Firm-level innovation in Africa: overcoming limits and constraints

1. Background and motivation for the special issue

From the perspective of developing countries, innovation is best understood as the process by which firms master and implement the design and production of goods and services which are new to them, irrespective of whether they are new to their competitors (Mytelka 2000). Innovation, then, in a developing country context, should be broadly defined to include processes of adoption and possibly modification of products and technologies that have first been developed elsewhere. The process defined here is not limited to technical functions but also includes organizational and marketing functions (Ernst 2007; UNU-INTECH 2004). These latter aspects are seen as extremely important for African firms (OECD/Eurostat 2005).

The literature on innovation in Africa is rapidly expanding (see, for instance, the special issue of this journal edited by Adebowale et al. 2014), as is the drive towards the measurement of innovation on the continent (AU-NEPAD 2010, 2014). Much of the literature identifies systemic weaknesses and other constraints that inhibit the innovation process on the continent. The characteristics of these constraints as well as concrete interventions to mitigate them are not yet fully understood. Nonetheless, it is interesting to observe that despite all the difficulties, a large share of firms are still able to innovate in the African context. Surprisingly, there are only a few systematic collections of works dedicated to understanding how firms on the continent learn to innovate despite extant limits and constraints. In view of the foregoing, this special issue is dedicated to improving our understanding of the innovation landscape in Africa. Specific issues of interest include:

1. The factors that enhance or inhibit innovation in African firms

2. The sources of (knowledge/information for) innovation

3. Policy options for overcoming constraints and facilitating firm-level innovation

4. The nature and roles of brokers and intermediaries in dealing with innovation constraints and in facilitating the innovation process

5. The role of interactive learning and acquisition of embodied technology in the innovation process

This special issue compiles papers from across the African continent, ranging from Tanzania and Ethiopia in the East to Nigeria in the West.¹ The six papers included in the collection adopt different but complementary theoretical and methodological approaches. This introductory article proceeds by highlighting some aspects of the innovation landscape in Africa, with particular emphasis on the prevalence of innovation and the distribution of information sources across countries. It then discusses the most important constraints to innovation in Africa and shows how the articles in this compilation advance the related research agenda. Finally, a number of issues on which knowledge remains limited are raised.

2. An overview of the innovation landscape in Africa

Table 1 presents the results from innovation surveys carried out in 11 African nations showing the share of firms that introduced, during the three years prior to the survey, innovations that are new to the firm but not necessarily new to the market. The data reported in the table are based on the OECD/Eurostat (2005) definition of an innovation as ‘the implementation of a new or significantly improved product (good or service), or process, a new marketing method, or a new organization method in business practices, workplace organization or external relations’. The data show that a large proportion of firms in African countries are innovative regardless of the level of development of the country. The rate of innovation in individual countries ranges from 40% in Egypt to 77% in Uganda. This suggests that innovativeness at the micro level may not necessarily depend on the current level of economic growth and that, even within the broad group of developing countries, poorer countries may be more innovative than the relatively richer ones. Underlying this pattern are the varied ways through which firms overcome the various economic and infrastructural constraints to innovation. Some of these ways are discussed in the contributions to this special issue.

Table 1. Rate of innovation in selected African countries.

Country	Reference period	% firms with an innovation
Egypt	2008–2010	40.1
Gabon	2010–2012	61.5
Ghana	2008–2010	72.5
Kenya	2008–2011	74.1
Lesotho	2010–2012	58.5
Nigeria	2008–2010	65.0
Senegal	2009–2011	63.1
South Africa	2005–2007	65.4
Tanzania	2008–2010	61.3
Uganda	2008–2010	77.0
Zambia	2008–2010	51.0

Source: Adapted from African Innovation Outlook II (AU-NEPAD 2014).

Note: Extensive dissimilarities in the survey methodologies, especially in sampling, across countries disallow statistically valid cross-country comparisons.

