Angel investments of small family business entrepreneurs: cross-country evidence

Introduction

Angel investors are a crucial source of capital, management experience, skills, and contacts for new entrepreneurs. An angel is commonly defined as an individual, frequently with high net worth and business experience, who directly invests part of their assets in new and growing private businesses to which they have no family connection—in exchange for ownership equity—acting alone or through semiformal networks (Croce et al. 2020; Cumming and Zhang 2019; Cumming and Zhang 2023; Edelman et al. 2017; Lerner et al. 2018; Mason et al. 2019; Preston 2007; Tenca et al. 2018). Angel investors are generally classified as affiliated or non-affiliated (Edelman et al. 2017). An affiliated angel is someone with whom the entrepreneur is acquainted, whereas an unaffiliated angel investor has no connection to the entrepreneur.

Unlike venture capitalists and crowd-funders, angels directly engage with entrepreneurs and start-ups by providing entrepreneurial experience, mentorship, networking, industrial knowledge, business strategy, and marketing (Blaseg and Hornuf 2024; Hsu et al. 2014; Solodoha et al. 2023). Evidence in this regard has indicated that several angel investors have a deep understanding of financial and technological markets and can foresee market trends. Therefore, angel investors have become essential for governments seeking to foster business creation and economic growth (Solodoha et al. 2023). Evidence from Huang et al. (2017) for the United States indicated that in 2016, angel investors provided approximately $24 billion to more than 64,000 start-ups in the country. The most experienced angel investors dedicated 15% of their investment portfolio to angel investments and provided $42,000 per start-up on average; meanwhile, those with less experience dedicated only 7% to angel investments and invested $25,000 on average. In particular, women investors provided relatively modest amounts of capital, at an average of $26,500, compared to men, who invested an average of $37,700.

Currently, family offices are an additional source of financing for innovative and fast-growing companies. These are investment vehicles for sustaining and growing the wealth of large family businesses, which currently account for nearly one-third of all capital invested in start-ups globally (PwC 2023). Family offices are typically conservative and risk-averse and, therefore, reluctant to make high-risk investments (Block et al. 2019; Wessel et al. 2014). In particular, evidence has indicated that single family offices are more averse to borrowing than private equity firms and behave in accordance with pecking order theory, which establishes that internal financing sources are preferred to external ones, such as debt and equity (Schickinge et al. 2022a).

This study focuses on the behavior of owners/managers of small family businesses as micro-angel investors. These are individual investors—not institutional investors or investment firms—who may have varying levels of experience in entrepreneurship, investing, or specific industries. Micro-angel investors typically make small investments, frequently as a side activity to their main occupation or as part of a diversified investment portfolio. These investments are usually smaller than those of traditional angel investors or venture capitalists. Beyond providing capital, micro-angel investors may offer mentorship, advice, and connections to help start-ups grow. Micro-angel investors may include friends, family members, or individual investors who are interested in supporting entrepreneurship and innovation but may not have the resources or risk tolerance to invest larger amounts (De Clercq et al. 2012; Ding et al. 2015; Honjo and Nakamura 2020; Fernandez 2024; Farrell et al. 2008; Maula et al. 2005). Moreover, they may participate in crowdfunding platforms or investment syndicates that pool small investments from multiple individuals to support start-ups (Deng et al. 2022; Estapé-Dubreuil et al. 2016).Footnote 1

The sample is derived from a unique dataset of family entrepreneurs gathered by the Global Entrepreneurship Monitor Adult Population Survey (GEM APS) 2018 as part of a special topic on family entrepreneurship (Kelley et al. 2020). The questions pertaining to this topic concern family participation in the management, ownership, and staff of a given business. Additionally, the APS regularly includes information on the survey participants’ angel investments. Therefore, family entrepreneurs’ angel investment decisions can be explored by combining these pieces of information. This topic is new to the literature because existing studies have focused on large family firms whose wealth is managed by family offices. For example, the family offices in Block et al.’s (2019) sample reported assets under management between $100 million and $200 million. By contrast, the family businesses of the APS 2018 were predominantly small and financed third-party start-ups through angel investments, which, on average, did not exceed $30,000.

According to the GEM APS 2018, businesses wherein family members participated in ownership, management, and staff contributed $8.9 million to businesses started by other entrepreneurs during the 3 years before the survey. Approximately 37%, 18%, and 20% of this amount went to start-ups of close family members, those of friends or neighbors, and those of strangers with good business ideas, respectively. By contrast, businesses with no family participation invested $13.9 million in the same period, of which approximately 21%, 27%, and 21% went to start-ups of close family members, those of friends or neighbors, and those of strangers with good business ideas, respectively. Overall, micro-angel investments, including those from businesses and individuals, reached about $132 million in the 49 sampled countries during the 3 years before the survey. Of this amount, men and women investors contributed about $92 million and $40 million, respectively.

I contribute to the extant literature by assessing the extent to which country-level rates of family businesses affect the likelihood of observing micro-angel investors and the amount invested in new ventures. Moreover, I contribute to the literature by examining whether family businesses prioritize endeavors started by extended family members and whether family businesses provide comparatively fewer financial resources than comparable non-family businesses. I hypothesize that ceteris paribus, family businesses have less financial slack in supporting third-party ventures because of their greater preference for internally generated funds. When investing in such ventures, family businesses favor start-ups of extended family members to increase their family capital through family social networks. In this sense, existing literature suggests that financing decisions of family firms consider control retention, risk aversion, and non-financial values and goals of owners/managers (Schickinger et al. 2022b). Theoretical foundations for such financing preferences have been noted in stewardship theory, the socio-emotional wealth perspective, and the theory of planned behavior (Michiels and Molly 2017).

Collectively, these findings highlight how the involvement of small family and non-family businesses in financing start-ups constitutes financial innovation within entrepreneurial ecosystems. Small business owners/managers are often deeply embedded in their local communities, thus allowing them to leverage their social capital to benefit start-ups through valuable guidance, mentorship, and financial backing. In turn, a thriving small business community often fosters a culture of innovation and creates a fertile ground for new start-ups to emerge. This culture can result in increased experimentation and the development of novel products and services, further enriching the entrepreneurial ecosystemFootnote 2 (Frimanslund et al. 2022; Wadichar et al. 2024).

The remainder of this paper is organized as follows: “Conceptual framework” section presents the conceptual framework for testing the research hypotheses. “Data, descriptive analysis, and methodological aspects” section presents the data, descriptive analysis of angel investments, and methodological tools. “Results and discussion” section presents and discusses the estimation results for family entrepreneurial participation (“Family participation” section), micro-angel investment decisions (“Becoming a micro-business angel” section), potential funding recipients (“Angel investment recipients” section), amounts invested by micro-angels (“Funding provided by micro-business angels” section), and two extensions of the average treatment effects (ATEs) of the national family business rate and institutional quality on the likelihood of financing family members/strangers and providing above-median funding (“Extensions: ATE of family business country rate and institutional quality” section). “Conclusions” section concludes by summarizing this study’s main findings, highlighting its shortcomings, proposing topics for future research, and suggesting its policy implications.

Conceptual framework

As mentioned in the Introduction, this study focuses on the behavior of small family businesses as informal investors. This topic is novel to the literature, which has centered on family offices, that is, legal entities that manage the wealth of one or some wealthy entrepreneurial families. In this sense, family offices are suitable vehicles for owner families seeking to reinvest their wealth after selling their family business or for entrepreneurial families that have accumulated substantial wealth in addition to the money invested in the family business (Block et al. 2019; Schickinger et al. 2022b). However, the family businesses covered by the GEM APS 2018 are generally small, domestically oriented, and lack family offices. Hence, if they invest, they do so as micro-business angels.

Existing theories of angel investors and family firms

The decision to become a business angel has been modeled using theories of social psychology, financial economics, and the institutional environment. For example, the theory of planned behavior (Ajzen 1991) provides a strong foundation for explaining individuals’ intentions to perform behaviors that require some form of premeditation. In the context of informal venture capital investment, the theory of planned behavior effectively elucidates the relationship between new firms’ growth intentions and realized growth (Lortie and Castogiovanni 2015). The theory of household portfolios, in turn, allows us to understand the demand for risky assets, such as business start-ups,Footnote 3 and partially explains the variance in informal investment activity (Maula et al. 2005).

Empirical evidence has indicated that angel investment decisions are rooted in and influenced by formal and informal institutional factors, such as social trust (Ding et al. 2015; Fernandez 2021, 2024; Xiao and Anderson 2022), legal protection (De Clerc et al. 2012), cultural differences (Cumming and Zhang 2019), and personal and corporate bankruptcy laws (Cumming and Zhang 2023). Moreover, angel investment decisions are influenced by gender, education, age, entrepreneurial experience, and business networks (Falcão et al. 2023; Honjo and Nakamura 2020).

