Comments on the Department of the Treasury Notice: Public Input on Expanding Access to Credit Through Online Marketplace Lending

Submitted electronically

Laura Temel
Marketplace Lending RFI
U.S. Department of the Treasury
1500 Pennsylvania Ave., NW., Room 1325
Washington, DC 20220

Via www.regulations.gov

Dear Ms. Temel:

The Milken Institute Center for Financial Markets would like to thank you for the opportunity to provide information about marketplace lending in response to the Treasury Department’s Request for Information (RFI).

The Milken Institute is a nonprofit, nonpartisan think tank determined to increase global prosperity by advancing collaborative solutions that widen access to capital, create jobs, and improve health. The Center for Financial Markets promotes financial-market understanding and works to expand access to capital, strengthen and deepen financial markets, and develop innovative financial solutions to the most pressing global challenges.

Marketplace lending represents a potentially promising development in small business and personal finance. In the wake of the recent financial crisis, bank credit has become more difficult and expensive to obtain for many customers. For example, traditional lending to small businesses has declined by as much as twenty percent compared to before the crisis. Personal credit also contracted, with the Federal Reserve Bank of Cleveland reporting that the total amount of money lent by banks as consumer finance loans declined by an average of two percent per quarter starting in the second quarter of 2007 to 2014, with money lent via credit-card declining by. percent over the same period. This reduction in credit access was driven by multiple factors, including economic uncertainty, regulatory concerns, tightening of loan criteria, industry consolidation and the fact that smaller loans were less profitable than larger ones, while requiring similar effort.

Marketplace lenders4 are leveraging innovative technological, financial, and legal structures to expand access to credit –- providing many borrowers with superior rates and terms,5 and allowing them to refinance traditional loans with lower terms.6 Morgan Stanley estimates that marketplace lenders originated approximately $7 billion dollars of consumer loans in 2014. They are also increasingly playing a role in providing funding to businesses, with approximately $4.6 billion dollars in loans originated in 2014,8 while simultaneously allowing investors to deploy their capital in profitable and socially productive ways.

While the promise of marketplace lenders is potentially significant, there are also risks. The innovations that make marketplace lending effective could also lead to problems for customers and regulators, since unlike traditional finance they are generally less well understood and untested in a variety of economic conditions. Likewise, there is the threat that current regulatory regimes, primarily implemented prior to the inception of marketplace lending, will show themselves to be obsolete and ill-suited to protecting customers or enabling innovation and competition that benefits everyone.

As such, creating a regulatory environment that provides for appropriate protection and robust competition is essential. To do this the following actions should be pursued:

• amend the relevant banking laws to codify the “valid when made” doctrine;

• develop a uniform and consistent national regulatory regime for non-bank lenders that fosters innovation and competition while addressing the risks posed by the lending of non-depository funds;

• provide lenders with guidance and appropriate safe harbors to avoid both inadvertent discrimination and meritless litigation; and

• evaluate whether risk retention as required under the Dodd-Frank Act is appropriate for marketplace lending given marketplace lenders’ competitive environment and current lack of systemic risk.

What follows is information responsive to certain questions asked in the RFI. Please do not hesitate to contact us if you have any questions.

Responses to the Request for Information
A. Market segmentation and its implication for policy and regulation

• The method of funding the loan and its disposition impact the type of investors who have exposure to the loans and the legal regimes the investment must comply with.

• The choice of pursuing a bank partnership to leverage banks’ preemption of certain state laws or direct engagement by the lender with state requirements can drive the structure of the loans, the interest rates marketplace lenders can potentially charge, and the compliance burden of the lenders.

“Marketplace lending” is a broad term that is used to describe multiple methods of innovative lending. While there are similarities common to most marketplace lending platforms, such as the use of the internet to advertise to potential borrowers and take loan applications, there are also significant differences that may be material from a policy or regulatory perspective. While the market could be segmented many different ways the methods marketplace lenders use to fund and sell, or retain, the loan and whether the marketplace lender pursues a bank partnership or state-by-state model are particularly relevant.

