11.1 Mortgage Crisis and Great Recession

When a Mortgage Left Minnesota and Came Back as a Foreclosure
Minnesota Geography | Housing, finance, and the Great Recession

When a Mortgage Left Minnesota and Came Back as a Foreclosure

The Twin Cities housing crisis was not only a national financial disaster that happened to arrive here. It traveled through a landscape already divided by race, credit, wealth, and neighborhood history.

Hero image: Minneapolis rally against a home foreclosure, November 2009. Fibonacci Blue, CC BY 2.0, Wikimedia Commons.

A house is profoundly local. It has an address, a sidewalk, a school district, perhaps a maple tree shading the yard. A mortgage can leave the neighborhood almost immediately. By the middle of the 2000s, ownership of a Minnesota loan, or the claim on its payments, could move from broker to lender to a mortgage pool to investors and a servicer. When defaults spread, the consequences returned to one very specific front door.

Reading path

Three questions to carry through the evidence

What can travel? Follow the mortgage contract and payment stream beyond the house.
What stays local? Trace the effects of default at a household, block, and city scale.
What can a map show? Read dated patterns carefully without treating them as proof of a single cause.

First, turn a house into a financial instrument

For most of the twentieth century, Americans often imagined a mortgage as a relationship between a household and a bank. The bank evaluated the loan, collected payments, and had a clear reason to care whether the borrower could repay. That model never described every mortgage, but it kept the chain of responsibility short. By the 1990s and early 2000s, the chain had become longer. Mortgage brokers found customers, lenders originated loans, secondary-market firms bought and bundled them, investors bought securities backed by payment streams, and separate companies serviced the accounts.

Each step could serve a purpose. Brokers connected borrowers with lenders. Securitization moved investment money toward mortgages. Selling loans let lenders make more loans. Investors gained a financial asset. Yet the arrangement could also move responsibility and risk farther from the original lending decision. The household still lived at one address. The financial claim attached to its mortgage had become widely distributed.

Interactive | Follow the mortgage

Watch one local loan enter a financial chain

Choose each stage or play the sequence. The point is not that every participant behaved badly. Responsibility and risk could become divided among institutions that never met the borrower.

HOMEHousehold
$$Broker
BANKLender
POOLLoan pool
BONDInvestor
BILLServicer
The house is local

A household takes out a mortgage on one property. At this moment the geography is simple: household, house, lender, payment.

The later crisis becomes easier to understand once the financial claim can move while the house cannot.
Foreclosed house in 2008
A national crisis with local addresses. A foreclosed California home in February 2008. The architecture changed from state to state; the financial mechanism did not. Brendel, CC BY-SA.
Minneapolis foreclosure protest in 2009
When finance comes back to the sidewalk. A Minneapolis rally against a home foreclosure in November 2009. Fibonacci Blue, CC BY 2.0.

Credit had a geography before subprime lending arrived

The mortgage crisis did not fall onto a blank metropolitan map. Racial covenants, redlining, unequal appraisals, discriminatory sales practices, and suburban exclusion had already shaped the Twin Cities. Historic redlining and discriminatory lending restricted access to conventional credit and investment. During the subprime era, communities long underserved by prime lenders were also heavily exposed to higher-cost lending.

The 2009 University of Minnesota Institute on Race and Poverty report Communities in Crisis captured this transformation while the foreclosure wave was still unfolding. Its central argument was not that every subprime mortgage was predatory or that every borrower of color had been treated identically. Rather, the report found a striking dual market. Prime lenders were less accessible in neighborhoods of color, people of color were much more likely to apply to lenders the report classified as subprime, and racial disparities persisted even among applicants with high incomes.

Chart showing subprime mortgage market growth 1994 to 2006
The market expands. The report's Chart 2 shows the rapid growth of subprime mortgage lending, especially after 2003. Institute on Race and Poverty, 2009.
Twin Cities mortgage denial rates by race and income
Income did not erase the disparity. In 2004-2006, very-high-income Black applicants had a 25 percent home-purchase denial rate, compared with 11 percent for the lowest-income White applicants. Institute on Race and Poverty, 2009.
Redlining restricted conventional credit and investment. Reverse redlining describes a different pattern: targeting communities for costly or risky credit. It does not mean that every subprime loan was abusive.