When analyzing angel investments in family businesses, considering a theory or conceptual framework elucidating the dynamics of such businesses is essential. Specifically, a key theory for the current study is the resource-based view of the firm, which suggests that the returns achieved by businesses depend largely on their resources. To provide a sustained competitive advantage, these resources must be valuable, difficult to imitate, and non-substitutable (Lerner and Levy 2009). In this sense, family businesses may create a competitive advantage from familiness qualities, such as strategic focus, customer orientation, family relationships, and operational efficiency (Collins 2012, chapter 1; Habbershon and Williams 1999; Tokarczyk et al. 2007). Consequently, family businesses may compensate for their relative weaknesses in human and financial capital by building a distinctive type of social capital, denoted as family capital (Levie and Lerner, op. cit.),Footnote 4 which can, in turn, be passed on to extended family members by creating new businesses.

Regarding family businesses’ financing decisions, the extant literature provides evidence of non-traditional perspectives that consider owner-manager preferences. Such preferences involve control retention, risk aversion, and non-financial values and goals (Michiels and Molly 2017). For example, Levie and Lerner (2009) concluded that family businesses prefer family funding and non-family debt to non-family equity, which is consistent with their desire to retain control, even if this results in undercapitalization. However, behavioral agency models suggest that debt financing decisions are not static; instead, they are driven largely by loss aversion (Schickinger et al. 2022a). Furthermore, socio-emotional considerations may influence wealth management decisions because of the balance between financial stability and family/generational cohesion (Schickinger et al. 2021).

Evidence from developing countries has indicated that small family businesses’ management styles affect their capital structures. For example, economic, institutional, and financial environment factors are relevant to family-run small businesses’ leverage decisions, but less relevant to small family businesses with an outside manager (Bas et al. 2022).

In sum, family businesses must look after their financial stability to secure their continuity. Additionally, family businesses must consider the economic, financial, and institutional constraints that may impact their financing sources. These aspects are particularly relevant for small businesses that rely primarily on their own capital and, to a lesser extent, on banks and other financial institutions (Fernandez 2021). Ultimately, such considerations impact small family businesses’ investments in others’ new ventures.

Table 9 of the Appendix presents a summary of the articles most closely related to this study. Specifically, these articles focus on angel investors (De Clercq et al. 2012; Ding et al. 2015; Fernandez 2024; Honjo and Nakamura 2020; Maula et al. 2005; Xiao and Anderson 2022), small family businesses (Bas et al. 2022; Levie and Lerner 2009), family offices, and financing decisions of family firms (Block et al. 2019; Michiels and Molly 2017; Schickinger et al. 2022a).

Research hypotheses

Considering the above discussion, I base my hypotheses on the theory of household portfolios and resource-based view of the firm (Levie and Lerner 2009; Maula et al. 2005). Accordingly, I postulate that family businesses favor investing in their continuity and future development to maintain their family capital. The latter offers a competitive market advantage, in line with the vision of the resource-based strategy. Hence, family—compared with non-family—businesses are less likely to support third-party start-ups as a survival mechanism. However, when they become angel investors, family businesses are inclined to finance extended family start-ups to pass on their influence and family capital. The latter represents a structural dimension of familiness, which allows an increase in the density and connectivity of family ties (Cano-Rubio et al. 2016).

Hence, I formulate the following hypotheses:

H1: Family business rates are inversely associated with micro-angel activity.

H2: When investing in start-ups, family businesses prioritize endeavors initiated by extended family members.

H3: When engaging in angel investments, family businesses provide relatively fewer financial resources than comparable non-family businesses.

Family businesses generally focus on profitability to avoid jeopardizing the financial and social well-being of their future generations (Block et al. 2019). Therefore, uncertain investment alternatives, such as new business creation, may not always align with such objectives. Additionally, different communication and work styles may make family firms reluctant to fund new ventures (Schickinger et al. 2022b). However, in some cases, family firms may consider investing in newly founded companies to possibly shape their business development or pursue their own business ideas. In this regard, large family firms usually invest in existing and new firms through special types of family offices, such as family business venture corporates or family investment companies.Footnote 5

Small family businesses may view new ventures started by close family members or relatives as vehicles for passing on family capital. That is, family social networks would provide a wealth of resources for extended family members and new firms. Such resources may include reputation and prior experience in the industry, personal relationships, and a pre-established level of trust that attracts outside investors (Collins 2012, chapter 1). Regarding capital provided to new business ventures, whether owned or managed by family members or strangers, small family businesses are likely to have limited financial resources. The reason could be their preference for internal financing over over-indebtedness owing to considerations of high financial costs and loss aversion (Schickinge et al. 2022a).

In this regard, Bas et al. (2022) concluded that the corporate financing decisions of small family firms in developing countries differ from those of large firms. Specifically, small firms in such countries do not follow pecking order or trade-off theories, as asset tangibility—a measure of collateral—may not be positively associated with short-term leverage. Bas et al. (2022) argue that in the short term, firms rely on sources of financing that do not require fixed assets as collateral, such as trade credit or bank overdrafts, whereas long-term financing is secured against fixed assets. Furthermore, owing to information asymmetry and high inflation, small firms usually face higher interest costs. Therefore, debt financing may be expensive for these firms.

Figure 1 depicts the conceptual framework used to test the three hypotheses: Family participation in a business may involve ownership, management, or staff. These three aspects are embodied in a family participation index, the values of which depend on whether family members are involved in all, some, or none of these matters. Using this index, we compute the percentage of family businesses in a given country. This index is expected to impact the likelihood of becoming an angel investor (H1), type of start-up funded (H2), and amount invested in the start-up (H3).

Fig. 1

Full size image

Conceptual framework. Note Family participation in a business may involve ownership, management and/or staff. These three aspects are reflected in a family participation index, whose values depend on whether family members are involved in all, some or none of these aspects. From this index, in turn, the percentage of family businesses in a given country can be calculated. This index is expected to affect the probability of becoming an angel investor (H1), the type of start-up funded (H2), and the amount invested in such a new business (H3)

Data

This study relied on information collected by the GEM APS 2018 (https://www.gemconsortium.org/data). The GEM is a networked consortium of national country teams associated with leading academic institutions that collect data on entrepreneurship directly from individual entrepreneurs. The APS is a comprehensive questionnaire administered to a minimum of 2000 adults in each GEM country designed to collect detailed information on respondents’ business activities, attitudes, and aspirations. Nascent entrepreneurs are those who have taken steps to start a new business but have not paid salaries or wages for more than 3 months. New entrepreneurs are those running an operational business for 3–42 months (3.5 years), while established entrepreneurs are those who have been in business for more than 3.5 years.

The GEM APS 2018 comprised 162,077 individuals across 49 countries, of which approximately 50% were located in Europe, 23% in Asia and Oceania, and 18% in Latin America and the Caribbean. The GEM APS 2018 included a special topic on family entrepreneurship, whereby survey respondents were asked three questions regarding family involvement in a given start-up or established business (Kelley et al. 2020). Specifically, start-ups were asked the following questions: (i) Will this business, for the most part, be owned by you and your family and relatives? (sufamily1) (ii) Will this business be primarily managed by you, your family, and relatives? (sufamily2). (iii) Do you expect that most people who will work for this business in the next 5 years will be from your family and relatives? (sufamily3). Current owners/managers were asked similar questions, as follows: (i) Is this business, for the most part, owned by you and your family and relatives? (omfamily1). (ii) Is this business primarily managed by you, your family, and your relatives? (omfamily2). (iii) Are most people working in this business from among your family and relatives? (omfamily3). Approximately 31,000 survey participants answered the questions on family participation.

To quantify family members’ involvement as business owners, managers, and/or staff, I built a family participation index based on the aforementioned questions. This index takes values of 0, 1, 2, or 3. Specifically, the index equals 0 when family members do not participate in the business as owners, managers, or staff; 1 when family members participate as owners only, managers only, or staff only; 2 when family members are owners and managers, owners and staff, or managers and staff; and 3 when family members are owners, managers, and staff.

The most frequent values of the index were 2 (41%) and 1 (29%), followed by 0 (17%) and 3 (13%). Regarding family businesses’ characteristics (i.e., those with a non-zero family participation index), 58% of the respondents were owned or managed by men, 69% were sole proprietorships, 87% were small (less than five employees), and 35% were exported. By contrast, 67% of non-family businesses were owned or managed by men, 53% were sole proprietorships, 80% were small, and 41% were exported.

Regarding micro-angel activities, the APS asks individuals whether they have personally provided funds for a new business started by someone else in the past 3 years, excluding any purchases of stocks or mutual funds (busang). As mentioned in the Introduction, such investments include those made in family members’ start-ups. In 2018, of the 160,873 survey respondents, 5.95% answered busang affirmatively. Approximately 30% of respondents were owners/managers of family businesses. Noteworthily, the APS does not collect specific information regarding the financed start-up, such as its size, economic sector, or domestic/international orientation. The only information collected is the micro-angel investor’s relationship with the investee (e.g., a close family member or stranger).

Table 8 of the Appendix presents all study variables’ complete descriptions and sources in alphabetical order. These variables were selected based on previous literature on micro-angel investment/small family businesses that used GEM survey data (De Clercq et al. 2012; Ding et al. 2015; Fernandez 2023; Honjo and Nakamura 2020; Levie and Lerner 2009; Maula et al. 2005). As Table 8 indicates, one informal institutional factor under consideration is trust (Ding et al. op. cit., Fernandez 2021; Maung 2024; Xiao and Anderson 2022). This is approximated by the proportion of people who answered “a lot” or “some” to the following question: “How much do you trust people in your neighborhood?” Considering that a large percentage of the micro-angel investments documented by the GEM APS 2018 involve family members or acquaintances, this seems to be a suitable metric for social trust. Furthermore, as Fig. 3 in the Appendix illustrates, this metric is positively and highly correlated (0.61) with property rights protection (De Clercq et al. 2012).

