What factors of Entrepreneurship have the most impact on the competitive position of sub-Saharan Africa?

(1)

on the competitive position of sub-Saharan Africa?

“A country-by-country analysis of Uganda and South Africa”

.

University of Groningen

Master Thesis IB&M, FINAL VERSION

Supervised by:

Dr. Rudi de Vries

June 13, 2016

(2)

List of abbreviations

APS – Adult Population Survey

EFC – Entrepreneurial Framework Conditions GCI – Global Competitiveness Index

GCR – Global Competitiveness Report GDP – Gross Domestic Product

GEM – Global Entrepreneurship Monitor GVC – Global Value Chain

IMF – International Monetary Fund NES – National Expert Survey

OECD - Organisation for Economic Co-operation and Development TEA – Total early-stage Entrepreneurial Activity

WEF – World Economic Forum

(4)

1. Introduction

Since the beginning of this century there has been increasing attention towards the development crisis in continents such as Asia and Africa (Robson, Haugh, Obeng, 2008). Significant difference between a developing continent as Africa1_{and a developed continent}

such as Europe remains primarily in terms of economic prosperity. “The everyday livelihoods of Africans have not kept pace with macroeconomic growth, and per-capita GDPs on the continent persistently lag behind the rest of the world” (Fal, 2013, p. 149).

In spite of these lacking economic developments, Africa’s potential on the other hand seems to be quite large. “Africa is the second largest continent in the world, the third largest in population and is rich in natural resources” (Koveos, Yourougou, Amoaku-Adu, 2011, p. 1). In terms of entrepreneurial activity, Africa is interesting as well. According to the World Bank (2008), the African region is among the most open regions with regards to world trade. In terms of trade as a share of GDP, Africa is only surpassed by the Middle East and North Africa region (World Bank, 2014). Africa thus leaves other developing regions such as the East Asia & Pacific region and the Latin America & Caribbean region in their wake. So even though Africa displays great potential in the field of entrepreneurship and

competitiveness, Africa has so far not been able to realise the potentials in terms of increasing economic development and in terms of an increase in their competitive position as well. But why is it interesting to research the relationship between entrepreneurship and

competitiveness? The level of entrepreneurial activities is an important determinant and can have significant effect on the competitive position of a country. The foundation behind this argument can be found when looking at the descriptive power that the concept of

entrepreneurship entails: “Entrepreneurship is a mechanism by which society converts technical information into products and services” (Shane and Venkataraman, 2000, p. 219). Additionally, entrepreneurship also identifies “temporal and spatial inefficiencies” in an economy. These economies than have the opportunity to formulate any responsive or mitigating measures. The last reason that Shane and Venkataraman provide is the process in which countries transform to more capitalist economies due to “isolated entrepreneurially driven innovation” (Shane and Venkataraman, 2000, p. 219) in products and processes.

1_{Whenever the term “Africa” is used in this research, I am referring to the sub-Saharan region of Africa. This}

(5)

Schumpeter already mentioned the importance of entrepreneurial innovation in 1934. Furthermore, Amoros et al (2011) state the following about entrepreneurship and competitiveness: “Entrepreneurship is a very important activity for a country’s

competitiveness and growth and a significant source of mobility” (p. 2). Additionally Birch (1979, 1987) emphasizes the importance of job creation as a result of increasing

entrepreneurship, through the establishment of new firms.

It is clear that in the past decades several scholars emphasized the importance of

entrepreneurship on the competitiveness of a country. However, it should be clear that there is no standard framework of entrepreneurship that would work equally well for every single country. Entrepreneurial dynamics are very different in for example the Netherlands than in a country such as Puerto Rico.

It is therefore essential to analyse the specific situation or dynamics of different countries in order to determine the best policy implications for each country. In this case I will focus on the possible policy implications relevant for African countries, because of the interesting situation Africa is in at the moment. In order to find out why Africa is underperforming despite of the present potentials in the field of competitiveness and entrepreneurship, I will analyse what specific entrepreneurial variables are most important in boosting the competitive position of African countries. The main question of this research will therefore be:

What factors of entrepreneurship have the most impact on the competitive position of countries located in the sub-Saharan part of Africa?

As will be explained in the methodology section, this research will be based on the research conducted by Amoros et al. (2011). This research will complement and contribute to previous research by conducting a country-by-country analysis. The two countries that will be

compared are Uganda and South Africa. For these two countries the availability of relevant data is most extensive. Uganda is interesting because of the high entrepreneurial activity while South Africa is interesting because of the economic development stage. This will be further elaborated in the methodology section of this paper.

In the first section of this research the relevant literature will be discussed. Within the literature the following four important concepts will be reviewed: the field of

entrepreneurship, the competitiveness of nations, entrepreneurship in Africa and the

(6)

will be discussed. The observed outcomes will be displayed in the results section which will lead to the discussion of the results. Concluding remarks can be found in the final section of this paper.

2. Literature review

In this section we will discuss the relevant theories for this research. In the first part of this section the subject of entrepreneurship will be explained. I will then discuss the concept of competitiveness. The last part of this section will consist of a discussion of the entrepreneurial and the competitive positions of the sub-Saharan part of Africa.

2.1. The field of entrepreneurship

One of the biggest challenges in the field of entrepreneurship is the formulation of a general definition of the concept (Shane and Venkataraman, 2000). Up until the article by Shane and Vankataraman was written in 2000, the authors claim that entrepreneurship was mostly described through the scope of the individual. This means that researchers mostly described the entrepreneurial activities of an entrepreneur (Venkataraman, 1997). But according to Shane and Venkataraman (2000) such an angle of incidence on the concept of

entrepreneurship is incomplete. The authors argue that entrepreneurship consists of two phenomena: “The presence of lucrative opportunities and the presence of enterprising individuals” (p. 218). Therefore the identification of an entrepreneur as someone who

establishes a new company or organization is incomplete. It does not consider the variation in the identification of entrepreneurial opportunities (Shane and Venkataraman, 2000).

According to the authors this definition is flawing because the measurement of

entrepreneurial opportunities is neglected. Therefore they opted for a new and more complete definition of the concept of entrepreneurship. This newly formulated definition of the field of entrepreneurship consists of the following three factors:

- The sources of opportunities;

- The process of discovery, evaluation and exploitation of opportunities; - The individuals who discover, evaluate, and exploit them.

