The Relationship between Government

The Causal Relationship between Trade
Openness and Government Size: Evidence
from Saudi Arabia
Khalid H. A. Al-Qudair
Associate Professor, Department of Economics, College of Administrative Sciences
King Saud University, Riyadh, Saudi Arabia
‫العالقة بين االنفتاح التجاري وحجم الحكومة في المملكة العربية السعودية‬
‫د‪.‬خالد بن حمد بن عبدهللا القدير‬
‫ملخص البحث‪:‬‬
‫يهدف البحث إلى دراسة طبيعة العالقة بين االنفتاح التجاري وحجم القطاع الحكومي في‬
‫المملكةةة العرةيةةة السةةعودية فةةي ايجةةا الطومةةا باسةةتكدام نمةةو‬
‫التكامةةا المرةةترتح وتحديةةد اتجةةا‬
‫العالق ةةة الس ةةببية ب ةةين االنفت ةةاح التج ةةاري وحج ةةم القط ةةاع الحك ةةومي ف ةةي ايج ةةا الطوم ةةا والق ةةير‬
‫باسةةتكدام نمةةو‬
‫متجهةةاح ت ةةحي الكطةةاخ وقةةد دك اكتبةةار التكامةةا المرةةترت لةةى وجةةود القةةة‬
‫توازنيةةة طوملةةة ايجةةا بةةين االنفتةةاح التجةةاري وحجةةم القطةةاع الحكةةوميخ كمةةا و ة اكتبةةار السةةببية‬
‫في ايجا الطومةا نن ننةات القةة سةببية اح اتجةا واحةد تتجةن مةن االنفتةاح التجةاري الةى حجةم‬
‫القطاع الحكومي وليس العكسخ كما نو‬
‫اكتبار السببية في ايجا الق ير ننن ال توجد القة‬
‫سببية بين االنفتاح التجاري وحجم القطاع الحكوميخ وب فة امةح يمكن القوك نن النتائج تؤمةد‬
‫فر ية التعويض التي تؤكد ان ننةات القةة سةببية موجبةة تتجةن مةن االنفتةاح التجةاري الةى حجةم‬
‫القطاع الحكومي في ايجا الطوما وليس في ايجا الق يرخ‬
‫‪-2 -‬‬
The Causal Relationship between Trade
Openness and Government Size: Evidence
from Saudi Arabia
Khalid H. A. Al-Qudair
Abstract:
This study examines the long run equilibrium relationship between trade openness
and government size in the Kingdom of Saudi Arabia using Cointegration technique
and the direction of causality relationship in the long and short runs between the
variables utilizing the Vector Error Correction Model (VECM). The Cointegration
test indicated the existence of the long run equilibrium relationship between trade
openness and government size. The causality test indicated that there is a unidirectional causal relationship that runs from trade openness to government size in the
long run not vice versa. In addition, the causality test indicates the absence of short
run causality between trade openness and government Size. Over all, it may be
concluded that the results provide support to the compensation hypothesis that entails
a positive causality that is running from the trade openness to the government size in
the log run not in the short run.
-3 -
I. Introduction
The direction of causality relationship between different time series has
become recently the concern of many studies using advanced econometric techniques.
The trade openness and government size are two major macroeconomic variables that
influence most economic variables; especially, economic growth. Many studies have
focused their attention on the causal relationship between trade openness and
economic growth, while other studies have focused on the causal relationship
between government size and economic growth. Recently, attention has been shifted
to the causal relationship between trade openness and government size. The purpose
of this paper is to investigate the causal relationship between trade openness and
government size in the Kingdom of Saudi Arabia over the period 1970-2001. The
evidence in either direction has important policy implications. The rest of the paper is
organized as follows. Section II presents a review of the relevant theoretical and
empirical works. Section III is devoted to the methodology used to test the
relationship between trade openness and government size, while Section IV describes
the data and discusses the empirical findings. Finally, Section V concludes.
II. Theoretical Background and Previous Empirical works
The effect of government size, on one hand, and trade openness, on the other,
on economic growth is a major concern of different empirical studies in
macroeconomics, pubic finance, and international trade.