Income did not eliminate the lending disparity

Mortgage underwriting considers more than income, including credit history, loan-to-value ratios, debt, and employment stability. The 2009 report says this explicitly. Its data cannot identify the reason for each denial or prove intent by a particular lender. Yet the reported racial differences were large enough that income alone could not explain the pattern. In the 2004-06 home-purchase application data, very-high-income Black applicants had a 25 percent denial rate, compared with 11 percent for the lowest-income White applicants.

The neighborhood pattern mattered too. High-income applicants were less likely to apply through prime lenders when the home was in a neighborhood with a larger share of residents of color. That pattern appeared for White applicants and applicants of color, although racial differences remained within every neighborhood type. The historical data document unequal outcomes across both applicant groups and places; they do not, by themselves, explain every lending decision.

Interactive | Hold income high and change the neighborhood

How likely was a high-income applicant to apply to a prime lender?

These are the report's 2004-06 home-purchase application rates for high- and very-high-income applicants. Change the neighborhood, then switch the applicant group.

81% applied to a prime lender
Prime lender
81%
Not prime
19%

For high-income White applicants in predominantly White neighborhoods, prime lenders dominated the application market.

Table of prime near-prime and subprime lender applications by race
The lender mix differed sharply by race. Table 1 shows that 55.5 percent of Black home-purchase applicants applied to lenders the report classified as subprime, compared with 12.1 percent of White applicants. An application does not reveal the terms or outcome of an individual loan. Institute on Race and Poverty, 2009, p. 21.
Mortgage denial rates by race and lending institution type in the Twin Cities
The disparity did not disappear inside prime banks. The report separated prime, near-prime, and subprime lenders and still found higher denial rates for applicants of color within each lender category. Institute on Race and Poverty, 2009.
Twin Cities subprime home purchase loans by race and income, 2004 to 2006
Income changed the odds, but did not erase the racial pattern. High-income Black and Hispanic homebuyers were still more likely to receive subprime loans than White borrowers in any income group. Institute on Race and Poverty, 2009.
Twin Cities home mortgage denial rates by census tract, 2004 to 2006
Before foreclosure, denial already had a geography. Across the wider Twin Cities region, the highest mortgage-denial rates clustered heavily in and around the central-city neighborhoods that also had large populations of color. Institute on Race and Poverty, 2009.
Share of originated Twin Cities mortgages acquired by people of color by census tract, 2004 to 2006
Now hold the metropolitan map still. This map shows the share of originated mortgages acquired by people of color, by census tract, in 2004-06. It makes it possible to compare the geography of lending with the geography of denial. Similarity does not by itself prove discrimination, but it identifies a question worth investigating. Institute on Race and Poverty, 2009, p. 19.

Compare two historical maps

The earlier maps show mortgage denial and the geography of borrowers of color. The next two use the same Hennepin-Ramsey frame to compare subprime lending in 2004-06 with foreclosure rates in 2007. Their boundaries do not match perfectly, nor should they. The repeated concentration in North Minneapolis, parts of St. Paul, Brooklyn Center, Brooklyn Park, and several inner-suburban tracts is still difficult to miss.

Interactive | Keep the geography still

Compare the lending pattern with the foreclosure rate

Use the same regional frame for two different historical measures. The map text names the year and unit. The original map carries the six-bin legend; click to enlarge it before comparing shades.

Share of mortgage loans that were subprime by census tract, 2004 to 2006
Where was subprime lending concentrated?
2004-06 • Percent of mortgage loans classified as subprime, by census tract

Subprime lending was especially common in North Minneapolis, central St. Paul, Brooklyn Center, Brooklyn Park, and several other inner-suburban tracts.

