• For Contributors +
• Journal Search +
Journal Search Engine
ISSN : 1225-8504(Print)
ISSN : 2287-8165(Online)
Journal of the Korean Society of International Agriculture Vol.33 No.1 pp.29-39
DOI : https://doi.org/10.12719/KSIA.2021.33.1.29

The Effect of International Technology Cooperation on Bilateral Trade Flows: A Case of The Korea-Africa Food and AgricultureCooperation Initiative (KAFACI)

Timothy Mtumbuka*,***, Jeongran Lee*†, Byungmo Lee*, Sinsuk Kang*, Hyeong Sik Eum*, Dayoung Park*, Jeong Jun Kim**
*Korea-Africa Food and Agriculture Cooperation Initiative (KAFACI), Rural Development Administration (RDA), 300 Nongsaengmyeong-ro, Deokjin-gu, Jeonju, Republic of Korea
**Department of Agricultural Environment, National Institute of Agricultural Sciences, RDA, 166 Nongsaengmyeong-ro, Iseo-myeon, Wanju-gun, Republic of Korea
***Ministry of Agriculture, Capital Hill, Box 30134, Lilongwe 3, Malawi
Corresponding author (Phone) +82-63-238-1126 (E-mail) kongsarang@korea.kr
November 13, 2020 December 8, 2020 December 11, 2020

Abstract

This paper used the gravity model of international trade to analyze the effects of international technology cooperation on bilateral trade flows between member and non-member countries of the Korea-Africa Food and Agriculture Cooperation Initiative (KAFACI). We used panel data of bilateral trade between 45 African countries from the United Nations commodity trade statistics database for the period between 2000 and 2018. To control for endogeneity, selection bias, and correlation within the panels, this study used the Heckman random-effect regression with sample selection technique. Although insignificant, the empirical results indicated a positive effect of international technology cooperation on bilateral trade flows between KAFACI member and non-member countries. Notwithstanding our findings, KAFACI might have significantly increased the bilateral trade flows between its members and main trading partners outside Africa, which were not explored in this paper. Besides, the level and rate of technology adoption by KAFACI member countries were also crucial for influencing the supply side of the economy to unlock trade, but they were not assessed in this study.

양자무역에서 국제기술협력의 효과: 한-아프리카 농식품기술협력협의체 사례 중심

타모씨 음투부가*,***, 이 정란*†, 이 병모*, 강 신숙*, 엄 형식*, 박 다영*, 김 정준**
*농촌진흥청 한-아프리카농식품기술협력 협의체
**국립농업과학원 환경개선미생물연구단
***말라위 농림부

INTRODUCTION

The Rural Development Administration (RDA) of the Republic of Korea is a government organization that facilitates agricultural technology development and dissemination to various stakeholders. The RDA also works in partnership with international bodies, including interna-tional research institutions and government agencies from both developed and developing countries in research projects, expert consultations and capacity building. Since the 1980s, the main focus of RDA with regards to international cooperation has been on transferring agricultural technology to developing countries and creating partnerships for the development of new technologies. To facilitate the development and transfer of technology to developing countries, the RDA introduced multilateral cooperation initiatives covering Asia, Africa and Latin America. The initiatives are named based on the region of collaboration as follows: (i) Asian Food and Agriculture Cooperation Initiative (AFACI); (ii) Korea-Africa Food and Agriculture Cooperation Initiative (KAFACI); and (iii) Korea- Latin America Food and Agriculture Cooperation Initiative (KoLFACI). In addition to multilateral cooperation initiatives, the RDA also introduced the Korea Program for International Cooperation in Agricultural technology (KOPIA) as a bilateral cooperation initiative for the development and sharing of agricultural technologies (RDA, 2015). However, this paper will only concentrate on the technology transfer and development collaboration between the RDA and African countries under KAFACI. As one of the cooperation initiatives under RDA, the KAFACI was established to promote intergovernmental and multilateral cooperation in a quest to improve food production, achieve sustainable agriculture and enhance extension services of African countries. To achieve its objectives, KAFACI promotes knowledge and information sharing on agricultural technologies among member countries. The initiative was officially inaugurated on July 6, 2010, in Seoul, the Republic of Korea with the participation of 16 member countries, namely Angola, Cameroon, Cote d’Ivoire, Democratic Republic of Congo (DRC), Ethiopia, Gabon, Ghana, Kenya, Malawi, Morocco, Nigeria, Senegal, Sudan, Tunisia, Uganda and Zimbabwe. Three more countries joined the initiative as follows: Comoros (2012), Rwanda (2015) and Zambia (2016). Currently, there are 20 member countries of KAFACI including the Republic of Korea. KAFACI envisions to ensure a vibrant and dynamic network among its member countries and enhance their partnership and leadership in the international community. Ultimately, the initiative aims to promote sustainable agricultural development in African member countries to improve people’s livelihoods, reduce food insecurity and contribute to economic development through technological cooperation in the agriculture sector (Cho, 2011;KAFACI, 2020).

