Groups, networks and business angels’ investment process.

Stefano Bonini

Stevens Institute of Technology

School of Business

1 Castle Point Terrace, Hoboken, NJ 07030, USA

Vincenzo Capizzi *

Department of Economics and Business Studies

Università del Piemonte Orientale

Via E. Perrone, 18, 28100, Novara, Italy

Mario Valletta

Department of Economics and Business Studies

Università del Piemonte Orientale

Via E. Perrone, 18, 28100, Novara, Italy

Paola Zocchi

Department of Economics and Business Studies

Università del Piemonte Orientale

Via E. Perrone, 18, 28100, Novara, Italy

This draft: July 6th 2016

JEL Codes: G24, G32, M13

Keywords: business angels, business angel networks, investment process, experience, monitoring,

networking, co-investment

____________________________________________________________________________________

* Corresponding Author. Tel.: +39 0321 375.438. Email: vincenzo.capizzi@uniupo.it

2

Abstract

This paper provides first-time evidence on the effects of angel groups and networks (AG/BAN)

membership on the investment choices of business angels. Looking at a proprietary dataset that collects

qualitative and quantitative information on over 800 investments by 625 business ange l s f rom 2008 to

2014, we show that AG/BAN membership generates valuable information, networking, monitoring and

risk reduction effects which, ultimately, affects the amount of financial resources available to invest as well as

the equity stake in the investee company. These results extend our knowledge of the behavior and

characteristics of business angels investing that are rapidly gaining prominence in the support of new ventures

and the development of global economies.

3

1. Introduction

In the last few years both academic and practitioners have devoted growing attention to

understanding the dynamics of business angels investment. Market data at both the US and

European level (US ACA, 2015; EVCA, 2014; EBAN, 2015; Kraemer-Eis et al., 2015;

OECD, 2016) show that business angels have become a major segment of the capital

market industry, capable of allocating financial resources to one of the riskiest asset class –

startup companies. As such, they are invaluable enablers of the development of economic

and social systems. Regulatory Authorities all around the world have been paying attention

to this investor class1 who nowadays is officially recognized as one component of the

shadow banking industry. Despite this recent attention our understanding of the features

of business angel investments is still very limited. In particular, little is known of the

differential investment behavior of business angels when they join semi-formal

organizations such as Angel Groups or Business Angel Networks. In recent contribution

Kerr et al. (2014) provide case-study based evidence of the impact of angel groups on later

round financing. Yet, their sample doesn’t allow to shed light on the differential (if any)

investment behavior of BAs within and outside a semi-formal organization. This paper

aims at filling this gap.

Business angels are high net worth individuals who invest their own money in small

unlisted companies, with no family connection, assuming typically a minority equity stake

as well as an active involvement in the portfolio companies (Mason, 2006). Furthermore, i t

is now widely accepted that business angels are amongst the most suitable actors of the

ecosystem for entrepreneurial businesses, considering their capability to fill the so-called

“funding gap” existing between demand and supply of early stage equity capital (Mason

and Harrison, 2000; Johnson and Sohl, 2012; Capizzi, 2015). First, business angels satisfy a

size of investment need (usually falling in the range 100k – 300k euros) which is not

typically considered interesting as well as profitable for venture capitalists, due to both its

relatively low cash flows’ generation potential and the relatively high costs for due

diligence, contracting and monitoring given the relevant adverse selection and moral hazard

issues affecting small-scale young businesses (Jeng and Wells, 2000; Carpenter and

Peterson, 2002; Mason, 2009). Second, alongside the capital injection, business angels

1 Also called “informal investors”, in order to differentiate them from venture capitalists and other financial

intermediaries investing on a professional and institutionalized basis capital raised from third parties (Wetzel,

1986; Freear et al., 1993; Landstrom, 1993; Mason, 1996; Van Osnabrugge, 2000)

4

provide non-monetary resources deemed highly valuable for entrepreneurs like industrial

knowledge, management experience, advice and mentoring, standing and personal

relationship networks (Harrison and Mason, 1992; Landstrom, 1993, Politis, 2008 ).

Correspondently, among the major expected outcomes from angels’ investments, a part

from monetary returns there are non-monetary benefits as well, such as such as playing an

entrepreneurial role, working with and mentoring highly-talented and creative people,

discovering new technologies, being invited to pitch presentations, interacting with other

angels and actors in the financial community, and the status from being perceived as a

sophisticated investor (Capizzi, 2015).

Over time, a growing number of angel investors started organizing themselves into groups

(also referred to as syndicates or networks or clubs, depending on their level of internal

structure), usually on a territorial or industrial basis, sharing presentation pitches from

potential entrepreneurs, due diligence over the potential investment opportunities,

transaction costs and investment deals to be implemented by group members (Mason,

2006; Sohl, 2007; Paul and Whittam, 2010; Gregson et al., 2013). These associations, called

Business Angels Networks (BANs), have grown to regional, national (for instance, ACA in

the US, BBAA in the UK, IBAN in Italy) and even continental proportions (among them,

EBAN and BAE in Europe) increasing also their internal structure and coordination

among the members as well as the quality and variety of the service provided (deal flow,

education and coaching, legal and advisory services). Thanks to BANs and angel groups

the informal venture capital market is nowadays much more visible and, hence, easier to

get access to by both demand and supply side (Mason, Botelho and Harrison, 2013).

However, there is still little evidence on the impact of BAN on the investment process of

business angels, most of all based on anecdotal evidence or case studies (May, 2002; Payne

et al., 2002; Mason, 2006; Sohl, 2012; Ibrahim, 2008; Brush et al., 2012; Kerr et a l , 2014 ;

Collewaert and Manigart, 2015; Croce et al. 2016). As such, the aim of this paper is to shed

some light over the major determinants of business angels’ investment decisions, focusing

on the differential role of investors’ membership to business angel groups and networks.

By comparing investment practices of business angels associated with a BAN to those of

business angels operating as individual investors, we investigate whether and how being

member of a semi-formal organization matters by either affecting the share of angels’

5

personal wealth invested in a given deal or affecting the amount of the equity stake in the

participated company.

Looking at a unique dataset that collects qualitative and quantitative information on over

800 investments by 625 business angels from 2008 to 2014, our paper for the first time

find evidence of different causal relationships for angels joining BANs when compared to

angels not members of any single group or network of business angels. In particular we

show that being part of an angel group does generate valuable information, monitoring,

networking and risk reduction effects that, ultimately, affect the amount of financial

resources available to invest as well as the equity stake to assume in the investee company.

The remainder of the paper is structured as follows: the second paragraph will derive the

research hypothesis to be tested from the literature dealing with business angels and the

informal venture capital. In the third paragraph will be presented the dataset and specified

the variables used to perform the empirical analysis, whose results are shown and

commented in the fourth paragraph. In the final paragraph will be presented authors’

conclusive remarks and suggestions for future research.

2. Hypotheses development and related literature

In this paper we focus on business angels investment choices trying to isolate the

differential role played by BANs when compared to the investment process of individual

informal investors. By shedding light on the impact of BAN on the asset allocation

decisions of business angels, it is possible to suggest policymakers new and more effective

measures aimed at stimulating entrepreneurship and, therefore, contributing to the

development and growth of economic and social systems (Baldock and Mason, 2015;

Kraemer-Eis et al., 2016).

The starting point of the research program is the identification of the phenomenon under

investigation, the amount of own risk capital invested by business angels. In this study the

authors investigate and measure business angels’ capital allocation decisions according to a

double perspective: the amount of capital invested as a share of a single business angel’s

personal wealth (“WEALTH%”), and the amount of capital invested as a share of the equity

capital of the investee company (“PARTICIPATION%”). The first measure is a commonly

6

used one in the literature dealing with both venture capitalists (Jeng and Wells, 2000;

Lerner, 1998) and informal investors (Harrison and Mason, 2002; Maula et al., 2005;

Wiltbank and Boecker, 2007), allowing to relate the investment decision making process to

the involvement of the business angels in the participated companies, to the value added

provided and to the nature of the negotiating process developed with the entrepreneurs. 2

The second measure, on the other hand, is useful in order to identify the perceived risk

drivers and their impact on the asset allocation decisions of informal investors.

We model the expected effects on these two measures as follows.

The role of business angel networks

One major evolutionary trend observed in the informal venture capital market over the last

two decades deals with the growing relevance of associations of business angels, either

structured or semi-structured, ranging from loose network of individual investors to formal

angel syndicates (Ibrahim, 2008; Mason, 2009; Paul and Whittam, 2010; Sohl, 2012;

Gregson et al., 2013; Mason, Botelho and Harrison, 2013). Many authors investigated the

black box of such associations, mainly according to a case study research methodology,

studying their internal structures and investment practices (Kerr et al., 2014) as well as the

kind of services and contribution provided to either the investee companies (Collewaert

and Manigart, 2016; Croce et al., 2016) or their members (Shane, 2000; Mason, 2007; Paul

and Whittam, 2010).

