Oil-Price Effects on the Real Business Cycle: Evidence from the G-7 Countries

European Journal of Economics, Finance and Administrative Sciences
ISSN 1450-2887 Issue 14 (2008)
© EuroJournals, Inc. 2008
http://www.eurojournals.com

Oil-Price Effects on the Real Business Cycle: Evidence from the
G-7 Countries


Abdullah Al-Salman
Department of Economics, P.O.Box 5486, Kuwait University
Safat 13055, Kuwait
E-mail: abdullas@cba.edu.kw

Khalifa H. Ghali
Department of Economics, P.O.Box 5486, Kuwait University
Safat 13055, Kuwait
E-mail: kghali@cba.edu.kw

Nayef Al-Shammari
Central Bank of Kuwait, P.O.Box 526 Safat, 13006, Kuwait
State of Kuwait
E-mail: nalshammari@CBK.GOV.KW


Abstract
This paper investigates the short run impact of changing oil prices on the business cycle of
the group-of-seven (G-7) countries. The data set is quarterly and covers the period 1970:1 –
2006:4 for most of the countries. We utilize the Granger-causality test to investigate the
causal relationships between the major determinants of the business cycle and their impact
on the short run real GDP in G-7 countries. In particular, we look at the effects of changes
in oil prices on the short run changes in real GDP across the G-7 countries. The empirical
results show that there is a short term neutrality of real GDP to oil-price changes for Italy,
Japan, and the UK. However, the oil effect is found to be obvious for the rest of the G-7
economies, especially Germany and France. On the other hand, changes of government
policies have played a role in mitigating the influence of high oil prices in Japan, Italy, and
France. In addition, the characteristics of the economy of the U.S, the U.K, Germany and
Canada have shaped the role of oil influence on their business cycles. These differences
show that there is a timing effect of changes in oil prices on the business cycle in some of
the G-7 economies.


Keywords: Business Cycle, Oil Price, G-7 Countries
JEL Classification Codes: E32,
E2

I. Introduction
Fluctuations in real business cycles have important implications in the short term. Macroeconomic
performance is empirically influenced by shocks in oil market. Changes in energy prices have become
a crucial component in determining the business cycle in industrialized economies. The difference in
responses of oil price shock may reflect the degree of severance in the oil importing economies at

---

75
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
which oil price hike is transmitted to such activities in the short term (Mussa, 2000). During the first
half of 2008, economists have raised issues concerning the recession in the U.S and the role of current
high oil prices. Shocks in energy prices may lead to drive the fluctuations of business cycle in the G-7
economies.
Historically, disturbances in oil market lead to economic distortions either for industrialized
countries or developing countries. Since 1972, the consequences of high oil prices have been
associated with recession, high inflation, low economic growth and low productivity. Recently,
economists argue that recent increase in oil prices may lead to push the U.S into recession. During the
first oil price crisis in 1974, high oil prices were caused for a slowdown in U.S labor productivity
(Hansen, 2001). It is also argued that high oil prices are associated with high inflation levels in U.S,
Japan and Europe (Leblanc and Chinn, 2004).
Although, crude oil price has experienced limited short run fluctuations between 1930s and
early 1970s period, it has gone through rapid cycles since 1972. Concerns have been raised in
investigating the impact of the short term effect of high oil price on economies of developed countries.
In most of industrialized countries, imported oil is considered to be a crucial input for their economies.
Therefore, any fluctuations in price of this input should lead to important consequences on the overall
economy of these countries. Economic activities are importantly linked to issues related to global
economy especially when the country is characterized by an open economy. Thus oil prices can be
endogenous components for the growth in these economies.
Recent empirical literature on the topic of business cycle relation with energy price shocks has
investigated the links through testing for existence of causality. However, the objective of this study is
to investigate the behavior of crude oil in short run. Particularly, this study examines the relationship
between crude oil price fluctuations and the real business cycle in the G-7 economies controlling over
the determinants of business cycle. It also shows the limiting effect of oil prices on real business cycle
in addition to the causality impacts. It is shown in our study that oil prices shocks differ from one
country to another for the G-7 case, in that it is transitory in some cases (such as Canada and France)
and permanent in others (such as Germany and U.S). It is also shown that the short term neutrality of
oil is found in Italy, Japan, and UK.
We use the techniques of cointegration and causality tests to investigate the causal relationship
among the major determinants of business cycles in the G-7 countries employing quarterly data over
the period 1970:1 to 2006:4. A general finding is that oil prices have a limiting effect on economic
growth of the seven major industrialized countries. However, this limiting effect varies from one
country to another across the G-7 economies. In particular, changes in government policies have
played a role in mitigating the influence of high oil prices in Japan, Italy, and France, while the
characteristics of the economy in the U.S, U.K, Germany and Canada have shaped the role of oil
influence on their business cycles
This paper is organized as follows. Section II contains a brief survey of the relevant literature.
Section III, presents the model and testing procedures. Section IV presents the data. The empirical
results are contained in section V. Section VI concludes.


