Download the full report (pdf)
This report now appears in the Federal Reserve of San Francisco’s "Community Development Investment Review" (pdf)
During what may become the largest financial crisis in the history of the United States—triggered by the U.S. housing market slide that began in 2007—the Center for American Progress made the point again and again that home prices would not recover until the federal government intervened in a more robust fashion to help individual homeowners cope with the cascading crisis.
This paper, written by New America Foundation’s Ellen Seidman and Andrew Jakabovics from the Center for American Progress, discusses asset disposition lessons learned from previous financial crises—including the activities of the New Deal-era Home Owners Loan Corporation, the Resolution Trust Corporation formed in the wake of the savings and loan crisis, and the Department of Housing and Urban Development’s more recent Asset Control Area program to dispose of foreclosed properties acquired through the Federal Housing Administration.
As Congress weighs the various strengths and weaknesses of creating a new Resolution Trust Corporation to address the current crisis, we believe this paper would be very useful. While much of the focus in the coming days will deal with acquisition of distressed mortgages and other debts, the inevitable need to have a strategy to deal with properties acquired through foreclosure by an RTC-like entity must also be in place from the outset.
While there is little to celebrate in the current foreclosure disaster, one potential silver lining in the large number of bank-owned properties is the opportunity to turn those properties into community assets. A recent conference hosted by the Furman Center for Real Estate and Urban Policy at New York University and sponsored by the Ford Foundation brought together policy experts and practitioners to share best practices for “Transforming Foreclosed Properties into Community Assets.” Most of the discussion focused on what can be done by partners working together at the local level. The current situation is not, however, the first time that the federal government has faced the challenge of turning foreclosed residential property into affordable housing.
In this paper, prepared for the NYU conference, we consider three earlier experiences with asset disposition by the federal government—the New Deal-era Home Owners’ Loan Corporation, the Resolution Trust Corporation, and the Department of Housing and Urban Development’s Asset Control Area program. In each case, the federal government was forced to deal with tge large-scale disposition of private-sector assets that passed into public hands as a function of federal funds put into an earlier, related transaction.
In the case of HOLC and ACA, default on federally guaranteed home loans triggered foreclosure and transfer of the property to the respective entities. Both programs were entirely focused on residential properties. In the case of the RTC, the properties in question were already owned by failed banks insured by the FSLIC or the FDIC, or were collateral on RTC-owned loans that proceeded to foreclosure. The vast bulk of the RTC’s loans and properties were commercial, although there was enough residential property to make the case worth studying. All three programs were challenged to maximize revenues from the disposition of the assets acquired, while also not overburdening local markets already in a weakened state. In the case of the RTC and ACA, the mandate included a third element: preservation and expansion of affordable housing. Here we provide an overview of the three programs, with a focus on their residential property disposition experiences. We conclude with some lessons we think we can take from these experiences, and pose a series of questions we believe they raise for our current situation.
The Home Owners’ Loan Corporation was established in 1933 and issued new loans through 1936. Homeowners would apply to HOLC to refinance their existing loans through one of 458 local offices around the country. At its peak in 1934 the HOLC employed approximately 20,000 people, in addition to contractors paid on a fee basis.
When a borrower applied to HOLC, the Corporation would appraise the property and offer the borrower a new loan based on the property’s current value. In exchange for the lien on the property, HOLC would offer the mortgagee a corporate bond backed by an explicit government guarantee. The homeowner was given a 15-year, fully amortizing loan. To put the scale of the program in perspective, one out of every five qualifying owner-occupied properties where a mortgage was outstanding in 1934 was refinanced through HOLC. Approximately 1 million loans were issued, with a total principal balance of $3.1 billion, or an average of slightly more than $3,000 per property.
As critics of HOLC have always been quick to point out, roughly 20 percent of the borrowers redefaulted to the point of losing their homes to foreclosure. HOLC disposed of 198,000 foreclosed properties, although the process was spread throughout the corporation’s existence. The maximum number of properties managed was roughly 103,000, from early 1938 through early 1939, after which numbers declined steadily, with only 6,000 properties owned by 1944.
Many of the properties acquired by HOLC through foreclosure required significant investment to make them saleable, above and beyond any tax liens that needed to be paid. This was due to a combination of aged stock and neglect by the homeowners foreclosed upon. Properties were often rehabilitated to bring them in line with comparable local property quality. Nationally, expenditures on rehabilitation averaged 11 percent of the original loan amount, or about 12 percent of the net sales proceeds. All together, for the 198,000 properties acquired by the HOLC, rehabilitation costs came to $89 million, or $451 per property. HOLC’s property management division relied on a combination of its own employees and contractors to fulfill a range of tasks.
HOLC engaged in a surprising amount of due diligence in advance of acquiring properties. When the various loss mitigation strategies employed by the Corporation seemed to fail, setting a property on the path to foreclosure, HOLC would reappraise the property using the three appraisal methods it used before originating a HOLC loan. The three methods were market comparables, replacement cost less depreciation, and capitalized rent stream. For the purposes of determining what to do with property acquired in foreclosure, emphasis was placed on the rental valuation—a critically important concept when, like today, accurate sale comparables are hard to come by. In many cases, HOLC worked with the borrower to try to sell the property before foreclosing. The fact that the property ultimately proceeded to foreclosure was an indication of continued weakness in the for-sale market, suggesting that renting the property would maximize value to the corporation.
HOLC properties were rented out on a month-to-month basis—which may have lowered the rents asked—by contract brokers, who were also responsible for property management. The brokers were allowed to make necessary repairs up to $25 without approval and spend up to $100 on an emergency basis, such as for a stove or heating. HOLC representatives would check properties annually to assess conditions. Maintenance costs on rented units totaled $26.8 million over the life of the corporation and equaled 19.3 percent of the gross rental income. As markets improved, rental properties were offered up for sale. When a determination was made to sell a HOLC-owned property, either directly after acquisition or after a period of renting, local brokers—often the same ones who managed properties—under commission-based contracts, listed and sold the properties. Advertising costs were paid by the brokers.
Prices were set based on appraisals that indicated estimated sales prices based on varying levels of rehabilitation. The decision to rehabilitate for-sale properties was based on the anticipated return on that investment. The minimum value HOLC would accept for a property was kept secret when the property was put up for sale. There was also HOLC oversight in setting the floor: any property offered for less than $1000 but that was profitable to HOLC or properties over $1000 where the price reflected less than a 35 percent loss were approved by HOLC’s regional property committee; losses greater than 35 percent needed to be approved by a Home Office Property Committee. The actual offering price was often far higher than the reserve, but it rarely exceeded HOLC’s costs.
HOLC was often forced to finance its own sales, both because as a government entity it could obtain better rates and because other lenders were unwilling to take on the risk. HOLC required a down payment that averaged 12.2 percent nationally (New York and New Jersey down payments were higher), and a year’s taxes and insurance in escrow, but the down payment would vary with creditworthiness and local standards. The down payment was designed to cover, at a minimum, HOLC’s anticipated costs should the new owner default on the loan.
From this summary, we think four HOLC innovations and practices stand out, as well as a few important lessons. First, the 15-year fully amortizing loan was a major breakthrough in mortgage finance. Second, the corporation was rigorous in its appraisal practices, both when initially making a loan and when putting a foreclosed property up for sale. In part this was a matter of risk management, but it was also an attempt not to add to a downward market spiral. Third, the agency had a very clear understanding both that it was a temporary entity and that real estate markets are highly localized. This led to the heavy reliance on a network of contract brokers and others. Fourth, when faced with the choice of continuing to hold on to a property for which it had a credit-worthy buyer or financing the buyer itself, it chose to self-finance.
As to lessons, we think a clear one is that when a government agency is tasked with the job of cleaning up after a major financial disaster, success is mitigating and spreading out over time the effects of the problem. The HOLC’s 20 percent default rate in a rising market, may look troubling today, but the question is “compared to what?” It almost certainly reduced the suffering of both homeowners and markets at a very troubled time. Moreover, best estimates are that profits from the HOLC’s lending operations were slightly larger than losses on foreclosures.
Turning now to the RTC, the Resolution Trust Corporation came into existence with the passage of FIRREA in August 1989, which transferred conservatorship of just over 250 failed thrifts from the FSLIC to the RTC. By the end of 1990, the RTC had taken over 531 thrifts with $278 billion in assets. This represented slightly more than two-thirds of the assets ultimately under the RTC’s responsibility. The volume of new assets dropped to $79 billion in 1991 and dropped again to $44 billion in 1992, the last year with any significant inflows to the RTC. Nearly half (48 percent) of the RTC’s assets were commercial and residential mortgages, with the other half a combination of REO (properties foreclosed upon by failed banks as well as bank real estate such as branch locations), other loans, securities, and other assets, including subsidiary corporations.
In contrast to HOLC, whose primary task was to refinance mortgages, with asset disposition a consequence of the inevitable failure of some of that financing, the RTC was about asset disposition from the start. Moreover, only about $3 billion of the $402 billion of assets that passed through the RTC consisted of residential real estate. The RTC’s enabling legislation emphasized maximizing returns and minimizing losses, minimizing the impact on local real estate, and maximizing affordable housing preservation. As should be obvious, at least two of these goals are inherently in conflict. While the RTC was initially expected to take many years to dispose of all its assets, the organization—staffed heavily with recruits from the bank regulatory agencies and under constant funding pressures—determined that maximizing returns meant moving quickly to reduce the properties’ drag on the market, especially in Texas. The RTC received its final funding from Congress in 1993 and closed up shop in 1995.
The RTC disposed of assets through a variety of channels—direct sales, auctions, securitization, and a small number of joint ventures with private firms. The corporation relied heavily on private firms to evaluate, package, and sell assets, which was both a matter of direction from Congress and a matter of necessity. Servicing of performing loans was contracted out to conventional mortgage servicers until disposition, while nonperforming loans, REO and other assets were offered to contractors to process for disposition.
From 1991 to 1993, 91 contractors won 199 contracts for the RTC’s Standard Asset Management and Disposition Agreements to handle assets of $48.5 billion, a small fraction of which was residential real estate. The contracts were for three years, with two optional one-year extensions, depending on the number of outstanding assets. The SAMDAs set up management, disposition, and incentive fees, but did not cover overhead costs. Firms bidding for SAMDA contracts would offer bids on their management and disposition fees; the incentive fee structure was fixed by the RTC for all contracts and paid a 20 percent bonus on the disposition fee for assets sold in the first year of the contract and 10 percent in the second year. The SAMDAs further required engaging other private sector firms and employees by mandating subcontracts for appraisals, REO brokerage, property management, and the like. In total, 12 services required subcontractors under SAMDA. A change to the program in January 1992 dropped disposition from contracts, as the RTC moved towards a national, multi-asset disposition model that improved net recovery and shortened holding times over individual asset sales.
The SAMDA program’s record was somewhat mixed, relative to similar contract programs run through the FDIC. Real estate expense ratios were higher under the SAMDA program, but the asset quality in the SAMDA program was worse, which accounts for some of the discrepancy. All together, the $48 billion in book value of assets disposed of through SAMDAs returned net collections of $19 billion, for a recovery rate of 41 percent. The overall expense-to-collection ratio was 19 percent. Real estate sales accounted for 40 percent of the book value reductions but 70 percent of the disposition fees.
In addition to disposing of real estate through SAMDA, the RTC used a direct sale approach, and later shifted to auctions as the volume of properties became too great to handle directly. To address concerns about flooding local markets with large numbers of distressed properties for sale, the RTC was under a mandate not to sell properties for less than 95 percent of value in direct sales or under the SAMDA contracts. Under the auction program rules established in March 1991, an absolute floor of 70 percent of value was set. In practice, the actual floor varied with each auction.
Residential real estate, a combination of single-family homes and multifamily buildings, was funneled through the RTC’s Affordable Housing Disposition Program (AHDP). Over all, the RTC sold 91,000 units of multifamily housing and another 28,000 single-family homes through the AHDP. Another 25,000 multifamily and 14,000 single family units were sold outside the AHDP. As noted, residential real estate represented about one half of one percent of the $402 billion in book value of assets that passed through the RTC. Seventy percent of the multifamily properties were in the South and West, and nearly 30 percent of the single family properties were in Texas.
Multifamily properties were initially sold through a clearinghouse process with sale to the highest bidder. Multifamily property sold through the AHDP had use restrictions for 40 years, during which 35 percent of the units needed to be rented to households below 80 percent of area median income. In addition, 20 percent of the units were to be rented to households below 50 percent of AMI. Single family houses were also deed restricted, but eligibility extended up to 115 percent of AMI. At closing, buyers were obligated to sign certificates of intent to occupy the property and to certify income eligibility. There was also a 1-year recapture provision that allowed the RTC to take 75 percent of the profits of a sale that took place within a year of closing. Single family properties could be sold directly to income-qualified households or to local housing nonprofits who were obligated to rehabilitate the properties and rent them or sell them to households meeting the eligibility requirements.
The RTC engaged in several activities to speed the disposition of its assets under the AHDP and improve the outcomes for buyers. First, the RTC engaged local nonprofits to provide pre-purchase counseling to prospective buyers as well as post-purchase seminars on owner responsibilities like maintenance, mortgage payments, and insurance. Nonprofits also provided technical assistance to public housing authorities seeking to purchase the RTC’s multifamily buildings. Second, the RTC established a seller financing program for both single and multifamily properties, in recognition of the difficulty many potential purchasers had in finding suitable financing, including through the FHA. RTC loans were offered for up to 97 percent of the value of single family homes, with the RTC also covering closing costs. Approximately 20 percent of single family sales involved RTC financing. In addition, the RTC would provide up to $5,000 to repair single family homes in inventory, in recognition of the fact that low- and moderate-income buyers would likely be unable to afford repairs.
In May 1992, the RTC changed how it sold multifamily properties, from sale to the highest bidder with use restrictions, to direct sales through a series of sales windows. Public agencies were given the first 30-day opportunity to buy a property. If the property was unsold after 30 days, nonprofits would be given a chance, and if it remained unsold after that time, the property would go into a clearinghouse for anyone to purchase within 90 days. Buyers would have to commit to the affordable unit set asides described previously. Only after a property failed to sell through the clearinghouse would the RTC place it for sale outside the AHDP. Under FIRREA, Congress established a similar 90-day marketing period for single-family houses for public agencies, nonprofits, and qualified buyers.
The innovation for which the RTC is best remembered is the commercial mortgage backed security. But while residential real estate was a relatively minor part of the RTC’s activities, the RTC was responsible for several important innovations in that area, especially with respect to protecting housing affordability. These include working through non-profits and local housing authorities, use restrictions, the tiered sale process, and systems to provide counseling to new homeowners. Like HOLC, the RTC provided some seller financing and assisted in upgrading many of its properties before or as part of their disposition.
The final asset disposition program we wish to discuss is HUD’s Asset Control Area program, the only one of these programs still in existence. The ACA program was initially authorized in 1998 as a pilot program to dispose of 40,000 FHA-foreclosed properties while stabilizing communities and reducing opportunities for speculative buying, superficially fixing and flipping properties. It has since become a permanent program.
Within each asset control area, housing intermediaries, originally acting on behalf of local governments, but now directly contracting with HUD for preferred bulk purchasing rights, agree to purchase FHA-owned properties, rehabilitate them, and resell them as affordable housing. The FHA sales price is determined by appraisal and discounted to account for necessary rehabilitation. Early ACA programs were in Chicago, Cleveland, and Los Angeles, among other places.
One of the central components of the ACA program is its narrow geographic focus. ACA program participants submit a plan to HUD that identifies specific census tracts for targeted investment, affordability requirements, counseling programs, marketing plans, sales projections, and quality control measures. Once a contract is reached with HUD, ACA participants are obligated to purchase all HUD-owned single-family homes within the designated area, up to an annual cap. For example, Enterprise’s Dallas ACA program committed to buying 100 homes per year during its two-year contract.
After receiving notice of HUD-owned properties in the area, the nonprofit has a short period of time in which to inspect the properties and itemize needed repairs. HUD then sells the properties to the nonprofit at a discount of at least 50 percent of as-is appraised value. Purchases are funded through credit facilities provided by local lenders, often with interest reserves funded by the city.
Once the nonprofit acquires the properties, it begins rehabilitation. Rehabilitation includes lead and asbestos abatement, improving energy efficiency, and repairing major building systems to provide a minimum 10-year service period. Substantial rehabilitation can cost as much as $150,000 per property, but it varies significantly by market. Enterprise reports that in Los Angeles, the average cost of acquisition and rehabilitation was $235,000 per property. The cost in Rochester was approximately one-third of that. In Dallas, rehabilitation alone costs an average of $25-$30,000, but there is a significant range, with new properties needing as little as $11,000 and older, wood frame properties with deteriorated foundations costing as much as $70,000.
