CRA Special Lending Programs.Increasing the flow of credit to lower-income households and communities has been the focus of many public-sector programs, such as those of the Federal Housing Administration and the Rural Housing Service. Government regulation of private-sector activities is often used to bolster such lending. The most prominent example of the latter approach is the Community Reinvestment Act (CRA CRA - C R Alpacas CRA - Cache Relative Area CRA - Calibration Requirements Analysis (US Navy) CRA - California Reading Association CRA - California Redwood Association CRA - California Rehabilitation Association - North, Inc. CRA - California Restaurant Association CRA - Call Routing Apparatus (UK Telecomms) CRA - Camden Recovery Act (Camden, New Jersey) CRA - Camera-Ready Art CRA - Campus Residents' Association CRA - Can't Remember Anything). The CRA was enacted in 1977 to encourage federally insured banking institutions (commercial banks and savings associations) to help meet the credit needs of their communities, including those of lower-income areas, in a manner consistent with their safe and sound operation. In responding to the CRA, banking institutions have sought to expand lending to lower-income populations in a variety of ways, but the approaches can be sorted into two broad types, both typically involving special marketing and outreach. In one approach, lenders have sought CRA-related customers who would qualify for market-priced loans under traditional standards (underwriting guidelines) for creditworthiness. In the second type of effort, lenders have sought customers by modifying their underwriting guidelines or loan pricing. To expand lending to lower-income populations through either approach, many banking institutions have developed or joined "CRA special lending programs," which seek out and assist such borrowers in a variety of ways. These programs vary greatly across banking institutions, differ widely in terms of their characteristics and how they are implemented, and can often be an important element of a banking institution's CRA-related lending activities. Although many institutions have offered special lending programs, some for many years, little systematic information is available about them. To further the understanding of these CRA special lending programs, this article provides new information on the nature of these programs, with particular emphasis placed on their characteristics and how these characteristics relate to the performance (delinquency and default rates) and profitability of the loans extended through them. BACKGROUND The CRA was enacted in response to concerns that banking institutions were, in some instances, failing to adequately seek out and help meet the credit needs of viable lending prospects in all sections of their communities. It directs the federal regulators of banking institutions (the Board of Governors of the Federal Reserve System, the Federal Deposit Insurance Corporation, the Office of the Comptroller of the Currency, and the Office of Thrift Supervision) to encourage the federally insured institutions they regulate to help meet community credit needs in a manner consistent with safe and sound operations. The CRA is likely to influence the behavior of a banking institution primarily through two mechanisms: an examination and ratings system and the formation of public opinion. Under the examinations and ratings system, regulators periodically visit the institution to assess the degree to which its lending is adequately serving its entire community. The CRA regulations guiding these examinations--jointly issued by the four federal banking agencies--emphasize an institution's record of serving the credit needs of low- and moderate-income populations within its CRA assessment area (see box "The CRA Regulations"). Each examination is followed by the assignment of a rating that is based on both quantitative and qualitative measures of the institution's performance. The CRA Regulations The regulations that implement the CRA set forth three tests by which the performance of most large retail banking institutions is evaluated: an investment test, a service test, and a lending test. The investment test considers a banking institution's qualified investments that benefit the institution's assessment area or a broader statewide or regional area that includes its assessment area.(1) A qualified investment is a lawful investment, deposit, membership share, or grant that has community development as its primary purpose. The service test considers the availability of an institution's system for delivering retail banking services and judges the extent of its community development services and their degree of innovativeness and responsiveness. Among the assessment criteria for retail banking services are the geographic distribution of an institution's branches and the availability and effectiveness of alternative systems for delivering retail banking services, such as automated teller machines, in low- and moderate-income areas and to low- and moderate-income persons. The lending test involves the measurement of lending activity for a variety of loan types, including home mortgage, small business, and small farm loans. Among the assessment criteria are the geographic distribution of lending, the distribution of lending across different types of borrowers, the extent of community development lending, and the use of innovative or flexible lending practices to address the credit needs of low- or moderate-income individuals or areas. For the lending test, the regulations implementing the CRA require the federal banking regulatory agencies to evaluate the geographic distribution of a banking institution's lending in two ways: (1) the proportion of all the institution's loans that are extended within its assessment area and (2) for loans within the institution's assessment area, their distribution across neighborhoods of differing incomes. In the latter measure, lending in low- and moderate-income neighborhoods is weighted heavily in CRA performance evaluations.(2) The CRA regulations also require the banking agencies to evaluate the distribution of a banking institution's lending within its assessment area across borrowers of different economic standing. This provision was added as part of revisions made to the CRA regulations in 1995. The exact definition of economic standing varies with the loan product being examined. For residential mortgage lending products, CRA assessments consider the distribution of loans across low-, moderate-, middle-, and upper-income borrowers, with a special focus on lending to low- and moderate-income borrowers.(3) For small business lending products, assessments consider the distribution of small loans (loans of $1 million or less) across businesses with differing levels of revenue, with a particular focus on loans to firms with annual revenues of $1 million or less. (1.) For purposes of evaluating CRA performance, each institution must delineate the geographic areas that constitute its CRA assessment area. For a retail-oriented banking institution, the institution's CRA assessment area must include the areas in which the institution operates branches and deposit-taking automated teller machines and any surrounding areas in which it originated or purchased a substantial portion of its loans. For a more complete description of these issues, see 12 CFR 228.41. (2.) The distribution of loans by neighborhood income is assessed for four income groups: low, moderate, middle, and upper. In a low-income area (typically a census tract), the median family income is less than 50 percent of the median family income for the broader area (such as a metropolitan statistical area or the nonmetropolitan portion of a state) as measured in the most recent decennial census. In a moderate-income area, the median family income is at least 50 percent and less than 80 percent of that for the broader area. In a middle-income area, the percentage ranges from at least 80 percent to less than 120 percent. And in an upper-income area, the percentage is at least 120 percent. (3.) Borrower income categories follow the same groupings as those for neighborhoods but rely on the borrower's income relative to that of the concurrently measured median family income of the broader area (metropolitan statistical area or nonmetropolitan portion of the state). An important aspect of the examination and ratings system is the statutory provision that requires regulators to consider the record of a banking institution in meeting the goals of the act when deciding on applications from that institution. In considering an application from an institution with a performance problem under the CRA, the regulators can--depending on the degree of the problem--potentially deny the application or require the institution to meet certain conditions in order to obtain approval. A second mechanism by which the CRA can influence the behavior of banking institutions is through the force of public opinion. In August 1989 the Congress amended the CRA to require each banking institution to allow public inspection of its examination ratings and supporting written evaluation. Such disclosure can influence the relationships that banking institutions have with potential investors, depositors, and borrowers. It may, for example, influence the nature and extent of public comments received on an application for a merger or acquisition. It may also influence decisions made by potential depositors, who may direct their funds to those institutions with the highest CRA performance ratings. Banking institutions thus have incentives to respond to the CRA. First, banking institutions have an incentive to engage in CRA-related activities to enhance their CRA performance rating. In addition, they have an interest in maintaining a good public image, which may be supported by a good CRA performance rating or by other CRA-related activities. Moreover, because of the potentially important role that CRA performance ratings and public comments can play in applications, such as for mergers and acquisitions, those banking institutions that anticipate making such applications are likely to be particularly sensitive to CRA considerations. In spite of a wealth of experience by banking institutions in undertaking CRA-related lending activities, little systematic information has been publicly available about those activities. For example, while banking institutions are known to use third parties to help reach certain targeted populations, little information is available on the nature and prevalence of these relationships. Also, there is reason to believe that the overall performance and profitability of CRA-related loans may differ from those of loans extended to other customers. The costs and possibly lowered revenues resulting from special marketing and outreach and from modified underwriting or loan pricing may make CRA-related loans less profitable than other loans. Moreover, the performance and profitability of CRA-related loans, whether or not they were originated through extra efforts or nontraditional standards, may differ from those of non-CRA-related loans simply because the two loan groups have differing characteristics. CRA-related loans might, for example, be smaller on average than other loans, which would make them relatively costly to originate and administer, or they might be less likely than other loans to be prepaid, a tendency that would also affect their profitability.