The Emergent Field of P2P Finance

Though the internet has since its inception held theoretically the power to ‘disintermediate’ established firms only in the last year, or two, the p2p concept has inspired entrepreneurs to develop new financial services.

Whilst in the 90’s innovation in the banking sector meant opening up online banking facilities they did not challenge the underlying intermediaries: banks.
The OpenBusiness team has looked in detail at five financial experiments online and provides an overview and comparison.
Two of those comparisons provide new ways to borrow and loan. Zopa a UK company lets individuals act in some ways like banks as they can give individuals loans. Prosper – a US company – does the same, but ‘disintermediates’ even more.
In addition Kiva and Fundable both innovate non-profit finance in different ways by engaging individuals directly.
Sellaband.com acts as an aggregator to finance music production. All of these services experienced strong growth in the last couple of months. OpenBusiness reported in the past repeatedly about p2p finance – see here, and here.
Almost ten years after eBay revolutionized consumer to consumer eCommerce individuals seem to be prepared to engage in financial transactions on a p2p basis, whilst entrepreneurs have started to offer the platforms facilitating financial activity in between individuals.

Zopa

Zopa – the Zone Of Possible Agreement – puts lenders and borrowers directly in touch and promises better rates for both by cutting out the middleman; though it only decreases the role of the middleman. The site acts as a facilitator for transactions and makes sure debts are repaid.

Who: Zopa was set up by many of the team who launched Egg, and is backed by Benchmark Capital, (who backed eBay), Wellington Partners, Bessemer Venture Partners (the VC firm which backed Skype), Tim Draper and The Rowland Family.
Borrowers: People can borrow from £1000 to £15000 under contracts of £10 each, in multiples of £100. Borrowers must submit to being credit-checked by Equifax, and to having the results made publicly available on their profile. On the basis of this and other information collected during registration, they are assigned to one of four different markets (A*, A, B and C). In the less ‘credit-worthy’ markets, borrowers will pay significantly higher interest rates, to account for the statistically greater ‘bad debt’ rate. The ‘market rate’ for each of the four classes of borrower is determined globally and not on an individual basis.
The funds are then reserved and after more checks with credit reference agencies the loan is approved by Zopa. The money is then paid directly into the borrower’s bank account, often within a few days. Any defaults or late payments will affect a borrower’s credit rating, as with a standard bank loan.
Lenders: Zopa lenders first transfer the amount they wish to lend into their Zopa holding account. This is a segregated account which is operated by Zopa and specified as containing money owned by Zopa members.
Zopa lenders can lend any amount from £10 to upwards of £25,000. Offers are made in amounts of £10 to each borrower, although the highest number of contracts any one borrower can have with a single lender as a result of his or her successful bids is 20. To lend more than £25,000 lenders need to first hold a Consumer Credit Licence. Lenders can choose their rates and loan lengths, and whether they want to lend in the A*, A, B or C markets. Zopa provide information – including market data and expected levels of bad debt – to help lenders choose their terms. Lenders will also earn 4.75% interest while they’re waiting for their money to be lent out.
Zopa estimate that lenders should make a 6-7% return per year if all the money repaid is lent out again (an average bad debt of 4% is already taken into account). This is 1% to 2% higher than the current best savings account. Returns are approaching the level of long term stock market returns (which are around 8%) but are more predictable.

Community: Members who have not lent or borrowed with each other are only identified on the site by their nicknames. If you have lent money, you will find out the real names (but not any of their contact details) of your borrowers on the quarterly statement. If you have borrowed money, you will see the real names (but not any of the contact details) of your lenders on your loan contract note. There is not a great deal of engagement within Zopa on a community or social level, as the primary aim is to save lenders and borrowers money by ‘cutting out the middleman’, not to create a community of lenders and borrowers.
Typical transaction: Zopa is the most classical ‘financial instrument’ of the emergent P2P finance services, and the typical transactions reflect that: relatively anonymous, diversified holdings in widely ranging amounts.
Business model: Zopa makes money by charging lenders and borrowers a fee. It charges borrowers 0.5% of their loan amount and lenders a 0.5% annual service fee. It also earns money through selling payment protection insurance to borrowers who want it (they have a commission based deal with Pinnacle Insurance), and through introducing people who can’t pass Zopa’s credit checking regime to other loan providers (again on commission through ‘preferred’ suppliers).
Zopa has received credit licenses from the Office of Fair Trading. It is authorised and regulated by the Financial Services Authority only in respect of its insurance mediation activities, although it must also comply with the “Higher Level Standards” that apply to all firms authorised and regulated by the FSA.

