Content uploaded by Isaiah Oluranti Olurinola
Author content
All content in this area was uploaded by Isaiah Oluranti Olurinola on Aug 17, 2017
Content may be subject to copyright.
Electronic copy available at: http://ssrn.com/abstract=2460959
1
INNOVATION IN AFRICA:
WHY INSTITUTIONS MATTER
OLUWATOBI Stephen*, EFOBI Uchenna**, OLURINOLA Isaiah*, ALEGE Philip*
*Department of Economics and **School of Business, Covenant University, Nigeria.
Abstract
Given the role that innovation plays as an engine for economic development, we examined the
enabling factor of institutions in Africa. Particularly, attention was given to determining the
equivalent effects of institutional development on innovation. A sample of 40 African countries
over the period 1996-2012 was employed; and our baseline equation was estimated using the
Systems Generalized Method of Moments estimation technique. The empirical result, reveals that
government effectiveness and regulatory quality are two institutional measures that have the most
equivalent impact on innovation. This result is robust in diverse respects.
Keywords: Africa, Human Capital, Innovation, Institutions
JEL Code: J24; O31; O32; O43
Electronic copy available at: http://ssrn.com/abstract=2460959
2
INNOVATION IN AFRICA:
WHY INSTITUTIONS MATTER
1. Introduction
Innovation, in this study is the creation and increase in the stock of knowledge. With respect to
economics, is a catalyst for growth and rapid advancements in diverse sectors of the economy such
as telecommunication, transportation, commerce and education. It include the outcome from the
useful/productive engagement of human capital and relevant resources. In the words of Joseph
Schumpeter, an economy can transform itself from within through innovation. On the other hand,
institutions is a systems of procedures, regulations and customs that shape socioeconomic activity
and behavior. It includes the regulatory framework in which economic operations and relationships
are governed.
Premixed on the fact that institutions are capable of directing economic operations and
relationships, then it implies that innovation – in order to achieve economic growth - will require
relevant institutions (Oyelaran-Oyeyinka and Sampath, 2006; Tebaldi and Elmslie, 2008a; Tebaldi
and Elmslie, 2008b; Tebaldi and Mohan, 2008; and Schiliro, 2010). This postulation is expected
to also apply to developing countries like Africa. If this idea is valid, then African countries can
enjoy accelerated economic growth for certain reasons. First, given the evidence from the Solow
growth model, innovation’s contribution to economic growth constitutes about 85% compared to
15% which factor inputs are responsible for. Second, stipulations made by the advocates of
endogenous growth theory suggest that innovation is not exogenous as neoclassical growth theory
suggests. Rather, it can be determined within the system (Romer, 1990; Romer, 1994). Thirdly,
innovation is needed to augment the state of human capital development in Africa. This implies
that there is the need to go beyond education to developing the innovation capacity of human
capital. Presently, African countries record a low score for innovative output when compared to
countries from other regions of the world. For instance, Statistics from the World Development
Indicators revealed that Africa’s contribution to world innovation in 2009, measured by the amount
of scientific and technical journal articles, is less than 1 percent (0.64%) compared to Europe
(36.84%), East Asia (24.17%), South Asia (2.72%) and Latin America (3.04%).
This study, therefore, answers two important questions. First, to what extent has the level of
institutional development in African countries affected her innovative outputs? Second, at what
point of institutional development is innovative output enhanced in African countries? In
discussing institutions, we pay particular attention to three broad indicators of institutional
development: corruption, government effectiveness and regulatory quality. The emphasis on these
measures of institutions is based on their expected direct impact on innovation. Focusing on the
regulatory quality for instance, an environment where there are lots of documentations and
administrative procedures to launch startup businesses that would have hitherto enhanced
innovative outputs, will inevitably impede innovative outputs. This is because the bureaucratic
protocols caused by the regulatory environment will either slow the process of innovation or create
3
an incentive for corrupt practices (Mahagaonkar, 2008). In the same vein, in countries where less
importance is given to the protection of intellectual property rights or there are long procedures
involved before a research output can be patented, there is likely to be less innovation. Basically,
these environments, which depicts poor regulatory quality, government ineffectiveness and
corruption, enhances opportunistic behaviors by economic agents, which hitherto will affect
innovative output.
The answers to these questions are very relevant because; theoretically, Solow growth model has
proved that the impact of innovation on economic growth is higher compared to factor inputs such
as physical capital. Therefore, it will be necessary to commit economic elements, in the form of
policy actions and improved institutions on innovation since it yields higher returns. Also, the
experience of emerging economies (such as South Korea and Singapore)
1
suggests that developing
economies can improve their growth rates by embracing innovation-driven growth; although this
may not be plausible if the quality of institution is low. Furthermore, considering the level of
institutional development in African countries (especially corruption, the extent of government
commitment and the quality of regulation), it will be of great importance to consider the extent to
which these frameworks can affect innovation and at what threshold of their development is
innovation enhanced. This will be relevant for policy action because policy makers can relate with
this analysis to adjust their framework for development. Fosu (2013) agrees with this fact by noting
that African countries (especially the resource-rich countries) need not experience slower growth
that are caused by political contestations, poor executive constraints, low political rights and civil
liberties. Therefore, pointing out what kind of institutions and at what threshold of development is
innovation enhanced, will be very relevant for the furtherance of development policies.
Given the perceived importance of innovation in facilitating growth, it is astounding that there is
no clear empirical study relating these concepts in direct linkage with the research questions posed
by this study. An exploration of literature showed that there is limited research on this subject with
respect to Africa and some of the empirical attempts were focused on the relationship between
social capital and innovation in Sub-Saharan Africa (Rijn, Bulte, and Adekunle, 2012); the effect
of social networks on innovation in Tanzania (Murphy, 2002); the role of corruption in impeding
innovation (Mahagaonkar, 2008); the relevance of innovation in the developing world (Fagarberg,
Srholec, and Verspagen, 2010); an African perspective of the innovation that enables access to
energy (Agbemabiese, Nkomo, and Sokona, 2012); and modelling the future of knowledge
economy in Sub-Saharan Africa (SSA) and Middle East and North Africa (MENA) [Asongu,
2013]. Mahagaonkar (2008) and Asongu (2013) seem to be the closest empirical study that directly
relates to the focus of this study. However, there are still gaps this study has identified to fill.
Mahagaonkar (2008), who explored a portion of the institutional setting in Africa, examined the
relationship between corruption and innovation at the firm level. Hence, the conclusions of the
1
Singapore and South Korea enhanced their international competitiveness by investment in high value-added research
and development and high-tech industries.
4
study may not be applicable at the macro level for African countries’ policy makers. Although he
employed the probit estimation technique and engaged instrumental variables to cater for
endogeneity problems that could arise as a result of the inclusion of corruption. However, his result
may not be reliable given the challenges involved in the selection of instrumental variables. For
instance, the author included firm’s perception of the efficiency of the government and faith in the
judiciary as instruments for corruption, which is problematic since these instruments correlate
with innovation. In support of this, Ardelini et al (2010) found strong correlation between
innovation and the legal system.
Asongu (2013), on the other hand tried to model the future of SSA’s and MENA’s knowledge
economy. Even though his study employed the system generalized method of moments (SGMM)
to address the challenges involved in the selection of instrumental variables and cater for reverse
causality and endogeneity problems, his conclusions will definitely be different from that of this
study. The reason is that he examined institutions and innovation as explanatory variables in his
model and did not investigate the relationship that can likely arise between the two (innovation
and institutions). Besides, his study combines samples that include the Middle East countries;
hence, his conclusions may not address the African peculiarity, which Asiedu (2002, 2006)
regarded as plausible only when African samples are used in empirical studies. Further, his
research question – the extent of convergence and how long it would take for economies with low
knowledge economy to catch up with those having higher knowledge economies in the two regions
– strikingly differs from that of this study. Finally, the magnitude to which institutions matter for
innovation in Africa was not clearly ascertained in the two studies, making it difficult for policy
implications.
