August 31, 2015

CROWDFUNDING – A SIGNIFICANT CONTRIBUTION TO FINANCIAL SYSTEM DEVELOPMENT IN DEVELOPING COUNTRIES?

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

ACKNOWLEDGMENTS We are grateful to the German Ministry for Economic Cooperation and Development (BMZ) for supporting this work and for comments on an earlier version of the study. We also appreciate helpful comments by KfW. The views expressed in this study are the ones of the authors and do not in any way represent the views of the institutions they are affiliated with or the BMZ. Likewise, all remaining errors remain our own.

We thank Fahad Malik for excellent research assistance.

AUTHORS Adalbert Winkler and Ulf Moslener Frankfurt School of Finance & Management Paper prepared for the Federal Ministry for Economic Cooperation and Development

Copyright © Frankfurt School of Finance & Management gGmbH 2015.

CITATION Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance (2015), Report Crowdfunding – Can it provide a significant contribution to financial system development in developing countries?, http://www.fs-unep-centre.org/

This publication may be reproduced in whole or in part and in any form for educational or non-profit purposes without special permission from the copyright holder, provided acknowledgement of the source is made. Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance would appreciate receiving a copy of any publication that uses this publication as a source. No use of this publication may be made for resale or for any other commercial purpose whatsoever without prior permission in writing from Frankfurt School of Finance & Management gGmbH.

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

TABLE OF CONTENTS

1.

INTRODUCTION ....................................................................................................................2

2.

SIZE, STRUCTURE AND KEY CHARACTERISTICS OF CROWDFUNDING................................3

2.1.

STRONG GROWTH AND SMALL SIZE: A SNAPSHOT OF CROWDFUNDING ............................... 3

2.2.

KEY CHARACTERISTICS OF CROWDFUNDING ............................................................................ 8

2.3.

CROWDFUNDING AND MICROFINANCE – KIVA.ORG .............................................................. 12

3.

FINANCIAL DEVELOPMENT AND CROWDFUNDING ..........................................................16

3.1.

FINANCIAL DEVELOPMENT – A STYLIZED VIEW ...................................................................... 16

3.2.

WHY DOES INTERMEDIATED FINANCE DOMINATE? - ANSWERS FROM THE THEORY OF BANKING ................................................................................................................................... 19

3.3.

BANKING AND CROWDFUNDING – MODELS AND REALITY (ADVANCED ECONOMIES) ....... 21

3.4.

BANKING AND CROWDFUNDING – MODELS AND REALITY (DEVELOPING COUNTRIES) ...... 22

4.

CROWDFUNDING’S POTENTIAL FOR FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES .........................................................................................................................25

5.

POLICY CONCULSIONS .......................................................................................................28

6.

REFERENCES ......................................................................................................................30

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

LIST OF CHARTS CHART 1: CROWDFUNDING - TOTAL FUNDING VOLUMES AND GROWTH ............................................... 4 CHART 2: CROWDFUNDING PLATFORMS - TOTAL NUMBER AND GROWTH ............................................ 4 CHART 3: CROWDFUNDING - REGIONAL DISTRIBUTION OF VOLUMES .................................................... 5 CHART 4: A CLOSER LOOK AT REGIONS WITH COMPARATIVELY LOW AMOUNTS OF CROWDFUNDING VOLUME ....................................................................................................................................................... 5 CHART 5: CROWDFUNDING - VOLUMES BY MODEL .................................................................................. 6 CHART 6: NUMBER OF SUCCESSFULLY FUNDED PROJECTS ON KICKSTARTER AS OF DECEMBER 2014, BY AMOUNT OF MONEY RAISED (IN USD) ................................................................................................. 7 CHART 7: CROWDFUNDING – WHERE THE MONEY GOES ......................................................................... 8 CHART 8: TRANSACTION COSTS OF DIRECT VERSUS INDIRECT FINANCE – THE TRADITIONAL VIEW ..... 9 CHART 9: PROJECTS PROPOSED AT KIVA.ORG AS OF MARCH 2015 - GEOGRAPHICAL DISTRIBUTION. 12 CHART 10: PROJECTS PROPOSED AT KIVA.ORG AS OF MARCH 2015 - DISTRIBUTION BY SECTOR ....... 13 CHART 11: CROWDFUNDING VIA KIVA .................................................................................................... 13 CHART 12: FINANCIAL AND ECONOMIC DEVELOPMENT......................................................................... 17 CHART 13: FINANCIAL DEVELOPMENT – A STYLIZED VIEW .................................................................... 18 CHART 14: EQUITY AS A PERCENT OF ASSETS FOR ALL INSURED COMMERCIAL BANKS, UNITED STATES, 1840 – 1989 .................................................................................................................................. 24 CHART 15: INTERNET PENETRATION VERSUS ECONOMIC DEVELOPMENT ............................................. 25 CHART 16: LEVEL OF TRUST ACROSS COUNTRIES .................................................................................... 27

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

LIST OF TABLES

TABLE 1: TRUSTEE RATINGS ...................................................................................................................... 15 TABLE 2: FINANCIAL DEVELOPMENT AND MECHANISMS FOR OVERCOMING ASYMMETRIC INFORMATION PROBLEMS ........................................................................................................................ 21 TABLE 3: KEY CHARACTERISTICS OF PRIVATE BANKERS IN THE EARLY STAGES OF FINANCIAL DEVELOPMENT - THE CASE OF GERMANY ............................................................................................... 23 TABLE 4: SIZE OF THE INFORMAL ECONOMY AND INCOME QUARTILES ............................................... 26

LIST OF ANNEXES

ANNEX 1: THE POPULARITY OF CROWDFUNDING ................................................................................... 35 ANNEX 2: DISTRIBUTION OF CROWDFUNDING PLATFORMS BY MODEL ................................................ 36

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

ABBREVIATIONS CFP

Crowdfunding platform

GDP

Gross Domestic Product

IFI

International Financial Institution

LAPO

Lift Above Poverty Organization

MFI

Microfinance Institution

NGO

Non Governmental Organisation

P2P

Peer to Peer

PPP

Purchasing Power Parity

SME

Small or Medium Enterprise

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

1. INTRODUCTION Crowdfunding refers to an open call by borrowers to investors via the internet. It represents direct, uncollateralized lending with a high degree of lot size transformation. As such, it contrasts with financial intermediation, i.e. uncollateralized lending to / depositing at a bank and collaterized (monitored) lending by the bank to final borrowers. Crowdfunding has captured the imagination of finance professionals, academics and the public. Over the last years google listings, academic literature databases and the financial media have seen a multiplication of entries featuring the term “crowdfunding” (see Annex 1) as news about projects successfully funded on the corresponding platforms have made headlines and funding volumes show strong growth. The rise in popularity of crowdfunding can be linked to two factors. First, crowdfunding promises a “democratization” of fund raising by small and young entrepreneurs, broadening their choices beyond family and friends, angel investors, venture capitalists and banks (Harrison 2013). This promise is exciting given the decline in trust in traditional financial service providers, notably banks, reflecting the global finance crisis.1 Second, the negative impact of the global financial crisis on small and young business finance from traditional sources, such as bank lending (Bruton et al. 2015), has triggered a search for new financing mechanisms and options. At the same time, investors hunt for yield in the current low interest rate environment, and some of them might think to find this yield on crowdfunding platforms. Indeed, some observers speculate that crowdfunding might contribute to a modification (Sviokla 2009) or even the demise of traditional finance, notably banking, at least in mature economies (Mayer 2015). Against this background, it is not surprising that there is interest in assessing the potential of crowdfunding for the developing world (World Bank 2013). From a financial sector perspective two arguments readily support efforts in this regard: First, crowdfunding can be seen as the online version of informal finance, i.e. financial transactions within the informal financial sector that still accounts for a substantial share of finance in developing and emerging market countries. Second, if crowdfunding were to establish itself as a reasonable alternative to traditional finance in mature economies, it could represent an avenue towards financial development and financial inclusion in developing countries where entrepreneurs’ access to finance is – irrespective of financial crisis effects – more limited than in mature economies. Thus, crowdfunding might offer the chance to bypass banks and other financial institutions as key providers of funds in developing and emerging countries (World Bank 2013). As a result, the slow and thorny process of financial institution building, a cornerstone of development finance since the mid-1990s (Krahnen and Schmidt 1994, Federal Ministry for Economic Cooperation and Development 2004) pursuing the goal of establishing sustainable financial intermediaries and thereby promoting financial inclusion in developing countries, might become less important. It could be supplemented or even replaced by efforts to foster crowdfunding. However, there are also critical voices on the sustainability and medium term contribution of crowdfunding to financial system development in mature economies, raising questions about its usefulness as an instrument to enhance financial inclusion and entrepreneurs’ access to finance in developing countries. These voices focus on the high risks associated with funding businesses with little screening and monitoring efforts by the funders, and no collateral and little reputation of the recipients. Naively applied crowdfunding opens doors for moral hazard and adverse selection behavior among recipients potentially implying that many of the projects being funded “are going to flop” (Dorff 2014). Subprime mortgage loans serve as a critical reminder: hailed as “democratization of credit” (Greenspan 1997, Gramlich 2004) in its early stages, it ended in crisis.2 Overall this would suggest that crowdfunding does not conform to responsible finance principles. This paper aims at answering the question whether crowdfunding represents an instrument that can make a significant contribution to financial development and inclusion in developing countries from a financial system perspective. It starts out by reviewing the state of play of crowdfunding in mature and developing countries (section 2). Concretely, we provide an overview of size, structure and key characteristics of crowdfunding. 1

For example, in Germany bankers rank 27th out of 32 professions in a survey on trust in professions (Gfk 2014).

2

Klein (2015) draws a direct comparison between subprime lending and crowdfunding.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

2

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Moreover, we have a detailed look at the most important crowdfunding initiative related to development finance, kiva.org. Section 3 takes a systemic view, i.e. it reflects on the potential of crowdfunding for financial development in developing countries by reviewing the empirical evidence and theory of financial development. In doing so we focus on mechanisms that have been used to mitigate moral hazard and adverse selection risks of finance in informal markets, intermediaries, notably banks, and organized markets. Section 4 combines the insights gained in sections 2 and 3. We find that financial underdevelopment does not enhance but hamper crowdfunding’s potential in developing countries. Indeed, the very arguments explaining financial underdevelopment in many countries, namely the repeated failure to establish and develop mechanisms mitigating adverse selection and moral hazard problems, also lead to the conclusion that crowdfunding faces a substantially stronger uphill battle in developing countries than in mature economies. Thus, even if internet penetration and online networks were to reach levels comparable to advanced economies, crowdfunding is unlikely to flourish in many developing countries as platforms, recipients and funders are unlikely to make use and enforce the very mechanisms of mitigating asymmetric information problems rejected in more traditional finance. However, without these mechanisms in place crowdfunding will not develop. We draw three policy conclusions from our analysis. First, at the current juncture crowdfunding does not (and should not) constitute a major field for development cooperation activities with regard to financial sector promotion. Second, financial institution building remains the right approach, even from the perspective of a possible enlarged role for crowdfunding in developing countries, as a higher level of financial development with strong institution does not hamper but raise the chances for crowdfunding to develop in these markets. Third, if crowdfunding platforms in developing countries were to develop, and if the founders were to ask for support from the development finance community, pilot projects involving donors, technical assistance providers and international financial institutions (IFIs) are conceivable. Those may be designed in order a) to learn about the concrete conditions in setting up such a platform in a developing county context, and b) to monitor, support and possibly enforce the use of mechanisms that limit moral hazard and adverse selection risks for local investors and thus enhance and safeguard the reputation of the platform.

