THE REGULATION OF MORTGAGE SERVICING: LESSONS FROM THE FINANCIAL CRISIS.
Literature on the appropriate regulation of mortgage servicers is scant, even though the aftermath of the 2007 to 2009 U.S. financial crisis revealed serious issues with U.S. servicers. Furthermore, the Financial Crisis Inquiry Commission (2010) did not deal at all with mortgage servicing, even though it was created by Congress to identify the causes of the crisis. We fill this major gap in the literature by considering the lessons to be learned about regulation of mortgage servicing from the crisis. A pivotal case against the largest subprime servicer in the country in the precrisis period, Fairbanks Capital Corporation (FB), provides information allowing such an investigation.
In a world where mortgage assets are securitized, a well-functioning mortgage servicing industry is essential to the health of the financial system. Servicers collect payments from homeowners, keep records of mortgage balances, pool the payments and remit them to investors, and manage escrow accounts. Investors in mortgage-backed securities (MBS) expect that the servicer of the underlying loans will pool the monthly payments, transfer interest and principal to investors, pay taxes and insurance, and keep accurate records of borrowers' balances so the investor can report the amount of its mortgage-related assets accurately. Nonetheless, the financial crisis and the foreclosure crisis that followed exposed major weaknesses, affecting even basic functions such as keeping accurate records of mortgage balances (Morgenson 2008a, 2011; Rudolph 2009). In October 2010, all 50 state attorneys general (SAGs) launched an investigation into the practices of the major mortgage servicers; there was also an ongoing federal investigation involving the Department of Justice and bank regulatory agencies resulting in a $26 billion settlement in 2012. (1) The 2010 Dodd-Frank financial reform legislation prompted commercial banks to reduce their mortgage servicing assets (e.g., Rexrode 2017). Nonbank specialty servicers acquired these servicing rights at a pace that appears in some cases to be faster than their ability to handle the increased volume. Hence, significant problems with mortgage servicing continued (e.g., Silver-Greenberg and Corkery 2014).
We show that these servicing problems have historical roots in the precrisis period and consider the implications for financial regulatory policy. We examine both the history and the wealth effects of the FB mortgage servicing litigation initiated in 2002, early in the housing boom. The litigation was settled very quickly at nominal cost to the servicer with the active involvement of federal regulatory agencies. The parties entered into the initial settlement agreement in November 2003; it was approved by a federal court 6 months later. The settlement provided the servicer a release of liability despite the large volume of complaints about its practices. The settlement, and this release of liability, provided major benefits to FB but allowed serious weaknesses in servicing and consumer protection to continue uncorrected. Regulators and the servicer then entered into a new agreement in 2007 that simply restated many of the provisions of the earlier agreement, a step that would have been completely unnecessary if the servicer had complied with the 2003 agreement.
FB, an approved Fannie Mae-Freddie Mac servicer, was rapidly expanding in the early 2000s by purchasing subprime servicing from other financial institutions such as Citigroup. FB was the target of class action lawsuits in more than ten major states, as well as in over 1,000 individual lawsuits (e.g., Eggert 2004). Allegations included assessing fictitious late fees and other fees, failing to keep accurate records of mortgage balances, misapplication of borrowers' funds, and other practices. The two federal agencies responsible for mortgage servicers were the Department of Housing and Urban Development (HUD) and the Federal Trade Commission (FTC). The PMI Group, which was traded on the New York Stock Exchange (NYSE) and was one of the nation's largest private mortgage insurance companies, owned 56.8% of FB's common stock on December 31, 2002. PMI acknowledged risks from the FB litigation in its Annual 10 K Reports.
We test the hypothesis that the settlement benefited PMI. Examining the stock price reaction of PMI to the settlement, as well as the contagion effects of the settlement on a major subprime lender, allows us to understand the impact of the litigation and settlement on shareholder wealth for major players in the subprime mortgage market. We estimate the wealth effects of different announcements from an initial litigation event in July 2002 up to approval by the court in May 2004. The cumulative abnormal returns (CARs) over the period are at least 7.59% using traditional event study methods, but range from 5% to 20% depending on which announcements are considered unanticipated and hence more informative (e.g., Eckel, Eckel, and Singal 1997). These results are consistent with our hypothesis that the $40 million settlement--distributed among a very large number of borrowers--benefited PMI. The cost was nominal, and 1,145 borrowers opted out of the settlement, an indication that informed borrowers considered the amounts offered inadequate in relation to their damages.
As the country's largest subprime servicer, FB was essential to the growth of the secondary market in subprime loans and subprime MBS. FB's website described the company as "a residential mortgage servicing company specializing in the underperforming mortgage market," a company that had "expertise in the effective management and resolution of sub-performing mortgage loans" (Cooks v. Fairbanks, Class Action Complaint).
Just as servicing allowed the secondary market in prime loans to develop over a very long period of time (e.g., Brown et al. 1992; Van Drunen and McConnell 1988), the subprime servicing industry was essential for the development of the secondary market in subprime mortgage loans, and FB was the leading firm in this industry. The primary market in both prime and sub-prime loans derives liquidity from the ability of lenders to sell loans in the secondary market, where they are generally packaged into MBS. Put simply, the subprime MBS market could not have developed without a subprime servicing industry, since securitization requires servicing. Our analysis illustrates the inadequate response of federal agencies to subprime mortgage market abuses in the period before this market virtually collapsed in 2007 to 2008. The collapse of the subprime MBS market triggered the broader collapse of other financial markets, and may have been the primary cause of the subsequent U.S. financial crisis and the Great Recession that followed.
