Export Sophistication Effects of Trade Facilitation Reform in Sub-Saharan Africa

Export Sophistication Effects of Trade Facilitation Reform
in Sub-Saharan Africa
Moukaila Mouzamilou Takpara1*, Constant Fouopi Djiogap2, Manfred Kouty3,

Bouraïma Sawadogo4
Abstract
This paper estimates the effects of trade facilitation reform on export sophistication,
measured by the Economic Complexity Index (ECI) in Sub-Saharan Africa (SSA) over
the period 2004-2017. It uses four indicators of trade facilitation capturing physical
infrastructure, ICT, business, and regulatory environment as well as border, and transport
efficiency. The empirical analysis using Ordinary Least Square, Two-Stage Least Square
and system Generalized Method of Moment shows that effective trade facilitation reform
targeted particularly at improving information and communication technology as well as
border, and transport efficiency contributes to promoting the export of sophisticated
goods in SSA. These findings suggest that reforms aimed at simplifying border
procedures should be implemented and accelerated in SSA countries. Hence, the
acceleration of the implementation of the World Trade Organization Trade Facilitation
Agreement will be an essential step toward achieving this goal.
Keywords: Trade facilitation; Export sophistication; Sub-Saharan Africa.
Jel Classification: Fl3; F14; 024.
1
PhD, Pan-African University - African Union Commission, Yaoundé – Cameroon, E-mail:
takpara2025@gmail.com,
*Corresponding author
2
Associate Professor in Economics, University of Yaoundé II, Cameroon, Department of Economics,
CEREG, E-mail: constantfouopi@yahoo.fr,
3
Assistant Professor in Economics, University of Yaoundé II International Relations Institute of
Cameroon, WTO-Co Chair, Yaoundé – Cameroon, E-mail: koutymanfred@yahoo.fr,
4
PhD Candidate, Pan-African University - African Union Commission, Yaoundé – Cameroon, E-
mail: bouraimasaw@gmail.com.
1
1- Introduction
Recent studies on international trade explain economic development and growth as a
process of developing information and learning to produce and export more
sophisticated products (Hidalgo et al. 2007; Saviotti and Frenken, 2008; Abdon and
Felipe, 2011; Felipe, 2012; Cristelli et a1. 2013; Haussmann et a1. 2014; Albeaik et a1.
2017). Indeed, since the 2000s, export sophistication5 has become a condition for
countries, particularly low-income countries to make their exports competitive. The
first theories on economic sophistication stem from the works on product classification
(Lall et al. 2006; Hausmann et al. 2007). Hausmann et a1. (2007) show that all types
of products do not have the same impact on economic development. These authors
measure economic sophistication by the level of productivity associated with a
country's exports and show that developing countries that manage to export products
like those produced and exported by high-income countries generate higher growth.
Increasing productivity is an issue of great importance to developing countries. They
have more difficulties improving their export basket (Harding and Javorcik, 2012).
Therefore, it becomes very important to understand the factors that determine a
developing country’s economic sophistication, since the latter is considered as an
important determinant of structural transformation. According to Lectard and Rougier
(2018), the sophistication and diversification signal the emergence of new industries
that are more capital-intensive and competitive enough to become exporters.
Since the seminal work of Hausmann et a1. (2007) on export sophistication, many
studies have examined the determinants of countries’ export sophistication, showing
positive effects of income level, human capital, productive investment, and foreign
direct investment (Cabral and Veiga, 2010; Spatafora et al. 2012; Weldemicael, 2012;
Zhu and Fu, 2013; Saadi, 2020). More recent works have provided evidence that it also
provides a substantial benefit to developing countries in terms of production stability
(da Costa Neto and Romeu, 2011). However, those studies on the determinants of
export sophistication are still limited, particularly for developing countries, especially
those in SSA (Zhu and Fu, 2013). According to Kamguia (2021), over the period 2000-
2010, Africa recorded the lowest level of export sophistication compared to Asian and
European countries, which are the most sophisticated ones. Jouini et al. (2016) argue
that African countries seem to suffer from a double handicap regarding their
productive and export structure: insufficient diversification coupled with low
sophistication. If export sophistication determines a country's growth and development
model, an interesting question arises: what contribution could trade policies,
particularly trade facilitation, make to export sophistication?
Trade Facilitation (TF) has become an important part of the current debate on trade
liberalization policy. It was at the heart of the World Trade Organization (WTO) negotiating
agenda since 2004. The ninth WTO Ministerial Conference held in Bali in December 2013 led
5
It is the export of new products and better-quality varieties of existing or new products (Lectard and
Rougher, 2018). It is also referred to the export of complex goods, i.e., goods of high value added, that cannot
be easily produced by other countries (Gnangnon, 2021).
2
to the adoption of the Trade Facilitation Agreement (TFA), which entered into force on 22
February 2017 following its ratification by two-thirds of the WTO membership. The TFA
contains obligations and disciplines that address many of the issues related to barriers
to global trade and trade costs (WTO, 2015). In a narrow sense, it is limited to the
logistics of moving goods through ports or the more efficient flow of documentation
associated with cross-border trade. But its broader definition includes infrastructural
aspects (Portugal-Perez and Wilson, 2012).
Most of the empirical studies on the effectiveness of TF reforms have generally
focused on countries’ export performance in terms of export volumes (Wilson et al.,
2003; Motsé and Sorescu, 2013; de Melo and Wagner, 2018) or sometimes, in terms of
export diversity (Lee and Kim, 2012; Persson, 2013; Feenstra and Ma, 2014; Beverelli
et al. 2015). It has been found that TF has contributed significantly to the increase in
exports at the intensive and extensive margins of countries (Beverelli et al. 2015).
While such an increase in exports can contribute to economic growth, recent studies
on trade and economic development point to the more critical role played by the
composition of a country's export basket in its economic performance (Hausmann et
a1., 2007; Minondo, 2010; Jarreau and Poncet, 2012; Grancay et a1. 2015). For these
authors, countries whose export basket contains more sophisticated products enjoy
higher economic growth. In other words, countries that are specialized in products with
high productivity, and sophisticated products will have greater export sophistication
and will tend to have higher and faster growth.
In this study, we contribute to the empirical literature on TF by assessing its
contribution to export sophistication in SSA countries. This paper is closed to that of
Spence and Karingi (2014) who analyzed the effects of TF on SSA countries’ export
sophistication using the EXPY index of Hausmann et a1. (2007). However, the value added
of our study compared to their work is threefold. First, unlike their study, our paper uses the
Economic Complexity Index (ECI). To the best of our knowledge, this paper presents
the first analysis of trade facilitation—export sophistication nexus for Sub-Saharan
African countries using the Economic Complexity Index. The ECI gives a holistic
measure of knowledge accumulation and sophistication in an economy, by combining
information on both the diversity of a country and the ubiquity of its products (Hidalgo
and Hausmann, 2009; Hartmann et al., 2017). Also, the focus on ECI which captures
economic sophistication6 as a potential outcome of fostering trade facilitation is dictated by the
fact that, it is an important driver of current economic growth as well as a good predictor of
countries' future economic growth and economic development patterns (Albeaik et a1. 2017;
Gnangnon; 2021). Second, our study covers a more recent period (2004-2017), unlike the
study by Spence and Karingi (2014) which covered only 2004-2007. The need to actualize
the work is particularly important due to the fact that, despite the enormous benefits that
may be derived from the effective implementation of trade facilitation reform, Africa's
level of trade facilitation remains low compared to other regions of the world (Sakyi
6
Some authors like Gnangnon (2021) refers “economic sophistication” to the concept of economic complexity.
According to Atasoy (2021), ECI shows the overall sophistication of the economy since it incorporates all
manufactured goods including exported goods.
3
et a1., 2018). Third, in our study, we test some theoretical channels through which trade
facilitation can affect export sophistication namely the intermediates imports and the foreign
direct investment channels.7
The analysis uses four comprehensive trade facilitation indicators (physical
infrastructure, ICT, business environment, border efficiency). Our results show that trade
facilitation indicators affect positively and significantly export sophistication in SSA. TF
induces greater export sophistication in countries that attract foreign direct investment flows,
the greater the level of FDI, the higher is the magnitude of the positive effect of trade facilitation
on export sophistication. Trade facilitation always exerts a positive effect on export
sophistication, but the magnitude of this positive effect decreases as countries experience
higher intermediates imports.
The remainder of the study is organized as follows. Section 2 provides a literature
review on the subject matter. Section 3 is devoted to data and stylized facts. Section 4
presents the econometric approach used in the study, and Section 5 analyzes and
discusses the results. Section 6 concludes with some policy implications.
2- Related literature review