This interpretation of entrepreneurship does not only consider the individual, but accounts for the identification and the implementation process of these opportunities as well.

Entrepreneurship can thus only occur when there are entrepreneurial opportunities.

(7)

methods can be introduced for a price higher than the cost of production (Shane and

Venkataraman, 2000). The discovery of such opportunities depends on whether an individual recognizes the opportunity or not. The probability of such an opportunity being recognized by an individual depends on: “The possession of the prior information necessary to identify an opportunity and the cognitive properties necessary to value it” (Shane and Venkataraman, 2000, p. 222). After the potential entrepreneur recognizes the opportunity, he or she has to decide whether to exploit the opportunity or not. Venkataraman (1997) argues that the

exploitation of these opportunities depends on the nature of the opportunity and the individual differences associated with valuing such an opportunity. The ‘nature of opportunity’ describes what the value of such an opportunity is expected to be. A cure for a terminal illness will probably have a higher expected value than a solution for not having snacks in the local gym. This expected value will however differ from person to person as well. Some individuals have access to greater financial capital, while others depend more on previous (entrepreneurial) experience. Another individual difference can for example derive from differing levels of optimism between individuals. In short, there are many different variables that can determine these individual differences.

Entrepreneurial activity can be exploited in different modes: through creation of new firms and through the sale of opportunities to existing firms. The most common assumption however is that the exploitation of entrepreneurial activity is executed through the establishment of start-ups.

According to Shane and Venkataraman (2000), the field of entrepreneurship has proven to be a difficult and complex concept that will probably impose an equivalent of difficult questions for those who are active within the field. Nevertheless, they state that the subject of

entrepreneurship is a very important and relevant field of study as well. With this research I hope to unravel important variables regarding entrepreneurship and thus shed some light on the complexity of this concept. This is captured by the first two sub-questions of this research:

1. What are the most important variables in the measurement of entrepreneurship? 2. What entrepreneurial variables are more and less important in different stages of

a nations’ development? 2.2. Competitiveness of nations

(8)

that time) a general misconception about the definition of national competitiveness.

According to general thinking at that time, the determination of national competitiveness was all about labor costs, interest rates, exchange rates, and economies of scale. Porter argues that this line of thinking is flawed as it is focused on short-term successes and it thus neglects the true sources of sustainable competitive advantage. He claims that the only meaningful concept of competitiveness at the national level is productivity. Porter (1990) states the following about productivity as a measure of national competitiveness: “The principal goal of a nation is to produce a high and rising standard of living for its citizens” (p. 76). The

standard of living is often measured in terms of the Gross Domestic Product (GDP) of a country (Berenger and Verdier-Chouchane, 2007). According to the OECD (2007) the GDP measure “is an aggregate measure of production equal to the sum of the gross values added of all resident institutional units engaged in production” (p. 346). The sum of the final uses of goods and services measured in purchasers’ prices, less the value of imports of goods and services, or the sum of primary incomes distributed by resident produces units.

2.2.1. Globalisation and the rise of Global Value Chains (GVC).

(9)

This process of production fragmentation has continued throughout the past decades and has led to the founding of sustainable Global Value Chains (GVCs) around the world (Gereffi and Fernandez-Stark, 2011). Gereffi et al. (2011) defines a GVC as follows: “The value chain describes the full range of activities that firms and workers perform to bring a product from its conception to end-use and beyond” (p. 4). The GVC methodology framework provides a helpful tool in the analysis of the complex structures of global industries and how they are organized. The framework also carefully analyses the actual value added per participant in the chain.

Access to GVCs can be particularly rewarding for lower income countries. GVCs can act as a stepping-stone for these countries as long as they are effectively integrated within the chain. According to Gereffi et al., successful integration within these chains is a vital condition for the development of lower income countries. In other words, participation of lower income countries within GVCs is crucial for the development of their competitive position. “Successful participation in GVCs depends on the ability to access GVCs, to compete successfully and to capture the gains in terms of national economic development, capability building and generating more and better jobs to reduce unemployment and poverty” (Gereffi et al, 2011). This means that (successful or unsuccessful) participation in GVCs will be reflected through the GDP measurement of a country as well (Gereffi et al, 2011). In line with previous research conducted in 1990 by Porter and in line with more recent research conducted by Gereffi et al. (2011), in this research I will measure the competitive position of a nation through the usage of the GDP per capita. The content of this sub-section leads to the following sub-question of this research:

3. What are the most important variables in the measurement of a nations’ competitive position?

2.3. Entrepreneurship in sub-Saharan Africa

As was already mentioned in the introduction of this research, the wealth of Africans continues to lag behind despite the encouraging trends and the increase in macroeconomic growth (Fal, 2013). Fal believes that the income gap between countries in Africa and countries in the rest of the world can be reduced by entrepreneurship as long as the

(10)

entrepreneurship in Africa, the Omidyar Network2 launched the Accelerating

Entrepreneurship in Africa Initiative in 2012. This initiative developed a survey in which 582 entrepreneurs participated. These entrepreneurs originated from 6 different countries in the sub-Saharan region of Africa. In addition to the surveys, 72 in-depth interviews were conducted which were benchmarked against 19 global peers. In the second phase of the research the results were interpreted and analysed by businesses, government officials and thought leaders. More than 300 relevant leaders were involved in this process. The

investigation was focused on finding the barriers and challenges of African entrepreneurship based on important pillars of the entrepreneurial environment: assets, business support, policy accelerators and motivation and mindset.

Assets

Entrepreneurial assets consist of three sub-factors: the source of financing, skills and talent and infrastructure. According to the results of this investigation, entrepreneurs are restrained by the lack of access to finance. The participating entrepreneurs argued that the cost of equity and debt capital hindered company formation and growth. “In some cases, banks required 150 percent of the borrowed amount in collateral” (Fal, 2013, p. 151). Additionally, government financing is constraint by bureaucracy and nepotism.

Regarding skills and talent, results showed that education in Africa is to heavily focused on preparing the African workforce for employment in establishing firms. Existing courses on entrepreneurship do not attend to the practical side of entrepreneurship, which are needed to start, manage, or work in potential entrepreneurial ventures.

Participants of the survey agreed that the infrastructure in sub-Saharan countries is simply inadequate, unreliable, costly and inefficient. Another constraint is found in the supply of electricity, which is deemed as inadequate and unreliable as well.

