Effects of Foreign Direct Investment and Trade Openness on Tax Earnings: A Study of Selected Sub-Saharan African Economies

Abstract

Every economy’s prosperity is determined by the quantity of tax income it receives. Over the years, studies have demonstrated that inflows from foreign investments and openness to international trade are important contributors to a country’s tax income. Based on this assumption, this study seeks to examine the impact of foreign direct investment (FDI) and open trade on tax income in a number of sub-Saharan African nations. The World Bank Development Indicators data on tax revenue, FDI, exports, imports, and exchange rates from 1990 to 2022 are used in the study. We also use the pooled mean group/panel autoregressive distributed lag approach to examine the data gathered for this inquiry. The results reveal that, in the long term, FDI has a significant negative impact on tax income; nevertheless, in the short run, Ghana’s tax revenue collection suffers while other nations profit from FDI. The results reveal that Nigeria’s exporting is detrimental to tax revenue collection, but South Africa’s export of goods and services is beneficial. However, imports and currency rates benefit Nigeria, Ghana, and South Africa in the near term. Thus, the research suggests improving tax rules and administration to prevent the movement of resources by foreign investors out of the host countries in order to avoid the imposition of huge tax burdens on their firms. Countries with low exports, such as Nigeria, are urged to enhance local manufacturing to meet international export standards in order to alleviate the continual negative balance of payments, which is primarily fixed by the adequate export of products and services.

1. Introduction

The justification for providing Foreign Direct Investment (FDI) motivation is centered on FDI’s direct and indirect benefits to the countries that receive it. Multinational companies are intended to contribute extra cash, increase the capacity for manufacturing, improve technological spread, and impart expertise in manufacturing and managerial capabilities. As such, the host economy benefits from higher levels of production, commerce, efficiency, and job opportunities as well as increased tax income (Becker et al. 2012). According to Serin and Demir (2022), inflows of foreign direct investment can broaden the base of taxation by establishing innovative manufacturing sectors, creating jobs directly or indirectly, and producing taxable activities across a variety of fields, including the acquisition and sale of items and amenities. Nevertheless, policymakers’ race-to-the-bottom strategies to encourage inflows of foreign direct investment or discourage them from leaving for other countries are likely to harm tax collections by limiting the base of taxes (Serin and Demir 2022). Globalization and the corresponding rise in the movement of capital have generated opportunities for possibly damaging tax rivalry among governments seeking to draw in international businesses (Gropp and Kostial 2001). Large multinational corporations can relocate their financial assets to decrease the amount they pay in taxes. Although this approach may not be bad by itself, the difficulties that national fiscal agencies face in taxing the capital of large international businesses can cause imbalances in investment and commercial arrangements. Major corporations’ capacity to reduce or eliminate their tax duty has profound consequences for a country’s fiscal framework, perhaps resulting in increasingly regressive tax structures, higher revenue shortfalls, or reduced services for the masses.

Many countries have established an array of incentives and policies to improve trade openness, collectively aiming to attract crucial foreign direct investment (FDI) for economic development (Onuogu et al. 2024). Fiscal motivations, particularly taxes, play an important role among the numerous rewards provided by authorities for recruiting multinational companies. Indeed, there has recently been a trend to lower taxes to encourage FDI from every continent (Devereux and Freeman 1995; Loretz 2008). According to empirical research, a one percent increase in the tax rate reduces foreign direct investment by an average of 3.7% (OECD 2008). A country’s commercial tax policy might have varying effects on nations around the globe. If an economy’s national burden of taxes is excessive compared to other nations, the tax baseline may migrate to jurisdictions with lower tax regimes, indicating outbound outflows of FDI (Gropp and Kostial 2001). Economies may also compete for inbound capital flow. Taxation may also influence corporations’ choices regarding earnings reporting. Empirical evidence shows that international companies invest substantial sums in price shifting alongside other overseas taxation-planning tactics to reduce tax liability.