II. Literature Review
There is a large body of studies that have investigated the link between energy prices and the business
cycle. Early work focused on the adverse business cycle effects of shocks in oil prices. While,
Hamilton (1983) showed that energy prices are countercyclical using a vector autoregression analysis
with a pre-1972 data, Serletis and Kemp (1998) found the energy prices to be procyclical. Many
empirical studies investigated the transitory and the permanent components of macroeconomic in G-7
economies. While studies such as Gregory, Head and Raynauld (1997), and Kose, Otrok, and
Whiteman (2003) test for the common fluctuations in the permanent component in the G-7, studies
such as Yoon (2004) and Kim and Nelson (2001) investigate both transitory and permanent

---

76
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
components of macroeconomic. The study by Yoon (2004) investigates the relationship between oil
prices and common permanent and transitory shocks. In his study, Yoon concluded that shocks in oil
prices are not to be a crucial determinant of common recessions in G-7 economies, with the exceptions
of energy prices shock in 1973 and 1979.
Other studies have investigated the issues related to relationship between oil prices and
recessions across G-7 economies. The issue of U.S recession and the role of oil prices was examined
by Raymond and Rich (1997) using the two-stage Markov Switching Model of Hamilton (1989). They
found that fluctuations in oil prices are not contributed determinant to the mean of low growth phases
of output. In line with this finding, Clements and Krolzing (2004) explored similar results for the U.S
as well. The issue of general price level and the relation with the oil price has been investigated by a
study of Abel and Bernanke (2001). They conclude that high oil prices fuel the general price levels.
Davis and Haltiwanger (2001) view the relationship between the oil price shocks and job creations and
losses. They find that, for industrial sectors, both shocks of oil price and monetary shocks lead to a
result of job losses than job creations. In addition, Keane and Prasad (1996) show the relationship
between the increasing oil prices and real wages. Their results confirm that high oil prices are
associated with raising the relative real wages of skilled labor.
On the other side, Hess (2000) agues that the period prior to 1980s observed a case at which
shocks of oil prices caused lower real GDP in U.S economy, whereas the period since then had not
shown any effect of oil prices shocks on U.S economic activity. He then concludes that outliers of oil
prices experienced short term effect and may not have a direct influence on the U.S economy. Similar
results have been found by Hooker (1996) during the same period. A study by Jimenez, Rodriguez and
Sanchez (2004) reports the negative impact of oil price on economic growth for oil importing
countries, whereas the effect is showed to be mixed for the oil exporting countries.


III. Methodology and Model Specification
III.1. Model Specification
We model the relationship between the GDP trends and its determinants at the short run using
macroeconomic variables as well as monetary variables. Sims (1980) applies a simple closed model
using factors of real output, money supply, and prices. Sims’s model was modified by Ghali (1998) by
adding exchange rate variable. Our model extends the work by both Sims (1980) and Ghali (1998)
through the inclusion of energy price factor.
The idea of including the oil prices matches with the uniqueness of G-7 economies as any
fluctuation in oil prices may affect the performance of their business cycles. Just recently with
booming oil prices, high oil prices are blamed for influencing the business cycles of these economies.
The determinants of the business cycle for the G7-economies can be indicated through several factors.
First of all, the macroeconomic variables such as the employment rate, inflation rate, and gross fixed
capital formation. In addition the monetary policy variables are represented by money supply and
interest rate. Factors of balance of payments and exchange rate policy are indicated by net exports and
the exchange rate volatility, respectively. Finally, the factor of energy prices is representing the
economy characteristic of G-7 economies.