The major subprime lender we consider in our contagion analysis is Countrywide. As the largest home lender in the United States before the crisis,
Countrywide is often considered one of the worst high-risk lenders; it nearly imploded during the crisis before being acquired by Bank of America (BOA). (2) Evidence of contagion effects suggests such lenders benefited from this settlement. Countrywide's stock price reaction to the first settlement-related announcement is 6.71%.
This paper contributes to the literature in three ways. First, while extensive literature exists on the economic consequences for shareholders of corporate litigation, none of these studies deal with mortgage servicing litigation. Second, our analysis fills the gap left by the Financial Crisis Inquiry Commission by showing how servicing was regulated in the precrisis period and how the failure to correct abuses contributed to the crisis. Third, we consider the appropriate regulatory structure for mortgage servicers. The 2010 Dodd-Frank Wall Street Reform and Consumer Protection Act transferred responsibility for regulating mortgage servicers to the newly created Consumer Financial Protection Bureau (CFPB). Campbell et al. (2011) and Campbell (2016) argue that there is a role for government in consumer protection because of asymmetric information and market power. Nonetheless, the agency has been the subject of controversy about its regulatory approach since its inception (e.g., Hayashi 2016). If the CFPB were abolished, as many of its critics in Congress have advocated, responsibility for regulating mortgage servicers would have to be placed somewhere. Policymakers should consider the history of mortgage servicing regulation, including that presented here. We analyze an important precrisis period when federal responsibility for regulating mortgage servicers was split between two federal agencies. With this type of arrangement it is possible that neither agency would consider it a priority to develop as much expertise in the industry as it might have if it had been the sole regulator; hence the activity would not get the attention it deserves.
The importance of these issues is underscored in testimony by then Federal Reserve Chairman Ben Bernanke before the Financial Crisis Inquiry Commission. Bernanke testified that the failure to regulate mortgage lending practices was "the most severe failure of the Fed in this particular episode" (Reddy 2010). Bernanke's comments might reasonably be extended to apply to mortgage servicing. Grind (2012) presents an analysis of the Washington Mutual failure which also identifies regulation that is split between two agencies as one of the causes of the largest bank failure in U.S. history.
II. RELATED LITERATURE
Lawsuits increase the probability of financial distress for the defendant, reflecting loss of reputation and increased likelihood of bankruptcy (Bhagat, Brickley, and Coles 1994; Cutler and Summers 1988; Engleman and Cornell 1988). The combined market value of the two firms declines as defendants lose more than plaintiffs gain at announcement. Settlement events increase the combined value of the firms. Settlements have a positive valuation effect for both parties as the cost of further litigation is removed and the defendant's reputation is somewhat restored. Importantly, incentives to settle differ substantially when a government agency initiates a lawsuit (Bhagat, Bizjak, and Coles 1998). Corporate settlements follow the net present value rule, but government agencies sue to establish a precedent and could choose not to settle. Settlement occurs for completely different reasons than private lawsuits. The studies in this area that we are aware of do not deal with mortgage servicing litigation.
The subprime mortgage market and high-risk mortgage lending grew substantially after 2000 (Gramlich 2007), and the quality of sub-prime lending deteriorated every year from 2001 to 2007 (Demyanyk and Hemert 2011). Servicers facilitated this process (Eggert 2004; Piskorski, Seru, and Vig 2010), and securitization has increased the influence that servicers have on mortgage termination (Adelino, Gerardi, and Willen 2009; Cordell et al. 2008; Pennington-Cross and Ho 2006).
Consumer financial markets are characterized by extreme asymmetric information between consumers and financial service providers (Campbell et al. 2011; Campbell 2016; Lusardi and Mitchell 2006, 2014). This "pervasive lack of basic financial literacy" (Campbell et al. 2011, 93) is greater among less educated and less affluent households, precisely the types of households for which FB serviced mortgages. As noted, the case for a role for government in consumer protection is based on asymmetric information and market power. Market power exists in servicing because borrowers do not choose their mortgage servicer (Eggert 2004) and cannot make a change if they are dissatisfied (unless they refinance).
III. FAIRBANKS AND THE LITIGATION
As an approved Fannie Mae-Freddie Mac servicer, FB was the largest subprime servicer in the United States in 2003, servicing $41 billion of mortgages. FB was a majority-owned subsidiary of the PMI Group, Inc., a provider of mortgage insurance, credit enhancement products, and lender services traded on the NYSE under the ticker symbol PMI. FB was accused of numerous violations of state and federal law in class action lawsuits in California, Florida, Georgia, Illinois, Massachusetts, Michigan, Ohio, Pennsylvania, Texas, and other states, as well as in several thousand individual lawsuits. Many SAGs also launched investigations and filed lawsuits. Allegations included violations of the Fair Debt Collection Practices Act (FDCPA), the Real Estate Settlement Procedures Act (RESPA), the Truth in Lending Act (TILA), the FTC Act, and various state consumer protection laws; deceptive, unfair and unconscionable debt collection practices; and assessing fictitious fees.
