Knight, financial institutions, and entrepreneurship in developing economies

Abstract Financial institutions in developing economies fail to provide entrepreneurs with access to finance to grow their businesses. This severely hampers economic development in these countries. We seek to explain why and develop an argument and model based on Knight's theory, which we augment in two ways. First, by describing problems embedded in financial institutions of developing economies, for which we use the Schumpeterian view that creative destruction requires new credit to fund entrepreneurial disruption and de Soto's finding that undocumented assets possessed by entrepreneurs in developing economies cannot be leveraged as collateral to access finance. Second, we use Williamson's hierarchical institutional model to distinguish vertical interactions. The model is illustrated using the case of Uganda, a developing country in Eastern Africa. Our analysis finds that Uganda suffers from intertwined and misaligned formal and informal institutions, limited extent of codified property, and sparse access to finance. The findings prompt policymakers in developing economies to consider problems with and within financial institutions.


Introduction
Much is known about the historical process through which Europe and North America escaped the 'Malthusian trap' and reached previously unimaginable standards of living. But, so far, economists have been unable to produce policy recommendations that enable this development in other parts of the world. There is wide recognition that economic development will be achieved in developing economies through entrepreneurship and expansion of small businesses (Calice et al., 2012;Karakire Guma, 2015;Morris et al., 1995;Singer et al., 2015). But, so far, even highly entrepreneurial developing countries are unable to generate much higher standards of living. This raises the question why countries in the so-called developing world, despite extensive entrepreneurialism and entrepreneurship, are unable to follow Europe and North America in generating much higher standards of living?
There are many attempted answers and clues to this important question, both theoretical and empirical. For example, Schumpeter (1934) famously theorizes that economic development is created through a process of creative destruction caused by the introduction of disruptive innovations through which entrepreneurs force change to the status quo (Schumpeter, 1947). This entrepreneurial process, to Schumpeter (1934), requires expansion of credit to finance innovation that can disrupt the economy's 'circular flow'. Innovation thrives under economic freedom (Ridley, 2020), which is a significant determinant of economic growth (e.g. Gwartney et al., 1999). However, economic freedom may not be necessary for economic growth. Williamson and Mathers (2011) find that certain cultures can support entrepreneurship and give rise to economic growth in the absence of economic freedom. Even so, as de Soto (2000) observed, there are institutional complications. What holds entrepreneurs back in many developing countries is a lack of capital but not possessions: as the institutional setting in these countries does not unambiguously assign property rights, possessions cannot be used for collateral or in other ways be leveraged to grow one's business beyond small-scale local trade. This suggests an explanation for the relative lack of growth-inducing 'systemic' entrepreneurship in developing countries (Sautet, 2013).
Although these different perspectives on entrepreneurship and growth differ and stress distinct aspects of the process, we find in Knight's Risk, Uncertainty and Profit (1921; henceforth RUP) a means to combine them in a unified institutional framework. Presently known principally for the distinction between risk and uncertainty, RUP is primarily an explanation for how and why entrepreneurs can earn profits: by bearing uncertainty through the formation of a business firm (cf. Bylund, 2016). Knight's argument rests on an institutional foundation as 'business could not be carried on at all without the protection of property and enforcement of contraction ' (1921: 194). Thus, the nature and quantity of entrepreneurship depend on the existing institutionsthe rules of the game (North, 1990)to determine how and where entrepreneurs acquire and allocate resources (Baumol, 1990). How and to what effect entrepreneurs carry out new combinations (Schumpeter, 1934), then, is to a great extent an institutional matter.
To Knight, formal institutions can offer an unbiased judicial system to uphold and enforce property laws, such that property owners are able to convert or exchange their property at will to allow for present consumption and acquisition of resources not owned (Emmett, 2007(Emmett, , 2011Hodgson et al., 2001). We augment Knight's discussion with the division of institutions into hierarchically interdependent institutional levels found in Williamson (2000). By distinguishing between institutions at different levels in a hierarchy, we can identify how they interrelate and uncover the nature of their interdependence. As Bylund and McCaffrey (2017) argue, entrepreneurial action takes place at a specific institutional level in response to the nature of the relevant institutions and institutional setting. Specifically, institutional uncertainty can compel entrepreneurs to act at an institutional level that they otherwise would not choose, which thereby affects the effectiveness and outcome of entrepreneurship (cf. Baumol, 1990). We further elaborate on the role of entrepreneurial access to finance, which Knight identified, using Schumpeter (1934) andde Soto (2000). The result is a conceptual model uncovering the institutional implications of access to finance for entrepreneurs.
We then use our conceptual model to analyze the African nation Uganda, which presents an interesting case of a developing yet highly entrepreneurial nation (Bischoff et al., 2020). Our analysis suggests that the country's weak institutions prevent entrepreneurs from accumulating capital and, therefore, to grow their businesses and contribute to the nation's economic growth. We also elaborate on constraints that relate to macroeconomic factors, market regulations, legal and political institutions, and within the financial institutions themselves (Calice et al., 2012), the interactions of which our model also uncovers.

