Fintech, Regulatory Arbitrage, and the Rise of Shadow Banks Greg Buchak, Gregor Matvos, Tomasz Piskorski and Amit Seru* This Version: SEPTEMBER 2017 Abstract We study the rise of shadow banks in the largest consumer loan market in the US. The market share of shadow banks in originating residential mortgages nearly doubled from 2007-2015. Shadow banks gained a larger market share among less creditworthy borrowers, with a significant share of loans being originated- to-distribute to GSEs. Difference in difference tests suggest that traditional banks contracted origination activity in markets in which they faced more capital and regulatory constraints; these gaps were partly filled by shadow banks. Shadow banks with predominately online mortgage application process, “fintech” lenders, accounted for roughly a quarter of shadow bank loan originations by 2015. Relative to non-fintech shadow banks, fintech lenders serve more creditworthy borrowers and are more active in the refinancing market. They appear to use different information in setting interest rates, consistent with a big data component of technology, and charge a convenience premium of 14-16 basis points. We use a simple model to decompose the relative contribution of technology and regulation to the rise of shadow banks. We interpret the variation in mortgage rates and market shares using the model and find that increasing regulatory burden faced by traditional banks and growth of financial technology can account, respectively, for about 70% and 30% of the recent shadow bank growth. Keywords: Fintech, Shadow Banks, Regulatory Arbitrage, Lending, Mortgages, FHA ______________________________ * Buchak is with the University of Chicago ([email protected]), Matvos is with the McCombs School of Business, University of Texas at Austin and NBER ([email protected]), Piskorski is with Columbia Graduate School of Business and NBER ([email protected]), and Seru is with Stanford GSB, the Hoover Institution, SIEPR and NBER ([email protected]). We thank Sumit Agarwal, Michael Cembalest, Stijn Claessens, John Cochrane, Darrell Duffie, Andreas Fuster, Holger Mueller, Chris Palmer, Thomas Philippon, Raghuram Rajan, Hyun Shin, Johannes Stroebel, Stijn Van Nieuwerburgh, Nancy Wallace, Luigi Zingales and seminar and conference participants at Bank of International Settlements, Chicago Fed, Kellogg, Hong Kong Monetary Authority, New York University, Stanford University, University of Wisconsin, NBER Summer Institute, FDIC 17th Annual Fall Research Conference, CFPB, and SITE conference on Financial Regulation for helpful comments. We thank Monica Clodius and Sam Liu for outstanding research assistance. First Version: November 2016. 1 I. Introduction In the last decade, the market for financial consumer products has undergone a dramatic change. Intermediation has shifted from traditional banks to less regulated shadow banks (Sunderam, 2015).1 This change has coincided with a shift away from “brick and mortar” originators to online intermediaries.2 Despite the scarcity of systematic evidence, regulators, policymakers, and academics have been engaged in an intense debate about the possible consequences of these developments.3 In this paper we undertake a first systematic examination of the evolution of shadow banking in the largest consumer loan market in the US, the ten trillion dollar consumer mortgage market. We study this market to explore the economic forces which could explain the drastic change in the nature of intermediation. We document that the market share of shadow banks in conforming mortgage origination has nearly doubled from roughly 30% to 50% from 2007-2015. 4 In the Federal Housing Administration (FHA) mortgage market, which serves less creditworthy borrowers, the change has been even more dramatic with market share of shadow banks increasing from 45% to 75% over the same period. Concurrently, “fintech” lenders, shadow banks with a predominately online mortgage application process, increased their market share rapidly, and accounted for roughly a quarter of shadow bank loan originations by 2015. Two leading classes of hypotheses have attempted to explain the decline in traditional banking: Increased regulatory burden on traditional banks, and disruptive technology. The idea behind the first explanation is that shadow banks exploit regulatory arbitrage. Banks are subject to an ever- increasing regulatory burden through heightened legal scrutiny and larger capital requirements. The increased burden has changed which products they can provide, and has increased the cost of their funding. Therefore, banks are withdrawing from markets with high regulatory costs. Shadow banks, facing substantially lower regulatory costs and related concerns, have stepped into this gap, giving rise to large gains in market share. The second hypothesis is that the shift from traditional banks is driven by changes in technology: Fintech shadow banks have gained market share because they provide better products, or because they provide existing products more cheaply, and their technology has disrupted the mortgage market. Consider Quicken Loans, which has grown to the third largest mortgage lender in 2015. 