The Community Reinvestment Act (CRA) was enacted in 1977 and revised in 1995. Its intent is to encourage depository institutions to help meet the credit needs of the communities in which they operate, including low- and moderate-income neighborhoods, in ways consistent with safe and sound banking operation. The legislation came as the result of widespread concerns among social activists that many banks drew deposits from the residents and businesses in their local communities and then used those deposits to make loans outside these communities. In addition, it was believed that the CRA would be a corrective tool to combat "redlining"-- banks' refusal to make loans in an area with a significant percentage of low- and moderate-income residents. (1, 4) This research, using Kentucky as a case study area, investigates whether the CRA helps increase money supply by increasing lending in economically distressed areas. An affirmative answer would imply that the CRA can be used as a legitimate policy option to enhance the regional money supply, and thus stimulate growth in economically distressed regions.

In an exchange economy, money is the most efficient vehicle for commodity exchange between places offering different goods and services, and between current consumption and future production. (3) By definition, an economically distressed area suffers from low income and resultant low purchasing power. This means that due to certain underlying causes, the area is unable to generate sufficient goods and services to be used in exchange with other areas to meet current consumption needs. (5) Regional policies that aim to alleviate economic stress in an area essentially tackle these underlying causes and build its future productive capacity by increasing the area's current public fund allocation. Therefore, from an exchange point of view, a successful regional policy, in essence, must create an effective and efficient exchange system between current spending and future production. From a money supply point of view, regional policy is a fund reallocation mechanism applied in order to build future growth in areas that are left out by the market. A vital part of any regional policy is the mechanism by which funds are allocated to regions or local communities.

There have been numerous federal programs designed to promote economic growth and development. In terms of the locations of assistance recipients, these programs can be largely classified as region-based, community-based, and individual clientele-based. Region-based programs target broad regions involving more than one state. A region-wide organization may be created to serve as the vehicle for planning and implementation of the development programs, and administration of federal grants. The first large scale regional project financed by the federal government was the Tennessee Valley project in the 1930s that combined flood control, energy development, irrigation, and navigation. Between the 1930s and 1970s, the federal government also sponsored major water projects in conjunction with basin-wide development in some of the nation's major river basins such as the Columbia, the Colorado, and the Missouri. True regional economic development programs began with the Area Redevelopment programs of the early 1960s. From the Appalachian Regional Commission in the mid 1960s, to the last batch of newly established Title V regional programs of the late 1970s, the entire country was covered, almost wall to wall, by 9 regional programs with their commissions serving as administrative agencies. By the early 1980s, all these regional commissions except the Appalachian Regional Commission disappeared as vehicles for coordinating area development efforts. (2)

The community-based programs target a geographical area smaller than a state, normally at the traditional neighborhood or municipality or sub-municipality level, or the county or sub-county level. A neighborhood development corporation, the municipal government, or a county council serves as the agent in program planning and implementation, budgeting, and administration. Currently, this is one of the most common forms taken by federal programs. Examples include the Community Development Block Grant, and the Empowerment Zones/Enterprise Communities program provided by the Department of Housing and Urban Development (HUD), the Public Works Development Facilities program, Planning Program for States and Urban Areas, Planning Program for Economic Development Districts, Indian Tribes, and Redevelopment Areas, and the University Center Program administered by the Economic Development Administration at the Department of Commerce, as well as the Community Express and U.S. Community Adjustment and Investment Programs offered by the Small Business Administration.

In contrast to the region-wide or community-wide focus of the previous two types of programs, the individual clientele -based programs are designed to provide assistance and direct benefit to individuals or individual organizations in need of development assistance. Examples include numerous farm loan, crop insurance, and agricultural exporting financing programs provided by the Department of Agriculture, and various business loan and loan guarantee, international trade loan, and Export Working Capital programs at the Small Business Administration. Some federal assistance lies between community-based and individual clientele-based programs. Examples include various Business programs offered by the Rural Business-Cooperative Services at the Department of Agriculture which provide loans and loan guarantees to both individual and rural communities.

