Relationship Between Export and Economic Growth: Evidence from West African Countries

Abstract

The export-led growth hypothesis (ELGH) postulates that export is a major driver of economic growth. This study tests this hypothesis and further analyses the determinants of exports in the case of West African countries. An annual panel data spanning from 2008 to 2018 was used. Findings from the system GMM and OLS estimations validate the ELGH in West Africa. The results also reveal that foreign direct investment, employment, remittances, land area and infrastructure are significant boosters of export while population, real effective exchange rate and taxes on international trade are detrimental to export performance in the region. The study recommends the relaxation of taxes especially on international trade to encourage businesses that produce to feed the export sectors, provide an enabling environment for businesses and also attract foreign investors.

JEL Codes: F14, F21, O55

Keywords

Export performance economic growth foreign direct investment West Africa

I. Introduction

The achievement of sustainable economic growth is a basic macroeconomic goal for almost all countries around the globe. The income level of individuals which is a major determinant of welfare is as well-argued to be influenced by growth at the aggregate level (Barro & Sala-i, 2004). Thus, economic growth is identified to have a significant association with the welfare of citizens (Oosterbaan et al., 2000). A variety of empirical studies have explored the relationship that exists between economic growth and variables such as trade openness, foreign direct investment, inflation, exchange rate, domestic credits to the private sector, domestic savings and investment (Ayibor, 2012; Dabel, 2016; Freund & Bolanky, 2008). Recommendations from such studies and some macroeconomic theories have informed the introduction of a series of interventions and policies geared towards growth attainment.

Trade openness particularly export is becoming increasingly relevant as it has proven to be a major catalyst through which economic growth and development can be improved (Didia et al., 2015). Not all writers, however, have been able to argue that there is a single direction in the impact of export on growth (Sannassee et al., 2014). Contrary to the export-led growth hypothesis (ELGH), the neoclassical theorists on trade reasoned that growth may Granger cause export. Thus, economic growth positively affects the factor endowment of an economy and makes the economy a more competitive one internationally. Again, the radicalists suggest an opposing view that less developed countries’ exports to the developed industrialised countries are instruments for exploiting the poorer countries. They argue that terms of trade of economies are worsened by growth in exports. Less developed economies import basic food items regardless of several goods exported. According to Sannassee et al. (2014), arguments have been made those efforts to expand export policies make an economy vulnerable to ideas of foreign demands as well as introducing competitions between less developed countries and developed countries.

The argument to some extent explains that export may not always be growth-enhancing for all economies hence the need for empirical tests. It is important to note that ELGH has been successful for instance in East Asian countries. It is, therefore, a wake-up call for West Africa and other regions. Though several studies have highlighted this relationship, this study contributes to existing knowledge by re-echoing the validation of the ELGH in West Africa, a region that is striving to grow through export enhancement. By way of contribution, the study also unearths some significant and dominant drivers of export performance for the attention of policymakers and actors in the sector. This is an extra effort to make known some crucial variables to be attended to in the quest to fully gain from export. For the West Africa sub-region, a study of this nature is in the right step owing to the attempts by member countries to improve on their export volumes and the relatively limited (if not total absence) of such a study focusing on the region. Thus, this effort is in line with the region’s agenda of expanding its quota of exports to deepen regional integration and stimulate structural transformation that enhances growth, creates employment and reduces poverty. More importantly, the literature on this subject area is supplemented with new findings on the dominant drivers of export performance after revealing the relationship between export and economic growth in the case of West Africa.

Export and Growth in West Africa

International trade in the Economic Community of West Africa (ECOWAS) is dominated by several products and as a result of the mass of fuels from the extractive industries, there is the generation of local value addition. Fuel represents three-quarters of West Africa’s exports, with Nigeria providing about 73% of it. Cocoa and cocoa food preparation are next, representing 5%, precious stones cover 3%, while edible fruit, rubber, secondarily cotton, wood and wood products, shellfish and fish all form about 1% each. With export destinations, about 28% of ECOWAS exports enter Europe, with 23% for the European Union. About 40% are accounted for by the Americas where 34% for the Free Trade Association of North America and Mexico, Canada and the United States account for 24%.

In most countries in West Africa, economic growth has not been sufficient to significantly reduce poverty. Farmers and firms in the region produce and trade in markets that are highly localised and unable to gain adequate economies of scale which is required in attracting broad-based investment that can induce growth and reduction in poverty. Several factors account for that. These include trade barriers along the borders such as taxes, insufficient reliance on power and communication networks, heavy dependence on family and informal financing sources. Also, trade in services which is needed to enhance growth in West Africa is hindered by regulatory, institutional and infrastructural constraints.

The strategy of USAID/West Africa is to address major constraints to competitiveness through regional organisations and private sector associations. It is also a plan to showcase the productive potential of West Africa as a means to activate greater investment in the region. The programme is implemented via the West Africa Trade Hub in Accra, Ghana, while coordinating closely with a network of partners of African regional private sector and public institutions such as ECOWAS and the West Africa Economic Monetary Union. The Trade Hub provides assistance to farmers and firms in meeting product quality standards and requirements of the markets and also to produce in large quantities. Currently, there have been efforts to ensure that the problem of information asymmetry on the foreign markets, hindrances to free movements, the situation of land-locked nations and the likes are addressed. These are conscious efforts to propel growth within the region.