(1) Despite widespread interest in the topic, little has been known about the performance and profitability of the loans that are made in conformity with the CRA regulation. To learn more about CRA-related lending activities, the Congress in November 1999 asked the Board of Governors of the Federal Reserve System to conduct a comprehensive study of the issue.(2) To this end, the Board conducted a special survey of the largest retail banking institutions to collect information on their lending experiences (see box "Participation in the Survey").(3) The survey was in two parts. Part A focused on an institution's total lending and its CRA-related lending in four broad loan product categories: one- to four-family home purchase and refinance lending, one- to four-family home improvement lending, small business lending, and community development lending. Participation in the Survey Participation by banking institutions in the Federal Reserve Board's Survey of the Performance and Profitability of CRA-Related Lending was voluntary. On January 21, 2000, each prospective respondent was mailed a copy of the questionnaire accompanied by a cover letter from Federal Reserve Board Chairman Alan Greenspan explaining the purpose of the survey and seeking voluntary cooperation in the study. The sample of institutions selected to participate in the survey consisted of roughly the largest 500 retail banking institutions--400 commercial banks and 100 savings associations. The sample was limited to the largest banking institutions because they account for the vast majority (estimated at more than 70 percent) of CRA-related lending nationwide. Survey responses were received from 143 banking institutions--114 commercial banks and 29 savings associations. Despite their relatively small number, the 143 survey respondents accounted for about one-half of the assets of the more than 10,000 U.S. banking institutions in existence as of December 31, 1999. Response rates varied markedly by the asset size of the institution. More than 80 percent of the largest surveyed banking institutions (assets of $30 billion or more as of December 31, 1999) returned a survey (27 out of 33 sampled institutions in this asset category). In contrast, only about 19 percent (72 out of 363) of the smallest surveyed banking institutions (assets of between $0.950 billion and $4.999 billion) responded. Institutions with assets of between $5 billion and $29.999 billion had a response rate of about 40 percent. In part B, the survey gathered extensive information on CRA special lending programs, defined as programs that banking institutions have established (or participate in) specifically to enhance their CRA performance, even if these programs may have been established for other reasons as well. Because these programs are often an important element of a banking institution's overall efforts to comply with the CRA, the survey collected information on many of their characteristics, including the performance and profitability of the lending extended under the programs. Responses to part B of the survey provide the data that form the basis of the analysis presented in this article. The analysis focuses primarily on CRA special lending programs exclusively offering home purchase and refinance loans, as survey responses indicated that most special lending programs were of this type. SURVEY RESPONSES REGARDING CRA SPECIAL LENDING PROGRAMS The Federal Reserve Board survey is the first systematic collection of information on the characteristics, performance, and profitability of CRA special lending programs from a broad base of institutions. As such, it provides a unique opportunity to learn about the characteristics of CRA special lending programs and relate these characteristics to the performance and profitability of programs. In the survey, banking institutions were asked to provide detailed information on the 1999 activity of their CRA special lending programs, defined as any housing-related, small business, consumer, or other type of lending program that the institution uses specifically to enhance its CRA performance.(4) For the survey, CRA special lending programs could include special programs offered or developed in conjunction with third parties, such as lending consortiums, nonprofit organizations, or government agencies that offer special lending programs in which an institution participates.(5) The survey was sent to the 500 largest retail banking institutions in existence at the end of 1999--400 commercial banks and 100 savings associations. Of these, 143 institutions responded (table 1).(6) Respondents offered or participated in 622 CRA special lending programs in 1999. Seventy-three percent of the responding institutions offered at least 1 CRA special lending program; on average the institutions with programs offered about 6 programs. To limit the burden of responding to the survey, the survey sought detailed information on only the 5 largest of a banking institution's CRA special lending programs (measured by dollar volume of originations in 1999), a restriction that produced detailed information for 341 programs. These 341 programs are estimated to account for 91 percent of the loan dollars that responding institutions extended under CRA special lending programs in 1999.
1. Banking institutions and CRA special lending programs covered
in survey, by size of institution, 1999
Item All reporting Size of banking institution
institutions (assets, in millions of
dollars)
950-4,999
Institutions
Number responding
to survey(1) 142 72
Offering at least
one program
Number 103 48
Percent 73 67
Number of programs
Among the five largest
at each institution(2) 341 138
Smaller than
the five largest
at each institution 281 31
Total
Number 622 169
Mean number per
institution offering
at least one program 6.0 3.5
Number of programs
among the
five largest at
each institution,
by type of
loan offered
One- to four-family
home, purchase and
refinance only(3) 247 98
Small business only 27 17
Other 67 23
One- to four-family
home, home
improvement only 17 7
Multifamily only 16 6
Consumer only 5 1
Commercial only 4 1
Other(4) 25 8
Programs among
the five largest at
each institution
operated by a
distinct unit
or department
of institution
Percentage of
institutions among
those with programs 67 60
Percentage of programs
among the five largest 63 56
Size of banking institution
(assets, in millions of dollars)
5,000-29,999 30,000 or more
Institutions
Number responding
to survey(1) 43 27
Offering at least
one program
Number 31 24
Percent 72 89
Number of programs
Among the five largest
at each institution(2) 116 87
Smaller than
the five largest
at each institution 139 111
Total
Number 255 198
Mean number per
institution offering
at least one program 8.2 8.3
Number of programs
among the
five largest at
each institution,
by type of
loan offered
One- to four-family
home, purchase and
refinance only(3) 83 66
Small business only 4 6
Other 29 15
One- to four-family
home, home
improvement only 6 4
Multifamily only 8 2
Consumer only 3 1
Commercial only 3 0
Other(4) 9 8
Programs among
the five largest at
each institution
operated by a
distinct unit
or department
of institution
Percentage of
institutions among
those with programs 77 92
Percentage of programs
among the five largest 75 80
(1.) Excludes one institution (in the middle size category) that did
not respond to the special lending portion of the survey For more
information on the sample size, see text bus "Participation
in the Survey."
(2.) Institutions were asked for detailed information on only
the live largest of their programs (measured by dollar volume
of 1999 originations)
(3.) Programs reported in this row and the remaining rows of this table
are from among the 341 reported by all institutions to be among their
5 largest Data in subsequent tables involve only the 247 programs
reported in this row (referred to hereafter as CRA special mortgage
programs).
(4.) Programs identified as such by survey respondents and programs
that offer more than one type of loan.
CRA special lending programs are often complex in design and can involve many features and a diverse group of market participants. As a consequence, the operation of some of these programs requires considerable training and experience. To facilitate the efficient implementation of these programs, many banking institutions establish distinct units or departments within the institution to run their CRA special lending programs. Among the banking institutions that offered at least one special lending program, 67 percent had at least one program operated by a distinct unit or department (table 1). Larger banking institutions in the sample were more likely than smaller institutions to offer programs through a distinct unit or department. Overall, of the special programs that each institution reported to be among its five largest, 63 percent were operated by a distinct unit or department. Before the survey was conducted, CRA special lending programs had been known to involve a range of credit products, but no information was available on the incidence of special lending programs across loan product categories. Results of the survey revealed that 72 percent of the programs (and 89 percent of the program dollars originated in 1999) for which banking institutions provided detailed information focused on one- to four-family home purchase and refinance loans. The next largest category of CRA special lending programs, comprising 8 percent of reported programs, focused on small business loans. The remaining programs cover a variety of loan products, none of which individually accounted for a substantial proportion of all programs. Because CRA special lending programs concentrating on home purchase and refinance loans constitute most of the CRA programs reported in the survey, the analysis in the remainder of this article (covering the data in table 2 and subsequent tables) focuses exclusively on these programs. The relatively small number of programs that were reported to focus on small business and other lending products precludes a comprehensive analysis of them. For simplicity, we will hereafter usually refer to CRA special lending programs that focus on home purchase and refinance loans as "CRA special mortgage programs."
2. CRA special mortgage programs, grouped by size of banking
institution and distributed by size and age of program, 1999
Percentage of programs
Item All-institutions Size of banking
estimate(1) institution (assets, in
millions of dollars)
950-4,999
Size of program (loan
dollars originated
in 1999)
500,000 or less 31 44
More than 500,000
to 2 million 28 32
More than 2 million
to 15 million 24 21
More than 15 million 18 3
Total 100 100
Year program established
Before 1990 6 5
1991-94 32 29
1995-97 43 43
1998-99 19 23
Total 100 100
Item Size of banking institution
(assets, in millions of dollars)
5,000-29,999 30,000 or more
Size of program (loan
dollars originated
in 1999)
500,000 or less 11 3
More than 500,000
to 2 million 25 8
More than 2 million
to 15 million 29 24
More than 15 million 35 65
Total 100 100
Year program established
Before 1990 7 10
1991-94 42 28
1995-97 41 47
1998-99 11 15
Total 100 100
NOTE. See table 1, note 3, for scope of data in this and subsequent
tables, Components may not sum to 100 because of rounding.
(1.) Average of values for the three asset-size categories after
adjustment; for tables 2-6, adjusted value (not shown in tables)
based on the rates of response to the survey; for tables 7-11,
adjusted value (not shown in tables) based on the rates of
response to the survey and to the particular question (for more
information, see text box "Calculating the `All-Institutions
Estimate'").
