Content uploaded by Hussein Salameh
Author content
All content in this area was uploaded by Hussein Salameh on Apr 11, 2020
Content may be subject to copyright.
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
5
ABSTRACT
The Kingdom of Saudi Arabia vision for 2030 is to replace foreign laborers with local ones to reduce Workers’
Remittances Outflows (WOR), which weaken economic growth in the short term but has no effect in the long term
according to previous literature review. The authors investigated the impact of WOR on the economy by employing
the Bai-Perron test and OLS regression. The results reveal that the structural breaks are in 1981 and 1987.
Furthermore, the results find that the impact of WOR and Government Expenditure (GE) are negatively significant,
while Inflation (INF) and Exports (EX) are positively significant on the economy. Moreover, the explanatory
power of the model is 87.3%. The authors conclude that rather than replacing foreign laborers with local ones, the
foreign laborers should be encouraged to invest locally. In addition, more procedures should be taken to encourage
exports. Moreover, government expenditure result’s is consistent with neoclassic economists’ explanation.
Finally, the outcome of exports and expenditure will increase inflation and cause its impact on economy. We
recommend that further research be conducted, using quarterly or monthly data, to pave the way for applying
ARDL methodology.
JEL Classification: C10; C40; E0; J60.
Keywords: Remittance; Worker; Outflow; Saudi Arabia.
*Corresponding author.
1. INTRODUCTION
The Kingdom of Saudi Arabia (KSA), The United States of America, (USA) and Russia are the countries with the
highest remittances outflows in the world. Research on remittance flow and its effect on economy have grown in
the last decades but most of them focused on remittance inflows and their effects on the receiving economies (their
impact are mixed). Few research studies have focused on the effect of remittances outflow on the economy; thus
this topic has become a hot one with the earliest study only conducted in 2013. The impact of remittance outflows
on the economy in Gulf Cooperation Council (GCC) countries weaken economic growth in the short term but has
no effect in the long term, Moreover, remittance outflow reduces local inflation (Edrees 2016).
This study will argue that the projected Balance of Payment in KSA indicates that cash inflow from oil exports
will decrease in the coming years. Accordingly, KSA government vision of 2030 recommends replacing the
foreign labor force with a local one and recommends improving the quality of education so the labor market can
be self-satisfied by its own citizens. Reducing the size of the foreign labor force will reduce the weakness in
economic growth in the short run while keeping the long growth as is. This study addresses the ignorance in the
literature on the impact of the outflow of remittances on remitting countries’ GDP such as KSA’s GDP, and the
positive and negative effects on the economy of the outflow of remittances.
The main purpose of this paper is to shed new light on the impact of remittance outflow on the KSA’s economy.
The present author’s research makes several major contributions to the literature. First, it is a continuation of
previous research on a hot topic for one of the countries with the highest remittances outflow, the KSA. A second
contribution is due to the stationary at the second difference for the dependent variable; the authors could not apply
Autoregressive Disturbed Lags ARD, but implement the Bai-Perron test to find the structural breaks and add two
intercept dummies for the two breaks. Finally, the authors apply OLS regression to find the long run relationship
Is Outflow of Workers’ Remittances Affecting the Kingdom of Saudi
Arabia’s Economy in The Long Run?
Abdulaziz ALDAARMI
Faculty of Business Administration, University of Tabuk, Saudi Arabia.
Hussein SALAMEH*
College of Administrative and Financial Sciences, King Khalid University, Saudi
Arabia. Email: salameh2272017@gmail.com.
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
6
between the variables. A third contribution is to find the impact of the other control variables (government
expenditure, inflation and exports) on the dependent variable, GDP.
This paper will be organized in the following manner: Section two summarizes the major previous studies that
investigate asset pricing models. Section three focuses on the methodology and data used in the present research.
Results and Discussion follow in section four. Finally, conclusions are presented in section five.