Economies in West Africa are different in many dimensions of development. The estimated real GDP growth for West Africa in 2018 was 3.3% (AEO, 2019). This is an improvement of 2.7% in 2017. Though faster than it was for Southern and Central Africa, GDP growth in West Africa between 2014 and 2017 lag behind for the whole of Africa. Lower commodity prices and reduced oil production in Nigeria (the largest economy in the region) resulted in sluggishness in the growth of the region (AEO, 2019). In 2018, investment and positive net exports were the main drivers (demand-side) of GDP growth.

Figure 1 shows the average export as a percentage of GDP for West African countries for the period 1980–2018. Mauritania recorded the highest export as a percentage of GDP (about 42%) for the period, while Burkina Faso recorded the lowest (about 14%) for the period. Indeed, the West Africa region has recently introduced an agenda of expanding its export quota as a means to deepen regional integration and structural transformation so that growth in the sub-region can be achieved. This study, therefore, focuses on West Africa as a means to provide empirical evidence to support and guide the achievement of the region’s agenda.

Figure 1.

Source: Author’s construction using data from World Development Indicators (WDI).

The remaining sections of this article are structured as follows: Section II reviews literature on the export–economic growth nexus and the drivers of export. The methodology is explained in Section III, the analysis and discussion are done in Section IV, while the conclusion and recommendation are presented in Section V.

II. Literature Review

Effect of Export Performance on Economic Growth

The export and growth nexus has for long been a subject of high interest to economists. In achieving economic growth, the ELGH has been implemented by most economies owing to the importance of the export. The ELGH theory suggests that export is one of the major drivers of economic growth. According to Medina-Smith (2001), export contributes to the overall growth of the economy and not only capital and labour. An extension of this idea is explained in the current endogenous growth model which explains the role exports play towards growth in the long run via technological innovations.

The relationship has been explained by several empirical studies as well. Thus, the short- and long-run relationship between export performance and economic growth in Nepal, a landlocked developing country was examined using annual data from 1980 to 2018 was examined by Thapa-Parajuli et al. (2021). The ARDL bounds testing approach to cointegration was used in determining the association between the regressors and economic growth. The results estimated indicate that there is a positive and statistically significant effect of export performance on economic growth. The Granger causality test also shows a unidirectional causality from economic growth to export. Again, Abogan and Akinola (2014) studied the impact of oil revenue on Nigeria’s economic growth. For the period between 1975 and 2009, the study made use of the error correction model and concentrated on the short-run and long-run effects. The findings of the study revealed that in the short term and long term, oil export has a statistically significant effect on Nigeria’s economic growth.

Using the vector autoregressive model as well as the Granger causality test, Lazarov (2019) analysed the effect of export sophistication on the growth of the Macedonian economy. The findings of the study revealed that export from Macedonia has witnessed a significant increase in trend in the last 20 years but the structure of the export is highly concentrated and made up of products with limited complexity levels. In addition, the findings revealed that for the period 1995–2017, there is a positive and statistically significant association between export sophistication and economic growth.

Malhotra and Kumari (2016) studied the impact of export performance on the economic growth of Japan, China and South Korea, the three major East Asian economies. The augmented Dickey–Fuller and the Phillips–Perron unit root tests were performed to check the stationarity of the annual data covering the period 1980–2012. The Johansen cointegration test for long-run relationship, the vector error correction model for short-run dynamics, the impulse-response functions as well as variance decomposition analysis (VDA) were used in analysing the relationship between the variables. The findings show that there is a cointegration for all variables for the East Asian economies. Thus, the results support the ELGH.

Drivers of Export Performance

In studying the factors determining export performance, a lot of variables have been considered. Both factors originating from within the countries under consideration (internal) and outside the countries being considered (external) have been probed by different studies. As the current study considers both categories of factors, the review of empirical studies took into consideration either the internal or external factors. For instance, a firm-level analysis on the drivers of export performance and competitiveness was performed by Brancati et al. (2018) during the current crisis in Italy. The findings explain that though structural factors have an impact on the major role of external competitiveness, other critical firm-level factors, specifically those that are linked to strategic profiles, technological capabilities as well as proactive behaviour. Again, Ngo et al. (2016) provide a conceptual model based on the institutional-based view that reveals the effect of several domestic institutional attributes on the performance of the export. The structural equation modelling (SEM) analysis was tested using data from 109 sampled Vietnamese exporters. The findings show that the institutional attributes (thus, stability, predictability, enforceability and specificity) positively increase export performance.

Also, in studying the internal and external determinants of export promotion, Tariq Majeed and Ahmad (2006) employed panel data for developing countries. The study found that foreign direct investment, GDP growth, official development assistance, savings, industrialisation, the number of televisions as well as the number of telephones are significant variables that positively affect the performance of the export. Consistent with theory, a significantly positive relationship was also found between the exchange rate and the performance of the export.