The most recent non-African studies, closest to the subject of this research include Blind (2012),
who examined the influence of regulations on innovations in 21 Organization of Economic
Cooperation and Development (OECD) countries, spanning from 1998-2004; and D’Este,
Lammarino, Savona, and Tunzelmann (2012) who examined the barriers to innovation by using a
sample from the survey of European Union firms for the period 2002-2004. We identify some
limitations from these studies. Apart from the fact that these studies focused on non-African
samples, the data used by Blind (2012) and D’Este, Lammarino, Savona, and Tunzelmann (2012)
were not recent and portend a concern to contemporary inquiries. Secondly, these studies did not
take care of possible endogenous relationships that can occur in their models. This is because they
were silent on the endogeneity of institutional variables, such as regulations, which was included
in their empirical models (Peters, 1995; Christin and Hug, 2006). The implication of the
endogeneity problem is that it affects the efficiency of the econometric estimations and will
influence the predictions of the relationship put forward in their studies. Most importantly, the
scope of these studies raises concerns about the applicability of their findings in African countries
who have diverse socio-economic and institutional structures.
The remainder of the paper is structured as follows: Section two gives an overview of institutions
and innovation in Africa. Section three discusses a review of literature, while the fourth section
5
describes the research method. The fifth section presents the empirical results and the sixth section
concludes with policy recommendations.
2. Overview of Innovation and Institutional Quality in Africa
In this section, we discussed the trend of the extent of innovation in African countries vis-à-vis
other countries from other regions of the world. The aim of this is to situate the African innovative
experience in the context of the global trend. We began by observing the overview of innovation
as presented in Table 2.1, noting that it can be measured using the number of scientific and
technical journal articles.
Table 2.1 Innovation (Scientific and Technical Journal Articles) around the World
1996
2001
2006
2007
2008
2009
Sub Saharan Africa-SSA
3908.6
3860
4615.6
4952.4
5074.4
5080.1
East Asia and the Pacific-EAP
89930.6
117690
161522.7
169109.4
182046.1
190578.9
Europe and Central Asia-ECA
240081.3
255860.3
282648.5
287422.2
291637
290424.4
Middle East and North Africa-MENA
9500.4
11452.6
15206.1
16628.2
17920.1
19167
South Asia-SA
10266.1
11380.8
17784.2
19385.6
20372.6
21432.3
Latin America and the Caribbean-LAC
10503.8
16074
21729.6
23337
24743.1
24032.6
Source: Computation from World Bank (2013)
Note: The years of reference were not selected based on any specific criteria.
From Table 2.1, Sub-Saharan African countries have consistently performed below other regions
of the world, in terms of the number of scientific and technical journal article publication around
the world. In all the periods, the volume of Africa’s scientific and technical journal articles was
many folds less than East Asia and the Pacific and Europe and Central Asia countries. It was also
more than two times less than the number for other countries in MENA, SA and LAC countries.
It is therefore obvious that Sub-Saharan Africa yields the least innovative contribution around the
world. Could this trend be traceable to the extent of the educational attainment by individuals in
these countries? To answer this question, we try to examine the factor input that can enhance
innovative acts. It includes the extent of education enrolment. We focus on gross tertiary
enrolment, because at this level of education, there are research and development activities, which
can enhance innovative output. Table 2.2 reports that African countries are also lagging behind
other regions of the world with regards to the gross enrolment rate in tertiary institutions. Other
countries from other regions of the world experienced a rising rate of tertiary enrolment rate, with
values that were well above 10 percent. The contrary was observed for African countries as they
had values that were below 10 percent for the entire period. This trend may not be disassociated
with the poor state of innovation in this region.
Table 2.2 School Enrollment, Tertiary (% gross)
1996
2001
2006
2007
2008
2009
2010
2011
Sub Saharan Africa-SSA
3.71
4.63
5.87
6.10
6.43
6.71
7.21
7.56
East Asia and the Pacific-EAP
11.25
17.24
24.37
25.24
26.08
27.72
29.11
30.11
Europe and Central Asia-ECA
38.52
47.23
55.95
56.82
57.50
58.49
60.04
60.32
Middle East and North Africa-MENA
17.24
21.09
24.43
25.95
27.91
28.66
30.48
31.34
South Asia-SA
5.59
8.33
10.13
11.50
13.02
14.17
15.67
15.88
Latin America and the Caribbean-LAC
18.09
24.43
32.42
35.52
38.52
39.60
41.17
42.32
Source: Computation from World Bank (2013)
6
Note: The years of reference were not selected based on any specific criteria.
Taking a step further, the extent of the development of the institutional framework was examined
for African countries in relation to countries of other regions of the world. African countries scored
low in the corruption control, government effectiveness and regulatory quality measures. The
scores were far behind those of other regions of the world. The issue of poor institutional quality
of African countries has remained a concern for her development process. This includes the quality
of policies and legal frameworks that are expected to encourage innovation and private sector
development.
Table 2.3 Indicators of Institutional Quality
Control of corruption
Government Effectiveness
Regulatory Quality
1996
2000
2005
2008
1996
2000
2005
2008
1996
2000
2005
2008
EAP
-0.43
-0.6
-0.53
-0.57
-0.3
-0.48
-0.46
-0.53
-0.35
-0.61
-0.56
-0.69
ECA
-0.7
-0.62
-0.52
-0.48
-0.58
-0.51
-0.37
-0.31
-0.59
-0.49
-0.32
-0.1
LAC
-0.35
-0.18
-0.16
-0.12
-0.34
-0.15
-0.14
-0.1
0.22
0.07
-0.07
-0.12
MENA
-0.46
-0.57
-0.55
-0.62
-0.45
-0.63
-0.63
-0.61
-0.64
-0.78
-0.73
-0.63
SSA
-0.63
-0.58
-0.68
-0.62
-0.66
-0.72
-0.78
-0.78
-0.65
-0.64
-0.75
-0.7
World
-0.03
-0.02
-0.02
-0.02
-0.04
-0.01
-0.01
-0.01
-0.05
-0.03
-0.02
-0.01
Source: Computation from World Trade Indicators (2010)
Notes: The values ranges from -2.5 (worst) to +2.5 (best) i.e. the higher the better.
The linkage between the low rate of innovation in African countries and the extent of her
institutional development can be clearly seen when considering cases of some African countries
that have improved innovative output and institutional frameworks. South Africa presents a good
case in point, with an average innovative output of 2425.38, which is almost half of the entire
output of SSA countries (see Table 2.1 and 5.1). Strikingly, this country also records a very high
institutional development, with an average values of 0.34, 0.57 and 0.50 for control of corruption,
government effectiveness and regulatory quality respectively.
The uniqueness of South African government is their spontaneity in addressing shortfalls in
institutional frameworks for achieving specific objectives. For instance, after the re-positioning of
South Africa in the transparency international from 54 in 2010 to the position 64/189 in 2011 and
further to 69/176 in 2012, the South African government spontaneously adopted the OECD’s
guidelines on reducing corruption. This guideline was expected to have drastic effect on corruption
in the private sector as well as tenders and procurement in the public sector. Out of the 40 countries
that have ratified or acceded to this guideline, South Africa is the only African country; this shows
their willingness to address/structuring an institutional framework that promotes development.
