Printer Friendly

Community banks and the importance of lending.

Community Banks and the Importance of Lending

This article examines the performing characteristics of high performing banks in a unique manner. High performance banks have been studied by the American Bankers Association Research and Planning Group [5,6,7] and by independent researchers [3,4,8,9,10,13], as well as the several Federal Reserve Banks, which publish profit statistics for banks in their districts. These studies, with some exceptions, examine the data on a year-to-year basis and report their analyses of the annual changes [3,4,8].

This study approaches the problem of identifying high performance characteristics in a different way. Average data for a five year period was used instead of the year-to-year data for previous studies. This approach has a smoothing effect on bank performance and concentrates on the characteristics that identify high performance in the long run. Additionally, this study concentrates on those banking variables related to lending and uses statistical techniques to identify those lending characteristics which are statistically significant in the determination of high performance.

The first step is to define high performance banking. In order to make the study manageable, the bank sample was limited to 251 Tennessee community banks. Tennessee was chosen because of its diversified economy and its diversified banking community of independent and holding company operations. In a comparison of bank performance in the region with the performance of other Tennessee banks, it can be shown that banking results in Tennessee closely resemble overall results for the Southeast U.S. [2,3,12]. In order to remove potential statistical bias, large banks with over $1 billion in assets and new banks less than five years old were eliminated from the study group.

The Federal Reserve Board's reports of condition and income computer tapes were used as the financial database for this study. The data were averaged for the five year period from 1983 through 1987. The banks in the study were not publicly traded. Measures of stock price performance and risk-adjusted rates of return could not be assessed to determine if the market agrees with the findings of this study.

As a first screen, only banks with a five year average return on assets (ROA) greater than the five year average ROA of all banks in the study qualified. The five year average ROA for the sample banks was 0.885. The second screen used was the loan to asset ratio. Only banks with an ROA higher than the five year average ROA for the sample and a five year average loan-to-asset ratio higher than the five year average loan-to-asset ratio (49.39 percent) for all of the banks were designated as high performers. The dual screen divided the banks into two groups: 65 high performers and 186 other banks (referred to as All Banks in the study). The banks were further subdivided into three asset size groups: less than $50 million, between $50 million and $100 million, and greater than $100 million.

The study used a univariate test of means to measure the statistical differences in performance between the two groups of banks. The means of 54 banking variables were tested for significant differences. Ten of these proved to be statistically significant in evaluating the importance of the credit function in high performing banks.

In order for a ratio to be designated "very significant," a dual test was imposed. The ratio mean for the high performing group was compared to the ratio mean for the other banks. If the mean was significantly "different" at the 99 percent level of confidence for the All Banks group, then the mean was further compared for each of the size groups. If the ratio mean was statistically different in all three size groups, it was found to be very significant and designated by an a superscript, as illustrated in Table 1. If the ratio mean failed the significance test in one of the three size categories, it was held only to be "significant" and designated by a b superscript. If the ratio mean failed the significance test in more than one of the size groups, it was not considered. The 15 significant ratios in Table 1 and Table 2 highlight the strengths of high performing banks. The values of very significant ratios and their significance levels are shown in Table 1. [Tabular Data Omitted]

Very Significant Ratios

Average Loans/Average Assets. This was a fundamental ratio since it was one of the two screens used. A bank that was a high earner and simultaneously a high lender was truly a high performance bank. High performers loaned 57.8 percent of total assets as compared to 46.4 percent for the other banks. This was a very large absolute difference and demonstrated clearly a difference in the lending policies between the two groups. The difference held for All Banks and for each of the size groups. The significance declined only slightly for the largest size group.

Net Charge-Offs/Average Loans. The key to profitable banking is being able to lend money and to hold losses down. Usually, as loan volume increases, loan losses rise proportionally. The remarkable high performing banks were able, on average, to reverse this relationship. Their higher loan percentage was coupled with a lower loan charge-off. High performers charged off 0.8 percent as compared to the other banks which charged off 1.1 percent. The most significant difference in the size groups was with the largest size. That group charged off only about one-half the amount as their other counterparts. The smaller size groups charged off fewer loans than the larger high performers and proportionally fewer than other banks in their group.

It is not possible to over emphasize the importance of controlling loan losses. A recent article by David Cates [2] pinpointed the importance of this ratio. Assume a $100 million size bank that is 60 percent loaned up. What is the impact of a 0.8 percent actual loan charge-off? Cates assumed that loan growth would be 20 percent and that the loan loss provision should cover both the anticipated loan growth and the existing charge-offs. Assuming a net interest margin of five percent on total assets and a loan/asset ratio of 60 percent ($100,000,000 x 0.6 x 0.008 x 1.20 = $576,000; $576,000/$5,000,000 = 11.5 percent), then high performers needed only 11.5 percent of net interest margin to cover the loan loss provision as compared to 15 percent needed by the typical bank according to Cates. Relatively speaking, the high performers in this study did very well, according to the Cates equation.