Also, using 448 large Brazilian manufacturing firms, Carneiro et al. (2011) explored how the characteristics of firms, the strategy of firms and the external environment have affected export performance in Brazil. A conceptual model was fit to empirical data using the approach of SEM. In validating the measurement models, an extensive set of procedures were employed. The findings showed that revenues from the export have a strong positive and negative relationship between the status of exporting activity and barriers existing in the host country, respectively. It was also found that profitability from export has a positive and strong relationship with export planning systematisation, while a moderate inverse relationship was found between barriers in the host country and psychic distance. Against the expected, no statistically significant relationship was found between business distance and revenues from export or profitability.

A current study by Haddoud et al. (2019) investigated the importance of both the internal and external factors that determine export performance in Algeria. The study empirically tested a model using data from 103 exporters that operate in Algeria. In general, the findings revealed that the performance of the export is influenced by the context from which the firms operate. Thus, exporters operating in Algeria were found to be driven by a variety of factors as compared to exporters in the developed world. For instance, the study showed that marketing capabilities, managerial resources as well as local relational resources influence exporters.

It is observed that most studies analysing the export and growth relationship also the determinants of exports focus on individual countries and also make use of country-specific variables. A relatively limited number of studies make use of panel data estimation. Again, more importantly, major drivers of export performance and even export and growth relationship differ from region to region, for instance, from developing to developed nation (Haddoud et al., 2019).

While the role of exports in economic growth and development cannot be overemphasised, key issues on the drivers of exports remain controversial. Characteristics of countries or regions which influence the relationships between significant variables such as export and growth differ from one another. This is partly because there are differences in resource endowment, technology and even regulations. Therefore, policy recommendations for countries or regions can best be drawn when the analyses are done on the specific country or region in question. On this subject, relatively limited studies have focused on West Africa, a region endowed in resources, and making effort to strive from the export-led growth strategy. Besides, the relationship between economic growth and exports is often influenced by the possible externalities in the domestic economy (Medina-Smith, 2001). This implies that investigating export as a means of economic growth or testing the ELGH may not be enough. There is, therefore, the need to further reveal the significant drivers of export towards which policy recommendation can be directed. Thus, in advancing the frontiers of knowledge on exports, this study seeks to achieve three key objectives; to determine the relationship between export and economic growth in West Africa, estimate the determinants of export performance in West Africa and identify the dominant drivers of export in West Africa. This is an effort to significantly contribute to the literature on the export-growth relationship and highlight on the significant drivers of export performance in West Africa to aid policy recommendation.

III. Data and Methods

Data

The study made use of secondary data sourced from the World Bank’s WDI. All West African countries (16 countries) were considered for the study using an 11-year period (2008–2018). Table 1 presents the descriptions of the variable and their expected signs. List of countries used for the study are presented in Table AI.

Table 1.

Variable Description and Apriori Signs.

Variables	Description	Apriori Signs
Export	Export of goods and services measured as a percentage of GDP
GDP growth	Annual percentage growth in total GDP	+
Credit	Domestic credit to the private sector measured as a percentage of GDP	+
Population	The population of ages from 15 to 64 years, measured as a percentage of the total population	+ or −
Real effective exchange rate	Nominal effective rate index (a measure of the value of a currency against a weighted average of several foreign currencies) divided by a price deflator or index of costs	−
Foreign direct investment	Foreign direct investment, net inflows measured as a percentage of GDP	+
Tax on international trade	Taxes charged on international trade, include import duties, export duties, profits of export or import monopolies, exchange profits and exchange taxes, measured as a current local currency unit	−
Land area	Geographical land size of exporting countries measured in kilometres squared	+
Remittances	Personal remittances received, measured as a percentage of GDP	+
Employment	The ratio of employment to population, 15 years and above the percentage of the total (ILO estimate)	+
Telephone	Number of fixed telephone subscriptions	+

Model Specification

In analysing the relationship between export and economic growth, we employ the Cobb–Douglas (C–D) production following Mankiw et al. (1992). Thus, assuming marginal contribution of capital and labour in output and in time, the C–D function is specified as

Y (t) = A (t) K {(t)}^{β} L {(t)}^{1 - β}, 0 < β < 1

(1)

$Y$ represents domestically produced output measured in real terms while $K$ and $L$ , respectively, represent capital and labour. Capital ( $K$ ) and labour ( $L)$ ) in this study are proxied by credit to the private sector ( $C R D$ ) and employment ( $E M P)$ ), respectively. The output is also proxied by GDP growth ( $G R O W T H$ ). We assume other variables ( $A (t))$ ) do affect domestic output aside from capital and labour. Hence, expanded Equation (1) is presented as

Y_{t} = δ_{0} + δ_{1} E X P_{i t} + δ_{2} C R D_{i t} + δ_{3} R E E R_{i t} + δ_{4} F D I_{i t} + δ_{5} E M P_{i t} + δ_{6} R E M_{i t} + δ_{7} I N F_{i t} + δ_{8} P O P_{i t} + e_{i t}

(2)

where $δ_{0}$ represents the constant term $E X P$ is export, $C R D$ is credit, $R E E R$ is real-effective exchange rate, $F D I$ is the foreign direct investment, $E M P$ is the employment, $R E M$ is the remittance, $I N F$ is the inflation, and $P O P$ is the population. Again, subscript $i$ represents the country and $t$ represents the time indices. Further, Equation (2) is linearised by logging the variables to obtain Equation 3.