One common allegation was that FB held monthly payment checks from borrowers until after the 15-day grace period to impose a late fee (generally 5% of the monthly payment). Unpaid late fees cause the next monthly payment to be insufficient, triggering a pyramiding of late fees. FB was also accused of misapplication of borrowers' funds, placing money in "suspense accounts" at no interest for over a year, charging borrowers for insurance they did not request or need, charging improper prepayment penalties, and a variety of similar practices.
PMI discussed its FB litigation risk extensively in its 2004 Annual 10K Report under both "Legal Proceedings" and "Risk Factors," indicating PMI management considered the FB litigation significant. The report also discusses criminal investigations by both HUD and the U.S. Department of Justice, a possible credit rating downgrade, and that the proposed settlement would require substantial changes in FB's operations. Clearly continued problems at FB would have a substantial negative effect on PMI.
On May 16, 2003, Curry et al. v. Fairbanks was filed in Massachusetts. The Curry case was a major class action lawsuit that was joined with United States of America v. Fairbanks 6 months later. On October 23, 2003, a preliminary order was granted in the U.S. District Court for the
District of Massachusetts approving a final judgment and order. This court order made the HUD-FTC settlement possible. On November 12,2003, in United States of America v. Fairbanks, the federal government sued Fairbanks for engaging in unfair or deceptive acts or practices in violation of the FDCPA, the RESPA, the TILA, and the FTC Act. The allegations in the 18-page complaint were essentially the same as those described above. The complaint noted the substantial growth of Fairbanks' servicing portfolio in the previous 3 years through acquisitions of subprime servicing from BOA and other servicers. The complaint also pointed out that "Fairbanks finished 2002 as the subprime mortgage industry's largest servicer, managing a portfolio that totaled almost $50 billion. Fairbanks services over 500,000 mortgage loans." (p. 4). This case was then joined with the Curry case. Curry et al. v. Fairbanks and United States of America v. Fairbanks thus became the same case.
The United States of America v. Fairbanks case was settled immediately. On the same day, November 12, 2003, the FTC and HUD announced a settlement with FB. The servicer did not admit any wrongdoing, but the settlement required changes in FB's operations and the creation of a $40 million redress fund for the benefit of affected borrowers, to remedy the violations of law alleged by the FTC and HUD. The founder and former CEO of FB was required to pay a personal fine of $400,000 into the fund. HUD Secretary Mel Martinez stated that FB had "engaged in a laundry list of predatory loan servicing practices," and called it a "record settlement" (Armas 2003; U.S. Department of Housing and Urban Development 2003). The settlement was contingent upon the final order of the Federal District Court in Massachusetts; this was effective on May 4, 2004. This is very small in comparison to recent settlements related to mortgage market abuses many of which are for amounts over $10 billion. On May 4, 2004, the settlement was approved by the U.S. District Court for the District of Massachusetts. The Court order approving the settlement was released the same day as the settlement. May 4, 2004 marks the end of the litigation.
Various Objections to the settlement were filed with the Massachusetts Court. (3) The Objections noted that FB received a very broad release of liability, and that many class members would receive minimal cash payments. (4) The Objections also pointed out that many class members entitled to relief would receive nothing if they did not respond to the one notice that was sent, that many class members who had moved (such as those who had lost their homes through foreclosure) would not receive any notice at all, that all discovery in the case was under seal, that class members who accepted the settlement were prohibited from speaking to the press about any aspect of the case, (5) and that the notice provided little financial information to allow borrowers to make an informed judgment as to whether to participate. As noted, 1,145 borrowers opted out of the settlement.
IV. EMPIRICAL ANALYSIS OF SETTLEMENT-RELATED EVENTS
Based on the literature summarized in Section II, we hypothesize that, as the majority owner of FB, PMI should lose from the litigation and benefit from the settlement. Table 1 presents the key dates in the Fairbanks litigation. We use the market model and an estimation window of 300 days that include both the pre- and postlitigation period. This approach takes into account possible risk changes surrounding the litigation. Specifically, the estimation window includes 150 days ending 60 days prior to the class action lawsuit in May 2003, and 150 days beginning 60 days following the settlement approval in May 2004. Over the event window, the CAR is 7.59%. The CAR associated with settlement-related events is 11.79%.
Figure 1 summarizes graphically the results based on traditional event study methodology. While the class action lawsuit was a major negative event for PMI, the subsequent legal procedures and in particular the settlement did not affect the company negatively.
We analyze key litigation dates separately in Table 2. Following previous literature, we use two alternative event windows: (1) five trading days before and after the official announcement of the event with the event window defined in the interval [--5,5]; and (2) 10 trading days before and after the official announcement of the event with the event window defined in the interval [-10,10]. Given the potential loss of power of the test in the presence of heteroscedasticity, we include the results of running Patell's (1976), and Boehmer, Musumeci, and Poulsen's (1991) tests, which rely on standardized abnormal returns and outperform nonstandardized tests in the presence of heteroscedasticity.