For the aim of this study the term Business Angel Network (BAN) is used to indicate an

association of business angels, whose member can join on a solicited or unsolicited basis,

that uses its resources to organize meetings for pitching events, training and mentoring,

lobbying purposes. Entrepreneurs are solicited to join the BAN through websites and

other networking activities taking place inside the community. There is no or limited

organized deal group processing and the association does not make investments or

recommend investments o members, rather each member decides whether or not to invest

on a deal-by-deal basis, typically finding co-investors (within or outside the BAN), sharing

due diligence, negotiations and term sheets.

2 As common in literature dealing with formal and informal venture capital, one major assumption in the

asset allocation process is the impossibility to rely upon the existing asset allocation techniques

developed by theory of finance and presuming to have traded securities as possible asset classes.

7

Consistently with Ibrahim (2008), Brush et al. (2012) and Mason, Botelho and Harrison

(2013), we argue that the membership to a BAN benefits the angel investors mainly

through the information and knowledge sharing effect taking place inside the community.

The possibility for unexperienced angels to get in touch with experienced angels is

particularly important inside the BAN, thus improving new investors human capital and

knowledge of how to implement effective value creating investment decision (Shane,

2000). Also the role of the so-called “gatekeepers”, the individuals who control access to

and manage much of the day-to-day operations of BANs (Paul and Whittam, 2010), is

crucial in favoring the sharing of information among BAN members.

Therefore, the investment made by BAN members, even if not in syndication with other

co-investors, should be a more informed and efficient one, leading to a wider capital

allocation investment choice. In other words, because of the services and contributions

provided by BANs to their members, it is reasonable to assume that BAN members, once

selected an investment opportunity and undertaken the investment decision making

process, will invest more of their personal wealth than non-BAN members. When

considering, on the other side, the participation in the investee company, BAN members

should benefit from the investment opportunities provided by the network inside the angel

communities, leading to invest in more companies, remaining unchanged the share of their

personal wealth allocated to this asset class. This would lead to a decrease in the equity

stake acquired in a given deal.

As such, our first research hypothesis is:

H1: The business angels’ capital allocation investment decision is affected by the membership to a giv en

BAN. The amount of angels’ invested capital is positively affected by BAN membership, whi le

the size of the equity stake is negatively affected by BAN membership.

BAN, groups and risk sharing

Among the many options available to business angels when valuing a given investment

opportunity, there is the possibility to make the deal either as an individual investor – the

“solo angel” – or co-investing with other angel investors, though the latter can be

implemented through different degrees of formal structures ranging from formal angel

syndicates to informal so called “club deals”.

8

From the one hand, each single investor, by co-investing in a given deal, can obtain a lower

equity stake in the target companies, though without giving up to the willingness to gain

active involvement in the participated companies themselves and, therefore, providing

value added contributions. The sum of the single equity positions of all the co-investors in

a given deal increase the possibility to play an active role in the investee companies, which

can also be bigger-sized than those affordable by solo angels (Paul and Whittam, 2010).

From the other hand, consistently with modern portfolio theory (Elton and Gruber, 1995),

the co-investment option is a completely rational diversification strategy aimed at reducing

the risk from a given equity investment opportunity. As a direct implication, business

angels choosing to share the risk of a given deal by co-investing with other ones, also

assuming to keep constant the share of their personal wealth devoted to the asset class of

private equity investments in early stage companies, can leverage on wider and better

diversified investment portfolios (Mason, Botelho and Harrison, 2013) as well as on the

possibility to get access to risk-reducing information (Aernoudt, 2005).

This leads to the following research hypothesis

H2: The business angels’ capital allocation investment decision is negatively affected by the possibility to

co-invest in a given deal.

As above mentioned, another yet agreed upon common features of business angels is their

active role played inside the investee company, in order to support the entrepreneur in the

value creation path through an hands on approach. Politis (2008) identifies four different

kind of value added contributions coming from angel investors, identifying a “sounding

board” role, a “monitoring” role a “resource acquisition” role and a “mentoring” role.

However, a number of surveys disclosed on a yearly basis by research centers (EIF,

OECD, IBAN, EBAN) and country federation of angels associations report the existence

of investors not interested or willing to play such an active role in the investee companies,

rather interested to capital gains and portfolio diversification benefits. For the purpose of

this study we call “passive” investors this latter category of business angels. We expect a

negative relationship between passive investors and the amount invested at least for non-

BAN members. However, for BAN-members the possibility to co-invest could generate

the opposite outcome. This leads to the following research hypothesis:

9

H3: The business angels’ capital allocation investment decision is affected by the expected active/passive

role to be played in a given deal.

The role of monitoring

Finance literature extensively investigated the role of monitoring as a way to reduce

asymmetric information and moral hazard problems stemming from any kind of securities

investment (Jensen and Meckling, 1976; Diamond, 1984; Aghion and Bolton, 1992).

As far as private equity investments are concerned, many authors evidenced how venture

capital organizations monitor investee companies and what are the major contingent

contracts, clauses and mechanisms used to reduce potential conflicts and incentives to

opportunistic behavior by entrepreneurs (Sahlman, 1990; Gompers and Lerner, 2000;

Triantis, 2001; Kaplan and Stromberg, 2003; Chemmanur et al., 2006; Cumming, 2008;

Wong et al., 2009; Cumming and Johan, 2013).

Dealing with business angels, specific contributions showed that they seldom adopt the

typical control and governance provisions of venture capital investors (Van Osnabrugge,

2000; Wiltbank and Boecker, 2007; Goldfarb et al., 2012; Bonini and Capizzi, 2016),

implementing monitoring mechanisms “non aggressive and striking in their informality” (Ibrahim,

2008). The major substitutes for contractual monitoring are represented by the angels’

knowledge of the industry coming from previous investment or managerial experience, the

existing interactions with entrepreneurs and the geographical proximity of the investee

company (Wong et al. 2009).

Consistently with the above mentioned arguments, we think the kind of monitoring taking

place in the informal venture capital market is a “soft” one, not based mainly on

contractual mechanisms, but rather on high involvement in the participated company

through company visits, interactions with entrepreneurs and other control techniques

based upon trust. We argue the higher such a soft monitoring effort, the lower the

investment risk perception by business angels in their investment decision making process.

Given the possibility to investigate the impact of the soft monitoring for both our sub-

samples of business angels – BAN members and non-BAN members – we expect different

causal relationships between monitoring and angel investments. In fact, BAN members

benefit from the screening support provided by BAN to their members as well as from the

10

information and knowledge sharing effect stemming inside the BAN, leaving the need for

higher monitoring effort to the investments perceived as riskier ones. This leads to less

informationally opaque investments when compared to those realized by non-BAN

member business angels, who don’t benefit from the soft information produced inside the

angel community and are supposed to compensate so greater an information asymmetry by

imposing higher level of monitoring. In this case, higher monitoring should not necessarily

be associated to higher investment risk.

We therefore hypothesize that:

H4: The business angels’ capital allocation investment decision is positively affected by the monitoring

effort put by non-BAN members and negatively affected by the monitoring effort put by BAN

members.

Controls

Following the extant literature we will test our hypotheses introducing a set of control

variables that are known to have a causal effect on the investment decision of business

angels. Mason and Harrison, 1996, Van Osnabrugge, 2000 Macht, 2011 explained the role

of experience, while Shane, 2000, Paul et al. 2007 showed the effect of age and education,

as well as the previous background, which could be a managerial one, an entrepreneurial

one or a financial one (Maula et al., 2005; Sudek, 2006; Morrissette, 2007; Collewaert and

Manigart, 2016). As far as age is concerned, consistently with studies dealing with investors’

utility functions in intertemporal portfolio choice model (Samuelson, 1997; Forsfalt, 1999),

it is possible to assume an increase in the risk aversion profile of business angels, leading to

a decrease in the share of their wealth available to riskiest asset classes, like entrepreneuria l

ventures. On the other hand, experience gained through past investments, high degree of

education and the amount of business angels’ personal wealth could act as counteracting

factors on their capital allocation investment decisions.

Additionally, we expect that the equity stake in the target company acquired by a business

angel is negatively affected by the size of the company itself (Mason and Harrison, 2000;

Van Osnabrugge, 2000), as well as by its stage in the life cycle (Wiltbank et al., 2006) and its

location (Sudek, 2006).

11

Finally, consistently with the above mentioned contributions investigating the determinants

of investments taking place both in the formal and informal venture capital markets, we do

consider in our model to be tested some industry (market capitalization, performance and

capital intensity) and macroeconomic (market interest rates) variables expected to play a

statistically significant role in the investment decision making process. In fact, it is

reasonable to assume the amount of capital to invest depends, among the others, also on

industry specific drivers as well as on the expected return from alternative asset class

investments.

Table 1 summarizes our research hypotheses and the predicted signs of both the

explanatory and control variables for each one of the proxies used as a measure of business

angels’ capital allocation process.

