The joint production of confidence – self-regulation in European crowdfunding markets

2.1 Literature review of the theory of self-regulation

In the legal scholarship and the scholarship of law and economics, self-regulation is understood as collective rules of conduct which do not originate from the government. These rules are meant to produce market outcomes which would not have been achieved if market participants were left to market forces alone (Ogus, 1995, p. 587). Self-regulation is taken to imply “the situation of a group of persons or bodies, acting together, performing a regulatory function in respect of themselves and others who accept their authority” (Black, 1996, p. 27). It helps achieving “public policy objectives at lower cost than government regulation” because of the information advantage which industry representatives have over regulators (Bowen, 2019, p. 257). As opposed to formal institutions of the state, self-regulation belongs to informal, private and social institutions, which may also serve to facilitate search and information by members of society (Dixit, 2009). Self-regulating institutions can imply one or several functions: formulation of rules, their interpretation, resolving disputes among group members and enforcement, which usually includes some form of sanctions (Black, 1996).

However, self-regulation does not necessarily imply an absence of government. Four ways in which state actors can relate to self-regulatory groups and bodies are (Black, 1996, p. 27):

1. Mandated self-regulation: the government delegates regulatory powers to self-regulatory groups;

2. Sanctioned regulation: the government approves rules drafted by the self-regulatory group;

3. Coerced self-regulation: as a response to a rising threat of government regulation not tailed to industry's needs, self-regulatory groups create standards;

4. Voluntary self-regulation: without government engagement of any kind, members of an industry or a profession create common standards for the behaviour of market participants.

Rules formulated by self-regulatory groups can be legally binding or non-binding (Ogus, 1995, p. 589). The breach of legally binding rules leads to private or public law consequences. The breach of non-binding rules leads to deprivation of certain rights within a group or association, such as ostracism. Sometimes non-binding regimes allow group members to opt out from certain norms, as long as a more appropriate behaviour can be demonstrated. Self-regulatory rules formulated as recommendations on a voluntary basis lead to no sanctions in case of non-compliance.

The question of the advantages and disadvantages of self-regulation is central in discussing the effectiveness of such a framework. Van den Bergh (2004, p. 3–8) argues that self-regulation is desired in industries in which market participants possess information advantage over regulators. This information advantage allows the creation of more suitable rules and achieves cost savings in relation to the enforcement of the rules. On the other hand, industry members lack democratic legitimacy to create and enforce rules. A lack of governmental oversight might lead to an abuse of self-regulatory powers to restrict competition.

The emergence of self-regulatory organisations depends on its advantages for the companies bound by the self-regulatory framework. Self-regulation, similarly to government regulation, is a result of a market failure in the form of information asymmetries, externalities or public goods. On the assumption of unconstrained competition among firms, which internalise all the costs of their activities, and well-informed consumers, who can choose the supplier with the best quality, regulatory rules would not be necessary. Under this assumption, firms that opt for low-quality standards would be forced to leave the market.

Self-regulation, as was described in relation to the emergence of law in primitive societies (Benson, 1988; Posner, 1980), is based on voluntary reciprocal arrangements. Members of an industry or a profession are willing to comply with self-regulatory norms if they can expect that other members will also adhere to such norms. The adherence to and sustainability of self-regulatory regimes depend on the ability of the group to enforce rules. The most common mechanisms of enforcement are the threat of expulsion of the non-compliant member and temporary suspension of membership rights.

Theories explaining the emergence of private ordering emphasise either the “coerced” or the “voluntary” nature of self-regulatory institutions, both are discussed in detail further.

One strand of economic literature emphasises that self-regulatory rules and organisations emerge because of a threat of government regulation. Maxwell et al. (2000) argue that firms are willing to reduce their harmful behaviour in order to save on political transaction costs associated with government regulations. These costs stem from organising themselves to influence policymakers once they enter into a political process (p. 586). Stefanadis (2003) develops a model that explains the emergence of self-regulation in an innovative industry in which new products are developed. Self-regulatory organisations can be regarded as a “club owned by industry participants” (Stefanadis, 2003, p. 5). Their incentive to introduce self-regulatory standards is to prevent the bureaucratic delays associated with ex-ante approval of new products in the system of statutory regulations. The advantage of an industry club formulating and enforcing regulatory rules is its superior knowledge about new technologies, which allows the club members to save on costs of delaying new products.

The other strand of economic literature assumes that self-regulation emerges independently of the government intervention on a voluntary basis to expand or complement existing regulation. Due to information asymmetries, consumers are unable to distinguish between the product qualities of different firms whose products typically represent credence or experience goods[1]. It might be profitable for a firm to offer higher standards, yet it might as well be prohibitively costly to produce credible signals of quality. Under conditions of uncertainty, adverse selection emerges (Akerlof, 1978). Gehrig and Jost (1995) explain that self-regulation of industry members serves as a signal of quality to outsiders. Industry members have strong incentives to enforce and punish any member that deviates, given that a failure to comply with industry standards weakens the signal of the entire club.

