Reflecting on perceived failure of entrepreneurship development initiatives to help ignite economic development in Malawi

Introduction Small and medium enterprises (SMEs) are considered the backbone of the economy in Malawi since the early 1980s (Masten & Kandoole, 1997). This followed growing acknowledgement of the importance of entrepreneurship in job creation (Birch, 1979) and economic development (Baumol, 1968; Kirzner, 1973; Leff, 1979; Leibenstein, 1968; Schumpeter, 1934). After attaining independence from the British Government in 1964 and becoming a Republic in 1966, Malawi was faced with a multitude of challenges. Its population was poor; the largest proportion of which lived in rural areas and depended on subsistence farming. This context has not changed much to date. Due to Abstract

lack of developed industry and resources for extraction, it was difficult to attract multinational corporations (MNCs) which was one of the popular approaches for developing countries to attain foreign direct investment (FDI) in capital and technology transfer for economic development (Williams et al., 2017). The Government of Malawi (GOM) therefore engaged in state entrepreneurship (Freeman, 1982) where state owned or supported enterprises were established to engage in various social and economic activities. Furthermore, the government supported development of privately owned enterprises through establishment of institutions which promoted the growth of the SME sector (Masten & Kandoole, 1997). The institutions provided a range of support services which include; start-up finance, enterprise start-up and enterprise management training, technical skills and start-up equipment, promotion of quality and standards, promotion of investment, trade and marketing, and infrastructure development.
At the turn of the twenty-first century, the government continued with its commitment towards promotion of private enterprise and support of SME sector growth. The Malawi Growth and Development Strategy (MGDS) which is the medium-term National Development Plan (NDP) since 2006, incorporates various initiatives for attainment of economic growth through private sector development (GOM, 2006;2017a). Energy, tourism, mining and industry (manufacturing) are identified as important areas for private sector development (GOM, 2017a). There are aims to improve private sector competitiveness, increase number of enterprises accessing export markets and increase the number of enterprises which contribute positively towards economic growth. As such, the government continues with programmes for the improvement of the environment for business by developing infrastructure, supporting youth skills development, providing start-up finance and equipment, restructuring SME support institutions and developing Rural Growth Centres (RGCs) to be catalysts of sustained economic growth in rural areas (GOM, 2006;2017a). However, the problem is that despite undertaking entrepreneurship development initiatives over the decades, the country continues to experience high levels of poverty and unemployment, 51.5% and 20.4%, respectively, and poor macro-economic performance (GOM, 2017b;2017c). Gross Domestic Product (GDP) growth averaged 4.1% between 2010 and 2019 (Macroeconomic trends, 2021) against the targeted 7.2% (GOM, 2017c). There are similar observations in Ethiopia (Meressa, 2020) and South Africa (Cassim et al., 2014) that efforts to support development of entrepreneurship and improvements in the environment for business have not resulted in the expected economic development and creation of jobs. It is therefore perceived that entrepreneurship development initiatives are failing to help ignite the expected economic development and thereby support the growing acknowledgement that entrepreneurship does not bring economic growth in developing countries but in developed and transition countries (Lafuente et al. 2018;Zaki & Rashid, 2016;Stam & van Stel, 2009;van Stel et al., 2005).
Existence of binding constraints in the environments for business in developing countries is presumed the main reason for ineffectiveness of entrepreneurship. Studies (Agwu & Emeti, 2014;Olawale & Garwe, 2010) including Doing Business of the World Bank highlight various challenges to SME sector growth in developing countries which include lack of access to finance, poor infrastructural services and weak legal and regulatory institutions. As such, SME policies in developing countries and entrepreneurship development initiatives undertaken focus on improving the environments for business (Republic of Namibia, 2016;Republic of Uganda, 2015;GOM, 2012). This study attempts to explain the perceived failure of entrepreneurship development initiatives in developing countries, particularly in Malawi, from the perspective of entrepreneurial activities undertaken. This is important because entrepreneurship is a concept with multiple perspectives (Bula, 2012). Although entrepreneurship is perceived primarily as creation of a new organisation or enterprise (Gartner, 1988;Scarborough, 2013), its neoclassical economic understanding which is at the centre of Economic Development Theory is that entrepreneurship is perception of opportunities and carrying out innovations (Kirzner, 1973;Schumpeter, 1934). Schumpeter (1934) asserts that creating new products, new production methods, new sources of supply of raw materials, new markets and new organisation of any industry are the entrepreneurial activities which bring economic changes and growth. Therefore, the understanding of entrepreneurship would have its implications on what is undertaken as entrepreneurship and its effects on the economy while holding all other factors constant.
Currently, entrepreneurship development initiatives aim to tackle binding constraints in the environments for business in Malawi (GOM, 2012) for general improvement of the performance of the SME sector. By reflecting on the perceived failure of initiatives from the perspective of entrepreneurial activities undertaken, the study offers a useful alternative to guide initiatives to support productive entrepreneurial activities which can influence economic development. This would eventually guide initiatives in the development of a requisite environment for business and thereby improve overall effectiveness of initiatives undertaken.
The study was necessary because entrepreneurship is still considered a strategy for economic development in Malawi and other developing countries (United Nations Conference on Trade and Development UNCTAD, 2015;GOM, 2012). And despite higher entrepreneurial behaviour in developing countries (Global Entrepreneurship Monitor GEM, 2018;Naude, 2011), Sheriff et al. (2016 stress that little is known about entrepreneurial activities in Schumpeterian sense, which are undertaken by enterprises to be able to reflect on the effects of entrepreneurship on economic productivity and development. Not much is known about innovations carried out by enterprises in developing countries especially in Sub Saharan Africa. Therefore, the study attempts to fill these knowledge gaps before we can inform on the reasons entrepreneurship does or does not contribute to economic growth in developing countries. Firstly, the study highlights knowledge about entrepreneurship in Malawi which guides entrepreneurship development initiatives undertaken and secondly, it measures innovations and their values carried out by enterprises in the country in order to understand productivity of entrepreneurship. Thirdly, it analyses the relative contribution of innovations towards aggregate entrepreneurship value in order to understand the type of innovations which contribute the most towards economic productivity; and lastly, it analyses the differences in value of innovations created by different types of enterprises in order to determine productive enterprises in the country. Knowledge generated allows the study to reflect on the effect of entrepreneurship development initiatives undertaken in the country. Therefore, the understanding of entrepreneurship in Malawi, entrepreneurial activities undertaken and their productivity, and the reasons for perceived failure of entrepreneurship development initiatives in the country become the study's contribution to knowledge about entrepreneurship undertaken in a selected sub-Saharan African country.