Once a property has been rehabilitated, it is offered for sale to low- and moderate-income buyers. Pricing is set based on a strict formula: acquisition costs, plus rehabilitation costs, plus a 15 percent markup. At the time of sale, the property is appraised, and the difference between the appraised value and sales price is captured by a soft second note held by HUD for three years. If the owner stays in the property for three years after purchase, the soft second is extinguished and the equity is transferred to the owner.
Eligibility is determined by the terms of the ACA agreement with HUD. Unlike HOLC or the RTC, which offered seller financing, ACA program participants are unable to finance prospective buyers. In many cases, the nonprofits or local government will offer down payment assistance or other subsidies, but the bulk of the purchase price must be met with other financing. For example, the city of Dallas offers buyers below 80 percent of AMI up to $10,000 in down payment assistance. To protect buyers from predatory lending practices, all financing arrangements must be approved by the nonprofit. Two-thirds of buyers in the Dallas ACA program rely on FHA financing. Buyers must complete pre-purchase counseling to be eligible to buy a home through the program.
Anecdotally, the ACA program has been effective in stabilizing communities that previously had high rates of foreclosures. The program’s success is now being challenged, however, by the fact that lenders are tightening credit standards, which makes it difficult for prospective buyers to purchase rehabilitated properties. This increases the risk to non-profits of participating in the program. The program is difficult to make work in highly distressed communities because of HUD’s rigid pricing structure as well as program rules that determine profit and loss on a per-property basis rather than at the portfolio level, which would allow a small measure of cross-subsidization. ACA programs rarely include the most distressed neighborhoods because there is no way for participants to cover their costs.
While the ACA’s rigid rules make it a less than perfect model for dealing with today’s foreclosure crisis, the program has provided those who have used it with valuable experience. ACA participants are starting to apply the knowledge and experience they gained under the ACA program to make bulk purchases of properties from private servicers.
What can we learn from these three experiences? We think there are three big lessons. First, achieving the balance among maximizing returns, stabilizing markets, maintaining or enhancing affordability, minimizing government outlays, and getting the job done quickly is hard—and pressures are constant to accomplish all five goals simultaneously. Second, the entity that undertakes the task must have multiple skills—asset manager, property manager, contract manager, financier, and coalition builder among them. Third, the job will require flexibility and innovation. Both the HOLC and the RTC, maligned as “cowboys” in their day, are in retrospect remembered as flexible and innovative entities. In contrast, the rigidity of the ACA program has limited its utility.
Any new disposition program will face not only the old questions, but new ones that reflect changes in both sensibility and law since even the early 1990s. For example, what will be the impact of modern landlord-tenant law on strategies that envision use of single-family properties temporarily for rental housing? Are income-based eligibility requirements appropriate in all cases, and if not, how should they be used? What is the effect of such criteria on the communities in which the homes are located; to what extent have these been mixed-income communities, and is the maintenance or expansion of mixed-income neighborhoods a goal? How do we best harness the capacity and discipline of the for-profit sector alongside the non-profit and governmental sectors? And finally, can we abide the “cowboy” flexibility and innovation that a temporary entity might be able to exercise—and can we find anyone in this day of “gotcha” politics and journalism who would take the job of leading such an entity?
To state the obvious, none of this will take place in a political vacuum. We began by acknowledging that the properties being disposed of came into public hands as a result of taxpayer funds used in an earlier transaction. With public funding comes not only public purpose but politics. Congressional pressure to liquidate the HOLC as its outstanding loan balances declined meant selling off mortgages to local banks sooner than a profit-maximizing strategy would have dictated. The RTC was under constant political pressure to get the job done quickly with no money, which definitely influenced the Corporation’s strategy and operations. And of course the RTC was subject to three arguably inconsistent mandates: maximizing profit, minimizing local market distortion, and enhancing affordable housing. It’s probably too much to ask, but there is something to be said for setting priorities in law, rather than leaving it to the bureaucracy to figure out.
In the end, the question is what are our overall goals. What types of communities do we envision ten years from now in the places with large numbers of foreclosed properties? Can we fundamentally change the boom and bust cycles in these communities that seem to recur in 20-year intervals? How can we use this crisis to improve our stock of quality, affordable rental housing? And—maybe biggest of all—what is the role of homeownership in America?
Download the full report (pdf)
This report now appears in the Federal Reserve of San Francisco’s "Community Development Investment Review" (pdf)