Email
Print Friendly
A A A

Community Development

Trends in Small Business and Small Farm Loans

The Community Reinvestment Act (CRA) of 1977 encourages commercial banks and savings institutions to help meet the credit needs of the communities they are chartered to serve. They must do so in a manner consistent with sound banking practices, particularly in neighborhoods that suffer from disinvestment and among lower-income residents and businesses that may not benefit from ready access to mainstream banking products.1 I look at temporal and geographic changes of these loans in the Federal Reserve Sixth District (hereafter, the district) in comparison to the United States.

CRA Data
A detailed account of CRA data is documented in Bostic and Canner (1998)2 and the information here is derived from that study. The institutions covered by the CRA data reporting requirements3 disclose information on the number and dollar amount of small business and small farm loans and community development loans. I limit my discussion to small business and small farm loans because geographic information on community development loans is not available in the data. (Community development loans are those not reported under home purchase, small business, small farm, or consumer loans. They qualify as community development loans if they are for projects to attract new or retain existing businesses in low- and moderate-income communities.)

For small businesses, the maximum loan size is $1 million and for small farms, $500,000. The small business and small farm loan data are grouped into three loan size categories. These are $100,000 or less, $100,001 to $250,000, and more than $250,000 up to the respective maximum for each loan type. Both the number of loans and the aggregate amount are reported for each category. The geographic location information (Census tract, county, etc.) of the firms and farms receiving the loans are also in the data. Borrower information such as income, gender, education, racial or ethnic background, and information from applications turned down or withdrawn are not reported under the CRA. The loan numbers and amounts are categorized as loan originations and purchases; the majority are originations. Only about 2 percent of small business loans and less than 1 percent of small farm loans were reported as purchases from another institution. Therefore, the following analysis is based only on loan originations. Even though the reported CRA data do not include all small business and small farm lending by all commercial banks and other financial institutions in the country, Bostic and Canner (1998) noted that reported institutions in CRA data represented a significant portion of such lending in 1996.4 This is also true for more recent years. Williams (2012)5 reports that the percentage of all regulated financial institutions in the United States submitting CRA information on small business loans increased from 60 percent in 2006 to 73 percent in 2010.

National and district trends
CRA data show about 5.3 million small business and farm loans were reported in the year 2000 for the nation (see chart 1), of which 5.1 million were small business loans and the remaining are farm loans. The district had 887,054 small business and farm loans (see chart 2), 863,916 of which were small business loans. The number of loan originations increased steadily for the country and the district from 2000 to 2004 and showed a sharp increase in 2006 and 2007. The number of total loans reported for 2007 was almost 14 million for the country and nearly 2.1 million for the district. The majority of these loans, ranging from 92 to 96 percent, regardless of the year, were for less than $100,000 for both the nation and the district.

chart 1

While the reasons for the sudden increase in the number of loans for both the country and the district in 2006 and 2007 are unclear, a similar trend was noted in overall small business loans reported by all depository lending institutions for the United States, not just CRA lending institutions (Williams, 2012). The sudden drop in the number of loans between 2007 and 2010 may have been due to weak economic conditions as a result of the Great Recession. Under recessionary conditions, local firms are subject to uncertain prospects and hence pose more credit risks, leading financial institutions to limit the amount of credit they are willing to offer to borrowers (Bostic and Canner, 1998). Small business owners often use the equity in their homes or other collateral to borrow. Therefore, the impact of housing value declines, problems with personal credit, and other negative market trends may have also affected the ability of small businesses and small farms to borrow.6 The number of loans in 2010 was the lowest over the 11-year time period (2000–2010) considered in this analysis. The number of total loans declined some 4.4 million in 2010 for the country, a decrease of 68 percent from 2007. The respective 2010 figure for the district was 568,634, a decrease of 73 percent from 2007. Most of these declines came from small business loans. The percent decreases in the number of farm loans were 33 percent for the country and 40 percent for the district, respectively.

chart 2

An aggregated total of $186 million in small business and farm loans under CRA (see chart 3 and the table), of which $174 million were small business loans, was reported for the year 2000 for the nation. The respective figure for the district was $27 million (see chart 4 and the table), of which $26 million were small business loans. Reflecting the trend in the number of loans, loan amounts increased steadily through 2007 except for the year 2005, which reported a temporary decrease from the previous year for both the country and the district. In 2007, the overall total small business and farm loans were $341 million and $52 million, respectively, for the United States and the district. Out of the total loans in 2007, the table shows small business loans were $328 million for the country and $51 million for the district. Similar to the trends in the number of loans, during the Great Recession and recovery period, the loan amounts also declined sharply. In 2010, total loan amounts for the United States were $187 million; total loan amounts for the district were $24 million. These declines reflect a 45 percent drop for the nation and 54 percent drop for the district, respectively, from 2007 levels. The 2010 loan amounts were the lowest for the 11-year time period (2000–2010) considered in this analysis.