In terms of the forms of the assistance provided, the federal development programs can be categorized as providing grants, loans, loan guarantees or insurance, and other forms such as tax incentives, and technical assistance. Federal grants require no repayment. Depending upon the program, such grants may be used for the acquisition of property, the construction of physical structures, infrastructure maintenance, the purchase of services, business start up capital, business operating expense, etc. Examples include HUD's Community Development Block Grant and Economic Development Initiative, and the Public Works Development Facilities program offered by the Economic Development Administration within the Commerce Department. Federal loans programs provide funds to those who would not be qualified for loans from conventional commercial lenders. Normally, these loans are provided via certain intermediaries such as conventional lending agencies, non-profit organizations, community-based development corporations, and collectives. These intermediaries receive funds from the federal agencies and make loans and conduct related loan administration to qualified recipients. Examples include the Business and Industry Direct Loan Program and the Intermediary Relending Program administered by the Department of Agriculture, and Community and Individual Investment Corporations financed by HUD. Loan guarantee and insurance programs provide financial backing to the applicants for loans from the conventional lenders. Without this backing, loans would be less likely to be granted due to the applicants' circumstances such as inexperience or lack of creditworthiness. The Economic Development Loan Guarantee Fund (Section 108) funded by HUD, farm loan guarantee programs offered by the Department of Agriculture, and most programs within the Small Business Administration belong in the loan guarantees or insurance category. The Empowerment Zones/Enterprise Communities programs funded by HUD use a mix of programs including grants, loan, loan guarantees, tax incentives, and technical assistance.

From a money supply point of view, various federal development programs classified in the previous section play the role of redirecting the geographic flow of money. The money is collected in the form of taxes from the public across the country and is reallocated via development programs to economically stricken regions, communities, and individuals.

Such a reallocation mechanism is the traditional way of providing public goods and became increasingly important from the 1930s until the late 1970s and early 1980s when the nation was swept by waves of deregulation. Taxation and fund reallocation via the public sector have the advantage that various programs have to go through pubic scrutiny in the form of legislation, public hearings, community participation, and public involvement. Such processes may increase the possibility that projects funded closely match local needs. However, these programs have disadvantages such as subjecting local development to pork barrels politics, the waxing and waning of political interest in publicly funded programs, and legislative cycles.

The Community Reinvestment Act established a new approach to financing development. It turns the financing of local development over to depository institutions and relies on their incentives and discipline to open a source of money to local communities. The CRA and its implementing regulations, require federal financial institution regulators to assess the record of each bank and thrift in helping fulfill their obligations to the community and to consider that record when evaluating applications for charters or for approval of bank mergers, acquisitions, and branch openings. CRA examinations are conducted by the federal agencies that are responsible for supervising depository institutions which include the Federal Reserve (FED), the Federal Deposit Insurance Corporation (FDIC), the Office of the Comptroller of the Currency (OCC), and the Office of Thrift Supervision (OTS). CRA examinations are generally conducted for every depository institutions every other year. Depository institutions that have been examined are given a written performance evaluation, which is made public. The evaluation plays a part in the approval review for any depository institution's merger and acquisition application. In accessing a bank's compliance with the law, federal regulators also take into account public opinion about the institution's CRA performance. As a result, the law establishes a structure for depository institutions and community groups to work together to improve the supply of credit and other banking services to under-served communities. Under the CRA's impetus, banks and thrifts that have an overall business strategy may have incentives to open new branches, provide expanded services, adopt more flexible credit granting criteria, make substantial commitments to local or community development organizations, and increase lending to under served segments of local economies and populations.

This study investigates whether the CRA has significantly increased lending in economically distressed areas using information on CRA loans published by the Federal Financial Institutions Examination Council (FFIEC). The federal regulatory agencies that conduct CRA examinations generally look at three main areas of performance or three performance tests: the lending test, the community development test, and the service test. The lending test focuses on loans, including small business, and small farm loans. The investment test evaluates the degree to which a bank makes qualified investments in its community, for the purpose of creating affordable housing or stimulating economic development. The service test assesses the level of a bank's retail delivery services, such as the accessibility of its branches and the extent to which it provides community development services such as credit counseling or small business technical assistance. The CRA implementation activities of different regulatory agencies are coordinated by the the FFIEC. The FFIEC also receives and publishes reports from these agencies concerning CRA loans.