Prakash and Potoski (2007) explain why companies adhere to voluntary rules whose purpose is to reduce negative externalities. Industry members who pursue voluntary programmes create a club, as an “institution that supplies impure public goods” (Prakash and Potoski, 2007, p. 776). Impure public or club goods are used within the meaning of the Buchanan theory (Buchanan, 1965) as goods that are exclusively used by the members of a club in a non-rivalrous way. In this context, club goods are “affiliation with the club positive brand reputation” (Prakash and Potoski, 2007, p. 777) and are only available to those industry members who incur the costs of producing positive (reducing negative) externalities. Both Gehrig and Jost (1995) and Prakash and Potoski (2007) emphasise that self-regulation overcomes the free-rider problem of public goods by introducing mechanisms of exclusion of those club members who do not incur the costs of protecting club's reputation.

Gorton and Mullineaux (1987) and Yue and Ingram (2012) argue that the reputations of different firms in the same industry are interrelated. A non-compliance by one member breaking the trust of consumers harms the reputation of the industry. Adherence to high-quality standards by competitors creates a positive image of the industry, splitting rewards among all members. Yue and Ingram (2012) claim that “reputation commons” cause the emergence of self-regulation in industries where “outsiders are not equipped with specialised knowledge to distinguish between the behaviours of the industry's individual members” (Yue and Ingram, 2012, p. 6). It is in the self-interest of market participants to cooperate with respect to setting and enforcing rules. Cooperation among industry members allows for mutual benefits that could not be attained through market forces.

The aforementioned theories were developed outside the crowdfunding context in relation to different financial and non-financial industries. They are nevertheless relevant to the crowdfunding market in Europe, given that this industry has been governed by a self-regulatory regime since a few years after its emergence. Platforms in various countries have adopted CoCs, which were drafted and promoted by crowdfunding associations. Self-regulatory regime emerged in very different regulatory contexts, sometimes in anticipation of crowdfunding-specific regulation and in other instances after, or, as a reaction to it. A widespread practice of adopting CoCs in crowdfunding markets in different European jurisdictions raises the question as to which of the aforementioned theories is best able to explain their purpose. Answer to this question has multiple implications for the regulators in these countries, given that the motivation behind this phenomenon is likely to affect the outcomes in these markets. Moreover, the content of self-regulatory rules is informative about which aspects of crowdfunding business are a potential source of market failure.

The following subsection discusses characteristics of crowdfunding markets which provide support for competing theories of self-regulation. The following subsection outlines a theoretical framework for understanding the purpose of CoC in the crowdfunding markets and discusses the motivation to focus on the institutional design of the CoCs.

2.2 Characteristics of the crowdfunding market and the theory of self-regulation

In an attempt to discern the motivation behind a widespread practice of adopting CoCs in different European jurisdictions, it is informative to look at some of the characteristics of the crowdfunding industry.

Several features provide support for the theory of “coerced self-regulation”. Crowdfunding platforms, which act as financial intermediaries between projects and investors, are part of the digital transformation of businesses and profit from the widespread use of social media (Cumming and Hornuf, 2018, p. 12). The novelty of this industry consists of providing new channels of communication and new ways to gather information, which allow investors to assess the viability of projects better. Technological tools to estimate the risk of projects and match it with investors' risk preference are commonly used. From the point of view of projects, the participation of a significant number of investors also generates information about their individual preferences for goods or services they offer. This is commonly denoted as “wisdom of crowds” (Surowiecki, 2005). Crowdfunding platforms have been experimenting with different business models and products, including different financial instruments, tools to pre-select projects and govern investments after the completion of a campaign. Regulatory delay in approving new business models and products may stifle these innovative efforts. This incentivises industry members to develop self-regulatory rules in the form of CoCs, which are more flexible and easily revised, allowing to keep pace with industry developments. In line with Stefanadis (2003, p. 5), self-regulation in crowdfunding, which pre-empts government regulations, can be attributed to continuous technological change in this industry and the requirement to ask for ex-ante permission from the regulator to introduce new products and services[2]. Similarly, (Lehner and Harrer, 2019, p. 89) argue that crowdfunding platforms have an incentive to work together to inform and influence legislation to enhance the current regulatory status. These arguments are consistent with empirical findings that the “changes to regulations” are ranked as one of the highest risk factors by platforms operating different business models (Ziegler et al., 2019, p. 53–54).

Support for theories of “voluntary self-regulation” comes from the nature of services that crowdfunding platforms offer, which can be characterised as “credence goods” (Darby and Karni, 1973, p. 69; Nelson, 1970). Investors are often not able to judge the quality of platforms' service neither before nor right after their experience on the platform. Crowdfunding platforms have a specific “look and feel” which make them indistinguishable from other platforms at first sight. The presentation of projects on a platform often includes a pitch video, a project description, a team section and a bar graph displaying the volume funded by the crowd[3]. The similarity of the websites' designs across the industry stems from the interaction between competitors duplicating successful website layouts. By providing user interfaces which are known to potential investors, thus, reducing the switching costs between platforms, market players increase the likelihood of investment. The layout can be replicated at low cost by a new entrant. Thus, the setting up of the platform does not serve as a credible signal of quality. The recognisable design increases the costs for an investor to infer the quality of platforms' services which are relevant for the selection of a platform such as fees, due diligence of projects, operational infrastructure, data protection standards and post-campaign services. The “credence” nature of these services is reinforced by the fact that retail investors typically pledge small amounts of money which do not justify high costs of platforms' comparison. As a result, platforms which offer high-quality service are unable to distinguish themselves from low-quality ones. In addition, they are likely to experience negative reputational spillovers from platforms which deceived investors (Akerlof, 1978).

