. International Journal of Economics and Financial Issues ISSN: 2146-4138 available at http: www.econjournals.com International Journal of Economics and Financial Issues, 2018, 8(1), 153-160. International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018 153 The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries Rawan F. Shubati1*, Taleb Awad Warrad2 1Department of Business Economics, University of Jordan, Amman, Jordan, 2Dean of Faculty of Business, Middle East University, Amman, Jordan. *Email: rawan.shubita@gmail.com ABSTRACT This study investigates the impact of international trade openness on government revenue in Middle East and North African (MENA) countries for the period of 2000-2015. More specifically, this study examines the relationship between government revenue and international trade openness, real gross domestic product (GDP) per capita, corruption level measure, and population. The study utilized panel data, covering the period of 2000-2015, for nine selected MENA countries. The results of the study, using the panel fully modified least squares, highlights the negative impact of international trade openness on government revenue. Moreover, the results indicate that countries with a higher real GDP per capita and lower corruption level have more government revenue while the total population plays a negative role in government revenues. Keywords: Trade Liberalization, Government Revenue, Gross Domestic Product JEL Classifications: E01, F1, H5 1. INTRODUCTION The role of trade openness as a proxy indicator of trade liberalization is one of the vital issues that have been investigated widely in the economic literature. Several studies argued that trade openness as an “engine of growth” can play a significant role in improving economic growth and the level of development. Especially in the long run, due to the ability of international trade in increasing productivity and encouraging greater efficiency, it has been revealed that internationally active countries tend to be more productive (Kim et al., 2013; Shahbaz, 2012; Dong, 2014). In general, developing countries suffer from a large deficit in both public budget and current account balance. In addition to insufficient domestic financial resources as a main source of fund. However, the different types of financial inflows play an important role in case of developing countries to finance government expenditure and achieve long-run developmental strategies (Todaro and Smith, 2009). Hence, the importance of this study comes from identifying the potential benefit of international trade openness in driving government revenue flows. Recent studies attempt to identify the potential factors that can be related to promoting domestic financial flows. Trade liberalization is considered by many studies as one of the major determinants of financial flows, especially government revenue. 2. BACKGROUND Several studies explored the potential linkage between international trade openness and public government revenue as an example of domestic financial flows. Direct and indirect tax types are considered as one of the major domestic financial resources that can be used to finance government expenditure. The relationship between international trade openness and government revenue has been widely investigated. According to previous literature, there are two different arguments about the impact of trade openness (free trade) on government revenue: The positive effect argument and the negative effect argument. The positive effect argument: Previous studies attempted to examine the main determinants of government revenue in Shubati and Warrad: The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018154 general and tax revenue. Specifically, several studies (Stotsky and Woldemariam, 1997; Bornhorst et al., 2009; Drummond et al., 2012) aimed to identify the potential factors that can be related to better the tax revenue inflows. The empirical evidence of these studies show that the degree of international trade openness plays a positive role in enhancing the tax revenue by increasing the productivity of output and promoting economic growth and hence, generating more government revenue. The outcomes of recent studies support the positive effect argument of international trade openness on government revenue in general, and tax revenue in particluar. Mushtaq et al., (2012) analyzed the determinants of tax revenue in Pakistan over a period of 1975-2010. This study used the following explanatory variables: Exchange rate, population, gross domestic product (GDP), and trade openness. The study’s results highlighted that international trade openness is one of the most important determinants of tax revenue in Pakistan. In addition, Jaffri et