Page 1 of 9
PUBLIC MONEY FOR SMALL BUSINESSES: AN EVALUATION OF EVANSVILLE'S METRO SMALL BUSINESS ASSISTANCE CORPORATION Laurence R. Steenberg University of Evansville ABSTRACT The Metropolitan Small Business Assistance Corporation (MSBAC) is a tax exempt certified development corporation. Its purpose is to use public money to aid the growth of small businesses in the Evansville Indiana area. In order to accomplish the mission described above, MSBAC must operate like a bank with both a risky loan portfolio and a social purpose. This paper seeks to evaluate MSBAC in terms of its own stated purposes through an analysis of its lending history between 1983 and 1993. That analysis shows that MSBAC performs it basic functions well. Worthy small business projects are gaining access to financing. The lending activity is being managed well. Loan losses are low and funds from repaid loans have been recycled to make additional loans. Analysis of the data also suggests that MSBAC should focus more upon the small loan segment of its business. Small loans create equal numbers of jobs more cheaply than large loans. Were more small loans to be made at the expense of some larger loans, the community would have more jobs at a lower cost. INTRODUCTION The Metropolitan Small Business Assistance Corporation (MSBAC) is a tax exempt certified development corporation. Its purpose is to use public money to assist the growth of small businesses in the Evansville, Indiana area. MSBAC deals with small businesses that are unable to obtain sufficient financing for capital improvement or expansion through conventional channels. Its resources come exclusively from government grants set up for this purpose. Often MSBAC combines its resources with private financial institutions and SBA loan guarantees. MSBAC loans are often collateralized by inventory, accounts receivable, equipment, real estate, and personal guarantees of the owners. When other institutions participate, MSBAC's security interests are subordinated to the private lenders. Loans are evaluated in terms of both the economic value of the business project and the number of jobs that the project will create or save. MSBAC looks to prospective cash flows as the primary source of repayment.
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 2 of 9
In order to accomplish the mission described above, MSBAC must operate like a bank with both a risky loan portfolio and a social purpose. It must discriminate between projects that are not realistic and projects that are worthy but not suitable for conventional financing. If MSBAC is too conservative, the capital needs of small business will not be met. If MSBAC is too lenient, it will waste public money. If it does not seek both job creation and business development, it will not serve the social purposes which are its foundation. During the past eleven years, MSBAC has attempted to fulfill its mission by participating in 139 loans to small businesses totalling $12,104,486. This activity has resulted in the creation or preservation of 1713 jobs within the local community. The cost to the taxpayer has been $102,521 in losses on nine defaulted loans. MSBAC's operating costs have been $831,807.(1) This paper seeks to evaluate MSBAC in terms of its own stated purposes. The lending history of MSBAC between, l 983 and 1993 was analyzed to find answers to the following questions. 1. Is MSBAC managing its share of the public's money well? 2. Is MSBAC successfully funding small businesses that could not get adequate financing from conventional sources? 3. Is MSBAC creating jobs through its activities? 4. Is the cost of the jobs so created reasonably low? MSBAC AS A LENDER The performance of a bank is evaluated in terms of profitability, loan quality, and risk. Profitability is measured by the rate of return on the assets employed or the rate of return on the equity capital of the banking corporation. Since equity is an idea that does not apply well to a tax exempt company, MSBAC will be evaluated by its rate of return on the total assets of the company (ROA). The assets of MSBAC primarily consist of cash and investments, loans receivable from small business clients, and a small amount of office equipment. During the period studied here, total assets have grown from less than $200,000 to more than $2,700,000 as shown on Exhibit 1. Grants from governmental agencies have financed all of this growth.(1) The rate of growth was rapid between 1983 and 1988, but has been slower since 1989.