Business support

Business support covers variables as business advisory services, government programs and incubators. Business advisory services are not widely available and are primarily located at urban centres. Availability is not the only problem. The quality of such services are severely lacking as well. The consequences are especially noticeable for early-stage businesses. In the

(11)

previous two decades African governments have increased support for entrepreneurs, but according to the participants of the survey the effects of these initiatives are hardly noticeable. Incubators are increasingly represented in countries in Africa and have had some successes in accelerating entrepreneurship. Most African entrepreneurs however claim that the amount of available incubators is insufficient.

Policy accelerators

The term policy accelerators incorporate the terms legislation and administrative burdens. In general the participating entrepreneurs in the monitor survey share a positive view towards African legislation. South Africa is the only sub-Saharan country in which legislation is seen as prohibitive towards entrepreneurship. This is due to the harsh penalties that are enforced by the government regarding noncompliance. The perceived positive view on legislation in other sub-Saharan countries is due to the ability of entrepreneurs to confine poorly enforced

regulations. Due to the informal climate entrepreneurs are able and allowed to operate below the radar and outside the constraints of the (formal) laws. If a company decides to do business in the informal climate, it will be economically excluded with only limited access to financial and consumer markets. Sixty percent of respondents believe it is acceptable to start and operate a new business in the informal climate. This results in a loss of potential tax revenues for the active governments.

According to African entrepreneurs administrative burdens continues to impede the development of business in the continent. Fal, (2011) complements to this information by stating that out of 183 countries ranked in terms of “ease of doing business” and “ease of starting a business” sub-Saharan countries are all ranked in the bottom half on both indices (p. 165). On a more positive note, African entrepreneurs do think that past reforms by

governments with regards to the ease of doing business have been fruitful. Furthermore African entrepreneurs all agree on the statement that the business environment in sub-Saharan countries has significantly improved in the last 10 to 20 years.

Motivation and mindset

(12)

general views on entrepreneurship are improving as well. It is increasingly seen as an

acceptable career choice. However in the view of many Africans, “successful” entrepreneurs are only judged based on their wealth and possessions rather than their business ideas. As a consequence, a lot of novice entrepreneurs are spending money on their image instead of their business ideas. Therefore, this could have negative consequences on their future businesses. 2.4. Competitive position of sub-Saharan Africa

As was mentioned before in this research paper Africa has experienced high economic growth in the last couple of years. Despite this high growth, African countries are not able to

strengthen their competitiveness position compared to the rest of the world. What is the reason for this stagnation in their development? In June 2015 the “Africa Research Bulletin” released the latest competitiveness report describing the latest trends and challenges in the field of competitiveness on the continent (Africa: Competitiveness Report, 2015). In a nutshell the report states that poor-quality institutions, bad infrastructure and poor health and education quality are the main culprits in the stagnation of competitiveness. The Global Competitiveness Index (GCI), which is established by the World Economic Forum (WEF), evaluates three key areas of economic activity. These areas are: Agricultural productivity, services sector growth and global and regional value chains (More about GCI index can be found in the methodology section). Stagnation is found in all sectors due to the above-mentioned culprits. On a positive note, Africa is performing better on labor and goods markets.

At the moment Africa is experiencing a rapid growth of the working-age population. This development will certainly lead to opportunities for investments in highly productive labour-intensive sectors, as this would probably generate much-needed employment opportunities for women and youth.

(13)

would certainly contribute to realising easy market access and give opportunity to improve production processes as well.

The Africa competitiveness report investigated trade statistics as well. These trade statistics suggests that the exports of services are very significant for the African economy as well. It is therefore important to further develop low-cost and high quality services, as this will help African countries to participate in global value chains. The Africa competitiveness report believes that it is essential for the competitiveness of African countries to actively participate in global value chains, as it will accelerate the economic transformation. It is however very important to mention that many supportive policies on trade facilitation, investments,

infrastructure and access to finance are needed in order to successfully integrate within GVCs. To conclude, African countries should work on a variety of competitive enhancing factors in different areas of the economy. African countries should work on increasing productivity in the agricultural sector and in the upcoming services sector as well. Productivity could be boosted by faster structural transformation. According to the OECD Structural transformation could further enhance job creation and social cohesion. In addition to the focus on

productivity, African countries should invest in infrastructure, adopt more dynamic trade procedures, increase regional integration and build more effective institutions.

Now that subjects as competitiveness and entrepreneurship are discussed through the scope of sub-Saharan Africa, the final two sub-questions can be introduced:

4. What are the most important entrepreneurial variables in relation to the competitive position of the Sub-Saharan part of Africa?

5. What are the most important entrepreneurial variables in relation to the competitive position of Uganda and South Africa?

(14)

Figure 1: Conceptual model. Entrepreneurship positively effects the competitive position of a nation and is moderated by stage of development and other variables.

The field of entrepreneurship and the competitive position of a country can be identified as the independent and the dependent variable of this research. As a country displays an increase in entrepreneurial activity, I expect that these activities will positively influence the

competitive position as well as positively influence economic development within the said country. I further expect that the ‘stage of economic development’ (in terms of GDP per capita) has a positive influence on the relationship between entrepreneurship and the competitive position of a country. This research revolves around this positive effect as I am aiming at clarifying what entrepreneurial factors will be important in different stages of competitive positions. Once these factors are identified, I will further analyse them in the context of Uganda and South Africa.

Sub-‐Saharan Africa

Entrepreneurship Competitive position

(15)

3. Methodology

In this research the methodology will be based on earlier research conducted by Amoros et al. (2011). Their research was based upon the countries located in Latin America. In this research the focus will be on African countries. ‘Entrepreneurship’ and ‘Competitiveness’ are both complex concepts; they are determined by many variables. This is proven by research conducted by the Global Entrepreneurship Monitor (GEM) and the World Economic Forum (WEF). The GEM and the WEF will provide all the necessary data for this research, which is in line with the research conducted by Amoros et al. (2011). In their research they analysed all the countries together, which gave aggregated information regarding groups of countries with the same level of GDP. The authors state this as an important limitation of their research and suggest that future research may want to focus on comparing the development of two

countries to generate even more interesting (and disaggregated) results. This thesis will therefore contribute to existing literature by extending previous research by conducting a country-by-country analysis.

The first part of the methodology section will explain the origin and background information on the used datasets in this research (GEM and WEF).