The innovation process is affected by a wide range of factors, including firm size and age, research and development (R&D) efforts, the quality or skill level of managers/employees, employee participation and motivation, managerial practices and inter-departmental cooperation and knowledge exchange, factors related to the firms’ network and its interactions with outside organizations, and factors specific to the industry. Cohen (2010), De Jong and Vermeulen (2007) and Bhattacharya and Bloch (2004) provide excellent overviews of related studies mostly in the context of developed countries. UNU-INTECH (2004) and Hadjimanolis (2000) provide related reviews for the developing country context. In the literature, one of the most widely acknowledged determinants of innovation is firm-level R&D efforts. Innovation does not necessarily proceed from formal R&D (OECD/Eurostat 2005), especially in the African context where few firms are at the technological frontier of their industry (Mytelka 2000). Yet, internal R&D provides a means to create new knowledge and enhances the firm’s capacity to exploit knowledge developed elsewhere (Cohen and Levinthal 1990). It is, therefore, not surprising to observe that in most African countries, many firms that innovate do not engage in formal R&D (AU-NEPAD 2010, 2014).² However, this raises questions about the important sources of knowledge for firm-level innovation in Africa. In addition, the capacity of African firms to generate novelty and the ability to exploit externally generated knowledge remain important empirical questions. These issues are taken up in some of the papers in this volume.

Table 2 shows data on the sources of information for innovation across eight countries. The data are derived from the same surveys as Table 1. As in other contexts – for instance, in Europe (Eurostat 2008) – the general indication is that innovation is a connected activity: several sources outside of the firm itself matter for innovation. The most important of these external sources are market-based: customers, suppliers and competitors. Firms in all the countries, with the exception of Kenya, make little use of information from universities or government and private research laboratories. The weak linkages between firms and these knowledge and research institutions is indeed an important constraint to innovation in African countries, especially in the case of science-based sectors.

Table 2. Sources of information for innovation in selected African countries.

Category	Information sources	Per cent of innovative firms
		Egypt	Kenya	Nigeria	Senegal	South Africa	Tanzania	Uganda	Zambia
Internal sources	Sources within same enterprise group	25.0	95.7	51.7	46.3	41.7	61.9	53.7	5.7
External sources
Market	Suppliers of equipment	17.3	88.0	39.3	43.9	21.3	32.1	26.1	5.7
	Clients or customers	20.0	89.7	51.7	14.6	41.2	66.7	49	10.9
	Competitors	12.0	80.3	30.0	12.2	11.4	27.4	23	3.3
	Consultants	8.9	53.0	14.6	12.2	4.6	16.7	12.2	0
Institutional	Universities or technical colleges	2.7	33.6	6.8	2.4	2.1	7.1	3.2	2
	Government or private research institutions	3.0	40.2	4.2	2.4	1.5	11.9	5.0	0.5
Other	Conferences	10.3	70.9	11.5	7.3	13.3	16.7	16.4	0
	Scientific journals	8.1	64.1	7.1	4.9	10.1	9.5	8.3	0
	Professional associations	4.8	71.8	20.2	4.9	5.6	20.2	11.3	0

Source: Authors’ compilation from African Innovation Outlook II (AU-NEPAD 2014)

Note: Extensive dissimilarities in the survey methodologies, especially in sampling, across countries disallow statistically valid cross-country comparisons.

3. Constraints to innovation in Africa and their implications for policy

Constraints to innovation in a country or continent often appear in the form of characteristics of its domestic innovation system and international value chain (see Annex A in OECD/Eurostat 2005). Such constraints have important implications for science, technology and innovation (STI) policy. For instance, latest available data from firms in African countries show that the innovation barrier most frequently experienced is lack of funds and high costs associated with innovation (AU-NEPAD 2010, 2014). Clearly, this calls for urgent action not only to improve access to credit but also to remove the deficits that push up the costs of innovation. In the rest of this section, we elaborate on some of the specific factors that inhibit innovation on the African continent. Our discussion relates to the economic infrastructure, local institutions, domestic capabilities and the policy context.

3.1. Limited economic infrastructure

The importance of economic infrastructure to innovation and development is well known (Foster and Briceño-Garmendia 2010; Ridley, Yee-Cheong, and Juma 2006; Salami, Kamara, and Brixiova 2010; Yepes, Pierce, and Foster 2009). However, at the global level, there is a significant gap in the amount of infrastructure needed to support economic activities and more so to deliver on the sustainable development goals (McKinsey 2016). The infrastructure gap in Africa is rather appalling. The gap in Africa is estimated to be $31 billion dollars per year, with power generation presenting the most significant challenge although some areas, particularly ICT, have seen a major expansion recently, arguably due to a growing injection of private capital in the ICT sector (Foster and Briceño-Garmendia 2010).