Entrepreneurship in developing countries
An entrepreneur is an individual willing to bear risk when exercising subjective judgment to direct resources for profit under high uncertainty (Cantillon, 1755;Knight, 1921;Mises, 1998): 'The entrepreneur buys productive services at a given time, and converts them into a product which he sells at a subsequent time' (Knight, 1921: xxxviii), during which 'period of production' the entrepreneur bears the risk of the investment. Capital thereby acts as a bridge that makes production possible yet does not take part in the production process because it is surrendered to purchase all the needed labor and materials.
Typically, an entrepreneur acts as capitalist (funder) of his own venture, but when empty-handed without own capital to risk, he must acquire external funds. This is possible only if he is able 'to convince anyone outside himself of any special fitness to exercise the function of an entrepreneur' (Knight, 1921: 279).
Capital can be conceived of as material goods and as a 'fundamental structure of society' (Knight, 1921: 319). The former is embedded in individual property held in the market that 'become commodities and are bought and sold' to satisfy wants (Knight, 1921: 272). Private ownership affords the owner control over the resources that is justified in the eyes of other members of society and, therefore, enforceable. Capital as the 'fundamental structure of society' determines the efficiency of financial institutions when availing capital to entrepreneurs to be transformed into capital goods to drive productive economic activity. Financial institutions, in particular banks, play the important role of accumulating capital by collecting surplus income of individuals desiring to postpone their expenditure, for which they, typically, pay interest. The funds are then offered to productive agents willing and capable to take on a loan repayable with interest. The organizing role of banks and financial institutions in the economy is an engine for formation and rapid expansion of corporations by lending capital loans to productive agents. Indeed, Knight (1921: 23) notes that 'financial institutions made it easy or at least generally possible for ability to secure capital when not in possession of it by direct ownership and made common the carrying on of business predominantly with borrowed resources'.
This problem of securing funding for a startup is faced by entrepreneurs everywhere, but is made worse in developing economies by the lack of reliable institutions to support and enforce agreements. Among developing countries, the African continent stands out as particularly problematic. For example, Munemo (2012) finds that new firm creation in African countries is 'less pronounced' than in other developing countries and Brixiova (2013) shows that access to financing stands out as the most distinctive constraint on firm activities in African as compared to non-African nations. Africa, therefore, appears to be particularly burdened by a lack of entrepreneurial access to financing, which raises questions about the comparative quality of African institutional settings with respect to finance.
Weakness in or lack of properly functioning financial institutions in most developing countries has been rationalized by legal origins, religion-based, disease endowment, and ethnic fractionalization theories (Emenalo et al., 2018). The theories explain the social and cultural norms rooted in formal institutions of former colonies that deter their effectiveness and impact on access to finance and economic development. The premise that embedded religious, ethnic, and cultural diversity in national views drive emphasis on retaining power in lieu of creating institutions that work for all people, impedes property rights and economic development (Beck et al., 2003;Emenalo et al., 2018;Zouache, 2018). Current financial disparities across developing economies are examined by financial system depth and access, that differently impact firms based on age or size (Beck et al., 2011;Emenalo et al., 2018). Weak institutions undermine people's ability to save due to lack of legally integrated structures that limit the accumulation of savings and/or conversion of assets into capital (Beck et al., 2011;Beck and Demirguc-Kunt, 2006;de Soto, 2000). This, Knight (1921: xlix) argues, is problematic as 'political institutions and free exchange and free enterprise are essential to the general framework of a truly moral order'. Therefore, politics needs to be separated from markets to include the poor majority, without which larger firms can use their power and political networks to access formal finance whereas entrepreneurs and small businesses are constrained by the lack of collateral (Beck et al., 2003). Moreover, institutional constraints such as socially embedded norms limit property ownership for women and communal property ownership are not easily translated into legal rights to meet formal requirements of acceptable collateral (Emenalo et al., 2018;Hodgson, 2015).
The institutional setting should further provide reliable enforcement of contract and protection of property so that owners willingly convert or exchange their property to facilitate present consumption and acquisition of resources not privately owned. Otherwise without formal guidelines 'business could not be carried on at all without the protection of property and enforcement of contraction' (Knight, 1921: 194).
Knight's exposition, which we have here summarized in very brief terms using entrepreneurship in developing economies as context, precedes the recent consensus that institutions, including property rights, shape, and support entrepreneurial activity by determining the level and type of entrepreneurship in the market (e.g. Baumol, 1990;Baumol and Strom, 2007;Bylund and McCaffrey, 2017;Henrekson and Sanandaji, 2011;North, 1990). It also suggests two important hurdles to expanding the extent of entrepreneurship that both relate to the financial institutional setting. First, the availability of capital limits entrepreneurs' ability to finance their ventures beyond their personal means. Whether to start new or expand existing businesses, insufficient freed-up capital caps a society's ability to support entrepreneurship. Second, without an institutional framework that properly codifies and secures property rights, the actions of both savers and entrepreneurs are circumscribed. There is less entrepreneurship as a result.