Ln G R O W T H_{t} = δ_{0} + δ_{1} Ln E X P_{i t} + δ_{2} Ln C R D_{i t} + δ_{3} Ln R E E R_{i t} + δ_{4} Ln F D I_{i t} + δ_{5} Ln E M P_{i t} + δ_{6} Ln R E M_{i t} + δ_{7} Ln I N F_{i t} + δ_{8} Ln P O P_{i t} + e_{i t}

(3)

We estimate a dynamic panel model using the GMM estimation technique. The GMM technique is used because it is capable of correcting possible issues of endogeneity, using the lag of both the dependent and independent variables as instruments. For instance, from economic theory, we expect a bi-causality between export and economic growth. Again, the GMM is known to provide a result that is more consistent in the presence of the collection effect. It is important to note that with the cross-sectional units (N = 16) being greater than the period (T = 11) under study, the GMM model is the most appropriate. The GMM model is specified in Equation (5):

\begin{array}{l} Ln G R O W T H_{t} = δ_{0} + δ_{1} Ln G R O W T H_{{(t - 1)}_{i t}} + δ_{2} Ln E X P_{i t} + δ_{3} Ln C R D_{i t} + δ_{4} Ln R E E R_{i t} + δ_{5} Ln F D I_{i t} + δ_{6} Ln E M P_{i t} \\ + δ_{7} Ln R E M_{i t} + δ_{8} Ln I N F_{i t} + δ_{9} Ln P O P_{i t} + e_{i t} \end{array}

(4)

where $G R O W T H_{t - 1}$ represents GDP growth in the previous year. The study further examines the determinants of export performance. We specify the export model taking into consideration the fact that export is influenced by both internal and external factors. FDI and real effective exchange rate represent the external factors while credit, population, tax on international trade, land area, employment and telephone subscriptions (used as a proxy for infrastructure) represent the internal factors. Thus, the simple linear model is specified as

EXP = f (C R D, P O P, R E E R, F D I, T O T, A R E A, R E M, E M P, T E L)

(5)

where $T O T$ represents the taxes on international trade, $A R E A$ represents the land area and $T E L$ represents telephone subscription. The export model estimated using the OLS and GMM estimation techniques are specified, respectively, in Equations (6) and (7) as follows:

Ln E X P_{i t} = β_{0} + β_{1} Ln C R D_{j t} + β_{2} {POP}_{i t} + β_{3} Ln R E E R_{i t} + β_{4} Ln F D I_{i t} + β_{5} Ln O T_{i t} + β_{6} Ln A R E A_{i t} + β_{7} Ln R E M_{i t} + β_{8} Ln E M P_{i t} + β_{9} Ln T E L_{i t} + ε_{i t}

(6)

\begin{array}{l} Ln E X P_{i t} = β_{0} + β_{1} Ln E X P_{{(t - 1)}_{i t}} + β_{2} Ln C R D i t + β_{3} P O P_{i t} + β_{4} Ln R E E R_{i t} + β_{5} Ln F D I_{i t} + β_{6} Ln T O T_{i t} + β_{7} Ln A R E A_{i t} + β_{8} Ln R E M_{i t} \\ + β_{9} Ln E M P_{i t} + β_{10} Ln T E L_{i t} + ε_{i t} \end{array}

(7)

where the subscript $i$ (=1,…, n) represents the West African country and $t$ (=1,…, $T$ ) the time (years), $ε_{i t}$ is the random disturbance term with zero mean and $E X P_{i t - 1}$ represents the lag of export for each country at time $t$

IV. Results and Discussions

Observations, mean, standard deviation, minimum and maximum are the examined descriptive statistics as shown in Table 2. The differences in the number of observations are a result of the non-availability of data for some countries in some years. Over the period, the average export for the region was about 28% of GDP with average GDP growth of about 5% even though some countries recorded negative growth for the period. Tax charged on international trade recorded the highest standard deviation for the period. This shows a great disparity in terms of international trade taxes across the countries under consideration.

Table 2.

Descriptive Statistics for Variables (2008–2018).

Variable	Obs.	Mean	Std Dev.	Min.	Max.
EXPORT	176	27.941	10.185	9.218	56.133
GROWTH	176	4.692	3.819	−20.599	20.716
CREDIT	173	19.616	13.541	2.66	65.742
EMP	176	61.087	10.607	40.942	79.353
AREA	176	384,059.06	439,232.16	4,033	1,267,000
TOT	73	2.077e + 11	3.197e + 11	79,931,058	1.186e + 12
TEL	172	139,772.34	191,955.86	0	1,482,000
FDI	176	6.488	13.176	−3.379	103.337
POP	176	53.984	3.659	47.183	66.613
REER	66	99.287	14.778	64.674	142.612
REMIT	167	5.515	4.491	0.261	21.81

Source: Author’s computation using data from WDI (2021).