In line with the voluntary self-regulation theory of Gehrig and Jost (1995) and Prakash and Potoski (2007), crowdfunding platforms have an incentive to coordinate their behaviour. They create club-like institutions which enforce rules for high-quality service. Such institutions are embedded in CoCs, which stipulate substantial rules of platforms' behaviour, as well as procedural rules of enforcement. Through the mechanism of a threat of exclusion of those platforms unwilling to adhere to the high standards, platforms ensure that high standards of service are achieved and benefits enjoyed by complying members. Enforcement can also be achieved through the withdrawal of specific rights, such as the right to make public their membership in the club-like institution. These rules also serve as a means to restrict competition by increasing barriers to market entry.

The emergence of CoCs in the early stages of industry development underpins an argument that it is a means to protect “reputation commons”. When CoCs were introduced in different European countries, the crowdfunding industry was going through a legitimisation process, which has a cognitive and socio-political component (Aldrich and Fiol, 1994, p. 648). Cognitive legitimisation refers to the public knowledge about crowdfunding activity, whereas socio-political legitimisation implies public acceptance of the industry as appropriate and economically viable. As shown by Jegelevičiūtė and Valančienė (2015), CoCs are one of the tools through which the crowdfunding industry is promoted by its members. The interdependence between the individual reputation of platforms and the reputation of the crowdfunding industry has been highlighted in several surveys. Platform operators rank the risk of a failure of a prominent competitor as very high (Ziegler et al., 2018) suggesting that “reputation commons” are threatened. However, the existence of separate CoCs for lending- and equity-based crowdfunding, such as in the United Kingdom, suggests that the fear of a failure does not necessarily transcend the business models of crowdfunding. Equity-based crowdfunding platforms might not be afraid of a failure of a lending platform, and vice versa.

Different characteristics of the crowdfunding market are able to underpin both “coerced” and “voluntary” theories of self-regulation and explain the widespread reliance on CoCs. Unlike in traditional financial markets, many aspects of the crowdfunding activity are left to platforms to choose, even in jurisdictions which have adopted bespoke crowdfunding regime – which are regulatory regimes that are tailored to the technological specifics of crowdfunding markets. This leeway of platforms is a result of incompleteness of existing regulations, uncertainty regarding inherent risks, or platforms' innovative efforts which can result in circumvention of applicable rules. In many instances, the applicability of certain laws that are not crowdfunding-specific is subject to ambiguous interpretations.

In the following subsection, this paper will discuss how the institutional design of CoCs can explain the emergence of self-regulation in crowdfunding markets.

2.3 Institutional design of codes of conduct and the theory of self-regulation

Mere features of the crowdfunding industry do not provide a definite answer which of the theoretical explanation is the most plausible. In contrast, the institutional design of the chosen self-regulatory regime provides better guidance. In the rest of this section, the paper argues that if the CoC is rationally designed, some of the institutional features can reveal the rationale for the adoption of CoCs. In relation to this, the paper outlines a simple theoretical framework for the study of CoCs in the crowdfunding industry, which builds upon competing theories of self-regulation discussed earlier.

Each of the three theories of self-regulation (“coerced theory”, “club theory”, and “reputation common theory”) emphasise a different economic problem that market players are trying to solve when adopting and adhering to self-regulatory rules. As a consequence, the institutional design of CoCs is expected to reflect the underlying rationale for their adoption.

The “coerced theory” of self-regulation emphasises the threat of government regulation. As discussed earlier, it is either savings on “political costs” or “costs of delaying the introduction of new products” that motivate market participants to introduce a self-regulatory regime. In order for the industry to be successful at pre-empting or deferring government regulation, the adopted self-regulatory framework must emulate the framework which a government would create. Deferring state regulation implies that CoCs are introduced before crowdfunding-specific regime, although it is conceivable that “coerced CoCs” appear with the intention of deferring further revisions of the crowdfunding-specific regime. Given that government regulations would be mandatory for all members of the industry, “coerced CoCs” are likely to be open for signing by all platforms (of a particular business model pertaining to the perceived scope of the regulation). Given that government regulation relies on centralised monitoring by the regulator and robust enforcement mechanisms, “coerced CoCs” are expected to have a similar institutional design. For instance, in a “coerced CoC”, a crowdfunding association is likely to take over the monitoring and enforcement role. As to the content of substantial rules of CoCs, a “coerced CoC” would create new obligations for platforms. Such a CoC would not emphasise rules already written in laws and regulations, as these are already enforced by the regulator.

Voluntary theories of self-regulation deny that self-regulation is primarily a result of a threat of government regulation. Instead, voluntary theories of self-regulation propose that it is in the best interest of market participants to create rules which would help them resolve the collective action problem inherent in public goods or the issue of overuse of commons. “Club goods” theories assume that self-regulations resolve the problem of free riding in relation to industry reputation by creating a club of close-knit market participants. These market participants comply with self-regulatory rules in order to benefit from a credible signal of quality. For the signal to be credible, the institutional design of the self-regulatory regime has to solve the issue of excludability of non-compliant industry members.