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 3 of 9
The income of MSBAC consists of interest, service fees, and grants, in agreement with the principles of fund accounting. Some of the grants made to MSBAC contained a requirement that the income generated from the grant be returned to the grantor at an appropriate time. Revenue is reported by MSBAC in its financial statements net of these returns. In order to evaluate MSBAC as a lender, the revenue as reported must be adjusted so that the resulting number compares to the data that a bank would report. Two adjustments were necessary. First, grants were removed from income since, in a taxable corporation, they would be treated as contributions of capital. Second, income returns to grantors were added back since they represent income from the loans made by MSBAC. These returns are more analogous to dividends than revenue reductions. The resulting modified income numbers represent the profits from the lending activity. Using adjusted income figures, MSBAC has been profitable in all but the first three of the eleven years reviewed here. Net income has ranged from a loss of $57 thousand to a gain of $73 thousand as shown on Exhibit 2. ROA was negative during the first three years but has varied from 1.6% to 2.6% since then as shown on Exhibit 3. For the eleven years taken together, income totaled over $345,000 for an average ROA of 1.6%.(1) These rates of return are better than a good bank, poorer than a good small loan company, but within the realm of possibilities for both. Loan quality has also been good over the eleven years studied. MSBAC has participated in 139 loans during the period which totaled $12,104,486. Nine of these loans have resulted in default amounting to losses of $102,521. More than 40% of the loss occurred in 1984. Since that time losses have remained below 1.3% of outstanding loans."' This number shows loan quality that many banks would be proud to report. Federal government data for the banking system in recent years show overall losses for commercial loans in the range of 0.5% to 1.0%. Risk is commonly measured in terms of loan to deposit ratio and loan quality. MSBAC seeks grants for the purpose of funding loans. These grants are the closest analogy to bank deposits. Given MSBAC's purpose, all of the grant money received in excess of operating expenses should be dispersed as loans. In addition, some of the money originally lent should have been repaid and lent again. Exhibit 4 shows, on a cumulative basis, the ratio of new loans to grants in excess of operating expenses. Since 1987, this ratio has been greater than
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 4 of 9
1, showing that some of the grant money was recycled after the initial loan was repaid. Exhibit 5 shows the cash position of MSBAC.(1) Liquidity has been rising during the entire eleven year period studied. This shows that the repaid loans are not invested as aggressively as the new money obtained from grants. MSBAC activities during 1994 have begun to address this problem. From a purely financial point of view, the performance of MSBAC has been good. Profitability has been high, losses have been low, and the money provided has been used to promote small business growth. The low rate of loss suggests that the lending standards are too conservative. The loss rates on unbankable loans look similar to data published by some banks for conventional commercial loans. Conservatism also shows in that money from repaid loans does not seem to be lent out as aggressively as does new money from grants. MSBAC AS A JOB CREATOR MSBAC operates as a publicly funded Certified Development Corporation. Presently, there are about 400 such corporations in operation dispersing a federal allocation of $1.5 billion. Many of these corporations, including MSBAC, also disperse additional money appropriated by state and local governments.(3) The stated purpose of the Certified Development Corporation program is the creation of new jobs. The reports submitted by these organizations to both the government and the public highlight job creation and preservation. Both the number of jobs and the cost of their creation is shown. The government reporting requirements define the number of jobs created as the sum of the new jobs and the existing jobs salvaged by the loan. The formula for the cost of job creation is the number of jobs divided by the total amount of the loan made.(2) The performance of MSBAC during the last eleven years is summarized in Exhibit 6. Using government definitions, 1,71 3 jobs have been created by $12,104,486 in loans for an average cost of $7,084/job.