3.1. Global Entrepreneurship Monitor (GEM)

According to the GEM itself they can be described as the “world’s foremost study of entrepreneurship” 3. The GEM launched their project in 1999 with the intention to find out why some countries are more ‘entrepreneurial’ than others. Since their founding they have collected 17 years of data. This data consists of 200.000 interviews a year, conducted in more than 100 countries. More than 500 specialists are involved in gathering data related to

entrepreneurship research. More than 300 academic and research institutions are involved in this process as well. The GEM investigates two main elements:

- The entrepreneurial behaviour and attitudes of individuals and - The national context and how that impacts entrepreneurship.

The collected data is composed by two surveys: The Adult Population Survey (APS) and the National Expert Survey (NES). According to the GEM the APS tracks the entrepreneurial attitudes, activity and aspirations of individuals while the NES monitors nine factors that are believed to have a significant impact on entrepreneurship. These nine factors are known as the Entrepreneurial Framework Conditions (EFCs). Participants of the NES survey are carefully

(16)

(17)

Figure 2: GEM conceptual framework, GEM Global Report (2013). 3.2. World Economic Forum (WEF)

For over 35 years the WEF has presented reports that described key factors and their underlying relations with regards to economic growth and a country’s present and future prosperity. According to the WEF the goal of establishing such a report is to reach a common understanding of the strengths and weaknesses of an economy so that stakeholders can work together to shape economic agendas that address challenges and enhance opportunities. The GCR report of 2015-2016 represents 140 economies and their corresponding competitive positions in terms of an index: The Global Competitiveness index (GCI). The GCI is compiled of 114 indicators that are grouped into 12 “pillars”. These pillars are: Institutions, infrastructure, macroeconomic environment, health and primary education, higher education and training, goods market efficiency, labour market efficiency, financial market

(18)

efficiency-driven economy is the next step in the development process. In this development stage the economy has become more competitive by an increase in industrialization and an increase in reliance on economies of scale in which capital-intensive large organizations are more dominant. Basic requirements are significantly improved and thus the focus shifts towards efficiency enhancers. In the last development stage, businesses are more knowledge-intensive, and the service sector expands substantially. Focus on further business development is built upon innovation and enhancing business sophistication.

Figure 3: Pillars Global Competitiveness Index, Global Competitiveness Report 2015-2016. The information on the aforementioned pillars is carefully collected through contributing organisations as the International Monetary Fund (IMF), the United Nations Educational, Scientific and Cultural Organization and the World Health Organization. Additionally, the WEF formulated the Executive Opinion Survey to capture concepts that require a more

qualitative assessment, or for which comprehensive and internationally comparable data is not available. According to the Global Competiveness Report of 2015-2016, the Executive

(19)

economies. For 6 other countries data from previous years were used so that in total 140 economies were covered in the latest issue of the GCR report. The average number of valid returned surveys averaged around the 90 per economy.

3.3. Data analysis

The goal of this research is to discover what entrepreneurial factors are important in different stages of competitive positions. As mentioned before, this research will follow the method of analysis conducted by Amoros et al. (2011) in which the authors were able to successfully analyse the relationship between entrepreneurship and the stages of competitiveness. The GCI data is very extensive and contains a lot of variables. It is therefore useful to investigate which variables have relative priority over others in enabling countries to attain their competitiveness goals (Amoros et al., 2011). Thus, I analysed the weight of relative importance that can be related to each variable for each country, in order to effectively measure this relative priority of variables. I expect that entrepreneurship will act as a key factor for countries that are trying to achieve their competitiveness-level goals. Therefore it is important to consider what variables are directly related to the entrepreneurship phenomena in the analysis of the ordinal classification of the variables. As Amoros et al. (2011) mentions, GCI sub indices are weighted subjectively. This could result in some sub-indexes being over- or underweighted. In order to find common structures among the variables independent from the categories that GCR defines, a factor analysis was conducted. In this way I will be able to identify a lower number of underlying structures.

The GCI variables will be grouped into different factors, and the next step is to analyse the effect of macroeconomic and entrepreneurship variables. According to Amoros et al. (2011) it is necessary at this stage to incorporate a discriminant analysis in order to effectively measure the aforementioned effect. By conducting a discriminant analysis it is possible to discover the differences in weight of importance that each factor contributes to a country’s ability to attain prosperity. Additionally the analysis allows for characterization and classification of different groups of countries (Moreno and Casillas, 2007).

3.4. Dependent and independent variables

As was previously discussed in the theory section as well, a countrys’ GDP per capita is probably the most useful measure when looking at their competitive position (Porter, 2005). Porter states the following about the GDP measure: “It reflects a country’s structural

(20)

differentiate based on the GDP per capita measure. Three groups will be identified; countries with low GDP per capita, countries with GDP per capita between certain thresholds and countries with high GDP per capita. The GDP per capita measure in this research can therefore be described as the categorical dependent variable (Amoros et al., 2011).

The independent variable of this research can be identified as entrepreneurship. Earlier in this methodology section a factor analysis was introduced to group certain variables to different factors of competitiveness. This process (the factor analysis) was repeated for the

indicators/variables of entrepreneurship. For this analysis the dataset provided by the GEM was used.

3.5. Country-by-country analysis: Uganda and South Africa

After the factor analysis and the discriminant analysis are completed, the findings will be further tested in the country-by-country analysis. This thesis will therefore extend and contribute to previous research. The methodology of this country-by-country analysis will depend on the outcome of the factor analysis and the discriminant analysis, as the two countries will be compared based on the most relevant entrepreneurial factors for their specific situation. In this analysis the GEM data will be used as well as relevant

complementary secondary data. A timeframe of 5 years will be applied to the analysis, so that changes in economic development can be effectively measured. Data is most extensively available between the years 2009 and 2014. Research will therefore be conducted in this timeframe. The country-by-country analysis should be as extensive as possible. I therefore chose to analyse the two sub-Saharan countries on which the most information is available. Furthermore, there were only two countries with enough information in the set timeframe of 5 years. These two countries are: Uganda and South Africa. Both Uganda and South Africa are very interesting to investigate as well. Uganda is in terms of early stage entrepreneurial activity one of the highest scoring countries in the world, as can be seen in figure 4. This figure shows the top 20 countries of the world with the highest ‘total early-stage

(21)

Figure 4: Global top 20 Total Early-stage Entrepreneurial Activity (TEA) in 2014, GEMconsurtium.org (2014).