Average Past Due Loans/Average Net Loans. This is the "bad loans" ratio. The numerator includes all past due and non-accruing loans reported by the bank. In the All Banks group, high performers had 41 percent fewer bad loans than other banks. High performers had a bad loan ratio of 0.70; others had a ratio of 1.18. The percent difference is: (0.70-1.18)/1.18 = 0.41 = 41 percent. The largest size group had the least significant difference. That group had 0.6 percent of their loans past due, 33.7 percent fewer than their lower performing counterparts.

Average Real Estate Loans/Average Assets. All of the components of the loan portfolio were tested for significance. Only the proportion of real estate loans proved to have a very significantly different mean. High performers had 23.4 percent of total assets in real estate loans as compared to 18.6 percent for the other banks. Real estate lending has great potential for being very risky. The fact that these high performing banks were able to emphasize real estate lending and, at the same time, hold loan losses down was a very considerable accomplishment. The significance of this ratio was very high for each group of banks. Not only was the volume significantly higher for the high performers, apparently the risk control referred to above was superior.

Average Liquid Assets/Average Assets. Liquidity is one of the first of a series of anti-lending ratios - anti-lending in the sense that it reflects an alternative use of liabilities that would otherwise be available to support the lending function. Liquidity was defined as the sum of cash due from banks plus U.S. securities and U.S. Government agency securities, net Federal Funds, and security repurchases. High performers maintained a lower liquidity position; 1.5 percent of their assets were "liquid" compared to 19.2 percent of the assets for the other banks. High performers were aggressive lenders and were able to manage their liquidity so well that they used fewer non-interest or low interest assets.

It was not surprising to find a lower volume of liquid assets in the high performing banks. Liquid assets earn relatively less than the less liquid loan portfolio. The other banks demonstrated a lesser ability to control credit risk in the loan portfolio. It seems natural that their volume of the more liquid and less risky assets included in this category would be higher. Lower liquidity undoubtedly annoys examiners but, all other things being equal, it does show better funds management.

Average Interest Income/Average Net Loans. This is an interesting measurement. High performers earned a relatively lower level of interest income than the other banks. The asset composition shows high performers have more loans and fewer liquid assets than their lower performing counterparts. Apparently, the liquid assets generate a greater percentage of their interest income for the low performers than the more risky loans. Pricing strategy may also be a factor. High performing banks may price their loans more conservatively than the other banks. A more aggressive (less conservative) pricing strategy may have been used by the lower performing banks. The ability of the high performers to better control risk was reflected earlier in this study. Although the other banks had significantly more interest income relative to loans, other factors prevented this supposed advantage from being carried through to the bottom line.

Net Interest Margin/Average Earning Assets. This is the basic operating ratio in banking. It is the difference between interest income and interest expense divided by average earning assets. All bank profitability follows from this relationship. In this case, high performers who were high lenders earned a better net interest margin (NIM) than the other banks. The high performing group had a ratio of six percent, ten percent higher than the 5.5 percent NIM for the other banks. Thus, banks that emphasized loans (higher loans to assets ratios) were able to secure a better margin through loans than the banks which emphasized investments. This was certainly not a surprise since the gross yield on loans should be higher. It was what the banks did with NIM which was the key factor. The evidence that the high performing banks had superior credit performance, as shown in the net charge-off ratio, demonstrated a very efficient use of the NIM.

Operating Income/Average Assets. After looking at the significant differences between the high performance banks and the other banks which showed lower net charge-offs, higher loan to asset ratios, and stronger net interest margins on earning assets, one would expect the high performers to have earned a higher level of operating income on their assets. The high performers earned 11.1 percent and the other banks earned 10.4 percent. This very significant difference reinforces the implication that high performance banks maintained strong and consistent management of their loan portfolio during 1983 through 1987.

Average Loan and Lease Income/Average Operating Income. This measurement holds the pieces together. High performers earned 20 percent more of the total income from lending than did the other banks. For reasons already stated, higher volume of loans, lower volume of liquid assets, higher earning assets, and better credit/risk management show the relative strength of lending in the determination of high performing banks. The strength of the difference was weakened somewhat as bank size increased, but the ratio retained significance for all size groups.

