Resilience in the Face of Foreclosures How National Actors Shape
Shared by: fanzhongqing
-
Stats
- views:
- 6
- posted:
- 5/17/2012
- language:
- English
- pages:
- 51
Document Sample
Resilience in the Face of Foreclosures:
How National Actors Shape Local Responses
Draft
Not for Quotation or Attribution
By
Todd Swanstrom,
University of Missouri, St. Louis
Prepared for the
Brookings-George Washington University-Urban Institute Conference
Urban and Regional Policy and Its Effects
May 20-21, 2010
1
Introduction
In recent years the United States has witnessed mortgage foreclosure rates not seen
since the Great Depression. According to one estimate, 13 million American homes will be
foreclosed upon by 2014.1 The problem may be getting worse. RealtyTrac reports that
foreclosure filings — default notices, scheduled auctions, and bank repossessions — were
reported on 932,234 properties in the first quarter of 2010, a 7 percent increase from the
previous quarter and a 16 percent increase from the first quarter of 2009.2
A mortgages foreclosure is basically the failure of a borrower to meet the terms of a
contract with the lender, at which point the property is sold to pay off the loan. If no one bids
high enough to pay off the loan, the lender takes over the property. Although mortgages are
private contracts, their failures have broad public effects. The recent rash of foreclosures has
had a huge impact on the national and even the international economy. Investors in mortgage-
backed securities lost hundreds of billions in equity capital when the mortgages went bad.
These investments were leveraged 10 to 25 times through a variety of complex financial
instruments, such as credit default swaps, which meant that foreclosed mortgages pulled
trillions of dollars of equity capital out of the American economy. As credit froze, the economy
plunged into a deep recession.3
Besides the macroeconomic effects, foreclosures also have tremendous effects on the
local neighborhoods, cities, and counties in which they foreclosed property is located. Research
has shown that foreclosures decrease the market value of nearby properties, disrupt social
1
Hatzius and Marschoun (2009); as reported in Standaert and Weed (2010).
2
RealtyTrac (2010). The Wall Street Journal, however, reported that delinquencies declined in February and
March (Simon, 2010).
3
Immergluck (2009, p. 159.
2
relations, increase crime, and impose fiscal stress on local governments.4 In this paper I say
little about the impact of foreclosures on the national economy and I only deal with the impact
on individuals and families indirectly. My focus is on the effects of foreclosures on places
(neighborhoods and cities) and the efforts by local actors to minimize damage to places from
the foreclosure crisis.
I frame my analysis using two broad literatures: regional resilience and federalism. I
both draw on these literatures and critique them using the example of local responses to
foreclosures. The literature on resilience, I argue, provides useful concepts for understanding
why some local actors have been able to respond more effectively to the foreclosure crisis than
others. On the other hand, I argue, the resilience literature tends to treat challenges, such as
foreclosures, as externally driven and intractable, when in fact local actors are often able to
shape the challenge before it hits them. In this case, local actors tried, but largely failed, to
address the root cause of the foreclosure crisis, subprime lending. The federalism literature,
likewise, give us tools for understanding how political actors engage in venue shopping to find
the most advantageous level of decision making and how decisions at different levels of the
American political system impact policy outcomes. I critique, however, the tendency in the
“new federalism” literature to treat central and local actors as locked in a zero-sum
competition. In contrast, I argue, central rules and policies can empower localities to respond
more effectively to the foreclosure crisis.
The “data” to support my argument comes from the literature on foreclosures and on
case studies of local resilience that I, along with colleagues from around the country,
4
For a synthesis of the research on the local spillover effects of foreclosures, see Kingsley, Smith & Price (2009).
3
conducted. As part of the Building Resilient Regions project, I conducted a series of interviews
in St. Louis and Cleveland and my co-authors did likewise in Chicago, Atlanta, Riverside-San
Bernadino and the East Bay area.5 I have also been fortunate to be present at two gatherings
organized by the National League of Cities at which local practitioners discussed their responses
to foreclosure. 6 I also draw on the growing literature and websites on local best practices by
such organizations as the National Housing Conference, Brookings, and Living Cities. The
evidence for my argument is mostly qualitative and based on case studies. My focus is on what
local actors have done -- on the policy process more than policy outcomes. I have little hard
evidence on whether these local actions were successful in preventing foreclosures or
minimizing their negative effects. Indeed, as Tom Kingsley has noted, “there is a lack of solid
research literature on such policies and their results.”7 With a few exceptions, we simply do
not know what local actions make a difference.8 Notwithstanding the dearth of hard empirical
evidence, I argue that interventions designed to prevent foreclosures and keep families in their
homes are more cost-effective than programs aimed at minimizing the costs of foreclosure
after they occur. Whether I am correct or not about the need for giving more priority to
prevention, we still need much more research on what kinds of prevention and neighborhood
stabilization work is most effective.
Nature of the Foreclosure Challenge: Subprime Lending and Market Disequilibrium
5
See Swanstrom, Chapple, and Immergluck (2009).
6
For the second gathering, in April 2010, I wrote a memo on the St. Louis response to foreclosures and five other
scholars wrote similar memos on their metropolitan areas. I draw freely on these memos. They are available from
the National League of Cities.
7
Kingsley (n.d.); Kingsley, Smith & Price (2009).
8
An important exception is the Urban Institute evaluation of the National Foreclosure Mitigation Program (NFMC)
foreclosure counseling program, which it concludes had positive outcomes. Although the program is federally
funded, it is usually implemented by local NeighborWorks organizations. See Mayer et al (2009).
4
The essence of the foreclosure challenge for places is that it creates an oscillating
imbalance between supply and demand for housing. A period in which demand far outstrips
supply and housing prices soar is followed by a vicious cycle of foreclosures and plummeting
housing prices that leaves a swath of social destruction in its wake. The foreclosure crisis
cannot be understood by linear or one-directional causal analysis. It must be understood in
terms of system dynamics: reinforcing loops of positive feedback that can lock people and
places into cycles of boom and bust that are difficult to control.9
According to economic theory, the free market is self-correcting ; negative feedback
acts like a thermostat guiding the system toward equilibrium: as demand for housing increases,
the supply of housing responds and satisfies the new demand, thus, moderating price increases
and bringing supply and demand back into balance. What happened in the run up to the
foreclosure crisis is that demand for housing, based on a flood of credit and lax underwriting
standards, soared. The supply of housing, being quite inelastic, failed to keep pace with rapidly
rising demand and housing prices began climbing rapidly in most areas. House prices increased
far beyond the pace of economic essentials; for example, monthly housing costs grew as a
proportion of income – a trend which in the long run was unsustainable.10 Normally, if
households take out loans that are unsustainable, they will default and quickly go into
foreclosure.11 In this case, housing price appreciation covered over the unsustainability of the
9
For an introduction to system dynamics thinking, see Richardson (1999).
10
U.S. Department of HUD (2010, p. 38).
11
Of course, another market mechanism that should have prevented the crisis was that risky mortgages destined
to fail should have been unable to be sold on the secondary market, thus choking off mortgage credit and
moderating demand. Federal Reserve Board Chairman Alan Greenspan clearly believed, at the time, that market
mechanisms would correct any imbalances and “the benefits of broadened homeownership are worth the risk [of
foreclosures].” (Greenspan, . He now has admitted he was wrong. The reason why is a long story involving the
5
loans: if borrowers were unable to meet monthly payments, they could simply refinance or sell
the home to pay off the loan. A self-reinforcing cycle of easy credit, rising demand, and soaring
prices was set in motion.
The rash of foreclosures that began in 2006-2007 set in motion a vicious cycle in the
opposite direction: foreclosures suddenly flooded the market with a supply of homes driving
down prices, which left more people vulnerable to foreclosure, further driving down prices
(Figure 1). In addition, burned by foreclosures, lenders tightened their underwriting standards,
further depressing effective demand for homes, reinforcing the cycle of rising foreclosures and
falling prices. And as home prices plummeted, consumer spending fell, increasing
unemployment and tipping more households into foreclosure.
There has been a great deal of debate in the literature on the underlying cause of the
foreclosure crisis. Some argue that it is a classical housing bubble that, when it burst, left
borrowers underwater and, then, according to option-based theory, borrowers ceded their
property to the lender in a so-called “ruthless” default.12 Others point to rising unemployment
and other triggering events as the primary causes of foreclosures. And still others assert that
the Community Reinvestment Act (CRA) pushed lenders into to make loans to borrowers who
lacked the economic resources to sustain homeownership.
A HUD “Report to Congress on the Root Causes of the Foreclosure Crisis” debunks all
three of these explanations.13 Although declining prices are strongly correlated with
foreclosures, they are not the root cause of the foreclosure crisis. Falling home prices simply
securitization of mortgages, the invention of complex new ways of insuring pooled mortgages, corrupt ratings
agencies, and other factors. See Immergluck (2009).
12
U.S. Department of HUD (2010).
13
U.S. Department of HUD (2010, p. 1).
6
took away one way for people who had unsustainable loans to avoid foreclosure (refinancing or
selling their home to pay off the loan). Past foreclosure waves were often caused by economic
weakness but the recent wave of foreclosures, the HUD report observes, occurred before
unemployment began to rise (p. 21). In fact, it is more correct to say that foreclosures caused
economic weakness, not that economic weakness caused foreclosures. (Now, of course, we
are facing a second wave of foreclosures, often of prime loans, driven by rising unemployment
and underemployment.14) And government regulations, like CRA, were not a significant cause
of the foreclosure crisis, the HUD report concludes, because the vast majority of loans that
went into foreclosure were originated by lenders who were not regulated by the federal
government.