III.2. The Econometric Methodology
We investigate the causal relationship between crude oil price fluctuations and the real business cycle
in the G-7 economies after controlling for the determinants of business cycle. We employ several tests
to examine the relationship between the real business cycle and its determinants. These tests are the
unit root tests, the cointegration test using the Johansen procedure (see Johansen 1988), and then
Granger causality test (see Granger, 1986).

---

77
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
Each series for each country was tested for nonstationarity using the Augmented Dickey- Fuller
(ADF). The test for a unit root contained a constant term and lagged difference terms (depending on
the series itself). Then we test for cointegration between real GDP and the determinants of the business
cycle using the Johansen procedure. Finally, the short-run effect of changing oil prices on the business
cycle is analyzed using the Granger-causality test procedure with different lag-structures.

III.3. Testing for Cointegration
To model the relationship between the business cycle in real GDP and its determinants we consider
that their short-run dynamics can be represented by a vector autoregression VAR specification as
follows (see Ghali 1998):
Xt = c + π1Xt-1 + … + πkXt-k + ΨDt + εt (1)
where X is a vector containing all the variables included in VAR, t is time, c is a vector of constants or
drift terms, πi, i = 1…k, are matrices of time-invariant coefficients, and ε is a vector of i.i.d errors with
a positive covariance matrix. D is a matrix of deterministic components such as trend and dummies.
The VAR(k) model defined in equation (1) is covariance stationary if all values of Y satisfying:
⏐I - π1Y - π2Y2 - … - πkYk ⏐ = 0 (2)
lie outside the unit circle.
If some or all the variables in X are I(1), then we can exploit the idea that there may exist co-
movements of these variables and possibilities that they will trend together towards a long-run steady-
state equilibrium (i.e cointegrated). Hence, their behavior can now be represented using a vector error-
correction (VEC) model that incorporates the short-run as well as the long-run dynamics,
p-1
∆Yt = c + Σ Γi ∆Y + ΠY + ε
t -p
t -p
t (3)
i=1
where Π = -(I - Σi πi), i =1,…p, is the long-run parameter matrix, Γi = -(πi + 1 + … + πi), i = 1, …, p –
1 are estimable parameters, ∆ is a difference operator and εt is a vector of impulses which represent the
unanticipated movements in Xt with εt ~ niid(0, ∑).
With r cointegrating vectors, Π has rank r and can be decomposed as Π = αβ´, with α and β
both are matrices. β is the matrix of cointegrating vectors and α are the adjustment coefficients which
measure the strength of the cointegrating vectors in the VEC model.
The Johansen (1988) approach uses a maximum likelihood procedure to test the cointegrating
rank r and estimate the parameters β and α. If the cointegration test reveals the existence of long-run
relationships between the variables in X, then several implications including “spurious” regression,
Granger non-causality, and dynamic simulations can be established.
One advantage of the cointegration methodology is that it illustrates the conflict that exists
between the equilibrium framework and the disequilibrium environment from which the data are
collected. As illustrated by the VEC model in (3), this conflict can be easily resolved by extending the
equilibrium framework into one that accounts for disequilibrium by including the equilibrium error
measured by (β´Xt -1). Once the equilibrium conditions are imposed, the model is now describing how
the system is adjusting towards its long-run equilibrium state. Since the variables are supposed to be
cointegrated, then in the short-run deviations from the long-run equilibrium will feed back on the
changes in the dependent variables in order to force their movements towards the long-run equilibrium
state. Thus, the adjustment coefficients α measure the proportion by which the long-run disequilibrium
(or imbalance) in the dependent variables are corrected in each short-term period.