Apart from cooperation among member countries, KAFACI also works in partnership with other organizations to ensure agricultural development in member countries. The main partner organizations include Alliance for a Green Revolution in Africa (AGRA), International Livestock Research Institute (IRLI), AfricaRice, Center on Conflict and Development (ConDev), International Trade Centre (ITC) and African Forum for Agricultural Advisory Services (AFAAS).

Through the famous green revolution and Saemaul movement program, the Republic of Korea has shared appropriate agricultural knowledge and technology for adoption in KAFACI member countries. The technologies are developed and shared in several forms including group training, on-the-job training, post-graduate programs and visiting fellowship programs (Cho, 2011;KAFACI, 2020). The RDA also dispatches scientists from its institutes to member countries for technical backstopping and provision of support services on technology development and application. The KAFACI technologies enable the member countries to develop new products and processes which are catalysts of international trade and country development. Based on both the technology gap theory (Posner, 1961) and product life cycle theory (Vernon, 1966) of international trade, technology developing countries or first adopters are expected to increase their volume of trade due to comparative advantage in the value chains with the high absorption capacity of technologies. The advantage stems from the technology gap that exists before the other countries copy the technologies and create substitutes for their domestic markets.

Several other studies have been conducted on bilateral trade using the gravity model of international trade and the results support the theoretical premise of technology in unlocking bilateral trade flows between trading partners. A study by Marquez-Ramos et al. (2010) found a positive correlation between technological innovation and export volumes. However, their findings did not exhibit a linear effect but it was observed that for a positive effect to occur, certain levels of technological advancement have to be achieved. More importantly, their conclusion highlighted the need to promote technology adoption for increased exports rather than sorely focusing on the acquisition and dissemination of technologies. Krugman (1982) analyzed the trends of trade as a result of varying technological gaps and intensities using the Ricardian model. The results showed that technological differences are critical for a country to have a comparative advantage in the production of a commodity over others. His findings confirmed the trade and production theories that assume the differences in technology levels among countries as well as their research and development capacities are crucial determinants of trade volumes and income level of a country.

Since 2010, KAFACI has been facilitating the transfer of agricultural technology to its member countries and creating partnership for new appropriate technology development. In line with the technology gap and product life cycle theories of international trade, adoption of these technologies by member countries will help them leapfrog and quickly improve their comparative advantage over non-member countries in production of specific products that intensively use KAFACI technologies in production process. Consequently, the level of bilateral trade flows between KAFACI members and non-member countries is expected to increase. Therefore, using bilateral trade data for 45 African countries, this paper aims to investigate if the international technology cooperation initiatives of the Republic of Korea has an effect on bilateral trade flows between KAFACI members and non-member countries.

METHODOLOGY

Model Specification

The gravity model has been adopted by vast studies on international trade such as Rose (2004), Helpman et al. (2008), Gómez-Herrera (2013), and Andrei (2017). By the same token, this study will apply the gravity model to determine the effect of international technology cooperation on the bilateral trade flows between KAFACI members and non-member countries in Africa. The model presumes that trade is positively correlated to the Gross Domestic Product (GDP) but negatively correlated to the distance between trading partners. The results of the gravity model on the two parameters have been consistent in many studies and they explain a considerable variation in international trade flows. This paper will build on the following basic gravity model introduced by Rose (2004):

$b t i j t = β 0 ( g d p i t × g d p j t d i s t i j )$
(1)

Where btijt is the log of real bilateral trade between countries i and j at time t, gdpit is the real GDP of country i at time t, gdpjt is the real GDP of country j at time t and distij is the distance in kilometers between countries i and j.

The specified augmented gravity model captures the effects of GDP per capita, land size, sharing a land border, use of the same language, regional economic/trade blocs and being colonized by the same country. The model consists of two equations, namely the outcome and sample selection as specified by (2) and (4), respectively.