Linking the relevance of South Africa’s ratification to innovation, some of the guidelines of the
OECD framework requires that enterprises do not, directly or indirectly offer/promise/give or
demand for a bribe or other forms of undue advantages in order to obtain or retain business or
other improper advantage (OECD, 2011). The implication of this on innovation is enormous. This
is because this kind of institutional failure (corruption), inhibits government’s responsiveness to
both indigenous and foreign entrepreneurs and creates bottlenecks to the implementation of their
7
goals and agendas (Mahagaonkar, 2008). The occurrences of these inhibits innovation and better
still increases the cost of innovating. For instance, in cases where the government officials collect
bribe for providing permits or securing licenses, two clear outcomes can be certain: either people
who are not willing to pay the bribe are discouraged from innovating or those willing to pay the
bribe pay higher to get the services to foster their innovation. The example given by Shleifer and
Vishny (1993), describing Mozambique’s bottle making factory, who had to resort to ordering a
unique technology ten times the cost of the technology actually needed as a result of secrecy
imposed on transactions in order to be able to inflate invoices, is instructive.
Despite this exposition, there is the need to provide a comprehensive account of empirical studies
that provides a clear understanding of the concepts – institutions and innovation. As earlier noted,
this will provide a distinct policy action for African countries and will be relevant for advancing
the academic literature on innovation.
3. Literature Review
There is a nascent literature on the linkage between institutions and innovation, it includes Blind
(2012) who embarked on the study to find out what types of institutions affect innovation. The
author identified six types of institutions, including competition legislation, price controls, product
legislation, environmental laws, intellectual property rights and legal and regulatory frameworks;
and adopted the endogenous growth approach as a conceptual analysis to examine the impact of
the different types of institutional settings on innovation. The analysis was based on a panel data
of 21 OECD countries, spanning from 1998 to 2004. The empirical result of the study showed that
institutional frameworks significantly impacts on innovation in OECD countries. Specifically,
legal and regulatory framework significantly impacts on the dynamics of innovation.
Barbosa and Faria (2011) explored a similar perspective by focusing on the relative significance
of variations of institutions in European countries and how these variations explain the differences
in their innovation drive. The result of their study identified that innovation is nurtured in more
advanced credit markets, while impeding institutions like stringent labor and product market
regulations inversely affect the intensity of innovation. This supports D’Este, Lammarino, Savona,
and Tunzelmann (2012), who suggested that regulation barriers limit the drive for innovation.
Further, their result supports the proposition that innovation is encouraged where intellectual
property rights are strengthened and protected. However, it is good to note, as Todtling and
Trippl’s (2005) study pointed out, that there is no one-size-fits-all innovation policy framework
since innovation activities may not be the same in all regions. Each region’s peculiarity demands
specific institutional framework that will address innovation issues related to such region.
Guan and Chen (2012) tried to model the relative efficiency of national innovation systems of 22
OECD countries. Their effort is novel since previous studies focused mainly on theorizing the
nature of national innovation system compared to empirical analysis. Their study, posits a
relational network data envelopment analysis (DEA) model used for measuring the efficiency of
innovation in a national innovation system. Their result showed that the innovation efficiency of a
8
national innovation system is dependent on commercial efficiency performance. This means that
gains and rewards from trading innovation outputs fosters innovation efficiency of national
innovation systems. This further suggests that innovation policy designs in OECD countries should
be aimed at enhancing commercial efficiency. Also, after employing a second-step partial least
squares regression (PLSR) to investigate the impact of institutional arrangements on the efficiency
of innovation systems, their result stipulates that institutional arrangements significantly affects
the efficiency performance of national innovation systems.
Tebaldi and Elmslie (2008a) clearly conducted an empirical investigation on the relationship
between institutions and innovation after which they discovered that property rights protection,
corruption control, effective judiciary system and market-oriented policies enhances the rate of
innovation. Their own study was based on a theoretical model they developed to show institutions’
impact on innovation as a driver for growth of GDP (Tebaldi & Elmslie, 2008b). Evidence was
also provided by their study stating that human capital cannot be ruled out. Hence, in the long run,
human capital accumulation is a vital factor that shapes institutions to affect innovation. While
Kumar, Mudambi, and Gray (2013) brought the element of internationalization into the
relationship between institutions and innovation. They postulated that the interrelationship among
internationalization of emerging market firms, rapidly increasing innovation, and the evolution of
institutions that affects these emerging markets are responsible for the expansion of emerging
market firms.
Pattit, Raj, and Wilemon (2012) tried to explore the long trend of the role of institutions in the
innovative process. They examined how institutions have affected the trend of innovation in the
United States since the mid-19th century. Their study reveals that formal and informal institutions
are crucial, especially at the initial stages of the market for innovation. They also found out that
formal and informal institutions triggered the pervasion of internal R&D lab in the course of the
20th century. These are reflections of the the relevance of institutions in facilitating or impeding
innovation. Whenever there is a favorable institutional environment, innovation thrives; and
whenever, there is an unfavorable institutional arrangement, innovation wanes. The exploration of
these long period shows that an economy can sustain growth and development by ensuring
consistent and favorable institutional arrangement that enables the innovation process. Malik
(2013) stepped up by exploring national institutional differences and how these affect knowledge
transfers and found that some factors of institutions facilitate international knowledge transfers
while some others impede knowledge transfers.
Asongu (2013) modelled the future of knowledge economy (including innovation) in Africa and
the Middle East countries. The author concluded that within a span of seven years, African
countries with low levels of knowledge economy have the tendency to catch up with those with
higher levels based on the justification that innovation matters. The author pointed out that the
pathway to long term economic prosperity can only be attained when countries encourage
innovation. However, Mahagaonkar (2008) clearly state that for African countries to drive at
improving their knowledge economy, there is the need to enhance her institutional framework to
9
achieve this broad goal. The author’s main focus is on corruption; implying that a reduction in the
extent of corruption will lead to improvement in Africa’s innovative outputs.
From the literature reviewed, it can be observed that there is a clear conclusion on the role of
institutions in the drive towards improving the innovative output countries. Currently the empirical
literature on institutions and innovation in Africa is still budding and these studies have reached
general conclusions that it matters. However, to what magnitude does it matter and what kind of
institutions matter? The empirical literatures were not able to give answers to these question and
this further raises a concern that speaks on the need for this study.
4. Research Method
This study aims at understanding the linkage between institutions and innovation in Africa. To
achieve this, we developed an empirical model that builds on the work of Tebaldi and Elmslie
(2008a), which is an empirical model that tests the relevance of institutions on innovation. Their
empirical model is presented as:
(1)
Where, is a measure of technical innovation, while
is the proportion of human capital
that is involved in research and development (R & D), which is expected to culminate in
innovation. refers to the level of institutions. The assumption of their model is based on constant
return to scale, which implies that the proportion of innovation output is relatively affected by the
proportion of institutional development.
For our study, the model was augmented with the inclusion of GDP per capita (Y) and Foreign
Direct Investment (FDI). The inclusion of GDP per capita in the model is expected to control for
the income effect in African countries, which is able to explain the capacity of economic agents to
engage in innovative activities such as R & D. FDI was also included in order to control for
knowledge spillovers, which has been attributed to FDI inflow into African countries. For instance,
Ahmed, Cheng and Messinis (2008) noted that the inflow of foreign investment into Sub-Saharan
African countries is able to enhance spillovers of technology and managerial skills, which are
beneficial for access into the global market. This was supported by other authors such as Dirk
(2006).