Economists recognize the role of government expenditure in the economy in
order to fulfill its vital functions. However, the role of government expenditure in
promoting economic growth was not widely accepted until Keynes published The
General Theory of Employment, Interest, and Money in 1936. Given the inefficiency
of private demand, expansionary fiscal policy can fill the gap between the aggregate
-4 -
demand and aggregate supply. The Keynesian income policies were; then, used
extensively in the Western Hemisphere in order to implement different economic and
social programs. However, these policies were questioned as they failed to overcome
the unemployment problem and contributed in accelerating inflation. In fact, some
studies including Landau (1983), Kormendi and Meguire (1985), and Peden and
Bradley (1989) among others found that government expenditure has a negative
effect on economic growth. Government through its involvement in economical,
social, legal, security and other programs has a negative effect on economic growth.
The large government sector may crowed out private investment and increase the
inefficiency in the economy which in turn has an adverse effect on economic growth.
Recently, the debate over the role of government expenditure in economic activities
has shifted to the effect of government size on economic growth. Landau (1983),
Peden and Bradley (1989), Barro (1989, 1997), Gwartney, Lawson, and Holcombe
(1998) have argued that the expansion of government expenditure beyond a certain
level has a negative effect on economic growth. However, Rubinson (1977), Ghali
(1998), and Rosen and Weinberg (1998) found that a larger government size
promotes economic growth. Moreover, Karras (1994) found that public and private
consumption are complementary. Moreover, he argued that the relationship depends
on government size. As government size increases, government consumption turns to
be substitutable to the private consumption rather than complementary. Karras (1996)
argued that the impact of government expenditure on private consumption depends on
government size. As government size increases, government consumption turns to be
substitutable to the private consumption rather than complementary. Furthermore,
Karras estimated the optimal size of government for the world as a whole to be
approximately twenty-three percent of GDP. On the other hand, Grossman (1988)
found that government expenditure in the U.S. has a positive effect on economic
growth, while the government size has negative effect on economic growth
With regard to the effect of trade on economic growth, the theory of
Comparative Advantage asserts that specialization in certain goods along with free
-5 -
trade will allow all parties, which are involving in trade, to reach a higher level of
welfare through efficient allocation of resources. Krueger (1978), and Edwards
(1997), among others, have found that more trade stimulates economic growth. In the
context of the new growth theory, Grossman and Helpman (1991) argued that trade
openness enables countries to increase their stock of capital and resources which in
turn would improve their productivities and utilize their available resources.
Considerable works have been done on the relationship between trade openness and
economic growth. These studies include: Bhala and Lau (1991) who found that trade
openness has a positive effect on economic growth. Sinha and Sinha (1996) found
that although there is a long-run relationship between trade openness and GDP in
India, the two time series are independent. Anoruo and Ahmad, (1997) found
evidence for bi-directional causality between economic growth and openness in the
ASEAN countries. Al-hoqubani (2004) investigated the nature of the relationship
between openness and economic growth in Saudi Arabia. The findings of the study
indicated that there is a unidirectional causality that runs from economic growth to
trade openness.
As has been mentioned in the introduction, in recent years economists have
shifted their attention to the causal relationship between government size and trade
openness. One view asserts that trade openness Granger cause government size.
Alesina and Perotti (1997) advocated the efficiency hypothesis which suggests that
free trade, along with the economic globalization, are the driving forces of
government expenditure. Both government expenditure and taxes have to rise in order
to overcome the adverse effects of international competitiveness on domestic
economy. Rodrik (1997, 1998) investigated the nature of the relationship between
trade openness and government size using cross-country data. He found that there is a
unidirectional causality that runs from trade openness to government size. Rodrik
argues that this evidence suggests that there may be a degree of complementary
between markets and governments. Moreover, he suggests that the causal relationship
between trade-openness and government size can be explained by compensation
-6 -
hypothesis. More dependency on foreign trade means that domestic economy is
dependent to some extent on the development of its trading partners, which in turn
give incentives for government to provide social insurance against international
competitiveness. Balle and Vaidya (2002) has found that public welfare and health
services expenditures are positively correlated with the level of the United States’s
trade openness. This suggests that trade openness in the U.S. has resulted in state
governments responding to the adverse effects of increased international trade activity
by providing greater social insurance.
Molana, Montagna, and Violato (2004) have tested the compensation
hypothesis that a positive causality is running from trade-openness to government
size using time series data for 23 industrialized OECD countries over the 1948-1998
periods. Their findings do not support the hypothesis.