Unsurprisingly, infrastructure network in Africa has increasingly lagged behind what can be found in other developing regions with a huge inequality in terms of rural–urban coverage (Poole and Buckley 2006; Yepes, Pierce, and Foster 2009; You 2008). In fact, Africa’s infrastructure deficit is higher than those of other developing countries with similar levels of per capita income, contradicting the notion that the limited infrastructure in Africa is a reflection of relatively low-income levels (Yepes, Pierce, and Foster 2009). Table 3, for example, compares the low-income countries in sub-Saharan Africa to those of other developing regions, showing that on all the measures of infrastructure presented, the low-income countries in sub-Saharan Africa have lower stock than their counterparts in other developing regions. Current projections on future infrastructure development, however, show that investment in Africa’s infrastructure will remain comparatively low even in relation to other developing or emerging economies (McKinsey 2016).

Table 3. Infrastructure endowment of low-income countries in sub-Saharan Africa and other developing regions.

Measure	Measurement unit	SSA low-income countries	Other low-income countries
Density of paved-road network	kilometres per 100 square kilometres	31	134
Density of total road network	kilometres per 100 square kilometres	137	211
Density of fixed-line	Subscribers per 1000 people	10	78
Density of mobile telephone	Subscriber per 1000 people	55	76
Internet density	Subscribers per 100 people	2	3
Electricity generation capacity	Megawatts per 1 million people	37	326
Electricity coverage	% of households with access	16	41
Improved water	% of households with access population	60	72
Improved sanitation	% of households with access	34	51

Source: Yepes, Pierce, and Foster (2009) and Foster and Briceño-Garmendia (2010).

The constraints that limited infrastructure pose are obvious. In general, inadequate infrastructure and poor management of the little available, in most cases by state-owned monopolies, create the need for local firms to generate their own supply. In fact, infrastructure services in Africa are about 100% more expensive compared to other regions which is a reflection of not only diseconomies of scale in production but also high profit margin due to lack of competition (Foster and Briceño-Garmendia 2010). The cost burden forces firms to divert resources that would otherwise be applied for expansion and innovation, thereby increasing the costs of growth and innovation. For instance, inadequate infrastructure has been a major constraint to productivity growth in small-holder farming in East Africa, a sector which accounts for over 75% of employment in this sub region (Salami, Kamara, and Brixiova 2010). Investing more in Africa’s infrastructure is critical to overcoming these challenges and to spurring innovation. China’s race to become a top innovation performer globally, following after the United States and Japan (Kim 2014), appears as a good lesson for Africa. While there are many determinants of innovation in China (Girma, Gong, and Görg 2008), it seems difficult to delink China’s experience from her massive investment in infrastructure over the past two decades. Between 1992 and 2013, China’s spending on infrastructure stood at an annual average of 8.6% of GDP, which was 3.6 percentage points higher than North America and Western Europe combined and 5.5 percentage points higher than Africa (McKinsey 2016). Increasing investment in infrastructure while addressing the rural–urban disparity in infrastructure coverage will be an important enabler of innovation (Africa Union 2015; UNECA 2014) and for promoting inclusivity in the innovation process and outcomes (Goyal 2016).

3.2. Weak systems/institutional factors

Strong innovation systems, be it at the national or sectorial level, are crucial for innovation performance. The functioning of an innovation system depends on its components – the organizations/actors and institutions – and relations among the components, which perform various innovation system activities (Edquist 2001; Edquist and Johnson 2005; Freeman 1987; Lundvall 1992). Being an essential component of the innovation system, institution’s role and impact are almost definitive (Edquist 2001); they are the rules of the game, determining how the game is played, who plays the game and even whether the game is played (Mudombi and Muchie 2014). The constellation of the characteristics of institutions also matter for innovation as it defines the economic properties of the knowledge produced (Foray 2005). Weak institutions with fragmented constellations of actors tend to constrain interactive learning, hindering innovation (Egbetokun, Siyanbola, and Adeniyi 2007; Iizuka, Mawoko, and Gault 2015; Muok and Kingiri 2015; Oyelaran-Oyeyinka, Laditan, and Esubiyi 1996) and affecting the potential of external regulation and control (government policy) of negative innovation externalities (Voeten and Naudé 2014). Furthermore, there is ample evidence that shortage of credit is a serious constraint to firm growth (Akoten, Sawada, and Otsuka 2006; Beck and Demirguç-Kunt 2006; Fisman 2001; Le 2012; Nguyen and Luu 2013; Rand et al. 2009).

