A Transaction Cost Economics and Agency Theory
Perspective on Equity Based Crowdfunding
MScBA Strategy & Innovation: master thesis Jeroen Langejan
Student number: 1272845 August 2013
Supervisors: R.A. Van der Eijk University of Groningen Faculty of Economics and Business Department of Innovation Management & Strategy
Prof. Dr. W.A. Dolfsma University of Groningen Faculty of Economics and Business
CONTENTS
EXECUTIVE SUMMARY ... 4 ABSTRACT ... 8 1. INTRODUCTION ... 9 1.1 Problem definition ... 10 1.1.1 Adverse selection... 10 1.1.2 Moral hazard ... 11 1.1.3 Hold up ... 11 1.2 Research question ... 11 1.3 Conceptual model ... 12 1.4 Research scope ... 12 1.5 Paper outline ... 13 2. LITERATURE REVIEW ... 14 2.1 Crowdfunding ... 14 2.1.1 Crowdfunding process ... 152.1.3 Inefficient capital market ... 16
2.1.4 Policies and regulations ... 17
2.2 Transaction cost economics ... 18
2.2.1 Modigliani and miller ... 18
2.2.2 Transaction costs ... 19
2.2.3 Optimal capital structure ... 19
2.3 Agency theory ... 20
2.3.1 Jensen and Meckling ... 20
2.3.2 Criticism on agency theory ... 21
2.4 Agency relations ... 22
2.5 Adverse selection ... 24
2.5.1 Signaling ... 25
2.5.2 Herding and information cascades ... 26
2.5.3 Wisdom of the crowd ... 27
2.6 Moral hazard and hold up ... 29
2.6.1 Group investments ... 29
2.6.2 Co-investments ... 29
2.7 Venture capitalists ... 30
3. METHODOLOGY ... 35
3.1 Exploratory case study ... 35
3.2 Validity, reliability and generalizability ... 35
3.3 Data collection ... 36
3.4 Data analysis... 37
4. RESULTS ... 38
5. DISCUSSION ... 40
5.1 Deal flow ... 40
5.2 Screening and evaluation ... 41
5.3 Due diligence ... 41
5.4 Non-financial support ... 41
5.5 BAs/VCs co-investing with the crowd ... 43
6. CONCLUSION ... 46
6.1 Managerial implications ... 47
6.2 Implications for theory ... 47
6.3 Recommendations for future research ... 48
6.4 Limitations... 48
REFERENCES ... 49
APPENDIX I: Previous research on crowdinvesting ... 60
APPENDIX II: Crowdinvesting platforms ... 63
APPENDIX III: Estimated market development crowdfunding ... 66
APPENDIX IV: Risk return profile of non-publicly traded companies ... 67
APPENDIX V: Investor interviews... Fehler! Textmarke nicht definiert. APPENDIX VI: Platform interviews ... Fehler! Textmarke nicht definiert. APPENDIX VII: Investor interview results ... 71
4
EXECUTIVE SUMMARY
This paper offers a critical review of the equity based crowdfunding model. The recent crowdfunding surge opens up a wealth of opportunities to entrepreneurs in search of financial capital. Additional advantages include the establishment of brand awareness, co-design and validation of product features, market testing, pre-selling of products, and customer feedback. However, due to its novelty, little is known about the effects and the risks of crowdfunding for equity. Through literature review, as well as in-depth interviews with professional investors and crowdfunding platform managers in Germany, the Netherlands, Belgium, and the UK, this paper offers novel insights, proposes avenues for future research, and presents solutions for sustainable growth.
Problem definition
Due to the lack of an established regulatory framework, most of the development of crowdfunding for equity is left to the market. This raises concerns about fraud, as well as poor investment decisions by an inexperienced crowd of small investors. In the absence of effective quality control mechanisms, information asymmetries between the crowd and the entrepreneur lead to problems of adverse selection (hidden information), moral hazard (hidden action), and hold up (hidden intention). In order to prevent market failure through erosion of trust, this study takes a transaction cost economics (TCE) and agency theory (AT) perspective on equity based crowdfunding. The central research question is:
How can information asymmetries, transaction-, and agency costs be mitigated between investors and entrepreneurs seeking financial capital via equity based crowdfunding platforms?
Crowdfunding
Crowdfunding can be defined as “a collective effort by people who network and pool their money together, usually via the internet, in order to invest in and support efforts initiated by other people or organizations”. Equity based crowdfunding distinguishes itself from donation, lending, and reward based crowdfunding through its revenue sharing instruments. The growth of crowdfunding (estimated $16,6 billion in 2014) can be attributed to the low transaction costs of the online placement process, the reduced costs of starting a business, and the decreased trust in financial institutions. Moreover, crowdfunding offers a solution for the early stage funding gap in equity investing caused by information inefficiencies and an inefficient capital market. Venture capital (VC) firms are increasingly investing further upstream, while business angel (BA) funding is difficult to access.
Theory
5
relevant for startup firms, which have a high failure rate overall. Financing through equity gives rise to problems of moral hazard and hold up. AT describes the screening- and monitoring costs that arise from an agent acting on behalf of a principal under conditions of incomplete and asymmetric information. This classical principal-agent problem helps conceptualize the problems of adverse selection, moral hazard and hold up for the various stakeholders in equity based crowdfunding, as further explained below.
Table 2: Agency relations
Principal – Agent relationship Agency Problems
Crowdinvestor - Entrepreneur Adverse selection: information asymmetries, lack of financial knowledge and resources in the crowd
Moral hazard: limited monitoring abilities, incompleteness of contracts Hold up: lack of decision making control
Entrepreneur - Crowdinvestor Adverse selection: unfulfillment of pledges Moral hazard: IP theft or copying of ideas
Crowdinvestor - Crowdfunding platform Adverse selection: lack of quality control in pre-selection of business propositions due to misalignment of interests and legal constraints Crowdfunding platform - Crowdinvestor Adverse selection: unfulfillment of pledges
Moral hazard: IP theft or copying of ideas, causing reputational damages Crowdfunding platform - Entrepreneur Adverse selection: constrained by resources and legal constraints to
pre-select high quality propositions, possible reputational damages as a result Entrepreneur - Crowdfunding platform Adverse selection: risk of unsuccessful fundraise, with possible
reputational damages and/or legal consequences
Adverse selection
Evidence from previous studies indicates that the crowd is able to distinguish and correctly act upon various signals of quality, thereby mitigating problems of adverse selection through rational herding behavior. Moreover, crowds are found to be remarkably accurate in predicting economic and financial outcomes. Several conditions must be met for the wisdom of the crowd to be effective: (i) each individual relies on private sources of information, (ii) individuals are not influenced in their decision by others and (iii) a mechanism is in place to collect and aggregate diverse opinions. Otherwise informational cascades may occur, in which individuals irrationally follow the decisions made by others. Platforms should seek a mix of independence and interdependence, limiting social influence while reaping the benefits of the collective knowledge of the crowd.
