CROWDFUNDING AS DEMOCRATIC FINANCE? UNDERSTANDING HOW AND WHY UK INVESTORS TRUST THESE MARKETS ¿“CROWDFUNDING” COMO FINANZAS DEMOCRÁTICAS? COMPRENDER CÓMO Y POR QUÉ LOS INVERSORES DEL REINO UNIDO CONFÍAN EN ESTOS MERCADOS

Can crowdfunding contribute to the rebalancing of the financial system via democratising investment? This paper begins to respond to this question by establishing how and why investors place trust in these markets. We offer two contributions. First, to theoretical debates on democratic finance; and second, to a more empirical body of crossdisciplinary research into popular investment via a qualitative analysis of 52 original interviews with investors in six UK crowdfunding markets. Our data is taken from a project with the UK’s Financial Conduct Authority to enhance investor protection in these markets. Deploying concepts from across economic sociology, we find that investors: mobilise embedded networks to establish trust in crowdfunding; are motivated by expectations of ‘blended returns’; prefer automated investment tools if they lack experience; and typically invest with funds they have earmarked as being prepared to lose. We conclude that enhanced investor protection is required for crowdfunding to help democratise finance.


Introduction
Over four decades of neoliberalism, separate and complex processes of financialization have produced a broad socio-economic and cultural shift from postwar Keynesianism to free-market fundamentalism in advanced capitalist systems (Davis and Walsh 2017). This has facilitated and accelerated the accumulation and concentration of financial power and vast profits in the hands of a small, super-wealthy group of neoliberal power elite (W. Davies 2016). As Sylvia Walby (2015:35) argues in relation to the events of 2008 and since: "Finance caused the crisis. More precisely, the failure of the state to regulate finance caused the crisis. Finance is intrinsically unstable; but this can be mitigated. The reduction in democratic control over finance led to the financial crisis".
With the urgent need to tackle the global Climate Emergency (Klein 2019;Pettifor 2019;Urry 2011), to overcome the democratic deficit (Crouch 2004), and to correct ever-widening inequalities (Blakeley 2019;Dorling 2018), the case for radical democratic reform of the global financial system becomes increasingly pressing. What shape should such radical reform take?
As the leading advocate for 'democratic finance ', Fred Block (2014:4) argues "there is an urgent need for ideas about how finance could be reorganized to disempower the existing financial elite", whom he sees as a direct threat to the viability of democracy itself. Block (2014:7) suggests the global financial system is deliberately constructed to ensure the majority of private savings and investments that pass through mainstream financial institutions (i.e. high-street banks, pension funds, etc.) are directed into a very narrow range of channels. These channels are controlled by a financial elite who extract significant transaction fees. Attempts to democratise finance need "to shrink the major financial institutions" and "to create new financial channels so that private savings could be directed to overcome the shortage of financing". These new channels must move money in a way that prioritises people and planet, targeting those areas of 'systematic underinvestment' by mainstream lenders, such as clean and renewable energy, retrofitting building stock, and large public infrastructure projects (Block 2014:10-11). If successful in creating the required regulatory space, gathering political support, and persuading the public to move their money out of traditional institutions, Block (2014:11) states, democratising finance could: "enhance the power of local communities, put greater emphasis on equality and social inclusion, and prioritize significant movement toward environmental sustainability. In short, democratizing finance fits the framework of a real utopia because it could simultaneously weaken the power of entrenched elites while moving society toward an economy that is subordinated to democratic political initiatives".
At a time when finance continues to be seen as an external and malevolent force that simply happens to people, rather than a system that people help to create and to sustain through their habitual uses of money, the pursuit of a more democratic finance shines a new light on those disruptive innovations already motivating people to use alternative channels of investment. Whereas Block advocates for a much larger sector of non-profit retail financial intermediaries as the base (i.e. mutual banks, cooperatives and credit unions), this paper contributes to these debates by asking: can crowdfunding contribute to democratising finance?
Noting that wealth disparity is in part caused by a lack of inclusive access to investment opportunities, Palladino (2019) has also looked to 'Fintech' innovations (e.g. crowdfunding) as attempting to create markets that enable ordinary investors to move their money in support of local social and medium enterprises (SMEs). As she states, "[t]hese new technologies open up the potential for a radically different approach to financial participation by lowering the costs of transactions and transmission of information" (Palladino 2019:575).
How realistic is it that crowdfunding can contribute to a democratic rebalancing of the financial system? And, if crowdfunding can create new financial channels to (re)direct private savings to social and environmental areas short of financing, how will people trust them at a time of acute economic uncertainty? This paper seeks answers by speaking with those who have already built and negotiated trust in crowdfunding. We do so via a qualitative analysis of 52 interviews conducted with 'sophisticated investors' in six UK crowdfunding markets using data taken from our 2017 project with the Financial Conduct Authority (FCA) 1 .
The paper proceeds as follows. Section 2 introduces the UK crowdfunding sector in order to establish context for an international readership, before briefly outlining four concepts drawn from across economic sociology that we used to frame our analysis of the data. Section 3 provides a methodological description of our study in consultation with the FCA. We present our qualitative analysis in Section 4, where we find that investors: mobilise embedded networks to establish trust in crowdfunding; are motivated by expectations of 'blended returns'; prefer automated investment tools if they lack experience; and typically invest with funds they have earmarked as being prepared to lose. We conclude by highlighting the potential of crowdfunding for democratizing finance, but caution that enhanced investor protection is required if the public are to build trust in the sector. https://doi.org/10.3989/ris.2020.78.4.m20.005

