Minority owned community banks: 1995-2004.

Edmonds, Jennifer Echols ; Robicheaux, Sara Helms

ABSTRACT

This paper documents the changes in minority ownership of banks over the past decade, 1995-2004. The number (26 to 62) and percentage (.93% to 1.78%) of minority owned banks has increased from 1995 to 2004. The Asian and Hispanic groups experienced the most growth while the African-American group experienced the least. We segment our sample of community banks over the past decade into minority owned banks (MOB) and non-minority owned banks (NMOB). We find significant differences in the bank characteristics MOBs versus NMOBs. We also examine each minority group separately and determine that there are unique characteristics exhibited by each group.

INTRODUCTION

A financial institution is considered to be minority owned if 51% or more of the voting stock is owned by one or more minorities including African-Americans, Hispanic-Americans, Asian-Americans, Native-Americans, and women. One exception to this requirement occurs if the bank's board of directors is mainly minority run or if the bank is serving a community that is predominately made up of minorities. Since a large number of minority owned banks are located in low-income areas, they are often viewed as being the providers of economic stimulation to surrounding underdeveloped communities. The minority owned bank's willingness to make loans in small amounts, connect with the community, and provide other services targeted to minority or low-income groups gives them access to a banking niche that would otherwise not be served. Some critics believe that lending money to individuals and/or businesses in these poor communities is too risky since most minority owned banks do not have access to large amounts of assets (e.g. Brimmer (1971)). Yet, advocates have lobbied for the creation of government sponsored programs that assist minority owned banks through funding.

The first government program created to aid minority-owned banks was the Minority-Bank Deposit Program (MBDP). It was designed by the Nixon administration and put into effect in October of 1970. The MBDP gives minority owned banks preferences to deposits that are made with funds from the federal government. It also encourages corporations to make deposits into minority owned financial institutions. This program was followed up with the Community Reinvestment Act which was passed by Congress in 1977. It was created in an effort to encourage banks to contribute to the communities that they served, especially low-income areas. The most recent act created to assist the minority banking community is the Community Development Banking Act. It was signed into law in 1994 by President Clinton with the intent of encouraging the creation of a group of community development banks. Beyond the reach of these programs, the FDIC has the ability to help minority owned banks survive even if they are being outperformed by their competitors if the MOB is viewed as being essential to the community that it serves.

It has been debated whether or not these programs have met their goals and contributed to the communities in which they operate. According to the majority of studies, minority owned banks are consistently less efficient and profitable than non-minority owned banks. Some studies attribute the lower levels of efficiency and profitability of minority owned banks to the volatile nature of the large amounts of government deposits that they hold. Other studies have attributed these discrepancies to anything from location in low-income communities to less-qualified key employees. The varying conclusions that have been drawn by previous studies leave room for further exploration of this topic. This paper will not directly address these conflicting views but will rather document the evolution of minority owned community banks over the past decade from 1995 to 2004. We limit our sample to U.S. commercial banks with total assets between $100 million and $1 billion because this is where the majority of minority owned banks exist.

The paper proceeds as follows. Section 2 reviews the existing literature. Section 3 describes the data used in our study, discuss changes in minority ownership over the past decade, and tests for differences in banking characteristics between minority owned banks and non-minority owned banks. Section 4 presents ideas for future research and section 5 concludes the paper.

LITERATURE REVIEW

The majority of the minority banking literature is related to the existence of government programs which are designed to provide deposits to minority owned banks either directly by an allocation of government funds or indirectly by encouraging deposits from corporations. By increasing the deposit base, these government programs have the effect of increasing the availability of loans to borrowers in the communities served by minority owned banks. The presence of these businesses in low-income communities not only helps to foster economic development but provides an example of leadership to the minorities living in the community. Furthermore, the minority bank is seen as providing services to low-income communities that would not be provided if they ceased to exist. Therefore, the goal of the government sponsored programs is not just to strengthen minority businesses, but also to contribute to the economic development of low to moderate income communities. We will proceed by discussing the literature related to these government programs to provide a background to the existence and progression of minority owned banks. Then we discuss the literature related to differences between minority owned banks and non-minority owned banks.

The Minority Banking Deposit Program was created to fuel the minority banking industry. It was believed that it would simultaneously help the minority banking industry and the economic situations of the local communities that these banks served. While the MBDP was created in an effort to eliminate discrimination against minority owned banks, it has since had the effect of encouraging government agencies and corporations to discriminate against non-minority owned banks when choosing which financial institutions to place their deposits.

