Content uploaded by Ghazi Al-Assaf
Author content
All content in this area was uploaded by Ghazi Al-Assaf on Mar 07, 2014
Content may be subject to copyright.
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-79-
1Assistant Professors of Economics, Department of Business
Economics, The University of Jordan .
a.altarawneh@ju.edu.jo
g.alassaf@ju.edu.jo
Received on 11/3/2013 and Accepted for Publication on
6/10/2013.
© 2014 DAR Publishers/The University of Jordan. All Rights Reserved.
Trade Openness and Real Investment in Jordan
An ARDL Bound Testing Approach
Alaaeddin A. Al-Tarawneh1 and Ghazi I. Al-Assaf 1
ABSTRACT
This paper investigates the relationship between trade openness and real investment in Jordan, over the period
1976-2010. An Autoregressive Distributed Lag (ARDL) approach is employed in order to explore the
association between investment and trade openness by estimating an investment function for the Jordanian
economy and incorporating trade openness as a determinant of investment. The Empirical results provide strong
evidence on the presence of a long-run stable investment function. In the short-run, trade openness has a positive
and significant effect on real investment; a consistent result with the literature. Therefore, the results confirm the
significant impact of trade openness on real investment found in the previous literature using a cointegration
technique. This finding emphasises the importance of the trade liberalisation effects on real investment
especially in the case of Jordan.
Keywords: Trade Openness, Investment, Cointegration, ARDL, Jordan.
INTRODUCTION
Economic theory stresses the importance of
investment for economic growth, development, and
economic stability. As a result, a number of reasons may
be proposed to justify a study of investment. Firstly,
investment is highly volatile; thus, investment
movements have important effects on the short-run
fluctuations of employment, income, and productivity.
Secondly, it is important in determining how much an
economy’s output is invested to increase future capacity.
In the long run, investment demand is considered to be
one of the important determinants affecting the future of
living standards of the population. Thirdly, investment
establishes some important issues relating to the
financial markets, and it has important feedback effects
on these markets. On the other hand, a few studies have
investigated the impact of investment on trade openness
especially in the case on developing economy.
Jordan has a strategic location. It is located at the
crossroads between Europe, Asia and Africa, and is
attached to the Red Sea through the Gulf of Aqaba port
and other ports via neighbouring countries. These
features make Jordan attractive for investors. However,
since the mid 1990’s, the Jordanian authorities passed
encouraging legislations in an attempt to create a proper
investment environment that attracts investors to invest
in Jordan, in addition to holding several bilateral and
multi-trade agreements with neighbouring and other
countries, and formatting appropriate infrastructure
needed for investment.
Among these agreements is the Free Trade
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-80-
Agreement (FTA) between Jordan and the United States.
It was the first convention of its kind to be signed with
an Arab country. The agreement provides for an ending
of all types of tariffs on industrial goods and farm
products. As the convention on the Euro-Jordanian
partnership, which plays an important role in
encouraging investment in Jordan, Jordan signed the
Euro-Mediterranean Association Agreement, which has
the status under which there is free trade with European
countries, as well as the free trade agreement with Arab
countries to facilitate and develop trade exchange
between Arab countries.
Investment in Jordan, defined in this study as the
change in the level of the fixed capital stock, has
fluctuated over time. This study attempts to investigate
whether there is a stable investment function for the
Jordanian economy. Determining the key features of
investment, with concern on the trade openness, enables
policy makers to identify the effect of trade policies on
real investment in Jordan.
The rest of this paper is structured as follows:
Section 2 reviews some of the literature in this field.
Section 3 discusses the main model used in the analysis.
Section 4 presents data and methodology adopted in this
paper. The empirical results are reported in Section 5,
while Section 6 is the conclusion.
2. Previous Empirical Work
Many empirical studies have used a number of
variables, suggested by the theory, in an attempt to
describe the determinants of investment in both
developed and developing countries. Shafik (1992)
provided a comprehensive survey of the empirical work
on estimating investment function before 1990s on
developing countries. She used an error-correction
model and co-integration to test Egyptian data for a
private investment function that takes into account some
features of developing countries. She found that a
number of variables, including mark ups, internal
financing, demand and the cost of investment, are the
main determinants of the private investment in Egypt.
Sun (1998), using data on China for the period 1953 to
1995, found that the long-run investment function can be
characterised by co-movements of real fixed investment,
grain output per capita, and energy supply per capita.