THE CHARACTERISTICS OF CRA SPECIAL MORTGAGE PROGRAMS The survey was designed to collect information that would shed light on the diversity of characteristics, both within and across banking institutions, among CRA special lending programs. In addition, because it was recognized that banking institutions may have established these programs for a variety of reasons that go beyond their efforts to enhance their CRA performance, the survey asked respondents to provide information on both the reasons for which they originally adopted these programs and the current benefits they receive from the programs. In table 1, data in the "all reporting institutions" column were taken from the 142 institutions responding to part B of the survey. In the analysis that follows (covering data reported in table 2 and subsequent tables), figures in the "all-institutions estimate" column are also based on the responses of the 142 institutions, but these responses have been weighted so that the figures represent an estimate of what the responses would have been if all 500 institutions had responded to the survey and provided answers to all applicable questions (see box "Calculating the `All Institutions Estimate'"). Calculating the "All-Institutions Estimate" The appropriateness of the "all-institutions estimate," reported in table 2 and subsequent tables, relies upon the validity of the assumptions needed to construct it. Key assumptions are those related to the treatment of sample and question nonresponse. The proportion of banking institutions that responded to the survey varied significantly by asset-size group (see preceding box "Participation in the Survey"); as a consequence, unless behavior is the same for institutions across different asset-size categories, simple averages based on the answers provided by respondents will distort the picture of what the survey responses would have been if all 500 institutions had provided answers to all applicable questions. To address this concern, the data in the "all-institutions estimate" column are calculated, in part, on the basis of adjustment factors reflecting the relative response rates for respondents in the three asset-size classes, The sample response adjustment factor for respondents with assets of $30 billion or more is 1.2 (or 33 / 27), that is, of the 33 institutions in the category, 27 responded). Similarly, the sample response adjustment factor for respondents with assets of $5 billion to $29.999 billion is 2.4 (or 104 / 44); and for respondents with assets of $0.950 billion to $4.999 billion, the adjustment factor is 5.0 (or 363 / 72).(1) An additional adjustment problem in calculating responses for the all-institutions estimate arises from the fact that some questions were not answered by a significant proportion of respondents. For questions with a significant number of nonresponses (tables 7-11), an additional adjustment factor, also based on asset size, was applied to correct for the varying propensities within the asset-size classes to answer questions. The general procedure used to calculate question-response adjustment factors was to assume that respondents within an asset-size category that did not provide an answer to a question would have the same response pattern as those that did. Thus, the number of respondents who answered each question was scaled up to represent those who did not answer. Respondents for whom a question was not applicable were not used in calculating the all-institutions estimates. For example, if 24 respondents were asked a question and 12 provided an answer, each of the 12 was multiplied by 2 to represent a total of 24 institutions. Question-response adjustment factors were calculated separately for each asset-size category because the responses varied by asset size. Data in the all-institutions estimate column in tables 7-11 are computed with the question-response adjustments in conjunction with the sample-response adjustments. For example, if the 12 respondents in the example above were large institutions, the total response adjustment for each of the 12 institutions that provided an answer would be 2 x (33 / 27), or 2.44. (1.) This procedure assumes that the respondents within an asset-size category are representative of all institutions in that category. The Size and Age of Individual Programs Survey responses indicate that in 1999 the dollar amount of loans extended under all CRA special lending programs made up a relatively small portion of total CRA-related lending in that year for most reporting institutions (see box "Survey Definition of a CRA-Related Loan"). In the case of home purchase and refinance loans, the proportion of CRA-related home purchase and refinance loan dollars that were extended under CRA special mortgage programs was only 4 percent for the median banking institution. Among the institutions that had CRA special mortgage programs, the proportion was 18 percent for the median institution. For about one-sixth of all institutions in the survey, however, CRA special mortgage programs accounted for more than 40 percent of their CRA-related home purchase and refinance loan dollars (data not shown in tables).(7) Survey Definition of a CRA-Related Loan In conducting the study of the performance and profitability of loans made in conformity with the CRA, the Board used the current CRA regulations as a guide in establishing a definition of a "CRA-related loan." As noted, the regulations require the banking agencies to evaluate the geographic distribution of lending and the distribution of lending across borrowers of different economic standing (see box "The CRA Regulations"). As a result, for purchase and refinance lending on one- to four-family homes, a CRA-related loan was defined to mean any loan made within the banking institution's CRA assessment area to a low- or moderate-income borrower (regardless of neighborhood income) or in a low- or moderate-income neighborhood (regardless of borrower income). In the aggregate, CRA special mortgage programs made up 21 percent of the total dollars of CRA-related home purchase and refinance loans originated by reporting institutions (and only 3 percent of the total dollars of home purchase and refinance originations).(8) Information reported also suggests that individual CRA special mortgage programs are generally small. For 1999, an estimated 31 percent of the CRA special mortgage programs reported in the survey had total originations of $500,000 or less, and about 28 percent had total originations of between $500,000 and $2 million; only 18 percent had originations of more than $15 million (table 2). The size of CRA special mortgage programs varied with the asset size of the banking institution, as programs tended to be larger for the largest banks in the survey (data not shown in tables). The median size of CRA special mortgage programs for large banks (those with assets of $30 billion or more) was about $36 million; for the smallest banks in the sample (those with assets of $0.950 billion to $4.999 billion) the median size of CRA special mortgage programs was about $680,000. Most of the CRA special mortgage programs that were reported in the survey were established relatively recently. More than half (62 percent) were established after the CRA regulations were modified in 1995 (table 2); only 6 percent of the programs were established before 1990. This pattern is consistent with the small size of many programs, as newer programs tended to be smaller. Reasons for Establishing CRA Special Mortgage Programs and Current Benefits Banking institutions cite many reasons for originally establishing or participating in CRA special mortgage programs (table 3). Responding to the credit needs of the local community and promoting community growth and stability are the two most frequently cited reasons. The third most frequently cited reason (for 76 percent of these programs) was to obtain a "Satisfactory" or "Outstanding" CRA rating. However, only 1 percent of CRA special mortgage programs are reported to have been established only to obtain a satisfactory or outstanding CRA rating. The fourth most frequently cited reason (also mentioned for more than half the programs) was to improve the institution's public image.
3. Reasons for establishing CRA special mortgage programs and their
current benefits to the banking institution, by size of
institution, 1999
Percentage of programs
Item All-institutions Size of banking
estimate institution (assets, in
millions of dollars)
950-4,999
Reasons for
establishing program
Help earn a CRA
rating of
"Satisfactory"
or "Outstanding"
Cited as only reason 1 1
Cited as one reason
among others 76 74
For a rating of
"Satisfactory" 37 40
For a rating of
"Outstanding" 52 42
Respond to credit
needs of local
community 95 93
Promote community
growth and stability 80 74
Improve institution's
public image 51 40
Earn additional
profits 46 39
Identify profitable
new markets 44 41
Maintain market
share in face of
increased competition 42 31
Minimize likelihood
of adverse public
comment on CRA record 31 22
Other 3 3
Current benefits
from program
Helps earn a CRA
rating of
"Satisfactory" or
"Outstanding"
Cited as only reason 1 1
Cited as one reason
among others 80 80
For a rating of
"Satisfactory" 41 44
For a rating of
"Outstanding" 53 45
Responds to credit
needs of local
community 94 92
Promotes community
growth and stability 87 83
Improves institution's
public image 54 47
Earns additional profits 42 38
Identifies profitable
new markets 37 33
Maintains market
share in face of
increased competition 50 38
Minimizes likelihood
of adverse public
comment on CRA record 38 30
Other 1 1
Item Size of banking institution
(assets, in millions of dollars)
5,000-29,999 30,000 or more
Reasons for
establishing program
Help earn a CRA
rating of
"Satisfactory"
or "Outstanding"
Cited as only reason 0 2
Cited as one reason
among others 86 68
For a rating of
"Satisfactory" 35 21
For a rating of
"Outstanding" 75 59
Respond to credit
needs of local
community 99 96
Promote community
growth and stability 92 91
Improve institution's
public image 76 65
Earn additional
profits 65 42
Identify profitable
new markets 53 41
Maintain market
share in face of
increased competition 61 64
Minimize likelihood
of adverse public
comment on CRA record 47 52
Other 4 3
Current benefits
from program
Helps earn a CRA
rating of
"Satisfactory" or
"Outstanding"
Cited as only reason 0 2
Cited as one reason
among others 86 73
For a rating of
"Satisfactory" 43 17
For a rating of
"Outstanding" 71 66
Responds to credit
needs of local
community 99 94
Promotes community
growth and stability 94 94
Improves institution's
public image 67 66
Earns additional profits 54 35
Identifies profitable
new markets 45 42
Maintains market
share in face of
increased competition 73 71
Minimizes likelihood
of adverse public
comment on CRA record 54 52
Other 2 0
NOTE. See notes to table 2, except that here components do
not sum to 100 because respondents could give more than one answer.
The pattern of reasons for establishing programs does not vary greatly by size of reporting institution in most cases; but large banking institutions were more likely than smaller institutions to cite a desire to improve their public image, to maintain their market share of lending, and to minimize the likelihood of adverse public comment on their CRA record. That only about three-fourths of CRA special mortgage programs were reportedly established to achieve a satisfactory or outstanding CRA rating may be somewhat puzzling, given that the survey explicitly asked institutions to report only on special lending programs that had as one of their documented purposes enhancement of the institution's CRA performance. One possibility is that some of the programs that support the CRA-related lending activities of institutions are not considered by the institutions to be "needed" to obtain a particular CRA rating. A second possibility is that the support of CRA-related activities is a documented purpose of some programs, but a relatively minor one. Banking institutions reported receiving a variety of current benefits from offering or participating in CRA special mortgage programs. Obtaining either a satisfactory or outstanding CRA rating was, again, the third most frequently mentioned benefit (for 80 percent of the programs), but also as before, this was cited as the only current benefit for just 1 percent of the programs. Responding to the credit needs of the local community, promoting community growth and stability, and improving the public image of the institution are also frequently cited current benefits of these CRA special lending programs. Features of CRA Special Mortgage Programs Almost all CRA special mortgage programs were targeted to populations that are emphasized in the CRA regulations: lower-income borrowers and borrowers in lower-income neighborhoods. Most programs targeted both of these populations (table 4). When only one population was targeted, it was much more likely to be lower-income borrowers than lower-income neighborhoods.