III.4. Testing for Granger Non-Causality
There are several procedures for testing Granger noncausality (GNC) using vector autoregressive
models. These were developed and studied by Toda and Phillips (1993, 1994), Toda (1995), Dolado

---

78
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
and Lutkepohl (1996), Zapata and Rambaldi (1997), and Yamada and Toda (1998). Recent surveys of
these procedures are in Caporale and Pittis (1999). A comparison of the properties of these procedures
based on simulation experiments can be found in Giles and Mirza (1999).
However, the choice between the alternative procedures depends on the pretesting for
integration and cointegration among the variables. That is, depending on the degree of integration of
the variables included in the system and depending on whether they are “sufficiently” cointegrated,
there is a particular specification of VAR that adequately ensures appropriate behavior of the test
statistic for testing the null of GNC. In particular, there are four different specifications. These are the
VAR in levels (VARL) model, the augmented VARL model, the vector error-correction model
(VECM), and the VAR in difference (VARD) model. In each one of these models, the statistic used to
test for GNC is a Wald statistic 1.
In general, if θ is an mx1 vector of parameters and R is a nonstochastic qxm matrix of known
parameters with rank q, then the Wald statistic to test H0: Rθ = 0 is
−
W = Tθˆ R
′ {′ ˆV
R
θˆ
1
( R }
′
θˆ
)
R (4)
ˆ ˆ
where θˆ is a consistent estimator of θ, V (θ ) is a consistent estimator of the asymptotic variance-
ˆ
covariance matrix of T (θ −θ ) , and T is the number of observations. Under appropriate conditions,
W is asymptotically distributed as a λ2(q) distribution under the null.
In the context of a VEC model, the null of GNC can be tested in three different ways using the
Wald statistic in (4). First, zero restrictions can be imposed on the α vector to test for weak exogeneity
of the variables. Toda and Phillips (1993, 1994) refer to these tests as long-run noncasality tests.
Second, zero restrictions can be imposed on the coefficients of the Γ matrix to test whether the lagged
coefficients of a variable are zero. This test is referred to as a short-run noncausality test. Finally, zero
restrictions can be imposed jointly on the coefficients in α and Γ to test for both short-and long-run
noncausality.


IV. Data Sources and Variables Definitions
Countries included in this study cover all the G-7 countries. These are Canada, France, Germany, Italy,
Japan, the United Kingdom, and the United States of America. The data are in quarterly base
throughout the period from 1970:1 to 2006:4. Quarterly data on the value of net exports, interest rates,
money supply, the consumer price index, and gross fixed capital formation are obtained from the
OECD database. Data on quarterly real GDP and employment are obtained from the IMF database
(International Financial Statistics). The nominal interest rate used for the G-7 countries is the deposit
rate in the case of Canada, Japan, France and the UK, whereas the federal rate, the call money rate and
the money market rate for the US, Germany, and Italy, respectively. The inflation is represented in the
data by the inclusion of the consumer price index using the 2000 base year. While the value of exports
is seasonally adjusted based on the FOB, the value of imports is seasonally adjusted based on the CIF.
The oil price variable is represented by the average of OPEC crude oil price on quarterly base. Data for
M1 is representing the money supply variable; however the data for M0 is used for the case of UK. The
money supply variable is not used in the data sample for France and Germany due to data availability.
The data for the gross fixed capital formation are seasonally adjusted. Data on the monthly nominal
exchange rates are obtained from the IMF database (International Financial Statistics).We measure
exchange rate volatility by the standard deviation of first differences of logs of monthly nominal
exchange rates on a quarterly base.



1 See Giles and Mirza (1999) for an extensive discussion of these models and a discussion of the comparative testing procedures.

---

79
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
V. Empirical Results
Before testing for the cointegration and granger causality, we investigate the time series properties for
all the variables included in the analysis. The results of testing the non-stationarity using the
Augmented Dickey-Fuller (ADF) test (Table 1), show that all variables are integrated of order one.
Across all countries, the log of oil prices is not stationary in level form, but is stationary in first
differences. This is shown in Table 2.