In the next sections literature review is presented on the concept of entrepreneurship and the hypotheses tested to determine a productive type of enterprise followed by a review of entrepreneurship development initiatives undertaken in developing and developed countries. The research methodology is then presented. Thereafter, results are presented and discussed with the focus on implications of knowledge about entrepreneurship, innovations carried out and productivity of types of enterprises in Malawi visa-vis entrepreneurship development initiatives undertaken in the country.

The concept of entrepreneurship
The meaning of entrepreneurship is difficult to pin down because scholarly fields provide different perspectives and scholars present theories focusing on different aspects of seemingly the same concept. Psychology studies of entrepreneurship focus on the traits of an entrepreneur or traits that underlie undertaking of entrepreneurial behaviour (McClelland, 1961), whereas economics studies view entrepreneurship as activity which underlies market dynamism. The activity which moves markets towards the state of equilibrium (Kirzner, 1973;Walras, 1954) or creates new disequilibrium altogether (Schumpeter, 1934).
Cantillon (1755) is considered the first economist to conceptualise entrepreneurship (Kuratko & Hodgetts, 2007). He theorises entrepreneurship as an activity undertaken to identify and exploit market discrepancies. The individual he calls an entrepreneur procures raw materials at certain prices in order to gradually rework them up for resell at uncertain prices for a profit. This individual is essentially self-employed and undertakes risks on capital employed with regard to uncertainty of future resell prices. Another prominent economist is Say (1816) whose theory of entrepreneurship focuses on the essence of entrepreneurial activities, the creation of goods and services. He defines entrepreneurship as coordination of factors of production to produce goods and services. But interpretation of the word 'coordination' leaves scholars with two viewpoints of the concept. First, coordination translates to management or uniting of the factors in the production process which Leibenstein (1968) refers to as routine entrepreneurship. Secondly, coordination refers to bringing together the factors of production to set up a new enterprise which produces the goods and services, and this is a popular perception of entrepreneurship (Dollinger, 2008;Gartner, 1988). Schumpeter (1934) is accredited with what Henrekson and Sanandaji (2014) consider the modern understanding of entrepreneurship. He asserts that entrepreneurship is central to economic development. Schumpeter (1934) theorises entrepreneurship as carrying out innovations in the form of new products, new production methods, new sources of supply of raw materials, new markets and new organisation of any industry. His theory broadens the outputs of entrepreneurial behaviour from creating goods and services or a new enterprise as in Say's (1816) theory to the other forms of innovation. His theory further broadens the context of the entrepreneur from an individual to whoever undertakes that special function even if it is other existing entity.
The prevailing meanings of entrepreneurship are derived from the theories put forward by Cantillon (1755), Say (1816) and Schumpeter (1934). Adam Smith and David Ricardo (Kirby, 2003) perceive entrepreneurship through Cantillon's (1755) theory but as risking capital used to undertake the entrepreneurial venture. This is where taking risks is derived as a key aspect in definitions of entrepreneurship (Hisrich et al., 2009;Scarborough, 2013). However, Say (1816) distinguished the functions of the entrepreneur and the capitalist based on how each is compensated. The capitalist gets interest, whereas the entrepreneur gets the profit after all other expenses are met. That means risk taking on capital is not a salient point when defining entrepreneurship even though capital is a key resource for entrepreneurship. Knight (1921) also isolates the meaning of entrepreneurship from Cantillon's (1755) theory as undertaking activities with uncertain outcomes. He furthermore distinguishes risk taking from bearing uncertainty in that the supplier of capital bears the risk but the entrepreneur bears the uncertainty of the entrepreneurial activity undertaken. Therefore, entrepreneurship is not about taking risks, but bearing uncertainty although the former is often highlighted in definitions of entrepreneurship. If an individual invests his/her own capital then he/she doubles as a capitalist and bears the risk as well as the uncertainty. Kirzner (1973) postulates entrepreneurship as perceiving and exploiting opportunities. His theory enters the space created by Schumpeter's (1934) theory. Schumpeterian entrepreneurship disrupts markets and thereby perceiving and exploiting the opportunities brought by the changes is entrepreneurship too according to Kirzner's (1973) theory and as explained by Drucker (1985). This involves imitation or transfer of knowledge therefore it is commonly referred to as imitative entrepreneurship (Schmitz, 1989). Stam (2013) observes that disruptive entrepreneurship in Schumpeter's (1934) sense is rare and far apart, but imitative entrepreneurship is common and would therefore be very useful for economic growth in developing countries (Schmitz, 1989).
A closer look at definitions of entrepreneurship highlighted by Mwatsika et al. (2018) shows that the concept is the same only that scholars try to pick on the point they feel it provides the best description of what entrepreneurship means. Entrepreneurship can be summarised as behaviour undertaken by an individual or other established entity which involves activities with uncertain outcomes. The entrepreneurial activities include perceiving entrepreneurship opportunities and undertaking new combinations of resources to create innovations as outputs of successful entrepreneurial behaviour. Schumpeter (1934) calls this enterprise and that means a new enterprise introduces innovations (new products, new methods of production, new sources of supply of raw materials, new markets and/or new ways of organisation). It is noteworthy that entrepreneurship opportunities are all about carrying out innovations (Shane & Venkataraman, 2000). There are therefore various standpoints for viewing entrepreneurship. Some perspectives capture the essence of entrepreneurship and others such as self-employment or a small and medium enterprise (Acs & Virgill, 2009) do not (Henrekson & Sanandaji, 2014;Hurst & Pugsley, 2011;Shane, 2009). However, perceptions of entrepreneurship have implications on entrepreneurship development initiatives undertaken.
