This article was downloaded by: [York University Libraries]On: 06 November 2014, At: 09:37Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Sociological FocusPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/usfo20

Race, Mortgage Lending and LoanRejections in a U.S. CityKevin Fox Gotham aa Tulane University , USAPublished online: 19 Nov 2012.

To cite this article: Kevin Fox Gotham (1998) Race, Mortgage Lending and Loan Rejections in a U.S.City, Sociological Focus, 31:4, 391-405, DOI: 10.1080/00380237.1998.10571116

To link to this article: http://dx.doi.org/10.1080/00380237.1998.10571116

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

RACE, MORTGAGE LENDING AND LOAN REJECTIONS IN A U.S. CITY'

KEVIN FOX GOTHAM SOCIOLOGICAL FOCUS Tu lane University Vol. 31 No. 4

October 1998

Tin· negative effect of neighborhood racial Komposition on mortgage lending has been docu­mented in recent years in several cities, even after controlling for income, condition of housing and related neighborhood and housing characteristics. Lending industry officials maintain that the racial distribution of mortgage lending is the unintended consequence of profit-based lending activi­ties. In contrast, community activists and civil rights groups claim that racial discrimination is the motivating force underlying mortgage lending practices. Data from the 1991 Home Mortgage Disclo­sure Act thJMDA) for the Kansas City metropolitan area are used to measure trends in mortgage lending and limn rejection ratios by local lenders. This study finds that Kansas City lenders reject minority applicants at higher rates than whites and reject high-income minorities as often as low-income whites. More importantly, results show that high-income African Americans are rejected at a higher rate than low-income whites, indicating that race of loan applicants plays a crucial role in the decision to approve a mortgage loan, even after controlling for inco?ne and other factors. Findings from Kansas City and other cities indicate that the low percentage of minority Uxxn applications from some lenders suggests not only meager marketing, but possible pre-screening of minority applicants where they are discouraged from completing an application form.

INTRODUCTION

An recent years, the impact of race on mortgage lending has become an important topic of scholarly research and investigation. A number of empirical studies have documented racial disparities in mortgage lending even after controlling for income, credit history of applicant, age of housing, market demand and a host of other factors (Squires, Velez and Taeuber 1991; Munnell, Brown, McEneaney and Tootell 1992; Dymski 1995). Prior research and debate on the relative salience of racial discrimina­tion and non-discriminatory market behavior as an explanation for racial disparities in mortgage lending has been quarrelsome and divisive. In a recent exchange with Goldstein and Squires (1995), Leven and Sykuta (1994) argue that the main problems in the li terature are the quality and reliability of data, methods of variable selection and research design and aggregation over geographic areas. Goldstein and Squires (1995) counter that existing research suffers from limited conceptualizations of dis­crimination and inappropriate use of controls, among other limitations. As a critical form of institutional discrimination in the lending market , racial disparities in mort­gage lending both reflect and reinforce the trajectories of uneven development and urban disinvestment in U.S. metropolitan areas. At issue is whether racial disparities in lending may be an unfortunate outcome of "efficiently functioning markets ," as

I wish to thank Michael Bates, Dustan Shepherd and the Kansas City Civil Rights Consortium for assistance in data collection and analysis. I thank Joel Devine for helpful suggestions on previous drafts.

391

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

392 SOCIOLOGICAL FOCUS

Leven and Sykuta (1994) argue, or stem from actual racial discrimination. As Goldstein and Squires (1995) and Shlay, Goldstein and Bartelt (1992) maintain, case studies of metropolitan housing markets that employ both quantitative and qualita­tive data sources are useful to address the question of race and mortgage lending.

This case study uses both quantitative and qualitative data sources to examine loan rejection ratios and the distribution of mortgage loans by granted lenders in the Kansas City metropolitan area. There are a number of reasons Kansas City repre­sents an ideal site for examining the social problem of racial inequalities in lending and the connection of such inequalities to uneven development and racial residential segregation. First, the social problems of urban disinvestment and racial residential segregation in Kansas City have been recognized and documented by scholars and local journalists and residents. According to a study released in 1988 by the Kansas City Consensus Race Relations Task Force (1988) and in 1996 by the Mayor's Task Force on Race Relations (1996), Kansas City is "one of the most segregated cities in the nation" with regard to neighborhoods, education and employment. A 1991 report by the Greater Kansas City Urban League found that "[h]ousing patterns in Kansas City show a continuation of . . . efforts to control the movement of the black population. . . . sellers of homes and landlords still use disguised tactics to prevent black families from purchasing or renting . . . For the most part, integration of housing has flowed one way — blacks move in, whites move out" (Williams 1988, Pp. 87-88). In recent years, Kansas City has been identified by scholars as one of the nation's hypersegregated metropolitan areas due to its high degree of residential segregation on a number of housing indices (Massey and Denton 1993; Denton 1994). Local public officials and journalists have acknowledged the deleterious effects of the racial segregation on area schools and housing, the growing urban disinvestment and job loss, and the increasing blight and rising poverty that occurred while downtown redevelopment and suburban growth have been taking place (Stephens 1985; Thomas 1990; Katerndahl 1992). These social problems are clearly linked to racial inequalities in lending that reflect and reinforce uneven development and urban disinvestment in Kansas City and throughout urban America.

In this study, I examine how one important element in the process of urban disinvestment — racial disparities in access to housing loans — reinforces and exacer­bates the trends of racial residential segregation and inequality in Kansas City and other U.S. cities. I begin by discussing recent scholarly research on race and mortgage lending. Next, I discuss the data and methodology used in this analysis of race, loan applications and mortgage lending. I then discuss the major findings of the study; lastly, 1 draw on other case studies of urban disinvestment in U.S. cities to examine why racial disparities in loan rejections persist despite decades of fair housing laws and assorted anti-discrimination statutes. Policy implications are discussed and rec­ommendations are offered to combat the racial effects of mortgage lending and urban disinvestment.