While trade facilitation directly affects trade flows, the channel through which it affects
export sophistication remains ambiguous. Therefore, researchers have conducted
extensive research on the link between trade facilitation and export sophistication in order
to find the theoretical mechanism linking both of them (Bas and Strauss-Kahn, 2015; Hu,
2020). According to these authors, there are two potential channels for the impact of trade
facilitation on export sophistication, one is intermediate goods imported from developed
countries, and the other is foreign direct investment (Hu, 2020). For the intermediate goods
imported from developed countries, trade facilitation can lead to the "technology spillover"
effect of imported intermediates, improve trade efficiency, make it easier for exporting
firms to obtain many imported advanced technology intermediates, and help them
accelerate their R&D and innovation capabilities. According to Blalock and Veloso,
(2007), when enterprises have technical "bottlenecks" in product quality improvement,
high-quality imports are an effective way to promote product quality improvement. In the
same vein, facilitating trade can also generate "learning" and "sharing" effects by
combining the externalities of industrial agglomeration, then making the "spillover" effect
an essential mechanism for export sophistication (Duranton and Puga, 2004). Ultimately,
export sophistication will increase, and the structure of trade will shift from "quantity
orientation to quality orientation. Bas and Strauss-Kahn (2015) have examined the effect
of import trade facilitation on the prices of imported inputs and export products, and
observed that TF measures will reduce the cost of high-quality intermediate products
imported from developed countries, and import equipment for domestic exports.
Intermediate products will promote quality improvement.
Concerning the foreign direct investment channel, TF reduces transaction costs,
import and export times as well as improving the access of advanced foreign products to
7
See the discussions on both channels in the next section.
4
the domestic market, and increases the level of competition in the domestic market.
Consequently, trade facilitation leads to the elimination of low-productivity firms for the
benefit of high-productivity firms, which will at the end increase the productivity of the
country as a whole (Baumol and Lee, 1991) cited by Hu et al. (2022). Also, FDI uses the
learning-by-doing effect produced by the implementation of trade facilitation, which can
shorten the negotiation time between export enterprises and foreign partners. This can
reduce investment costs, and improve local business models for enterprises located in
export regions; fully utilizes the comparative advantages of related industries and absorb
advanced technologies from FDI, thereby increasing the productivity and sophistication
level of exports. Dunning (2000) confirms this theoretical prediction. For this author,
foreign producers could bring their own technology and know-how to produce
sophisticated goods that the host country could not produce due to inadequate labor,
capital, and technology. The quality of the goods produced will be impacted directly. FDI
could also promote export sophistication by taking advantage of knowledge spillovers
from foreign firms to locals (Lesher and Miroudot, 2008). According to Hausmann and
Rodrik (2003), this indirect effect, that causes the surge in efficiency and productivity, is
called “cost discovery". In the same vein, for Gnangnon (2021), FDI inflows positively
influence economic complexity since they are fundamental for the acquisition and
transfer of knowledge8. Trade facilitation can reduce the fixed costs of firms engaged
in trade activities and lower the productivity threshold that meets export standards,
thereby promoting export diversification and export sophistication (Hu, 2020).
Few studies have also highlighted the role of trade facilitation on export
sophistication. (Felipe, 2012; Weldemicael, 2012; Spence and Karingi, 2014;
Chakraborty and Mukherjee, 2016; Nguyen, 2016; Lapatinas, 2019; Atasoy, 2021;
Hu, 2020; Hu et a1., 2022). For example, Lapatinas (2019) has studied the effect of
the internet on economic sophistication. The author has found a positive effect of the
internet on economic sophistication measured by the economic complexity index. The
results of Atasoy (2021) confirm those of (Lapatinas, 2019). Indeed, in his study on
the role of digitalization captured by the use of ICT in export sophistication in 61
countries including 33 emerging countries and 28 developed countries during the
period 1995-2017, Atasoy (2021) has found that exports become more sophisticated
as digitization progresses. In another dynamic, Hu (2020) has examined the impact of
trade facilitation and R&D innovation on the sophistication of manufacturing exports
of Russia and Central and Eastern European countries over the period 2003-2017. The
author has found that trade facilitation and R&D innovation contributed significantly
to export sophistication of Russia and Central and Eastern European countries,
especially manufacturing after the 2009 financial crisis. For Central European
countries and EU member states, the impact of trade facilitation and R&D innovation
on export sophistication is significantly higher than for Eastern European countries
and non-EU states. The study of Hu et a1. (2022) confirms the finding in European
countries. In fact, Hu et a1. (2022) have examined the impact of trade facilitation on
8
See empirical study for more details about the relationship between FDI and export sophistication (Spatafora
et al. 2012; Zhu and Fu, 2013; Cabral and Veiga, 2010).
5
the export technological sophistication of the European transition economies from
2004 to 2020. With doing business data, they found that trade facilitation significantly
impacts the export technological sophistication of transition economies in general.
In the case of Africa, using the ’EXPY’ index, Spence and Karingi (2014) have
observed that in Sub-Saharan Africa, physical infrastructure and border and transport
efficiency promote export sophistication. Analyzing the effect of TF on export
sophistication, they found that a 1% improvement in physical infrastructure increases
export sophistication by 0.69%, while a 1% improvement in border efficiency
increases export sophistication by 0.29%.
Finally, there is a literature also on aid for trade and export sophistication
(Gnangnon 2021; Luu Hai; 2021). Indeed, aid for trade flows for economic
infrastructure for example which can be associated with the hard aspect of TF could
reduce trade costs and hence, promote the export of complex products. For these
authors, developing roads and port infrastructure reduces trade costs and improves
trading firms’ competitiveness, as well as incentivizes them to invest in R&D that is
aimed at identifying new export and complex products. Additionally, an improvement
in people's access to ICT tools reduces costs and supports knowledge and innovation
improvement.
Summing up, from the existing literature, most of the above studies linking trade
facilitation with export sophistication have focused on some selected countries and other
regions in the world. Studies on Africa are scarce, (e.g., Spence and Karingi; 2014).
However, Spence and Karingi (2014) have used the EXPY index which has some
shortcomings. Our study aims to fill these gaps.
In light of the foregoing, we hypothesize that as FDI inflows are essential for export
sophistication, countries that attract more FDI are likely also those that export
sophistication products. Therefore, we expect greater FDI to be associated with greater
export sophistication. Overall, trade facilitation will likely lead to export sophistication in
countries that attract more FDI (theoretical channel 1). Also, countries that import
intermediates products are likely also those that export sophisticated products. We expect
trade facilitation to result in greater export sophistication in countries that import
intermediates products (theoretical channel 2).
3- Data and stylized facts

In this section, we present the data and analyze the facts between TF and export sophistication.
3-1 Data
The export sophistication effects of trade facilitation reform are analyzed using
international institutions’ databases. The study focuses on 22 Sub-Saharan African
countries and covers the period 2004-2017. The sample and period are justified by the
availability of data on trade facilitation9 and those on the Economic Complexity Index.
9
Many studies using trade facilitation variables have encountered this challenge (see for example Portugal-
Perez and Wilson (2012); Sakyi et al. (2018).
6
The definition of variables is provided in appendix 3.
Export sophistication: It is our dependent variable and captures the level of
sophistication of an economy. To measure the export sophistication, authors use
various indexes (for example the product classification of Lall et a1. (2006), the EXPY
index of Hausmann et a1. (2007), etc.). The EXPY measure of Hausmann et a1. (2007)
aims to capture the level of productivity associated with a country's exports. However,
this measure of export sophistication has been criticized because its computed using
the income level of countries (Hidalgo and Hausmann, 2009; Andreoni, 2011). For
Hidalgo and Hausmann (2009), it mechanically raises a problem of circularity,
according to which "rich countries export products from rich countries". To address
this shortcoming of the EXPY measure of export sophistication, another variable used
in the literature is the Economic Complexity Index (ECI) from the Atlas of Economic
Complexity of Harvard University. First introduced by Hausmann et a1. (2014), ECI
is a holistic measure of knowledge accumulation and sophistication in an economy.
The ECI measures the sophistication of a country's productive structure by combining
information on both the diversity of a country (the number of products it exports) and
the ubiquity of its products (the number of countries that export these products)
(Hidalgo and Hausmann, 2009). The idea behind ECI is that sophisticated economies
are diverse and export products that, on average have low ubiquity because only a few
different countries can produce these sophisticated products. Similarly, less
sophisticated economies are expected to produce a few ubiquitous products (exported
by several countries). ECI exploits this variation in a country’s product diversity and
ubiquity to create a measure of the country's productive structure that incorporates
information on product sophistication (Hartmann et al., 2017). In view of its holistic
nature, we use the ECI in this study. ECI takes higher values when the economy
becomes more sophisticated10. In 2017 for example, Japan leads the index with an ECI
score of 2.18 and South Sudan is the worst performer with its ECI score of −2.1411.
Trade facilitation: This is our variable of interest. It should be noted that the
literature has proposed various indicators that make it possible to understand trade
facilitation reforms. These measures bring together different elements that determine
the extent to which the environment is conducive to economic and business
performance (Wilson et al. 2003; Francois and Manchin, 2013; Portugal-Perez and
Wilson, 2012 etc.). The four indicators used in this study are categorized in two
groups (hard and soft infrastructure12) including:
i) Physical infrastructure which measures the level of development and
quality of roads, ports, airports, and rail infrastructure. It is made up of four
sub-indicators: quality of the road network infrastructure, quality of the
railway infrastructure, quality of the port infrastructure, and quality of the
air transport infrastructure (airport connectivity).
10
Note that, ECI is consistent with the existence of income differentials between countries (Minondo
and Requena-Silvente, 2013).
11
https://oec.world/en/rankings/eci/hs6/hs96?tab=rank, consulted on 14/09/2022.
12
See Portugal and Wilson (2012) for details.
7
ii) Information and Communication Technology is interpreted as the extent to
which an economy uses ICT to improve efficiency and productivity, as well
as to reduce transaction costs. It contains indicators on the availability, use,
absorption, and priority given by the government to ICT;
iii) Business and regulatory environment measures the level of development of
regulations and transparency. It is based on indicators of irregular payments,
government transparency, and anti-corruption measures;
iv) Border and transport efficiency aims to quantify the level of customs and
inland transport efficiency which is reflected in the time, cost, and the
number of documents required for export and import procedures. 13
Following previous studies, we use GDP/per capita, human capital, FDI,
intermediates imports, financial development, the average tariff, and natural resources
as other explanatory variables14.
3-2 Stylized facts

Figure 1: Evolution of export sophistication in Sub-Saharan Africa (average 2004-2017)
Source: Authors based on the Atlas of Economics Complexity data.