2. LITERATURE REVIEW
The effect the outflow of remittances has on real GDP has received considerable attention in the recent literature.
Many attempts have been made to find the control determinants of the real GDP. A survey of previous studies that
have been done on this issue is presented in this section.
In Saudi Arabia Alkhathlan (2013) showed that the effect of WOR on economic growth is negative and statistically
insignificant in the long term, while it is negative and statistically significant in the short term. Therefore, his
recommendation for decreasing WOR is to convince foreign workers to consume and invest their money in the
KSA. Moreover, he showed that the effect of GE and EX on economic growth is positive and statistically
significant. From their side, Haddad & Choukir (2015) suggested that time-varying impulse responses of non-oil
GDP depend on the magnitude of structural volatilities of remittance outflows. Therefore, monetary policies must
consider this result with investment, current account balance (CAB), and remittance outflows. Finally, the results
recommend offering monetary incentives to encourage keeping foreign workers’ earnings inside the host country
to promote investment, and to enhance current account surplus. Moreover, in 2017, they found that there is one
co-integrating vector and the common trends model is identified to three permanent shocks, interpreted remittance
shock (real external shock), non-oil GDP (domestic real shock), (the inflation domestic nominal shock) and
investment (transitory shock). The impulse response functions showed that the remittances shocks have short-run
negative effects on non-oil GDP, negative long-run effects on inflation, and shorter positive response on
investment while negatively on the long run. Therefore, to reduce the negative effect of the outflow of remittances
on non-oil GDP, inflation, and investment, more incentives are required to encourage foreign workers to invest in
the KSA.
Rahmouni & Debbiche (2017) found that both in the long and the short term, the outflow of remittances has no
significant effect on GDP which could be attributed to its decreasing percentage relative to GDP. Abdel-Rahman
(2017) showed that the effect of per capita GDP on the levels of per-worker remittances is positive indicating that
remittances from the KSA are pro-cyclical, increasing during booms and declining during recessions. The authors
found that wages affect remittances per worker positively and that differential return has an inverse relationship
with the dependent variable. They further found that political, economic, and financial risks’ variables measure
the degree of government stability and that the degree of law and order has a significant impact on remittances.
Finally, the effect of socio-political stability is negatively significant on per worker remittances when using
composite risk variable.
Tackling the effect of remittances outflow on GDP in the GCC countries, Edrees (2016) revealed that foreign
workers positively affect economic growth. On the other hand, outflow remittances negatively impact the
economic growth of the United Arab Emirates, the KSA, and Qatar. Kaabi (2016) tested the effects of remittance
outflows and inflation on GDP in the GCC countries and his results showed that remittance outflows affect growth
in GDP adversely only in the KSA. In addition, he showed that growth in remittance outflows affects inflation
only in Bahrain. Naufal & Genc (2014) showed that the Gulf region’s remittance outflows are 50% larger than the
amounts remitted from the United States. Moreover, they found that the Gulf region’s remittances outflows are
linked to local labor policies because it decided the source of foreign labor (cheap foreign labor and strong
exchange power of local currency). Finally, they explore the potential role of remittance outflows on the local
economies and on the foreign workers’ countries.
Investigating this issue in Germany, Baasa & Melzer (2012) showed that stronger remittances outflows depreciate
the real exchange rate and give incentives to reallocate resources towards the tradable goods sectors. This translates
into a converse Dutch disease phenomenon.
Tackling this issue in Russia, Naufal & Genc (2017) concluded that even though there is large size of remittance
outflow, this outflow is a small proportion of the Russian Federation’s economy, a fact that should encourage
investments in the local economy by foreign labor. Accordingly, this could perhaps necessitate legal arrangements
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
7
(permanent residence), tax benefits, and investment incentives (targeting high skilled workers). Finally, the
proportion of remittance outflows is expected to grow.