In the crowdfunding context, “club good” rationale for CoCs entails that a membership will be limited to few platforms willing to adhere to high-quality standards, which will be clearly distinguishable from the rest of the industry. A right to display a logo or other label associated with a CoC is likely to be prominently displayed. Unlike “coerced CoCs” which are likely to appear before state regulations, CoCs that create a club of quality platforms are likely to appear both before and after state regulations. A self-regulatory club can emerge in a regulated environment as well because incomplete, unclear or insufficient laws and regulations are a threat to market participants aiming to provide high-quality services. In order for exclusion to be a credible threat, such CoCs are likely to introduce strong monitoring and enforcement mechanisms. However, unlike “coerced CoCs”, decentralised monitoring, in which platforms supervise compliance of each other, is likely to be also introduced. This is due to the fact that small membership base leads to reduced costs of monitoring as well as transaction costs of coordinating monitoring efforts. Since excludability lies in the definition of a club, CoCs that serve as a club-like institution are expected to have strong enforcement mechanisms in place which would ensure compliance. Given that self-regulatory clubs serve as a quality signal, this signal might as well be used to demonstrate compliance with existing state regulations. As such, substantial rules of CoC are likely to restate platforms' obligations, which are already required by existing laws and regulations.

“Reputation commons” theory considers self-regulation as a means to overcome the issue of overuse of commons. In the crowdfunding market, industry reputation is being “overused” because platforms cannot internalise all the costs they create for the industry reputation when lowering the quality of their service. Low-quality service of platforms creates spillover effects for the entire industry, given the interconnectedness of the reputation of all industry members. As a consequence, CoCs are likely to create rules which will help to get all market participants “on board”, by increasing the benefits of cooperation or increasing the costs of deviation. Similarly to other voluntary theories of self-regulation, the adoption of CoCs is independent of the time of introducing state regulations in a particular jurisdiction. Inclusive membership of such CoCs also dictates monitoring and enforcement mechanisms. Due to a large number of members, monitoring is likely to be delegated to a crowdfunding association or another designated body. However, their monitoring powers might be weak due to the resistance of smaller platforms to more intrusive methods of detecting non-compliance. As a result of a compromise of a large number of actors, rules are likely to reflect the common ground among them. Consequentially, weak monitoring is often accompanied by weak enforcement. As to the content of CoCs substantial provisions, both new rules and reference to existing rules are likely to be included, since industry reputation is equally at stake when existing laws are not adequately implemented.

Table I summarises the theoretical framework outlined earlier, emphasising common institutional dimensions between existing theories as well as their distinctive dimensions.

In conclusion, the three competing theories share many institutional characteristics. The main argument of distinction between “coerced” and “club goods” theories is exclusive membership, decentralised enforcement and reference to existing laws, which are likely to appear in relation to club-like institutions. Self-regulatory regimes which emerge as a response to a threat of government intervention are also unlikely to have in place weak monitoring and enforcement mechanisms inherent in the institutional design of CoCs that aim to protect reputation commons. Finally, the difference between the two voluntary theories of self-regulation lies in the inclusiveness of membership and efficacy of monitoring and enforcement mechanisms.

The rest of the paper relies on this theoretical framework to discuss what the institutional design of CoCs adopted in different European jurisdictions suggests about the economic rationale for self-regulation in crowdfunding.

4.1 Origins of codes of conduct

Most crowdfunding markets in Europe emerged before a bespoke national crowdfunding regime. A bespoke crowdfunding regime is a regulatory framework designed specifically for crowdfunding activities, platforms and other market participants (such as issuers/borrowers and investors/lenders). Bespoke regimes can differ with respect to the scope and extensiveness of tailored rules.

Existing rules tailored to traditional financial intermediaries are not necessarily applicable and suitable to a nascent industry such as the crowdfunding industry. Yet crowdfunding markets face the same issues that led to extensive regulations in capital markets and banking: market failures associated with asymmetric information and negative externalities. The magnitude of these issues is unknown to policymakers. In some countries, they deferred the adoption of a crowdfunding-specific regulatory regime until the industry reached a certain stage of maturity (Wenzlaff, 2019). In some other countries, regulators introduced bespoke-crowdfunding regimes to stimulate its development. Some empirical evidence shows that there is a high correlation between adequacy of crowdfunding regulations as perceived by platforms and volumes per capita (Ziegler et al., 2019, pp. 55–56). Even in countries with high volumes per capita, some of the relevant questions often remained unaddressed by regulations or gave rise to ambiguities in interpretation. This has created an incentive for platforms to take pre-emptive actions.

Crowdfunding platforms in different countries, usually under the auspices of national or transnational crowdfunding associations, introduced a self-regulatory regime in the form of CoC. Table V provides an overview of the association, the country of origin, the CoC and its denotation in this paper, the year of adoption and the year of adoption of the respective crowdfunding regulatory framework[4].

The study identified nine relevant CoCs in crowdfunding markets in Europe, as listed in Table V. Two of the CoCs are transnational in character (European CoC and Nordic CoC), while the remaining seven are country-specific. The European CoC explicitly states that it does not constitute a court-enforceable regulation and that it does not overrule provisions of national CoCs. No such provisions can be found in the Nordic CoC.

Some CoCs are explicit about the possibility of foreign-based platforms operating in that country to join the crowdfunding association and adhere to the CoC (for instance, UK Crowdfunding CoC and UK Lending CoC). The United Kingdom is the only country where two CoCs were identified (lending- and equity-based crowdfunding).

CoCs of most of the countries were adopted under the umbrella of crowdfunding associations at the national or transnational level. In two countries, the CoC was published by an organisation whose membership is open to companies other than crowdfunding platforms. In Estonia, the CoC was issued by Finance Estonia, an organisation that gathers members from various Fintech industries. In Austria, the CoC was issued by a subsection of the Austrian Chamber of Commerce (Wirtschaftskammer Österreich) – Financial Intermediary (Fachverband Finanzdienstleister), which established a working group on crowdfunding (Arbeitsgruppe Crowdfunding).