Figure 1 above shows the evolution of export sophistication for SSA countries during the
period 2004-2017. The analysis in this figure shows that most countries in SSA (90% of the
sample) have negative ECI indexes (ranging between -2 and 0). This reflects the fact that
most SSA countries in the sample export less sophisticated goods. Among those countries,
Nigeria, Mauritania, Mozambique, Ethiopia, Burkina Faso, and Cameroon are the worst
performer with scores ranging between -2 and -1. The two countries that perform relatively
well with positive ECI index among SSA countries are Eswatini (0.468) and South Africa
(0.078). These two countries seem to export relatively sophisticated goods compared with
13
See details about the four TF indicators in appendix 2.
14
See the discussion on their expected effects in the empirical specification sub section.
8
other SSA countries in the sample. It should be noted that, over the period, the values of
ECI range between -2.424 and 1.220, while the average ECI index of the SSA zone over the
study period is (-0.764) (see appendix 2).
Figure 2 analyzes the correlation between the four trade facilitation indicators and export
sophistication. It can be observed that there is a positive and statistically significant
correlation between trade facilitation indicators and export sophistication, which suggests
that they are moving positively in the same direction. This is confirmed by the positive
and statistically significant correlation between trade facilitation indicators and export
sophistication at the 1% level (see the correlation coefficients in appendix 1). There is a
disparity between countries. Indeed, countries with a high trade facilitation score, seem to
favor export sophistication in SSA and vice versa. For example, countries (Eswatini and
South Africa) with a high physical infrastructure score have a high index of export
sophistication. Conversely, countries such as Mauritania and Nigeria have low average
physical infrastructure scores and negative average export complexity indices close to -2
(low complexity goods). Thus, it can be said that an improvement in physical infrastructure
contributes to the improvement in export sophistication. Also, having efficient borders
with simplified procedures in terms of time and documents for export/import operations
seems to favor export sophistication in SSA countries because countries like Mauritius,
Eswatini, Namibia, Botswana, and South Africa, which have efficient borders, have also a
relatively high economic complexity index.
Figure 2: Correlation between trade facilitation and export sophistication (average 2004-2017)
Source: Authors based on WEF data and Atlas of Economics Complexity data.
9
4- Econometric approach
In this section, we present the empirical specification (4-1); next, we present the estimation
technique (4-2).
4-1 Empirical specification
To study the effects of trade facilitation on export sophistication, we focus on the
Economic Complexity Index (ECI) as the main measure of export sophistication but
repeat the analysis using the EXPY measure of Hausmann et al. (2007) for the
robustness checks. We follow Lapatinas (2019); Atasoy (2021) and Luu Hai (2021),
and estimate the baseline Equation (1) below:
𝑬𝑪𝑰𝒊𝒕 = 𝜷𝟏 𝑮𝑫𝑷𝒑𝒄𝒊𝒕 + 𝜷𝟐 𝑬𝒅𝒖𝒊𝒕 + 𝜷𝟑 𝑭𝑫𝑰𝒊𝒕 + 𝜷𝟒 𝑰𝒏𝒕𝑴𝒊𝒕 + 𝜷𝟓 𝑭𝒊𝒏𝑫𝒆𝒗𝒊𝒕 +
𝜷𝟔 𝑰𝒏𝒔𝒕𝒊𝒕 + 𝜷𝟕 𝑻𝒂𝒓𝒊𝒇𝒇𝒊𝒕 + 𝜷𝟖 𝑵𝒂𝒕𝑹𝒆𝒔𝒊𝒕 + 𝜷𝟗 𝑻𝑭𝑰𝒊𝒕 + 𝜼𝒕 + 𝝁𝒊 + 𝜺𝒊𝒕 (1)
Where ECIit represents the Economic Complexity Index, which measures the
sophistication of exports of country i at period t. It is computed using the methodology
described in Hidalgo and Hausmann (2009). We include in the model a set of control
variables that are deemed to influence export sophistication. These variables include GDP
per capita (denoted GDPpc); the level of education (denoted Edu), the net foreign direct
investment inflows (denoted "FDI"); the intermediates imports (denoted "IntM"); the
financial development (denoted "FinDev"), the institutional quality (denoted "Inst"); the
tariff applied (denoted "Tariff"), the natural resources dependence (denoted "NatRes");
TFI; represents the vector of trade facilitation indicators (physical infrastructure, ICT,
trade and regulatory environment, border efficiency); ηt are the time-fixed effects; μi; is
a vector representing country fixed effects; εit is the error term.
GDP per capita: It is a determinant of export sophistication suggested in the
literature. According to the supply and demand theories, when GDP per capita
increases, consumption preferences change (Aghion and Howitt, 1990; Fiorillo, 2001).
The potential risks of diversification decrease with increasing income, and export
diversification increases with increasing GDP per capita (Imbs and Wacziarg, 2003).
By construction, the measurement of export sophistication is strongly correlated with
the income level of exporting countries. We, therefore, expect a positive effect of per
capita GDP on export sophistication.
Foreign direct investment: it is captured by the ratio of net foreign direct investment inflows
to GDP, and measures knowledge and technology spillovers. By facilitating the transfer of
knowledge, technology, and managerial skills, FDI can promote the production and
export of more complex goods by destination countries (Hausmann, 2016).
Intermediates imports: It includes all parts and accessories, as well as industrial
primary and processed intermediate products. We retain the narrow definition of
intermediates imports adopted by the WTO15, which includes the HS subheadings
corresponding to codes 42, 53, 111, 121, 21, and 22 of the Broad Economic Categories
(BEC) classification. Since the import of intermediate products can contribute to
production, we, expect that it would promote exports of goods, including sophisticated
ones.
15
See the WTO report 2015 titled: International Trade Statistics 2015; www.wto.org/statistics.
10
Human capital: It refers to all the capacities learned by individuals which increase
their productive efficiency. Education level is generally positively associated with
productivity and export sophistication in the literature (Hidalgo and Hausmann, 2009;
Faruq, 2011). Indeed, investment in education and health improves the quality of the
labor force, increases productivity, and thus promotes export sophistication. In this
study, we use the enrollment rate as a human capital index and expect a positive effect.
Quality of institutions: The literature recognizes institutional factors as one of the
most important determinants of export sophistication. Among the set of quality of
institutions indicators reported by Kaufmann et al. (2010), we choose the control of
corruption indicator as a proxy for the quality of institutions. Indeed, reducing
corruption within a government can reduce uncertainty and increase producers’
incentives to invest and innovate; which improves export diversification and
sophistication. It should be noted that corruption within the bureaucracy in countries
hinders their trade. We expect it to affect positively export sophistication.
Access to credit: Access to credit is an important component of financial development
that facilitates technological innovations through its intermediary role. The literature
postulates that as firms’ access to credit improves, their chances of diversifying and
upgrading their exports increase. Thus, there is a positive correlation between financial
development and the sophistication of economies (Fang et al. 2015; Saadi, 2020). The
ratio of private sector credit to GDP, which refers to the financial resources provided
to the private sector by financial firms, is used in this study.
The average applied tariff: this is the average of the tariff rates actually applied,
weighted by the import shares of products corresponding to each partner country. It is
a proxy for trade policy and higher tariffs are likely associated with low levels of
export sophistication.
Natural resources: The existence of natural resources in a country is often seen as a
constraint on a country's export diversification and sophistication. A country's
production and export of natural resources benefit relatively little from innovation and
productivity gains, so an economy highly specialized in these sectors would generally
experience only fairly slow productivity improvements and consequently very slow
diversification and sophistication processes. It is also argued that the effect of natural
resources on export sophistication could depend on whether countries rely on these
resources or not (Gnangnon, 2021). Total natural resource rents as a percentage of GDP
is used as a proxy for natural resources dependence in this study like Luu Hai (2021).
4-2 Estimation technique