Descriptive analysis

Table 1, Panels (a)–(d), indicates statistics of angel investments in US dollars made by survey respondents within the 3 years before the APS 2018. Specifically, Panel (a) reports mean, median, and total figures in US dollars according to the relationship between the money recipient and investor. Evidently, most micro-angel investors corresponded to the affiliated type referred to in the Introduction. Specifically, close family members received the most funding (41%), followed by friends or neighbors (23%) and, thereafter, strangers with a good business idea (14%). Overall, micro-angel funding amounted to $130.7 million over a 3-year period.

Table 1 Descriptive statistics of angel investments according to the GEM APS 2018 Full size table

Panel (b) reveals funding allocated to the recipient type versus the index value of family involvement in different business dimensions. Zero corresponds to a non-family business. As reported, non-family businesses invested predominantly in business start-ups with friends or neighbors (27%), close family members (21%), and strangers with a good business idea (21%). These findings contrast with those of survey respondents with non-zero family index values who favored close family members (37–40%). Moreover, family businesses with index values of 1 or 2 provided a considerably lower percentage of strangers with a good business idea (13–16%). The largest group of micro-angels comprised respondents with a family index value of 2 (1095 individuals), who invested approximately $22 million over a 3-year period. Regarding average investments, micro-angels belonging to non-family businesses were those that invested the most ($31,118), followed by those belonging to family businesses with an index value of 1 ($26,032). These figures are consistent with the investments of less-experienced business angels reported by Huang et al. (2017).

Figure 2a, b depict the average and median micro-angel investments made by family and non-family business entrepreneurs, respectively, according to their region of residence. For example, in North America, the average micro-angel investment made by non-family business entrepreneurs reached nearly $60,000, while that made by family business entrepreneurs reached nearly $38,000. In turn, the median investments in North America were noticeably lower—specifically, $30,000 among non-family business entrepreneurs and around $8000 among family business entrepreneurs. This discrepancy between average and median investments was also observed in other regions.

Fig. 2

Full size image

Angel investments in US dollars made by non-family and family business entrepreneurs. a Average investments, b Median investments. Note Family business entrepreneur is one that is involved in ownership, management, or staff or in some combination of such categories

In alignment with existing literature (Fernandez 2024; Honjo and Nakamura 2020; Huang et al. 2017), the APS 2018 suggests a gender gap in micro-angel investment activity.Footnote 6 Indeed, as Panel (c) indicates, the mean, median, and total informal funding provided by women was less than that provided by men in all geographic regions worldwide. For instance, in a 3-year period, European male micro-angels invested a total of $50 million compared with about $20 million for European female micro-angels. Concerning total investment, the largest gender gap was found in North America, with a male-to-female ratio of 2.8, whereas the smallest was found in Africa (1.02). However, in the latter region, female and male micro-angels allocated comparatively low amounts of money to start-ups.

Although this is not a topic pursued in this study, cultural aspects impact how male and female micro-angels invest. Specifically, Panel (d) reveals the impact of power distance and masculinity. The former is the extent to which the less powerful members of organizations and institutions accept and expect that power is distributed unequally; the latter is the extent to which the use of force is endorsed socially (Hofstede et al. 2010). Specifically, the table reports the impact of living in a country with a power distance above the sample average or masculinity above the sample average. The above-average power distance has a greater impact on the variation in men’s and women’s micro-angel investments than the above-average masculinity. Investments tend to be significantly lower in countries where power is unequally distributed than in those where it is not. This finding may not be surprising because a high power distance encourages bureaucracy, which, in turn, hinders entrepreneurial activity (Shah et al. 2015). One exception is that, in these societies, women seem to invest more in strangers’ start-ups.

Finally, Panel (e) reveals a list of countries sampled by the GEM APS 2018, ordered from the highest to the lowest average micro-angel investment. The top 5 countries were South Korea ($90,560), United Arab Emirates ($75,958), Luxembourg ($72,638), Switzerland ($69,917), and Cyprus ($67,494). However, average values were driven by outlying observations, as they were considerably above median values. For example, the median investment in Luxembourg reached only $11,686. Concerning total angel investment, the sampled countries that invested the most in a 3-year period were Spain ($15.4 million), Chile ($10.3 million), and China ($9.4 million).

Methodological aspects

The statistical modeling in “Results and discussion” section is based on logistic, ordered logistic, and multinomial logistic regressions (Liu 2016; chapters 3, 9, and 12), robust regressions (Andersen 2008), self-selection models (Greene 2018, chapter 19), and matching (Stuart 2010). Considering that logistic regression and its generalizations are well-known statistical techniques, I briefly refer to robust regressions, self-selection models, and matching. Specifically, the latter two allow for the accommodation of endogenous decisions. That is, choices are driven by interactions between economic agents and underlying economic environment, including market forces, institutional arrangements, and policy variables.

Robust regressions

Robust regression is a statistical technique used to model relationships between variables in the presence of outliers or violations of classical linear regression assumptions. Specifically, robust regression methods aim to provide more reliable parameter estimates by down-weighting the influence of outliers or modeling the data in a manner that is less affected by extreme observations. As mentioned earlier, the mean and median micro-angel investments differ significantly because of outlying observations. Therefore, ordinary least squares is an inadequate technique for modeling investment drivers.

Specifically, I used a robust regression model with Huber weights/bi-weights. This technique is a combination of a least-squares objective function and a least-absolute value objective function, such that observations are down-weighted as they proceed farther out on the tails (Andersen 2008, pp 18–19). For \(y_{i} = {\mathbf{x}}_{{\text{i}}}^{\prime } {{\varvec{\upbeta}}} +\upvarepsilon _{{\text{i}}}\), such that \(\upvarepsilon _{{\text{i}}} \left( {{\varvec{\upbeta}}} \right) = y_{i} - {\mathbf{x}}_{{\text{i}}}^{\prime } {{\varvec{\upbeta}}}\), a robust estimator of β solves:

$$ {{\varvec{\upbeta}}}_{{\mathbf{r}}} = \arg \;\min_{\upbeta } \mathop \sum \limits_{{{\text{i}} = 1}}^{{\text{n}}}\uprho \left( {\upvarepsilon _{{\text{i}}} \left( {{\varvec{\upbeta}}} \right)} \right) $$

(1)

where \(\uprho \left( {\text{z}} \right) = \left\{ {\begin{array}{*{20}l} {{\text{z}}^{2} ,} \hfill & {{\text{if}}\;\left| {\text{z}} \right| < {\text{c}}} \hfill \\ {\left| {2{\text{z}}} \right|{\text{c}} - {\text{c}}^{2} ,} \hfill & {{\text{if}}\;\left| {\text{z}} \right| \ge {\text{c}}} \hfill \\ \end{array} } \right.\) and \({\text{c}} \approx 1.345\).

Self-selection models

A self-selection model is a statistical framework used to analyze situations wherein individuals or entities choose between different groups or treatments based on their characteristics or preferences. That is, individuals or entities select themselves into particular categories or treatment groups based on observable or unobservable attributes (Greene 2018, pp 938–939). The key concept in self-selection models is that individuals or entities possess some degree of autonomy in choosing their actions or treatments, which can precipitate biases in estimating causal effects if not properly accounted for. Specifically, let us consider a two-stage process wherein the first stage comprises the decision to become a micro angel investor, and the second stage comprises how much money to invest in a given start-up.

Specifically, let the selection equation be:

$$ z_{i}^{*} = {{\varvec{\upgamma}}}^{\prime } {\mathbf{w}}_{{\text{i}}} + {\text{u}}_{{\text{i}}} $$

(2)

where \(z_{i} = \left\{ {\begin{array}{*{20}l} 1 \hfill & {{\text{if}}\;z_{i}^{*} > 0} \hfill \\ 0 \hfill & {{\text{if}}\;z_{i}^{*} \le 0} \hfill \\ \end{array} } \right.\) and P (\(z_{i}\) = 1) = Φ (γ′wi).

\(z_{i}^{*}\) is a latent variable, wi is a vector of observable characteristics, \(u_{i}\) is an error term, and Φ(.) is the cumulative distribution function of a standard normal.