Hence, the augmented version of Tebaldi and Elmslie’s (2008a) model is presented in an explicit
form as:
Innovationit = β0 + β1Human_Capitalit + β2Institutionsit + β3 GDP_Percapitait +
β4FDI_Spilloverit + vit (2)
Innovation is measured by the amount of scientific and technical journal articles. The other
measures of innovation include royalty payments and receipt, patents granted and trademark office
per million people (The World Bank, 2008). Apart from data unavailability, the use of scientific
and technical journal articles is considered a better measure of innovation in Africa for the
10
following reasons: 1) scientific and technical journal articles will capture more output from
innovation compared to other measures because innovative individuals from a wider discipline can
readily express their ideas through scientific journal publications. For instance, profitable
innovative ideas that emanate from non-engineering disciplines can easily be stored for
retrieval/referral, instead of being sidelined because it is not patent-worthy. Such innovative ideas
may not require patenting; hence, scientific and technical journal articles will be a veritable
platform for expression of such. 2) The process of securing a patent and trademark, such as
documentations and requirements, is cumbersome especially in most African countries. In Nigeria,
for instance, the process still involves bureaucratic exigencies, which cause delay in securing the
protection of innovative ideas. Some innovative outputs and ideas may therefore end up insecure
and stolen. Others may end up becoming obsolete and unnecessary before they are registered. 3)
The motivation for patenting usually is profits. Hence, innovative outputs are patented to license
out and possibly earn royalties from them. This profit drive leaves out innovative ideas that may
not reflect any potential for profits initially.
Human capital is measured by tertiary enrolment rate as a percentage of the gross enrolment. This
measure considers the percentage of the total enrolled individual in tertiary institutions. We would
have considered other measures of human capital such as personnel involved in R & D, as was
used by Tebaldi and Elmslie (2008a) or adult literacy rate as used by Zanakis and Becerra-
Fernandez (2005), who studied the competiveness of nations by focusing on knowledge discovery.
However, these data were very scanty and missing for most of the sampled years. Despite the fact
that other measures of human capital have been used by other studies, we argue that the use of
tertiary enrolment rate as a percentage of gross enrolment is suitable for the study of innovation
for African countries because most innovative output, reflected in the scientific and technical
journal, are prevalent in tertiary institutions.
Institutions include those measures that reflect the regulations, policies and formal structure that
enhance innovation. We are interested in three measures of institutions. They include control of
corruption, government effectiveness and regulatory quality, which are scored from -2.5 (poor
institutional quality) to +2.5 (better institutional quality)
2
. The relevance of these measures of
institutions are based partly on Blind’s (2012) emphasis that the role of regulatory frameworks and
government effectiveness in implementing regulations on innovation in OECD countries, will
determine the extent of innovation in the countries. Likewise, corruption as a measure of
institutions is relevant in this study, following anecdote evidence on the situation of African
countries. For instance, corruption will impede the translation and transmission of government
expenditure to actual societal outcomes.
2
Control of corruption reveals the extent to which public power is exercised for private gain, including both petty and
grand forms of corruption, as well as ‘capture’ of the state by elites and private interests; Government effectiveness
shows the quality of public services, the quality of the civil service and the degree of its independence from political
pressures, the quality of policy formulation and implementation, and the credibility of the government's commitment
to such policies; Regulatory Quality reflects perceptions of the ability of the government to formulate and implement
sound policies and regulations that permit and promote private sector development (World Bank, 2012).
11
We do not neglect the argument of Rodrik (2004) that most institutional measures, including the
ones employed by this study, are based on surveys of both domestic and foreign investors, thus
capturing perceptions and opinions rather than the actual institutional development. Despite this
obvious challenge in measuring institutions, we observe that for a panel data analysis, the only
measures that best capture institutions are derived from these kinds of survey. Instances include
measures from World Governance indicators, Freedom House variables (political right and civil
liberty) and the International Country Risk Guide (ICRG). Also, we are simply interested in
finding out an indication of the kind of institutions that will affect innovation and this makes these
measures not so problematic. More so, these measures have gained popularity in scientific
literature (e.g. Asiedu, 2006; Catrinescu et al, 2009; Asiedu and Lien, 2011; Osabuohien and Efobi,
2013; Efobi, 2014).
GDP per capita was measured as the current US dollars of the gross domestic product per capita
of the sampled countries in the various years. As earlier noted, this variable is able to capture the
income effect, which follow logical reasoning that the generation and implementation of
innovation is capital-intensive and will require capital for its execution. Likewise, the FDI inflow
variable efficiently captures the effect of knowledge spillovers, which usually occurs with FDI
inflows into African countries and other developing countries (Dirk, 2006). FDI inflow is
measured by the net FDI to GDP ratio.
To be sure that the estimated results are not spurious, we will employ alternative econometric
techniques in estimating equation (2). Firstly, as a check, the Feasible Generalized Least Square
technique will be applied due to its allowance for the presence of sampled countries
heteroscedasticity and within panels’ autocorrelation, and provides panel-corrected standard
errors. The essence of applying this technique is to test the consistency of the behaviors of the
variables that were included in the model. This implies that the output from the estimation becomes
reliable when the signs and coefficients remain consistent and do not behave differently when
other estimation techniques are applied.
The endogeneity concerns cannot be neglected in econometric models. For instance, the right hand
side variables in our model include measures of institution, which are popularly known to be
externally determined and always correlates with the error term (Osabuohien and Efobi, 2013).
This raises possible endogeneity concern in the econometric model and will require other
estimation techniques other than the FGLS. Conventionally, the introduction of instrumental
variables would have helped in resolving this issue. However, candidate instruments must be such
that it is highly correlated with the instrumented variable, but not with the error term. This poses
an overburdening pressure in identifying such instruments.
In view of the above mentioned challenge, this study resolved to use the System GMM estimation
technique, which has been favored by some recent studies (e.g. Rao, Tamazian and Kumar, 2010;
Asiedu and Lien, 2011; Bandyopadhyay et. al., 2014). The System GMM technique addresses
issues of endogeneity by using internal instruments, which has been noted to be more efficient
(Blundell and Bond, 1998, 2000). This technique includes reasonable stationarity restrictions on
12
its initial condition process; includes additional moment conditions unlike other dynamic
estimation tools like the Difference GMM - DGMM; and it is robust to heteroscedasticity and
distributional assumptions (Cheong and Wu, 2013; Bandyopadhyay et. al., 2014). The only
challenge with this technique is the re-assurance that the internal instruments included in the
estimation process are not over-identified. To validate this, the test for autocorrelation [AR (1) and
AR (2)] and Sargan test for instrument over-identification suffices. As a rule of thumb, it is
expected that the probability value of the AR (1) test should be ≤ 0.05 and that of the AR (2) be ≥
0.05. Likewise, the probability value for the Sargan test is expected to be ≥ 0.05. Once these rules
are met, then it is evident that the internal instruments applied are not over-identified.
The SGMM equation type for equation (2) is as follows:
Innovationit = αInnovationit-1 + β1Human_Capitalit + β2Institutionsit + β3 GDP_Percapitait +
β4FDI_Spilloverit + ηi + vit (3)
Where the other variables are as earlier defined and the lag of the explained variable has ‘α’
coefficient and ‘η’ is the unobserved country-specific effects and the error term is ‘v’.
We used data for 40 countries around the world (see Table A1 in the appendix), for the period
1996-2012. The choice of the sample was based on data availability. Our original sample contained
56 African countries but we excluded those countries that do not have data for at least a period of
five years. We did this to enhance the smoothing of the data and to allow for fluctuations in the
data set.
5. Result and Discussion
The data for this study was sourced from the World Bank’s World Development Indicators and
World Governance Indicators, 2013 versions. Our sample consists of 40 African countries for the
period 1996-2012. The sampled countries and the period were selected based on data availability.