From all the sub-Saharan countries, South Africa can be seen as one of the most developed countries in terms of national income. Botswana is the only sub-Saharan country that

(22)

4. Results

In this section the results of the analysis will be presented. As is mentioned previously in this paper, the factor analysis is needed because of the subjective weightings that are established by the WEF. These subjective weightings formulated the current model of the GCI as was presented in figure 3 of this thesis (p. 17). In order to maintain the conceptual value of the model as presented in figure 3, the model should be investigated without the subjective weightings (Amoros, et al., 2011). In addition to ruling out subjective weightings by the WEF, this analysis enables the establishment of conceptual relationships between the

variables that are grouped together by the conducted factor analysis in this thesis. The results of the factor analysis can be found in the rotated component matrix of Appendix 2. At the start of the analysis procedure no limits in terms of a maximum amount of components was set. The factor analysis, as performed in SPSS 22, is typically stopped before the eigenvalue surpasses 1. For clarity, any outcomes below 0.200 are excluded from the analysis, as they will not have a significant impact on the results. By excluding outcomes below 0.200 the rotated component matrix in Appendix 2 became easier to interpreted. In my view this is desirable, due to the large amount of variables that are included in this analysis. The original factor analysis yielded a total of 16 measured components. However, many of these

components (components 5 to 16) consisted of only 1 affiliated variable, which did not

substantially correlate with their affiliated components as well. Furthermore, said components often correlated with another component for which the correlation rate was almost the same in both cases. This means that the impact of these variables on the analysis is minimal. For the sake of clarity, I therefore chose to limit the components of the factor analysis to 4 underlying themes (which is consistent with Amoros et al. 2011). The 4 components of the factor

analysis conducted in this research together explain 66% of the variance, as can be seen in the table below (Table 1). Although this may not seem particular high, I chose to proceed the analysis with 4 components anyway. This is due to the fact that other analyses with different amounts of components did not reveal any additional underlying themes.

(23)

Component

Initial Eigenvalues

Total % of Variance Cumulative %

1 52.683 47.039 47.039

2 9.521 8.501 55.540

3 7.325 6.540 62.080

4 4.182 3.734 65.813

Table 1: Explained variance by the constructed components in the factor analysis.

The Kaiser-Meyer-Olkin (KMO) measure has to score at least 0.600 or higher whereas the Bartlett’s test has to be significant below 0.05. The results of these two tests can be found in Table 2. Both statistics meet the demanded requirements.

Kaiser-Meyer-Olkin Measure of Sampling Adequacy. .815 Bartlett's Test of Sphericity Approx. Chi-Square 35115.870 Sig. .000

Table 2: KMO and Bartlett’s results as determined by SPSS. 4.1.1. Interpretation of the observed factor components.

The interpretation of observed components within the factor analysis is quite subjective. Hence, the researchers in question have to re-categorize (and rename) factors within the components that are revealed by the factor analysis. This is even more relevant for this analysis because of the large amount of variables that are involved. In this subsection the underlying themes within the 4 different components will be discussed.

(24)

47% (see Table 1). The most prominent underlying factors are efficiency, higher education, business environment and innovation. The efficiency factor is measured by variables that are related to labor market efficiency, goods market efficiency, financial market efficiency and government efficiency. The innovation factor consists of variables such as patent applications, capacity for innovation, quality of scientific research institutions, FDI and technology

transfer. The Business environment factor is related to variables such as local supplier quality and quantity, extent of market dominance, state of cluster development, extent of marketing and nature of competitive advantage for example. The last prominent underlying factor, higher education, consists of variables related to quality of the education system, quality of math and science education, quality of management schools, extent of staff training and the availability of research and training services.

Component 2 can be identified as the General welfare component. This component describes factors that are related to Health, macroeconomic development, education enrolment,

technological readiness and electricity and telephone access. Together these factors account for 8,5% of the total variance. The most prominent variables in this category are related to Health. The second most important variables describe electricity and phone line connections. Examples of such variables are: impacts of malaria, tuberculosis and HIV/AIDS on business, infant mortality, life expectancy, mobile telephone subscriptions, quality of electricity supply and fixed telephone lines.

Component 3 can be identified as the Innovative and open mindset component. This

component consists of factors that boost and support innovation. Examples of such variables are: University-industry collaboration, overall company spending on R&D and the availability of scientists and engineers. Additionally, this component covers variables related to the

‘openness’ of a country. Examples of such variables are: Value chain breadth, willingness to delegate authority and the control of international distribution. This component explains 6,5% of the total variance.

Component 4 can be identified as the Competition component. This final component describes variables that are related to competition and labor market flexibility. This

(25)

If the 4 components found in this research are compared with the original pillars constructed by the WEF, a couple of differences between the two can be identified. The WEF pillars and the established components in this research can be found in table 3.

WEF Pillars Components identified in this research: Basic requirements: Institutions,

Infrastructure, Macroeconomics, Health and Primary Education

1. Business enhancers: Efficiency, Business environment, Innovation, higher education,

transport infrastructure, safety and security, Ethics and the financial market

Efficiency enhancers: Higher education and training, market efficiency and technological readiness.

2. General welfare: Health, macroeconomic development, education enrolment, technological readiness and electricity and telephone

Innovation and sophistication factors: Business sophistication and innovation.

3. Innovative and open mindset: Innovative mindset, openness of market, ICT and air transportation

4. Competition: Foreign and domestic competition, market size and labor market flexibility.

Table 3: Comparison between the WEF Pillars and the 4 components found in this research.

It is clear to see that the basic requirements category of the WEF closely resembles

component 2 identified as ‘general welfare’ in this thesis. The difference however is that basic requirements would be separated between ‘business enhancers’ and ‘general welfare’ in this research. Additionally, the category efficiency enhancers would be divided into ‘business enhancers’ and ‘general welfare’ as well. Lastly, innovation and sophistication factors can be found in ‘business enhancers’ and ‘innovative and open mindset’. The aim of this factor analysis was to categorize all the variables of the GCI into a couple of underlying themes that are purely based on correlating variables.

4.1.2. Discriminant analysis

Now that these underlying themes are identified, the next important issue to uncover is the importance and impact of each factor on the specific economic situation of a country.