Entrepreneurship Intention Models; the Theory of Planned Behaviour (Ajzen, 1991) and Shapero's model of Entrepreneurial Event (Shapero, 1982), present entrepreneurship as behaviour which is influenced by psychological and sociological factors that affect the individual's attitudes, perceived feasibility, propensity to act and entrepreneurship intentions. Perceiving entrepreneurship as behaviour first directs attention towards the study of traits and important characteristics of successful entrepreneurs. That informs entrepreneurship development initiatives especially education and training on traits and characteristics to develop in order to enhance the supply of productive entrepreneurs in the economy. However, there are many traits of successful entrepreneurs highlighted in literature (Amiri & Marimaei, 2002) that not all of them have been empirically tested on their usefulness.
Secondly, perceiving entrepreneurship as behaviour prompts to seek understanding of the activities undertaken which constitute entrepreneurship. It is in this vein that entrepreneurship is defined as a process (Kuratko & Hodgetts, 2007;Stevenson & Jarillo, 1990) meaning that entrepreneurial behaviour involves various activities which are undertaken in a particular order. Shane and Venkataraman (2000) highlight three phases of the entrepreneurship process which include identification of entrepreneurship opportunity, evaluation and exploitation. The exploitation phase involves mobilising resources, organising and managing the entrepreneurial activity. Timmons (1989) explains that the entrepreneurship process starts with an individual who using his/her knowledge, skills and expertise creates the entrepreneurship opportunity and then mobilises the resources for its exploitation. Timmons (1989) affirms that being able to put together a capable team is a key factor for success. On the other hand, Moore (1986) and Ward (2005) present that the entrepreneurship process starts with existence of entrepreneurship opportunity which arises from market discrepancies. An individual with appropriate skills and cognitive abilities recognises or discovers the entrepreneurship opportunity and then mobilises resources and organises to exploit the opportunity. Obviously, Timmons (1989) perceives entrepreneurship opportunity as subjective whereas Ward (2005) and Moore (1986) perceive entrepreneurship opportunity as objective arising from the change brought into the markets by disruptive innovation. Nonetheless, perceiving entrepreneurship as a process highlights entrepreneurship opportunity as a key construct (Shane & Venkataraman, 2000). Understanding the entrepreneurship process would inform entrepreneurship development initiatives on requisite skills to engage with the process and on appropriate support to successfully undertake each phase, for example availability and accessibility to key resources for successful exploitation of opportunities. Provision of appropriate support is broader in scope such that it covers the requirement to improve the environment for business touching on varying aspects relating to entrepreneurship or undertaking business in general. Theories such as 'Jack of all trades' (Lazear, 2005) portray that an individual would require a range of skills to undertake the entrepreneurship process successfully especially where sole proprietorship is concerned. The O Ring theory (Fabel, 2001) presents the requirement for individuals with varying but complimenting skills to successfully undertake entrepreneurship and this is more applicable in corporate entrepreneurship (Morris et al. 2008).
Perceiving entrepreneurship as carrying out innovations places more emphasis on skills development in areas of expertise and undertaking of Research and Development (R&D) in various fields for the development of innovations. Countries which invest heavily in R&D create more new knowledge and entrepreneurship opportunities which spill over in the economy with multiplier effect on entrepreneurial behaviour (Stam, 2013). Although Kukoc and Regan (n.d.) note that huge investments in R&D in some European countries did not translate into the expected economic development. Which means R&D expenditure may not directly result in productive entrepreneurship. Nonetheless, for countries which cannot afford huge investments in R&D, entrepreneurship development initiatives can focus on imitation and transfer of knowledge to local industries.
Perceiving entrepreneurship as creating new enterprise in literal sense or being selfemployed puts much emphasis on promoting and supporting start-ups, and not all of them could be productive. Therefore, entrepreneurship development initiatives need appropriate guidance to support productive entrepreneurship which can bring economic growth. Entrepreneurship is undertaken through particular organisation or enterprise and in order to identify types of enterprises which undertake productive entrepreneurship, six null hypotheses (H o ) were tested. First, an entrepreneur assumes many forms. He may be a private business man (sole proprietor) or a Limited Liability Enterprise . But in Malawi, entrepreneurship development initiatives focus on supporting sole proprietorships. Therefore, the study tested the first null hypothesis (H o 1) to find out if there are any differences in productivity of entrepreneurial activities undertaken between the two categories as follows: H o 1 There are no differences between sole proprietorship enterprises and limited liability enterprises in the value of innovations carried out.
Secondly, enterprises can be distinguished based on motivation as either necessitymotivated or opportunity-motivated enterprises (Bell, 2013). Olafsen and Cook (2016) observe that necessity motivated enterprises are more prevalent in developing countries while opportunity-motivated enterprises are more prevalent in developed countries. Noting the differences in effects of entrepreneurship on economic growth between developing and developed countries, the study tested the second null hypothesis (H o 2), to determine if there were differences in productivity of entrepreneurship, as follows: H o 2 There are no differences between necessity-motivated enterprises and opportunity-motivated enterprises in the value of innovations carried out.
Third, entrepreneurial behaviour is commonly associated with new enterprises even though old enterprises can undertake entrepreneurial activities too. In Malawi, entrepreneurship development initiatives focus on supporting new enterprises and not many programmes are undertaken to support entrepreneurial behaviour in old enterprises. Therefore, the study tested the third null hypothesis (H o 3) to determine if there are differences in productivity of entrepreneurial activities between new and old enterprises as follows: H o 3 There are no differences between new enterprises and established enterprises in the value of innovations carried out.