CONTEMPORARY RESEARCH ON RACE AND MORTGAGE LENDING

Racial discrimination in mortgage lending patterns, loan applications and homeowner's insurance coverage has been identified as a major factor in the process of urban disinvestment and uneven development within U.S. metropolitan areas.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

RACE, MORTGAGE LENDING AND LOAN REJECTIONS IN A U.S. CITY 393

Despite the passage of numerous anti-discrimination laws and fair housing statutes, a number of scholarly analyses and government studies have uncovered a negative rela­tionship between the race of loan applicants and mortgage lending activity (Listokin and Casey 1980; Hula 1984; Squires and Velez 1987; Squires, Velez and Taeuber 1991; Dymski 1995). For example, in a study of changing patterns of insurance agency locations within Milwaukee, Squires, Velez and Taeuber find (1991) that "the racial composition of neighborhood is associated with agency location even after the effects of family income, condition of housing, and number of dwellings are controlled" (p. 567). Similarly, in an analysis of mortgage lending activity in Chicago, Listokin and Casey find that "racial composition of the neighborhood has a significant relationship with the volume of loans given to an area . . . even after controlling for economic variables" (1980, p. 157). Other research has concluded that the distribution of homeowner's insurance policies is negatively associated with central city location and positively cor­related with white suburban communities (Squires and Velez 1987, 1988). A 1991 Federal Reserve study of 6.4 million home mortgage applications by race and income documented that commercial banks rejected African American applicants twice as often as whites nationwide. In some cities, such as Boston, Philadelphia, Chicago and Minneapolis, the Federal Reserve study reported African Americans being rejected three times more often than whites (Munnell et al. 1992). Such findings corroborate other studies of mortgage lending in U.S. cities that document the persistence of racial disparities even after taking into consideration family income and wealth, age and condition of housing, residential turnover and other factors related to credit worthiness of residents and security of property (Bradbury, Case and Dunham 1989; Shlay 1985; Department of Housing and Urban Development 1991; Caskey 1994; Brown and Bennington 1993; Dymski and Veitch 1994; Dymski 1995; Yinger 1995).

Other researchers and lending industry officials challenge the validity of racial discrimination as an explanation for the racial disparity in rejection ratios and in the distribution of mortgage loans. For example, in an analysis of mortgage lending in New York, Schäfer (1978) finds no statistically significant relationship between race, income and several other factors in accounting for racial disparities in mortgage lending. Similarly, Benston and Horsky (1979) find no relationship between racial disparities in city and suburban distribution of mortgage loans. Interestingly, Barth, Cordes and Yezer (1983) maintain that what may appear to be racial discrimination on the part of lenders may, in fact, be the unintended consequence of nondiscrimina-tory profit-oriented market behavior in the aggregate. Tomer (1980) also discounts the influence of race, arguing that racial differences in mortgage lending are due to lending industry actions "which are biased (in an economic sense) against lending in older urban areas, especially those whose residents are considered low in socio-economic status" (p. 210). One recent study of mortgage lending in Detroit finds "no significant race effect" in determinants of lending patterns (Horner, Lynch and Perle 1994). Using data from a large midwestern lending institution, Leven and Sykuta (1994) find that "race does not emerge as significant" when controlling for measures of ability to repay the loan, risk of default, potential fault amount and personal charac­teristics. Hula (1992) warns that the "process by which discrimination does occur may be very complex and circuitous" and that quantitative methods, by themselves, "may not be sufficient to capture them." That few mortgage loans are available for inner-city minorities is taken as a given by scholars, government officials and the lending indus-

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

394 SOCIOLOGICAL FOCUS

try. At issue is whether racial discrimination or other socioeconomic factors such as age of housing, income, demand and other factors explain racial disparities in the dis­tribution of mortgage loans and loan rejection ratios.

A key point of contention in debates over the influence of race in mortgage lending is the importance of disparate treatment (intentional discrimination) versus disparate impact (institutional discrimination) upon minority applicants. The former refers to actions and behaviors that are based on conscious intent to discriminate while the lat ter are actions, decisions and policies tha t have a discriminatory effect without necessarily being based on discriminatory intent . According the ant i-discriminatory statutes and guidelines of the Departments of Justice and Housing and Urban Development (HUD), a policy or criterion that adversely affects certain racial groups must be a "businesses necessity for that factor to be considered nondiscrimina-tory." As Goldstein and Squires (1995) put it, if a lender wishes to use a factor in evaluating the creditworthiness of a potential mortgage applicant "that has the effect of placing minority applicants at a disadvantage (i.e., a fact that disqualifies a higher proportion of minority than nonminority applicants), tha t lender must be able to dem­onstrate that the use of the factor constitutes a business necessity" (p. 582). That is, if charged with disparate negative actions against a minority group, the lender must demonstrate that the various discriminatory criteria used to evaluate creditworthi­ness cannot be formulated and implemented in a nondiscriminatory fashion (e.g., there is not a less discriminatory way to achieve the goal of equality). Examples of underwrit ing practices and mortgage evaluation criteria that potentially have an adverse impact upon minorities are tiered pricing, minimum loan amounts, meager market ing, pre-screening of applicants, etc. The challenge for researchers is to uncover and document these more subtle and covert forms of lender behavior that can systematically have a disparate negative impact upon minority groups.