Critics of the MBDP such as Brimmer (1971) believe that the program has led to greater inefficiency among minority owned banks. Brimmer believes the types of lending activities that minority owned banks are involved in is restricted by the volatile nature of the government deposits. Boorman (1974) finds a significant difference between the amount of federal deposits at minority banks and non-minority banks. He finds that in 1971, federal deposits made up 10% of total deposits at minority owned banks compared to only about 2% at non-minority owned banks. More recently, Iqbal and Ramaswamy (1999) find federal deposits account for 4.6% of the total deposits of minority banks compared to only 0.5% in non-minority banks. These percentages only represent the amount of deposits that are directly given to the minority owned banks from government funding and do not represent the additional deposits that the banks receive from corporations as a result of government encouragement. It is believed that these percentages would be significantly higher if the deposits from private donors could be singled out. Yet, even from the government funding percentages, it is clear that minority banks continue to rely heavily on deposits from the federal government. Iqbal and Ramaswamy propose that this dependency on funding from the federal government creates deposit instability which is a recurring problem for minority owned banks.

Another way that the government has stepped in to aid minority-owned banks is through the passage of the Community Reinvestment Act (CRA) requires every bank to be evaluated on how well it has helped meet the credit needs of its local community. Each bank is evaluated by use of a CRA audit and is given one of four ratings: Outstanding, Satisfactory, Needs to Improve, or Substantial Non-Compliance. While this process of evaluation was originally intended to occur every 18 months, the passage of the Glass-Leach-Bliley law in 1999 altered the requirements. Since the law went into effect in 2000, the only banks that are required to be audited every 18 months are those banks that receive a rating of either Needs to Improve or Substantial Non-Compliance. The banks that receive an Outstanding rating are to be audited every 5-years and those receiving a Satisfactory rating are to be audited every 4-years. Critics of the new law believe that it will encourage neglect among the banks that are audited less frequently. Matasar and Pavelka (2004) examine the most recent CRA ratings available and indicated that no minority bank received a rating of Substantial Non-Compliance and that only one minority bank received a rating of Needs to Improve.

A bank's CRA rating is required by The Financial Institutions, Reform, Recovery and Enforcement Act of 1989 to be readily available to the public so that potential customers can monitor how much a bank is contributing to the community. While the CRA does not encourage or require banks to get involved in risky loans that might hurt their performance, minority owned banks often participate in these types of activities causing their operating costs to be significantly higher than non-minority owned banks. The scarcity of creditworthy customers in these low-income communities has greatly increased the competition between minority and non-minority banks for the business of the limited number of customers with good credit. Lawrence (1997) points out that this competition often leads the larger non-minority banks to offer services at lower prices than the minority banks can afford, and the result is that the non-minority banks capture the only creditworthy customers from the low-income communities. Therefore, the minority bank's high rejection rate of minority applicants may be attributable to the fact that they receive applications from the least credit-worthy individuals of the community.

The Community Development Banking Act (CDBA), signed into law by Clinton in 1994, created the Community Development Financial Institutions Fund which is aimed at financing specialized, community development financial institutions. It is an alternative to the Community Reinvestment Act and differs by providing subsidies directly out of general tax revenues. According to the CDBA, the government set aside $382 million from general tax revenue to contribute over a span of four years to finance community development banks that are located in and serve low-income communities. The CDBA requires that the banks match every dollar received from government funding with a dollar raised from private capital. Its intent is to promote economic and community development by contributing to the liquidity of community banking institutions. Advocates of the CDBA believe it is less costly when a tax is spread over the entire society as opposed to being levied on a single industry. Since the CRA imposes a tax on the banking industry alone, it is often viewed as inefficient (Lacker (1995)).

Next, we will discuss literature relating to comparisons between minority owned banks and non-minority owned banks. When studying the different characteristics between minority owned banks and non-minority owned banks, the two types of banks must be compared in a similar context. In the past there have been several methods that have been used in order to create a control group of non-minority banks to compare with minority banks. One method of creating these control groups is through the use of zip code clusters. Clair (1988) was the first to use this methodology to compare banks that were located in the same zip code or in a zip code adjacent to the bank's main office. Those in favor of using zip code clusters believe that it is more appropriate to use this method, because it matches banks within a narrower geographical region. The narrower the region being used, the more likely it is that both the non-minority and minority banks exist in similar conditions and face comparable market trends. Clair (1983) and Lawrence (1997) find in their studies that when a narrower geographical region is used there are fewer significant differences between the performances of minority and non-minority owned banks.