Lenders take losses Fewer Loans Fewer
Homebuyers
Subprime
Foreclosures More houses Declining Home
loans
for sale Values
Higher Less Consumer
unemployment Spending
Figure 1.
Vicious Cycle of Foreclosures
Rejecting the bursting housing bubble, the faltering economy, and misguided federal
regulations as primary causes, the HUD Report to Congress concludes that “highly risky loans …
were the root cause of the current crisis.” In short, the recent foreclosure crisis was driven by a
14
Edmiston (2009).
7
huge increase in risky loans that were doomed to fail. The market share of risky loans
(subprime, Alt-A, and home equity loans) increased from 16 percent in 2003 to 48 percent in
2006.15 Collapsing housing prices suddenly left people vulnerable to foreclosure, but the
collapse of housing prices was not the driver of the crisis. In fact, the authors of the HUD report
stress, rising prices were themselves significantly caused by the flood of easy credit and loose
underwriting standards. When enough borrowers defaulted on their loans and went into
foreclosure, the whole house of cards collapsed into a vicious cycle in which supply exceeded
demand and plummeting house prices and rising unemployment exposed more people to
foreclosure (Figure 1). In short, the rapid rise of risky subprime loans was the driver of the
foreclosure crisis.
The Spatial Dimension of the Foreclosure Challenge
Local actors have a great deal of responsibility for dealing with the costs of foreclosure
but relatively little power over foreclosures. The positive feedback loops discussed above that
have driven foreclosures can also take root in neighborhoods where the foreclosures first
appear, setting in motion processes of disinvestment and decay. Although local actors bear
many of the spillover costs of foreclosure, subprime lending was permitted and encouraged by
state and federal governments, which created what Ed Gramlich called “a gigantic hole in the
supervisory safety net”.16 Subprime lending was further enabled by Wall Street investors who
bought up subprime loans without properly evaluating the risks. Local governments and actors
had responsibility for cleaning up the mess from the spillovers of foreclosures, but they had
little power to prevent subprime lending.
15
U.S. Department of HUD (2010, p. 25).
16
Gramlich (2007, p. 21).
8
The literature on the local spillover effects of foreclosures is not huge but it is
enough to identify the major negative externalities of foreclosures with confidence.
Foreclosure spillovers fall into three categories: 1) declining property values; 2) crime and
social disorder; 3) Local government fiscal stress and deteriorating services.17 This is not the
place to discuss the full range of local spillovers, but a quick survey will demonstrate that the
local effects are significant indeed.
Declining Property Values
The impacts of foreclosures can be Scholarly research has consistently found a negative
impact on the market value of homes located within approximately 1/8 of mile (660 feet) of a
foreclosure. The Center for Responsible Lending estimates that by 2012 ninety-two million
households will suffer declines in property values totaling $1.2 trillion.18 One of the most
broadly based studies covering 628,000 repeat sales transactions in 13 states found a negative
impact of 1.3 percent on properties located within a 300 foot radius of a foreclosure and a drop
of 0.6 percent within the 660 foot radius.19 A study of St. Louis County foreclosures between
1998 and 2007 found a drop of about 1.0 percent in the sales prices of properties located
within 1/8th of a mile of a foreclosure.20 In a study of Chicago, Immergluck and Smith (2006)
found a drop in market value within the 1/8 of a mile radius that varied from 0.9 percent to 1.8
percent. Using the conservative Immergluck and Smith estimate (0.9 percent), the Center for
17
This categorization is taken from Kingsley, Smith, and Price (2009), a valuable synthesis of the research on the
local spillover effects of foreclosures.
18
Standaert and Weed (2009, p. 72).
19
Harding, Rosenblatt & Yao (2008).
20
Rogers and Winter (2009).
9
Responsible Lending (CRL) calculated that on average homeowners located near a foreclosure
will lose about $5,000 in value. 21
Social Disorder and Crime
By creating involuntary moves, foreclosures disrupt the community fabric and deplete
social capital. Children are in many ways the most disturbing innocent victims of foreclosure.
When they are pulled out of one school and put in another their learning is disrupted.
According to one estimate, 1.952 million children were impacted by foreclosures as a result of
subprime loans made in 2005-2006.22 This imposes huge costs on families, communities, and
schools. Frequent residential moves can increase violent behavior in high school by 20 percent
and reduce the chance of graduating from high school by more than 50 percent.23 Foreclosures
can also damage the health of children, negatively affecting diet and healthy body weight. In
addition, foreclosures put stress on families, leading to higher rates of divorce, child abuse, and
addictive behaviors.
Foreclosures are also associated with crime. According to “broken windows” theory,
signs of disorder in a neighborhood, as small as a broken window that is not fixed, encourage
crime.24 Foreclosures, especially when the home lies vacant and not properly maintained (such
as tall grass or trash out front), can create a sense of disorder in a neighborhood and encourage
crime. Research has confirmed this connection. A study of Chicago found that for each 1
percentage point increase in the foreclosure rate, the number of violent crimes in a census tract
21
Immergluck and Smith (2006A); Center for Responsible Lending (2008).
22
Lovell & Isaacs (2008).
23
Rumberger (2003).
24
Kelling & Coles (1996).
10
will increase by 2.33 percent.25 A study by the Charlotte-Mecklenburg Police Department found
that both violent and property crime rates were higher in the high-foreclosures neighborhoods
compared to neighborhoods that had not yet experienced high rates of foreclosure.26
Local Government Fiscal Stress and Deteriorating Services
The impact of foreclosures on local governments is double-edged—both on the revenue
and expenditure side. On the revenue side, most analysts agree that foreclosures will decrease
revenues. The impact of foreclosures on tax revenues, especially property taxes, is significant.
Once assessments catch up with declining property values, governments will be forced to
increase local tax rates to maintain property tax revenue. Additionally, local governments will
have to deal with increased tax delinquencies and failure to pay utilities.
At the same time that revenues are declining, local governments are bound to face
increased costs as employees miss work, productivity declines, and families seek additional
social services. These indirect costs are difficult to estimate but researchers have measured the
direct costs of foreclosures on local governments. The best scholarly study on the cost of
foreclosures to municipalities examined Chicago, quantifying costs of foreclosure based on five
scenarios.27 If a property goes into foreclosure and is quickly put back on the market, the
researchers estimate it will cost the local municipality only $430. At the other end, if the
foreclosure leads to vacancy, abandonment, and fire, ultimately requiring demolition, the cost
to the local municipality soars to $34,199. Based on the Apgar and Duda study, the Joint
Economic Committee of the U.S. Congress estimated average costs at $19,227.
25
Immergluck & Smith, 2006B.
26
Bess (2008).
27
Apgar & Duda (2005).
11
Subprime
Foreclosures Glut of Declining Home
loans
houses on Values
market
Vacancy &
Rising Crime abandonment
Government Fiscal Stress:
Rising Taxes & Declining Services
Figure 2.
Foreclosures and Reinforcing Processes of
Neighborhood Decline
As the Apgar and Duda study suggests, the costs of foreclosure vary significantly from
place to place. The negative impact of foreclosures on places depends on foreclosure density
and the strength of the local housing market. Dispersed foreclosures in strong housing markets
have minimal effects. Concentrated foreclosures in weak housing markets can have massive
effects. In strong markets, the housing market will soak up the relatively small number of
foreclosed properties and quickly put them back on the market. Other than an unusually large
number of “For Sale” signs, a casual observer driving through the neighborhood would see little
change.28 On the other hand, concentrated foreclosures in weak markets can create positive
feedback loops that reinforce neighborhood decline (Figure 2).
28
But the costs to individuals and families can still be considerable. Indeed, in strong market metropolitan areas
foreclosures may be harder on families because affordable replacement housing is more difficult to obtain than in
a weak market. For a review of the relatively thin scientific research on the effects of foreclosures on families and
individuals, see Kingsley, Smith & Price (2009).
12
Reinforcing processes of neighborhood decline are rooted in the fact that real estate
markets are highly social or interactive. The self-correcting equilibrium of market economics is
based on rational actors making independent decisions. If the housing prices fall below the
inherent characteristics of the neighborhood (condition of stock, amenities, location relative to
jobs, etc.), then presumably buyers, seeing an opportunity, will jump in and bring prices back up
to equilibrium. But as the growing literature on behavioral economics has taught us, market
actors do not always act rationally or independently.29 Economic decision makers are prone to
social contagion in which they follow the herd rather than make cold-hearted rational decisions
based on the merits of each case. This is especially true in the case of real estate markets.
Admittedly, market fundamentals cannot be ignored indefinitely. Market fundamentals always
reassert themselves in the long run. But as Keynes once remarked: “In the long run we’re all
dead.” It is like having a thermostat that brings your home back to 70 degrees, but only after
forcing you to swelter in 90 degree heat for days.
This herd-like destabilizing behavior was clearly evident in both the inflating of the
housing bubble and its subsequent collapse. Real estate markets are especially prone to social
contagion because the value of property depends so much on what happens to the parcels
surrounding it. The guiding principle of real estate is “location, location, location.” When
foreclosures occur in sufficient numbers in a neighborhood with weak market demand, they
can tip the neighborhood into a reinforcing cycle of disinvestment. Just as housing prices
climbed beyond economic fundamentals during the period of loose credit, housing prices can
29
For an insightful application of behavioral economics to the foreclosure issue, see Barr, Mullainathan and Shafir
(2008).