al.s. (2015) study investigated the potential linkage between trade liberalization and tax revenue in Pakistan during the period of 1982-2013 by using the autoregressive distributed lag model. The empirical study evidence showed a positive impact of trade liberalization on tax revenue over the study period. The results indicate that in the long- run, a 1% increase in international trade openness leads to 0.35% increase in tax revenue. Based on this, the study recommends the policymakers to reduce the restrictions on international trade and take advantage of the positive role of international trade openness in enhancing the total tax revenue by reducing tax evasions and tax exemptions in Pakistan. Gnangnon (2017) indicated that trade openness has a positive impact on tax revenue, based on the panel data for 169 countries among which 37 where least developed countries during the period of 1995-2013. The study showed that the positive role of trade liberalization depends on the level of development of the country and the level of its domestic trade liberalization policy. The general argument indicates that restriction in the trade policy weakens the positive impact of trade openness on government revenue and hence, the tax revenue. In addition, a higher level of development measured by the real GDP increases the positive impact of trade openness on tax revenue, especially in the long-run. Lutfunnahar (2007) explored the main determinants of tax revenue performance in Bangladesh by using tax revenue as a percentage of the GDP as a dependent variable. He argued that Bangladesh has a low tax ratio and the study results suggested that increasing the trade openness will lead to better tax revenue performance. The study results highlighted that international trade related positively to tax revenue and is considered to be one of the main variables that determine the tax effort. Chaudhry and Munir (2010) used the same explanatory variables as Lutfunnahar’s (2007) study to investigate the determinants of tax revenue in Pakistan from 1973 to 2009. The empirical evidence showed that international trade openness, and international financial flows (external debt, foreign aid), and political stability are the major determinants of tax efforts in Pakistan. The study also showed that foreign trade played a key role in enhancing the tax revenue as a percentage of the GDP ratio, since Pakistan increased the degree of trade liberalization over the study period and which contributed positively to the tax revenue performance. They argue that Pakistan suffered from a high budget deficit with low tax ratio (tax revenue/GDP), but after boosting the trade openness the tax ratio increased. In general, the empirical evidence for most of the studies that analyzed the main determinants of tax revenue support the positive impact of trade openness on the performance of tax revenue. However, this positive effect of trade openness on government revenue depends on several factors, such as the structure of trade liberalization and the impact of the existing free trade structure on each component of government revenue, in addition to the import and export price elasticity (Frankel, 1999). In addition, findings from several studies on international trade openness and non-tax resource revenue showed that free trade positively affects the domestic non-tax revenue resources as a percentage of the GDP by, enhancing the mobilization of these types of domestic revenue (Crivelli and Gupta, 2014; Brun et al., 2015; Thomas and Trevifo, 2013). 2.1. Negative Effect Argument On the other hand, several studies argued that trade liberalization can lead to a reduction in government revenue; the rationale behind this negative argument is based on the fact that developing countries rely heavily on indirect tax revenue sources, such as import tariffs. So, when countries increase their degree of openness to international trade, it induces a reduction of the restrictions on imports tariff tax and hence, it decreases the tax revenue. The general negative argument of trade liberalization on tax revenue, based on recent studies, argues that there is a potential loss in government revenue as result of decreasing the tax on international trade to facilitate free trade (Khattry and Rao, 2002). The empirical evidence from recent studies (such as Khattry and Rao, 2002) showed that increasing the degree of international trade openness contributes negatively to the total tax revenue in low-middle income countries. The study shows that the structural characteristics of low-middle income countries, such as the size of population, the degree of urbanization, and the age dependency ratio and others explain the negative effect of international trade openness on