The amount invested, yi, is observable only if \(z_{i} \) = 1; that is, the individual became a micro angel:

$$ {\text{y}}_{{\text{i}}} = {{\varvec{\upbeta}}}^{\prime } {\mathbf{x}}_{{\text{i}}} +\upvarepsilon _{{\text{i}}} $$

(3)

where \(\left( {{\text{u}}_{{\text{i}}} ,\upvarepsilon _{{\text{i}}} } \right) \sim {\text{N}}\left[ {\left( {\begin{array}{*{20}c} 0 \\ 0 \\ \end{array} } \right),{ }\left( {\begin{array}{*{20}c} 1 &\uprho \\\uprho & {\upsigma _{\upvarepsilon }^{2} } \\ \end{array} } \right)} \right]\), such that:

$$ {\text{y}}_{{\text{i}}} |z_{i}^{*} > 0 = {{\varvec{\upbeta}}}^{\prime } {\mathbf{x}}_{{\text{i}}} +\upbeta _{\uplambda }\uplambda _{{\text{i}}} + {\text{v}}_{{\text{i}}} $$

(4)

where βλ = ρσε, \(\updelta _{{\text{i}}} =\uplambda _{{\text{i}}} \left( {\uplambda _{{\text{i}}} - \upalpha _{{\text{i}}} } \right)\), \(\upalpha _{{\text{i}}} = - {{\varvec{\upgamma}}}^{\prime } {\mathbf{w}}_{{\text{i}}}\), \(\uplambda _{{\text{i}}} = \frac{{\upphi \left( {{{\varvec{\upgamma}}}^{\prime } {\mathbf{w}}_{{\text{i}}} } \right)}}{{{\Phi }\left( {{{\varvec{\upgamma}}}^{\prime } {\mathbf{w}}_{{\text{i}}} } \right)}}\), \({\text{Var}}\left( {{\text{v}}_{{\text{i}}} |{\mathbf{w}}_{{\text{i}}} , {\mathbf{x}}_{{\text{i}}} } \right) =\upsigma _{\upvarepsilon }^{2} \left( {1 -\uprho ^{2}\updelta _{{\text{i}}} } \right)\), and ϕ(.) where denotes the density function of the standard normal.

If ρ = 0, the error term of the selection equation (ui) is uncorrelated with the error term of the equation of interest (εi), such that Eq. (4) becomes yi = β′xi + εi. That is, the equation of interest can be estimated independently of the self-selection mechanism.

Matching methods

Matching techniques, frequently used in observational studies and causal inferences, aim to simulate a randomized controlled trial by creating comparable groups of participants or units based on observed characteristics. These techniques are valuable when random assignment to treatment and control groups is neither feasible nor ethical. Specifically, matching techniques help mitigate selection bias by creating comparison groups that are more similar in observed characteristics, thereby improving the validity of causal inferences in observational studies (Stuart 2010, pp 1–4).

I focused on nearest-neighbor matching, whereby each treated individual is matched with one or more control individuals with the closest values of the observed covariates. Matching can be performed with or without replacement depending on the specific method used. Specifically, the distance between the vectors of the covariates of individuals i and j, xi and xj, is parameterized by the vector norm:

$$ {\text{D}}_{{{\text{ij}}}} = \left\| {{\mathbf{x}}_{{\text{i}}} - {\mathbf{x}}_{{\text{j}}} } \right\|_{\Sigma } = \left\{ {({\mathbf{x}}_{{\text{i}}} - {\mathbf{x}}_{{\text{j}}} )^{\prime } \Sigma^{ - 1} ({\mathbf{x}}_{{\text{i}}} - {\mathbf{x}}_{{\text{j}}} )} \right\}^{1/2} $$

(5)

where Σ is a given symmetric and positive-definite matrix. For example, under the Mahalanobis metric, Σ is given by the sample variance–covariance of X in the pooled treatment and control groups if the focus of interest is the ATE. That is, the average difference in outcomes between individuals who receive treatment and those who do not, holding all other factors as constant.

Results and discussion

This section is divided into four subsections. “Family participation” section models the family involvement degree in a given business. “Becoming a micro-business angel” section models the probability of becoming a micro-angel investor when controlling for the country rate of family businesses. “Angel investment recipients” section focuses on the drivers of financial support for businesses started by extended family members or unrelated entrepreneurs, wherein the baseline is not investing. “Funding provided by micro-business angels” section focuses on the determinants of the amount invested in a given start-up. “Extensions: ATE of family business country rate and institutional quality” section presents extensions.

Family participation

Table 2 presents an ordered logistic model for the family participation index, which takes the values of 0 (non-family business), 1, 2, or 3. The predictor variables include survey respondents’ demographics—specifically, age (baseline: under 35 years old), gender (baseline: female), tertiary (or higher) education, household income (baseline: lowest 33 percentile), and number of household members; entrepreneurial culture conditions: risk aversion, self-efficacy, and opportunity alertness; business characteristics: ownership structure, number of employees, international orientation (exporting business), and economic sector (baseline: extractive sector); and, country-level characteristics: trust (baseline: low), cultural support for entrepreneurship, income group (baseline: low income), and geographic region (baseline: Africa). I select these predictors based on related literature (Ding et al. 2015; Honjo and Nakamura 2020; Levie and Lerner 2009; Uriarte et al. 2023). Parameter estimates are expressed as odds ratios relative to the baseline category, such that a given predictor variable is positively (negatively) associated with the dependent variable if its odds ratio is greater (less) than 1.

Table 2 Ordered logistic regression for family participation Full size table

Per the estimation results, a higher degree of family participation is positively and strongly associated with demographics such as age; middle 33-percentile income; household size; entrepreneurial culture conditions, such as risk aversion and opportunity alertness; business characteristics, such as small size (less than five employees); and country-level characteristics, such as intermediate and high levels of trust and cultural support for entrepreneurship. For example, for a small business, the odds of moving to an upper level of the family participation index are 50% higher than those for a larger business. In turn, in a high-trust country, the odds of greater family participation in a business are 36% higher than those in a low-trust country.

Meanwhile, in certain geographic regions such as Latin America and the Caribbean and North America, the odds of more family participation are approximately three times higher than those in Africa, the baseline region. For example, in Latin America, the founding of family businesses has largely resulted from immigration waves. Family businesses in the region have adopted organizational forms that are flexible to the institutional, financial, and market environments, as well as to existing technologies. In this sense, the essence of the largest family businesses in Latin America is the ability and intention to recombine resources over time to ensure the survival of the family in the business (Müller and Sandoval-Arzaga 2021, p 5).

Conversely, if owners/managers have higher educational levels, are men, have single proprietorship, and reside in a more developed country, this may discourage the participation of family members and relatives in the company. For example, for a male entrepreneur, the odds of moving to an upper level of family participation are approximately 28% lower than those for a female entrepreneur. Likewise, for an entrepreneur with tertiary or higher education, the corresponding odds are 26% lower than for a less-educated entrepreneur. Additionally, the odds for all economic sectors relative to the extractive sector are against family involvement. Thus, businesses that exhibit greater familiness will survive in relatively unattractive niches that probably require low levels of human capital. In this sense, family altruism indicates that the owners/managers of these businesses will persist in these niches for the sake of their families or because they have no other alternative to earn a living because of their low levels of human capital (Levie and Lerner 2009; Fernandez 2023).

Becoming a micro-business angel

Table 3 presents a logistic model elucidating the likelihood of becoming a microbusiness angel. The predictor variables were demographics (age, gender, education, and household income level), entrepreneurial culture conditions (risk aversion, personal network, self-efficacy, and opportunity alertness), employment status, business owner/manager status, country cultural support for entrepreneurship, trust level and related interactions, country rate of family businesses, and country income group. The rate of family businesses in a country is computed from the information provided by the GEM APS 2018 and corresponds to the percentage of businesses with a non-zero family index value in a given country. For example, 64% of the French businesses sampled were family businesses.

Table 3 Logistic regression for the decision to become a micro-angel investor Full size table

As Table 3 indicates, the odds of becoming a micro angel were higher for male and older individuals. Indeed, the odds for men were 20% higher than those for women and 17% higher for people over 55 than for those under 35. Simultaneously, accounting for tertiary or higher education, employment, and relatively high levels of household income also facilitates becoming a micro-angel. Moreover, the estimation results indicate that having experience as a business owner or knowing an entrepreneur (personal network) encourages supporting business start-ups. This is consistent with the findings of Maula et al. (2005) and Honjo and Nakamura (2020). For example, the latter two authors reported that, in a sample of Organisation for Economic Co-operation and Development countries, being a business owner/manager and knowing other entrepreneurs had the greatest impact on the odds of becoming a micro-angel investor.

Regarding country-level trust, noteworthily, higher levels of trust do not necessarily encourage micro-angel activity. As noted earlier, national trust is positively associated with the strength of property rights (Fig. 3). Hence, in countries with more property rights protection and higher levels of trust, individuals are likely to prioritize conventional over informal investments (Cumming and Zhang 2019; Ding et al. 2015). Interestingly, the interaction of trust with cultural support for entrepreneurship fosters micro-angel activity. For example, in countries that encourage entrepreneurial activities, an intermediate level of trust increases the odds of becoming a micro angel by 20%. Related literature has indicated that cultural support for entrepreneurial motivation may increase growth aspirations among early-stage entrepreneurs (Tominc and Rebernik 2007).

By contrast, for the extreme case of a one-unit increase (i.e., 100-percent increase) in the country rate of family businesses, the odds of becoming a micro angel would decrease by 61%. That is, the odds of micro-angel activity are 61% lower in a country with 100% of family businesses relative to one with 0% of such businesses. This evidence supports H1, substantiating the inverse association between family business rates and micro-angel activity. As expected, micro-angel activity was more likely to occur in richer countries. For example, in a high-income country, the odds of becoming a micro-angel are approximately 34% higher than those in a low-income country (baseline).