In their article, du Toit and Moolman (2004) found
that, in South Africa, investment depends on the interest
rate (the user cost of capital), international position, and
domestic and foreign financial constraints. Ismihan et al.
(2005) used data from Turkey covering the period 1963
to 1999. The main findings they made were that capital
formation and growth are seriously affected by the
macroeconomic instability of the Turkish economy.
Anoruo et al. (2007) estimated a neoclassical model of
investment for Bangladesh covering the time period
1973 to 2004. They found that there is an equilibrium
relationship between the investment-output ratio, real
output, and real interest rate.
Heim (2008) identified seven major variables that
have a significant effect on the investment demand in the
US economy for the period 1960 to 2000. These
variables are the crowd out problem, depreciation,
growth rate of GDP, interest rate, growth in stock values,
exchange rate changes, and company profitability.
Moore (2010), using panel data for 107 developing
countries over the period 1970 to 2006, tested
McKinnon’s hypothesis that the rate of return on money
does matter for investment in developing countries. His
findings appear to support this hypothesis under a
number of conditions.
The linkage of trade openness and investment has
traditionally been an important topic in developing
countries. It attracted attention due to the vital role of
trade openness in the development in many economies.
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-81-
A number of recent papers have empirically examined
the relationship between trade, economic growth, and
Foreign Direct Investment (FDI), using a wide variety of
econometric techniques. Only very few studies have
focused on the linkage between trade openness and level
of investment.
One of the very beginning efforts to explore the
impact of trade openness on investment is done by
Baldwin and Seghezza, (1996). This study is one of the
first to discuss the role of trade liberalization in fostering
the capital accumulation and affecting domestic
investment. They employed a model using the Three
Stages Least Squares and found that trade barriers
depresses investment and thereby slows economic
growth. Trade openness was found to have a positive
impact on the level of investment by allowing domestic
agents to import relatively cheaper capital goods from
foreign countries.
Harrison (1996) also investigated the association
between trade openness and growth, and extended the
analysis to examine the relationship between openness
and investment. He found, using a panel data, trade
openness affects domestic investment positively.
Another panel data study was conducted by Salahuddin
and Islam (2008) who found a positive relationship
between trade openness and investment. They have
extended the investigation pertaining determinants of
gross investment in developing countries by including
investment.
A very recent study by Oladipo (2011) employed
cointegration techniques to examine the impact of trade
liberalization (openness) on long run economic growth.
The empirical results, obtained from a quarterly data on
Mexico, showed that investment play a significant role
in determining the economic growth along with the trade
openness in Mexico. The study has highlighted the
interrelationship between trade liberalization,
investment, human capital, and growth. The researcher
found all these variables are statistically cointegrated.
As can be noticed from the previous review of the
main literature on the relationship between trade
openness and investment, there have been a few studies
connecting trade openness to the level of investment,
and there is no research exclusively devoted to the case
of Jordan. In the present paper, we intend to fill in this
gap in the literature concerning the links between trade
openness and real investment by employing a long span
of data in the case of a small developing country
(Jordan) and using a cointegration analysis based on the
ARDL approach.
3. Model of Investment
Net investment can be defined as the net change in
the capital stock (K) since the previous time period (Kt -
Kt-1) and equals total investment (It) minus replacement
investment (δKt-1),
where (δ) is the capital stock depreciation rate and is
measured to be constant.
The next short sub-sections will discuss a number of
different theories of investment behaviour, starting with
the Keynesian theory and the accelerator models,
followed by the neoclassical model of investment, and
Tobin’s q model.
Keynesians and the Accelerator Model
Keynesians consider investment as one of the key
determinants of macroeconomics activity. Keynes
(1936) emphasised the essentially unstable nature of
investment, because investment is based on expectations.
However, he derived an investment demand function
which was inversely linked to the interest rate, which
then formed a key transmission mechanism between
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-82-
monetary policy and the real side of the economy.
The Keynesian theory assumed that firms would
implement an investment project only if the discounted
flow of expected future revenue from that project (ER)
exceeded the total cost (C). Thus the firm invests if the
net return (NR) is positive.