Funding and Retention or Disposition of the Loan

The four primary methods marketplace lenders use to fund loans are: 1. Fractionalization of loans; 2. Traditional securitizations of loans; 3. The sale of whole loans; and 4. “Balance sheet lending.” Marketplace lenders may use one or a combination of these methods (a hybrid model), with each method presenting its own unique advantages and challenges.

1. Fractionalization – Securitization is the creation of a security based on the performance of an underlying asset, in this case a loan, which allows marketplace lenders to obtain investment funds and move the credit risk of a loan off of the lenders’ books (at least partially).

Some marketplace lenders who focus on loans to consumers and allow the general public to invest use an innovative means of securitization called fractionalization. In fractionalization, the marketplace lender allows potential investors to select which loans they want to support and pledge to contribute a portion of the total loan amount (sometimes as little as $25) to that specific loan. After enough support had been pledged to fund a loan, the marketplace lender has its partnering bank issue a loan to the borrower. The marketplace lender then purchases the entire loan from the bank and issues debt securities (notes), which are tied to the specific loan’s performance, to the investor, consistent with the amount invested. This can occur using a registered offering of securities under a shelf-registration as permitted by Securities Act Rule 415, with the marketplace lender providing supplements to its shelf prospectus with anonymized details of each loan and borrower, which allows the general public to invest. These notes generally do not provide the investor with any recourse against the borrower or lender, or ownership interest in the marketplace lender. Similarly, marketplace lenders will also establish trusts that acquire loans and offer “certificates” through investment advisors to accredited investors in private placement transactions. These certificates are similar to the notes in that their performance is tied to underlying loans and the investor generally has no recourse in event of non- payment to the assets of the lender or trust. However, certificates charge an asset-based management fee rather than a servicing fee like the notes.

2. Traditional Securitization - Marketplace loans are also used in more traditional securitizations by large institutional investors. The lender will sell loans to an institutional investor, who may take a pool of existing loans and create a special purpose vehicle that issues debt securities with the loans as collateral. The securities vary as to the riskiness of the collateral and the rate of return. The marketplace lender is normally responsible for the servicing of the underlying loans and charges a fee to do so.

Marketplace lenders may use proprietary ratings to indicate the perceived riskiness of loans, though it is increasingly common for securitizations to be rated by rating agencies. To improve the rating of the securities, the securitizer may utilize credit enhancement, including establishing a reserve fund or extra collateral to diminish the risk of loss to investors and improve the credit rating. Because securitizations rely on the marketplace lender to service the loans, investors are exposed to the risk that the marketplace lender will cease operations, or otherwise be unable to continue to service the loans. As such, securitization agreements usually have contingency provisions to provide backup servicing of the loans. Additionally, as discussed in greater detail below, under Dodd-Frank, marketplace lenders will likely be required to retain some portion of the securitization as a form of risk retention.

3. Sale of whole loans – In response to the demand of “qualified” investors, primarily large institutions like banks and hedge-funds, many marketplace lenders have begun to sell entire loans. The sale of whole loans differs from securitization in that, unlike securitization, the actual loan is transferred to a new party, rather than the marketplace lender keeping the loan and selling a new interest connected to the loan – though the lender generally continues to service the loan and charge the purchaser a fee for doing so. As such, the transaction does not require the creation of a new security tied to the loan and, depending on the details, may therefore mean that the transaction would not fall under the purview of securities law, as is the case, for example, with traditional bank sales of originated loans. In some cases, these whole loan sales can be used by institutional buyers to fill the pools used for the securitizations described above.

The sale of whole loans offers several advantages for both the marketplace lender and investors. First, as previously mentioned, whole loan sales do not require fractionalization of the loans and, under certain circumstances, may not constitute a security at all, potentially leading to a lower compliance burden. Secondly, the sale of whole loans can simplify administration of the loan since it minimizes the number of parties involved. Finally, it provides an efficient means for investors to purchase and manage larger amounts of debt from borrowers they consider to be good risks. As appetite for marketplace loans has increased among institutions, whole loans have become a larger and larger part of funding. For example, in the fourth quarter of 2014, of the $1.4 billion dollars of loans originated by Lending Club, $800 million were funded through whole loan sales.