DATA AND METHODOLOGY

A number of data sources are used to answer whether racial discrimination or other socioeconomic factors explain racial disparities in loan rejection rates and the distribution of mortgages in metropolitan Kansas City. The number and dollar amount of all loans made by the Federal Housing Administration (FHA) and conven­tional mortgages for each census tract in the Kansas City Metropolitan Statistical Area (MSA) were collected from the 1991 Home Mortgage Disclosure Act (HMDA) reports for Kansas City MSA lenders. The federal Home Mortgage Disclosure Act of 1975 and the Community Reinvestment Act of 1977 require federally regulated depository insti tut ions to keep track of the number and dollar amounts of their housing loans. Within the ten-county Kansas City metropolitan area, 290 lending inst i tut ions originated or purchased housing loans in 1991, according to HMDA reports. I calculated the number of loan applications received by each lender, noted whether they were approved or rejected and ranked each of these lenders from highest to lowest minority-to-white loan rejection ratios (where there were at least four minority rejections). All housing loan originations are included (e.g., home purchase, refinance, multi-family, home improvement, subsidized). I calculated loan rejection ratios on the basis of all applications, not jus t for those in which a final lender acceptance or rejection was reported.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

RACE. MORTGAGE LENDING AND LOAN REJECTIONS IN A U.S. CITY 395

Early studies of lending discrimination by Shlay (1985) and Squires and Velez (1987) pointed out several limitations of HMDA data, including the absence of infor­mation on the number of applications or rejections. However, federal changes in the Home Mortgage Disclosure Act in the late 1980s now require lenders to make public the race and income of loan applicants and rejections, the type and amount of the loan application and to disclose by census tract the number and amount of loans they are making. Thus, while some limitations to this data set persist (see Squires and Velez 1987, Pp. 219, 230), systematic data are now available on demand, making it possible to assess whether the low number of loan mortgages in predominantly minority census tracts is due to the failure of minorities to apply for such loans or the failure of lenders to grant the loans. The HMDA reports also allow for the control of race, income and related socioeconomic variables in explaining loan rejection rates.

Second, statistical analysis of 1991 HMDA data is supplemented with 1990 cen­sus data, housing reports and analyses from local planning agencies and newspaper accounts to examine racial disparities in loan rejection rates in the Kansas City MSA. Although there is an abundance of quantitative studies of the influence of race and mortgage lending, there have been relatively few studies that combine both quantita­tive and qualitative data sources. The advantage of combining HMDA reports and census data with qualitative data is that it provides a way of examining, in detail, processes of urban disinvestment and racial discrimination in lending tha t could be difficult to infer from one data source alone. Early investigations of racial lending pat­terns were based almost exclusively on national-level data. More recent analyses by Caskey (1994), Avery (1991) and Squires and colleagues (Squires and Velez 1987; 1988; Squires, Velez and Taeuber 1991; Squires 1993) use multiple data sources to examine racial disparities in the distribution of home mortgage loans, insurance red­lining and bank representat ions in low-income minority areas . By using both quantitative and qualitative data sources, this study at tempts to expand on recent analyses to examine how the process of lending discrimination constitutes a critical form of urban disinvestment.

FINDINGS

One major finding of this study is tha t racial minorities, especially African Americans, are rejected for mortgage loans at a much higher rate than white appli­cants. According to 1991 HMDA data, lenders in Kansas City rejected 15 percent of white applicants, 36 percent of minority applicants and 42 percent of African Ameri­can applicants. The ratio of minority-to-white rejection rates represents one way of determining the likelihood that a minority applicant will be rejected or accepted for a mortgage loan. For example, a lender tha t rejected 10 percent of applications from whites and 25 percent of applications from minorities would have a ratio of 2.5. Thus, a minority family applying for a mortgage loan through this lender would be 2.5 times more likely to be rejected than would a white family. According to 1991 HMDA data, ten lenders have minority rejection ratios between 1.2 and 1.99, seven between 2.0 and 2.9, one between 3.0 and 3.99 and one between 4.0 and 4.99. Interestingly, nine lenders have African American rejection ratios between 1.3 and 1.99, seven between 2.0 and 2.99, one between 3.0 and 3.99, one between 4.00 and 4.99, two between 5.0 and 5.99 and one above 6.0. In the Kansas City MSA, the minority-to-white rejection

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

396 SOCIOLOGICAL FOCUS

ratio is 2.43, indicating that minorities are 2.43 times more likely to be rejected for a mortgage loan at area lenders than are whites. Interestingly, African American loan applicants are 2.80 times more likely than whites to be rejected. As of 1992, this 2.80 African American-to-white rejection ratio was the second highest in the nation, behind Chicago (3.08) and ahead of Philadelphia (2.76), Atlanta (2.63) and Detroit (2.61) (Wall Street Journal 11/30/92).

Many lenders recognize the racial disparities in mortgage applicant rejection rates but maintain that this is due to other factors besides racial discrimination by individual lenders. Lenders contend that other factors such as the applicant's credit-worthiness, and, above all, income determine whether or not they will receive a mortgage. A loan applicant who has a lower income may be less likely to have the resources to repay a loan and therefore be more likely to be rejected. Thus, many lenders argue that because minorities as a group are poorer than whites, it is income and not race that explains the higher rejection rates for minorities. Table 1 ranks 18 Kansas City lenders with the highest high-income minority to low-income white rejec­tion ratios. This ranking includes lenders having at least two high-income minority rejections where minority applicants with relatively high incomes of over $40,870 (120 percent of metropolitan median income) are rejected at a higher rate than white appli­cants with lower incomes of under $27,240 (80 percent of metropolitan median income). This table shows that a number of local lenders, including GMAC Mortgage Corporation, Capitol Federal and Federal Employees Credit Union, reject high-income minorities at a higher rate than low-income whites. Overall, Kansas City area lenders rejected higher-income minorities 0.93 times as often as low-income whites, about equally. However, high-income African Americans were rejected 1.18 times more often than low-income whites, or about 30 percent of the time.

Table 2 shows statistically significant differences, based on Chi-square tests, at all income levels for all metropolitan lenders. Using the figures from Table 2, Graph 1 shows rejection rates at all income levels for Kansas City area lenders. Not surpris­ingly, the graph shows that lower-income applicants are indeed rejected more often than higher-income applicants. However, the graph also reveals much higher loan rejection rates for minorities than for whites, regardless of income. This graph, and Table 2, show that applicant income does not explain all the disparity in rejection rates between whites and minorities.