The rate of return, r, required to equate total revenue
with total cost (or NR=0) was termed the marginal
efficiency of capital by Keynes. When this rate exceeds
the market interest rate, which represents the opportunity
cost, the investment project would be undertaken. A
decrease in the market interest rate would result in the
marginal investment of projects becoming more
profitable, thus the total demand for capital in the
economy would increase. However, Keynes argued that
the supply of capital would be unable to meet these
increases in demand caused by changes in the interest
rate, and from this he constructed an investment
marginal efficiency curve that was inversely connected
with the market interest rate.
Changes in monetary policy, that caused interest rate to
change, would affect the investment volume in the
economy, Keynes argued. Thus, he emphasised that interest
rate is the main transmission mechanism between monetary
policy and investment. However, his theory did not ignore
the importance of other factors, such as prices and wages, in
determining the demand for investment.
The naive accelerator model, elaborated by Clark
(1917), is based on the assumption of a fixed capital-
output ratio. This implies that prices, wages, interest
rates, and taxes may have indirect impacts on the capital
stock. The naive accelerator model defines the optimal
capital stock (K*t) as a constant proportion of output (Yt):
where µ represents the capital-output ratio.
Furthermore, while the capital stock is always optimally
adjusted in each period (i.e. K*t = Kt) net investment
(NIt) will be equal to:
The flexible accelerator model, derived by Chenery
(1952) and Koyck (1954), which is a more general form
of the naive accelerator models, is based on the gap
between the existing capital stock level (Kt-1) and the
desired capital stock (K*t). is a constant proportion
and lies (0< <1), then net investment is:
Output, funds, costs and other variables may be
included in the model as determinants of the desired
capital stock, proportional to output.
Neoclassical Model
The neoclassical approach has gained importance in the
investment literature following the work of Jorgenson, who
presented a ‘neoclassical theory of optimal accumulation of
capital’. The main assumptions behind the neoclassical
approach are that there is perfect certainty. In other words,
all agents in the market perform with the same certain
expectations about the future, and there is a perfect capital
market, though all agents in the capital market are price
takers (Jorgenson, 1971).
The main feature of the neoclassical theory of
investment is that it is based on a model of optimisation,
in which the desired capital stock is determined by
interest rates, output, capital prices and tax policies.
Assuming a firm produces output, Y, by using two
inputs, K and L. Jorgenson’s approach modelled the
firm’s net worth maximisation objective as its optimal
objective, which equals the sum of the discounted value
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-83-
of the cash flow of profits from time zero, subject to a
neoclassical production function: .
Then the optimisation problem is:
, and is is the interest rate at
time s, It is gross investment at time t, the output is sold at
price pt and the inputs are bought at the prices wt and qt
respectively. Under a perfectly competitive market, the firm
is a price taker, therefore the firm has to choose Lt, Kt and It
to maximise the discounted value of cash flow of the firm.
The Lagrangian multiplier yields the conditions for
solving the optimisation problem, for capital and labour:
where MPK,t and MPL,t represented the marginal
products of K and L, respectively.
Under the theoretical conditions for profit
maximisation, firms will employ a set of inputs; for each
input the marginal product of employing another unit of
the input should equal the marginal cost of employing
that additional input, thus the additional real user-cost of
capital or real wage.
The user-cost of capital, rt, represents the true cost to the
firm of holding its assets in the form of capital stock. It is
composed of three elements. The first is the cost taking place
from the interest rate foregone from not investing elsewhere,
the opportunity cost. The second element is a cost caused by
the depreciation of the capital stock. The third is the possible
capital gain (loss) occurred as a result of a change in the
market value of the capital over the period.
Tobin’s q Investment Theory
Tobin (1969) generalised a cash-flow model and
presented a framework for an investment model where
net investment depends on the ratio of the market value
of additional capital stock to its replacement costs. This
ratio is known as marginal q. The naive form of the
Tobin’s q model is specified by:
It implies that whenever marginal q deviates from
unity, this indicates that there are incentives to
investment or disinvestment the capital by the firm. The
marginal valuation approach is therefore basically a
predictor of investment, the underlying principle of this
theory is to provide a link between monetary policy and
the real side of the economy. For this purpose, the
market interest rate, which measures the opportunity cost
of investment, is seen as a crucial determinant of
investment, without, of course, ignoring the importance
of other factors.