government revenue. Cagé and Gadenne’s (2014) study indicated that international trade openness induces extra financial costs related to international trade tax cuts in developing countries. However, in developed countries, increasing the degree of trade openness does not have a negative effect on government revenue. This is because the high-income developed countries have efficient tax management systems which can compensate any decline in the tax burden in foreign trade by imposing tax on domestic transactions. The main conclusion of this study is that international trade openness leads to a reduction in the total tax revenue, especially in the long-run. 3. DATA, METHODOLOGY AND ECONOMETRIC MODEL 3.1. The Econometric Model Models based on economic theory and previous empirical studies, such as Brun and Gnangnon, (2017), will be estimated in order to Shubati and Warrad: The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018 155 achieve the study objective and test the main study hypotheses. To analyze the impact of international trade openness on government public revenue, a log linear form, such as the following, is used: logY1it = α0+α1logx1it+α2logx2it+α3logx3it+α4logx4it+ԑit Where: • logY1it: Denotes government public revenue, excluding grants, as a percentage of the GDP. • logx1it: The logarithm of international trade openness of a country, measured by the sum of exports and imports for goods and services in terms of dollars, as a percentage of the GDP measured in terms of dollars. • Logx2it: The logarithm of the real GDP per capita, measured in terms of dollars. • Logx3it: The logarithm of the total population. • Logx4it: The logarithm of the corruption level. • α0: The intercept. • α1,….,α4: The estimated parameters for the independent variables. • ԑit: The error term. 3.2. Study Sample The study used the annual database, covering the period of 2000- 2015, for selected countries from the Middle East and North African (MENA) regions. The selected countries are: Algeria, Bahrain, Egypt, Iraq, Jordan, Kuwait, Lebanon, Morocco, Oman, Qatar, Saudi Arabia, and Tunisia. The country selection was based mainly on the data availability. 3.3. Study Hypotheses The following null hypotheses were formulated: H0: There is no significant relationship between international trade openness and government revenue. H01: There is no significant relationship between government revenue and the real GDP per capita. H02: There is no significant relationship between government revenue and corruption level. H03: There is no significant relationship between government revenue and total population. 3.4. Study Variables Measurement In order to test the study’s hypotheses, the model variables need to be defined and measured. As for the study dependent variable, there are two main types of financial flow: The first one is the domestic financial flow, while the other one is the international financial flow. According to Shukla and Glenday (2001), domestic financial flows are when government public revenue is considered as one of the main tools for fiscal policy, and government revenue is defined as the domestic financial inflow that the government received from different sources as follows: 1. Taxable sources (direct tax): Taxes imposed on different sources of income and wealth and paid by mainly by the public (individuals and corporations), indirect tax: Sales tax (tax on goods and services). 2. Non-taxable sources: Income from government-owned corporations, capital inflows in the form of external loans and debts from international financial institutions, and foreign aid from other counties. The general argument on trade openness, as one of the study’s independent variables, is based on the previous literature (Akubu et al., 2015; Hysa et al., 2014; Kose et al., 2009) which denotes that trade openness, as a proxy for trade liberalization, has a positive role in promoting and enhancing economic growth and trade openness, and thus, is considered as one of the main determinants of economic growth and is considered as an engine of growth. There are different measures for international trade openness, the most popular one being the sum of total exports and imports as a percentage of the GDP. Based on several researches (Brun and Gnangnon, 2017; Masood et al., 2016; Pattayat, 2016; Chaudhry and Munir, 2010), the following control variables are expected to be included: Real GDP per capita, total population, corruption index. Real GDP per capita: According to the World Bank, the real GDP per capita can be used as a proxy indicator for social welfare and used as a benchmark to compare the standard of living between the countries. Real GDP per capita is calculated as the total economic output of a country divided by its total population, adjusted for inflation. C o r r u p t i o n l e v e l i n d e x : A c c o r d i n g t o t h e Wo r l d w i d e Governance Indicators, a corrupt practice is the offering, giving, receiving or soliciting, directly or indirectly, of anything valuable to influence the actions of another party improperly. 