Table 2 reports daily abnormal returns both in standardized and nonstandardized form for PMI's stock using the single factor market model. Six out of the nine announcements in the FB case lead to statistically significant abnormal returns in the stock price of PMI in the (--5, +5) window. Eckel, Eckel, and Singal (1997) note that financial markets attach greater importance to unanticipated announcements. Our results indicate that the main events that convey information not anticipated by market participants in the FB litigation include: (1) the Curry et al. v. Fairbanks event date on May 16, 2003, which negatively surprised the stock price of PMI (the statistically significant AR is -1.54%); the event signaled to investors the effective endorsement by the federal government of at least some of the claims of the alleged victims; and (2) the three consecutive events that followed the preliminary approval of the nationwide settlement, up to and including the final court approval of the settlement. The individual ARs are, respectively, 2.56%, 2.60%, and 0.98%. Additionally, we reject the joint null hypothesis of no abnormal returns over the main events. The average CAR for PMI's stock when using a single factor model is 8.68% (it is 10.06% when using a four factor market model).
The results of Table 2 are summarized graphically in Figure 2. We plot the daily average CAR for the settlement-related announcements over the (-5, +5) day window around the announcements. Whereas abnormal returns are positive prior to the announcements (days -5 to 0), they become visually larger over days 0 to +5. Overall, the analysis of abnormal stock returns around the FB litigation indicates significant positive wealth effects for FB's parent company.
V. SPILLOVER EFFECTS
Countrywide was once the largest home lender in the country and is often considered to have been the most egregious and aggressive lender in the precrisis period. Countrywide's incentive structure encouraged putting borrowers into high-risk, high-up-front-fee loans that they were often unlikely to be able to repay. Monthly payments increased by as much as 70% at the first interest rate reset. The New York Times reports that judges expressed shock at Countrywide's abuse of the foreclosure process and bankruptcy system by attempting to collect more than it was owed upon foreclosure. (6)
Since the subprime mortgage market could not have functioned without mortgage servicers, subprime lenders should have benefited from a favorable settlement for servicers. A finding of positive spillover effects on a large sub-prime lender is consistent with the notion that the Fairbanks litigation benefited these lenders by demonstrating that such practices either would not be heavily penalized by regulators or were actually encouraged in order to bolster homeownership. Gramlich (2007), Becker, Stolberg, and Labaton (2008), and the Financial Crisis Inquiry Commission (2010) review numerous federal policies designed to bolster homeownership in the precrisis period. The encouragement of homeownership facilitated the growth of higher-risk subprime lending and the creation of large volumes of subprime MBS. The loans that backed these securities defaulted in large numbers during the crisis.
We investigate whether FB's settlement announcement had spillover effects on aggressive mortgage lenders. We focus on Countrywide (CFC) because an initial correlation analysis of the daily contemporaneous stock returns of a group of 18 subprime lenders and PMI (not shown) reveals a statistically significant correlation between PMI and CFC. We find that Countrywide's stock price reaction to the first settlement-related announcement is 8.89% on the announcement day and 6.71% over window (-5, +5). The CAR from the first to the last announcement day is 7.66%. Thus, event study methods confirm the presence of significant spillover effects, despite the fact that PMI was a mortgage servicer and Countrywide was a subprime mortgage lender. The positive spillover effects support our argument that the settlement, which took place early in the precrisis period, bolstered the minimally regulated subprime mortgage market, in which Countrywide was a substantial player.
To evaluate the magnitude of the spillover effect from PMI's stock to Countrywide's stock after the event, we perform impulse-response analyses using a nonstructural vector autoregressive regression analysis (VAR). Impulse-response functions trace the response of a variable of interest, that is, in our case the innovation in the stock return of Countrywide, given an exogenous shock or innovation in some other variable, that is, in our case the "exogenous" innovation in the stock price of PMI. Although not shown, we find that a one standard deviation movement in the stock return of PMI resulted in a 0.78% shock to the stock return of Countrywide, fading away after three trading days. This result is economically and statistically significant.
The literature suggests that both parties should benefit from settlement. We cannot quantify the effect of the settlement on the borrowers, but the Objections to the settlement discussed above, and our calculation of average awards suggests that, contrary to what the literature would predict, most borrowers who did not opt out probably lost.
VI. ANOTHER AGREEMENT 4 YEARS LATER
The FTC, HUD, and FB (under the new name of Select Portfolio Servicing) entered to a "Modification" of the 2003 settlement in 2007. The FTC stated that it had evaluated FB's compliance with the provisions of the 2003 settlement. The 34-page modified 2007 settlement contains an extremely long list of prohibited practices, many of which were the same practices that were prohibited by the original settlement. (7) This suggests that FB had failed to comply with many of the terms of the original agreement, including those related to foreclosure and the failure to provide accurate and timely information to borrowers about amounts actually owed. For example, the new agreement restated the requirement that consumers be provided with monthly statements showing the outstanding balance on their mortgage loan. The inclusion of this provision suggests that in some cases FB may not have been following standard mortgage servicing practice 4 years after the settlement, which logically raises serious questions about the settlement's effectiveness. Put simply, a new agreement would have been unnecessary if the earlier agreement had been effective. The 2007 agreement also added the requirement for an annual audit for a period of 10 years by an independent accounting firm of FB's (now Select Portfolio Servicing's) practices in servicing loans.