This is a historical lending measure, not a map of current lending conditions. It does not prove why any individual loan was made.
Relationship between borrowers of color and foreclosure rates
The curve bends upward. Foreclosure rates rose sharply in tracts with higher shares of borrowers of color. This is correlation, not proof that race itself caused foreclosure. Institute on Race and Poverty, 2009.
Relationship between subprime lending and foreclosure rates
The financial mechanism appears beside it. Tracts with higher subprime lending rates also had much higher foreclosure rates. Institute on Race and Poverty, 2009.

The shock was concentrated, not evenly shared

North Minneapolis had been shaped by earlier parts of the metropolitan housing story: racial exclusion, disinvestment, urban renewal, unequal access to conventional mortgages, and persistent gaps in household wealth. In the report's mid-2000s comparison, 49 percent of North Minneapolis home-purchase loans and 47 percent of refinance loans were subprime. Elsewhere in the metropolitan area, the respective figures were 14 and 20 percent.

The foreclosure burden was similarly unequal. In the report's 2007 comparison, roughly two foreclosure sales occurred for every hundred owner-occupied units across Hennepin and Ramsey Counties; in North Minneapolis the figure was about twelve per hundred. Clusters also appeared in central St. Paul, Brooklyn Park, Brooklyn Center, and other communities. The pattern documents a historic shock, not the whole story of any neighborhood.

Vacant boarded house in North Minneapolis
Vacancy can be one afterlife of foreclosure. A boarded and condemned house on Logan Avenue North in 2016. Not every vacant house resulted from foreclosure; the image illustrates a possible landscape consequence, not an explanation by itself. Tony Webster, CC BY 2.0.
Near North Minneapolis neighborhood in 2022
A neighborhood is never only its crisis. Near North Minneapolis in 2022. The foreclosure map records one historical shock, not the totality of the community. Fibonacci Blue, CC BY 2.0.

A foreclosure is not one event

For the household, foreclosure can mean losing a home, equity, stability, and perhaps a school or social network. For the block, a vacant house can mean maintenance problems, vandalism, weaker comparable sales, or an investor purchase. For a city, hundreds or thousands of distressed properties mean administrative costs and a weaker property-tax base. The report cited research finding that a conventional single-family foreclosure reduced nearby property values in Chicago, then used those findings to estimate enormous spillover losses in Minneapolis and St. Paul.

This is where the word crisis can mislead us. It sounds like a moment. In geographic terms, foreclosure is more like a wave moving outward from one address through several systems.

Interactive | Follow the ripple

One lost house can touch several geographies

Advance through the sequence. The effects vary enormously from case to case, but foreclosure does not stop at the property line.

1HouseholdPayment shock, job loss, or an unaffordable loan produces default.
2PropertySheriff sale or lender repossession changes ownership and occupancy.
3BlockVacancy and distressed sales can weaken nearby values and stability.
4CityInspection, policing, demolition, and lost tax base create public costs.
5BuyerAn investor may purchase the house at a lower post-crisis price.
6TenureAn owner-occupied home may return to the market as a rental property.
Begin at the mortgage payment

The national financial crisis becomes local when a household can no longer make the payment attached to one address.

How the housing boom became a financial crisis

The national financial crisis had many causes, and no one Minnesota neighborhood caused it. Mortgage underwriting weakened, home prices were assumed to keep rising, adjustable and otherwise risky loans proliferated, securities transformed thousands of mortgages into investment products, rating failures obscured risk, and financial firms became deeply exposed to housing debt. When home prices stopped climbing, refinancing became harder. Defaults rose. Securities built from those loans lost value, credit tightened, construction collapsed, unemployment surged, and a housing problem became the Great Recession.

The Community Reinvestment Act did not force lenders to create the subprime boom. Federal Reserve research found that CRA-related lending was a small share of higher-priced mortgages and did not find CRA to have played an important role in the subprime expansion. The Twin Cities evidence points to a different question: why were conventional prime institutions often less accessible in the neighborhoods that later received so much expensive credit?