4. CRA special mortgage programs, grouped by size of banking
institution and distributed by targeted market, 1999
Percentage of programs
Target All-institutions Size of banking institution
estimate (assets, in millions of
dollars)
950-4,999
Lower-income targets
Neighborhoods 6 6
Borrowers 22 24
Neighborhoods and
borrowers 69 66
Other 3 4
Total 100 100
Target Size of banking institution
(assets, in millions of dollars)
5,000-29,999 30,000 or more
Lower-income targets
Neighborhoods 4 8
Borrowers 18 17
Neighborhoods and
borrowers 76 76
Other 2 0
Total 100 100
NOTE See notes to table 2.
Third parties played a role in about three-fourths of CRA special mortgage programs (table 5). Third parties involved in the programs included public entities at all levels of government and a range of for-profit and nonprofit private-sector firms and organizations. Some programs (31 percent in the survey, not shown in tables) involved the active participation of multiple third parties.(9)
5. Third-party involvement in CRA special mortgage programs, by size
of banking institution, 1999
Percentage of programs
Size of banking
institution
(assets, in
Third-party types and activities All-institutions millions of
estimate dollars)
950-4,999
Any 76 72
Type of third party (percentage
of programs with third-party
participation)
Nonprofit organization 47 46
Local government 35 30
State government 30 19
Fannie Mae, Freddie Mac 24 13
Federal Home Loan Bank 22 20
Financial institution consortium 21 21
Federal government 17 13
Other 2 3
Third-party activities (percentage
of programs with third-party
participation)
Services 70 73
Pre-loan education or
counseling for applicants 57 60
Identification of potential
borrowers 49 47
Screening of potential
applicants 33 34
Post-loan education or
counseling for borrowers 28 27
Underwriting 16 23
Assistance in servicing account 15 17
Second review of loan
applicants 3 4
Subsidies 71 70
Grants for down payment or
other purposes 60 59
Subsidized interest rates 27 24
Subsidized fees 24 27
Tax relief (credits or
exemptions) 3 1
Assumption of risk 49 49
Subordinate mortgages 36 33
Credit guarantees 16 20
Miscellaneous 13 6
Purchase of broker loans 12 4
Other 1 1
Size of banking
institution (assets, in
Third-party types and activities millions of dollars)
5,000-4,999 30,000 or more
Any 87 73
Type of third party (percentage
of programs with third-party
participation)
Nonprofit organization 40 71
Local government 43 46
State government 47 48
Fannie Mae, Freddie Mac 43 33
Federal Home Loan Bank 25 23
Financial institution consortium 19 25
Federal government 21 31
Other 0 2
Third-party activities (percentage
of programs with third-party
participation)
Services 63 75
Pre-loan education or
counseling for applicants 47 67
Identification of potential
borrowers 49 63
Screening of potential
applicants 26 40
Post-loan education or
counseling for borrowers 25 38
Underwriting 6 6
Assistance in servicing account 10 15
Second review of loan
applicants 0 6
Subsidies 69 81
Grants for down payment or
other purposes 58 71
Subsidized interest rates 36 19
Subsidized fees 21 15
Tax relief (credits or
exemptions) 3 17
Assumption of risk 42 65
Subordinate mortgages 35 60
Credit guarantees 8 8
Miscellaneous 29 17
Purchase of broker loans 28 17
Other 1 0
NOTE. See notes to table 2, except that here components do not sum
to 100 because respondents could give more than one answer.
Although their roles vary across programs, third parties conduct a wide range of activities that contribute to the implementation of CRA special lending programs, including activities that reduce the costs and risks of default that banking institutions might otherwise incur in extending credit to the populations served by the special programs. The most frequently cited activities were providing grants for down payments or other purposes, providing pre-loan education or counseling to loan applicants, and helping lenders identify prospective borrowers. Large banking institutions were more likely than smaller institutions to use third-party services for applicant screening and for grants to cover the loan down payment, while smaller institutions were more likely to use third-party underwriting services, credit guarantees, and subsidies to borrowers for fees they incur in obtaining mortgage credit. Apart from the efforts of third parties, many features of CRA special mortgage programs directly involved the banking institutions themselves (table 6). The most frequently mentioned were more flexible underwriting criteria, a second review of loan applicants to determine qualifications, special outreach and marketing activities, waived or reduced fees, pre-loan education or counseling to applicants, and reduced interest rates. The proportion of CRA special mortgage programs that offered any given feature varied somewhat across institution size classes, although the smallest institutions were less likely to conduct the two major services-type activities--special outreach and marketing and pre-loan education or counseling--and less likely to conduct a second review of applicants.
6. Program features and underwriting variances provided by institutions
in their CRA special mortgage programs, by size of banking institution,
1999
Percentage of programs
Size of
banking
institution
(assets, in
millions of
Feature or underwriting variance All-institutions dollars
estimate
950-4,999
Program feature(1)
Services 67 58
Special outreach and marketing
activities 52 40
Pre-loan education or
counseling for applicants 45 38
Post-loan education or
counseling for borrowers 8 9
Other 3 4
Subsidies 72 74
Waived or reduced fees 51 56
Reduced interest rates 41 45
Waived PMI (private mortgage
insurance) 30 33
Grants for down payment or other
purposes 23 24
Special financial incentives to
loan officers or brokers 21 15
Altered terms 88 89
More flexible underwriting
criteria 76 80
Second review of loan applicants 55 48
Longer term of loan 10 12
Underwriting variances(2)
Yes 83 87
Variances (as a percentage of
programs with any variances)
Lower down payment 85 86
Alternative measures of credit
quality (such as rent payments) 79 76
Higher debt ratios 77 83
Lower cash reserve requirement 72 70
More flexible requirements for
employment history 58 58
Lower standards for credit
history 45 52
Provisions waived or reduced
PMI or credit guarantee 40 43
Collateral 2 2
Compensating balances 8 7
Less documentation 14 16
Other 4 4
Size of banking institution
Feature or underwriting variance (assets, in millions of dollars
5,000-29,999 30,000 or more
Program feature(1)
Services 86 83
Special outreach and marketing
activities 74 79
Pre-loan education or
counseling for applicants 60 52
Post-loan education or
counseling for borrowers 5 11
Other 1 5
Subsidies 65 80
Waived or reduced fees 40 46
Reduced interest rates 30 41
Waived PMI (private mortgage
insurance) 21 39
Grants for down payment or other
purposes 19 24
Special financial incentives to
loan officers or brokers 28 46
Altered terms 84 88
More flexible underwriting
criteria 65 80
Second review of loan applicants 68 70
Longer term of loan 10 0
Underwriting variances(2)
Yes 70 91
Variances (as a percentage of
programs with any variances)
Lower down payment 83 88
Alternative measures of credit
quality (such as rent payments) 88 82
Higher debt ratios 59 70
Lower cash reserve requirement 72 78
More flexible requirements for
employment history 57 58
Lower standards for credit
history 22 38
Provisions waived or reduced
PMI or credit guarantee 26 45
Collateral 2 3
Compensating balances 9 12
Less documentation 5 20
Other 3 3
NOTE. See notes to table 2, except that here components do not sum to
100 because respondents could give more than one answer.
(1.) Responses to part B, question 14, "What special features or
services does your banking institution provide in connection with the
program?"
(2.) Responses to part B, question 17, "Are your banking institution's
customary underwriting standards ... altered under [the] program?"