Moral hazard and hold up
6
control the group’s investments post-funding and (ii) a co-investment construct alongside professional investors. Studies on VCs and BAs show that the level of their post-investment involvement is critical to the success of their investments. As such, platforms should seek out engaged investors.
Methodology
An exploratory research method is used for a multi-case study on professional equity investors and equity based crowdfunding platform managers. Theory and data from semi-structured in-depth interviews are contrasted through an iterative process to shape propositions that serve as input for future research.
Results
The investors stressed the importance of formal due diligence requirements and generally expected a high rate of insolvency. The platform managers varied in their responses. Some stressed the importance of expert third party involvement, others claimed support by professional investors to be overrated. The largest intra group contrast is found on the ability of the crowd to make sensible investment decisions. The platform managers were very positive about the crowd's abilities, whereas the investors were more critical. Both respondent groups were positive overall on the added value of the crowd and were keen to the idea of professional investors co-investing with the crowd.
Propositions
Proposition 1: The deal flow quality of crowdfunding platforms will greatly improve through resolution of regulatory constraints and increased awareness of crowdfunding as a viable alternative means of raising startup capital
Proposition 2: Individual sophisticated investors are unable to significantly outperform a collective crowd of unsophisticated investors in predicting future business success
Proposition 3: Crowdfunding platforms with no due diligence requirements will produce significantly lower ROI for investors than crowdfunding platforms with minimal to high due diligence requirements
Proposition 4: Crowdfunded businesses that receive no professional operational support will produce a significantly lower ROI than those that do
Proposition 5: Crowdfunded businesses in which professional investors co-invest a significant amount with the crowd will generate a higher ROI, compared to crowdfunded businesses in which professional investors hold no significant share
Managerial implications and implications for theory
7
Harmonize pan-European regulations to enable cross-border investing. Clarify regulations to provide legal certainty about crowdfunding.
Promote transparency and cooperation between all stakeholders. High traceability and clear accountability creates and sustains trust.
Initiate and support research into the best practices of crowdfunding.
Raise awareness among informal investor networks about the benefits of crowdfunding. Harness the wisdom of the crowd through diversity of opinion, independence,
decentralization, and aggregation.
Implications for theory and recommendations for research
Contrasting results are found with regards to (ir)rational herding behavior by the crowd. It is critical to understand under which circumstances irrational herding takes place to prevent misuse and problems of adverse selection. The investment behavior by the crowd on online crowdfunding platforms should be further investigated in qualitative research. In addition, further research is needed on best practices, possible learning curves, the effectiveness of various online investment groups, as well as the attributes of (un)successful crowdfunded businesses.
Limitations
This study hinges largely on an aggregation of opinions and views and a comparison with previous findings from related research fields, which may not transfer well into crowdfunding. Coupled with the restricted scope of the study, these factors limit the generalizability of the results.
8
ABSTRACT
The research goal of this paper is to mitigate the problems of adverse selection, moral hazard and holdup that result from information asymmetries between entrepreneurs, crowdinvestors and crowdfunding platforms in equity based crowdfunding. The agency- and transaction costs that arise between the stakeholders are viewed through an agency theory (AT) and transaction cost economics (TCE) perspective in a multi case exploratory research study. In depth interviews with professional equity investors (BAs and VCs), as well as platform managers, reveal a lack of sufficient quality control mechanisms for venture capital seeking firms on equity based crowdfunding platforms, which is likely to result in a high insolvency rate. In order to improve pre-investment selection and post-investment monitoring and control, a solution is sought through a co-post-investment construct with BAs or VCs. Sharing a significant stake in the investment gives the professional investor the necessary incentives to screen out high quality business propositions and provide post-investment support, increasing the expected return on investment for crowdinvestors. In turn, the crowd's involvement improves the likelihood of success of the venture capital seeking firm through market- and product validation, as well as increased exposure.
9
1. INTRODUCTION
Crowdfunding is growing rapidly as a new means of financing a wide variety of projects online, giving anyone the opportunity to bankroll their dream. The most recent development in this modern democratized approach to finance is crowdfunding in the form of equity (also referred to as crowdinvesting). While banks and venture capital investors are increasingly moving their business further upstream, crowdfunding is emerging as a whole new asset class for early stage companies. Apart from raising capital, crowdfunding offers many other advantages to startup firms, including the establishment of brand awareness, co-design and validation of product features, market testing, pre-selling of products, and customer feedback. It has the power to disrupt the financial industry, disintermediating traditional financing models. The lack of a developed regulatory framework, however, has prompted concerns about fraud and bad investment decisions by the crowd. In order to provide solutions for sustainable growth, this paper offers an agency theory and transaction cost economics perspective on equity based crowdfunding through in-depth interviews with professional investors and crowdfunding platform managers in an exploratory multi-case study.
The success of crowdfunding has been overwhelming, close to doubling in volume every year as of 2009 and reaching a total funding volume of $2.7 billion in 2012 (Massolution, 2013). The recent passage of the JOBS (Jumpstart Our Business Startups) Act, permitting equity crowdfunding in the US, will add to an expected growth of $5.1 billion in 2013 (Massolution, 2013). The largest contributing factors to the growth are (i) the evolution of social media and web 2.0 technologies, facilitating access to the crowd at low cost and dramatically decreasing the costs of starting a business, (ii) the success of crowdsourcing and reward based crowdfunding carrying over to equity based crowdfunding, (iii) a decreased trust by the crowd in traditional financial institutions due to the systemic disruptions in the financial world, and (iv) a widespread squeeze on retail credit resulting from the recent economic downturn.
10
However, equity based crowdfunding differs from the more traditional capital markets in a number of ways and the lack of established rules and regulations threatens the sustainability of the model. Common regulatory approaches to financial intermediation typically lag behind the establishment of the common market itself (De Buysere, Gadja, Kleverlaan, & Marom, 2012). This paper therefore seeks to resolve the conflicts of interests that arise between the various stakeholders, adding to the increasingly important topic of financial consumer protection in the regulatory and academic debate (see e.g. Hildebrand, Puri, & Rocholl, 2013; Inderst & Ottaviani, 2009).
1.1 Problem definition
Investor protection is critical for equity based crowdfunding to become a sustainable asset class. Although few cases of misuse have been reported, fraudulent behavior is found to co-evolve with the introduction of new technological developments (e.g., Hazen, 2012; Rothchild, 1999). Moreover, small investors typically lack the resources and the incentives to secure their investments, exposing them to high risk. Similar to venture capital backed firms (see e.g., Keuschnigg, 2004; Lerner, 2001; Teten, AbdelFattah, Bremer, & Buslig, 2013), success rates for crowdfunded businesses are likely to drop significantly in the absence of adequate quality control mechanisms. In order to prevent the erosion of trust and the formation of a typical lemons market (Akerlof, 1970), pre-investment screening and post-investment monitoring and quality control will be critical to the functioning of the model.