1 We use the term “shadow bank” to refer to non-bank (non-depository) lenders, consistent with the definition of the Financial Stability Board (FSB), whose members cover G20 national regulators, the International Monetary Fund, the World Bank, and the Bank of International Settlements. See also Adrian and Ashcraft (2016). 2 Goldman Sachs Report, March 3, 2015: “The Future of Finance: The Rise of the new Shadow Bank.” 3 Bank of International Settlements, 2017: “FinTech credit. Market structure, business models and financial stability implications.” http://www.bis.org/publ/cgfs_fsb1.pdf 4 See Figures 1-3. 2 The Quicken “Rocket Mortgage” application is done mostly online, resulting in substantial labor and office space savings for Quicken Loans. The “Push Button. Get Mortgage” approach is also a more convenient and faster way for internet savvy consumers to obtain binding rate quotes and electronically provide documentation.5 Additionally, fintech lenders may be better able to screen potential borrowers, leveraging alternative sources of information and the big data approaches inherent in technology based lending. To examine whether increased regulatory burden is a driving force behind the decline of traditional mortgage banking, first we compare lending of banks to all shadow banks, irrespective of their fintech affiliation. By 2015 shadow banks originated about 50% of loans in the conforming market and 75% of loans insured by the FHA. The FHA loans allow lower income and less creditworthy households to borrow money for the purchase of a home with as little as 3.5% of the property value as down payment. The large prevalence of shadow banks’ suggests that their advantage over traditional banks is especially strong in a market dominated by riskier borrowers. Traditional banks also lose market share in areas with larger minorities shares. Given that several enforcement actions and lawsuits had specifically targeted banks’ treatment of less creditworthy and minority borrowers, this evidence is consistent with shadow banks expanding in segments where regulatory burden has risen substantially. The differences between shadow and traditional banks are not limited to customer characteristics, but also extends to differences in financing of mortgages. Traditional banks’ share of mortgage originations held on their balance sheet has declined since 2007, but has not declined below 30%. Shadow banks, on the other hand, primarily finance mortgages using the originate-to-distribute model, especially later in the sample. Moreover, the securitization predominantly involves products associated with government sponsored enterprises (GSE). Next, we more directly link the rise of shadow banks to increased regulatory burden of traditional banks by focusing on three specific increases in this burden: increased capital requirements, increased regulatory burden related to mortgage servicing rights (MSR), and mortgage-related lawsuits, exploiting a differences in differences approach. We find a larger growth of shadow banks in counties in which a larger share of traditional banks had to build up capital reserves over the last decade. Additionally, regions with a larger share of lenders subject to legal actions, and a larger share of lenders involved in mortgage servicing activity also saw a larger growth in market share of shadow banks. This evidence is consistent with the idea that traditional banks are retreating from markets with a larger regulatory burden, and that shadow banks fill this gap. 5 https://www.nerdwallet.com/blog/mortgages/quickenloansandrocketmortgagereview/ [Accessed on 11/8/2016] 3 These results suggest that the lower regulatory burden provides a cost advantage to shadow banks. We next examine if these cost advantages are passed through to consumers: i.e., is the change we observe in the market limited to quantities, or is it also reflected in prices of mortgages? We find that differences in pricing, on average, are negligible. This average pattern hides interesting differences between loans originated by fintech and non-fintech lenders. Non-fintech lenders, which do not hold a technology advantage over traditional firms, do offer lower interest rates, suggesting that they pass some regulatory cost savings to customers. In addition, shadow banks loans are also more likely to prepay and default conditional on observables. Accounting for this higher risk further continues to suggests that, on average, shadow banks pass some regulatory cost savings to borrowers. Regulation is not the only possible reason why the market share of traditional banks may have declined over time. To assess the role that