Exhibit 6 Job Creation By MSBAC Year No. Total Loans Made 1983 Loan Average Size No. of Created $120,765 292 $6,204 Loan/Job
Amount Of Of Loan Jobs
15 $1,811,477
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 5 of 9
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993
22 $1,721,051 10 19 14 $981,790 $2,42.557 $890,361
$78,230 226 $98,179
$7,615
59 $16,841
$133,819 153 $16,618 $63,597 39 $22,830 $2,567
12 $1,586,400 9 5 11 9 13 $496,750 $195,000 $867,000 $311,000 $731,100
$132,200 618 $55,194 $39,000 $78,818 $34,556 $56,2381
20 $24,838 39 $5,000
35 $24,771 41 91 $7,585 $3,828
Total 139 $12,134,486
$87,298 1,713 $7,084
Calculating the cost of a created job in this way is misleading. MS@AC disburses money in the form of loans. MSBAC fully expects the loans to be repaid with interest. If MSBAC made a loan, jobs were created, and the loan repaid with interest, it can be argued that there was no cost to the public other than the cost of operating MSBAC. The opportunity cost to the government of the grant money was less than the interest charged to the small business borrower. The cost of job creation would be better defined as the value of defaults suffered by MSBAC added to its cost of operation. Defaults between 1983 and 1993 totaled $102,521. If we add an estimated provision for future losses from current loans equal to 50% of all loans not now current, losses from the program would be $144,715. The cost of operations during those eleven years was $831,807.(1) Combining these numbers, the cost of a job created or salvaged by MSBAC is $570. Compared to the cost of unemployment insurance, the cost of welfare, or the anticipated employment tax revenue, this is an attractive result. MSBAC has participated in projects ranging in size from $2,665 to $820,000. (1) This span includes businesses of widely differing nature and sophistication. The data from all of these projects were analyzed to determine whether job creation depended on the size of the project. Two linear regressions were used. The results are summarized in Exhibit 7.
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 6 of 9
The first regression attempted to predict the number of jobs created as a function of the loan size. The coefficient of determination (R2) is 0.078 which is quite low. The scatter diagram shown on Exhibit 8 confirms that loan size is a poor predictor of the number of jobs created. The second regression attempted to predict the cost of job creation as a function of loan size. In this case, the coefficient of determination is 0.28 (R=0.53) and the slope of the regression line is positive (0.11). Exhibit 7 Summary of Regression Results Regression #1 Independent Variable Dependent Variable Y Intercept X Coefficient R^2 Number of Observation Pairs 128 128 2.84 6.20e-05 .078 .28 Regression Loan Size, $
Loan Size, $ 5885 .11
Number of Jobs Cost of Jobs
The cost of job creation increases as loan size increases. The scatter diagram shown on Exhibit 9 confirms this. Small projects are more efficient creators of jobs. Each of these conclusions suggests that the social goals of MSBAC are best served when its funds are applied to small projects requiring small loans. For any total amount of loan volume, smaller average loan size creates more jobs at a lower cost.
MSBAC AS A SOURCE OF CAPITAL In order for a small business to request funding from MSBAC, it must first be turned down by a bank. Sometimes, banks refuse the entire project and the business turns to MSBAC for financing. On other occasions, a bank agrees to lend some of the required amount and the business looks to MSBAC for the rest. Conventional wisdom suggests that those projects partially financed by a bank carry lower risk. This belief relies upon the willingness of the bank, after its analysis, to participate with MSBAC. The bank demands that .MSBAC take a subordinate position and, as a result, views the MSBAC funds as similar to equity. The Chi Square method was used in a 2 x 2 contingency table to see if defaults were related to bank participation. The results of this test are shown on Exhibit 10. As can be seen, the null hypothesis cannot be rejected.
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 7 of 9
In other words, the loans rejected by banks are no more risky than those accepted by banks in participation with MSBAC. Exhibit 10 Summary of Chi Square Tests NULL HYPOTHESIS NO. 1: Businesses which must borrow from MSBAC alone are no more likely to default than businesses that may borrow from both banks end MSBAC. Loans by Loans by Banks and MSBAC 3 67 73 63 66 9 130 139 All Loans
MSBAC Alone Defaults Good Loans All Loans 6
Chi Square calculated: 1.124 Chi Square for 5% significance level: 5.9 Null hypothesis cannot be rejected NULL HYPOTHESIS NO. 2: The rate of default on small loans is no higher than the rate of default on larger loans. Loans <425,000 Loans >$25,000 All Loans Defaults Good Loans All Loans 7 73 80 2 57 59 9 130 139
Chi Square calculated: 1.9261 Chi Square for 5% significance level: 5.9 Null hypothesis cannot be rejected Another belief commonly held is that small loans are more risky than large loans. This belief, too, was tested from the MSBAC data using a 2 x 2 contingency table. The dividing line between small and large was set first at $25,000. The table compared loan size and risk measured by default. The results are shown on Exhibit 10. Again, the null hypothesis cannot be rejected and it appears that there is no relation between loan size and default. Similar result was obtained using $15,000 as the dividing line instead of $25,000. The issue of small loans being riskier was also analyzed using a T-test. The mean size of defaulted loans was compared to the mean size of loans which did not default. The results, summarized on Exhibit 11, show that the average
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 8 of 9
defaulted loan is significantly smaller than the average non-defaulted loan at a high level of significance. In other words, loans which default tend to be smaller. Exhibit 11 Results of T-Tests NULL HYPOTHESIS NO. 3: The mean size of loans which defaulted is the same as the mean size of loans which did not default. Defaulted Loans Number Mean size, Standard Deviation, T Statistic: 9 Good Loans 126 $26,222 $44,198 2.7372 0.0105 All Loans 135 82,621 161,698
Probability > T:
Hypothesis no. 3 is rejected The low default rate for all MSBAC activity affects both Chi Square results. The total number of defaults during the eleven years studied is 9 out of 139 loans. With such small numbers, tests based on nominal data are not very powerful. The low default rate makes it difficult to conclude that any one segment defaults at a significantly higher rate than any other segment. It is possible to conclude, however, that since the entire group of 139 loans were rejected by banks, the bank loan review standards are very conservative. It is also possible to conclude that MSBAC loan review standards are very conservative, and to ask whether the community would be better served if more loans were made and the loss rate allowed to increase a little. CONCLUSIONS MSBAC is performing it basic functions well. The basic premise supporting M SBAC, and other similar organizations, is that many worthy small business projects cannot obtain funding through conventional channels. The data for MSBAC clearly support this premise, at least for the Evansville, Indiana area. MSBAC operation has allowed over $12 million in small business financing that would not have occurred otherwise. As a direct result, the community has benef itted from the creation of more than 1700 jobs at a real cost of less than $600/job. The lending activity is being managed well. Loan losses are low and funds from repaid loans have been recycled to make additional loans. Viewed as a
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004
Page 9 of 9
high risk lender, the performance numbers for MSBAC are good. Since the basic purpose of MSBAC is social, it is fair to ask whether the numbers are too good. Analysis of the data suggests that MSBAC should focus more upon the small loan segment of its business. Small loans create equal numbers of jobs more cheaply than large loans. Were more small loans to be made at the expense of some larger loans, the community would have more jobs at a lower cost. The data further suggest that the risk of small loans of this kind is no worse than larger loans. Since loan losses are very low, a modest relaxation of credit standards to attract more small loans would be a reasonable strategy for MSBAC to adopt. Should MSBAC elect to consider this strategy, additional study should be done. Records of rejected loans should be analyzed to determine the rate of rejection, the credit worthiness of the rejected loans compared with the accepted loans, and the criteria used for rejection. Acknowledgement: The author wishes to thank Dr. Dale Hockstra, Professor of Management at the University of Evansville, for his invaluable assistance in the analysis and presentation of the statistical test data. REFERENCES Annual Reports: Metro Small Business Assistance Corporation; for the years 1983 through 1993. Private Conversations with Ms. Debra Lutz, Executive Director of The Evansville Metro Small Business Assistance Corporation. Private Conversation with Ms. Gail Hepler, Financial Analyst, Small Business Administration; Office of Rural Affairs and Economic Development. U.S. Department of Housing and Urban Development Community Development Block Grant Program: Grantee Performance Reports for the Period 01-01-79 through 12-31-91.
file://C:\Documents%20and%20Settings\Dr.%20Don%20Bradley\Desktop\1995\SB... 6/3/2004