4. Data and Methodology
4.1. Data
This section analyses the empirical effects of the
economic fundamentals on aggregate investment in
Jordan, using annual observations over the period 1976-
2010. The objective here is to investigate whether a
stable investment function exists, with concern on the
role played by the trade openness in such functions. The
variables included are the natural logarithm of real
investment, LRIt; the natural logarithm of real Gross
Domestic Product (GDP), LRYt; the Natural Logarithm
of Real Credit, LRCRt; the user cost of capital, UCCt;
and natural logarithm of trade-to-GDP ratio, LOpent.
Real investment is approximated by using gross fixed
capital formation in constant prices. Real GDP is
calculated as nominal GDP deflated by the GDP
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-84-
deflator. A positive growth rate represents an increase in
aggregate demand, which in turn increases investment,
and then we expect a positive sign with investment.
Real credit is measured by the domestic credit
provided to the private sector deflated by the GDP
deflator. An increase in credit increases investment, and
then we expect a positive sign with investment.
The user cost of capital can be constructed as follows:
where is the price of capital measured by the gross
capital formation deflator, it is the average of interest rate
on loans and advances, is the depreciation rate, and Pt is
the GDP deflator. We assume a negative relation between
user cost of capital and investment.
Trade liberalisation or the degree of economy
openness, represented as trade-to-GDP ratio, is
calculated as the sum of exports and imports divided by
GDP. An increase in trade liberalisation expressed by
decreasing barriers gives incentives for investment and
may affect their expectations about the new markets
available. We expect a positive sign with investment.
Most of the data are obtained directly from the World
Development Indicators of the World Bank and
International Financial Statistic of the IMF.
Figure (1) represents plots of the variables. Table (1)
shows descriptive statistics and Table (2) summaries the
expected signs and the theoretical reference.
Gross fixed capital formation in real term had a
positive trend during 1976-1981, from JD 512.8 millions
in 1976 to JD 1328.0 millions in 1981 or an annual
average growth of 21.7 percent, which is the case for
GDP and credit, these upward being the result of the
expansion that occurred in the economy. The increasing
demand for Jordanian products came from the
neighbouring oil exporting countries, which were
affected by the oil price shock in the late 1970s.
Investment started to decrease sharply, about 14.6
percent, because of the recession in the economy until
the mid-1980s, followed by an increase until 1989. Then
it decreased again reflecting the increase in the price
level in the late 1980s. During 1991-1994 investment
started to follow a positive trend, from JD 755.8 millions
in 1991 to JD 1391.6 millions in 1994 or on annual
average growth of 23.8 percent, reflecting the
improvements in the economy during the second reform
programme, which started in 1992. These improvements
did not last long because of the recession in the mid
1990’s. During the 2000s, investment followed an
upward trend reflecting the openness that occurred in the
economy after Jordan became a member in the World
Trade Organisation (WTO) in 2000. Investment
increased from JD 1059.6 millions in 2001 to JD 2140.6
millions in 2008, an annual average of 11.2 percent.
In addition, the figure shows the general trend in
trade openness indicator for Jordan over the period
1976-2010. It can be clearly seen that Jordan had
experienced a spectacular increase in the volume of both
exports and imports over that period. This led to increase
the level of international trade transactions with foreign
world due to the expansion of establishing international
trade agreements with various partners especially during
the last two decades. It is also noticed that trade
openness indicator in Jordan is fluctuating over time and
influenced by various exogenous factors.
During the oil boom in the Gulf Countries over the
1970s, trade openness indicator had experienced a clear
increase. However, over the first half of 1980s, this
indicator had dropped gradually as a result of the rapid
growth in the Jordanian GDP compared to the growth in
exports and imports during that period. After that,
Jordanian exports and imports had experienced rapid
growth rates, particularly over the years 1987 and 1988.
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-85-
This increase had continued steadily over the period
1989-1990.
The consequences of the 1989-economic crisis had
badly affected the level of trade openness in Jordan
during the period 1991-1999, and caused trade openness
to decrease in Jordan during that period. After that, trade
openness indicator had experienced a tremendous
increase. This is arising from implementing different
trade agreements, where many of these agreements came
into enforce in 2001, 2002, and after, such as Jordan-US
FTA and Jordan- EU Association Agreements and other
trade agreements.
The global financial crisis and the US recession,
during the year 2008 and after, caused the Jordanian
international trade level to decreased, affecting the level
of trade openness in the years 2008 and 2009. However,
both exports and imports have gradually increased
compared to the GDP, and as a results trade openness
indicator improved again in the year 2010 and after.