13
also fall further after foreclosures than would be expected given the economic fundamentals of
the area.
The recent wave of foreclosures flooded across a metropolitan landscape that varied
tremendously in market strength. Some metropolitan areas, driven by robust job growth and
high levels of immigration, have a perpetual shortage of housing relative to demand. As a
result, the so-called sand states experienced huge run ups of housing prices in the boom
followed by precipitous price declines. Because of the underlying strong market, however,
metropolitan areas in Florida, Arizona, and California have experienced relatively low levels of
vacancy and abandonment due to foreclosures. Even in the strongest regional housing
markets, however, there are pockets of market weakness. Exurban areas with so-called “drive
‘til you qualify” mortgages did suffer from a build-up of REOs.30
Weak market metros are more vulnerable to foreclosures. When combined with
imbalances between supply and demand within regions, the spillover effects of foreclosures
can be severe. Many central cities in the Rustbelt have suffered from an imbalance of supply
and demand for decades resulting in vacancy and abandonment at the end of the filtering
chain. In many metropolitan areas, more houses are built on the suburban fringe than there
are new households in the region. In the 1990s, for example, the City of Buffalo produced 3.89
more units of housing than there were new households formed in the region.31 The inevitable
result was vacant and abandoned housing. Concentrated foreclosures can tip transitional
neighborhoods over into a cycle of disinvestment, vacancy, and abandonment.32
30
Immergluck (2009).
31
Bier & Post (2006).
32
Goldstein (2008); Mallach (2008).
14
Foreclosures are not randomly distributed across metropolitan areas. Subprime
lending, which is highly correlated with foreclosures, was highly concentrated. Foreclosures are
correlated with each other in space. Subprime loans, the root cause of the crisis, were not sold
they were aggressively marketed – especially in minority neighborhoods. They were not
marketed to the poorest areas but to moderate and middle-income minority neighborhoods.
Many of these neighborhoods had experienced revitalization, often due to the hard work of
community development corporations. Ironically, this hard work in pushing up market values
may have attracted subprime lenders.33 Concentrated foreclosures can set in motion a cycle of
decline that can wipe out decades of community development that built up assets: physical
capital, social capital, and financial capital.34
Abandonment
Vacancy
Sheriff’s
Sale
Foreclosure
Notice
Risky lending Delinquency
33
For example, Slavic Village, the epicenter of the foreclosure crisis in Cleveland, had one of the most active and
successful CDCs in the area.
34
Homes equity represents 60 percent of the total wealth of middle class households. Before the recent wave of
foreclosures, African American household wealth was less than one-tenth of white households (Shapiro 2004, p. 47
and 107). There is every reason to believe that the wave of foreclosures wiped out proportionally more home
equity among African American households than among whites. (See Kochar and Gonzalez-Barrera (2009).)
15
Magnitude of
Spillovers
Stronger Housing Market Weaker Housing Market
Figure 3. Rising Spillovers in Weak Market Areas
In conclusion, the fundamental challenge of foreclosures for places is that they can
cause severe imbalances of supply and demand for housing. Foreclosures have substantial local
spillover effects and when they are concentrated in weak market areas, they can set in motion
destructive cycles of decline. As Figure3 depicts, the weaker the housing market the greater
the likelihood that foreclosures will result in vacancy and abandonment and the greater the
magnitude of the spillover effects. Weak market areas are especially vulnerable to social
contagion from foreclosures. In such areas, minimizing the spillover effects after foreclosures
have occurred is difficult. Local actors may be forced to let the “disease” run its course but in
the process much of the physical, social, and household assets accumulated in that place will be
destroyed. Inoculating places against the contagion would make more sense. But as we will
demonstrate in the following section, local actors, who are responsible for dealing with the
spillover effects of foreclosures, had little power to control risky lending, the primary cause of
the recent wave of foreclosures.
Foreclosure Prevention: Controlling Subprime Lending
16
The most effective way local actors could have responded to the foreclosure crisis
would have been to limit subprime lending within their borders. When cities acted to regulate
subprime lending, however, courts ruled that they lacked home rule powers for that purpose or
that state laws preempted local laws.35 Similarly, when states stepped in regulate subprime
lending, they were preempted by federal laws. Finally, competing federal regulators engaged in
a “race to the bottom” that gutted regulation of subprime lending. Successful venue shopping
by the mortgage lending industry within the federal government and across federal, state, and
local governments undercut efforts to regulate subprime lending.36
The origins of subprime lending go back to a 1980 federal law which phased in the
abolition of state usury limits on first mortgages by depository institutions.37 Part of the
rationale was that allowing higher interest rates would increase access to credit by higher risk
households and give them a chance to become homeowners. In 1982 another federal law
allowed mortgage companies, that originate loans and sell them to investors, to opt for federal
regulations rather than the regulations of the state in which they were located.38 The vast
majority of subprime loans were originated by mortgage companies. In 1994 the federal
government did enact the Home Ownership and Equity Protection Act (HOEPA) which
prohibited certain loan terms and practices on high-cost loans. The definition of “high cost”,
however, was so high (generally, 8 percent above the rate on comparable Treasury securities)
35
Frug and Barron (2008, p. 196).
36
The story of the failed efforts of local, state and federal regulators to limit subprime lending is told in U.S.
Department of HUD (2010) and more completely in Immergluck (2009). I draw freely from these accounts. For the
concept of “venue shopping” see Baumgartner and Jones (1993).
37
Depository Institutions Deregulation and Monetary Control Act (DIDMCA).
38
Alternative Mortgage Transaction Parity Act (1982)
17
that it applied to very few loans. In 2006, for example, less than 0.1 percent of refinancing and
home improvement loans fell under the law.39
Subprime lending, greenling as opposed to redlining, began to emerge as a problem in
the 1990s and in 1999 North Carolina became the first state to enact comprehensive anti-
predatory lending legislation. The law was modeled on HOEPA but set the threshold much
lower so that it applied to large segment of subprime loans. Other states followed suit so that
by 2007 only seven states had no mini-HOEPA statutes on the books to regulate subprime
lending. 40 These laws were generally ineffective, however.41 Many state anti-predatory
lending laws were weak but another reason why they were ineffective is that they were
preempted by federal regulators. Federal regulators receive part of their funding from fees
paid by the institutions they regulate. Financial institutions have some freedom to choose
which regulator they will fall under and they shop for the best deal. Federal regulators
competed to see who could offer the most preemption of state mortgage regulations – and
therefore attract the greatest number of members and highest fees. In 1996, the Office of
Thrift Supervision (OTS) issued a sweeping regulation preempting all state laws regulating
mortgage lending for federal savings institutions. Under pressure from its members, in 2004
the Office of the Comptroller of the Currency (OCC) issued a similarly sweeping preemption
regulation for all national banks. Both agencies exempted mortgage banking operating
39
Avery, Brevoort & Canner (2007) as reported in U. S. Department of HUD (2010, p. 34).
40
Bostic, et al (2007) as reported in U. S. Department of HUD (2010)).
41
According to one study passage of these laws was actually associated with an increase in subprime lending
(Bostic, et al, 2007).
18
subsidiaries, which were more involved in subprime lending, from state regulation as well. In
2007, the U.S. Supreme Court upheld this extension of preemption powers. 42
As is often the case in our federal system, when the federal and state governments
withdrew from effective regulation of subprime lending, local governments moved in to fill the
gap. A 2005 study identified eighteen local anti-predatory lending laws.43 For example, Chicago
and Cook County enacted anti-predatory lending laws in 2000 and 2001, respectively, which
sought to withdraw municipal business and deposits from firms that engaged in predatory
lending. However, most local laws were preempted by state legislation or state court ruled that
they lacked authority to regulate financial institutions. An excerpt from a dissenting opinion in
a California Supreme
Court case that invalidated Oakland’s pioneering predatory lending legislation clearly expresses
why local governments should have authority in this sphere:
Oakland’s particular interest in regulating subprime loans goes beyond merely
protecting its particularly vulnerable citizens …. Predatory home mortgage lending has
enormous impacts on targeted neighborhoods. … “Foreclosures, especially in low- and
moderate-income neighborhoods turn what might be typically viewed as a consumer
protection problem … into a community development problem, in which increased
foreclosures lead to property abandonment and blight.”44
42
Immergluck (2009, p. 179). Besides refusing to regulate the origination of subprime loans, the federal
government also promoted the securitization of subprime loans which provided massive new capital streams for
subprime lenders and also made it more difficult for subprime loans to be modified, as I discuss later. Immergluck
(2009) tells this story in some detail.
43
Pennington-Cross and Ho (2005).
44
Quoted in Frug and Barron (2009, p. 197).
19
In short, government efforts to regulate subprime lending, the root cause of the
foreclosure crisis, were gutted by state preemption of local laws, federal preemption of state
laws, and a race to the bottom by federal regulators. Most subprime loans were originated by
mortgage brokers who were lightly regulated by the states and federal government. After
foreclosures led to huge losses by investors in pools of mortgages containing subprime loans,
subprime lending largely came to a halt, because investors refused to buy them any longer. By
that time, however, subprime loans were going into default all over the nation. Preventing
foreclosures and their massive spillovers on local communities is much more difficult after
subprime loans have been originated than before. In any case, that is the situation local
authorities found themselves in.