We included only those African countries that had available data that is consistent for at least five
(5) years. Apart from the reason earlier given, this approach was necessary to avoid any form of
breaks in the trend of our data which can affect the behaviors of the variables.
Table 5.1 reports the mean of each sampled African country. The mean results show that South
Africa performed best in terms of innovation while Comoros is the least innovative of all the
countries studied. From the ranking, Nigeria emerged the most innovative country after South
Africa, Egypt and Tunisia. Linking the level of innovation and institutional development, we will
use South Africa as a focus, which ranked very high in terms of control of corruption (0.34),
government effectiveness (0.57) and regulatory quality (0.50). Taking this further, Botswana has
a very strong institutional rating, with control of corruption, political stability and regulatory
quality ranking the best among the sampled countries, while government effectiveness measure
was third best. This country also performed fairly well with regards to innovation compared to
other African countries, as the country accounts for about 51 scientific and technical journal
articles and ranking 14th in this category. Tunisia performed well in innovation and also had a high
13
institutional rating. Contrary to this is Nigeria, which had a high innovation ranking and poor
institutional rating.
14
Table 5.1 Mean of Innovation and Institutions of all Sampled Countries
Innovation
Corruption
Control
Government
Effectiveness
Regulatory
Quality
Innovation
Corruption
Control
Government
Effectiveness
Regulatory
Quality
All Sample
Algeria
309.34
-0.62
-0.63
-0.79
Lesotho
2.94
-0.07
-0.27
-0.56
Angola
3.09
-1.30
-1.18
-1.26
Libya
22.89
-0.99
-1.08
-1.46
Benin
28.69
-0.67
-0.46
-0.38
Madagascar
29.96
-0.16
-0.66
-0.45
Botswana
51.44
0.89
0.54
0.61
Malawi
45.08
-0.51
-0.56
-0.47
Burk. Faso
33.75
-0.22
-0.65
-0.23
Mali
17.81
-0.54
-0.80
-0.40
Burundi
3.66
-1.10
-1.30
-1.25
Mauritania
4.29
-0.35
-0.54
-0.44
Cameroon
109.65
-1.04
-0.82
-0.81
Mauritius
15.21
0.51
0.66
0.60
Cape Verde
0.56
0.50
0.07
-0.17
Morocco
401.24
-0.17
-0.11
-0.16
Cent. Afr Rep.
5.15
-1.06
-1.45
-1.13
Mozambique
16.12
-0.48
-0.48
-0.44
Chad
3.39
-1.20
-1.20
-1.03
Namibia
16.86
0.31
0.16
0.18
Comoros
0.51
-0.84
-1.58
-1.40
Niger
18.43
-0.84
-0.82
-0.60
Congo, D. R.
8.36
-1.46
-1.71
-1.61
Nigeria
401.31
-1.11
-1.02
-0.89
Congo, Rep.
13.65
-1.05
-1.25
-1.21
Rwanda
5.83
-0.21
-0.48
-0.66
Djibouti
0.52
-0.48
-0.36
-0.34
Senegal
65.62
-0.29
-0.27
-0.24
Egypt
1614.91
-0.53
-0.91
-0.70
South Africa
2425.38
0.34
0.57
0.50
Eritrea
5.32
-0.12
-1.18
-1.73
Swaziland
4.43
-0.28
-0.75
-0.51
Ethiopia
110.02
-0.70
-0.69
-1.05
Tanzania
104.17
-0.67
-0.49
-0.42
Ghana
86.05
-0.11
-0.06
-0.12
Togo
8.94
-0.88
-1.36
-0.78
Guinea
3.35
-0.93
-1.06
-0.98
Tunisia
472.34
-0.02
0.40
-0.05
Kenya
251.46
-0.96
-0.56
-0.23
Uganda
94.21
-0.83
-0.49
-0.07
Source: Computation from World Bank (2013)
15
Having observed the performance of the countries, we are not able to exactly identify the linkage
between institutions and innovation as some countries like Nigeria rank high for innovation and
low for institutional quality. We therefore report a correlation analysis to observe the bivariate
linkage between these variables. Figures 5.1a, b and c report the association between the measure
of innovation and the three indicators of institutional quality (control of corruption, government
effectiveness and regulatory quality). From Figure 5.1a, a positive and significant association was
observed between the measure of innovation and control of corruption. The bivariate association
connotes that a 1 unit improvement in the quality of institutions (i.e. reduction in corruption), will
lead to a 166 unit improvement in the extent of innovation (proxy by scientific journal publication).
Although, we cannot draw an efficient inference as to the cause and effect dynamics between the
variables, since this figure only explains a bivariate relationship; however, possible explanations
for this relationship can be traceable to the fact that corruption in the society tends to advance
opportunistic behavior among economic agents and then increase transaction cost. In as much as
corruption persist, innovation becomes stiffened since innovation is not rewarded because of
opportunistic behaviors of capital owners (i.e. those that ought to reward innovation).
Figure 5.1a Relationship between Innovation and Control of Corruption
A positive association was also observed for innovation measure and government effectiveness.
The straight line equation as presented in the top of Figure 5.1b reveals that innovation will
experience a positive change of 222 units if the extent of government effectiveness improves by 1
unit. In the case of government effectiveness, it is expected that in situations where the government
-500
0
500
1000
1500
2000
2500
3000
-2 -1.5 -1 -0.5 0 0.5 1
Scientific and Technical Journal Articles
corruption
Y = 261. + 166.X
16
have quality policy formulations and implementations, and are committed to such policies that
creates conducive atmosphere for innovation; consequently, innovation will be boosted. Vivid
example is the case of South Africa, where the government is clearly involved in financially
supporting innovative products and/or processes through the development stage of the idea to the
pre-production prototype stage. This is apart from other initiatives such as the development of
young South African entrepreneurs through access to funding and assistance through the Youth
Technology Innovation Fund, which was launched by the Technology Innovation Agency (TIA)
in 2008. In retrospect, South Africa has maintained a steady growth in research and development
over the past decade, with gross expenditure on research and development (GERD) growing
fivefold from R4 billion in 1997/98 to R21 billion in 2008/09 (Department of Science and
Technology South Africa, 2011). These government involvement and initiatives has immeasurably
improved innovative output in South Africa.
Figure 5.1b Relationship between Innovation and Government Effectiveness
Figure 5.1c further corroborates the outlook from the other Figures. The scatter plots reveal that
as institutional quality (regulatory quality) increases, the extent of innovation will also be
improving. The straight line equation, as displayed at the top of the Figure, shows that a unit
improvement in regulatory quality will result in a positive change in the extent of innovation by
214 units. As earlier mentioned, there are lots of innovation in countries with regulatory
environments that reduces lots of documentations and administrative procedures to launch
startups. Besides, in the case where there are long procedures involved before a research output
can be patented, there is likely to be less innovation. These explains why a positive relationship is
exhibited between innovation and regulatory quality.
-500
0
500
1000
1500
2000
2500
3000
-2 -1.5 -1 -0.5 0 0.5
scientific and Technical Journal Articles
Government Effectiveness
Y = 313. + 222.X
17
Figure 5.1c Relationship between Innovation and Regulatory Quality
The summary from the correlation analysis reveals that there is a likely positive influence of
institutional measures on innovation. This is not far-fetched considering that institutions will
provide an enabling environment that fosters innovation. However, we are not able to base our
conclusion on the bivariate relationships; hence we go further to present a multivariate analysis
that considers the influence of other variables (covariates) and minimizes any possible endogeneity
problems.