Figure 1. Data plots (1976-2010)
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-86-
Table 1. Descriptive statistics
LRI
LRY
LRCR
UCC
LOPEN
Mean
7.009
8.366
-0.476
45.288
4.796
Standard Deviation
0.330
0.470
0.261
12.822
0.126
Sample Variance
0.109
0.221
0.068
164.401
0.016
Count
35
35
35
33
35
Table 2. Expected signs of the model
Expected sign
Theoretical reference
LRY
Positive
Anoruo et al. (2007), Heim (2008)
LRCR
Positive
Shafik (1992)
UCC
Negative
du Toit and Moolman (2004), Anoruo et al. (2007)
LOPEN
Positive
Baldwin & Seghezza (1996), Harrison (1996), Salahuddin and Islam (2008)
4.2. Methodology
Following the literature on investment, a general model
is considered to formulate empirically the investment
function in Jordan, which could be expressed as:
Using the technique of 'General-to-Specific' it will
estimate a number of models and then choose the most
accurate one, which represented the investment function
in Jordan.
These models are estimated depending on an
Autoregressive Distributed Lag (ARDL) approach by
Pesaran and Shin (1999), where this procedure allows us to
apply the model regardless of the stationarity of the
variables. The results of this approach are equivalent to the
results of the Error-Correction Models (ECM) (Hassler and
Wolters, 2006). ARDL is adopted for a mixture of
stationary and non-stationary variables, the advantage of
ARDL over the ECM is that it can be applied irrespective
of whether the regressors are I(0) or I(1).
The autoregressive distributed lag (ARDL) approach,
developed by Pesaran and Shin (1999) and Pesaran et al.
(2001), is also known as the ARDL bounds test. The
ARDL approach has numerous advantages which make
it preferable over other methods in estimating the long-
run co-integration relationships. The main advantage is
that it is not necessary for testing the unit root of the
variables, where the ARDL can be applied irrespective
whether regressors are I(0) or I(1).
Following Pesaran and Shin (1997) and Pesaran et al.
(2001), a general ARDL (p,q) model can be presented as
follows:
where yt represents the dependent variable, xt is a
vector of explanatory variables and , are
uncorrelated error terms with zero mean and constant
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-87-
variance.
The model can be rewritten as:
Where
, and L is
the lag operator. For simplicity define:
Then yt can be expressed as:
Where
, and represents the
error term.
The long-run co-integrating vector can be expressed
as:
Using the lag and first differences of y and x, we
obtain:
Where is a deterministic variable, j is the number
of explanatory variables, r is the number of lags selected
based on the information criteria, and is a white noise
disturbances.
The implementation of this technique involves two
stages. First one, test for the existence of co-integration
relationship among yt and xjt variables by the bounds
test, using a Wald-test (F-test). The test null hypothesis
that there is no co-integration relationship among the
variables, and can be conducted as a joint significance
test on lagged level variable’s coefficients as follows:
The computed Wald test gives two sets of critical
values, bounds, one set based on the assumption that all
variables in the ARDL model are I(1), and the other set
assumes that all variables are I(0). If the calculated F is
higher than the upper critical bound, then the null
hypothesis is rejected, therefore a co-integration
relationship between the variables exists. If the test statistic
is below the lower critical bound, then the null hypothesis
cannot be rejected. And if the calculated F-test is between
the bounds, then the test cannot give a conclusive inference.
In the second stage, if the long-run relationship
exists, then the long-run and short-run coefficients of the
equation (18) can be estimated.
The investment function can be presented as follows:
Where ∆ denotes the first difference, and p=1,2 is the
number of lags determined by information criteria, and k is
the number of independent variables, and DEPt represented
a vector of the explanatory variables of the real investment.
5. Empirical Results
5.1. Unit Root Tests
According to figure (1), most variables have a trend
over the sample period. Therefore, a constant and trend
have been included in the unit root test.
A visual inspections of the data confirmed that all
variables were I(1), except for LRCR and LOpen which
are I(0). The Augmented Dickey-Fuller (ADF), Phillips-
Perron (PP) and Kwiatkowiski-Phillips-Schmidt-Shin
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-88-
(KPSS) unit root tests confirmed the stationary
hypothesis for the first difference. Table (3) summarises
unit root tests results.