Foreclosure Prevention in the Wake of Subprime Lending: Loan Modifications
Strong arguments can be made for limiting foreclosures both from the viewpoint of
private investors and the public. Given the transaction costs and loss in home value following a
foreclosure, the investors holding mortgage-backed securities would often be better off if the
loan were modified to enable the borrower to stay in the home making lower monthly
payments. Many loan modifications are win-win for the lender and the borrower. The public
has an additional interest in preventing foreclosures, given their large spillover effects,
especially for weak market areas with concentrated foreclosures. Despite the powerful private
and public motives for limiting foreclosures, the rate of loan modifications has been
disappointingly low. Local actors, particularly housing nonprofits, aided by federal grants, have
been resilient in responding to the need for foreclosure prevention by creating housing
counseling networks with broad public outreach. However, vertical disintegration of the
20
mortgage market has pulled key decisions about loan modifications into the hands of remote
servicers constrained by hyperrigid mortgage pools.45
Investors in mortgages lose a great deal when the mortgage goes through foreclosure.
Estimating these losses is difficult because they can vary so much in different parts of the
country and under different scenarios. Clearly, however, the losses are high. Loss severities
have been estimated at about 50 percent for prime mortgages and 70 percent for subprime
mortgages.46 According to one study of subprime loans going through foreclosure, about 42
percent of the loss was due to legal fees, sales commissions, maintenance expenses, and
missed mortgage payments. About 44 percent of the loss was a deadweight loss – a decline in
the value of the home for which there is no compensating gain by anyone else.47 Part of the
loss is due to the general decline in prices. Part is due to vandalism and stripping. And a
significant part is due to the stigma that a foreclosure sale puts on a property – which is
estimated at about 28 percent below equivalent standard sales.48 A study of 900 subprime
loans estimated the cost to investors of foreclosure at over 50 percent. With an average
principal balance of $190,000 that means a loss of about $95,000 on each foreclosure. That
leaves a lot of room to work out a win-win loan modification with the borrower.
The deadweight loss to private investors is much more severe in weak market
metropolitan areas. In the City of Cleveland, for example, most properties that go through
foreclosure are taken over by the bank, so-called real-estate owned properties (REOs). In 2007
properties sold by banks in Cleveland fetched only 13 percent of their estimated market value
45
Hyperrigid is not my term but is take from Gelpern & Levitin (2009).
46
Cordell, et al (2009, p. 7).
47
Cordell et al (2008, p. 12)
48
Cordell et al (2009, p. 8).
21
before foreclosure filing. In 2008, 80 percent of properties sold out of REO in Cleveland’s East
Side sold at extremely distressed prices of less than $10,000.49 In the Cleveland area property
losses by foreclosed properties now exceed a billion dollars. Investors should be highly
motivated to prevent foreclosures and cut their losses.
The public sector has an additional motive to prevent foreclosures because of the
massive negative spillovers discussed earlier. It would especially make public policy sense to
prevent foreclosures in transitional areas where risky loans are concentrated and their
foreclosure could set in motion reinforcing cycles of decay. Some of these external costs are
quantifiable, such as loss by nearby homeowners, but many of these costs are difficult to
quantify. The existence of nonmonetary benefits of stable homeownership, however, is
indicated by the fact that homeowners put a value on staying in their homes even when it does
not make economic sense. Studies have shown that only a small minority of homeowners who
are underwater, owing more on their homes than they are worth, default and go into
foreclosure. A study of all homeowners in Massachusetts, for example, found that in the early
1990s only 6.4 percent who were underwater ended up in foreclosure.50 This suggests that
homes are not just a financial investment but have substantial “use values” that are
nonfungible and nonportable, such as belonging to a community, accumulated social capital,
and a satisfying daily round.51 Parents, with good reason, are reluctant to pull their children
out of local schools in order to achieve a financial benefit. In short, the economic and
49
Coulton, Schramm & Hirsch (2010).
50
Cited in U.S. Department of HUD (2010, pp. 15-16). The HUD report cites a number of other studies to support
this finding.
51
For an explication of the concept of use values, see Logan and Molotch (1987).
22
noneconomic spillovers of foreclosure make a powerful argument for public interventions to
prevent foreclosures.
The main argument against public intervention to prevent foreclosures is the risk of
creating a moral hazard: helping homeowners who took on risky debt could encourage more
risky behavior in the future. This is a valid concern. It is very difficult to estimate the effect of
loan modifications on future household behavior but the substantial benefits of loan
modifications seem to outweigh concerns about moral hazards. First, the explosion of risky
mortgages appears to be less related to risky consumer behavior and more connected to
deceptive lending practices by mortgage brokers. Sudden risky behavior by borrowers does
not explain the timing of the crisis; the emergence of risky and exotic mortgage products does.
Many borrowers did not know what they were agreeing to when they signed a mortgage.
Education of borrowers about the risks subprime lending is helpful but given the complexity of
mortgage instruments, the best way to limit future risky lending is to regulate risky and
unsustainable lending practices, such as negative amortization, yield-spread premiums, no-doc
loans, prepayment penalties, and exploding ARMs. Given the seriousness of the consequences
of concentrated foreclosures it makes sense to act quickly to minimize the damage. A
firefighter does not first ask whether the homeowner was smoking in bed before putting out
the fire.
In fact, local actors have been highly resilient in responding to the challenge to keep
people in their homes. Resilience can be defined as the ability to respond to a challenge by 1)
redeploying assets or expanding organizational repertoires; 2) collaborating within and across
public, private, and nonprofit sectors; 3) mobilizing or capturing resources from external
23
sources.52 Using this definition local housing nonprofits have been quite resilient. They shifted
employees from housing rehabilitation and other activities to foreclosure counseling. This
effort was greatly aided by the National Foreclosure Mitigation Counseling (NFMC) program
funded by the federal government and administered through NeighborWorks America. NFMC
counseling agencies provide free counseling to homeowners who are in trouble, not only
helping them to avoid foreclosure but also advising them in the event of a foreclosure on how
to find replacement housing and access social services. Since 2007 approximately $410 million
has been invested in this program mostly for grants to local counseling agencies that receive
between $150 and $350 per client depending on the extent of the counseling.53 As of
November 2009 762,284 clients had been served.54 Standards have been established to insure
quality. Counselors become HUD-certified by receiving training in the intricacies of the home
financing, usually funded by scholarships.
A good example of resilient local efforts to prevent foreclosures is the Homeownership
Preservation Initiative (HOPI) in Chicago. Begun in 2003, HOPI included Neighborhood Housing
Services of Chicago, many local CDCs, nonprofit counseling organizations, legal services
providers, the City of Chicago, and local foundations. HOPI used the City of Chicago’s 311
nonemergency hotline to link homeowners to counselors. By May 2008 HOPI reported it had
prevented 1,700 foreclosures. Foreclosure prevention has been vigorous in other metropolitan
areas, as well. In 2005 Cuyahoga County launched a foreclosure initiative that included nine
housing nonprofits, numerous municipalities and a number of lenders. Funds from Temporary
52
Swanstrom, Chapple and Immergluck (2009, p. 4).
53
For helping with HAMP applications, counseling agencies can now receive up to a maximum of about $500, it is
my understanding.
54
As reported in the Fourth Congressional Report accessed at:
http://www.nw.org/network/nfmcp/congressional_reports.asp.
24
Assistance to Needy Families (TANF)and fees from tax delinquent properties were shifted to
foreclosure counseling. The County invested almost $2.5 million in the program up to 2008 and
raised about a half a million dollars from local foundations and banks. Between March 2006
and February 2007 the program reported preventing 1,497 foreclosures. 55 The City of St.
Louis has now invested more than $1 million in foreclosure prevention, including the creation
of a foreclosure rescue fund that can make small payments to help homeowners secure a loan
modification. Foreclosure counseling agencies in St. Louis partnered with the local public
television station (KETC) to create programs and a local website to encourage people to seek
help. The KETC initiative was later duplicated in twenty-five cities around the country with the
help of grants from the Corporation for Public Broadcasting.56
The capacity of local actors to engage in foreclosure prevention varies significantly.
CDCs have been “first responders” in the wake of the foreclosure epidemic. Unfortunately,
many suburban areas hard-hit by foreclosures lack robust networks of CDCs.57 Chicago
responded to this gap by creating a regional HOPI, RHOPI, which was led by NHS, Chicago
Community Trust, and the Federal Reserve Bank of Chicago. The RHOPI foreclosure counseling
task force developed a report mapping gaps in access to housing counseling agencies, which
was used to better target counseling resources. Significant foundation and government
funding for counseling in the Chicago area has filled many gaps. But this is not true in other
metropolitan areas.58
55
Swanstrom, Chapple & Immergluck (2009).
56
Swanstrom (2010).
57
Swanstrom, Chapple & Immergluck (2009) provides GIS maps of the location of housing nonprofits in
metropolitan areas, showing their central city concentration.
58
Swanstrom, Chapple & Immergluck (2009).
25
In 2009, the Urban Institute (UI) published preliminary findings of its evaluation of the
NFMC counseling program.59 While not meeting the gold standard of double-blind
experimental research, the study uses rigorous methods to isolate the effects of counseling. UI
examined approximately 61,000 households that received foreclosure counseling and
compared them to a control group of 61,000 similar households that did not receive foreclosure
counseling. The researchers found that the program had modest effects on homeowners who
were two to three months delinquent on their loans when they started counseling, but more
dramatic effects helping homeowners “cure” an existing default. Counseled homeowners were
60 percent more likely to “cure” a foreclosure than those who had not received counseling.