Previous studies used zip-code clusters to compare the efficiency of minority owned banks to that of non-minority owned banks. One approach to comparing the efficiency of these two types of banks is through the use of a frontier model. Elyasiani and Mehdian (1992) judge efficiency under the frontier model by determining how the bank managers could maximize revenue or minimize cost. Specifically, they focus on the levels of costs incurred by the minority owned banks versus the non-minority owned banks. In their study they control for regional, regulatory, size, and maturity differences, and find that there was not a significant statistical difference between the efficiency of minority-owned banks and non-minority owned banks. Hasan and Hunter (1996) follow a similar approach in their study of the production efficiency of minority-owned banks versus non-minority owned banks. They compare minority owned banks and non-minority owned banks to a group of best-practice banks. They define best-practice banks as any bank, regardless of ownership, that is able to produce its financial services at the lowest cost while also allocating any inputs or factors of production efficiently. Their results indicate that the minority and women owned banks operate at a lower level of efficiency than the average non-minority bank. This result is partly based on the fact that minority and women owned banks experienced significantly higher operating costs. Lawrence (1997) points out that under information theory, financial institutions may discriminate in an effort to lower costs. If they are prohibited from discriminating they may be faced with higher costs and therefore be viewed as operating inefficiently.

In an early study by Boorman (1974), he finds when comparing costs of non-minority banks to minority banks, the minority banks had higher employee expenses, lower interest costs, and significantly larger loan losses than non-minority banks. Boorman also finds that while minority owned banks and non-minority owned banks have similar levels of efficiency, minority owned banks do not have much potential to stimulate the economies of the communities they are located in and that minority banks are less than half as profitable as non-minority banks. His study determines that this large difference could be attributed to the large loan losses experienced by minority banks. Brimmer (1971), Gardner (1984), and Boorman and Kwast (1974) also indicate in their research that minority banks are less profitable than non-minority banks.

In a recent study by Iqbal and Ramaswamy (1999), the frontier model is used to determine output efficiency by measuring technical and allocative efficiency, which measure whether banks are producing the maximum amount of possible outputs and using the correct proportions of those outputs. Their results indicate that minority owned banks are not acting efficiently in the technical or allocative areas.

DATA, VARIABLE SPECIFICATION, AND METHODOLOGY

We gather our community bank data from the Annual Reports of Condition and Income (Call Report) required by the Federal Deposit Insurance Corporation. We use the definition of community banks as defined by Carter and Sinkey (1998). We examine US commercial banks with total assets between $100 million and $1 billion. There are 4,397 banks and/or bank holding companies in this category between 1995 and 2004. We choose the dates for our sample to cover the most recent 10 year period. The Call Reports classify banks as being owned by either a Caucasian man, black, woman, Hispanic-American, Asian-American, Native American, Eskimo, Aleut, Arab-American, or Multiracial. In our sample we find at least one bank owned by each minority group except for the Eskimo, Aleut and Arab-American.

This paper examines the changes in minority ownership at the community bank level over the past ten years. Chart 1 shows the growth in minority banks as a percentage of all banks. Here you can see that in 1995, .93% (26) of community banks were owned by minorities. That percentage grew to 1.78% (62) by 2004. Although these are very low percentages, the number has more than doubled in a decade. Chart 2 examines which minority group of minorities has experienced the most growth from 1995-2004. Here you can see that the Asian (11 to 27) and Hispanic (5 to 13) minority groups have seen the largest growth. The number of women owned banks has increased from zero existing in 1995, to three existing in 2004. It is interesting to note that the African American group has remained fairly stable in their ownership of community banks, ranging from 10 to 13.

[GRAPHIC 1 OMITTED]

[GRAPHIC 2 OMITTED]

In addition to simply examining growth among each minority group, it is interesting to examine the change in the characteristics of these banks. We start by examining changes in size. We divide the entire sample of community banks into quartiles based on total assets. Chart 3 shows the number of minority owned community banks in each quartile. You can see from this chart that quartile 4, contains the largest community banks, and has seen an increase in the number of MOBs. In 1995, only 3 minority owned banks existed in quartile 4. By 2004, 21 MOBs were included in quartile 4. We also observe a less significant increase for quartile 3 from 8 banks to 13 banks. This chart indicates that not only are minority owned banks increasing in number but they are also increasing in size which should increase their ability to compete in their local markets.

