Printer Friendly



Banking is an integral part of all global business and important to a local economy. With 94% of banks in the United States being community banks, failure of the banks can detrimental to not only the bank investors but also the economy in the region. Loss of liquidity from bank failure directly impacts business and individual lending. It can cascade into a wider contagion, affecting the local, regional, country, and ultimately the world economy (Diamond & Rajan, 2005, Cifuentes, Ferrucci, & Shin, 2005). Per the FDIC, "Community banks focus on providing traditional banking services in their local community. They obtain most of their core deposits locally and make many of their loans to local businesses. (FDIC, p. 1-1)" The global downturn in 2008-9 resulted in many banks failing. The ability to predict which banks might fail could be very helpful to banks, regulators, investors and their local communities.

A review of the literature concerning bank capital regulations and reporting requirements show little consensus, and often direct contradiction. Some authors such as Barber, Chang and Thurston (1996) claim the requirements are too high and while other they are too low. Certain studies show the requirements could increase the risk profiles of banks, whereas others claim that they decrease the riskiness of banks. For banking ratios, some studies including the one by Badea and Matei (2015) posit the predictive capacity of the ratios and yet others suggest they are shortsighted. Many authors such as Altman (1968, 1977, and 1983) and Taffler (1984) have put forward tests using ratio analysis along to predict corporate and banking failure. However, in almost every study, the research compares the failed companies' ratios to those of a random sample of all companies. In this, the failed company has worse ratios compared to good banks. Intuitively, there would be an expectation that a failed company would have worse ratios. As significant distinction of this study was to compare the ratios of the failed banks to a group of banks that were below regulatory standards and would be reported as "troubled". This study tries to answer why certain banks, with poor financials, would fail while others in the same class would not?

The Federal Depositor Insurance Corporation's (FDIC) Capital adequacy, Asset quality, Management, Earnings, Liquidity, and Sensitivity (CAMELS) score is a rating for each bank and is touted as the most relevant indicator for prediction of commercial bank failure post-downturn (Baek, Balasubramanian, & Lee, 2015, p. 107). The score is released to the individual bank to help with the management and/or the rehabilitation of the bank. The FDIC regulators accumulate a list of "troubled banks" for special tracking by regulators. However, to prevent the "run on banks" issue and to allow the rehabilitation for the banks, the FDIC does not publish a full list of their "troubled bank" ratings. A failure of a large bank or large number of smaller banks may cause a contagion that spreads to the industry (Baek, Balasubramanian, & Lee, 2015, p. 95). In this, many authors have claimed success in finding a predictor or proxy for the CAMELS score that can identify which banks will fail in the future. However, in reality, the true success of these proxies is shown to only be successful in a tight window of time. It is true, if you group the five percent of banks with the lowest return on equity, return on assets or other ratios together, it is highly probable that the banks that actually fail will be in that group. But this does not mean that the proxy is always successful in predicting bankruptcy failure.

Other than the unpublished FDIC CAMELS score, arguments have arisen in the literature on what would provide the best information in the current banking environment. Some authors recommend capital ratios as the best proxy for the troubled bank score. Other authors recommend the change in the income as the best indicator. Financial accounting ratios, leverage ratios and cash flows have all been endorsed as the best proxy. Additionally, some parties including IDC Financial Publishing, Inc. have recommended models with proxies for the various components of the CAMELS score. The IDC Financial Publishing version includes proxies to form their own CAMEL score which includes capital risk, adequacy of capital and reserves, margins, earnings, and liquidity.

Former Federal Reserve Chairman Ben Bernanke said, "Capital is important to banking organizations and the financial system because it acts as a financial cushion to absorb a firm's losses." Numerous authors have claimed the regulations are unnecessary, too high, overreaching, etc. This study compares the ratios for the forty-two banks that failed in the years 2013-2016 versus a random sample of "troubled" banks.


Peltzman (1970) explained the difference between banking and most other industries relate to the relative importance of the capital to banks. The more the capital a bank has, the more the value of its assets can fall before depositors incur losses. However, Peltzman found that a simple capital model, such as could be applied to any other industry, can explain the investment of capital in banking. The author says that there is no evidence that bank investment behavior is affected by the standards set by the government regulations. Per Peltzman (1970), the banks substitution of deposit insurance for bank capital has caused an illusion of a bank capital problem.

Kahane (1977), Kareken & Wallace (1978), and Sharpe (1978) studied necessity of deposit insurance and the relevance of capital requirements due to the risk created. They reviewed the effectiveness of the capital standards in banks' solvency and found depositors have no incentive to control the risk profiles at the banks due to the deposit insurance protection in complete markets. Stein (2012) expanded on this idea. He wrote, with sufficient collateral, banks prefer to fund themselves with short-term debt because it is riskless money and it can generate a cheap source of financing creating a benefit to society and banks. However, this increases the risk of failure. Therefore, financial stability regulations are necessary to control, monitor the system, and to create the use of central banks. Due to the significant shadow banking sectors, tools such as expanded reserve requirements and haircut regulations are necessary to secure the monetary system.

