Transcript Zhirui Zhu 2012

Identifying Supply and Demand Elasticities of Iron Ore
Author: Zhirui Zhu, Faculty Advisor: Professor Gale A Boyd
ABSTRACT
This paper utilizes instrumental variables and joint estimation to
construct efficiently identified estimates of supply and demand
equations for the world iron ore market under the assumption of perfect
competition
With annual data spanning 1960-2010, I found an upward sloping
supply curve and a downward sloping demand curve. Both of the
supply and demand curves are efficiently identified using a 3SLS
model. The instruments chosen are strong and credible. Point
estimation of the long-run price elasticities of supply and demand are
0.45 and -0.24 respectively, indicating inelastic supply and demand
market dynamics
Back-tests and forecasts were done with Monte Carlo simulations. The
results indicate that 1) the predicted prices are consistent with the
historical prices, 2) world GDP growth rate is the determining factor in
the forecasting of iron ore prices
METHODS & DATA
Supply and Demand Structural Equations
Demand: ln (QtD) = β0 + β1 ln(Pt) + β2 ln (Pt-1) +β3
ln (Scrap Pricet) + β4 Demand Shockt + β5 ln (GDPt)
+μ
Supply: ln (QtS ) = γ0 + γ1 ln (Pt) + γ2 ln (Pt-1) + γ3 ln
(Interest Ratet) + γ4 Timet + γ5 Supply Shockt + ν
Market Clearing: QtD = QtS
Instruments Selection
Fig 3 World iron ore price and production, U.S. scrap price, world GDP, China iron ore
imports and U.S. real interest rate plots (1962 – 20010)
Annual Supply and Demand using Iron Ore
Production for Quantity
OLS
2SLS
3SLS
Wd Ore Pd
Wd Ore Pd
Wd Ore Pd
Demand
Function
Ore Price (t)
-0.0693
-0.463
-0.928***
(-0.61)
(-1.71)
(-4.33)
The current iron ore trade market is dominated by the Big Three - Vale,
Rio Tinto and BHP Billiton. Three companies control about 70% of the
seaborne market in 2010
In 1970’s, European steelmakers determined the market demand with the
Japanese buyers taking over in the 80’s and 90’s. In the 2000’s, Chinese
steelmakers started to play a dominant role in the buyer’s market
Since post-World War II, iron ore prices had been decided behind closed
doors in negotiations between mining companies and steelmakers
RESULTS - BACK TESTING (1962-2010)
Fig 4 Iron ore price and predicted prices, quantities and predicted quantities (19622010)
Effect of instruments on annual iron ore
price
OLS
OLSOLSDemand Supply
Ore
Ore
Ore
Price (t) Price (t) Price (t)
Ore Price (t-1)
0.479*** 0.653*** 0.923***
(5.06)
(6.85)
(8.13)
Demand
Shifters
Scrap Price
0.0612
(1.57)
0.0811
(1.77)
0.0910**
0.464***
0.608***
0.744***
(6.21)
(5.01)
(7.65)
0.435***
0.419***
0.404***
(12.62)
(10.42)
(11.69)
Ore Price (t-1)
0.0917
(0.99)
0.353
(1.86)
0.688***
(4.45)
P-value for all
demand
shifters
Constant
9.039***
(8.54)
0.95
9.894***
(7.56)
0.93
10.76***
(9.77)
0.88
Supply
Shifters
Scrap Price
Demand
Shock05
Wd GDP
R2
Supply
Function
Ore Price (t)
GLOBAL IRON ORE INDUSTRY
In 2004, Chinese steelmakers obtained the right to negotiate the
ironstone benchmark price for the first time. However the benchmark
price was mainly settled between the Big Three and the Japanese
steelmakers. Meanwhile, the global ore import price from 2004 to 2008
had increased 18%, 71%, 20%, 9% and 96% respectively
RESULTS – OLS, 2SLS, 3SLS
(2.66)
0.322**
(3.22)
0.601***
(3.88)
0.900***
(6.84)
Interest Rate
-0.0418*
(-2.02)
-0.0410
(-1.81)
Time
0.0261***
(16.81)
Supply Shock
Wd GDP
Demand
Shock05
0.0572
(1.15)
0.0609
(1.21)
0.206
(1.24)
-0.0504
(-1.30)
0.600*** 0.366***
(6.24)
(3.80)
0.000***
0.0171**
(-2.90)
Time
0.00523
(1.60)
Interest Rate
-0.00538
(-0.32)
-0.00295
(-0.16)
-0.0122
(-0.82)
Supply Shock
0.172*
(2.18)
-0.0959
(-1.45)
0.0247***
(13.77)
0.0231***
(15.11)
P-value for all
supply shifters
-0.175***
(-3.36)
-0.148*
(-2.56)
-0.138**
(-2.99)
Constant
Ore Price (t-1)
0.0680
(0.67)
-0.190
(-1.26)
-0.450***
(-3.48)
Constant
18.76***
(93.55)
0.94
18.69***
(84.84)
0.93
18.48***
(97.84)
0.89
R2
Durbin–
Watson
N
1.04
47
1.54
47
-3.240
(-0.76)
47
2.492
(1.96)
48
0.231
(0.51)
47
N
P-value from
joint test of all
0.00*** 0.00*** 0.00***
coefficients
(Prob > F)
Price 2elasticities of supply and demand
adj. R
0.90
0.82
0.81
Coef.
S.D.
Z
Demand
β1+β2
-0.241**
0.090
-2.68
Supply
γ1+γ2
0.450***
0.045
9.93
1.94
47
0.0034**
RESULTS – FORECASTS (2011-2020)
The iron ore annual pricing system officially ended in 2010 and moved
to a spot market system
Fig 1 Iron Ore Pricing Diagram
Fig 2 Iron Ore Imports by Country
Covariates Sensitivity Tests
Fig 5 Covariates sensitivity tests (1962-2010)
CONCLUSTION
3SLS yields efficient and consistent coefficient estimators
The annual model indicates that the long-run iron ore supply curve
appears to be upward sloping while the long-run demand curve for iron
ore appears to be downward sloping
The simulation results indicate that the annual model captures most of
the historical price fluctuations