[GRAPHIC 3 OMITTED]

We also compute summary statistics on the minority owned and non-minority owned banks each year. The variables that we calculate and make comparisons between are defined as follows:

TA = Total Assets

EC = Equity Capital divided by Total Assets

PS = Dollar amount of Preferred Stock divided by Total Assets

DIV = Dividends divided by Total Assets

NIM = Net Interest Income divided by Total Assets

NLCO = Net Loan Charge-Offs divided by Total Assets

BHC = Bank Holding Company

ROA = Net Income divided by Total Assets

ROE = Net Income divided by Total Equity

Cash/TA = Cash divided by Total Assets

Loans/Deposits = Total Loans divided by Total Deposits

CommL/TL = Commercial Loans divided by Total Loans

USgovtdep/TD = US government deposits divided by Total Deposits

Provforloan/TOE = Provisions for loan losses divided by Total Operating Expenses

Nonfedsec/TS = Non-federal Securities divided by Total Securities

TA/#Emp = Total Assets divided by Number of Employees

Total assets is used to represent firm size. Equity capital represents a firm's financial stability as viewed by regulatory agencies. Asset liquidity represents the bank's on-balance sheet assets. Preferred stock is included to determine if the source of financing differs by bank ownership type. Dividends are included because the amount that a firm pays in dividends can indicate whether the firm has sufficient funds to enable them to pay out surplus cash to investors and be viewed positively by creditors. Net interest margin is a measure of the banks spread between interest income and interest expense. Net loan charge-offs is a proxy for credit risk. The bank holding company dummy variable indicates whether or not the bank is part of a holding company (1, it is part of a BHC and 0, it is not). ROA and ROE are measures of the bank's return on assets and equity and are used as indications of performance. Cash to total assets is a measure of liquidity. Loans to deposits and commercial loans to total loans give us information on the banks' portfolio composition. United States government deposits to total deposits measures volatility of the deposit base (Bates and Bradford (1980)). Provision for loan losses is the expense a bank incurs due to expected loan losses and it directly reduces net income and ROA. Non-federal securities to total securities measures the portion of investments made in non-federal securities such as mortgage backed securities, asset-backed securities, and foreign debt securities. Total assets divided by number of employees indicates the amount of assets managed per employee and is a measure of operational efficiency.

PRESENTATION OF SUMMARY STATISTICS

The final aspect of this study is to compare the characteristics of MOB to NMOB. These statistics appear in Table 1. (Tables 1, 2 and 3 are at the end of the article). The most notable difference is that the size of NMOB is much larger than the MOB. This is not surprising and has been cited by other researchers. However, one interesting observation is that the size gap narrows as the years progress. For example, in 1995 the difference in average total assets for MOB compare to NMOB was almost $54 million. By 2004, the difference had shrunk to less than $10 million. The ratio of equity to total assets is lower for MOB indicating they have less equity capital per asset than NMOB. Additionally, MOB have higher ratios of preferred stock and lower dividend ratios. These two results indicate that NMOB are more financially secure than MOB and therefore MOB may be viewed as riskier investments.

MOB have a larger net interest margin, which can be viewed in two different ways. It is favorable in that it indicates higher profitability for MOB. However, the higher the NIM is it is more likely that riskier loans are being made which would result in larger interest revenue but could lead to larger loan losses. Not surprisingly, then, we find net loan charge-offs at MOB are higher than at NMOB. Both profitability measures, ROA and ROE are smaller for MOB. In addition to the credit concerns discussed earlier, the lower profitability of MOB can also be traced to the ratio of cash to total assets, which is larger at MOB. This indicates that MOB keep larger sums of cash on hand and are more liquid, however, the downside of cash is it earns a very low return. Also contributing to lower profitability is the fact that MOB have lower loan to deposit ratios. Since loans are where community banks make most of their money, this is clearly a problem for the MOB.

As expected, US government deposits are larger for MOB and is of concern because of the volatile nature of government deposits. Provisions for loan losses to total operating expenses are also larger for MOB, which is indicative of the riskier loans made by MOB. MOB invest in a larger proportion of federal securities than NMOB. While these securities are less risky, they also have lower returns which hurts profitability. Finally, total assets to number of employees is smaller for MOB indicating they have fewer assets managed per employee, which could be due to their higher cost deposits.

Now that we have described these differences, we must test for statistical significance. Since the data set is broken into two subsets (MOB and NMOB), the Cochran and/or Satterthwaite t-test must be used because the subsets have unequal variances. As shown in table 2 all the variables are significantly different at the 10% level or less. In fact, all but 3 of the variables were significantly different at the 1% level (provisions for loan losses to total operating expenses, US government deposits to total deposits, and net loan charge-offs).