INSERT TABLE 1 HERE

3. Sample data and variables

Our data are obtained from sequential surveys administered by IBAN to its associates and

other unaffiliated BAs. Surveys are conducted during the period January – March of each

year beginning in 2008. Each survey was designed in order to collect information on

previous year’s operations. Each survey is completed in a four-step process: at the

beginning of January IBAN forwards the survey’s website link to its associates and other

known BAs.3 By the first week of March data are collected (step 1). Non-responding BAs

are contacted by email and phone to solicit the survey completion (step 2) while an IBAN

team reviews the data to identify incomplete, wrong or unverifiable answers (step 3) which

is further checked through direct follow-up calls (step 4). This process is a fairly common

survey technique called sequential mixed mode, which is a particular methodology that

employs a different survey mode in a sequential way (e.g. the non-respondents to a mail

survey are subsequently contacted and requested to answer the questionnaire through a

different mode, such as a phone interview). Evidence shows that a mixed mode survey

3 See IBAN website (www.iban.it) for the survey questionnaire.

12

approach significantly improves the response rate (De Leeuw, 2005 and Dillman et al.,

2009).

After the completion of the surveys, IBAN received 625 full responses by investors who

performed at least one investment, for a total 808 investments during the 2008 – 2014 time

period. The overall number of angel unique investors is 490, of which the 35% are BAN

members. When considering the overall number of investments, the 54% of them refers to

BAN members, thus implying the membership to an angel community should presumably

affect angels’ capital allocation choices.

The authors are aware, as a major and common issues affecting literature dealing with

informal venture capital, that despite the high response rate sample representativeness can

still be an issue. Research on BAs shares the key difficulty of finding reliable data given that

investments are not publicly recorded and most investors strive for anonymity , while

others don’t even know to be business angels (Mason and Harrison, 2008; Capizzi , 2015).

Therefore, it is important to emphasize that the data used for the analysis do not represent

the full extent of the business angel activity taking place in the investigated arena, rather

produce a reliable estimate of the visible market of the informal venture capital in Ita ly, as

confirmed by the legitimation of IBAN itself, which is the only recognized and authorized

source of information by research centers both at the national level (Bank of Italy, Bocconi

University, Politecnico Milan) and at the international level (EIB, BAE, OECD).

In the following Table 2 we present the temporal and industry distribution of the final

sample data.

INSERT TABLE 2 HERE

The target ventures belong to various industries, among which the most represented ones

are: Manufacturing of Food and Beverage products (20%), Biotech (17.1%) and Cleantech

(13.1%). The geographical distribution of ventures is quite concentrated, since

approximately the 88% of them is located in Italy.

Table 3 reports summary statistics on the participation to groups and networks and the

conditional distribution of the 2 dependent variables built in order to measure the

investigated phenomenon which is the business angels’ capital allocation decision:

13

PARTICIPATION%, which is computed as the amount invested in a venture as a share of its

net-asset-value, and WEALTH%, which is the share of a BA’s financial wealth invested in a

venture.

INSERT TABLE 3 HERE

The business angels joining an angel community (herein after “BAN members”) are almost

54% of the sample, giving the possibility to empirically investigate the role of the business

angel networks in the investment decision making process.

The descriptive statistics related to the dependent variables show that the relative incidence

of BAs’ investments widely varies in the sample, both in terms of participation in the

venture and in terms of personal wealth of the BAs. Nevertheless, the majority of the

investments are relative small. In fact, the participation in a venture is lower than 10% in

half of the cases and lower than 20% in three quarters of them and the share of wealth

invested in a venture is lower than 15% in the 50% of the investments and lower than 17%

in three quarters of them.

Being member of an angel community significantly indeed affects the amount of wealth

invested in a venture. This evidence emerges by performing a two-group-mean comparison

test on the variables PARTICIPATION% and WEALTH%, between those BAs which are

members of a Business Angel Network (BAN) and those which are not members of any

BAN. As presented in Table 1, the test shows that being a member of an angel community

positively affect the share of wealth invested in the investee company, while it does not

influence the size of the participation in the investee company.

The present empirical analysis aims at investigating the existence of a causal relationship

between the two above mentioned dependent variables and some explanatory variables

related to the research hypotheses developed in the previous section. We also introduce in

our model a series of independent variables related to the business angels’ features, the

investee firms and to industry and macroeconomic factors. Table 4 provides a description

of the proxies built for each variable as well as summary statistics. It is worth noting that,

in order to manage the volatility observed for the variables Net_Asset_Value and Co-

14

Investors, largely depending on the nature of the sample, we winsorized them at the 95 th

percentile of their respective distributions.

INSERT TABLE 4 HERE

Our first research hypothesis (H1) deals with the role of angel communities, who propose

investment opportunities and also support their members in different stages of the decision

making process, arguably making their members feel more confident in deciding where and

how much to invest. Therefore, we rely on a dummy variable (BAN_MEMBERSHIP).

discriminating BAN members from non-BAN members, expected to be positively related

to the share of personal wealth invested in a given deal by each single angel.

As for the hypothesis dealing with co-investing (H2), the survey directly provides the

required metric. The data presented in Table 4 show that co-investments are very frequent.

More specifically, the 70% of the investments are characterized by the presence of at least

two co-investors. Since co-investing favors risk-sharing, the expected relationship between

both the dependent variables and the explanatory variable CO-INVESTORS is negative.

To test our third research hypothesis (H3), we use a dummy variable (PASSIVE INVESTOR)

directly related to the answer given by the surveyed angels to a specific question: the va lue

1 signs the investment decision was exclusively driven by capital gain motivation. This

variable should be negatively related to both the dependent variables.

The survey also offers interesting evidences regarding the role played by BAs in the

monitoring of the participated firms, allowing us to test the last research hypothesis (H4).

We built the ordinal variable MONITORING, which graduates the frequency of the visits that

a BA made in a participated venture, from 1 to 5, where 1 means a very limited

involvement (no or very few visits) and 5 means a very high involvement (a constant

presence in the firm). Though the survey collects this information ex-post, asking the

effective involvement in the participated firms by BAs, we believe that they a lready know

the future degree of involvement in a venture also at the time when the investment

decision is taken. Moreover, it is likely that it influences the choice concerning the amount

to invest. In particular, a higher degree of monitoring is expected to decrease the

investment risk perceived by a BA. Therefore, we assume the variable MONITORING as a

15

proxy of the agreed degree of monitoring at the time when the investment decision was

taken. The expected sign for the variable MONITORING is positive, when related to both the

dependent variables.

As far as the angel specific control variables are concerned, for AGE, EDUCATION and

BACKGROUND (which could be a managerial one or an entrepreneurial one) we again rely

on specific dummies making reference to the answer given by the surveyed angels in the

questionnaire. Furthermore, we measure the EXPERIENCE considering the number of

investments made in the past, consistently, consistently with Hsu (2004) and Capizzi

(2011). We believe that more experienced BAs find easier to identify worthy investment

opportunities. Moreover, a greater experience could also increase self-confidence. For both

of the reasons, more experienced BAs are likely to invest greater amounts than less

experienced BAs. Thus, we expect a positive relationship between the dependent variables

PARTICIPATION% and WEALTH% and the variable EXPERIENCE.

Moving to the firm specific control variables, as far as the size of the investee venture is

concerned, we expect the share of participation to decrease as the size of a firm increases.

In contrast, since it is likely that larger-sized firms are perceived as less risky investments, it

is possible that the relationship between the dependent variable WEALTH% and the

NET_ASSET_VALUE variable is positive. As one can see from Table 3, the net-asset-value

of the ventures varies from 1.000 euro to more than 160 millions of euro. However, 75%

of the ventures show a net-asset-value lower than 1,400,000 euro. It is possible to

argument the such a high variability depends on the great heterogeneity of ventures

belonging to the sample in terms of industry and stage of the life cycle.

Approximately the 36% of the investments mapped in dataset are addressed to seed

projects, while in the other cases the target firms are start-up, early growth enterprises,

turn-around and buy-out projects. Since, investing in a seed enterprise is likely to be riskier

than investing in a well-established entrepreneurial project, the expected relationship

between the dummy SEED and the dependent variables PARTICIPATION% and WEALTH% is

negative.

Dealing with the geographical location of the investee companies, foreign ventures

represent only the 12% of the projects financed. We presume, consistently with previous

mentioned literature, that the hogher the geographical, cultural and juridical distance

16

between the residence of the BA and the location of the venture, the lower the amount

invested in a single venture. Thus, the expected sign for the dummy FOREIGN is negative.

Looking at the financial wealth of Bas, we observe a wide range of values (from 250,000 to

7,500,000 euro), however, for the three quarters of them it is lower than 1,450,000 euro.

The wealthier BAs could both choose to invest higher amounts in single ventures, in order

to detain higher participations, and to select a higher number of investment opportunities ,

in order to achieve a better portfolio diversification. Thus, the sign of the relationship

between the dependent variables PARTICIPATION% and WEALTH% and the WEALTH

variable is not plain.