First, the cost of acquiring financial capital is analyzed from a transaction cost economics (TCE) perspective. Transaction costs include the costs of searching, information gathering, bargaining, monitoring and enforcement. Subsequently, agency theory (AT) is utilized to analyze the principal-agent relationships between the most important stakeholders: the crowdfunders, the entrepreneurs and the crowdfunding platforms. AT describes the screening and monitoring costs that arise from an agent acting on behalf of a principal under conditions of incomplete and asymmetric information. The theory is frequently applied to explain the separation of ownership and control between the shareholders and the managers of a firm and thereby provides for a helpful perspective on crowdfunding. The principal-agent relationships relevant to the research can be found in the conceptual model in figure 1. The transaction costs and the informational asymmetries connected to these relations give rise to problems of adverse selection (hidden information), moral hazard (hidden action) and hold up (hidden intention).
1.1.1 Adverse selection
11
accurately be determined. Information asymmetries in crowdfunding arise due to the inherent higher knowledge of the entrepreneurs about their business model. Crowdfunding platforms refrain from screening activities due to legal constraints, whereas small investors usually lack the expertise and/or the time to properly vet a business proposition. Moreover, it would be highly ineffective to perform extensive due diligence on small investments. This is further aggravated by the entrepreneurs’ hesitancy to make detailed business information public, for fear of competitors copying their ideas. In other words, crowdinvestors currently face difficulties determining the value of a business, giving rise to problems of adverse selection.
1.1.2 Moral hazard
A second problem that arises from information asymmetries is moral hazard. Arrow (1963) was one of the pioneers of the theory, focusing on the risk of opportunistic behavior between two parties that enter into a contractual relationship. The ability of a small investor to monitor the behavior of an entrepreneur is constrained due to a lack of resources and/or incentives. As contracts are necessarily incomplete (Hart, 1995), other quality control measures are needed to resolve the monitoring problem in crowdfunding.
1.1.3 Hold up
The hold up problem plays a central role in the study of firm boundaries and transaction cost theory, originated by Coase (1937) and later adapted by Williamson (1979). It describes a situation in which one party gains increased bargaining power over another when relation specific investments are made. In crowdfunding, investors suffer from reduced bargaining power post-investment, having little mechanisms of control to secure their interests.
1.2 Research question
Mitigating these problems will be critical to prevent market failure and help build a sustainable crowdfunding model. As such, the following research question is formed:
How can information asymmetries, transaction- and agency costs be mitigated between investors and entrepreneurs seeking financial capital via equity based crowdfunding platforms?
12
entrepreneurial firms produce better results than those without the aid of professional support (e.g. Gompers, & Lerner, 2001; Keuschnigg, 2004).
1.3 Conceptual model
The problems of adverse selection, moral hazard and hold up for the most relevant internal stakeholders of the research are identified in the conceptual model in figure 1 and individually discussed in section 2.4. VCs and BAs, as well as regulators, are deemed critical to the functioning of the crowdfunding model and serve as important external stakeholders.
1.4 Research scope
The scope of the research is limited to equity based crowdfunding platforms, accessible to non-accredited investors in the general public. This excludes white label platforms, other types of crowdfunding, as well as platforms that are restricted to sophisticated investors. The intricate national and international judiciary systems with regards to financial law are largely left outside the scope of this paper. The main focus of the study is on the principal-agent relationships between the crowdinvestors, the entrepreneurs, and the crowdfunding platforms.
Crowdfunding
platforms
VCs / BAs
Project owners
Crowdfunders
Regulators
signaling effects / support / € policies / regulationslobbying policies / regulations / signaling effects lobbying € € € / signaling effects € / support / screening rewards / € monitoring / screening / feedback signaling effects / € signaling effects / Feedback Adverse selection Hold up Moral hazard
13 1.5 Paper outline
14
2. LITERATURE REVIEW
2.1 Crowdfunding
Crowdfunding is a subset of the wider defined concept of crowdsourcing. Howe (2006) first originated the term crowdsourcing as an open call to the general public by a company to outsource functions previously performed by employees. Saxton & Kishore (2013, p. 3) elaborate on Howe’s definition by explicitly adding advanced technologies to the definition, “crowdsourcing is a sourcing model in which organizations use predominantly advanced internet technologies to harness the efforts of a virtual crowd to perform specific organizational tasks”. The three key elements that can be distinguished in crowdsourcing are the (i) crowd, (ii) outsourcing and (iii) advanced internet technologies. Outsourcing and crowdsourcing are similar in that they both employ outside parties to fulfill their business objectives. However, where outsourcing traditionally relies on the contracting of professional third party companies, crowdsourcing solicits the help from an undefined, non-professional, and heterogeneous online crowd (Saxton & Kishore, 2013). Lastly, the crowdsourcing model relies on advanced web based technologies to involve the crowd in the production of goods and services. The minimal learning curves and the low cost of online storage, software and hardware, enable any company, regardless of size, to employ the capabilities and collective knowledge of the crowd.
Crowdfunding differentiates itself from crowdsourcing by drawing in financial resources. Ordanini, Miceli, Pizetti and Parasuraman (2011, p. 443) offer a useful description of crowdfunding by defining it as, “a collective effort by people who network and pool their money together, usually via the internet, in order to invest in and support efforts initiated by other people or organizations”. Crowdfunding itself is not new, nonprofit organizations have been soliciting funds from the crowd for more than a century. Moreover, traditional private placements, which have been in use for decades, are actually very similar. What sets crowdfunding apart is the reduction in transaction costs through the online placement process, allowing smaller investments to be made across a wide variety of private offerings. Crowdfunding would not even be possible without the dramatic decrease in costs for creating a new business.
15
2.1.1 Crowdfunding process
Crowdfunding is typically facilitated by online platforms which connect entrepreneurs with potential investors from the crowd. The crowdfunding process varies across the different platforms, broadly following these phases:
Project application: the entrepreneur applies at a crowdfunding platform, at which point the platform checks whether the project meets their required demands. The platform typically provides advice on the sales pitch, the rewards offered, the duration of the campaign and the target capital threshold. Platforms do not perform any due diligence, because of liability issues. Some platforms require due diligence by third parties.
Fundraising: when the platform and the project entrepreneur are satisfied with all the details, the campaign goes live. The entrepreneur now has a limited time frame to raise the required capital, with most platforms advising a running time of 30 days. The entrepreneur raises attention via social media and answers questions from the crowd.
End of the campaign: most platforms release the funds only when the target threshold is reached, otherwise all pledges are returned to the investors. Platforms typically charge a success fee of 5 to 10% of the total funding amount.