The next question we address is whether the different minority groups have different bank characteristics. Table 3 shows these results. The minority group with the largest number of banks in the sample is Asian, followed by Black and Hispanic. There are only 2 multiracial banks in our sample so we will not use them for this comparison. The NMOB have the most average total assets followed by Asian, Hispanic, Black, Women, Native American and Multiracial. Preferred stock is the largest for the Black and Hispanic groups while the other minority groups do not differ much from the NMOB. Very few Asian and Native American owned banks are part of a bank holding company while all of the women and multiracial banks are part of a BHC. Return on assets and equity is the highest for the women owned banks and lowest for the black and multiracial owned banks. US government deposits to total deposits is extremely high for the black owned banks and very low for the Asian owned banks. Provision for loan losses to total operating expenses is similar for all groups. Assets per employee is highest for the Asian and Hispanic owned banks and lowest for the Black owned banks. These results show us that the type of minority owned bank is a significant indicator of its efficiency and performance. This also indicates that future research endeavors should seriously consider differentiating beyond Minority ownership into race/gender. Most research to date (one exception is Lawrence (1997)) groups all minority owned banks together to compare against non-minority owned banks. These results indicate that this may not make for the best comparisons.

FUTURE RESEARCH

Previous research has been done to compare the efficiency of minority owned banks to that of non-minority owned banks. Yet, these studies have failed to specifically look at how government funding contributes to the efficiency or inefficiency of minority owned banks. Also, most of the studies have failed to specify how the minority groups differ. Therefore, further research needs to be done that compares the efficiency and profitability of each type of minority-owned bank with government funding, minority-owned banks without government funding, and non-minority owned banks without government funding.

We hypothesize that the financial institutions with government funding will be less efficient than financial institutions without government funding. Yet, since the minority owned banks are often grouped collectively in a heterogeneous manner, the results can be hard to interpret. In order to ensure more reasonable results, Clair's (1988) methodology of zip-code clusters should be used to compare banks that are located in the same zip code or in a zip code adjacent to the bank's main office. Minority banks should be grouped together not only based on location, but also based on ethnicity, size, and age. Boorman (1974) originally pointed out that it was important to analyze banks based on their relative age through time series analysis in order to distinguish whether minority-owned banks were becoming more efficient with time. The results from his study indicated that minority owned banks were increasing their levels of efficiency with time even though they were still lagging behind the non-minority owned banks. It is important to use this type of analysis again to determine whether this improvement has continued or slowed in the passing years.

CONCLUSION

The number (26 to 62) and percentage (.93% to 1.78%) of minority-owned banks has increased in the past decade, 1995-2004. This paper shows the growth in minority ownership by various racial groups. The Asian and Hispanic groups experienced the most growth while the Black group experienced the least. We segment our sample of community banks over the past decade into minority owned banks (MOB) and non-minority owned banks (NMOB). We show the averages of various bank characteristics for each group for 1995 through 2004. Then we test the significance of the differences in these characteristics and found that every characteristic was significantly different for MOB versus NMOB. Finally, we look at each minority group separately to determine if there were unique characteristics exhibited by each group. We find that each minority groups differs in most of the bank characteristics.

While this paper does not address whether government funding of minority owned banks should continue to exist, it documents the changing environment found among minority owned banks. It is left for future research to examine and determine if government funding, which encourages minority ownership, is supporting continued inefficiencies in MOB or whether the growth documented here in minority ownership indicates a change for the better in the future profitability and efficiencies of MOB.

REFERENCES

Boorman, J.T. (1974). The prospects for minority-owned commercial banks: A comparative performance analysis, Journal of Bank Research, 5, 263-279.

Brimmer, A.F. (1971). The Black Banks: An Assessment of Performance and Prospects. Journal of Finance, 26, 379-406.

Carter, D. & J. Sinkey. (1998).. The use of interest rate derivatives by end-users: The case of large community banks. Journal of Financial Services Research, 14(1), 17-34.

Clair, R. T. (1988). The Performance of Black-Owned Banks in Their Primary Market Areas. Federal Reserve Bank of Dallas Economic Review: 11-20.

Elyasiani, E. & S.Mehdian. (1992). Productive Efficiency Performance of Minority and Nonminority-Owned Banks: A Nonparametric Approach. Journal of Banking and Finance, 16: 933-948.