Econometric Results

There are indications of valid and exogenous instruments used in the estimations. This is evident by the p values of the Hansen and Sargan test of over-identifying restrictions, presented in Table 3. Export, real effective exchange rate, foreign direct investment, employment and remittances are all positive and statistically significant. Credit, inflation and population though with the expected signs were not statistically significant. Improved FDI increases domestically produced outputs for domestic consumption as well as export purposes. This helps to improve the country’s trade balance. Again, the government does generate revenue from taxes paid by the foreign investors which are then used for activities that will aid the growth of the economy. In addition, appreciation of the local currency enables businesses who import raw materials and other machines for production to have them at favourable prices. It will as well improve the purchasing power of the citizenry hence promoting growth. This argument is in line with studies by Dabel (2016). Also, a unit increase in remittances increases economic growth by about 0.23 units holding all other variables constant. Remittances are known to serve as a catalyst for the financial market and the development of monetary policy in developing countries. The credit constraints of the poor are improved, making way for capital allocation which subsequently accelerates economic growth. A study by Cazachevici et al. (2020) concludes that there is a positive mean effect of remittances on growth though the effect is economically small. The coefficient of the main explanatory variable (export) implies that all other things being equal, a unit increase in export, leads to about 0.41 units increase in economic growth. This is consistent with the postulates of the ELGH and empirical studies by Abogan et al. (2014) and Chemeda (2001). It is important to note that exports form a major component of aggregate demand. This implies that an increase in export will improve aggregate demand hence a rise in the level of economic growth. Improvement in export performance provides more revenue to the businesses. These can then lead to improvement in capital investment spending via the accelerator effect. The productive capacity of the country is increased which induces growth in the economy. More export also implies a high level of output from factories within a country and a larger number of people getting employed to keep such factories functioning. The ongoing discussion confirms that the ELGH is proven and applicable in West Africa. Thus, a theoretical implication for the West Africa sub-region to leverage on. Just as its success has significantly improved the growth of the East Asian economies, a positive signal is, therefore, given to the policy makers in West African countries in their quest to achieve economic growth. The development of local industries is critical for the implementation of the ELGH and the efforts in being self-sufficient hence lowering the dependence on developed countries. This is also expected to help grow local industries, put the local currency in a desired position and improve overall growth in the domestic economies.

Table 3.

Effects of Export on Economic Growth.

Dependent Variable: Log(GROWTH)
Variables	Coefficient	Robust Std Error	p Values
Log(GROWTH) (−1)	−0.034	0.130	.793
Log(EXP)	0.414**	0.112	.021
Log(CRD)	−0.114	0.799	.886
Log(REER)	3.560***	0.652	.000
Log(FDI)	0.145*	0.081	.074
Log(EMP)	7.072**	3.113	.023
Log(REM)	0.257***	0.024	.000
Log(INF)	−0.050	0.040	.218
Log(POP)	7.939	9.724	.414
Constant	−75.841	50.560	.134
Wald chi²(9) = 150.17.No. of instruments = 11.Prob. > chi² = 0.0000.No. of obs. = 46.
Arellano–Bond test for AR(1) in first differences: z = −1.69, Pr > z = 0.090.Arellano–Bond test for AR(2) in first differences: z = −1.64, Pr > z = 0.101.Sargan test of overid. restrictions: chi²(1) = 4.70, Prob. > chi² = 0.130.Hansen test of overid. restrictions: chi²(1) = 0.00, Prob. > chi² = 1.000.

Source: Estimated using data from WDI (2020).

Notes: ***, ** and * indicate statistically significant at 1%, 5% and 10% level, respectively.

Table 4 presents the findings on the determinants of exports. The estimated model for this relationship does not suffer from instrument proliferation, as the number of instrument (13) is less than the number of observations and cross-sectional groups (Roodman, 2009). The p values of the Hansen and Sargan test of over-identifying restrictions further show that there are valid and exogenous instruments used in the estimations (Goodman, 2009). Again, there is evidence of the absence of autocorrelation in Arellano–Bond tests for both the first and second lag. It is shown that the previous year’s export has a positive and statistically significant effect on the current year’s export. The coefficients of foreign direct investment, land area, remittances and telephone subscription (used as a proxy for infrastructure) are all positive and statistically significant. Thus, these variables are noted to induce export performance in West Africa. The signs for foreign direct investment and telephone subscription are consistent with the findings of Tariq Majeed and Ahmad (2006), though Sharma (2000) found no statistically significant effect of FDI on export from India. FDI relating positively with export is in line with the FDI-led export growth strategies which come with twin objectives of seizing the benefits of FDI inflows and growth in export, particularly, when the motive is to take advantage of the comparative advantage of the region. Communication facilities are of much importance in the modern world. Expanding such facilities have a favourable impact on exploring and accessing the world market hence making way for international trade or improved volume of exports. The positive coefficient for the land area is in line with the findings of Kulu (2019) who explained that a larger land area induces export from member states of the West Africa Monetary Zone to the United States of America. Thus, the land is identified as one of the main factors of production (alongside labour and capital) in classical economics. It is the backbone of most countries in Africa that are agricultural economies hence providing a substantial economic benefit through exports and other social benefits (Wu, 2008). With the availability of land, producers have the option to either expand or diversify in their activities depending on the demand, especially from the international market.