United States homeownership by race and ethnicity 1976 to 2006
Homeownership rose before the subprime explosion. The report used this chart to argue that increased homeownership among people of color began before high-cost subprime lending dominated the market. Institute on Race and Poverty, 2009.

Federal mortgage rules changed after the crash

After the collapse, national policy changed. Congress enacted the Dodd-Frank Wall Street Reform and Consumer Protection Act in 2010 and created the Consumer Financial Protection Bureau. The new mortgage rules required lenders, for most residential mortgages, to make a reasonable determination that borrowers could actually repay their loans. The Qualified Mortgage framework restricted or discouraged several risky loan features, while later HMDA rules expanded the information lenders report about mortgage applications and loans.

None of that erased inequality or made mortgages risk-free. It changed the machinery. The rules were designed in part to reconnect underwriting with repayment capacity and to make the market easier to monitor than the one the 2009 report described.

Interactive | The rules change

What changed after the crash?

2007-09Foreclosures and financial failures turn a housing bust into the Great Recession.
2010Dodd-Frank creates a new federal framework and the CFPB.
2013-14Ability-to-Repay and Qualified Mortgage rules take effect.
2015-18HMDA reporting expands information about mortgage terms and outcomes.
2025-26Fair-lending and mortgage-data rules change again. Read the linked primary sources for their separate scope.

Policy changes after a crisis

Public arguments about regulation often change as a crisis recedes. Rules can duplicate one another, impose real costs, and make safe credit harder to provide. At the same time, the mortgage collapse did not come from one villain or one obviously foolish loan. It emerged from incentives scattered across brokers, lenders, securitizers, rating agencies, servicers, and investors, each able to pass some risk onward.

Current changes have separate agencies and limited scopes. In 2026, the Consumer Financial Protection Bureau amended Regulation B and stated that the Equal Credit Opportunity Act does not recognize disparate-impact liability. The Office of the Comptroller of the Currency also rescinded its Fair Housing Home Loan Data System rule for national banks. Neither action alone recreates or prevents a housing crisis. The historical question is more practical: which data, rules, and public questions help people recognize harmful patterns before losses reach a neighborhood?

Rules cannot prevent every failure. They help determine which information is collected, which practices receive scrutiny, and which warning signs can be seen before losses reach a neighborhood.

The foreclosure wave receded. The ownership map did not reset.

If the crisis had been merely a temporary interruption, we might expect the housing geography eventually to spring back to its earlier form. That is not what later research found. In Hennepin and Ramsey Counties, Urban Institute researchers found White household homeownership remaining near 70 percent from 2000 through 2018, while Black homeownership fell from about 31 percent in 2000 to 24 percent in 2010 and 21 percent in 2018. During the same period, the number of Black households grew dramatically, but almost all of that net household growth was among renters.

The same research found another transformation after the Great Recession. Single-family rentals in the two counties more than doubled, from an estimated 22,000 in 2005 to more than 48,000 in 2020. Investor ownership rose especially sharply in North Minneapolis and parts of central St. Paul. The trends do not show that every foreclosed house became a rental, but they show a major shift in who owned the region's homes.

Urban Institute chart of homeownership by race in Hennepin and Ramsey counties
The long aftermath. Urban Institute analysis shows Black homeownership in Hennepin and Ramsey Counties falling from about 31 percent in 2000 to 21 percent in 2018, while White homeownership stayed near 70 percent. Urban Institute.
Urban Institute map of change in investor ownership 2005 to 2020
The new ownership geography. Investor acquisition of single-family properties increased especially in North Minneapolis and parts of central St. Paul from 2005 to 2020. Urban Institute.
22k
Estimated single-family rentals in Hennepin and Ramsey Counties in 2005.
48k+
Estimated single-family rentals by 2020.
29.1%
Twin Cities Black homeownership in the 2024 ACS one-year estimate, which the Minneapolis Fed cautions is noisy.