Some lenders contend that lack of demand for mortgages by minorities explains the relatively few mortgages made in minority neighborhoods. A 1996 report from the Mayor's Task Force on Race Relations found that "[m]any minority individuals do not even think about trying to access these service providers. Others are understandably intimidated and/or lack the knowledge needed to traverse the often completed process" (Mayor's Task Force on Race Relations 1996). However, HMDA data suggest that many lenders don't market loans to minorities and fail to decide on loan applications for minorities at a higher rate than they do for whites. Although 14 percent of Kansas City's metropolitan households are minority and 9 percent of all loan applicants are minority (and 6 percent African American), many local lenders have extremely low rates of minority applications, even though they claim to serve the entire metropolitan area (Mid-American Regional Council 1993; City Development Department 1995).

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

Len

der

Nor

hank

G

MA

C

Nor

th A

mer

ican

Se

ntin

el

Gua

rant

y B

ank

Blu

e R

idge

Ban

k H

ome

Stat

e B

ank

Sunb

elt

Mor

tgag

e E

mpi

re R

ealty

C

apit

al F

eder

al

Cit

izen

s B

ank

Mid

-Con

tine

nt

Uni

ted

Mis

sour

i Fe

dera

l E

mpl

oyee

s C

U

Bro

ther

hood

Ban

k So

urce

One

Mor

tgag

e Se

curi

ty

Gre

en T

ree

Tot

al

TA

BL

E 1

HO

USI

NG

LO

AN

S BY

HIG

H-I

NC

OM

E M

INO

RIT

Y/L

OW

-IN

CO

ME

WH

ITE

RE

JEC

TIO

N R

AT

IO

Whi

te

Per

cent

A

ppli

cant

s R

ejec

ted

55

285

765

200 34

231 29

53

1 31

3346

26

0 14

2 13

60

138 43

22

7 79

6 53

6

14.5

7.

4 2.

4 12

.5

17.6

19

.5

17.2

10

.0

51.6

5.

2 22

.3

12.7

25

.4

14.5

37

.2

8.4

8.4

82.6

Und

er 8

0 M

edia

n I

App

lica

nts

12

35

83

42 7 51

10

146 6

238 56

42

417 25

18

26

21

5 21

6

r t A

rea

ncom

e P

erce

nt

Rej

ecte

d

8.3

2.9

2.4

19.0

14

.3

25.5

20

.0

7.5

50.0

9.

2 28

.6

26.2

34

.1

24.0

50

.0

23.1

12

.1

84.3

Min

orit

y

Tot

al

App

lica

nts

6 28

42 8 14 9 20

51 3

109 18

57

42

8 59

28

19

95

60

Per

cent

R

ejec

ted

83.3

35

.7

7.1

50.0

50

.0

55.6

45

.0

17.6

10

0.0

14.7

33

.3

31.6

57

.7

45.8

64

.3

26.3

24

.2

90.0

Ove

r 12

0% A

rea

Med

ian

App

lica

nts

2 14

17 3 5 3 7 19 3 76 (1

13

103 24 ö n 22

13

Inco

me

Per

cent

R

ejec

ted

100.

0 21

.4

11.8

66

.7

40.0

66

.7

42.9

15

.8

100.

0 17

.1

44.4

38

.5

48.5

33

.3

60.0

27

.3

13.6

92

.3

Min

orit

y/

Whi

te

Rej

ecti

on

Rat

io

5.73

4.

85

3.04

4.

00

2.83

2.

85

2.61

1.

77

1.94

2.

84

1.49

2.

49

2.27

3.

16

1.73

3.

14

2.88

1.

09

Hig

h-In

c.

Min

orit

y/

Low

-Inc

. W

hite

R

ejec

tion

R

atio

12.0

0 7.

50

4.88

3.

50

2.80

2.

62

2.14

2.

10

2.00

1.

85

1.56

1.

47

1.43

1.

39

1.20

1.

18

1.13

1.

10

> n m

2 O

ΪΒ > r*

Z σ z c § o c- O z 23 5 z en

Z > G

01

o

Sour

ce:

1991

HM

DA

Rep

orts

for

Kan

sas

City

MS

A. T

able

ada

pted

fro

m C

ivil

Rig

hts

Con

sorti

um (

1993

). Le

nder

s ra

nked

on

the

basi

s of

Hig

h-In

com

e M

inor

ity/L

ow-I

ncom

e W

hite

Rej

ectio

n R

atio

. All

hous

ing

loan

orig

inat

ions

are

incl

uded

(e.

g., h

ome

purc

hase

, ref

inan

ce, m

ulti-

fam

ily,

hom

e im

prov

emen

t an

d su

bsid

ized

).

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

398 SOCIOLOGICAL FOCUS

Table 3 shows six Kansas City metropolitan area lenders with the lowest percentage of total applications that are minority. All the lenders listed on Table 3 have fewer than 7 percent minority applicants, and Capitol Federal, the largest lender in the metropolitan area (according to number and volume of loans), has fewer than 4 percent minority applications and fewer than 1 percent African American applica­tions.