4. EMPIRICAL ANALYSIS This section presents the empirical portion of the study. The analysis of the results will be conducted according to the following three steps: 1. Examine the stationarity of each variable using panel unit root tests. 2. Examine the long run-relationship between the study variables. 3. Depending on 1 and 2, a proper method of estimation will be chosen. 4.1. Panel Unit Root Tests Before testing the existence of a cointegration relationship between the study variables, the stationarity of the variables are tested by applying the different types of panel stationarity tests at both the levels differently. The results with constant, constant and trend are presented in Tables 1-4, respectively. It is important to detect whether the study variables have unit root. There are various methods that can be used to test the stationarity of variables. Summary is considered as a formal test method that includes the major types of panel unit root test. The general hypotheses for a panel unit root test are as follow: Shubati and Warrad: The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018156 • Null hypothesis: Panel data has a unit root (non-stationary). • Alternative hypothesis: Panel data does not have a unit root (stationary). Tables 1 and 2 present the testing of the null hypothesis of the unit root of the study variables by applying a summary test type at level with the individual effects and individual linear trends, respectively. Based on the probability value, we can take the decision regarding the stationarity of the study variables. If the probability value is <5%, we reject the null hypothesis and accept the alternative hypothesis. This means that the study variables at level do not have a unit root. On the other hand, if the probability value is more than 5%, we fail to reject the null hypothesis and reject the alternative hypothesis, meaning that the study variables have a unit root. If the results are mixed, we take the decision based on the majority of test method results. Hence, if most of the methods show that the study variable has a unit root, the decision will be that the study variable is non-stationarity. On the other hand, if most of the methods show that the study variable does not have a unit root, the decision will be that the study variable is stationarity. According to Table 1, government public revenue (TGR) and the international trade openness of a country (TOP) has a unit root Table 1: Panel unit root test: Level Variables Exogenous variables Methods Statistic value P value Decision TGR Individual effects Levin, Lin and Chu −1.4291 0.0765 Non-stationary Im, Pesaran and Shin W-stat −0.57878 0.2814 Non-stationary ADF-Fisher Chi-square 19.9816 0.3339 Non-stationary PP-Fisher Chi-square 22.6035 0.2063 Non-stationary TOP Individual effects Levin, Lin and Chu −1.59068 0.0558 Non-stationary Im, Pesaran and Shin W-stat 0.27319 0.6076 Non-stationary ADF-Fisher Chi-square 11.8965 0.8525 Non-stationary PP-Fisher Chi-square 11.1831 0.8864 Non-stationary COR Individual effects Levin, Lin and Chu −2.30818 0.0105 Stationary Im, Pesaran and Shin W-stat −1.46503 0.0715 Non-stationary ADF-Fisher Chi-square 25.9767 0.1003 Non-stationary PP-Fisher Chi-square 25.2025 0.1194 Non-stationary POP Individual effects Levin, Lin and Chu −1.70468 0.0441 Stationary Im, Pesaran and Shin W-stat 0.63738 0.7381 Non-stationary ADF-Fisher Chi-square 32.4612 0.0194 stationary PP-Fisher Chi-square 2.69370 1.0000 Non-stationary GPC Individual effects Levin, Lin and Chu −3.885 0.0001 stationary Im, Pesaran and Shin W-stat −1.09644 0.1364 Non-stationary ADF-Fisher Chi-square 21.5038 0.2548 Non-stationary PP-Fisher Chi-square 12.9913 0.7921 Non-stationary Table 2: Panel unit root test: Level Variables Exogenous variables Methods Statistic value P value Decision TGR Individual effects Levin, Lin and Chu −1.1058 0.1344 Non-stationary individual linear trends Breitung t-stat 0.32464 0.6273 Non-stationary Im, Pesaran and Shin W-stat 0.89496 0.8146 Non-stationary ADF-Fisher Chi-square 12.1809 0.8378 Non-stationary PP-Fisher Chi-square 15.4666 0.6297 Non-stationary TOP Individual effects Levin, Lin and Chu −1.32458 0.0927 Non-stationary individual linear trends Breitung t-stat 0.42042 0.6629 Non-stationary Im, Pesaran and Shin W-stat 0.27277 0.6075 Non-stationary ADF-Fisher Chi-square 16.2454 0.5754 Non-stationary PP-Fisher Chi-square 13.9501 0.7323 Non-stationary COR