The consequences of weak systems of innovation are enormous. Mudombi and Muchie (2014) have argued that the institutional set up at the national (and international) levels has been less favourable for any meaningful innovation in Africa. Oyelaran-Oyeyinka, Laditan, and Esubiyi (1996) found that innovation in Nigeria’s manufacturing sector has largely been ad hoc and incremental partly due to weak linkages between the firms and public R&D organizations in Nigeria. Through a case study of cassava and maize-processing industries in Nigeria and Kenya, Ndichu et al. (2015) found that innovation and firm learning with respect to energy efficiency in agro-industrial sectors in sub-Saharan Africa (SSA) tend to emanate from informal mechanisms while universities and public research institutes have remained as less important sources of knowledge. Similarly, Osoro et al. (2016, this issue) found that knowledge sources that are external to the firm, particularly R&D, have limited impact on firm-level innovation in Tanzania. Institutional rigidities in Africa countries have also affected the uptake of more appropriate technological innovations (capital goods) from China in favour of those from advanced countries which are largely unsuitable for Africa’s factor endowment (Agyei-Holmes 2016, this issue; Atta-Ankomah 2014).

Of related concern are issues of institutional quality including rule of law, control of corruption and voice and accountability. A large literature has addressed the impact of such institutional quality factors on growth and development (e.g. Acemoglu, Johnson, and Robinson 2005; Glaeser et al. 2004; La Porta et al. 1999). Reviewing several iterations of the World Bank’s Doing Business Report, Davis and Kruse (2007, 1102) pointed out the initial conclusion of regulation in developing countries being more cumbersome, even sometimes completely outdated, in all aspects of business activity. Besides, compared to developed countries, the business registration process is usually long, expensive and often corrupted (World Bank and International Finance Corporation 2006). Many firms therefore prefer to remain in the informal sector, partly to avoid the burden of the formalization process and partly to also avoid the responsibilities such as taxation and regulation that come with being in the formal sector. In addition, many studies have revealed the negative effects of corruption in the form of informal payments on firm growth (Fisman and Svensson 2007; Rand and Tarp 2012; Ufere et al. 2012). In combination, informality and corruption exert a negative influence on firm growth (Rand and Tarp 2012).

3.3. Limited capabilities

Despite the fact that Africa accounts for 12% of the world population, less than 1% of the world’s research output comes from Africa (Mwiti 2015). This somehow points to the extent to which global innovative capabilities are skewed, not only to the disadvantage of Africa but also to that of the entire global south. This pattern has historical roots. Developing countries’ share in global R&D was as low as 2% in the late 1960s, sparking a UN-led campaign to deepen R&D activities in the developing countries and culminating in the birth of a radical document called the Sussex Manifesto in the early 1970s (Bell 2009; Kaplinsky 2011) that challenged the global division of labour in innovation. Developing countries’ share in global R&D has increased since this period, attaining 21% at the beginning of the twenty-first century, although China and South East Asia account for a large part of the increase (Ely and Bell 2009). Correspondingly, East Asian countries such Korea, China, Thailand and Singapore advanced significantly in the twentieth century while Africa has largely missed out (Lee, Juma, and Mathews 2014).

Since the 1970s, our understanding of how innovative capabilities are developed has deepened and transcended an often one-dimensional focus on formal R&D. There has been a recognition that capability development is largely embedded in interactions and feedbacks between R&D, design and production and marketing processes, representing a departure from the linear models of innovation (Bell 2009; Lundvall and Johnson 1994; Von Hippel 2005). This improvement in our understanding has unveiled additional dimensions of the challenges constraining innovation in low-income settings such as Africa. We now know that capabilities built up through interactive learning processes internal to the firm and through interactions between firms and their clients are as crucial as those derived from access to external R&D organizations. (Caloghirou, Kastelli, and Tsakanikas 2004; Cohen and Levinthal 1990; Jensen et al. 2007; Yam et al. 2011). Both internal and external sources of innovation capabilities are conditioned by factors such as institutions (the rules and norms), the state of economic infrastructure, the level and nature of education and the dynamics/relations within systems of innovation (Assink 2006; Mudombi and Muchie 2014; Spielman et al. 2008). As noted earlier, there are significant gaps in all of these elements in the context of Africa countries, even when compared to other developing regions.