Entrepreneurship is predominantly perceived as profit oriented, but there is a growing dimension where entrepreneurship is undertaken essentially to provide social benefits, termed 'social entrepreneurship' (Gawell, 2013). The study therefore tested the fourth null hypothesis (H o 4), to determine if there are differences in productivity of entrepreneurial activities undertaken for profit or provision of social benefits as follows: Page 8 of 24 Mwatsika J Innov Entrep (2021) 10:40 H o 4 There are no differences between profit-oriented enterprises and non-profit-oriented enterprises in the value of innovations carried out.
Furthermore, size of an enterprise informs perceptions of entrepreneurship. SMEs are perceived synonymously with entrepreneurship (Acs & Virgill, 2009) even though an enterprise of any size can undertake entrepreneurial activities (Drucker, 1985). With entrepreneurship development initiatives predominantly focused on supporting SMEs in Malawi, the study tested the fifth null hypothesis (H o 5), to determine if there are differences in productivity of entrepreneurial activities undertaken by enterprises of different sizes, as follows: H o 5 There are no differences between micro-to-small and medium-to-large enterprises in the value of innovations carried out.
Lastly, growth orientation helps to categorise enterprises as either lifestyle (subsistence) or high growth (Burns, 2016). While there are differences in productivity of entrepreneurship between developed and developing countries (Stam & van Stel, 2009), Olafsen and Cook (2016) observe that lifestyle (subsistence) enterprises are more prevalent in developing countries. The study therefore tested the sixth null hypothesis (H o 6) to determine if there are differences in productivity between the growth-oriented and subsistence-oriented enterprises in Malawi as follows: H o 6 There are no differences between subsistence-oriented enterprises and growthoriented enterprises in the value of innovations carried out.
Determining the level of innovation in enterprises in Malawi and the differences in value of innovations carried out in types of enterprises would allow the study to reflect on the focus of entrepreneurship development initiatives undertaken and their effectiveness while holding constant the influence of the environment for business.

Entrepreneurship development initiatives
Since studies (Birch, 1979;Chen, 2014;Decker et al., 2014;Lee & Xin, 2015;Schumpeter, 1934) support that entrepreneurship is important for economic growth and job creation, entrepreneurship is one of the popular strategies for economic development in both developing and developed Organisation for Economic Cooperation and Development (OECD) countries (OECD, 2019;2004;UNCTAD, 2015). Countries undertake initiatives to develop entrepreneurship. We define entrepreneurship development as a process of enhancing entrepreneurial behaviour for the creation of innovations which ignite economic development. Therefore, entrepreneurship development initiatives involve strategies and activities undertaken to enhance and support productive entrepreneurship which contributes positively towards economic growth and job creation. Entrepreneurship development initiatives are undertaken by key stakeholders of entrepreneurship (Mwatsika, 2018) and usually focus on improving areas within the environment for business which are identified as constraining entrepreneurial behaviour and carrying out of innovations.
The environment for business comprises factors which affect failure or continued existence of an organisation (Smit et al. 2007). Mwatsika (2018) categorises the environment for business into macro-and micro-environmental factors. Macro-environmental factors include: political, legal and regulatory factors; economical factors; societal factors; natural environmental factors and technological factors whereas micro-environmental factors include; industry and market forces (competitors, customers, suppliers, substitutes, private and public institutions, and the communities) (Porter, 1998). An organisation's internal environment which comprises strategies, structure, systems, shared values, staff skills and style of management (Gokdeniz et al., 2017) is also very important. Mwatsika (2018) provides a review on the influence both macro and microenvironmental factors have on entrepreneurship where studies support that each of the factors affect productivity of entrepreneurship in various ways. Studies undertaken in different developing countries highlight poor performance of factors in the environment for business which negatively affect entrepreneurship and SME sector growth. The challenges include; poor macro-economic performance (high inflation, high interest rates, high tax rates, foreign exchange rates), unfavourable societal factors (crime, corruption, poor business ethics), poor strategic infrastructure services (electricity, water, sewage, transport, telecommunication), lack of access to modern technologies for production and information and communication, and lack of availability or accessibility to finance which is regarded as a major obstacle to SME sector growth (World Bank Group, 2020; Agwu & Emeti, 2014;GOM, 2012;Olawale & Garwe, 2010). These challenges constrain entrepreneurship, increase costs of production, and make SMEs uncompetitive and unable to access international markets. As such, developing countries have been guided by studies such as Doing Business of the World Bank Group to determine specific areas for improvement. A quick look at SME policies for Malawi, Namibia, Rwanda, Uganda and Zambia shows that the policies target to enhance SME sector growth and focus in similar areas which include improving; (1) legal and regulatory frameworks to facilitate ease of starting up an enterprise; (2) the physical infrastructure for business; (3) business development services; (4) access to finance; (5) access to markets, and (6) enhancing institutions which support SME development (GOM, 2012;Republic of Namibia, 2016;Republic of Rwanda, 2010;Republic of Uganda, 2015;Republic of Zambia, 2008). Strategies for SME sector growth include entrepreneurship development because carrying out innovations improves enterprise productivity, competitiveness and growth. Therefore, entrepreneurship development initiatives also aim to inculcate a culture of entrepreneurship through education and training in universities and technical colleges targeting the youth who are the segment of the population experiencing high levels of unemployment (Republic of Namibia, 2016; Republic of Rwanda, 2010; Republic of Zambia, 2008).
Entrepreneurship development initiatives in Malawi have focused on provision of finance, education and training, and improvement of the environment for business for decades. Public but politically driven programmes offer start-up capital for youth and women economic empowerment whereas Micro Finance Institutions (MFIs) and Savings and Credit Cooperatives dominate the provision of micro-credit finance. SME lending in commercial banks remains underdeveloped with concerns on both sides about risks of lending to SMEs on the one hand and high interest rates and unfavourable terms of borrowing on the other. Capital markets remain undeveloped in Malawi Page 10 of 24 Mwatsika J Innov Entrep (2021) 10:40 (GOM, 2017a). But there are programmes to facilitate trade and investment as well as quality and standards in the country. Universities, technical colleges and the SME development institution offer entrepreneurship education and training in their curricula while the Technical and Entrepreneurial Vocational Education and Training Authority (TEVETA) supports technical skills development and provision of start-up equipment. Malawi has high rates of unemployment (GOM, 2017c), therefore the aim of entrepreneurship education and training is to enable graduates to start and run their own enterprises. While the ministries of trade and industry focus on formulation and facilitation of policy implementation, the Government overall focuses on development of strategic infrastructure (transport, telecommunication and energy) and legal and regulatory frameworks. Furthermore, the Government through the ministry of local government and rural development is constructing rural growth centres to be catalysts of economic development in rural areas where some of the initiatives involve construction of agroprocessing facilities, roads, markets and bus depots.