Individual effects Levin, Lin and Chu −0.65163 0.2573 Non-stationary individual linear trends Breitung t-stat −0.46709 0.3202 Non-stationary Im, Pesaran and Shin W-stat −0.80078 0.2116 Non-stationary ADF-Fisher Chi-square 22.6818 0.2031 Non-stationary PP-Fisher Chi-square 18.5060 0.4228 Non-stationary POP Individual effects Levin, Lin and Chu −20.3304 0.0000 Stationary individual linear trends Breitung t-stat 2.32242 0.9899 Non-stationary Im, Pesaran and Shin W-stat −26.2292 0.0000 Stationary ADF-Fisher Chi-square 159.097 0.0000 Stationary PP-Fisher Chi-square 6.55016 0.9934 Non-stationary GPC Individual effects Levin, Lin and Chu 0.71640 0.7631 Non-stationary Individual linear trends Breitung t-stat 2.33490 0.9902 Non-stationary Im, Pesaran and Shin W-stat 3.20938 0.9993 Non-stationary ADF-Fisher Chi-square 10.1648 0.9264 Non-stationary PP-Fisher Chi-square 3.15998 1.0000 Non-stationary Shubati and Warrad: The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018 157 which is at level with the individual effects. With regard to Levin, Lin and Chu tested that the corruption level (COR), real GDP per capita (GPC) are stationary at level but the majority results for other test types show that COR, GPC have a unit root. With regard to ADF - Fisher Chi-square and Levin, Lin and Chu tested that the total population (POP) is stationary, but the majority results for other test types show that POP has a unit root. According to Table 2, government public revenue (TGR), international trade openness of a country (TOP), the corruption level (COR), GDP per capita (GPC) have a unit root at level with the individual effects and individual linear trends. With regard to Levin, Lin and Chu, Im, Pesaran and Shin W-stat and ADF - Fisher Chi-square tests, the total population (POP) is stationary but the majority results for other test types show that POP has a unit root. Tables 3 and 4 present the testing of the null hypothesis of the unit root of the study variables by applying the summary test type at 1st difference with the individual effects and individual linear trends, respectively. According to Table 3, government public revenue (TGR), international trade openness of a country (TOP), the corruption level (COR), real GDP per capita (GPC), total population (POP) are stationary at first difference with the individual effects. According to Table 4, the government public revenue (TGR), international trade openness of a country (TOP), the corruption level (COR) is stationary at first difference with the individual effects and individual linear trends. With regard to Breitung t-stat, the total population (POP) and real GDP per capita (GPC) have a unit root but the majority results for other test types show that the POP and GPC are stationary. The main conclusion is that the study data are stationary after taking the first difference into consideration. 4.2. Panel Cointegration Test It is critical to examine the existence of a cointegration relationship between the study variables in order to check whether these variables have a long-run stable relationship. Developing a panel cointegration model is the main target of this study. To achieve this target, there is a precondition for running the panel cointegration model, in other words (cointegration regression), and that the study variables must be non-stationary at level. However, when the variables are converted into first difference, it will become stationary. The general hypotheses for panel unit root test are as follow: • Null hypothesis: There is no cointegration in the model. • Alternative hypothesis: There is a cointegration in the model. According to the Kao panel cointegration test (Engle Granger based), there is one deterministic trend specification which is the individual intercept. Table 5 presents the outcome of this test for the study model. If the probability value is <5%, we can reject the null hypothesis and accept the alternative one, meaning that the study variables are cointegrated. On the other hand, if the probability value is more than 5%, we will fail to reject the null hypothesis and reject the alternative one, meaning that the study variables are not cointegrated. With regard to the Kao test outcomes in Table 5, the probability value is <5%, meaning that the study variables for each model have a long-run, stable relationship. 