Table 3. Unit root tests results
ADF
PP
KPSS
Level
1st Difference
Level
1st Difference
Level
1st Difference
LRI
-2.906
-4.525*
-2.906
-4.456*
0.061
LRY
-3.097
-3.666**
-2.403
-4.089**
0.093
LRCR
-4.289**
-1.962
-6.672*
0.242
0.075
LOpen
-3.309***
-2.667
-3.932**
0.050
UCC
-2.486
-5.804*
-2.598
-5.805*
0.075
-ADF, Augmented Dickey-Fuller; PP, Phillips-Perron; KPSS, Kwiatkowiski-Phillips-Schmidt-Shin. For ADF Schwarz information
criterion used to select the lag length and the maximum number of lags was set to be 8. For PP and KPSS Barlett-Kernel was used as
the spectral estimation method and Newey-West used to select the bandwidth.
-ADF & PP critical values: 1% -4.263, 5% -3.558, 10% -3.212, KPSS critical values: 1% 0.216, 5% 0.146, 10% 0.119.
-*Significant at 1%, **significant at 5%, and ***significant at 10%.
For the level variables, under ADF and PP the null
hypothesis of a unit root cannot be rejected at the 5%
significance level, except for LRCR and LOpen which can be
rejected at level 5% and 10%. However, the null that the first
difference of the variables has unit root is rejected at the 5%
level. While according to the KPSS test, the null hypothesis
of stationarity can be rejected at level 5% significant, except
for LRCR which cannot be rejected at level 5%. However,
the null that the first difference of the variables is stationary
cannot be rejected at the 5% significance level.
5.2. Co-integration Tests
In order to check the existence of a co-integration
relationship among the variables, the bounds test,
Pesaran et al. (2001), is implemented, which is based on
testing the null hypothesis of no co-integration
relationship among the variables. The test uses the F-
statistic depend on Wald test on equation (19):
Where the null and the alternative hypotheses are
constructed as follows:
H1: At least one is not zero
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-89-
Table 4. Bound test results
F statistic
Critical values*
Sig level
Lower bound
Upper bound
3.87
5%
2.69
3.83
10%
2.38
3.45
* The critical values are obtained from Pesaran et al. (2001), table CI(v).
From Table (4), the calculated F-statistics for all
models are exceeding the upper critical bound at the 5%
level of significance. Thus the null hypothesis of no co-
integration can be rejected, so only one co-integration
relationship exists in these models.
The long-run relationship can be estimated as:
R2= 0.883, RSS= 0.322, F(Prob.)= 70.74 (0.00)
The above estimation represents the long-run
relationship between investment and its determinants.
Real investment depends positively on real income and
real credit, and negatively on the user cost of capital. All
signs and coefficients magnitude seem to be consistent
with the economic theory. One can notice that the
coefficient of trade openness is not shown in the long-
run relationship as it has not a significant effect.
5.3. Estimation Results
Table 5. Short run results of investment equations
Dependent variable: ∆LRI
Model 1
Model 2
LRIt-1
-0.365
(0.099)
-0.379
(0.092)
∆3LRY t
0.715
(0.176)
0.844
(0.171)
∆LRCRt-1
0.485
(0.314)
∆∆LRCR t
0.771
(0.281)
0.409
(0.167)
∆LUCC t
-0.759
(0.106)
-0.598
(0.114)
∆LOpen t
0.394
(0.151)
LRCR t-1
0.692
(0.215)
0.641
(0.190)
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-90-
Dependent variable: ∆LRI
Model 1
Model 2
LUCCt-1
-0.260
(0.102)
-0.228
(0.095)
Constant
3.777
(1.040)
3.714
(0.958)
SE
0.155
0.132
R2
0.840
0.864
2
0.791
0.882
F-statistic
17.21
20.81
Prob(F-statistic)
0.000
0.000
Table (5) represents the results of the ARDL
estimation of the short run for two suggested models
(just significant variables included which driven from a
general model). The results of these models appear to be
similar. Nonetheless, the explanatory variable added in
the model 2, trade-to-GDP ratio (∆Lopent), increases the
explanatory power; as adjusted R2 increased from 79
percent to 88 percent.
The coefficient of LRIt-1, that measures the speed of
adjustment, appears to be negative and less than 1 in
magnitude, about -37 and -38 percent, and they are
statistically significant at 1% level. Thus, it seems that
the model takes about three years to be adjusted.
The real income and real credit have a positive effect
on real investment. The real income does not vary
among the models, and it is less than unity; however the
real credit has also a stable effect among the models.