Table A2 in the Appendix reveals the correlation results of the main variables that are included in
our econometric estimations. We examined this in order to determine the extent of
multicollinearity that is likely to occur in the combination of our explanatory variables. From the
Table, we do not observe any possible multicollinearity problem except for institutional measures.
This therefore implies that all the other variables can be combined in our empirical model except
for institutional quality variables, which can only be included in the model one after the other.
To begin the empirical analysis, we present a benchmark regression analysis using the FGLS
estimation technique. As earlier stated, this technique was presented in order to check the
consistency of the signs and significant values of our explanatory variables, when alternative
technique is applied. Table 5.2 presents the FGLS results. As a benchmark check, the variables of
interest (measures of institution), exhibited a positive relationship with innovation. This suggest
the expected outcome when an estimation technique that inculcates tools for handling endogeneity
is applied in the estimation process. In this study, the SGMM technique was applied.
-500
0
500
1000
1500
2000
2500
3000
-2 -1.5 -1 -0.5 0 0.5
Scientific and Technical Journal Publication
Regulatory Quality
Y = 301. + 214.X
18
Table 5.2 FGLS Estimation (Dependent Variable: Innovation)
1
2
3
4
5
FDI Spillover (Net FDI Inflow % GDP)
0.032
(0.566)
0.063
(0.351)
0.052
(0.388)
0.031
(0.600)
0.051
(0.402)
Income (GDP Per Capita)
-0.301**
(0.021)
-0.251
(0.146)
-0.648*
(0.001)
-0.789*
(0.000)
-0.640*
(0.000)
Human Capital (tertiary enrolment rate as a
percentage of the gross enrolment)
0.141*
(0.000)
0.126*
(0.000)
0.137*
(0.000)
0.158*
(0.000)
0.143*
(0.000)
Institutional Quality (Control of Corruption)
0.079
(0.742)
Institutional Quality (Government Effectiveness)
1.094*
(0.000)
Institutional Quality (Regulatory Quality)
1.514*
(0.000)
Institutions a
1.114*
(0.000)
Constant
4.062*
(0.000)
3.858*
(0.000)
6.931*
(0.000)
7.876*
(0.000)
6.797*
(0.000)
R2
118.59
85.32
113.67
142.48
108.29
P Value
(0.000)
(0.000)
(0.000)
(0.000)
(0.000)
Note: The explained variable (innovation) and the explanatory variables (FDI and Income) were all logged to bring
them to comparable state. This is apart from the measures of institutions and human capital measure. The values in
parenthesis are the probability values. *, **, and *** represent the significant levels at 1, 5 and 10 percent levels of
significance.
We present the SGMM estimation in Table 5.3. A diagnostic test was carried out on the validity
of our instruments. We checked this by considering the probability values of our AR (2) and Sargan
test and we expect that the probability values of both tests should be above 0.05. The statistics test
in all the columns in Table 5.3 shows that the instruments were not over-identified. The results
from Table 5.3, in all the columns, disclose that innovation has a spillover effect. This is based on
the positive and significant effect of the lagged values of innovation on current level of innovation.
This implies that a current development of innovative act, will affect its future development.
The Table also reveals that an increase in the level of human capital in African countries will
influence the level of innovative development. This is seen in the positive and significant levels of
human capital variable. This is expected as the extent of innovation is embedded on the level of
development of human capital in the country. This finding aligns with other empirical studies such
as Blind (2012). The behavior of GDP per capita and FDI in Table 5.3 turned out unexpected
3
.
This contradicts our expectation but we suspect some explanations for this. For instance, the GDP
per capita was negative, implying that as income per head of the country improves the extent of
innovation will reduce significantly. The issue of motivation and value placed on scientific
enquiries is brought to bear, where increased income reduces motivation for engaging in scientific
enquiries in the African context. This therefore inhibits the extent of innovative outcomes in
African countries. Likewise, a possible explanation for the negative coefficient of FDI variable is
3
The behaviour of these variables do not raise much concern since they were logged. Alternative estimations were
conducted, where the variable GDP and FDI were included in separate models. However, there was no difference in
their signs. This was not reported for conciseness.
19
the particular situation in African countries, where FDI inflow is predominantly resource-driven
and market-seeking FDIs (Asiedu, 2006; Gerlach and Liu, 2010; Asiedu and Lien, 2011). These
kinds of FDI do not have a positive effect on innovation compared to efficiency-seeking FDIs,
which African countries have not been able to sufficiently attract.
Table 5.3 SGMM Estimation (Dependent Variable: Innovation)
1
2
3
4
Lag of Innovation (Scientific and Technical Journal
Articles)
0.364*
(0.000)
0.374*
(0.000)
0.376*
(0.000)
0.397*
(0.000)
Human Capital (tertiary enrolment rate as a percentage
of the gross enrolment)
0.100*
(0.000)
0.088*
(0.000)
0.105*
(0.000)
0.096*
(0.000)
Institutional Quality (Control of Corruption)
0.312*
(0.000)
Institutional Quality (Government Effectiveness)
0.819*
(0.000)
Institutional Quality (Regulatory Quality)
0.973*
(0.000)
Institutions a
0.821*
(0.000)
Income (GDP Per Capita)
-0.002*
(0.000)
-0.003*
(0.000)
-0.004*
(0.000)
-0.004*
(0.000)
FDI Spillover (Net FDI Inflow % GDP)
-0.018*
(0.000)
-0.015*
(0.003)
-0.019*
(0.000)
-0.016*
(0.000)
Constant
1.998
(0.000)
2.346
(0.000)
2.408
(0.000)
2.292
(0.000)
AR (1)
0.001
0.002
0.001
0.001
AR (2)
0.428
0.127
0.178
0.253
Sargan Test
0.530
0.460
0.591
0.498
Note: Same as Table 5.2
Our main variables of interest are the institutional variables, which were all positive and significant
in all the columns. The magnitude of the coefficients of the variables reveals that government
effectiveness, in making policies and enforcing them, and regulatory quality have higher effect on
innovative capacity of African countries. A combined development of the three forms of
institutional quality also yielded a high influence of 0.821. The extent of government commitment
to make policies and ensure the compliance is encouraged for the fostering of innovation in African
countries. This is the major drawback in African countries because some of the political leaders
lack political will to enforce policies that were created by them. In some other cases, policies are
made for self-interest (Jo-Ansie, 2007) and these political leaders thrive on such policies for the
benefit of a particular interest group. In this case, the adverse effect on innovative output is
immense because their attention are diverted from making policies that fosters education and
innovative capacity.
As a robustness, we applied another measure of innovation. In this case, an output measure that
was captured as ICT export. ICT export is able to portray the extent to which innovative activities
are translated into tangible outputs that can be exported and reflected in the gross national products.
Torbjorn (2013), on behalf of UNCTAD, spoke about the contribution of the ICT industry to the
volume of innovative output in a country: suggesting that a countries innovative capacity can be
seen by building a vibrant ICT sector. Other measures of ICT (such as internet, technology
20
utilization in the manufacturing sector) from our databank do not really show the extent of
innovation in the country because these measures only capture the extent of usage of technology
but does not reflect the innovative output of the human resource in the country. Since this is a
robustness check, then ICT export clearly suffices as a supporting explained variable to the one
earlier used in Table 5.3.
The SGMM estimation technique was applied on equation 3, with the explained variable being
ICT export. From the Table, the behavior of the main explanatory variable remained consistently
positive and significant as it is in Table 5.3. As earlier noted, the association between the variables
‘government effectiveness’ and ‘regulatory quality’ on the measure of innovation was higher
compared to that of ‘control of corruption’. Therefore, the conclusion that the result is robust to
the usage of other measure of innovation.