Establishing trust: Everyone looking to borrow is credit-checked and risk-assessed by Equifax, and people judged not credit-worthy will be prevented from borrowing at Zopa. The rest are put into either the A*, A, B or C market. This allows borrowers to “get a rate that’s right for them”, and means lenders can manage their risk level. Lenders are encouraged to diversify risk by spreading money across a range of borrowers. When a person lends £500 or more, her money is spread across at least 50 borrowers. A collections agency chases missed payments on each lender’s behalf. Zopa’s model is close to that used by banks and other financial institutions.
In the event of a total business failure, the loan agreements still stand because Zopa is not a party to any loan contracts; it only provides the mechanism for agreeing them. The repayments will continue to be collected by a collections agency that is appointed by Zopa lenders to collect missed payments. The costs of collections activity will not vary if Zopa has failed.
Zopa also have a number of online and offline procedures to catch unusual or suspicious behaviour on the site and can immediately suspend the membership of anyone whose intentions do not look completely honourable. These procedures are not elaborated.

Performance: Zopa has been relatively successful, and has recently secured an additional $12.9m of investment to expand its business in the UK and to launch in the US.

Problems or limitations: The forced diversification and lack of meaningful contact between lender and borrower may mean that users could feel somewhat estranged from one another. Lenders know the ‘class’ of borrower (e.g. A*, B) but they cannot lend more than £200 to any one borrower. The bond is ultimately legal and not social — the devices for reclaiming money are drawn from the banking industry, such as risk evaluation and collection agencies. The potential social aspects of the service are not exploited to the same extent as they are in competing services (such as Prosper) since personal interaction is not of great importance to the model.

Prosper

The US-based Prosper bills itself as the ‘eBay of personal finance’. The service is similar in many ways to Zopa, but with somewhat more emphasis placed on personal interactions and community lending. People who need money request it, and other people bid for the privilege of lending it to them. Prosper aims to make sure everything is safe, fair and easy.

Who: Prosper’s CEO and co-founder, Chris Larsen, was formerly the CEO, Chairman and Founder of E-LOAN, an online consumer lender “dedicated to providing consumers with a fast, transparent, and low cost way to obtain mortgage, auto and home equity loans.”
Borrowers: Prospective borrowers register with the site and allow the company to review their credit history. Then they post a loan request of up to $25,000, along with an upper limit for the amount of interest they are willing to pay. Loans are not secured by collateral and are paid off over three years at a fixed rate, with no prepayment penalty. Once the bidding is complete, and if enough lenders bid enough money to finance the loan at a single rate acceptable to the borrower, Prosper transfers the money to the borrower’s account and establishes a monthly repayment system that withdraws money from the borrower’s checking account. Should a borrower default, Prosper hires a collection company on the lender’s behalf and alerts credit bureaus.
Lenders: All of Prosper’s loans are 3-year unsecured loans. People who want to lend set the minimum interest rate they are willing to earn and bid in increments of $50 to $25,000 on loan listings they select. People who lend can easily diversify using “standing orders”, which automatically make many small loans to different borrowers. Lenders essentially deposit their money with Prosper — which holds it in an interest-bearing account with Wells Fargo— and either review the loan requests individually or fill out a form permitting Prosper to allocate money to borrowers who meet certain criteria. Chief among those criteria is the borrower’s rating from the credit reporting bureau Experian, but borrowers can also join or create groups with defined interests or characteristics that, they hope, will make them more attractive to some lenders.
Community: Unlike Zopa, there is a greater emphasis on personally selecting and lending to particular borrowers. Loans can be fully funded by one person, so it is possible to lend an individual up to $25,000 (in Zopa the limit is just £200.) As a result there is much more interaction between lenders and borrowers. The ‘groups’ are an important part of this process, and make the service seem less like just another financial instrument.
Among the groups on Prosper are aficionados of the Porsche 914 model, associates and employees of a Berkeley cafe and Vietnamese-American students. Prosper’s group leaders receive a commission on the group’s lending and borrowing activities, which they sometimes share among the group. When you join a responsible group with a good payment history, you get a good reputation by association, and lenders are more likely to offer good interest rates. But, belonging to a good group puts some pressure on you, too. If you stop making your loan payments, you not only tarnish your own reputation, but the group’s as well. If you’re part of a group, the theory is that you’ll perform better as a borrower than if it was an impersonal bank or credit card company.
Typical transaction: Loans requested range widely between about $2000 and $25000. The average loan amount as of April 2007 is just shy of $5000. Most lenders lend between $50-200 to any one borrower, and the vast majority of lenders have a total portfolio size of less than $2000.