III. Methodology
The causal relationship between trade openness and government size known
as Granger causality is concerned with the relevance of past information of each one
of variables in predicting the value of the other (Granger, 1969, 1988).
The causality test relationship between the trade openness and government size
requires three steps. First, the time series would be analyzed in order to determine the
order of integration. Second, the long run relationship between the trade openness and
the government size is investigated. Finally, the short run dynamics as well as the
causality relationship in the short run and the long run between the trade openness
and the government size would be investigated.
-7 -
Unit Root Test:
Most of time series have unit root as many studies indicated including Nelson
and Plosser (1982), and as proved by Stock and Watson (1988) and Campbell and
Perron (1991) among others that most of the time series are non-stationary. The
presence of a unit root in any time series means that the mean and variance are not
independent of time. Conventional regression techniques based on non-stationary
time series produce spurious regression and statistics may simply indicate only
correlated trends rather than a true relationship (Granger and Newbold, 1974).
One of the most widely used unit root test is the Augmented Dickey-Fuller
(ADF) unit root test (Dickey and Fuller, 1979, 1981).
Alternatively, Phillips (1987) and Phillips and Perron (1988), PP, have proposed
a nonparametric method to correct a wide variety of serial correlation and
heteroskedasticity. Perron (1989, 1990) demonstrates that if a time series exhibits
stationary fluctuations around a trend or a level containing a structural break, then
unit root tests will erroneously conclude that there is a unit root.
The unit root test and the order of the integration would be preformed on both
the original series of the trade openness and the government size and their differences
using the PP unit root test.
Cointegration Test:
The Johansen (1988) and Johansen and Juselius (1990) proposed a procedure to
test for cointegration which is preferable to the Engle-Granger (1987) procedure since
it allows feedback effects among the variables under investigation. The procedure is
based on likelihood ratio (LR) test to determine the number of Cointegrating vectors
in the regression.
-8 -
Error Correction Model and Causality Tests:
Having established the long run equilibrium relationship between the trade
openness and the government size, the short run adjustments are estimated using the
Vector Error Correction Model (VECM). The error correction model is based on the
two following equations:
m
n
i 1
j 1
Opent   0  1et 1  i Opent i   j Govsizet  j   t
m
n
i 1
j 1
Govsizet  0  1ut 1   i Govsizet i    j Opent  j  ut
(1)
(2)
Where et 1 and t 1 represent the error-correction terms lagged residuals from the
cointegration relations that will capture the speed of the short run adjustments toward
the long run equilibrium.
Furthermore, the VECM, equations (1) and (2), allows to test for short run as
well as the long run causality between the trade openness and the government size.
The short run causality is based on a standard F-test statistics to test jointly the
significance of the coefficients of the explanatory variable in their first differences.
The long run causality is based on a standard t-test.
IV. The Empirical Findings
The variables that are used in the model to investigate the nature of the
relationship between the openness and the government size in Saudi Arabia are:
Trade Openness (Open) measured by real total trade/ real GDP and government size
(Govsize) measured by real government consumption/ real GDP in natural log forms.
The annual data employed in this study covers the period from 1970-2001 obtained
from the 38th annual report of Saudi Arabian Monetary Agency.
-9 -
Properties of the Time Series:
The first step in constructing the cointegration model and testing the Granger
causality relationship is to test the stationarity of the series over time and to determine
the degree of integration based on PP unit root test. The analysis of time series
showed that the time series of the trade openness (Open) and the government size
(Govsize) are not stationary at their levels at the 5% level of significance. However,
the series are stationary at their first differences at the 5% level of significance, which
indicate that the series are integrated of degree one (I (1)).
Table (1): PP Unit Root Test
First difference
with intercept
and Trend
First difference
with intercept
Level with
intercept and
trend
Level with
intercept
Variable
-4.07
-4.18
-2.43
-1.22
Open
-8.72
-8.49
-2.71
-2.24
Govsize
Critical values:
Intercept
Intercept and Trend
At (1%) level of Significance
-3.66
-4.29
At (5%) level of significance
-2.96
-3.57
At (10%) level of significance
-2.62
-3.22
Cointegration Test:
Since the series of Open and Govsize are integrated of degree one i.e. I (1), they
may be cointegrated if there exist some linear combination of the series that can be
tested for stationarity i.e. (I (0)).
Johansen (1988) and Johansen–Juselius (1990) procedure is used to test for
Cointegration between Open and Govsize. Table (2) presents the result of the
Johansen – Juselius cointegration technique in order to examine the existence of the
long run relationship between Open and Govsize. Table (2) presents the result of
Johansen–Juselius test where the null hypothesis of no cointegration between Open
and Govsize is rejected at the (5%) level of significance. However, the null
-10-
hypothesis of the existence of at most one cointegration equation is accepted at (5%)
level of significance. This result indicates that there exists at least one cointegration
equation between the two time series.
Table (2): Johansen Cointegration Test
Eigenvalues
Likelihood
Ratio
5% Critical
Value
1% Critical
Value
Null
hypothesis
"No. of CE"
0.51
26.52
19.96
24.60
None
0.17
5.59
9.24
12.97
At most one
The Vector Error Correction Model (VECM):
Since the cointegration test revealed that there exists a long run equilibrium
relationship between Open and Govsize, following Johansen– Juselius (1990), we can
use a vector error-correction model (VECM) in which an unconstrained VAR is used
in order to investigate the short run dynamics and to assess the direction of Granger
causality in both the short and the long run as well. The inclusion of the error terms in
the Granger causality test equations (1) and (2) will enable us to distinguish between
short run and long run causality.
-11-
Table (3): Estimates for VECM Regression
Regresses
et 1
 Ln Open
 Ln Govsize
-0.005
(-0.76)
 t 1
-0.06
(-4.73)
 Ln Open -1
0.26
0.26
(1.34)
(0.65)
-0.14
0.26
(- 0.67)
(0.41)
-0.04
- 0.15
(-0.53)
(- 0.43)
-0.09
-0.19
(-1.30)
(-0.48)
R2
0.14
0.56
F
1.01
7.82
S. E.
0.09
0.17
Log likelihood
32.38
12.45
 Ln Open-2
 Ln Govsize-1
 Ln Govsize-2
(Terms in brackets are t – ratios).
Short Run Dynamics:
Table (3) presents the VECM estimations. The lagged error term coefficient
(et 1 ) is statistically insignificant, while the coefficient of (  t 1 ) is negative and
statistically significant. The value of  t 1 indicates the speed of adjustment of any
disequilibrium towards the long-run equilibrium. six percent of the disequilibrium in
Govsize is corrected each year. Moreover, the significant error term in the Govsize
equation supports the existence of a long run equilibrium relationship between Open
and Govsize.
-12-
Long and Short Run Causality Tests:
Table (3) presents the results of both the short run Granger causality test based
on a standard F-test statistics that tests jointly the significance of the coefficients of
the explanatory variables in their first differences as well as the long run Granger
causality test based on a standard t test statistics that test the significance of the error
terms lagged one period.
Long run Causality Test:
The coefficient of the error term in the Open equation (1) based on t-test
statistics from the VECM is statistically insignificant which means that the error term
(et 1 ) does not contribute in explaining the changes in Open equation. However,
the causality test based on the standard t test statistics from the VECM indicates that
causality runs from Open to Govsize since the coefficient of the error term in Govsize
equation (2) is statistically significant and negative based on standard t-test which
means that the error term (  t 1 ) contributes in explaining the changes in Govsize.
Therefore, there is unidirectional causality running from openness (Open) to
Government size (Govsize) in the long run.
Short Run Causality Test:
The coefficients of the explanatory variables in their first differences in the
Open equation are jointly statistically insignificant based on F-test statistics.
Although, the coefficients of the explanatory variable in their first differences in the
Govsize equation are jointly statistically significant based on standard F-test statistics,
none of the coefficients of the explanatory variables based on t-test statistics are
statistically significant. The significance explanatory variables based on F-test
statistics in the Govsize equation do not mean that Open Granger cause Govsize, but
rather a reflection of the significance of the error term (  t 1 ). Therefore, there is no
-13-
causality between Open and Govsize in the short run. In order to confirm the result of
the short-run causality between Open and Govsize based on VECM estimates, a
standard Granger causality test is run based on F-statistics.