To examine the contribution of trade facilitation to export sophistication, it
is important to choose an estimation strategy that addresses challenges due to the
heterogeneity of African countries: cross-sectional dependence, heteroscedasticity,
and serial correlation. It is argued that these challenges are an important problem in
panel data, and ignoring them in the estimation may provide invalid test statistics.
For example, in the modern economy, cross-sectional dependence cannot be denied
11
due to higher economic, financial, and trade integration. For Khan et al. (2020), it
cannot be ignored also in the backdrop of shared global shocks and shared global
spillover impacts among countries. We first estimate the equation (1) with Ordinary
Least S quares (OLS), and we take into account these issues raised above by using the
country-year cluster in the regression16.
It is possible also that model 1 could suffer from the reverse causality
between export sophistication and trade facilitation indicators. Reverse causality
bias can occur when better infrastructure (hard and soft) influences export at an
extensive margin including the export of sophisticated products, the return gained
from exports can also be used to invest in hard and soft infrastructure development.
In addition, the inclusion of institutional and economic variables in the same
equation raises a reverse causality (Faruq, 2011). In other words, institutions may
affect economic outcomes including export sophistication, while economic outcomes
may also affect the quality of institutions. Also, export sophistication is not
measured by certainties, which could rise a measurement bias.
To address the endogeneity issues mentioned above, we employ an instrumental
variable estimation: a Two-Stage Least Square (2SLS) following Banerjee et al.
(2022) who have employed lag one and lag two as instruments and derived the
predicted values. Thus, equation (1) can be specified as:
̂ 𝑰𝒊𝒕 + 𝜼𝒕 + 𝝁𝒊 + 𝜺𝒊𝒕
𝑬𝑪𝑰𝒊𝒕 = 𝜷𝟏 𝑿𝒊𝒕−𝟏 + 𝜷𝟐 𝑻𝑭 (2)
̂ 𝑰𝒊𝒕 = 𝜷𝟏 𝑿𝒊𝒕−𝟏 + 𝜷𝟐 𝑻𝑭𝑰𝒊𝒕−𝟏 + 𝜼𝒕 + 𝝁𝒊 + 𝜺𝒊𝒕
𝑻𝑭 (3)
̂ 𝑰𝒊𝒕 = 𝜷𝟏 𝑿𝒊𝒕−𝟏 + 𝜷𝟐 𝑻𝑭𝑰𝒊𝒕−𝟐 + 𝜼𝒕 + 𝝁𝒊 + 𝜺𝒊𝒕
𝑻𝑭 (4)
Where 𝑇𝐹𝐼𝑖𝑡−1 is the one-period lagged value and 𝑇𝐹𝐼𝑖𝑡−2 is the two-period lagged value of
TFI in country i and at time t. 𝑋𝑖𝑡−1 are lag-one period of our control variables in order to
minimize their potential endogeneity except for variable natural resources that is exogenous.
The rest of other variables remain the same as in Equation (1). Equations (3) and (4) are the
first stage equations where we isolate the effect of trade facilitation. In the second stage, the
̂𝑖𝑡 ) derived from Equations (3) and (4) are inserted into
fitted values of trade facilitation (𝑇𝐹𝐼
Equation (2) to address the endogeneity issue between export sophistication and trade
facilitation.
16
The two-way clusters correct the potential presence of heteroscedasticity and autocorrelation. The diagnostics
tests in appendix 4 confirm their presence.
12
5- Results
This section first presents and interprets the results of our estimations and then
performs robustness checks of the findings.
5-1 Presentation and interpretation of results
The results of the effect of trade facilitation reform on export sophistication with
OLS and 2SLS are presented in Table 1 and Table 2 respectively. It should be noted that
our preferred estimator is two-stage least square. The diagnostic tests that allow for the
checking of the consistency of the two-stage least square approach are reported at the bottom
of Table 2. First, we note the rejection of the null hypothesis of under-identification since the
p-values of Kleibergen and Paap, (2006) are zero for all specifications. Our model is therefore
correctly identified. Second, when we compare the Cragg-Donald Wald F statistics values to
the critical values of Stock and Yogo, (2005) to determine instrumental variable bias and size
bias, we reject the weak instrument null hypothesis since the values of the statistics are greater
than the critical values of Stock and Yogo, (2005). The Hansen’s p-values also are greater than
10% for all specifications. Based on the above, the two-stage least square estimator is
appropriate for conducting the empirical analysis.
Our variables of interest have the expected sign according to economic theory, with
varying statistical significances. The trade facilitation index (average) and three out of the
four trade facilitation indicators affect positively and significantly export sophistication in
SSA. The coefficient of the trade facilitation index is positive and statistically significant
at 5% level. Its coefficient is 1.30 (column 1) implying that, a l % improvement in the
average trade facilitation index is roughly associated with a 1.30-point increase in the
export sophistication in SSA. For the specific TF indicators, we note that physical
infrastructure positively affects export sophistication in SSA at 1% threshold. Its
coefficient is positive and significant and indicates that all else being equal, a 1%
improvement in the quality of road, rail, airport, and port infrastructures leads to a 1.14-
point increase in the export of sophisticated or complex goods. Indeed, this result suggests
that having better physical infrastructure reduces the transport costs associated with
inputs, and increases productivity and the production of quality goods, thus
encouraging firms to export more sophisticated products. This result is consistent with
the finding in the literature (Spence and Karingi, 2014). These authors found in their
work that physical infrastructure positively and significantly affects export
sophistication in SSA. The information and communication technology plays an
important role on export sophistication in SSA. The coefficient of ICT is positive and
statistically significant at the 5% level and reveals that an increase in the availability
of technology and its utilization in the production process of 1 % is associated with an
increase in export complexity in SSA by 0.73-point, all other things being equal. The
use of ICT enables innovation, which favors the introduction of new, more
sophisticated production methods. This result supports those found in the literature
that ICT promotes export sophistication (Rodriguez-Crespo and Martinez-Zarzoso,
2019; Lapatinas, 2019; Atasoy, 2021). The border and transport efficiency indicator is
found to be an important factor in explaining exports of sophisticated goods in SSA.
13
Indeed, it has a positive and statistically significant contribution at the 1% level and
indicates that a 1 % reduction in border procedures leads to a 1.21-point increase in
exports of sophisticated goods, all else being equal. In other words, simplifying border
procedures, i.e., reducing the number of days and the number of export/import
documents promotes the export of sophisticated goods. This result is consistent with
those found by Spence and Karingi, (2014). The business and regulatory environment
affects positively but not significantly the export of sophisticated goods. Among the
four trade facilitation indicators, it appears that border and transport efficiency, and
physical infrastructure are the ones that show the highest positive and significant effect
on export sophistication.
With respect to the control variables, the results support mostly the theory and
reveal the crucial role of the credit to the private sector in promoting export
sophistication. This variable has the expected sign in all specifications (columns 1-
5) and is significant at 1% in all specifications (table 2). Theoretically, granting credit
to firms makes it possible to finance not only physical capital but also other factors of
production with significant fixed costs. For Fosu and Abass (2019), a firm must have
sufficient liquidity through credit to sustain itself until future returns are realized,
given that returns from fixed costs are not immediate. Human capital captured by
the enrollment rate has the expected sign and is significant at 1% level in all
specifications (columns 1-4). The result shows that it contributes positively to export
sophistication. This result confirms the economic theory that workers are likely to be
more productive when they are more educated and skilled (Faruq, 2011). As a result,
countries with a larger stock of human capital specialize in producing and exporting
higher-quality products (Schott, 2004; Hummels and Klenow; 2005). This finding is
consistent with that of Faruq (2011), who found that improving human capital through
education is associated with better export sophistication. In contrast, GDP per capita
contributes negatively to export sophistication. This result contradicts the economic
literature. Hidalgo and Hausmann (2009) have shown in their study that countries tend
to converge towards the level of income dictated by the complexity of their productive
structures. Control of corruption affects positively and significantly export
sophistication at 1% level in columns 3 and 4 (Table 2). Thus, controlling corruption
increases the ability of producers to improve the quality of their exports. The natural
resources variable for its part is negative and significant at 1% level in all
specifications. According to economic theory and as shown in the results, natural
resources contribute negatively to export sophistication. This finding is consistent with
the finding by Yue and Zhou (2018) that abundant natural resources are good
predictors of the comparative disadvantage of goods. The outcomes also reveal that
FDI inflows contributes to promoting export sophistication. The coefficient of the
related variable is positive in all specifications and significant at 10% level in column
1 and at 5% in column 4. The role of intermediate imports, as well as FDI, will be
further e x p l o r e d in the analysis using their interactions terms with TF.
14
Table 1: Export sophistication effects of trade facilitation reform: OLS Results
Dependent Variable: Economic Complexity Index (ECI)
VARIABLES 1 2 3 4 5 6
Log (pc GDP) -0.191 -0.284* -0.104 -0.289* -0.209*** 0.063
(0.129) (0.140) (0.106) (0.136) (0.068) (0.120)
Secondary enrollment rate 0.011* 0.010* 0.013** 0.012* 0.002 0.009
(0.006) (0.006) (0.005) (0.006) (0.004) (0.005)
Foreign Direct Investment 0.010* 0.009* 0.006 0.006 0.010* 0.004
(0.005) (0.005) (0.004) (0.005) (0.005) (0.003)
Intermediates import -0.234 0.097 1.468* 0.313 0.516 0.526
(0.701) (0.662) (0.827) (0.614) (0.592) (0.657)
Private_sector_credit 0.007*** 0.006*** 0.003 0.007*** 0.006*** 0.003*
(0.001) (0.001) (0.002) (0.001) (0.001) (0.002)
Control_Corruption 0.221* 0.096 -0.076 0.212** -0.086 -0.036
(0.120) (0.118) (0.106) (0.095) (0.065) (0.122)
Log (Tariff) 0.552 0.212 -0.036 0.047 0.796 -0.280
(0.726) (0.872) (0.755) (0.814) (0.671) (0.693)
Natural resources -0.029*** -0.030*** -0.038*** -0.032*** -0.025*** -0.027**
(0.009) (0.008) (0.011) (0.008) (0.007) (0.010)
Trade Facilitation Index 1.290***
(0.417)
Physical Infrastructure 1.337*
(0.735)
Technology (ICT) 0.615***
(0.165)
Business and Regulatory Environment 0.153
(0.250)
Border and Transport Efficiency 0.945**
(0.427)
Constant -0.320 -0.361 -2.366*** -0.259 0.630 -2.711**
(0.682) (0.589) (0.771) (0.573) (0.430) (1.106)
Observations 234 234 234 234 234 234
R-squared 0.800 0.821 0.784 0.821 0.883 0.732
Number of countries 22 22 22 22 22 22
Note: Robust (clustered on country-year) standard errors in parentheses *, **, *** represent significance at
10%, 5% and 1% respectively. Regressions always included country and time dummies.
Source: authors.
15
Table 2: Export sophistication effect of trade facilitation reform: IV-2SLS Results
Dependent variable: Economic Complexity Index (ECI)
VARIABLES 1 2 3 4 5
Log (pc GDP) -0.341*** -0.072 -0.267*** -0.266*** -0.114*
(0.071) (0.057) (0.079) (0.065) (0.060)
Secondary enrolment rate 0.012*** 0.015*** 0.010*** 0.013*** 0.012***
(0.002) (0.002) (0.002) (0.003) (0.002)
Foreign Direct Investment 0.007* 0.004 0.004 0.008** 0.005
(0.003) (0.004) (0.004) (0.004) (0.004)
Intermediates import -0.098 1.332*** 0.291 -0.223 0.513
(0.364) (0.488) (0.392) (0.358) (0.469)
Private_sector_credit 0.006*** 0.003*** 0.005*** 0.008*** 0.004***
(0.001) (0.001) (0.001) (0.001) (0.001)
Control_Corruption 0.115 -0.093 0.175*** 0.225*** 0.022
(0.072) (0.068) (0.061) (0.076) (0.074)
Log (Tariff) -0.144 0.265 0.506 0.159 0.065
(0.591) (0.733) (0.585) (0.599) (0.814)
Natural resources -0.032*** -0.034*** -0.028*** -0.035*** -0.036***
(0.005) (0.006) (0.005) (0.006) (0.006)
Trade Facilitation Index 1.304**
(0.517)
Physical Infrastructure 1.146***
(0.362)
Technology (ICT) 0.727**
(0.350)
(0.275)
Border and Transport Efficiency 1.214***
(0.402)
Constant -0.012 -2.535*** -0.230 -0.001 -1.715***
(0.461) (0.450) (0.472) (0.398) (0.487)
Observations 203 203 203 203 203
R-squared 0.813 0.772 0.833 0.800 0.766
Underindentification test 58.38 37.51 40.60 33.94 61.19
Prob>LM 0.000 0.000 0.000 0.000 0.000
Weak Identification Test
170.551 229.943 62.122 145.990 190.633
(Cragg-Donald Wald F stat)
Stock-Yogo weak ID test critical values
10% maximal IV size 19.93 19.93 19.93 19.93 19.93
15% maximal IV size 11.59 11.59 11.59 11.59 11.59
20% maximal IV size 8.75 8.75 8.75 8.75 8.75
25% maximal IV size 7.25 7.25 7.25 7.25 7.25
Hansen_stat (overidentification test) 1.987 2.540 0.992 0.479 0.585
Hansen P_value 0.159 0.111 0.319 0.489 0.444
Note: Robust standard errors in parentheses *, **, *** represent significance at 10%, 5% and 1% respectively.
Regressions always included country and time dummies.
Source: authors.
16
Turning now to the results in Table 3, we have found across columns 1 that the
overall trade facilitation index exerts a positive and significant effect on export
sophistication, while the interaction term (associated with the variable capturing the
interaction between the overall trade facilitation variable and FDI) is positive and
significant at 5%. The positive effect suggests that the overall trade facilitation index
induces greater export sophistication in countries that attract foreign direct investment.
Among the four trade facilitation indicators, the ICT indicator has a positive
and significant effect on export sophistication at 5 % level, while the interaction term
between this variable and FDI is also positive and significant at 5% level. Taking
together, these two outcomes suggest that ICT induces greater export sophistication in
countries that attract FDI. In addition, the greater the level of FDI, the higher is the
magnitude of the positive effect of ICT on export sophistication. Similarly, the
positive effect of ICT on export sophistication rises as the degree of FDI increases.
Also, the border efficiency indicator has a positive and significant effect on
export sophistication at 5% level while the interaction term between this variable and
FDI is positive and significant at 10% level. These two outcomes reveals that the
reduction of trade cost by simplifying trade procedures induces greater export
sophistication in countries that attract FDI.
Results in table 4 show the interaction terms between trade facilitation and the
intermediates imports. Column (1) in Table 4 shows that the interaction term
associated with the overall trade facilitation index and the intermediates imports
(TF*IM) is negative and significant at 10% level, while the coefficient of the overall
trade facilitation is positive and significant at 10% level. These results suggest that
over the full sample, there may be a level of the intermediates imports above which
the effect of the overall trade facilitation changes sign and becomes negative. This
level of the intermediate import is 1.820 [= exponential (3.475/5.802)], which is far
higher than the maximum value (0.664, see appendix 2) of the intermediate’s imports
in the full sample. Thus, on average, over the full sample, the overall trade facilitation
always exerts a positive effect on export sophistication, but the magnitude of this
positive effect decreases as countries experience higher intermediates imports. This
result is confirmed when looking at the different trade facilitation indicators, which
are positive and significant, while the interactions terms associated with them are
negative and significant. In fact, physical infrastructure, border efficiency indicators
affect positively export sophistication at 5% and ICT is significant at 1%, while the
interactions terms affect negatively the export sophistication at 5% for physical
infrastructure and ICT. The level of the intermediate’s imports above which the effect
of the trade facilitation indicators changes sign and becomes negative is 1.845 [=
exponential (2.175/3.552)] for physical infrastructure and 2.237 [=exponential (2.321
/ 2.883)] for ICT. With the above, we can say that, trade facilitation exerts greater
effect on export sophistication through intermediates imports at a certain level. Thus,
we can conclude that both variables are complementary over the full sample.
17
Table 3: Estimation results: Interaction between Trade facilitation and FDI
VARIABLES 1 2 3 4 5
Log (pc GDP) -0.179*** -0.094 -0.466*** -0.266*** -0.164**
(0.051) (0.066) (0.093) (0.065) (0.070)
Secondary enrollment rate 0.003** 0.016*** 0.013*** 0.012*** 0.018***
(0.001) (0.002) (0.003) (0.003) (0.003)
Foreign Direct Investment (FDI) -0.054** 0.006 -0.002 0.005 -0.006
(0.025) (0.006) (0.006) (0.005) (0.007)
Intermediates import 0.206 1.090** -0.219 -0.306 -0.714
(0.349) (0.535) (0.430) (0.353) (0.442)
Private_sector_credit 0.006*** 0.003*** 0.009*** 0.008*** 0.004***
(0.001) (0.001) (0.001) (0.001) (0.001)
Control_Corruption 0.074 -0.042 0.159** 0.214*** 0.159*
(0.063) (0.080) (0.067) (0.078) (0.083)
Log (Tariff) 0.816 0.212 0.773 0.110 -0.236
(0.567) (0.746) (0.566) (0.595) (0.819)
(0.004) (0.006) (0.005) (0.006) (0.007)
Trade Facilitation Index (TF) 0.902*
(0.522)
Physical Infrastructure (PI) 1.070***
(0.370)
(0.249)
Business and Regulatory Environment (RE) 0.128
(0.277)
Border and Transport Efficiency (BE) 0.736**
(0.327)
TF*FDI 0.129**
(0.065)
PI*FDI 0.005
(0.029)
ICT*FDI 0.036**
(0.017)
RE*FDI 0.025
(0.021)
BE*FDI 0.019*
(0.011)
Constant -0.250 -2.559*** 1.091* 0.083 -1.140*
(0.387) (0.582) (0.642) (0.389) (0.590)
Observations 203 203 203 203 203