Post-investment: the entrepreneur typically provides progress updates and communicates with investors; directly, through the platform, or via social media. Platforms provide no direct post-investment support due to liability issues.
Table 1: Stakeholder motivators
Stakeholder Incentives Benefits Costs
Crowdfunding platform Extrinsic Revenues Market share Marketing costs Operational costs Entrepreneur Extrinsic Financial capital
Market testing Customer feedback Brand awareness Feature validation
High transaction costs due to many small investors Large commitment in time and effort
Theft of IP
Crowdinvestor Extrinsic and intrinsic
Social, material and financial return
Financial risk, loss of funds
16
Only 7% stated not to trust this new fundraising method. The incentives, benefits and costs of the various stakeholders are discussed in table 1.
2.1.3 Inefficient capital market
Many startup firms fail to raise critical early-stage capital, which has at least two causes. The first is information inefficiency, characterized by a failure to match capital sources with investment opportunities. Entrepreneurs often struggle to find equity investors (Ibrahim, 2008). Likewise, investors often lack sufficient information about potential investment opportunities (Sjostrom, 2004; Sohl, 1999). The major sources of small business capital are typically concentrated in certain geographical areas and tend to invest mostly locally (Bradford, 2012). Consequently, the search for startup capital is often a time-consuming process, with missed market opportunities and unsuccessful business attempts as a result. Crowdfunding offers a solution to this problem, connecting entrepreneurs with a large crowd of small investors around the globe.
Figure 2. Equity funding gap
The second cause is an inefficient capital market. Startup firms lack the collateral and operating history to access the sources of capital that are available to established companies. Coupled with the high risk involved with startup investing, entrepreneurs typically cannot access debt financing from banks and face difficulties raising startup capital from private equity investors. Consequently, they are left with bootstrapping methods and the help of friends and family (F&F) to finance the earliest stages of development. As these financial resources are quickly exhausted, entrepreneurs have to search for other sources of capital to realize significant growth. However, VCs have been increasingly moving their business upstream in recent years (Ernst & Young, 2013), investing less, at later stages of firm development, and on tougher terms. This widens the equity gap with F&F and BA funding (see figure 2). Crowdfunding is an asset class that complements angel investing and helps resolve the early stage and later stage funding gaps. First, it is suitable for the seed phase, where F&F are unable to provide sufficient funding and the funding amounts are too small for BAs to get involved. Second,
€100k €1M €2M
F&F
BAs
VCs
17
crowdfunding projects have been able to raise capital above the level that BAs tend to invest, bridging the gap with VCs and raising capital to realize significant growth. Third, crowdfunding enables access to new sources of capital by small investors. Fourth, crowdfunders typically have intrinsic motivations to invest, sharing a sense of ownership in the project, while contributing to the creative process. Consequently, they are willing to accept lower rewards than traditional equity investors for high risk projects. See figure 3 for a graphical representation.
Figure 3. Risk-reward ratios for startup funding
2.1.4 Policies and regulations
Most countries allow donation, reward and lending based crowdfunding, as well as various forms of revenue and profit sharing arrangements without voting rights. The selling of equity shares with voting rights, however, is restricted in most OECD countries and to date only allowed in the UK, Germany, Ireland, France, Switzerland, Australia and the Netherlands. Due to the passing of the JOBS Act, the US will likely follow in 2013. The way in which the various platforms operate and the financial constructs they use, differs greatly across each country and platform. The key differences between the various equity based platforms and the constructs they employ are summarized in appendix II.
18
readily available secondary market, crowdfunders face liquidity risk on their investments. Crowdfunding platforms typically do not facilitate such transactions, due to the risk of having to register as an organized securities market or alternative trading system. Thirdly, financial constructs with a pre-specified repayment timeframe (common in Germany) increases the risk of business failure when the company has not yet sufficiently developed.
As equity crowdfunding is a relatively new phenomenon, relatively little research has been performed on the subject of relevant control mechanisms. Despite increased attention by policymakers, regulators, investors and platform owners, the mechanisms and dynamics of crowdfunding are not yet well understood (Griffin, 2012). It makes sense, therefore, to first turn to the relating theory on traditional equity investing. First, the capital structure of a firm is discussed from a TCE perspective. Subsequently, AT is used to explain the problems of adverse selection, moral hazard and hold up between the crowdfunders, platforms and project owners.
2.2 Transaction cost economics
Cheap and easy access to external finance is crucial for firms in their early stages of development. Without sufficient capital a business is unlikely to get off the ground. In order to give insight into the role of crowdfunding as a new means of raising startup capital, it is important to first take a look at corporate financing as a whole. TCE offers a useful perspective to analyze the capital structure of a firm, which is largely determined by the capital cost of acquiring debt and/or equity. This in turn will shape the way in which contracts are formed and actions are governed. Modern corporate finance literature often cites the Modigliani and Miller theorem (1958), when explaining the capital structure decisions of a firm.
2.2.1 Modigliani and miller
19
investment increases. Ultimately, the debt-equity ratio of a firm is a function of the transaction costs of acquiring debt.
2.2.2 Transaction costs
Williamson (1985) attests that transaction cost theory is useful in discussing corporate financial structure. Transaction cost theory is a trade-off theory, which states that the relative cost of different financing options determines the capital structure of a firm. The firm’s choice between debt and equity is analogue to the choice between market exchange and vertical integration. Instead of viewing equity and debt as mere financial instruments, they are regarded as alternative governance structures of the firm, where individual investment projects are distinguished in terms of their asset specificity. In this sense, specific assets refer to non-redeployable projects, such as market and product development, which are best financed through equity. Non-specific assets, on the other hand, are investment projects with a general purpose, redeployable and best financed through debt. In sum, when the redeployability of an investment project increases, the debt-equity ratio will increase as well. The institutional environment in which the firm operates, also affects the firm’s capital structure. For instance, in countries with underdeveloped stock markets or weak legal enforcement, transaction costs increase (McLean, Pontiff, & Watanbe, 2009). When creditor protection improves, leverage will decrease and firms tend to rely more on long-term debt in their capital structure (Gianetti, 2003).
A key element in transaction cost theory with regards to the debt-equity ratio is the governance structure of the firm. Debt is similar to contracting in that it relies on discrete and easily enforced rules, preventing post contractual opportunism. Equity is more resembling of internal administration and can be easier adapted to changing circumstances (Alchian & Woodward, 1988). Financing through equity helps to prevent opportunism, but it is also more costly to set up. The two most important types of potential opportunism are moral hazard and hold up. Moral hazard can arise when the owner of a firm is unable to monitor what the manager of the firm is doing. The weaker the firm level corporate governance, the easier it will be for the manager to seek personal gain, thereby reducing firm value. This classical principal-agent problem will be discussed more thoroughly in the next section. Hold up opportunism arises when two parties refrain from working together due to concerns of giving the other party increased bargaining power, which can occur when transactions involve specific assets. In the case of equity, the hold up problem arises when the manager refuses to pay out a fraction of the firm’s free cash flows to shareholders or uses the funds to invest in negative net present value projects, while shareholders lack the necessary means of control (Saravia & Chen, 2008).