(26)

or more groups (Klecka, 1980). In this analysis three groups will be analyzed. Group 1 consists of countries that are relatively poor, countries that have intermediate success are located in group 2 and group 3 consists of relatively rich countries. As mentioned in the methodology part of this research, the successfulness of a country will be determined based on GDP per capita. This will be the categorical dependent variable. The grouping thresholds used in this research will be based on the thresholds established by the WEF (Table 4). In this research I ignore transition stages, as I am specifically interested in the relation between the variables and the countries that are poor, intermediate or rich4. The following conditions are applied in this analysis:

- Countries with a GDP per capita below 2000 US$ are considered as poor.

- Countries with a GDP per capita between 3000US$ and 9000US$ are considered as intermediate performing countries.

- Countries with a GDP per capita above the 17000US$ threshold are considered as rich countries. Stage 1: Factor-driven Transition from stage 1 to stage 2 Stage 2: Efficiency-driven Transition from stage 2 to stage 3 Stage 3: Innovation-driven GDP per capita (US$) thresholds < 2000 2000-2999 3000-8999 9000-17000 >17000

Table 4: Income thresholds for stages of development. Source: Global competitiveness report 2013-2014, p. 10. Now that the grouping variable and its thresholds are clearly identified, it is time to determine the independent variables for the discriminant analysis. The most important variables of the factor analysis (so the most important variables of the 4 established components) were inserted into the analysis as independent variables. Due to the fact that many variables are involved in this analysis, I chose to analyze them all together instead of using a stepwise method in SPSS. The stepwise method in SPSS only selects variables that keep the Wilks lambda statistic as close to zero as possible. The closer the Wilks lambda statistic is to zero, the more the variable contributes to the discriminant function. The stepwise method only

(27)

selected 12 variables in this analysis, whereas I am interested in the effects of all the variables that resulted from the factor analysis. In Appendix 3 the results of the discriminant analysis can be found. The structure matrix displays the most important discriminating variables between the three categorical dependents (Poor, intermediate and rich countries). On the left side of the matrix all the independent variables can be found. On the right side of the table function 1 and function 2 are displayed. A summary of the discriminant analysis can be found in table 5. Function 1 describes the discriminating effect of the independent variables between poor countries and intermediate successful countries. Function 2 describes the discriminating effect of the independent variables between intermediate successful countries and rich

countries. This means that the structure matrix explains what independent variables are most important in the different economic developments stages.

Function

1 2

Factor 2 Availability of latest technology .320*

Factor 2 Fixed telephone lines .303*

Factor 2 Quality of electricity supply .249* Factor 2 Secondary education enrolment .248* Factor 2 Firm-level technology absorption .228*

Factor 1 Extent of marketing .263*

Factor 1 PCT patent applications .259*

Factor 1 Government procurement of advanced technology .250*

Factor 1 Reliability of police services .244*

Factor 1 Quality of the education system .208*

Table 5: Summary of the discriminant analysis with the 5 most important variables for function 1 and 2 (Source: Appendix 3).

To determine whether the functions 1 and 2 explain the group membership correctly, the Wilks’Lamba test can be conducted. Results of this test can be found in table 6. If the p-value is less than 0.05, we can conclude that the corresponding function explains the group

(28)

Wilks' Lambda Test of

Function(s)

Wilks'

Lambda Chi-square df Sig. 1 through 2 .003 411.000 154 .000

2 .145 136.908 76 .000

Table 6: Wilks’ Lambda test of both functions 1 and 2.

Now that we have established that both functions explain group membership well, it is time to focus on the independent variables (and the before mentioned components of the factor analysis) that score high on either function 1 or function 2. In the structure matrix each variable is accompanied with its component group as well so that we can see which

components (in addition to the variables) explain the differences between the three economic development stages. When looking at function 1 in the structure matrix, we can immediately see that variables grouped under General welfare (component 2) are prevailing. In the General welfare category we find variables related to health, macroeconomic development, education enrolment, technological readiness and electricity and telephone. This component closely resembles the established category “Basic requirements” by the WEF. Function 1 solely describes the discriminating effect between poor countries and intermediate successful countries. The results therefore definitely make sense, as poor countries should first try to improve their basic requirements in order to enter the next economic development stage. In this analysis general welfare measures a significant amount of these basic requirements and thus this component seems the best factor in explaining the difference between poor and intermediate successful countries. Other variables that seem to explain the difference between poor and intermediate successful countries are variables such as venture capital availability, Internet access in schools, quality of overall infrastructure, irregular payments, tertiary education enrolment and soundness of banks.

(29)

difference between poor and intermediate successful countries. In other words, the gap

between intermediate and rich countries consists of a larger amount of variables in which each variable accounts for a smaller portion of the gap (compared with the gap between poor and intermediate successful countries). In table 7 the first 5 variables of function 2 are listed. As can be seen, these variables all belong to the business enhancers category. The five listed factors in the table have the highest scores in the explanation of function 2. If we look back at the pillars constructed by the WEF, the countries with intermediate success are considered as efficiency-driven economies. If these economies want to bridge the gap towards the rich countries, they should invest in business sophistication and innovation (according to the WEF). In the discriminant analysis the first listed variable is related to business

sophistication, namely ‘Extent of marketing’. Variable 2 and 3 are related to innovation: Patent applications and government procurement of advanced technology. So far these three variables are in line with what we would expect in explaining differences between countries that have intermediate success and countries that are successful. The scores on these factors tell us that these variables are also important in explaining differences between poor countries and intermediate successful countries but the same variables are even more important in explaining differences between countries with intermediate success and rich countries. The fourth variable of the table seems to be a bit out of place. I would expect that the effect of the reliability of police services would be more significant between poor countries and countries with intermediate success (function 1 instead of function 2). However, this variable also scores relatively high on function 1. This means that this particular variable seems to be important in explaining differences between all the three development stages. The last variable in the table measures the overall quality of the education system of a country. This variable is important as well in explaining differences between intermediate and rich countries.

Besides the 5 listed variables in table 7, there are many more variables that seem to be

(30)

Factor 1 Extent of marketing .203 .263*

Factor 1 PCT patent applications .215 .259*

Factor 1 Government procurement of advanced technology .204 .250* Factor 1 Reliability of police services .185 .244* Factor 1 Quality of the education system .132 .208*

Table 7: The first five variables that explain function 2. Source: Structure matrix, Appendix 3. Next to the variables that explain the functions 1 and 2, a couple of variables also seem to be equally important in discriminating between the transition stages (from poor to intermediate and from intermediate to rich). Examples of such variables are listed in table 8. These variables can be considered as evenly important when countries transition from poor to intermediate and from intermediate to rich countries.