Table 5.4 SGMM Estimation (Dependent Variable: Another Measure of Innovation Output-ICT
Export to Total Merchandise Export)
1
2
3
4
Lag of Innovation (Scientific and Technical Journal
Articles)
0.025
(0.349)
0.008
(0.2835)
0.017
(0.464)
-0.006
(0.809)
Human Capital (tertiary enrolment rate as a
percentage of the gross enrolment)
0.050*
(0.000)
0.053*
(0.000)
0.066*
(0.000)
0.061*
(0.000)
Institutional Quality (Control of Corruption)
0.377*
(0.000)
Institutional Quality (Government Effectiveness)
1.000*
(0.000)
Institutional Quality (Regulatory Quality)
0.889*
(0.000)
Institutions a
1.006*
(0.000)
Income (GDP Per Capita)
-0.147*
(0.000)
-0.343*
(0.000)
-0.302*
(0.000)
-0.346*
(0.000)
FDI Spillover (Net FDI Inflow % GDP)
0.028*
(0.000)
-0.005
(0.398)
0.025*
(0.008)
0.022*
(0.001)
Constant
0.377*
(0.000)
2.842*
(0.000)
2.317*
(0.000)
2.764*
(0.000)
AR (1)
0.050
0.005
0.056
0.055
AR (2)
0.717
0.339
0.598
0.296
Sargan Test
0.524
0.6532
0.599
0.580
Note: Same as Table 5.2
We progressed to find out the point of institutional development for which innovation is enhanced
in African countries. In doing this, we examined the relative impact of the measures of institutions
(corruption control) on innovation by first using Democratic Republic of Congo (the most corrupt
country in our sample) and Botswana (the least corrupt), as benchmarks. Columns 1 and 2 of Table
5.4 report the average values of all the institutional quality measures for the two countries, for the
period 1996-2012. The third column reports the estimated coefficients for the baseline
specification. The fourth column shows the equivalent effect of a change in institutional quality
on innovation. The result in the Table shows that a decrease in the extent of corrupt practices from
the contemporaneous level in Democratic Republic of Congo (DR Congo) to that of Botswana has
about 2.022 percent positive effect on innovation.
21
The relative impact of the extent of government effectiveness on innovation was examined by
using Democratic Republic of Congo (the worst scored in our sample, in this category) and
Botswana (the highest scored), as benchmarks. The result reveals that an improvement of DR
Congo government effectiveness to that of Mauritius, will improve innovation by 5.174 percent.
While Table 5.4 also reveals the relative impact of the extent of regulatory quality on innovation
by using Eritrea (the least score for regulatory quality) and Botswana (the highest score for
regulatory quality). The result shows that an improvement of Eritrea to Botswana will improve
innovation by 6.056 percent.
Table 5.4 Estimated Equivalent Effect of a Change in Measures of Institutions vis-a-vis Innovation
DR Congo
Botswana
Estimated
Coefficient a
Equivalent Effect on
Innovation b
Control of Corruption
-1.464
0.894
0.312
2.022
DR Congo
Mauritius
Government Effectiveness
-1.706
0.656
0.820
5.174
Eritrea
Botswana
Regulatory Quality
-1.729
0.611
0.973
6.056
Notes: a is the absolute values of the estimated coefficients from columns 1, 2 and 3 of Table 5.2; b are the
equivalent effects of a change in the measures of institutions from the level of the worst performer to the best
performer in each category. For the case of control of corruption, the equivalent effect is given by [(-1.464-
0.894)×0.312]/0.364, where 0.364 is the estimated coefficient of Innovation (column 1 of Table 5.2). In this case,
we use the lag value of innovation.
The relative impact of the measures of institution on innovation is higher when government
effectiveness and regulatory quality is improved compared to the improvement of corruption. All
these measures of institution yields positive and significant effect on innovation but at varying
magnitudes. The implication of these results is that African countries should focus on the
improvement of their government effectiveness and regulatory quality to yield a higher rate of
innovation. This is because their equivalent effects were higher compared to corruption variable.
This implication does not downplay corruption, since the control of the extent of opportunistic
behavior is a prerequisite for enhancing innovation; however, the need for strengthening
government commitment and regulatory framework for enhanced innovation is paramount in
Africa.
For emphasis, government effectiveness include the improved quality of public officers and the
degree of their independence from political pressures, the quality of the policy formulated by these
officers and the implementation, and the credibility of their commitment to such policies. On the
other hand, regulatory quality include the ability of the government to formulate and implement
sound policies and regulations that permit and promote private sector development. As a matter of
fact, if these institutional frameworks are developed, the incentive for opportunistic behaviour and
the barrier to innovative activities will be reduced because public officers are committed to making
and pursuing policies that deters hindrances to innovation.
22
6. Conclusion
After examining the impact of institutions on innovation in Africa, it was recorded that the control
of corruption, government effectiveness and regulatory quality positively affects the rate of
innovation in Africa. Hence, an improvement in the quality of institutions in Africa will advance
the rate of innovation in Africa. And since, innovation is a recognized engine for economic growth,
we expect that commitment to advancing the quality of institutions will enable Africa catch up
with advanced economies.
It is clear from our result that as countries in Africa with weak institutions make efforts to reduce
corruption and improve government effectiveness and regulatory quality, the rate of innovation
will change positively. African policy makers can therefore consider innovation-driven growth by
intense commitment to improving institutions that encourages innovation. We lay emphasis on the
extent of government effectiveness and regulatory quality. These two institutional measure had the
most equivalent effect and this implies that an improvement in this regard, will have a significant
impact on innovation.
Subsequent studies on the subject matter of innovation and institutions in Africa can consider how
institutions actually affects innovation. Understanding how institutions affect innovation provides
insight into the mechanisms of each stage on institutional development and how each translate into
improvements in innovation. Understanding this process will help identify what point of the
institutional development process is and is not culminating in enhancements in innovation. This
will make institutional reforms more strategic and purposeful.
23
Acknowledgement
This research is part of a broader PhD research of the first author and the research funding from
Covenant University is gracefully acknowledged. The two anonymous reviewers’ comments were
also very valuable and this we acknowledge.
24
APPENDIX Table A1: List of Sampled Countries
Algeria
Comoros
Lesotho
Niger
Angola
Congo Democratic Republic
Libya
Nigeria
Benin
Congo, Rep.
Madagascar
Rwanda
Botswana
Djibouti
Malawi
Senegal
Burk. Faso
Egypt
Mali
South Africa
Burundi
Eritrea
Mauritania
Swaziland
Cameroon
Ethiopia
Mauritius
Tanzania
Cape Verde
Ghana
Morocco
Togo
Central African Republic
Guinea
Mozambique
Tunisia
Chad
Kenya
Namibia
Uganda
Table A2: Correlation Matrix
Innovation
Human
Capital
GDP Per
Capita
FDI %
GDP
Corruption
Control
Government
Effectiveness
Regulatory
Quality
Innovation
1.000
Human Capital
0.466
1.000
GDP Per Capita
0.083
0.076
1.000
FDI % GDP
-0.171
-0.092
-0.029
1.000
C/orruption Control
0.163
0.233
0.118
-0.023
1.000
Government Effectiveness
0.367
0.325
0.219
-0.032
0.811
1.000
Regulatory Quality
0.365
0.161
0.181
-0.021
0.716
0.867
1.000
25
REFERENCE
Agbemabiese, L., Nkomo, J., & Sokona, Y. (2012). Enabling Innovations in Energy Access: An
African Perspective. Energy Policy, 47(1): 38-47.