3. METHODOLOGY
3.1 Data Selection and Data Collection
In our data base, the aim was to construct data collected from World Bank for the dependent variable, Gross
Domestic Product (GDP), and the other control variables (government expenditure, Exports and Inflation), Data
for the main independent variable (outflow of workers' remittances) is collected from Thomson Data Stream.
Yearly time series data were collected from 1972 to 2016. After assessing the availability and quality of data (all
variables were in US dollars) and the most important periods for which the data were available, we arrived at 45
observations for each of the variables of the model.
3.2 The Model
Our paper aims to track the effect of outflow of workers' remittances and other control factors on the real GDP in
the KSA. We based our model on theory and previous literature; as such, we used a model that derived GDP
behavior. Accordingly, a functional form is used. We intended to analyze how the outflow of workers' remittances
as an independent variable and how the other control variables affect the GDP in the KSA. These other control
variables consist of government expenditure, inflation, exports, imports, ratio of the sum of imports and exports to
GD, total enrollment in tertiary education, regardless of age, expressed as a percentage of the total population,
previous year’s investment and the size of foreign workers. Due to a lack of availability of data, we exclude
imports, ratio of the sum of imports and exports to GD, total enrollment in tertiary education, regardless of age,
expressed as a percentage of the total population, previous year’s investment and foreign workers. To apply the
model in the current study, the equation based on modern time series data can be specified as follows:
GDPt = α + β1WORt + β2 GEt + β3 INFt + + β4EXPt
3.2.1 Definition and Measurements of Terms in the Models
(GDP) denotes the Gross Domestic Product (current US $): measured as natural logarithm of GDPt (Gross
Domestic Product this year) minus natural logarithm of GDPt-1 (Gross Domestic Product previous year)
GDP = LnGDPt –LnGDPt-1
(WOR) denotes the outflow of workers' remittances (current US $): According to Alkhathlan, the relationship
between outflows of workers' remittances and economic growth is a negative but statistically insignificant
relationship in the long term while it is negative and statistically significant in the short run. This variable measured
as natural logarithm of (outflows of workers’ remittances divided by Gross Domestic Product) this year minus
natural logarithm of (outflows of workers’ remittances divided by Gross Domestic Product) previous year
WOR = Ln(WOR/GDP)t - Ln(WOR/GDP)t-1
(GE) denotes General government final consumption expenditure (current US $): According to Alkhathlan,
Naufal & Genc, the relationship between government expenditure and economic growth is positive and statistically
significant in the long term. Moreover, the effect of this variable on GDP is equivocal. In fact, while Keynesians
defend its positive effect, neoclassic economists consider that it has a harmful effect in the LT growth and Neo
Ricardian approach pleads for its neutrality Rahmouni & Debbiche (2017). This variable is measured as natural
logarithm of (government expenditure divided by Gross Domestic Product) this year minus natural logarithm of
(government expenditure divided by Gross Domestic Product) previous year
GE = Ln(GE/GDP)t - Ln(GE/GDP)t-1
(INF) denotes inflation: According to Alkhathlan measured by the growth rate of the consumer price index (CPI),
the relationship between inflation and economic growth is positive and statistically significant in the long term.
This variable is measured as natural logarithm of (Consumer Price Index divided by Gross Domestic Product) this
year minus natural logarithm of (Consumer Price Index divided by Gross Domestic Product) previous year
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
8
INF = Ln(INF/GDP)t - Ln(INF/GDP)t-1
(EXP) denotes exports: According to Alkhathlan the relationship between exports and economic growth is positive
and statistically significant in the long term. This variable is measured as natural logarithm of (Exports divided by
Gross Domestic Product) this year minus natural logarithm of (Exports divided by Gross Domestic Product)
previous year
EX = Ln(EX/GDP)t - Ln(EX/GDP)t-1
3.3 Econometrics & Statistical Techniques (Eviews, SPSS)
3.3.1 Unit Root Test
Before conducting estimation and in order to avoid possible spurious regression, it is necessary to distinguish
stationary from non-stationary variables. The first step undertaken would be to establish the order of integration
of variables used in the model. This is accomplished by applying firstly the Augmented Dickey-Fuller (ADF) test
on each of the series in the estimated equations, standard unit root tests. The well-known ADF test for a unit root
in yt, omitting a linear deterministic trend is:
∆yαβy
δ∆y ε
Where ∆ is the difference operator, ε is a white noise disturbance term with variance σ2, and t= 1, …, T indexes
time. The∆yt-i terms on the right-hand side of equation allow for serial correlation and are designed to ensure thatε
is white noise. The empirical evidence suggests that there is no time trend in the data. The ADF (parametric test)
test and PP (non parametric test) has a null hypothesis of non-stationarity (random series) against an alternative of
stationarity (non random series).