There has been progress and reversal rather than one smooth line. The Minneapolis Fed reported that the regional Black homeownership rate reached 34.3 percent in 2023, then fell to 29.1 percent in the noisier one-year 2024 estimate. More important than any single year's bounce is the scale of the continuing divide: in 2024 the Minneapolis-St. Paul metropolitan area had the largest Black-White homeownership gap among the Fed's twelve peer metros, 46.9 percentage points.

Webber Park Library in North Minneapolis
Recovery is also a landscape. North Minneapolis contains new public investment, institutions, businesses, and homes alongside the scars of foreclosure and disinvestment. Webber Park Library, 2017. Tony Webster, CC BY 2.0.

The house never left

This is perhaps the oddest feature of the whole story. The mortgage could be sold, pooled, securitized, rated, insured, traded, serviced, and moved through institutions thousands of miles away. The house remained where it had always been. So did the street, the neighborhood, the school, and the household trying to make the payment.

That is why the Great Recession belongs in a Minnesota geography course rather than only an economics course. The national financial system did not strike every landscape equally. It encountered a metropolitan geography already patterned by segregation, uneven wealth, unequal access to conventional credit, and the legacy of earlier housing policy. Then the crash rearranged property ownership again. The original report could see the first half of that process in 2009. Later data make more of the second phase visible.

A mortgage can leave Minnesota. A foreclosure cannot. It eventually has to happen somewhere.
Main points

Keep three ideas with you

A loan can travel. Ownership of a mortgage or its payment stream can move through institutions far from the home.
The house stays put. Default, vacancy, ownership change, and public costs occur at local addresses and in specific neighborhoods.
Maps require dates and limits. The 2004-07 evidence shows a historic pattern of unequal outcomes, not current lending conditions or a single-cause explanation.

Key terms

Securitization Turning mortgage payment streams into financial securities that can be sold to investors.
Mortgage servicer The company that collects payments and manages a mortgage account, which may differ from the original lender.
Subprime lender In the 2009 report, a lender classified by the share of its loans that were subprime. That category does not describe the terms of every individual loan.
Reverse redlining Targeting a community for costly or risky credit, often after that community has been underserved by conventional lenders.
Foreclosure sale A legal sale after a borrower defaults, which can change ownership and occupancy at one address.
Homeownership gap A difference in the share of households that own homes across racial or ethnic groups.

Sources, dates, and limits

The maps and charts in this reading are intentionally historical. The Institute on Race and Poverty report analyzed 2004-06 mortgage data and 2007 foreclosure sales while the crisis was still unfolding. Its mortgage panels use HMDA data for conventional, first-lien, owner-occupied one- to four-unit homes; its foreclosure panels show Hennepin and Ramsey County sheriff sales in 2007. They are snapshots of the mortgage boom and crash, not maps of lending conditions today. HMDA does not contain complete credit-file or underwriting information, so the patterns document unequal outcomes without proving the reason for every lending decision. The Urban Institute and Minneapolis Fed material at the end shows later ownership patterns and current estimates.

Primary Twin Cities source:
Communities in Crisis: Race and Mortgage Lending in the Twin Cities, Institute on Race and Poverty, University of Minnesota, February 2009. Charts and maps reproduced from the course source document; see especially pp. 11-12, 16, 18-24, and 30-35.
Current regional context:
Federal Reserve Bank of Minneapolis, Twin Cities region housing indicators, 2026. The 2024 Black-homeownership figure is a one-year ACS estimate and should be read with the article's caution.
2026 policy changes:
CFPB Regulation B; OCC rescission of its Fair Housing Home Loan Data System rule. The sources identify different agencies, legal bases, and scope.
Enlarged figure

Comments

Popular posts from this blog

1.1 Minnesota: A History of the Land 1 Ordering the Land

2.1 Minnesota A History of the Land 2 When Minnesota Began Spending Its Inheritance

1.2 Minnesota: A Pollen Grain, a Prairie Fire, and a Surveyor’s Chain