The CRA loan data published by the FFIEC include the number and amount of small business loans and farm loans made by depository institutions within all counties. Since CRA examinations are conducted by several independent regulatory agencies and the depository institutions examined are distributed across the entire nation, the published CRA loans actually represent a random sample of CRA lending by depository institutions on a yearly basis. This allows a statistical study comparing regions with different characteristics to determine whether or not CRA loans increase lending in economically distressed areas, as is one of its major legislative intents.

The study area selected is the state of Kentucky. All Kentucky counties are grouped into one of three different categories the metropolitan, non-Appalachian, and Appalachian counties (Figure 1). The rationale for such groupings is to control for counties of various economic development status, such as non-Appalachian vs. Appalachian, and to control for counties with different economic structure, such as metropolitan vs. non-metropolitan. The Appalachian counties are those non-metropolitan counties that are also defined as Appalachian by the Appalachian Regional Commission (ARC). There are 44 counties in this category. There are 54 rural non-Appalachian counties. In addition, there are 22 metropolitan counties, comprised of 17 non-Appalachian metropolitan counties, and 5 metropolitan counties designated as Appalachian by ARC.

In 1997, Kentucky counties received a total of $3682 million in CRA loans. Of this total, $3369.6 million were small business loans and $312.4 million were farm loans. A breakdown along the lines of the three regions is given in Table 1.

Sources: Community Reinvestment Act, FFEIC

Given that in 1997, Appalachian counties accounted for 22.7% of Kentucky's total population, and for 16.6% of the total deposits in Kentucky depository institutions, their CRA loans were generally underrepresented in the sample except for the number of farm loans. In comparison, non-Appalachian counties as a whole accounted for 29.05% of Kentucky's population and for 28.7% of the total deposits, and their CRA loans were all over-represented in the sample. Metropolitan counties accounted for 48.3% of the total population and for 54.7% of depository institution deposits. Their CRA loans were over- represented in terms of the amount of small business loans but underrepresented in terms of the farm loans in the sample. The overall picture seems to favor non-Appalachian rural counties, but to discriminate against Appalachian counties.

*Significant at 0.01

Statistical tests on the means of per capita CRA loans among the three regions reveal a similar pattern. First of all, one way analyses of variance indicates that there are significant differences among the three regions in terms of mean per capita farm loans, small business loans, and the total CRA loans at the 0.01 level (Table 2).

In addition, pairwise comparisons of the mean per capita CRA loans reveal that in term of farm loans, small business loans, and total CRA loans, Appalachian counties had a statistically significantly lower average than non-Appalachian counties (Table 3 a). Appalachian counties had a significantly lower average than Metropolitan counties with respect to per capita small business loans and total CRA loans (Table 3 b). In addition, with respect to per capita small business loans, farm loans, and total CRA loans there were no significant differences between non-Appalachian and Metropolitan counties (Table 3 c). This last finding is particularly significant. Since metropolitan counties generally have more business opportunities, and thus attract more lending than rural communities, insignificant differences between non-Appalachian rural counties and metropolitan counties in receiving CRA loans may indicate a significant increase in lending in non-Appalachian rural counties.

a. Appalachian vs. non-Appalachian
 

	Mean per capita small business loans  ($)	Mean per capita farm loans ($)	Mean total CRA loans ($)
Appalachian	252	54	306
non-Appalachian	560	172	731
degree of freedom	96	96	96
t value	-3.26*	-2.68*	-3.63*

b. Appalachian vs. Metropolitan
 

	Mean per capita small business loans  ($)	Mean per capita farm loans ($)	Mean total CRA loans ($)
Appalachian	252	54	306
Metropolitan	652	107	760
degree of freedom	64	64	64
t value	-4.54*	-1.17	-4.07*

c. Non-Appalachian vs. Metropolitan
 

	Mean per capita small business loans  ($)	Mean per capita farm loans ($)	Mean total CRA loans ($)
Non-Appalachian	560	172	731
Metropolitan	652	107	760
degree of freedom	74	74	74
t value	-0.70	1.10	-0.18

Significant at 0.01.