Overall, it is observed that SME policies in developed OECD countries focus in similar areas; improving institutional and regulatory frameworks, providing supportive environment and scaling up SME capacity (OECD, 2019) just like in developing countries. However, the main difference is that particular entrepreneurship development initiatives undertaken in developed countries focus on enhancing innovative start-ups and supporting high growth SMEs (HGSMEs). For instance, there are public support packages which aim to enhance provision and access to risk finance for undertaking R&D in SMEs, whereas in developing countries the focus is on providing micro-credit to support youth and women enterprising initiatives which alleviate poverty. SMEs continue to lack access to finance for developing innovations (Olawale & Garwe, 2010). Furthermore, developed countries focus on developing digital networks, large research and computing infrastructure and platforms for technology transfer (OECD, 2019) whereas the majority of infrastructure development for business in developing countries focus on improving structural services (electricity, water, transport) and constructing markets or industrial parks to provide SMEs with appropriate infrastructure for production and value addition (Republic of Namibia, 2016; Republic of Rwanda, 2010; Republic of Uganda, 2015; Republic of Zambia, 2008). While developing countries are struggling to improve the use of modern technologies in SMEs' production, developed countries are advancing absorption of high end technologies within specific business ecosystems (OECD, 2019).
As much as the differences in entrepreneurship development initiatives demonstrate disparities in economic development between developed and developing countries, there also seems to be differences in knowledge about entrepreneurship which guides initiatives undertaken. In developed countries entrepreneurship is progressively been perceived as undertaking innovations which is considered the modern perception of the concept (Henrekson & Sanandaji, 2014;Davidsson & Henrekson, 2002). However, although the SME policies in developing countries provide guiding definitions of the SME sector, they are silent on the definition or understanding of what entrepreneurship really means. Therefore the focus of entrepreneurship development initiatives on innovative start-ups and HGSMEs in developed countries is based on growing acknowledgement that not all SMEs are entrepreneurial (Shane, 2009) and that HGS-MEs which contribute significantly towards new jobs and economic productivity are the Page 11 of 24 Mwatsika J Innov Entrep (2021) 10:40 manifestation of carrying out innovations (Audretsch, 2012;Decker et al., 2014;Stam & van Stel, 2009). Since high growth is a phase in an enterprise's life cycle (Churchill & Lewis, 1983), the challenge of entrepreneurship development initiatives becomes how to identify SMEs with high growth potential for attention. Nonetheless, differences in the focus of entrepreneurship development initiatives towards HGSMEs would result in productivity differences and the contribution entrepreneurship makes towards economic growth and development between developed and developing countries. Studies by Zaki and Rashid (2016), Stam and van Stel (2009) and van Stel et al. (2005) found that entrepreneurship does not contribute towards economic growth and development in developing countries but in developed and transition countries despite other studies (GEM, 2018;Naude, 2011) showing that entrepreneurial behaviour is higher in developing countries than developed countries. Further observations by Meressa (2020) and Cassim et al. (2014) in Ethiopia and South Africa, respectively, which also reflect the context of Malawi, support the notion that indeed entrepreneurship does not bring economic development in developing countries despite entrepreneurship development initiatives undertaken for decades. It is against this background that the study sought to reflect on the reasons entrepreneurship development initiatives in Malawi have failed to help ignite economic development. Schumpeter's (1934) theory guided the understanding of entrepreneurship and its measurement in the study. New products, new methods of production, new markets and new enterprises were the indicators of entrepreneurship undertaken and their values were the measure of its productivity. A new product was defined as an improved, imitated or new brand (Kotler & Armstrong, 2012) and the exchange value realised when sold was considered its value (Bowman & Ambrosini, 2003). Similarly, a new method of production was defined as any improved, replicated or one newly developed by an enterprise and its total investment value was considered its value. On a new market, Schumpeter's (1934) definition was applied and the exchange values of products sold in the new market represented its value (Bowman & Ambrosini, 2003). A new enterprise was defined as the one with a payroll above zero which did not exist in the previous year (Godin, Clemens & Veldhuis, 2008). The present worth of the enterprise represented its value (Miciula et al., 2020). The aggregate value of innovations represented the entrepreneurship value, which contributes towards economic productivity and growth.

Methodology
A cross-sectional survey was undertaken in three cities and three rural growth centres in Malawi where 337 enterprises participated. The study first assessed knowledge about entrepreneurship by asking respondents to define or explain entrepreneurship from the top of their mind (Lee, 2011). Secondly, innovations carried out by enterprises studied were recorded together with the values realised. Analyses of the data collected involved descriptive statistics of the study sample, content analyses of the definitions of entrepreneurship and comparison of mean values of innovations carried out by enterprises. Furthermore, non-parametric tests (Chi-square test, 2 independent samples tests-Mann-Whitney U tests) were conducted to test the study hypotheses. To compare the relative contribution of innovations towards entrepreneurship value, dummy variables were created for each innovation where a dummy value of 1 represented that an Page 12 of 24 Mwatsika J Innov Entrep (2021) 10:40 enterprise had undertaken a particular innovation and a dummy value of 0 represented that an enterprise did not undertake a particular innovation. A multiple linear regression was then used to determine the relative contribution of innovations towards economic productivity in the country. The multiple linear regression model was specified as: where LNY i is the outcome variable taking a natural logarithmic form of the sum of values of innovations. D j (j = 1,…,n) are binary dummy variables for innovations undertaken which include: D 1 = new products dummy (NPSD), D 2 = new production methods dummy (NPMD), D 3 = new markets dummy (NMKD) and D 4 = new enterprises dummy (ENTD). X j (j = 1,…,n) are a set of covariates, the independent variables of the study sample which include enterprise age and age of respondent and µ is the error term. For measurement and interpretation, the outcome variable and enterprise age are ratios, whereas dummy variables and age of respondent are in nominal scale. The influence of entrepreneurship development initiatives was reflected upon the prevalence of productive enterprises which carried out innovations, overall contribution of innovations towards economic productivity and the knowledge about entrepreneurship which guides entrepreneurship development initiatives undertaken in the country.

Results and discussion
The results of the statistical descriptive analysis of the study sample are presented in Table 1. It important to note the higher prevalence of enterprises which are sole proprietorship (94.4%), above one year old (91.1%), necessity motivated (81%), profit making (96.4%), subsistence oriented (97.9%) and in the micro-and small size category (96.7%). The low prevalence of opportunity-motivated (19%), growth-oriented (2.1%) and limited liability enterprises (5.6%) overall highlight the lack of medium-tolarge enterprises in industries of Malawi.
On the first objective of the study, definitions of entrepreneurship in enterprises were analysed in order to reflect on the knowledge which guides entrepreneurship initiatives in the country. Respondents were requested to define or explain entrepreneurship from the top of their mind (Lee, 2011). The analyses involved summarising the definitions, condensing descriptive words and uniting them into themes which describe what entrepreneurship means to the respondents. Results of the Content analyses of the definitions of entrepreneurship are presented in Table 2. The results show that respondents perceive entrepreneurship as creating jobs in the economy, starting and managing one's own business and being self-employed.
A larger proportion of the respondents (36.2%) do not know the meaning of entrepreneurship and these are mostly those with low education (primary and secondary school education). Only 0.9% of the respondents, within other (4.2%), perceive entrepreneurship as carrying out innovations. This demonstrates that entrepreneurship in the country is predominantly perceived through classical economic theories (Cantillon, 1755;Say, 1816) as starting and managing an enterprise, being self-employed.
On study objective number two, mean values of innovations carried out by enterprises were analysed to determine: (1) innovations carried out; (2) values of innovations carried out and (3) productivity of entrepreneurial activities undertaken. Details of new products, new production methods, new markets and new enterprises were collected together with values realised. The results of the comparison of mean values of innovations carried out by enterprises are presented in Table 3.  Page 14 of 24 Mwatsika J Innov Entrep (2021) 10:40 Of the enterprises studied, 25% created new products, 2% created new production methods, 1.2% entered new markets and 8.9% were new enterprises. As observed in Table 3, the mean values of the innovations carried out were low. Overall, very few of the enterprises studied created innovations. Creating new markets and new methods of production are the least undertaken entrepreneurial activities demonstrating the lack of investment in new markets and production facilities (processes, equipment and technologies) among enterprises in Malawi. The results show that when entrepreneurship is perceived through neoclassical economic theories (Kirzner, 1973;Schumpeter, 1934) as identifying opportunities and carrying out innovations, then it is very low among enterprises studied.
In further statistical analyses, the bivariate test of the relationship between aggregate entrepreneurship value created and the values of innovations and the relationship among the categories of innovations showed that entrepreneurship value was significantly correlated to values of all innovations (p = 0.001). However, value of new products highly correlated with the value of new markets (r = 0.912, p = 0.001), whereas the values of all other innovations were not highly correlated with each other (r < 0.70).
A multiple linear regression in semi-logarithmic form was undertaken on study objective number three to analyse the relative contribution of innovations to the aggregate entrepreneurship value created. This was undertaken to understand how the categories of innovations (new products, new production methods, new markets and new enterprises) contributed towards economic productivity. In the regression, a dummy value of 1 represented that an enterprise had undertaken a particular innovation and a dummy value of 0 represented that an enterprise did not undertake that innovation. The results of the multiple linear regression are presented in Table 4.
The semi-elasticity due to the change in the dummy regressors was derived by following the method suggested by Halvorsen and Palmquist (Gujarati & Porter, 2009) of exponentiating the regression coefficient through a natural base transformation followed by the subtraction of 1 to obtain the relative change and finally multiplying the difference by 100 to obtain the semi-elasticity. The results therefore show that the median value of an enterprise which created new products (D 1 = 1) is 175% higher than that of the enterprise which did not create new products (D 1 = 0) holding all other factors constant. The median value of an enterprise which entered new markets (D 3 = 1) is 300% higher than that of the enterprise which did not enter new markets (D 3 = 0) while all other factors remain the same and the median value of a new enterprise (D 4 = 1) is 186% higher than that of an old enterprise (D 4 = 0) while all other factors remain the same too. The results are statistically significant at 5% level of significance. Lastly, the median value of an enterprise which created new methods of production (D 2 = 1) is 156% higher than that of the enterprise which did not create new methods of production (D 2 = 0) and the result is statistically significant at 10% level of significance.
The interpretation of the regression analyses, Table 4, is that enterprises which carry out innovations create more value and therefore contribute more towards economic production than enterprises which do not carry out innovations. The results show that entering new markets contributes the most entrepreneurship value followed by establishing new enterprises, new products and new production methods. Therefore in order to help ignite economic growth, entrepreneurship development initiatives would be expected to focus on assisting SMEs to carry out innovations; enter new markets, establish productive enterprises, create new products and invest in new methods of production (processes, equipment and technology).
In order to determine the type of new enterprise which would create more value, 2 independent samples tests (Mann-Whitney U tests) were undertaken to test the null hypotheses on the differences between categories of enterprises in the value of innovations carried out. The first null hypothesis was that there are no differences between sole proprietorship and limited liability enterprises in the value of innovations carried out. The Mann-Whitney U test results, Table 5, show that there are statistically significant differences between limited liability enterprises and sole proprietorships in the values  of new markets (NMKV) (p = 0.007), new methods of production (NPMV) (p = 0.001) and new enterprises (ENTV) (p = 0.006). There are no statistically significant differences between the categories in the value of new products (NPSV) (p = 0.694). The results reject the first null hypothesis of the study. There are differences between limited liability enterprises and sole proprietorships in the value of innovations carried out. Since the results show that limited liability enterprises create significant value of innovations, they become the category which requires the attention of entrepreneurship development initiatives. The second null hypothesis was that there are no differences between necessitymotivated and opportunity-motivated enterprises in the value of innovations carried out. The Mann-Whitney U test results, Table 6, show that there are statistically significant differences between opportunity-motivated enterprises and necessity-motivated enterprises in the values of new markets (p = 0.001), new production methods (p = 0.001) and new enterprises (p = 0.002) but there are no statistically significant differences in the value of new products (p = 0.592) between the categories.
The results reject the second null hypothesis, there are differences between opportunity-motivated enterprises and necessity-motivated enterprises in the value of innovations carried out. Opportunity-motivated enterprises create significant values of innovations and therefore necessary for attention in the country.
The third null hypothesis was that there are no differences between new and old enterprises in the value of innovations carried out. The Mann-Whitney test results, Table 7, show that there are no statistically significant differences between new enterprises and established enterprises in the values of new products (p = 0.388), new production methods (p = 0.547), new markets (p = 0.202) and enterprises values (p = 0.94).
Page 17 of 24 Mwatsika J Innov Entrep (2021) 10:40 The results fail to reject the third null hypothesis. There are no differences between new and established enterprises in the value of innovations carried out. That means both new and old enterprises can undertake productive entrepreneurial activities in the economy which require support of entrepreneurship development initiatives.
The fourth null hypothesis was that there are no differences between profit-oriented and non-profit-making enterprises in the value of innovations carried out. The Mann-Whitney U test results, Table 8, show that there are statistically significant differences between profit-oriented enterprises and non-profit-oriented enterprises in the value of new production methods only (p = 0.001). There are no statistically significant differences between the categories in the values of new markets (p = 0.608), new products (p = 0.668) and new enterprises (p = 0.134). The results overall fail to reject fourth null hypothesis. There are no differences between profit-oriented enterprises and non-profitoriented enterprises in the value of innovations carried out. Therefore, both profitoriented and non-profit-oriented enterprises can carry out innovations for economic growth and entrepreneurship development initiatives would need to support both categories.
The fifth null hypothesis was that there are no differences between micro-to-small enterprises and medium-to-large enterprises in the value of innovations carried out. The Mann-Whitney U test results, Table 9, show that there are statistically significant differences between medium-to-large enterprises and micro-to-small enterprises in the values of all innovations; new products (p = 0.035), new production methods (p = 0.001), new markets (p = 0.002) and new enterprises (p = 0.001). The results reject the fifth null hypothesis. There are differences between medium-to-large enterprises  Table 9 Differences between medium-to-large enterprises and micro-to-small, enterprises in the value of innovations carried out Page 18 of 24 Mwatsika J Innov Entrep (2021) 10:40 and micro-to-small enterprises in the value of innovations carried out and that make medium-to-large enterprises more desirable for economic development.
The last null hypothesis was that there are no differences between subsistence-oriented and growth-oriented enterprises in the value of innovations carried out. The Mann-Whitney U test results, Table 10, show that there are statistically significant differences between growth-oriented enterprises and subsistence-oriented enterprises in the values of new products (p = 0.001), new markets (p = 0.021) and new enterprises (p = 0.047), but there are no statistically significant differences in the value of new production methods (p = 0.163). The results reject the sixth null hypothesis. There are differences between growth-oriented enterprises and subsistence-oriented enterprises in value of innovations carried out which make growth-oriented enterprises more desirable since they create significant values of innovations.
The 2 independent samples tests (Mann-Whitney U tests) overall provide a description of a productive enterprise as the one which is opportunity-motivated, growth-oriented and limited liability in the medium-to-large size category. It is necessary to note that it is enterprises which are less prevalent in the sample which are found to be more productive than the type of enterprises which are highly prevalent. The objective of the enterprise (profit-oriented or non-profit-oriented) and age of the enterprise (new or old) are factors which do not necessarily affect undertaking of innovations according to the findings.
Overall, the study finds that: (1) entrepreneurship is perceived as starting and managing one's own business, being self employed; (2) few enterprises carry out innovations and of low value; (3) carrying out innovations contribute more towards economic production and entering new markets is the most productive entrepreneurial activity; and (4) opportunity-motivated, growth-oriented limited liability enterprises in the medium-to-large size category are the productive type of enterprise, but there are very few in Malawi. The results are hereby used to reflect on the perceived failure of entrepreneurship development initiatives to help ignite economic development in Malawi.
The study first agrees with Bell (2013) and Olafsen and Cook (2016) that necessity motivated and subsistence-oriented enterprises, respectively, are most prevalent in developing countries and further agrees with Shane (2009) that the most prevalent enterprises in developing countries are unproductive. It highlights the prevalence of poverty, unemployment and other economic challenges which push individuals towards undertaking of self-employment. Based on the study findings, the perceived failure of entrepreneurship development initiatives to help ignite economic development can be attributed to two factors: (1) the knowledge about entrepreneurship guiding the initiatives undertaken, and (2) lack of focus on innovation that would help SMEs become productive for economic growth.