However, many authors challenged the application of the banking capital research in incomplete markets. These challenges led to studies that used a portfolio application instead of complete market approach including Pyle (1981), Hart & Jaffee (1974), Santomero &Watson (1977), Koehn & Santomero (1980), and Kim & Santomero (1988). These studies assumed a risk-aversion for the bankers. With this assumption, the research showed an incentive for the bankers to reduce leverage to offset the risk. Koehn & Santomero (1980) wrote regulatory capital constraints are designed to protect depositors and the banking system as a whole. The rationale given for these constraints usually point to their result being a lower probability of failure for banks. However, the authors found that higher capital requirements can cause the opposite effect.

Some authors challenged the use of capital as a tool for risk reduction. Barber, Chang & Thurston (1996) argued that increases in capital to asset ratios reduce the riskiness of the bank's equity capital but increases the probability of bank failure. Prescott (2001) wrote that the stability of banks could be accomplished with cheaper financial instruments such as convertible debt or warrants with high strike prices.

Rochet (1992) challenged the incomplete market research for banks. In this research, bank capital is treated the same as other capital. This implies that banks can change their bank capital ad hoc through purchase and sale of their stock in the markets. However, this is limited through regulation. Rochet (1992) found that in the case of undercapitalized banks, the bank is not risk-averse and would choose a higher risk profile. He found that risk-based capital regulations may not be enough to restrain a bank's appetite for risk). Other authors including Furlong & Keeley (1989) and Keeley & Furlong (1990) found the opposite. They questioned the impact of capital requirements on bank solvency and found that an increase in capital standards decreased the bank's incentive to take risk.

Orgler and Taggart (1983) described the fundamental differences between commercial banks and nonfinancial firms that contribute to a large disparity in their degree of leverage. Banks primary raise their funds in the form of deposits that offer different combinations of interest and other services such as liquidity, safety and bookkeeping. Additionally, the special function of deposit insurance from the FDIC creates a unique role as principal for the FDIC in all insured banks. The substantial and regulatory powers of the FDIC reduce agency costs and risks for private bank investors.

Diamond & Rajan (2000) expanded on this idea. They wrote that optimal bank capital structure trades off effects on liquidity creation, costs of bank distress, and the ability to force borrower repayment. Their model explains the decline in bank capital over the last two centuries and identified overlooked consequences of having regulatory capital requirements and deposit insurance. Requiring more capital makes a bank safer but also increases the bank's effective cost of capital. Also, capital requirements have subtle effects, affecting the flow of credit and even making the bank riskier. These effects emerge only when the capital requirements are seen in the context of the functions the bank performs rather than in isolation. For deposit insurance, when some deposits are uninsured, when a bank gets in trouble, those depositors uninsured will run on the bank further deteriorating the capital available.


This study used the data available from the Bank Financial Quarterly Reports published by IDC Financial Publishing, Inc. Per the IDC report, the data presented includes key financial ratios important in the evaluation of banks. Using predictor variables from the years 2010 through 2012, the analysis then looked at which banks failed during the following three years from 2013 through 2016. It was hoped that certain precursors to ultimate bank failure would become apparent, and stronger, as time moved forward.

Dependent variable for the testing was bank failure, defined as bank receivership during the years 2013-2016. Logistics regression analysis (LRA) was chosen as the appropriate statistical procedure due to the discrete or binary nature of the dependent variable (failure or no failure), and to the fact that it relaxes the requirement of normality. Statistical analysis was performed to determine which independent variables impact bank failure.

Various capital ratios, loan riskiness, and accounting variables were tested based upon predictors specified from the literature. The study specifically tested the following items:

* Tier 1 capital as a percent of total assets (T1)

* Tier 1&2 capital as a percent of risk-based capital (T2)

* Tier 1 capital as a percent of risk-based capital (T3)

* Loan loss reserve as a percent of tier 1 capital (LLL)

* Loan risk as a percent of tier 1 capital (LR)

* Balance sheet cash flow percent of tier 1 net worth (CF)

* Net income as a percent of average assets (ROA)

* Net income as a percent of average total equity (ROE)

* Increase in loan loss reserve as a percent of earning assets (LLR)

* Operating profit margin (OP)

* Leverage Spread defined as the NOPAT ROEA less the cost of adjusted debt (LS)

* Leverage Multiplier defined as the ratio of adjusted debt to adjusted equity (LM)