The outcome equation is modeled as follows:

$l n ( b t i j t ) = β X i j t + v i j + μ i j t$
(2)

Where Xijt represents the covariates modeling the flows of bilateral trade, vij represents the random effects at panel level and μijt is the observation-level error. The full outcome model is specified as follows:

$l n ( b t i j ) t = β 0 + β 1 l n ( d i s t i j ) + β 2 l n ( p g d p i j ) t + β 3 l n ( p g d p p c i j ) t + β 4 l n ( p l a i j ) + β 5 b o t h i n i j t + β 6 o n e i n i j t + β 7 l a n d l i j + β 8 l a n d b i j + β 9 c o m l a n g i j + β 10 R E B i j + β 11 c c o l o n i j + v i j + μ i j t$
(3)

The covariates include: distij – log of distance between countries i and j; pgdpijt – product of real GDPs for countries i and j; gdppcijt – log product of real GDP per capita for countries i and j; plaij - log product of land areas for countries i and j; bothinijt – a dummy variable and is equal to 1 if both countries i and j are KAFACI members at period t and zero otherwise; oneinijt – a dummy variable and is equal to 1 if either country i or j is KAFACI member at period t and zero otherwise; landlij – denotes the number of landlocked countries in each pair of i and j (0/1/2); landbij – a dummy variable and is equal to 1 if both countries i and j share a land border and zero otherwise; comlangij – a dummy variable and is equal to 1 if countries i and j share a common official language and zero otherwise; REBijt – a dummy variable and is equal 1 if both countries i and j belongs to the same regional economic or trade bloc and zero otherwise; ccolonij – a dummy variable and is equal to 1 if both countries i and j were colonized by the same country and zero otherwise; and sreligionij – a dummy variable and equal to 1 if both countries i and j share the same religion and zero otherwise.

Equation (4) presents the selection equation where dy is the latent variable. It is only observed if there is some level of bilateral trade flows between countries i and j or not, thus dy =1 if btijt >0 and dy=0 otherwise. The variable Zijt represents the covariates that affect dy which include Xij variables and an additional dummy variable capturing the same religion (sreligionij) between countries i and j. It is equal to 1 if both countries i and j share the same religion and zero otherwise.

$d y i j t = 1 ( α Z ijt + s ij + ε ijt > 0 )$
(4)

The random-effects vij and sij have a bivariate normal distribution with a zero mean and variance of σ2. Likewise, the observation-level selection errors μijt and εijt also have a bivariate normal distribution with zero mean and variance of σ2. The full specification of the selection equation is as follows:

$d y i j t = α 0 + α 1 l n ( d i s t i j ) + α 2 l n ( p g d p i j ) t + α 3 l n ( p g d p p c i j ) t + α 4 l n ( p l a i j ) + α 5 b o t h i n i j t + α 6 o n e i n i j t + α 7 l a n d l i j + α 8 l a n d b i j + α 9 c o m l a n g i j + α 10 R E B i j + α 11 c c o l o n i j + α 12 s r e l i g i o n i j + s i j + ε i j t$
(5)

Using the variance adaptive Gauss-Hermite quadrature approximation with the abscissa and weight pairs for each panel equal to akij and wkij, respectively; where k =1,…….,q., then we can generate the log likelihood for all panels as follows:

$l n L = ∑ i j = 1 N ( l n ∑ k 2 = 1 q ⋯ ∑ k 2 = 1 q [ { ∏ t = 1 N i j f ( b t i j t , d y i j t | ( v i j , s i j ) ′ = L ∝ k ) } { ∑ s = 1 2 W k s } ] )$
(6)

The conditional mean for ln(btijt) is given by the following equation:

$E ( l n b t i j t | X i j t ) = β X i j t$
(7)

RESULTS AND DISCUSSIONS

Prior to the estimation of the models, we conducted some preliminary tests. Summary statistics for the correlation between covariates showed no significant association except for the correlations between GDP and land area (- 0.64), distance and sharing land border (-0.55), and between distance and regional trade/economic blocs (-0.54), which were moderately correlated. An analysis of the residuals and fitted values of the model confirms the presence of heteroskedasticity because the variance of the residuals is not constant as presented in Figure 1 below. Therefore, linear estimation methods are not appropriate for analyzing the data.