Ahmed, A., Cheng, E., & Messinis, G. (2008). The Role of Exports, FDI and Imports in
Development: New Evidence from Sub-Saharan African Countries. Victoria University:
CSES Working Paper No. 39.
Anderlini, L., Felli, L., Immordino, G., Riboni, A., (2013). Legal Institutions, Innovation and
Growth. International Economic Review, 54(3): 937-956.
Asiedu, E. (2002). On the Determinants of Foreign Direct Investment to Developing Countries: Is
Africa Different? World Development, 30(1): 107-119.
Asiedu, E. (2006). Foreign Direct Investment in Africa: The Role of Natural Resources, Market
Size, Government Policy, Institutions and Political Instability. World Economy, 29(1): 63-
77.
Asiedu, E., & Lien, D. (2011). Democracy, Foreign Direct Investment and Natural Resources.
Journal of International Economics, 84: 99-111.
Asongu, S. A. (2013). Modeling the Future of Knowledge Economy: Evidence from SSA and
MENA Countries. Munich: MPRA.
Bandyopadhyay, S., Sandler, T., & Younas, J. (2014). Foreign Direct Investment, Aid, and
Terrorism. Oxford Economic Papers, 25-50.
Barbosa, N., & Faria, A. P. (2011). Innovation Across Europe: How Important are Institutional
Differences? Research Policy, 40(9): 1157-1169.
Blind, K. (2012). The Influence of Regulations on Innovation: A Quantitative Assessment for
OECD Countries. Research Policy, 41(2): 391-400.
Blundell, R., & Bond, S. (1998). Initial Conditions and Moment Restrictions in Dynamic Panel
Data Models. Journal of Econometrics, 87(1): 115-143.
Catrinescu, N., Leon-Ledesma, M., Piracha, M., & Quillin, B. (2009). Remittances, Institutions
and Economic Growth. World Development, 37(1): 81-92.
Cheong, T. S., & Wu, Y. (2013). Globalization and Regional Inequality in China. Discussion
Paper. 13.10. The University of Western Australia.
Christin, T., & Hug, S. (2006). Political Institutions and Ethnic Conflict Resolution: Dealing with
the Endogenous Nature of Institutions. Available at
www.unige.ch/ses/spo/static/simonhug/piecr.pdf: Paper presented at the ISA Conference
in San Diego.
D'Este, P., Lammarino, S., Savona, M., & Tunzelmann, N. v. (2012). What Hampers Innovation?
Revealed Barriers versus Deterred Barriers. Research Policy, 41(2): 482-488.
26
Dirk, W. (2006). Foreign Direct Investment and Development: An Historical Perspective.
Background Paper for the World Economy and Social Survey, Overseas Development
Institute.
Efobi, U.R., (2014). Politicians’ Attributes and Institutional Quality in Africa: A Focus on
Corruption. Journal of Economic Issues (Forthcoming in March 2014 Issue).
Fagarberg, J., Srholec, M., & Verspagen, B. (2010). Innovation and Economic Development.
Handbook of the Economics of Innovation, 2: 833-872.
Fosu A. K. (2013). Institutions and African Economies: An Overview. Journal of African
Economies, 22(4): 491-498.
Gerlach, A., & Liu, P. (2010). Resource-Seeking Foreign Direct Investment in African
Agriculture: A Review of Country Case Studies. FAO Commodity and Trade Policy
Research Working Paper. Working Paper No. 31.
Guan, J., & Chen, K. (2012). Modeling the Relative Efficiency of National Innovation Systems.
Research Policy, 41(1): 102-115.
Jo-Ansie, V. (2007). Political Leaders in Africa: Presidents, Patrons or Profiteers? ACCORD
Occassional Paper Series, 2(1): 1-38.
Kumar, V., Mudambi, R., & Gray, S. (2013). Internationalization, Innovation and Institutions: The
3 I's Underpinning the Competitiveness of Emerging Market Firms. Journal of
International Management, 19(3): 203-206.
Mahagaonkar, P. (2008). Corruption and Innovation: A Grease or Sand Relationship. Jena: Jena
Economic Research Papers, No. 2008, 017.
Malik, T. H. (2013). National Institutional Differences and Cross-Border University-Industry
Knowledge Transfer. Research Policy, 42(3): 776-787.
Murphy, J. T. (2002). Networks, Trust, and Innovation in Tanzania's Manufacturing Sector. World
Development, 30(4): 591-619.
OECD (2011). OECD Guidelines for Multinational Enterprises: Recommendations for
Responsible Business Conduct in a Global Context. OECD Ministerial Meeting.
Osabuohien, E., and Efobi, U., (2013). Africa’s Money in Africa, South African Journal of
Economics, 81 (2), 291-306.
Oyelaran-Oyeyinka, B., & Sampath, P. G. (2006). Rough Road to Market: Institutional Barriers
to Innovations in Africa. Maastricht: UNU-MERIT.
Pattit, J. M., Raj, S., & Wilemon, D. (2012). An Institutional Theory Investigation of US
technology Development Trends since the mid-19th Century. Research Policy, 41(2): 306-
318.
27
Peters, M. (1995). Decentralized Market and Endogenous Institutions. Canadian Journal of
Economics, 28(2): 227-260.
Rao, B. B., Tamazian, A. & Kumar, S. (2010). A Systems GMM Estimate of the Feldstein and
Horioka Puzzle with Data from the OECD Countries. Economic Modelling, 27 (5): 1269-
1273
Rijn, F. V., Bulte, E., & Adekunle, A. (2012). Social Capital and Agricultural Innovation in Sub-
Saharan Africa. Agricultural Systems, 108: 112-122.
Rodrik, D. (2004, November 5). Getting Institutions Right. Retrieved from wcfia.harvard.edu:
http://files.wcfia.harvard.edu/807__ifo-institutions%20article%20_April%202004_.pdf
Romer, P. M. (1990). Endogenous Technological Change. The Journal of Political Economy,
98(5): 71-102.
Romer, P. M. (1994). The Origin of Endogenous Growth. The Journal of Economic Perspectives,
8(1): 3-22.
Schiliro, D. (2010). Investing in Knowledge: Knowledge, Human Capital and Institutions for Long
Run Growth. In M. J. Arentsen, W. Rossum, & A. E. Steenage, Governance of Innovation
(pp. 33-50).
Shleifer, A., & Vishny, W. (1993). Corruption. NBER Massachusetts: NBER Working Paper No.
4372.
Tebaldi, E., & Elmslie, B. (2008a). Do Institutions Impact Innovation? Munich: Munich Personal
RePEc Archive.
Tebaldi, E., & Elmslie, B. (2008b). Institutions, Innovation and Economic Growth. Munich:
Munich Personal RePEc Archive
Tebaldi, E., & Mohan, R. (2008). Institutions-Augmented Solow Model and Club Convergence.
Munich: Munich Personal RePEc Archive
The World Bank. (2008). Measuring Knowledge in the World's Economies. World Bank Institute.
Todtling, F., & Trippl, M. (2005). One Size Fits All? Towards a Differentiated Regional
Innovation Policy Approach. Research Policy, 34(8): 1203-1219.
Torgjorn, F. (2013). Promoting the ICT Sector: The Importance of Internationally Comparable
Data. Available at witsa.org/witsa-wp-site/wp.../Fredriksson_UNCTAD_-GPATS-
2013.pdf (Retrieved 27 June 2014)
World Bank (2013). World Development Indicators. Available at
http://databank.worldbank.org/data/views/variableSelection/selectvariables.aspx?source=
world-development-indicators
Zanakis, S., & Becerra-Fernandez, I. (2005). Competition of Nations: A Knowledge Discovery
Examination. European Journal of Operation Research, 166(1): 185-211.
28