4.3. Model Estimation In light of the results of the panel stationarity and cointegration tests, the model cannot be estimated directly with the panel ordinary least squares (OLS). To avoid the problem of spurious regression, a panel fully modified OLS (FMOLS) was used to estimate the model’s long-run parameters. The results of applying panel fully modified least squares are shown in Table 6 below: According to the previous studies, there are different types of factors that have a potential impact on the government revenue. These factors include both the supply side and demand side factors (Brun and Gnangnon, 2017). This study analyzed the potential role of some of these factors, such as: 4.3.1. International trade openness The study used international trade openness as a traditional measure for trade liberalization by considering the summation Table 3: Panel unit root test: 1st difference Variables Exogenous variable Methods Statistic value P value Decision D (TGR) Individual effects Levin, Lin and Chu −2.69274 0.0035 Stationary Im, Pesaran and Shin W-stat −2.85569 0.0021 Stationary ADF-Fisher Chi-square 39.5585 0.0024 Stationary PP-Fisher Chi-square 95.7312 0.0000 Stationary D (TOP) Individual effects Levin, Lin and Chu −5.13671 0.0000 Stationary Im, Pesaran and Shin W-stat −3.68057 0.0001 Stationary ADF-Fisher Chi-square 45.4309 0.0004 Stationary PP-Fisher Chi-square 85.7624 0.0000 Stationary D (COR) Individual effects Levin, Lin and Chu −6.82269 0.0000 Stationary Im, Pesaran and Shin W-stat −5.55652 0.0000 Stationary ADF - Fisher Chi-square 65.3645 0.0000 Stationary PP-Fisher Chi-square 96.7489 0.0000 Stationary D (POP) Individual effects Levin, Lin and Chu −15.9662 0.0000 Stationary Im, Pesaran and Shin W-stat −15.5127 0.0000 Stationary ADF-Fisher Chi-square 158.534 0.0000 Stationary PP-Fisher Chi-square 7.12886 0.9890 Stationary D (GPC) Individual effects Levin, Lin and Chu −2.54573 0.0055 Stationary Im, Pesaran and Shin W-stat −2.15906 0.0154 Stationary ADF-Fisher Chi-square 30.8497 0.03 Stationary PP-Fisher Chi-square 34.444 0.0111 Stationary Shubati and Warrad: The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018158 of the total export and import as a percentage of the GDP. The results of the model are reported in Table 6. The results show that the international trade openness (TOP) coefficient is associated negatively with the government revenue (TGR) as a dependent variable; meaning that a 1% increase in international trade openness leads to 0.228% decline in the government revenue. This relationship has a significance level of 1%, wherein the results are consistent with the outcomes of Khattry and Rao’s (2002) study. This result supports the negative perspective of the relationship between international trade openness and government revenue. The rationale behind this negative argument is based on the fact that international liberalization induced a reduction in the taxes imposed on international trade transactions and hence, reduced the government revenue. Previous studies (Agbeyegbe et al., 2006) have argued that this negative impact can be offset only if the countries are able to recover these losses in the foreign trade tax from other resources, such as domestic tax sources, imposed on domestic transactions. Table 4: Panel unit root test: 1st difference Variables: Exogenous variables Methods Statistic value P value Decision D (ODA) Individual effects Levin, Lin and Chu −2.22671 0.0130 Stationary Individual linear trends Breitung t-stat −2.25999 0.0119 Stationary Im, Pesaran and Shin W-stat −1.9271 0.0270 Stationary ADF-Fisher Chi-square 27.0990 0.0404 Stationary PP-Fisher Chi-square 98.4920 0.0000 Stationary D (FDI) Individual effects Levin, Lin and Chu −8.43573 0.0000 Stationary Individual linear trends Breitung t-stat −5.56191 0.0000 Stationary Im, Pesaran and Shin W-stat −7.58889 0.0000 Stationary ADF-Fisher Chi-square 77.0248 0.0000 Stationary PP-Fisher Chi-square 91.2491 0.0000 Stationary D (TGR) Individual effects Levin, Lin and Chu −8.24354 0.0000 Stationary Individual linear trends Breitung t-stat −5.36828 0.0000 Stationary Im, Pesaran and Shin W-stat −5.50619 0.0000 Stationary ADF-Fisher Chi-square 58.514 0.0000 Stationary PP-Fisher Chi-square 77.7774 0.0000 Stationary D (TOP) Individual effects Levin, Lin and Chu −5.86559 0.0000 Stationary Individual linear trends Breitung t-stat −3.28127 0.0005 Stationary Im, Pesaran and Shin W-stat −2.58814 0.0048 Stationary ADF-Fisher Chi-square 34.8833 0.0098 Stationary PP-Fisher Chi-square 91.3201 0.0000 Stationary D (COR) Individual effects Levin, Lin and Chu −6.78376 0.0000 Stationary Individual linear trends Breitung t-stat −5.8113 0.0000 Stationary Im, Pesaran and Shin W-stat −6.01492 0.0000 Stationary ADF-Fisher Chi-square 65.051 0.0000 Stationary PP-Fisher Chi-square 91.8147 0.0000 Stationary D (POS) Individual effects Levin, Lin and Chu −2.48914 0.0064 Stationary Individual linear trends Breitung t-stat −2.441 0.0073 Stationary Im, Pesaran and Shin W-stat −2.39683 0.0083 Stationary ADF-Fisher Chi-square 33.9099 0.0129 Stationary PP-Fisher Chi-square 70.9330 0.0000 Stationary D (POP) Individual effects Levin, Lin and