A more trade
90
iberalization, positively affect the
real investment, about 40 percent. In other words, any 1
percent increase in the trade-to-GDP ratio, increases the
real investment by about 40 percent. This result confirms
the finding at the previous empirical literature.
The log of the user cost of capital lagged one period
appears to be consistent with the theory in both models;
it has a magnitude between -23 percent and -26 percent.
Thus any increase in interest rate by 1 percent, may
decrease investment by about 23 and 26 percent.
The goodness of fit of these models is relatively high, and
the overall models are significant. The regression
specifications fit well and pass all diagnostic tests against
serial correlation, autoregressive conditional
heterosedasticity, non-normal residual, heterosedasticity, and
incorrect functional form. Table (6) reports the diagnostic
tests. Figure (2) shows residuals, actual, and fitted lines.
Table 6. Diagnostic tests
Model (1)
Model (2)
ARCH 1-1 test
F(1,21)= 0.342 [0.714]
F(1,21)= 0.120 [0.668]
Normality test
χ2 (2)= 0.167 [0.920]
χ2 (2)= 3.283 [0.194]
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-91-
Model (1)
Model (2)
Hetero. Test
F(14,8)= 0.275 [0.983]
F(14,8)= 0.356 [0.956]
RESET test
F(1,22)= 1.071 [0.311]
F(1,22)= 0.278 [0.604]
Model (1)
Model (2)
Figure 2. Actual, Fitted and Residuals lines
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-92-
6. Conclusion
As investment theories emphasise the importance of
investment for economic growth, development, and
economic stability, this research looks at the potential
determinants of investment for Jordan, trying to formulate an
investment function and attempting to capture the role played
by trade openness in the real economy. It employs a number
of variables depending on the theory in an attempt to capture
their effect on investment in Jordan covering the time period
1976 to 2010, using the Autoregressive Distributed Lag
(ARDL) model, by Pesaran and Shin (1999), which is the
procedure used when some of the variables, being used in the
analysis, are I(1) and others are I(0). In order to check for the
existence of a long run relationship among the variables
exists in the model, the bounds test is implemented, and it is
found that there is strong evidence that at least one co-
integration relationship exists.
The OLS estimation is used to estimate the long-run
relationship between investment and its determinants
including trade openness. In the long-run, real
investment depends positively on real income and real
credit, and negatively on the user cost of capital.
The speed of adjustment appears to be negative and
less than 1, about -37 percent and -38 percent for both
models, respectively, and they are statistically significant
at 1% level. This speed of adjustment coefficients is
considered to have about three years to adjust.
The coefficient of the trade-to-GDP ratio, that
represents the trade openness, is found positive and
significant; a consistent result with the literature. This
finding emphasise the importance of the trade
liberalisation effects on the real investment in Jordan.
Trade openness has a relatively important role in
stimulating capital formation and increasing the level of
investment in the case of a small economy as in Jordan.
Giving the positive impact of trade openness on the level
of investment in Jordan, policies should be aimed at
enhancing trade liberalisation and focusing on
implementing bilateral trade agreement with the foreign
world in order to increase investment, where
cointegration analysis employed in this paper shows that
trade liberalisation is accompanied by a large change in
the composition of investment in the case of Jordan.
It is also shown that a strong cointegration
relationship exists among trade openness and real
investment. This fact highlights the broadly ignored
effects of trade liberalisation when formulating the
determinants of investment equation in many countries.
Therefore, international trade variables should be
considered when investigating the behaviour of
investment especially in developing countries.
Acknowledgement
This research was conducted under the support of the
WTO-Chair Programme at the University of Jordan.
Also I would like to thank the referees for their valuable
comments which improve the work.
REFERENCES
Anderson, G. J. 1981, A New Approach to the Empirical
Investigation of Investment Expenditures, the Economic
Journal, Vol. 91, No. 361, 88-103.
Ang, James B. 2009, Private Investment and Financial Sector
Policies in India and Malaysia, World Development, Vol.
37, No. 7, 1261–1273.
Anoruo, Emmanuel, Saten Kumar, and Bill DiPietro 2007, A
Co-integration Analysis of Investment Output Ratio in
Bangladesh, MPRA, working paper, No. 19338.
Aysan, Ahmet, Mustapha Nabli, and Marie-Ange Veganzones-
Varoudakis 2007, Governance Institution and Private
Investment: Application to the Middle East and North
Africa, The Developing Economies, XLV-3, 339–377.