chart 3

chart 4

table

As for the distribution of these funds among various types of loan categories, in 2000, for the country, 35 percent of the total amount was for loans of less than $100,000, 20 percent went to loans of $100,000 to $250,000, and 45 percent was for loans of more than $250,000 (see chart 5). The figures for the district for the same year followed a similar pattern: 38 percent went to loans of less than $100,000, 20 percent for $100,000 to $250,000, and 42 percent for over $250,000 (see chart 6). The share of the dollar amount dedicated to smaller loans in the nation rose steadily from 2001 to 2007, from 31 to 44 percent, then declined to 32 percent in 2010. A similar trend was noted for the district: the share increased from 31 percent in 2001 to 41 percent in 2007, declined to 27 percent in 2009, then increased slightly (to 29 percent) in 2010. The trend in the distribution of the middle loan amounts showed a similar pattern for both the United States and the district. This category of loans showed a slightly decreasing trend from the beginning of the decade up until 2007, then an increasing trend. As for the larger loans, the dollar amounts of these loans for both the country and the district were pretty stable through 2004, showed a downward trend through 2007, and then increased during the recession and recovery period.

chart 5

chart 6

Distribution of per capita lending by county types
I examine credit flows over time in specific geographies and the distribution of lending across geographic areas. I also look at credit flows and distribution in counties in general and in various county types based on the U.S. Department of Agriculture's rural urban continuum codes.7 This section is based on loan amounts, normalized by the county population, in order to take into consideration the population in a locality. The trend in loan amounts per capita did not differ much between the country and the district over the time period considered; the United States slightly lagged behind the district through 2007 (see chart 7). Per capita total small business and small farm loans for the United States increased from $651 in 2000 to $1,119 in 2007, before decreasing to $598 in 2010. The respective figures for the district are $675 in 2000, $1,140 in 2007, and $513 in 2010.

chart 7

Next, I examine the time trends of these per capita loans for urban and rural counties in the United States and the district. Chart 8 shows that per capita dollar amounts for rural counties were much lower than urban counties for both the country and the district. The per capita loan amount for U.S. metro counties in 2000 was $538 and $548 for the district. For the same year, nonmetro U.S. counties reported $114, nonmetro district counties, $128. These figures increased for metro counties until 2007, before declining in 2010 to the lowest reported levels for the time period. The loan amounts per capita for nonmetro counties showed an overall downward trend since 2004 with a slight upward tick in 2007.

chart 8

Credit flows by LMI counties
This section examines credit flows by low- and moderate-income (LMI) counties, defined as where the median income is less than 80 percent of the state's median income. For the United States, the share of the total population living in LMI counties was about 12 percent in 2000 and remained around this level through 2010. The institutions covered by the CRA reporting requirements extended about 9 percent of the total number of small business loans and about 8 percent of the total aggregated small business loan funds in LMI counties in 2000. These numbers stayed remarkably stable for the rest of the decade. In 2000, LMI counties received around 26 percent of small farm loans both in numbers of loans and in dollar value. These shares stayed stable through the time period. The share of population living in LMI counties in the district was about 11 percent in 2000. This share declined to about 10 percent in the latter half of the decade. The share of small business loans (in number and value) to LMI counties in the district remained around 6 percent through the reporting period. As for the small farm loans in the district, the shares of both the number and the amount of loans extended to the LMI counties in 2000 were 26 and 34 percent, respectively. These figures went up in 2010 to 32 percent of number of farm loans issued and 37 percent for the dollar amount of farm loans.

Conclusion
As stated in Bostic and Canner (1998), the CRA data do not provide any information on local credit demands and many other factors that may affect the differences in time trends and local loan volumes. The purpose of gathering CRA lending data is to better define the market context and peer activity related to an individual bank's performance in a given year. Therefore, the findings in this report are not intended to present a definitive analysis of lending activity and should be interpreted within the limited scope of the data.

Analysis of the data and geographic comparisons suggest that, overall, the number and dollar amount of loans originated by CRA reporting institutions show similar trends and patterns for the country and the district. There was a steady increase in the number and the dollar amount of loans originated from 2000 to 2007 and a sharp decrease during the Great Recession and recovery period. The decline during the recession and recovery period as a percent of 2007 numbers was sharper for the district than the country for both the number and the amount of loans.