TABLE 2

HOUSING LOAN REJECTION RATES BY INCOME

Minority/ White Minority White Minority White

Income Loan Loan Percentage Percentage Rejection Chi-(Thousands) Applications Applications Rejection Rejection Ratio Square

0-4 Ö-9 10-14 IS-19 20-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60-64 65-69 70-74 75-79 80-84 85-89 90-94 95-99 100-104 105-109 110-114 115-119 120-124 125-129 130-134 135-139 140-144 145-149 150-154 155-159 160-164 165-169 170-174

34 200 839

1406 2258 2492 2946 3070 2857 2525 2158 1622 1667 1112 1083

720 626 442 403 354 365 195 176 120 151 94

103 76 84 42 60 36 43 41 29

14 88

283 339 388 361 346 290 245 173 163 101

96 59 50 33 24 21 17

9 15 8 6 9 9 6 2 3 4 3 7 6 3 1 1

26.5 38.0 38.6 30.4 21.9 19.9 18.1 14.6 13.9 11.6 10.8 8.9 8.4 9.4 9.6 8.2 9.1 7.5 7.4 8.2 4.9 5.1 7.4 7.5 8.6 7.4 9.7

11.8 4.8 7.1 1.7 2.8 7.0 4.9

13.8

57.1 67.0 52.7 48.4 44.1 36.0 35.3 35.9 23.3 28.3 28.2 20.8 20.8 25.4 30.0 27.3 20.8 19.0 11.8 11.1 13.3 12.5 16.7 22.2 11.1 16.7 0.0 0.0 0.0 0.0

14.3 0.0 0.0 0.0 0.0

2.16 1.76 1.36 1.59 2.01 1.81 1.95 2.46 1.67 2.43 2.63 2.33 2.48 2.69 3.12 3.33 2.29 2.55 1.58 1.36 2.70 2.44 2.26 2.96 1.29 2.24 0.0 0.0 0.0 0.0 8.57 0.0 0.0 0.0 0.0

2.421 9.349

20.472 41.278 81.911 49.229 62.523 89.275 18.022 41.493 43.141 14.216 16.710 17.632 20.474 13.656 3.388 3.754 0.674 0.420 2.260 0.928 0.868 2.283 0.047 0.480 0.252 0.420 0.216 0.157 3.761 0.193 0.263 0.053 0.166

Source: Kansas City MSA 1991 HMDA Reports. Table adapted from Civil Rights Consortium (1993). A Chi-Square of 3.481 indicates a probability of less than 5 percent that the relationship between race of the appli­cant at each income level is statistically unrelated to the disposition of the loan applicant.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

GR

APH

1

LOAN

REJ

ECTI

ON

S BY

RA

CE

AN

D IN

CO

ME

S o so

H o > o 171

r™

m

Z σ z o

Perc

ent

of

App

licat

ions

Re

ject

ed 70

Whi

te

■ M

inor

ity

60

50

-I 1

1 1

1 1

1 1

1 1

-i—

-i-

| |

1-

| |

1 1

1014

|5 1

9 20

24

25 2

9 1(1

M

35-1

9 40

-44

4.1.4»)

50-5

4 55

59

(.1) f>4

r.5-69

70-7

4 75

-79

R0-R

4 R5

.89

<X,.Q4

gs q

g |0

0

Ann

ual

Inco

me

(Tho

usan

ds)

PC

m

e-·

PJ

n H o z z > c en

O

H

-S

Sour

ce: 1

991

ΗΜΠΑ

Rep

orts

for

Kan

sas

Citv

MSA

. Gra

ph a

dapt

ed f

rom

Civ

il Ri

ghts

Cons

ortiu

m 1

1992

i.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

400 SOCIOLOGICAL FOCUS

TABLE 3

LENDER MARKETING TO MINORITIES

Percentage of Total Applications Which

Lender Are Minority

Capital Federal Savings and Loan 3.3 Sentinel FS and LA 4.0 Superior National Bank 4.9 Mark Twain Mortgage 5.2 Corinthian Mortgage 6.2 Farm and Home Savings 6.4

Source: 1991 HMDA Reports for the Kansas Sity MSA. Table adapted from Civil Rights Consortium < 19921.

There is also evidence indicating that some lenders fail to decide on loan appli­cations for minorities at a higher rate than for whites. Table 4 lists the ten lending institutions with the highest percent of minority applications left undecided. As this table shows, on the average, lenders across the metropolitan area have about eight percent of applications at the undecided stage, regardless of race. However, some lenders have substantially higher rates of undecided applications from minorities than others. The failure to approve or decline the application may be due to a number of factors, such as inadequate information, or socioeconomic factors, or may be a disguised form of lender rejection. In 1993, one former lending official in Kansas City explained to a local journalist how he "got rid" of loan applications he did not want to approve for a mortgage loan: "Most of my rejects were minorities. I probably could have gotten mortgages for most of them," he replied. Some of his disguised techniques for rejecting loans included "Making minority applicants fill out the complicated loan application form themselves, whereas we helped fill out applications for white borrowers"; "Creating endless and needless paperwork requests until the borrower gave up"; and "[l]etting an incomplete loan application sit for two or three months" {Kansas City Star 8/3/93). According to this former lending official, these kinds of techniques invariably make up a high proportion of "undecided" minority loan applications. While there are no systematic data available, the possibility that many undecided loan dispositions could in fact be disguised rejections suggests that some reported minority-to-white loan rejection ratios may understate the actual racial disparity in access to mortgage loans.

DISCUSSION

Some lenders maintain that because HMDA data do not include information on applicant credit histories, neighborhood property values and other factors, conclusions about racial discrimination as an explanation for higher rejection rates for minorities are unwarranted. In response to lender complaints and the limitations of HMDA data, in 1991 and 1992 the Federal Reserve Board completed studies of Kansas City, Boston and several other cities asking all local lenders to provide additional information on financial and employment variables relevant to the mortgage lending decision, including credit history, age, education, employment history and thirty-four additional

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

RACE, MORTGAGE LENDING AND LOAN REJECTIONS IN A U.S. CITY 401

variables (Kansas City Star 10/26/91; Munnell et al. 1992; Oliver and Shapiro 1995, Pp. 19-20, 137-141). In Kansas City, the Federal Reserve Board study found tha t applicants with incomes greater than 120 percent of the median income for the area were rejected 20 percent of the time, while white persons with less than 80 percent of the median income, the lowest income category, were rejected only 18 percent of the time. The results of the Boston study indicated tha t minority applicants, on average, have greater debt burdens, higher loan-to-value ratios and weaker credit histories than white loan applicants and that these disadvantages account for a large portion of the racial disparity in mortgage loan rejections, even after controlling for a host of additional variables. Including additional information on applicant financial and credit histories and property characteristics did reduce the minority-to-white rejection ratio in Boston from 2.7 to 1.6. However, controlling for additional financial, economic and property variables did not entirely eliminate the disparity, since the adjusted ratio revealed that minority applicants in Boston were 60 percent more likely to be rejected for a mortgage loan than whites (Munnell et al. 1992). Thus, in the end, a statistically significant gap remains that is associated with race.