The alteration of customary underwriting standards by banking institutions was a part of a large majority (83 percent) of special mortgage programs. The most frequently cited underwriting variances offered were lower down payments; the acceptance of alternative measures of credit quality, such as rent and utility payment histories, in lieu of more traditional measures of credit risk; lower cash reserve requirements; and higher debt-to-income ratios. A large proportion of programs (58 percent) also allowed additional flexibility when reviewing an applicant's employment history. The opportunity for borrowers to qualify for credit using these additional underwriting flexibilities suggests that loans made under CRA special mortgage programs may have elevated rates of delinquency and default. Banking institutions may offset these apparent additional risks through steps they often take in conjunction with these underwriting variances, such as pre-loan education and counseling and enhanced monitoring of borrower payment patterns. PERFORMANCE AND PROFITABILITY OF CRA SPECIAL MORTGAGE PROGRAMS Performance and profitability are important issues to consider in evaluating the long-term viability of CRA special mortgage programs and the effect of these programs on the financial condition of the banking institutions that offer them. Performance To assess the performance of CRA special mortgage programs, the survey focused on delinquency rates and net charge-off rates, which are closely related to default rates (see box "Measures of Performance"). The survey used two measures of delinquency--the percentage of loan dollars 30-89 days past due and the percentage of loan dollars 90 days or more past due or nonaccruing--that, like net charge-off rates, are commonly used in the industry and are regularly tracked and disclosed in regulatory reports filed by banking institutions.(10) Both delinquency measures are calculated as of December 31, 1999, and the net charge-off rate is calculated over the calendar year 1999. Measures of Performance Given a definition of performance in terms of delinquency or default, one can measure performance in either of two ways. One method is to consider performance at the loan level by calculating the percentage of loans that are delinquent or in default. The second method is dollar-based: The dollars or costs (in dollars) associated with delinquent or defaulted loans are summed and compared with the total dollars of loans outstanding. This article uses the dollar-based measure of performance. For the definition of loan performance, many people are familiar with the terms "delinquency" and "default." Delinquency occurs when a borrower fails to make a scheduled payment on a loan in a timely manner and in full. Because loan payments are typically due monthly, the lending industry customarily categorizes delinquent loans as either 30, 60, 90, or 120 or more days late depending on the length of time the oldest unpaid loan payment has been overdue.(1) Technically, default occurs at the same time as delinquency; that is, a loan is in default as soon as the borrower misses a scheduled payment. However, the term "default" is not generally used this way in the mortgage market, where it has, instead, a variety of other definitions. Among them are these four: * A lender forecloses on the property to gain title to the asset securing the loan * The borrower chooses to give the lender title to the property securing the loan "in lieu of foreclosure" * The borrower sells the property securing the loan obligation and makes less than full payment on the obligation * The lender renegotiates or modifies the terms of the loan and forgives some or all of the delinquent principal and interest payments. Loan modifications may take many forms, including a change in the interest rate on the loan, an extension of the length of the loan, and an adjustment of the principal balance due. Regardless of the definition of default used, a dollar-based measure of it could be computed, but the measure would not take into account the losses associated with default, which may be more or less than the loan amount. The actual losses are the unpaid principal and interest plus ancillary out-of-pocket costs, such as those of collection, less any amounts recovered. As a consequence, a related dollar-based measure of default--net charge-offs--is often used instead. For a given loan, the net charge-off is the total dollars owed at default (including the ancillary out-of-pocket costs) minus any subsequent recoveries. The institution-based net charge-off rate is calculated by summing its loan-level net charge-offs over a period of time (a year, for example) and dividing this amount by the average outstanding loan balances (including delinquent loans) over the period. In this article, the institutional net charge-off rate is used as the measure of default. (1.) For purposes of reporting on delinquency experience in the Report of Condition and Income (for commercial banks) and the Thrift Institution Financial Report (for savings associations), institutions typically group delinquent loans into three broad categories: 30-89 days past due and still accruing interest. 90 days or more delinquent and still accruing interest, and nonaccruing. The relative performance of CRA special mortgage programs varied with the measure of performance considered. For delinquencies, survey responses indicated that, on average, CRA special mortgage programs had lower rates than those for overall CRA-related home purchase and refinance lending but higher rates than those for an institution's total home purchase and refinance lending (table 7). For example, the mean rate for loans that were delinquent 90 or more days or nonaccruing was 1.00 percent for CRA special mortgage programs, 1.42 percent for overall CRA-related home purchase and refinance lending, and 0.78 percent for total home purchase and refinance lending.
7. Performance of CRA special mortgage programs, by size of banking
institution 1999
Percentage of loan dollars per program
Size of banking institution
(assets, in million
Program performance All-institutions of dollars)
measure estimate
950-4,999 5,000-29,999
Delinquencies(4)
30-89 days
Mean 1.94 1.31 3.04
Median .50 .00 1.88
90 or more days or
nonaccruing
Mean 1.00 .59 1.72
Median .07 .00 .91
Net charge-offs(5)
Mean .19 .09 .41
Median .00 .00 .00
Size of banking
institution MEMO: All-institutions
(assets, in estimate(1)
millions of
dollars) All
Program performance CRA-related All mortgage
measure 30,000 or more mortgage loans(3)
loans(2)
Delinquencies(4)
30-89 days
Mean 3.07 2.95 1.86
Median 2.01 2.40 1.44
90 or more days or
nonaccruing
Mean 1.70 1.42 .78
Median 1.06 .90 .53
Net charge-offs(5)
Mean .36 .23 .14
Median .05 .05 .02
NOTE. Results are for loans held in institution's portfolio. See also
notes to table 2 and text box "Measures of Performance."
(1.) Only institutions that reported performance of their CRA special
mortgage program loans as well as of all their CRA-related mortgage
loans and of their total mortgage loans. The weights for calculating
the all-institutions estimate here are the number of CRA special
mortgage programs offered by the respondents.
(2.) All of institution's CRA-related home purchase and refinance
loans whether or not part of a CRA special lending program.
(3.) All of institution's home purchase and refinance loans, whether
or not CRA-related.
(4.) At year-end 1999.
(5.) Total net charge-offs of program loan dollars during 1999 divided
by average program loan dollars outstanding during 1999.
On the other hand, CRA special mortgage programs performed better than total home purchase and refinance lending when performance was assessed using median values. For example, the median per program rate for loans that were delinquent 90 or more days or nonaccruing was 0.07 percent for CRA special mortgage programs and 0.53 percent for total home purchase and refinance lending. For net charge-offs, the zero rate for more than half of the CRA special mortgage programs could possibly reflect the relative newness of many of the programs as well as the influence of a number of other factors, including more intensive screening of prospective borrowers, sometimes by third parties, greater efforts to work with delinquent borrowers, and policies encouraging increased forbearance for such programs. The performance of these programs appears to vary with the asset size of the banking institution operating the program. On average, CRA special mortgage programs at large banking institutions had higher delinquency and charge-off rates than programs at smaller institutions. For example, at year-end 1999, the mean 30-89 day delinquency rate for the CRA special mortgage programs of large banking institutions was 3.07 percent, while the mean for smaller institutions was 1.31 percent. Profitability The survey sought information on the profitability of CRA special mortgage programs using return on equity (ROE) as the preferred measure of profitability (see box "Measuring Profitability"). Discussions with banking institutions in advance of the survey suggested that some of them might have difficulty calculating an ROE for individual loan programs. Consequently, the survey also collected detailed qualitative information on profitability as well: Banking institutions were asked if each individual CRA special mortgage program was "profitable," "marginally profitable," "break even," "marginally unprofitable," or "unprofitable." The same question was asked for overall CRA-related home purchase and refinance lending and total home purchase and refinance lending. Only the qualitative data are provided here because they were in fact far more frequently reported than were the quantitative data. Measuring Profitability Measuring the profitability of lending activities offers special challenges. First, the profit on a loan or program can be calculated in various ways. For the survey, profit from a lending activity was measured using a comprehensive definition that included all "revenues and costs associated with overhead, origination, and servicing costs; pricing; delinquency, default and losses; prepayment; loan sales and purchases; and related customer account business." Although overhead was not defined, it was intended to include the costs of permanent and working capital (sometimes referred to as a hurdle rate). Total dollars of profit may not be a meaningful measure of profitability, as programs may differ in size, for example. Therefore, profitability is typically expressed as a rate, with return on equity (ROE) and return on assets (ROA) both commonly used. Calculating the ROE or the ROA for a program requires the allocation of equity or assets, respectively, to it. The ROA is commonly used because it can often be more easily calculated for a given point in time. However, the ROA cannot be used to compare programs among institutions that have varying propensities for selling their loans. For example, a banking institution that sells most of the loans it originates, and thus has few assets, may misleadingly appear to be extraordinarily profitable when measured using the ROA. Thus, comparing the ROA across programs in which loans are sold at different rates can be misleading. Consequently, ROE was selected as the more appropriate measure of profitability to be used in the survey. According to respondents, the majority (64 percent) of CRA special mortgage programs were either profitable or marginally profitable in 1999 (table 8). Twenty-two percent of the programs were considered either marginally unprofitable or unprofitable. Experience varies across reporting banking institutions grouped by asset size. Compared with large and medium-sized institutions, small institutions (assets of between $0.950 billion and $4.999 billion) reported that a higher percentage of their CRA mortgage programs were either profitable or marginally profitable and that a lower percentage were either marginally unprofitable or unprofitable in 1999. For example, small institutions reported that 72 percent of their CRA special lending programs were either profitable or marginally profitable; large institutions reported that only 40 percent of their programs were either profitable or marginally profitable.
8. CRA special mortgage programs, grouped by size of banking
institution and distributed by profitability category of program,
1999
Percentage of programs
Size of banking
institution (assets,
in million of
Program profitability All-institutions dollars)
measure estimate
950-4,999 5,000-29,999
CRA special mortgage
programs alone
Profitable 29 34 19
Marginally profitable 35 38 35
Break-even 14 18 5
Marginally unprofitable 15 7 29
Unprofitable 7 4 12
Total 100 100 100
Relative to all
CRA-related
mortgage loans(4)
Lower 32 31 33
Same 50 55 43
Higher 17 15 24
Total 100 100 100
Relative to all
mortgage loans(5)
Lower 59 49 77
Same 36 51 4
Higher 5 0 20
Total 100 100 100
Size of banking
institution MEMO: All-institutions
(assets, in estimate(1)
millions of
dollars) All
CRA-related
Program profitability mortgage All mortgage
measure 30,000 or more loans(2) loans(3)
CRA special mortgage
programs alone
Profitable 20 37 61
Marginally profitable 20 40 33
Break-even 15 1 7
Marginally unprofitable 29 18 0
Unprofitable 16 4 0
Total 100 100 100
Relative to all
CRA-related
mortgage loans(4)
Lower 38 ... ...
Same 42 ... ...
Higher 20 ... ...
Total 100 ... ...
Relative to all
mortgage loans(5)
Lower 78 ... ...
Same 18 ... ...
Higher 4 ... ...
Total 100 ... ...
NOTE. Results are for estimates of 1999 profitability. See also notes
to table 2 and text box "Measuring Profitability."
(1.) Only institutions that reported profitability of CRA special
mortgage program loans as well as of all their CRA-related mortgage
loans and of their total mortgage loans. The weights for calculating
the all-institutions estimate here are the number of CRA special
mortgage programs offered by the respondents.
(2.) All of institution's CRA-related home purchase and refinance
loans, whether or not part of a CRA special lending program.
(3.) All of institution's home purchase and refinance loans, whether
or not CRA-related.
(4.) Data derived from comparing the profitability category
("profitable" through "unprofitable") in which respondents placed CRA
special mortgage program loans with the category in which they placed
all of their CRA-related home purchase and refinance loans.
(5.) Data derived from comparing the profitability category
("profitable" through "unprofitable") in which respondents placed CRA
special mortgage program loans with the category in which they placed
all of their home purchase and refinance loans.
... Not applicable.
This pattern--smaller institutions being more likely to report that their programs were profitable--is consistent with the broader pattern observed for all CRA-related mortgage lending and could be the result of a number of factors. As one example, the pattern is consistent with the view that smaller banking institutions have better knowledge of their local markets and more familiarity with local borrowers, which could result in less risky loan portfolios derived from better assessments of the risks associated with prospective borrowers. As a second example, the pattern is also consistent with the view that smaller institutions are less likely to be involved in mergers and hence are less subject to some of the lending incentives associated with the CRA. In this view, smaller institutions would be less inclined to provide services that adversely affect profitability. The accuracy of conjectures such as these is unknown. The profitability of CRA special mortgage programs can also be gauged using two types of comparisons with the profitability of two broader groups of loans examined in the analysis of performance: all CRA-related home purchase and refinance lending and total home purchase and refinance lending. The first type of comparison looks at the profitability distribution of CRA special mortgage programs against the profitability distribution of each of the two broader groups of loans. This comparison finds that while 64 percent of CRA special mortgage programs were reported to be at least marginally profitable (table 8), 77 percent of all CRA-related mortgage lending and 94 percent of overall home purchase and refinance lending programs were reported to be at least marginally profitable (table 8, memo). A second approach compares the 1999 profitability of CRA special mortgage programs with that of other loan groups within each institution. That is, survey responses by an individual institution regarding the profitability of each loan group were compared and rank ordered. If the responses indicated that the profitability category (profitable, marginally profitable, break even, marginally unprofitable, unprofitable) of two groups was the same, their relative profitability was considered to be equal. If an institution's responses placed two loan groups in different profitability categories, then relative profitability was judged based on which group was placed in the higher profitability category. The results indicate that respondents placed 50 percent of their CRA special mortgage programs in the same profitability category as their overall CRA-related home purchase and refinance lending. Of the comparisons revealing a difference between CRA special mortgage programs and overall CRA-related home purchase and refinance lending, roughly twice the percentage of CRA special mortgage programs were placed in a lower profitability category than were placed in a higher profitability category (32 percent versus 17 percent). Differences are more prominent when comparisons are made between the profitability of CRA special mortgage programs and total home purchase and refinance lending. Here, nearly 60 percent of CRA special mortgage programs were placed in a lower profitability category. Performance and profitability are generally thought to be positively correlated, and thus one would expect that rank orderings of groups of loans by the two criteria should be similar. But in the case of CRA special mortgage programs, the similarities did not hold. For profitability, CRA special mortgage programs tended to be less profitable than CRA-related and total home purchase and refinance lending. Conversely, for performance, CRA special mortgage programs performed better on average than overall CRA-related home purchase and refinance lending and only slightly worse than overall lending by most measures of performance. This apparent inconsistency may be a consequence of additional, perhaps costly, steps that institutions take as a part of their CRA special mortgage programs to identify and work with potential borrowers both before and after the loan is extended. These efforts, which can include enhanced marketing, counseling, and more intensive monitoring of loan payments, may result in better loan performance but may also lower the profitability of the loans. RELATIONSHIP BETWEEN PROGRAM FEATURES AND PROGRAM PROFITABILITY The features of CRA special mortgage programs, how banking institutions deliver the services associated with them, and the characteristics of the banking institutions themselves all may influence the profitability of these programs. The previous analysis indicated that in 1999 the profitability of CRA special mortgage programs varied significantly with the size of the banking institution that operated them. Further analysis (not shown) suggests that the profitability of these programs also varied with program size, measured by dollars of 1999 loan originations. Consequently, the following analysis of the relationship between program features and profitability categorizes programs by their size (large and small) and by the size of the banking institution that instituted the program.(11) Too few small programs were reported on by large institutions to support analysis, so figures for this subcategory are not reported here; the small number of such programs likely reflects the focus of the survey, which asked banking institutions to report detailed information on only their five largest programs. Once program size and banking institution size are taken into account, the profitability of CRA special mortgage programs does not appear to have varied significantly with the reason for which a program was established or the benefit afforded by a program (tables 9.A, B, and C). For almost every reason cited by banking institutions for creating a program and for almost every benefit, the proportion of programs reported to be profitable and unprofitable is quite similar to that of the "all programs" category (first line of tables). The comparison holds even for programs that were specifically established to achieve a "Satisfactory" or "Outstanding" CRA rating.
9. Selected characteristics of CRA special mortgage programs, grouped
by size of banking institution and size of program, and distributed
by profitability category of program, 1999
A. All banking institutions
Percentage of programs
All programs
Characteristic
Profitable Break-even Unprofitable
All programs 64 14 22
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 64 15 20
Respond to community credit
needs or promote
community growth and
stability 63 14 23
Improve institution's public
image 61 10 29
Earn additional profits or
identify profitable
new markets 74 7 19
Maintain market share in
face of increased
competition(1) 63 12 25
Minimize likelihood of
adverse public comment
on CRA record 60 10 30
Year program established
Before 1990 38 22 40
1991-94 59 11 31
1995-97 63 17 20
1998-99 75 15 10
Program features and
underwriting variances
provided by bank(2)
Services 65 12 23
Subsidies 61 16 23
Altered terms 66 13 21
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 64 15 22
Altered standards for
credit quality and
employment history 66 12 22
Waivers or reductions in
security requirements 58 17 25
Multiple variances 64 14 22
Small programs
Characteristic
Profitable Break-even Unprofitable
All programs 67 16 17
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 65 18 16
Respond to community credit
needs or promote
community growth and
stability 66 17 17
Improve institution's public
image 64 10 26
Earn additional profits or
identify profitable
new markets 86 3 10
Maintain market share in
face of increased
competition(1) 73 12 15
Minimize likelihood of
adverse public comment
on CRA record 65 6 29
Year program established
Before 1990 (*) (*) (*)
1991-94 67 12 22
1995-97 57 21 22
1998-99 80 14 6
Program features and
underwriting variances
provided by bank(2)
Services 72 14 14
Subsidies 62 17 21
Altered terms 72 14 15
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 64 18 19
Altered standards for
credit quality and
employment history 67 11 21
Waivers or reductions in
security requirements 61 18 21
Multiple variances 68 16 16
Large programs
Characteristic
Profitable Break-even Unprofitable
All programs 60 11 29
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 63 11 26
Respond to community credit
needs or promote
community growth and
stability 59 11 30
Improve institution's public
image 57 10 33
Earn additional profits or
identify profitable
new markets 64 10 25
Maintain market share in
face of increased
competition(1) 55 12 33
Minimize likelihood of
adverse public comment
on CRA record 55 14 32
Year program established
Before 1990 44 10 46
1991-94 45 9 46
1995-97 69 12 18
1998-99 (*) (*) (*)
Program features and
underwriting variances
provided by bank(2)
Services 58 10 32
Subsidies 59 15 25
Altered terms 60 12 28
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 64 11 26
Altered standards for
credit quality and
employment history 65 12 24
Waivers or reductions in
security requirements 52 15 33
Multiple variances 60 11 29
B. Smaller institutions
Percentage of programs
All programs
Characteristic
Profitable Break-even Unprofitable
All programs 67 14 20
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 68 15 17
Respond to community credit
needs or promote
community growth and
stability 66 14 20
Improve institution's public
image 64 9 27
Earn additional profits or
identify profitable
new markets 77 6 18
Maintain market share in
face of increased
competition 67 11 22
Minimize likelihood of
adverse public comment
on CRA record 65 9 27
Year program established
Before 1990 46 16 38
1991-94 62 10 29
1995-97 64 18 19
1998-99 77 16 7
Program features and
underwriting variances