Finally, as proxies of the control variables, consistently with extant literature, we selected

the average 5-years Eurirs (5Y_EURIRS), the industry market capitalization as measured by

the MSCI (MSCI_YEAR_INV), the industry price-to-book value ratio (INDUSTRYPBV) and

the industry net capital assets-to-sales ratio (CAPITAL INTENSITY).

4. Methodology and Results

Determinants of business angels’ capital allocation choices

The first analysis investigates the determinants of the share of personal wealth invested in a

venture by a BA. To this end, we run a OLS regression between the dependent variable

WEALTH% and a set of explanatory variables related to the venture, the investor and the

investment decision. We also add to some model specification time and industry fixed-

effects. We manage heteroskedasticity by computing the natural logarithm of the

dependent variable and of the explanatory variables NET_ASSET_VALUE, WEALTH and

EXPERIENCE and by using White heteroskedasticity-consistent standard errors and

covariances.

Equation (1) comprises a fully balanced model with time fixed-effect.

WEALTH% = f (BAN_MEMBERSHIP, CO-INVESTORS, NET_ASSET_VALUE, SEED,

MSCI_YEAR_INV, EURIRS5Y, YEARt;) (1)

Equation (2) adds to the previous model fixed industry-effects.

WEALTH% = f (BAN_MEMBERSHIP, CO-INVESTORS, NET_ASSET_VALUE, SEED,

MSCI_YEAR_INV, EURIRS5Y, YEARt; INDUSTRYi) (2)

17

Equation (3) comprises all the explanatory variables described in Table 2 and controls:

WEALTH% = f (BAN_MEMBERSHIP, CO-INVESTORS, PASSIVE_INVESTOR, SOFT-MONITORING,

AGE, EDUCATION, WEALTH, ENTREPRENEUR, MANAGER, EXPERIENCE, NET_ASSET_VALUE,

SEED, FOREIGN, MSCI_YEAR_INV, EURIRS5Y, YEARt; INDUSTRYi) (3)

Finally, since the two-group-mean comparison test on the dependent variable WEALTH%,

presented in Table 3, shows that being member of an angel community affect the share of

wealth invested in a venture, we also separately run equation (3) for the sub-samples of

BAN members and non-BAN members.

Table 5 presents the results of the analysis. The model is significant in all the specifications

and shows a R-squared of 25% for the base model in column (3) and above the 30% for

the sub-samples of BAN members and non-BAN members, in column (4-5).

INSERT TABLE 5 HERE

First, it emerges that being member of an angel community increases the share on wealth

invested by approximately 14%, consistently with our H1 research hypothesis.

Second, as far as H2 is concerned, other conditions being equal, one-unit increase in the

number of co-investors reduces by 2% the amount of money invested in a venture.

However, by comparing BAN members with non-BAN members, we observe some

interesting differences highlighting the differential role played by co-investing in their

investment decisions, as it comes out that the decision about the amount to invest is

affected by the presence of co-investors only for the sub-sample of the BAN members,

implying that there could be a positive effect played by trust inside a given angel

community, whereas non BAN member could be prevented from co-investing maybe

because fearing potential issues of free riding and opportunistic behavior. Thus, H2 is

confirmed only for angel member of an angel community.

Third, as expected, for business angels acting as passive investors we find a negative

relationship with the capital invested, confirming H3 hypothesis; such a relationship,

however, is statistically significant only for non-BAN member angels. Maybe, in the case of

BAN members, the possibility to benefit from co-investing, other angels’ experience,

18

mentoring and information provided by the BAN gatekeepers give incentives which

ultimately positively affect asset allocation investment decisions of those angels driven

mainly by capital gain motivations.

Fourth, the SOFT MONITORING variable shows a positive significant sign for the group of

BAs not affiliated to an angel community and a negative sign for the BAN members ,

though not significant. This evidence is consistent with our H4 hypothesis, and seems a

further proof of the quality of the contribution in terms of deal flow and screening

provided by the BA networks to their members. In fact, it is likely that BAN members

impose higher level of monitoring only to ventures that are more opaque. If this is true, the

negative sign is related to the perceived investment risk (which require more monitoring).

In contrast, since non-BAN members do not benefit from the soft-information given by

angel communities, they probably compensate this greater information asymmetry by

imposing more extensively high level of monitoring. In this case, higher monitoring is not

necessarily associated to higher risk.

In order to be sure that the above results do not depend on differences related to personal

characteristics between the two groups of investors, we also run, as a robustness check, a

logistic equation with the dummy BAN_MEMBERSHIP as dependent variable and a set of

personal characteristics as independent variables. As displayed in Table 6, only the fact of

being a manager shows a significant relationship with the affiliation to an angel community.

BAN_MEMBERSHIP = f (AGE, EDUCATION, GENDER, WEALTH, EXPERIENCE,

ENTREPRENEUR, MANAGER) (4)

BAN_MEMBERSHIP = f (AGE, EDUCATION, GENDER, WEALTH, EXPERIENCE,

ENTREPRENEUR, MANAGER, CENTER OF ITALY, SOUTH OF ITALY (5)

INSERT TABLE 6 HERE

Furthermore, results show that BAs not member of an angel community, and thus not

benefitting from the prescreening analyses that BANs make for their members as well as

from the knowledge sharing effect taking place inside the BAN themselves, consider a

wider set of information. In fact, the variables related to the net-asset value of the firm and

the level of the medium-term interest rates are significant for this group of investors, while

19

they are not for the other sub-sample. In particular, the amount invested by non-BAN

members decrease as the net-asset-value of the venture increases. This evidence, however,

does not subtend that they invest in ventures characterized by higher dimensions. In fact,

the regression (equations 6 and 7) presented in Table 7, that we run as robustness check,

shows that the firms proposed by BANs have, on average, a larger size that those identified

by single investors.

PROPOSAL = f (NET_ASSET_VALUE, SEED, INDUSTRY PBV, NET CAPEX/SALES,

MSCI_YEAR_INV) (6)

PROPOSAL = f (NET_ASSET_VALUE, SEED, FOREIGN, INDUSTRY PBV, NET

CAPEX/SALES, MSCI_YEAR_INV) (7)

INSERT TABLE 7 HERE

Finally, dealing with the control variables, it emerges that the amount of wealth invested in

a venture depends on the personal characteristics of BAs, while it is not influenced by the

firms’ characteristics. In particular, the dependent variable increases as the experience of

the BA grows and diminishes as her/his age and her/his financial wealth decrease. In

addition, managers and entrepreneurs invest 10% more than other categories of workers.

Once again, we observe different investment behaviors between BAN members and non-

BAN members as far as the education of the investor is considered. Even without

developing a specific hypothesis about that, it is reasonable to deduct that the information

and knowledge sharing effect taking place inside a community allow to compensate for the

limited education of a given angel investors, who, otherwise, would be prevented from

investing more capital if asked to act as a solo angel.

Determinants of business angels’ investments in the investee firms

The second part of the empirical analysis explores the factors affecting the amount

invested in a venture by a BAs. For this purpose, we estimate the relationship between the

dependent variable PARTICIPATION% and the same set of explanatory variables previously

used, by running a new standard OLS regression.

20

Similarly to the approach used for the dependent variable WEALTH%, we manage

heteroskedasticity by computing the natural logarithm of the dependent variable and of the

explanatory variables PARTICIPATION%, NET_ASSET_VALUE, WEALTH and EXPERIENCE.

We manage heteroskedasticity by considering the natural logarithm of the variables

PARTICIPATION%, NET_ASSET_VALUE, WEALTH and EXPERIENCE, and we also estimate

White heteroskedasticity-consistent standard errors and covariances.

Equation (8) represents a fully balanced model with fixed time-effects.

PARTICIPATION% = f (NET_ASSET_VALUE, SEED, BAN_MEMBERSHIP, CO-INVESTORS,

MSCI_YEAR_INV, INDUSTRY PBV, NET CAPEX/SALES, TIMEt) (8)

Equation (9) comprises the whole set of explanatory variables described in Table 4.

PARTICIPATION% = f (NET_ASSET_VALUE, SEED, FOREIGN, BAN_MEMBERSHIP, AGE,

EDUCATION, WEALTH, ENTREPRENEUR, MANAGER, EXPERIENCE, PASSIVE_INVESTOR, CO-

INVESTORS, MONITORING, MSCI_YEAR_INV, INDUSTRY PBV, NET CAPEX/SALES, TIMEt) (9)

Equation (10) adds to the previous model Industry fixed-effects

PARTICIPATION% = f (NET_ASSET_VALUE, SEED, FOREIGN, BAN_MEMBERSHIP, AGE,

EDUCATION, WEALTH, ENTREPRENEUR, MANAGER, EXPERIENCE, PASSIVE_INVESTOR, CO-

INVESTORS, MONITORING, MSCI_YEAR_INV, INDUSTRY PBV, NET CAPEX/SALES, TIMEt,

INDUSTRYt) (10)

Table 8 presents the results of the model. The analysis shows a high explanatory power: the

adjusted R2 is greater than 50% in the model specification (2) and (3) and the majority of

the independent variables are significant.