Angel investment recipients

As discussed in “Descriptive analysis” section, the GEM APS 2018 classified the money provided by micro-angel investors according to their relationship with recipients. Approximately 41% of micro-angel funding went to close family members (Table 1a). Based on these findings, Maula et al. (2005) and Ding et al. (2015), I distinguish between family and non-family angel investments because financial support given to extended-family start-ups is generally referred to as love money and is not strictly considered angel investment (Edelman et al. 2017); the interpersonal relationships between investors and entrepreneurs can bias investment decisions such that inferior-quality ventures are eventually financed (Ding et al. 2015). That is, the determinants of angel investments would differ depending on whether the funding recipient is an extended family member.

I denote family angel investments as those allocated to close family members or relatives (Table 1a). Consequently, the dependent variable has three categories—specifically, 1 if the survey respondent did not make angel investments, 2 if the survey respondent made angel investments in a business started by a close family member or some other relative, and 3 if the survey respondent made such investments in a business started by someone other than a close family member or relative. The predictor variables are analogous to those in the previous section. However, in addition to the country rate of family businesses, I considered the family participation index modeled in “Family participation” section.

The dependent variable is modeled through a multinomial logistic regression wherein the baseline category is 1, that is, if the survey respondent did not make micro-angel investments. Table 4 presents the estimation results. Parameter estimates are expressed as relative risk ratios, which can be interpreted as the change in the odds or relative risk of being in a particular category relative to the baseline for a one-unit change in a predictor variable while holding the other predictor variables constant. As in a binary logistic model, a positive (negative) regression coefficient in a multinomial logistic model corresponds to an odds ratio greater (less) than 1 (Liu 2016, chapter 12).

Table 4 Multinomial logistic model for micro-angel investments Full size table

Table 4 reports the following two formulations: Columns (1) and (2) include the country rate of family businesses as a predictor, while columns (3) and (4) include the family participation index. Collectively, column (1) reveals that micro-angel investments in extended-family start-ups are positively associated with investors’ age, owner/manager status, risk aversion, personal network, self-efficacy, opportunity alertness, tertiary or higher education, employment status, and household income. Among these predictors, personal network has the greatest impact. Indeed, the odds for a well-connected micro-angel of investing in extended family start-ups (relative to the baseline of not investing) are 2.6 times as large as the odds for an investor who lacks such entrepreneurial networks. Regarding country-level predictors, extended-family micro-angel investments are negatively associated with trust but positively associated with country income group. In this case, the interaction terms between trust and cultural support for entrepreneurship are insignificant.

As Column (2) indicates, some of these conclusions hold for micro-angel investments in non-family members, with some noteworthy exceptions. First, male or younger micro-angels are more likely to invest in non-family members’ start-ups. Second, the interaction between trust and cultural support positively and significantly impacted such investments. That is, in entrepreneurial countries with more social trust, people are encouraged to invest in start-ups run by entrepreneurs who are not family members—consistent with the findings of Ding et al. op. cit. Third, the national family business rate negatively impacts micro-angel investments in non-family members, whereas it exerts no impact on investments in family members.

Concerning magnitude, the estimation results suggest, for example, that the odds of a male micro-angel investing in non-family member start-ups (relative to the baseline) are 1.55 times as large as the odds for a female micro-angel. However, the statistical model suggests that angel investments are primarily allocated to extended family members in countries with a high proportion of family businesses. Indeed, the odds of investing in non-family members’ start-ups for an individual located in a country where all businesses are family-run or owned are only 0.12 times as large as the odds for an individual located in a country without family businesses.

Columns (3) and (4) report the model specifications with the family participation index as a predictor variable. In this case, the sample was constrained to individuals who answered the family participation questions. As the first three rows of Column (3) reveal, higher levels of familiness imply higher odds of investing in businesses started by extended family members. For example, for owners/managers of a business with a family participation index of 3, the odds of financing a family member are twice the odds of owners/managers of a non-family business (i.e., one with a family participation index = 0). Conversely, column (4) suggests that as familiness increases from 1 to 3, the odds of financing non-family member start-ups decrease slightly from 0.82 to 0.78. For the other predictor variables, most findings regarding their association with the dependent variable remained unchanged. One exception is household income, which is barely significant.

Therefore, these findings complement those of Maula et al. (2005) and Ding et al. (2015) in that the determinants of micro-angel investments differ depending on the funding recipient. Furthermore, these findings suggest that small family businesses allocate their informal investments differently than non-family businesses, thus supporting H2. In this regard, Maula et al. underscored that financing provided to unrelated entrepreneurs is likely driven by proactive investment decisions, while financing given to close family members is likely to be a reactive investment motivated by the need to support businesses owned by close family members.

Funding provided by micro-business angels

In this section, I model the amount of money allocated to angel investments by conducting a preliminary analysis wherein I modeled a two-stage process whereby an individual first decides whether to become an angel investor and then determines how much money they contribute. As discussed in “Self-selection models” section, the reason for modeling the two stages jointly is that unobservable factors impacting the likelihood of becoming a micro angel may be correlated with unobservable factors that influence how much money is invested.

Table 10 of the Appendix presents the fitted model. Column (1) corresponds to the decision to become a micro-angel, whereas Column (2) corresponds to the amount invested once the individual becomes a micro-angel. The US dollar investment is expressed as a natural logarithm because of its right-skewness (Table 1a, d). In Column (1), most predictor variables correspond to personal- and country-level characteristics. Column (2) also includes the characteristics of the business owned/managed by the micro-angel investor, such as familiness degree, economic sector, size approximated by the number of employees, and international orientation. As presented at the bottom of the table, the likelihood ratio test of the independent equations is insignificant at the 1% level (p value = 0.146).

Therefore, I model monetary investments independent from the decision to become a micro-business angel. This route has the advantage of using a regression technique that is robust to outlying observations (“Robust regressions” section). Therefore, US dollar investments can fit without the need for any transformation, which facilitates interpretation. Table 5 presents the estimation results. The model specifications in Columns (1) and (3) include the family participation index as a predictor, whereas those in Columns (2) and (4) include the country rate of family businesses. Another distinction is that columns (2) and (4) include a binary variable as a predictor of a micro-angel having exited their own business in the past 12 months (exit). This is meant to capture a micro-angel’s past entrepreneurial experience (Uriarte et al. 2023). To maintain consistency with Column (2) of Table 9, the remaining predictors of the model are considered analogous.

Table 5 Robust regression model for USD funds provided by micro business angels Full size table

The two family metrics suggest that family businesses invested less in start-ups. However, the family participation index is significant only when it equals 2. Companies with families involved in management and ownership, for example, or any other pair of combinations of management, ownership, and staff. Columns (1) and (3) reveal that family businesses, on average, invested between $617 and $659 less than their non-family counterparts. Columns (2) and (4) illustrate the impact of an increase in the country’s rate of family businesses on angel investments. For example, a 10-percentage point increment in this country’s rate would decrease angel investment by $981–$1029.

Regarding the micro-angel characteristics, gender, age, and household income are significant predictors of their investments. For example, micro angels belonging to the top 33-percentile income level invested $874–$1059 more than those belonging to the lowest 33-percentile income level. In turn, male micro-angels invested $424–$472 more than female micro-angels, while micro-angels over 55 invested $1357–$2098 more than those under 35. However, entrepreneurial experience (exit) is significant only in column (4). Specifically, a former business owner/manager was willing to invest $486 more than someone lacking that experience. Overall, household income appeared to make the largest difference.

Regarding the characteristics of the businesses managed/owned by micro-angel investors, size, international orientation, and economic sector impacted their investment. For example, owners/managers of businesses with fewer than five employees invested $575–$812 less than the owners/managers of larger businesses. Interestingly, micro angels from all economic sectors other than the extractive sector invested less—particularly in the business services and consumer-oriented sectors (Column (2)). That is, when other predictors are held constant, primary sector owner/manager micro-angels were more supportive of new business ventures.

Regarding the country-level variables, social trust and income positively impacted angel investments. For example, an angel from a medium-trust country invested $887–$1324 more than an angel from a low-trust country. That is, once an individual becomes a micro-business angel, higher levels of trust encourage them to invest more. That is, trust improves information transmission, collaboration, and sanctioning mechanisms in a society through moral obligations (Ding et al. 2015; Xiao and Anderson 2022). Similarly, an angel from a high-income country invested $2045–$2472 more than that from a low-income country.

Thus far, I have considered total investments without distinguishing between recipient types. To this end, I follow the model specification in Column (4) and include the interaction terms between the two new binary variables. One equals 1 if the funding recipient is a close family member or relative. The other equals 1 if the country’s rate of family businesses is above the median of all sampled countries. Therefore, the regression model has the following three interaction terms: non-family recipient × above median family business rate, family recipient × below median family business rate, and family recipient × above median family business rate; here, being the non-family recipient × below median family business rate interaction term is the baseline. Table 6 presents the estimation results.

Table 6 Robust regression model with interaction terms between investee and national family business rate Full size table

In countries with above-median family business rates, non-family members received $1370 less than in countries with below-median family business rates. Likewise, in countries with above-median family business rates, family members received $632 less than in those with below-median family business rates. That is, micro-angels consistently underinvest in countries with a high prevalence of family businesses, thus supporting H3. However, family members received $1066 more than non-family members in countries with below-median family business rates. That is, extended family members were favored, even in countries with low familiness ratios.