R-squared 0.759 0.779 0.808 0.802 0.688
Prob>LM 0.000 0.000 0.000 0.000 0.000
Weak Identification Test (Cragg-Donald
Wald F stat) 165.401 207.722 89.968 145.032 265.302

Hansen P_value 0.532 0.114 0.747 0.830 0.822
Source: authors.
18
Table 4: Estimation results with the interaction term between Trade facilitation
and Intermediates imports
VARIABLES 1 2 3 4 5
Log (pc GDP) 0.057 -0.179*** -0.286*** 0.041 -0.166**
(0.075) (0.053) (0.085) (0.063) (0.068)
Secondary enrollment rate 0.012*** 0.011*** 0.014*** 0.015*** 0.018***
(0.002) (0.003) (0.003) (0.003) (0.003)
Foreign Direct Investment 0.003 -0.005 -0.002 -0.000 0.002
(0.004) (0.004) (0.005) (0.004) (0.004)
Intermediates import (IM) 3.974** 1.142* 1.387** -0.115 0.057
(1.841) (0.691) (0.620) (0.637) (0.640)
Private_sector_credit 0.003* 0.005*** -0.000 0.004*** 0.004***
(0.002) (0.002) (0.004) (0.001) (0.001)
Control_Corruption -0.146* 0.233*** 0.176** 0.118 0.161*
(0.083) (0.059) (0.078) (0.113) (0.085)
Log (Tariff) -0.455 0.644 0.961 -0.165 -0.207
(0.799) (0.639) (0.634) (0.873) (0.835)
(0.005) (0.005) (0.005) (0.006) (0.006)
Trade Facilitation Index (TF) 3.475*
(2.076)
Physical Infrastructure (PI) 2.175**
(0.937)
Technology (ICT) 2.321***
(0.777)
Business and Regulatory Environment (RE) -1.104
(0.981)
Border and Transport Efficiency (BE) 1.452**
(0.649)
TF*IM -5.802*
(3.489)
P*IM -3.552**
(1.807)
ICT*IM -2.883**
(1.458)
RE*IM 1.540
(1.831)
BE*IM -1.381
(1.008)
Constant -4.181*** -1.118* -1.131* -1.965*** -1.588**
(1.043) (0.600) (0.670) (0.617) (0.784)
Observations 203 201 201 201 201

R-squared 0.745 0.768 0.791 0.759 0.698
Prob>LM 0.000 0.000 0.000 0.000 0.000
12.016 45.147 28.186 27.551 89.205
Wald F stat)
Hansen P_value 0.765 0.179 0.675 0.343 0.909
10%, 5% and 1% respectively. Regressions always included country and time dummies.
Source: authors.
19
5-2 Robustness of the results
We run several robustness tests to gain additional insights into the effects of trade
facilitation reform on export sophistication. To do so, we test the robustness by using an
alternative measure of export sophistication as well as an alternative estimation method.
5-2-1 Alternative measure of export sophistication
We use the export sophistication (EXPY index) measure of Hausmann et a1. (2007) in
place of the economic complexity index. All other variables remain the same.
The estimation results are presented in table 5 . We can see that with the alternative
measure of export sophistication, the average trade facilitation index and two out
of four trade facilitation indicators positively and significantly affect export
sophistication. Indeed, ICT, border, and transport efficiency contribute positively and
significantly to the export of sophisticated goods. Although physical infrastructure as
well as the business environment affects positively export sophistication, their
coefficients are not significant. Compared with the results in table 2, the coefficients
of TF indicators are small in table 5. This can probably be explained by the fact that
the ECI index takes into account the differential in income level and give real results
than ‘EXPY’. Regarding the control variables, gross domestic product per capita
positively and significantly affects export sophistication (column 1; 2; 3; 4; 5). This is
understandable since ‘EXPY’ is computed by the income level of countries, and it's in
line with (Hidalgo and Hausmann, 2009). The results of other control variables are
similar to the results in table 2 with differences in some cases.
20
Table 5: Export sophistication effects of trade facilitation reform: Alternative measure
Dependent Variable: log (EXPY)
VARIABLES 1 2 3 4 5
Log (pc GDP) 0.094** 0.157*** 0.085*** 0.130*** 0.080**