provided by bank
Services 68 12 21
Subsidies 64 16 20
Altered terms 70 13 18
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 67 15 19
Altered standards for
credit quality and
employment history 70 11 19
Waivers or reductions in
security requirements 63 16 21
Multiple variances 67 14 20
Small programs
Characteristic
Profitable Break-even Unprofitable
All programs 67 17 17
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 65 18 16
Respond to community credit
needs or promote
community growth and
stability 66 17 17
Improve institution's public
image 64 10 26
Earn additional profits or
identify profitable
new markets 87 3 9
Maintain market share in
face of increased
competition 72 12 15
Minimize likelihood of
adverse public comment
on CRA record 65 6 29
Year program established
Before 1990 (*) (*) (*)
1991-94 66 12 22
1995-97 57 22 21
1998-99 80 14 6
Program features and
underwriting variances
provided by bank
Services 72 15 14
Subsidies 62 17 21
Altered terms 71 14 15
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 63 18 19
Altered standards for
credit quality and
employment history 67 11 21
Waivers or reductions in
security requirements 61 18 21
Multiple variances 68 16 16
Large programs
Characteristic
Profitable Break-even Unprofitable
All programs 66 10 24
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 72 9 19
Respond to community credit
needs or promote
community growth and
stability 64 10 26
Improve institution's public
image 64 7 29
Earn additional profits or
identify profitable
new markets 67 8 25
Maintain market share in
face of increased
competition 60 10 30
Minimize likelihood of
adverse public comment
on CRA record 64 12 24
Year program established
Before 1990 55 0 45
1991-94 52 5 44
1995-97 72 13 15
1998-99 67 22 10
Program features and
underwriting variances
provided by bank
Services 63 8 28
Subsidies 67 14 19
Altered terms 67 11 22
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 73 9 18
Altered standards for
credit quality and
employment history 76 10 14
Waivers or reductions in
security requirements 70 9 21
Multiple variances 65 10 25
C. Large institutions
Percentage of programs
All programs
Characteristic
Profitable Break-even Unprofitable
All programs 40 15 45
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 35 16 49
Respond to community credit
needs or promote
community growth and
stability 41 15 44
Improve institution's public
image 38 18 45
Earn additional profits or
identify profitable
new markets 53 18 28
Maintain market share in
face of increased
competition 41 18 41
Minimize likelihood of
adverse public comment
on CRA record 31 17 52
Year program established
Before 1990 0 50 50
1991-94 29 21 50
1995-97 58 8 33
1998-99 (*) (*) (*)
Program features and
underwriting variances
provided by bank
Services 42 16 42
Subsidies 39 17 43
Altered terms 40 13 48
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 42 15 44
Altered standards for
credit quality and
employment history 40 17 43
Waivers or reductions in
security requirements 21 25 54
Multiple variances 42 15 43
Large programs
Characteristic
Profitable Break-even Unprofitable
All programs 38 15 46
Reason for program
establishment or benefit of
program
Help earn a CRA rating of
"Satisfactory" or
"Outstanding" 32 17 51
Respond to community credit
needs or promote
community growth and
stability 39 16 45
Improve institution's public
image 38 18 45
Earn additional profits or
identify profitable
new markets 55 19 26
Maintain market share in
face of increased
competition 38 19 43
Minimize likelihood of
adverse public comment
on CRA record 31 17 52
Year program established
Before 1990 0 50 50
1991-94 23 23 54
1995-97 59 9 32
1998-99 (*) (*) (*)
Program features and
underwriting variances
provided by bank
Services 40 16 44
Subsidies 37 19 44
Altered terms 37 13 50
Underwriting variances
Changes in required
down payment,
debt ratios,
and cash reserves 39 15 46
Altered standards for
credit quality and
employment history 39 17 44
Waivers or reductions in
security requirements 19 26 56
Multiple variances 40 16 44
NOTE. Covers each institution's estimate of 1999 profitability of its
five largest programs. For the small programs, 1999 loan originations
were less than $2 million; for the large programs, originations were
at least $2 million. For smaller banking institutions (tables 9.B and
9.C), year-end 1999 assets were less than $30 billion; for larger
institutions, assets were at least $30 billion. "Profitable" includes
"marginally profitable"; "unprofitable" includes "marginally
unprofitable." See also notes to table 2; values for each set of three
profitability categories may not sum to 100 because of rounding.
(1.) Also includes reasons reported under "other" in the survey.
(2.) For detailed list, see table 6.
(*) Data received on live or fewer programs.
Programs established as a source of additional profits or to identify profitable new markets are an exception to the pattern. For each combination of program size and institution size, except large programs at smaller banking institutions (table 9.B), the proportion of programs established to be a profit source that is reported to have been profitable was substantially higher than the proportion in the all-programs category. Another exception is small CRA special mortgage programs at smaller banking institutions (table 9.B) that were established, at least in part, to improve an institution's public image or to minimize the likelihood of adverse public comment on the institution's CRA record; these programs were more likely to be unprofitable than the all-programs category for smaller institutions. The profitability of CRA special mortgage programs appears to vary with the age of the program, with newer programs generally being more profitable than older programs. As noted earlier, CRA special mortgage programs carry a wide range of features and underwriting variances (table 6). For the most part, no close relationship appears to exist between the features or variances and program profitability (tables 9.A, B, and C). Even when a banking institution offered some form of subsidy, such as a reduced interest rate or a fee waiver or reduction, CRA special mortgage programs generally were not reported to be more unprofitable than the all-programs category for the same size category of program and institution. These service categories reflect, however, only the existence of a feature or subsidy and not necessarily its amount or extent. The survey collected no further information on these items. Responses indicated that the participation of third parties in CRA special mortgage programs was sometimes related to program profitability, depending on the type and number of third parties involved and the nature of their role in the program (tables 10.A and B). Overall, among programs with a third party, 67 percent were profitable compared with 54 percent of programs with no third party. But CRA special mortgage programs at large banking institutions (table 10.A) were less likely to be profitable (more likely to be unprofitable) when a third party was involved in the program than when one was not. By contrast, programs were more frequently profitable when programs with third parties were conducted by smaller banking institutions (table 10.A) and when third parties participated in small programs (table 10.B), although the frequency of unprofitability was about the same for both groups.
10. Third-party involvement in CRA special mortgage programs, grouped
by size of banking institution and size of program and distributed by
profitability category of program, 1999
A. By size of banking institution
Percentage of programs
All-institutions estimate
Third-party
types and activities
Profitable Break-even Unprofitable
All programs 64 14 22
Presence of third parties
Yes 67 9 23
No 54 26 19
Type of third
party when program
has only one type
of third party
Any 67 5 28
Government 75 3 22
Government-sponsored 47 10 43
Nonprofit 61 7 31
Other 91 0 9
Type of third party
when program
has multiple types
of third parties
Any 68 16 16
Government 70 14 16
Government-sponsored 73 13 14
Nonprofit 67 16 17
Other 80 16 4
Third-party activities(1)
Services 67 9 24
Subsidies 70 12 19
Assumption of risk 74 10 17
Smaller
Third-party
types and activities
Profitable Break-even Unprofitable
All programs 67 14 20
Presence of third parties
Yes 71 9 20
No 54 28 18
Type of third
party when program
has only one type
of third party
Any 71 5 24
Government 78 2 20
Government-sponsored 49 10 41
Nonprofit 67 7 26
Other (*) (*) (*)
Type of third party
when program
has multiple types
of third parties
Any 72 16 12
Government 75 13 12
Government-sponsored 74 13 13
Nonprofit 73 17 11
Other 82 15 3
Third-party activities(1)
Services 71 9 20
Subsidies 75 10 15
Assumption of risk 83 7 10
Large
Third-party
types and activities
Profitable Break-even Unprofitable
All programs 40 15 45
Presence of third parties
Yes 37 15 48
No 56 11 33
Type of third
party when program
has only one type
of third party
Any 21 11 68
Government 33 17 50
Government-sponsored (*) (*) (*)
Nonprofit 13 13 75
Other (*) (*) (*)
Type of third party
when program
has multiple types
of third parties
Any 48 19 33
Government 50 21 29
Government-sponsored 71 12 18
Nonprofit 48 16 36
Other 67 22 11
Third-party activities(1)
Services 37 15 48
Subsidies 38 18 44
Assumption of risk 32 19 48
B. By size of program
Percentage of programs
All program
Third-party
types and activities
Profitable Break-even Unprofitable
All programs 64 14 22
Presence of third parties
Yes 67 9 23
No 54 26 19
Type of third
party when program
has only one type
of third party
Any 67 5 28
Government 75 3 22
Government-sponsored 47 10 43
Nonprofit 61 7 31
Other 91 0 9
Type of third party
when program
has multiple types
of third parties
Any 68 16 16
Government 70 14 16
Government-sponsored 73 13 14
Nonprofit 67 16 17
Other 80 16 4
Third-party activities
Services 67 9 24
Subsidies 70 12 19
Assumption of risk 74 10 17
Smaller
Third-party
types and activities
Profitable Break-even Unprofitable
All programs 67 16 17
Presence of third parties
Yes 74 9 17
No 48 36 16
Type of third
party when program
has only one type
of third party
Any 75 3 22
Government 76 0 24
Government-sponsored 64 0 36
Nonprofit 72 8 20
Other (*) (*) (*)
Type of third party
when program
has multiple types
of third parties
Any 70 24 6
Government 73 20 7
Government-sponsored 72 22 7
Nonprofit 67 33 0
Other 71 24 5
Third-party activities
Services 73 10 17
Subsidies 76 11 12
Assumption of risk 83 10 7
Large
Third-party
types and activities
Profitable Break-even Unprofitable
All programs 60 11 29
Presence of third parties
Yes 59 9 32
No 62 15 23
Type of third
party when program
has only one type
of third party
Any 54 9 37
Government 74 6 20
Government-sponsored 22 24 54
Nonprofit 33 5 62
Other (*) (*) (*)
Type of third party
when program
has multiple types
of third parties
Any 66 10 24
Government 67 10 23
Government-sponsored 74 8 18
Nonprofit 66 9 25
Other 91 6 3
Third-party activities
Services 59 9 32
Subsidies 62 12 26
Assumption of risk 66 9 26
NOTE. See general note to table 9.A.