Additionally, the Urban Institute study concluded that loan modifications received by
homeowners through the NFMC program resulted in significantly reduced monthly mortgage
payments after a loan modification (-$454), thereby further decreasing the likelihood of future
defaults and foreclosures.
Research suggests that foreclosure counseling can improve outcomes and help keep
families in their homes. However, the success of foreclosure counseling at best only chips away
at the margins of the problem. The Urban Institute study applied its estimates to the
approximately 260,000 clients who received counseling in calendar year 2008. They calculate
that 880 additional loans would have gone into foreclosure in the absence of counseling and
2,195 households that were in foreclosure at the time they entered counseling were able to
“cure” the foreclosure as a result of the counseling. They use the estimate of a HUD report that
59
Mayer, et al (2009). Collins, et al (2009) conclude that when offered in conjunction with state
foreclosure prevention policies or programs counseling reduces delinquency rates and foreclosure
filings. They also report on one other study that found a positive association between counseling and
foreclosure cure rates.
26
each foreclosure creates costs of about $37,000. Assuming that each homeowner prevented
from going into foreclosure and each homeowner in foreclosure cured is a foreclosure avoided,
then NFMC counseling had benefits of $113,810,000. As I discuss above, the benefits of
foreclosure prevention are considerably higher than this if you count the nonmonetary costs to
households and places. In any case, even at the maximum cost per client served of $350, NFMC
counseling cost only $91 million – leaving a positive benefit-cost ratio. The benefits of
taxpayer-funded counseling are significant. One study found that about half of households that
went through foreclosure never even communicated with their servicer.60 Early
communication increases the likelihood of staying in your home and foreclosure counseling
increases this. The authors conclude that “the availability of the service [foreclosure
counseling] is low relative to the need.”61 Clearly, we should invest more in foreclosure
counseling. Metropolitan areas that have invested in foreclosure counseling have probably
reaped social benefits far beyond the costs.
The problem with foreclosure counseling is that it does not get close to addressing the
magnitude of the problem. The Urban Institute estimates that 3,095 foreclosures were
prevented through foreclosure counseling. In a year when about 2 million households went
through foreclosure, that is clearly not enough. The main reason for the low rate of foreclosure
prevention is that lenders are reluctant to modify many loans.
In order to address the problem, the Obama Administration launched the Making Home
Affordable (MHA) Program in March 2009 which was designed to help 7-9 million households.
60
Cutts & Merrill (2008, p. 211).
61
Cutts & Merrill (2008., p. 252).
27
Over 85 percent of all mortgages are now covered by the program.62 Intended to strengthen
the ability of homeowners and local counselors to negotiate sustainable mortgage
modifications, the program has two main components—a Home Affordable Refinance Program
(HARP) for homeowners whose loans are held by the government-sponsored entities, and a
Home Affordable Modification Program (HAMP) for modifying loans held by conventional
lenders. Of the two, HAMP has become the more important program for most local
homeowners, and it has also been the more complex and controversial program for the federal
government to implement. The main components of the program are incentives to servicers to
modify loans and requirements to bring monthly payments down to 31 percent of household
income. The bar graphs below show that although over 1.1 million trial modifications had been
initiated by March 31st, but only 230,801 permanent modifications had been completed. The
rate of loan modifications is growing but at this rate it will never come close to meeting initial
expectations.
62
Cordell, et al (2009).
28
Source: Making Home Affordable (2009)
The main reason for the slow pace of loan modifications and foreclosure prevention lies
with the servicers and the nature of mortgage securitization. The resilience literature teaches
that specialization and scale can increase efficiency at the same time that they reduce
resiliency. For example, companies with diverse and redundant supply chains will be less
29
vulnerable to an interruption in the supply chain.63 Local housing nonprofits were small and did
not have a highly developed internal division of labor and therefore they were able to shift
organizational routines and employees to address the foreclosure challenge. Servicers were
just the opposite. Over time, the servicers had become large, specialized organizations highly
efficient at collecting and distributing mortgage payments. “As a result of the consolidation in
the industry, servicers have realized large economies of scale in payment processing and
collections, so that the costs of servicing have trended down over time.”64 Loss mitigation is
not subject to economies of scale and therefore did not fit the business model of servicers.
Servicers were not prepared to take on the task of loss mitigation and loan modifications. They
lacked qualified staff who understood loan underwriting. Moreover, they were distant from
local real estate markets and therefore not in a good position to understand how much
properties were really worth.
The whole process of securitization of mortgages and the pooling and servicing
agreements (PSAs) that govern them make loss mitigation extremely difficult. Over 90 percent
of mortgages initiated in recent years have been securitized.65 Gelpern and Levitin call the PSAs
“Frankenstein contracts” – brilliant human creations that come back to haunt us because we
have no way of stopping the destruction they wreak. “The continuing foreclosure epidemic
holds an important lesson for the future: even where rigidity makes perfect sense for the
contracting parties, widespread barriers to modification can unleash catastrophic social
63
Sheffi (2007).
64
Cordell, et al (2008, p. 15).
65
Gelpern & Levitin (2009, p. 1081).
30
consequences.”66 We want contracts to be difficult to break, the authors stress, however,
“private contracts must not be read to interfere with legitimate public policymaking.”67 They
cite numerous instances, including corporate bonds and policies initiated in response to the
foreclosure crisis of the Great Depression, where contracts were renegotiated in order to
protect the parties involved and the broader society.
The rigidities of PSAs come in many different forms. PSAs sometimes simply limit the
number of mortgages that can be modified. Another problem is that the mortgage pools are
owned by thousands of investors from all over the world and in many cases modifying loans
requires the consent of each investor to modify its right to receive principal and interest
payments. Modifying a particular loan can affect different investors different because they
invest in different “tranches” or slices of the overall pool. The result can be so-called “tranche
warfare.”
Servicers also have more incentives to proceed to foreclosure or stretch out the process
than they do to modify a loan.68 Servicers are paid a servicing fee which is based on the unpaid
principal balance of the loans in the pool and therefore, other things being equal, they do not
want to refinance loans or reduce the principal owed. Services make money on loans in
default, through late fees and “process management fees.” If a loan goes through foreclosure,
servicers recover all their expenses before any of the investors get paid. Servicers get paid
nothing for the loss mitigation work that they perform. As the HUD report to Congress
concluded: “[T]here is growing consensus that the rules governing securitization can and do
66
Gelpern & Levitin (2009, p. 1080).
67
Gelpern & Levitin (2009, p. 1134).
68
My discussion of servicer incentives relies on Thompson (2009).
31
limit the flexibility of servicers to pursue modifications, even in situations where an aggressive
modification would benefit both the borrowers and the investors.”
In sum, local actors have often been resilient in pursuing foreclosure prevention by
establishing foreclosure counseling networks and engaging in broad collaborations to reach out
to distressed homeowners. Part of the work of foreclosure prevention must be local. Loan
modifications need to be tailored to the individual. Face-to-face counseling is more effective at
generating the trust and exchange of information necessary for a successful modification. The
willingness of a lender to modify a loan will be based partly on the value of the home in a
foreclosure sale and this will require knowledge of local real estate markets. In addition,
targeting outreach to homeowners facing foreclosure requires local knowledge of subprime
lending patterns and methods of reaching the most vulnerable neighborhoods.
There is also a place for central actors to provide resources, training, and standard
procedures – all of which benefit from economies of scale. The federal government’s grants for
local foreclosure counseling have been crucial in providing resources for counseling in hard-hit
localities. The federal government has also assured quality control through training and
certification that has increased trust toward foreclosure counselors. The Obama
Administration’s Making Home Affordable program is well-structured. By providing a common
set of procedures for modifying loans it helps overcome the ad hoc quality that characterized
early loan modification efforts. It even establishes a net present value (NPV) calculation that
should make the decision making transparent as to whether a loan should or should not be
modified.
32
However, the resources for counseling, standard procedures, safe harbor legal
protections, and incentives for servicers have simply not been enough to break through the
perverse incentives of the servicers and rigidities in the PSAs.69 Decisions about which loans to
modify have been pulled out of the hands of knowledgeable local actors and placed in the
hands of national servicers who lack the expertise and the local knowledge to act effectively.
The hands of the counselors have been tied by “hyperrigid” national servicers. Local
foreclosure counselors report that servicers are difficult to contact, lack proper training, lose
paperwork, and delay decision making. The NPV calculation, which is supposed to be
transparent, can in fact be manipulated by servicers who can put higher recidivism rates and
other debatable data into the formula thus reducing the net present value of a modification.70
Just as the authority to regulate predatory lending was pulled out of the hands of local actors
and given to state and federal governments, decision making authority to modify mortgages
has been pulled out of the hands of local decision makers (public and private) and given to
national servicers who are insulated from public accountability by private contracts. Once
again, local actors have responsibility for the consequences of foreclosures but little authority
over them.
The Fog of Foreclosures: Data and Local Empowerment
If my analysis of the foreclosure challenge is correct, interventions need to be targeted
in time and space. We need to act as early in the process as possible before subprime loans
have forced people out of their homes. Second, we need to concentrate our efforts on weak
69
So-called cram-down legislation that would enable courts to include mortgages in bankruptcy proceedings would
probably help to motivate servicers.
70
I would argue that the decision of whether to modify a loan should not just rest on whether it is beneficial to
investors but should take into account the costs to society as well.