2.

SIZE, STRUCTURE AND KEY CHARACTERISTICS OF CROWDFUNDING

2.1. STRONG GROWTH AND SMALL SIZE: A SNAPSHOT OF CROWDFUNDING Crowdfunding has seen spectacular and rising growth in recent years. Growth rates of global funding volumes have increased from 72% in 2011 to 167% in 2014, leading to a rise in total funding volumes from about USD .85 bn in 2010 to USD 16.2 bn in 2014. Growth is expected to stay at elevated levels of 112% in 2015, mainly driven by growth in Asia, notably China, which would lead to a total funding volume of USD 34.4 bn (Chart 1). World Bank (2013) expects that the global market will reach a volume of USD 93 bn by 2025.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

3

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Chart 1: Crowdfunding - total funding volumes and growth

Source: Massolution (2015), authors’ compilation

In parallel, the number of crowdfunding platforms has increased from 100 in 2007 to 1250 in 2014 (Massolution 2015). However, platform growth peaked in 2012 (+80%) and has become more modest in recent years (Chart 2). Chart 2: Crowdfunding platforms - total number and growth

Source: Massolution (2015), authors’ compilation

Despite strong growth, crowdfunding accounts for an extremely small size of the overall financial sector. Indeed, total global crowdfunding volumes in 2014 are about twice as large as the 2013 balance sheet of the savings bank of Frankfurt (Frankfurter Sparkasse).3 Crowdfunding is an activity mainly conducted in mature economies. In 2014, close to 80% of the total volume was generated in North America and Europe (Chart 3).4 Of the remaining 20 3

The gap between traditional finance and crowdfunding can also be illustrated by comparing global crowdfunding volumes with the volume of total outstanding credit by US banks to the private non-financial sector. The latter amounted to USD 25.1 trillion at the end of the third quarter 2014 (https://research.stlouisfed.org/fred2/series/CRDQUSAPABIS), which is more than 1,500 times larger than global crowdfunding volumes in 2014.The comparison is for illustrative purposes only. Strictly speaking, a valid comparison would have to adjust for the different time dimensions of both indicators, i.e. the balance sheet size of the Frankfurter Sparkasse and the volume of outstanding credit by US banks are stock variables, while volumes funded on crowdfunding platforms represent flow variables. Size and direction of the adjustment depends on the average maturity of the funds provided.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

4

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? percent, the bulk comes from Asia, notably China with its dominance in P2P lending. Similar evidence can be found for the number of crowdfunding platforms across regions (Massolution 2013, Massolution 2015). Moreover, with the exception of the China driven developments in Asia, this regional distribution has been basically stable over time. Chart 3: Crowdfunding - regional distribution of volumes

Note: „Others“ include Asia for 2009-2011. *2015 values represent predictions. 2012 „Asia“ and 2009-2011/2013-2015 „Others“ do not represent missing values, but very low funding volumes (USD 0.03, 0.02, 0.03, 0.05, 0.05, 0.11, 0.18 bn respectively). Source: Massolution (2013), Massolution (2015), authors’ compilation

By contrast, Latin America and Africa have barely participated in the rise of crowdfunding over recent years (Chart 4). Volumes have either stagnated or even declined and expectations for 2015 do not suggest a strong catching-up. Thus, claims according to which “crowdfunding is quickly becoming a global financing methodology” (Massolution 2015, 20) have to be substantially qualified. Asia, mainly due to P2P lending in China, is the only emerging market region seeing strong growth and (comparatively) significant volumes. Chart 4: A closer look at regions with comparatively low amounts of crowdfunding volume

Source: Massolution (2015), authors’ compilation

4

Kshetri (2015) refers to a “US-centric phenomenon” (p. 100).

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

5

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? There are two main crowdfunding models: a) Non-financial crowdfunding either in the form of donation- or reward-based crowdfunding; b) Financial crowdfunding in the forms of lending-, equity and royalty-based crowdfunding (Massolution 2015). As the terms indicate non-financial crowdfunding implies that the funder does not get a financial return or compensation for providing funds. If it is not an outright donation, rewards can take the form of small gifts, special treatment of the funders (e.g. mentioning the funders’ names when presenting the project funded or the project’s output), and pre-sales of the product that was funded (Hemer 2011). Creative industries provide the main example for these kind of rewards and related crowdfunding activities. By contrast, financial crowdfunding mimics in principal traditional debt and equity finance, i.e. recipients have to pay interest and return the funds to funders in the lendingbased model, while funders participate in profits and losses of the project or business when providing capital via the equity-based model. Originally, donation- and reward-based transactions represented the majority of crowdfunding transactions. However, in recent years, P2P lending has emerged as the by far most important crowdfunding activity on a global level, substantially ahead of donations-, reward- and equity-based funding (Chart 5). P2P lending volume amounted to USD 11.1 billion in 2014, i.e. more than 68% of the total volume. In 2015, the share of P2P lending is expected to rise to 72% of a a total of 34.4 bn. Fitch Ratings (2014) and PwC (2015) expect P2P lending volumes rising to USD 114 bn and USD 150bn over the medium term, respectively. Chart 5: Crowdfunding - volumes by model

Source: Massolution (2015), authors’ compilation

The rise in global P2P lending has been driven by the US, the UK and China (Grant Thornton 2014). Just in the US P2P lending amounted to about USD 5.4 bn in 2014. Small-scale consumer lending, mainly to prime and near prime borrowers, is dominating P2P lending, often with the purpose of refinancing more traditional consumer debt, e.g. credit card debt (Bruett 2007, Fitch Ratings 2014). By contrast, business lending remains underdeveloped. For example, on Lending Club, one of the two leading US P2P platforms business loans accounted for about 3.5% of total lending in the period mid-2007- 2012 (Mach et al. 2014) Funding of P2P loans is increasingly provided by non-crowd investors, notably institutional investors such as hedge funds or pension funds. There are estimates according to which the share of this investor group in total US P2P lending was approximately 80% in 2014 (Morse 2015). Transactions on P2P platforms show on average much larger volumes (USD 48 million per platform) than transactions on platforms pursuing the donation- (USD 6.8 million per platform) and reward-based (USD 3.7 million) crowdfunding model. This reflects the fact that there is a larger number of reward and donation based platforms than lending platforms (Annex 2) despite the drastically lower volumes. Moreover, in particular in North America, P2P lending is Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

6

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? dominated by a few platforms. In 2014, “the top two North American P2P lenders control almost 95% of their respective market, while the top ten largest North American CFPs across all models transacted 88% of the total North American funding volume.” (Massolution 2015, 19) Overall, i.e. including all crowdfunding models, most projects funded are small, with SME loans and equity-based funding representing the main exceptions. Average “campaign-sizes” are less than USD 4,000 for consumer lending-, reward- and donation-based crowdfunding, while SME loans and equity-based funding campaigns reach average sizes of about USD 93,000 and USD 262,000 respectively (Massolution 2015). Evidence from Kickstarter, a platform providing for rewardand donation-based funding only, illustrates this pattern: 86% of all projects funded between 2009 and 2014 had a volume of less than USD 20,000 (Chart 6). Chart 6: Number of successfully funded projects on Kickstarter as of December 2014, by amount of money raised (in USD)

Source: Statista (Kickstarter)

Business and entrepreneurship funding accounts for the largest share of crowdfunding volume, with a rising trend (Chart 7). However, funding for social causes (18.9%) and funding of creative industries (Films & Performing Arts, Music & Recording Arts, Fashion, Art (general) and publishing arts – 21.2%), the early drivers of crowdfunding (Hemer 2011), remain an important part of the industry.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

7

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Chart 7: Crowdfunding – where the money goes

Source: Massolution (2015), authors’ compilation

2.2. KEY CHARACTERISTICS OF CROWDFUNDING In its essence, crowdfunding involves uncollateralized lending with a high degree of lot size transformation, i.e. many small contributions – “the crowd” – finance one project, one household or one business (Herzenstein et al. 2011). Lot size transformation is a key function of finance (Bodie and Merton 1995), characterizing banks as well as markets. Thus crowdfunding represents a new form to perform the function of finance. The peculiarity of crowdfunding with regard to lot size transformation is the open call to investors via the internet, distinguishing crowdfunding from traditional informal finance, where this call is transmitted to a small group of people, i.e. family and friends. 5

Thus, crowdfunding via the internet significantly expands the geography of informal finance via “matching systems” provided by online platforms which lenders can use “to generate portfolio recommendations and minimize lending risks.” (Emekter et al. 2015, 55) Available evidence suggests that the average distance between funders and recipients can be as wide as 3,000 miles (Agrawal et al. 2013). Crowdfunding differs from direct finance performed on organized markets, as the latter are highly regulated by laws and regulations aimed at providing transparency and investor protection. However, in order to generate trust in the crowdfunding mechanism, investors, recipients and platforms – supported and pushed by governments and regulatory agencies – are in a process of drafting and implementing regulations and supervision thereby limiting the “open call” character of crowdfunding. It is argued that in particular, investors with purely financial motivations and geared towards (small) business finance call for stricter regulation, also because of the higher risk of business than consumer finance (Mach et al. 2014). Investors are said to seek “greater limits on the amount entrepreneurs can raise, and lower thresholds for audited financial statements, more education, greater portal due diligence, and other protections to mitigate risks.” (Cumming and Johan 2013, 376). Moreover, in stark contrast to organized markets and to banks, crowdfunding platforms do not generate liquidity of the assets created, i.e. there is no maturity transformation. Investors are typically unable to sell their claims on the secondary market. Low search and communication costs as well as the ability of funders to keep risk exposures low by funding in small increments have been key success factors of

5

In the Crowdfunding Industry Report crowdfunding is defined as “any kind of capital formation where both funding needs and funding purposes are communicated broadly, via an open call, in a forum where the call can be evaluated by a large group of individuals, the crowd, generally taking place on the Internet.” (Massolution 2015, 34)

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

8

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? crowdfunding (Agrawal et al. 2013). Together they have reduced the transaction cost advantage of intermediated (indirect) finance, for example via banks, traditionally named as a key argument in explaining the existence and superiority of bank intermediated finance compared to direct finance on open markets (Chart 8). However, there is also the view expressed that potential borrowers refrain from entering the crowdfunding market due to substantially higher transaction costs associated with investor management involving many creditors.6 As a result, when (re-)gaining the option accessing traditional sources of funding borrowers prefer the more traditional funders, such as banks. Chart 8: Transaction costs of direct versus indirect finance – the traditional view

Transaction costs of direct finance: m * n * c With c = transaction costs per transaction Relevant for traditional informal finance, capital market finance and crowdfunding

Transaction costs of intermediated finance: (m + n)*c Relevant for banks and other financial intermediaries, i.e. regulated microfinance institutions

Source: authors’ compilation

Funding is highly skewed. Many borrowers looking for funds do not get access to finance. On reward- and donation based platforms success rates are in the range between low single digits to upward of 40% (Massolution 2015, 62 f.). Success ratios are also small in P2P lending: Analyzing differences between consumer and business lending on Lending Club Mach et al. (2014) use a dataset with “more than 670,000 rejected loan applications and just under 100,000 funded loans” over the period mid-2007 – 2012. Individual crowdfunding success very much depends on early contributions. Moreover, there is strong evidence that funding propensity increases with accumulated capital indicating herding behavior by investors (Herzenstein et al. 2011, Lee and Lee 2012, Massolution 2015). This suggests that crowdfunding is subject to the “known market failure in financial markets”, namely the “[e]xternality of monitoring, selection, and lending”, i.e. that someone is more willing to lend money to a project if he knows that a third party has already decided to lend money“ (Stiglitz et al. 1993). Family and friends play a key role in the early stage of funding, suggesting similarities between crowdfunding and informal finance, between crowdfunding and more traditional ways of entrepreneurial finance, such as business angel investment (Conti et al. 2013), and between crowdfunding and microfinance (Lee and Lee 2012). Early contributions by family and friends send a quality signal to other investors allowing for a successful crowdfunding operation. Without this signal funders are likely to free-ride on the screening and monitoring efforts of others and postpone their investment. As a result, the market fails “as everyone waits and nobody invests.” (Agrawal et al. 2013).