Business model: Prosper generates revenue by collecting a one-time 1% or 2% fee on funded loans from borrowers (depending on credit grade), and assessing a 0.5% or 1.0% annual loan servicing fee to lenders.

Establishing trust: Prosper obtains the borrower’s Experian credit score, and assigns one of seven credit grades, from AA to HR. Borrower credit grades are posted with their listing to help lenders plan their bidding. Further information is also provided, such as delinquencies, number of credit lines, debt-to-income ratio and debit or credit card utilisation. Additional data, which is self-reported by most borrowers when registering, includes occupation and income. This information combines with the personal and group interactions that Prosper enables, to give lenders and borrowers a credible sense of risk and trust in other people.

Performance: Prosper has 270,000 members and $61 million in loans to date. Their open API has spawned dozens of websites focused on Prosper users, and these communities and Prosper groups are very active. Prosper has raised approximately $20 million in investment for further expansion.

Problems or limitations: Like Zopa, Prosper’s contractual and enforcement structure is founded on specific territorial laws. Growing the network transnationally is impracticable due to regulatory considerations. Given the size of the US banking market this is not a fundamental problem, but it is necessarily a brake on growth, especially given the geographically dispersed nature of many online communities and affinity groups.

Sellaband

Sellaband.com is a music business based on profit sharing and bringing together artists and music lovers directly. Fans can become ‘believers’ in bands by paying $10 for a ‘part’ in an artist they like. Once an artist has reached the goal of $50,000 (5,000 parts), Sellaband uses this money to record a CD, providing the artist with a studio, an A&R manager and a producer. ‘Believers’ receive a copy of the CD and share in some of the profit of the band.

Who: Sellaband is the idea of the Dutch entrepreneur Pim Betist, and has been managed by Johan Vosmeijer, a former Sony BMG executive.

Musicians: The money reaches bands indirectly: once $50,000 is raised, the funds are held in a Sellaband account, as a budget for the band to spend ‘as they like’. Sellaband has a stable of A&R reps, recording studios and CD pressing plants, which the musicians are encouraged to utilise.
Believers: Investing in bands allows fans the chance to ‘be in business’ with bands they like. The more tangible return is a special edition of the CD that is sent to all believers once produced. Additionally, believers share in a one-thirds split of the advertising revenues from the website. The amount received depends on the market share their band has on Sellaband’s download portal. In addition, believers will share 50% of the net profits from so-called ‘regular versions’ of CDs that are sold at gigs.
Once a year the totals are added up and any returns are deposited in the believers’ Sellaband account. Users can withdraw money from this account or use the funds to buy parts in other musicians.
Community: Believers can maintain a blog, upload photos and videos and post messages to network with other believers and artists. These community aspects are limited to users who have bought at least one part. Musicians have similar options, and it seems to be a prerequisite of success for bands to engage heavily with the Sellaband community.
Typical transaction: Each part costs $10, with additional transaction costs that are proportionately reduced the more parts a ‘believer’ purchases. Many believers just buy one part in various artists, but some invest heavily, with some believers buying over $5,000 in parts in artists like Nemesea and Clemence (two artists who have succeeded in raising $50,000).

Business model: The advertising revenues generated are split evenly between the artist, the believers and Sellaband. The publishing income of the songs that artists record with Sellaband are divided between the artist, Sellaband, producer and A&R manager, in the following way: Artist 60%, Sellaband 30%, Producer 5%, A&R manager 5%. Sellaband doesn’t touch any of the $50,000 that is given.
Sellaband is for-profit, and registered in Germany as a company. Due to the structure of the returns, it is not subject to financial regulations.
Establishing trust: Sellaband retains controls of the funds pledged by believers and provides these as a ‘budget’ for bands to use as they see fit, within certain parameters. A sense of participation that is more than purely financial is provided, as bands are expected to blog and engage with their believers. However, believers have no input into the recording of artists’ music, and have no say over the end result.