Extensions: ATE of family business country rate and institutional quality

In this section, I focus on the following two extensions: First, I quantify the impact of living in a country with an above-median family business rate on the odds of financing family members and strangers and those of providing above-median funding. Second, considering that I assessed only the informal institutional factors of social trust, I now examine the impact of the quality of formal institutions on the aforementioned odds. To these ends, I used nearest-neighbor matching (NNM). As discussed in “Matching methods” section, NNM is used to match items or observations based on their similarity or proximity to each other. In essence, NNM uses the average of the individuals that are most similar but obtain the other treatment level (e.g., below-median family business country rate) to predict the unobserved potential outcome (e.g., financing strangers).

Panels (a)–(g) of Table 7 present the estimation results. The covariates used for matching are investor age, gender, risk aversion, self-efficacy, personal network, opportunity alertness, tertiary education, household income, household size, economic sector, trust, cultural support for entrepreneurship, internationally oriented business, and country income group. Panel (a) indicates that, in countries with a high prevalence of family businesses, the odds of micro-angels financing family members are 7.2 percentage points higher. A 95% confidence interval for this ATE ranges from 1.8 to 12.7 percentage points. By contrast, in such countries, the odds of micro-angels financing strangers with good business ideas are 3.7 percentage points lower. The associated 95% confidence interval is (− 6.3, − 1.1) percentage points.

Table 7 Nearest-neighbor matching estimators: impact of national family business rate and institutions Full size table

Concerning monetary investments, panel (c) shows that in countries with a high prevalence of family businesses, the likelihood for micro-angels of providing above-median funding to any type of entrepreneur is 6 percentage points lower, with an associated 95%-confidence interval of (− 10.9, − 1.0) percentage points. Interestingly, as panel (d) indicates, in such countries, micro angels are − 9.4 percentage points less likely to provide above-median funding to family members. These odds range from − 16 to − 2.8 percentage points with a 95% confidence level—consistent with those of the robust regressions in Table 6.

In Panels (e)–(g), I repeat the analysis considering the ATE of high institutional quality (see Table 8 for the definition). The odds of financing family members decrease by 16 percentage points, whereas those of financing strangers with good business ideas increase by 11 percentage points in countries with strong institutions. Moreover, the odds of providing above-median funding to start-ups increase by 39 percentage points in such countries. This evidence complements that of De Clercq et al. (2012) in that a protective legal environment stimulates micro-angel activity focused on higher-quality start-ups. De Clercq et al. op cit. highlighted that when legal rights are respected, the illegal use of start-ups’ intellectual capital by third parties is unlikely, and the potential returns from micro-angel investments increase. Simultaneously, higher legal protection levels can reduce the cost and time required to select and evaluate investment opportunities.

Conclusions

Micro-angel investors provide small amounts of funding to start-ups or early-stage companies in exchange for equity ownership. They typically invest smaller sums of money than traditional angel investors, with investments ranging from a few hundred to a few thousand dollars. Overall, micro-angel investors contribute to the start-up funding ecosystem by providing early-stage capital and support to entrepreneurs, helping fuel innovation and economic growth. This study focused on micro-angel investments of around 30,000 family and non-family business owners/managers from 49 countries worldwide, drawn from the GEM APS 2018.

This survey included three questions regarding the participation of family members in the ownership, management, and staff of established or new businesses. Such questions were not asked in previous or later versions of the GEM APS; therefore, they constitute a unique set of information. Based on these three questions, I built a family participation index that takes numerical values ranging from 0 (no family participation) to 3 (full family participation). Based on this index, I computed the national family business rate, measured as the percentage of businesses that had a family participation index value greater than or equal to 1 in a given country—that is, businesses wherein a family is involved in at least one of the following activities: ownership, management, or staff.

Based on this national family business rate, an approximation of social trust, and several aspects of the entrepreneurial process, which are regularly covered by the GEM APS, I modeled the likelihood of becoming a micro-angel investor in new businesses started by extended family and non-family members. Additionally, I modeled the determinants of monetary investments in start-ups and analyzed the impact of institutional quality on the odds of financing family members and unrelated entrepreneurs. To this end, I based my empirical approach on household portfolio theory and the resource-based view of the firm. The former allows us to understand the determinants of household investments in risky assets, whereas the latter establishes that family businesses can create a competitive advantage from their social capital. I postulate that family capital can be passed on to new ventures initiated by family members.

Main findings

Small business owners/managers possess localized knowledge and insights into market trends, customer preferences, and operational challenges within their specific communities. This localized expertise is invaluable for start-ups seeking to establish themselves in similar markets. Beyond financial investment, small business owners/managers can mentor start-up founders by providing them with practical advice and strategic directions. This mentorship helps build the start-up team’s capacity, enhancing its ability to manage and grow its business. Experienced mentors can impart lessons learned from their successes and failures, providing start-ups with a more realistic view of the entrepreneurial journey (Falcão et al 2023).

This study demonstrates that family and non-family business entrepreneurs adopt different investment strategies. Indeed, the results revealed that angel activity is lower in countries with high rates of small family businesses (Hypothesis 1) and that family businesses favor start-ups of family members and close relatives (Hypothesis 2). Furthermore, after controlling for various factors, I conclude that family businesses tend to provide less financing than comparable non-family businesses, irrespective of whether their relationship with the entrepreneur is supported (Hypothesis 3). I infer that this could be explained by the financing decisions of family businesses, which favor internal resources, such as retained earnings, over bank debt and other financing options (Bas et al. 2022; Schickinger et al. 2022a). That is, within this theoretical framework, family businesses, particularly small ones, would have less financial leeway to support new third-party businesses than comparable non-family businesses.

Regarding countries with strong institutions, I conclude that start-ups of entrepreneurs with good business ideas are more likely to be funded than those of family members. That is, in such countries, micro-angel investors are more selective in their investment choices because their legal rights are more likely to be respected, indicating that micro-angel investments are guided by not only altruistic motivations but also property rights considerations (Cumming and Zhang 2023; De Clercq et al. 2012).

Limitations and future research

Considering that most family businesses surveyed by the GEM APS 2018 were small, the findings are not necessarily generalizable to larger family businesses. Furthermore, the GEM APS 2018 considered only three questions regarding family engagement in a business. However, no information was gathered on socio-emotional aspects, such as the identification of family members with the firm or bonding social ties (Gómez-Mejías and Herrero 2022), which could have enriched the analysis.

Another limitation of the GEM APS 2018 is that it does not provide information regarding the financial aspects of the sampled businesses. The only available information is household income, which can be considered an approximation of business profitability (Peroni et al. 2016). However, no further information was provided to assess indebtedness or financial constraints (Ergün and Doruk 2020). A more general limitation of the GEM APS 2018 is that it probably captures a broader and less selective population of start-ups than formal venture capital funds or angel investors (Edelman et al. 2017). In this sense, understanding the characteristics of start-ups financed by micro-business angels would be enlightening. As mentioned above, blood ties may lead angels to finance unsuitable business ventures.

Hence, future research on micro-angel investments in small family businesses should ideally consider a dynamic context wherein researchers can infer changes in family capital arising from such investments. Additionally, future research should quantify how the financial constraints of small family businesses limit increases in family capital through micro-angel investments.

Furthermore, extensive literature exists on the impact of culture on entrepreneurial activities and innovation (Shah et al. 2015). In this regard, how power distance and individualism, for example, may impact micro-angel investments by family and non-family entrepreneurs is worth exploring. As discussed in “Descriptive analysis” section, an unequal distribution of power may discourage investment by men and women.

Policy implications

Generally, small business owners/managers may invest in varied start-ups, diversifying their portfolios, and spreading risk. This diversification could lead to more sustainable financial support for ecosystems. Moreover, their investments can frequently come with more flexible terms compared to traditional venture capital, rendering it easier for start-ups to access the necessary funds (Tenca et al. 2018).

In this respect, this study revealed that high national rates of small family businesses would inhibit angel investments, particularly those in strangers’ start-ups. As previously discussed, family businesses frequently exhibit a preference for internally generated funds over external financing owing to their unique characteristics and values. This preference can impact their ability to support third-party ventures and shape their investment choices, particularly favoring ventures initiated by extended family members. This insularity can translate into less openness to external advice and innovation, preferring to rely on internal knowledge and traditions (Biel and Ślutsarczyk 2023, chapter 1).

Therefore, policymakers should encourage angel investment in a broad spectrum of new businesses. This is because, although investments in new extended family businesses can be a mechanism to expand family capital, this might come at the expense of the quality of the funded businesses (Ding et al. 2015). Fostering investment in innovative new ventures can be achieved, for example, through tax incentives, to the extent that the elasticities of tax-related investments may significantly impact the propensity to make micro-angel investments (Maula et al. 2005). Moreover, effective tax policy incentives help reduce financial risk perception and, consequently, increase the number of investors (Solodoha et al. 2023). However, such a policy should ensure that tax incentives do not attract inexperienced investors who channel investments into more marginal business enterprises (Harrison et al. 2020). Another important consideration is policy clarity. For example, ambiguity in its formulation can reduce the number of angel investors per company (Solodoha et al. op. cit.)