4. Balance sheet lending – Balance sheet lenders make loans from company funds and hold the loans on the company’s books, where they receive the proceeds from the borrower paying their loans. Investors invest in the marketplace lender directly, and receive either an equity interest in the lender or become a creditor of the lender, where repayment is based on the general financial health of the lender as opposed to the performance of a specific loan. Unlike securitizations and whole loan sales, balance sheet lenders retain all of their loan portfolio’s credit risk.

Unlike traditional bank lending, which uses customer deposits to fund loans, marketplace lenders use investor or company funds exclusively, and do not serve as a depository institution. This means that unlike banks, whose customers frequently deposit funds for safe-keeping and ease-of-access, the entities that fund marketplace loans do so with the intention of receiving a significant return and the understanding that their funds are at risk of loss.10 This difference argues against using bank regulations as a default for marketplace lenders since banks and marketplace lenders offer different services and pose different risks. While regulations relating to lending (e.g. fair lending and disclosure laws) may be applicable to both marketplace lenders and bank lenders, regulations relating to safety and soundness may not be.

Even among marketplace lenders the differences have significant policy implications. The method of funding can dictate what body of laws and level of scrutiny are appropriate. For example, under certain circumstances the sale of whole loans may not constitute the sale of a security, while the sale of fractional interests in a loan or group of loans does. Likewise, the nature of the investor can matter, with the sale of interests to the general public potentially justifying different consumer protections than the sale to large institutional buyers.

Licensing
Lending is a regulated activity at both the federal and state level. Marketplace lenders therefore need to ensure that their loans are made in compliance with relevant regulations. There are generally two paths that marketplace lenders take: partnering with a bank or pursuing state-by-state licensing.

1. Bank partnership - Some marketplace lenders partner with a bank to do the actual origination of the loan, which the platform then purchases and services. The marketplace lender will identify, underwrite, and approve the borrower and source of capital for the loan. If the bank approves, it will then create the loan, which is purchased by the marketplace lender. The marketplace lender services the loan going forward. In certain “whole loan” transactions, the loan purchaser finds the borrower through the marketplace lender’s marketplace and the bank then originates and sells the loan directly to the purchaser with the marketplace lender providing loan servicing for a fee. The lender and the bank each have responsibility for compliance with anti-money laundering (AML) and “know your customer” (KYC) regulations and similar requirements.

A bank partnership offers several important advantages. First, it allows a lender to move to market very quickly because it can rely on the bank’s charter and the ability to operate across state lines that banks enjoy under federal law, rather than taking the time to obtain licenses from each state the lender wishes to do business in. Secondly, under federal law, banks are able to “export” the rate of interest they can charge under the law of the state they are located in for loans it makes nationwide.11 This allows lenders to avoid having to comply with usury laws on a state-by-state basis, which would increase compliance costs and could prevent the marketplace lender from charging a profitable rate of interest. This arrangement also allows the lender to compete with traditional bank products under similar terms and circumstances. This arrangement may be particularly appropriate for marketplace lenders focused on issuing large numbers of relatively small and risky loans nationally -- the type of product where scale, consistency, and the ability to charge a relatively high interest rate are most important.

2. State requirements - The other primary path for marketplace lenders is to comply with the rules and requirements, including obtaining state lender licenses where necessary, of the states the marketplace lender intends to extend credit to. While this method may be more time-consuming and costly than the bank partnership method, and may prevent the lender from charging as high an interest rate in certain states because they are limited to the maximum allowed under that state’s law, as opposed to exporting the interest rate of the bank partner’s home state, the method is less controversial with state regulators and may be more legally stable in light of recent legal challenges such as Madden v. Midland Funding.

Whether a marketplace lender partners with a bank or pursues state-by-state compliance (some lenders do both) can impact the lender’s speed to market, capabilities, and the structure and participants of the loan. For example, lenders who partner with a bank have to structure the loan process to be routed through the bank. These structural differences may create different legal rights and risks that are relevant for regulators to consider.

B. The opportunities and risks of marketplace lenders’ use of data

• The reliance on new and proprietary data, and the self-modifying nature of some marketplace lenders’ algorithms may increase the risk of inadvertent discrimination.