According to Mahdavi (2008), once countries that are developing open their markets to foreign commerce, they face substantial problems in obtaining revenue for the budget to meet their developmental objectives. Although Weisbrot and Baker (2003) put forward the heavy reliance of these nations on budgetary income, the majority of which is generated by taxes, it is exacerbated by the decrease in foreign trade taxation while adopting moderate and over-time trade liberalization strategies. However, the domestic mobilization of resources is a significant concern for emerging nations. Although many research investigations have sought to emphasize various techniques for increasing income from taxes, the role of foreign direct investment (FDI) inflows in this endeavor has attracted little scrutiny. Commercial liberalization, whether by means of global or regional arrangements, is unavoidable and could further decrease foreign trade tax income (Ho et al. 2023). A number of current and past studies (ALshubiri 2024; Cagé and Gadenne 2018; Gnangnon and Brun 2019; Gondo et al. 2021; Jemiluyi and Jeke 2023; Kastrati and Vokshi 2022; Khattry and Rao 2002; Khattry 2003; Serin and Demir 2022; Shrestha et al. 2021) have found that international trade access and FDI have detrimental effects on fiscal earnings, particularly tax receipts in both advanced and emerging economies. Several other researchers (Binha 2021; Camara 2023; Ha et al. 2022; Ho et al. 2023; Gaspareniene et al. 2022; Mahmood and Chaudhary 2013) have argued that FDI and trade openness help improve tax revenue collections in many countries worldwide. Based on this backdrop and ongoing dichotomy, this study considers the role of FDI and trade liberalization in improving tax revenue collection in selected sub-Saharan African countries.

Objectives of the Study

The study specifically seeks to accomplish the following:

i.

Determine the effect of foreign direct investment on tax revenue collection in certain sub-Saharan African countries.

ii.

Examine the influence of exportation to other countries on tax revenue growth in selected sub-Saharan African countries.

iii.

Evaluate the level at which the importation of goods and services affects tax receipts in the particular sub-Saharan African countries identified in this study.

iv.

Confirm the impact of exchange rate fluctuations on tax revenue collection in designated sub-Saharan African countries.

2. Literature Review

2.1. Conceptual Review

A range of economic factors shaping the global economy has underscored the critical roles of free trade and foreign direct investment as fundamental drivers of economic growth (Chibalamula et al. 2023; Yeboah 2024). Nguyen and Darsono (2022) affirm that tax income constitutes one of the most significant variables affecting the prosperity of a nation. Overseas direct investment is a critical component that fosters national competitiveness and boosts the economy by transferring technological advances, developing new management skills, facilitating overseas commerce, and increasing productivity among companies (Gaspareniene et al. 2022). FDI contributes to the general advancement of an economy by serving as an essential source of capital, creating employment opportunities, boosting domestic capital, fostering exports, updating technology, and improving competence and operational effectiveness (Agrawal and Khan 2011; Bayar and Ozturk 2018; Magombeyi and Odhiambo 2017). FDI is often seen as having a positive economic influence; however, it may also have detrimental consequences on national economic growth (Gaspareniene et al. 2022). The growing recognition of these underlying facts by African leaders is contributing to an increased value of the continent’s sectoral fragmentation, while the separation from global markets is increasingly seen as a reasonable strategy (Oloyede et al. 2021).

2.2. Theoretical Review

Nguyen et al. (2013) offer a brief theoretical foundation of the significance of FDI in tax collection. The contributors created an empirical representation of FDI’s effects on tax income by identifying three key channels: demand generation, technical spread, and competitiveness. The business development effect is based on the theory that FDI generates income from taxes through direct and indirect consumption. FDI directly generates demand by procuring supplies regionally, resulting in downstream consumption (Nguyen et al. 2013). FDI creates a secondary market demand for materials by increasing the production levels of local intermediary businesses and encouraging the emergence of fresh enterprises. The spike in activity in the country of residence caused by multinational enterprises suggests a corresponding increase in tax income. Technology spread can have both beneficial and harmful effects. Indigenous enterprises’ capacity to absorb information significantly influences the impact of FDI technology transfers on tax collection. Firms with a high ability to absorb information can use transmitted information to improve and increase productivity (Jemiluyi and Jeke 2023). Firms with limited capability to absorb information struggle to keep up with changing technology trends through heavy investments in technology. The entire process leads to competition and a reduction in tax collection, as firms consider investment in technology a financial priority. Competition between native firms and foreign investments may have both beneficial and negative effects. By establishing retroactive connections with the remaining sectors of the financial system, FDI has the potential to produce beneficial spillovers in the economy hosting it. Greater effectiveness can lead to improved performance and domestic output, resulting in a higher tax income. Unfavorable competitiveness can hurt the host economy by reducing regional investment, leading to decreased production and lower tax income (Haddad and Harrison 1993; Nguyen et al. 2013).