Hurdle 1: capital availability
The issue of lacking availability of capital was expounded by Schumpeter (1934), who argued that disruptive entrepreneurship, and the pursuit of economic novelty in general, might require issuance of new credit. The entrepreneur, to Schumpeter (1934: 66), introduces a new product, new method of production, opens new markets, or creates a new organization. Such new combinations are introduced into and ultimately challenge the circular flow of the economy and, thereby, drive economic development through breaking out of the economy's equilibrium state. However, Schumpeter observes, for these attempts at novelty to be possible they must be financed to be feasible, and this means taking capital out of the circular flow. This poses a problem, since market equilibrium suggests full or nearfull use of existing capital, leaving little to none for experimenting with entrepreneurial innovations. Without slack and idle resources (Bradley et al., 2011;Tan and Peng, 2003), an economy lacks the means to support entrepreneurial innovations and, as a consequence, the means to engender economic development. Economic development, Schumpeter (1947) argued, is an 'organic process' of evolutionary change. He notes: A systemany system, economic or otherthat at every given point of time fully utilizes its possibilities to the best advantage may yet in the long run be inferior to a system that does so at no given point of time, because the latter's failure to do so may be a condition for the level or speed of long-run performance. (Schumpeter, 1947: 83) Similar to Knight (1921), Schumpeter (1934) recognizes that the entrepreneur may himself possess the needed capital but, if not, must borrow it. An efficient and properly regulated banking system making credit available to entrepreneurs in the form of loans is therefore required to support entrepreneurial action in the economy. As entrepreneurs often, especially in developing countries, personally lack the funds necessary to acquire what they need to start or grow their business, capital access is vital for the functioning of the system. It is the creation of new credit, Schumpeter (1934: 95) argues, that is 'a means of diverting the factor of production to new uses, or of dictating a new direction of production' thereby acting as a bridge between the entrepreneur and the foreseen product. This payment fund of new credit offers the entrepreneur purchasing power beyond what is otherwise accessible and thereby provides new entrepreneurs a means to reallocate productive means from their previous to new innovative uses.
Individual entrepreneurs, and therefore the frequency of entrepreneurship, rely on the total funds available to them, whether in the form of personal property or loans obtained. These sources are related not only because the entrepreneur can draw from both to finance his business. Although credit is provided in the form of currency, and therefore purchasing power, personal property typically needs to be converted to money first. This requires access to markets that determine the monetary value of the property possessed and, therefore, the purchasing power the entrepreneur can wield in the bidding for productive resources. Therefore, financing an entrepreneur is dependent on the existing institutional structure 'which permits protection (through, for example, patents and secret processes) to the innovator' (Schumpeter, 1934: 113).

Hurdle 2: codified property
None of the capital-acquiring transactions noted above, which are easily recognizable as necessary for entrepreneurs to pursue starting and growing their business, would be possible without also enjoying formal ownership of property. Without proper institutional support and formalized property rights existing capital has very limited use. Developing countries typically have weak formal institutions that compete with informal institutions, thereby creating high uncertainty in the market (e.g. de Soto, 2000). This is problematic, since entrepreneurship flourishes in economies with well-established institutional framework where judicial, property, and financial institutions can enact compliance and enforcement of laws to protect both ownership and transactions (Baumol and Strom, 2007;Sobel, 2008). de Soto (2000) argues that developing countries have not realized entrepreneurial success because they are incapable of producing capital that holds the power to increase productivity of labor. This makes financing innovation almost impossible hence entrepreneurial ingenuity and enthusiasm unexploited unable to accelerate economic development (de Soto, 2000).
The key problem here is accumulated undocumented assets, whose physical attributes and economic value cannot be determined. For example, people have houses built on land without title or unincorporated businesses. What is missing in developing countries is the ability for entrepreneurs to utilize the potential of their properties to be deployed in new production as capital (de Soto, 2000: 40). They suffer from a 'bad legal and administrative system' (de Soto, 2000: 164) and from a legal infrastructure that properly codifies and upholds property rights (Hodgson, 2015), which means the undocumented assets without property titles cannot be used as collateral when seeking loans from financial institutions. de Soto (2000: 46) states that 'what the poor lack is easy access to the property mechanisms that could legally fix the economic potential of their assets so that they could be used to produce, secure or guarantee greater value of the expanded market'. They, therefore, are deprived of the means to pursue entrepreneurship or grow their firms (Bischoff et al., 2020;Bu and Cuervo-Cazurra, 2020), as Knight (1921) also recognized, and keep them in their family (Boudreaux, 2014).
Notably, within weak institutions, incongruence between formal institutions and economic activity allows for existence of alternative informal institutional framework, 'extralegal systems' within which entrepreneurs operate in the market, yet increase uncertainty when starting and growing new businesses (Baumol, 1990;de Soto, 2000;Mair et al., 2012). The 'extralegal systems' permit complex informal entrepreneurship to invisibly exist without legal registration but at the same time limit or make impossible access to formal credit from financial institutions and, therefore, their contribution to economic development. Entrepreneurs in extralegal systems use informal agreements embedded in social contracts to acquire needed resources.