VII. POLICY IMPLICATIONS
The FTC regulated mortgage servicers before the Dodd-Frank legislation in 2010, but the HUD also had a role because of RESPA, the primary law regulating mortgage servicers (Eggert 2004). When responsibility is split, it is possible that neither agency would have as much incentive or resources to develop major expertise in the area, compared to a situation in which it has sole authority.
What was the appropriate federal response to FB? What were the alternatives available to these agencies? How should regulators deal with future problems in mortgage servicing? The pivotal FB case indicates that servicers need to be held to a high standard; the size of the MBS market makes servicers central to the health of the entire financial system. Most importantly, FB could have been prevented from purchasing additional servicing until it had clearly demonstrated over a period of time the ability to handle their existing portfolio of servicing. Given that regulatory resources are limited, regulators should identify the worst financial practices, publicize them, and sanction the firms. In September 2014, the CFPB followed this approach when it prevented a mortgage servicer engaged in alleged abusive practices from acquiring any additional mortgage servicing rights until they demonstrate the ability to handle such servicing, and levied a $37.5 million fine against the parent bank holding company (Dayen 2014). This could have been done with FB. The FB case reveals that considerable operational risk can be created when servicers expand rapidly. Two depositions of FB employees in one case (Gullo v. Fairbanks) provide testimony suggesting that in the Hatboro, Pennsylvania office of FB there was extreme disorganization; some borrowers' files were missing and there was no employee training. Also, the requirement for an annual audit of FB's servicing practices added in 2007 could (and should) have been imposed in 2003, given the evidence on the nature of these practices.
A more stringent approach would have been to require divestiture of some servicing rights that were recently acquired. Regulators also could have taken steps to ensure that the redress fund created by the settlement was adequate to meet the legitimate claims of homeowners who had been damaged. A random sampling of the claims would establish an estimate of average damages, which would have revealed the inadequacy of the fund, even after allowing for some claims that may have been unsupported. More than one settlement notice could, and should, have been provided to each FB customer, especially since plaintiffs who had lost their homes in foreclosure would certainly have had a different address than the one on file at the servicer. As noted, this provision allowed FB to dismiss many claims without allowing the homeowners to make their case, creating a significant release of liability for FB. In the future, more thorough settlement procedures should be followed.
VIII. SUMMARY AND CONCLUSIONS
In a world where mortgage assets are securitized, a well-functioning mortgage servicing industry is essential to the health of the financial system. The 2007 to 2009 U.S. financial crisis, and the foreclosure crisis that followed, have exposed major weaknesses in servicing, including even such basic functions as keeping accurate records of mortgage balances. We consider the federal response to alleged abusive practices by the largest subprime servicer in the United States, Fairbanks Capital (FB), in the 5-year period immediately before the 2007 to 2009 U.S. financial crisis, that is, beginning early in the housing boom. FB was the target of class action lawsuits in more than 10 large states, and over 1,000 individual lawsuits. FB's practices were under investigation by major SAGs, and were also the subject of ongoing federal investigations. Allegations included assessing improper fees, failing to keep accurate records of mortgage balances, misapplication of borrowers' funds, foreclosing when loans were current except for failure to pay the fictitious fees, and other such practices. FB was growing rapidly by acquiring servicing from other servicers, even though the large volume and severity of these complaints indicate that the quality, reliability, and integrity of its existing servicing was being called into question.
Government agencies sue to establish a precedent (Bhagat, Bizjak, and Coles 1998). The federal response to FB was to sue and immediately administer a nominal settlement that created a small redress fund for affected borrowers, but gave the servicer a very broad release of liability. The November 2003 settlement was negotiated between FB and the two federal agencies responsible for regulating mortgage servicers, the HUD and FTC.
Objections to the settlement filed with the Court noted the very broad release of liability, and that many class members would receive very small cash payments and many class members entitled to relief would receive nothing if they did not respond to the one notice that was sent, and there was little financial information to allow borrowers to decide whether to participate. Nonetheless, the federal judge who approved the settlement specifically noted the support of the FTC for the settlement as one basis for his decision.
A 2007 "Modification" of the settlement prohibited many of the practices that also had been prohibited by the 2003 settlement. A new agreement would have been unnecessary if the earlier settlement had been effective. The Modification required an annual audit of FB's records and practices, something that could have been required earlier. By 2007, the damage had already been done; the subprime mortgage market had imploded and the U.S. financial crisis would soon follow. We suggest that a more substantive response to servicing abuses earlier may have prevented some of the later abuses, and may have caused the subprime mortgage market to develop differently.
While extensive literature exists on the economic consequences for shareholders of corporate litigation, we know of no studies dealing with the U.S. mortgage servicing litigation. We analyze the wealth effects for shareholders of FB's parent firm, PMI, from the settlement. We examine the stock price reaction of PMI, as well as the contagion effects on a major subprime lender, Countrywide, to determine the impact of the litigation and settlement on major players in the subprime market that ex post continued to display alleged improper behavior.