(1.) For list sec Table 5.
(*) Data received on live or fewer programs.
Programs that exclusively involved government entities as third parties generally appeared to be more profitable than other programs. Overall, 75 percent of CRA special mortgage programs that involved only a government entity were reported to be profitable (table 10.A), compared with 64 percent of all CRA special mortgage programs. Although half of such programs at large banks were reported to be unprofitable, large banks reported only a small number of third-party programs that exclusively involved a government entity (data not shown). The results for other types of third-party involvement are not consistent. For example, large programs (table 10.B) that exclusively involved a government-sponsored entity (Fannie Mae, Freddie Mac, or the Federal Home Loan Banks) were less likely to be reported as profitable than large programs considered as a group (22 percent versus 60 percent). However, large programs that involved a government-sponsored entity as well as other third parties were more likely to be profitable than large programs considered as a group (74 percent versus 60 percent). Overall, the number of third parties involved in a CRA special mortgage program did not seem to bear a strong relationship to program profitability. As noted, third parties conduct a wide range of activities in support of CRA special mortgage programs (table 5), and these activities appear to be related to program profitability (tables 10.A and B). Third-party assumption of risk was positively related to program profitability. This relationship was strongest for programs at smaller banking institutions and for small programs. For example, 83 percent of the programs at smaller banks that involve the assumption of risk by a third party were reported to be profitable (table 10.A), compared with 67 percent for all programs at such banks. Third-party provision of services and subsidies also appeared to contribute to program profitability, although these relationships were not as strong as that for the assumption of risk. Survey results indicate that characteristics of a banking institution besides its size are related to the profitability of CRA special mortgage programs (tables 11.A, B, and C). Banking institutions with overall profitability above the median were more likely to have had large programs that were profitable than were institutions with overall profitability below the median (table 11.A).(12) However, the reverse is true for small programs at smaller banking institutions (table 11.B).
11. Selected characteristics of institutions with CRA special mortgage
programs, grouped by size of banking institution and size of program
and distributed by profitability category of program, 1999
A. All banking institutions
Percentage of programs
All programs
Characteristic
Profitable Break-even Unprofitable
All programs 64 14 22
Profitability of banking
institution(1)
Above median 71 3 27
Below median 62 17 21
Merger by banking
institution in
year program
was established
(for mergers since
1990)(2)
Yes 60 9 31
No 66 17 17
Structure of banking
institution
Multibank holding
company 60 9 32
Other 69 20 11
CRA rating of banking
institution
"Satisfactory" 67 12 21
"Outstanding" 60 16 24
Small programs
Characteristic
Profitable Break-even Unprofitable
All programs 67 16 17
Profitability of banking
institution(1)
Above median 55 5 40
Below median 69 19 12
Merger by banking
institution in
year program
was established
(for mergers since
1990)(2)
Yes 68 7 25
No 66 23 11
Structure of banking
institution
Multibank holding
company 60 10 31
Other 74 23 3
CRA rating of banking
institution
"Satisfactory" 70 15 15
"Outstanding" 63 18 19
Large programs
Characteristic
Profitable Break-even Unprofitable
All programs 60 11 29
Profitability of banking
institution(1)
Above median 80 1 18
Below median 51 15 34
Merger by banking
institution in
year program
was established
(for mergers since
1990)(2)
Yes 45 11 44
No 66 11 23
Structure of banking
institution
Multibank holding
company 59 8 33
Other 61 15 24
CRA rating of banking
institution
"Satisfactory" 63 9 28
"Outstanding" 57 13 30
B. Smaller banking institutions
Percentage of programs
All programs
Characteristic
Profitable Break-even Unprofitable
All programs 67 14 20
Profitability of banking
institution
Above median 73 2 26
Below median 65 17 18
Merger by banking
institution in
year program
was established
(for mergers since
1990)
Yes 65 8 27
No 68 17 16
Structure of banking
institution
Multibank holding
company 63 7 29
Other 70 21 10
CRA rating of banking
institution
"Satisfactory" 69 13 17
"Outstanding" 63 15 22
Small programs
Characteristic
Profitable Break-even Unprofitable
All programs 67 17 17
Profitability of banking
institution
Above median 57 5 39
Below median 69 19 13
Merger by banking
institution in
year program
was established
(for mergers since
1990)
Yes 68 7 24
No 66 23 11
Structure of banking
institution
Multibank holding
company 60 10 31
Other 74 23 3
CRA rating of banking
institution
"Satisfactory" 70 15 15
"Outstanding" 63 18 18
Large programs
Characteristic
Profitable Break-even Unprofitable
All programs 66 10 24
Profitability of banking
institution
Above median 84 0 16
Below median 57 15 28
Merger by banking
institution in
year program
was established
(for mergers since
1990)
Yes 55 11 34
No 69 9 21
Structure of banking
institution
Multibank holding
company 69 4 27
Other 63 16 21
CRA rating of banking
institution
"Satisfactory" 69 10 21
"Outstanding" 63 10 27
C. Large banking institutions
Percentage of programs
All programs
Characteristic
Profitable Break-even Unprofitable
All programs 40 15 45
Profitability of banking
institution
Above median 55 9 36
Below median 36 16 48
Merger by banking
institution in
year program
was established
(for mergers since
1990)
Yes 33 10 57
No 48 20 32
Structure of banking
institution
Multibank holding
company 39 16 45
Other 45 9 45
CRA rating of banking
institution
"Satisfactory" 44 4 52
"Outstanding" 37 23 40
Large programs
Characteristic
Profitable Break-even Unprofitable
All programs 38 15 16
Profitability of banking
institution
Above median 60 10 30
Below median 33 17 50
Merger by banking
institution in
year program
was established
(for mergers since
1990)
Yes 30 11 59
No 48 20 32
Structure of banking
institution
Multibank holding
company 37 17 46
Other 45 9 45
CRA rating of banking
institution
"Satisfactory" 39 4 57
"Outstanding" 38 24 38
NOTE. See notes to table 9.A.
(1.) Return on equity in 1999 compared with 1999 return on equity of
the 500 largest retail banking institutions.
(2.) Includes acquisitions by an institution's holding company.
Merger activity also appears to matter. In all combinations of program size and banking institution size, programs that were established in a year in which the banking institution was engaged in mergers or acquisitions were more likely to have been reported as unprofitable than programs established in years in which the banking institution did not merge. Whether or not the banking institution is part of a multibank holding company appears to matter only for small programs at smaller banking institutions (table 11.B): Such programs at institutions that are part of multibank organizations tend to be less profitable (or more unprofitable) than programs at independent institutions. Finally, with the exception of large programs at large banking institutions (table 11.C), a banking institution's CRA performance rating does not appear to be strongly related to program profitability. For large programs at large institutions, the percentage of programs reported as profitable for institutions with outstanding CRA ratings was the same as for institutions with satisfactory ratings. However, somewhat surprisingly, the proportion reported to have been unprofitable is much higher for those with satisfactory ratings than for those with outstanding ratings. LIMITATIONS OF THE ANALYSIS The survey and resulting data provide new and systematic information about the characteristics and the role of CRA special mortgage programs in the lending activities of banking institutions. In particular, the information provides opportunities to determine the factors that influence the performance and profitability of these programs and to better understand the role of specific program features and arrangements with third parties. However, the survey data do not address all issues in this regard, and the foregoing analysis has some important limitations. First, by design, the survey collected detailed information on only a subset of CRA special lending programs from a fairly narrow group of lenders and programs. Only the five largest programs at each of the largest 500 retail banking institutions were covered in the survey; the characteristics and profitability of smaller programs at these institutions or of programs at other banking institutions may have differed. Second, the survey collected performance and profitability information for 1999 only. However, 1999 lending experiences may not be representative of those of other years. For example, program performance may have been better in 1999--a year of ongoing economic expansion--than might have been expected |