3.3.2 Bai-Perron Structural Breaks Test
The scholars consider a standard multiple linear regression model with T periods and m potential breaks (producing
m+1 regimes). For the observations Tj, Tj+1, …., Tj+1 -1 in regime j we have the regression model
yt = Xt’ β + Zt’δj + εt ………..( 1 )
for the regimes j= 0, ……, m. Note that the regressors are divided into two groups. The X variables are those whose
parameters do not vary across regimes, while the Z variables have coefficients that are regime-specific.
While it is slightly more convenient to define break dates to be the last date of a regime, we follow EViews’s
convention in defining the break date to be the first date of the subsequent regime. We tie down the endpoints by
setting T0 = 1 and Tm+1 = T +1.
Once the number and identity of the breakpoints is determined, the model may be estimated using standard
regression techniques. We may rewrite the equation specification as a standard regression equation Zt’δj
yt = Xt’ β + Zt’
δȷ
+ εt …………( 2 )
with fixed parameter vectors β and where δ
= ( δ0’ , δ1’, ……., δm’ ) where Zt’
is an expanded set of regressors
interacted with the set of dummy variables corresponding to each of the m+1 regime segments.
The breakpoints may be known a priori or they be estimated using a variety of approaches. The breakpoint
estimation methods that we consider may broadly be divided into two categories: global maximizers for the
breakpoints and sequentially determined breakpoints.
Global Maximization
Bai and Perron (1998) describe global optimization procedures for identifying the m multiple breaks and associated
coefficients which minimize the sums-of-squared residuals of the regression model Equation (1).
If the desired number of breakpoints is known, the global m-break optimizers are the set of breakpoints and
corresponding coefficient estimates that minimize the sum-of-squares for that model.
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
9
If the desired number of breakpoints is not known, we may specify a maximum number of breakpoints and employ
testing to determine the “optimal” number of breakpoints. The various test approaches are outlined in detail in
“Global Maximizer Tests”, but briefly speaking, involve:
Global tests of l breaks versus none (Bai-Perron 1998). The test of l versus no breaks procedure may be applied
sequentially beginning with a single break until the null is not rejected. Alternately, it may be applied to all breaks
with the selected break being the highest statistically significant number of breaks, or it may employ the
unweighted or weighted double maximum statistics (UDmax or WDmax).
Information criteria based model selection of the number of breaks (Yao 1988; Liu, Wi, and Zidek 1997), where
we minimize the specified information criteria with respect to the number of breaks
Sequential tests of l+1versus globally determined breakpoints. The procedure is applied sequentially, beginning
with a single break, until the null is not rejected. This approach is a modified Bai (1997) method in which, at each
test step, the l breakpoints under the null are obtained by global optimization, and the candidate breakpoints are
obtained by sequential estimation.
Sequential Determination
Bai (1997) describes an intuitive approach for obtaining estimates for more than one break. The procedure
involves sequential application of breakpoint tests.
Begin with the full sample and perform a test of parameter constancy with unknown break. At each stage, test
for breakpoints in breakpoint tests in each subsample. Add a breakpoint whenever a subsample null is rejected.