The erosion of the domestic tax base and the shifting of profits, commonly referred to as BEPS, exemplify the methods employed by multinational corporations to utilize transfer pricing and thin capitalization strategies to relocate profits from African nations. BEPS encompasses tax planning techniques that multinational enterprises adopt to take advantage of gaps in tax legislation, thereby artificially transferring profits to jurisdictions with minimal or no taxation to evade tax obligations (OECD 2024). The OECD/G20 BEPS Project provides governments with the necessary regulations and tools to combat tax avoidance, ensuring that profits are taxed in the locations where the economic activities that generate them occur and where value is created (OECD 2024).

2.3. Empirical Review

2.3.1. Does FDI Have Any Negative Connection or Effect on Tax Earnings?

ALshubiri (2024) researched the impact of foreign direct investment inflows on tax income in 34 industrialized and developing nations between 2006 and 2020. The outcomes reveal a strong adverse long-term and favorable short-term association between FDI inflows and earnings from taxes in industrialized nations. At the same time, there was a strong optimistic, long-term association and a negative short-term association between FDI inflows and developing country tax collections. Jemiluyi and Jeke (2023) examine the influence of foreign direct investment on generating tax income in South Africa. The study finds that FDI has a significantly negative impact on tax income. Duong and Nguyen (2022) investigated whether grafts, underground economic status, and foreign direct investment impacted revenue from taxation gathering in BRICS countries. According to the investigation, controlling bribery had a significant beneficial influence on the collection of taxes, but the magnitude of the informal economy had a deteriorating association with BRICS nations’ tax revenues. The findings also reveal that the amount of FDI has a significantly negative impact on these economies’ tax revenues. Furthermore, the study found that GDP per capita and farming have a negative impact on revenue from taxes. In contrast, the overall Governance Indicator had a rather large positive impact on BRICS countries’ earnings from taxes.

Serin and Demir (2022) utilized the system GMM estimator to analyze the impact of FDI on business tax collections in 35 OECD countries from 2005 to 2020. The investigation confirms that the consequences of FDI on corporate taxes are restricted and detrimental. Gondo et al. (2021) examine the influence of external FDI on Indonesian local investment between 1980 and 2018. Using a vector error correction model, this study discovered that FDI has a considerable negative impact on national investment.

2.3.2. Does FDI Have Any Positive Relationship or Influence on Tax Revenue?

Feld and Heckemeyer (2011) conducted a meta-analysis of FDI and taxes using 46 research papers and found no association between tax and FDI. Odabas (2016) investigated the causal correlation between taxation and FDI receipts in seven developing EU countries from 1996 to 2012. They observed a one-way causal link between FDI earnings and tax collection. Nguyen et al. (2022) utilized data from 1986 to 2020 and applied the autoregressive distributed lag bounds checking approach to investigate the relationships among economic growth, foreign direct investment, exports of goods, and importation in Vietnam. The analysis found a long-run association, demonstrating that FDI greatly boosted economic development, whereas exports and imports had no statistically significant influence on economic expansion.

Mahmood and Chaudhary (2013) sought to determine the influence of foreign direct investment on Pakistan’s tax income. Foreign direct investment and GDP per person employed were utilized as independent variables, with tax revenue as the dependent factor. The analysis identifies long- and short-term linkages in the model. Foreign direct investment and GDP per person employed have a favorable and considerable effect on tax revenue. Thus, the analysis finds that foreign direct investment contributes positively to Pakistan’s tax collection. Khan and Hye (2014) examined the influence of trade liberalization on FDI inflows in Pakistan using time-series data from 1971 to 2009. The findings revealed that liberalization measures, such as the financial liberalization index, trade permeability, and actual interest rates, had an adverse impact on FDI inflows in Pakistan. Tax income from products also has a detrimental effect on FDI. On the other hand, gross fixed capital formation, amenities, and price increases all had a favorable impact on Pakistani foreign direct investment. Market strength has little effect on FDI.

Balıkçıoğlu et al. (2016) used a dataset of Turkish production companies from 2004 to 2012 to study the impact of FDI on tax collected in Turkey. This study focuses on the differences among companies with various technological capabilities. The findings reveal that foreign affiliations enhance the organization’s tax payments. The investigation indicated that FDI had greater effects on the taxation of technologically advanced enterprises than on that of moderate or low-technology organizations. Gnangnon (2017) investigated the effects of FDI inflows on administrative income, particularly overall non-resource tax revenue and non-resource corporation tax revenue. This research is based on an imbalanced panel dataset of 172 nations (both industrialized and emerging) between 1980 and 2013. Empirical evidence demonstrates that the impact of FDI inflows on each of these two categories of government income is proportional to the magnitude of FDI inflows, represented as a percentage of the host country’s GDP.

