Transcript Factoring Emerging Markets into the Relationship Between

Steven Landgraf
WPPI Energy and Marquette University
Abdur Chowdhury
Marquette University
USAEE/IAEE North American Conference
Tuesday Oct 11th, 2011
 Commodity
price bubble (2003 – 2008)
• Record high oil and natural gas prices
• Ultra low interest rates, 2003-2004
• Accelerated EM economic growth
• “Financialization” of commodity markets
Q2 1995
Q4 1995
Q2 1996
Q4 1996
Q2 1997
Q4 1997
Q2 1998
Q4 1998
Q2 1999
Q4 1999
Q2 2000
Q4 2000
Q2 2001
Q4 2001
Q2 2002
Q4 2002
Q2 2003
Q4 2003
Q2 2004
Q4 2004
Q2 2005
Q4 2005
Q2 2006
Q4 2006
Q2 2007
Q4 2007
Q2 2008
Q4 2008
Q2 2009
Q4 2009
Q2 2010
S&PGSCI Index
(January 1 1970 = 100)
800
700
600
500
400
300
200
100
 Commodity
price recovery post financial
crisis (2009- )
• Near-zero interest rate policies in advanced
countries
• Massive injections of liquidity during the
financial crisis (Quantitative Easing).
 QE2 (late 2010 to mid 2011)
•
Backlash in the media against QE2
– Coincided with run-ups in prices of oil, gold,
food, etc.
– Roubini: “Wall of liquidity” chasing assets in EMs
•
Continued strong performance of EMs
after 2008 fueled commodity demand
•
Not much “global” research incorporates
BRIC influence
•
Frankel (1986, 2008)
– Overshooting model
– Monetary variables and commodity prices related
•
Sousa and Zaghini (2004, 2006)
– Global monetary shocks have long-run impacts on
domestic prices
•
Rüffer and Stracca (2006)
– “Excess liquidity” impacts prices in advanced
countries
•
Belke et. al. (2010)
– Expansionary shocks increase relative prices
 Does
excess liquidity positively impact
commodity prices?
 Which
effect is more prominent:
• Demand channel?
• Excess liquidity?
 Do
the results change if emerging market
data is included in the global aggregate?
 ADV
– aggregates 10 advanced
economies and the euro zone economies
 ALL
– aggregates the BRIC countries in
addition to the countries in ADV
 Output
(demand channel) impacts
commodity prices – robust result
 Excess
liquidity, interest rate – mixed
results
• Interest rate – little influence
• Shocks to excess liquidity more prominent than
shocks to output in ALL – opposite of ADV
•
•
•
•
Vector Error-Correction
Granger causality
Impulse Response Function (IRF)
Variance Decomposition (VDC)
GDP
Sum of GDP (“demand channel”)
MON
Sum of broad money supply divided by GDP sum
(“excess liquidity”)
INT
GDP-weighted average of S-T (3 mo) interest rate
CPI
COM
GDP-weighted average of headline CPI
Commodity price index
Q2 1995
Q4 1995
Q2 1996
Q4 1996
Q2 1997
Q4 1997
Q2 1998
Q4 1998
Q2 1999
Q4 1999
Q2 2000
Q4 2000
Q2 2001
Q4 2001
Q2 2002
Q4 2002
Q2 2003
Q4 2003
Q2 2004
Q4 2004
Q2 2005
Q4 2005
Q2 2006
Q4 2006
Q2 2007
Q4 2007
Q2 2008
Q4 2008
Q2 2009
Q4 2009
Q2 2010
Global Excess Liquidity
1.05
1
0.95
0.9
0.85
0.8
0.75
0.7
ADV
ALL

Log-difference (except interest rate)

Sample: 1995Q2 to 2010Q3

Sourced mostly from IMF
• Supplemented with World Bank data for some BRIC
countries
• Aggregation methodology follows Sousa and Zaghini
(2004)
 PPP exchange rates

Commodity Index: S&P GSCI
• 66% weighted with energy commodities
 Lag
length (Info criterion and LM test)
• ADV: 2
• ALL: 3
 Unit
root tests: stationary in 1st diff.
 Cointegration: tests
suggest its presence
• Appropriate to use a VEC vs. a VAR
• Long-run equilibrium exists between variables
 Demand
channel (GDP) robustly impacts
commodity prices whether or not BRICs are
included.
• Granger, IRF, VDCs support
• Structural relationship between output and
commodity prices
• Prices also respond positively to positive shocks
 Neither
interest rates nor excess liquidity
Granger cause commodity prices
 Positive
shocks (1 std. dev.)
• ADV: increase commodity prices 2 quarters out
• ALL: increase commodity prices 6 and 7 quarters
out
 VDCs
show discrepancies between ADV
and ALL
 Excluding
BRIC data overestimates
impact of demand channel and
underestimates excess liquidity
Shock to GDP
Quarter ADV
ALL
1
12%
23%
2
30%
32%
3
27%
27%
4
34%
24%
5
36%
24%
6
34%
21%
7
37%
17%
8
37%
18%
9
36%
16%
10
37%
17%
Shock to MON
Quarter
ADV
ALL
1
1%
1%
2
9%
7%
3
8%
9%
4
12%
25%
5
11%
25%
6
11%
34%
7
11%
42%
8
11%
42%
9
11%
46%
10
11%
46%
Percent Variance of COM due to GDP
e to RGDP
7
40
35
30
ADV:
25
20
15
10
5
0
8
9
1
10
3
4
5
6
7
8
9
10
Percent
PercentVariance
Variance of
of COM
COM due
due to
to GDP
CPI
e totoRGDP
ue
CPI
77
2
35
60
30
50
25
40
20
ALL:
30
15
20
10
5
10
00
88
99
10
10
11
2
2
33
44
55
66
77
88
99
10
10
Percent
Variance
COM
due
CPI
Percent
Variance
of of
COM
due
toto
MON
eoto
CPI
MON
ADV:
45
14
40
12
35
10
30
8
25
20
6
15
4
10
25
00
88
99
11
1010
to
to MON
INT
99
33
44
55
66
77
88
99
10
10
Percent
PercentVariance
Varianceof
ofCOM
COMdue
dueto
toMON
INT
950
45
8
40
7
35
6
30
525
420
315
210
15
00
ALL:
88
22
1010
11
22
33
44
55
66
77
88
99
10
10
 BRIC
country economies impact
commodity prices in a way not captured
by using advanced country data
 Global
liquidity shocks have a great
impact on prices when country sample is
expanded beyond advanced countries
 Suggests
a diminished role of advanced
countries in impacting prices
 Central
Banks should continue to closely
monitor emerging market monetary
policy when considering effects on
commodity and energy markets
 Research
coming from a global
standpoint should not exclude emerging
markets from analyses
• Subject to data availability
 Interest
rates not shown to have a
measurable impact on commodity prices
• Contrasts with the literature
• Average a good measure? Some use LIBOR
 Relatively
low degrees of freedom
 Data quality
 Evidence of monetary impacts not
overwhelming