Hasan, I. & W. C. Hunter. (1996). Management Efficiency in Minority-and Women-Owned Banks. Economic Perspectives, 20.2: 20-29.

Iqbal, Z., K. V. Ramaswamy & A. Akhigbe. (1999). The Output Efficiency of Minority-Owned Banks in the United States. International Review of Economics and Finance, 8: 105-14.

Lacker, J. M. (1995). Neighborhoods and Banking. Federal Reserve Bank of Richmond Economic Quarterly 81: 13-38.

Lawrence, E. C. (1997). The Viability of Minority-Owned Banks. The Quarterly Review of Economics and Finance, 37: 1-21.

Matasar, A. B.& D. D. Pavelka. (2004).."Minority Banks and Minority Communities: Are Minority Banks Good Neighbors? International Advances in Economic Research, 10: 43-58.

Jennifer Echols Edmonds, Birmingham-Southern College

Sara Helms Robicheaux, Birmingham-Southern College

Table 1: Summary Statistics

We compute summary statistics on the minority owned and non-minority
owned banks each year. The variables that we calculate and make
comparisons between are defined as follows: TA = Total Assets,
EC = Equity Capital divided by Total Assets, PS = Dollar amount
of Preferred Stock divided by Total Assets, DIV = Dividends divided
by Total Assets, NIM = Net Interest Income divided by Total Assets,
NLCO = Net Loan Charge-Offs divided by Total Assets, BHC = Bank
Holding Company, ROA = Net Income divided by Total Assets, ROE =
Net Income Divided by Total Equity, Cash/TA = Cash divided by Total
Assets, Loans/Deposits = Total Loans divided by Total Deposits,
CommL/TL = Commercial Loans divided by Total Loans, USgovtdep/TD =
US government deposits divided by Total Deposits, Provforloan/TOE =
Provisions for loan losses divided by Total Operating Expenses,
Nonfedsec/TS = Non- federal Securities divided by Total Securities,
TA/#Emp = Total Assets divided by number of Employees

 Year 1995 1996

 Group MOBs NMOBs MOBs NMOBs

Sample Size 26 2765 32 2825
TA (in thousands) 186920 240773 200548 241080
EC 0.0840 0.0954 0.0840 0.0953
PS 0.0012 0.0003 0.0021 0.0003
DIV 0.0019 0.0067 0.0017 0.0064
NIM 0.0419 0.0415 0.0416 0.0413
NLCO 0.0049 0.0025 0.0037 0.0030
BHC 0.5000 0.8448 0.5313 0.8499
ROA 0.0056 0.0120 0.0075 0.0122
ROE 0.0452 0.1288 0.0700 0.1313
CashTA 0.0542 0.0504 0.0605 0.0496
Loansdeposits 0.5701 1.9407 0.6064 3.6027
commLtotalL 0.2621 0.1748 0.2733 0.1758
USgovtdepTdep 0.0596 0.0375 0.0374 0.0361
ProvforloanTopexp 0.0498 0.0304 0.0379 0.0365
nonfedTsec 0.4397 0.6509 0.4620 0.6467
TAnumemp 2161.89 2559.81 2265.76 2614.96

 Year 1997 1998

 Group MOBs NMOBs MOBs NMOBs

Sample Size 35 2825 37 2874
TA (in thousands) 204488 246388 196674 241981
EC 0.0920 0.0969 0.0838 0.0960
PS 0.0020 0.0003 0.0054 0.0002
DIV 0.0010 0.0066 0.0013 0.0067
NIM 0.0424 0.0411 0.0412 0.0398
NLCO 0.0032 0.0026 0.0029 0.0026
BHC 0.5714 0.8545 0.5946 0.8636
ROA 0.0079 0.0124 0.0072 0.0121
ROE 0.0927 0.1234 0.0850 0.1288
CashTA 0.0506 0.0482 0.0470 0.0474
Loansdeposits 0.6629 2.6016 0.6344 1.5476
commLtotalL 0.2321 0.1735 0.2344 0.1749
USgovtdepTdep 0.0290 0.0364 0.0641 0.0370
ProvforloanTopexp 0.0383 0.0368 0.0445 0.0381
nonfedTsec 0.4623 0.6400 0.6096 0.6931
TAnumemp 2431.54 2712.81 2506.67 2758.56