technology has played in the decline of traditional banking, we focus on technology differences between shadow banks. Doing so allows us to hold the regulatory differences between different lenders fixed. In particular, we collect information on a shadow bank’s online presence, to classify their lending operations as fintech or non-fintech. We then examine markets in which fintech lenders have grown faster than, and other ways in which fintech mortgages differ from their non-fintech counterparts. Fintech firms accounted for about a quarter of shadow bank loan originations by 2015. This simple fact suggests that on-line origination technology may have played an important role in the decline of traditional banks during the last decade. Fintech lenders serve a different segment of the mortgage market than non-fintech shadow banks. Fintech lenders are much less likely to serve less creditworthy FHA borrowers. Additionally, fintech originations are heavily tilted towards refinancing, differing from non-fintech shadow banks. One possible reason is that the more standardized tasks involved in mortgage refinancing are the best fit for fintech technology.6 Fintech lenders also differ from non-fintech shadow banks in loan pricing. As discussed earlier, non-fintech shadow banks offer lower interest rates than traditional banks, suggesting that they pass through a part of regulatory cost savings to customers. Fintech lenders, on the other hand, charge higher interest rates relative to traditional banks, consistent with the notion that fintech consumers are willing to pay for the convenience of transacting online. Finally, we explore whether there is indeed a difference in the technology used to set mortgage interest rates between fintech and non-fintech lenders. We find that standard variables for predicting interest rates, such as FICO and LTV, explain substantially less variation in interest rates of fintech lenders relative to non-fintech lenders. These results are consistent with the narrative that technology based lending 6 In refinancing, the fintech lender benefits from many on-the-ground activities having already taken place at the time of purchase, such as a title check, structural examination, negotiations between buyer and seller, 4 uses substantially different information, potentially based on big data they obtain in addition to standard variables. Taken together, our results suggest that two factors have contributed to the precipitous decline of traditional banks’ market share in the residential mortgage market over the last decade: Additional regulatory burden faced by banks, and technology related to on-line lending platforms. To decompose the relative contribution of technology and regulation to the rise of shadow banks we present and calibrate a simple quantitative model of mortgage origination. In the model, traditional banks, non-fintech shadow banks, and fintech shadow banks compete for borrowers. To capture the stylized facts that we document, these lenders differ on three dimensions: regulatory burden, convenience, which we model as a difference in quality, and potential differences in costs of making loans. Pricing, firm entry and markups are determined endogenously for each type of lender. We interpret the variation in mortgage rates and market shares using the model to identify the relative importance of different factors in the decline of traditional banking. Our estimates imply that traditional banks have slightly lower shadow cost of funding and provide higher quality products than shadow banks. Despite these advantages, they lose market share during this period because of large increase in the regulatory burden after 2010. This period coincides, among others, with passage of the Dodd-Frank Act, a formation of Consumer Financial Protection Bureau, and the new Basel III capital rules imposing more onerous limits on the amount of mortgage servicing payments that can count towards the bank regulatory capital. We also estimate a substantial increase in perceived quality and convenience of on-line origination platforms by borrowers that occurred during 2009-2012 period. Overall, using this simple quantitative model, we find that increasing regulatory burden can account for about 70% of shadow bank growth during 2008-2015 period with advancement in on-line lending technology accounting for another 30%. II. Related Literature Our paper ties together separate strands of the literature relating to residential mortgage lending, banking regulation, and the growing role of financial technology. The Structure of the Residential Mortgage Market Many papers have studied the changing structure of the mortgage origination chain, with particular attention paid to the originate-to-distribute model and the costs and benefits thereof (e.g., Berndt and Gupta 2009, Piskorski et al. 2010, Keys et al. 2010 and 2013, Purnanandam 2011). The focus has primarily been on the run-up to the financial crisis, rather than on the immediate aftermath and recovery following the crisis. 