Other mechanisms might involve government programs that provide entrepreneurial training and seed capital (Uriarte et al. 2023; Yigitcanlar et al. 2019), incentivize the creation of corporate accelerators (Seitz et al. 2023), or support non-corporate-led accelerators, such as universities (Basco et al. 2018). In this respect, university accelerators offer the potential for collaboration and association with the university community, which offers expertise and resources that are challenging to find elsewhere (Seitz et al. 2023).

Availability of data and materials

The dataset used in this study is available at https://www.gemconsortium.org/data.

Notes

Micro-angel investments fall into the category of informal venture capital, that is, the provision of capital by individuals or groups outside traditional institutional channels, such as venture capital firms or banks, to support early-stage and growing businesses. Unlike formal venture capital, which typically involves professional investors managing pooled funds, informal venture capital often involves individual investors who provide funding on a more ad-hoc basis (e.g., Amore et al. 2023).

That is, the structures that regulate how start-ups and entrepreneurs interact with other stakeholders (Wadichar et al. 2024).

Household portfolios comprise a variety of assets—financial (e.g., liquid accounts, stocks, bonds, and mutual fund shares) and real assets (e.g., primary residence, real estate investment, and private businesses), as well as liabilities held by households, such as mortgages and consumer debt (Haliassos 2018).

Alternative theories of family businesses include agency theory, stewardship theory, and the socio-emotional wealth (SEW) model (e.g., Wu et al. 2007; Young et al. 2008; Michiels and Molly 2017; Swab et al. 2020; Gómez-Mejía and Herrero 2022: Reina et al. 2023). However, these are not considered here.

Statistics show that during 2021, around 6500 family offices worldwide invested $1617 million in seed-stage companies, $2026 million in early-stage companies, and $2449 in later-stage companies (PwC 2023).

There is related literature that shows a disparity in the amount of angel funding received by male and female entrepreneurs. According to Poczter and Shapsis (2018), in entrepreneurial pitch competitions female teams receive less capital and provide more equity than their male counterparts.

Abbreviations

APS:

Adult Population Survey

ATE:

Average treatment effect

GEM:

Global Entrepreneurship Monitor

NNM:

Nearest-neighbor matching

References

Ajzen I (1991) The theory of planned behavior. Organ Behav Hum Decis Process 50(2):179–211. https://doi.org/10.1016/0749-5978(91)90020-t

Article

Google Scholar

Amore MD, Conti A, Pelucco V (2023) Micro venture capital. Strateg Entrep J 17(4):886–924. https://doi.org/10.1002/sej.1478

Article

Google Scholar

Andersen R (2008) Modern methods for robust regression. Sage Publications, Inc.

Book

Google Scholar

Bas T, Muradoglu YG, Phylaktis K (2022) Capital structures of small family firms in developing countries. Rev Corp Finance 2(4):745–790. https://doi.org/10.1561/114.00000029

Article

Google Scholar

Basco D, Gutierrez CI, Graf M (2018) The role of public policy in supporting business accelerators. Santa Monica, CA: RAND Corporation. https://www.rand.org/pubs/working_papers/WR1271.html

Biel M, Ślutsarczyk B (2023) Family business and management. Objectives, theory, and practice. Routledge Advances in Management and Business Studies, New York

Google Scholar

Blaseg D, Hornuf L (2024) Playing the business angel: the impact of well-known business angels on venture performance. Entrep Theory Pract 48(1):171–204. https://doi.org/10.1177/10422587231153603

Article

Google Scholar

Block J, Fisch C, Vismara S, Andres R (2019) Private equity investment criteria: an experimental conjoint analysis of venture capital, business angels, and family offices. J Corp Finan 58:329–352. https://doi.org/10.1016/j.jcorpfin.2019.05.009

Article

Google Scholar

Cano-Rubio M, Fuentes-Lombardo G, Hernández-Ortiz MJ, Vallejo-Martos MC (2016) Composition of familiness: perspectives of social capital and open systems. Eur J Fam Bus 6(2):75–85. https://doi.org/10.1016/j.ejfb.2016.12.002

Article

Google Scholar

Collins L (2012) The family business. In: The modern family business. Palgrave Macmillan, London. https://doi.org/10.1057/9781137001337_1

Croce A, Ughetto E, Bonini S, Capizzi V (2020) Gazelles, ponies, and the impact of business angels’ characteristics on firm growth. J Small Bus Manag 59(2):223–248. https://doi.org/10.1080/00472778.2020.1844489

Article

Google Scholar

Cumming D, Zhang M (2019) Angel investors around the world. J Int Bus Stud 50:692–719. https://doi.org/10.1057/s41267-018-0178-0

Article

Google Scholar

Cumming D, Zhang M (2023) Bankruptcy law and angel investors around the world. J Int Bus Stud. https://doi.org/10.1057/s41267-023-00627-w

Article

Google Scholar

De Clercq D, Meuleman M, Wright M (2012) A cross-country investigation of micro-angel investment activity: the roles of new business opportunities and institutions. Int Bus Rev 21(2):117–129. https://doi.org/10.1016/j.ibusrev.2011.02

Article

Google Scholar

Deng L, Ye Q, Xu D, Sun W, Jiang G (2022) A literature review and integrated framework for the determinants of crowdfunding success. Financ Innov 8:41. https://doi.org/10.1186/s40854-022-00345-6

Article

Google Scholar

Ding Z, Au K, Chiang F (2015) Social trust and angel investors’ decisions: a multilevel analysis across nations. J Bus Ventur 30(2):307–321. https://doi.org/10.1016/j.jbusvent.2014.08.003

Article

Google Scholar

Edelman LF, Manolova TS, Brush CG (2017) Angel investing: a literature review. Found Trends® Entrep 13(4–5): 265–439. https://doi.org/10.1561/0300000051

Ergün B, Doruk ÖT (2020) Effect of financial constraints on the growth of family and nonfamily firms in Turkey. Financ Innov 6:28. https://doi.org/10.1186/s40854-020-00188-z

Article

Google Scholar

Estapé-Dubreuil G, Ashta A, Hédou JP (2016) Micro-equity for sustainable development: selection, monitoring and exit strategies of micro-angels. Ecol Econ 130:117–129. https://doi.org/10.1016/j.ecolecon.2016.06.0

Article

Google Scholar

Falcão R, Carneiro MJ, Moreira AC (2023) Why do business angels invest? Uncovering angels’ goals. Cogent Bus Manag 10:2. https://doi.org/10.1080/23311975.2023.2234134

Article

Google Scholar

Farrell E, Howorth C, Wright M (2008) A review of sampling and definitional issues in informal venture capital research. Ventur Cap 10(4):331–353. https://doi.org/10.1080/13691060802151986

Article

Google Scholar

Fernandez V (2021) The role of trust and social commitment in start-up financing. Int Rev Financ Anal 75:101722. https://doi.org/10.1016/j.irfa.2021.101722

Article

Google Scholar

Fernandez V (2023) Family entrepreneurship around the world. Int Rev Financ Anal 89:102808. https://doi.org/10.1016/j.irfa.2023.102808

Article

Google Scholar

Fernandez V (2024) Micro-angel investments of men and women: the role of institutions. Res Int Bus Financ 70:102378. https://doi.org/10.1016/j.ribaf.2024.102378

Article

Google Scholar

Frimanslund T, Kwiatkowski G, Oklevik O (2022) The role of finance in the literature of entrepreneurial ecosystems. Eur Plan Stud 31(2):372–391. https://doi.org/10.1080/09654313.2022.2055962

Article

Google Scholar

Gómez-Mejía LR, Herrero I (2022) Back to square one: the measurement of Socioemotional Wealth (SEW). J Fam Bus Strat 13:100480. https://doi.org/10.1016/j.jfbs.2021.100480

Article

Google Scholar

Greene W (2018) Econometric analysis, 8th edn. Pearson Education

Google Scholar

Habbershon TG, Williams ML (1999) A resource-based framework for assessing the strategic advantages of family firms. Fam Bus Rev 12(1):1–25. https://doi.org/10.1111/j.1741-6248.1999.00001.x

Article

Google Scholar

Haliassos M (2018) Household portfolios. In: The new Palgrave dictionary of economics. Palgrave Macmillan, London. https://doi.org/10.1057/978-1-349-95189-5_2098

Harrison RT, Bock AJ, Gregson G (2020) Stairway to heaven? Rethinking angel investment policy and practice. J Bus Ventur Insights 14:e00180. https://doi.org/10.1016/j.jbvi.2020.e00180

Article

Google Scholar

Honjo Y, Nakamura H (2020) The link between entrepreneurship and informal investment: an international comparison. Jpn World Econ. https://doi.org/10.1016/j.japwor.2020.101012

Article

Google Scholar

Hofstede G, Hofstede G, Minkov M (2010) Cultures and organizations: software of the mind, 3rd edn. McGraw Hill

Google Scholar

Hsu DK, Haynie JM, Simmons SA, McKelvie A (2014) What matters, matters differently: a conjoint analysis of the decision policies of angel and venture capital investors. Ventur Cap 16(1):1–25. https://doi.org/10.1080/13691066.2013.825527