Baldwin, R. E., and E. Seghezza 1996, Testing for trade-
Jordan Journal of Economic Sciences, Volume 1, No.1 2014
-93-
induced, investment-led growth, NBER Working Paper
No. W5416.
Bean, C. R. 1981, An Econometric Model of Manufacturing
Investment in the UK, the Economic Journal, Vol. 91,
No. 361, 106-121.
Chenery, H.B. 1952, Overcapacity and the Acceleration
Principle, Econometrica, 20
Bleaney, M. F. 1996, Macroeconomic stability, investment and
growth in developing countries, Journal of Development
Economics, 48(2), 461–477.
Clark, Maurice J. 1917, Business acceleration and the law of
demand: A technical factor in economic cycle, Journal of
political Economy, 25, 217–235.
du Toit, Charlotte, and Elna Moolman 2004, A neoclassical
investment function of the South African economy,
Economic Modelling, vol. 21, 647–660.
Harrison, A. E. 1996, Openness and growth: A time-series,
cross-country analysis for developing countries, Journal of
Development Economics, 48(2), 419–447.
Heim, John J. 2008, The Investment Function: Determinants of
Demand for Investment Goods, Rensselaer, working
papers in economics, NY.
Ismihan, Mustafa, Kivilcim Metin-Ozcan, and Aysit Tansel,
2005, The Role of Macroeconomic Instability in Public
and Private Capital Accumulation and Growth: the Case of
Turkey 1963–1999, Applied Economics, vol. 37, 239–251.
Johgwanich, Juthathip and Archanun Kopaiboon, 2008,
Private Investment: Trends and Determinants in Thailand,
World Development, Vol. 36, No. 10, 1709–1724.
Jorgenson, D. 1971, Econometric Studies of Investment
Behaviour: A Survey, Journal of Economic Literature,
91.
Keynes, J.M. 1936, The General Theory of Employment,
Interest and Money, London, UK: Macmillan.
Koyck, L.R. 1954, Distributed Lags and Investment Analysis,
Amsterdam, North Holland Publishing Company.
Moore, Tomoe 2010, A Critical Appraisal of McKinnon’s
Complementarity Hypothesis: Does the Real Rate of
Return on Money Matter for Investment in Developing
Countries?, World Development, Vol. 38, No. 3, 260–269.
Nazmi, Nader, and Miguel Ramirez 1997, Public and Private
Investment and Economic Growth in Mexico,
Contemporary Economic Policy, vol. 15, 65-75.
Oladipo, O.S., 2011, Does trade liberalization cause long run
economic growth in Mexico? An empirical investigation,
International Journal of Economics and Finance, vol. 3,
63–74.
Pesaran, M.H., Shin, Y. 1999, An Autoregressive Distributed
Lag Modelling Approach to Cointegration Analysis, In
Econometrics and Economic Theory in the 20th Century,
The Ragnar Frisch Centennial Symposium (S. Strom,
ed.), chapter 11, Cambridge University Press, Cambridge,
371-413.
Pesaran, M.H., Shin, Y. and Smith, R. 2001, Bounds testing
approaches to the analysis of level relationships, Journal
of applied econometrics, 16(3), 289-326.
Shafik, Nemat 1992, Modelling Private Investment in Egypt,
Journal of developing economics, vol. 39, 263-277.
Salahuddin, M., and M. R. Islam 2008, Factors affecting
investment in developing countries: A panel data study,
Journal of Developing Area, vol. 42, no. 1, 21–37.
Sun, Laixiang 1998, Estimating Investment Functions Based
on Co-integration: The Case of China, Journal of
Comparative Economics, vol. 26, 175–191.
Tobin, James, 1969, A General Equilibrium Approach to
Monetary Theory, Journal of Money, Credit and
Banking, 1, 15-29.
Voth, Hans-Joachim 1995, Did High Wages or High Interest Rates
Bring Down the Weimar Republic? A Co-integration Model of
Investment in Germany, 1925-1930, the Journal of Economic
History, Vol. 55, No. 4, 801-821.
Trade Openness and… Alaaeddin A. Al-Tarawneh and Ghazi I. Al-Assaf
-94-
ARDL
1
1
ARDL67910262
ARDL
6
a.altarawneh@ju.edu.jo
g.alassaf@ju.edu.jo 66/3/0263 1/62/0263