Table 1 presents the results of Heckman random-effect regression with sample selection, where the logarithm of real bilateral trade (lnbt) is the dependent variable for the outcome equation and the latent variable (dy) is the dependent variable for the selection equation. The results show that there is no significant correlation between observation- level errors (e. dy, e. lnbt) for the outcome and the selection models in our uncontrolled models 1 and 4 but there is a significant negative correlation in models (3) and (6) where random effects are constrained to be independent. On the other hand, the correlation coefficients between the random effects of the outcome and the selection models {Corr. (dy[pairid], lnbt[pairid]))} are positive and significantly different from zero in both models (1) and (4) with the coefficients of 0.746 and 0.671, respectively. According to StataCorp (2019), if at least one of the correlation coefficients is significant then it can be concluded that sample selection is endogenous. Therefore, our results confirm the presence of endogeneity in sample selection i.e. unobserved individual-level factors that increase the probability of reporting positive bilateral trade flows tend to increase the level of bilateral trade between trading partners.

The estimation technique is consistent with other estimation techniques that have used the gravity equation to analyze trade volumes in terms of the direction of the effect for the main covariate (Rose, 2004;Gómez-Herrera, 2013;Westerlund & Wilhelmsson, 2011;Haq et al., 2012;Andrei, 2017). Sharing of the common border, language and colonizer as well as belonging to the same economic or trade bloc significantly increase the level of bilateral trade. Likewise, the level of real GDPs for the trading partners also significantly increases the level of bilateral trade. Other variables such as the real GDP per capita, distance between trading partners and landlocked partners are also significant in explaining the variability of bilateral trade but the effect is negative. Models (1), (2) and (3) use the full sample covering the period from 2000 to 2018. The only difference is that model (2) omits the random effects from the selection model while model (3) constrains the random effects to be independent. The significance of the coefficient estimates is consistent across the three models. However, model (2) with no random effects from the selection model has none of the correlation coefficients significantly different from zero. Models (4), (5) and (6) present the results for the period of interest from 2010 when the KAFACI was established to 2018. Most coefficients of the covariates are consistent with the full sample models (1), (2) and (3) in terms of significance but the magnitude of the effect is different.

Though not significant, our variables of interest, onein exhibit a positive effect of international technology cooperation on the level of bilateral trade flows. Therefore, the international technology cooperation positively affected the flow of bilateral trade between KAFACI members and non-member countries [model (4)]. The insignificance of the effect may be attributed to the fact that most of the African countries trade more with countries from other continents than among themselves through non-reciprocal trade agreements such as the General Agreement on Trade and Tariff (GATT), African Growth Opportunity Act (AGOA) and the Generalized System of Preferences (GSP) (Garth & Van Biessebroeck, 2010;Rose, 2004).

All the coefficients of the models in table 1 are consistent with available literature except for the level of real GDP per capita which shows an inverse relationship with the flow of bilateral trade. Contrary to other studies (Rose, 2004;Kalirajan, 2007;Mnasri & Salem, 2019; and Santos and Tenreyro, 2006), this study found that the level of real GDP per capita reduces the levels of bilateral trade between partner countries. This can be attributed to the non-achievement of demographic dividends by the sampled countries. The sample of study includes the Sub- Saharan countries and other developing countries in Africa whose levels of population growth and unemployment are high. According to Dao (2012), Headey and Hodge (2009) and Sachs (2008), rapid population growth has a significant negative impact on the economic growth of these developing countries. This reduces the growth rate of GDP per capita if the population growth rate is higher than the GDP growth rate hence negatively affects the level of bilateral trade between partner countries in our model.

Regional blocs and member countries

Table 2 below presents the list of member countries for each trade/economic bloc in Africa, including KAFACI and the respective dates of establishment.

The last model considers the whole period of study from 2000 to 2018 and the results are similar to those of the preceding model (2010-2018). The IGAD, AMU, ECCAS and KAFACI had no effect on the level of bilateral trade among their respective member countries. The findings in Table 3 are consistent with the findings of Turkson (2012) who used time-series data from 1960 to 2006 to analyze the effect of regional trade agreements within Sub-Saharan Africa on bilateral trade flows. He found that SADC and ECOWAS had a significant positive effect on bilateral trade among member countries while EAC and IGAD produced inconclusive results.