The institutional system
The aforementioned problems interact to make entrepreneurship, and especially scalable entrepreneurship, all but impossible. However, the interaction of institutions and entrepreneurship is not oneway but has been widely recognized as a bilateral relationship (e.g. Baumol, 1996;Baumol and Strom, 2007;Bylund and McCaffrey, 2017;Henrekson and Sanandaji, 2011). Although entrepreneurs operate within institutions, they react to the institutional setting: they can take evasive actions to avoid existing institutions that they find unreasonable or inapplicable to challenge the status quo, abiding action to comply with existing institutions, or altering action intended to change the institutions rather than, as evasive entrepreneurship, aim to avoid and thereby indirectly challenge them (Henrekson and Sanandaji, 2011). This applies also in developing economies (Elert and Henrekson, 2020) even though expected economic results may not be realized because of weak formal institutions, which are often contradicted by existing informal institutions (Helmke and Levitsky, 2004).
As a result, contributions of entrepreneurial activity in the market are influenced by the effectiveness of the formal and informal institutional interrelationships (Baumol and Strom, 2007;Khanna and Palepu, 1997;Mair and Marti, 2009). Because of weak institutions and institutional voids, enormous entrepreneurial potential remains unexploited (Khanna and Palepu, 1997;Olthaar et al., 2017). The lack of facilitative institutions, primarily in finance, also undermines the interrelation between entrepreneurship and the institutional setting. There is, as a result, continued failure to develop strong formal institutions with streamlined systems that monitor and regulate financial activity to manage uncertainty in the market. Banks encounter numerous constraints related to macroeconomic factors, market regulations, legal and political institutions, and within the banks themselves when processing loans (Beck et al., 2008;Beck and Demirguc-Kunt, 2006;Calice et al., 2012). Access to formal substantial capital becomes limited to fewer borrowers and most entrepreneurs therefore mobilize only small savings and family contributions (Amoako-Adu and Eshun, 2018; Beck et al., 2008;Biekpe, 2004;Morris et al., 1995).
The institutional embeddedness, if not dependence, of entrepreneurship has been widely documented (Baumol, 1990;Bylund and McCaffrey, 2017;Harper, 2018). For an institutional setting to be supportive of entrepreneurial action requires, as both Schumpeter (1934) and de Soto (2000) observed alongside Knight (1921), more than the economic freedom to pursue entrepreneurial ventures. In other words, the institutional setting is much broader in scope than what directly affects the entrepreneur in their decision-making, including the financial system that provides investment capital and a property system that codifies, assigns, and enforces property rights.
Furthermore, as suggested by the discussion above, institutions are interrelated, such that access to finance through banks, which requires supportive financial institutions, in turn depend on codified property rights. Institutions are therefore best understood as existing at different levels in a hierarchy. Williamson (2000) conceptualizes the institutional system of the economy in four distinct levels. Figure 1 illustrates the levels, their respective aims, and interrelationships. They can be conceived as levels of abstraction, each determined by their respective order of economizing arising from human actions. Although action takes place with respect to any of the levels, the impact of an individual action typically lessens the further up in the hierarchy the aimed change.
Individual action takes place at the lowest level L4, from which institutions emerge, arising in the form of standards and expectations. The price mechanism, for example, and its 'invisible hand' allocation of resources toward their more highly valued uses emerges from the mass of actions taken without requiring prior design. The higher level L3 comprises designed attempts to overcome costs and shortcomings at the lowest level. This is where we find organizations and other forms of governance adopted to align productive incentives beyond what is possible at L4. Moving to L2, we find the institutional environment with formal institutions such as property rights and laws. The top level L1 is the social embeddedness level at which culture, norms, and societal values reside.
Institutions at these levels are vertically interdependent in the sense that institutions at a higher level constitute boundary conditions for institutions (and actions) at lower levels. Although action is still possible beyond this boundary, it is exceedingly costly (Elert and Henrekson, 2016). Similarly, moving up in the hierarchy is associated with higher cost. A core function of institutions is to lower transaction costs for abiding action, thereby 'providing structure to everyday life' by bringing about standardization, increased predictability, and, therefore, lower uncertainty (North, 1990: 3).
If institutions at different levels are misaligned, the institutional uncertainty (Bylund and McCaffrey, 2017) can be detrimental for entrepreneurship. For example, an entrepreneur acting within a legal system that is at odds with a society's core values and norms will find it difficult to both serve customers, who espouse those values, and comply with laws, which are based on contradictory values. To overcome such issues, entrepreneurs are forced to take actions aiming at changing institutions at a higher level, and they may therefore choose to exit. We find examples of this in developing economies where the state, through legislation, or international bodies of expertise (cf. Easterly, 2014) attempt to change society in some particular direction that is incompatible with society's informal institutions.
The hurdles discussed above exist at specific levels within the institutional system. We will use the case of Uganda to demonstrate how the institutional structure impacts and affects the opportunities for entrepreneurship and, therefore, economic development.