33
market areas with concentrated foreclosures which can “tip” vulnerable neighborhood into
processes of contagious abandonment that destroy accumulated place-based assets. For these
reasons, information on the timing and location of the foreclosure is necessary for effective
local interventions. Information is not sufficient for effective responses, which also require
adequate resources, administrative capacity, and political will. In this section I argue that
federal and state policies concerning data are crucial to resilient responses. Without
standardized and sophisticated data, local actors are often in a “fog” – groping forward and
guiding action based on incomplete and inconsistent data. I will only make a few observations
about neighborhood stabilization in the wake of foreclosures; it is simply too large a topic to
cover here. But the following analysis of data and public policy provides an exemplar of central-
local relations that can be extended to the foreclosure policy system more generally.71
Foreclosure Sheriff’s New Owner
Loan Delinquency Vacancy Abandonment
Origination Notice Sale REO
REO Sale
Figure 4. The Stages of Foreclosure
71
I rely heavily in this section on the insights of Newman (2010).
34
The foreclosure challenge is a complex process that rolls out over time and space.
Figure 4 shows nine possible data points over time. There are many more. After a REO sale, for
example, local actors need to know whether the property was bought by an owner-occupier, a
local housing nonprofit, a local investor, or a national speculator. The foreclosure challenge is
constantly changing so responders need up-do-date data to adjust their policy responses to
constant change. If we know the location and timing of different kinds of loans, we can
anticipate foreclosure trends. Subprime loans began to reset to higher interest rates in two
years but Alt-A loans often had resets of about five years. To anticipate when neighborhoods
will be hit with more delinquencies and foreclosures, we need to know how many Alt-A loans
were originated and when they will reset. We also need to be able to disaggregate the data to
small geographies. Housing markets are local and processes of contagious abandonment are
very local. Research has shown that foreclosures damage property values in a circle with a
radius of about 1/8th of a mile or 660 feet. In order to track the spillover effects of foreclosures
we need data at the census tract level and smaller – even down to the block or parcel level.
Good data is essential for bringing the issue to the attention of the public and placing it
on the policy agenda. Kermit Lind called local CDCs “the canary in the mine.” 72 However,
stories of individual suffering are not enough. CDCs are viewed as having a stake in magnifying
the problem. A more objective way to document the problem is needed. As Deborah Stone
has observed, “The process of counting something makes people notice it more….”73 In metro
area after metro area, counting foreclosures, showing their growth over time and spread across
72
Lind (2008, p. 240).
73
Stone (1998, p. 187).
35
the region has been crucial in garnering public attention. GIS maps can be visually arresting. In
St. Louis advocates used “heat maps” of foreclosures to show that the problem is spreading
across the city boundary to the suburbs.
Just counting foreclosures is difficult because of data inadequacies. National data on
subprime lending was not collected until 2004 when HMDA reporting was expanded to include
information on high-cost loans. In retrospect, this data was crucial in putting the issue before
the public and making the connection between subprime loans and foreclosures. This data,
however, does not include ARMs and does not tell who originated the loan -- among other
flaws. Data on completed foreclosures is even weaker. Because the foreclosure process is
regulated by the states, the process varies and the data recorded is often not comparable. In
order to get meaningful data, a great deal of labor-intensive cleaning and compiling of data is
necessary. Lacking a national data depository on foreclosures, the private sector has stepped
in. RealtyTrac is a private firm that is often relied upon for foreclosure counts, but its data is
notoriously uneven and unreliable. Local actors frequently use it anyway because it is free.
McDash Analytics, now Core American Logic, has compiled much more accurate data but it is
very expensive. One of the major reasons why local actors often operate in a “fog of
foreclosures” is because mortgages and foreclosures are essentially a private transaction and
public access to meaningful data is often blocked.
Good data is also needed after loan origination in order to prevent people from going
into foreclosure and being forced from their homes. Research has shown that contacting
people early in the process facilitates better outcomes. Probably because of the stigma of
foreclosure, people often do not take action to avoid foreclosure by talking to their servicer or
36
seeking counseling. As a result, it is important to reach out to people who are facing
foreclosure. In order to do this, counseling agencies need data on where loan delinquencies
are concentrated. Thus, for example, it would be useful to have data on loan delinquencies as a
precursor to a foreclosure notice. The Mortgage Bankers Association conducts a survey of
mortgage delinquencies but it is not disaggregated to small geographies to be useful to
foreclosure counseling agency outreach efforts.
Later in the process it would be very useful if foreclosure counselors had information on
how servicers conduct their net present value (NPV) calculations. If counselors knew how this
calculation was made, they could concentrate their energies on clients who had a good chance
of having their loan modified. The purpose of the NPV calculation under the Obama
Administration’s HAMP program is to standardize the loan modification process and make it
transparent. Unfortunately, the NPV calculation is far from standard and transparent. Large
servicers, whose servicing book value exceeds $40 billion, may use their own default and
redefault rates and they do not have to publicly defend them. Even smaller servicers have the
right to adjust the standard discount rate set by the federal government, up to 2.5 percentage
points, which can have a huge effect on whether a loan should be modified or not.74 Housing
counselors report that they do not understand how the NPV calculations are made for their
clients. They are operating in a fog.75
Data is equally important in the post-foreclosure process. Here, once again, the fog can
be thick. Data on the number and duration of REO properties is an excellent indicator of
market recovery. The accumulation of REO properties in specific areas indicates a weak market
74
Cordell et al (2009, p. 28).
75
Winter and Swanstrom (2010).
37
and possible contagious abandonment. Dan Immergluck, one of the most active researchers on
foreclosures, states the problem bluntly: ”Data on foreclosures and post-foreclosure
properties, such as REO, are not compiled on a regular, uniform bases by any public agency at a
multistate level.”76 Forced to use a private national data base, Immergluck identified that REO
properties tend to accumulate in central cities in weak market metros but in strong market
metros REOs have accumulated in distant “drive-‘til-you-qualify” suburbs. After a property is
sold out of REO, local governments often do not even know who owns it and therefore they
cannot hold the owner accountable for basic costs – such as grass cutting or board up of vacant
properties. Vacancy data would also be very useful in determining which neighborhoods are
vulnerable to contagious processes of abandonment and decay. The United States Postal
Service has data on vacancies but it has many flaws.
The federal government’s Neighborhood Stabilization Program (NSP) requires extensive
data and skillful data analysis in order to be effective. The small size of the program relative to
the need makes strategic targeting of the program all that much more important. Originally
funded at 3.92 billion nationwide, a second phase of competitive grants (NSP2) received an
additional $1.92 billion . This money was slated to be primarily used to buy foreclosed
properties, fix them up, and put them back on the market. To give an idea of the gap between
resources and need, St. Louis County received a total of $16.5 million ($9.3 as a direct allocation
from the federal government and $6.2 million allocated from the State of Missouri). This is
enough money to do 100 properties at $100,000 per home – in a county which had over 4,000
completed foreclosures in 2009. This means the funds could treat less than 2.5 percent of the
76
Immergluck (2008, p. 3)
38
foreclosed properties.77 Spreading the funds around, the so-called peanut-butter approach,
risks wasting taxpayer’s money when rehabilitated homes are overwhelmed by market
dynamics.
The logic of NSP targeting is clear: funds should be concentrated on “transitional”
neighborhoods where foreclosures could tip a previously stable neighborhood into decline.78
This is where the negative spillovers of foreclosure can mount and where policy can get the
biggest bang for the bucks. Identifying transitional neighborhoods is not an easy task. The
Reinvestment Fund in Philadelphia has pioneered using cluster analysis to identify transitional
neighborhoods.79 This requires collecting a wide range of housing data at the block group or
neighborhood level. A statistical analysis is then performed (cluster analysis) to identify which
neighborhoods are performing similarly. Neighborhoods can be arrayed from the strongest to
the weakest, with transitional areas in the middle. Cluster analysis requires a wide array of
parcel-based data and a skilled analyst. The technique is not automatic. It requires knowledge
of local markets in order to identify the number of clusters and final results must be evaluated
in the context of developments not captured by the data, such as where a new school or
grocery store is being built. The obstacles here are both the availability of the data, some of
which must be procured from county assessors who often are technologically backward and
refuse to allow public access to the data.
It worth reiterating that just doing good data analysis is not enough. Targeted funding
must overcome what Anthony Downs called the “law of political” dispersion –the tendency to
77
Of course, if the properties are sold the NSP funds can be recycled so that more properties can be helped over
the long run.
78
See Goldstein (2008) and Mallach (2008).
79
Goldstein & Closkey (2006).
39
spread funds around to enough ward or political jurisdictions to create a minimum-winning
coalition.80 Not only may it make political sense to spread some of the funding around, the
analysis also suggests a rationale for spending funds outside transitional neighborhoods. There
are appropriate strategies and legitimate uses of funds in very strong and weak market
neighborhoods beset by foreclosures: strong market neighborhoods would benefit from
increased housing code enforcement to make sure that foreclosed properties do not create
visual blight and weak market neighborhoods would benefit from land banking and rightsizing
to set the stage for future market recovery.
Under the leadership of the Northeast Ohio Community and Neighborhood Data for
Organizing (NEO-CANDO) at Case Western Reserve University, the Cleveland metropolitan area
has developed one of the best data collection and analysis systems in the nation. NEO CANDO
gathers neighborhood data for 17 counties in Northeast Ohio and makes it available to the
public on its website. Established in 1992, NEO CANDO has a wide support from area
foundations and played a key role in the rethinking of community development in Cleveland
from a bricks and mortar orientation to one focused strategically on fortifying markets in a
handful of transitional neighborhoods. Neighborhood Progress Inc. has developed a Land
Assembly Project that targets funds to six CDCs to create “neighborhoods of choice”. Cuyahoga
County’s application to the federal government for Neighborhood Stabilization Program funds
supports this strategy.