2.2.3 Optimal capital structure
20
able to earn more money on loan capital than it pays in interest. On the other hand, when a company is too highly leveraged, the increasing interest costs will start hurting the firm’s bottom line. In short, there is no one best way to finance a firm. The debt-equity relationship varies across industries, the type of business and the stage of development of the firm. For example, in asset intensive industries (e.g. mining, rail, telecommunications) the use of long term debt increases. Banks are more willing to finance fixed assets (property, plant, equipment) than intangibles (goodwill, R&D and advertising). Startup and early stage development firms, on the other hand, are financed more often through equity. As they lack collateral and steady cash flows to ensure regular interest payments, startups’ access to debt finance is limited (Berger & Udell, 1998; Gilson, 2003). They have to rely more on equity investors to gain access to capital. Contrary to lenders, shareholders often forego regular short term (dividend) payments, in the hopes of a large future pay off upon exit. The optimal capital structure will therefore be different for each firm and consists of a mix of debt, preferred stock and common equity. In general, companies should have lower debt and higher equity levels, because investors are more likely to contribute financial resources to firms with strong balance sheets. The decisions and the actions of such investors can prove insightful for the way in which the crowdfunding model can be optimally structured. In order to understand the way in which VCs and BAs control their investments, their relationship with the entrepreneur is viewed through the lens of AT.
2.3 Agency theory
One of the assumptions in the Modigliani and Miller theorem is that firm managers are presumed to act in the best interest of investors. However, when managers have a low equity stake in a company, they do not bear the full cost of expenditures. Consequently, agency costs in the form of moral hazard opportunism arise (Jensen and Meckling, 1976) as managers will be increasingly inclined to use company money for personal gain (e.g. a larger than optimal office, company car, etc.). A second crucial theme in AT is that of asymmetric information between equity investors and firm managers, which will be further discussed below. The relevant literature on AT is too extensive to provide an exhaustive review; this section therefore provides a general overview of the theory and the most representative theoretical works.
2.3.1 Jensen and Meckling
21
actions that would harm the principal’s interests. As a result, both parties will incur positive monitoring and bonding costs. The divergence that remains between the agent’s actions and those actions that would maximize return for the principal is referred to as residual loss (Jensen & Meckling, 1976). In summary, agency costs can be defined as the sum of (1) the monitoring costs of the principal, (2) the bonding costs of the agent, and (3) the residual loss.
Jensen and Meckling (1976, p. 310) view organizations as “legal fictions which serve as a nexus for a set of contracting relationships among individuals”. In this view, contracts stipulate the control mechanisms to mitigate agency costs between the principal and agent. The theory is quite general, it exists in all organizations, in all cooperative efforts, at every level and has been applied in various research fields; including marketing, finance, accounting, economics, political science, organizational behavior and sociology (Cuevas-Rodriguez, Gome-Meija, & Wiseman, 2012). The theory can be linked to utilitarianism as a model for human behavior, in which individuals seek to maximize their utility given a certain set of preferences (Rothbard, 1998). Two streams of literature with regards to AT can be identified: positivist and principal-agent. Principal-agent theory is more abstract and mathematical, applicable to any agency relationship. The positivist stream, the view taken in this paper, focuses primarily on the relationship between owners and managers of large corporations and has a more descriptive nature (Eisenhardt, 1989a).
2.3.2 Criticism on agency theory
22
satisficing behavior. Since it is difficult to model such a comprehensive behavioral view, AT effectively ignores these aspects (Hendry, 2005). Cuevas-Rodriguez et al. (2012) and Shapiro (2005) argue for a more behaviorist approach, taking into account the social context of the principal-agent contract, in which agents are motivated to act responsibly. Hendry (2005) adds to that by putting forward that governance mechanisms should not be designed to deal merely with problems of self-interest, control and accountability. They should include problems of honest incompetence and ways in which to enhance competence, principals should be concerned with ensuring good performance, not with ensuring accountability for performance. The following section discusses the relevant agency relations in more detail.
2.4 Agency relations
The most important stakeholders in crowdfunding are the crowdfunders, the project entrepreneurs and the crowdfunding platforms. One or more problems of moral hazard, hold up and adverse selection can be identified for all six principal-agent relationships (see figure 1 and table 2).
Table 2: Agency relations
Principal – Agent relationship Agency Problems
Crowdinvestor - Entrepreneur Adverse selection: information asymmetries, lack of financial knowledge and resources in the crowd
Moral hazard: limited monitoring abilities, incompleteness of contracts Hold up: lack of decision making control
Entrepreneur - Crowdinvestor Adverse selection: unfulfillment of pledges Moral hazard: IP theft or copying of ideas
Crowdinvestor - Crowdfunding platform Adverse selection: lack of quality control in pre-selection of business propositions due to misalignment of interests and legal constraints Crowdfunding platform - Crowdinvestor Adverse selection: unfulfillment of pledges
Moral hazard: IP theft or copying of ideas, causing reputational damages Crowdfunding platform - Entrepreneur Adverse selection: constrained by resources and legal constraints to
pre-select high quality propositions, possible reputational damages as a result Entrepreneur - Crowdfunding platform Adverse selection: risk of unsuccessful fundraise, with possible
reputational damages and/or legal consequences
Crowdinvestors (P) - Entrepreneurs (A)
23
crowdinvestors to monitor the entrepreneurs’ activities, coupled with the necessarily incompleteness (or absence) of investment contracts (Hart, 1995). The entrepreneurs may decide to use funds for private benefits, over which the investors have no control. Lastly, as crowdinvestors have no decision making authority or ability to renegotiate predetermined conditions, they suffer from problems of hold up when entrepreneurs do not act in the best interest of the crowd. Entrepreneurs could, for example, forego a profitable exit opportunity in order to reap the intrinsic and extrinsic rewards of their job as CEO of the firm.
Entrepreneurs (P) – crowdinvestors (A)
Conversely, entrepreneurs suffer from problems of adverse selection and moral hazard as well. They have no information about the creditworthiness or the intentions of the crowd. Pledges are usually not charged before completion of the project and may fall through. In addition, entrepreneurs share private information about their business with the crowd, opening themselves up to competitors looking to copy their ideas. Intellectual property (IP) rights cannot always be acquired and can sometimes be circumvented. Non-disclosure agreements are usually relatively ineffective as well (Collins & Pierrakis, 2012). This is especially problematic for early stage high tech projects, for which IP protection is a large concern.