Saini and Singhania (2018) explore the possible factors of foreign direct investment in both developed and developing countries. The outcomes varied among the regions. In advanced economies, FDI prioritized policy-related drivers such as GDP expansion, openness to trade, and the liberty index, but in developing countries, FDI exhibited a beneficial association with fiscal variables comprising gross fixed capital formation, trade openness, and efficiency factors. According to Basheer et al. (2019), various national budgetary and macroeconomic parameters, including surplus money shortfalls and FDI net infusions, have a major impact on tax income. Binha (2021) demonstrated that foreign direct investment significantly affected the growth of Zimbabwe’s tax base. The beneficial effects of FDI on tax income indicated that it facilitated advantageous technological transfers, which improved the productivity of local enterprises, thus leading to an increase in tax collection by the Zimbabwean government.

Manh and Sang (2022) investigate the influence of foreign direct investment as a means of budgetary and operational decentralization on economic growth in Vietnam, an example of a rising economy. The findings reveal that budgetary decentralization, as measured by increased tax income and development expenditure on investments, has a favorable and major influence on foreign investment attractiveness. Kastrati and Vokshi (2022) investigated the effect of the tax burden on promoting foreign direct investment and growth in Kosovo. The study tests the comparative evaluation technique results by examining the influence of taxation on recruiting FDI and driving economic growth. The findings revealed that Kosovo was less successful than other nations in the area with regard to implementing financial strategies designed to draw foreign direct investment and drive revenue growth. Omodero et al. (2022) examine the effectiveness of ICT taxes and FDI in improving overall tax revenue. The research revealed that ICT taxes had a positive and significant influence on tax revenue, but FDI inflows had an inconsequential beneficial effect on earnings from taxes.

Ha et al. (2022) use an equal number of data points from eight nations to identify the factors that influence tax income in Southeast Asia. The study found that economic openness, FDI, proportion of foreign debt to GDP, and percentage worth contributed by industry to GDP all had beneficial effects on tax collection, but government development aid had negative consequences. Gaspareniene et al. (2022) examined the importance of FDI and its influence on tax income and productivity, with an emphasis on the EU economy. An independent study was undertaken to investigate the link between inbound and outbound FDI and tax income, using data from 1999 to 2019. The results show that outbound FDI had a large positive influence on overall tax income. Conversely, inward FDI has a negative effect on tax revenue.

Camara (2023) used a system GMM estimator to provide empirical evidence of the critical impact of FDI inflows on tax revenue mobilization for 90 developing countries between 1996 and 2017. The statistics clearly demonstrate that FDI inflows significantly increased tax income. Nonetheless, this impact was not detected in resource-exporting nations, where tax revenues appear substantially unresponsive to foreign direct investment inflows. Hani and Warad (2023) seek to illuminate the realities of tax revenue and identify its drivers in Jordan. This was accomplished by examining the influence of several economic and non-economic variables on tax revenue. The economic variables were real GDP, the index of consumer prices, investment from abroad, and commercial liberalization; non-monetary elements included malfeasance and tax code revisions. The analysis reveals that real GDP and CPI are the primary drivers of tax revenue in Jordan. It was also shown that grafts have a considerably harmful influence on tax income, but changes to tax legislation have no obvious or small influence on tax earnings.

2.3.3. How Does Trade Openness and FDI Affect Tax Revenue and GDP Growth?

Ho et al. (2023) used statistical data from 29 emerging economies with rapid GDP growth from 2000 to 2020 to examine the influence of tax income on financial improvement in the setting of increasing international trade access. The study finds that tax income has a beneficial impact on the overall expansion of the economy. In addition, the study discovered that trade openness strengthened the favorable association between revenues from taxes and economic expansion, whereas extreme trade openness diminished performance. Abdulrazak and Abdikani (2023) evaluate the uneven impact of FDI on GDP growth in Somalia between 1977 and 2021. According to this study, an increase in FDI boosted Somalia’s growth rate, but an adverse change had the opposite effect. The Wald assessment revealed that foreign direct investment had an unbalanced influence on revenue growth in immediate and distant futures. Furthermore, openness to trade and rising prices greatly slowed economic expansion in both the short and the long term. The accumulation of assets has increased long-term economic development.