We find that the parent firm, PMI, experienced abnormal stock returns from the beginning to the end of the litigation events of at least 5%; some of the estimates reach as high as 20%. The clustering and size of significant cumulative abnormal returns around the settlement date points to this date as the major turning point in investors' expectations--the returns become significantly positive. Furthermore, there is evidence of statistically significant positive spillover effects from PMI to Countrywide around the settlement announcement date. This indicates that a major subprime lender benefited from the settlement, most likely because it bolstered the minimally regulated subprime mortgage market in which Countrywide was a major participant.
Regulation of mortgage servicers is a vital public policy issue given the importance of this activity to the entire financial system. Such servicers must be held to a high standard. Since two federal agencies shared responsibility for regulating mortgage servicers at this time, it is possible that neither agency had as much incentive to develop significant and detailed expertise in this industry as it might have if it had been the sole regulator. Hence, this split authority likely contributed to the ineffectiveness of the settlement.
Adeline, ML K. Gerardi, and P. S. Willen. "Why Don't Lenders Renegotiate More Home Mortgages? Rede-faults, Self-cures, and Securitization." National Bureau of Economic Research Working Paper No. 15159, 2009.
Armas, G. C. "Mortgage Firm's Practices Decried." South Florida Sun Sentinel, November 13, 2003, 1D.
Becker, J., S. G. Stolberg, and S. Labaton. "White House Philosophy Stoked Mortgage Bonfire." New York Times, December 21. 2008, Al.
Bhagat, S., J. Bizjak, and J. L. Coles. "The Costs of Inefficient Bargaining and Financial Distress." Journal of Financial Economics, 35, 1994, 221-47.
--. "The Shareholder Wealth Implications of Corporate Lawsuits." Financial Management, 27(4), 1998, 5-27.
Boehmer, E., J. Musumeci, and A. B. Poulsen. "Event Study Methodology under Conditions of Event-Induced Variance." Journal of Financial Economics, 30, 1991, 253-72.
Brown, S., L. S. Hayre, K. Lauterbach, R. Payne, and T. Zimmerman. "Analysis of Mortgage Servicing Portfolios." Journal of Fixed Income, 2, 1992, 60--75.
Campbell. J. Y. "Restoring Rational Choice: The Challenge of Consumer Finance." American Economic Association. Richard T. Ely Lecture (January). American Economic Review. 106.2016, 1-30.
Campbell, J. Y., H. E. Jackson, B. C. Madrian, and P. Tufano. "Consumer Financial Protection." Journal of Economic Perspectives, 25, 2011, 91 -114.
Cordell, L., K. Dynan. A. Lehnert, N. Liang, and E. Mauskopf. "The Incentives of Mortgage Servicers: Myths and Realities." Federal Reserve Board Working Paper No. 2008-46, 2008.
Cutler, D., and L. W. Summers. "The Costs of Conflict Resolution and Financial Distress: The Texaco-Pennzoil Litigation." RAND Journal of Economics, 19, 1988, 157-72.
Dayen, D. "A Regulator Finally Gets It Right: CFPB Hits a Bad Bank Where It Hurts." The Fiscal Times. October 3, 2014.
Demyanyk, Y., and O. V. Hemert. "Understanding the Sub-prime Mortgage Crisis." Review of Financial Studies, 24(6), 2011, 1848-80.
Eckel, C, D. Eckel, and V. Singal. "Privatization and Efficiency: Industry Effets of the Sale of British Airways." Journal of Financial Economics, 43, 1997, 275-98.
Eggert, K. "Limiting Abuse and Opportunism by Mortgage Servicers." Housing Policy Debate, 15, 2004, 753-84.
Engleman, K., and B. Cornell. "Measuring the Cost of Corporate Litigation: Five Case Studies." Journal of Legal Studies, 17. 1988,377-99.
Federal Trade Commission News Release. "FTC, Subprime Mortgage Servicer Agree to Modified Settlement: Agreement with Former Fairbanks Capital Provides Additional Consumer Benefits." 2007. Accessed August 2, 2007. https://www.ftc.gov/.
Financial Crisis Inquiry Commission. The Financial Crisis Inquiry Report. New York: Perseus Books, 2010.
Gramlich, E. M. Subprime Mortgages: America's Latest Boom and Bust. Washington, DC: The Urban Institute Press, 2007.
Grind, K. The Lost Bank: The Story of Washington Mutual--The Biggest Bank Failure in American Histoiy. New York: Simon and Schuster, 2012.
Hayashi. Y. "Case Fuels Fight on Consumer Watchdog." Wall Street Journal, January 25, 2016.
Lusardi, A., and O. S. Mitchell. "Financial Literacy and Planning: Implications for Retirement Well-being." Pension Research Council Working Paper No. 1, 2006
--. "The Economic Importance of Financial Literacy: Theory and Evidence." Journal of Economic Literature. 52,2014,5-44.
Michaelson, A. The Foreclosure of America: The Inside Story of the Rise and Fall of Countrywide Home Loans, the Mortgage Crisis and the Default of the American Dream. New York: Berkley Publishing Group. 2009.
Morgenson, G. "Inside the Countrywide Lending Spree." New York Times, August 26, 2007, CI.