(Alternately, one could test only the single subsample which shows the greatest improvement in the sum-of-
squared residuals.) If any of the tests reject, add the specified breakpoint to the current set.
Repeat the procedure until all of the subsamples do not reject the null hypothesis, or until the maximum number
of breakpoints allowed or maximum subsample intervals to test is reached.
Perform refinement so that breakpoints are re-estimated if they are obtained from a subsample containing more
than one break. This procedure is required so that the breakpoint estimates have the same limiting distribution
as those obtained from the global optimization procedure.
If the number of breakpoints is pre-specified, we simply estimate the specified number of breakpoints using the
one-at-a-time method.
3.3.3 The ordinary least squares (OLS) technique
Ordinary Least Squares is a mathematical approach used for prediction, the objective from this analysis is
developing a statistical model to predict the dependent variable from the values of the independent variables. Also
it is used to find if the independent variables have a significant effect on the dependent variable.
4. RESULTS
This section provides the detailed results of the study. The first part shows the normality test results (Descriptive
Statistics), while the second part provides unit root test. In part three we discuss the structural breaks of the
dependent variable. Finally, in the fourth we introduce the OLS regression results.
4.1 Descriptive Statistics
Table No. 1 below presents the descriptive statistics of the annual observations for variables during the study
period. Interestingly, the real numbers of the mean for the variables are reported as follows: gross domestic product
(GDP) is 164632.3 (million dollars), Worker Outflow Remittances (WOR) as a percentage of GDP is 5.09%,
Government Expenditure (GE) as a percentage of GDP is 23.23%, Inflation (INF) represented as Consumer Price
Index is CPI 73.49. Finally, Exports (EX) as a percentage of GDP is 42.32%.
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
10
The Jarque-Bera values for Gross Domestic Product, Workers Outflow Remittances and Exports are insignificant
to reject the null hypothesis of normal distribution meaning that all the variables have normal distribution.
Meanwhile the Jarque-Bera values for Government Expenditure and Inflation are significant at 1% to reject the
null hypothesis of normal distribution. This means that these variables don’t have normal distribution.
Table 1. Descriptive Statistics
Ln (GDP) Ln (WOR/GDP) Ln (GE/GDP) INF= Ln(CPI) Ln (EX/ GDP)
Mean 12.01147 1.627368 3.145293 4.297201 3.745279
Median 11.93975 1.634071 3.193206 4.307886 3.766017
Maximum 13.53626 2.593289 3.561686 4.807281 4.360452
Minimum 9.176179 0.132191 2.13198 3.198547 3.143842
Std. Dev. 0.953442 0.594432 0.287936 0.315195 0.305354
Skewness -0.51917 -0.6266 -1.17886 -1.55702 -0.07377
Kurtosis 3.874168 3.040273 4.979862 6.763342 1.940899
Jarque-Bera 3.454322 2.947762 17.77259 44.73743 2.143991
Probability 0.177788 0.229035 0.000138 0.000000 0.342325
Observations 45 45 45 45 45
4.2 Unit Root Test
In Augmented Dickey-Fuller (ADF), the Gross Domestic Product RGDP is presented in (LnGDPt,- LnGDPt-1) and
the data of Workers Outflow Remittances and Government Expenditure and Exports are expressed in terms of the
natural logarithm of this year’s percentage of Gross Domestic Product minus the natural Logarithm of the previous
year. Finally, the natural logarithm of consumer product index CPI this year minus the natural logarithm of the
previous year is employed as a measure of Inflation.
The results in Table No. 2 show that we can reject the null hypothesis because there is unit root, and we can accept
the alternative hypothesis because there is no unit root for all the variables (GDP ( Stationary a the level t I(0) at
1% significance level and at the first difference I(1) at 1% significance level ), Worker Remittances Outflow
(stationary at the level t I(0) 1% significance level and at the first difference I(1) at 1% significance level),
Government Expenditure (stationary at the level t I(0) 1% significance level and at the first difference I(1) at 1%
significance), Inflation (stationary at the level t I(0) 10% significance level and at the first difference I(1) at 1%
significance level), and Exports (stationary at the level t I(0) 1% significance level and at the first difference I(1)
at 1% significance level ).