Chu −21.2729 0.0000 Stationary Individual linear trends Breitung t-stat 2.88624 0.9981 Non-stationary Im, Pesaran and Shin W-stat −17.1031 0.0000 Stationary ADF-Fisher Chi-square 89.9424 0.0000 Stationary PP-Fisher Chi-square 6.34448 0.9946 Non-stationary D (DOC) Individual effects Levin, Lin and Chu −2.97528 0.0015 Stationary Individual linear trends Breitung t-stat 1.17272 0.8795 Non-stationary Im, Pesaran and Shin W-stat −1.28149 0.1 Non-stationary ADF-Fisher Chi-square 28.5417 0.0453 Stationary PP-Fisher Chi-square 35.7808 0.0075 Stationary D (GPC) Individual effects Levin, Lin and Chu −3.85369 0.0001 Stationary Individual linear trends Breitung t-stat 0.08055 0.5321 Non-stationary Im, Pesaran and Shin W-stat −2.03294 0.0210 Stationary ADF-Fisher Chi-square 34.0085 0.0126 Stationary PP-Fisher Chi-square 48.4338 0.0001 Stationary Table 5: Kao residual cointegration test Cointegration test Statistic value P value ADF −4.16252 0.0000 Table 6: Estimation result for the model using FMOLS Variable Coefficient standard error t-statistic P value TOP −0.227934 0.014005 −16.27549 0.0000 GPC 0.054014 0.001778 30.3778 0.0000 POP −0.002044 0.000634 −3.224717 0.0016 COR −0.047592 0.010698 −4.448652 0.0000 Source: Authors calculation using Eviews. FMOLS: Fully modified ordinary least squares Shubati and Warrad: The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018 159 4.3.2. The real GDP per capita The real GDP per capita considers a proxy measure for the level of development. Real GDP per capita (GPC) has a positive and significant impact on the government revenue at 1% significant level, meaning that a 1% increase in the real GDP per capita leads to a 0.054% increase in the government revenue. These results are consistent with the outcomes of (Agbeyegbe et al., 2006) other studies which argued that a high real GDP per capita reflects a higher capacity for paying taxes and hence, this leads to more tax revenue collection. 4.3.3. Total population The relationship between the total population (POP) and government revenue (TGR) is negative and significant at 1%, meaning that a 1% increase in the total population results in 0.002% decline in the government revenue. Therefore, we conclude that the total population has a weak negative effect on the government revenue. These results are consistent with the outcomes of (Khattry and Rao, 2002; Bahl, 2003) other studies, which argued that in case of the faster growth in population, it is difficult for the government to capture the new taxpayers, especially in the short-run. Based on this, we can conclude that a higher size in the total population leads to lower tax revenue. 4.3.4. The level of corruption The impact of the corruption level (COR) on government revenue is negative and significant at 1%, meaning that a 1% increase in the corruption level leads to 0.048% decline in the government revenue. This result is consistent with the outcomes of Bird et al. (2008). The study argued that a lower corruption level contributes positively in the mobilization of the government revenue and hence, enhances the performance of tax collection. With regard to this, a lower corruption level leads to more government revenue. 5. CONCLUSION The main goal of this study is to empirically test whether the existence of international trade openness as a proxy measure for trade liberalization is associated with providing more domestic financial flows measured by government public revenue in MENA countries, during the period of 2000-20015. Consequently, the study looks deeply into more related elements by exploring the main factors that determine domestic financial flows. The study used international trade openness as a traditional measure of trade liberalization by taking the summation of the total export and import as a percentage of the GDP. The empirical results indicated that the existence of international trade openness induced a reduction in the government public revenue. These results are consistent with the outcomes of Khattry and Rao’s research (2002). Based on the above discussion, we conclude that the empirical results of the study support the negative perspective of the relationship between international trade openness and government revenue. The rationale behind this negative argument, based on the fact that developing countries rely heavily on indirect tax revenue sources such as import tariffs. Therefore, when countries increase the degree of openness to international trade, it induces a reduction in the restrictions on import tariff tax and hence, it decreases the tax revenue. Hence, the main conclusion of this study is that there is a potential loss in government revenue as result of decreasing the tax on international trade for facilitating free trade