Table 1:
Unit Root Tests (Level)

Real
Interest
Gross Fixed
Money
Net
Country
CPI
Oil Price
Employment
GDP
Rate
Capital Formation
Supply
Export
Canada 0.471 -3.534
-3.977
-2.903 -2.571
-2.565
-2.727
-3.669
France -2.291
-2.655
-0.979
-1.298 NA
NA
-2.556
-0.918
-1.418
Germany -1.703 -2.838 -1.413
-2.012 -3.104
-2.564
-3.606
-0.887
Italy -1.235
-1.475
-2.335 -0.776 NA
A
-2.564 -1.631 0.983
Japan -1.495
-1.427
-1.912 -1.856 -3.806
-2.556
-2.470
-1.276
UK -1.583
-3.060
-1.635 -2.101 -2.147
-2.080
2.080
-1.540
USA -0.514
-3.502
-7.000 -2.029 -1.826
-2.564
-3.701
0.526
Note: T-statistic values are reported

Table 2:
Unit Root Tests (First Difference)

Real
Interest
Gross Fixed
Money
Net
Country
CPI
Oil Price
Employment
GDP
Rate
Capital Formation
Supply
Exports
Canada -10.059 -8.661
-11.652
-7.101
-10.019 -9.032 -5.138 -12.158
France -5.282
-9.876
-10.310
-1.963 NA
NA
-6.007 -6.160 -12.328
Germany -12.935 -6.936 -19.349
11.735
-5.615 -9.031
-6.421
-13.417
Italy -9.050
-7.453
-10.867 -3.758 NA
NA
-9.031 -5.143
-14.848
Japan -4.376
-9.375
-12.403 -3.261 -3.486
-6.007
-4.084
-1.977
UK -11.741
-8.721
-7.306 -4.666 -3.091 -3.275
-3.275 -6.529
USA -9.367
-4.862
-7.161 -7.518 -1.755
-9.031
-5.268
-4.318
Note: T-statistic values are reported

Next we test for cointegration between the variables using the Johansen (1988) procedure 2.
Then using the Granger representation theorem, we use the Granger Causality test to investigate the
intertemporal effect of oil prices on the real business cycle. We do this for the group of the seven
countries using different lag orders. The results of these tests are represented in Table 3. The Granger
causality test is employed up to twenty lagged terms for each country explaining whether changes in
oil prices are causing the real GDP to change, or if changes in a real GDP cause the oil prices to
change. The results reported in Table 3 indicate that, in general, changes of oil prices have an impact
on the real GDP across the G-7 economies. The short term neutrality of oil is found in Italy, Japan, and
the UK, whereas the oil effect is significant for the rest of G-7 economies especially Germany and
France.

2 To conserve space we only present the results of the Granger-causality tests. Results of cointegration are readily available
from the authors upon request.

---

80
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
Table 3:
Granger Causality Tests (Null Hypothesis: Oil Prices do not Granger Cause Real GDP) Country