Shrestha et al. (2021) investigate the influence of foreign commerce on fiscal diversity, which is an essential problem for resource-constrained economies. From 1996 to 2014, an autoregressive distributed lag model was applied to panel data from thirteen different countries. Estimates found a detrimental long-term relationship between fiscal resource earnings and freedom of trade. The analysis demonstrated that resource-rich nations that push trade expansion typically deconstruct the resource-dependent composition of government income. Alege and Osabuohien (2015) analyzed the relationship between international trade and exchange rates in sub-Saharan African (SSA) countries. Their empirical findings revealed that both exports and imports were largely inelastic to changes in exchange rates. The anticipated effects of currency depreciation in the SSA region did not materialize, likely due to the specific economic structures and export compositions present. Consequently, rather than reducing imports, currency depreciation only served to aggravate the balance of payments issues.

3. Methodology

This study includes tax revenue statistics for 33 African countries: Botswana, Burkina Faso, Cabo Verde, Cameroon, Chad, the Republic of Congo, the Democratic Republic of Congo, Côte d’Ivoire, Egypt, Equatorial Guinea, Eswatini, Gabon, Ghana, Guinea, Kenya, Lesotho, Madagascar, Malawi, Mali, Mauritania, Mauritius, Morocco, Namibia, Niger, Nigeria, Rwanda, Senegal, Seychelles, Sierra Leone, South Africa, Togo, Tunisia, and Uganda. The analysis applies the widely used approach for categorization of government income outlined in the OECD Analytical Instructions to African countries, allowing for comparisons of revenues from taxes and structures not only throughout the continent as a whole, but also with the OECD, Latin America and the Caribbean, and Asia and the Pacific. The data on African nations contained in this article are also available in the OECD’s Global Revenue Statistics database, which is a valuable resource for analyzing domestic resource mobilization. Data for this inquiry were gathered from the OECD Centre for Tax Policy and Administration, OECD Development Centre, African Union Commission (AUC), and African Tax Administration Forum (ATAF).

Table 1. Measurement of variables.

Variable Codes	Description	Measurement	Source
LNTXR	Tax revenue	Billions of USD	OECD
LNFDI	Foreign Direct Investment inflows	Billions of USD	WDI
LNEXP	Exports	Billions of USD	WDI
LNIMP	Imports	Billions of USD	WDI
LNXGR	Exchange rate	In percentage	WDI

Source: Research output, 2024.

describes the variables used in this study and their sources.

Model Specification

This study investigates the effect of foreign direct investment and trade openness on tax earnings in sub-Saharan Africa using selected countries that comprise Nigeria, Ghana, Kenya, and South Africa. The study period is from 1990 to 2022. For empirical estimation, the model is established as follows:

$$LNTXRt-1={\alpha }^0+{\alpha }^{1}LNFDIt-1+{\alpha }^{2}LNEXPt-1+{\alpha }^{3}LNIMPt-1+{\alpha }^{4}LNXGRt-1+\mu t$$

(1)

$NTXRt$ represents the total tax revenue in sub-Saharan Africa and serves as the response variable in this study. $LNFDIt$ is the foreign direct investment inflow, $LNEXPt$ is the amount of export of goods and services in sub-Saharan Africa, $LNIMPt$ indicates the amount of import of goods and services, $LNXGRt$ is the exchange rate that is built into the model as a control variable, ₀ is the numerator, ₁–₄ is the co-integrating matrix to be anticipated, and $\mu t$ is the traditional random error. With the exception of the exchange rate, which is collected as a percentage, all data on relevant factors are given in billions of US dollars before conversion to natural log form. The pooled mean group/autoregressive lag model is used in econometrics to calibrate the longstanding relationship between two or more parameters, based on stability metrics. To determine the short and long-run effects of FDI and commerce openness on tax revenue collection in Africa, pooled mean group (PMG) is specified as follows:

$$∆\gamma it=\sum _{k-1}^{p-1}\lambda \ast ik \Delta yi,t-k+\sum _{k-1}^{q-1}{\delta }^{\prime }ik\Delta xi,t-k+\phi i\left(yi,t-1+{\beta }^{\prime }iXit\right)+wi+\epsilon it$$

where:

The following apply:

Xit = Vector of explanatory variables for group, which may be I(0), I(1), or mix;

$\lambda ik, {\delta }^{\prime }ik$ = Short-run coefficients

$\lambda ik$ = Coefficients of lagged dependent variables (Scalers)

${\delta }^{\prime }ik$ = Coefficient vectors

$\phi i$ = Group-specific error-correction coefficients

$\beta i$ = Vector of long-run coefficients;

$wi$ = Group-specific fixed effects error term; $\epsilon it$ = Error term.