What is Crowdfunding?
Crowdfunding is 'a way of financing projects, businesses and loans through small contributions from a large number of sources, rather than large amounts from a few' (Baeck et al. 2012:3). In practice, individuals deposit money via an online crowdfunding platform, committing that money to a specific project, business or loan, and have that relationship mediated by the platform. Its popular association with donation-based contributions to charity appeals, creative arts projects, or local independent business ventures has created an image of crowdfunding as dominated by a 'hipster millennial crowd' of aspiring social entrepreneurs. It is an image that is hard to shake off (Reiser and Dean 2017). Operating akin to an economy of gift exchange (Mauss [1954] 2001), typically the promoter is a friend, a relative, or socially connected in some way, either physically or virtually (e.g. through social media) (Borst et al. 2018). For many, crowdfunding remains simply another form of charitable giving.
Contrary to this image, crowdfunding is also a serious form of investment that helps to bypass traditional bank lending criteria and enables more direct funding from highly-dispersed 'lenders/investors' to highlydifferentiated categories of 'borrowers/projects' via an online platform or smartphone app. These new financial channels involve the use of various debt and equity business models and often raise hundreds of thousands (sometimes millions) in investment (Angerer et al. 2017;Belleflamme et al. 2014;Mollick 2014;Lehner 2013). The types of investment facilitated through crowdfunding vary, from equity (shares) in a business through to peer-to-peer (P2P) loans for consumer goods (Langley 2016;Ahlers et al. 2015;Cholakova and Clarysse 2015). Sums are often allocated more transparently to social and environmental projects systematically underfunded by mainstream lenders. This reflects a growing motivation amongst investors to pursue a 'blended return' of financial and non-financial outcomes.

Typology of UK Crowdfunding Platforms
To give a sense of scale, the total value of the overall alternative finance market in the UK grew 35% to £6.2bn during 2017, up from £4.6bn in 2016 and from £3.2bn in 2015 (CCAF 2018). This growth and maturation of the sector is driven by sophisticated P2P crowdfunding models, which facilitate loans either to retail borrowers (i.e. peerto-consumer, or P2C) or to businesses (i.e. P2B). As such, a highly diverse ecosystem of crowdfunding platforms now operates in the UK (Davis and Braunholtz-Speight 2016;Langley 2016) and worldwide (CCAF 2020). Together, they provide capital to virtually every sector of the economy and life stage of a company or project, from clean and renewable energy through to community and social enterprises. UK platforms tend to develop a focus on a specific type of finance (e.g. donation, debt or equity), but then diversify via their focus on a specific sector of the economy, such as charity, real estate, or infrastructure funding. Broadly speaking, UK crowdfunding platforms can be categorised as follows:

Donation/Rewards-Based Crowdfunding Platforms
These platforms facilitate the financing of individuals, charities or other smaller non-profit organisations. Investors see themselves as 'donors' and participate principally because they believe in a specific cause. They do not receive a financial return on their money, but may receive non-financial rewards. Donation-based platforms facilitate investment in everything from creative arts projects through to civic crowdfunding ventures tied to public infrastructure. Leading UK donation platforms are: Crowdfunder 2 , who specialise in enabling individuals to back socially-useful projects and activities; and SpaceHive 3 , who enable investment in 'place-based' opportunities, such as improving derelict land or bringing a community asset back into use.

Debt Security or Loans-Based Crowdfunding Platforms
These platforms facilitate the provision of debt finance to organisations and companies, bypassing the need for traditional banks. Depending on the specific model of the platform, investors lend money via a loan or a debt security (i.e. bond / debenture). Investors see themselves as 'lenders', receiving interest on money lent. If smooth, their capital is returned as either a single payment or as an annuity over the life of the investment. Platforms that deal in loans or debt securities are currently regulated under two related but separate regimes. Debt securities sit within the EU-derived Markets in Financial Instruments Directive (MiFID) 4 regime, whereas loans are governed by UK specific legislation introduced in 2014 and updated in 2019 5 . It is generally understood that debt security platforms face higher regulatory standards. The debt category of crowdfunding is the most populated and diverse, which reflects the wide variety of use cases for debt financing within the economy. Leading UK platforms are: Abundance Investment 6 , a debt security platform focused on providing shortand long-term debt to infrastructure companies and public sector organisations; and Funding Circle 7 , which is a loan-based platform focussed on providing working capital and growth capital to the UK's SME sector.

Equity-Based Crowdfunding Platforms
These platforms support equity-based capital raising by new or established businesses. Investors selfidentify as such and allocate capital to a given opportunity in exchange for transferable shares. Currently, the sector is focussed primarily upon the 'early-stage' or 'start-up' phase of company growth, so investors typically hope their shares will increase in value. Crowdcube 8 are a leading UK equity crowdfunding platform for entrepreneurs of start-ups and growing businesses to connect with potential investors. Seedrs 9 was the UK's first regulated equity-based crowdfunding platform, enabling investors to buy shares in early-stage high-growth businesses.