Article

Google Scholar

Huang L, Wu A, Lee MJ, Bao J, Hudson M, Bolle E (2017) The American angel: the first in-depth report on the demographics and investing activity of individual American angel investors. https://www.hbs.edu/faculty/Pages/item.aspx?num=57843

Kelley D, Gartner WB, Allen M (2020) Global entrepreneurship monitor family business report. Babson Park: Babson College Press, Babson Park, MA. https://www.gemconsortium.org/report/20192020-global-entrepreneurship-monitor-gem-family-entrepreneurship-report-2

Lerner J, Schoar A, Sokolinski S, Wilson K (2018) The globalization of angel investments: evidence across countries. J Financ Econ 127(1):1–20. https://doi.org/10.1016/j.jfineco.2017.05.012

Article

Google Scholar

Levie J, Lerner M (2009) Resource mobilization and performance in family and nonfamily businesses in the United Kingdom. Family Business Rev 22(1):25–38. https://doi.org/10.1177/0894486508328812

Article

Google Scholar

Liu X (2016) Applied ordinal logistic regression using Stata. Sage Publications Inc

Google Scholar

Lortie J, Castogiovanni G (2015) The theory of planned behavior in entrepreneurship research: what we know and future directions. Int Entrep Manag J 11(4):935–957. https://doi.org/10.1007/s11365-015-0358-3

Article

Google Scholar

Mason C, Botelho T, Harrison R (2019) The changing nature of angel investing: some research implications. Ventur Cap 21(2–3):177–194. https://doi.org/10.1080/13691066.2019.1612921

Article

Google Scholar

Maula M, Autio E, Arenius P (2005) What drives micro-angel investments? Small Bus Econ 25(5):459–475. https://doi.org/10.1007/s11187-004-2278-4

Article

Google Scholar

Maung M (2024) The bright sight of social trust and entrepreneurial finance. Int Rev Econ Financ 92:778–795. https://doi.org/10.1016/j.iref.2024.02.046

Article

Google Scholar

Michiels A, Molly V (2017) Financing decisions in family businesses: a review and suggestions for developing the field. Fam Bus Rev 30(4):369–399. https://doi.org/10.1177/0894486517736958

Article

Google Scholar

Müller CG, Sandoval-Arzaga F (2021) Family business heterogeneity in Latin America. A historical perspective. In: Palgrave studies in family business heterogeneity. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-030-78931-2_1

Peroni C, Riillo CAF, Sarracino F (2016) Entrepreneurship and immigration: evidence from GEM Luxembourg. Small Bus Econ 46(4):639–656. https://doi.org/10.1007/s11187-016-9708-y

Article

Google Scholar

Poczter S, Shapsis M (2018) Gender disparity in angel financing. Small Bus Econ 51:31–55. https://doi.org/10.1007/s11187-017-9922-2

Article

Google Scholar

Preston S (2007) Angel financing for entrepreneurs: early-stage funding for long-term success, 1st edn. Jossey-Bass

Google Scholar

PwC (2023) How family offices globally are investing in a brighter future. https://www.pwc.com/gx/en/services/family-business/family-office/family-office-startup-investments-worldwide-2023.html

Reina W, Pla-Barber J, Villar C (2023) Socioemotional wealth in family business research: a systematic literature review on its definition, roles and dimensions. Eur Manag J 41(6):1000–1020. https://doi.org/10.1016/j.emj.2022.10.009

Article

Google Scholar

Schickinger A, Bierl PA, Leitterstorf MP, Kammerlander N (2021) Family-related goals, entrepreneurial investment behavior, and governance mechanisms of single family offices: an exploratory study. J Fam Bus Strat. https://doi.org/10.1016/j.jfbs.2020.100393

Article

Google Scholar

Schickinger A, Bertschi-Michel A, Leitterstorf MP, Kammerlander N (2022a) Same same, but different: capital structures in single family offices compared with private equity firms. Small Bus Econ 58:1407–1425. https://doi.org/10.1007/s11187-021-00448-x

Article

Google Scholar

Schickinger A, Bertschi-Michel A, Kammerlander N (2022) Family offices as startup investors: A synergetic relationship of the old and new economy? In: Lingelbach D (ed) De Gruyter handbook of entrepreneurial finance. De Gruyter, Berlin, Boston, pp 265–288. https://doi.org/10.1515/9783110726312-018

Chapter

Google Scholar

Seitz N, Lehmann EE, Haslanger P (2023) Corporate accelerators: design and startup performance. Small Bus Econ. https://doi.org/10.1007/s11187-023-00817-8

Article

Google Scholar

Shah R, Gao Z, Mittal H (2015). Culture. In: Innovation, entrepreneurship, and the economy in the US, China, and India, pp 55–65. https://doi.org/10.1016/b978-0-12-801890-3.00004-6

Solodoha E, Rosenzweig S, Harel S (2023) Incentivizing angels to invest in start-ups: evidence from a natural experiment. Res Policy 52(1):104634. https://doi.org/10.1016/j.respol.2022.104634

Article

Google Scholar

Stuart EA (2010) Matching methods for causal inference: a review and a look forward. Stat Sci 25(1):1–21. https://doi.org/10.1214/09-STS313

Article

Google Scholar

Swab RG, Sherlock C, Markin E, Dibrell C (2020) “SEW” What do we know and where do we go? A review of socioemotional wealth and a way forward. Fam Bus Rev 33(4):424–445. https://doi.org/10.1177/0894486520961938

Article

Google Scholar

Tenca F, Croce A, Ughetto E (2018) Business angels research in entrepreneurial finance: a literature review and a research agenda. J Econ Surv 32:1384–1413. https://doi.org/10.1111/joes.12224

Article

Google Scholar

Tokarczyk J, Hansen E, Green M, Down J (2007) A resource-based view and market orientation theory examination of the role of “familiness” in family business success. Fam Bus Rev 20(1):17–31. https://doi.org/10.1111/j.1741-6248.2007.00081.x

Article

Google Scholar

Tominc P, Rebernik M (2007) Growth aspirations and cultural support for entrepreneurship: a comparison of post-socialist countries. Small Bus Econ 28:239–255. https://doi.org/10.1007/s11187-006-9018-x

Article

Google Scholar

Uriarte S, Espinoza-Benavides J, Ribeiro-Soriano D (2023) Engagement in entrepreneurship after business failure. Do formal institutions and culture matter? Int Entrep Manag J. https://doi.org/10.1007/s11365-023-00829-6

Article

Google Scholar

Xiao L, Anderson A (2022) The evolution of Chinese angels: social ties and institutional development. Br J Manag 33:69–87. https://doi.org/10.1111/1467-8551.12495

Article

Google Scholar

Wadichar RK, Manusmare P, Burghate MA (2024) Entrepreneurial ecosystem: a systematic literature review. Vision 28(2):143–156. https://doi.org/10.1177/09722629221093866

Article

Google Scholar

Wessel S, Decker C, Lange KSG, Hack A (2014) One size does not fit all: entrepreneurial families’ reliance on family offices. Eur Manag J 32(1):37–45. https://doi.org/10.1016/j.emj.2013.08.003

Article

Google Scholar

Wu Z, Chua JH, Chrisman JJ (2007) Effects of family ownership and management on small business equity. J Bus Ventur 22:875–895. https://doi.org/10.1016/j.jbusvent.2006.07.002

Article

Google Scholar

Yigitcanlar T, Sabatini-Marques J, da-Costa EM, Kamruzzaman M, Ioppolo G (2019) Stimulating technological innovation through incentives: perceptions of Australian and Brazilian firms. Technol Forecast Soc Change 146:403–412. https://doi.org/10.1016/j.techfore.2017.05

Article

Google Scholar

Young MN, Peng MW, Ahlstrom D, Bruton GD, Jiang Y (2008) Corporate governance in emerging economies: a review of the principal–principal perspective. J Manag Stud 45:196–220. https://doi.org/10.1111/j.1467-6486.2007.00752.x

Article

Google Scholar

Download references

Acknowledgements

I am thankful to the anonymous reviewers for their constructive comments.

Funding

Financial support from Programa de Apoyo a la Investigación (PAI) 2023, Universidad Adolfo Ibañez, and FONDECYT Grant No. 1240098 is greatly acknowledged.

Authors and Affiliations

Universidad Adolfo Ibáñez, Escuela de Negocios, Santiago, Chile

Viviana Fernandez

Authors Viviana Fernandez

View author publications

Search author on:PubMed Google Scholar

Contributions

Writing and statistical analysis.

Corresponding author

Correspondence to Viviana Fernandez.

Competing interests

The author declare that they have no competing interests.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

See Tables 8, 9, 10 and Fig. 3.

Table 8 List of variables in alphabetical order Full size table

Table 9 Existing literature closely related to this study Full size table

Table 10 Self-selection model to become a micro angel and subsequently invest Full size table

Fig. 3

Full size image

Source: tcdata360.worldbank.org

Social trust versus property rights. Notes (1) Trust is approximated by the share of people who answered, “a lot” or “some” to the question: “How much do you trust people in your neighborhood?” Source: ourworldindata.org. (2) Property rights protection is on a scale of 1–7 (best).

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions