The City of Philadelphia's Residential Mortgage Foreclosure Diversion Program: Addressing the Rising Tide of Foreclosure

Abstract

As the foreclosure crisis was intensifying in 2008, the Philadelphia Court of Common Pleas stepped in and created the Philadelphia Residential Mortgage Foreclosure Diversion Program. Together with the City of Philadelphia and a steering committee of stakeholders, the particulars of this program were shaped and reshaped at various times after it began. Philadelphia was among the first cities to create a program, although there are many today. Diversion Programs vary greatly in structure and eligibility criteria, but what is common is the general lack of data descriptive of what the intervention has accomplished. Our analysis of 28,000 Court Orders on 16,000 foreclosure cases shows that (a) the program annually addresses more than 60% of Philadelphia's foreclosure case filings, (b) approximately 70% of eligible homeowners participated, (c) 35% of those who participated reached an agreement with their lender/servicer, and (d) of the first year's agreements, 85% of homeowners remained in their homes 20+ months post-agreement. All of this was accomplished without significant delay in case processing time or expenditure of Court resources.

Keywords:

• data
• foreclosure
• neighborhood

Notes

^1. See U.S. Department of the Treasury (2012).

^2. See Office of the Comptroller of the Currency, US Department of the Treasury (2012).

^3. Although not studying diversion programs, Gerardi, Lambie-Hanson, and Willen (2011) call into question the efficacy of two borrower protections—judicial foreclosure requirements and a Massachusetts right-to-cure law—that are designed to avoid unnecessary foreclosures. The authors conclude that these are ineffective in improving outcomes for homeowners in foreclosure. They conclude that these protections do not increase the likelihood of curing a delinquency. Moreover these protections extend “the period of separation of ownership and control of the residential property” thereby adversely impacting the communities within which properties are located. The Gerardi et al. finding makes the sustainability of results achieved in mediation programs, of any kind, a critical factor to understand.

^4. See Clifford (2012).

^5. See Cohen and Jakabovics (2010).

^6. See Harvard Negotiation Law Review (2012).

^7. See American Bar Association (2012).

^8. For a detailed description of these arguments, see Brief Amici Curiae of the National Association of Consumer Advocates, the U.S. Public Interest Research Group, and The Center for Responsible Lending in Support of Respondents in Buckeye Check Cashing, Inc. v. John Cardegna. 546 U.S. 440 (2006).

^9. See Barnum and Rosenberg (2011).

^10. See Walsh (2012) and Collins and Urban (2011).

^11. Walsh at 17–19.

^12. Collins at 15.

^13. See, for example, Streitfeld (2011) and Blankenship (2008).

^14. See Mortgage Bankers Association (2012).

^15. See Cohen and Jakabovics (2010, p. 3).

^16. See United States Department of Justice, Access to Justice Initiative (2010).

^17. Cohen and Jakabovics (2010, p. 4).

^18. Access to Justice at 4.

^19. Walsh (2012, p. 34).

^20. On July 1, 2011, owing to a failure of the Pennsylvania Legislature to fund HEMAP, the Pennsylvania Housing Finance Agency stopped accepting applications for the program. See Hudson (2011). HEMAP was restarted in August of 2012 when Pennsylvania dedicated $66.5 million from the state-federal settlement with mortgage servicers to recapitalize the program.

^21. For a description of the various approaches and the conditions in which they are appropriate, see http://courts.phila.gov/pdf/forms/civil/Residential-Mortgage-Foreclosure-Diversion-Program-Materials.pdf (accessed July 9, 2012).

^22. See http://www.saveyourhomephilly.org/.

^23. CLS and PLA represent low income Philadelphians in civil legal matters; VIP facilitates pro bono representation of low income Philadelphians in civil matters.

^24. The phrase book and page is a reference to the way, historically, property deeds and mortgages were stored. Deeds and mortgages are now assigned a document id number when filed with the Recorders Office. Each unique identification number is important because it is the most certain means to link the foreclosure filing to other databases in order to identify the lending institution associated with the mortgage at the time of origination.

^25. There is no comparable Order to “Failed to Appear” if the lender's attorney does not attend the conciliation conference or if the lender or their attorney is unprepared for the conference.

^26. Status is a code that is no longer accepted by the Court.

^27. When a bankruptcy is filed, the Diversion Program loses jurisdiction of the case.

^28. The Diversion Program's Steering Committee voted to alter the Order form to include information on the nature (e.g., HAMP, HEMAP, traditional loan modification) of the agreement, although complete financial details are not included in the new Order form.

^29. The “closed/inactive/inactive” status of a case is one that is very difficult to determine in the Diversion Program. That is because there is extended time lag between the effective end of a case and when an Order is issued and recorded to close the case. Accordingly, a rule was established for estimating a closed/inactive case so that when there is no activity 120 or more days after the last known Order has been issued on a case, the case is coded as closed/inactive. This is admittedly imperfect as evidenced by the fact that a number of cases remain in a limbo status (e.g., continued, status, and so on).

^30. Participation rates vary greatly by jurisdiction, legal structure, and program design. For example, the NY state program initially applied only to residential foreclosures associated with certain subprime loans, and now applied to all residential foreclosures, reports a default rate of 25.5%. In CT, all homeowners are eligible but must respond in order to preserve that right. CT reports an approximate 38% participation rate. RI's program allows for automatic homeowner eligibility and has an estimated 10% participation rate. Marion County, IN, allows all homeowners to participate in their program. A reported 30–40% participates in mediation (Walsh, 2010 and 2011).

^31. The Prothonotary assigns a Case ID number to each case that includes the month and year of filing. Estimating from the 15th of that month, we can measure the length of time from filing to the end of Diversion, and in some cases to the resolution of the foreclosure action.

^32. As noted previously, there is little comparative data on the impact of mediation programs. Walsh (2012) notes that in New York, average foreclosure time frames are nearly 1,000 days. This time frame he attributes not to homeowners dragging out the process but to the lenders/servicers failing to file all appropriate paperwork to allow the process to go forward.

^33. There is no stated time outlined for completing of foreclosure process, from time of filing to the eventual sale of the property. A 10-month period is the minimum time period from filing to completing of foreclosure absent any defense mounted by the homeowner. For a chart detailing the entire process as it existed before the introduction of the Diversion Program (see Urban Affairs Coalition, 2012).

^34. Cases not going through the Diversion Program represent something of an open question. We believe that this pool of cases includes a mixture of the following: (1) cases who may be officially residential but not actually so; (2) cases who are estimated to be owner occupied, but are actually not; (3) property owners who, for one reason or another, did not petition into the Diversion Program and any selection bias that may attach to these cases; (4) cases who the Court and Prothonotary failed to properly identify as eligible and the homeowners who did not come forward and petition into the Diversion Program.

^35. Stated differently, the latter group would have been eligible for the Diversion Program and thus its presence was likely influential on the substance of the negotiated agreement. For whatever reason, however, the homeowners and plaintiffs in this group achieved their agreement prior to a conference.

^36. Moran's I ranges from .00 to 1.00; values near 0 reflect a lack of spatial autocorrelation (i.e., the degree to which values of a variable are similar in neighboring spaces). A positive spatial autocorrelation would exist if, for example, areas with high number of foreclosures have neighboring areas also having high number of foreclosures. In this instance, the computed Moran's I values are essentially zero.