Uganda: entrepreneurship without economic development
Uganda makes a good case to illustrate our argument. It is a developing nation of sub-Saharan Africa, with one of the highest rates of entrepreneurial activity in Africa (Ács et al., 2015;Baughn et al., 2006;Bischoff et al., 2020;World Bank, 2019). Entrepreneurship and/or small business ownership is conceived as remedy to high unemployment rates among Uganda's youths (Langevang et al., 2012), and driver of economic development (Elert and Henrekson, 2020).
Uganda is one of the countries observed and analyzed in the annual Global Entrepreneurship Monitor (GEM) dataset. The Global Entrepreneurship Index ranked Uganda's entrepreneurship ecosystems 131 of 136, indicating limited support for healthy entrepreneurial activity (Ács et al., 2015). Despite the uncertainty faced by entrepreneurs, the International Labor Organization estimate modeled 78.2% of the total employable persons to be engaged in self-employment, above 76% of the sub-Saharan Africa, and indicative of a preponderance of entrepreneurship being local and ineffective with respect to economic growth rather systemic and growth-generating (Sautet, 2013). Similar numbers were reported by Uganda Bureau of Statistics (UBoS) (2017), that more than three quarters of Ugandans were engaged in some form of entrepreneurial activity. However, most entrepreneurship activity is factor-driven, representing the earliest stage of economic development that depends largely on subsistence agriculture and unskilled labor (World Bank, 2019). This suggests lacking institutional support, if not a presence of institutional barriers, for entrepreneurs to expand beyond small-scale, local ventures. Knight (1921) theorized five 'elements of progress', one of which, growth of population, is principal to Ugandans' entrepreneurship potential (p. 144). Uganda's 44 million population, with a 3.6% growth rate that is one of the highest globally, and 80% below the age of 35, are the bedrock for the nation's entrepreneurship potential (World Bank, 2019). The rapidly growing population guarantees long-time labor supply and a consuming unit for produced goods, central for a society's economic growth (Knight, 1921). Knight recognized a society's population and education as 'long-term conditions of the supply of labor' to achieve higher levels of production efficiency (p. 158). This suggests that Uganda's the large population facilitates and can support increases in productive entrepreneurship and quality labor supply to production through entrepreneurship education. However, the World Bank (2019) reported that 33.5% of Uganda's youth were not in education or employment training, suggesting that the youth lack technical business skills to successfully become entrepreneurs. Bischoff et al. (2020) find, in line with Knight, that entrepreneurship education mitigated the negative relationship between capital constraints and business creation in Uganda.
We will now illustrate Knight's discussion on formal institutions as judicial systems that facilitate exchange of property, into hierarchically interdependent institutional levels per Williamson's (2000) model.

L1: social norms
Institutional history informs us of the culture history facts and products that influence economic activity (Knight, 1921). To understand the challenges faced at L3 related to capital acquisition, one needs to familiarize with Uganda's institutional history. Pre-colonial period in Uganda, informal institutions governed by kings and chiefs owned land, which they gifted loyal subjects (Ellis et al., 2005;Musinguzi et al., 2020). During British rule and the signing of the 1900 Buganda agreement, cultural institutions gained autonomy over land that had occupants (cf. Goodfellow and Lindemann, 2013;Musinguzi et al., 2020). Although cultural institutions gained legitimacy, overlapping problems emerged between registered owners of gifted land and the peasants who occupied it.
Currently, disputes over legal land titles have heightened uncertainty over legal land titling. For example, Buganda kingdom's Mailo land accounts for 20% of the country's land and home of the metropolitan area, is key to Uganda's economic development (Musinguzi et al., 2020). However, in some areas bewilderment surmounts whether Buganda kingdom or the state should issue titles.
L1 institutions (i.e. religion and culture) determine family structure, gender roles, and property ownership, which in Uganda can conflict with formal rules on lower levels (primarily L2). For example, although the state can implement systems to encourage savings, the entitlement of family members to receive financial support makes state efforts unattainable (Daspit and Long, 2014;Khayesi et al., 2014). Furthermore, the state efforts to promote gender equality are frustrated by socially constructed, culturally based gender biases that place women as male subordinates and usually unrightfully property owners (Ellis et al., 2005;Karakire, 2015).

L2: the rules of the game
Knight viewed formal institutions as unbiased judicial system, upholding and enforcing property laws. We expect that, for a functioning economy, property rights be supported by legal systems that encourage others to abide by and avoid interference or conflict over a legally titled property (de Soto, 2000;Hodgson, 2015). As the functions of financial institutions are founded in formal property institutions and free contracting providing the institutional background that motivates economic behavior, bank or financial institutions are active intermediaries distributing economic resources from capitalist with productive means and entrepreneurs forming new combinations.