--. "Lender Tells Judge It Created Letters." New York Times, January 8, 2008a, C3.
--. "U.S. Seeks Sanctions against Lender." New York Times, March 1, 2008b, B3.
--. "The Foreclosure Machine: An Industry Thrives on Housing Woes." New York Times, March 30, 2008c.
--. "Countrywide's Buyer Isn't Blinking." New York Times. June 8, 2008d, C1.
--. "Illinois Files Suit against Countrywide." New York Times, June 25, 2008e, C1.
--. "Waiting 7 Years for 2 Answers." New York Times, February 27, 2011, C1.
Ovide, S. "Bank of America-Countrywide: Worst Deal in History?" Wall Street Journal, June 29, 2011.
Patell, J. "Corporate Forecasts of Earnings per Share and Stock Price Behavior: Empirical Tests." Journal of Accounting Research, 14, 1976, 246--76.
Pennington-Cross, A., and G. Ho. "Loan Servicer Heterogeneity and the Termination of Subprime Mortgages." Federal Reserve Bank of St. Louis Working Paper No. 2006-24A, 2006.
Piskorski, T., A. Seru, and V. Vig. "Securitization and Distressed Loan Renegotiation: Evidence from the Subprime Mortgage Crisis." Journal of Financial Economics, 97, 2010, 369-97.
Reddy, S. "Bernanke Defends Record on Lehman." Wall Street Journal. September 3, 2010, A6.
Rexrode, C. "Citi to Leave Business of Servicing Mortgages." Wall Street Journal, January 31, 2017, B10.
Rudolph, J. C. "Judges' Frustration Grows with Mortgage Servicers." New York Times, September 4, 2009.
Silver-Greenberg, J., and M. Corkery. "Loan Complaints by Homeowners Rise Once More." New York Times, February 19, 2014, Al.
U.S. Department of Housing and Urban Development. "HUD and FTC Announce Record RESPA Settlement with Fairbanks Capital for Alleged Abuses of Borrowers." Press release, November 12, 2003. http://www.hud.gov.
Van Drunen, L. D., and J. J. McConnell. "Valuing Mortgage Servicing." Journal of Real Estate Finance and Economics, 1, 1988,5-22.'
Cooks et al. v. Fairbanks Capital Corporation (Class Action, Circuit Court of Cook County, Chicago, Illinois, Case No. 02 CH 4382).
Curry et al., individually and on behalf of all others similarly situated v. Fairbanks Capital Corporation (Civil Action No. 03-10895-DPW [National Class Action]).
Gullo v. Select Portfolio Servicing f/k/a Fairbanks Capital Corporation (Seventeenth Judicial Circuit. Broward County, Florida, Case No. 03-020055 CA 8).
In re Pearl Maxwell, Pearl Maxwell v. Fairbanks Capital Corporation (2002 WL 1586325 (Bankr.D.Mass.).
United States of America v. Fairbanks Capital Corporation (Civil Action No. 03-12,219-DPW). United States District Court for the District of Massachusetts. (1) Notice of Proposed Class Action Settlement and (2) Order Preliminarily Approving Stipulated Final Judgment and Order as to Fairbanks Capital Corporation and Fairbanks Holding Company, November 21, 2003.
United States of America v. Fairbanks. Objection to Proposed Settlement Agreement, Filed April 12, 2004. Pacer.gov.
United States of America v. Select Portfolio Servicing, Inc. and Thomas J. Basmajian (Civil Case No. 03-2,219-DPW). United States District Court for the District of Massachusetts. Modified Stipulated Final Judgment and Order, September 4, 2007.
JAMES E. MCNULTY, LUIS GARCIA-FEIJOO and ARIEL VIALE (*)
(*) We thank an anonymous referee for very valuable comments.
McNulty: Professor, Department of Finance, Florida Atlantic University, Boca Raton. FL 33431. Phone (561) 451-0934, Fax (561) 883-9394, E-mail email@example.com
Garcia-Feijoo: Associate Professor, Department of Finance, Florida Atlantic University, Boca Raton, FL 33431.
Phone (954) 236-1239, Fax (561) 297-3978, E-mail firstname.lastname@example.org
Viale: Associate Professor, Rinker School of Business, Palm Beach Atlantic University. West Palm Beach, FL 33401. Phone (561) 803-2462, Fax (561) 803-2455, E-mail email@example.com
BOA: Bank of America
CARs: Cumulative Abnormal Returns
FB: Fairbanks Capital Corporation
FDCPA: Fair Debt Collection Practices Act
FTC: Federal Trade Commission
HUD: U.S. Department of Housing and Urban Development
MBS: Mortgage-Backed Securities
NYSE: New York Stock Exchange
SAGs: State Attorneys General
(1.) Banks have defended their practices, arguing that the deficiencies are merely technical issues, and stated that their reviews of their own records have failed to detect any borrowers who were foreclosed on improperly.
(2.) The Wall Street Journal has characterized the BOA-Countrywide merger as perhaps the worst merger in U.S. corporate history for the acquiring institution because of the Countrywide litigation (Ovide 2011).
(3.) United States of America v. Fairbanks, Objection to Proposed Settlement Agreement, Filed April 12, 2004 (available at www.Pacer.gov).