Factor 1 Production process sophistication .200* .192

Factor 1 Local supplier quantity .198* .198

Factor 1 State of cluster development .193* .179 Factor 1 Foreign market size index .172* .171

Factor 1 Extent of staff training .152* .147

Table 8: Variables with relatively the same discriminating effect in function 1 and 2. Source: Structure matrix, Appendix 3.

Now that the important underlying components and its corresponding variables are identified and the discriminating effects of these components and variables between different economic development stages are established by the discriminant analyses, it is time to move on to the final step in this analysis. In this final step another discriminant analysis will be conducted. Only now the focus will be on entrepreneurship indicators. In this research two indicators of entrepreneurship will be used. These indicators were established by the Global

(31)

of entrepreneurial activity, different variables and different goals are relevant. In necessity-driven entrepreneurship, the entrepreneurs are forced to become entrepreneurs because they do not see any other options and thus they are trying to survive. This is completely different in the case of Opportunity-driven entrepreneurship. In this case entrepreneurs see opportunities on which they act and try to make profit. According to the GEM necessity-driven

entrepreneurship is most common in poor countries, while opportunity-driven

entrepreneurship is prevailing in intermediate and rich countries. The structure matrix with the results of this analysis can be found in table 9. However, we can not draw any conclusions from the statistics presented in function two. According to the Wilks’ Lambda test the

function is insignificant (0.952 as visualized in table 9). This means that the variables are bad discriminating factors in the comparison of the two development stages. Therefore the only conclusion of this analysis is that NEC is prevalent in poor countries in comparison with intermediate and successful countries.

Structure Matrix Function 1 2 Necessity-driven Entrepreneurship (NEC) .996 * _.092 Opportunity-driven Entrepreneurship -.653 .758* *. Largest absolute correlation between each variable and any discriminant function

Table 9: Structure matrix of the discriminating effects of NEC and OPP on the different economic development stages.

The structure matrix indicates that Necessity-driven entrepreneurship is a significant variable in explaining the differences between poor countries and countries with intermediate success. This means that NEC is prevailing in countries that are poor. Opportunity driven

entrepreneurship is the prevailing discriminating variable between countries with intermediate welfare and rich countries. This is in line with what we would expect.

Wilks' Lambda Test of

Function(s)

Wilks'

Lambda Chi-square df Sig.

1 through 2 .568 36.431 4 .000

2 1.000 .004 1 .952

(32)

4.1.3. Discussion part 1: Factor and Discriminant analysis.

Now that all the statistical tests are done, I will briefly reiterate the most important findings. The results of the factor analysis can be seen in table 11. The table describes the 4 main components and their underlying themes. These components are important in the

determination of a country’s competitive position. The category business enhancers and its variables may act like the most important category as it explains almost half of the total variance. Entrepreneurial factors are most prevailing in the categories Business enhancers and Innovative and open mindset.

Business enhancers General welfare Innovative and open mindset Competition - Efficiency - Business environment - Innovation - Higher education - Transport infrastructure - Safety and security - Ethics - Financial market - Health - Macroeconomic development - Education enrolment - Technological readiness - Electricity and telephone - Mindset towards innovation - Openness of market - ICT - Air transportation - Foreign and domestic competition - Market size and labor market flexibility.

Table 11: Results of the factor analysis.

In the discriminant analysis it became clear that the differences between poor and intermediate successful countries is mostly due to variables related to General welfare, whereas the main differences of countries with intermediate success and rich countries is mostly due to business enhancers. This is according to what we would expect. The last discriminant analysis showed that there are different motivational backgrounds between development stages with regards to entrepreneurship. The structure matrix in table 9 shows that the differences between poor and intermediate successful countries can be explained by necessity-driven entrepreneurship, which is much higher in poor countries.

4.2. Country-by-country analysis: Uganda and South Africa

(33)

in stage 2 according to the WEF (so it is an intermediate successful country). Uganda is a country with very high levels of entrepreneurship, but is positioned in the lowest category of GDP per capita by the WEF (so Uganda is amongst the poor countries). For this reason, it is expected that this analysis will yield interesting results.

The structure of this final analysis will be determined by the outcomes of the statistical analyses. Derived from the analyses we found that ‘general welfare’ best explains the differences between poor and intermediate successful countries, whereas the component ‘business enhancers’ best explains the differences between intermediate successful and rich countries. Therefore in the country-by-country analysis we will compare Uganda and South Africa based on both components (General welfare and business enhancers). Even though general welfare should explain most of the differences between the two, I also want to consider ‘business enhancers’ as it explains such a significant part of the total variance. Within both components, I will obviously consider the results from the discriminant analysis as well. Therefore the first subsection will describe the development of Uganda based on the most prominent underlying themes of the ‘General welfare’ and the ‘Business enhancers’ categories as well as the most important variables that came out of the discriminant analysis. Important to notice is that relevant variables will be discussed quick and briefly. This means that we will quickly move from subjects like infant mortality to subjects such as ‘fixed telephone lines’. This is however necessary and desirable due to the limited space of this paper and for the sake of clarity as well. In this analysis we will rely on as many sources as possible, however it is difficult for some indicators to find relevant information over a period of 5 years. Therefore in some cases we have to rely on the information that is provided by the WEF. In the second subsection South Africa will be described based on the same components. The last subsection will consist of a comparison of findings between the two countries. 4.2.1. General welfare and Business enhancers of Uganda

(34)

other regions in the world (as of 2014). In comparison, life expectancy in the European Union is 81. So in terms of life expectancy, sub-Saharan Africa is lagging behind. Life expectancy in Uganda is, as of 2014, set on the age of 58.

Figure 5: Life expectancy of Sub-Saharan Africa and Uganda. Source: World Bank. This is slightly below the sub-Saharan average of 59. However, a positive note can be observed when looking at the development of the life expectancy in the timeframe of 2009-2014. This positive development can be seen in figure 5. From figure 5 we can derive that the life expectancy of Uganda has increased from 55 to 59 in the period between 2009 and 2014.

(35)

When looking at the infant mortality rate Uganda is performing slightly better (relatively speaking). Uganda is performing better than the average sub-Saharan African countries and is nearing the world average. Infant mortality rate declined from approximately 52 1000 live births in 2009 to 39 per 1000 live births in 2014.

With regards to health, it is clear to see that sub-Saharan Africa is lagging behind the rest of the world. The focus and attention of Uganda should definitely go to trying to increase the life expectancy, while it is very low compared to other countries and regions of the world.

