SICK PAPES SPECIAL ON RACIAL DISPARITIES IN MORTGAGE LENDING:
Interview with Jim Campen, author of âChanging Patterns XXI: Mortgage Lending to Traditionally Underserved Borrowers & Neighborhoods in Boston, Greater Boston and Massachusetts,â a report prepared for the Massachusetts Community & Banking Council.
The United States of America was built on genocide and slavery, and severe racial inequalities are built into the structure of our government and society in countless ways. Some forms of structural racism are more obvious than others; the mass incarceration of people of color for non-violent offenses, the murder of unarmed black men by police who face no legal consequences, and the savage inequalities in our educational system are extremely visceral manifestations of our countryâs deeply racist structures. There are other ways that racial inequalities are perpetuated that are less obvious, yet have equally destructive effects on social justice. Today we focus on one particular form of structural racism: access to mortgage loans.
In order to buy a house, you need a loan. Thus, there is a direct relationship between who gets loans, and the racial segregation of our neighborhoods and cities, and also on the ability of people to accumulate wealth via home-ownership. Our country has a long history of explicit racism in its housing policy, which has effectively âlocked [black people] out of the greatest mass-based opportunity for wealth accumulation in American history.â Today, there continue to be extreme disparities in the availability and the quality of the loans available to black people and Latinos. Here is something I learned during this interview: if you divide up Massachusetts into census tracts, and ask how many of the predominantly (>75%) non-white areas are, on average, âmiddle- or upper-incomeâ (very conservatively defined), the answer is: 4%. If you ask, how many of the >75% white census tracts are middle or upper-income, the answer is: 91%. Are you kidding me.Â
For the past 21 years, Jim Campen (aka my dad) has published an annual Sick Report entitled Changing Patterns: Mortgage Lending to Traditionally Underserved Borrowers & Neighborhoods in Boston, Greater Boston and Massachusetts. The 2014 report was released last week, and we sat down with a computer connected to Dr. Campen to discuss the relationship between mortgage lending and social justice.Â
Sick Papes: Can you explain some of the history of how racial discrimination in mortgage lending affects us?Â
Jim Campen: For neighborhoods to prosper, residents need to be able to borrow to purchase, maintain, and improve their homes. But ever since the government first began to promote homeownership early in the New Deal of the 1930s, the availability and cost of mortgage loans has been dramatically different in black neighborhoods than in white ones. Two major periods stand out. During the decades following World War II, suburbs grew as banks funneled money to white families abandoning central cities while refusing the make loans to the people â mainly people of color â who were left behind in urban neighborhoods. Because the areas where banks refused to lend were sometimes outlined in red on bank and insurance company maps, this geographic neglect came to be known as âredlining.â By the middle of the 1960s, even as the Civil Rights Movement focused its attention on Jim Crow laws and legal discrimination in the South, the central cities in the north were largely poor and black. Â
Inner city residents fought back against redlining and eventually succeeded in getting two important laws enacted in the mid-1970s. The Home Mortgage Disclosure Act of 1975 required lenders to begin reporting information on where they were and were not making loans. (These data provide the basis for reports like mine.) And the Community Reinvestment Act (CRA) of 1977 required banks to serve the credit needs of all the neighborhoods in areas where they did business (âreinvestmentâ because community groups demanded that banks reinvest their deposits in their own neighborhoods rather than sending them to the suburbs.) Â
The CRA didnât have much impact during the Reagan years, but the Clinton administration took it much more seriously, and flows of mortgage credit to borrowers and neighborhoods of color greatly increased during the 1990s. However, the resulting increases in housing wealth in these neighborhoods turned out to be short-lived as a result of the second major episode of lending discrimination â which turned the nature of lending on its head. Instead of refusing the make mortgage loans in neighborhoods of color (redlining), lenders targeted these same neighborhoods for exploitative, predatory loans (reverse redlining). These subprime loans ended up destroying more housing wealth than had been created between 1990 and 2005. Many people got rich making these loans (and many others got rich selling securities based on them), but neighborhoods of color where predatory loans were concentrated were hit disproportionately hard by the tsunami of foreclosures. . Â
The story is obviously a lot more complicated than this, with lots of other factors playing a role, but there can be no doubt that discrimination in mortgage lending has played a central role in bringing about the highly segregated nature of U.S. metro areas.
SP: Just how unequally is wealth distributed in Boston and/or Massachusetts?