in developing countries. REFERENCES Agbeyegbe, T.D., Stotsky, J., WoldeMariam, A. (2006), Trade liberalization, exchange rate changes, and tax revenue in Sub-Saharan Africa. Journal of Asian Economics, 17(2), 261-284. Akubu, M., Akanegbu, B. (2015), The impact of international trade on economic growth in Nigeria: 1981-2012. European Journal of Business, Economics and Accountancy, 3(6), 26-36. Bahl, R.W. (2003), Reaching the Hardest to Tax: Consequences and Possibilities. Paper Presented at the ‘Hard to Tax: An International Perspective’ Conference, Andrew Young School of Policy Studies: Georgia State University. Bird, R.M., Martinez-Vazquez, J., Torgler, B. (2008), Tax effort in developing countries and high income countries: The impact of corruption, voice and accountability. Economic Analysis and Policy, 38(1), 55-71. Bornhorst, F., Gupta, S., Thornton, J. (2009), Natural resource endowments and the domestic revenue effort. European Journal of Political Economy, 25, 439-446. Brun, J., Gnangnon, S. (2017), Does Trade Openness Contribute to Driving Financing Flows for Development? World Trade Organization, WTO Working Paper. Brun, J.F., Chambas, G., Mansour, M. (2015), Tax Effort of Developing Countries: An Alternative Measure. In: Editors. Financing Sustainable Development Addressing Vulnerabilities, Ch. 11. Paris: Economica. FERDI. Cagé, J.,Gadenne, L. (2014), Tax Revenues, Development, and the Fiscal Cost of Trade Liberalization, 1792-2006, Working Paper. Crivelli, E., Gupta, S. (2014), Resource Blessing, Revenue Curse? Domestic Revenue Effort in Resource-Rich Countries. IMF Working Paper, WP/14/5. Chaudhry, I., Munir, F. (2010), Determinants of low tax revenue in Pakistan. Pakistan Journal of Social Sciences (PJSS), 30(2), 439-452. Dong, F. (2014), The impact of economic openness on standard of living and income inequality in eight countries. Emerging Markets Review, 21, 234-264. Drummond, P., Daal, W., Srivastava, N., Oliveira, L.E. (2012), Mobilizing Revenue in Sub-saharan Africa: Empirical Norms and Key Determinants. IMF Working Paper, No. 12/108, Washington, DC. Frankel, J.A. (1999), No Single Currency Regime is Right for All Countries or At All Times. NBER. Working Paper No. 7338. Cambridge, MA. Gnangnon, S. (2017), Structural economic vulnerability, openness and bilateral development aid flows. Economic Analysis and Policy, 53, 77-95. Hysa, E., Albania, T. (2014), Trade effect on economic growth of Balkan countries. Economic Theories-International Economic Relations, 11, 109. Jaffri, A., Tabassum, F., Asjed, R. (2015), An empirical investigation of the relationship between trade liberalization and tax revenue in Pakistan. Pakistan Economic and Social Review, 53(2), 317-330. Khattry, B., Mohan, J.R. (2002), Fiscal Faux Pas? An analysis of the revenue implications of trade liberalization. World Development, 30(8), 1431-1444. Kim, D., Lin, S., Suen, Y. (2013), Investment, trade openness and foreign direct investment: Social capability matters. International Review of Economics and Finance, 26, 56-69. Shubati and Warrad: The Effects of International Trade Openness on Government Revenue: Empirical Evidence from Middle East and North African Region Countries International Journal of Economics and Financial Issues | Vol 8 • Issue 1 • 2018160 Kose, M., Prasad, E., Terrones, M. (2009), Does openness to international financial flows raise productivity growth? Journal of International Money and Finance, 28, 554-580. Lutfunnahar, B. (2007), A Panel Study on Tax Effort and Tax Buoyancy with Special Reference to Bangladesh. Working Paper 715: Policy Analysis Unit (PAU) Research Department Bangladesh Bank. Masood, S., Ilyas, M., Abdul Rehman, C., Sheikh, M. (2016), Determinants of foreign direct investment: Evidence from Pakistan. Pakistan Journal of Social Sciences (PJSS), 36(1), 165-176. Mushtaq, S., Bakhsh, K., Hassan S. (2012), Estimating impact of trade liberalization on tax revenue in Pakistan. Journal of Agriculture & Social Sciences, 8, 48-150. Pattayat, S. (2016), Examining the determinants of FDI inflows in India. Theoretical and Applied Economics, 60(2), 225-238. Shahbaz, M. (2012), Does trade openness affect long run growth? Cointegration, causality and forecast error variance decomposition tests for Pakistan. Economic Modeling, 29, 2325-2339. Shukla, G., Glenday, G. (2001), Public finance in open economies. Duke Center for international development: Duke University. Stotsky, J.G., WoldeMariam, A. (1997), Tax effort in Sub-Saharan Africa. IMF Working Paper, WP/97/107. Thomas, A., Treviño, J.P. (2013), Resource Dependence and Fiscal Effort in Sub-Saharan Africa. IMF Working Paper, WP/13/188. Todaro, M., Smith, S. (2009), Economic Development. 10th ed. Essex, UK: Pearson Education.