Fortunately, life expectancy has already increased in the last couple of years. In terms of infant mortality Uganda is performing much better. Their infant mortality rate, however, lies far below the sub-Saharan average.

Figure 7: Quality of electricity supply. Source: GCI index, WEF.

Next important factor on the list is the quality of the electricity supply and the amount of fixed telephone lines. In the GCI index the quality of the electricity supply is rated a 2,46 on a scale from 1 to 7, which is very low. Uganda lies below the sub-Saharan average score of 2,9. “Average GCR” stands for the world average.

From 2010-2011 the quality of electricity supply slightly declined after which it went up again after 2013. Needless to say, Uganda should invest in a better supply of electricity as they are far behind world average and even the sub-Saharan Africa average. The same development can be observed when looking at other sources like the World Bank and the World Energy Outlook (WEO). Another similar situation can be identified when we look at the fixed telephone lines per 100 people of the total population.

(36)

Figure 8: Fixed telephone lines per 100 of total population. Source: GCI index, WEF.

In figure 8 we can see that again there is a large difference between sub-Saharan Africa and the world average. Uganda is yet again below the sub-Saharan Africa average. Data from the World Bank is less precise than the WEF and states that the ratio is 1/100 people that have a fixed telephone line in Uganda and was constant in the last 5 years. According to the WEF the same situation can be observed when looking at variables that measure ‘individuals that are using Internet’ and ‘fixed broadband subscriptions’. However, the average of individuals that are using Internet is slightly higher than the sub-Saharan average.

Figure 9: Availability of technology. Source: GCI index, WEF.

The last important factor that we will discuss is ‘the availability of technology’ and is visually represented in figure 9. According to the WEF the availability of technology has slightly declined since 2011-2012 in Uganda and in the rest of the world. It is clear to see that the availability of latest technology in sub-Saharan Africa as a whole has increased since

(37)

2010. They are slowly catching up with the rest of the world, which is obviously a positive development. However, a slight decline can be noticed as well when looking at the

development of Uganda (from 2011-2012 to 2013-2014).

Now that prominent variables of the General Welfare category are discussed, it is time to shift the focus towards the Business enhancers category. There are several important sub-factors when looking at the efficiency of a country. These factors consist of government efficiency, financial market efficiency and general market efficiency. According to the United Nations Global Development network (UNDP) the government of Uganda is already focusing on improving the overall/general efficiency of the country. The UNDP further states that they supported the government of Uganda with an amount of US$21.6m to be used for the

strengthening of democratic governance and reducing the poverty. According to the financial market report of 2013-2014, which was published by the Daily monitor (which is Uganda’s leading independent daily newspaper), Uganda has to “step up” with regards to the efficiency of the financial market. The report claims that the financial sector is still constrained by slow market growth and by challenges arising in the financing market infrastructure. Together these constraints result in a lack of availability of financial services, a lack of financing through the local capital market and a lack of access to credit. In order to improve this financial system Uganda needs to invest in revamping public infrastructure, fully leverage ICT uptake (which was already mentioned in the previous subsection) and further improve quality of public and private institutions. In this research we further use the Worldwide Governance Indicators (WGI) to determine the government efficiency in Uganda. The WGI uses the following indicators: voice and accountability, political stability and absence of violence/terrorism, regulatory quality, rule of law and control of corruption. I compared Uganda to sub-Saharan Africa averages as a whole and the results can be found in appendix 4. According to these indicators Uganda is performing below sub-Saharan Africa average in the fields of ‘political stability and absence of violence/terrorism’, ‘voice and accountability’ and ‘control of corruption’. Uganda is scoring above sub-Saharan Africa average in the field of ‘regulatory quality’ and ‘rule of law’. Three of the five indicators are scoring below average compared to sub-Saharan Africa.

(38)

can be seen in the figure, the gap between Uganda and the world average is very large. Uganda is even scoring below the average in sub- Saharan Africa as a whole.

Higher education is therefore another issue that definitely needs the attention of African governments as they are significantly lagging behind.

Figure 10: Aggregated score on higher education.. Source: GCI index, WEF. Last but not least, the situation of Uganda with regards to business environment and

innovation will be discussed. In order to obtain a clear and complete picture of the business environment in Uganda, it is in my opinion useful to look at the economic freedom of Uganda. In figure 11 we can observe that Uganda has seen a steady decline in economic freedom in the past few years. Uganda can now be identified as a ‘mostly unfree’ country.

(39)

According to heritage.org the decline is due to the slowed momentum of reforms within the country. World Bank adds that power outages are one of the biggest challenges in doing business in Uganda as these outages are costing many start-ups a lot of money. It must be said that this particular observation was made back in 2006, but in the previous subsection we have seen that the quality of electricity supply hasn’t improved since 2009. This means that the problem of outages would still be relevant as of today. Heritage.org further state that the business environment impedes the formation of new businesses. In other words, the business environment hinders entrepreneurship in the country. Despite of the business environment, Uganda seems to perform quite well in the field of innovation. “In 2014, Uganda was

classified as an innovation achiever for the second time by the Global Innovation Index (GII)” (Ecuru, Kawooya, 2015, p. 148). This means that the GII score of Uganda is significantly higher in relation to their GDP. Additionally, the GII of Uganda is significantly higher than other low-income countries. According to Ecuru and Kawooya the positive innovation performance is related to the “wider mix of socioeconomic policies, which over the years have remained stable and predictable” (Ecuru, Kawooya, 2015, p. 152). These policies resulted in the attraction of foreign direct investment and also led to increased favourable conditions to learning and innovation. Uganda is ranked 111th of the total of 141 evaluated countries in the GII. However, in order to maintain and improve their GII position the cost and ease of doing business should decrease significantly.

4.2.2. General welfare and Business enhancers of South-Africa

Now that the relevant factors and variables for Uganda are discussed, it is time to shift the focus towards South Africa. The structure of this subsection will be the same as the structure in the subsection ‘general welfare of Uganda’ and ‘business enhancers of Uganda’.

(40)

Figure 12: Life expectancy in South Africa. Source: World Bank.

When looking at infant mortality (per 1000 live births), South Africa is performing much better (figure 13). South Africa is very close to the world average and is clearly surpassing the sub-Saharan Africa average.

Figure 13: Infant mortality rate (per 1000 live births). Source: World Bank.