By implication, the model provides assumptions as follows:

$$∆\gamma it=\sum _{k-1}^{p-1}\lambda \ast ik \Delta yi,t-k+\sum _{k-1}^{q-1}{\delta }^{\text{'}}ik\Delta xi,t-k+\phi i\left(yi,t-1+{\beta }^{\prime }iXit\right)+wi+\epsilon it$$

where the following apply:

The pooled mean group (PMG) long-run coefficients assume that $\beta i$is the same across groups (i.e., cross-sections), while the short-run (${\delta }^{\prime }ik$) and co-integrating ($wi$) coefficients vary across groups. The dynamic fixed effect (DFE) assumes that $\beta i,$ $\lambda ik, {\delta }^{\prime }ik$ is the same across groups. The mean group (MG) assumes that that $\beta i,$ $\lambda ik, {\delta }^{\prime }ik$ varies across the groups.

4. Results

Section 4 provides insights into the outcomes of the data analysis carried out in this research.

The trends in the figures used in this study from to 1990 to 2022 are displayed in Figure 1. A closer look shows that the tax revenues of Kenya and South Africa were higher in 2017 and 2021, respectively. For FDI, Ghana experienced the highest foreign investment inflows compared to other countries. South Africa has the highest exports and imports of goods and services, while the exchange rate fluctuation is at its peak in Nigeria from 2018 to 2022.

We used the descriptive statistics listed in

Table 2. Descriptive statistics.

	TXR	EXP	FDI	IMP	XGR
Mean	10.87	8.941	7.227	9.075	4.495
Median	11.60	8.838	6.082	8.809	4.488
Maximum	14.41	11.81	15.17	11.76	4.890
Minimum	6.216	5.697	0.000	5.986	4.204
Std. Dev.	2.796	1.793	4.264	1.654	0.168
Skewness	−0.233	0.075	0.795	0.144	0.239
Kurtosis	1.484	1.873	2.423	1.939	2.071
Jarque–Bera	9.223	4.733	10.51	4.431	4.008
Probability	0.993	0.093	0.522	0.109	0.135
Sum	956.5	786.7	636.0	798.6	395.6
Sum Sq. Dev.	680.2	279.7	1581	238.0	2.465
Observations	88	88	88	88	88
Correlation matrix
TXR	1.000
EXP	0.183	1.000
FDI	−0.378	0.347	1.000
IMP	0.261	0.678	0.371	1.000
XGR	−0.274	−0.113	−0.088	−0.063	1.000

Source: Authors’ calculation, 2023.

to describe the nature of the dataset employed in this investigation. The major aim of this test was to check the normality of the datasets used in this study before applying the model. Again, we also deemed it appropriate to look out for the presence of multi-collinearity by checking the type of relationship existing among the variables under study. From Table 2, it is confirmed that the datasets are normally apportioned, as validated by the Kurtosis and Jarque–Bera p-values of 0.993, 0.093, 0.522, 0.109, and 0.135 for TXR, EXP, FDI, IMP, and XGR, respectively. All Jarque–Bera p-values are above the 0.05 threshold of significance, while the kurtosis is between 2 and 3 acceptable values.

Having established that the datasets are normal and applicable for this investigation, we also went further to check the unit root of the series to avoid spurious regression analysis and to guide the selection of a suitable econometric tool for the analysis.

The results of the unit root test, which was performed using various criteria, are summarized in

Table 3. Unit root test.