6

Crowdfunding platforms involved in equity-based crowdfunding aim at reducing these costs by raising “funds in a Special Purpose Vehicle, which becomes the nominal shareholder of the entire issue and places a single party to represent the interest of the crowdfunding contingent.” (Massolution 2015, 44)

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

9

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Recent research suggests that acquaintances, group membership and friendships built within crowdfunding platforms (and online media in general) may substitute for family and friends in providing early funding and hence enhance the chance of funding success (Lin et al. 2013, Colombo et al. 2015, Freedman and Jin 2014). However, there is conflicting evidence on whether and to what degree the signal provided by these expressions of social capital is of reliable quality, i.e. whether and to what extent social capital acquired online is a good indicator of credit risk. Moreover, Morse (2015, 15), referring to a paper by Lu et al. (2012), argues that online connections might also involve negative externalities: “when a borrower friend defaults, the likelihood that the borrower will default more than doubles.” 7 P2P platforms, benefitting from each transaction by receiving a fee from borrowers successfully tapping the crowd, aim at fostering participation by limiting credit, moral hazard and adverse selection risks which funders are exposed to. These activities involve various measures. For example, platforms screen borrowers according to credit scores and admit only borrowers to the platform presenting reasonable credit risk. They also report defaulting borrowers to credit agencies and “hire a collection agency to collect the funds on behalf of the lender.” (Emekter et al. 2015, 83). Some platforms, like kiva.org, also make use of intermediaries between funders and end-borrowers, who – by having skin in the game, i.e. participating in losses – have an incentive to screen and monitor applicants. However, platforms still stop short of having “skin in the game” themselves, keeping the role of a broker and refraining from an intermediary role (Greenbaum and Thakor 2007). The long-term sustainability of the crowdfunding model is likely to depend on the success of these efforts. This is because crowdfunding platforms – in particular P2P platforms – are subject to substantial reputation risk, also characterizing more traditional finance, such as banks or markets (Gorton and Mullineaux 1987, Calomiris and Gorton 1991, Gorton 2008). This risk – if it were to materialize – could provide a decisive blow to the industry. Platform reputation risk arises if investment decisions by lenders do not only reflect the assessment of the credit risk of the respective borrowers, but also trust in the platform as such, i.e. in its screening and monitoring efforts geared towards limiting lenders’ credit risk (Bruett 2007). Thus, individual failures of investments may lead investors to stop investing at the platform and – due to spillover and contagion effects – to invest in P2P lending at all (Fitch Ratings 2014). Platform reputation risk characterizes all decentralized market places, including online market places such as eBay (Nosko and Tadelis 2015). While there are ideas to design systems that would facilitate and enhance reputation building through online platforms (Dellarocas 2010), these will likely be less effective in the case of crowdfunding as investors do not repeatedly interact with a certain borrower – like buyers do with a certain seller on eBay (Agrawal et al. 2013). 8 Thus, the individual borrower is less able than the individual seller to build-up a reputation of its own, providing protection against spillover and contagion effects that may derive from defaults of other, unrelated borrowers. Relevance and size of platform reputation risks are difficult to assess as crowdfunding has not been subject to a negative credit quality shock yet (like the decline in house prices with regard to subprime lending, Gorton 2008).9 For sure, risks are at elevated levels, as despite all efforts de facto risk costs have been found to be substantial, reflecting borrower incompetence, fraud, moral hazard risks as well as project risk (Agrawal 2013). In P2P lending non-performing loan ratios have developed in line with standard consumer finance loans (Demyanyk and Kolliner 2014). However, as stressed by Fitch Ratings (2014), the good performance of P2P lending in recent years might simply reflect the fact that loan quality has not been put to a test yet by a recession or a rise in

7

Thus, the merits of online friendships in signalling credit quality are similar to those of friendships in traditional microfinance group lending. While there are studies suggesting that self-selected groups, i.e. where people know each other, show better repayment behavior, other studies suggest that it invites to collusive behavior (Armendáriz and Morduch 2010). 8 The same point can be made when comparing crowdfunding to banks: Crowdfunding platforms face more difficulties than banks in establishing long-term relationships with their investors as crowdfunding does not involve as many repeated transactions as in the case of banking as banks do provide a whole range of financial services to their customers (Fitch Ratings 2014). 9 However, as long as crowdfunding does not involve maturity transformation, i.e. the provision of liquidity, the financial stability implications of such a shock are likely to be more subdued: Investors cannot run on platforms or borrowers. Hence, the main effect would be a significant decline or total stop of activity on platforms.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

10

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? interest rates. However, in a less benign environment credit risks may quickly become visible. Thus, the long-term success of crowdfunding remains an open issue. Donation- and reward-based platforms are less exposed to reputation risks as funders are not only influenced by financial but also by non-financial considerations when providing capital to recipients. Moreover, these non-financial considerations are very much directed to the individual recipient or project being supported, limiting spillover and contagion risks. Non-financial considerations include (Agrawal et al. 2013): a) getting access to investment opportunities with (seemingly) great potential, b) gaining early access to new products (reward-based crowd funding), c) attaching value to the funding act as such by becoming part of the entrepreneurial activity that is funded. (Bruett 2007) Non-financial considerations are also a driving force for fund recipients when turning to crowdfunding platforms. These include (Agrawal et al. (2013): a) information about demand for the product offered to fund, b) input funders may provide on how to improve the product or the associated business plan. Some large corporations even engage in crowdfunding activity purely for these reasons as they face no challenges in raising capital in the traditional financial sector (Massolution 2015) Transaction volumes on platforms following non-financial models of crowdfunding are of small size and are likely to remain small. The evidence from mature economies indicates that non-financial funding is mainly attracted by creative industries and by activities linked to social causes as only for these activities there is a reasonably large investor base willing to provide capital primarily based on non-financial motivations. Moreover, due to the inherent characteristic of being non-financial they are largely irrelevant from a financial development and inclusion perspective. 10 They offer much more broad-based forms of fundraising, but not finance.11 Financial models of crowdfunding have boomed in recent years and are expected to continue growing at a rapid pace over the medium term, at least in mature economies, notably the US, and in Asia, notably China. This development has triggered questions whether crowdfunding will become a substantial part of the financial sector, somewhat challenging traditional finance, notably intermediated finance via banks. Financial models are also most relevant from a financial development and inclusion perspective: can these models be imported to developing countries and make a significant contribution for financial development and inclusion, in particular for small and young firms lacking access to finance? There are several characteristics of (financial models of) crowdfunding in advanced economies that have to be taken into account when assessing its potential as a driver of financial inclusion and development in developing countries. 1. Most of P2P lending represents consumer lending involving borrowers with relatively high income levels and comparatively good credit risk (based in track records summarized in credit scores). Thus, it does not represent lending to the poor or to borrowers without access to the formal financial sector, which is a key target group in development finance, notably microfinance. 2. Business lending as a purely financial activity is underdeveloped and just emerging, outstanding exceptions widely covered in the media notwithstanding. 3. Creative industries have been able to tap the crowd, however mainly via donationand reward-based models. 4. Institutional investors are increasingly involved as funders of transactions on P2P lending platforms. Thus, the recent rise of crowdfunding – at least partly – is driven by the non-crowd.

10

It should be emphasized that the distinction between financial and non-financial models is not clear-cut. For example, being non-profit or social ventures might serve as a quality signal for funders facilitating crowdfunding (Lehner 2013, Pitschner and Pitschner-Finn 2014). Vice versa some contracts in P2P lending are written that do not include an interest rate, as “impact investors” provide funds just with goal of supporting struggling entrepreneurs (Massolution 2015). A prominent example is the microfinance platform kiva.org. 11 This is demonstrated by the historical example, often referred to in the crowdfunding literature, namely the funding of the Statue of Liberty.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

11

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? 5. Financial crowdfunding models are subject to the familiar moral hazard and adverse selection problems characterizing financial transactions under asymmetric information. Accordingly, the industry has been developing the familiar mechanisms to mitigate these problems notably signaling, reputation, screening and monitoring. This might also include the involvement of traditional intermediaries in the crowdfunding transaction chain. An important example of the latter is kiva.org, the most widely used crowdfunding platform for microfinance.

2.3. CROWDFUNDING AND MICROFINANCE – KIVA.ORG Kiva Microfunds is an American nonprofit organization based in California. Founded in 2005 its mission is to alleviate poverty through microloans by giving financially underserved people access to capital. Thus, Kiva runs online crowdfunding platforms which allow the users to lend microloans to financially underserved entrepreneurs, students, families and villages around the world. Chart 9 provides information on the geographical distribution of projects proposed as of March 2015. It shows that Africa and Asia account for the bulk of projects and Latin America has a significant share. Thus, in contrast to the geographical distribution of crowdfunding in general, kiva has a clear developing countries focus. Chart 9: Projects proposed at Kiva.org as of March 2015 - geographical distribution

Source: www.kiva.org, authors´ compilation

Users can browse among different profiles of candidates who are in need of a loan and choose who they would like to finance. Kiva users can also specifically search for potential borrowers by region or sector, with agriculture, retail trade and food being the sectors most commonly proposed for funding (Chart 10).The lenders may not charge any interest and in case of a loan default they lose their capital. Kiva also does not charge any interest or fees and covers its own costs almost entirely through grants, loans and donations from its users, corporations and national institutions.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

12

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Chart 10: Projects proposed at Kiva.org as of March 2015 - distribution by sector

Note: “Health”, “Arts” and “Manufacturing” do not represent missing values, but very low project counts (8, 1, 1 respectively). Source: www.kiva.org, authors´ compilation

The Kiva organization has two basic approaches for their crowdfunding platforms. The main approach Kiva started with relies on local field partners who act like an intermediary between kiva lenders and kiva borrowers (Kiva.org). The second is a P2P approach with direct lending from the crowd to borrowers (Kivazip.org).