Table 4.

Determinants of Exports Performance Using GMM and OLS.

Dependent Variable: Log(Export)
Variables	GMM	OLS
Log(EXP) (−1)	0.531** (0.275)
Log(CRD)	0.089 (0.161)	0.139 (0.183)
Log(POP)	−7.869 (4.795)	−13.525*** (2.851)
Log(REER)	0.132 (0.350)	−1.088** (0.405)
Log(FDI)	0.060*** (0.013)	0.034 (0.034)
Log(TOT)	−0.068*** (0.017)	−0.067 (0.040)
Log(AREA)	0.431** (0.219)	0.632*** (0.995)
Log(EMP)	−0.493 (3.353)	1.636*** (0.224)
Log(REM)	0.106*** (0.005)	0.175*** (0.040)
Log(TEL)	0.023** (0.040)	−0.026 (0.088)
Constant	30.588 (28.560)	49.837*** (12.459)
Arellano–Bond test for AR(1) in first differences: z = −0.84, Pr > z = 0.401.Arellano–Bond test for AR(2) in first differences: z = −1.50, Pr > z = 0.134.Sargan test of overid. restrictions: chi²(2) = 1.32, Prob. > chi² = 0.516.Hansen test of overid. restrictions: chi²(2) = 0.00, Prob. > chi² = 1.000.
	Wald chi²(10) = 2,212.72No. of instruments = 13Prob. > chi² = 0.0000No. of obs. = 25	Prob. > chi² = 0.0000R² = 0.9272No. of obs. = 29

Source: Estimated using data from WDI (2021).

Notes: *** and ** indicate statistically significant at 1%, and 5% level, respectively.

Employment used as a proxy for labour is found to be positive and statistically significant in the OLS model. This is consistent with Pfaffermayr (1996) who argues that labour force can be utilised properly especially in the agricultural and industrial sectors to increase production levels hence export volumes. Thus, all other things being equal, as the size of labour force increases, the corresponding output increases. With increased output, domestic consumption can improve as well as the volume available to be exported.

Tax on international trade is negative and statistically significant (though statistically insignificant in the OLS model). Thus, a percentage increase in tax on international trade leads to about 0.1% fall in export all other variables are held constant. Taxes on international trade include duties and tariffs on exports and so on which are noted to discourage export. Thus, taxes are found among the barriers to the performance of export (Carneiro et al., 2011). Taxes are also seen as a cost to production hence their increments erode the capacity of export by most producers especially those with a small size or new in the business. With increasing taxes on international trade, local producers may be forced to market their outputs in the local market. Again, the population is found to be negative and statistically significant in the OLS model. An increase in population will reduce the volume of export if it is accompanied by a rise in domestic demand for outputs (Sharma, 2000). Export performance is also found to relate negatively with the real effective exchange rate (as found in the OLS model). The real effective exchange rate is an index measured such that an increase depicts appreciation of the local currency. Appreciation of the local currency (increase in the real effective exchange rate) therefore adversely affects export performance (Sharma, 2000; Tariq Majeed & Ahmad, 2006). When the local currency appreciates against a foreign currency, a relatively little amount of money can be made when the received amount (in foreign currency through trade) is converted to a local currency. Thus, depreciation benefits exporters more than appreciation of a local currency.

Dominance Analysis

The general dominance analysis shows the additional unique variance that is contributed by each independent variable (Azen & Budescu, 2003). Also, it breaks the R-squared and recognises the contribution of each variable by using a step-wise approach (Koomson et al., 2016; Nathan et al., 2012). The dominance analysis is presented in Table 5 showing the standardised dominance statistics and the resultant ranking. In terms of relative importance, it is observed that land area is ranked the highest, followed by the tax on international trade, credit, employment, remittances, telephone subscription and so on. Relatively, the land area and taxes on international trade are essential variables to be focused on in the quest in improving export volumes for overall growth in the economy. Availability of land is very significant for agricultural production, especially in Africa. Indeed, after production has taken place, taxes on international trade also interplay. High taxes tend to discourage producers from exporting their products. Foreign direct investment though significant in the econometric estimations is ranked the last but one in terms of relative importance. Thus, the role FDI plays in developing countries such as West African states in terms of export promotion remains a debate. It is argued that this depends on the motive behind such an investment. If capturing of the domestic market (that is tariff-jumping investment) is the main motive for the investment, then the contribution to export may be minimal or absent (Tariq Majeed & Ahmad, 2006).

Table 5.

General Dominance Weight.

Variables	Standardised Domin. Stat.	Ranking
Log(CRD)	0.1209	3
Log(POP)	0.0617	7
Log(REER)	0.0250	9
Log(FDI)	0.0447	8
Log(TOT)	0.2302	2
Log(AREA)	0.2475	1
Log(EMP)	0.1067	4
Log(REM)	0.0822	5
Log(TEL)	0.0810	6

Source: Author’s computation using STATA 15 package.