Co-operatives and Societies for the Benefit of Communities
A smaller UK market exists for both cooperative (Co-ops) and community benefit society (Ben Comm) business models. Although there are differences between the two, in practice they are extremely similar. Technically, a co-operative is run for the benefit of its members, whereas a benefit society is run for the benefit of the community. Both use withdrawable shares, known as 'community shares'. This model is distinct from traditional equity investing as the share offers are currently exempt from FCA rules. The model is underpinned by the idea of equality in terms of governance, with one shareholder getting one vote regardless of investment level, rather than a voteper-share as with traditional equity models. Ethex 10 are the UK's leading platform in this sector, having pioneered the concept of 'positive investing'.
As this outline suggests, there is no single 'type' of crowdfunding investor. Rather, the diversity of platforms in the UK sector reflects the broad blend of motivations that individuals have for their moneyranging from the philanthropic to the self-interested, from the constructive to the speculative -and which investments they are prepared to trust.

Towards an Economic Sociology of Crowdfunding
One explanation for the growth in crowdfunding activity is precisely the wider 'crisis of trust' in mainstream finance pointed to earlier, and which remains largely unchanged over a decade on from the events of 2008 (Tooze 2018;Mirowski 2013). The perceived lack of accountability and democratic control over finance has seen crowdfunding platforms increasingly position themselves as 'alternative, disruptive, or democratizing', as compared to more traditional finance, a positioning that has been repeatedly questioned (Tooker and Clarke 2018:60;Nelms et al. 2018:12;Langley and Leyshon 2017;Maurer 2008;Aitken 2006). And yet, many people around the world evidently do trust crowdfunding and are motivated to invest, so the questions we asked were: how and why?
In seeking answers, we first note that Bandelj (2015:237) has stated that "little economic sociology explicitly examines the dynamics of how trust in economic relations is built and negotiated […] Therefore, crucial work remains to be done on how trust is achieved, mistrust is overcome, or what the consequences are of trust violations, or betrayal, in economic transactions". Our contribution to this 'crucial work' is to analyse empirically 'how trust is achieved' and 'mistrust is overcome' in a single bounded sector, namely UK crowdfunding markets.
Existing literature notes that crowdfunding investors are motivated by: a commitment to communities of interest; the exciting challenge of an innovative venture; the opportunity to advance social status amongst peers; and, most obviously, a desire to make money (Lehner et al. 2015;Belleflamme et al. 2014;Seltzer and Mahmoudi 2012;Brabham 2010). Clauss et al. (2018) agree that individuals will support a given crowdfunding project according to their social status and preferences, but add that previous investment experience is also a significant factor in shaping behaviour. In non-equity crowdfunding, Cecere et al. (2017) note the importance of altruism and the 'warm glow' effect of supporting social and cultural projects, whilst in equity-based crowdfunding, 'being excited about a specific company or project' has been ranked by investors as more important than high financial returns (OXERA 2015:4). This blend of motivations is also strongly attested to by a growing literature on crowdfunding for social or environmental causes (Lehner and Nicholls 2014;Horisch 2015;Holstenkamp and Kahla 2016).
Given this, we decided against adopting a single theoretical framework for approaching our study. Instead, we read across economic sociology and applied three key concepts when interpreting our data: embeddedness, expectations, and earmarking. Following Clauss et al. (2018), we added experience as a fourth concept that proved to be significant in our analysis.
As the economy is not separate from society and its institutions, Granovetter (1985) argues that the specific social relations in which particular economic actions are embedded are crucial for producing and sustaining trust. It is this embeddedness that enables economic transactions to take place (Krippner et al. 2004;Polanyi 1957). We use this concept to frame our analysis of how and why crowdfunding investors trust the platforms they invest through, and the businesses or individuals they invest in.
The concept of expectations points to the 'temporal order' of capitalism and the specific time horizons that shape economic activities (Beckert 2015;Bourdieu 1979). For example, attitudes to risk and reward are framed by shorter-and longer-term expectations that encourage more or less 'speculative' attitudes towards finance and investment (Adkins 2018;Konings 2018; https://doi.org/10.3989/ris.2020.78.4.m20.005 Staheli 2013). We apply this concept to our analysis of how crowdfunding investors are motivated by the promise of mixed returns beyond a headline rate.
As is widely known in economic sociology, Zelizer's (1994) concept of earmarking reveals how money is interpreted socially and culturally, such that our engagement with money and markets is 'relational' rather than rational. For example, the extent to which we interpret any money we may receive as a "gift, payment, or entitlement" is crucial in terms of how we then choose to allocate it. This was a vital insight when approaching our data as it allowed us to understand the decisions investors take about which 'pot' of money, and which percentages of their overall portfolio, they are prepared to earmark for crowdfunding.
As will become clear, we also deploy the theoretically underdeveloped concept of experience to capture other areas of our data. In so doing, we hope to encourage others to undertake the 'crucial work' on trust called for by Bandelj. Each of these four concepts is used to frame the analysis of our data in Section 4. Before then, we offer a description of our study and our justification of method.