Performance: Four bands have succeeded in raising $50,000 and recording a CD since the site’s launch in August 2006. There are also over 15 bands that have raised over $10,000.
Problems or limitations: The mechanism for splitting profits from CD sales is somewhat ill-defined. This is probably because of the need to avoid legal regulations, but could create problems as the service scales.
Another criticism is that the structure of the site tends — at this point at least — to favour bands with a mainstream sound. This might change as Sellaband grows and is able to connect niche musicians with enough believers of like-minded tastes. But the constraints that Sellaband puts on bands, by offering rather generic industry tools, studios and producers, might not appeal to more experimental or innovative artists who want to do their own production or who are purely electronic musicians.

Fundable

Fundable allows groups of people to pool funds to make purchases or raise money. Similar to online auctions, Fundable’s pages, called “group actions,” are created by people who use the service. Each group action gives a description of how much money needs to be collected and what it will do. Once enough pledges (not payments) have been collected, Fundable turns them into real payments and sends the total to the group action’s organizer.

Who: Louis Helm and John Pratt co-founded Fundable in 2005.

Group leaders: Group leaders may receive money they have collected either through their paypal account or (for a $10 flat fee) by cheque.

Community: Social transactions are at the heart of Fundable, and many users will be dealing exclusively from people drawn from their own social network.
Typical transaction: The majority of the group actions tend to be fundraisers, such as sponsored walks or pleas for charity. There are also people doing things like recording albums, and promising a CD to investors in the style of Sellaband.
Fundable sets no limits on how many people a group leader expects to recruit, and if the group leader permits, people can pledge more than the contribution amount. The value of the average pledge has risen since the service launched. The first completed pledges were for amounts between $50 and $1000 — with most in the range of $200-500. Recently completed group actions have been for greater amounts: the typical action raises around $1000, and rarely less than $500. The smallest individual pledges are typically for $10 or $20.
Business model: Fundable acts as an escrow between groups of people and the person who collects their money. It is for-profit, and makes money by charging an 8.9 percent fee for its services. No fee is charged if a pledge expires without reaching its goal, and it is free to register and to set up a group action.
Establishing trust: Fundable allows group leaders with an established online auction reputation to display a link to their eBay rating on their group actions.
In most cases, the weight of 20 or 25 waiting contributors creates considerable pressure for organisers to deliver on what they promise. However, as in eBay, people are expected to ask questions and verify the trustworthiness of the organizer before making a pledge.
Fundable holds pledges until the group action reaches its collection goal, and then collects payment. If a group action’s collection falls short of its target pledge amount on deadline, Fundable deletes the pledge and no money is collected. This lets users participate in a group purchase or fundraiser without worrying about what other people will do. No one pays until and unless everyone else makes a pledge. As a further measure to ensure trust in the system, Fundable reviews most group actions that seek over 40 contributors.
Performance: Fundable is still in its infancy, but there have been hundreds of successfully completed group actions, and the engagement and enthusiasm of users suggests great potential for growth.
Problems or limitations: If you are recruiting from your immediate social network, it is very hard to get more than 20-25 people involved in a group action. Collecting pledges from more people than this involves planning and effort, and if a group action seeks more than 40 participants, the organiser must have access to a large audience. This is a reflection of the difficulty of raising money within real-world social networks, an objective for which Fundable is nonetheless very helpful.
Where Fundable falls short is in providing series of feedback mechanisms between group actions and funders, and degrees of involvement between group leaders and contributors.

Kiva

Kiva lets individuals connect with and loan money to unique small businesses in the developing world. By choosing a business on Kiva.org, lenders can “sponsor a business and help the world’s working poor make great strides towards economic independence.” Throughout the course of the loan (usually 6-12 months), lenders receive email journal updates from the business they’ve sponsored. As loans are repaid, they get their loan money back.

Who: The team behind Kiva worked at TiVo, PayPay and Google before developing the service in late 2004. Kiva partners with many existing microfinance institutions (MFIs), who work on the ground in developing countries.
Borrowers: Before an entrepreneur appears on Kiva they have first been vetted by a Field Partner for loan application approval. Each of Kiva’s Field Partners use their own application procedure which Kiva has reviewed and approved. This ensures that loan funds are actually going to genuine entrepreneurs who will use the loan for the purpose they specified.
Kiva does not send loan funds directly to the entrepreneurs; instead, each loan is managed by a microfinance institution. The full value of the loan goes to the entrepreneur, but the Field Partners do charge interest. However, Kiva requires Field Partners to fully disclose their interest rates, and doesn’t partner with organisations that charge exorbitant rates. Kiva claims that allowing MFIs to charge interest enables them to bear transaction costs and currency risk, and achieve self-sustainability.Kiva is also the first organization PayPal is supporting by providing free payment processing, reducing the transaction costs significantly.