The mandate of financial institutions in Uganda is distribution of financial resources to potential users and the public by extending credit and monitoring marketing to reduce risk (BoU, 2019). All forms of financial institutions (e.g. commercial banks, credit units, or micro finance) are regulated by the Bank of Uganda. At the end of 2019, Uganda had 24 registered commercial banks, and 553 branches serving the nation's population (BoU, 2019). Effective capacity of financial institutions to extend credit is limited by the ineffective property rights protection, created by weak formal institutions conflicting with informal institutions (Branch, 2007;Goodfellow and Lindemann, 2013;Lindemann, 2011). Uganda further suffers from inconsistent enforcement of formal property institutions, which is seen in, for example, cancelation of documented land titles, making the institutional setting unpredictable and therefore also unreliable. For example, Uganda land commission (ULC) issued 175 land titles between 2017 and 2019 to investors that National Forestry Authority (NFA) annulled (Ladu, 2020).
Entrepreneurship in Uganda is further negatively affected by the intertwined political and economic institutions monopolized by individual with political networks and information (Elert and Henrekson, 2020). Political institutions and the market do not function independently to create a free market as Knight theorized. Political constraints imposed on entrepreneurial activity make breaking out of the circular flow inconceivable while following the rules of the game as is (Wegner, 2019). Political power over the rules of the game dictate how they are applied, rarely uniformly, even when accessing formal finance at L3 in the absence of secure property rights. For decades, the Ugandan government uses tax incentives to appeal to foreign and local investor, to increase job creation and boost economic development. For example, Chinese investors are attracted by tax policies, vast natural resources, land, and labor. The job market has grown as the government predicted, despite increased uncertainty over land when local occupants are evicted without compensation. Moreover, some investors are not well vetted or monitored, coming in as 'big investors' but start small businesses that compete with (or outcompete) local entrepreneurs (South China Morning Post, 2017), which discourages the growth of local businesses.
Uganda is an agricultural country and 81% of the adult population engages in agriculture, contributing over 30% to the country's GDP (UBoS, 2018). In 2001, National Agriculture advisory services (NAADS) was initiated to accelerate decentralized economic growth and agriculture entrepreneurship. NAADS members decide whether to participate in the program to leverage government advisory services, grants, technology, and new crops and livestock (Benin et al., 2011). However, farmer groups cannot bargain market prices for their produce due to limited direct access to international markets. The farmers are mediated by individuals with political power serving personal and political financial interests (Pearl Times Reporter, 2020). Benin et al. (2011) find that despite NAADS, there was no significant difference in growth between NAADS and non-NAADS farmers.

L3: banks and the financial system
Knight equated capital to property with the assumption that an entrepreneur may possess individual capital or borrow it when not in possession of it, in exchange of private property owned. However, uncertainty imposed at L2, renders assets possessed by entrepreneurs unable to be used as collateral in exchange for formal capital (de Soto, 2000). L2 uncertainty compels financial institutions to design costly measures to offset risk, such as high-monitoring costs, which cause them to charge high interest rates on borrowing.
It is not enough, of course, for entrepreneurs to hold capital goods. Capital must be held as money to be paid for labor or be possessed in rightfully documented property that can be turned into liquid capital (cf. de Soto, 2000). However, there are lots of undocumented property in Uganda that is kept in unplanned houses, unregistered businesses, fields of crops, livestock, or motor bikes, without legal titling. All such property is dead capital unless there is a market for exchange that make these assets indirectly liquid. This process is greatly hampered by the market's limited capacity to evaluate property, since it depends on few qualified and skilled valuers (Franzsen, 2002). This greatly limits the liquidity of owned assets even if ownership can be established. Moreover, banks are limited in valuing assets of the poor and cannot accept them as collateral due to the absence of nationally established standards for all forms of assets. In 2012, 50% of financial institutions could not lend to firms because they were unable to define or register claimed property (Calice et al., 2012). Without a commonly available framework for evaluating and registering private property, assets remain invisible in the market and cannot contribute to production (de Soto, 2000).
This added uncertainty is offset by banks through charging 20% average weighted lending interest rate, the highest in the East African region (World Bank, 2019). As a result, banks do not serve the whole market and cannot serve it well. On the one hand, the poor lack formal titles that make their assets unusable as collateral (Amoako-Adu and Eshun, 2018;Biekpe, 2004;de Soto, 2000). On the other hand, large businesses encounter unique challenges related to stringent repayment terms and ambiguous international banking laws. For example, in 2020, the high court determined that Diamond Trust Bank (DTB) Uganda and its parent DTB Kenya illegally lent Ham Enterprise Ltd (Ugandan business) a syndicated loan (Anyanzwa, 2020), and DTB Uganda was ordered to compensate all loan repayments withdrawn from the borrower.
Banks thus struggle to effectively serve both the poor and wealthy to engage in economic activity (Chowdhury et al., 2018). Notably, only 25% of Uganda's population have access to formal financial institutions (i.e. banks) or semi-formal institutions (i.e. credit institutions and micro-deposit taking institutions) (Akileng et al., 2018;Munyegera and Matsumoto, 2016). Currently, mobile banking has become a faster and cheaper alternative for the poor to save, send, and receive money using a registered SIM card. In 2019, Bank of Uganda reported 277 million mobile phone transactions conducted by the 27.1 million registered customers of the mobile services. Although Knight viewed such progress in technology as an element of progress, mobile services that serve the poor majority do not offer generous loans for investment (Baganzi and Lau, 2017;Munyegera and Matsumoto, 2016).