Lag(s) Italy France
Canada Japan Germany USA
UK
1
0.2620 0.0003** 0.0446** 0.4057 0.5288 0.1040 0.0584*
2
0.5153 0.0038** 0.1669 0.2447 0.3497 0.2123 0.2907
3
0.0847*
0.0152** 0.3216 0.6994 0.1632 0.2707 0.5829
4
0.1804 0.0256** 0.5056 0.9234 0.0082** 0.1343 0.5650
5
0.3690 0.0078** 0.2790 0.9439 0.0282** 0.0736* 0.2550
6
0.3970 0.0204** 0.5749 0.9795 0.0267** 0.1313 0.1769
7
0.3916 0.0319** 0.7902 0.9854 0.0436** 0.2022 0.0746*
8
0.3994 0.0555* 0.7163 0.9881 0.0360** 0.2802 0.1091
9
0.4063 0.0644* 0.6524 0.9891 0.0396** 0.4478 0.2080
10
0.4023 0.1008* 0.8369 0.9686 0.0590* 0.1673 0.4497
11
0.3990 0.1777 0.7985 0.9798 0.0053** 0.1884 0.4378
12
0.5251 0.2928 0.7592 0.9454 0.0090**
0.0411**
0.5166
13
0.6066 0.3466 0.6662 0.9503 0.0103**
0.0643* 0.3505
14
0.8032 0.2210 0.7499 0.9674 0.0177**
0.1001* 0.2995
15
0.5058 0.1943 0.7694 0.9296 0.0301**
0.0995* 0.4693
16
0.4821 0.2247 0.8590 0.9105 0.0532* 0.1104 0.5434
17
0.1606 0.1541 0.8849 0.7121 0.1006* 0.0969* 0.5162
18
0.1636 0.1137 0.8139 0.5675 0.1301 0.0079**
0.6155
19
0.0756* 0.1255 0.7850 0.6882 0.0976*
0.0171**
0.4769
20
0.0649* 0.0054** 0.7901 0.7798 0.0936* 0.0180** 0.5561
Notes: P- values are reported
** Significance at 5% level, * Significance at 10% level

The Granger-causality test results also show a significant impact of oil prices on the business
cycle of U.S, Germany, and France. In the case of the U.S, the impact of oil price seems to start at the
twelfth quarter then disappears in the eighteenth quarter, suggesting that the impact of high oil prices
has a long run influence on U.S business cycle. The long run impact of oil prices in the U.S business
cycle might be explained by the stagflation effect. This means that, in the long run, oil price shocks
may lead to a stagflation impact on the macroeconomy of an oil importing country, which may lead to
slow down the growth rate of the economy and, then increase the inflation pressure. Also the policy
response of U.S monetary and fiscal policy, the oil shock’s persistence, and the size of oil shock may
delay the impact of high oil prices on the business cycle.
The short term impact of oil prices on the German business cycle seems to begin from the
fourth quarter and remains until the fifteenth quarter. There are several reasons for a clear effect of
high oil prices on German business cycle. First, Germany depends heavily on imports to meet its need
of energy due to the fact that it suffers from lack of domestic fossil-fuel resources. In addition, with
high standard of energy efficiency and environmental protection in West Germany versus highly
centralized energy sector in East Germany, merging the fundamentally different energy sectors of the
East and West was the main Germany energy policy after the reunification of Germany in 1990. This is
in turn may lead to implement such reforms in the energy sector in Germany to meet that needs. Also,
being the third largest oil importer in the world and a minor oil producer might be a reason to have the
German economy vulnerable to permanent oil shocks which last almost for 12 quarters, based on the
result.
Finally, France seems to have the highest short term effect of oil prices on its real business
cycle. The short run effect of oil prices on the business cycle is quite obvious for France, suggesting a
first quarter impact of oil price which lasts up to the seventh quarter. An explanation for that driven by
the French energy policy which lasts over a decade with securing the energy supply and protecting the
environment, which led to long term alternative sources of energy(mainly nuclear sources). This
suggests that France is subject to a short run exposure to high oil prices.
The short run neutrality of oil prices in Japan is quite realistic because Japanese oil dependency
has been decreased dramatically over time due to the Japanese efforts for diversification of energy
sources and significant increase in energy consumption efficiency. The use of policy changes is also

---

81
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
tending to play a role in Italy. Neutrality of oil prices in Italy may be justified by certain changes in
policies made by the government. Consumer spending represents a significant share in the Italian GDP,
thus the influence of high energy prices may weaken the consumer spending which in turns holds back
the growth of the economy. The policy changes include for example freezing energy tariffs, reducing
fuel prices at the pump in a purpose of offsetting the impact of rising of oil prices on growth. In the
case of the UK, the impact of high oil prices on its business cycles may be mitigated by the fact that
UK is a net oil exporter.
On the other hand, the short run effect of oil prices on the business cycle of Canada seems to be
noticeable. The effect starts from the first quarter and disappears afterward suggesting that high oil
prices have a one-time effect of the business cycle of Canada due to the fact that Canada is a net-
importer of oil.