(4.) One of the objectors argued that many class members would receive cash payments of less than $200. $40 million distributed equally among 500.000 borrowers (the total used in the court documents described in Section III) is $80 per borrower, but not all borrowers were affected. However, less than $40 million was available for redress because of attorneys' fees. One borrower with a much larger loan than most of the class (and hence arguably more damages) was offered approximately $3,500 but opted out of the settlement.
(5.) Despite an extensive search, we could not find an objection to the settlement filed by any state attorney general, although many had sued Fairbanks earlier. The prohibitions on discussing the settlement may account for this.
(6.) Morgenson (2007, 2008a, 2008b, 2008c, 2008d, 2008e) discusses lending practices at Countrywide. One theme is that employees routinely encouraged borrowers to purchase homes they could not afford, locked them into high-interest-rate loans with high prepayment penalties, and the commission structure rewarded these actions. Michaelson (2009), a former Countrywide executive, also discusses the extremely aggressive sales culture at Countrywide prior to the crisis.
(7.) See Federal Trade Commission (2007) and United States of America v. Select Portfolio Servicing (2007). Both documents are available at www.FTC.gov under Press Releases for August 2, 2007.
TABLE 1 Fairbanks Litigation Dates Date Description July 16,2002 Maxwell case settled. A Massachusetts Bankruptcy Court Judge comments that Fairbanks "in a shocking display of corporate irresponsibility, fabricated amounts owed out of thin air." May 16, 2003 Curry et al. v. Fairbanks case filed in Massachusetts. The Curry case was a major class action lawsuit that was later joined with United States of America v. Fairbanks. October 23, 2003 Preliminary order approved in U.S. District Court for the District of Massachusetts approving a final judgment and order. This court order made the HUD-FTC settlement possible. November 12, 2003 HUD-FTC settlement announced December 1,2003 Various objections to the settlement filed December 10, 2003 U.S. Court for the Massachusetts District entered an order granting preliminary approval of the nationwide class action settlement of servicing disputes with Fairbanks and a Preliminary Injunction staying all servicing disputes between Fairbanks and class members April 6, 2004 and April 8, 2004 Further objections to settlement filed May 4, 2004 The HUD-FTC settlement approved by the U.S. District Court for the District of Massachusetts. The court considered the objections to the settlement, but cited the support of the FTC as an important basis for approving the settlement. The court order approving the settlement is released the same day as the settlement. TABLE 2 Abnormal Returns Based on the Single-Factor Market Model Event date Event Description Window B&H(%) AR(%) July 16, 2002 Maxwell case settled (-5, +5) -9.97 -1 79 (**) (-10,+10) -4.77 -0.48 May 16, 2003 Class action lawsuit (-5,+5) -9.87 -1.54 (**) (-10,+10) -15.59 -1.53 (**) October 23, 2003 Preliminary approval (-5, +5) 5.39 0.82 (**) (-10,+10) 3.05 0.36 November 12,2003 Settlement announced (-5. +5) -0.40 -0.07 (-10,+10) -3.80 -0.37 December 1,2003 Objections filed (-5, +5) -0.73 -0.09 (-10,+10) -1.16 -0.12 December 10, 2003 Preliminary approval (-5, +5) 0.65 0.10 (-10,+10) -1.37 -0.12 April 6, 2004 Objections filed (-5, +5) 14.87 2.56 (**) (-10,+10) 16.24 1.96 (**) April 8, 2004 Objections filed (-5.+5) 15.01 2.60 (**) (-10,+10) 20.19 2.38 (**) May 4, 2004 Settlement approved (-5, +5) 5.39 0.98 (**) (-10,+10) 9.29 1.23 (**) Event date StdAR(%) July 16, 2002 -1.87 (**) -0.50 May 16, 2003 -1.52 (**) -1.55 (**) October 23, 2003 0.80 (**) 0.36 November 12,2003 -0.06 -0.38 December 1,2003 -0.09 -0.12 December 10, 2003 0.10 -0.12 April 6, 2004 2.60 (**) 2.03 (**) April 8, 2004 2.64 (**) 2.46 (**) May 4, 2004 1.03 (**) 1 32 (**) Notes: This table reports the reaction of the PMI stock price to the nine announcements comprising the Fairbanks litigation process. The results are shown for event windows of 5 and 10 days prior to and subsequent to the event. B&H denotes annualized buy and hold abnormal returns starting from the first event. AR and Std. AR denote daily abnormal returns and standardized daily abnormal returns, respectively. Standardized abnormal returns have been adjusted to be distributed as a standard normal variate with zero mean and unit variance. (*) Statistical significance at the 5% level and "statistical significance at the 1% level, using a two-tail test.
|Printer friendly Cite/link Email Feedback|
|Author:||Mcnulty, James E.; Garcia-Feijoo, Luis; Viale, Ariel|
|Publication:||Contemporary Economic Policy|
|Date:||Jan 1, 2019|
|Previous Article:||WILL GERMAN BANKS EARN THEIR COST OF CAPITAL?|
|Next Article:||LIABILITY AND ANTIFRAUD INVESTMENT IN FINTECH RETAIL PAYMENT SERVICES.|