to
coincide with the Rio Tinto Investor seminar.

Executive summary

. Commodity markets are entering a fifth straight year of growth with mineral and metal
prices at levels well above their long term average and in many cases above levels at
the start of this year.
. Firm global economic activity led by China is expected to support strong increases in
demand for most metals and minerals over 2008 and 2009.
. With low stocks and a likely continuation of supply side difficulties, most commodity
prices are expected to remain well above their long run trend over the short and medium
term.
. It is too early to suggest that the current price cycle has peaked across the range of
commodities.
. While the central case is positive, we are mindful of the short term risks associated with
the predicted slowdown in the US economy.
o But, it is important to recognise that the United States is now significantly less
important in world commodity demand than it was just five years ago.
o Additionally our analysis suggests that even a sharp slowing in the US economy
would have only a small impact on Chinese and Indian economic growth and
consequent demand for commodities.
. Viewed from a longer run perspective recent history and the IMF’s forecasts suggest that
we are currently going through a period of global growth not seen since the period of fast
growth and reconstruction in OECD economies following World War 2.
. Specifically, there has been a structural shift favouring rapid growth in developing
countries with large populations such as China and India. Growth in these economies will
be resource intensive as they industrialise and urbanise.
. The implications for commodity markets are nothing short of profound. Projections for
iron ore, aluminium and copper suggest that demand could double and even triple over
the next 25 years.
. In time production can be expected to expand to meet faster growth in demand at more
sustainable prices. But that pricing environment is expected to be significantly stronger
than would be implied by historical trends.
o It is expected that prices of many minerals and metals will remain elevated above
trend for longer than has been the case in the past because of constraints on the
speed with which production capacity can be expanded over the next few years.
o Also most prices are expected to assume significantly higher average levels over
the very long run than has been the case historically due to structural increases
in industry costs.
. We present case studies relating to markets for aluminium, copper and iron ore – three
commodities that are expected to be drivers of the industrialisation and urbanisation
process in developing countries.
Iron ore
o Substantial growth is expected in the demand for iron ore reflecting expected strong
growth in steel demand related to the processes of ongoing industrialisation and
urbanisation in the developing world.
o Reflecting the current tight market spot prices have risen sharply over the last few
months with Indian ores currently selling in China at around $190/tonne double their
price at the beginning of 2007. Australian and Brazilian ores are selling at a
substantial discount to this spot rate given current freight rates.
o A substantial amount of high cost production will be required to meet growing
demand over the long term. This strongly suggests the possibility of higher long run
prices & higher margins for traditional lower cost producers.
Aluminium
o Prices are currently in the range of $2450-2550/t – levels supported by industry cost
structures.
o Aluminium consumption has grown the fastest of all non-ferrous metals over the last
5 years and is forecast to grow rapidly over the next 20 years.
o There has been enormous recent growth in Chinese consumption and production but
aluminium has benefited from increasing intensity in many other regions including the
OECD.
o Constraints on China’s domestic bauxite production suggest that the country’s
massive investment in aluminium capacity will remain reliant on imported bauxite.
This combined with China’s high power cost environment mean that Chinese
aluminium capacity will continue to be high cost on a global scale.
o Additionally, Chinese production will also be disadvantaged by a stronger currency
as the RMB edges toward fair value over time.
o The implied increase in the marginal cost of production for alumina and aluminium
means that their prices are unlikely to revert to the lower levels implied by historical
trends
Copper
o Reflecting the tight market situation, copper prices are currently in the range of
$6400/t-$7000/t – about three times higher than their average level through the 1990s
and well above levels achieved in the early part of this decade.
o Prices could remain near current levels as long as production growth continues to
under-perform against the underlying demand trend creating a need to ration
supplies.
o Strong Chinese demand growth is expected next year and on the supply side the
likelihood of ongoing disruptions and possible constraints on the availability of
sulphuric acid affecting SxEw operations are issues.
o The importance of investment funds in exchange traded commodity markets means
that large price movements could take place on the back of commodity specific
speculative shifts or broader shifts in investor sentiment – well in advance of any
fundamental change in physical markets.
o Looking to the long run, strong demand growth prospects are based on the expected
resource intensive development of economies such as China and associated
investment in power distribution networks and other infrastructure.
o On balance, we believe that, as for many other commodities, there has been a
structural shift in copper costs supporting the expectation of significantly higher long
run prices than would be implied by historical trends.

A fundamental shift toward fast and resource intensive growth

As 2007 draws to a close, resource markets are entering a fifth straight year of cyclical
strength with virtually all minerals and metals prices at levels significantly above their long
run historical trends and in many cases above start of year levels.

Looking forward, global GDP growth is expected to be firm in 2008 and 2009 with rapid
growth in China and other developing countries expected to reduce any drag from slower
growth in OECD countries. Such conditions should create a basis for continued strong
underlying commodity demand over the medium term. At the same time, growth in
production for a number of commodities is expected to remain relatively constrained. In this
context, it is entirely possible that some commodity prices may not have reached their
cyclical peaks as yet.

Indeed, illustrating the significance of current commodity market developments in a historical
context, if as expected, most prices remain well above trend over 2008 and 2009 then prices
will have been above trend for 6 years. This would be a 3-in-100 year event for many
commodities.

Viewed in a broader setting, the current strength in most resource markets can be seen as
the result of a fundamental shift in economic forces that is leading to rapid growth in
developing economies with large populations. For example, recent history and the
International Monetary Fund’s projections for future growth would suggest that we are
currently going through a period of global growth not seen since the period of rapid
economic development and reconstruction that followed the Second World War.

China and India provide the most significant recent examples of the current growth
phenomenon but they are not alone. For instance, many countries in the Middle East and
ASEAN are also on fast growth paths. In such economies growth tends to be resource
intensive. In particular, the processes of urbanization and infrastructure development that
accompany early-mid stage productivity growth and industrial development require
increasing utilisation of resources such as steel (and therefore raw materials such as iron
ore), aluminium, copper and energy. At a global level such a resource intensive development
pattern is expected to persist for at least another two decades leading to sustained strong
global demand growth for many commodities.

Prices are determined by both demand and supply and by its nature minerals and metals
production can be difficult to accelerate. But accelerating production in cyclically high
markets can also be very profitable causing a ‘rush’ to meet demand. The current ‘rush’ has
manifested itself in a number of forms including: increasing construction costs and project
delays; higher levels of disruption as existing production systems are stretched; higher
average production costs due to labour and other input cost increases; and higher industry
marginal costs as increasingly expensive production is drawn in to meet demand.

In time, it is expected that sufficient commodity supply growth will be induced to cause prices
to revert down toward more sustainable long-run levels. But the demand and supply
phenomena just described suggest that it will most likely take longer for commodity prices to
return to long-run levels than would have been the case if historical reversion rates had
applied. At the same time long-run prices and in some instances margins are expected to be
significantly higher than would be implied by historical trends.

Medium term expectations are for strong global GDP growth

Against a backdrop of record high oil prices, a rapidly slowing US housing market and a
credit crunch precipitated by the US sub-prime mortgage crisis, the IMF is nevertheless
projecting global GDP growth in 2008 of about 4.8 per cent (in PPP terms)1. This projection
accounts for slowing growth in advanced economies but relatively fast growth in the
developing world. In 2009 a recovery in activity in advanced economies and continued
strong growth in developing economies are projected to generate global growth of around
5 per cent1. Viewed in a historical context such growth rates are high and therefore provide a
positive setting for underlying commodity demand in the medium term.
World GDP Growth (PPP basis)
0
1
2
3
4
5
6
7
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012
Percent YoY
Advanced economies
Developing economies
Source: IMF

China’s GDP has continued to surge, growing at 11.5 per cent (y-o-y) in the first three
quarters of 2007. In this period industrial production grew by about 18 per cent, nominal
fixed asset investment grew by about 26 per cent and the trade surplus soared to about
US$200bn. At the same time inflationary pressures, mainly related to food and energy
prices, have been increasing with consumer and producer price inflation in October at 6.5
per cent and 3.2 per cent respectively. In this setting the government has introduced
measures to restrain liquidity. But even with this tightening, GDP is expected to grow rapidly
at between 10 per cent and 11 per cent in 2008 and around 10 per cent in 2009 based on
strong domestic demand and a strong but moderating contribution from net trade2.

The US economy is expected to grow at around 2 per cent in 2008 as residential
construction continues to fall and private consumption growth slows. On the other hand, a
weaker dollar is expected to generate an improved contribution to GDP from trade. In 2009
residential construction is expected to start growing again and the positive effects of recently
lowered interest rates on investment and disposable incomes should lead to a recovery in
economic activity; GDP growth in the range of 2.5 per cent to 3 per cent is expected in that
year3.

Japan’s growth has been volatile during 2007. It grew strongly by 0.7 per cent (q-o-q) during
Q1, declined by 0.4 per cent in Q2 and then beat analysts’ expectations to grow by 0.6 per
cent in Q34. The contraction in Q2 was partly attributable to reduced residential construction
resulting from the implementation of a stricter building code. In 2008 and 2009 steady growth
in consumer demand and a rebound in residential construction are expected to offset slower
net exports to see overall GDP grow at around 2 per cent5.

The Indian economy grew by more than 9 per cent in each of the first two quarters of 2007
and growth of between 8 per cent and 9 per cent is expected in 2008 and 20096. Capacity
constraints in many parts of the economy are creating inflationary pressures and in response
the Central Bank has tightened monetary policy. While this reduced inflation it probably also
contributed to reduced industrial production growth during the second half of this year.

Expected limited global economic risk from a further US slowdown

Markets have been nervous about the impact of slowing US growth on commodity markets
and speculation about this has had negative effects on exchange traded prices. But in terms
of commodity demand generally, the importance of the United States has declined
substantially relative to that of China since 2000 and in the specific case of seaborne iron
ore, the US is a negligible market participant. In that context, the key issue for the health of
commodity markets over the medium term is the magnitude of any negative spillover effect
from a slowing US economy on economic activity in the rest of the world and China in
particular.

Copper & Aluminium consumption 2007
China
25%
Other –
OECD
40%
US
11%
Other –
Non-
OECD
24%
China
32%
US
15%
Other –
OECD
34%
Other –
Non-
OECD
19%
Copper Aluminium
Source: CRU

One macroeconomic linkage is clear. With slower US private consumption and a weaker
currency, US demand for exports from other regions can be expected to decline and its own
exports to increase. In terms of GDP accounting, this would reduce the net contribution of
the United States to aggregate demand in the rest of the world. But it is easy to exaggerate
the potential flow on effects of this possibility on global economic activity including in Asia.
For example, modelling suggests that a sharp reduction in US consumption and residential
investment during 2008 to levels consistent with a US recession and a weaker US exchange
rate would be expected to reduce Chinese growth by less than a percentage point. This
would still leave scope for Chinese growth at levels approaching 10 per cent. For India, the
impact of any further slowdown in the US would be expected to be smaller because of
India’s more limited exposure to world trade.

The modelling captures the likelihood that, governments and central banks in countries
affected by any US slow down could boost economic activity through monetary and fiscal
responses. Some commentators have noted that such economic pump priming would favour
construction and infrastructure development, which in turn is likely to be a positive for
commodity demand.

Moreover shifts in trade flows and policy responses are only part of the picture. A focus on
these aspects alone ignores any shift in financial flows favouring countries with better
investment prospects. For example modeling based on a framework that focuses on
international financial dynamics suggests that the flow on effects of any slowing in US
growth could be reduced substantially as financial resources shift away from the United
States toward other locations including developing Asia.

Some commodity specific examples further illustrate the point. Chinese consumption of steel
is believed to be affected only marginally by fluctuations in external demand as China’s steel
industry is overwhelmingly focused on meeting needs from the domestic construction sector.
Even in the case of copper, consumption is mostly driven by domestic construction and
infrastructure development which together account for the majority of China’s copper
consumption. Exposure to external conditions arises from China’s position as a major global
supplier of household appliances containing copper components. China has also grown its
exports of semi-fabricated products and copper tubes in recent years, although it remains a
net importer of semis.

It is difficult to put a precise figure on the amount of copper embodied in Chinese trade to the
United States. But even if 20 per cent of total Chinese copper consumption were related to
exports and given that the United States generally accounts for around 20 per cent of
China’s total merchandise exports, the direct exposure of China’s total copper demand
growth to any slowing in US economic activity would be very limited.

Currencies

Since the start of this year the US dollar has weakened appreciably against most other
currencies. Recent falls in the greenback have been driven by perceptions of increased
riskiness in US asset returns in the wake of the sub-prime mortgage crisis; and expectations
that the US Federal Reserve may cut interest rates to address growth concerns while other
central banks may raise rates to combat inflationary pressures.

The currencies of many commodity exporting countries have also been affected by upgrades
to market expectations about future commodity prices and M&A activity. The Australian
dollar has gained about 13 per cent against the US dollar since the start of the year. The
Chilean Peso has gained about 5 per cent, the Brazilian Real has gained about 16 per cent
and the Canadian dollar has appreciated by about 19 per cent7. Such exchange rate shifts
have increased average US dollar production costs for many commodities. But at the same
time, US dollar weakness provides support to prices of commodities that are denominated in
US dollars but with large non-US consumption and cost bases.

Over time, market based exchange rates are likely to fluctuate on ever-shifting speculation
about relative interest rate policies and ongoing concerns about risk and structural
imbalances and commodity prices in the case of large commodity exporters. In the case of
managed currencies, policy and economic pressures on governments or the emergence of
unsustainable foreign exchange flows will have the greatest influence on outcomes. Future
rates of appreciation in the Chinese RMB are of special importance for commodity markets.
Chinese RMB expected to continue to strengthen providing a basis for higher long run prices
A range of empirical analyses and the evidence of burgeoning trade surpluses and foreign
exchange reserves suggest that the RMB is significantly undervalued. The extent of possible
undervaluation is shown in the following chart based on research commissioned by Rio
Tinto. The straight line labeled ‘Fair Value’ shows a statistically determined relationship
between real exchange rates and per capita incomes. This empirical analysis backs the
theoretical argument (known as the Balassa Samuelson effect) that as developing countries
become richer their real exchange rates can be expected to strengthen relative to those of
the richest countries. The Chinese real exchange rate has been and remains below the
estimated fair value line suggesting substantial scope for ongoing revaluation.

Chinese currency is likely to strengthen
0.0
10.0
20.0
30.0
40.0
50.0
60.0
70.0
80.0
90.0
100.0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6
Real Per Capita Income (Relative to the US)
Real Exchange Rate (Index: PPP=100)
Korea 1960 to 2004
1998 Asian Finanial Crisis

‘Fair Value’ real exchange rates as a function of per capita income

Based on simple estimated relationship using pooled data for
50 countries, 4 time periods, pooled data. ‘Fair Value increases as incomes increase
China 1995-2007
The Gap between Chinese Real
Exchange rates and ‘Fair Value’
illustrates scope for real exchange rate
appreciation

Source: Center for International Economics
The Chinese authorities have progressively allowed the currency to appreciate in nominal
terms against the US dollar. Pressure on the RMB to strengthen further is expected to
continue, both to address political frictions associated with the trade surplus and constraints
on monetary policy associated with a managed currency.

Importantly this process of revaluation has been increasing the US dollar costs incurred by
trade exposed industries including metals and minerals producers. In aluminium and iron ore
Chinese producers are already among the highest cost producers in the industry and
therefore any currency appreciation would tend to increase marginal industry costs for those
commodities over time. In turn, this cost increase provides a basis for higher global prices.

Long run economic developments and implications for commodity markets

Long Run Growth and Development Entering a New Elevated Phase

It is important to reiterate that the IMF’s growth projections for 2008 and 2009 are high when
viewed in a historical context. This results from an expected structural shift (rather than
cyclical move) in global economic activity based on the ongoing economic emergence of
China and other developing economies such as India. Indeed, the expected shift would take
world growth to levels not seen since the period of rapid growth and reconstruction in OECD
countries just after the Second World War. The longer term implications of this shift for
commodity markets are potentially profound.

Developing world’s share of the world economy is
expected to almost double
World GDP
0
50
100
150
200
250
300
350
2005 2025 2050
Trillion $ (2000 terms, PPP basis)
Rest of world
India
China
Eastern Europe
South America
Asia Pacific Developed
Western Europe
North America
Source: World Bank for base 2005 data; Global Insight for growth estimates to 2025;

Rio Tinto estimates for 2050 data

Developed world Developing world

Our revised analysis of longer run Chinese growth prospects suggests that GDP growth can
be expected to average levels approaching 9 per cent per annum in the period to 2015
providing a sustained basis for strong commodity demand growth. In a similar vein, a recent
study carried out by the Development Research Centre under China’s State Council
concluded that China’s industrialization stage will last into the 2020’s with potential GDP
growth rate at around 10 per cent in the period to 2015 and 8 per cent from 2015 to 2020. A
study by Australian National University academics, Garnaut and Song, suggests that China
is reaching a ‘turning point’ in its growth creating potential for highly resource intensive
growth in excess of 10 per cent over the next two decades.

Indian growth has lagged behind China’s despite both countries having had similar per
capita incomes in 1980. In more recent years Indian growth has accelerated and importantly
it has become less variable. Studies by Rio Tinto suggest that India has the potential to grow
at a sustained rate of around 10 per cent for at least a decade if key economic reforms are
undertaken. Studies by banking research groups produce similar results with long run growth
potential estimated to be in the 8 per cent-10 per cent range8. Most of this research points
out that continued reforms that free up the ability of Indian industry to respond to price
signals remain crucial both to allow more rapid allocation of resources to their most profitable
uses and reduce inflationary risks. In India’s case it is arguable that the turning point
favouring highly resource intensive growth, identified for China by Garnaut and Song, is still
to be reached.

Commodity demand expected to grow strongly

The shift toward faster and more resource intensive global growth led by developing
countries has led to strong rises in commodity demand over recent years. But per capita
consumption remains relatively low in those countries suggesting scope for strong growth
well into the future. For example, while China accounted for 60-90 per cent of the increase in
global demand for steel, aluminium and copper between 2000 and 2006 its per capita
consumption of those metals remains well below that of many OECD countries9. For
example in 2006 Chinese per capita steel consumption was about 60 per cent of that in the
OECD average while its per capita copper and aluminium consumption was at around onethird
of OECD levels. The implication is that Chinese demand for these commodities – and
associated raw material inputs – has substantial scope to continue to grow rapidly for some
time. Indian per capita consumption is a fraction even of China’s consumption, suggesting
scope for rapid sustained demand growth over the longer run.

Empirical analysis based on historical patterns of commodity consumption and differences in
commodity consumption between countries shows a relationship between per capita
incomes and commodity consumption – typically known as ‘commodity S-curves’. The Scurves
suggest that as incomes grow per capita consumption of commodities increases. At
first growth is more rapid as economies industrialise, build urban environments and
commercial infrastructure and then growth slows for rich countries as consumption per head
approaches saturation. Such S-curves are shown for a range of commodities in the graph
below.
0%
5%
10%
15%
20%
25%
30%
0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000
0
10
20
30
40
50
60
Aluminium
GDP per capita (in 2000 US$)
Expenditure per capita
US$ (2007 terms)
2007 Population
Distribution
Copper
World average income per capita
2007 2022
Source: Global Insight for population distribution; Rio Tinto estimates for commodity expenditure profiles.
Note: Expenditure profiles are based on Rio Tinto estimates of global income and consumption relationships and
average real terms prices between 1990-2006. Iron ore and hard coking coal expenditure calculated based on
crude steel demand projections, assuming all met by blast furnace production at historic average export prices.
Iron ore
Nickel
Hard coking coal
Economic and demographic development
will generate commodity demand growth
10

The analysis shows that per capita consumption of aluminium, copper and steel making raw
materials tend to pick up at a relatively early stage of industrial development. For aluminium
the empirical analysis also suggests that demand in developed countries has not yet
achieved saturation implying scope for more broadly based future per capita global
consumption growth. For commodities such as iron ore and coking coal it is important to
recognise that the S curves are for total consumption and not for consumption of seaborne
material. So, for countries in which domestic production of raw materials is constrained,
growth prospects for imports of seaborne materials may be greater than for demand in
aggregate.

The chart shows that the largest proportion of the world’s population has low average per
capita rates of commodity consumption consistent with relatively low incomes. As per capita
incomes grow larger numbers of people will consume increasing commodity volumes and
aggregate global consumption of commodities can be expected to rise at a fast rate. For
example, based on the S-curve analysis and assuming a plausible positive scenario for
global growth over the next 2 to 3 decades, the seaborne iron ore market could triple in size
relative to today’s level.

Of course the analysis does not suggest that demand will increase at a steady rate over
time. Macroeconomic cyclicality along with stocking and destocking patterns can be
expected to play major roles in determining demand outcomes for different commodities in
any given year. For example, in China’s case with virtually all sectors of the economy
growing rapidly, there is a chance that developments in some parts of the economy could
move out of phase with developments in other parts. This suggests a possibility that at times
economic imbalances could emerge, generating cycles of strong growth and capacity
building followed by periods of slower growth, catch up and consolidation. The implication is
that Chinese growth and associated commodity demand growth can be expected to have a
cyclical pattern over time.

Supplies are stretched and capacity expansions have been delayed
The supply side of the mining industry continues to face challenges in its response to fast
demand growth. These challenges have been exacerbated by a prolonged period of
underinvestment throughout the sector due to slow demand and low prices during the 1990’s
and into the early 2000’s. As a result, the industry entered the current cycle with reduced
capabilities at all stages of project development, from exploration through to construction
and operations. In this setting, the industry has become increasingly stretched in its attempt
to meet stronger demand and in many cases there is little slack in the system to compensate
for disruptions such as those related to events of nature and downtime for maintenance.
There are few signs that constraints faced by the supply side are easing in the short term –
as indicated for example by shipping freight rates which have reached new record levels in
recent months. Operating and capital costs have continued to rise in 2007 and supply has
once again underperformed significantly especially in the copper market.

Some of the supply challenges are medium term in nature. These are typically related to:
increased demand for materials and equipment, leading to higher input costs and longer
lead times; and bottlenecks in supporting infrastructure (ports, rail and shipping).
11
November 16, 2007 Presentation title 11
Current
delivery time
Acute shortages constrain the
supply side
Normal
delivery time
Grinding mills
Reclaimers
Power generators
Tyres
Barges
Year 1
Wagons
Locomotives
Draglines
Large haul trucks
Rope Shovels
Crushers
Year 2 Year 3 Year 4
Source: Rio Tinto estimates

Other constraints will take much longer to address. First, the industry is moving increasingly
toward the development of resources that in the past were either considered to be too
complex or low-grade or in regions with high country risks and poor infrastructure. This is
presenting challenges to mining companies not only because new skills and technologies
are required to develop those resources, but also because of the increased capital intensity
and delivery risks associated with many such projects. Second, the industry and its
suppliers and contractors are facing acute shortages in the labour market – especially in
relation to skilled professionals such as experienced mining engineers with project
management experience – leading to increased project costs and delays.

Importantly the mining industry is not alone in the struggle to access material and human
resources. The upstream and downstream oil sector as well as the chemical industry have
also stepped up their capital spending plans in recent years and are competing for similar
parts and equipment, construction workers and the services of EPCM contractors. The
competitive environment for such inputs has led to capital cost escalation for mining projects
and longer lead times to deliver new capacity. As a result, the commodity prices required to
induce development of future resources are likely to face upward pressure.

Natural resources constraints facing the mining industry extend to its access to supporting
resources such as energy and water. In this context, environmental considerations in the
development of new projects present a key set of long-term resource related challenges.

Additionally, it is becoming apparent that regulatory approvals are becoming an increasingly
significant barrier to rapid supply expansion. Such constraints are likely to persist and
perhaps become more significant over the longer term.

Commodity case studies

We present case studies relating to markets for aluminium, copper and iron ore – three
commodities that are expected to be drivers of the industrialisation and urbanisation process
in developing countries.
Ship loaders
12
Iron ore
Current pressures in the iron ore market are intense as reflected by spot prices, which have
increased sharply over the last several months. Spot Indian ores are currently selling in
China at now at around $190/tonne, double their price at the start of the year. After taking
into account current freight rates, Australian fine ores sold at benchmark prices (of around
$50/tonne) trade at a substantial discount to these spot prices. Brazilian ores, which have
significantly higher transportation costs to the growing Asian market, sell at a lower but still
significant discount10.
40
60
80
100
120
140
160
180
200
Jan 07 Mar 07 May 07 Jul 07 Sep 07 Nov 07
Australia CFR China
Brazil CFR China
Indian exports
Domestic concentrate
Iron ore prices delivered into China (US$/t, dmt)
Source: CCCMC and Rio Tinto analysis (Indian exports), Mysteel, CUSTEEL and Rio Tinto analysis (domestic
concentrate), benchmark prices and Clarkson spot freight rates (Australia and Brazil)
On the demand side, steel production has grown rapidly leading to strong growth in iron ore
trade. Chinese crude steel production has continued to rise by over 18 per cent y-o-y
despite the levying of export taxes11. While these taxes and the weaker US economy have
discouraged exports, this has been more than offset by renewed strength in domestic
demand. Evidence of this is suggested by domestic Chinese steel prices which reached a
new high in mid-October.

Strong demand for iron ore is not limited to China, however. Annual Japanese crude steel
output this year is expected to hit a new record level for the first time in 33 years and the
German Steel Federation has recently raised its forecast of domestic steel production for this
year. While North American steel producers have cut back output this has little impact on
the seaborne iron ore market as nearly all domestic production relies on either domestic ores
or scrap.

Increases in demand for iron ore have continued to outpace the ability of low cost producers
to add additional supply. Over the first three quarters of the year Chinese iron ore imports
rose by 15 per cent12 – less than the rise in steel production This means that in high-grade
equivalent terms the amount of Chinese iron ore production required to meet domestic
demand is expected to be around 350 million tonnes this year. Much of this is produced at a
relatively high cost. Chinese costs (in US dollar terms) have also been affected by a
strengthening RMB and this pressure is expected to persist while the RMB remains
undervalued. At the same time, as well as having to mine lower grade ores, there is an
increasing reliance on new more remote and therefore more expensive supply from the far
north eastern parts of China. Costs of Indian ores have also increased due to new taxes and
a progressively strengthening rupee. There has only been limited progress on the major
infrastructure investments required in India to make its exports more competitive and at the
same time exports compete against strongly rising domestic demand for ores. Most
importantly the escalation in freight rates has substantially increased the cost of landing ore
in Asian markets from all destinations. The overall implication is that current high prices
have, in all probability, been supported by a rising and steepening industry marginal cost
structure.

2007 Iron ore cost curve delivered
at spot freight rates

20
40
60
80
100
120
140
160
180
200
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100
USD/tonne CFR
Mtpa
Cumulative production (million tonnes)
Source: CRU (mining operating costs), Tex report (spot freight rates), Rio Tinto estimates
Over the longer run, continued strong growth in demand for steel in developing countries
and developed parts of the Middle East is expected to result in substantial growth in
seaborne iron ore trade over the next two decades. Given the large volumes of high cost
production currently in operation and expectations for continued demand growth, any
reversion of prices to lower long run levels can be expected to take place over an extended
period. Additionally it is expected that a substantial amount of high cost production from
China and India will continue to be required to meet growing demand over the long term.

This strongly suggests the possibility of higher long run prices and margins for the traditional
lower cost producers.

Aluminium

Spot aluminium prices have moderated by about 10 per cent since the middle of 2007 and
are currently moving in the range of $2450/t-$2550/t13. Prices at around these levels are
supported by production costs at the highest cost smelters. Forward prices have increased
in relation to spot prices reflecting a market expectation that production with high marginal
costs could be required to meet demand for primary metal over the medium term.

Aluminium has experienced the fastest consumption growth of all non-ferrous metals over
the past five years and it is forecast to continue to enjoy one of the most rapid growth
profiles over the next two of decades. CRU projects consumption to grow by more than 140
per cent over the period to 203014. One reason for the recent growth is that China’s
economic development is highly aluminium intensive. But at the same time there have been
worldwide gains in intensity of use and favorable substitution across a wide range of
applications.

The strong and sustained growth in aluminium demand is starting to stretch the resource
base that has been the foundation of the development of the aluminium industry – largescale
good-quality bauxite deposits and competitively priced stranded energy.

In the case of bauxite the escalation in demand for aluminium is being met increasingly from
high cost and low-scale bauxite deposits in China and opportunistic mining operations in
Indonesia. Such sources of supply are unlikely to provide a long term solution for the
industry’s rapidly growing bauxite needs and have already led to stronger prices for traded
ore. This implies that significant investment in new large scale bauxite mines are likely to be
required if the industry is to meet demand projections.
0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
2000
2001
2002
2003
2004
2005
2006
2007
2017
2000 forecast:
+3.7% p.a.
actual outcome:
+6.0% p.a.
average of CRU and
Brook Hunt forecasts:
+5.5% p.a.
Bauxite potential bottleneck to meet demand growth
Source: CRU, Brook Hunt, Rio Tinto.
Aluminium demand
million tonnes
= 30 million
tonnes
bauxite gap
met by
opportunistic
and fragmented
Indonesian
(+14mt) and
Chinese (+16mt)
bauxites
= need for
additional
130 million
tonnes of
bauxite
8 new Weipa
size mines
This is a challenge, especially if Indonesian supplies not sustainable.

Meanwhile, high energy prices combined with a greater integration between regional energy
markets through the development of LNG and gas-to-liquids projects could increase the
costs of power available to greenfield smelters around the world. Together with a likely
growing trend towards the introduction of pricing mechanisms or tax regimes for carbon
emissions, sustainable stranded hydropower sources have become more valuable. This in
turn may increase the value of existing aluminium capacity linked to such power sources.

In the context of growing demand and constrained supply, the long run pricing environment
for the industry will be heavily influenced by the evolution of costs. Turning first to alumina,
refineries relying on imported bauxite supplies, such as in Europe and the US Gulf coast,
have traditionally occupied the top-end of the alumina cost curve. High energy prices and
rising delivered bauxite costs have increased the competitive disadvantage of these
refineries over the past five years. The Chinese industry is currently adding significant non15
integrated alumina capacity drawing on its capital cost advantage. These refineries are
rapidly joining US and European alumina refineries toward the top of the cost curve. The
resulting increase in the marginal cost of production means that alumina prices are unlikely
to revert to the lower levels implied by long run historical trends even if some higher cost
integrated capacity is eventually replaced by lower cost production.

As with alumina, in the case of aluminium new Chinese smelters are fundamentally changing
the shape of the industry cost curve. The rapid increase in Chinese smelting capacity since
the start of this decade reflects the moderate barriers to entry in building smelters in China
due to low capital costs and short build times. However, this new capacity has come in at the
top-end of the operating cost curve mainly reflecting relatively high power costs.
Consequently, the industry aluminium cost curve has shifted up since 2003 and become
steeper. This has provided a new significantly higher base for prices.

Aluminium costs under upward pressure
Global aluminium costs
750
950
1150
1350
1550
1750
1950
2150
2350
2550
0% 25% 50% 75% 100%
2006 curve 2003 curve
2007 $/t
% production
Source: CRU for 2003 and 2006 cost estimates, Rio Tinto estimates of 10% Rmb appreciation impact.

Impact of further 10% appreciation of Chinese Rmb from today’s levels

A key point to note is that the gradual appreciation of the Chinese currency should also
translate into higher US dollar production costs for Chinese smelters – all other things being
equal. To illustrate this point, the effect of a further 10 per cent appreciation of the Chinese
RMB on the aluminium cost curve is shown on the chart above. With Chinese smelters
predominantly in the third and fourth quartiles, the top end of the curve would shift up in such
a scenario creating an even higher basis for aluminium prices and higher margins for
smelters in the lower cost quartiles.

Copper
Copper stocks have been at critically low levels since a surge in consumption in 2004
depleted available inventories. From that point, stocks have been constrained by supply’s
inability to match a stronger underlying demand growth trend related mainly to Chinese
growth. Reflecting the tight market situation, copper prices are currently moving in the range
of $6400/t-$7000/t15 or about three times higher than their average level through the 1990s
and well above levels achieved in the early part of this decade.

Unlike for iron ore and aluminium, the scope for opportunistic and high-cost sources of
supplies to help bridge supply shortages has been limited for copper. Current prices are
therefore significantly above marginal costs of supply. Short term copper prices are instead
supported by the need to induce those with the least ‘willingness to pay’ for copper to reduce
their consumption. In this context most of the switch away from copper has so far occurred in
the plumbing sector to the advantage of plastics. This means that prices could remain near
current levels as long as production growth continues to under-perform against the
underlying demand trend creating a need to ration supplies.

In terms of demand, most analysts are projecting flat copper consumption outside of China
in 2007. But within China copper consumption growth is projected to grow by 15 per cent this
year. Calculations of apparent demand are pointing to growth well in excess of that number,
although this is thought to reflect the reversal of a destocking phase in China during 2006.
Overall global demand is expected to record its strongest growth since 2004, rising this year
by about 3.5-4.0 per cent16. Looking forward, even with a projection of high underlying
average demand growth in China, demand for material can be expected to fluctuate
unpredictably over periods of months on stocking and destocking cycles generating price
volatility.

On the supply side copper miners have faced many of the challenges and bottlenecks
discussed earlier. Strikes and unforeseen disruptions from weather related events and
accidents have also affected the performance of existing copper mines. It is estimated by
Brook Hunt that actual global mine output over the past three years has underperformed
market expectations by a cumulative 2.5 to 3 million tonnes of copper. This is equivalent to
annual losses of 4 per cent to 6 per cent which compare with losses in normal years closer
to 2 per cent.

The medium term outlook for copper will be highly dependent on whether disruptions
continue to run at high levels. Third quarter production reports from copper mining
companies suggest that this remains an ongoing issue. In addition recent reports have
pointed to potential shortages of sulphuric acid which could constrain SxEw operations in the
short term. This source of supply has accounted for a high proportion of primary production
growth over the past two years.

The influence of investment funds activity could also be a factor affecting medium term
prices in the copper market. In particular, some analysts are suggesting that additional
demand associated with long only funds means that stocks levels associated with a market
in equilibrium will need to be higher than in the past. In any case, the likely continued
importance of investment funds in exchange traded commodity markets means that large
price movements could take place on the back of commodity specific speculative shifts or
broader shifts in investor sentiment – well in advance of any fundamental change in physical
markets.

Looking to the long run, CRU projects that copper demand will more than double over the
next 25 years. Growth prospects are based on the expected resource intensive development
of economies such as China and the associated investment in power distribution networks
and other infrastructure. Additionally, in a high-energy price and carbon conscious world
copper can be expected to benefit from any global drive for increased energy efficiencies
and any shift towards the development of local distribution networks around sources of
renewable energy.

Ultimately, the industry should be able to surmount bottlenecks in equipment and supplies.
However, some of the supply challenges are likely to be of a longer-term nature. These
include declining ore grades, an increasing shift towards underground operations and the
need for the industry to access and develop deposits in countries with higher risk profiles
such as the DRC. Meanwhile upward pressure on capital costs for copper projects is likely
to remain. This suggests that future long run prices and margins may be sustainable at
levels well above long run historical averages without encouraging excess capacity.
Historically copper prices have tended to trend towards the industry’s marginal cash cost of
production. But reflecting the cointegrated nature of costs and prices, cash costs have
continued to move up driven by higher labour rates and bonuses, increased royalty
payments, and stronger prices for supplies, services and energy. Exchange rate changes
have also been significant in key copper mining regions, exacerbating the upward pressure
on cost. Overall, cost increases have been felt more strongly at the margin and we estimate
9th decile costs to be back near or even above levels last seen at the start of the previous
decade. While some of the recent cost pressures are likely to subside in the longer term, we
believe that, as for many other commodities, a structural shift in the copper cost curve has
occurred supporting an expectation of significantly higher long run prices than would be
implied by historical trends.

Conclusion – faster long run average demand growth, extended medium term price
elevation and higher long run prices

Continued firm global economic activity led by rapid resource intensive growth in China is
expected to support strong increases in demand for most metals and minerals over 2008
and 2009. At the same time with low stocks and a likely continuation of supply side
difficulties, most commodity prices can be expected to average well above their long run
trend over this period. It is too early to suggest that the price cycle has peaked across the
range of commodities.

While the central case is positive, it is important to remain mindful of macro-economic risks
especially relating to US housing and credit markets. However it is also important not to
exaggerate these risks as our modelling suggests that they may not have a significant
impact on the developing economies that have been the growth engines of commodity
demand. It is also important to keep in mind that price movements from month to month will
be influenced by stocking and destocking and investment funds’ activities that may be only
indirectly related to economic growth.

Over the longer run strong resource intensive growth from China, India and other developing
countries should continue to provide momentum to commodity demand. Indeed, recent
history and the IMF’s projections for future growth would suggest that we are currently going
through a period of global growth not seen since the period of fast growth and reconstruction
in OECD economies following the Second World War. The implications for commodity
markets are profound.

In time production can be expected to expand to meet faster growth in demand at more
sustainable prices. However, it is expected that prices of many minerals and metals will
remain elevated above trend for longer than has been the case in the past because of
constraints on the speed with which production capacity can be expanded over the next few
years. Also most prices are likely to assume higher levels than has been the case historically
due to structural increases in industry marginal costs.

Sources:
1: IMF World Economic Outlook, October 2007
2: Global Insight, Interim China forecasts, November 2007
3: Global Insight, Interim USA forecasts, November 2007
18
4: Cabinet Office, Government of Japan for historical quarterly GDP growth estimates,
Development of real GDP, November 2007
5: Global Insight, Interim Japan forecasts, November 2007
6: Global Insight, Interim India forecasts, November 2007
7: Data downloaded from Ecowin database, November 2007
8: Lehman Brothers, India: Everything to play for, October 2007; Goldman Sachs, India’s
Rising Growth Potential, January 2007
9: References to steel from IISI, World Steel in Figures, September 2007; for references to
aluminium CRU The Long Term Outlook for Aluminium, 2007 Edition; for references to
copper CRU Copper Quarterly, October 2007
10: CCCMC and Rio Tinto analysis (Indian exports), Mysteel, CUSTEEL and Rio Tinto
analysis (domestic concentrate), benchmark prices and Clarksonspot freight rates (Australia
and Brazil).
11: IISI Media Release, November 2007
12: CRU Steelmaking raw materials Monitor, November 2007
13: LME data downloaded from Ecowin database, November 2007
14: CRU The Long Term Outlook for Aluminium, 2007 Edition
15: LME data downloaded from Ecowin database, November 2007
16: CRU Copper Quarterly, October 2007