(0.039) (0.054) (0.030) (0.033) (0.040)
Secondary enrollment rate -0.001 0.007*** 0.001 -0.000 0.005***
(0.002) (0.001) (0.002) (0.002) (0.001)
Foreign Direct Investment 0.004* -0.006* -0.006* 0.004** -0.006*
(0.002) (0.004) (0.003) (0.002) (0.004)
Intermediates import -0.876*** -0.140 -0.803*** -0.953*** -0.258
(0.238) (0.252) (0.195) (0.268) (0.229)
Private_sector_credit 0.000 0.001 0.001*** 0.001** 0.001*
(0.000) (0.001) (0.000) (0.000) (0.000)
Control_Corruption -0.060* 0.003 0.055** -0.019 0.023
(0.035) (0.036) (0.026) (0.064) (0.035)
Log (Tariff) -0.255 0.151 0.114 -0.154 0.048
(0.389) (0.473) (0.346) (0.386) (0.418)
Natural resources 0.015*** 0.004 0.008*** 0.014*** 0.003
(0.002) (0.003) (0.002) (0.002) (0.003)
Trade Facilitation Index (TF) 0.608**
(0.257)
Physical Infrastructure 0.121
(0.237)
(0.098)
(0.313)
(0.178)
Constant 8.725*** 7.413*** 8.824*** 8.653*** 7.986***
(0.266) (0.375) (0.209) (0.291) (0.312)
Observations 203 203 203 203 203

R-squared 0.810 0.796 0.874 0.811 0.807
Prob>LM 0.000 0.000 0.000 0.000 0.000
165.277 181.817 182.747 142.787 190.633
Wald F stat)
Hansen P_value 0.390 0.288 0.680 0.403 0.297
Source: authors.
21
5-2-2 Alternative estimation method
Our third robustness test is performed using an alternative estimation method.
There is a concern with the OLS as well as 2SLS estimations. First, and foremost,
it is possible that export sophistication is persistent, which is addressed by
including his lagged variable in model 1.
Thus, to take into account the dynamic aspect of the model, we employ the system-
Generalized Method of Moment (GMM) (Blundell and Bond, 1988). We check the
relevancy conditions of this estimator with diagnostic tests (Arellano-Bond (AB) test
and the Hansen test).
The regression results with the system GMM estimator are globally valid. Our results
with the system GMM estimator presented in table 6 satisfy the conditions for the
use of GMMs, particularly with respect to the autocorrelation test and the validity of
the instruments. Indeed, the Arellano and Bond AR (2) test of no second-order
autocorrelation in the residuals is acceptable (the p-value > 0.10 for all estimates).
Regarding the validity of the instruments, Hansen's tests in all estimates indicate that the
internal instruments used are satisfactory (p-value > 0.10) and the ratio of the number of
instruments to the number of countries (N) is less than 1 in all regressions (number of
instruments < number of countries). Thus, there is no proliferation of instruments. The
results presented in table 6 consolidate the first results we obtained with the static model
(table 2). First, we note that in all specifications, the initial export sophistication is positive
and significant, suggesting a persistence of the export sophistication process in the long
term. The coefficients of our key variables with syst-GMM are in line with the economic
theory. The overall trade facilitation index as well as the four TF indicators have a positive
effect on export sophistication. In fact, the overall TF index is positive and significant at
1% level. Physical infrastructure and the business environment affect positively but
not significantly export sophistication, while the coefficient of ICT as well as the
border and transport efficiency indicator are positive and significant at 5% and 1%
respectively. Among the four trade facilitation indicators, border and transport
efficiency seem to play a key role in exporting sophisticated goods. Indeed, efficiency
in border and transport which mean simplification of documents as well as reducing
number of days required for export/import by 1% increase export of sophisticated
goods by 1.14-point. By controlling for endogeneity issue, our results give more
insight and are consistent. The control variables globally are similar to the result in
table 2. Although this result gives once again more insight into the effects of TF on
export sophistication, it should be noted that we do not have a large sample as required
for GMM (Roodman, 2009).
22
Table 6: Estimation results correcting for the persistence of export sophistication
VARIABLES 1 2 3 4 5 6
L.ECI_initial 0.444*** 0.547*** 0.359*** 0.488*** 0.373*** 0.424**

(0.080) (0.074) (0.089) (0.135) (0.131) (0.156)
Log (pc GDP) 0.059* -0.031 -0.034 -0.122 0.113** -0.048
(0.034) (0.045) (0.046) (0.087) (0.053) (0.048)
Secondary enrollment rate 0.002* 0.002** 0.002* 0.004** 0.003* 0.002
(0.001) (0.001) (0.001) (0.001) (0.001) (0.002)
Foreign Direct Investment 0.002 0.006* -0.000 -0.001 0.007 -0.004
(0.003) (0.003) (0.003) (0.003) (0.006) (0.004)
Intermediates import 0.208 0.293 0.306 0.392* 0.588 0.254
(0.227) (0.346) (0.260) (0.222) (0.480) (0.452)
Private_sector_credit 0.002*** 0.000 0.003*** -0.001 0.002* 0.002**
(0.000) (0.001) (0.001) (0.002) (0.001) (0.001)
Control_Corruption 0.051 -0.100 0.060 0.094 -0.043 -0.076
(0.053) (0.065) (0.055) (0.062) (0.165) (0.048)
Log (Tariff) -0.073 0.931*** -0.521 -0.509 3.696 -0.243
(0.604) (0.329) (0.497) (0.556) (3.339) (0.695)
Natural resources -0.026*** -0.021*** -0.033*** -0.014 -0.028*** -0.025**
(0.004) (0.006) (0.008) (0.010) (0.010) (0.009)
Trade Facilitation Index 1.456**
(0.579)
Physical Infrastructure 0.172
(0.236)
(0.502)
(0.911)
(0.297)
Constant -0.920** -1.084*** -0.526 -0.505 -2.182** -0.997*
(0.333) (0.329) (0.339) (0.521) (0.858) (0.519)
Observations 219 219 219 219 219 219