JC: Iâm not familiar with studies of the distribution wealth distribution locally or statewide, although I donât see any particular reason to think that it is very different from the appalling inequalities at the national level that have been thoroughly documents and widely publicized. But I can offer two striking examples of racial disparities in this part of the world. Â
First, the homeownership rate in Massachusetts is 69.0% for whites, but only 32.0% for people of color â that is, the white homeownership rate is 2.15 times greater. This is the second highest disparity ratio in the U.S; smaller only that that of New York (66.2% vs. 28.5%, for a ratio of 2.32.). The national figures are 72.0% for whites and 46.2% for people of color, for a disparity ratio of 1.56. Â
Second, three tables in my new report show the incredibly stark income differences among neighborhoods with different percentages of minority residents.  (Table 14 is for the city of Boston, Table 15 is for Greater Boston, and Table 16 is for Massachusetts.)  These tables use standard income categories to classify census tracts (areas with typically 3,000 â 6,000 residents) as low-income, moderate-income, middle-income, or upper-income on the basis of the median family income (MFI) in the census tract as a percentage of the MFI in the relevant metropolitan are. In the city of Boston, for example, there are 44 predominantly-minority census tracts (that is, more than 75% minority residents) and 39 predominantly-white census tracts. Of the predominantly-minority tracts, 91% are low- or moderate-income, 9 are middle-income, and none are upper-income. Of the predominantly-white tracts, in contract, only 3% are low- or moderate-income, 25% are middle-income, and almost three-quarters (72%) are upper-income. For the state as a whole, just 4% of the 98 predominantly-minority tracts are middle- or upper-income (none of these are upper-income) compared to 91% of the 952 predominantly-white tracts. [editorâs note: sorry for the redundancy of this fact, but I just couldnât believe it.]
SP: You have consistently documented that black people and Latinos are much more likely to be denied a mortage. What do you think are the reasons for this increased higher denial rate? Are there racist bankers sitting behind a desk, personally denying loans based on race? I would (naively) imagine that there are computer programs that decide whether to approve loans, and I canât imagine these programs are allowed to explicitly account for race in the decision, right?
JC: The key question that I think you are getting at is: what percentage of the disparities in denial rates are a result of discrimination by mortgage lenders? That is a very interesting question that is impossible to answer on the basis of publicly-available information. Â
My own belief is that the majority of the disparities are the result of black-white differences in the incomes, wealth, education, employment stability, and other factors that are legitimately relevant in judging a potential borrowerâs credi âworthiness (that is, the likelihood of their making timely monthly payments on the mortgage loan that they are applying for) â differences that have resulted from our nationâs long and disgraceful history of oppressing, exploiting, and discriminating against black people and from the continuing institutional racism throughout American society. Â
Having said that, discrimination by lenders almost certainly plays a role. One type of evidence for this is the statistical studies that have been made in the few cases for researchers had access to enough data about applicants to reach a valid conclusion on whether or not race had independent explanatory power in accounting for higher black denial rates. The most important such study was conducted by researchers at the Federal Reserve Bank of Boston in the early 1990s and found that, taking into account over thirty other variables, denials rates for blacks and Latinos were 60% higher than for equally credit-worthy whites. (Munnell et al., âMortgage Lending in Boston: Interpreting HMDA Data,â American Economic Review, 1996). Â
The other type of evidence is that from âpaired testingâ studies. In such studies, trained âtestersâ with appearances and profiles that are similar (except for the difference being tested for â e.g., black/white or male/female) are sent into the same situation and follow the same general âscript.â Testing is relatively easy to do for shoppers in jewelry store, applicants for rental housing, and seekers of a taxi, but is much harder in the case of mortgage lending â because lenders generally require lots of real information about the applicant (Social Security number, credit score, employment history, tax returns) and the property being purchased. As a result, testing in the mortgage area can only be done for the initial stages of the process, when potential borrowers first approach lenders to make an inquiry. Nevertheless, HUD has sponsored major mortgage testing studies more or less every decade and consistently found discrimination. Closer to home, and on a much smaller scale, when I was on the board of the Fair Housing Center of Greater Boston, we did a local study in 2006 involving twenty pairs of potential mortgage borrowers. The results were quite clear: âOverall, the Fair Housing Center found differences in treatment which disadvantaged the homebuyer of color in nine of the twenty matched paired tests conducted, or 45 percent.Â
SP: If you control for other economic variables, is there still a large difference in the rate of loan denial to blacks and Latinos?Â
JC: In this new report, I used the HDMA data to investigate whether the racial differences in denial rates could be explained as a result of black and Latino applicants having lower incomes. The answer is: not very much of the difference. In Greater Boston, for example, where the overall black/white denial rate disparity ratio in 2013 was 3.14 (19.3% vs. 6.1%), the disparity ratio considering only those with incomes between $91K and $120K was 2.69, and the disparity ratio for those with incomes over $120K was 2.86.