Table 2. Unit Root Test for Stationary
Augmented Dickey Fuller ADF
Test Significance Level
Variables Level 1st Diff. 1% 5% 10%
Gross Domestic Product (GDP) -4.093318*** -10.42539*** -2.618579 -1.948495 -1.612135
Worker Remittances Outflow (WOR) -5.185291*** -5.376604*** -2.618579 -1.948495 -1.612135
Government Expenditure (GE) -7.908505*** -6.275321*** -2.618579 -1.948495 -1.612135
Inflation (INF) -1.923253* -5.208868*** -2.618579 -1.948495 -1.612135
Export (EX) -5.332085*** -8.637796*** -2.618579 -1.948495 -1.612135
Significance at 1% * , 5% ** , 1% ***
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
11
4.3 Test of Structural Breaks for Dependent Variable
The authors use the Bai – Perron L + 1 versus L test to determine the optimal breaks. Table no 3 results show that
F statistics on 1 vs. 2* is more than the critical value while F statistics on 2 vs 3* is less than the critical value;
therefore there are two break points according to the table below in 1981 and in 1987:
Table 3. Regression Results
Ordinary Least Squares Regression GDP (Depenedent Variable)
HAC standard errors & covariance (Prewhitening with lags = 1, Quadratic -Spectral kernel, Andrews
bandwidth = 1.2658)
Variable Coefficient t-Statistic P value
C 0.099989 2.227155 0.0311
Multiple breakpoint tests: Bai-Perron tests of L+1 vs. L sequentially determined breaks
Breakpoint variables: C
Break test options: Trimming 0.15, Max. breaks 5, Sig. level 0.05
Test statistics employ HAC covariances (Prewhitening with lags = 1, Quadratic-Spectral kernel, Andrews
bandwidth), assuming common data distribution
Sequential F-statistic determined breaks 2
Break Test F-statistic Scaled F-statistic Critical Value**
0 vs. 1 * 10.02927 10.02927 8.58
1 vs. 2 * 15.32873 15.32873 10.13
2 vs. 3 0.527031 0.527031 11.14
Breaks Dates
Sequential Repartition
1 1981 1981
2 1987 1987
Due to those structural breaks the authors add two intercept dummy variables in the model (D1: zero for the years
before 1981 and one for the years 1981 and after it. D2: zero for the years before 1987 and one for the years 1987
and after it). Accordingly, our new model is as follows
GDPt = α + β1D1t + β2D2t + β3WORt + β4 GEt + β5INFt + + β6EXPt
4.4 Regression Results
Highlighting table 4, the adjusted R2 is 87.3% which means that the independent variables explain 87.3% of the
variations in the GDP but not all of it. This means that there are other variables which explain the dependent
variable.
Furthermore, we can reject the main null hypothesis; there is no significant effect for all the variables on the GDP
and accept the alternative hypothesis which indicate that there is significant effect for all the variables on the GDP.
We based our rejection on the fact that P-value of F-statistics is less than 1% (1- confidence level (99%)), so it
falls within the rejection area.
Regarding the other hypothesis, we can reject the null hypothesis which indicate that there is no significant effect
for the workers outflow remittances on the GDP as the P-value is less than 1% (1-confidence level (99%)) which
means that workers outflow remittances significantly affects the GDP. Moreover, the coefficient of the workers
outflows remittances (independent variable) is -0.272790 (significant at 1 percent). This means that workers
outflow remittances significantly affects the GDP (negative relationship).