5 Bank-like activities taking place outside of traditional deposit-taking institutions have attracted considerable attention in the literature and at Federal banking regulators (see Adrian and Ashcraft (2016) for an exhaustive summary). The literature (e.g., Bord and Santos 2012) has primarily focused on the maturity transformation role of banks taking place outside of banks. Our paper instead focuses on mortgage origination taking place outside the traditional banking system, and its accompanying regulatory structure. In this regard our paper is also related to the recent literature investigating the industrial organization of the residential mortgage market (e.g., Stanton, Wallace, and Walden 2014, 2017). Banking Regulation and GSEs Our paper relates to a large literature has examined the role of government programs undertaken during the financial crisis. (e.g., Mayer et. al. 2014, Haughwout et. al. 2016, Agarwal et al. 2015 and 2017). Like Agarwal et. al. (2014), Lucca et. al. (2014), Granja et al. (2014), Piskorski et al (2015), Fligstein and Roehrkasse (2016), we study lawsuits arising out of the financial crisis and capital constraints. We make use of geographical heterogeneity in regulatory burdens to show that shadow banks, facing relatively lower regulatory pressure in heavily regulated markets, gain market share. Because shadow banks rely heavily on GSEs and FHA guarantees, our paper relates to literature studying GSEs and their role in mortgage lending. GSEs were established to promote housing ownership, particularly in underserved areas, and a number of papers (e.g., Elenev et al. 2016; Hurst et al 2015; Bhutta 2012; Acharya et. al. 2011) have studied their role in income redistribution and house ownership, finding mixed results. Our paper suggests that increased regulatory burden of traditional banks combined with GSEs and FHA guarantees may have contributed greatly to the rise of the shadow banking sector. Financial Technology Our paper connects to the growing literature on financial technology, e.g., Philippon (2015, 2016) and Greenwood and Scharfstein (2013). To our knowledge, ours is the first paper that performs a detailed analysis on fintech and non-fintech firms operating within the residential mortgage industry in an effort to explore what technological advantages fintech lenders have over non- fintech ones. Using a methodology similar to Rajan et al. (2015), we document that fintech lenders appear to use substantially different methods to set interest rates. Philippon (2015) documents that advances in financial technology have failed to reduce intermediation costs. In that spirit, our paper shows fintech lenders in fact offer higher interest rates than non-fintech lenders. However, consumers’ willingness to use more expensive fintech lenders may also reflect more convenient services offered by these lenders. 6 Finally, Philippon (2016) proposes that fintech can offer a way towards structural change in the financial industry, because political economy considerations can stifle change in the traditional part of the sector. Our paper advises caution: while fintech lenders do enter to help fill the gap left by the banks, they have done so by having relied almost exclusively on explicit and implicit government guarantees as customers. III. Data and Lender Classification III.A Description of Datasets We combine and use the following datasets in our paper. HMDA: We use mortgage application data collected under the Home Mortgage Disclosure Act (HMDA) to examine loan-level and area-level lending patterns. HMDA records the vast majority of home mortgage applications and approved loans in the United States. The data provides, among other things, the application outcome, the loan type and purpose, the borrower's race, income, loan amount, year, census tract, and importantly for our purpose, the originator’s identity. Due to mergers and name changes, the identification of HMDA lenders changes over time, and to overcome this limitation, we manually linked lenders across years. HMDA further records whether the originator retains the loan on balance sheet or sells the loan within one year to a third party, including to a GSE. If the originator retains a loan through the end of the calendar year before selling it, we would observe this as a non-sale. Fannie Mae Single-Family Loan Performance Data: This dataset provides origination and performance data on a subset of Fannie Mae’s 30-year, fully amortizing, full documentation, single-family, conforming fixed-rate mortgages that are the predominant conforming contract type in the US.7 This loan-level monthly panel data has detailed information on a rich array of loan, property, and borrower characteristics (e.g., interest rates, location of the property, borrower credit scores, LTV ratios) and monthly