Return on Assets. The aggressive lending (lending a high percentage of assets and a high percentage of deposits) coupled with good credit management contributed strongly to the significantly higher ROA of the high performance banks, 1.2 percent compared to the 0.8 percent of the other banks (a 48 percent difference). The highest ROA was earned by the $50 to $100 million banks. With a five year average of 1.2 percent, they out performed the total sample of high performers by 3.4 percent.

Significant Ratios

Out of any pool of loanable funds, a bank has a choice of lending money or investing the funds. Lending has a number of advantages. The act of lending usually increases the economic strength of the community, promoting the growth of the community and the bank. Lending typically creates deposits at the originating institution because the borrower usually receives a demand deposit account from the lender.

Unfortunately, lending has adverse affects as well. Lending is labor intensive and tends to raise overhead costs. It requires an elaborate staff, floor space, and equipment. Lending is, above all, risky. High lending exposes banks to credit losses and related higher expenses.

Investing, on the other hand, is not labor intensive, and it requires no floor space. One person with a telephone can manage millions of dollars of investments in U.S. securities and state and local government securities. The cost requirements are absolutely minimal. The investment function is concerned with the liquidity and profitability of the bank. It is not directly involved with the people in the community as the lending function is. In that respect, it can be said that the investment function, when compared to the lending function, does relatively little for the community or for the banks' deposit growth.

The following ratios were very significant for the total population of banks and significant for at least two of the three size categories. In all cases, significance levels dropped in the third asset size category, greater than $100 million. The drop in significance may have been related to the small number of banks in the larger than $100 million group: 20 high performers and 25 other banks. Table 2 shows the values of the significant variables and their level of significance.

Average U.S. Government Securities/Average Assets. High performing banks did not use their funds to buy a large portfolio of investment securities. The ratio was closely related to the liquidity position discussed above. In this sense, the high performers were willing to trade extra return for extra risk associated with a larger loan portfolio. This translated into a higher return on equity.

The high performers held 17 percent in U.S. securities relative to assets. This compared to 24.8 percent for other banks. Large high performing banks and small high performing banks held equal relative volumes of U.S. Government securities, 17 percent of assets.

Average Equity/Average Risk Assets. The ratio subtracts from total assets all cash and related items plus direct obligations of the United States. This leaves a balance of assets which have some credit risk. High performers had an average equity to risk asset ratio of 11.5 percent compared to 14 percent for the other banks. Other things being equal, this absolute difference translated into an increase of 21.5 percent in return on equity (ROE) assuming an average ROA of 0.885. This was a strong indication of better risk management by the high performing banks, a direct result of the better credit policies.

Allowance for Loan Losses/Average Loans. With the exception of the large banks in the study, the high performance banks had a lower percentage (1.1 percent) of their net loans in the allowance account. Their stricter credit policies and, apparently, better credit management provided them with a high quality loan portfolio which did not require a higher allowance for loan losses. The other banks had allowances for 1.4 percent of their net loans, or about 27 percent higher than the high performance banks. With these significantly higher loan losses, the high performing banks managed to earn 50 percent higher ROA than other banks.

Average Non-Interest Income/Average Non-Interest Expense. This ratio measures expense control. The ability to improve performance while maintaining lower than average expenses is well indicated in these high performance banks. In high performing banks, 25 percent of the non-interest expenses are covered by service charges, fees, administrative fees, and other sources of non-interest income. The implication is that these banks are managing better their expenses (wages, fees, benefits, fixed assets, etc.) or they are pricing their products effectively. High performance banks with assets greater than $100 million were not significantly different from their counterparts of other banks. The two groups generated statistically similar coverage of their non-interest expenses from non-interest income.

Return on Equity. The aggressive lending measured by the significant difference in the loans/assets ratio and the good credit management of the high performing banks contributed strongly to the very significant differences in ROE. High performance banks averaged a ROE of 14.4 percent, 101 percent higher than other banks in the sample. This relationship held for the small banks and diminished somewhat in significance for the mid-size banks. There was no significant difference in the ROE for the large banks.


This study used a five year average of banking variables. Other research which found some of the same variables to be important in a one year analysis were identified. However, many of the variables which others considered important in the short term were not statistically significant in the long run, as determined by this study. This study statistically tested each of the high performance characteristics that the others did not. Table 3 compares the findings of three popular studies of bank performance with this study.