VI. Conclusion
This study provides evidence of the short run impact of high oil prices on the business cycle of the G-7
economies. The evidence presented here shows that the impact of oil prices has different implications
across G-7 countries. We herein used the Granger causality test to investigate the causal relationship
between real GDP and its major determinants, in particular oil prices and that for the G-7 countries
using quarterly data over period from 1970:1 to 2006:4.
The findings of this paper show that the limiting effect of oil prices on real GDP varies from
one country to another across the G-7 economies. While, the oil effect is found to be significant for the
G-7 economies especially Germany and France, the short term neutrality of oil is found in Italy, Japan,
and UK. Particularly, changes of the government policy have played a role in mitigating the influence
of high oil prices in Japan, Italy, and France. The characteristics of the economy in the U.S, U.K,
Germany and Canada have shaped the role of oil influence on their business cycles. These differences
reflect the timing effect of high oil prices on the business cycle in some of the G-7 economies. The
short term one time effect (first quarter only) of high oil prices on the short run real GDP is found to be
in Canada. However, the economy of Germany is subject to permanent oil shocks which last almost 12
quarters. The impact of high oil prices has a long run influence on U.S business cycle, while France is
vulnerable to a short run oil impact that lasts up to seven quarters.

---

82
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
References
1]
Abel, Andrew B., and Bernanke, Ben S. (2001) “ Macroeconomics”, Addison Wesley, ISBN 0-
201-44133-0.
2]
Baltagi, Badi (2005), Econometric Analysis of Panel Data, John Wiley and Sons Ltd, Third
Edition, New York.
3]
Caporale, G. and Pittis, N. (1999), “Unit Root Testing Using Covariates: Some Theory and
Evidence”, Oxford Bulletin of Economics and Statistics, 61(4): 0305-9049.
4]
Clements, Michael P., and Krolzig, Hans-Martin (2004) “ Can regime-switching models
reproduce the business cycle features of US aggregate consumption, investment and output? ”,
International Journal of Finance & Economics, John Wiley & Sons, Ltd., 9(1):1-14.
5]
Davis, Steven J, and Haltiwanger, John. (2001),“ Sectoral Job Creation and Destruction
Responses to Oil Price Changes”, Journal of Monetary Economics, 48(3): 465-512.
6]
Dolado, Juan, and Lütkepohl, Helmut, (1996), “Making wald tests work for cointegrated VAR
systems”, Econometric Reviews, Taylor and Francis Journals, 15(4): 369-386.
7]
Fuller, W. A. (1976) “Introduction to Statistical Time Series”, New York: John Wiley.
8]
Ghali, Khalifa (1998), "Business Cycles in Developing Economies: The Case of Tunisia”,
Economia Internazionale 1(1): 171-188.
9]
Giles, Judith A., and Mirza, Sadaf (1999), “Some Pretesting Issues on Testing for Granger
Noncausality ", Econometrics Working Papers, 9914.
10]
Granger, C.W.J. (1986) “Developments in the Study of Cointegrated Economic Variables”,
Oxford Bulletin of Economics and Statistics, 48, 213-228.
11]
Gregory, A., Head, A., and Raynauld Jacques (1997), “Measuring World Business Cycles”,
International Economic Review, 38: 677-701.
12]
Hamilton, James, (1983), “Oil and the macroeconomy since World War II”, Journal of Political
Economy 91:228-248.
13]
Hamilton, James, (1997),“ Macroeconomic impacts of energy shocks (Book Review)”, Journal
of Economic Literature, 26(4).
14]
Hamilton, James, (1989), “A New Approach to the Economic Analysis of Nonstationary Time
Series and the Business Cycle”, Econometrica, Econometric Society, 57(2): 357-384.
15]
Hansen, Bruce (2001), “The New Econometrics of Structural Change: Dating Breaks in U.S.
Labor Productivity”, Journal of Economic Perspective 15:4: 117-128.
16]
Hess, Gregory D. (2000), “ Oil and the Macroeconomy”, Bradley Policy Research Center,
William E. Simon Graduate School of Business Administration, University of Rochester.
17]
Hooker, Mark A. (1996), “This Is What Happened to the Oil Price-Macroeconomy
Relationship: Reply”, Journal of Monetary Economics, 38(2): 221-222.
18]
Jimenez-Rodriguez, Rebeca, and Sanchez, Marcelo(2004),“Oil price shocks and real GDP
growth: empirical evidence for some OECD countries”, European Central Bank, Working
Paper Series: 362.
19]
Johansen, S. (1988) “Statistical Analysis of Cointegration Vectors”, Journal of Economic
Dynamics and Control, 12, 231-254.
20]
Keane, Michael P, and Prasad, Eswar S, (1996), “ The Employment and Wage Effects of Oil
Price Changes: A Sectoral Analysis”, The Review of Economics and Statistics, MIT Press,
78(3): 389-400.
21]
Kim, Chang-Jin, and Nelson, Charles R, (2001), “ A Bayesian Approach to Testing for
Markov-Switching in Univariate and Dynamic Factor Models”, International Economic
Review, Department of Economics, University of Pennsylvania and Osaka University Institute
of Social and Economic Research Association, 42(4): 989-1013.
22]
Kose, M., Otrok, C., and Whiteman, C. (2003),“International Business Cycles: World, Region,
and Country-Specific Factors”, American Economic Review, American Economic Association,
93(4): 1216-1239.