Alternative estimation techniques for the gravity model

In order to analyze the sensitivity of the covariates across different estimation techniques, our augmented gravity equation was also estimated using ordinary least squares (OLS), Heckit model and Pseudo Poisson maximum likelihood (PPML) techniques. Table 4 provides the estimation results which are consistent with our results from the Heckman random-effect regression with sample selection model in terms of significance of the parameters though the magnitude is different. Results from OLS regressions in both periods give higher estimates compared to the Heckman and PPML estimators. In presence of heteroscedasticity, the OLS estimates are inconsistent because truncation of zero values leads to loss of information and the addition of a constant to the values of bilateral trade alters the conditional distribution of the bilateral trade (Gómez-Herrera, 2013). The results from both Heckman sample selection equations (Table 4) are consistent with those of our Heckman random-effect regression with sample selection (Table 1) in terms of significance and direction of effects on bilateral trade flows. Likewise, the significance and direction of most parameters for the PPML are consistent with results from other models in Table 4 and Table 1. However, the magnitude and associated standard errors of the PPML are significantly different because the dependent variable is introduced in levels rather than taking a log transformation (Gómez-Herrera, 2013;Santos & Tenreyro, 2006).

All the model results have robust standard errors clustered by pairid presented in brackets. The regressand is the logarithm of the product of bilateral trade between countries i and j except for the PPML regression where the dependent variable is categorized into levels. All the models include time fixed effects.

적 요

한-아프리카 농식품 기술협력 협의체 (KAFACI)는 아프리카 농업 공통현안을 연구과제를 통하여 해결하고자 2010년에 출 범하여 현재 한국 포함 20개국으로 구성되었다. 본 논문은 지 난 10년간 회원국에서 수행한 결과를 바탕으로 KAFACI 회원 국과 비회원국 간의 양자 무역 흐름에 대한 국제 기술 협력의 영향을 분석하였다. 분석은 UN 상품 무역 통계 데이터베이스 에서 2000년에서 2018년 동안 45 개 아프리카 국가 간의 양 자 무역 패널 데이터를 이용하였으며 국제 무역 중력 모델을 적용하였다. 또한, 내생성, 표본 선택 편의, 패널내의 상관관계 를 통제하는 표본 선택 기법을 적용한 핵크만 랜덤 효과 회귀 모형을 사용하였다. 분석결과 경험적 결과는 국제 기술 협력 이 KAFACI 회원국과 비회원 국 간의 양자 무역 흐름에 긍 정적인 영향을 미친다는 것을 보여준다. 이러한 연구 결과 이 외에도 KAFACI는 회원국과 아프리카 이외의 주요 무역 파트 너 간의 양자 무역 흐름을 크게 늘렸을 가능성이 있지만 여기 서는 다루지 않았다. 또한 KAFACI 회원국의 기술 채택 수준 과 비율은 무역을 잠금 해제하기 위해 경제의 공급 측면에 영 향을 미치는데 중요하지만 여기에서는 다루지 않았다.

ACKNOWLEDGMENTS

This study was supported by the grant of KAFACI Outcome Analyst Program, Rural Development Administration of Korea.

Figure

Heteroskedasticity in the data.