Post-estimation Tests

Table 6 presents various post-estimation tests in checking the robustness of the OLS results. The p values of 0.142 and 0.6411 in the Ramsey test for omitted variables and Breusch–Pagan/Cook Weisberg test for heteroscedasticity, respectively, show the problem of omitted variables and heteroscedasticity are absent in the OLS estimation. Thus, the model is correctly specified.

Following Behera and Dash (2018), the Pesaran CD test was performed in checking the cross-sectional dependence (as evident in Table A2). Against the null hypothesis of cross-sectional independence, we reject the null hypothesis for all variables except for economic growth. This indicates that there is a serial correlation among the variables which may originate from macroeconomic common shocks which impact West African countries differently. It is important to note that the robust standard errors were used in both the GMM and OLS models to correct serial correlation and any potential heteroscedasticity.

Table 6.

Post-estimation Tests.

Ramsey RESET test using powers of the fitted values of log(EXP)H₀: model has no omitted variablesF(3, 16) = 3.42Prob. > F = 0.142Breusch–Pagan/Cook–Weisberg test for heteroscedasticityH₀: Constant varianceVariables: fitted values of log(EXP)chi²(1) = 0.22Prob. > chi² = 0.6411

Source: Author’s computation using STATA 15 package.

V. Conclusions and Recommendations

The export–growth nexus cannot be over-emphasised owing to its dynamics and the growing success being enjoyed by countries and regions adopting it. Thus, conclusions being drawn from the literature are that the relationship to some extent is region-specific. The current study aims at examining the relationship between export and economic growth as well as factors influencing export performance in West Africa. Using the GMM estimation technique, a positive association between export and economic growth was found, therefore, validating the ELGH in the region. The study further showed that FDI, employment, remittances, infrastructure and land area are boosters of export performance while export from the West Africa region is also affected adversely by population, real effective exchange rate and taxes on international trade. The dominance analysis performed also showed that land area and tax on international trade are relatively the most important variables in explaining export performance. Having the ELGH validated in the economies of West Africa, the study, therefore, recommends that taxes especially on international trade should be relaxed to encourage the activities of West African businesses in the export sector and also attract other businesses to feed the export sector of the region. There is a need to work hard to recover the business environment in the region as a means to attract foreign direct investment. Thus, investors are mostly attracted by the position of a country on the global competitiveness ranking. Again, countries in the West Africa region are encouraged to undertake the development of infrastructure through communication facilities as it is known for sustaining export performance. These activities are envisaged to go a long way to improve export performance and realisation of sustainable economic growth.

Appendix A

Table A1.

Countries Used for the Study.

Benin	Guinea	Nigeria
Burkina Faso	Guinea Bissau	Senegal
Capo Verde	Liberia	Sierra Leone
Cote d’ Ivoire	Mali	Togo
Gambia, The	Mauritania
Ghana	Niger

Table A2.

Pesaran (2004) CD Test.

Variable	CD test	p Value	Corr.	Abs (Corr.)
LGROWTH	0.20	0.840	0.002	0.336
LEXP	3.78	0.000	0.121	0.421
LCRD	7.07	0.000	0.214	0.635
LPOP	20.06	0.000	0.628	0.728
LFDI	1.65	0.099	0.054	0.367
LEMP	4.20	0.000	0.116	0.709
LINF	5.63	0.000	0.200	0.437

Source: Author’s computation using STATA 15 package.

Notes: Variables such as tax on international trade, remittances and telephone subscriptions did not appear because they have insufficient observation to perform the Pesaran test. Also, land area is a time-invariant variable hence unable to undergo the Pesaran test.

Footnotes

Declaration of Conflicting Interests

The author declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Evans Kulu

References

Abogan

O. P.

, Akinola

, & Baruwa

(2014). Non-oil export and economic growth in Nigeria. Journal of Research in Economics and International Finance , 3(1), 1–11.

AEO. (2019). Macroeconomic performance and prospects. Regional integration and structural transformation in West Africa . Africa Development Bank Group.

Ayibor

R. E.

(2012). Trade liberalisation and economic growth in Ghana [Doctoral dissertation]. Department of Economics, University of Cape Coast.

Azen

, & Budescu

D. V.

(2003). The dominance analysis approach for comparing predictors in multiple regression. Psychological Methods , 8(2), 129–148. https://doi.org/10.1037/1082-989x.8.2.129

Behera

D. K.

, & Dash

(2018). The impact of macroeconomic policies on the growth of public health expenditure: An empirical assessment from the Indian states. Cogent Economics & Finance , 6(1), 1435–1443

Barro

R. J.

, & Sala-i

M. X.

(2004). Economic growth. Aufl . Massachusetts Institute of Technology (MIT).

Brancati

, Brancati

, Guarascio

, Maresca

, Romagnoli

, & Zanfei

(2018). Firm-level drivers of export performance and external competitiveness in Italy (No. 087). Directorate General for Economic and Financial Affairs, European Commission.