SP: For many years prior to the 2008 financial crisis, you were already reporting about the risks of âsub-prime loansâ and âpredatory loans,â and emphasizing that lenders were seeking out people of color for these types of loans. Did the financial crisis have worse effects on people of color than it did on whites?Â
JC: Yes, much worse. One indicator is the relative decreases in homeownership rates between 2008 and 2013. While the white homeownership rate fell by 1.7 percentage points (from 75.0% to 73.3%), the Latino homeownership rate fell by 3.0 percentage points (49.1% to 46.1%), and the black homeownership rate dropped by 4.1 percentage points (from 47.9% to 43.8%). [Joint Center, The State of the Nationâs Housing, 2014, Table A-2]
And a 2013 study by the Urban Institute found that whereas white families had an average wealth about four times greater than that of nonwhite families before the recession, their average wealth after the recession was six times greater than that of nonwhites. [www.urban.org/publications/412802.html]Â
SP: Iâve noticed that, in general, your graphs donât have error bars and you donât report statistical tests. Can you explain why?Â
JC: Thatâs because Iâm working with the entire âuniverseâ of available data, not just a sample. When you do a poll of 1,500 likely voters, this sample may not be an accurate representation of all voters in the state or nation â and statistical tests, confidence intervals, margins of error, and so forth are quantifications of how likely the sample data is to accurately reflect the data in the entire universe. If youâre studying 50 fruit flies, the question is how likely your findings are to hold true for all fruit flies. But HMDA data is it not a random sample from a larger universe of mortgage loans. When HMDA data shows that the denial rate for blacks in Greater Boston in 2013 was 3.14 times the denial rate for whites, that is based on all 768 black applicants and all 25,741 white applicants that are reported in the HMDA data. There are various reasons why these denial rates might not be exactly correct, but sampling error is not one of them â and sampling error is what error bars and statistical tests are all about.Â
SP: What prompted you to start publishing this study in the first place?
JC: The Civil Rights Movement was at its peak during my formative years â I was in eighth grade when Rosa Parks refused to go to the back of the bus in Montgomery and a senior in college when Martin Luther King Jr led the March from Selma to Montgomery â and Iâve been concerned with issues of racial injustice and racial inequalities ever since. I first got interested in studying mortgage lending patterns in the early 1990s. After a year of struggles between community and groups and local banks following the January 1989 leak of a draft report by the Boston Fed that found that mortgage lenders were discriminating against black neighborhoods in Boston, the Mass. Bankers Association finally agreed to a program that included increasing mortgage lending to borrowers and neighborhoods of color. Their program also established the Massachusetts Community and Banking Council (MCBC) to maintain continuing dialogue between banks and community groups. I was hired by MCBC to prepare a report on how mortgage lending patterns had â or had not â changed in the first three years of the program, and in 1995 I released a report called Changing Patterns. MCBC seems to have liked it, since theyâve asked me to prepare annual reports ever since. The new report is the twenty-first in this annual series. Â
SP: What are some ways we can fight to change this? What are some of the groups working to fight this, on the legal level and on the street level?Â
JC: The struggle for greater economic justice in the area of mortgage lending is part of the overall struggle for greater economic justice throughout American society. Anything that results in higher incomes, reduced unemployment, better education, and less incarceration for people of color will enhance their ability to obtain good mortgage loans.Â
In terms of mortgage lending specifically, there are many groups engaged in a variety of efforts at both the national and local level.  Nationally, the establishment of the Consumer Financial Protection Bureau (CFPB) as part of the Dodd-Frank financial reform legislation of 2010 was a major victory, and the CFPB has already done a lot to reduce unfair, deceptive, and abusive mortgage lending practices. Thus, a top priority of national groups working in this area â such as Americans for Financial Reform, the Center for Responsible Lending, and the National Consumer Law Center â is to protect the CFPB from unrelenting attacks by Republicans in Congress who would like to end its ability to aggressively enforce consumer protection and fair lending laws. Â
On the local level, there are local and statewide groups around the country working in this area both through advocacy work and grassroots organizing. Many of these groups keep in touch with each other through the National Community Reinvestment Coalition, National Peopleâs Action, the PICO National Network, and other umbrella organizations. For several years in the 2000s, seven groups from around the country produced a joint annual report on some aspect of racial disparities in mortgage lending in our seven cities â Boston, Charlotte, Chicago, Cleveland, Los Angeles, New York City, and Rochester NY. I did the Boston analysis on behalf of the Massachusetts Affordable Housing Alliance (MAHA), a wonderful local organization that combines home- buyer and home-owner education classes, grass roots organizing (through its Home Buyers Union), and advocacy work with the Boston and Massachusetts governments (full disclosure: Iâm a long-time MAHA board member [editorâs full disclosure: heâs also a great dad]). Â
MAHA helped persuade Massachusetts to establish and promote a new Massachusetts Homeownership Compact that encourages banks to pledge to increase their lending through two publicly-supported affordable mortgage lending programs. Both the MassHousing Mortgage Program and the ONE Mortgage Program devote a high percentage of lending to exactly the kind of affordable prime mortgage loans to neighborhoods and borrowers of color that my report shows are much too scarce. The ONE Mortgage Program that MAHA has championed is particularly affordable (the âmortgage comparison chartâ at www.mymassmortgage.org shows that a lower-income borrower buying a $250,000 home would have a monthly payment of $1,004for a ONE mortgage, compared to $1,482 for an FHA mortgage) â and more than half of ONE mortgage loans in the state have gone to black or Latino borrowers. Â
SP: Thanks, dad! I love you.