TABLE 4

UNDECIDED LOAN APPLICATIONS

Lender

Percent of Minority Appl icants Left

Undec ided

8 50 17 15 18 17 14 18 16 11 15

Percent Higher Than for White

Applicat ions

0 52 55 24 24 96

435 32 50 29

None for Whites

Kansas City Average Charles F. Curry Co. Citizen's State Bank American Bank GMAC Mortgage Co. Home Federal Savings Sears Mortgage Co. Security Financial Co. Sunbelt National Mortgage United Missouri Mortgage Home State Bank of Kansas

Source: 1991 HMDA Reports for Kansas City MSA. Table adapted from Civil Rights Consortium (1992).

Why do racial disparities exist in mortgage loan rejection rates even after con­trolling for the effects of income and other factors? Lending industry officials argue that no rational lender would turn down a qualified applicant because s/he is a mem­ber of a minority group. The results of the Federal Reserve Board study in Boston suggest that minorities with untainted records are almost (97 percent) certain of being approved for a mortgage loan. However, the vast majority of loan applicants do not have perfect credentials, and lenders exercise considerable discretion in weighing a myriad of objective and subjective factors in deciding whether to approve a loan appli­cation (Oliver and Shapiro 1995, Pp. 19-20, 137-141) Applying for a mortgage loan is a multilayered and complicated process tha t involves several interrelated steps where differential t reatment can occur. Thus, empirical studies such as Leven and Sykuta's (1994) t ha t rely only on dichotomous variables (in which applications are either

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

402 SOCIOLOGICAL FOCUS

approved or denied) and do not consider applications that are withdrawn or discrimi­nation prior to the submission of an application are limited. Insufficient marketing, pre-application screening, verification of application information, appraisal of the property and a host of other objective and subjective factors can conspire to limit the number of minority applicants and generate racial disparities in mortgage lending. Like the real estate and insurance industries, the modus operandi of the lending industry is shaped by both external and internal factors, including government regu­lations, institutional guidelines set by real estate and banking boards, underwriting requirements and, more importantly, tolerance of informal patterns of racial bias by lenders (Squires 1993; Feagin 1994).

As in many other U.S. metropolitan areas, racial inequality in the distribution of housing, schools and mortgage money are central features of Kansas City. Since the 1970s, investigations by the Kansas City Star, the Kansas City Business Journal, the local branch of the Association of Community Organizations for Reform Now (ACORN) and other groups such as the Greater Kansas City Civil Rights Consortium, the Kansas City Consensus and the Kansas City, Missouri, Human Relations Commission have uncovered an undeniable pattern of discriminatory lending patterns and insur­ance underwriting (City Development Department May 1976; Bi-State Committee on Education 1977; Mid-American Regional Council November 1980; Missouri Depart­ment of Insurance 1/9/95; 5/30/95; Civil Rights Consortium 9/93; Business Journal 7/17/89; Kansas City Star 5/18/90; 10/22/91, p. A-10; 10/26/91; 11/19/91, p. D-l; 11/5/94, p. Bl; 5/31/95, p. Bl; 2/16/96, p. C-l). An analysis of 1987 and 1988 HMDA data by the Kansas City Business Journal shows that only 2.3 percent of the more than $98 million residential dollars loaned by the metropolitan area's eight largest banks in 1988 went to areas where minorities made up 50 percent of more of the population. Of the total of 2,717 residential loans made by the eight largest banks in 1988, only 183, fewer than 7 percent, went into predominantly minority areas {Business Journal 7/17/89). A review of HMDA data by the Kansas City Star for years 1990 through 1994 indicates that these racial disparities are continuing (Kansas City Star 10/3/93, p. K-l; 11/5/94, p. Bl; 12/18/95). Such findings corroborate findings from other cities indicating differential rates of mortgage lending by race (U.S. News and World Report 4/17/95, p. 51; The Nation 4/14/94, p. 450; USA Today 1/27/97, p. B-l).

CONCLUSION

The findings of this case study offer tentative support to the argument that race continues to play a significant role in mortgage lending. Goldstein and Squires (1995), Shlay, Goldstein and Bartelt (1992) and a number of other scholars argue that significant racial disparities in access to mortgage loans and other housing resources exist because of entrenched institutional discrimination in the housing and lending markets. But are African Americans and other minorities rejected at disproportionate rates compared to white applicants because of their race or because they live (or are trying to purchase a home) in a predominantly minority census tract? Are whites applying for loans in minority census tracts rejected at roughly similar rates to African Americans applicants? Here the evidence is mixed. Horner, Lynch and Perle's (1994) analysis of lending patterns in Detroit suggests that race does not play a role and that "normal economic determinants of mortgage lending predominate in the

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

RACE, MORTGAGE LENDING AND LOAN REJECTIONS IN A U.S. CITY 403

market." An analysis of mortgage lending in Milwaukee by Squires and Velez (1987) indicates that "the effects of race are accounted for primarily by geography. Separate analyses of city and suburban census tracts revealed, however, that race has adverse although different effects in these two areas" (p. 227). The important finding here is that race is "negatively but differentially associated with lending" (p. 227) in the city and suburbs. However, more recent analyses of HMDA data by the Federal Reserve Board in various cities suggest that overall, minorities residing in predominantly minority neighborhoods are rejected at roughly similar rates to minorities residing elsewhere (Munnell et al. 1992). Future research could address whether "neigh­borhood" redlining may not be as prevalent as it once was. It may be that the race of the applicant plays a far greater role in the lender's decision than the racial composition of the neighborhood in which the property is located.