CoCs have appeared in countries with most developed crowdfunding market, in terms of either total volumes (the United Kingdom, France, Germany, Netherlands, Nordic Countries, Austria) or volumes per capita (Estonia) (Ziegler et al., 2019). The relationship between the maturity of a crowdfunding market and the existence of CoCs and crowdfunding associations is twofold. CoCs may facilitate the growth of the industry by creating trust among different stakeholders. One of the purposes of crowdfunding associations is to provide institutional coordination among its members to fill out the gaps in regulatory frameworks, thus helping to improve the crowdfunding ecosystem (Kshetri, 2015, p. 102). However, the costs of operating a crowdfunding association and enforcing rules of CoCs become easily justifiable in a consolidated market. Given the number of different factors that affect the development of crowdfunding industry (Rau, 2017), the question of reverse causality between self-regulation and market development would merit further research beyond the scope of this paper.

The CoCs in different European countries originated between 2013 and 2017. Most crowdfunding markets were still nascent, yet facing strong growth. CoCs were either anticipatory or reactionary with respect to crowdfunding regulations. Out of 11 national markets covered in this study, including Nordic countries, five still do not operate under a crowdfunding-specific regulatory regime (Estonia, Norway, Sweden, Denmark and Iceland), yet they are governed by a CoC. Among the jurisdictions with tailored crowdfunding regulations, the United Kingdom saw the emergence of the two CoCs prior to the adoption of statutory regulations. Similarly, the European CoC was created in anticipation of the European Crowdfunding Regime. In line with its Fintech action plan, the European Commission presented a proposal for a regulation on crowdfunding service providers in March 2018, which is currently under review. UK lending-based crowdfunding market is specific for the fact that a bespoke-crowdfunding regime was also revised after the two revisions of the UK Lending CoC in 2019. Therefore, this market features very dynamic interactions between self-regulatory and statutory regimes.

In France, Germany, Netherlands, Austria and Finland, CoCs were created following the adoption of crowdfunding regulations. In several instances, the CoCs were also revised. As mentioned earlier, the UK Lending CoC was revised twice, in 2015 and 2018. The French CoC was revised in 2016. In Germany, the CoC published in 2016 was complemented by the Reporting Guidelines in 2017. In addition to the Guidelines, in 2019, the German Crowdfunding Association produced a list of criteria for receiving a “transparency certificate” or “label” of the association.

The timing of the emergence of CoCs in relation to crowdfunding bespoke regimes may point at the economic rationale for self-regulation in crowdfunding. However, this relationship is complex. By the very definition of anticipatory self-regulatory regimes, it is meant to prevent the forthcoming government intervention. Given that four out of nine CoCs were introduced or revised after crowdfunding-specific regulations, voluntary theories of self-regulation cannot be ruled out. However, as seen in the UK lending-based crowdfunding market, it is also possible that a CoC precedes revisions of the existing crowdfunding regulation. In other words, reactionary CoCs suggest that voluntary theories are likely to prevail in those countries in which revisions of bespoke regimes are not “on the horizon”. Further analysis of the institutional design of different CoCs is necessary to discuss which of the theories of self-regulation is the most convincing in crowdfunding.

4.2 Inclusiveness of codes of conduct

This section examines features of national and transnational CoCs which determine how inclusive membership in self-regulatory institutions is. Inclusiveness in this paper as the scope of business models covered by CoCs, preconditions for becoming a member and the right to display a logo for those platforms who fulfil certain criteria. Although crowdfunding can refer to different business models of operating a platform, the CoCs covered in the analysis of this paper make an explicit or implicit reference to the type of business model addressed.

The paper identified three types of CoCs in relation to covered business models:

1. CoCs aimed at all business models (Nordic CoC);

2. CoCs aimed at and restricted to only one business model (UK Lending CoC, Austrian CoC);

3. CoCs aimed at all business models, yet provisions mainly appropriate for financial-return business models (European CoC, UK Crowdfunding CoC, Estonian CoC, Dutch CoC, German CoC, French CoC).

The UK Lending CoC is the only example of self-regulation restricted to lending-based platforms. The restrictiveness to lending models can only be inferred indirectly, given that the Art. 4 of the Rules of Association stipulate that “any legal entity that operates a Platform is eligible for Membership of the Association provided that entity does substantive business in the UK and meets the Membership criteria published by the Board and amended from time to time.” The membership criteria lay down credit risk and operational risk management procedures and rules on a sound and responsible business model, many of which are meaningless for non-lending platforms. To the knowledge of the authors, the UK Lending Association is the only association for peer-to-peer lending platforms in Europe. However, some lending platforms have joined industry associations in the banking sector and thus have to adhere to the CoCs of these organisations. For instance, in Germany, the Banking Association expanded their membership criteria to include lending platforms. In the Netherlands, the Corporate Finance Association opened their membership to platforms which sign the CoC. In contrast, the Austrian CoC is limited to “operators of crowdinvesting platforms […] within the meaning of § 2 no. 5 AltFG”, which essentially limits the scope of the CoC to equity-based crowdfunding, since this model is covered by the bespoke-crowdfunding regime in Austria.