Lenders: Kiva’s loans do not provide a financial return on investment, but lenders do get the investment back at the end of the loan term of investment — which range from 6-12 months. There are no tax implications because there is no possibility of earning interest. For lenders, it’s a sustainable, high impact, high engagement way to get involved with just a little amount of money, and carries minimal financial risk.
Unless you’re extremely rich, an accredited investor, or a big institution, you can’t really invest in microfinance institutions, so lenders and charitable donors are limited to giving to organizations that already have a lot of money. The smaller organizations that are working hard to serve their communities and to loan money to people are typically capital-poor. Kiva allows individuals to access a long tail of organisations and individuals, which can potentially be more rewarding and efficient for lenders.

Community: Sponsorship has always been a high overhead business. Kiva creates a similar interpersonal connection at much lower costs due to the instant, inexpensive nature of internet delivery. The individual nature of Kiva’s loans and the feedback that lenders receive all the way through the course of their loan makes for more personal and social transactions.

Typical transaction: Varies widely between $200 – $1200. The majority of loans are a smaller range of $800 – £1000.

Business model: Kiva is a non-profit social venture, that currently has financial support from a number of angel investors, including Silicon Valley donors, and corporate sponsors including Microsoft Research. However, Kiva says “self-sustainability is critical and we intend to be fully self-sustainable by 2008. This will be achieved through the implementation of a number of income streams which may include optional transaction charges to lenders and low debt capital fees to Field Partners.”
Kiva’s loans are personal agreements between lender and borrower. There is no note or security involved.
Establishing trust: Kiva partners with existing microfinance institutions. In this way, they gain access to outstanding entrepreneurs from impoverished communities world-wide. The partners are experts in choosing qualified borrowers, and usually have many more promising projects than funds. Through Kiva, the partners upload their borrower profiles directly to the site so that users can lend to them.
Users receive emails throughout the loan term updating them on the progress of the business, and letting them know each time a repayment is made. Each project also has a business page on the website, with a journal which can be commented upon.
Loans are not guaranteed, but are statistically low-risk. Microfinance loans worldwide are generating repayment rates of 97%. To date, Kiva’s repayment rate is 100%.
Performance: Kiva has processed over $6 million in loans, distributed to more than 60,000 people. It is growing fast — in April alone over $1 million was loaned. The current repayment rate is 100% (although of course many loans have not come to term yet).

Problems or limitations: Kiva doesn’t fully cut out the middleman — it does bring investors and recipients closer together but there is another layer of interest-collecting MFIs on the ground. However this is not necessarily a drawback. Part of the aim of Kiva is to help not only borrowers but MFIs as well.

Conclusions:

Most striking are the differences in between p2p Finance Services in the kind and degree of mediation they provide between the investor and the investee.
Some models may not ‘disintermediate’ entirely, but rather exploit some characteristics of networked organisation while also retaining elements of hierarchy — e.g., in vetting or enforcement.
It seems, however, the potential power of P2P Finance will be best facilitated by a model that extends as much process as possible out to its users, minimising central control over functions and processes that have traditionally been carried out by expensive, centralised hierarchies.

Clearly, this goes to the question of the kinds of relationships set up within the finance system. In a system of loans such as Zopa and Prosper, enforceable agreements based on a clear risk proposition are necessary, and this implies a degree of centralisation necessary to establishing credit checks, hiring credit enforcement agencies and so forth.
However, as Zopa has shown, such functions too may be decentralised by outsourcing them to third parties.
In this regard, lessons can be learned from Prosper, who alongside their traditional enforcement systems have seen that encouraging borrowers to form affinity groups gives rise to self-enforcement effects derived from peer-pressure.
The obligation to service a debt may be more keenly felt if others whom the borrower regards as peers are both aware of the debt, and have activities that to some extent are affected by it.
Such a ‘peer pressure’, self-enforcement mechanism could allow a P2P finance system to lessen reliance on traditional instruments and instead develop a set of distributed, social
controls.
In addition, looking at Sellaband, it seems individuals are ready to engage in direct financial investment into musicians. It remains to be seen, if the above outlined ‘procedural innovations’ will be carried over to other areas such as ‘crowd-investment’ into other cultural products, or even, start ups, or the production of tangible goods.

Comments

Responses are currently closed, but you can trackback from your own site.