Table (4) presents pairwise Granger causality test that confirm the earlier finding
based on VECM estimates of the absence of short run causality between Open and
Govsize.
Table (4): Results of Pairwise Granger Causality Test: number of lags = 2
Null hypothesis
F-statistics
Probability
Δ Openness does not Granger cause
0.06
0.94
0.38
0.69
Δ Govsize
Δ Govsize does not Granger cause
Δ Openness
V. Conclusion
The goal of this paper is to investigate the long run relationship between the
openness and the government size in the Kingdom of Saudi Arabia using
cointegration technique and the short run dynamics as well the direction of causality
in both the long and the short run based on ECVM. Data properties were analyzed to
determine their stationarity using the PP unit root test which indicated that the series
of Open and Govsize are I (1). The cointegration test based on Johansen - Juselius
technique indicates that although the two time series of Open and Govsize may be in
disequilibrium in the short run, there is a long run equilibrium relationship between
them.
The long run causality test based on the standard t test statistics from the
VECM indicates that there is a unidirectional causality runs from Open to Govsize
not vice versa.
-14-
The short run causality test between Openness and Government size based on
F-test statistics from the VECM indicated the short run causality.
Pairwise Granger
causality test confirmed the absence of the short run causality based on VECM
estimates.
The policy implication of the results suggests that although the international
trade represents a considerable component of GDP, the government expenditure in
Saudi Arabia does not provide social insurance to the domestic economy against
international competitiveness in the short run. However, the government expenditure
in Saudi Arabia increases in the long run as a result of international trade in order to
overcome the negative effects of international competitiveness. Over all, it can be
concluded that the results provide support to the compensation hypothesis that entails
a positive causality running from trade openness to government size in the log run not
in the short run.
-15-
References
Alesina, A. and Perotti,R., (1997)., "The Welfare State and Competitiveness",
American Economic Review, 87, PP. 921-39.
Al-hoqubani, M.,(2002)." Causality Between Openness And Economic Growth:
Empirical Evidence From Saudi Arabia ", The journal of Economic and
Legal Researches, College of Law, University of Monofia (April).
Anoruo, E., Ahmad, Y., ( 1997), Openness and Economic Growth: Evidence
From Selected ASEAN Countries ", The Indian Economic Journal, 47 (3),
PP. 110-117.
Balle, F., and Vaidya, A., (2002). " A Regional Analysis of Openness and
Government Size ", Applied Economic Letters, 9 (5), PP. 289-92.
Barro, R.J. (1989): " A cross-country study of growth, saving, and government"
in National Saving and Economic Performace (eds) B.D. Berneheim and J.
Shoven, University of Chicago Press and NBER, Chicago.
Barro, Robert J. (1997). Determinants of Economic Growth: A Cross-Country
Empirical Study, Cambridge: MIT Press.
Bhala, S., and Lau, L. J., (1991), "Openness, Technological Progress, and
Economic Growth in Developing Countries", Background Paper for World
Development Report.
Campbell, J. Y., and P. Perron (1991), "Pitfalls and Opportunities: What
Macroeconomists Should Know about Unit Roots", in NBER
Macroeconomics Annual, ed. by O. J. Blanchard, and S. Fischer, pp. 141–
201. MIT Press, Cambridge, MA.
Dickey, D. A., and Fuller, W. A (1979), "Distribution of the Estimators for
Autoregressive Time Series with a Unit Root", Journal of the American
Statistical Association, 74, PP. 427-431.
Dickey, D. A., and Fuller, W. A (1981),"Likelihood Ratio Statistics for
Autoregressive Time Series with a Unit Root", Econometrica, 49, PP.
1057-1072.
Edwards, S., (1997), Openness, Productivity and Growth: What Do We Really
Know? National Bureau of Economic Research Working Paper,
http://papers.nber.org/papers/w5978.
Engle, R.F., and Granger, C.W.J. ( 1987), "Co-Integration and ErrorCorrection: Representation, Estimation and Testing", Econometrica, 55,
PP. 251-276.
Ghali. K. (1998): " Government size and economic growth: Evidence from a
Multivariate cointegration analysis", Applied Economics, 31, 795-987.
-16-
Granger, C. W. J., (1969), "Investigating Causal Relations by Econometric
Models and Cross-Spectral Methods", Econometrica, 37, PP. 242-38.