Crowdinvestors (P) - crowdfunding platforms (A)
In a principal-agent relationship between crowdinvestors and crowdfunding platforms respectively, a problem of adverse selection may occur. When platforms inaccurately signal the quality of their screening and support activities, crowdinvestors may be led to believe that a project is not as risky as it really is. With many platforms currently fighting for market share, their short term interests to build traffic and create attention are not properly aligned with the interests of the crowd. Especially in these early stages of crowdinvesting, with no reports of failure, everybody seems excited to participate and invest. The high demand from the crowd carries the danger of a hype in which platforms host too many weak investment proposals. After the first companies become insolvent, the crowd will realize the inherent risk involved with venture capital investing. This carries self-correcting behavior, with the problem of adverse selection becoming less severe as platforms become more established. The effect can already be seen on reward based platform Kickstarter, which started to decline projects for the pre-selling of hardware products. This is most likely due to the long delays associated with these projects, leading to dissatisfaction in the crowd. Similar to online retailing, dominant players in crowdfunding will emerge that will gradually implement control- and quality disclosure mechanisms. Such mechanisms are inevitably necessary to prevent market failure (Dranove & Jin, 2010).
Crowdfunding platforms (P) – crowdinvestors (A)
24
creditworthiness of investors can be checked relatively easy, the problem of securing IP rights is more difficult to solve and abuse could negatively impact the reputation of the platform. However, the incentives of the platforms and the entrepreneurs on securing IP rights are perfectly aligned.
Crowdfunding platforms (P) - Entrepreneurs (A)
In the principal-agent relationship between platforms and entrepreneurs, problems of adverse selection and moral hazard arise as well. Platforms have to balance the risk-reward ratio of hosting a project, with possible negative reputational effects when it turns out to be unsuccessful. Although platforms may be willing to accept higher risks than crowdinvestors in their quest to claim market share, as previously discussed, platforms’ and crowdinvestors’ long-term interests are more tightly aligned in sifting out low quality campaigns. Platforms should be encouraged to be as open and transparent as possible and share their information on the entrepreneurs with the crowd, so they can jointly benefit from more effective screening.
Entrepreneurs (P) - crowdfunding platforms (A)
Conversely, entrepreneurs may suffer from adverse selection when a platform fails to generate sufficient attention for a project. An unsuccessful campaign signals low market interest and can negatively impact the project owner’s future success, whether in selling their end product or in attracting additional funding. In addition, entrepreneurs may be subject to prosecution in case of insolvency due to incompliance with financial legislations. The platforms Companisto and Seedmatch in Germany, for instance, circumvented the requirement to publish a financial prospectus (for fundraises over €100,000 and more than 20 shares) through a specific financial construct which can be described as a profit sharing loan (partiariches Nachrangdarlehen). However, if the construct is deemed an investment vehicle by the court in case of insolvency, the entrepreneurs can be held personally liable.
The information asymmetries and misaligned interests between the various stakeholders will be difficult to resolve. As entrepreneurs will be hesitant to make detailed information public, information sensitive projects will very likely not be suitable for crowdfunding. For less information sensitive projects, it is critical to determine how the crowd decides which projects to back, whether they make rational or irrational funding decisions, and to what extent they will be able to effectively screen out weak investment propositions. The following section draws on related literature research to address these issues for the problem of adverse selection.
2.5 Adverse selection
25
door to misuse and fraud. As crowdfunding is a relatively new phenomenon with limited scientific research to date, little can be said about the effectiveness of the crowd to select good business proposals. However, a wide body of related works is available on the effects of popularity indicators and adoption rates for consumption decisions, which helps to explain selection behavior by the crowd. Evidence from these studies indicates that the crowd is able to distinguish and correctly act upon various signals of quality, thereby mitigating problems of adverse selection through rational herding behavior.
2.5.1 Signaling
Signaling theory in economic transactions, originated by Spence (1973), explains how an agent conveys information to a principal in situations of asymmetric information. As the crowdinvestor (principal) is typically far less knowledgeable than the entrepreneur (agent) about the value of a business and its likely success, it is critical that the crowd is able to correctly interpret these signals and that the cost of acquiring the signals is sufficiently high. When both conditions are met, only high quality business propositions will be successfully funded, preventing fraud and misuse. Examples of such quality signals are historical financial data, human capital (education level, team size, age), firm characteristics (firm size, legal form, ownership structure) and industry characteristics (industry type, population density). The correlation between these signals and startup success is further explained in appendix IV.
Mollick (2013) attests that the crowd is able to correctly predict high potential projects, finding evidence that crowdfunders act on (costly) quality signals in a similar fashion as VCs. Mollick analyzed over 2000 hardware, software, video game and product design projects on Kickstarter that match areas of VC investment. He finds past success, third party endorsements and preparedness all to be strongly related to successful project funding. Past success (Gompers, Lerner, Scharfstein, & Kovner, 2010), demonstrated preparedness by entrepreneurs (X.P. Chen, Yao, & Kotha, 2009) and the strength of their social network (Stam & Elfring, 2008) have all been found to be effective in other investment settings, confirming rational decision making behavior by the crowd. Even though individual crowdinvestors are less knowledgeable than sophisticated investors (Bradford, 2012), Mollick’s findings indicate sound investment decisions by the crowd. As such, adverse selection does not appear to be as problematic as some critics make it out to be (Rattner, 2013).
26
Matthew Effect (Merton, 1957), where high quality projects are likely to be actively promoted to other potential investors. Important to note is the influence quality signals have on narrow-appeal products specifically. Tucker and Zhang (2011) studied hyperlink click-through rates and found low volume niche products to benefit greatly from quality signals, more so than high volume broad-appeal products. 1 It could therefore be argued that crowdfunding projects are highly susceptible to quality signals, which connects to the literature of observational learning.
2.5.2 Herding and information cascades
Classic analytical and empirical studies on observational learning find that individuals tend to follow their peers when judging the quality of a product (Banerjee, 1992; Bikhchandani, Hirshleifer, & Welch, 1992; Y. Chen, Wang, & Xie, 2011; Hirshleifer, Subrahmanyam, & Titman, 1994). This leads to what is referred to as herding behavior, where ‘everybody does what the rest is doing’ (Banerjee, 1992). Herding can be observed in a wide variety of fields and is typically found in situations where it is difficult for users to assess the quality of a product. Information cascades can develop as a result, in which users rely on decisions made by their peers to the point that it trumps any private information they may have.