• The online collection and storage of data, as well as the breadth and robustness of the data sets compiled on borrowers and investors, may create a risk of the data being compromised and of harm to borrowers and lenders if the data is compromised.

• The relatively new and untested nature of marketplace lenders pricing models, and the lack of experience with a variety of economic and interest rate environments may create a risk of false confidence that could lead to greater than expected losses if circumstances change.

Along with the use of investment of company -- instead of depository -- funds, the innovative use of data distinguishes marketplace lenders from more traditional lenders. Many marketplace lenders utilize proprietary algorithms, datasets that include factors not traditionally included in lending decisions (like social media presence), and machine learning (algorithms that self-modify to become more effective) to inform lending decisions. Lenders believe that this use of data and technology enables them to make decisions more efficiently and accurately than traditional methods – allowing the lender to provide the borrower with a decision more quickly (even if the decision is “no”), “score” borrowers who do not have sufficient history to be accurately evaluated under traditional systems, and more correctly price risk, oftentimes at reduced costs to the borrower when compared to traditional bank lending.

While each marketplace lender’s process is different, there is a general move towards greater automation and data accumulation compared with more traditional lending processes. These trends pose potential risks of inadvertent discrimination, data security breaches, and failure to properly anticipate and price risk.

Inadvertent Discrimination
 Federal laws prohibit basing lending decisions on certain prohibited criteria. There are two broad types of discrimination – discriminatory intent, where the lender chooses to discriminate, and disparate impact,12 where the lender does not intend to discriminate but its choices disproportionately impact protected classes of people (e.g., a lender does not take race into consideration but its use of certain facially neutral criteria results in a disproportionate number of applicants from racial minority groups being denied credit).

While there is absolutely no evidence of intentional discrimination, the use of machine learning may create a risk where, unbeknownst to the lender, the algorithm begins using criteria that serve as proxies for protected classes that result in a disparate impact in the lender’s decision making. While the lender would want to avoid this, without intensive monitoring of the algorithm the disparate impact may only become apparent in retrospect as a pattern develops, therefore making it difficult for lenders to intercede before many borrowers are affected and the lender is exposed to potential liability. While this problem exists in traditional lending as well, the selection of non-traditional data sets and partial automation of decision making (both the credit decision, and by virtue of machine learning, the “recipe” for the credit evaluation) may exacerbate the risk.

Data security
Marketplace lenders strive to accumulate a large amount of data on potential borrowers and lenders, both because it is required by law (e.g. KYC requirements) and because it helps them to better evaluate overall credit risk. This information is primarily obtained, directly or in conjunction with other services,13 over the internet and stored by the lender. As such, the data obtained by the lender is information rich, joining and correlating data across the life of the borrower, as well as others like the borrower’s family members or employees. As such, the information is highly sensitive where any breach could potentially imperil both the company and the people behind the company.

False Certainty
While marketplace lenders are confident their use of proprietary algorithms help them price risk more effectively, the industry is still fairly young and many within the industry have not yet experienced going into an economic downturn or a more “normal” interest rate environment. It is unclear whether the models used by marketplace lenders are well suited to handling these changes, or whether deficiencies have been masked by general economic improvement since the crisis and abnormally cheap credit. While these concerns are by no means limited to marketplace lenders, their reliance on proprietary data and algorithms may make them uniquely vulnerable to losses if their models fail to accurately price risks.

While the use of data to expand access and improve costs is one of the most promising aspects of marketplace lending, it is not without risk. As such, while government regulation should not stifle or curtail innovative uses of data, regulators should be aware of the potential problems.

C. What role can government play in facilitating positive innovation

• Creating clarity around the long standing “valid when made” doctrine by amending the relevant banking laws will avoid potentially destructive ambiguity and vindicate the reasonable expectations of marketplace lenders and their customers.

• Establishing a uniform and specific regulatory regime for marketplace lenders would allow marketplace lenders to compete on an even playing field and avoid unnecessarily burdensome, complex, and artificial structures while allowing regulators to have appropriate tools and jurisdiction to protect customers.