---

83
European Journal of Economics, Finance And Administrative Sciences - Issue 14 (2008)
23]
LeBlanc, Michael and Menzie Chinn (2004), “Do High Oil Prices Presage Inflation? The
Evidence from G-5 Countries”, Business Economics 39:2:.
24]
Mackinnon, J. G. (1991) “Critical Values for Cointegration Tests”, in R.F. Engle and. C.W.J.
Granger (eds) Long-Run Economic Relationships: Readings in Cointegration, Oxford
University Press, Oxford, 1993, 267-276.
25]
Mussa, Michael (2000), “The Impact of Higher Oil Prices on the Global Economy”, IMF
Working Papers.
26]
Raymond, Jennie, and Rich, Robert, (1997),“ Oil and the Macroeconomy: A Markov State-
Switching Approach”, Journal of Money, Credit, and Banking, 29(2): 193-213.
27]
Serletis, Apostolos and Todd Kemp (1998), “The cyclical behavior of monthly NYMEX energy
prices”, Energy Economics, Elsevier 20(3): 265-2571.
28]
Sims, C. (1980) “Comparison of Interwar and Postwar Business Cycles: Monetarism
Reconsidered”, American Economic Review, 70, 250-257.
29]
Toda, Hiro Y, and Phillips, Peter C B, (1993), “Vector Autoregressions and Causality”,
Econometrica, Econometric Society, 61(6): 1367-1393.
30]
Toda, Hiro Y, and Phillips, Peter C B, (1994), “Vector Autoregression and Causality: a
Theoretical Overview and Simulation Study”, Econometric Reviews, Taylor and Francis
Journals, 13(2): 259-285.
31]
Toda, Hiro Y, (1995), “Finite sample performance of likelihood ratio tests for cointegrating
ranks in vector autoregressions”, Econometric Theory, 11: 1015-1032.
32]
Yamada, H., and Toda, H.Y. (1998), “ Inference in Possibly Integrated Vector. Autoregressive
Models: Some Finite Sample Evidence”, Journal of Econometrics, 86: 55-95.
33]
Yoon, Jae. Ho, (2004), “Oil and the G7 Business Cycle Friedman's Plucking Markov Switching
Approach”, Econometric Society 2004 Far Eastern Meetings, Econometric Society, 773.
34]
Zapata, Hector O, and Rambaldi, Alicia N, (1997), “Monte Carlo Evidence on Cointegration
and Causation”, Oxford Bulletin of Economics and Statistics, Department of Economics,
University of Oxford, 59(2): 285-298.

---