All other CoCs are open for signing by any type of crowdfunding platform, independently of the business model. For instance, provisions of the Nordic CoC are very general and aimed at all crowdfunding models. The transnational character of the association makes it primarily a tool to foster awareness and legitimise crowdfunding activity. Since members of the Nordic CoC face different regulatory regimes in their home jurisdictions, a more detailed CoC would not be sensible. In contrast to the Nordic CoC, all remaining CoCs (European CoC, UK Crowdfunding CoC, Estonian CoC, Dutch CoC, German CoC, French CoC) contain provisions which indicate that they primarily target financial-return models, in particular, equity-based crowdfunding platforms. For instance, the Reporting Guidelines and the Transparency Certificate of the German CoC contain provisions which are only relevant for equity-based crowdfunding. Thus, it is not surprising that their signatories are overwhelmingly equity-based platforms.

The fact that provisions of CoCs are often more tailored to financial-return crowdfunding (lending- and equity-based crowdfunding), and address donation and reward-based crowdfunding only in general terms, does not indicate that CoCs are meant to constraint membership, in line with “club goods” theories of self-regulation. In the case of donation-based and reward-based crowdfunding, the risks of market failures are simply less pronounced. As a consequence, both the threat of government regulation and reputation spillovers are more likely in relation to financial-return models.

Similarly, the existence of separate CoCs for lending and equity-based platforms in the United Kingdom might as well indicate that reputation spillovers between lending and equity crowdfunding are limited. This is in line with arguments in the literature that reputation externalities are less intense “the more numerous, distant, and heterogeneous are the members” (Barnett, 2006, p. 1763). Alternatively, the limitation of a CoC to a specific business model, such as in the United Kingdom and Austria, may mirror separate regulatory regimes at the state level. In sum, neither explicit nor implicit restrictions of CoCs to a specific business model are enough to conclude that CoCs are driven by an intention to create a club-like institution. However, such restrictions are indicative of the fact that potential market failures are not universal across different models.

In addition to business models, CoCs differ with respect to which platforms are allowed to join a self-regulatory regime. The paper identified three ways in which CoCs restrict who can become their signatory:

1. only members of the crowdfunding association;[5]

2. only platforms licensed by the state supervisory authority;

3. only platforms who fulfil certain criteria (such as annual volumes invested in the platform).

Seven out of nine CoCs apply exclusively to the members of crowdfunding association (European CoC, UK Lending CoC, French CoC, German CoC, Nordic CoC, Dutch CoC, UK Crowdfunding CoC, UK Lending CoC). In Austria, all crowdfunding platforms are required by the law to join the Chamber of Commerce, making the access to CoCs inseparable from membership in the association. The only true exception is the Estonian CoC (Best Practice Guidelines). It is not necessary to be a member of Finance Estonia in order to receive the label of the Best Practice Guidelines, but to follow the Guidelines and report it to Finance Estonia, who conducts checks on the annual basis and decides whether to renew the label.

In essence, restricting access to CoCs to members of the crowdfunding association is not indicative that platforms were aiming at creating club-like institutions. In the case of a large number of platforms, decentralised monitoring and enforcement are too costly to implement. The implementation of CoCs critically depends on the operation of a crowdfunding association which receives monitoring and enforcement powers from the members. Linking adherence to CoCs to adherence to crowdfunding associations is simply a way to cover the costs of implementing a self-regulatory regime.

In two cases, only platforms with a state licence are allowed to join a CoC. The Austrian CoC restricts the access to platforms that are licensed under the bespoke-crowdfunding regime in Austria (Alternativfinanzierungsgesetz). The rules for joining the Nordic Crowdfunding Association require prospective members to provide a statement of authorisation in their home market. In contrast, the UK Crowdfunding CoC specifically states that its rules apply to platforms regardless of whether they are authorised by the state authority.

The Dutch CoC is the only example of restricting membership to platforms which are already operating and fulfil certain requirements. The Dutch CoC stipulates that in order to become a member, platforms need to have at least five projects and must have raised at least 500,000 euros. While not part of CoCs, some statues of crowdfunding associations foresee similar membership criteria. For instance, the statute of the German association requires a platform to have been in operation for at least one year, to have a minimum of five crowdfunding projects and to have raised in total at least 500,000 euros. The rationale for limiting the membership to platforms that fulfil certain criteria is to intensify the quality signal of the club-like institutions. The more constrained the membership, the lower the costs associated with effective enforcement, and thus, the stronger the signal.

Overall, in relation to precondition for platforms to adhere to a self-regulatory regime, restricting membership is an exception rather than a rule. This indicates that, in relation to this criterion, there is more support for the “coerced theory” and “reputation commons theory” than for “club goods theory” of self-regulation.

However, there are three cases in which a specific right is conferred only to platforms who fulfil additional requirements in terms of quality of their service. This right refers to a display of a logo upon verification by the association that additional criteria are met by the platform. As described earlier, the label of the Best Practice Guidelines in Estonian crowdfunding market is only available to those platforms who regularly report to Finance Estonia, which conducts checks on the annual basis and decides whether to renew the label. Similarly, the transparency certificate (logo) of the German crowdfunding association is only available to those members of the association who comply with additional transparency requirements and provide various data to the association once per year, which enable the association to check compliance performance. The German case example is specific for explicitly distinguishing between standard members and those who also adhere to higher transparency standards. Display of a crowdfunding association logo, which is not subject to the fulfilment of any specific criteria verified by the association, is a common practice in various countries. Sometimes this right is explicitly mentioned in the CoC (European CoC, UK Crowdfunding CoC, German CoC, Estonian CoC). Some other associations provide a logo to their members, even though this is not referenced explicitly in the CoC.