• Providing guidance and safe harbors to marketplace lenders regarding their use of data and algorithms to price risk so they can avoid both inadvertent discrimination and needless litigation would help protect borrowers from both a lack of access and unnecessarily increased costs.

Lending is a highly regulated activity, and while market forces are driving marketplace lenders to constantly innovate to better serve customers, government activity will necessarily impact the trajectory of the industry. The best thing government policy can do to facilitate innovation is to provide clear, accessible, and minimally distortive regulation, which will allow lenders and consumers to understand their rights and obligations, protect borrowers and investors, and limit deadweight costs created by artificialities that lenders may resort to in order to comply with outmoded regulations. Specifically, the government should address issues surrounding licensing for marketplace lenders to provide them with an appropriate and uniform licensing regime and provide guidance to help marketplace lenders avoid disparate impact issues.

Licensing
As mentioned above there are two primary models for licensing marketplace lenders: partnering with a bank or complying with state-by-state requirements, including licensure where needed. While both of these methods work, they each have problems, and reflect a regulatory regime that is outdated and inapt for the economic reality of marketplace lending.

An important distinction between marketplace lending and more traditional sources of lending is that marketplace lending is a creature of the internet, unlike banks and other traditional lending that began before the internet, usually with a physical branch network and an inherent geographic limitation to their range of service. While many firms in these industries developed into national players – and the regulatory environment adapted to reflect and encourage this – many other firms remain local due to limited capability and interest in extending their reach. This is not the case with marketplace lenders who are inherently internet-based businesses. As such, they can provide their services with equal capability nationwide, and enjoy significant economies of scale due to lenders’ ability to access lending capital more cheaply as they grow.

The primary limit to a marketplace lender’s reach is regulatory, with some states requiring licensing to operate within their boundaries unless there is a federal preemption. To address this limitation, marketplace lenders generally either have to get licenses on a state-by-state basis, increasing compliance cost and complexity, or partner with a bank to leverage the bank’s ability to preempt certain state laws. Both of the current options are suboptimal because they introduce needless costs and complication that harm borrowers and investors.

Marketplace lenders are inherently engaged in interstate commerce. Congress therefore could, consistent with the Constitution, preempt state regulation and create a uniform federal system that would fit the unique needs of marketplace lending, encourage competition and innovation, and protect consumers. Such a regime would also negate the need for arrangements that only serve a compliance purpose, rather than an economic function.

Moving marketplace lending licensing to a more rational and efficient posture will require both a short- and long-term solution. First, as mentioned previously, the bank-partnership model has been cast into doubt by recent litigation. This poses a threat to marketplace lending, as well as other common lending activities, and should be addressed. Secondly, a national lending charter should be developed to provide the uniform and efficient regulation necessary to protect consumers and increase innovative competition.

1. Bank partnership - As mentioned previously, the Madden v. Midland Funding case, decided in May 2015 by the U.S. Second Circuit, held that non-banks who purchase debt from nationally-chartered banks cannot take advantage of the banks’ ability to export their home state interest rate limitations to other states and are instead limited by the laws of the borrower’s state of residence.14 This ruling calls into question the validity of the bank partnership model because marketplace lenders using the model are not banks but do purchase the debt from the bank and rely on their ability to receive interest income at the rate set by the bank when originating the loan.

The court reached this conclusion by looking at the National Bank Act (NBA) and did not consider the “valid when made” doctrine, a longstanding common law doctrine which holds that a loan that is not usurious when it is made cannot become usurious later by virtue of the loan being sold to another party.15 Likewise, the court in Madden did not reach an opinion on whether a choice of law provision in the lending contract, which would dictate that the contract was to be governed under the law of the bank’s home state, would take precedence over state usury laws, and remanded the issue to the trial court. The defendants are seeking review by the U.S. Supreme Court.

While the Supreme Court may take the case, the trial court may find that the choice of law provision controls (a decision which will likely be appealed regardless of how the court decides), or the Second Circuit may limit the holding in subsequent cases. The ruling creates ambiguity that will take a considerable amount of time to resolve through litigation, and calls into question the validity of significant numbers of outstanding marketplace and traditional loans and securitizations.