Number of countries 22 22 22 22 22 22
Number of instruments 21 21 19 20 16 18
AR2 p-value 0.189 0.874 0.756 0.321 0.415 0.627
Hansen p-value 0.584 0.499 0.650 0.275 0.431 0.463
10%, 5% and 1% respectively. Regressions always included time dummies.
Source: authors
23
6- Conclusion and policy recommendations
In this study, we analyze the effect of trade facilitation on export sophistication in Sub-
Saharan Africa. We first review the theoretical and empirical links between trade facilitation
and export sophistication. Theoretically, we found that trade facilitation can contribute to
export sophistication through indirect channels, notably intermediates imports and FDI.
Empirically, we found that few studies link trade facilitation and export sophistication and there
are almost no studies devoted to the specific case of Sub-Saharan Africa. Next, we reviewed
the state of trade facilitation and export sophistication in SSA by analyzing the stylized facts
as well as the correlation between the different indicators of trade facilitation and export
sophistication. The results from the estimation with 2SLS have mostly met the expectations of
the economic and econometric literature. The indicators of physical infrastructure, ICT, and
border and transport efficiency contribute favorably to the export of sophisticated products in
Sub-Saharan Africa. Given the significance and high coefficient of border and transport
efficiency and ICT, they appear to contribute more to export sophistication in SSA among the
four TF indicators. Finally, we tested the robustness of our results with an alternative measure
of export sophistication as well as an alternative estimator, namely the GMM which allows us
to take into account the dynamic nature of the model as well as the endogeneity problem. The
results obtained with the GMM confirm those obtained in the static model.
This study shows that trade facilitation affects positively export sophistication in SSA.
However, to fully benefit from this positive effect, several challenges need to be addressed.
Thus, we formulate some recommendations for policymakers. First, border and transport
efficiency play a key role in the export of sophisticated goods. To this end, reforms aimed at
simplifying border procedures should be implemented and accelerated in SSA countries. These
reforms include the establishment of one-stop borders for export-import operations and the
installation of single windows in SSA countries, which will reduce the time that economic
operators spend at the borders. In order to attract more foreign direct investment especially in
the current context of the African Continental Free Trade Area, countries should develop
measures such as the publication of information on trade activities and customs procedures at
borders. This will enhance the transparency and predictability of customs administration
operations. The acceleration of the implementation of the WTO Trade Facilitation Agreement
will be an essential step to achieving these objectives.
Moreover, with the increased use of ICT in the production process, better quality of
exports can be expected to grow in digitized economies. This work highlights once again the
importance of digitalization in facilitating trade. SSA countries that export less sophisticated
products have the opportunity to improve their export baskets if exporters are able to adapt to
the new digital age. Governments could support this process by providing the right incentives
and implementing effective digitalization strategies.
Finally, physical infrastructure remains an important factor in the export of
sophisticated products in SSA. There is a need to invest more in the development of
infrastructures. Since building physical infrastructures requires big investments, SSA countries
need to explore new financing mechanisms including public-private partnerships. Also, the Aid
for Trade initiative launched a few years ago seems to be a good niche that could help SSA
countries in building infrastructure with the infrastructure category covered by this initiative.
24
References
Abdon, A. and Felipe, J. (2011). The product space: What does it say about the opportunities
for growth and structural transformation of sub-Saharan Africa?
Aghion, P. and Howitt, P. (1990). A model of growth through creative destruction.
Albeaik, S., Kaltenberg, M., Alsaleh, M., and Hidalgo, C. A. (2017). Improving the economic
complexity index. arXiv preprint arXiv :1707.05826.
Andreoni, A. (2011). Productive capabilities indicators for industrial policy design. UHIDO Working
paper series, (17/11).
Atasoy, B. S. (2021). The determinants of export sophistication: Does digitalization matter?
International Journal of Finance & Economics, 26(4) :5135—5159.
Banerjee, R., Maruta, A. A., & Donato, R. (2022). Does higher financial inclusion lead to better health
outcomes? Evidence from developing and transitional economies. Economics of Transition
and Institutional Change.
Bas, M. and Strauss-Kahn, V. (2015). Input-trade liberalization, export prices and quality
upgrading. Journal of 'International Economics, 95(2) :250—262.
Baumol, W. J. and Lee, iL S. (1991). Contestable markets, trade, and development. The World
Bank Research Observer, 6(1) :1—17.
Beverelli, C., Neumueller, S., and Teh, R. (2015). Export diversification effects of the WTO trade
facilitation agreement. World Development, 76 :293—310.
Blalock, G. and Veloso, F. M. (2007). Imports, productivity growth, and supply chain learning.
World Development, 35(7) :1134—1151.
Blundell, R. and Bond, S. (1998). Initial conditions and moment restrictions in dynamic panel
data models. Journal of econometrics, 87(1) :115—143.
Cabral, M. H. and Veiga, P. (2010). Determinants of export diversification and sophistication
in sub- Saharan Africa.
Chakraborty, D. and Mukherjee, S. (2016). How trade facilitation measures influence export
orientation? empirical estimates with logistics performance index data. Journal of
Economics Library, 3(4) :554—569.
Cristelli, M., Gabrielli, A., Tacchella, A., Caldarelli, G., and Pietronero, L. (2013). Measuring
the intangibles: A metrics for the economic complexity of countries and products. PloS
one, 8(8) : e70726.
da Costa Neto, M. N. C. and Romeu, R. (2011). Did export diversification soften the impact of
the global financial crisis ? International Monetary Fund.
de Melo, J. and Wagner, L. (2018). Les coûts à commercer en uemoa impacts potentiels de la
mise en œuvre de l'accord sur la facilitation des échanges (afe).
Dunning, J. H. (2000). The eclectic paradigm as an envelope for economic and business theories
of mne activity. International Business Review, 9 :163—190.
Duranton, G. and Puga, D. (2004). Micro-foundations of urban agglomeration economies. In
Handbook o/ regional and urban economics, volume 4, pages 2063—2117. Elsevier.
Fang, Y., Gu, G., and Li, H. (2015). The impact of financial development on the upgrading of China’s
export technical sophistication. International Economics and Economic Policy, 12(2) :257-
280.
Faruq, H. A. (2011). How institutions affect export quality. Economic s ystems, 35(4) :586—606.
Feenstra, R. C. and Ma, H. (2014). Trade facilitation and the extensive margin of exports. The
Japanese Economic Review, 65(2) :158—177.
Felipe, J. (2012). Inclusive growth, full employment, and structural change: Implications and
policies for developing Asia. Anthem Press.
25
Fiorillo, F. (2001). Rate of growth and sector specialization coevolution in a kaldorian export-
led growth model. Structural Change and Economic Dynamics, 12(1) :91-114.
Fosu, A. K. and Abass, A. F. (2019). Domestic credit and export diversification: Africa from
a global perspective. Journal of African Business, 20(2) :160—179.
Francois, J. and Manchin, M. (2013). Institutions, infrastructure, and trade. World development,
46: l65— 175.
Gnangnon, S. K. (2021). Effect of productive capacities on economic complexity. Journal of
Economic Integration, 36(4) :626—688.
Grancay, M., Grancay, N., and Dudas, T. (2015). What you export matters: does it really?
Contemporary Economics, 9(2) :233—244.
Harding, T. and Javorcik, B. S. (2012). Foreign direct investment and export upgrading. Review
of Economics and Statistics, 94(4) :964—980.
Hartmann, D., Guevara, M. R., Jara-Figueroa, C., Aristarán, M., and Hidalgo, C. A. (2017).
Linking economic complexity, institutions, and income inequality. World development,
93 :75—93.
Hausmann, R. (2016). Economic development and the accumulation of know-how. Welsh Economic
Review, 24 :13—16.
Hausmann, R., Hidalgo, C. A., Bustos, S., Coscia, M., and Simoes, A. (2014). The atlas of
economic complexity: Mapping paths of prosperity. Mit Press.
Hausmann, R., Hwang, J., and Rodrik, D. (2007). What you export matters. Journal of economic
growth, 12(1) :1—25.
Hausmann, R. and Rodrik, D. (2003). Economic development as self-discovery. Journal of
development Economics, 72(2) :603—633.
Hidalgo, C. A. and Hausmann, R. (2009). The building blocks of economic complexity.
Proceedings of the national academy of sciences, 106(26) :10570—10575.
Hidalgo, C. A., Klinger, B., Barabasi, A.-L., and Hausmann, R. (2007). The product space
conditions the development of nations. Science, 317(5837) :482—487.
Hu, Y. (2020). Trade facilitation, R&D innovation, and export sophistication of manufacturing
industries Evidence from Russia and central-eastern European countries.
Hu, Y., Jiang, M., Sun, S., and Dai, Y. (2022). Does trade facilitation promote export technological
sophistication? evidence from the european transition countries. SWGR Open, 12(2)
:21582440221106675.
Hummels, D. and Klenow, P. J. (2005). The variety and quality of a nation's exports. American
economic review, 95(3) :704—723.
Imbs, J. and Wacziarg, R. (2003). Stages of diversification. American economic review, 93(1)
:63-86.
Jarreau, J. and Poncet, S. (2012). Export sophistication and economic growth: Evidence
from China. Journal of' development Economics, 97(2) :281—292.
Jouini, N., Oulmane, N., and Péridy, N. (2016). North African countries’ production and
export structure towards a diversification and export sophistication strategy. Canadian
Journal of' Development Studies Revue canadienne d'études du développement, 37(2)
:217-238.
Kamguia, B. (2021). Do terms of trade improve export sophistication in Africa? new empirical
evidence and transmission channel.
Kaufmann, Daniel Kraay, Aart, and Massimo Mastruzzi, “The Worldwide Governance Indicators:
Methodology and Analytical Issues,” World Bank Policy Research Working Paper No.
5430, Washington, D.C., 2010.
Khan, F. N., Sana, A., and Arif, U. (2020). Information and communication technology (ict) and
26
environmental sustainability: a panel data analysis. Environmental! Science and Pollution
Research, 27(29) :36718-36731.
Kleibergen, F., & Paap, R. (2006). Generalized reduced rank tests using the singular value
decomposition. Journal of Econometrics, 133(1), 97‑126.
Lall, S., Weiss, J., and Zhang, J. (2006). The ‘sophistication" of exports: A new trade measure.
World development, 34(2) :222-237.
Lapatinas, A. (2019). The effect of the internet on economic sophistication: An empirical analysis.
Economics Letters, 174 :35-38.
Lectard, P. and Rougier, E. (2018). Can developing countries gain from defying comparative
advantage? distance to comparative advantage, export diversification and sophistication,
and the dynamics of specialization. World Development, 102 :90-110.
Lee, H.-y. and Kim, C.-S. (2012). The impact of trade facilitation on the extensive and intensive
margins of trade: An application for developing countries. Journal of East Asian Economic
Integration, 16(1) :67— 96.
Lesher, M. and Miroudot, S. (2008). Fdi spillovers and their interrelationships with trade.
Luu Hai, D. (2021). Aid for trade and export sophistication in recipient countries. The journal of
international trade & economic development, 30(4) :530—548.
Minondo, A. (2010). Exports' quality-adjusted productivity and economic growth. The Journal of
International Trade & Economic Development, 19(2) :257-287.
Minondo, A. and Requena-Silvente, F. (2013). Does complexity explain the structure of trade?
Canadian Journal of Economics /Revue canadienne d'économique, 46(3) :928-955.
Moisé, E. and Sorescu, S. (2013). Trade facilitation indicators: The potential impact of trade
facilitation on developing countries’ trade.
Nguyen, D. X. (2016). Trade liberalization and export sophistication in vietnam. The Journal of'
International Trade & Economic Development, 25(8) :1071-1089.
Persson, M. (2013). Trade facilitation and the extensive margin. The Journal of' International
Trade & Economic Development, 22(5) :658—693.
Portugal-Perez, A. and Wilson, J. S. (2012). Export performance and trade facilitation reform:
Hard and soft infrastructure. World development, 40(7) :1295-1307.
Rodriguez-Crespo, E. and Martinez-Zarzoso, I. (2019). The effect of ict on trade: Does product
complexity matter? Telematics and Informatics, 41 :182—196.
Rodrik, D. (2006). What's so special about China’s exports? China & World Economy, 14(5) :1—19.
Roodman, D. (2009). How to do xtabond2: An introduction to difference and system gmm in stata.
The stata journal, 9(1) :86—136.
Saadi, M. (2020). Remittance inflows and export complexity: New evidence from developing and
emerging countries. The Journal of Development Studies, 56(12) :2266-2292.
Sakyi, D., Bonuedi, I., and Opoku, E. E. 0. (2018). Trade facilitation and social welfare in Africa.
Journal of African Trade, 5(1-2) :35-53.
Saviotti, P. P. and Frenken, IL (2008). Export variety and the economic performance of countries.
Journal of Evolutionary Economics, 18(2) :201-218.
Schott, P. IL (2004). Across-product versus within-product specialization in international trade. The
Quarterly Journal of Economics, 119(2) :647-678.
Spatafora, M. N., Anand, R., and Mishra, M. S. (2012). Structural transformation and the
sophistication of production. International Monetary Fund.
Spence, M. and Karinigi, S. (2014). Impact of trade facilitation mechanisms on export
competitiveness in Africa.
Srholec, M. (2007). High-tech exports from developing countries: A symptom of technology
spurts or statistical illusion? Review of world economics, 143(2) :227-255.
27
Stock, J., & Yogo, M. (2005). Identification and Inference for Econometric Models. Cambridge
University Press.
Wagner, J. (2007). Exports and productivity: A survey of the evidence from firm-level data. World
Economy, 30(1) :60-82.
Weldemicael, E. (2012). Determinants of export sophistication. The University of Melbourne.
Wilson, J. S., Mann, C. L., and Otsuki, T. (2003). Trade facilitation and economic development: A
new approach to quantifying the impact. The World Bank Economic Review, 17(3) :367-389.
WTO. (2015), Speeding up trade: benefits and challenges of implementing the WTO Trade
Facilitation Agreement, World Trade Report, 157p.
Yue, J. and Zhou, S. (2018). Democracy's comparative advantage: Evidence from aggregated trade
data, 1962—2010. World Development, 111 :27-40.
Zhu, S. and Fu, X. (2013). Drivers of export upgrading. World Development, 51 :221-233.
28
Appendix
1- Correlation matrix between trade facilitation indicators and export sophistication
Information and Business and Border and
Export Physical
Communication Regulatory Transport
Sophistication Infrastructure
Technology Environment Efficiency
Export Sophistication 1.0000
Physical Infrastructure 0.6445*** 1.0000
Information and
0.3437*** 0.5686*** 1.0000
Communication Technology
Business and Regulatory