Furthermore, we can reject the null hypothesis which indicates that there is no significant effect for the government
expenditure on the GDP, because the P-value is less than 1% (1-confidence level (99%)). Moreover, the coefficient
of the government expenditure (independent variable) is -0.395632 (significant at 1 percent). This means that
government expenditure significantly affects the GDP (negative relationship).
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
12
In addition, we can reject the null hypothesis which indicates that there is no significant effect for the Inflation on
the GDP as the P-value is less than 1% (1-confidence level (99%)). Moreover, the coefficient of the Inflation
(independent variable) is 0.715591 (significant at 1 percent). This means that Inflation significantly affects the
GDP (positive relationship).
Finally, we can reject the null hypothesis which indicates that there is no significant effect for the Exports on the
GDP as the P-value is less than 1% (1-confidence level (99%)). The coefficient of the Exports (independent
variable) is 0.404836 (significant at 1 percent). This means that Exports significantly affects the GDP (positive
relationship).
On the other hand, Granger and Newblod (1974) assume that the model is spurious if R square is greater than
Durbin-Watson statistic, where the null hypothesis of the model is spurious and the alternative hypothesis of the
model is non-spurious or sense regression. We reject the null hypothesis for test which assumes spurious regression
and accept the alternative hypothesis of non-spurious or sense regression, where R2 value is less than Durbin-
Watson statistic as shown in the table below.
Table 4. Regression Results
Ordinary Least Squares OLS Regression Dependent Variable: GDP
Variable Coefficient t-Statistic Prob.
C 0.274366*** 6.600800 0.0000
D1 -0.212865*** -4.055698 0.0002
D2 -0.014393 -0.414902 0.6805
WOR -0.272790*** -3.305229 0.0021
GE -0.395632*** -3.866402 0.0004
INF 0.715591*** 2.972091 0.0051
EX 0.404836*** 3.958296 0.0003
R-squared 0.890453 Adjusted R-squared 0.873157
F-statistic 51.48076 Prob (F-statistic) 0.000000
Durbin-Watson stat 2.082654
Significance at 1% * , 5% ** , 1% ***
5. CONCLUSION
The Kingdom of Saudi Arabia has one of the highest Workers’ Remittances Outflows in the world. Empirical
modern studies provide evidence that its Workers Remittances Outflows weaken economic growth in the short
term but have no effect in the long term. According to the KSA vision of 2030 to replace the foreign work force
with a local one and to improve the quality of education so it will be self satisfied by its citizens which will reduce
the world in Workers’ Remittances Outflows WOR. In addition, due to the ignorance in the literature of the impact
of Worker Remittances Outflows on remitting countries GDP such as KSA and the dual effect of remittances
outflow through its negative impact by the outside flow of funds and its positive in participating in economic
growth. Our research contributes in shedding new light on the impact of Workers’ Remittance Outflow on the
KSA’s economy and checking its direction and magnitude. To investigate the effect of Workers’ Remittances
Outflow on GDP we could not employ Autoregressive Distributed Lags ARDL methodology because the data of
the dependent variable was stationary at the second difference I(2); therefore we used Bai-Perron to find the
structural breaks and to add two intercept dummy variables as independent variables to the model. Finally, we
implemented Ordinary Least Squares OLS regression to find the long-term relationship among the variables; this
method is an advanced econometric technique and is often a preferred method to depict actual effects because it
provides a more real-world context than other forms when trying to check the impact of Workers’ Remittances
Outflow on GDP.
The results showed that the adjusted R2 is 87.3% which means that the independent variables explain 87.3% of the
variations in the GDP but not all of it. Moreover, the Bai-Perron test results found that there are two structural
breaks at 1981 and 1987, therefore we add two intercept dummy variables to the model as independent variables.
In addition, the OLS regression results revealed that workers’ remittances outflow and Government Expenditure
Journal of Economic & Management Perspectives, 2019, Volume 13, Issue 2, 5-13.