The sophisticated regional data system in Cleveland is more the exception than the rule.
This is because under the present rules of the game developing a sophisticated regional data
80
Downs (1980, p. 530).
40
system is expensive, labor-intensive, and politically challenging. The first problem is that the
private nature of foreclosure data and the fragmented collection of parcel-level data at the
state and county level means that every metropolitan area must reinvent the wheel, cleaning
and organizing the data in an expensive and time-consuming process. Because there is no
profit in doing it, creating a regional data utility requires significant public or foundation
funding. Moreover, much of the parcel-level data is controlled by county assessors who are
concerned about tax rates and assessment practices, not public policy. Many are underfunded
and the data is not digitized, requiring laborious hand entry. Some counties sell the data.
Acquiring access to this data at a reasonable cost and on a regular basis requires cutting
political deals that must be negotiated with each county.81
The solution to the problems of data on foreclosures is a policy of “targeted
transparency.”82 According to the originators of the concept, targeted transparency policies
include five characteristics:
Mandated public disclosure
By corporations or other private or public organizations
Of standardized, comparable, and disaggregated information
Regarding specific products or practices
To further a defined public purpose.83
81
Newman (2010).
82
Fung, Graham & Weil (2007).
83
Fung, Graham & Weil (2007, p. 6).
41
When problems are widely dispersed or locally variable a policy of targeted transparency may
make more sense than government regulation. Government acts as “stewards of transparency
policy by compelling disclosure of needed information when participants cannot obtain it,
fostering common definitions and accurate metrics, and providing feedback and analysis to
encourage transparency improvement.”84
The originators of the concept cite the 1975 Home Mortgage Disclosure Act (HMDA) as a
successful example of targeted transparency. Motivated by fears of redlining and
discrimination in lending, HMDA required federally regulated lenders to report on the loans
they made by census tract; late they were required to report on the race of applicants and loan
recipients. The 1977 Community Reinvestment Act (CRA) required regulated lenders to “meet
the credit needs of their communities.” Advocates of “targeted transparency” stress that the
information needs to be embedded in the decision making routines of organizations.
Community-based organizations used HMDA data to challenge the banks and during the formal
period of the challenge the lending institution was prohibited from merging or opening new
branches. In order to avoid negative publicity and stop the challenge, banks often cegotiated
community lending agreements in which they agree to lend to low-income and minority
neighborhoods. According to the National Community Reinvestment Coalition between 1992
and 2000 the value of CRA agreements totaled $1.09 trillion.85 A study by the Federal Reserve
concluded that CRA opened up new profitable business opportunities for banks and CRA loans
generally performed as well as non-CRA loans.86
84
Fung, Graham & Weil (2007, p. 17).
85
Community Reinvestment Coalition 2001; as cited in Fung, Graham & Weil (2007, p. 251).
86
Board of Governors of the Federal Reserve System (2000); as cited in Fung, Graham & Weil (2007, p. 252).
42
HMDA, originally designed to address the problem of redlining, could be reformed and
expanded to address the problem of greenlining. First, HMDA needs to be extended from
federally regulated institutions to all mortgage lending. Data needs to be reported in a
common format and standardized way to track mortgage lending from loan origination to REO
sales so that it can be disaggregated to small geographies, preferably census tracts or smaller.
In addition, data should be reported in real time so that actors can keep pace with fast-
changing trends. The internet enables data to be shared almost instantaneously with users. In
addition, HUD should work to require county assessors to share parcel-level data in a
standardized and digitized form. There is no reason why every county in every state should
record and store data in different ways. HUD could condition grants on developing statewide
common property reporting requirements. In order to facilitate digitizing data HUD should
consider competitive grants to states or county assessors.87
Once the necessary data is made available in a common format and on a timely basis
there still is the problem of regions developing the capacity to analyze and manipulate the data
in meaningful ways. Regional data clearinghouses, like NEO CANDO, have developed in
metropolitan areas across the country, usually led by universities. Community foundations
have played a key role in developing regional data clearinghouses. However, many regions lack
strong community foundations. National organizations like the National Vacant Property
Campaign, the National Neighborhood Indicators Project, and Living Cities have played the role
of national intermediaries in promoting regional data capacity, but capacity varies
87
HUD will also need sophisticated regional data system to evaluate NSP grants. This will require an understanding
of the NSP interventions and an analysis of housing market dynamics that will enable researchers to isolate policy
effects from normal neighborhood trajectories. See Kingsley, Smith & Price (2009, p. 43).
43
tremendously across regions. HUD should consider start-up grants for regional data
clearinghouses. Modest amounts of money could leverage large increases in technical and
analytical capacity.
Conclusion: The Lessons for Central-Local Relations
Foreclosures are the new urban renewal. Once again, millions of people are being
displaced, roiling the social structure and increasing social tensions. This time, however,
instead of government, the private sector is taking the lead and decision making processes are
much more diffuse. The particular challenge of foreclosures places is that they can destabilize
the balance between supply and demand for housing, setting in motion reinforcing processes of
neighborhood decline that wipe out place-based social, physical, and financial assets. The
nature of the foreclosure challenge requires that it be addresses as early as possible and that
interventions be targeted on concentrated foreclosures in weak market areas. He tragedy of
the foreclosure crisis is that early and targeted interventions were hobbled by existing federal
and state policies.
The most effective action by local actors would have been to intervene early in the cone
of causality, stopping subprime lending before it locked homeowners into unsustainable
mortgages. As we saw, however, preemption of local action by states, preemption of state
antipredatory policies by the federal government, and a race-to-the-bottom by federal
regulators gutted effective regulation of subprime lending, the root cause of the foreclosure
crisis.
Local housing nonprofits were in many ways resilient in responding to the threat of
foreclosures from subprime lending -- shifting employees, expanding organizational repertoires,
44
mobilizing new resources, and establishing innovative partnerships with public and private
organizations. Aided by federal NFMC grants local foreclosure counseling agencies have been
able to facilitate loan modifications and keep some families in their homes. However, rigid
policies by servicers and PSAs that serve almost as suicide pacts have prevented foreclosure
counseling from scaling up to address the magnitude of the problem.
Finally, faced with the mounting spillovers from completed foreclosures local actors
have struggled to cope, but they lack the data and the analytical capacity to identify transitional
neighborhoods where intervention could make the biggest difference, the resources to change
market dynamics, and often the civic and administrative capacity to implement sophisticated
multi-sectoral and cross-silo revitalization policies.
According to a recent Pew Research Center Survey, “Americans are more skeptical of
Washington than ever.”88 Even in wake of the foreclosure crisis and the abuses on Wall Street
public opinion has shifted toward wanting a smaller government with fewer services. Clearly,
we need strong federal regulations to prohibit predatory lending practices across-the-board.
But we also need to shape federal policies that empower local actors to solve their own
problems. Even though positive perceptions of the personal impact of local governments on
their daily lives are down, a majority (51 percent) still view local government’s impact as
positive compared to only 38 percent for the federal government.89 In this paper I have
discussed how the federal and state governments can support resilient local responses to the
foreclosure crisis. First, we need to give local governments the home rule powers to regulate
88
Kohut (2010).
89
Pew Research Center Survey available at: http://pewresearch.org/pubs/1569/trust-in-government-distrust-
discontent-anger-partisan-rancor.
45
mortgage lending that can have huge negative spillover effects within their borders. Effective
responses to foreclosures also require supportive state and federal policies. Higher level
governments need to provide the infrastructure, like interstate highways and water mains,
leaving it to local governments to connect the feeder roads and the water spigots. Standard
information and uniform reporting practices may appear to centralize power in the hands of
federal bureaucrats but, in fact, they create transparency and help to empower local actors.
Lacking supportive state and federal policies, local actors will never be able to bring their
resilient responses up to scale. Far from taking power away from local actors, more vigorous
federal and state policies can empower localities and enhance democratic accountability.
46
References
Apgar, W. C. & M. Duda. 2005. Collateral Damage: The Municipal Impact of Today’s Mortgage
Foreclosure Boom. A report prepared for the Homeownership Preservation Foundation, May 11.
Avery, Robert B., Kenneth P. Brevoort, and Glenn B. Canner. 2007. “The 2006 HMDA Data,”
Federal Reserve Bulletin, Vol. 93, December.
Barr, Michael S., Sendhil Mullainathan, and Eldar Shafir. 2008. “Behaviorally Informed Home
Mortgage Credit Regulation,” in Borrowing to Live: Consumer and Mortgage Credit Revisited,
ed. by Nicolas P. Retsinas and Eric S. Belsky. Washington, D.C.: Brookings Institution Press.
Baumgartner, Frank R. and Bryan D. Jones. 1993. Agendas and Instability in American Politics.
Chicago, IL: University of Chicago Press.
Bier, Thomas & Charlie Post. 2006. “Vacating the City: An Analysis of New Home Construction
and Household Growth,” in Redefining Urban & Suburban America: Evidence from Census 2000,
edited by Alan Berube, Bruce Katz & Robert E. Lang. Washington, D.C.: Brookings Institution
Press.
Board of Governors of the Federal Reserve System. 2000. The Performance and Profitability of
CRA-Related Lending.