Whether it is because of the race of the applicant or the racial makeup of the census tract, the fact of racial disparity in mortgage loan rejection rates reflects and reinforces the process of urban disinvestment, uneven development and racial inequality. Oliver and Shapiro (1995, p. 144-145) and Franklin (1991) maintain that banks, insurance companies and other financial institutions calculate risk factors in such a way as to "induce impersonal decisions that reinforce segregation." Impor­tantly, the differential mortgage loan rejection rates for whites and racial minorities, especially African Americans, cannot be explained away in terms of income, demand, risk or other seemingly "objective" socioeconomic and demographic factors that lenders claim as the basis for racial loan disparity rates. Thus, findings from Kansas City and other cities indicate that race intersects with a host of subjective evaluations that lenders use to evaluate the creditworthiness of the loan applicant. While state and federal laws make official discrimination illegal, informal patterns and institutional­ized mechanisms of lending discrimination remain a persistent and undeniable char­acteristic of American society. As a number of scholars have recognized, the everyday activities and prescribed norms of public agencies, private firms and social networks of actors within the housing and lending industries have a differentiated and negative impact on members of racial minority groups (Bratt, Hartman and Meyerson 1986; Goering 1986; Tobin 1987; Squires 1993; Massey and Denton 1993; Bullard, Grigsby and Lee 1994; Dymski and Veitch 1994; Feagin 1994). Persistent racial discrimination in the distribution of mortgage money exists not because of a few isolated bigots but because the majority of participants in the lending industry still harbor strong anti-black stereotypes including the belief that racial minorities are poor credit risks and racially mixed or predominantly black and minority neighborhoods are of lesser value than all-white neighborhoods.

The racial disparities in loan rejection rates identified in this and other studies suggest that more ambitious efforts, in addition to more effective enforcement of fair housing laws and anti-discrimination statutes, need to be initiated to address the problems of urban disinvestment. Several cities have instituted comprehensive training programs and hiring and promotion practices that encourage lenders to accept and appreciate racial and ethnic diversity, provide counseling and education on homeownership and familiarize loan officers with the special needs of first-time minority customers (Bradford and Cincotta 1994; Grigsby 1994; Keating 1994; Reed 1994; Smith 1994). State regulators and lending industry officials could establish a compensation structure based not simply on the number and size of loans granted, a

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

404 SOCIOLOGICAL FOCUS

practice that tends to disadvantage applicants who need more time or assistance or who are requesting small loans. In addition, underwriting and lending policies could be designed with appropriate safeguards that define and monitor the use of under­writing standards and practices to ensure they are applied consistently, without regard to race or ethnicity. While there is no guarantee that such policies, by themselves, would ensure racial minorities are treated fairly, these and other efforts by community groups and government agencies and regulators could begin to address the general problem of racial inequality in access to mortgage money and other hous­ing resources.

Kevin Fox Gotham is Assistant Professor of Sociology at Tulane University in New Orleans. His research interests are in urban sociology, political economy and historical sociology. He is currently exploring the his­torical development of the real estate and housing industry in the United States since the 1930s.

REFERENCES

Avery, R. 1991. "Deregulation and the Location of Financial Institution Offices. Federal Reserve Bank of Cleveland." Economic Review 27(3):30-42.

Barth, J.R., J.J. Cordes and A.M.J. ΥβεβΓ. 1983. "An Analysis of Informational Restrictions on the Lending Decisions of Financial Institutions." Economic Inquiry 21:349-60.

Benston, G.J. and D. Horsky. 1979. "Redlining and the Demand for Mortgages in the Central City and the Suburbs." Journal of Bank Research 10:72-87

Bradbury, K.L., K.E. Case and C.R. Dunham. 1989. "Geographic Patterns of Mortgage Lending in Boston, 1982-1987." New England Economic Review September and October:3-30.

Bradford, C. and G. Cincotta. 1992. "The Legacy, the Promise, and the Unfinished Agenda." Pp. 228-280 in From Redlining to Reinvestment: Community Responses to Urban Disinvestment, edited by Gregory Squires. Philadelphia: Temple University Press.

Brown, J. and C. Bennington. 1993. Racial Redlining: A Study of Racial Discrimination by Banks and Mortgage Companies in the United States. Washington, DC: Essential Information.

Bullard, Robert D., J. Eugene Grigsby and Charles Lee (eds). 1994. Residential Apartheid: The American Legacy. Los Angeles: CASS Urban Policy Series.

Caskey, John P. 1994. "Bank Representations in Low-Income Minority Urban Communities." Urban Affairs Quarterly 29(4): 617-638.

Civil Rights Consortium. 1993. Race and Mortgage Lending in Kansas City. Kansas City, MO: Kansas City Civil Rights Consortium. September.

Denton, Nancy A. 1994. "Are African-Americans Still Hypersegregated?" Pp. 49-81 in Residential Apartheid: The American Legacy, edited by Robert D. Bullard, J. Eugene Grigsby III and Charles Lee. Regents of the University of California: CAAS Publication.

Department of Housing and Urban Development. 1991. Housing Discrimination Study: Incidence of Discrimination and Variation in Discriminatory Behavior. Contract HC-5811. Washington, DC: U.S. Government Printing Office.

Dymski, Gary Arthur. 1995. "The Theory of Bank Rekindling and Discrimination: An Exploration." The Review of Black Political Economy. 23:39-56.

Dymski, Gary A. and John M. Veiteh. 1994. "Taking it to the Bank: Race, Credit, and Income in Los Angeles." Pp. 150-179 in Residential Apartheid: The American Legacy, edited by Robert D. Bullard, J. Eugene Grigsby III and Charles Lee. Los Angeles: CAAS Urban Policy Series.