Method
As the UK regulator, in the summer of 2016 the Financial Conduct Authority (FCA) circulated an online questionnaire that was completed by over 22 000 self-certifying 'sophisticated investors' in UK crowdfunding markets. Response data from this survey was compiled by the FCA into a single spreadsheet for initial analysis by researchers at the Cambridge Centre for Alternative Finance (CCAF) 11 . In early 2017, we were subsequently granted unique access in order to conduct an independent review of this data and to highlight any limitations. We concluded that the online survey had missed certain areas: what had motivated investors to move their money into crowdfunding; how trust in crowdfunding markets and platforms was negotiated and achieved; and how they had experienced crowdfunding investments.
We proposed that a small-N qualitative approach would allow us to correct these gaps by investigating in depth how and why investors had decided to trust these markets. Framed in terms of democratizing finance, we also wanted to understand how their experiences might shed new light on how members of the wider public might be sufficiently protected. In so doing, might more people be encouraged to move their money via crowdfunding platforms to begin to correct systematic underinvestment in vital green and social projects?
In order to access data on trust and motivations, a qualitative study was necessary (Byrne 2012;Mason 2002). The authors decided to adopt a semi-structured approach, developing an interview schedule of 35 open-ended questions that allowed us to explore 'investor motivations', 'investor behaviour' and 'trust and confidence'. This schedule was developed and then shared with the FCA on the understanding that our qualitative analysis of this interview data would correct knowledge gaps in their online survey. Ultimately, the aim was to help to form a part of the evidence-base for their post-implementation review of UK crowdfunding regulations (FCA 2019).
Following this consultation with the FCA, we decided to focus our interviews on four investor practices: carrying out due diligence on crowdfunding platforms, entrepreneurs and investment products; understanding the 'wind down' process in the event of any future platform failure; knowledge and use of platform 'contingency funds'; and understanding and use of 'automated investment' tools provided by platforms. We also added three new closing questions, inviting interviewees to provide a score on a simple Likert scale (from 0 to 5) to signal how far they trusted five different types of financial organisation to act in the investor's interest. These organisations were: high-street banks; mutuals and building societies; small and medium sized enterprises (SMEs); large multinational corporations; and crowdfunding platforms.
In the next stage of the research process, we emailed over 500 respondents to the FCA's online survey to invite them to participate in our follow-up study. Response rates were low (c.10%). The sample of investors from which our data is drawn is therefore 'self-selecting' and we were unable to control for diversity (Emmel 2013), beyond trying to maximise a spread across the six crowdfunding markets we describe below.
As we anticipated, the majority of our sample is male, white, over-55 years of age, and retired. Our sample does include minor diversity with respect to age (one, 18-24; five, 35-44; nine, 45-55; and thirtyseven, over-55), and four female investors, but isolating age and gender variables revealed little during our analysis. For example, prior experience of investing was a far more significant factor in understanding behaviour, as previously indicated, and this cut across both age and gender. We do, however, endorse Walby's (2015) proposal that further research is needed into the 'gendered regime' in finance.
Between 23 January and 10 February 2017, the authors conducted 52 interviews with investors across six UK crowdfunding markets: 20 with P2B investors; 14 with P2C lenders; 6 with equity-based crowdfunders; 5 with securities investors; 5 with real estate lenders; and 2 with real estate equity investors. These six market categories were established in the FCA's online survey, so we retained these categories to improve consistency. As is clear, our sample was skewed towards P2B and P2C markets. We granted full anonymity to each interviewee as a condition of their participation. As such, respondents are identified throughout this article only by broad reference to these six market categories.
In testing these assumptions, analysis of our data reveals that prior experience of investing is a significant factor in deciding which crowdfunding markets to trust. Although most investors we studied recommended that more consistent and standardised information ought to be provided by all platforms, this view was concentrated amongst P2B investors who had experienced previous defaults by borrowers. Reflecting on how they had achieved trust in crowdfunding, investors openly named platforms that gave limited information about what exactly they were investing in, desiring far more information about a given property (Real Estate), a small-medium enterprise (P2B), or a consumer loan (P2C) before allocating funds. More experienced investors were wary of trusting platforms that provided limited information: "… one of the main things that regulation ought to do is to improve consistency and comparability across loans and across platforms […] I mean you have to understand what assumptions the platforms are making and I don't think most people have got the time to do that […] So for the sector to avoid getting a bad name there really ought to be some standard approaches to sharing information on default rates, on interest rates, on loan-to-value, on a whole range of … standard metrics" (Real Estate investor).
More inexperienced P2C lenders, however, reported feeling intimidated by large amounts of complex financial information. Instead, they preferred to trust those crowdfunding platforms that provided more summative information about a prospective borrower. Contrary to the rational calculation of that "anthropological monster" (Bourdieu 2005:209), homo economicus, these P2C lenders were opting to trust platforms that were providing less information, which they found simple and easy to grasp. Interestingly, we also heard that crowdfunding investments made in haste, and thus based on limited due diligence and 'knowing less' in a market context (Dorn 2012), appeared to deliver similar returns to those capital allocations made with slower, more careful decision-making.
Instead of relying solely upon information provided by the platforms, many investors leveraged networks in which they were already embedded (social capital) and drew upon wider norms and values (cultural capital) in their assessment of crowdfunding (Bourdieu 1986). Examples included attending company or sector-level events, such as platform or product launch nights, as well as academic and civil society conferences. More experienced investors also spoke of proactively arranging face-to-face meetings with those leading crowdfunding platforms, either individually or as part of a wider group. As one P2B investor explained: "… and they were like real people. When I spoke to [crowdfunding entrepreneur], he sort of told me what they were trying to do […] So, we were at a meeting in London and we sort of told him what we thought Each interview lasted around 60 minutes, was mainly conducted via Skype (occasionally via a landline), with data captured by the CallNote software package. Our audio files were sent to the TRINT transcription service and subsequently shared amongst each author for independent coding (Fielding 2008). A social science colleague at the University of Leeds with no involvement in our study was employed as an independent fourth coder to enhance the reliability of the data by limiting author bias. For analytical integrity, the authors conducted this coding process independent of the FCA and our analysis was completed before sharing our indicative findings. We submitted a formal report to the FCA in May 2017 that informed subsequent rule changes to place new marketing restrictions on P2P platforms, designed to protect new or less-experienced investors, and providing exemption for direct offers to the kinds of 'sophisticated investors' involved in our interviews (FCA 2019).
In close, we concede that our analysis of the data which follows is based on a relatively small and selfselecting sample of UK investors. Nevertheless, we suggest that the following analysis -framed by those four concepts outlined in the previous section -contributes to the 'crucial work' necessary for economic sociology to understand better the dynamics of how trust in economic relations is built and negotiated.

Embeddedness: Building and Negotiating Trust
Crowdfunding platforms expect investors to conduct their own due diligence under the broadly accepted principle of 'buyer beware'. Existing research suggests that this process involves weighing up the balance of risk and return in a process of 'rational' calculation. Lehner et al. (2015:172) suggest that this isn't easy, however, because crowdfunding actually provides "little opportunity for due diligence", pointing to its popular image of relying upon friends and family for funding (Borst et al. 2018;Agrawal et al. 2015).
Likewise, it has been suggested that "platforms see little incentive or advantage in providing structured, transparent access to their project data" (R.Davies 2015:349) and want to retain 'information asymmetries' between the platform and its investors to limit opportunities for product and marketing imitation by competitors (Clauss et al. 2018;Hall and Lerner 2010). In contrast, Langley (2016:309) observes that UK crowdfunding platforms are actually transparent about their historic and projected rates of default. Ahlers et al. (2015) suggest that investors in crowdfunding markets simply do not have the time, resources or willingness to analyse each entrepreneurial venture and its business model in close detail. as a group. Different people told them exactly what we thought -him and say four or five others. And they sat in a room at the back and listened to what we were saying. And he said that's how the platform was built up as well, by listening to what people said. And so that was quite good […] Well, it gave me confidence because I knew that they'd already worked for [Platform A], and also I liked the software idea that they had. And it just seemed they knew what they were talking about" (P2B Investor).
We thus agree with Clauss et al. (2018) that the most reliable predictor of financial success in crowdfunding is the 'human capital' and 'embeddedness' (Granovetter 1985) of the entrepreneurs leading the platforms, once again underlining the 'relational' aspect of economic behaviour (Zelizer 1994). For many investors, achieving trust required the development of 'offline' social relations and exposure to those forms of 'entrepreneurial storytelling' outlined in the quote above (Manning and Bejarano 2017;Bernadino and Freitas Santos 2016). As with traditional financial organisations, 'putting a face' to an investment remains important, especially for experienced investors: "I just want to check that the people are real and the business actually exists" (Securities Investor). Being able to verify online information in this way also extended to the underlying asset attached to an investment opportunity: "So, some of these companies say, 'we're going to use [your funds] to build this', for example, they're going to build a wind turbine. But what they really mean is we're going to pay off the loans that we borrowed to pay to buy the wind turbine or whatever. I think you need to know specifically where the money is going and then you want to know how you will extract your money out of it" (Securities Investor).
Perhaps unsurprisingly, establishing trust in the existence and valuation of an underlying asset was of greatest concern to Real Estate crowdfunding investors, who reported checking that an advertised property genuinely existed. Whilst loan-to-value ratios were used as a means of assessing risk, they told us that such ratios could be misleading if the property value used to calculate them was unreliable. We were told that one of the principal barriers to trusting a Real Estate crowdfunding project was a lack of clarity about the valuation methodology, especially the 'hidden' relational networks that connected the property valuers, the property developers and the crowdfunding platform:

"… [and you worry about] this nefarious valuation […]
Many of the valuations appear to be done by a surveyor with some sort of related party interest. None of that's declared on the website" (Real Estate investor).
In leveraging wider networks, investors also reported trawling online forums for personal accounts of positive and negative experiences of crowdfunding platforms and investments. This included forums provided (i.e. managed and monitored) by platforms themselves, as well as popular independent forums such as the UK's FrankP2P website 12 . As Mollick (2014) has shown, this culture of online forums provides the opportunity for investors to share their emotions -i.e. perceptions, feelings, and quality assessments -with respect to a given platform, as well as to test claims made publicly by a platform against investor experiences. This can lead crowdfunding investors to exhibit those same 'herding' trends common to other financial markets (Kuppuswamy and Bayus 2013), but both lesser and more experienced investors in our sample viewed these forums as providing a vital means of social support when trying to calculate who to trust for the best mix of risk and return.

Expectations: Pursuing Blended Returns
In all six markets, interviewees told us they were motivated to invest in crowdfunding by a desire to diversify their overall portfolio, to minimise losses elsewhere, and to benefit from slightly higher rates of return relative to high-street banks. Investors reported avoiding those platforms that appeared to offer very high rates, however, expecting that this was an indication that only borrowers in distress and/or with sub-prime projects were operating via that platform.
Their expectations of what appeared to be the correct 'strike' point of a good rate varied, but was significant in adjusting their assessment of risk and, therefore, how far to trust an investment: "I think […] one of the reasons why I'm less attracted to the high return loan products on offer is, y'know, if you look and they give you 7-8, or 10-12, percent […] I know there must be riskier borrowers involved there" (P2C Lender).
P2C Lenders in our sample managed risk by spreading investments across multiple mini-investments (often providing as little as £10 towards a target), in order to provide further protection against non-repayment by individual borrowers. This ability to spread very small investments was one of the most common reasons given for 'feeling safe' in crowdfunding markets. In terms of our wider argument, the following quote also reveals the importance some investors attach to the idea of crowdfunding as 'alternative, disruptive, or democratizing' (Langley and Leyshon 2017) when compared to more traditional finance. "I like [platforms] because they split the money into lots of tiny little bits, so you get the security of spreading your money across things. I like the idea that there's an alternative for people to go to who want to borrow money, rather than from banks or loan sharks. And I think that they are quite sensible alternatives now. They're well enough established to be a good alternative. If you want to borrow some money to go and buy a car or something, [Platform X] and [Platform Y] are sensible places to go look. They come up with good prices […] and they also give quite good returns compared to banks, so I'm happy to carry on [lending]" (P2C Lender).
Of course, the diversification of very small amounts of capital was not always seen as a 'rational' approach. As one more experienced investor caustically remarked: " there's no point having a diversified portfolio if all you have is diversified shit!" (Real Estate Investor).
In reflecting on their decision to move money into crowdfunding, investors did report a mix of 'extrinsic' and 'intrinsic' motivations (Allison et al. 2015). The desire to contribute a more 'social' return was expressed in two main ways. First, several investors wanted "after the financial crisis, just to give the banks a kicking!" (P2C Lender). Second, investors also sought to push back against narrow, financialonly returns, noting their frustration with crowdfunding platforms providing the same crude assessment levels as traditional institutions (i.e. 'High, Medium, Low' risk). P2B investors and P2C lenders especially wanted crowdfunding to facilitate more refined choices that would allow for allocating funding to projects or businesses that would realise social and ethical outcomes, as well as financial returns. This was relayed to us as investments in a social enterprise, a local community project, in renewable energy infrastructure -those areas of 'systematic underinvestment' outlined by Block (2014).
"The limited choice that I have is whether to pick which associated risk group my money is being lent out to […] Obviously, I know certainly with the money I've got invested that it's going out in personal loans, and obviously there's an ethical association with that. I don't want to be involved with people getting into debt. But yes, I think that they [more 'social'/'ethical' investment options] would be quite a thing!" (P2C Lender).
This mix of motivations helps to explain why people are moving their money into crowdfunding, which our data shows is based upon far more than seeking optimal financial returns. The expectation that crowdfunding platforms enable investors to have a more direct say in precisely where and how much of their capital is allocated to a given borrower or social / environmental project was a major factor, with obvious implications for democratizing finance (Block 2014; Palladino 2019).

Experience: Using Automated Investment Tools
We were surprised, therefore, to learn that many of our interviewees were content to surrender this new found financial agency in favour of using 'automated investment' tools provided by the platforms. Common in high-frequency trading (MacKenzie 2018;Preda 2017;Coombs 2016), and with problematic consequences (Borch 2016), these tools rely upon complex algorithms to allocate capital automatically to investment opportunities as per a set rate of interest and default-risk.
Reliance upon automated systems in crowdfunding is increasing. CCAF (2016) collected and analysed quantitative survey data from European crowdfunding markets to show that 82% of P2C lending and 38% of P2B investments were arranged by automated processes in 2016. This growing preference for automation appears to undermine the very principle of 'peer-to-peer' (P2P) lending, as it (re)introduces an intermediary responsible for allocating capital within certain agreed parameters -in this case, an algorithm rather than a portfolio manager. If responsibility for investment decisions is so willingly surrendered, then the use of automated tools does seem to undermine the positioning of crowdfunding as 'alternative, disruptive, or democratic' (Langley and Leyshon 2017).
Drawing on our data, one explanation for this is that relative levels of experience amongst investors operating in crowdfunding markets are significant. Individual selection was more common in P2B and Real Estate markets, where prior experience of investment was higher, with auto-bidding tools more common for P2C lenders, where experience was limited. Automation was also preferential for those investors without the available time to engage in protracted and detailed analysis of their investment options, especially given the small amounts of capital they trusted in these markets relative to their overall portfolio (Ahlers et al. 2015).
The preference for selecting investments amongst P2B investors was typically because they felt their expertise could in effect "beat" any automated algorithm and thus yield higher financial returns. Those P2B investors in our sample, however, also reported enjoying the feeling of supporting small businesses and creating jobs, thus realising more social / nonfinancial outcomes that they felt were inadequately catered for by markets elsewhere. This was true also for Equities and Securities Investors, who avoided automated tools because they preferred the timeconsuming process of browsing websites and selecting opportunities, which was precisely why they were participating in crowdfunding. As these markets were an additional field of activity, separate from their main sources of income, they reported enjoying the process of 'gamble and jeopardy' in seeing whether the business they had selected succeeded or failed. "I take opportunities where I can analyze it myself and I don't necessarily believe in the [automated] investments […] But for myself, I rather prefer to set-off and be the master of my destiny, and analyze stuff myself, and look at stuff myself" (Equities Investor).
The time it takes to realise returns from Equities and Securities investments was also a factor in rejecting automated tools. These investors were thinking for the longer-term in the crowdfunding sector, hoping that their chosen investment "becomes the new Facebook" (Equities Investor). Significantly for the goal of democratizing finance, they believed that makhttps://doi.org/10.3989/ris.2020.78.4.m20.005 ing careful selections in order to provide 'patient capital' for businesses and projects over the longer-term (Mazzucato 2018;Klinger-Vidra 2016) was of greater importance than simply investing 'idle cash' for a quick financial return. Likewise, Real Estate investors in our sample wanted to dedicate time to selecting opportunities in property, depending upon the number and size of investments that each was making: "I enjoy it, so I probably spend somewhere between an hour and two hours each day just keeping an eye on what's going on and picking up new loans, moving money around" (Real Estate Investor).
As the following quote indicates, a process of 'earmarking' money -either by platform or by investment opportunity -is a significant factor in understanding investor behaviour in crowdfunding markets.
"[Platform B] is for some money that I don't need to think about. I don't necessarily feel the need to manage everything actively. Some of the money that is there is in [Platform B's] rolling short-term account, and it's there because I think I might need it in a year's time. I suspect that I will probably run down my five-year lending on [Platform B]. Historically, I've had a mixture of five-year money and rolling money, with the rolling money being stuff that I think I might actually need" (P2B Investor).
P2C lenders, however, were split in their use and preference for automated investments, perhaps because these investments were seen as a higheryielding substitute for 'idle cash' that would otherwise be accruing almost zero interest in high-street bank accounts. Some P2C lenders reported being attracted to crowdfunding precisely because of the fully automated processes: "I only look for an [automated] one […] I hand over completely the responsibility to the platform. I mean what interested me was that they spread the risk across multiple people, and that they had a good [contingency] fund" (P2C Lender).
Other P2C lenders disagreed: "No, I am absolutely not looking for a fully [automated] service! And I do, absolutely, prefer to do it myself […] I guess there's a bit of mental accounting where I think I should put some of it in a risky box for the long-term. And then some of it is in a safe spot, there for emergencies. Then at the same time I consider myself 'irrational', still having an ISA with savings, the same thing as cash. It's not rational, but I still have it. So, I'm definitely not rational in this. Yeah, so I do a little bit of everything. I try to stay diversified so I wouldn't invest too much into one pot" (P2C Lender).
Engaging with both the rational and emotional expectations of potential investors is clearly important in order to scale (Berezin 2009;Preda 2009;Galak et al. 2011). Increasing automation may compromise crowdfunding's capacity to democratize finance, however, instead accelerating wider processes that seek to remove human influence over economic decision-making in an increasingly decentralized and financialised economy (Lash and Dragos 2016;Vigna and Casey 2016). If investment parameters are automated around narrow criteria of only financial risk and return, crowdfunding could further entrench -rather than radically rebalance -power relations in the global financial system.

Earmarking: Prepared to Lose
More hopefully, then, it is interesting that interviewees across all six markets reported making crowdfunding investments without any real expectation of high financial returns. Investors ensured a safe level of financial protection by choosing to operate with funds they had earmarked as being 'prepared to lose'. Often these funds were unearned, acquired via a 'windfall event' such as an inheritance, an unexpected yield from traditional investments, or else by recycling the interest realised from existing crowdfunding ventures. In this sense, the money being used was variously interpreted through the prism of "gift, payment, entitlement" established by Zelizer (1994). Consequently, investors 'earmarked' these funds differently and regarded this money as appropriate for more 'social' investments, with crowdfunding seen as the most direct and transparent mechanism for achieving this.
Investors were also happy to 'play around', testing the robust nature of platform processes and promises, accepting that things were more likely to go wrong with 'alternative' investments. They reported that their limited expectations of high financial returns shaped their behaviour in these markets, regarding these investments as an opportunity to 'take a punt' on a particular business, asset, or social / environmental concern. Some investors even preferred to allocate funding to particular UK regions for which they held a sentimental attachment, raising interesting questions about the potential of crowdfunding to facilitate citizen-led and 'place-based' financial structures, such as community municipal investments (Davis and Cartwright 2019;Wright 2016).
As an example, some investors admitted that decisions were made in a matter of seconds, driven by an urgency to invest 'idle cash' they had earmarked for that purpose. In a telling quote, one time-pressed P2B investor revealed a rapid approach to investing based solely upon how a trusted platform assessed the 'quality' of the borrowers (i.e. into risk categories of A, B, and C): "Yeah […] I spend quite a while just on some, and then other times it varies, and there are times I think I have no time to go into this, so, I'll just, 'tick, tick, tick, tick' -that will be B or that's a C. And I found actually that sometimes the As were okay, except you got lower rates. I think that Bs were worse than the Cs. The second one down is the second highest risk and that was quite surprising, because you're getting more [return] off the Cs than the Bs. And the Bs defaulted more than the Cs!" (P2B Investor).
Likewise, experienced Equities and Securities investors spoke of being relatively disinterested in conducting extensive due diligence precisely because they were prepared to lose in these markets: "There's no guarantees with any of it. So, y'know, the little investor may make a good profit; they may make a total loss. But the idea is you don't invest your shirt on it. You just put in pocket money really" (Equities Investor).

And again:
"It's a tiny proportion that I invest in crowdfunding equities, as a proportion of my total portfolio, but I'd like to increase it if my confidence in it grows, as I think it will. But at the moment, it's just how much spare money that I've got" (Securities Investor).
Similarly, most P2B Investors had strict rules about the maximum they would invest in any single loan, especially if they had previously experienced defaults when straying from that principle: "In the beginning, I put more and more in. And then I sort of started to get [the level of investment] back down. I think it's too much exposure, so I've reduced it now. As I put more in, I took more risk. That was a big mistake because I would put £600 [GBP] out to one guy and then he'd put it another loan for something else. And I lent out another £600 [GBP] to him again for something else, because he had a good story. If it hadn't been for him, I wouldn't have lost £1,200 [GBP]! So, I've been [cheated] a couple of times. So, then I felt like I'm not going to lend that amount of money again -it's just going to be, maximum £20 [GBP]" (P2B Investor) When prompted during interviews, it was surprising how few investors in our sample understood the 'wind down' process should a platform collapse. Most were simply unconcerned, assuming that their legal rights must be strong and trusting that the subsequent administration of any platform collapse would ensure their investments were protected. Others were more worried, citing concerns over how much time the process could take and the amount of additional fees an administrator might demand for this service (a concern that was most acute in Real Estate markets).
Most P2B investors did not consider 'wind down' a danger, but recognised that there was a remote material risk that a platform could fail, which was perceived to be separate from the risk of losing their investment (hence diversifying their investments across platforms as well as markets). Both P2B Investors and P2C lenders did not fear losing a significant amount of their investment this way, at least in part because they believed portfolios would be either 'wound down' by a 3rd party platform or acquired by another: "My expectation is that, y'know, an ideal outcome would be another platform steps in, and so buys the book or picks the book up and agrees to man-age it. I think that's optimistic. I think a more realistic proposition is that essentially administrators would come in. Lawyers would come in and their loan book would get wound up over there. Lenders would or should expect to take a 'haircut'. Now, is that haircut 2 percent, 5 percent, 20 percent? I don't know. I have no way of really judging that" (P2B Investor).
We interpret these relatively relaxed attitudes to 'wind down' in line with our findings that investors in crowdfunding markets are typically using funds they have earmarked as being prepared to lose.

Conclusions
Radical reform of the global financial system is pressing. Greater democratic control over finance would begin to disempower a financial elite who appear increasingly unaccountable to the public. One way to democratize finance is to create new financial channels that redirect money so as to overcome systematic underinvestment in those areas of the economy capable of tackling urgent social and climate emergencies. Crowdfunding is one such channel that has emerged to offer greater accountability and transparency over how and where money is invested. But, if democratizing finance requires the wider public to move their money into crowdfunding, how is trust in these platforms and alternative investments to be achieved? This paper has responded to this question, and the need for economic sociology to understand better the dynamics of how trust in economic relations is built and negotiated, by establishing how and why existing investors already trust crowdfunding. Analysing the data from our project with the UK's Financial Conduct Authority (FCA), we found that investors: mobilise embedded networks to establish trust in crowdfunding; are motivated by expectations of 'blended returns'; prefer automated investment tools if they lack experience; and typically invest with funds they have earmarked as being prepared to lose.
To conclude, we offer three closing remarks as a contribution to further research in this area. First, the four concepts we used to frame our analysis are clearly interrelated. Engagement with crowdfunding markets was clearly influenced by pre-existing social ties, as more experienced investors were embedded in networks that enabled them to leverage social and cultural capital to negotiate and build trust in crowdfunding. These investors also used money earmarked specifically for crowdfunding, which in turn shaped their motivations and impacted upon their expectations of a more complex blended return of social, environmental and economic outcomes. Taken together, we suggest that our data adds further empirical weight to Zelizer's (1994) insight that our engagement with money and markets is 'relational' rather than rational. https://doi.org/10.3989/ris.2020.78.4.m20.005 Second, we suggest that it is possible to interpret crowdfunding as providing a form of 'less impatient capital' (Mazzucato 2018;Klinger-Vidra 2016). Whilst different to State-led 'patient capital', the different 'time horizons' (Beckert 2015;Bourdieu 1979) of economic activities in these markets mean that investors do not always expect high returns in a short timeframe. Instead, investors shift their expectations in order to utilise crowdfunding as way of rolling 'idle cash' or earmarking 'pots' of windfall money to social and/or environmental projects over a longer period. The UK sector has already demonstrated crowdfunding's capacity for facilitating long-term financing in areas of 'systematic underinvestment' by mainstream lenders (e.g. clean and renewable energy, SMEs, and retrofitting building stock). Thus, as a form of democratic finance, crowdfunding can be mobilised for public infrastructure projects providing the outcome is transparently positive in social and/or environmental terms.
Third, if crowdfunding is to provide a way of rebalancing the wider financial system, then there needs to be a shift in the power dynamics operating inside these markets. Currently, those projects receiving funding -and consequently the type of society slowly being created via these investments -reflect the specific worldview of a narrow demographic of already well-resourced and well-networked individuals. Without a greater democratisation of investment (Palladino 2019), the outcomes of crowdfunding activities will increasingly tend only to reflect the ethics of participating white men, over-55 years of age, and retired. Thus, we reiterate that it is vital for future research to pursue an analysis of the 'gendered regime' of finance (Walby 2015). We therefore agree with Langley and Leyshon (2017) that, until participation diversifies considerably, it is sensible to caution against any hasty celebration of crowdfunding as already being 'alternative, disruptive, or democratic'. If evermore members of the wider public are to be encouraged to shift their funds into crowdfunding, in pursuit of a democratic finance, then we conclude that enhanced investor protection is needed first so that the victims of neoliberal financialisation are not further exploited.