Bostic, Raphael W., Kathleen C. Engel, Patricia A. McCoy, Anthony Pennington-Cross, and Susan
M. Wachter. 2007. State and Local Anti-Predatory Lending Laws: The Effect of Legal
Enforcement Mechanisms. Working paper. Social Science Research Network.
http://ssrun.com/abstract=1005423.
Center for Responsible Lending. 2008. Subprime Spillover: Foreclosures Cost Neighbors $202
Billion; 40.6 Million Homes Lose $5,000 on Average.
Collins, J. Michael, Ken Lam, and Christopher Herbert. 2009. State Mortgage Foreclosure
Policies and Counseling Interventions: Impacts on Borrower Behavior. Available at:
http://ssrn.com/abstract=1475505.
Cordell, Larry, Karen Dynan, Andreas Lehnert, Nellie Liand, and Eileen Mauskopf. 2008. The
Incentives of Mortgage Servicers: Myths and Realities. Finance and Economics Discussion
Series, Division of Research & Statistics and Monetary Affairs, Federal Reserve Board,
Washington, DC.
Cordell, Larry, Karen Dynan, Andreas Lehnert, Nellie Liand, and Eileen Mauskopf. 2009.
Designing Loan Modifications to Address the Mortgage Crisis and the Making Home Affordable
47
Program. Finance and Economics Discussion Series, Division of Research & Statistics and
Monetary Affairs, Federal Reserve Board, Washington, DC.
Coulton, Claudia, Michael Schramm, and April Hirsch. 2010. Foreclosure Crisis and Response in
Cleveland and Cuyahoga County. Internal memo prepared for Metropolitan Response to the
Foreclosure Crisis: A Dialogue between Researchers and Practitioners, Center on Urban Poverty
and Community Development, Mandel School of Applied Social Sciences, Case Western Reserve
University.
Cutts, Amy Crews and William A. Merrill. 2008. Interventions in Mortgage Default: Policies and
Practices to Prevent Home Loss and Lower Costs. Freddie Mac Working Paper #08-01.
Downs, Anthony. 1980. “Using the Lessons of Experience to Allocate Resources in the
Community Development Program,” in Housing Urban America, 2md edition, ed. by Jon
Pynoos, Robert Schafer, and Chester W. Hartman. New York: Aldine Publishing, pp. 522-535.
Gramlich, Edward. M. 2007. Subprime Mortgages: America’s Latest Boom and Bust.
Washington, D.C.: Urban Institute Press.
Edmiston, Kelly. 2009. “Characteristics of High-Foreclosure Neighborhoods in the Tenth District.
Economic Review: Federal Reserve Bank of Kansas City, Second Quarter, pp. 51-75.
Frug, Gerald E. and David J. Barron. 2008. City Bound: How States Stifle Urban Innovation.
Ithaca, NY: Cornell University Press.
Fung, Archon, Mary Graham, and David Weil. 2007. Full Disclosure: The Perils and Promise of
Transparency. New York: Cambridge University Press.
Gelpern, Anna and Adam J. Levitin. 2009. “Rewriting Frankenstein Contracts: Workout
Prohibitions in Residential Mortgage-Backed Securities,” Southern California Law Review, Vol.
82, pp. 1075-1152.
Goldstein, I. (2008). Targeting Neighborhood Reinvestment Strategies: Using TRF's Markt Value
Analysis to Strategically Target the A. St. Louis, MO: Presentation to St. Louis Federal Reserve
Strengthening Neighborhoods in Weak Markets Conference.
Goldstein, I., & Closkey, C. S. (2006). Market Value Analysis: Understanding Where and How to
Invest Resources. Bridges , 1-2.
Greenspan, Alan. 2007. The Age of Turbulence: Adventures in a New World. New York: Penguin.
Hatzius, Jan and Michael A. Marschoun. 2009. Home Prices and Credit Losses: Projections and
Policy Options, Global Economics Paper No. 177, at 16 Goldman Sachs Global ECS Research.
48
Ho, Ghiang and Anthony Pennington-Cross. 2005. The Impact of Local Predatory Lending Laws,
Working Paper 2005-049B Federal Reserve Bank of St. Louis.
Immergluck, Dan. 2009. Foreclosed: High-Risk Lending, Deregulation, and the Undermining of
America’s Mortgage Market. Ithaca, NY: Cornell University Press.
Immergluck, Dan & Geoff Smith. 2006A. “The External Costs of Foreclosure: The Impact of
Single-Family Mortgage Foreclosures on Property Values,” Housing Policy Debate, 17 (6), pp.
57-79.
Immergluck, Dan & Geoff Smith. 2006B. “The Impact of Single-Family Mortgage Foreclosures
on Neighborhood Crime,” Housing Studies, 21 (6), pp. 851-866.
Kelling, George and Catherine Coles. 1996. Fixing Broken Windows: Restoring Order and
Reducing Crime in Our Communities. New York: The Free Press.
Kingsley, G. Thomas. n.d. “Policies to Cope with Foreclosures and Their Effects on
Neighborhoods. Forthcoming in Howard Wial and Hal Wolman, eds.
Kingsley, Thomas G., Robin Smith, and David Price. 2009. “The Impacts of Foreclosures on
Families and Communities,” report prepared for the Open Society Institute, Urban Institute,
Washington, D.C. (May).
Kochar, Rakesh and Ana Gonzalez-Barrera. 2009. Through Boom and Bust: Minorities,
Immigrants and Homeownership. Pew Hispanic Center Report, Washington, DC: Pew Research
Center.
Lind, K. J. 2008. “The Perfect Storm: An Eyewitness Report from Ground Zero in Cleveland’s
Neighborhoods,” Journal of Affordable Housing, Vol. 17, No. 3 (Spring), pp. 237-258.
Logan, John R. and Harvey Molotch. 1987. Urban Fortunes: The Political Economy of Place.
Berkeley, CA: University of California Press.
Lovell, Phillip & Julia Isaacs. 2008. “The Impact of the Mortgage Crisis on Children,” First Focus
(May), pp. 1-5.
Making Home Affordable. 2009. Making Home Affordable Program: Servicer Performance
Through March 2010. http://www.makinghomeaffordable.gov/docs/report.pdf.
Mallach, A. (2008). How to Spend $3.2 Billion: Stabilizing Neighborhoods by Addressing
Foreclosed and Abandoned Properties. Philadelphia: Federal Reserve Bank of Philadelphia.
Mayer, Neil, et al. 2009. National Foreclosure Mitigation Counseling Program Evaluation:
Preliminary Analysis of Program Effects. Washington, DC: Urban Institute (November 2).
49
National Community Reinvestment Coalition. 2001. CRA Commitments. Washington, DC:
National Community Reinvestment Coalition.
Newman, Kathe. 2010. “Using Publicly Available Data to Understand the Foreclosure Crisis.”
Journal of the American Planning Association, Vol. 76, No. 2, pp. 1-12.
RealtyTrac, “Foreclosure Activity Increases 7 Percent in First Quarter,” April 15, 2010; at:
http://www.realtytrac.com/contentmanagement/pressrelease.aspx?channelid=9&itemid=8927
accessed 4/19/10.
Richardson, George P. 1999. Feedback Thought in Social Science and Systems Theory. Waltham,
MA: Pegasus Communications.
Rogers, William & Will Winter. 2009. “The Impact of Foreclosures on Neighboring Housing
Sales,” Journal of Real Estate Research Vol. 31, No. 4, pp. 455-480.
Rumberger, Russell. 2003. “The Causes and Consequences of Student Mobility,” Journal of
Negro Education, 72 (1), pp. 6-21.
Shapiro, Thomas M. 2004. The Hidden Cost of Being African American: How Wealth
Perpetuates Inequality. New York: Oxford University Press.
Sheffi, Y. 2007. The Resilient Enterprise: Overcoming Vulnerability for Competitive Advantage.
Cambridge, MA: MIT Press.
Simon, Ruth. 2010. “Delinquencies for Mortgages Decline Again.” Wall Street Journal (April 19).
Standaert, Diane M. and Sarah K. Weed. 2009. “Secure Transactions: Restoring Our
Communities with Responsible Lending,” Journal of Affordable Housing, Vol. 19, No. 1, Fall, 71-
87.
Stone, Deborah. 1998. Policy Paradox: The Art of Political Decision Making. New York: W. W.
Norton.
Swanstrom, Todd, Karen Chapple & Dan Immergluck. 2009. Regional Resilience in the Face of
Foreclosures: Evidence from Six Metropolitan Areas. Building Resilient Regions Working Paper
2009-05.
Swanstrom, Todd. 2010. Foreclosure Challenge and Response: St. Louis City and County.
Internal memo prepared for Metropolitan Response to the Foreclosure Crisis: A Dialogue
Between Researchers and Practitioners, National League of Cities, February 10.
50
Thompson, Diane E. 2009. Why Servicers Foreclose When They Should Modify and Other
Puzzles of Servicer Behavior: Servicer Compensation and its Consequences. Boston, MA:
National Consumer Law Center, October.
U. S. Department of HUD. 2010. Report to Congress on the Root Causes of the Foreclosure Crisis.
Washington, D.C.: U. S. Department of HUD.
Winter, Will and Todd Swanstrom. 2010. The Effectiveness of Foreclosure Counseling in St.
Louis: A Report Prepared for the Metro St. Louis Foreclosure Intervention Task Force and
Beyond Housing. Public Policy Research Center, University of Missouri, St. Louis, March 15.