Feagin, Joe R. 1994. "A House Is Not a Home: White Racism and U.S. Housing Practices." Pp. 17-48 in Residential Apartheid: The American Legacy, edited by Robert D. Bullard, J. Eugene Grigsby III and Charles Lee. Los Angeles: CAAS Urban Policy Series.

Franklin, Raymond S. 1991. Shadows of Race and Class. Minneapolis: University of Minnesota Press. Goldstein, Ira and Gregory Squires. 1995. "Obfuscating the Reality of Lending Discrimination through

Deceptively Rigorous Statistical Analysis: Comment on Leven and Sykuta." Urban Affairs Review 30:580-591.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14

RACE, MORTGAGE LENDING AND LOAN REJECTIONS IN A U.S. CITY 405

Grigsby, J. Eugene. 1994. "African American Mobility and Residential Quality in Los Angeles." Pp. 122-149 in Residential Apartheid: The American Legacy, edited by Robert D. Bullard, J . Eugene Grigsby, III and Charles Lee. Los Angeles: CAAS Urban Policy Series.

Homer, Jeffrey, Kathryn Lynch and Eugene D. Perle. 1994. "Perspectives on Mortgage Lender and Redlining." Journal of the American Planning Association 60:128-161.

Hula, R. 1984. "The Allocation of House Credit: Market vs. Non-Market Factors." Journal of Urban Affairs 6:29-18.

. 1992. "Neighborhood Development and Local Credit Markets." Urban Affairs Quarterly 27:249-67. Kansas City Consensus Race Relations Task Force. 1988. Creating Equality through Economic Development.

Kansas City, Missouri: Kansas City Consensus. Katerndahl, Dean Hill. 1995. Intercity Class Segregation in U.S. Metropolitan Areas. Unpublished Ph.D.

Dissertation. Department of Sociology. University of Kansas. Lawrence, KS. Keating, W. Dennis. 1994. Suburban Racial Dilemma: Housing and Neighborhoods. Philadelphia: Temple

University Press. Leven, Charles L. and Michael E. Sykuta. 1994. "The Importance of Race in Home Mortgage Approvals."

Urban Affairs Quarterly 29(3): 479-489. Listokin, D. and S. Casey. 1980. Mortgage Lending and Race. New Brunswick, NJ: Center for Urban Policy

Research, Rutgers University. Massey, Douglas S. and Nancy A. Denton. 1993. American Apartheid: Segregation and the Making of the

Underclass. Cambridge, MA: Harvard University Press. Mayor's Task Force on Race Relations. 1996. Interim Report. Kansas City, MO: Urban League of Greater

Kansas City. Munnell, Alicia H., L. E. Brown, J . McEneaney and G.M.B. Tootel. 1992. Mortgage Lending in Boston:

Interpreting HMDA Data. Boston: Boston Federal Reserve Board. Working Paper No. 92-97. Oliver, Melvin L. and Thomas M. Shapiro. 1995. Black Wealth, White Wealth: A New Perspective on Racial

Inequality. New York: Routledge. Reed, Veronica. 1994. "Fair Housing Enforcement: Is the Current System Adequate?" Pp. 222-236 in

Residential Apartheid: The American Legacy, edited by Robert D. Bullard, J. Eugene Grigsby III and Charles Lee. Los Angeles: CAAS Urban Policy Series.

Schafer, R. 1978. Mortgage Lending Decisions: Criteria and Constraints. Cambridge: Harvard-MIT Jo in t Center for Urban Studies.

Shlay, Anne. 1985. Where the Money Flows: Lending Patterns in Washington, DC, Maryland, Virginia SMSA. Chicago: Woodstock Institute.

Shlay, Anne B., Ira Goldstein and David Bartelt. 1992. "Racial Barriers to Credit: Comment on Hula." Urban Affairs Quarterly 28(1): 126-140.

Smith, Shanna L. 1994. "The National Fair Housing Alliance at Work." Pp. 237-256 in Residential Apartheid: The American Legacy, edited by Robert D. Bullard, J . Eugene Grigsby III and Charles Lee. Los Angeles: CAAS Urban Policy Series.

Squires, Gregory D. 1993. "The Political Economy of Housing: All the Discomforts of Home." Pp. 129-157 in Urban Sociology in Transition, edited by Ray Hutchison. Research in Urban Sociology. Vol. 3. JAI Press.

Squires, Gregory D. and William Velez. 1987. "Neighborhood Racial Composition and Mortgage Lending: City and Suburban Differences." Journal of Urban Affairs 9(3): 217-232. . 1988. "Insurance Redlining and the Process of Discrimination." Review of Black Political Economy

16:63-75. Squires, Gregory D., William Velez and Karl E. Taeuber. 1991. "Insurance Redlining, Agency Location and the

Process of Urban Disinvestment." Urban Affairs Quarterly 26(4): 567-588. Stephens, G. Ross. 1985. "Politics in Kansas City." Pp. 216-24 in Missouri Government and Politics, edited by

Richard J. Hardy and Richard D. Dohm. Columbia, MO: University of Missouri Press. Thomas, John Clayton. 1990. "Kansas City: A City in the Middle." Pp. 158-169 in Snowbelt Cities:

Metropolitan Politics in the Northeast and Midwest Since World War II, edited by Richard M. Bernard. Bloomington: Indiana University Press.

Tomer, John F. 1980. "Community Control and the Theory of the Firm." Review of Social Economy 38(2): 1 9 1 -214.

Williams, Deborrah D. 1991. "The Status of Housing in Kansas City's Black Community." Pp. 81-102 in The State of Black Kansas City, 1991. Kansas City, MO: Urban League of Greater Kansas City.

Yinger, John. 1995. Closed Doors, Opportunities Lost: The Continuing Costs of Housing Discrimination. New York: Basic Books.

Dow

nloa

ded

by [

Yor

k U

nive

rsity

Lib

rari

es]

at 0

9:37

06

Nov

embe

r 20

14