Duan, Gu and Whinston (2009), for instance, researched download rates for software products and found informational cascades to significantly impact product adoption. Prior downloads strongly influenced decisions by subsequent users, whereas user ratings had no significant impact on product adoption for popular products. User reviews were only found to be significant for the adoption of products that were less popular. In addition, informational cascades acted as a stronger predictor of decision behavior than network effects, word-of-mouth effects and product diffusion. Likewise, Simonsohn and Ariely (2008) discovered irrational herding behavior in online Ebay auctions, whereas Salganik, Dodds and Watts (2006) studied social influence in music preferences and found the success of songs to be unpredictable and only partly determined by quality, suggesting irrational herding behavior. Similarly, Ward and Ramachandran (2010) studied crowdfunding for music projects on Sellaband. Investors in their study were strongly influenced by peer effects and common information indicators such as blogs and top-5 popularity lists, more so than individual information sources. These studies indicate that individuals are strongly influenced in their decisions by prior users, resulting in both rational and irrational herding behavior.
A caveat is that these studies focused on binary adoption measures (adopt versus non adopt) and common popularity indicators, whereas crowdfunding projects allow for more flexible funding and
1
27
typically make richer information available. In a study on peer-to-peer lending platform Prosper, for instance, Zhang and Lio (2012) reported rational herding behavior. Higher herding momentum was found to be correlated with lower default rates, implying that herding leads to sound lending decisions. Similarly, Yum, Lee and Chae (2012) on a study on peer-to-peer lending in microfinance, found evidence that lenders show herding behavior when little information on borrowers’ creditworthiness is available and rely on personal judgment upon receiving more quality signals. Kappuswamy and Bayus (2013) found no evidence for irrational herding in their study on Kickstarter projects and found future funder support to be negatively related to past funder support, “A high level of initial project interest in the first few days is quickly followed by decreasing support over most of the funding cycle” (p. 16). Lastly, Kim and Viswanathan (2013) studied the role of reputable investors in a crowdfunding market for mobile applications on Appbackr and found successful funding to be positively associated with ex-post app sales, indicating rational herding. However, even when people act rational, information cascades can lead to the adoption of inferior alternatives (Bikhchandani et al., 1992), which begs the question how ‘smart’ the crowd is.
2.5.3 Wisdom of the crowd
When a group of people make a joint decision, their aggregated judgment is often found to be as good as, and sometimes better than, the most experienced individual in that group. When individuals are unbiased in their decisions, their estimation errors tend to cancel each other out, leading to more accurate aggregate estimates than those of expert decision makers. This phenomenon is known as the wisdom of the crowd. Surowiecki (2004) lays out the following requirements for the wisdom of the crowd to be effective, (i) diversity of opinion, (ii) independence (to prevent information cascades), (ii) decentralization, (iv) aggregation (mechanisms to collect individual private judgments), (v) sufficient group size, and (vi) motivation to participate. Similarly, Seltzer and Mahmoudi (2013) offer a useful overview on the topic of crowdsourcing and emphasize the need for a well-defined problem, a diverse and heterogeneous crowd, as well as having clear benefits for the crowd to participate. Although most scientific research on the wisdom of the crowd is centered on numeric estimates or multiple choice selections, Yi, Steyvers, Lee and Dry (2012) demonstrate that the effect transfers into high dimensional practical decision making problems as well. Lakhani, Jeppese and Panettai (2007), for instance, have empirically shown online communities to be more effective at solving R&D problems than in house research teams.
28
correlations between variables (Ray, 2006, paras. 3-4). Although Hill and Ready-Campbell (2011) reported evidence in favor of expert decision makers, in a study on online user generated stock voting. The authors found the crowd to outperform the S&P 500 with their stock picking decisions, while experts yielded an even higher return.
Problematic is that most crowdfunding platforms violate the condition of independence between individual decision makers. Investors are able to see who invested before them, while platforms increasingly make profile information public. Evidence from social psychology indicates that through interaction, individuals tend to alter their behavior to correspond with the consensus of the group (Lorenz et al., 2011). Social influence thereby has a negative impact on the accuracy of group decisions. It significantly diminishes group diversity by promoting a convergence of estimates. A second effect is related to range reduction, where a small group of experts negatively influence individual decision makers in their estimates. This could lead to the clustering around a wrong value, as it is shown to move the position of the truth to the outer regions of the range, providing less reliable results (Lorenz et al., 2011 para. 9020). Furthermore, the convergence of estimates will lead to a false belief of collective accuracy. These effects are more profound for complex decisions, characterized by uncertainty, such as evaluations of business propositions.
29 2.6 Moral hazard and hold up
Platforms are both legally and practically constrained to monitor a company post-funding, while individual crowdinvestors have insufficient incentives or resources to control their investment. In the absence of third party quality control, investors have to rely on protection in the form of investment contracts. Following contract theory (Hart, 2003), contracts are complete when all parties can specify all rights and duties for any possible future event. Most crowdfunding projects are characterized by high uncertainty, for which it is impossible to predict every future outcome. As such, contracts are necessarily incomplete and provide insufficient protection against problems of moral hazard and hold up. Two possible solutions to this problem are (i) a group investment construct with an appointed group leader who is incentivized to control the group’s investments post-funding and (ii) a co-investment construct alongside professional investors.
2.6.1 Group investments
Group forming and the appointment of group leaders in lending based crowdfunding is shown to produce higher quality investment decisions, as long as group leaders’ interests are aligned with those of the rest of the group. Hildebrand et al. (2013) have shown that group leaders strongly influence community participation and that their behavior can be influenced through incentive schemes.2 The authors demonstrate that group leaders have a stimulating effect on participation rates, which can be either harmful or beneficial to the group, depending on the incentives given. It follows that similar outcomes can be accomplished in equity based crowdfunding, as long as group leaders are knowledgeable and have significant skin in the game to actively manage their investments. Most members of the crowd are not very well financially informed (Bradford, 2012) and lack the necessary knowledge to understand the risks involved with crowdfunding. However, investment expertise is no guarantee for success. Moreover, crowdinvestors will likely become more experienced over time. Freedman & Jin (2008) found lenders on the platform Prosper.com to improve their rate of return over a two year period. Due to the limited scope of the research, the remainder of the paper focuses an a co-investment construct with professional investors.
2.6.2 Co-investments
A second solution to protect against moral hazard and holdup is to co-invest alongside professional equity investors. As long as the sophisticated investor owns a significant stake, the crowd will benefit from their added quality control. The two most important groups of sophisticated investors that provide startup capital have traditionally been venture capitalists and business angels. The next sections review how these investors typically resolve information asymmetries.
2 In their research on online lending platform Prosper.com, they investigated the effects of incentives for group leaders (reputable lenders) on
30 2.7 Venture capitalists
AT in VC investing is covered extensively in scientific research. Due to its similarities with the crowdfunding model, the VC perspective is useful to understand the principal-agent problem in crowdfunding. VCs invest funds from non-active investors, who act as limited partners, in high growth high-return investment opportunities with an investment horizon of two to seven years (Cumming & MacIntosh, 2003). They tend to be very selective, accepting less than 1% of the business plans they receive (Bradford, 2012). In addition to financial resources, VCs add value through their network of suppliers, distributors, investors and specialists, and offer the entrepreneurs financial, legal, strategic, marketing and managerial advice (Cumminga & Johan, 2008; Mayer, Schoors, & Yafeh 2005). VCs are experts at screening startup firms and helping them grow, while solving problems of opportunism and asymmetric information (Kaplan & Strömberg, 2001).