To address this, Congress and the Executive can amend the NBA, FDIA, FCUA, and HOLA to explicitly incorporate the valid when made doctrine. This would not change the rules retroactively, since the doctrine is already part of the common law, nor would it subvert the reasonable expectations of both borrowers and lenders. Instead, such a change would merely eliminate any uncertainty about the validity of this debt.

While failure to address the problem raised by Madden would probably not kill marketplace lending, it might force lenders and their partners to create new and more complex structures that, while technically compliant, were even less related to the economic reality of the transaction and instantly subject to litigation. Lenders could also pursue state licenses, but this would be costly, time consuming, and potentially difficult for new entrants relative to the bank-partnership option. A legislative fix would quickly provide needed clarity and consistency to the market.

2. National lender charter - While addressing the problem raised in Madden is an important short- term step, it does not address the underlying problem that current non-bank lending regulation has not kept up with technological and business innovation. While marketplace lenders could apply for bank charters and become banks, a bank charter is excessive and unnecessary if marketplace lenders want to focus exclusively on lending. A new uniform regulatory regime that allows marketplace lenders to compete in the lending market on a level playing field with banks (and vice-versa), encourages innovation and entry into the market, and calibrates the scope and nature of regulation to the actual risks posed by marketplace lending would allow marketplace lenders to avoid artificial or inefficient structures, provide them with clarity, and allow for effective regulation of the growing field.

While this change would require significant work the following general framework would be a good place to start the discussion.
Under charter a marketplace lender could:

• Make loans directly to consumers and businesses.

• Advertise loans and solicit borrowers.

• Set interest rates and terms for their loans that could be transferred to loan purchasers or securitizers with interest rate export powers similar to banks.

• Service debt.

• Initiate collections.

• Seek investors and obtain investment to fund loans. National lenders could not:

• Take deposits.

• Access the Federal Reserve discount window.

• Offer depository services such as checking accounts, savings accounts, debit or prepaid card services, or ATMs. Though lenders can partner with banks to provide these services.

• Pay interest on investor money except as part of an active loan or investment in the lender itself (i.e., there would be no brokerage “cash account” equivalent).

•  Represent itself as a bank or as a place offering deposit or safeguarding services.

• Engage in investment banking except for its own securities (both securities tied to specific loans or
  pools of loans and those of the lender itself).

National lenders would be responsible for:

• Performing necessary checks such as KYC and AML on borrowers and investors. (They may use outside vendors for this but retain responsibility).

• Complying with fair lending laws.

• Complying with relevant privacy laws.

• Maintaining client relationships.

• Providing appropriate disclosures to both borrowers and investors.

The appropriate primary regulator for national lenders would most likely be one of the bank chartering regulators. While the chartering regulator should create a regulatory regime that protects borrowers and investors it should develop the regime based on the nature of the lenders and their industry, rather than just analogizing to banks as a default. For example, while banks use large amounts of leverage and maintain deposits, which justifies a capital retention requirement, marketplace lenders do not. Therefore, a capital retention requirement may not be as justified or relevant. Rather, to protect borrowers and lenders it would be more sensible to require that a lender have backup trustees in place to handle servicing in the event that the lender ceases operations. In addition to the chartering regulator, the Securities and Exchange Commission (SEC), Federal Trade Commission (FTC), and in the case of consumer loans, the Consumer Financial Protection Bureau (CFPB), would regulate the portions of marketplace lending that intersect with their jurisdictions.

Because marketplace lending is such a rapidly innovating field regulators should avoid overly proscriptive regulation that favors or requires particular models. Rather, regulators should favor principles based regulation that focuses on an end result of transparency and protection while allowing market forces and consumer choice to drive specific implementations.16 The regime should also include scalable disclosures to the regulator based on the lenders size or age while disclosures to borrowers and investors should be consistent for all lenders. This would allow smaller and newer lenders to begin operations (where they will have minimal activity and information to disclose) without facing a crushing regulatory burden. Then, as marketplace lenders grow and accrue more activity and information, they can provide regulators with more data.