 Year 1999 2000

 Group MOBs NMOBs MOBs NMOBs

Sample Size 39 2930 36 2980
TA (in thousands) 195170 247075 205682 248998
EC 0.0775 0.0935 0.0805 0.0970
PS 0.0050 0.0002 0.0045 0.0002
DIV 0.0020 0.0069 0.0039 0.0068
NIM 0.0415 0.0400 0.0433 0.0392
NLCO 0.0033 0.0024 0.0033 0.0023
BHC 0.6154 0.8696 0.6111 0.8701
ROA 0.0069 0.0123 0.0093 0.0118
ROE 0.0895 0.1381 0.1179 0.1262
CashTA 0.0535 0.0475 0.0447 0.0445
Loansdeposits 0.6514 1.9176 0.6869 1.4039
commLtotalL 0.2259 0.1774 0.2180 0.1761
USgovtdepTdep 0.0620 0.0356 0.0664 0.0346
ProvforloanTopexp 0.0479 0.0373 0.0369 0.0374
nonfedTsec 0.6183 0.7004 0.5938 0.6937
TAnumemp 2470.80 2861.96 2724.84 3103.64

 Year 2001 2002

 Group MOBs NMOBs MOBs NMOBs

Sample Size 39 3092 44 3217
TA (in thousands) 211405 255312 243038 260947
EC 0.0865 0.0973 0.0885 0.1001
PS 0.0034 0.0002 0.0026 0.0002
DIV 0.0029 0.0061 0.0030 0.0060
NIM 0.0392 0.0374 0.0389 0.0380
NLCO 0.0029 0.0026 0.0024 0.0028
BHC 0.6667 0.8768 0.6818 0.8788
ROA 0.0084 0.0109 0.0088 0.0115
ROE 0.1036 0.1158 0.1044 0.1208
CashTA 0.0570 0.0471 0.0522 0.0478
loansdeposits 0.6994 0.9942 0.7321 0.9945
commLtotalL 0.2113 0.1710 0.1744 0.1633
USgovtdepTdep 0.0662 0.0377 0.0563 0.0420
ProvforloanTopexp 0.0479 0.0466 0.0648 0.0595
nonfedTsec 0.6960 0.7430 0.6585 0.7603
TAnumemp 2755.36 3245.59 3161.88 3508.8

 Year 2003 2004

 Group MOBs NMOBs MOBs NMOBs

Sample Size 53 3324 62 3415
TA (in thousands) 258671 263576 258448 268431
EC 0.0946 0.0994 0.0948 0.1003
PS 0.0022 0.0002 0.0017 0.0002
DIV 0.0031 0.0065 0.0046 0.0058
NIM 0.0380 0.0366 0.0390 0.0367
NLCO 0.0023 0.0028 0.0017 0.0022
BHC 0.6604 0.8842 0.6935 0.8837
ROA 0.0107 0.0115 0.0108 0.0116
ROE 0.1157 0.1197 0.1162 0.1193
CashTA 0.0609 0.0451 0.0467 0.0387
loansdeposits 0.7736 1.0670 0.8069 1.0036
commLtotalL 0.1639 0.1601 0.1481 0.1553
USgovtdepTdep 0.0531 0.0452 0.0418 0.0433
ProvforloanTopexp 0.0585 0.0574 0.0461 0.0490
nonfedTsec 0.6989 0.7716 0.6618 0.7715
TAnumemp 3171.75 3605.89 3416.83 3788.18

Table 2: MOBs versus NMOBs T-test Statistic

Since the data set is broken into two subsets (MOBs and NMOBs),
the Cochran and/or Satterthwaite t-test must be used because the
subsets have unequal variances. These tests determine if these
differences found in Table 1 were significant. The variables are
defined as: TA = Total Assets, EC = Equity Capital divided by
Total Assets, PS = Dollar amount of Preferred Stock divided by
Total Assets, DIV = Dividends divided by Total Assets, NIM = Net
Interest Income divided by Total Assets, NLCO = Net Loan Charge-
Offs divided by Total Assets, BHC = Bank Holding Company, ROA =
Net Income divided by Total Assets, ROE = Net Income Divided by
Total Equity, Cash/TA = Cash divided by Total Assets, Loans/
Deposits = Total Loans divided by Total Deposits, CommL/TL =
Commercial Loans divided by Total Loans, USgovtdep/TD = US
government deposits divided by Total Deposits, Provforloan/TOE =
Provisions for loan losses divided by Total Operating Expenses,
Nonfedsec/TS = Non-federal Securities divided by Total Securities,
TA/#Emp = Total Assets divided by number of Employees