Table : TABLE 3 : A Comparison of Important Variables In Different Studies
Variable In This Study
Total Expenses/Assets No
U.S. Government/Assets Yes
Net Loans/Assets Yes
Fixed Assets/Assets No
(Pre-tax inc + OLL) Net Loan Loss No
Wages/Assets No
Assets/Employees No
Wages/Employees No
Occupancy Expense/Opr Income No
Allowance for Loan Losses No
Net Interest Margin Yes

Using the criteria of ROA and average loans to average assets, a series of "very significant" and "significant" lending related ratios were identified. These ratios represented the strengths of the high performance banks and distinguished them from other banks. In general, the correlation of performance (profitability) to high lending seemed to be related to a bank's ability to manage credit risk. It was not just how much the bank loaned, but how carefully the bank managed the lending function. Managing credit risk allowed the well managed bank to operate with a lower level of capital. This in turn, resulted in a higher ROE.

To improve bank performance, management of the lending function is significantly important. The study reveals the following characteristics of lending need to be specifically addressed by banks wanting to improve total performance. * Determine the credit needs of the market area and work

to meet those needs within the constraints of good credit

management. High performing banks in the study

maintained 58 percent of their assets in loans. * Manage the liquidity position to provide the money

estimated to be needed. Keep investments in liquid

assets to a minimum. * Develop realistic credit control policies and procedures.

These will include credit rating systems designed to

reduce/control the volume of past due and non-accrual

loans. High performing banks in the study had "bad

loans" of less than one percent of net loans. Recognizing

the need to review the loan files will help determine the

importance of charge-off and collection policies. Good

collection policies keep the net charge-off statistic low.

High performers had net charge-off of only 0.76 percent

of their five year average of loans. * Real estate lending was important in the study. All size

groups held an average of 23 percent of their assets in

real estate loans. Obviously, the high performance

banks are recognizing the needs of developers and home

buyers in their communities and are meeting that need. * Expense control, in general, helps all areas of the bank.

The lending area is no exception. Efficient cost control

systems and effective pricing policies work together to

strengthen the net interest margin. High performance

banks in the study averaged six percent NIM over the

five years. This reflects management's attention to these

important details, in all size groups included in the study.

The ratios listed in Tables 1 and 2 can be used in developing the appropriate strategies for improving the efficiency of the community bank. The information in this report can be used by corporate treasurers to assist them in determining how to evaluate the performance of a bank with which they now do or are considering doing business with.



[1.] Barry, L.M. "District Bank Performance in 1987: Bigger is Not Necessarily Better." Review, Federal Reserve Bank of St. Louis, Vol. 20, No. 2. March/April 1988, pp. 39-48. [2.] Cates, D. "Required Reading." American Banker, April 13, 1988, p. 4. [3.] Dince, R.R. and D. Holdren. "Ringing Bells in Tennessee: Traits of High Performance Banks." The Tennessee Banker, September 1988, pp. 21-23. [4.] ______. "High Performance Banking in the IBAA." Independent Banker, Vol. 38, No. 10, October 1988, pp. 25-28. [5.] Ford, W.F. "Using |High Performance' Data to Plan Your Bank's Future." Banking, October 1978, pp. 40-48. [6.] ______ . and D.A. Olson. "How to Manage High Performance Banks." Banking, August 1976, pp. 36-38. [7.] ______ . "How 1,000 High-Performance Banks Weathered the Recent Recession." Banking, April 1978, pp. 36-48. [8.] Holdren, D.P. "IBAA Agricultural Banking (1984-1987)." Independent Banker, Vol. 39, No. 3, March 1989, pp. 25-30. [9.] Nejezchleb, L.A. "Declining Profitability at Small Commercial Banks: A Temporary Development or a Secular Trend?" Banking and Economic Review, June 1986, pp. 9-21. [10.] Olson, D.A. "How High-Profit Banks Get That Way." Banking, May 1975, pp. 46-58. [11.] Poelker, J.S. "Developing a Comprehensive Financial Frame-work for Business Planning." Bank Administration, December 1988, pp. 44-50. [12.] Staats, W. "Good Management More Than Bottom-Line Profit." The Tennessee Banker, August 1988, pp. 18-21. [13.] Wall, Larry D. "Commercial Bank Profitability: Still Weak in 1987." Economic Review, Federal Reserve Bank of Atlanta, July/August 1988, pp. 28-42. [14.] Whalen, G. and J. B. Thomson. "Using Financial Data to Identify Changes in Bank Condition." Economic Reviews, Federal Reserve Bank of Cleveland, Vol. 24, No. 2, 1988, pp. 17-26.
COPYRIGHT 1991 St. John's University, College of Business Administration
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1991 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Holdren, Don P.
Publication:Review of Business
Date:Mar 22, 1991
Previous Article:The effects of cognitive style on the design of expert systems.
Next Article:The volatility of short-term interest rates.

Related Articles
Statements to the Congress.
Rural banks and their communities: a matter of survival.
CRA opportunities important in the current finance environment.
N.H. business will weather the storm.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters