The behavior of base m etals prices So Paulo Novem ber 5 , 2 0 0 9 - - PowerPoint PPT Presentation

1

The behavior of base m etals prices

São Paulo Novem ber 5 , 2 0 0 9

2

Agenda

• Unique features
• The price behavior
• Facts and fantasies
• Minerals and metals for the long run

3

Unique features

4

Unique features

• Commodity investing
• Commodities and the weather
• Commodities and inflation
• Price elasticities
• Cyclical drivers

5

The price behavior

6

Price performance

• Key features
• Asymmetric volatility
• Price co-movement

7

Fat tails: large numbers of extreme values

129.30 Nickel 17.50 Freight 8.94 15.01 Kurtosis1 Copper Aluminum

1 - The kurtosis coefficient measures the magnitude of the extreme values of the distributions. If returns are normally distributed, then the kurtosis should be three.

8

1,000 2,000 3,000 4,000 5,000 6,000 7,000 1 9 8 8 1 9 8 9 1 9 9 1 9 9 1 1 9 9 2 1 9 9 3 1 9 9 4 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9

US$/ m etric ton

• 48%
• 42%
• 36%
• 30%
• 24%
• 18%
• 12%
• 6%

0% 6% 12%

Daily returns Source: Vale and LME

Alum inum prices

Unpredictability and volatility clustering

9

2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 18,000 20,000 19 88 19 89 19 90 19 91 19 92 19 93 19 94 19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09

US$/ m etric ton

• 40%
• 30%
• 20%
• 10%

0% 10% 20%

Daily returns Source: Vale and LME

Copper prices

Unpredictability and volatility clustering

10

15,000 30,000 45,000 60,000 75,000 90,000 1 9 8 8 1 9 8 9 1 9 9 1 9 9 1 1 9 9 2 1 9 9 3 1 9 9 4 1 9 9 5 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9

US$/ m etric ton

• 50%
• 40%
• 30%
• 20%
• 10%

0% 10% 20%

Daily returns Source: Vale and LME

Nickel prices

Unpredictability and volatility clustering

11

Brief summary of econometric tests

• Autocorrelation in returns are not significant

 unpredictability

• Autocorrelation in square returns are significant

 volatility clustering

12

S&P 5 0 0 x VI X

600 700 800 900 1,000 1,100 1,200 1,300 1,400 1,500 1,600 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 10 20 30 40 50 60 70 80 90 S& P 500 VI X

Source: Bloomberg

Stock prices are inversely related to volatility

13

1.90% 2.08% Copper 2.40% 1.64% 1.73% Dow n cycle 2.97% 2.06% 0.86% Up cycle Nickel Alum inum S&P 5 0 0 Volatility asym m etry¹ 1987 - 2009

¹ One month moving average of the standard deviation of daily returns Source: Vale

Volatility asymmetry: base metals prices are positively related to volatility

14 Source: Vale

Co-m ovem ent betw een returns: sim ple correlation Jan/ 04 - Oct/ 09

Prices co-movement in the short-term influenced by common factors

0.57 Soybeans 0.53 0.76 Corn 0.21 0.31 0.29 Oil 0.24 0.36 0.24 0.50 Thermal Coal 0.32 0.20 0.18 0.29 0.19 Iron ore 0.20 0.29 0.29 0.56 0.45 0.26 Aluminum 0.27 0.27 0.23 0.63 0.37 0.28 0.72 Copper 0.15 0.22 0.22 0.34 0.18 0.33 0.53 0.54 Nickel Wheat Soybeans Corn Oil Thermal coal Iron

• re

Aluminum Copper

15 1- We can say that the degree of the co-movement is low because there is only one co-integration vector. To find a strong co-movement between n variables it would be necessary to have n-1 co-integration vectors. Source: Vale

Price co-movement in the long term: specific market fundamentals prevail

Co-integration analysis March 1 9 8 7 – October 2 0 0 9 X X X X Low degree of co-m ovem ent 1 High degree of co-movement Metals prices Food and metals prices Energy and metals prices All commodities

16

The relationship between iron ore and maritime freight prices

• There is a high correlation betw een

m aritim e freight and iron ore prices.

• They also behave in a sim ilar m anner over

the long term .

• There is a com m on trend w hich influences

the behavior of both prices in the sam e direction.

17

The relationship between iron ore and maritime freight prices

Analysis of com m on trend influences 87.9% 94.6% 1 0-year 85.8% 93.1% 5-year 69.8% 93.5% 2-year 33.7% 91.7% 1-year % of iron ore price variations explained by com m on trend % of freight price variations explained by com m on trend Horizon

18

• 0 .8
• 0 .6
• 0 .4
• 0 .2

0 .0 0 .2 0 .4 0 .6 0 .8 1 .0 1 9 8 3 1 9 8 5 1 9 8 7 1 9 8 9 1 9 9 1 1 9 9 3 1 9 9 5 1 9 9 7 1 9 9 9 2 0 0 1 2 0 0 3 2 0 0 5 2 0 0 7 2 0 0 9

Base metals and oil Base metals and food

1- Measured by 1-year correlation of price returns Source: Vale

Co-m ovem ent¹

Price co-movement is likely to increase during financial crises

Base metals  LMEX Food  CRB food

19

Facts and fantasies

20

Facts and fantasies about minerals and metals prices

• Mean reversal
• Financial investments
• The Chinese iron ore stockpiling

21

20 40 60 80 100 120 140 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000

¹ Nominal prices adjusted by the September 2009 US PPI Source: Vale

Real iron ore prices¹ US$ cents/ fe-dmt

Iron ore prices are not doomed to revert to the mean

22

Base metals prices are not doomed to revert to the mean

1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 5,500 6,000 1987 1992 1997 2002 2007 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 1986 1991 1996 2001 2006 10,000 20,000 30,000 40,000 50,000 60,000 1987 1992 1997 2002 2007

Alum inum price¹ US$/ ton

¹ Monthly nominal prices adjusted by the September 2009 US PPI Sources: Vale and LME

Copper price¹ US$/ ton Nickel price¹ US$/ ton

23

Unit root test: Augm ented Dickey-Fuller 1 9 8 7-2 0 0 8

Brief summary of econometric tests

0.14 0.93 0.66 0.28 Augmented Dickey- Fuller test P-value Alum inum Copper Nickel I ron ore Prices1 Null hypothesis: price has a unit root

¹ Annual real prices adjusted by US PPI Sources: Vale

• Unit root tests do not support the mean reversal

hypothesis

24

However, metals price volatility reverts to the mean

Nickel price volatility¹

¹ Standard deviation of 20-trading day moving average of daily returns. Sources: Vale and LME

0 % 2 % 4 % 6 % 8 % 10 % 12 % 14 % 16 % 18 % 198 7 1 989 19 91 1993 1 995 19 97 1999 2 001 20 03 2005 2 007 200 9

25

Metal prices are determined by fundamentals

• 40
• 30
• 20
• 10

10 20 30 40 50 60 1995 1997 1998 2000 2002 2004 2006 2008 US$/ ton 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000 000 num ber of contracts

Copper: prices and long positions

Source: Vale, NYMEX and CFTC

Net long positions Spot price

26

Metals prices are determined by fundamentals

0.00064 9.90226 380 Spot price does not Granger causes long positions 0.66215 0.41271 380 Long positions does not Granger causes spot price Probability F-Statistic Obs Null hypothesis Granger causality test March 1995 – October 2009

• Financial investments are caused by prices, it does

not cause prices.

27

20 40 60 80 100 120 140 Sep-08 Nov-08 Jan-09 Mar-09 May-09 Jul-09 Aug-09 Oct-09

Mt

5 10 15 20 25 30 35 40

Days

Days of iron ore consum ption

There is no evidence of stockpiling

Stocks at the ports

20 40 60 80 100 120 140 Sep-08 Nov-08 Jan-09 Mar-09 May-09 Jul-09 Aug-09 Oct-09

Mt

10 20 30 40 50 60

Days

Days of iron ore im ports

Stocks at the ports

28

Minerals and m etals for the long run

29

Is there a bubble?

• 60.8

72 184 Thermal coal

• 61.5

494 1,283 Wheat4

• 40.7

978 1,649 Soybeans4

• 47.0

77 145 Oil³

• 41.1

1,326 2,251 Platinum²

• 65.9

18,465 54,200 Nickel

• 26.8

6,575 8,985 Copper

• 55.6

1,903 4,290 Aluminum

• 55.1

92 205 Iron ore Δ% Current price¹ All-time high

¹ October 30, 2009 ² US$ per oz ³ US$ per barrel

4 US$ per bushel

US$/ ton

30

• 46%
• 56%
• 65%
• 26%
• 30%
• 32%

38% 102% 80% 5% 5% 13% Current recession Average of last 5 recessions Current recovery² Average of last 5 recoveries

Aluminum Copper Nickel

Commodity prices in global recessions and recoveries¹

¹ Recessions are measured as the percentage change of metal prices from the beginning of each recession to trough; recoveries are measured as the percentage change of metal prices from the trough to the end of each recession. Global recessions: 1974, 1981-82, 1990-91, 1998 and 2001. ² Percentage change from last trough until September 2009. Source: Vale and IMF

This time metal prices were much more volatile than in the recessions of the last 40 years

31

Global industrial production grow th % 3mma, saar¹

• 40
• 30
• 20
• 10

10 20 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 2006 2009

¹ 3-month moving average, seasonally adjusted annualized rate Source: Vale and JP Morgan

The sharp drop of metal prices was caused by the steep decline of manufacturing output …

32

• 1 0
• 8
• 6
• 4
• 2

2 4 6 8 Jan- 0 8 May- 0 8 Sep- 0 8 Jan- 0 9 May- 0 9 Em erging economies Developed economies

… while the strong recovery from the recessions was determined by the fast EM-led growth of IP

¹ Seasonally adjusted Source: Vale and JP Morgan

I ndustrial production % 3mma, sa¹

33

31.7 34.5 30.3 9.7 59.2 61.2 46.0 53.8

Aluminum Copper Nickel I ron ore¹ 1993 2008

Share in global consumption %

¹ Seaborne trade Chinese market share Source: Vale and WBMS

Emerging economies have become the dominant consumers of commodities

34 Sources: Vale, Tex Report and WBMS

Developed economies Emerging market economies China 110% 161% 97%

• 10%
• 61%

3%

Alum inum Copper Nickel

74% 106% 106%

Steel

3.5% 96.5% 62.2%

5% 95%

China Rest of the world

Contribution to global consumption growth 2000 -2008 Contribution to growth in iron ore seaborne trade 2000 -2008

Emerging economies have been the drivers of global consumption growth of minerals and metals

35

• 4
• 3
• 2
• 1

1 2 3 4 5 6 7 8 9 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 2010E 2014E

Developed econom ies Em erging econom ies

Source: Vale and IMF

Real GDP grow th % annual Decoupling

Unless economic policies deteriorate, we expect convergence to go on, thus underpinning a long minerals and metals cycle

Decoupling

36

www.vale.com

rio@vale.com

Vale: a global leader