3.4. Policy constraints and weak government support

African countries’ policy landscape and the support it renders to innovation capability building raise a lot of concern. Iizuka, Mawoko, and Gault (2015) point out several challenges including weak governance systems for policy-making, implementation, monitoring and experimentation; lack of coherence in policy across national, regional and continental levels; and limited continuity, accountability and transparency of policy. Focusing on agribusiness innovation systems in Tanzania, Mpagalile, Ishengoma, and Gillah (2008) showed that there are major laxities in policy implementation, including limited communication of policy to firms within the agribusiness value chains, while some policies regarding trade, energy, land and labour appears counterintuitive in relation to innovation. There are also issues around policy lock-in during implementation. The innovation systems approach has been officially adopted by many developing countries, including those in Africa with science and technology (S&T) policy having evolved to cover innovation (I) policy and focusing more on capability building; however, the practice still tends to follow a conventional innovation paradigm, and fails to address systemic problems and failures (Intarakumnerd and Chaminade 2007). Iizuka, Mawoko, and Gault (2015) provide an assessment of the extent to which STI policy objectives and priorities of Southern and Eastern African countries have been achieved. Their analysis summarized in Table 4 shows that ‘S&T’ are more advanced in the countries studied compared to ‘I’ as the objectives of strengthening linkages with the private sector and building network of researchers are far behind the others.

Table 4. ST&I policies of the 11 African countries.

Source: Iizuka, Mawoko, and Gault (2015).

Note: The degree of progress in the chart only refers to the specific country’s plan; hence, these evaluations cannot and should not be used comparatively across countries.

4. Outline of the special issue

The broad objectives of this special issue are to shed light through sound empirical evidence on different factors outlined above that facilitate or constrain innovation in Africa, and to identify policy options for overcoming the constraints. The first article by Olawale Adejuwon focuses on a structural element in the African innovation ecosystem, that is, brokers. The illustrative sector chosen is small-scale agriculture, a sector that employs the most labour and contributes the most to real GDP in most African countries. The paper argues that the limited development and diffusion of appropriate innovations in the context of small-scale agricultural production in sub-Saharan Africa can be attributed to the lack of cohesiveness among actors within agricultural innovation systems. Actors can, however, be linked through brokerage. The paper identifies brokerage content that is relevant to the African context and recommends brokers who are embedded in the innovation system by their core functions.

The need to make systemic changes to overcome even broader system-wide challenges is the subject of the paper by Andrew Agyei-Holmes. It builds upon the discussion of Adejuwon in the context of the agriculture sector. The specific problem discussed is how agricultural technologies, particularly from China, are imported and diffused within the Tanzanian agriculture sector. The technology transfer process faces challenges related to institutional and systemic weaknesses. Based on firm-, farm- and government-level data on importation, distribution, usage and maintenance of tractors in Tanzania, the paper offers a deeper understanding of the challenges and proposes some pragmatic solutions.

As already discussed above, the innovation process is difficult in many African countries partly due to weak domestic capabilities. African firms often have low levels of absorptive capacity and hence, have difficulties assimilating knowledge developed elsewhere. This paper makes an assessment of the level of absorptive capacity among Nigerian manufacturing and service firms, and links this to product innovation. The paper by Ukpabio et al. assesses the influence of absorptive capacity on firms’ product innovation, using innovation survey data on Nigerian manufacturing and services firms. A key result of the paper is that the factors associated with the build-up of absorptive capacity in the two sectors differ; hence, there is no one-cap-fits-all solution to the accumulation of capabilities across sectors in Africa. This result underscores the need for sector-specific policies that will enhance firms’ competences and capabilities and drive national competitiveness.