L4: level four
For most entrepreneurs in Uganda, the lack of legal status of their enterprise limits access to valuable information on bidding or contracting for formal projects. Furthermore, without legal membership in trade unions and business associations, entrepreneurs are excluded from bargaining better prices, taxes, interest rates, and access to capital. For example, at the end of 2019, 46% of all businesses in Uganda were formally registered businesses whereas 54% were unregistered (UBoS, 2017). The red tape and many fees associated with business formalizing makes it unattractive, such that small business owners prefer to act informally to avoid registration fees and tax evade.
Although technology typically improves entrepreneurship through increased knowledge diffusion, most entrepreneurs in Uganda have had their online business activity disrupted or undermined by policies and government action. For instance, on the eve of the January 14, 2021 presidential and parliamentary elections, Uganda Communication Commission (UCC) on behalf of government shut down the internet for 5 days in the entire country. The unprecedented decision left many businesses dependent on internet access dysfunctional (Ladu, 2021). Online shopping and delivery stores experienced massive losses, whereas extra fees were incurred in goods storage. Moreover, in 2018, the government imposed a 1% tax on mobile money transactions, a policy that mostly affected the poor and small business owners.

Weak institutional support and uncertainty
Our discussion above clearly indicates significant weakness in the institutional support for entrepreneurship in Uganda. Access to finance stands out as the banking and financial system in L3 appears ill equipped to provide entrepreneurs with means to start and expand their businesses. Banks are further undermined in their entrepreneurship-supporting role by Uganda's inconsistent and unreliable legal framework and its enforcement (both residing in L2). This increases uncertainty and costs in the banks' operations and also limits the scope of services that can legally and profitably be offered. Banks' servicing of entrepreneurs is further hampered by the limited availability of capital that can be offered as loans (hurdle 1) as Uganda overall has little surplus and therefore savings, which suggests an economy that is, in at least this respect, similar to a circular flow (Schumpeter, 1934). In addition, the legal framework undermines entrepreneurs who are seeking finance by requiring difficult and costly licensing for legal protection whereas, at the same time, failing to protect property rights (hurdle 2). As banks can rely on neither the entrepreneurs' business status nor their possessions as collateral, there is little to no motive for offering them loans. As a result, the loans that are provided to businesses primarily end up financing large and politically connected business, which further strengthens the barriers for entrepreneurship and growth prospects for small business.
In addition, misalignment between institutions at different levels causes institutional uncertainty (Bylund and McCaffrey, 2017) that undermine the predictability and thus reliability of the institutional system. Cultural norms with a firm historical and traditionalist basis (Emenalo et al., 2018), which dominate in the Ugandan countryside, sometimes differ significantly from the legal framework that pertains to regulate economic action. Similarly, inconsistencies and ambiguities within L2 cause the legal framework to be further unpredictable, thereby prompting actors, whether banks or entrepreneurs, to rely on extra-legal measures, which typically will be based in the cultural norms of L1. This increases the cost of business but also further undermines specific legal institutions which thereby increases uncertainty.
If entrepreneurs also perceive government structures as unfair and limiting, they (re)organize to do business beyond what is recognized by the formal institutions. In other words, they choose economic action in the 'black' market (Rohac, 2014), a decision can be prompted by not having access to the regular market but also, for example, due to cumbersome registration bureaucracies and high costs associated with running a business (de Soto, 2000;Ellis et al., 2005;Singer et al., 2015). Moreover, information asymmetry on existing resources limits the exploitation of profitable opportunities in Uganda, thereby leaving many untapped. Hence, we find that the problem of entrepreneurial access to finance has its basis in weak institutions at levels L2 and L3, specifically in terms of hurdles 1 and 2, and is further augmented by institutional inconsistencies and the institutional uncertainty that they cause. As a result of this institutional situation, small businesses rely primarily on small personal savings and family contributions, which severely hampers their ability to expand both in organizational size and geographically. Entrepreneurs are, therefore, due to the aforementioned institutional constraints, destined to do business first and foremost using means without institutional support in L2 and L3. Instead, they choose to rely on L1 norms to operate through cash-based contracts for minimal profit margins.

Concluding discussion
We have analyzed how entrepreneurship in the developing nation Uganda is impacted by institutional predicaments identified by Knight, Schumpeter, and de Soto. We find that Knight's discussion in RUP provides a constructive framework for understanding entrepreneurship as uncertainty-bearing and the role of the firm. But, the framework also emphasizes the important role of institutions and points explicitly to the importance of financing for entrepreneurship to be a feasible force of change and, consequently, cause economic growth. With Knight as starting point, we find in Schumpeter and de Soto specific arguments that clarify the meaning of and provide nuance to Knight's emphasis on finance.