The European CoC goes even further in terms of establishing mechanisms that provide credible signals of quality for a limited number of platforms. Art. 2.4 grants rights to ECN to deliver an independent ranking on compliance performance by members of the association. However, these rankings can be published on the ECN website subject to a member's prior acceptance. The ECN has not yet implemented a rating system of compliance. While disclosure of compliance would allow certain industry members to capture “ability rents” (Gorton and Mullineaux, 1987, p. 461), it would also deter a number of platforms from joining the association.

Another interesting provision which creates difference among platforms is known as “comply or explain”. Only two out of nine CoCs (the UK Lending CoC and the Estonian CoC) contain this provision. The “comply-or-explain” procedure can be found in Art. 4 of the Operating Principles of the UK Lending CoC, which also stipulates that waivers have to be approved by the Board Members of the Association and are published on the website of the association. In contrast, the “comply-or-explain” procedure found in the Reporting Form for Compliance in the Estonian CoC forces the platform to publish a deviation from the CoC on its platform. In contrast to the right of a logo, this provision serves to exemplify a non-complying platform, thus, preventing to weaken the quality signal of other members.

In conclusion, it is not a widespread practice to develop specific preconditions for membership as well as conferring certain rights only to a subsample of crowdfunding platforms. Most CoCs opted for a large and unrestricted membership base supporting the argument of either “coerced” or “reputation commons” theory. However, those few exceptional examples described earlier indicate that club-like self-regulatory institutions also emerge in certain crowdfunding markets. The case of German crowdfunding association reconciles both the “club” theory and “reputation commons” theory perspective since it ensures minimum standards for all industry members while at the same time allowing a stronger quality signal from a subsample of platforms. The findings of this section suggest that the rationale for CoCs might not be universal across different markets.

4.3 Monitoring and enforcement of codes of conduct

This section discusses how associations monitor and enforce rules created through CoCs. As to monitoring, which serves to detect platforms who fail to comply with self-regulatory rules, the paper identified three types of obligations borne by platforms:

1. Information duty – platforms are required to provide the association with information about their compliance;

2. Denouncing obligation – platforms are required to notify the association about unlawful behaviour of other platforms (including non-compliance with CoC);

3. Audits – platforms are required to allow audits and reviews by the association.

Table VI provides an overview of monitoring tools in different CoCs. The monitoring mechanisms are put in order according to how strong (efficient) they are, from the weakest (none) to the strongest (audits). It is possible to make a distinction between centralised (information duty and audits) and decentralised forms of monitoring (denouncing duty). A distinction should also be made between information duty and auditing, on the one hand, and denouncing the breach of CoCs by other platforms, on the other. The former mechanisms entail centralised monitoring by the self-regulatory organisation, whereas the latter involves decentralised monitoring by other industry members. Decentralised monitoring is expected to produce efficient outcomes in terms of detecting a breach, only in close-knit communities with limited membership. Both the costs of monitoring and free-riding problem increase with a broad membership.

French, Dutch and Nordic CoCs do not foresee any monitoring mechanisms. This does not exclude the possibility that such a mechanism is foreseen in other rules of association such as statutes. Only the German CoC foresees the information duty. Art. 4 of the Transparency Certificate Guidelines of the German CoC requires the platforms to provide information about the owners, the latest auditor's report, the data security officer and procedures in case of platform insolvency. This information is shared with a designated officer in the association, but it is not published externally or shared among the platforms.

Three CoCs foresee an obligation to denounce unlawful behaviour and non-compliance with the CoC. Article 15 of the UK Crowdfunding CoC describes and limits the denouncing obligation at the same time. The article states that the ability of platforms to denounce negative behaviour is limited by confidentiality obligations towards its users. Similarly, the Austrian CoC in Art. 14 ensures that deviations from the CoC are being brought to the attention of the Subcommittee of the Financial Intermediaries in the Chamber of Commerce. Since the Subcommittee of Financial Intermediaries is higher in the hierarchy than the Working Group for Crowdfunding Platforms in Austria, this provision entails a severe threat to the platform which breaches the CoC.

Audits are usually done by the associations themselves or by external agents. Four CoCs (European CoC, UK Crowdfunding CoC, UK Lending CoC, Estonian CoC) foresee audits and require the platform members to agree to be audited. The UK Lending CoC in Art 7 stipulates that a formal inquiry can result in an audit through the association – the audit is the consequence of an earlier notification of misconduct, submitted by a competitor of the platform. The Estonian CoC in Art. 11 of the Best Practices Guidelines stipulates that the label is reviewed once per year. The same is the case for the Transparency Certificate of the German Crowdfunding Association.

In sum, monitoring mechanisms are very diverse across crowdfunding markets. Except for the Austrian CoC, crowdfunding associations opted for a centralised system or mixed systems (in the United Kingdom). This is in line with a broad membership discussed earlier. Audits, as the most intrusive monitoring mechanism, are foreseen in a few markets. It is worth noting that European CoC, German Transparency Certificate, UK Lending CoC and Estonian CoC are also self-regulatory regimes with exclusive membership as discussed in subsection 2.2. This correlation suggests that, in a few cases, self-regulatory regimes are meant to serve as club signals. Other cases of weak monitoring support the “reputation commons” theory.

Rules on enforcement are another critical component of CoCs. As Buchanan (1965, p. 1965) points out, “the theory of clubs is […] a theory of optimal exclusion”. The exclusion devices must be in place to ensure that no one can benefit from club goods created through self-regulation without contributing as a club member. Clubs without means to withdraw or suspend membership are not efficient in enforcing the rules. Similarly, the “coerced” theory lies on the premise that the association can impose sanctions on non-complying industry members.

The paper finds that four (UK Crowdfunding CoC, UK Lending CoC, French CoC and Estonian CoC) out of nine CoCs contain membership withdrawal or suspension clauses. The procedures for a suspension or withdrawal of membership are quite diverse. The UK Crowdfunding CoC in Article 16 delegates the authority to the directors of the association. The UK Lending CoC in Art. 7 allows for the suspension of the membership only after a formal inquiry and hearing of the member. The Estonian CoC in Art 11 2nd paragraph foresees the expulsion of members if they do not comply with the best practice guidelines and that no compensation can be retrieved from the association. It is important to note that enforcement of the rules is stronger if it entails a label that can be withdrawn, making the expulsion from an association visible to investors as well.

Another related mechanism of enforcement is a dispute resolution mechanism. A dispute can arise either between members and the association or between members. Four (European CoC, UK Lending CoC, French CoC, Estonian CoC) out of nine CoCs foresee a dispute resolution mechanism. The European CoC in Art 2.5 describes the mechanism for a dispute resolution between members, motivating that “parties will always seek a negotiated solution”. The members of the European association can also appoint an external mediator. The UK Lending CoC Article 7 of the Rules outlines a similar complaint procedure. Complaints by members can be resolved without a formal inquiry, but if a significant breach is detected, then a formal inquiry will be launched, which can result in expulsion from the association. The French CoC in Art 11 is unique insofar it stipulates that the French Association can become a mediator for disputes between platforms and between platforms and investors or between platforms and projects. Art 10 of the Best Practice Guidelines of the Estonian CoC stipulates that the association will publish dispute cases.

Overall, the strong enforcement mechanisms are foreseen in less than half of CoCs discussed. In relation to findings of the prevalence of inclusive membership and weak monitoring from previous subsections, the theory of “reputation commons” has the most explanatory power. However, once again case examples such as Estonian CoC and UK Lending CoC which, in addition to restrictive membership clauses and strong monitoring, have strong enforcement mechanisms in place bring evidence for the “club goods” theory.

4.4 Content of codes of conduct

The following overview (Tables VII and VIII) of substantial rules contained in CoCs in Europe, and the coherent systematisation of provisions, may facilitate the future in-depth study and cross-country comparisons. Due to a large number of provisions appearing under different wordings and sporadically in several CoCs, this section will not indicate where provisions appear and how they differ. Such an extensive analysis merits a deeper reflection on the underlying regulation in each country, which is beyond the framework of this paper.

Beyond the provisions pertaining to the relation between platform and associations, the paper identified three broad categories of substantial provisions in CoCs:

1. Provisions related to the transparency of platforms' behaviour;

2. Provisions related to the business conduct of platforms;

3. Provisions related to the protection of investors (beyond transparency).

Even though transparency, business conduct and investor protection relate to each other in manifold ways, the list of provisions in Tables VII and VIII allow a first glimpse of the topics covered in CoC.

The main concern of both regulators and industry participants are potential market failures due to information asymmetries. Platforms compete not only in terms of fees but also in terms of the quality of their service.

As discussed in subsection 2.2, important aspects of platforms' operation are unobservable by their customers. Screening and selection of projects are often not made transparent. Contracts between platforms and project owners are not made public. The outcome of platforms' due diligence of financial documents and legal structure of the projects is not always published. Management of conflicts of interest is not easily discerned for individual investors. Services in the post-investment stage and overall performance of projects' cannot easily be identified by investors.

All CoCs contain a mixture of rules which increase transparency and rules that establish standards of service quality. The systematisation of provisions contained in CoCs sheds light on services which platforms choose to compete on. For instance, greater transparency is often required with respect to fees charged, project selection criteria and platforms' due diligence. For these services, in the platforms' view, information remedy is sufficient. On the contrary, for some services platforms choose to regulate quality directly. For instance, rules of business conduct are of this type. Some of the provisions of CoCs, such as anti-money laundering or data security, are already required by the law. The rationale for including them into a CoC is to provide additional monitoring and enforcing mechanism.

While the primary purpose of CoCs is to reduce information asymmetries between platforms and their clients, many of the provisions of CoCs indirectly reduce the information advantage of project owners over investors. This is done through mandating platforms to have a mechanism in place to induce greater disclosure by project owners. Trust in the platforms then translates into trust in the projects (Moysidou and Hausberg, 2019). In this way, platforms indirectly “communicate the legitimacy of crowdfunding ventures” (Lehner and Harrer, 2019, p. 85).

The content analysis of the provisions of CoCs does not allow to decide whether “coerced” theories are more plausible than “voluntary” theories. The provisions which restate the obligation to comply with state laws and regulations already in place support the explanatory power of “voluntary” self-regulation theories, since “coerced CoCs” would focus on newly developed provisions in order to defer or stifle regulation.

In the final section, the paper will discuss how to extend the research of the CoCs based on an analysis of the provisions.