As for the distribution of the funds among various types of loan categories, similar trends were shown for the United States and the district for all three types of loans. The hardest-hit category during the recession and recovery period was the smallest loan category of under $100,000; the $100,000 to $250,000 and over $250,000 loans remained somewhat stable. Loans of less than $100,000 may have decreased in 2008 and 2009 due to CRA lending institutions being more careful due to loss or lack of collateral (such as decline of real estate values). For example, the Federal Reserve Senior Loan Officer Opinion Survey on Bank Lending Practices reported that a significant share of domestic banks (65 percent in January 2009) tightened lending standards on commercial and industrial loans during 2008 and 2009.8 Laderman and Reid (2010) suggest that bank failures and consolidations may have played a role in the decline of small business loans in recent years. 9

On the other hand, even though CRA reporting data do not show such factors, discouraged borrowers may have played a role in the decreasing trend in loan numbers and amounts. For example, the Atlanta Fed's Small Business Survey reported that some businesses did not seek credit due to unfavorable terms or the anticipation of a denial.10

The trends in loan amounts per capita in counties did not differ much between the country and the district and followed a similar trend over the time period considered. The urban and rural comparisons showed that per capita dollar amounts of small business loans for rural counties are much lower compared to urban counties for both the United States and the district. Although this indicates that the number and dollar amount of small business loans originated in rural counties are not distributed to match the distribution of population, more analysis is needed to see if the loan distribution compares with small businesses in rural counties. The analysis also showed that only a tiny fraction of total small business loans, whether measured in number of loans or in dollar amount, originated in LMI counties. While a more accurate analysis of the distribution of loans will require small business information in LMI counties, the present analysis hints at the possibility that, despite the CRA, availability of and access to small business credit in LMI communities is still a policy issue. In contrast, the share of the number and amount of small farm loans originated in LMI counties ranged from 25 percent to 35 percent for both the country and the district, probably because LMI counties are disproportionately rural.

While the above analyses on time trends and geographic distributions of CRA loans are useful for benchmarking, data on small business dynamics and factors affecting such dynamics are required to be able to assess the impact of CRA loans on small business and small farm outcomes and economic performance in LMI communities. I continue to research the effects of small business loans on small businesses creation and destruction at the state level.

By Anil Rupasingha, research adviser and economist in the Atlanta Fed's community and economic development group

_______________________________________

1 The regulations that enforce CRA compliance were revised in 1995 and among the changes were institutional reporting requirements. Under these reporting requirements, banking institutions are required to collect and report annually the number and amount of small loans, including the locational information of these loans to small businesses and small farms (firms with gross annual revenues of $1 million or less) and any community development loans. The new data permit analysis of trends of the loans reported under the act and the impact of these loans on various economic outcomes. The data also allow for a more systematic evaluation of bank performance outcomes under the CRA, in the context of peer activity during the same time period, to complement the qualitative evaluation of product and service development and deployment strategies.

2 Bostic, R.W., and G.B. Canner. 1998. "New information on lending to small businesses and small farms: The 1996 CRA data." Federal Reserve Bulletin (Jan):1-21.

3 All institutions regulated by the Office of the Comptroller of the Currency, Federal Reserve System, Federal Deposit Insurance Corporation, and the Office of Thrift Supervision that meet the asset size threshold are subject to data collection and reporting requirements. The asset size threshold that triggers data collection and reporting for all agencies for 2013 is $1.186 billion as of December 31 of each of the prior two calendar years. Please see Community Reinvestment Act data.

4 CRA reporting institutions extended about two-thirds of all small business loans and about one-fifth of all small farm loans granted by such institutions in 1996 (Bostic and Canner, 1998).

5 Williams, V. 2012. Small Business Lending in the United States 2010–2011. Office of Advocacy U.S. Small Business Administration.

6Due to the lack of information on credit applications in small business and small farm lending data reported here, it is not possible to know how supply and demand has changed for these loans.

7 Metropolitan (urban) and nonmetropolitan (rural) definitions are from the USDA 2003 Rural-Urban continuum codes (rucc). Please visit U.S. Department of Agriculture Economic Research Service website for details.

8 Please see the Fed Board's Senior Loan Officer Survey on bank lending practices for more details. January 2010 of this survey reported that although banks started relaxing some strict standards on many loan types in the fourth quarter of 2009, they are yet to undo the tightening that occurred in 2008 and 2009.

9 Laderman, E. and C. Reid. 2010. The Community Reinvestment Act and Small Business Lending in Low- and Moderate-Income Neighborhoods during the Financial Crisis. Working Paper 2010-05. Federal Reserve Bank of San Francisco.

10 Please see the Atlanta Fed's small business survey for more details.