The prevailing knowledge about entrepreneurship found in the study is that entrepreneurship is starting and running one's own business; being self-employed. It reflects the perception which has guided entrepreneurship development initiatives in Malawi since the early 1980s (Masten & Kandoole, 1997). However, starting and running one's own business, being self-employed is not a salient issue even though the perception can be grounded in classical economic theories (Cantillon, 1755;Say, 1816). Entrepreneurship is about perceiving opportunities and undertaking new combinations of resources to create innovations which ignite economic development (Schumpeter, 1934;Shane & Venkataraman, 2000). The study found that it is enterprises which carry out innovations that contribute more towards economic productivity and that creating new export markets and new enterprises are the most productive entrepreneurial activities. The study further found that productive enterprises are opportunity-motivated, growth-oriented limited liability enterprises in the medium-to-large size category, but are few in numbers. The higher prevalence of enterprises which do not carry out innovations could be attributed to the prevailing perception of entrepreneurship since starting any business is perceived an entrepreneurial act. The findings provide the realisation that the theoretical understanding of the concept of entrepreneurship is important in practice with regard to entrepreneurial activities undertaken and support provided. Entrepreneurship development initiatives have failed to help ignite economic development in the country because of focus on unproductive enterprises. The initiatives do not focus on supporting the creation of innovations which can ignite economic development due to inadequate understanding of the concept of entrepreneurship.
Although the MGDS III (GOM, 2017a) aims to increase the number of enterprises accessing export markets and contributing positively towards economic growth, the implication of the context highlighted by the study findings is that initiatives will continue to fail to help ignite economic development in the country. There is need to adjust the understanding of the concept of entrepreneurship to neoclassical economic perspectives and guide initiatives accordingly to support innovation in SMEs and creation of a conducive business environment. Promotion of entrepreneurship, education and training, SME finance policy and business infrastructure development need to be guided by the understanding of entrepreneurship as perception of opportunities and carrying out innovations. This understanding of entrepreneurship should inform the type of business environment required for productive enterprises to flourish and thereby guide initiatives to tackle respective binding constraints. Currently, there are no initiatives in Malawi to support SME R&D skills development for new products development, SME investment in modern production and processing equipment and technologies or access to new export markets. It is the lack of focus on supporting the growth of productive SMEs which cause entrepreneurship development initiatives to fail to create conducive environment and help ignite economic development.
If the study findings reflect the context of other developing countries in sub-Saharan Africa, for instance Ethiopia (Meressa, 2020) and South Africa (Cassim et al., 2014) then adjusting the perceptions of entrepreneurship to neoclassical economic theories and directing entrepreneurship development initiatives towards the support of innovation in SMEs should be considered. It is presumed that despite decades of efforts to improve the environments for business in sub-Saharan Africa and support of the SME sector to lead in economic development, the results have been disappointing because overall the enterprises supported are unproductive. However, because data used in the study are from a single Sub Saharan African country, it limits the strength of the position reached. Therefore further studies are required in other developing countries in Sub Saharan Africa to assess the guiding understanding to entrepreneurial activities undertaken, innovations carried out by local enterprises and their values so that further reflections are made on appropriateness of entrepreneurship development initiatives undertaken. Zaki and Rashid (2016) and Stam and van Stel (2009) found that entrepreneurship does not bring economic development in developing countries. Although other studies (Adusei, 2016) have argued to the contrary, observations in Ethiopia and South Africa (Cassim et al., 2014;Meressa, 2020) support the former. Nonetheless, entrepreneurship remains one of the pronounced strategies for economic development in Sub Saharan Africa and we need to start finding solutions to make it an effective strategy. This study has offered a reflection on the reason entrepreneurship may not influence economic development in developing countries. Although the environment for business does influence entrepreneurship undertaken in a country, it is the knowledge about entrepreneurship which guides what is undertaken and creation of the requisite environment to support entrepreneurship.

Conclusion
From the literature reviewed entrepreneurship has multiple perspectives which would guide entrepreneurship development initiatives undertaken. The findings show that the prevailing understanding of entrepreneurship is starting and managing one's own business; being self-employed, and it reflects the understanding of entrepreneurship which guides entrepreneurship development in the country. Entrepreneurship is promoted among the youth, women and marginalised individuals as starting and operating one's own business. Entrepreneurship education and training, micro-credit finance and most infrastructure development in rural growth centres for instance, focus on helping many people to start and run their own businesses. This is against the background of high poverty and unemployment levels in the country. As such there are many sole proprietorship, necessity motivated, subsistence-oriented micro and small enterprises which do not carry out innovations. Overall, entrepreneurship is low in the country when perceived through neoclassical economic theories as identifying opportunities and carrying out innovations, and it does not contribute significantly towards economic growth. It is therefore concluded that entrepreneurship development initiatives in Malawi fail to help ignite economic development because the grounding knowledge about entrepreneurship does not guide initiatives to support productive enterprises and there are no efforts to encourage innovation among SMEs. The study therefore offers the alternative reflection on the reasons Zaki and Rashid (2016) and Stam and van Stel (2009) found that entrepreneurship does not influence economic development in developing countries.
Although the SME policy focuses on providing an enabling legal and regulatory environment, finance, education and training, and developing infrastructure, entrepreneurship development initiatives miss the essence of entrepreneurship which is carrying out innovations. As such, the expectations to increase the number of enterprises accessing export markets and making contributions towards economic development as expressed in MGDS III (GOM, 2017a) would not be realised.
It is therefore recommended that; first, the key stakeholders of entrepreneurship development in the country adjust their understanding of the concept. Entrepreneurship in neoclassical economic perspective (Kirzner, 1973;Schumpeter, 1934) is identifying entrepreneurship opportunities and carrying out innovations. This is the modern understanding of entrepreneurship (Henrekson & Sanandaji, 2014) which informs development initiatives towards providing support for innovative start-ups and HGSMEs that make significant contributions towards job creation and economic growth (OECD, 2019;Decker et al., 2014;Audretsch, 2012). Secondly, it is recommended that with the modern perspective of the concept, entrepreneurship development initiatives focus on assisting SMEs to improve in R&D skills for development of new products aligned with opportunities in international markets, and develop market research capabilities to pursue new export markets. Furthermore, support will be required to help SMEs invest in modern production facilities (processes, equipment and technologies) in order to have capabilities and capacities of local industries that meet international quality standards and therefore increased number of SMEs that are able to enter new export markets. Overall, policy, education and training, entrepreneurship finance and infrastructure development would need to focus on SME innovation, competition and growth to enable enterprises make significant contributions towards economic development in Malawi.