payment history (e.g., delinquent or not, prepaid). The loans in our data were acquired between January 1, 2000 and October 2015. The monthly performance data runs through June 2016. The Freddie Mac Single Family Loan-Level Dataset: Similar to the Fannie Mae data, this dataset contains a subset of loan-level origination, monthly loan performance, and actual loss data of fully amortizing, full documentation, single family mortgages. Included in the dataset are 30-year fixed- 7 The dataset does not include adjustable-rate mortgage loans, balloon mortgage loans, interest-only mortgage loans, mortgage loans with prepayment penalties, government-insured mortgage loans, Home Affordable Refinance Program (HARP) mortgage loans, Refi Plus™ mortgage loans, and non-standard mortgage loans. Also excluded are loans that do not reflect current underwriting guidelines, such as loans with originating LTV’s over 97%, and mortgage loans subject to long-term standby commitments, those sold with lender recourse or subject to other third-party risk-sharing arrangements, or were acquired by Fannie Mae on a negotiated bulk basis. 7 rate mortgages originating between January 1999 and September 2015 and purchased by Freddie Mac. Also included are 15- and 20-year fixed-rate mortgages originating between January 2005 and September 2015. The monthly loan performance data runs until March 2016 for all the loans provided.8 Combining the Fannie Mae and Freddie Mac datasets gives us coverage of the majority of conforming loans issued in the United Sates during the period of our study. The FHA Dataset: This data provided by the U.S. Department of Housing and Urban Development (HUD) contains single-family portfolio snapshots of loans insured by the FHA. The FHA program is intended to aid borrowers with particularly low credit scores who may otherwise be unable to borrow from conventional lenders. The data begins in February 2010, and is updated monthly through December 2016. The FHA data records product type (adjustable or fixed-rate), loan purpose (purchase or refinance), interest rate, state, county, MSA, and importantly for our purposes, the originating mortgagee. Notably absent from the FHA data are borrower FICO scores, so while by the nature of the program, FHA borrowers have low credit scores, we cannot directly control for borrower credit score within the FHA data. For this reason, when studying loan interest rates and outcomes, we focus our analysis primarily on the loans from Fannie Mae and Freddie Mac databases. US Census Data: We use county-level demographic data from the US Census and American Community Survey between 2006 and 2015. We collect population, population density, racial and ethnic characteristics, education, income and poverty, and homeownership statistics. Regulatory Burden of Depository Institution Data: In studying the market share of shadow banks we investigate whether shadow banks are likely to enter areas where the traditional banking system faces heightened regulatory scrutiny. We draw on a number of data sources to measure these regulatory burdens between 2006 and 2015. In particular, we use bank balance sheet data from the bank call reports, from which we calculate bank capitalization. Lawsuit Settlements Data: Finally, following Piskorski et al. (2015) and Fligstein and Roehrkasse (2016), we collect lawsuit settlements arising out of the financial crisis brought against banks, lenders, and mortgage servicers. We construct a timeline of settlements and settlement amounts by year and bank by aggregating data from a number of sources. From Law3609, a news service that covers all aspects of litigation, we collect data on lawsuit settlements associated with RMBS, mortgage foreclosures, fraud, deceptive lending, securitization, refinancing, and robo-signing. The Law360 data spans 2008 through 2016. From the SEC, we collected all legal actions taken by the 8 Not included are ARMs, balloon loans, mortgages with step rates, relief reliance mortgages, government-insured mortgages, affordable loan mortgages such as Home Possible® Mortgages, mortgages delivered to Freddie Mac under alternate agreements, mortgages associated with Mortgage Revenue Bonds, and mortgages with credit enhancements other than primary mortgage insurance. 9 https://www.law360.com/faq 8 SEC regarding misconduct that led to, or arose from the financial crisis. 10 The SEC data spans 2009 through 2016. From SNL Financial, now a part of S&P Global Intelligence, we collect a timeline of major bank settlements arising out of the financial crisis between 2011 and 2015. 