Although patents are widely regarded as an inadequate indicator of innovation particularly in Africa, they nevertheless remain a useful measure. This is so for three reasons, at least. First, patents embody a considerable amount of new knowledge. Second, because they are tangible, patents are some of the most tractable quantities in the innovation process. Third, as a consequence of foregoing reasons, patents offer a nice way to track knowledge flows. These attributes are exploited in the paper by Francesco Lamperti et al. Using patent data and a non-parametric approach, the paper examines persistence of innovation and knowledge flows at the firm level in Africa. This sort of analysis is crucial in order to discriminate between different possible drivers of innovative processes and for guiding public policies aimed at promoting innovation. The paper finds some degree of persistence and a positive impact of knowledge flows from OECD countries.

Beyond knowledge embodied in patents, firms acquire knowledge for innovation from various sources including acquisition of external R&D or new machinery and equipment. This has become a contemporary issue in the economics and management of innovation. The next paper in this issue by Otieno Osoro and colleagues sheds light on the relationship of different external knowledge sources with product and process among a sample of Tanzanian firms, that is, the effect of knowledge sources on firm-level innovation. The analysis reveals, among other things, that product innovation is more constrained by a lack of external knowledge than process innovation. They also show that the joint effect of internal and external knowledge on innovation exceeds the separate effects of internal and external knowledge. This suggest that firms benefit more in terms of innovation by complementing internal knowledge with externally generated knowledge.

The paper by Abdi Yuya and Keun Lee, the last in this special issue, takes the analyses of the preceding paper further by focusing on the knowledge acquired through imported technologies and by participating in international markets. The paper examines how imported technologies and exporting enhance firm-level productivity in the Ethiopian manufacturing sector. The results indicate that exporting, greater use of imported inputs and new capital goods significantly improve productivity and catch-up among firms. The positive productivity effects of imported inputs and new capital goods are higher for exporters than non-exporters. In sum, the findings of this paper suggest that improving access to imported inputs, encouraging investment in new capital goods and strengthening export orientation among manufacturing firms can help accelerate technology transfer and build local innovation capabilities.

5. Issues for further research

The papers in this special issue point to the need for further research in a number of key areas, to deepen our understanding of constraints to innovation in the African context. First, there is a clear need for rigorous analyses on the impact of infrastructure on innovativeness and performance: such analyses are still rare. Datta (2012), Rud (2012) and Ghani, Goswami, and Kerr (2015) lay a formidable groundwork based on Indian data in this regard. Shiferaw et al. (2015) is one of the rare studies in the African context. Another pressing need is for more research on the role of networks and collective action. In this regard, new tools (e.g. dynamic social network analysis as applied in Giuliani 2010) and new data sources (e.g. innovation surveys as championed in Africa by the New Partnership for African Development (NEPAD)) should enable rigorous analyses of networks and interactive learning in Africa. The earlier works of Giuliani and Bell (2005); Oyelaran-Oyeyinka (2005) and Egbetokun (2015) are instructive in this aspect. Poor financing and corruption as barriers to innovation are also under-researched in the African context. It is well known that funding constraints and exposure to corruption inhibit innovation firm performance but the evidence from Africa is very thin. Richer analyses like those by Akoten, Sawada, and Otsuka (2006), Rand and Tarp (2012) and Lorenz (2014) are highly desirable.

In all of the above, the need for effective data collection cannot be overemphasized. The recent efforts under the NEPAD African Science, Technology and Innovation Indicators Initiative (ASTII) are notable, but unless the data are collected longitudinally, sample attrition is discouraged or at least attenuated and the resulting microdata are accessible,³ progress will be very limited.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCiD

Abiodun Egbetokun http://orcid.org/0000-0002-2069-7648

Notes

1. These papers were selected after a thorough double-blind review process. Two independent reviewers assessed each paper and several of the papers had to undergo multiple review rounds. We are immensely grateful to all the scholars who gave their time and expertise to the review process.

2. This also partly explains why marketing and organisational changes are at the heart of the innovation process on the continent.

3. The OECD Eurostat Innovation statistics and indicators (http://www.oecd.org/innovation/inno/inno-stats.htm) is an established example to follow. In Africa, a donor-funded project (http://pedl.cepr.org/content/creating-micro-level-dataset-innovation-nigeria) has taken the first steps in this direction by making the data from the existing two waves of innovation surveys in Nigeria openly accessible (https://goo.gl/gYdpLx).