Journal of Economic & Management Perspectives ISSN 2523-5338 © International Economic Society
http://www.econ-society.net
13
negatively affect the Gross Domestic Product and significant. Furthermore, Inflation and Exports positively affect
the Gross Domestic Product and significant.
Based on these results, several points can be highlighted. First, the literature review showed that Workers Outflow
Remittances has a negative impact (significantly or insignificantly) on the KSA’s economy because it reduces
consumption and investment inside the KSA. It is recommended to encourage workers to immigrate to the KSA
or to give them long legal stays so they may invest locally. Even though workers’ outflow remittances reduce
inflation, its impact is negative. Our research shows also that its impact is negatively significant, a result similar
to that of previous studies. However, the Saudi authorities should allow migrants a long legal stay and should
facilitate procedures for them to invest locally which will reduce remittances outflow. Building the KSA’s
capacities of its own citizens requires changing the culture and attitude of generations, as well as substantial
improvement in the processes of the entire educational system. Secondly, the literature review revealed that exports
and expenditure enhance the general purchasing power in the economy and our study agreed with previous studies
concerning exports because the impact of exports was shown to be positive and significant (therefore more
procedures must have taken to increase exports and encourage it). Concerning the government’s expenditures in
the KSA, it is consistent with what neoclassic economists would consider as having a harmful effect on long term
economic growth. Finally, the impact of inflation is positive and significant because the volume of exports far
exceeds that’s of the value of government expenditure which will enhance the growth in economy (Alkhathlan,
2013). Accordingly, we recommend that further research be performed which investigates the impact of Workers
Outflow Remittances but by using quarterly or monthly data to deal with the big changes in time and to fix the
stationary of the data at the second difference which would pave the way to apply Autoregressive Distributed Lags
ARDL methodology.
REFERENCES
Alkhathlan, K. (2013) “The nexus between remittance outflows and growth: A study of Saudi Arabia” Economic
Modelling, Vol. 33, pp. 695–700.
Abdel-Rahman, A. (2017) “The Determinants of Foreign Worker Remittances in the Kingdom of Saudi Arabia”
Working paper.
Baasa, T. & Melzer, S. (2012) “The Macroeconomic Impact of Remittances: A sending country perspective”,
NORFACE MIGRATION, Discussion Paper No. 2012-21.
Edrees. A. (2016) “The Impact of Foreign Workers, Outflow Remittances on Economic Growth in Selected GCC
Countries: ARDL Approach” Arabian Journal of Business and Management Review, 6:5.
Haddad, H. &Choukir, J. (2017) “Short- and long-run effects of remittance outflow shocks on the Saudi Arabian
economy” International Journal of Economics and Business Research, Volume 14, Issue 2, pp. 194–213,
DOI: 10.1504/IJEBR.2017.10007794.
Haddad, H. &Choukir, J. (2015) “The Time-varying Reponses of Saudi Arabia Economy to Workers Remittance
Outflows Shocks” Arabian Journal of Business and Management Review, 5:6, DOI: 10.4172/2223-
5833.1000155.
Kaabi, F. (2016) “The Nexus between Remittance Outflows and GCC Growth and Inflation” Journal of
International Business and Economics, Vol. 4, No. 1, pp. 76-85, DOI: 10.15640/jibe.v4n1a7.
Naufal, G. & Genc, I. (2014) “The Story of Remittance Flows from the GCC Countries” Gulf Research Center,
Gulf Labour Markets and Migration, GLMM - EN - No. 5.
Naufal, G. & Genc, I. (2017) “Impact of Remittance Outflows on Sending Economies The Case of the Russian
Federation” United Nations Economic and Social Commission for Asia and the Pacific.
Rahmouni, O. & Debbiche, I. (2017) “The effects of remittances outflow on economic growth in Saudi Arabia:
Empirical evidence” Journal of Economics and International Finance, Vol. 9(5), pp. 36-43, DOI:
10.5897/JEIF2017.0828.