To Schumpeter, economies can grow only through the disruptive force of entrepreneurs who break out of the regular market's assumed circular flow. This requires, argues Schumpeter, not only access to financing, but also creation of new credit. Markets tend to make use of productive resourcescapitalto the degree technological know-how and entrepreneurial ingenuity make possible. For this reason, new entrepreneurs must outbid current users of capital in order to bring their innovations to market. Therefore, they need new purchasing power, which places financial institutions at the core of entrepreneurially driven economic growth processes. Similarly, de Soto shows that poor countries typically lack the means for entrepreneurs to take advantage of such credit because they lack titles to the property they own. As a result, entrepreneurs in poor countries cannot gain access to the credit that Schumpeter showed is necessary for the realization of entrepreneurial innovation.
We use Williamson's institutional system model to illuminate the hierarchical interdependencies within the institutional setting for entrepreneurship in Uganda. Doing so, we find support for the aforementioned ideas: limited access to finance (Schumpeter, 1934) and lack of codified property (de Soto, 2000). We find that the problems exist specifically at levels L2, the formal rules of the game, and L3, governance, and that they interact to worsen the situation for entrepreneurs. Furthermore, the misalignment of institutions at different levels causes institutional uncertainty that, as a result, undermines entrepreneurship and causes many entrepreneurs to deal in accordance with L1 cultural norms and without means to expand their businesses. Our analysis therefore corroborates Knight's focus on financing as an important means for entrepreneurship, but also suggests an institutional explanation for the limited economic development in Uganda. It also lends support to Sautet's (2013) argument that the lack of economic growth in developing nations can be explained by entrepreneurship being overwhelmingly local rather than systemic and therefore not contributive to the nations' economic growth. Following Knight, and as indicated in our case study, we find an explanation for this in the ineffectiveness of the financial institutions, which provide very limited access to credit and do not unambiguously assign property rights to entrepreneurs' possessions.

Little access to finance
The credit available to entrepreneurs at L3 results from the interaction of banks and entrepreneurs. Banks fulfill one of their key roles and conduct due diligence to determine business formality, forecasted revenue, and production site before issuing capital. They select entrepreneurs to finance based on the feasibility and profitability of the innovation. However, information asymmetry during loan screening on property ownership and formal credit history, and guarantors, signals entrepreneurs as riskier borrowers compared to incumbent firms (Amoako-Adu and Eshun, 2018; Chowdhury et al., 2018;Singer et al., 2015). That explains the 6% of banking system users that have accessed formal credit (Akileng et al., 2018).
Consequently, entrepreneurs bootstrap resources from personal savings, family, and friends, to increase their productive capacity (Beck et al., 2008). Intriguingly, expected investment by family and friends is embedded in social norms at L1 capital (Chowdhury et al., 2018). Also, due to imperfect intercommunication, entrepreneurs are ignorant of all possible financing options, pay different prices for capital, and some options are too complex to understand (Gollin and Rogerson, 2014). However, we note that access to financial services has been growing, amplified by the popularity of mobile money services through telecommunication networks facilitate online transaction more conveniently (Baganzi and Lau, 2017).
No properly codified property to serve as collateral As banks conduct due diligence to offer loans to entrepreneurs, it is difficult for them to trace and substantiate asset ownership. The office of the chief government valuer is responsible for valuing property on behalf of government, but national standards are yet to be established within Uganda's constitution that banks can utilize. The lack of guidelines delineating all types of property limits acceptable collateral by banks and individuals undercapitalized by failure to draw economic value from their property (de Soto, 2000). Challenges in codifying property relate in part to social norms at L1 and political involvement and attempted direction at all institutional levels (Williamson, 2000), which contribute to institutional uncertainty (Bylund and McCaffrey, 2017).
Certified valuers are expensive and focus on valuing mostly large and capital intensive assets such as buildings, factories, and sophisticated machinerythe types of property not possessed by the poor (Amoako-Adu and Eshun, 2018). In a functioning formal property system, records and titles present economically meaningful value of assets that banks can use when evaluating collateral (de Soto, 2000;Emenalo et al., 2018). Poor borrowers' limited financing options and weak bargaining power is exploited by informal moneylenders charging high interest rates. When undocumented property is accepted, it is undervalued and the loan amounts consequently much lower than the value of the property used as collateral.
Knight and economic development Knight's (1921) entrepreneur can gain access to finance by persuading individuals or financing entities of his ability to exercise the function of an entrepreneur. In contrast to entrepreneurs in many developing countries, they typically must have ongoing businesses before they can be considered as borrowers. That means the institutional environment does not make their ability sufficient to secure capital, as Knight (1921) theorized. Limited capital and undocumented private property, which therefore cannot serve as collateral leave vast entrepreneurial opportunities unexploited in developing countries (de Soto, 2000). We have shown that this is the case in Uganda in line with the arguments of Schumpeter and de Soto where banks are challenged when extending capital to entrepreneurs without titled property in weak property laws, and how these issues exist at two distinct institutional levels that interact and thereby augment the problem. This explanation suggests the problem may be systematic and in the form of lacking institutional support for expansive entrepreneurship rather than, as with e.g. Sautet (2013), a matter of entrepreneurs being unable to trust in banks and other providers of financing.
In sum, our analysis offers valuable insights into how weak institutions impede entrepreneurship, especially systemic, in developing countries and thereby stifle economic growth and development.