Granger, C. W., (1988), "Some Recent Development in a Concept of
Causality", Journal of Econometrics, 39, PP. 199-211.
Granger, C.W. J., and Newbold, P. (1974), "Spurious Regression in
Econometrics", Journal of Econometrics, 2, PP. 111-120.
Grossman, G. M. and Helpman.E.,( 1991). Innovation and Growth in the Global
Economy. Boston: MIT Press.
Grossman, P.J. (1988): " Government and economic growth: A nonlinear
relationship" Public Choice, 56, PP. 193-200.
Gwartney, J., Lawson, R., and Holcombe, R. (1998): " The size and functions of
government and economic growth", Report for the Joint Economic
Committee of the Congress (April).
Johansen, S. (1988). "Statistical Analysis of Cointegration Vectors", Journal of
Economics, Dynamics and Control, 12, PP. 231-54.
Johansen, S., and Juselius, K. (1990). "Maximum Likelihood Estimation and
Interference on Cointegration with Application to the Demand for
Money", Oxford Bulletin of Economics and Statistics, 52, PP. 169-210.
Karras, G. (1994). “Government Spending and Private Consumption: Some
International Evidence”, Journal of Money, Credit and Banking, 26, PP. 922.
Karras, G. (1996). The Optimal Government Size: Further International
Evidence on the Productivity of Government Services, Economic Inquiry,
34 (April), PP. 193-201.
Keynes, J. M., "The General Theory of Employment, Interest and Money"
,(New York: Harcourt Brace Jovanovich, Inc., 1936).
Kormendi, R. and Meguire, P. G. (1985), " Macroeconomic Determinants of
Growth: Cross-Country Evidence", Journal of Monetary Economics, 16,
PP. 141-163.
Krueger, A., (1978), " Foreign Trade Regimes and Economic Development:
Liberalization attempts and consequences ", Cambridge, MA: Ballinger
Publication company for National Bureau of Economic Research (NBER)
NY.
Landua, D. (1983): " Government expenditure and economic growth: A cross
country study", Southern Economic Journal, 49, PP.783-792.
Molana, H., Montagna, C., Violato, M.(2004) " On the Causal Relationship
between Trade Openness and Government Size: Evidence from 23 OECD
Countries",Working Paper No. 164, April 2004.
-17-
Nelson, C. and Plosser, C. (1982), "Trends and Random Walks in
Macroeconomic Time Series; Some Evidence and Implications", Journal
of Money Economics, 10, PP. 139-162.
Peden, E.A. and Bradley, M.D. (1989): " Government size, productivity, and
economic growth: The post-war experience", Public Choice, 61, PP. 229245.
Perron, P. (1989), "The Great Crash, the Oil Price Shock, and the Unit Root
Hypothesis", Econometrica, 57, PP. 1361-1401.
Perron, P. (1990), "Testing for a Unit Root in a Time Series with a Changing
Mean", Journal of Business and Economic Statistics, 8 (2), PP. 153-162.
Phillips, P. C. B. (1987), "Time Series Regression with a Unit Root",
Econometrica, 55 (2), PP. 277-301.
Phillips, P.C.B., and Perron, P. (1988), "Testing for a Unit Root in Time Series
Regression", Biometrika, 75, PP. 335-346.
Rodrik, D. (1997), " Trade, Social Insurance, and the Limits to Globalization",
NBER, Working Paper No. 5905.
Rodrik, D. (1998). “Why do more open economies have bigger governments?”,
Journal of Political Economy, 106 (5), PP. 997-1032.
Rosen, S. and Weinberg, B. (1998), " incentives, efficiency and government
provision of public services", Annual World Bank Conference on
Development Economics 1997, the World Bank, PP. 139-166.
Rubinson, R. (1977): " Dependency, government revenue and economic growth
1955-70", Studies in Comparative International Development, 12, PP.328.
SAMA, "Annual Report", The Saudi Arabia Monetary Agency, Vol. 38, 2002.
Sinha, D. and Sinha, T., (1996), "Openness and Economic Growth: Time Series
Evidence From India" , Applied Economics, 21-28.
Stock, J. H. and Watson, M. W. (1988), "Testing for Common Trends", Journal
of American Statistical Association, 83, PP. 1097-1107.
-18-