Table

Effects of international technology cooperation on bilateral trade flows

Membership to economic, trade, and cooperation blocs

Analysis of disaggregated regional blocs and varying periods

Comparison of alternative estimation models

Reference

1. Andrei, S. 2017. Estimating Gravity Models of International Trade: A Survey of Methods, HSE Economic Journal, National Research University Higher School of Economics, 21(2): 224- 260.
2. Bartus, T. and Roodman, D. 2014. Estimation of Multi-Process Survival Models with cmp. Stata Journal. 14: 756-777.
3. Cho, G. 2011. South Korean Strategy for Agricultural Technology Transfer to Developing Countries: Case of Rural Development Administration, Korea center for U.S.-Korea Policy, Oct 17-18.
4. Dao, M. 2012. Population and Economic Growth in Developing Countries. International Journal of Academic Research in Business and the Social Sciences. 2.
5. Freund, C. and Weinhold, D. 2002. The Internet and International Trade in Services. American Economic Review. 92: 236- 240.
6. Garth, F. and Van Biesebroeck, J. 2010. Trade Growth under the African Growth and Opportunity Act, The Review of Economics and Statistics, MIT Press, 92(1): 128-144.
7. Gómez-Herrera, E. 2013. Comparing Alternative Methods to Estimate Gravity Models of Bilateral Trade, Empirical Economics, Springer, 44(3): 1087-1111, June.
8. Haq, Z. , Meilke, K. and Cranfield, J. 2012. Selection Bias in a Gravity Model of Agrifood Trade. European Review of Agricultural Economics. 1-30.
9. Headey, D. and Hodge, A. 2009. The Effect of Population Growth on Economic Growth: A Meta-Regression Analysis of the Macroeconomic Literature. Population and Development Review. 35: 221-248.
10. Heckman, J. 2012. The Common Structure of Statistical Models of Truncation, Sample Selection and Limited Dependent Variables and a Simple Estimator for Such Models. NBER Book Chapters.
11. Helpman, E. , Marc, M. and Yona, R. 2008. Estimating Trade Flows: Trading Partners and Trading Volumes. Quarterly Journal of Economics 123 (2): 441-487.
12. KAFACI,2020. Korea-Africa Food and Agriculture Cooperation Initiative (KAFACI), retrieved from https://kafaci.org (Accessed: 15 September, 2020).
13. Kalirajan, K. 2007. Regional Cooperation and Bilateral Trade Flows: An Empirical Measurement of Resistance. The International Trade Journal. 21: 85-107.
14. Keller, W. 2004. International Technology Diffusion. Journal of Economic Literature, September 2004, 42(3): 752-782.
15. Krugman, P. 1982. Trade in Differentiated Products and the Political Economy of Trade Liberalization. Import Competition and Response, edited by Jagdish N. Bhagwati, 197-221. Chicago: University of Chicago Press.
16. Loungani, P. , Mody, A. and Razin, A. 2002. The Global Disconnect: The Role of Transactional Distance and Scale Economies in Gravity Equations. Scottish Journal of Political Economy. 49: 526-43.
17. Marquez-Ramos, L. and Martnez-Zarzoso, I. 2010. The Effect of Technological Innovation on International Trade. A Nonlinear Approach. Economics. 4: 1-37.
18. Martínez-Zarzoso, I. , Vidovic, M. and Voicu, A. 2014. EUAccession Effects on Sectoral Trade: A Helpman-Melitz-Rubinstein Approach with Panel Data.
19. Miran, B. , Atis, E. , Bektaş, Z. , Salali, E. and Cankurt, M. 2013. An Analysis of International Raisin Trade: A Gravity Model Approach.
20. Mnasri, A. and Nechi, S. 2019. New Approach to Estimating Gravity Models with Heteroscedasticity and Zero Trade Values. SSRN Electronic Journal.
21. Posner, M.V. 1961. International Trade And Technical Change, Oxford Economic Papers, Oxford University Press, 13(3): 323-341.
22. Ronald, J. 1970. The Role of Technology in the Theory of International Trade, NBER Chapters, in: The Technology Factor in International Trade, 73-94, National Bureau of Economic Research, Inc.
23. Roodman, D. 2009. Estimation Fully Observed Recursive Mixed- Process Models with CMP. Center for Global Development, Working Paper Number 168.
24. Rose, A. K. 2004. Do We Really Know That the WTO Increases Trade? American Economic Review, 94 (1): 98-114.
25. Rural Development Administration.2015. Greater Sharing of Agricultural Technology with the World. Available at: www.rda.go.kr/foreign/ten/wwd_sha.jsp (Accessed: 15 September, 2020).
26. Sachs, J. 2008. Common Wealth: Economics for a Crowded Planet, Penguin Press HC ISBN 978-1-59420-127-1.
27. Santos, S. J. and Tenreyro, S. 2006. The Log of Gravity. The Log of Gravity. 88: 641-658.
28. StataCorp.2019. Stata statistical software: Release 16. College station, TX: StataCorp LP.
29. Tafirenyika, M. 2014. Intra-Africa Trade: Going Beyond Political Commitment, Africa Renewal, 28(2): 5-6, https://www.un.org/africarenewal/sites.
30. Turkson, F. 2012. Trade Agreements and Bilateral Trade in Sub-Saharan Africa: Estimating the Trade Effects of the EU-ACP PTA and RTAs.
31. Vernon, R. 1966. International Investment and International Trade in the Product Cycle Quarterly Journal of Economics, 80: 190-207.
32. Westerlund, J. and Wilhelmsson, F. 2011. Estimating the Gravity Model Without Gravity Using Panel Data. Applied Economics. 43: 641-649.
33. Xing, Z. 2018. The Impacts of Information and Communications Technology (ICT) and E-commerce on Bilateral Trade Flows. Int Econ Econ Policy 15: 565-586.