Post-investment, conflicting interests with the entrepreneur lead to agency costs. For instance when entrepreneurs (1) underperform, (2) use firm resources to reap private benefits, (3) adopt strategies with too little/much risk relative to expected return, or (4) leave the firm at a time when it is difficult and costly to find replacement (Kaplan & Stromberg, 2003; Klausner & Litvak, 2001). Although VCs cannot eliminate problems of opportunism altogether, they can mitigate these problems in a number of ways. First, they can seek out high quality propositions through pre-investment screening and due diligence checks. Second, through post-investment monitoring they can collect information and advise the entrepreneur once the project is underway. Although VCs control board seats less frequently than commonly assumed (Kaplan & Strömberg, 2003), they often secure indirect control through the appointment of managing directors (Fried & Ganor, 2006). VCs thereby acquire substantial power over other participants in the business, allowing them to more easily monitor and influence performance of the entrepreneurs. Third, they can spread their risk by investing in a portfolio of startups or collaborate with other VCs in joint investments. Fourth, they can use incentive-aligning compensation schemes. Delaying funding until the entrepreneur achieves predefined milestones has shown to be an effective means of reducing information asymmetries and agency costs (Gompers & Lerner, 2001; Kaplan & Strömberg, 2003). Gilson (2003) attests that entrepreneurs signal performance quality by agreeing to these terms, effectively reducing screening costs. In addition, by aligning entrepreneurs’ interests with those of the VC, the entrepreneur’s holdup potential and the cost of monitoring by the VC are reduced. It effectively gives a strong performance incentive to the entrepreneur to obtain subsequent funding, thereby reducing agency costs. Lastly, VCs typically devise elaborate investment contracts to control their investments (Hart, 2001). VC investment contracts generally use the following safeguard measures:
Preferred stock: Kaplan and Strömberg (2003) have found that VCs often rely on the use of
31
Preferred stock is paid first in the event of liquidation, providing the VC with substantial protective rights. In addition, the issuance of preferred stock provides tax advantages and it provides entrepreneurs with stronger incentives to generate value (Fried & Ganor, 2006). Negative covenants: Negative covenants require VC approval for important company
decisions. They thereby prevent entrepreneurs from behaving opportunistically in situations where VCs lack control of the board (Ibrahim, 2008).
Exit rights: Exit rights include redemption (or put) rights, demand registration rights, and
conversion rights (Ibrahim, 2008). The interests of VCs and entrepreneurs can differ with regards to planning an exit, giving rise to agency costs. VCs need to provide returns for their fund investors and consequently prefer earlier exits. Conversely, entrepreneurs might be better served with later exits, enjoying the social status and psychic benefits of managing the firm, as well as steady salary income (Klausner & Litvak, 2001).
Keuschnigg (2004) found startup firms who are backed by VCs to outperform those who are not. Gompers and Lerner (2001) found that startups unable to attract venture capital within the first three years showed a failure rate of 90%, compared to a failure rate of 33% for those that did receive funding. However, VCs have not been able to generate the 20% annual returns they typically claim to generate for their limited partners and a recent Kaufmann study deemed the limited partner VC model to be broken (Mulcahy, Weeks & Bradley, 2012). Table 3 below compares VC returns with public market indexes in the US for the last 10 years.
Table 3: U.S. Venture Capital Index and selected benchmark statistics
Index 1-Year 5-Year 10-Year
U.S. Venture Capital Index 4.92 4.79 7.36
Early Stage Index 5.66 4.67 6.38
Late & Expansion Stage Index 4.11 8.15 10.96
Multi-Stage Index 2.44 3.87 8.03
Dow Jones Industrial Average 13,37 6.50 8.94
NASDAQ Composite 5.69 7.47 9.31
S&P 500 13.96 5.81 8.53
Source: National Venture Capital Association, data as of March 31, 2013
32 2.8 Business angels
Following Mason (2006: p. 3), BAs can be defined as “high net worth individuals who invest their own money, along with their time and expertise, directly in unquoted companies in which they have no family connection, in the hope of financial gain.” BAs help startup firms move through the critical early stages of development, to the point where it becomes attractive for VCs to invest. They typically invest in the range of $30,000 up to $2 million (Sohl, 2003), while providing informal advice in areas such as marketing, business development and securing intellectual property rights. Their advice is often seen as important as their financial capital (Ibrahim, 2008; Ramadani, 2012). Most BAs are ex-entrepreneurs with years of hands-on experience who actively participate in the development of the firm and have been quoted for fun and altruistic motives, besides pure financial gain (Christensen, 2011; Paul, Whittam, & Johnston, 2003). Sohl (1999) found the efforts of BAs to be mostly complementary rather than competitive with VC investments and estimated that they provide around 80% of the financial resources for early stage firms. Although difficult to estimate due to its informality, studies suggest that the capital market for BAs is larger than the capital market for VCs (Mason, 2006; Ramadani, 2012; Wiltbank, 2005). BAs thereby fund significantly more startups than VCs due to their smaller investments. Both Google and Amazon profited from early angel investments, before receiving large capital injections from VCs to help them grow out to the multi-billion dollar companies they are today (Ibrahim, 2008).
Kerr, Lerner and Schoar (2010) find BA funding to have a large and significant impact on startup survival and growth, while DeGennaro and Dwyer (2010) attest that BAs are able to earn similar returns as VCs. These results indicate that BAs have a highly positive influence on the success of the firms they back, although verifiable statistics on their returns are hard to find. The Angel Investor Performance Project reveals that average BA returns are 2.5 times their investment, although 50-75% of BAs earn less than they invest (Wiltbank & Boeker, 2007). BAs increasingly spread their risk by co-investing in groups, similar to VCs (Ibrahim, 2008). Interestingly, they exert fewer control rights than VCs (Goldfarb, Hoberg, Kirsch, & Triantis, 2012; Wong, Bhatia, & Freeman, 2009). This appears counterintuitive, as BAs face significant risks by investing in the earliest stages of development of a firm. It would therefore be expected that they seek more protection through elaborate investment contracts.
The angel capital market is characterized by (1) informality, with BAs operating behind the scenes, and (2) secrecy, to prevent them from receiving a flood of funding requests. Consequently, they are hard to find, both for entrepreneurs and academics.3 The number of studies on BA contract design is therefore limited. These studies are consistent, however, in that angel investment contracts use less restrictive means of control than those of VCs, employing none of the safeguard measures VCs use to
3