Apparently, on a per capita basis, the rural non-Appalachian region as a whole benefited significantly more from the CRA than the Appalachian region. That is, under CRA, non-Appalachian counties received more lending from depository institutions than their population share would suggest compared with the Appalachian counties. On the other hand, metropolitan areas seem to be in an advantageous position in attracting more loans under CRA than their population share would suggest.

The above conclusions raise the hope that CRA functions as a tool to attract more lending in regular rural communities than would otherwise occur, but also raise doubt concerning the use of CRA to increase the money supply in economically distressed areas. CRA provides incentives to private institutions to provide lending in local communities including those in economically distressed areas. The main incentive for these institutions to conform to the CRA requirements is related to the regulatory approval process with regard to consolidation and other corporate actions. As a result, institutions with geographic expansion ambitions are more likely to conform to the CRA requirements. A simple correlation analysis is conducted between the amount of deposits controlled by a county's depository institutions but located out of the county, on the one hand, and an CRA index on the other hand. The CRA index is constructed by finding the difference between two percentages. The first percentage is the percent of a county's CRA loans in respect to the entire state CRA loans. The second percentage is the percent of the county's deposit with respect to total deposits in the state. The difference between the two percentages represents the extent to which CRA loans are over-represented in a county when compared with the amount of deposits. The correlation coefficient is a significant 0.67. In comparison, the correlation between the CRA index and the amount of the deposits controlled by out of county institutions is an insignificant -0.07. These results seem to suggest that depository institutions with businesses outside their home county respond more readily to the CRA. This may be due to the fact that geographic expansion minded institutions partly count on their CRA performance as a way to gain regulatory agencies' approval of their strategy of geographic expansion. From this point view, CRA achieves its intended mission of using CRA requirements as leverage to modify institutions' lending practice. On the other hand, since counties housing expansion minded institutions tended also to be counties receiving more CRA loans than their share in deposits would suggest, this may imply that a county is more likely to receive CRA loans from local institutions than from outside institutions. In other words, the source of CRA funds originated locally, instead of being drawn from outside. This is largely different from what the CRA activists intended for the Act. If the extra CRA lending is financed by diverting funds from securities purchase, which would reduce lending to the national capital market, CRA still has the effect of increasing local money supply. However, since Appalachian counties did not experience a significant increase in CRA lending, any effect of increasing money supply in economically distressed areas may be marginal.

This study finds that CRA requirements seem to have the effect of increasing lending and thus money supply in rural communities. However, it fails to find evidence of significantly increased local lending in economically distressed rural counties. This study also finds that CRA loans are mainly financed by local instead of outside sources. These findings suggest that CRA may be used as a tool for altering lending behaviors for those institutions that intend to gain favorable regulatory rulings from government agencies. From this point of view, CRA may have the effect of increasing money supply and promoting growth in local communities. However, the long-term effects of the CRA on firms are not clear. Institutions that do not intend to consolidate would have no incentive to conform to CRA requirements. If these institutions increase securities purchase at the national capital market as a response to the increased securities rates due to reduced purchase from CRA compliance institutions, they would reduce local lending. As a result, the overall balance of CRA in improving local lending may not change much. In addition, CRA alone does not seem to constitute a viable tool for increasing money supply and thus promoting growth in economically distressed areas. In these areas, policies that directly promote public projects via fiscal means may have the most immediate effects.

1. Bradford, C. and G. Cincotta. 1992. The Legacy, the Promise, and the Unfinished Agenda of the Community Reinvestment Act. http://www.cpn.org/sections/topics/community/civic_perspectives/legacy1.html

2. Bradshaw, M. 1992. The Appalachian Regional Commission: Twenty-Five Years of Government Policy. Lexington: the University Press of Kentucky.

3. Drake, P.J. 1980. Money, Finance, and Development. Oxford: Martin Robertson.

4. Kuttner, R. 1997. Lessons of the Community Reinvestment Act. http://www.prospect.org/columns/kuttner/bk971111.html

5. Santomero, A.M. and D.F. Babbel. 1997. Financial Markets, Instruments, and Institutions. Chicago: McGraw-Hill, Companies, Inc.