Carneiro

, Rocha

A. D.

, & Silva

J. F. D.

(2011). Determinants of export performance: A study of large Brazilian manufacturing firms. Brazilian Administration Review , 8(2), 107–132. https://doi.org/10.1590/S1807-76922011000200002

Cazachevici

, Havranek

, & Horvath

(2020). Remittances and economic growth: A meta-analysis. World Development , 134, 105021. https://doi.org/10.1016/j.worlddev.2020.105021

10.

Chemeda

F. E.

(2001). The role of exports in economic growth with reference to Ethiopian country (No. 374-2016-19785).

11.

Dabel

(2016). Trade openness and economic growth: Evidence from Ghana [Doctoral dissertation]. University of Cape Coast.

12.

Didia

, Nica

, & Yu

(2015). The gravity model, African Growth and Opportunity Act (AGOA) and US trade relations with sub-Saharan Africa. Journal of International Trade and Economic Development , 24(8), 1130–1151. https://doi.org/10.1080/09638199.2014.1000942

13.

Freund

, & Bolaky

(2008). Trade, regulations, and income. Journal of Development Economics , 87(2), 309–321. https://doi.org/10.1016/j.jdeveco.2007.11.003

14.

Goodman

(2009). How to do xtabond2: An introduction to difference and system GMM in STATA. STATA Journal: Promoting Communications on Statistics and STATA , 9(1), 86–136. https://doi.org/10.1177/1536867X0900900106

15.

Haddoud

M. Y.

, Nowinski

, Jones

, & Newbery

(2019). Internal and external determinants of export performance: Insights from Algeria. Thunderbird International Business Review , 61(1), 43–60. https://doi.org/10.1002/tie.21972

16.

Koomson

, Annim

S. K.

, & Peprah

J. A.

(2016). Loan refusal, household income and savings in Ghana: A dominance analysis approach. African Journal of Economic and Sustainable Development , 5(2), 172–191. https://doi.org/10.1504/AJESD.2016.076095

17.

Kulu

(2019). Effects of the Africa growth and opportunity on export of member states of the West Africa monetary zone to the United States of Africa [Doctoral dissertation]. University of Cape Coast.

18.

Lazarov

(2019). Empirical analysis of export performance and economic growth: The case of Macedonia. International Journal of Trade and Global Markets , 12(3/4), 381–393. https://doi.org/10.1504/IJTGM.2019.101541

19.

Malhotra

, & Kumari

(2016). Export performance and economic growth in East Asian economies application of cointegration and vector error correction model. Eurasian Journal of Business and Economics , 9(18), 135–152. https://doi.org/10.17015/ejbe.2016.018.08

20.

Mankiw

N. G.

, Romer

, & Weil

D. N.

(1992). A contribution to the empirics of economic growth. Quarterly Journal of Economics , 107(2), 407–437. https://doi.org/10.2307/2118477

21.

Medina-Smith

E. J.

(2001). Is the export-led growth hypothesis valid for developing countries? A case study of Costa Rica . United Nations.

22.

Nathan

L.L.

, Oswald

F.L.

& Nimon

(2012). Interpreting multiple linear regression: a guide book of variable importance, Practical Assessment, Research & Evaluation , 17(9), 1–19

23.

Ngo

V. D.

, Janssen

, Leonidou

L. C.

, & Christodoulides

(2016). Domestic institutional attributes as drivers of export performance in an emerging and transition economy. Journal of Business Research , 69(8), 2911–2922. https://doi.org/10.1016/j.jbusres.2015.12.060

24.

Oosterbaan

M. S.

, van Steveninck

T. D.R.

, & Windt

(Eds.) (2000). The determinants of economic growth . Kluwer Academic Publishers.

25.

Pfaffermayr

(1996). Foreign outward direct investment and exports in Austrian manufacturing: Substitutes or complements? Weltwirtschaftliches Archiv , 132(3), 501–522. https://doi.org/10.1007/BF02707510

26.

Roodman

(2009). A note on the theme of too many instruments. Oxford Bulletin of Economics and Statistics , 71(1), 135–158. https://doi.org/10.1111/j.1468-0084.2008.00542.x

27.

Sannassee

R. V.

, Seetanah

, & Jugessur

(2014). Export-led growth hypothesis: A meta-analysis. Journal of Developing Areas , 48(1), 361–385. https://doi.org/10.1353/jda.2014.0018

28.

Sharma

(2000). Export growth in India: Has FDI played a role? https://www.econ.yale.edu/~egcenter/

29.

Tariq Majeed

, & Ahmad

(2006). Determinants of exports in developing countries. Pakistan Development Review , 45(4), 1265–1276.

30.

Thapa-Parajuli

, Aryal

, Alharthi

, & Paudel

R. C.

(2021). Energy consumption, export performance and economic growth in a landlocked developing country: The case of Nepal. AIMS Energy , 9(3), 516–533. https://doi.org/10.3934/energy.2021025

31.

(2008). Land use changes: Economic, social, and environmental impacts. Choices , 23, 6–10.