 Variable Name Group Mean T-Value P-Value

TA MOB 221835 -4.29 <.0001
 NMOB 252122
 -30287
EC MOB 0.0875 -7.23 <.0001
 NMOB 0.0972
 -0.01
PS MOB 0.0029 5.34 <.0001
 NMOB 0.0002
 0.0027
DIV MOB 0.0027 -14.26 <.0001
 NMOB 0.0064
 -0.004
NIM MOB 0.0404 3.37 0.0008
 NMOB 0.039
 0.0014
NLCO MOB 0.0029 1.74 0.083
 NMOB 0.0026
 0.0003
BHC MOB 0.6253 -10.04 <.0001
 NMOB 0.8685
 -0.243
ROA MOB 0.0086 -8.46 <.0001
 NMOB 0.0118
 -0.003
ROE MOB 0.098 -5 <.0001
 NMOB 0.1249
 -0.027
CashTA MOB 0.0526 3.41 0.0007
 NMOB 0.0464
 0.0061
Loansdeposits MOB 0.6986 -6.31 <.0001
 NMOB 1.666
 -0.967
CommLtotalL MOB 0.2054 5.2 <.0001
 NMOB 0.1697
 0.0357
USgovtdepTdep MOB 0.0531 2.22 0.0268
 NMOB 0.0388
 0.0144
ProvforloanTopexp MOB 0.0481 1.76 0.0784
 NMOB 0.0435
 0.0046
NonFedTsec MOB 0.5865 -4.38 <.0001
 NMOB 0.7028
 -0.116
TAnumemp MOB 2797.9 -3.95 <.0001
 NMOB 3105.7
 -307.8

Table 3: Bank Characteristics by Minority Group

This table addresses whether the different minority groups have
different bank characteristics. The variables are defined as:
TA = Total Assets, EC = Equity Capital divided by Total Assets,
PS = Dollar amount of Preferred Stock divided by Total Assets,
DIV = Dividends divided by Total Assets, NIM = Net Interest Income
divided by Total Assets, NLCO = Net Loan Charge-Offs divided by
Total Assets, BHC = Bank Holding Company, ROA = Net Income divided
by Total Assets, ROE = Net Income Divided by Total Equity,
Cash/TA = Cash divided by Total Assets, Loans/Deposits = Total
Loans divided by Total Deposits, CommL/TL = Commercial Loans
divided by Total Loans, USgovtdep/TD = US government deposits
divided by Total Deposits, Provforloan/TOE = Provisions for loan
losses divided by Total Operating Expenses, Nonfedsec/TS = Non-
federal Securities divided by Total Securities, TA/#Emp = Total
Assets divided by number of Employees

Race Non-Minority Black Women Hispanic

Sample Size 30247 121 17 81
TA (in thousands) 252122 193472 165598 240170
EC 0.0972 0.0855 0.1050 0.0863
PS 0.0002 0.0072 0.0004 0.0032
DIV 0.0064 0.0026 0.0046 0.0029
NIM 0.0390 0.0426 0.0449 0.0375
NLCO 0.0026 0.0035 0.0017 0.0019
BHC 0.8685 0.8017 1.0000 0.6790
ROA 0.0118 0.0065 0.0137 0.0102
ROE 0.1249 0.0759 0.1298 0.1217
CashTA 0.0464 0.0567 0.0640 0.0565
loansdeposits 1.6660 0.6080 0.8067 0.6466
commLtotalL 0.1697 0.2100 0.2032 0.2046
USgovtdepTdep 0.0388 0.1248 0.0258 0.0533
ProvforloanTopexp 0.0435 0.0407 0.0469 0.0433
nonfedTsec 0.7028 0.6431 0.1182 0.8134
TAnumemp 3106 1988 2849 3174

Race Asian Native American Multiracial

Sample Size 167 15 2
TA (in thousands) 248582 133349 103492
EC 0.0874 0.0915 0.0929
PS 0.0003 0.0000 0.0000
DIV 0.0026 0.0017 0.0056
NIM 0.0397 0.0394 0.0460
NLCO 0.0032 0.0012 0.0049
BHC 0.4790 0.0667 1.0000
ROA 0.0088 0.0105 0.0066
ROE 0.098 0.1170 0.0709
CashTA 0.0449 0.0729 0.0397
loansdeposits 0.7788 0.6728 0.8738
commLtotalL 0.2075 0.1607 0.1458
USgovtdepTdep 0.0008 0.0939 0.0045
ProvforloanTopexp 0.0558 0.0419 0.1009
nonfedTsec 0.4180 1.0000 --
TAnumemp 3258 2173 2412