INSERT TABLE 8 HERE

In general, the outcomes suggest that BAs are very well conscious about the equity risks of

their investments and, because of that, they rationally manage their risk exposures, a lso by

taking benefit of the risk-reduction effect coming from the membership to angel

community, the co-investments and the monitoring effort.

21

More in detail, BAN_MEMBERSHIP (H1) does negatively affects business angels’ investment

decisions, because, as we argue, a given community provide the possibility to better

diversify the risk by providing more investment opportunities, each one after a deal flow

and screening process more efficient when compared to that undertaken by solo investors.

Furthermore, we find positive evidence for H2 hypothesis, in that as the number of co-

investors increases the individual participation shows a 7% decrease: co-investing appears

an effective way for pursuing risk-minimizing investment decisions, also benefitting from

portfolio diversification upsides.

On the other side, when the main motivation is just and explicitly the search for capital

gain, (when the dummy PASSIVE_INVESTOR is equal to 1) the dependent variable shows a

18% reduction, consistently with our H3 hypothesis.

Dealing with H4 hypothesis, data show that the share of participation in a given investee

company increases by more than 20% as the degree of soft monitoring increases, once

again confirming the relevance of monitoring mechanisms, even if non-contractual based,

as usually agreed upon between entrepreneurs and business angels (Ibrahim, 2008).

As far as the angel specific control variables, the model results display a progressive

reduction of the amount invested in a venture as the age of the investor increases. It also

emerges that, less educated BAs show a greater risk exposure. The degree of experience in

BAs’ investments positively affects the amount invested. The same evidence emerges if the

BA is a manager or an entrepreneur. In contrast, no significant evidences emerge for the

level of financial wealth.

Dealing with the firm specific control variables, the analysis shows that the share of

participation in a venture decreases as the net-asset value of the firm increases. Moreover,

the participation diminishes by more than 30% if the target company is located abroad.

Finally, regarding the market conditions and the sector specific indexes, the outcomes

highlight that BAs invest growing amounts in presence of positive market trends and

prefer sectors characterized by a lower price-to-book value and a higher capital intensity.

5. Conclusive remarks and suggestions for future research

22

In this paper we presented evidence of the existence of a number of factors producing a

significant impact on the business angels’ investment. More in detail, by making reference

to an unique dataset which provides data on the Italian informal venture capital market

over a seven-year time period, we contributed to extant finance literature by exploring for

the first time the differential investment behavior of business angels when they join semi-

formal organizations such as Angel Groups or Business Angel Networks, compared to the

investment decisions of business angels investing alone outside a given angel community.

The main results of the empirical analysis allow us to understand the relevant role played

by angel communities who indeed affect the business angels’ investment decisions. This

means that angel investors do recognize and appreciate the contribution provided by BAN

in terms of deal flows, screening, mentoring, information and knowledge sharing.

Furthermore, only for BAN members the possibility to co-invest appear to be a factor

significantly affecting their investment decisions, giving them the possibility, on the one

hand, to benefit from risk-reduction effects and, on the other hand, to still continue playing

an active role in the investee company. This means that there is a role for angel

communities in spreading information, reducing incentives to opportunistic behaviors and

enhancing trust among investors.

For non-BAN members, instead, one major significant driver of the investment decision is

the capital gain motivation or, better, the unwillingness to play an active role in the investee

company (passive investor behaviour), which is negatively related to the amount of capita l

invested, even due to the lower probability to effective give rise to club deals with other

solo investors.

Finally, the soft monitoring – not based on corporate governance and contractual

provisions typically adopted by venture capitalists – plays a different role in the business

angels’ capital allocation decisions depending on the membership to an angel community.

Non-BAN members must rely on monitoring in order to reduce information asymmetries

and incentives to opportunistic behaviours of entrepreneurs: the higher the possibility to

monitor the investment with non-contractual mechanisms, the higher the amount of

capital they are available to invest. BAN members, on the other side, can compensate

monitoring with other benefits provided by the angel community aimed at managing and

minimizing the investment risk perception, like the screening support provided by BAN to

23

their members, the knowledge sharing effect stemming inside the BAN, the probability to

gain more active involvement in the participated companies thanks also to co-investments.

Angel communities, thus, can decrease the need for individual monitoring effort, increasing

at the same time the members’ confidence in the investments made.

Our findings have relevant implications for academicians, policy makers and entrepreneurs

seeking finance, given the insight on business angels’ investment decisions. Stimulating and

strengthening the networks could be a major goal in the future for the growth of

entrepreneurship and the recovery of employment level of a given economic and social

system.

However, in order to both continue shedding light over the differential investment

behaviors of angels joining a given business community when compared to solo angels,

and further investigate the beneficial effects coming from the BANs, the academic

community needs to get a deeper access to the black box of the angel communities

themselves, sharing at a national and international level common and homogeneous data

about their operations and decision making processes.

24

References

Aernoudt, R. (2005), Business Angels: The Smartest Money for Starters? Plea for a Renewed Policy Focus on Business Angels. International Journal of Business, Vol. 19, No. 5.

Aghion, P., and Bolton, P. (1992), An Incomplete Contracts Approach to Financial Contracting. Review of Economic Studies, Vol. 59, No. 3, 473-494.

Baldock, R., and Mason, C. (2015), Establishing a new UK finance escalator for innovative SMEs: The roles of the enterprise capital funds and angel co-investment fund. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 17, No. 1-2, 59-86.

Bammens, Y., and Collewaert, V. (2013), Trust between entrepreneurs and angel investors. Exploring positive and negative implications for venture performance assessments. Journal of Management, Vol. 50, No. 6, 1070-1096.

Bonini, S., and Capizzi, V. (2016), The effects of private equity investors on the governance of companies. In Gabrielsson J. (ed.), Handbook of Research on Entrepreneurship and Corporate Governance, Edward Elgar Publishing, Camberley, UK, 2016 (forthcoming).

Brush, C. G., Edelman, L. F., & Manolova, T. S. (2012). Ready for funding? entrepreneurial ventures and the pursuit of angel financing. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 14, No. 2-3, 111-129.

Capizzi, V., (2015), The returns of business angel investments and their major determinants. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 17, No. 4, 271-298.

Carpenter R.E., and Petersen, B.C. (2002), Capital Market Imperfection, High-Tech Investment and New Equity Financing. The Economic Journal, Vol. 112, No. 477, F54-F72.

Collewaert, V., and Manigart, S. (2016), Valuation of Angel-Backed Companies: The Role of Investor Human Capital. Journal of Small Business Management, Vol. 54, No. 1, 356-372.

Coveney, P., and Moore, K. (1998), Business Angels: Securing Start-Up Finance. Chichester, UK: Wiley. Croce, A., Tenca, F. and Ughetto, E. (2016), How business angel groups work: rejection criteria in

investment evaluation. International Small Business Journal, forthcoming. Cumming, D., and Johan, S. (2013), Venture Capital and Private Equity Contracting: An International

Perspective, 2nd Edition. Amsterdam, The Netherlands: Elsevier Science Academic Press.

De Leeuw, E.D. 2005. To mix or not to mix data collection modes in surveys. Journal of Official

Statistics, Vol. 21, 233–255.

Delmar, F., and Shane, S. (2006), Does experience matter? The effect of founding team experience

on the survival and sales of newly founded ventures. Strategic Organization, Vol. 4, No. 3, 215-

247.

Diamond, D.W. (1984), Financial Intermediation and Delegated Monitoring. The Review of Economic

Studies, Vol. 51, No. 3, 393-414.

Dillman, D. A., Smyth, J. D., Christian, L. M., & Dillman, D. A. (2009). Internet, mail, and mixed-mode

surveys: The tailored design method. Hoboken, N.J: Wiley & Sons.

Elitzur, R., and Gavious, A. (2003), Contracting, signaling, and moral hazard: A model of

entrepreneurs, ‘angels,’ and venture capitalists. Journal of Business Venturing, Vol. 18, No. 6, 709-

725.

Elton, E.J., and Gruber, M.J. (2005), Modern Portfolio Theory and Investment Analysis. Fifth Edition.

Hoboken, N.J: Wiley & Sons.

EBAN (2015), Activity Report 2015. Available at: www.eban.org.

EVCA (2014), EVCA Report 2014, Available at: www.evca.eu.

Erikson, T., and Sørheim, R. (2005), ‘Technology angels’ and other informal investors. Technovation,

Vol. 25, No. 5, 489-496.

Feeney, L., Haines, G. H., and Riding, A. L. (1999), Private investors’ investment criteria: Insights

from qualitative data. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 1,

No. 2, 121-145.

25

Festel, G. W., and De Cleyn, S. H. (2013), Founding angels as an emerging subtype of the angel

investment model in high-tech businesses. Venture Capital: An International Journal of

Entrepreneurial Finance, Vol. 15, No. 3, 261-282

Forsfält, T. (1999), The Effects of Risk Aversion and Age on Investments in New Firms. Research

Papers in Economics, Stockholm University, Department of Economics No 1999:18.