Kiva.org Kiva.org works with 296 field partners in 86 countries. Kiva provides lenders with a risk rating of all partner institutions, ranging from one to five (highest rating) stars. In April 2015, 25 out of the 296 institutions had a rating at or above 4 stars, 150 institutions were unrated and 22 institutions were either inactive or paused.12 Many Kiva partners are microfinance institutions (MFIs)13, of which 150 are reporting to Mix Market. This indicates that many MFIs are among the better and best managed MFIs worldwide (Krauss and Walter 2009) as Mix Market analysts review and cross-check information provided to them by the reporting MFIs. Chart 11: Crowdfunding via Kiva

Offers for funding at kiva.org

Usually pre-disbursed loans

Source: Ly and Mason (2012a, 644), authors´ compliation

12

Ly and Mason (2012a, 645) find for their sample that the majority of projects (70.2%) posted by kiva partners originate from partners with a rating of four or five stars. 13 Field partners can also be schools, NGOs and social enterprises.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

13

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Field partners screen and monitor borrowers and pre-disburse loans carrying an interest rate of 33% p.a. on average. After disbursement they create an account for them on Kiva to address the lenders. Individual lenders have to provide an amount of at least USD 25. Once the lender has found a project or entrepreneur she would like to finance she transfers the amount she chooses to Kiva via PayPal14. Kiva collects the amounts from different lenders for the specific project and as soon the Kiva community has crowdfunded the entire amount required by the borrower, the money is forwarded to the field partner effectively refinancing the pre-disbursed loans (Chart 11)15. The repayment rate is said to be 98.75%, suggesting that the financial intermediary solution to the asymmetric information problems involved in crowdfunding works. As a result, lenders – formally bearing the specific risk of the project they fund – seem to ignore idiosyncratic project risks relying on kiva’s and kiva partner’s screening and monitoring abilities.16 Thus, Ly and Mason (2012b) find no evidence that projects proposed from partners with a higher rating, record a faster funding speed. However, partner ratings seem to mitigate borrower and project risks. For example, while in general lenders prefer to lend to borrowers that are culturally and socially similar to themselves (Galak et al. 2011), this effect is weakened when the respective borrowers are presented from kiva partners with high ratings (Burtch et al. 2014). While the average Kiva lender issues about 10 loans, most lenders provide few or no loans (Chen et al. 2015). Thus, kiva faces a “challenge in terms of lender engagement” (Chen et al. (2015, 2). The limited number of lenders may also account for the observation that an increase in the number of projects posted at kiva leads to slower funding speeds. This effect gains strength if projects competing for capital are close substitutes, i.e. projects are from the same sector, the same region or from the same field partner (Ly and Mason 2012a). Kiva lenders engage in prosocial lending as they do not receive any interest on their loan. Thus, kiva lenders are “social investors” or “philanthropic citizens” (Ly and Mason 2012a, 2012b). Thus, they are likely to be influenced by the character as well as the narratives of projects and people to be funded, in eceonomic as well as social or development terms. Research increasingly addresses lender motivation and decision making (Ly and Mason 2012b, Moss et al. 2015, Allison et al. 2015). Results are mixed. For example, Moss et al. (2015) find that entrepreneurial characteristics presented and conveyed by the borrower increase the likelihood of funding. By contrast, results by Allison et al (2015, 53) suggest that “lenders respond positively to narratives highlighting the venture as an opportunity to help others, and less positively when the narrative is framed as a business opportunity.” Since its launch in 2005 Kiva claims to have funded 857,791 loans totaling USD 686,306,450 involving 1,586,771 borrowers (74% female borrowers) by 1,265,555 users of the Kiva community. 17 [However, as most loans are short-term loans, Kiva’s outstanding portfolio at year-end is considerably smaller. According to its financial report the outstanding loan amount in 2013 (2012) was USD 44.4 million (35.3 million). For the sake of comparison: on average MFIs reporting to Mix Market had an outstanding loan portfolio of USD 81.1 million (USD 78.8 million) in 2013 (2012). Thus, the total outstandind funding provided via kiva.org amounted to about half of the loan portfolio managed by a single MFI reporting to Mix Market on average. Kivazip.org Kiva Zip started in 2011 and follows the established P2P approach as users can directly lend to borrowers and pass the field partners. Thus, borrowers do not pay any interest on the loans received. Moreover, borrower and lenders can communicate directly with each other, which is not possible on Kiva.org. Users have to lend at least USD 5. Like in Kiva.org the users can browse through different profiles of potential borrowers. Kiva Zip so far operates with borrowers from Kenya and the US only. Loan maturity reaches 2 years on average and average loan sizes are USD 300 in Kenya and USD 5,000 in the US.

14

PayPal forgoes its fees for Kiva.org and Kivazip.org.

15

Kiva transfers money to its field partners only once a month to limit transaction costs.

16

Moreover, as suggested by Roodman (2009), kiva partners might bail out borrowers with repayment difficulties in order to keep its risk rating high and thereby ensuring a continuous inflow of funds at a zero interest rate. 17 Group lending explains the difference between number of loans and number of borrowers,

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

14

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Until March 2015 USD 7,017,153 have been lent on Kivazip.org with a repayment rate of 89.3%, i.e. considerably below the rate observed at Kiva.org. In total 47,338 individual lenders have funded 8,572 loans to small businesses. The role of the intermediary in mitigating problems related to asymmetric information is taken over by a trustee. A trustee is an organization or an individual from the borrower´s community who signals the creditworthiness of the borrower but is not liable in case of a default. The trustee is profiled on kiva.org, i.e. for each trustee there is a track record providing information on his reliability in assessing the creditworthiness of borrowers. Based on this track record Kivazip.org determines the number of borrowers and the loan size a given trustee may support (Table 1). Thus, a trustee and her respected tier (rating) provides a signal to lenders and hence increases the chances of borrowers getting a loan financed. Table 1: Trustee ratings Tier

Requirements

Endorsements Allowed

Max 1st Loan Size

Pilot

Approved Trustee application, > repayment rate

3

USD 5.000

1

10 repayments made, > 80% repayment rate

10

USD 5.000

2

25 repayments made, > 85% repayment rate

20

USD 10.000

3

50 repayments made, > 90% repayment rate

30

USD 10.000

Source: www.zip.kiva.org/faq

Signaling is enhanced for US Kiva Zip borrowers as they have to further invite lenders from their own network and the network of their trustee first before they can have access to the rest of the Kiva Zip community. For loans up to USD 2,500 borrowers have to be funded by at least 10 lenders from within their network. For loans above USD 2,525 borrowers have to be funded by at least 20 lenders from her or her trustee’s network. Kiva City Program. Kiva City was launched in 2011 in Detroit in cooperation with Visa and the Clinton Global Initiative. It aims at small businesses and entrepreneurs from US cities economically stressed. Lenders at least have to lend USD 5 with 0% interest. Potential borrowers can seek a field partner who can predisburse the loan and will post their profile online to collect the money from Kiva. The field partners screen the borrowers and charge them interest. Most of the recent Kiva City profiles use Kivazip.org so that the lenders can directly finance the borrower with a 0% loan. Some examples of Kiva City are: Kiva City Newark, Kiva City Pittsburgh, Kiva City Philadelphia and Kiva City Louisville. Kiva Labs. Kiva Labs was launched in 2013 with the aim of exploring new ways, i.e. products, schemes and technologies, of issuing microloans at affordable, lower interest rates. Kiva Labs also supports clean energy projects as well as implementing mobile information and other transformative technologies.

Box: Other crowdfunding initiatives in microfinance* Lendwithcare (www.lendwithcare.org, since 2010). Lendwithcare is a non-profit microfinance platform from CARE International UK. Although at times it is labeled as a P2P micro-lending platform it works with MFIs who take care of the monitoring and screening of the borrowers. Till March 2013 GBP 2,050,000 were raised consisting of 58,000 loans; the average lender lent GBP 35 and continued to lend 5.8 times. As of April 2015 23,434 lenders are part of the Lendwithcare community. Vittana (www.vittana.org, since 2008). Vittana is a nonprofit organization based in Washington, which helps students in developing countries to fund their academic education. There is a pre-disbursement of Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

15

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? funds which is then covered after collecting the funds through the site. A close family member has to cosign the loan. As with Kiva, a field partner is the intermediary between the site and the costumer. The repayment rate is said to be at 98%. Milaap (www.milaap.org, since 2010). Milaap, based in India, has the aim to offer its clients loans with interest rates which are lower than the average MFI loan. The interest rates of Milaap are between 12%18%. Milaap charges its field partners a 5% fee on the loans. Funds with a total volume of USD 3.88 million have been disbursed; 25,352 loans have been funded with a 99.75% repayment rate. Zidisha (www.zidisha.org, since 2009). Zidisha is a nonprofit organization based in Virginia, which allows lenders to directly lend to entrepreneurs. Zidisha had funded USD 2.1 million to about 6,000 borrowers. For the first loan borrowers have to pay USD 12 in case the loan is funded. Lenders bid on the conditions with each other. The interest rate has to meet the borrowers’ requirement or be lower. A borrower on average pays 10.6% interest (this includes a 5% p.a transaction fee). 82.2% of the borrowers have been paying back or have already paid back in time. 4.7% have been written off. MicroPlace (www.microplace.com, 2006-2014). Microplace ran a platform which allowed users to make investments by purchasing securities from microfinance security issuers. It was bought by ebay and was closed according to ebay in 2014 because of its lack of impact. Wokai (www.wokai.org, 2007-2012). Wokai was a nonprofit based in California, which helped entrepreneurs in rural China. It financed itself through individual donors, corporate sponsors, fundraising events and grants. It closed in 2012 due to “fundraising roadblocks”. The average loan size of Wokai loans was USD 300.. The field partners who disbursed the loans and monitored final borrowers charged between 15% to 20% interest to their clients. Coufenzi (will probably launch in 2015). Coufenzi is a collaboration between Grameen China and JD.com. JD is one of Chinas biggest online direct sales company. Grameen China is part of the Grameen Trust network, a non-profit NGO network originally founded by Nobel Peace Prize winner Muhammad Yunus. The collaboration aims to use JD.com´s crowdfunding platform to establish Grameen Chinas ground operations. FINCA Global (www.finca.org, since 1984). FINCA is a non-profit microfinance organization based in Washington DC and founded in 1984. It provides the unbanked with access to loans, savings, insurances and money transfers. FINCA has 1.8 million active clients in 23 countries. Although it is not an online crowdfunding platform, visitors of the site can donate for specific projects. * This box basically relies on information from the websites of the institutions covered.

3.

FINANCIAL DEVELOPMENT AND CROWDFUNDING

3.1. FINANCIAL DEVELOPMENT – A STYLIZED VIEW Financial development, depicted by the development of the formal financial sector, in particular banks and capital markets, correlates with economic development (Chart 12). This is indicated by the different dimensions of financial system development, i.e. the sum of private sector credit issued by banks, the outstanding volume of public and private bonds, as well as stock market capitalization – all of them measured as a percentage of GDP. Ratios are significantly higher in high income than in upper middle income countries, in upper middle income countries than in lower middle income countries, and in lower middle income countries than in low income countries.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

16

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Chart 12: Financial and economic development

Source: Worldbank, authors’ compilation

The evidence suggests that banking, i.e. intermediated finance, is the first step in formal financial sector development, superseding direct finance in the informal financial sector.

 Many low and lower middle income countries basically rely on banking as the only part of formal financial sector finance, i.e. they have not been able yet to develop meaningful public or private bonds markets or stock markets (Čihák et al. 2012). Bars indicating (public and private) bonds as well as stock market development in low and lower middle income countries in Chart 12 are based on a few countries only as most countries in the respective income groups do not report data for these subsectors of the formal financial system.18

 Informal finance has remained an important part of the financial sector in many developing countries (Rutherford 2000, Madestam 2014).

 Reviewing financial development as it historically emerged in today’s mature economies suggests that banking dominated the early stage of financial development (Lamoreaux 1986, Caprio and Vittas 1997), i.e. capital markets gained significant size after banking had been reasonably established. This includes countries where capital markets have played a more important role, i.e. countries with financial systems characterized as market-based, notably the United States and the United Kingdom (Sylla 1998), while most of continental Europe and Japan have been categorized as bank-dominated financial systems. Overall, modern cross-country datasets and long-term, historical case studies of today’s mature economies indicate a sequence of financial development (Chart 13): 1. banks emerge out of / follow informal direct finance and retained earning finance, 2. capital markets emerge out of / follow banks, and 3. insurance emerge out of / follows banks and capital markets.

18

The stock market bars for low income countries are based on 2 (1990) and 7 (2011) observations. No low income country has yet reported information about bonds market depth. 3 (4) lower middle income countries report data on public bond market development in 1990 (2011), while for private bonds market development the number shrinks to 1 (1990) and 3 (2011).

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

17

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Chart 13: Financial development – a stylized view

Source: Authors’ compilation based on views expressed in Terberger (2007)

The crowdfunding industry might be next step on the financial development ladder for advanced economies (Mayer 2015): due to technological advances reducing transaction costs of direct lending (Morse 2015) crowdfunding emerges from the failure of bank-intermediated finance, as evidenced by the global financial crisis (Mach et al. 2014). Following up on thisview, some studies (e.g. World Bank 2013) argue that crowdfunding might provide a kind of fast track to financial development in developing countries. Indeed, crowdfunding is portrayed as having the potential of at least partly “leapfrogging” steps 3 and 4 of the ladder characterizing financial development in mature economies. This raises questions a) whether such a path of financial development is likely or possible, b) whether such a path of development would be beneficial for developing countries and c) whether development policy, notably the development community, should support this development (and if so, how?) The view that “the developing world has the potential to leapfrog developed countries” (World Bank 2013) with regard to financial development is largely based on the fact that developing countries are financially underdeveloped leaving “substantial reservoirs of entrepreneurial talent” in the respective countries without (proper) access to finance. This funding gap could be filled by crowdfunding. However, while there is little doubt that this funding gap exists, it remains unclear why crowdfunding is a superior instrument compared to banks and traditional capital makets in filling this gap. There is no theory of financial development which could provide direct guidance when assessing the claims by the crowdfunding optimists (de la Torre et al. 2011). Most importantly, there is no theory providing explicit support for the view that historical financial development patterns characterizing mature economies of today reflect fundamental factors suggesting that developing countries will have to go through the same patterns over time. Thus, it might be the case that financial development patterns in today’s developing countries could differ from those in mature economies. Clearly, leap-frogging is possible. However, there is a widely accepted theory of banking, grounded on the theory of financial contracting under asymmetric information. This theory argues that banks employ superior, i.e. more cost-efficient mechanisms in addressing problems of asymmetric information between borrowers and lenders than direct finance which explains the dominance of bank-intermediated finance in advanced as well as in developing countries. Thus, implicitly the theory of banking provides arguments hinting that it Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

18

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? can be expected for banks to emerge out of direct, informal finance and hence become the main source of funds for the real economy compared to direct finance, superseding direct finance in the informal financial sector. Moreover, the same theory is able to explain why some firms are able to and will return to the open market, i.e. direct finance, laying the ground for the development of (liquid) bonds and stock markets. Clearly, this theory would face a severe challenge if developing countries were to leapfrog these stages of financial development to any significant extent. We make use of this theory in assessing the potential for crowdfunding in developing countries by raising and answering the following question: Which advantages of banks (intermediated finance in general) in comparison to direct finance are significantly reduced by crowdfunding? To what extend is this suggesting that crowdfunding might become a viable and significant option for financial development substituting or significantly complementing bank-intermediated finance in advanced economies and thereby leapfrogging financial development patterns in developing countries? Against the background of the evidence compiled on the crowdfunding industry and summarized in section 2, we then continue by asking whether conditions in developing countries are conducive for crowdfunding to exploit this new opportunity for financial development (section 4). Based on this we will address the question whether and how the development community could support this new path of financial development.

3.2. WHY DOES INTERMEDIATED FINANCE DOMINATE? - ANSWERS FROM THE THEORY OF BANKING There are two arguments suggesting that intermediated finance by banks is superior to direct finance by markets: liquidity (maturity transformation – Diamond and Dybvig 1983) and transaction costs (lot size transformation – Diamond 1984). As crowdfunding has been identified as providing a new form of lot size transformation, our focus is on the latter argument (and hence on the latter model).19 Transaction cost arguments related to lot size transformation are of high relevance for assessing the potential of crowdfunding as an alternative to bank-intermediated finance. Lot size transformation implies that many lenders have to monitor a single borrower in order to ensure that the borrower does not engage in moral hazard activities (i.e. pursues a riskier project than proposed to the lenders). Monitoring efforts are needed as it is assumed that borrowers are unable to use other mechanisms, notably signaling, to mitigate moral hazard risks. These alternative mechanisms are a) putting up collateral, b) co-financing with an equity participation and c) reputation (involving non-pecuniary costs in case of failure). As crowdfunding is uncollateralized lending involving lot size transformation and inherently subject to a high degree of asymmetric information (as most borrowers and lenders do not know each other),20 it has to rely on either reputation as a signalling mechanism or on monitoring efforts by investors and/or platforms in order to mitigate asymmetric information problems and to provide a viable and cost-efficient alternative to bank-intermediated finance. Indeed, the review of key crowdfunding characteristics has provided a list of concrete activities by recipients and platforms performing this function, stopping short of platforms themselves credibly signaling (by taking “skin in the game”) the quality of end-borrowers. The rise of crowdfunding in mature economies over the recent years suggests that – to a limited degree in terms of size – the industry has been successful in developing these mechanisms. The competitive advantages of banks compared to direct finance are twofold (Diamond 1984):

 First, banks economize on monitoring costs by serving as delegated monitors. Investors refrain from directly financing borrowers but use banks to engage in lot

19

If banking theory is correct, the very fact that crowdfunding does not provide for maturity transformation rules out that the industry will ever be able to supersede traditional banking. 20 This is in strong contrast to informal finance within family and friends, where lending takes place in an environment characterized by comparatively symmetric information reducing screening and monitoring costs. Moreover, the social context implies that financial transactions within the family and among friends puts the reputation (social capital) of borrowers at stake further reducing moral hazard risks.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

19

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? size transformation on their behalf because – as suggested in Chart 8 – this substantially reduces monitoring costs. 21

 Second, banks do not have to be monitored by depositors because by lending to many borrowers they can diversify credit risk.22 Thus, banks are in a much better position than any single borrower in signaling low credit risk. As a result, (in the Diamond (1984) model) the deposit contract between banks and depositors has neither to be collateralized nor do depositors have to monitor banks. Banks can rely on their reputation and the associated costs of reputation loss in case of failure in mitigating asymmetric information problems vis-à-vis their borrowers.23 They can do so as they will never have to bear these costs if they hold a diversified portfolio and monitor. The emergence of banks significantly expands the funding options of borrowers that are unable to rely on family and friends in providing the needed funds due to a lack of resources. The alternative mechanism, the professional informal financial sector, is very expensive (Rutherford 2000), explaining why the informal financial sector is widespread but small as there are only a few borrowers able to pay the associated high rates. Moreover, since high rates trigger moral hazard and adverse selection behavior by borrowers (Akerlof 1970, Stiglitz and Weiss 1981) only a few lenders are willing to provide funds. The market fails. Finally, high interest rates in the informal financial sector indicate that monitoring costs are substantial, i.e. that economizing on these costs, as done by banks, is decisive to get financial development started. 24 Thus, the development of banking represents a first step in the democratization of credit, as it provides firms with a new alternative of funding that goes beyond the informal financial sector, family and friends – with a low degree of asymmetric information but limited resources –, and the professional informal financial sector, e.g. moneylenders, charging high rates (either reflecting high risks or high monitoring costs) with limited funds as well. Financial institution building as a development policy tool builds on this line of reasoning and has been successful in fostering financial inclusion, for example via the establishment of microfinance institutions (Helms 2006). Financial inclusion is defined as providing firms and households in developing countries with this new funding option. In this regard financial inclusion via financial institution building mimics the efforts by cooperatives and savings banks in mature economies in expanding access of formal sector finance to the population at large (Guinanne, 2002, Schmidt et al. 2015) Borrowers return to the open capital market when they have accumulated reputation themselves by punctually serving debt over time via banks (Diamond 1991). Thus, by performing banking activities, the banking industry itself establishes the basis for some firms and companies to return to the open capital market. By demonstrating their creditworthiness via banks, these firms acquire reputation. Accordingly, investors decide to finance these firms without the involvement of a bank as reputation replaces costly monitoring efforts as the main instrument to mitigate asymmetric information problems.25 As a result, firms with a built-up reputation experience a further democratization of credit, as they do no longer have to rely on either family and friends or on banks as a source of funds only. Financial development can thus be interpreted as development of cost-effective mechanisms, such as intermediated finance, in overcoming asymmetric information problems with severe implications on size and efficiency of financial systems (table 2). The sufficient condition for crowdfunding to establish itself as a viable alternative to bank-intermediated 21

Monitoring costs are reduced from mxnxc to (m+n)*c (with m = number of borrowers, n = number of funders per borrower, depicting the extent of lot size transformation, and c = monitoring costs). 22 In the model (Diamond 1984) assumptions are set in a way that the bank does not face any credit risk if it performs the delegated monitoring role. 23 In Diamond (1984) these costs are referred to as delegation costs. 24

Alternatively, it can be argued that moneylenders have a comparative advantage compared to banks in screening and monitoring their clients. This explains why despite banking sector development the informal sector remains viable and co-exists with formal sector finance in many developing countries (Madestam 2014). 25 By way of example: Households are ready to directly provide funds to Siemens, without knowing anything about Siemens, most importantly its current financial status. They only know that Siemens has served and repaid debt for more than 150 years, making Siemens a reputable borrower.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

20

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? finance at a relevant scale is demonstrating its ability to address moral hazard and adverse selection risks and to do so in a cost-efficient way. Table 2: Financial development and mechanisms for overcoming asymmetric information problems Stage of financial development

Key mechanisms in mitigating asymmetric information problems

Costs

Size

Low

Small

Screening, monitoring

High

Small

Monitoring (via final borrowers), reputation (via depositors)

Low

Large

Low

Large (for those with built-up reputation), Non-existent (for borrowers without reputation)

Informal financial sector - Family and friends - Professional service providers (moneylenders , deposit collectors) Banks

Symmetric information, reputation

Capital markets Reputation

Source: authors’ compilation

3.3. BANKING AND CROWDFUNDING – MODELS AND REALITY (ADVANCED ECONOMIES) In reality, as demonstrated recently in the global financial crisis, banks can fail suggesting that diversification benefits are limited and monitoring of borrowers by banks is far from perfect. Crowdfunding benefits from the doubts in the superiority of banks in mitigating problems of asymmetric information. These doubts relate to the asset and liability side of banking. On the asset side some borrowers do not get access to finance as banks have become more risk-averse and monitoring technologies seem to be too costly when lending to certain borrower groups, i.e. small and young businesses.26 On the liability side, investors, after having seen that banks can fail and that outright defaults have only been prevented by strong interventions of governments and central banks, conclude that – under certain circumstances– direct lending might be the more cost-effective investment approach. The low interest rate environment and related search for yield behavior by investors provides an additional impetus for experimenting with new forms of lending (Morse 2015, Massolution 2015). However, doubts about the ability of banks to perform the delegated monitoring function in a way that it is perceived as being superior to direct finance, is only a necessary but not a sufficient condition for the emergence and medium to long-term sustainability of crowdfunding as a new form of direct finance. Concerns about the ability and costeffectiveness of direct finance in dealing with asymmetric information problems remain (Morse 2015) independently of banks’ apparent weaknesses in this respect. The sufficient condition for crowdfunding establishing itself as a viable alternative option to bank-intermediated finance with relevant scale is demonstrating its ability to address

26

Blaseg and Koetter (2015) provide evidence suggesting that borrowers from banks that became distressed during the global financial crisis, were more likely to turn to crowdfunding than other borrowers.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

21

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? moral hazard and adverse selection risks and to do so in a cost-efficient way. As surveyed in section 2, this process is under way, but – as argued above – not yet tested. Measures taken up to now include

 reducing information asymmetries by making use of the new and low-cost opportunities provided by the internet and its technology,

 establishing borrower reputation via traditional (family and friends) and new ways (social media connections and links) or by the product being offered (creative industries, social entrepreneurship),

 establishing platform reputation by limiting the pool of borrowers to (comparatively) highquality borrowers via screening and by getting engaged in investor protection activities,

 emerging regulation and supervision. In the course of this process, tendencies of re-intermediation emerge, i.e. the original, but “naïve” crowdfunding model, i.e. a platform creating an open market where all borrowers freely meet and interact with all investors, is replaced by models of cooperation or even strategic partnerships (Fitch Ratings 2014) between crowdfuning platforms and more traditional financial institutions, These tendencies are likely to become stronger over time (Morse 2015), suggesting that a further expansion of crowdfunding in terms of volumes and outreach will depend on a successful re-intermediation process.27 If this process fails, crowdfunding is likely to remain a small niche of financial markets only.28 Kiva.org, the most well known platform with a developing finance purpose, provides an example for this by having always relied on intermediaries, i.e. partner institutions (mainly MFIs), in conducting its crowdfunding activities. This can be explained by the fact that asymmetric information problems loom so large in direct cross-border funding to unknown borrowers with no reputation that meaningful volumes cannot be raised (as demonstrated by comparing kiva.org with kiva.zip.org, see section 2). The market fails and is replaced by charity.

3.4. BANKING AND CROWDFUNDING – MODELS AND REALITY (DEVELOPING COUNTRIES) As noted above a key difference between mature and developing countries is the underdevelopment of intermediated finance in the latter group of countries. This raises the question why banking has not taken off in developing countries, or – vice versa – how could and did banking take off in today’s advanced economies. Financial history of mature economies reveals that early banks relied on three mechanisms when starting their activities: 1. Low degree of asymmetric information with borrowers served. Early banking often emerged out of trading and merchant activities, i.e. private bankers had – given the conditions prevailing at that time – information about economic (and social) activities of a broad range of trading partners. Local lending to these partners characterized the asset side of banks in early banking (Lamoreaux 1991, Wang 2008). 2. Bank owners had accumulated reputation either through their commercial or social activities long before entering banking. Bank founders and owners were people with reputation acquired through economic and social (public) activities. For example, in Germany, many bankers had been successful traders and businessmen and active in public life, with links to the political rulers of those days, providing reputation they could build upon when founding their banking business (Table 3). 27

Depending on the concrete form of re-intermediation it could actually raise rather than lower the financial stability risks. The subprime crisis, while directly involving assets in the total amount of more than USD 500 bn, got out of control because of the (intransparent) links to the banking sector (Gorton 2008). 28 The niche is defined by the number of funders that do not invest for financial considerations alone and by the ability of borrowers to develop and make use of reputation in order to mitigate asymmetric information problems. Morse (2015) indicates that this niche might turn large if “big data” would allow reliable, individual credit scoring on a large scale. However, he also notes such a development would trigger new questions such as privacy, monopoly power and discrimination that have not been answered yet. Alloway (2015) presents a balanced discussion on whether credit profiling via “digital scorecards” based on almost any facet of borrowers’ personalities, many of the seemingly unrelated to credit quality, is likely to lower or raise financial inclusion via crowdfunding and other forms of finance.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

22

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Table 3: Key characteristics of private bankers in the early stages of financial development - the case of Germany Private Banks Metzler

Bethmann

Foundation of the Start of the banking business (retail-) company

Other relevant criteria

1674: textile mil

1760: first documented loan transaction

Active in the municipal selfadministration

1700: Foundation of an own retail business trading spices and colour/chemicals commodoties

1731: first documented loan transaction

Commercial transactions were very successful and the founder (Jacob Adami) became one of the most reputable and wealthy citizen of Frankfurt 1755: Johann Phillipp Bethmann was elected as a city council of Frankfurt 1768: Johann Jacob Bethmann was nominated as imperial consul

Rothschild

Family business in merchandise trade in Frankfurt since middle of 16th century

Since the beginning of the 18th 1769: Approbation to the century: bill of exchange business “Fürstlich Hessen-Hanauischen Hoffaktor” (a kind of treasurer/ 1798: Initiation of the bill of “finance minister” of the local exchange – and loan transactions ruler) for Prince Wilhelm 60ies of the 18th Contacts to Prince Wilhelm von 1800: Conducting bond century: foundation Hessen-Kassel transactions on behalf of the of a retail- and bill Elector/ ruler Wilhelm von Hessenof exchange Kassel, the “most significant business (Meyer capitalist at that time” Amschel Rothschild)

Source: authors’ compilation, based on Klein (1982)

3. Strong capitalization. Historically, equity was a key mechanism banks used to mitigate moral hazard risks (Chart 14). Indeed, in early banking equity ratios were substantially higher than today. Only over time, by building-up reputation in the banking business itself, banks were able to increase leverage, getting closer to the highly leveraged banks the existence of which banking theory aims to explain (by simplifying that banks do not hold any equity). In parallel, banking regulation and supervision, combined with government interventions in the form of establishing a central bank and deposit insurance, have supported uncollateralized and unmonitored lending by depositors to banks (Goodhart 1988, Dewatripont and Tirole 1994).

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

23

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? Chart 14: Equity as a percent of assets for all insured commercial banks, United States, 1840 – 1989

Source: The Department of the Treasury (1991)

Many developing countries have been unable to develop their banking sectors as many banks have been founded without the needed backing by equity, reputation and knowledge of the respective founders. Partly this reflects the legacy of colonial times where an endogenous development of local banking was severely hampered, also because the non-financial private sector lacked depth, wealth and reputation (Newlyn and Rowan 1955). However, it also reflects ill-conceived financial sector policies, such as the reliance on government-owned banks (La Porta et al. 2002), deliberate attempts of financial sector repression (McKinnon 1973), naïve liberalization efforts that ignored the need for mechanisms to address incentive problems related to asymmetric information (Diaz-Alejandro 1985) as well as weak property and creditor rights (de la Torre et al. 2011). In recent years, several developing and emerging market countries invited foreign-owned banks to develop local banking sectors, making use of their reputation, expertise and capitalization. This holds in particular for countries in Central, Eastern and South-Eastern Europe that have joined or have the perspective to join the European Union as well as some Latin American countries (Cull et al. 2010, Claessens and Van Horen 2012). Overall this suggests that in many developing countries mechanisms to mitigate problems of asymmetric information in a cost efficient way have been made less use of. Financial underdevelopment and the lack of progress in climbing the ladder of financial development characterizing mature economies is an expression of the failure to account for and address information asymmetries which make financial transactions inherently more complex than most transactions involving goods and services.29 However, the analysis of the crowdfunding market in advanced economies suggests that the use of mechanisms to mitigate and overcome asymmetric information is key to develop crowdfunding beyond a niche activity. This process is ongoing in various forms, including a reapproachment with traditional financial institutions, i.e. banks. It remains to be seen whether this process will be successfully completed. This raises severe doubts about crowdfunding’s potential as a means of financial sector development in developing countries. This holds in particular with regard to small and young 29

Of course, as Akerlof (1970) has demonstrated with the „market for lemons“, i.e. the used car market, these challenges are not confined to financial transactions.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

24

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? business finance where information challenges are most acute. Moreover, it implies that supporting activities in the developing world which are targeted at mitigating the problems related to asymmetric information are useful irrespective of the question to what extent crowdfunding might replace other forms of more formal finance.

4.

CROWDFUNDING’S POTENTIAL FOR FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES Combining the analysis of the crowdfunding market pursued in section 2 with the financial development perspective outlined in section 3, we conclude that the potential for crowdfunding contributing in a significant way to financial development and inclusion in emerging markets and developing countries is limited. This holds in particular for developing countries as they lack almost all ingredients that contributed to rise of crowdfunding in advanced economies, notably the US, over the last years. Prospects are better for emerging market economies, but mainly because they show a higher level of financial development than their developing country peers. The arguments supporting this proposition can be summarized as follows: 1.

Internet penetration is low in many developing countries. There is a positive non-linear relationship between internet penetration and economic development, measured by GDP per capita in PPP terms (Chart 15). Thus, the most fundamental precondition from a “hardware” perspective is not given in many developing countries, providing the first explanation why crowdfunding has basically been a mature economy event. However, internet penetration is a conditio sine qua non for online reputation to develop in the first place.

Chart 15: Internet penetration versus economic development

Source: WorldValueSurvey, World Bank, authors’ compilation.

2.

Most developing countries lack the kind of borrowers that have been tapping the crowd in mature economies over the last years. Creative industries play a marginal role in most developing countries (De Beukelaer 2014). Moreover, there are few prime and near prime consumers in need of refinancing their consumer debt. Finally, the business population is largely composed of firms and companies operating in the informal sector, i.e. firms that are either unable or unwilling to provide information about thrie creditworthiness to the public. As Table 4 suggests there is a strong negative correlation between per capita income levels and the size of the informal sector, irrespective of how informality is measured. Thus, even if internet penetration were significantly higher, most businesses would not provide the kind of information about their activities and their creditworthiness, for example their credit rating, online. However, this is the kind of information crowdfunding platforms in mature economies

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

25

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? intermediating significant funds between lenders and borrowers demand from their potential borrowers in order to generate capital supply. 30 i

Table 4: Size of the informal economy and income quartiles Country

Measure of informality

Income quartile

GDP/ population

% GDP informal (World Economic Forum)

% Tax evasion (Enterprise Survey)

% Self % GDP employment informal (electricity consumption)

Registered firms/ population (1.000 s)

Bottom Second Third Top Sample mean Difference 1 vs. 4 quartile

429 1.362 4.002 20.348 10.015 -19.919*

35,4 33,7 27,6 17,3 27,6 -18,1*

29,0 23,3 19,7 8,2 22,5 -20,8*

46,4 35,7 23,1 13,3 26,5 -33,1*

38,9 42,7 31,3 17,6 29,0 -21,4*

3,2 8,2 28,7 41,8 24,7 38,7*

Observations

185

125

95

133

57

83

st

th

Note: Table 1 presents various measures of the size of the informal sector, with 185 countries grouped by the quartile of per capita income.*, ** and *** indicates significance at the 1,5, and 10 percent levels, respectively Source: La Porta and Shleifer (2014)

3.

Investors (depositors) within developing countries are unlikely to make use of crowdfunding platforms. In many countries a large share of the population does not even have a bank account. This does not only reflect high transaction costs but a lack of trust in banks. However, this does not imply that crowdfunding has a better chance to attract funding / investing in developing countries because crowdfunding assets are – compared to bank depositis – illiquid and highly risky assets, with credit risk for the reasons referred to above likely to be even higher than in mature economies. Moreover, given the small scale of wealth of most “would be depositors” it is also unlikely that non-financial considerations will play an important role in raising funds for borrowers, at least for the vast majority of the population. Finally, institutional investors, such as hedge funds or pension funds, that have been increasingly dominating the supply of capital on mature economy P2P lending platforms, operate on a mauch smaller scale in emerging markets and developing countries, if they do exist at all.

4.

The lack of financial development in terms of traditional institutions, banking and capital markets provides an extra hurdle for crowdfunding in developing countries. As investors have repeatedly made the experience that even banks are unable or unwilling to address asymmetric information problems and to enforce contracts, it is unlikely that they will invest money in a system where these problems are substantially more severe and mechanisms that might address these problems are untested. Thus, the view that underdeveloped banking sectors provide a rich opportunity for crowdfunding as the alternative investment option in form of intermediated finance is weak, is plausible on first glance only (Chart 16). This is even more the case as the general level of trust is lower in developing than in mature economies. Given that trust is a key component of social capital, crowdfunding is likely to face a much stronger uphill battle in developing countries than in mature economies (Kshetri 2015). 31

30

The microfinance industry, which emerged over the last fourty years, can be interpreted as an concentrated effort to generate this information either via group lending among borrowers or via unconventional individual lending between borrowers and microfinance institution (Armendáriz and Morduch 2010). 31 This does not imply that a lack of trust makes it impossible to fund opaque and small borrwers, including poor households, in these countries. Indeed, the microfinance industry shows that it is possible. However, it is possible because microfinance institutions extensively employ mechanisms to mitigate and overcome asymmetric information between the institution and borrowers. Thus, credit is not granted on tne basis of trust but on the basis of an appropriate credit technology. ,

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

26

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

Chart 16: Level of trust across countries

Source: World Value Survey, authors´ compilation.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

27

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES? The experience of kiva.org does not contradict, but rather confirms the claim that crowdfunding’s potential within developing countries is limited. First, funding via kiva.org is cross-border funding. Moreover, funds intermediated via kiva.org basically originate from mature economies (Burtch et al. 2014). Thus, kiva.org does not significantly contribute to financial development within emerging markets and developing countries. Second, these funders are guided by limited financial considerations only, as they do not earn any interest on their loans. While a zero interest rate has become a familiar experience when investing in high graded assets in mature economies, it clearly does not properly compensate for the risk in lending to MFIs in emerging markets and developing countries and even less for the risk of the final borrowers of those MFIs. Third, given the interest of kiva partner institutions to receive further funding at zero rates, they have a proper incentive to screen and monitor the borrowers. Thus, by including MFIs as intermediaries kiva.org addresses asymmetric information problems. Fourth, by carefully selecting partner institutions kiva.org can minimize credit risk. In doing so the platform can rely on the key achievement of microfinance, namely to demonstrate that micro businesses and poor households are creditworthy if served by proper lending technologies (Cull et al. 2008). Overall, the success of kiva.org – keeping in mind that intermediated volumes are small compared to the microfinance universe – does not build on crowdfunding as an instrument to lend funds to micro and small businesses, but it (partly) builds on the success of financial instution building in microfinance. It makes use of the success of MFIs by providing a global platform for lending / fundraising to those MFIs.

5.

POLICY CONCULSIONS We have shown that the potential for a significant contribution by crowdfunding to financial development and financial inclusion is limited. The policy conclusions we derive from this assessment are the following:

 At the current juncture crowdfunding does not (and should not) constitute a major field for development cooperation activities with regard to financial sector promotion. This does not mean that the development of the industry should not be monitored, in advanced as well as developing countries. As much as we have stressed the potential downsides of crowdfunding, in particular with regard to the possible implications of severe platform reputation losses, we have also indicated that many observers expect the industry to grow, at least in advanced economies and China. With more experience from mature economies, the current dismal state of affairs of crowdfunding in most developing countries might turn into a temporary phenomenon, even if we have presented a range of arguments suggesting that the short to medium-term prospects of the industry in developing countries are rather subdued. For example, successful platforms from mature economies – assuming that they will win the reputation game – might decide to expand their activities across borders. Thus, like mature economy retail banks in the past some platforms might establish themselves in emerging and developing countries, making use of their reputation advantage.

 Financial institution building remains the right approach, even from the perspective of a possible enlarged role for crowdfunding in developing countries. There are three reasons for this: a) a higher level of financial development with strong institutions does not hamper but raise the chances for crowdfunding to develop in these markets; b) strong financial institutions are the preferred intermediaries of the most successful crowdfunding platform currently employed in development finance, kiva.org; c) if crowdfunding were to follow the reintermediation process expected by several observers in mature economies, the link between the two sectors will become even stronger. For example, it might well be the case that crowdfunding – after internet penetration has achieved a critical mass – will be a useful mechanism to complement financial development via banks, providing a more sophisticated link between informal and formal finance than it exists in developing countries today.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

28

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

 If crowdfunding platforms in developing countries were to develop, or if serious proposals would be made to set up such platforms, for example with the goal of providing funds to small businesses, and if the founders were to ask for support from the development finance community, pilot projects involving donors, technical assistance providers and development finance institutions (DFIs) are conceivable. These projects would serve the purposeto learn about the concrete conditions in setting up such a platform in a developing county context. Concretely, DFIs could monitor, support and possibly enforce the use of mechanisms that limit moral hazard and adverse selection risks for foreign and local investors and thus enhance and safeguard the reputation of the platform. Indeed, such pilot projects might inform about the instrument’s capabilities in raising qualified loan demand. For example potential borrowers might become motivated to present their ideas according to standards set by the platform, as it opens a funding opportunity from domestic and international investors. Having said this, donors and DFIs should be aware of the fact that any engagement with crowdfunding platforms involves risks, notably reputational risks. Our policy conclusions are strictly limited to a crowdfunding assessment from a financial development perspective. Thus, under certain conditions crowdfunding may serve well as an instrument of fundraising for development or development finance purposes. However, it is unlikely to make a substantial contribution to financial development and inclusion in developing countries in the near and medium term.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

29

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

6.

REFERENCES Allison, T.H., Davis, B.C., Short, J.C., Webb, J.W., (2015). Crowdfunding in a Prosocial Microlending Environment: Examining the Role of Intrinsic Versus Extrinsic Cues. Entrepreneurship Theory and Practice, 39 (1), pp. 53-73. (DOI: 10.1111/etap.12108). Agrawal, A.K., Catalini, C., Goldfarb, A., (2013). Some Simple Economics of Crowdfunding. National Bureau of Economic Research, NBER Working Paper No. 19133. Akerlof, G.A., (1970). The Market for "Lemons": Qualita¬tive Uncertainty and the Market Mechanism, Quarterly Journal of Economics, 84 (3): pp. 488 – 500. Alloway, T., (2015). Big data: Credit where credit’s due, Financial Times, 4 February. Armendáriz de Aghion, B., Morduch, J., (2010) The Economics of Microfinance. Massachusetts Institute of Technology. Belleflamme, P., Lambert, T., Schwienbacher, A., (2013). Individual crowdfunding practices. Venture Capital, 15 (4): pp. 313-333. Blaseg, D. and M. Koetter (2015). Friend or foe? Crowdfunding versus credit when banks are stressed. Frankfurt School mimeo. Bodie, Z., Merton, R.C., (1995), A Conceptual Framework for Analyzing the Financial Environment, in: Crane, D. B. (ed.), The global financial system: a functional perspective, Boston, pp. 3 – 32. Bruett, T., (2007). Cows, Kiva, and Prosper.Com: How Disintermediation and the Internet are Changing Microfinance. Community Development Investment Review, 3 (2): pp. 44-50. Bruton, G., Khavul, S., Siegel, D., Wright, M., (2015). New Financial Alternatives in Seeding Entrepreneurship: Microfinance, Crowdfunding, and Peer-to-Peer Innovations. Entrepreneurship Theory and Practice, 39 (1), pp. 9-26. (DOI: 10.1111/etap.12143). Burtch, G., Ghose, A., Wattal, S., (2014). Cultural Differences and Geography as Determinants of Online Prosocial Lending. MIS Quarterly, 38 (3): pp. 773-794. Calomiris, C.W. and Gorton, G. (1991). The Origins of Banking Panics, In R.G. Hubbard (ed.). Financial Markets and Financial Crises, Chicago, pp. 109-173, The University of Chicago Press. Caprio, G., D. Vittas (1997). Financial history: Lessons of the past for reformers of the present, in: Caprio, G. and D. Vittas (eds.), Reforming financial systems: historical implications for policy, Cambridge, New York and Melbourne, pp. 1-21. Chen, R., Chen, Y., Liu, Y., Mei, Q., (2015). Does team competition increase pro-social lending? Evidence from online microfinance. Games and Economic Behavior. (DOI: 10.1016/j.geb.2015.02.001). Dellarocas, C., (2010). Online Reputation Systems: How to design one that does what you need. MIT Sloan Management Review, 51 (3). Čihák, M., Demirgüç-Kunt, A., Feyen, E., Levine, R., (2012). Benchmarking Financial Development Around the World. Policy Research Working Paper 6175, World Bank, Washington, DC. Claessens, S., Van Horen, N., (2012). Foreign Banks: Trends, Impact and Financial Stability. IMF Working Paper No. 12/10. Colombo, M.G., Franzoni, C., Rossi-Lamastra, C., (2015). Internal Social Capital and the Attraction of Early Contributions in Crowdfunding. Entrepreneurship Theory and Practice, 39 (1): pp. 75-100. Conti, A., Thursby, M., Rothaermel, F.T., (2013). Show Me the Right Stuff: Signals for High-Tech Startups. Journal of Economics & Management Strategy, 22 (2): pp. 341-364. Cull, R., Demirguc-Kunt, A., Morduch, J., (2008). Microfinance Meets the Market, World Bank Policy Research Working Paper No. 4630, Washington DC. Cull, R., Peria, M.S.M., (2010). Foreign Bank Participation in Developing Countries – What Do We Know about the Drivers and Consequences of This Phenomenon? World Bank Policy Research Working Paper No. 5398, Washington DC.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

30

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

Cumming, D., Johan, S., (2013). Demand-driven securities regulation: evidence from crowdfunding. Venture Capital, 15 (4): pp. 361-379. De Beukelaer, C., (2014). Creative industries in “developing” countries: Questioning country classifications in the UNCTAD creative economy reports, Cultural Trends, 23 (4): pp. 232-251. (DOI: 10.1080/09548963.2014.912043). De la Torre, A., Feyen, E., Ize, A., (2011). Financial Development – Structure and Dynamics. World Bank Policy Research Working Paper No. 5854, Washington DC. Demyanyk, Y., Kolliner, D., (2014). Peer-to-Peer Lending is Poised to Grow, Federal Reserve Bank fo Cleveland, https://www.clevelandfed.org/en/Newsroom%20and%20Events/Publications/Economic%20Trends/2014/Peer-to-Peer%20Lending%20is%20Poised%20to%20Grow.aspx, accessed 26 April 2015. Dewatripont, M., Tirole, J., (1994). The prudential regulation of banks, MIT Press. Diamond D.W., Dybvig P.H., (1983). Bank runs, deposit insurance, and liquidity. Journal of Political Economy, 91(3): pp. 401–419. Diamond, D.W., (1984). Financial Intermediation and Delegated Monitoring. The Review of Economic Studies, 51(3): pp. 393-414. Diamond, D.W., (1989). Reputation Acquisition in Debt Markets, Journal of Political Economy, 97: pp. 828-862. Diamond, D.W., (1991). Monitoring and Reputation: The Choice between Bank Loans and Directly Placed Debt, Journal of Political Economy, 99: pp. 689-721. Diaz-Alejandro, C., (1985). Good-Bye Financial Repression, Hello Financial Crash, in: Journal of Development Economics, 19: pp. 1- 24. Dorff, M.B., (2014). The Siren Call of Equity Crowdfunding. Journal of Corporation Law, 39 (3): pp. 493-524. Emekter, R., Tu, Y., Jirasakuldech, B., Lu, M., (2015). Evaluating credit risk and loan performance in online Peer-to-Peer (P2P) lending. Applied Economics, 47 (1): pp. 54-70. Federal Ministry for Economic Cooperation and Development (2004). Sectoral Policy Paper on Financial System Development, Bonn. Fitch Ratings, (? Sheehy, B., Flanders, N.) (2014). Peer-to-Peer Lending. Global Industry Overview, Special Report. Financial Instituions U.S.A.. Freedman, S., Jin, G.Z., (2014). The Information Value of Online Social Networks: Lessons from Peerto-Peer Lending. National Bureau of Economic Research, NBER Working Paper No. 19820. (DOI: 10.3386/w19820). Galak, J., Small, D., Stephan, A.T., (2011). Microfinance Decision Making: A Field Study of Prosocial Lending. Journal of Marketing Research, XLVIII (Special Issue 2011): pp. 130-137. GFK Verein, (2014). Trust in Professions 2014. Goodhart, C.A.E. (1988). The Evolution of Central Banks. Cambridge, London: The MIT Press. Gorton, G. and Mullineaux, D.J. (1987). The Joint Production of Confidence: Endogenous Regulation and Nineteenth Century Commercial-Bank Clearinghouses, Journal of Money, Credit, and Banking, 19 (4): pp. 457-468. Gorton, G., (2008). The Panic of 2007. National Bureau of Economic Research, NBER Working Paper No. 14358. (DOI: 10.3386/w14358). Gramlich, E.M., (2004). Subprime Mortgage Lending: Benefits, Costs, and Challenges, Remarks at the Financial Services Roundtable Annual Housing Policy Meeting, Chicago, Illinois, May 21, 2004. Grant Thornton (2014). Alternative Lending: A Regulatory Approach to Peer-to-Peer Lending. http://www.grant-thornton.co.uk/Documents/financial-services/Alternative-Lending-regulatoryapproach-to-Peer-to-Peer-lending.pdf.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

31

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

Greenbaum, S.I. and A.J. Thakor (2007) Contemporary Financial Intermediation, 2nd ed., April 2007, Butterworth Heinemann; Elsevier Science. Greenspan, A., (1997). Consumer Credit and Financial Modernization, Remarks at the Economic Development Conference of the Greenlining Institute, San Francisco, California. Guinnane, T.W., (2002). Delegated Monitors, Large and Small: Germany's Banking System, 1800-1914, Journal of Economic Literature, 40 (1): pp. 73-124. Harrison, R., (2013). Crowdfunding and the revitalisation of the early stage risk capital market: catalyst or chimera? Venture Capital, 15 (4): pp. 283-287. Helms, B., (2006). Access for all, Building Inclusive Financial Systems, The World Bank, Washington DC. Hemer, J., (2011). A snapshot on crowdfunding, Working papers firms and region, No. R2/2011. Herzenstein, M., Dholakia, U.M., Andrews, R.L., (2011). Strategic Herding Behavior in Peer-to-Peer Loan Auctions. Journal of Interactive Marketing, 25: pp. 27-36. Klein, E., (1982). Deutsche Bankengeschichte – Band I, Von den Anfängen bis zum Ende des alten Reiches (1806). Frankfurt am Main. Klein, M.C., (2015). What will P2P lending look like 5 years from now? Financial Times, 19 February. th

Krahnen, J.-P., R.H. Schmidt (1994). Development Finance as Institution Building, Westview Press, Boulder et. al. Krauss, N., & Walter, I. (2009). Can microfinance reduce portfolio volatility? Economic Development and Cultural Change, 58 (1): pp. 85–110. Kshetri, N., (2015). Success of Crowd-based Online Technology in Fundraising: An Institutional Perspective. Journal of International Management, 21 : pp. 100-116. La Porta, R., Lopez-De-Silanes, F., Shleifer, A., (2002). Government Ownership of Banks, Journal of Finance, 57 (1): pp. 265 - 301. La Porta, R., Shleifer, A., (2014). Informality and Development. The Journal of Economic Perspectives, 28 (3), pp. 109-126. Lamoreaux, N.R., (1986). Banks, Kinship and Economic Development: The New England Case, in: Journal of Economic History, XLVI (3): pp. 647-667. Lamoreaux, N.R., (1991). Information Problems and Banks‘ Specialization in Short-Term Commercial Lending: New England in the Nineteenth Century, in: Temin, P. (ed.), Inside the Business Enterprise: Historical Perspectives on the Use of Information, Chicago, pp. 154-195. Lee, E., Lee, B., (2012). Herding behavior in online P2P lending: An empirical investigation. Electronic Commerce Research and Applications, 11: pp. 495-503. Lehner, O.M., (2013). Crowdfunding social ventures: a model and research agenda. Venture Capital, 15 (4): pp. 289-311. Lin, M., Prabhala, N.R., Viswanathan, S., (2013). Judging Borrowers by the Company They Keep: Friendship Networks and Information Asymmetry in Online Peer-to-Peer Lending. Management Science, 59 (1): pp. 17-35. Lu, Y., Gu, B., Ye, Q., Sheng, Z., (2012). Social Influence and Defaults in Peer-to-Peer Lending Networks. Conference on Information Systems and Technology, (CIST 2012) White Paper. Ly, P., Mason, G., (2012a). Competition Between Microfinance NGOs: Evidence from Kiva. World Development, 40 (3): pp. 643-655. (DOI: 10.1016/j.worlddev.2011.09.009) Ly, P., Mason, G., (2012b). Individual Preferences Over Development Projects: Evidence from Microlending on Kiva. Voluntas, 23 (4): pp. 1036-1055. (DOI: 10.1007/s11266-011-9255) Mach, T.L., Carter, C.M., Slattery, C.R., (2014). Peer-to-peer lending to small businesses. Finance and Economics Discussion Series Division of Research & Statistics and Monetary Affairs, Federal Reserve Board.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

32

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

Madestam, A., (2014). Informal finance: A theory of moneylenders. Journal of Development Economics, 107: pp. 157-174. Massolution, (2012). 2012CF The Crowdfunding Industry Report. Massolution, (2015). 2015CF The Crowdfunding Industry Report. Mayer, T. (2015). Der Untergang der klassischen Banken. Frankfurter Allgemeine Zeitung, 4th April. McKinnon, R.I. (1973), Money and Capital in Economic Development, Washington DC. Morse, A., (2015). Peer-to-Peer Crowdfunding: Information and the Potential for Disruption in Consumer Lending. National Bureau of Economics Research, NBER Working Paper No. 20899. (DOI: 10.3386/w20899). Moss, T.W., Neubaum, D.O., Meyskens, M., (2015). The Effect of Virtuous and Entrepreneurial Orientations on Microfinance Lending and Repayment: A Signaling Theory Perspective. Entrepreneurship Theory and Practice, 39 (1): pp. 27-52. (DOI: 10.1111/etap.12110). Newlyn, W.T., Rowan, D. C., (1955). Money and Banking in British Colonial Africa, Blackwell Publishing, pp. 72-153. Nosko, C., Tadelis, S., (2015). The Limits of Reputation in Platform Markets: An Empirical Analysis and Field Experiment. National Bureau of Economic Research, NBER Working Paper No. 20830. (DOI: 10.3386/w20830) Pitschner, S., Pitschner-Finn, S., (2014). Non-profit differentials in crowd-based financing: Evidence from 50,000 campaigns. Economics Letters, 123: pp. 391-394. Porta, R.L., Shleifer, A., (2014). Informality and Development. The Journal of Economic Perspectives, 28(3): 109-126. PWC, (2015). Peer pressure: How peer-to-peer lending platforms are transforming the consumer lending industry. http://www.pwc.com/en_US/us/consumer-finance/publications/assets/peer-topeer-lending.pdf. Roodman, D. (2009). Kiva Is Not Quite What It Seems, Center for Global Development, http://www.cgdev.org/blog/kiva-not-quite-what-it-seems, accessed 30 April 2015. Rutherford, S. (2000), The Poor and Their Money, Oxford, Oxford University Press. Schmidt, R.H., Seibel, H.D. and Thomes, P., (2015). From Microfinance to Inclusive Finance, Sparkassen and Cooperative Banks - Evidence from Local Banking Networks, Wiley, forthcoming. Stiglitz, J.E. and A. Weiss (1981). Credit Rationing in Markets with Imperfect Information, in: AER, 81 (3): pp. 393 – 410. Stiglitz, J. E., Jaramillo-Vallejo, J., and Park Y.C. (1993). The Role of the State in Financial Markets. World Bank Research Observer, Annual Conference on Development Economics Supplement, pp. 19-52. Sviokla, J., (2009). Forget Citibank – Borrow from Bob. Harvard Business Review, February, p. 25. Sylla, R., (1998). U.S. Securities Markets and the Banking System, 1790-1840, Review – Federal Reserve Bank of St.Louis, pp. 83-98. Terberger, E., (2007). Banken oder Kapitalmärkte: Welche Struktur des Finanzsystems ist geeignet für Entwicklungsländer? In: Hemmer, H.-R. (ed.), Zur Wirksamkeitsdebatte in der Entwicklungszusammenarbeit (EZ), Tagungsband, 10. Entwicklungsökonomisches Seminar vom 17.-20. Januar, http://verein-eff.de/resources/tagungsband_2007.pdf, accessed 27 April 2015. th

The Department of the Treasury (1991). Modernizing the Financial System, Washington DC. Wang, T.C. (2008). “Paying Back to Borrow More: Reputation and Bank Credit Access in Early America.” Exploration in Economic History, 45: pp. 477-488. World Bank (2013). Crowdfunding´s Potential for the Developing World. infoDev, Finance and Private Sector Development Department. Washington DC.

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

33

CROWDFUNDING – CAN IT PROVIDE A SIGNIFICANT CONTRIBUTION TO FINANCIAL DEVELOPMENT AND INCLUSION IN DEVELOPING COUNTRIES?

Annex

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

34

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

Annex 1: The popularity of crowdfunding

Source: authors’ compilation.

All listings are until the 31st December of the respective year. Google search options was used to set the respective dates (dates were set as follows: the starting date of the search was left blank in order to get the first possible date; the date till which the listings were requested was set to the 31st December of the respective year). Search Tag Year Search option "Also within the full text". Growth in %

2006

Search option "In the Titel". Growth in %

Search Tag Year Listings Growth

EBSCO - Listings for the Word "crowdfunding" from 2006 - 2015* "Crowdfunding" 2007 2008 2009 2010 2011 2012

2013

2014

2015*

1

4 400,00%

51 1275,00%

125 245,10%

392 313,60%

1.679 428,32%

10.641 633,77%

27.971 262,86%

58.137 207,85%

68.244 117,38%

0

0

4

15 375,00%

91 606,67%

356 391,21%

2.583 725,56%

6.789 262,83%

12.582 185,33%

14.441 114,78%

Financial Times - Listings for the Word "crowdfunding" from 2006 - 2015* "Crowdfunding" 2007 2008 2009 2010 2011 2012 0 0 4 5 13 51 125,00% 260,00% 392,31%

2013 179 350,98%

2014 335 187,15%

2015* 396 118,21%

2006 0

* Status 15th Apri l , 2015

Source: authors’ compilation

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

35

CROWDFUNDING – A CONTRIBUTION TO FINANCIAL DEVELOPMENT IN DEVELOPING COUNTRIES?

Annex 2: Distribution of crowdfunding platforms by model

Source: Massolution (2015), authors’ compilation

Frankfurt School - UNEP Collaborating Centre for Climate & Sustainable Energy Finance

36

FS-UNEP Collaborating Centre Frankfurt School of Finance & Management Sonnemannstrasse 9-11 60314 Frankfurt am Main http://fs-unep-centre.org www.frankfurt-school.de E-Mail: [email protected] Phone: +49 (0)69 154008-604 Fax: +49 (0)69 154008-670