Variables	Levin, Lin, and Chu t	Im, Pesarann, and Shin W-Stat	ADF-Fisher Chi-Square	PP-Fisher Chi-Square	Order of Integration
LNTXR	−4.595 (0.000)	−5.213 (0.000)	43.53 (0.000)	299.9 (0.000)	I(1)
LNFDI	−2.259 (0.011)	−1.928 (0.027)	16.59 (0.035)	15.86 (0.044)	I(0)
LNEXP	−6.372 (0.000)	−5.973 (0.000)	44.92 (0.000)	45.81 (0.000)	I(1)
LNIMP	−4.405 (0.000)	−2.454 (0.007)	18.71 (0.016)	22.06 (0.005)	I(0)
LNXGR	−4.380 (0.000)	−4.938 (0.000)	37.70 (0.000)	44.65 (0.000)	I(1)

Source: Authors’ calculation, 2023.

. The results show that all criteria applied established that LTXR, LNEXP, and LNXGR are stable at order one or first difference, whereas LNFDI and LNIMP are stationary at level or order zero. Therefore, in accordance with Pesaran et al. (2001), the combination of stationarity level at order zero and one shows the need for the application of PMG ARDL to check co-integration in the model.

The study determined that the series are stationary at orders zero and one, implying that there is a strong possibility of a joint long-term equilibrium association among all of these factors. Pedroni’s test was used to determine the presence of any mutual long-term relationships among variables (Pedroni 2004).

Table 4. Pedroni residual co-integration test.

Series: LNTXR LNEXP LNFDI LNIMP LNXGR
			Weighted
	Statistic	Prob.	Statistic	Prob.
Panel v-Statistic	−0.588	0.722	−0.907	0.818
Panel rho-Statistic	−0.337	0.368	−0.508	0.305
Panel PP-Statistic	−2.727	0.003	−3.012	0.001 ***
Panel ADF-Statistic	−0.806	0.209	−0.818	0.206
	Statistic	Prob.
Group rho-Statistic	0.774	0.780
Group PP-Statistic	−1.858	0.031 ***
Group ADF-Statistic	−0.220	0.412

Authors’ calculation, 2024.

presents the results of the Pedroni test. The results of three of the parameters used showed a p-value of less than 0.05, indicating the presence of a collective long-run relationship. Hence, we employed the PMG ARDL estimation approach.

Table 5. PMG ARDL estimation result.

Dependent Variable: D(LNTXR)
Variable	Coefficient	Std. Error	t-Statistic	Prob. *
	Long-Run Equation
LNFDI	−0.179	0.089	−1.998	0.051 **
LNIMP	7.258	1.975	3.675	0.000 ***
LNEXP	−7.515	2.427	−3.095	0.003 ***
LNXGR	0.131	0.645	0.204	0.839
	Short-Run Equation
ECT	−0.337	0.323	−1.045	0.300
D(LNTXR(−1))	−0.205	0.150	−1.365	0.178
D(LNFDI)	0.025	0.0235	1.065	0.291
D(LNIMP)	−1.014	1.195	−0.849	0.399
D(LNEXP)	1.418	1.417	1.000	0.321
D(LNXGR)	0.068	0.393	0.175	0.862
C	4.141	3.756	1.102	0.275

Source: authors’ calculation, 2024 (*** indicates significance level at 1%; ** shows significance level at 5%; * implies significance level at 10%).

displays the results for both long- and short-run PMG ARDL estimations. It is revealed that in the long run, FDI has a statistically significant and negative impact on tax revenue in the sub-Saharan African countries selected for this investigation. The t-statistic is −1.998 with a p-value of 0.051, which implies that at the 5% level of significance, foreign direct investment in the selected regions of Africa significantly and negatively affects the tax revenue collection of the countries. The result has given credit to the findings of other authors (ALshubiri 2024; Cagé and Gadenne 2018; Gnangnon and Brun 2019; Gondo et al. 2021; Jemiluyi and Jeke 2023; Kastrati and Vokshi 2022; Khattry and Rao 2002; Khattry 2003; Serin and Demir 2022; Shrestha et al. 2021). The result for exports of goods and services also indicates a significant adverse consequence (t-statistic = −3.095; p-value = 0.003) on tax revenue, while the importation of products and services becomes positively significant (t-statistic = 3.675; p-value = 0.000). The results corroborate the GMM result in Appendix B. In this case, the tendency that sub-Saharan African economies import more than they export is obvious, and the implication is that it leads to an unfavorable balance of payment as well as high currency fluctuation. Thus, trade openness becomes less beneficial because importation overrides and dominates the framework. The correction of the entire anomaly experienced in the long run is shown by the result of the ECT (−0.337) which indicates that the negative effects of FDI and trade openness on tax revenue in the previous year can return to equilibrium in the current year with an adjustment of 33.7%.