11 III.B Lender Classification Central to this paper is the classification of mortgage lenders as banks or shadow banks, and within shadow banks, as fintech or non-fintech. We perform this classification manually. The Fannie Mae, Freddie Mac, and FHA data identify each loan’s originator if the originator was among the top-50 originators in the reporting period. HMDA identifies all originators. We classify the identified lenders in the Fannie Mae, Freddie Mac, and FHA data. Additionally, we classify the largest lenders in HMDA that are not identified in the Fannie, Freddie, or FHA data, so that our classified sample covers 80% of total originations by value in 2010. Robustness with respect to lender classification is discussed in Section IX. The classification of “bank” versus “shadow bank” when doing so is straightforward: a lender is a “bank” if it is a depository institution; a lender is a “shadow bank” if it is not a bank. This definition of banks is consistent with the definition of the FSB, which defines banks as “All deposit-taking corporations” and shadow banks as “credit intermediation involving entities and activities outside of the regular banking system.”12 Because our focus is mortgage origination, our measurement of shadow banking falls squarely within the FSB definition. FSB members comprise both national regulators of G20 countries, as well as international financial institutions, such as the International Monetary Fund, the World Bank, and the Bank of International Settlements, as well as, and international standard-setting and other bodies such as Basel Committee on Banking Supervision. Therefore our measurement of shadow banks has broad regulatory agreement. The classification of fintech and non-fintech is less straightforward: we manually classify a lender as a fintech lender if it has a strong online presence and if nearly all of the mortgage application process takes place online with no human involvement from the lender. For example, an applicant to Quicken Loans, the prototypical fintech lender, can be approved for a loan with a locked-in interest rate with no human interaction; the borrower meets a Quicken Loans loan officer for the first time only at closing (see Appendix A5). An applicant at a non-fintech firm, on the other hand, interacts with a human loan officer much earlier in the process, even if the process begins online. For instance, a borrower may input her name and location online, and then be directed to phone a local loan officer to continue. A lender using this process is classified as a non-fintech lender. 10 https://www.sec.gov/spotlight/enf-actions-fc.shtml 11 https://www.snl.com/InteractiveX/Article.aspx?id=33431645 12 http://www.fsb.org/wp-content/uploads/global-shadow-banking-monitoring-report-2015.pdf 9 Appendix A1 shows the list of main lenders in each of these three categories. Appendix A8 provides more details on the classification process. IV. Institutional Background A. Banks, Shadow Banks and Fintech This section provides an overview of the institutional details and history of shadow banking before and after the financial crisis. We use the term shadow banking broadly to refer to non-bank financial intermediaries that engage in activities which have traditionally been the business of banks.13 The key difference between shadow and traditional banks is that shadow banks do not take deposits, which frees them from a large amount of regulatory oversight directed at traditional banks. B. History of Shadow Banking in the Retail Mortgage Market Although this paper focuses on the rise of shadow banking in mortgage origination after the crisis and the factors that contribute to the rise, it is important to note that in the run-up to the financial crisis, shadow banks’ share of mortgage origination was quite high. Goldman Sachs estimates that among the top 20 lenders, shadow banks originated roughly 30% of all mortgages for the years 2004—2006 and mostly specialized in loans issued without government guarantees (e.g., non- agency subprime loans). The market share of shadow bank lenders was heavily concentrated. Countrywide Financial accounted for more than half of the shadow banks’ share of originations.14 Shadow bank originators do not take deposits. Instead, they rely almost exclusively on making loans that are originated for sale, and earn revenue through the sale of mortgage servicing rights (MSR)—the capitalized value of future cash flows from the mortgages, a small amount of interest income between origination and sale, and servicing income.15 Because shadow banks rely so heavily on sale of mortgages to third parties, they are particularly sensitive to the financial health of these third parties. The potential buyer depends on the originated product: conforming loans are typically sold to Fannie Mae and Freddie Mac. Government-insured loans, such as FHA or VA mortgages, are typically sold to Ginnie Mae. Non-conforming loans, such as jumbo or subprime mortgages, were typically securitized into non-agency MBS, although after the crisis, the secondary market for most jumbo mortgages essentially vanished. As we document, traditional banks are also a purchaser of shadow bank mortgages. 13 GS Report, Pg. 5 14 GS Report, Pg. 51 15 GS Report, Pg 51. 10