Freear, J., Sohl, J. E., and Wetzel, W. E. (1992), The investment attitudes, behavior and

characteristics of high net worth individuals. In Churchill, N. C, Birley, S., Bygrave, W. D.,

Muzyka, D. E., Wahlbin, C., and Wetzel, W. E., Jr. (eds.), Frontiers of Entrepreneurship Research

1992. Babson Park, MA: Babson College, 374-387.

Freear, J., Sohl, J. E., and Wetzel, W. (2002), Angles on angels: Financing technology-based

ventures – A historical perspective. Venture Capital: An International Journal of Entrepreneurial

Finance, Vol. 4, No. 4, 275–287.

Gaston, R. J. (1989), The scale of informal capital markets. Small Business Economics, Vol. 1, No. 3, 223-230.

Goldfarb, B., Hoberg, G., Kisch, D., and Triantis, A. (2012), Does Angel Participation Matter? An

Analysis of Early Venture Financing. Robert H. Smith School of Business Research Paper Series,

No. 06-072, May 2012.

Gompers, P., and Lerner, J. (2004), The Venture Capital Cycle. Cambridge, MA: The MIT Press.

Gregson, G., Mann, S., & Harrison, R. (2013). Business angel syndication and the evolution of risk

capital in a small market economy: Evidence from Scotland. Managerial and Decision Economics,

Vol. 34, No. 2, 95-107.

Harrison, R. T., and Mason, C. M. (1992), International perspectives on the supply of informal

venture capital. Journal of Business Venturing, Vol. 7, No. 6, 459-475

Harrison, R. T., and Mason, C. M. (1994), The informal venture capital market in the UK. In Hughes, A. and Storey, D. J. (eds.), Finance and the Small Firm. London, UK: Routledge, 64-111.

Harrison, R. T., and Mason, C. M. (eds.) (1996a), Informal Venture Capital: Evaluating the Impact of Business Introduction Services. Hertfordshire, UK: Prentice Hall.

Harrison, R. T. and Mason, C. M. (1996b), Why ‘business angels’ say no: A case studies of opportunities rejected by an informal investor syndicate. International Small Business Journal, Vol. 14, No. 2, 35-51.

Harrison, R. T. and Mason, C. M. (l996c), Informal venture capital: A study of the investment process, the post-investment experience and investment performance. Entrepreneurship Regional Development, Vol. 8, No. 2, 105-126.

Harrison, R. T. and Mason, C. M. (2002), Barriers to investment in the informal venture capital

sector. Entrepreneurship Regional Development, Vol. 4, No. 3, 271-287.

Harrison, R. T. and Mason, C. M. (2008), Venture Capital: An International Journal of Entrepreneurial

Finance, Vol. 10, No. 4, 309-330.

Harrison, R. T., Mason, C. M., and Robson, P. J. A. (2010), Determinants of long-distance investing

by business angels in the UK. Entrepreneurship and Regional Development, Vol. 22, No. 2, 113-137.

Hsu, D. K., Haynie, J. M, Simmons, S. A., and McKelvie, A, (2014), What matters, matters

differently: A conjoint analysis of the decision policies of angel and venture capital investors.

Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 16, No. 1, 1-25.

IBAN (2008), Annual Survey 2007. Available at: www.iban.it. IBAN (2009), Annual Survey 2008. Available at: www.iban.it. IBAN (2010), Annual Survey 2009. Available at: www.iban.it. IBAN (2011), Annual Survey 2010. Available at: www.iban.it. IBAN (2012), Annual Survey 2011. Available at: www.iban.it. IBAN (2013), Annual Survey 2012. Available at: www.iban.it. IBAN (2014), Annual Survey 2013. Available at: www.iban.it.

26

IBAN (2015), Annual Survey 2014. Available at: www.iban.it. Ibrahim, D. M. (2008), The (not so) puzzling behavior of angel investors. Vanderbilt Law Review,

Vol. 61, No. 5, 1405-1452.

Jeng, L. A., and Wells, P. C. (2000), The determinants of venture capital funding: evidence across countries. Journal of Corporate Finance, Vol. 6, No. 3, 241-289.

Jensen, M. C. and Meckling, W.H. (1976), Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, No. 3, 305-360.

Johnson, W. C., and Sohl, J. (2012), Angels and venture capitalists in the initial public offering

market. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 14, No. 1, 27-42.

Kaplan, S. N., Sensoy, B. A., and Strömberg, P. (2009), Should investors bet on the jockey or the horse? Evidence from the evolution of firms from early business plans to public companies. Journal of Finance, Vol. 64, No. 1, 75-115.

Kerr, William R., Josh Lerner, and Antoinette Schoar (2014), The Consequences of Entrepreneurial

Finance: A Regression Discontinuity Analysis. Review of Financial Studies, Vol. 27, 20-55.

Kraemer-Eis, H., Signore, S., and Prencipe, D. (2016), The European venture capital landscape: an

EIF prespective. EIF Research & Market Analysis, working paper, No. 2016/34.

Kraemer-Eis, H., Lang, F., Torfs, W., and Gvetadze, S. (2015), European Small Business Finance

Outlook. EIF Research & Market Analysis, working paper, No. 2015/32.

Landström, H. (1993), Informal risk capital in Sweden and some international comparisons. Journal of Business Venturing, Vol. 8, No. 6, 525-540.

Landström, H. (1995), A pilot study on the investment decision-making behavior of informal

investors in Sweden. Journal of Small Business Management, Vol. 33, No. 3, 67-76.

Lerner, J. (1998), “Angel” financing and public policy: An overview. Journal of Banking & Finance,

Vol. 22, Nos. 6-8, 773-783.

Macht, S. (2011a), The role of investee company managers in business angels’ involvement: Empirical insights from dyadic data. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 13, No. 3, 267-293.

Macht, S. A. (2011b), Inexpert business angels: How even investors with ‘nothing to add’can add

value. Strategic Change, Vol. 20, Nos.7-8, 269-278.

Macht, S. A., and Robinson, J. (2009), Do business angels benefit their investee companies?

International Journal of Entrepreneurial Behaviour & Research, Vol. 15, No. 2, 187-208.

Madill, J. J., Haines, G. H., Jr., and Riding, A. L. (2005), The role of angels in technology SMEs: A

link to venture capital. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 7,

No. 2, 107-129.

Manigart, S., and Struyf, C. (1997), Financing high technology startups in Belgium: An explorative

study. Small Business Economics, Vol. 9, No. 2, 125–135.

Manigart, S., De Waele, K., Wright, M., Robbie, K., Desbrières, P., Sapienza, H. J., and Beekman,

A. (2002), Determinants of required return in venture capital investments: A five-country

study. Journal of Business Venturing, Vol. 17, No. 4, 291-312.

Mason, C. M. (2006), Informal sources of venture finance. In Parker, S. C. (ed.), The Life Cycle of

Entrepreneurial Ventures. New York, NY: Springer, 259-299

Mason, C. M. (2008), The Real Venture Capitalists: A Review of Research on Business Angels. Working

Paper WP08-06, Hunter Center for Entrepreneurship, University of Strathclyde, September

2008, 1-55.

Mason, C. M. (2009), Public policy support for the informal venture capital market in Europe: A

critical review. International Small Business Journal, Vol. 27, No. 5, 536-556.

Mason, C. M., Botelho, T., and Harrison, R. T. (2013), The Transformation of the Business Angel

Market: Evidence from Scotland. Available at: http://dx.doi.org/10.2139/ssrn.2306653.

27

Mason, C. M., and Stark, M. (2004), What do investors look for in a business plan? A comparison

of the investment criteria of bankers, venture capitalists and business angels. International Small

Business Journal, Vol. 22, No. 3, 227-248.

Mason, C. M., and Harrison, R. T. (2000), The size of the informal venture capital market in the

United Kingdom. Small Business Economics, Vol. 15, 137-148.

Maula, M., Autio, E., and Arenius, P. (2005), What drives micro-angel investments? Small Business

Economics, Vol. 25, No. 5, 459-475.

May, J., 2002, Structured Angel Groups In The USA: The Dinner Club Experience. Venture Capital:

An International Journal of Entrepreneurial Finance, Vol. 4, No. 3, 337-342.

Mitteness, C. R., Baucus, M. S., and Sudek, R. (2012), Horse vs. jockey? How stage of funding

process and industry experience affect the evaluations of angel investors. Venture Capital: An

International Journal of Entrepreneurial Finance, Vol. 14, No. 4, 241-267.

Morrisette, S. G. (2007), A profile of angel investors. The Journal of Private Equity, Vol. 10, No. 3, 52-

66.

OECD, (2016), Business and Finance Outlook. OECD Publishing. Available at

http://www.oecd.org.

OECD, (2012), Financing High-Growth Firms: The Role of Angel Investors. OECD Publishing.

Available at: http://dx.doi.org/10.1787/9789264118782-en.

Paul, S., Whittam, G., and Johnston, J. B. (2003), The operation of the informal venture capital

market in Scotland. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 5, No.

4, 313-335.

Paul, S., Whittam, G., and Wyper, J. (2007), Towards a model of the business angel investment

process. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 9, No. 2, 107-125

Paul, S., and Whittam, G. (2010), Business angel syndicates: An exploratory study of gatekeepers.

Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 12, No. 3, 241-256.

Payne, W. H. and Mccarty, M. J. (2002), The Anatomy Of An Angel Investing Network: Tech

Coast Angels. Venture Capital: An International Journal of Entrepreneurial Finance, Vol. 4, No. 31-

336.

Politis, D. (2008). Business angels and value added: What do we know and where do we go? Venture

Capital: An International Journal of Entrepreneurial Finance, Vol. 10, No. 2, 127-147.

Riding, A. L., Madill, J. J., and Haines, G. H., Jr. (2007), Investment decision making by business

angels. In Landström, H. (ed.), Handbook of Research on Venture Capital. Cheltenham, UK:

Edward Elgar, 332-346.

Samuelson, P. A, (1997), How Best to Flip-Flop If You Must: Integer Dynamic Stochastic

Programming for Either-Or. Journal of Risk and Uncertainty, Vol. 15, No. 3, 183-90.

Shane, S. (2000), Prior Knowledge and the Discovery of Entrepreneurial Opportunities.

Organization Science, Vol. 11, 448–469.

Shane, S. (2008), The Importance of Angel Investing in Financing the Growth of Entrepreneurial Ventures, SBA

Office of Advocacy working paper, Washington, D.C., No. 331.

Shane, S. A. (2009), Fool’s Gold?: The Truth Behind Angel Investing in America. New York, NY: Oxford

University Press.

Sohl, J. E. (1999), The early-stage equity market in the USA. Venture Capital: An International Journal of Entrepreneurial Finance Vol. 1, No. 2, 101-120.

Sohl, J. E. (2003a), The U.S. angel and venture capital market: Recent trends and developments. Journal of Private Equity, Vol. 6, No., 2, 7-17.

28

Sohl, J. E. (2003b), Angel investing: A market perspective. In May, J. and O’Haloran, E. (eds.), State

of the Art: An Executive Briefing on Cutting-Edge Practices in American Angel Investing. Charlottesville,

VA: Darden Business Publishing, 2-14.

Sohl, J. E. (2006), Angel investing: Changing strategies during volatile times. Journal of Entrepreneurial

Finance and Business Ventures, Vol. 11, No. 2, 27-48.

Sohl, J. E. (2007), The organisation of the informal venture capital market. In Landström, H. (ed.),

Handbook of Research on Venture Capital. Cheltenham, UK: Edward Elgar, 347-368.

Sudek, R., Mitteness, C., and Baucus, M. (2008), Betting on the horse or the jockey: The impact of

expertise on angel investing. Venture Capital: An International Journal of Entrepreneurial Finance,

Vol. 14, No. 4, 241-267.

US Angel Capital Association, (2015), Introducing Angel Investing as an asset class for all investors.

The Go Beyond Investor Report. Available at: http://www.angelcapitalassociation.org

Vanacker, T. Collewaert, V., and Paeleman, I. (2013), The relationship between slack resources and

the performance of entrepreneurial firms: The role of venture capital and angel investors.

Journal of Management Studies, Vol. 50, No. 6, 1070-1096.

Van Osnabrugge, M. (2000), A comparison of business angel and venture capitalist investment

procedures: An agency theory-based analysis. Venture Capital: An International Journal of

Entrepreneurial Finance, Vol. 2, No. 2, 91-109.

Wetzel, W. E. (1981), Informal risk capital in New England. In Vesper, K. H. (ed.), Frontiers of Entrepreneurship Research. Wellesley, MA: Babson College, 217-245.

Wetzel, W. E. (1983), Angels and informal risk capital. Sloan Management Review, Vol, 24, No. 4, 23-34.

Wetzel, W. E. (1986), Entrepreneurs, Angels and Economic Renaissance. Lexington, MA: Lexington Books.

Wetzel, W. E. (1987), The informal venture capital market: Aspects of scale and market efficiency. Journal of Business Venturing, Vol. 2, No. 4, 299-313.

Wetzel, W. E., and Freear, J. (1993), Promoting informal venture capital in the United States:

Reflections on the history of the venture capital network. In Harrison, R. T., and Mason, C.

M. (eds.), Informal Venture Capital: Information, Networks and Public Policy. Hemel Hempstead, UK:

Woodhead-Faulkner.

Wiltbank, R. (2005), Investment practices and outcomes of informal venture investors. Venture

Capital: An International Journal of Entrepreneurial Finance, Vol. 7, No. 4, 343-357.

Wiltbank, R., Dew, N., Read, S., and Sarasvathy, S. D. (2006), What to do next? The case for non-

predictive strategy. Strategic Management Journal, Vol. 27, No. 10, 981-998.

Wiltbank, R. and Boeker, W. (2007), Returns to angel investors in groups. Report, Ewing Marion

Kauffman Foundation and Angel Capital Education Foundation, November.

Wiltbank, R., Read, S., Dew, N., and Sarasvathy, S. D. (2009), Prediction and control under

uncertainty: Outcomes in angel investing. Journal of Business Venturing, Vol. 24, No. 2, 116-133.

Wong, A., Bhatia, M., and Freeman, Z. (2009), Angel finance: The other venture capital. Strategic

Change, Vol. 18, Nos. 7-8, 221-230.

29

Table 1

Causal relationships to be tested: predicted signs

Explanatory Variables

Expected sign

Dependent variable =

PARTICIPATION%

Dependent variable = WEALTH% Whole sample BAN members Non BAN

members BAN_MEMBERSHIP + -

CO-INVESTORS - - - - PASSIVE_INVESTOR - - - - SOFT-MONITORING + + - + Firm Specific Controls

SIZE - + LIFE CYCLE - -

ARENA - -

Angel Specific Controls AGE - - PERSONAL WEALTH +/- +/- EDUCATION +/- +/- ENTREPRENEURIAL

BACKGROUND +/- +/-

MANAGERIAL

BACKGROUND +/- +/-

EXPERIENCE + +

Market and Industry Specific Controls MARKET INDUSTRY

CAPITALIZATION +/- +/-

MARKET INTEREST

RATES +/-

INDUSTRY

PERFORMANCE -

INDUSTRY CAPITAL

INTENSITY +

30

Table 2

Sample distribution

PANEL A – YEAR DISTRIBUTION

Year Investments

2008 95 2009 142 2010 137 2011 159 2012 162 2013 58

2014 57 Total 810

PANEL B – INDUSTRY DISTRIBUTION

Industry

Investments

BIOTECH

137

CLEANTECH

105

COMMERCE AND DISTRIBUTION

81

ELECTRONICS

75

FINANCIAL SERVICES

27

FOOD & BEVERAGE

23

ICT (SW AND HW, APP WEB AND MOBILE)

167

MECHANICAL ENGINEERING

60

MEDIA & ENTERTAINMENT

80

TELECOMMUNICATIONS & SIMILAR SERVICES

23

TEXTILE & APPAREL

25

Total

803

31

Table 3

Dependent variables: descriptive statistics.

Total sample BAN members Non-BAN members

Dependent variable = PARTICIPATION% Mean 14.74 14.87 14.59 Maximum 100 100 100 Minimum 1 1 1 Standard deviation 19.54 18.30 20.93 No. observation 808 436 372

Two-group mean-comparison test It-statI 0.197 Dependent variable = WEALTH% Mean 15.48 17.09 13.67 Maximum 60 60 60 Minimum 5 5 5 Standard deviation 11.80 13.13 9.80

No. observation 669 354 315 Two-group mean-comparison test It-statI 3.480***

Table 4

Summary statistics

Description Obs. Mean Std.Dev

.

Min Max Dummy=1

percentage

BAN_MEMBERSHIP Dummy =1 if the BA is a BAN member 810 54.1

CO-INVESTORS Number of co-investors 809 4.30 4.99 0 15

PASSIVE INVESTOR Dummy =1 if the investment is exclusively driven by capital gain

motivations

668 22.0

SOFT-MONITORING Ordinal variable ranging from 1 to 5, where 1 means monitoring

very low or absent and 5 means monitoring very high, with a

constant presence in the firm

668 2.75 1.25 1 5

Angel specific controls

AGE Age of the BA 668 48.32 9.40 28 71

LOW-EDUCATION Dummy = 1 if the BA holds a high school diploma or a lower

educational qualification

668 6.7

WEALTH (in euro) BAs’ financial wealth in the year of the investment 669 1.480.682 1515.29 250.000 7.500.000

ENTREPRENEUR Dummy =1 in case of prevalent working occupation as

entrepreneur

668 37.7

MANAGER Dummy =1 in case of prevalent working occupation as manager 668 16.8

EXPERIENCE Number of BA’ investments in lifetime 668 6.36 4.01 0 26

Firm specific controls

NET_ASSET_VALUE

(in euro)

Enterprises’ net asset value in the year of the BA’s investment 806 1389.67 2281.66 1,430 8,928,570

SEED Dummy = 1 if the BA has invested in a seed enterprise 810 35.7

FOREIGN Dummy = 1 if the BA has invested in a foreign enterprise 711 12.1

Market and industry controls MSCI_YEAR_INV Industry market capitalization as measured by the MSCI, in the

investment year

810 58.93 10.01 45.78 81.81

5Y_EURIRS Average 5y Eurirs, in the investment year 810 2.251 1.01 0.73 4.31

INDUSTRY PBV Industry price-to-book value, in the investment year 810 3.05 1.36 0.71 8.62

NET CAPEX/SALES Industry net capital assets to sales, in the investment year 810 0.80 3.18 -4.47 22.96

Table 5

Regression Results (dependent variable: WEALTH%)

The regressions are all conducted with the ordinary least square method (OLS), using White heteroskedasticity -consistent

standard errors and covariances. The dependent variable, WEALTH%, is the share of Angels’ Wealth invested in each BA

enterprise. Equation (1) comprises a fully balanced model with time f ixed-ef f ect. Equation (2) adds to the previous model f exed

industry-ef f ects. Equation (3) comprises all the explanatory variables described in Table 2. We also run equation (3) for the two

sub-samples originated by grouping BAs on the basis of the BAN_MEMBERSHIP dummy. The t-stat are reported in brackets

under each coef f icient.

Indipendent Variables

Whole Sample BAN Member

Non-BAN Member

(1) (2) (3) (4) (5)

BAN_MEMBERSHIP 0.109 0.106 0.139

(2.32)** (2.23)** (2.93)*** CO-INVESTORS -0.026 -0.026 -0.016 -0.037 -0.005 (6.38)*** (6.25)*** (2.98)*** (4.83)*** (0.85) PASSIVE INVESTOR -0.042 0.024 -0.176 (0.72) (0.27) (2.29)** SOFT-MONITORING 0.057 -0.050 0.154

(2.01)** (1.52) (4.58)*** AGE -0.015 -0.010 -0.017

(5.58)*** (2.70)*** (3.81)***

LOW-EDUCATION 0.038 0.208 -0.190 (0.49) (1.78)* (1.83)*

WEALTH -0.049 -0.039 -0.094 (1.64) (0.93) (2.73)***

ENTREPRENEUR 0.081 0.026 0.152 (1.56) (0.36) (2.31)** MANAGER 0.064 0.289 -0.103 (1.03) (2.64)*** (1.40) EXPERIENCE 0.198 0.280 0.160 (4.79)*** (4.36)*** (3.32)*** NET_ASSET_VALUE -0.009 -0.003 0.009 -0.011 0.039 (0.47) (0.15) (0.45) (0.39) (1.55) SEED 0.006 -0.007 -0.050 -0.022 -0.143 (0.11) (0.12) (0.91) (0.31) (1.78)* FOREIGN -0.006 -0.007 0.030 (0.09) (0.06) (0.34) MSCI_YEAR_INV -0.013 -0.013 -0.018 -0.029 -0.022 (3.10)*** (3.23)*** (4.33)*** (4.79)*** (3.57)*** 5Y_EURIRS 0.116 0.118 0.096 0.146 0.162 (2.60)*** (2.66)*** (1.68)* (1.87)* (2.34)** _CONS 3.138 3.026 3.815 4.496 4.045 (14.68)*** (14.16)*** (12.15)*** (10.81)*** (7.75)*** YEAR YES YES YES YES YES INDUSTRY NO YES YES YES YES

R2 0.08 0.11 0.25 0.32 0.36 Observation 668 668 569 292 277 * = significant at 10% level; ** = significant at 5% level; ***= significant at 1% level

34

Table 6

BA’s investments: probability of being proposed by a BA network

Logistic regression with the dummy PROPOSAL as dependent variable, which corresponds to 1 if the project f inanced by an Angel

has been proposed by a BA network or an Investors’ club. The table reports results of equation (4). The z-scores are reported in

brackets under each odds-ratio.

Indipendent Variables

Odds-ratio (z-score)

(1) (2) NET_ASSET_VALUE 1.14

(2.40) ** 1.16

(2.43) ** SEED -1.95

(-0.30) 1.07 (0.36)

FOREIGN 0.004 (0.02)

INDUSTRY PBV 1.07 (1.09)

1.01 (0.23)

NET CAPEX/SALES -0.97 (-1.14)

-0.96 (-1.34)

Prob chi2 0.0639 0.7529

Pseudo R2 0.0106 0.0037

Observations 668 569 * = significant at 10% level; ** = significant at 5% level; ***= significant at 1% level

35

Table 7

Probability of being a BAN member

Logistic regression with the dummy BAN_MEMBERSHIP as dependent variable, which assumes the value 1 in case of BAN

membership. The table reports results of equation (5). The z-scores are reported in brackets under each odds-ratio.

Indipendent Variables

Odds-ratio (z-score)

(1) (2) AGE 1.005

(0.60) -0.998 (-0.22)

EDUCATION -0.959 (-0.13)

-0.846 (-0.52)

GENDER -0.867 (-0.45)

-0.929 (-0.23)

WEALTH -0.962 (-0.45)

-0.977 (-0.27)

EXPERIENCE -0.979 (-1.03)

-0.982 (-0.89)

ENTREPRENEUR 1.112 (0.50)

1.202 (0.85)

MANAGER -0.581 (-2.05) **

-0.593 (-1.94) *

CENTER OF ITALY - 1.044 (0.24)

SOUTH OF ITALY - -0.987 (-0.04)

Prob chi2 0.2664 0.3342

Pseudo R2 0.0108 0.0125 Observations 668 655 * = significant at 10% level; ** = significant at 5% level; ***= significant at 1% level

36

Table 8

Regression Results (dependent variable: PARTICIPATION%)

The table reports results of equation 4. The regressions are all conducted with the ordinary least square method (OLS), using

White heteroskedasticity-consistent standard errors and covariances. The dependent variable, PARTICIPATION%, is the amoun t

invested in a venture by an Angel as a share of its net -asset-value. The explanatory variables are described in Table 2. In column

(1) we show a fully balanced model with time f ixed-ef f ect. In column (2) we add the full set of explanatory variables. In column

(3) we add the interaction term between BAN_MEMBERSHIP and SOFT-MONITORING. In column (4) we run equation

(8) with also industry f ixed-ef f ects, which represents the base model The t-stat are reported in brackets under each coef f icient.

Independent Variables (1) (2) (3) (4)

BAN_MEMBERSHIP -0.114 -0.157 0.317 -0.147

(1.83)* (2.20)** (1.74)* (2.05)**

CO-INVESTORS -0.093 -0.069 -0.069 -0.067

(16.47)*** (9.33)*** (9.62)*** (8.96)***

PASSIVE INVESTOR -0.176 -0.173 -0.176

(2.36)** (2.33)** (2.33)**

SOFT-MONITORING 0.213 0.295 0.215

(5.87)*** (6.17)*** (5.67)***

AGE -0.009 -0.009 -0.009

(2.37)** (2.23)** (2.12)**

LOW-EDUCATION 0.310 0.253 0.340

(2.38)** (1.89)* (2.56)**

WEALTH 0.044 0.038 0.046

(1.12) (0.99) (1.20)

ENTREPRENEUR 0.345 0.352 0.351

(4.64)*** (4.76)*** (4.63)***

MANAGER 0.333 0.351 0.332

(3.47)*** (3.60)*** (3.38)***

EXPERIENCE 0.100 0.115 0.092

(1.72)* (2.01)** (1.54)

NET_ASSET_VALUE -0.244 -0.251 -0.252 -0.249

(10.08)*** (10.47)*** (10.72)*** (10.23)***

SEED -0.048 -0.117 -0.124 -0.128

(0.71) (1.49) (1.59) (1.60)

FOREIGN -0.315 -0.302 -0.325

(3.06)*** (2.89)*** (3.06)***

MSCI_YEAR_INV 0.012 0.011 0.010 0.011

(3.96)*** (2.08)** (2.02)** (2.08)**

INDUSTRY PBV -0.040 -0.065 -0.054 -0.049

(1.52) (2.32)** (1.90)* (0.83)

NET CAPEX/SALES 0.029 0.035 0.036 0.024

(2.41)** (2.75)*** (2.87)*** (1.35)

BANMEMB*MONIT -0.169

(2.70)***

_CONS 3.476 2.870 2.651 2.709

(12.29)*** (6.05)*** (5.60)*** (4.95)***

YEAR

INDUSTRY

YES

NO

YES

NO

YES

NO

YES

YES

R2 0.47 0.55 0.56 0.56

N 798 569 569 569