Disparate Impact
 As discussed above, marketplace lenders’ reliance on data, proprietary credit models, and machine learning may create a risk of inadvertently making lending decisions that have a disparate impact. To minimize this risk, the government should provide lenders with guidance and safe harbors regarding the use of data to help them avoid inadvertent discrimination while allowing them to confidently pursue valid, accurate, and innovative risk pricing models. Additionally, government regulators should work to build their internal technological competences so they can better understand and appreciate how the lenders operate. This will allow them to both better determine when intervention is appropriate and when it is not, minimizing needless litigation.

D. The role of risk retention

• Risk retention requirements are a powerful tool to align incentives between the originator and purchaser of a debt and are best suited for cases where other incentives are insufficient or unavailable, or where there is significant systemic risk.

• Marketplace lending operates in a highly competitive environment with strong market feedback signals. It also poses very limited systemic risk.

• Risk retention requirements could force some marketplace lenders to fundamentally change their business model in a way that would disadvantage their customers for little benefit.

Risk retention requirements are used to align the incentives of an originator of a debt with the purchaser of the debt and avoid scenarios where an originator may be tempted to compromise underwriting standards or exploit information asymmetries with the buyer to generate profit while transitioning all of the risk to the buyer. This dynamic in the broader asset-backed securities (ABS) market is considered to be a major cause of the recent financial crisis which led to section 15G of the Securities Exchange Act, as added by section 941 of the Dodd-Frank Act. This provision requires securitizers of certain ABS and the originators of certain self-liquidating collateral that forms the basis of ABS to retain a portion of the risk.

While ABS based on marketplace loans are likely covered by Dodd-Frank it is unclear whether risk retention is a good fit for marketplace lending. Marketplace lending is a highly competitive space where participants face pressure from traditional lenders, other existing marketplace firms, and new entrants. Marketplace lenders also need to convince investors to provide capital on a constant basis. Additionally, unlike banks, which maintain multiple lines of business that can provide revenue, marketplace lenders tend to be lenders exclusively. As such, if a lender develops a reputation for underestimating risk it will face prompt and significant market discipline as investors move to alternatives.

Marketplace lending’s lack of structural significance also weakens the justification for risk retention. First, marketplace lending is currently relatively small compared to traditional lending, with approximately $12 billion17 in loans originated in 2014 compared to approximately $430 billion in loans originated by banks and other FDIC insured institutions over the same period.18 Secondly, unlike banks, marketplace lenders use only risk (as opposed to depository) capital and are generally at very low or no leverage. As such, marketplace lenders pose a limited threat that any financial reversals will extend beyond investors to broader systemic problems.

Additionally, risk retention will force a change in business model that may not provide sufficient benefit to outweigh the cost. In particular, whole loan sales should remain exempt from risk-retention requirements. In addition to the reasons discussed above, requiring sellers of whole loans to retain risk would force a fundamental and unnecessary change to their business plan. To comply with risk retention, the whole loan would need to be fractionalized so that a portion could be retained. This would increase the cost and complexity of the transaction and require compliance with the securities laws. This would transform the nature of the transaction and disadvantage investors who prefer whole loan sales to more traditional securitizations.

Risk retention is a powerful and expensive tool that can be used to address serious potential problems. However, in cases where other market or regulatory options are available to appropriately mitigate risk, those should generally be considered the first option, especially when risk retention will force artificial changes to the structure of transactions for limited benefit.

Conclusion
Marketplace lending is a potentially promising option for consumer and small business credit. While this innovation presents an opportunity, it is not without risk. Thoughtful, tailored regulation to help mitigate those risks without subjecting the industry to outdated, inappropriate, or overly-onerous rules can help marketplace lending reach its potential. We hope that the Treasury Department will consider this information, as well as the comments from other stakeholders, and act, or refrain from acting, as appropriate to enable the proper regulatory environment to develop.

Thank you again for the opportunity to provide this information. If we can be of any further assistance please do not hesitate to contact us.

Respectfully,

Brian R. Knight
Associate Director, Financial Policy, Center for Financial Markets Milken Institute
[email protected]