0.3545*** 0.4932*** 0.2549*** 1.0000
Environment
Border and Transport

0.4562*** 0.458*** 0.5807*** 0.3628*** 1.0000
Efficiency
*** significance at 1%.
2- Descriptive statistics
Variable Obs Mean Std. dev. Min Max
Export Sophistication (ECI) 308 -0.76413 0.561458 -2.42371 1.21975
Overall TF indicator 308 0.461768 0.143001 0.236041 0.803065
Physical_Infrastructure 308 0.435155 0.212433 0.099813 0.958784
Technology_ICT 308 0.448351 0.210831 0.059395 1
Business_Environment 308 0.332142 0.154289 0.045806 0.877447
Border_Efficiency 308 0.631423 0.160155 0.309531 1
GDP_pc 308 2141.116 2321.565 136.4663 10484.91
School_Enrollment 272 103.5443 18.06197 53.8744 149.3075
FDI 308 3.909849 4.946428 -1.44269 39.4562
Intermediates Import 299 0.483574 0.085662 0.205295 0.664063
Private_Sector_Credit 270 28.2207 32.82442 3.678724 160.1248
Control_Corruption 308 -0.43142 0.584954 -1.4049 1.159934
Tariffs 308 0.0559031 0.03916 0 0.1535791
Natural_Resources 308 8.128203 5.749212 0.001171 32.79626
29
3- Summary of model 1 variable
Variables Definitions Sources Expected signs
It is measured by the Economic Complexity Index (ECI). Atlas of N/A
Higher values of ECI indicate a complex or sophisticated Economic
Export Sophistication (ECI) goods while lower values show exports of unsophisticated Complexity
commodities
It captures the level and quality of road, port, airport and
Physical Infrastructure (PI) rail infrastructure (ranges from 1= extremely WEF +
underdeveloped, to 7 = well developed)
Information and It concerns the use of ICT to improve efficiency and
Communication Technology productivity and to reduce transaction costs (ranges from WEF +
(ICT) 1 to 7 = best)
Business and Regulatory RE measures the level of development of regulations and
Environment (RE) transparency (range from 1=low to 7=high). WEF +
Border and Transport It captures the customs and inland transport efficiency
reflected in the time and number of documents. DB +
Efficiency (BE)
GDP per capita Gross domestic product per capita (current $) WDI +
Human Capital It measured by the enrollment rate WDI +
Foreign Direct Investment Ratio of the net foreign direct investment as a % of GDP WDI +
It represents the import of the Broad Economic Categories UNCOMTRA

Intermediates Imports
(BEC) products as a share of total imports DE
It is measured by the credit to the private sector, which +
Financial development refers to the financial resources provided to the private WDI
sector by financial corporations
It is captured by the control of corruption which measures
how the public power is exercised for private gain,
Institution WGI +
including minor and major forms of corruption. It values
range from -2.5 to +2.5
Tariffs It is a weighted average rate applied to all products WDI -
It is the sum of oil rents, natural gas rents, coal rents (hard -
Natural resources WDI
and soft), mineral rents and forest rents
Source: Authors
30
4- Diagnostics tests
Econometric Problem Tests F-Stat P-Value Null Hypothesis
Accepted/Rejected
Serial Correlation Wooldridge test 15.928 0.0007 Rejected
Modified Wald test for

Heteroscedasticity GroupWise
heteroskedasticity 1023.43 0.0007 Rejected
Pesaran (2015) test for

Cross-sectional
weak cross-sectional
dependence
dependence 0.0000 1.000 Accepted
Source: Authors
5- Trade facilitation indicators and the contribution of each variable in (%) using
principal component analysis
Aggregate indicators Variables/indices F1
Quality of port infrastructure 23,896
PHYSICAL Quality of the airport infrastructure 23,542
INFRASTRUCTURE Quality of the road infrastructure 29,333
(PI) Quality of the railway infrastructure 23,228
Availability of ICT 37,776
INFORMATION AND Level of ICT absorption 35,393
COMMUNICATION
TECHNOLOGY (ICT) Use of ICT 26,831
Transparency of government policies 21,186
Public trust in politicians 27,703
BUSINESS AND
Irregular payments and bribes 24,387
REGULATORY
ENVIRONMENT (RE) Government favoritism to business 26,725
Number of documents to export 19,498
BORDER AND Number of documents to import 20,866
TRANSPORT Number of days to export 29,295
EFFICIENCY (BE) Number of days to import 30,341
Note: Each variable was standardized to values that range from 0 to 1 to facilitate comparison.
F1 represent the weight of each sub-indices.
Source: Authors based on data from WEF and Doing Business.
31
6- Economic Complexity Index method
ECI is calculated from export data connecting countries to the products in which they
have Revealed Comparative Advantages (RCA) (Hidalgo and Hausmann, 2009).
The Revealed Comparative Advantage (RCA) of a country c in a product p is:
𝑋𝑐𝑝 / ∑𝑃 𝑋𝑐𝑝
𝑅𝐶𝐴𝑐𝑝 = Ʃ (1)
𝑐 𝑋𝑐𝑝 / ∑𝑐𝑝 𝑋𝑐𝑝
where Xcp is the total export of country c in product p. RCA is larger than 1 (indicating
that a country has comparative advantage in a product), if a country's export of a
product is larger than what would be expected from the size of the country's export
economy and the product's global market.
RCA are used to define a discrete matrix Mcp which is equal to 1 if country c has RCA
in product p and 0 otherwise.
Mcp= 1 if RCAcp ≥1
Mcp= 0 if RCAcp <1
The matrix Mcp allows to define the diversity of a country and the ubiquity of a
product, respectively, as the number of products that are exported by a country with
comparative advantage, and the number of countries that export a product with
comparative advantage.
Diversity = kc,0 = ΣpMcp
Ubiquity = kp,0 = ΣcMcp
Next, a matrix can be defined that connects countries exporting similar products,
weighted by the inverse of the ubiquity of a product (to discount common products),
and normalized by the diversity of a country:
1 𝑀𝑐𝑝 𝑀c’𝑝
𝑀𝑐c’ = 𝑘 ∑𝑝 𝑘𝑝,0
(2)
𝑐,0
Finally, the Economic Complexity Index (ECI) is defined as:

𝐾𝑐 −(𝑲)
𝐸𝐶𝐼𝑐 = 𝑠𝑡𝑑(𝐾)
(3)
where Kc is the eigenvector of 𝑀𝑐c’ associated with the second largest eigenvalue, the
vector associated with the largest eigenvalue is a vector of ones (Haussmann et a1.
2014; Hartmann et al., 2017).
Source: Adapt from Hartmann et al. 2017.

7- List of countries included in the sample
South Africa (ZAF), Botswana (BWA), Burkina Faso (BFA), Cameroun (CMR), Cote d’Ivoire (CIV),
Eswatini (SWZ), Ethiopia (ETH), Ghana (GHA), Kenya (KEN), Madagascar (MDG), Malawi (MWI),
Mali (MLI), Mauritius (MUS), Mozambique (MOZ), Namibia (NAM), Nigeria (NGA), Senegal (SEN),
Tanzania (TZA), Uganda (UGA), Zambia (ZMB), Zimbabwe (ZWE), Mauritania (MRT),
Source: Authors.
32

Export Sophistication Effects of Trade Facilitation Reform in Sub-Saharan Africa

Uploaded by

Copyright:

Available Formats

Export Sophistication Effects of Trade Facilitation Reform in Sub-Saharan Africa

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Export Sophistication Effects of Trade Facilitation Reform in Sub-Saharan Africa

Uploaded by

Copyright:

Available Formats

Export Sophistication Effects of Trade Facilitation Reform

Moukaila Mouzamilou Takpara1*, Constant Fouopi Djiogap2, Manfred Kouty3,

2- Related literature review

3- Data and stylized facts

3-2 Stylized facts

Source: Authors based on the Atlas of Economics Complexity data.

4-2 Estimation technique

Observations 203 203 203 203 203

Hansen_stat (overidentification test) 0.391 2.500 0.104 0.0462 0.0506

Observations 203 201 201 201 201

Log (pc GDP) 0.094** 0.157*** 0.085*** 0.130*** 0.080**

Observations 203 203 203 203 203

L.ECI_initial 0.444*** 0.547*** 0.359*** 0.488*** 0.373*** 0.424**

Observations 219 219 219 219 219 219

Export Sophistication 1.0000

Physical Infrastructure 0.6445*** 1.0000

Business and Regulatory

Border and Transport

*** significance at 1%.

Human Capital It measured by the enrollment rate WDI +

It represents the import of the Broad Economic Categories UNCOMTRA

Tariffs It is a weighted average rate applied to all products WDI -

Modified Wald test for

Pesaran (2015) test for

Finally, the Economic Complexity Index (ECI) is defined as:

Source: Adapt from Hartmann et al. 2017.

You might also like

Log (pc GDP) 0.094 0.157* 0.085* 0.130* 0.080**

L.ECI_initial 0.444* 0.547* 0.359* 0.488* 0.373* 0.424