For country-specific results in

Table 6. Panel ARDL country-specific short-run estimation.

	Nigeria	Ghana	Kenya	South Africa
ECT	21.85 *	−178.4 *	−14.17 *	−354.5 *
D(LNTXR(−1))	−12.41 *	−0.701	1.976	−10.47 *
D(LNFDI)	12.26 *	−9.978 *	9.586 *	425.5 *
D(LNEXP)	−10.80 *	0.076	1.081	4.633 ***
D(LNIMP)	14.93 *	5.010 ***	−1.142	8.534 *
D(LNXGR)	4.508 ***	−9.909 *	−0.101	−36.42 *
C	4.809 ***	18.23 *	0.487	11.47 *

Source: Authors’ computation, 2024 (See Appendix A) (* Significant at 1%; *** Significant at 5%).

, Ghana’s short-run disease will be normalized in the long run at an adjustment speed of 178.4%, Kenya’s at 141.7%, and South Africa’s at 354.5%, but Nigeria’s is positive at 218.5%. However, in the short run, FDI positively impacts tax revenue in Nigeria, Kenya, and South Africa, whereas the result indicates that it has a negative effect on Ghana’s tax revenue. Nigeria’s level of exportation is harmful to tax revenue collection, while South Africa’s export of goods and services is beneficial to tax revenue at the 5% significance level, while the others are not tangibly represented. In addition, importation and exchange rates benefit Nigeria, Ghana, and South Africa in the short term.

The criteria for the model selection are shown in Figure 2 and Table 6. The study used AIC, which gives the best model at 2, 1, 1, 1, 1, as indicated in Figure 2, and the lowest criterion of −0.948, as indicated in

Table 7. Model selection criteria table.

Model	LogL	AIC *	BIC	HQ	Specification
3	65.91	−0.948	−0.114	−0.614	ARDL(2, 1, 1, 1, 1)
4	79.33	−0.883	0.427	−0.358	ARDL(2, 2, 2, 2, 2)
1	58.33	−0.858	−0.144	−0.572	ARDL(1, 1, 1, 1, 1)
2	72.46	−0.811	0.379	−0.334	ARDL(1, 2, 2, 2, 2)

.

Diagnostic tests were conducted (results shown in Figure 3 and

Table 8. Residual cross-section dependence test.

Test	Statistic	d.f.	Prob.
Breusch–Pagan LM	12.54	6	0.051
Pesaran scaled LM	0.734		0.463
Pesaran CD	1.389		0.165

Authors’ calculation, 2024.

) to assess the reliability and efficacy of the selected model. The cross-sectional dependence test determines whether various cross-sections (countries) are mutually reliant and interrelated. According to (Baltagi 2005), there might be a relationship between the cross-sections, causing a number of issues with the model. Table 8 clearly demonstrates that all the calculated probabilities of the three parameters are greater than the critical point of 0.05. Essentially, we fail to reject the null assumption that the cross-sections are not reliant on one another in the long run, which reinforces the outcomes of the PMG estimation technique, which supposes that the cross-sections are uniform over time. Figure 3 also shows that the objectives of this study were well articulated and achieved.

5. Conclusions

This study investigates the effects of foreign direct investment and trade openness on tax revenue collection in selected sub-Saharan African countries, including Nigeria, Ghana, Kenya, and South Africa. From the results of this study, FDI exerts a strong negative influence on tax revenue in the long run, but in the short run, as shown in Table 6, Ghana is also adversely affected, while other countries benefit from FDI only in the short run. The policy implication is that sub-Saharan African countries should learn to maintain a fiscal strategy to keep FDI inflows an advantageous endeavor, both in the short and long run. The fact is that these foreign investors come with initial attraction and try as much as possible to comply with all the tax regulations of their host economy. Over time, they devise ways to transfer resources from their host countries to avoid taxation. This is accomplished through exportation of their products to countries that practice tax havens or countries with less tax burden, thereby taking advantage of transfer-pricing strategies. Therefore, this study recommends favorable tax policies that prevent foreign firms from employing transfer pricing strategies to decrease the controlling company’s aggregate tax liability. The overall fiscal policies in sub-Saharan African countries should be geared towards boosting tax revenue and not diminishing it through harsh tax reforms and policies. The study also recommends an increase in industrialization to boost the export of goods and services to avoid adverse balance of payment and frequent currency fluctuations.