* Return on earning assets (ROEA)

* Return on financial leverage (ROFL)

The analysis includes the significance of each variable over each of the years 2010 to 2012, and a mean of the three years. Also, included is the logistic regression Odds Ratio (O.R.) for each independent variable (Equation 1) By this, one can determine the impact of a unit change in a variable on the odds of a bank ultimately failing. While most studies would look at the ratios in aggregate, this study considered the timeliness of the predictor to give an early warning to regulators. The bank data used in the analysis includes the forty-one U.S. commercial banks that failed during the period of 2013-2016. The comparison group includes one hundred forty-one randomly sampled banks from the 540 banks with a Tier 1&2 capital as a percentage of risk-based assets less than the 8% as required to be adequately capitalized.

The logistic model is expressed as

Log (P/1-P) = a + bx (1)

Where P = probability of no failure and 1-P the probability of failure.:

From (1), the odds P/(1-P) = exp (a + bx) (2)


Based upon the literature and suggested predictor variables, the research hypotheses for the analysis, using the logistic model in Equation (1), were:

H1: Tier 1 capital as a percentage of total assets (T1) is significantly related to the probability of bank failure

H2: Tier 1&2 capital as a percent of risk-based capital (T2) is significantly related to the probability of bank failure

H3: Tier 1 capital as a percentage of risk-based assets (T3) is significantly related to the probability of bank failure

H4: Loan loss reserve as a percent of tier 1 capital (LLL) is significantly related to the probability of bank failure

H5: Loan risk as a percent of tier 1 capital (LR) is significantly related to the probability of bank failure

H6: Balance sheet cash flow percent of tier 1 net worth (CF) is significantly related to the probability of bank failure

H7: Net income as a percent of average assets (ROA) is significantly related to the probability of bank failure

H8: Net income as a percent of average total equity (ROE) is significantly related to the probability of bank failure

H9: Increase in loan loss reserve as a percent of earning assets (LLR) is significantly related to the probability of bank failure

H10: Return on earning assets (ROEA) is significantly related to the probability of bank failure

H11: Leverage Multiplier defined as the ratio of adjusted debt to adjusted equity (LM) is significantly related to the probability of bank failure

H12 Return on financial leverage (ROFL) is significantly related to the probability of bank failure

H13: Leverage Spread defined as the NOPAT ROEA less the cost of adjusted debt (LS) is significantly related to the probability of bank failure

H14: NOPAT ROE is significantly related to the probability of bank failure

H15: The IDC CAMEL score is significantly related to the probability of bank failure

H16: Operating profit margin (OP) is significantly related to the probability of bank failure


Binary logistic regression analysis was processed for each of the sixteen ratios for 2010, 2011, 2012, and the three-year mean between the failed banks and the troubled banks to evaluate the predictive capacity of a ratio to the level of statistical significance. A relationship is shown to be statistically significant when the p-value is less than or equal to the .05 .

As shown in Table 1 for 2010, only Tier 1 capital as a percentage of risk-based assets (T3) was significantly related to bank failure. From the odds ratio (O.R), it is seen that an increase ib T3 by one unit increased the odds of not failing by 17.9% . For 2011, only Tier 1 capital as a percentage of total assets (T1), the combination of Tier 1&2 capital as a percent of risk-based capital (T2), and loan risk (LR) were significantly related to bank failure. In particular, the Odd in 2011 shows that a 1 unit increase in the Tier 1 or Tier 2 Capital ratios increase the odds of not failing by 28.1% and 17.2% respectively. Likewise, a 1 unit increase in the Loan Risk ratio decreases the odds of a bank not failing by 3%

As shown in Table 2, for 2012 many of the predictors became significant. These were: T1, T3, LLL, CF, ROA, ROE, NOPAT ROE, OP, ROEA, LM, ROFL, and LS. As seen from the regression coeffcients and the odds raios, all (except LLL and LM) had a positive relationship with the probability or odds of no failure. LLL and LM had a negative relationship with the probability of no failure. As explained above, the odds ratio gives the percent increase or decrease in the odds of no failure per unit increase in a given ratio.

Looking at the data in aggregate, the three-year mean Tier 1 capital as a percentage of total assets (T1), the combination of Tier 1&2 capital as a percent of risk-based capital (T2), CF, ROE, and OP had a positive significant relationship with the probability or odds of no failure. On the other hand, LLL, LM and LR had a negative relationship with the probability of no failure. Looking at the Odds Ratio for each significant variable in the Mean Scores, the effect of a 1 unit increase in Tier 1 capital has the largest effect, 35% increase in the odds of a bank not failing. Hence, even a very small increase in T1 makes a large difference in decreasing the odds of failure.

The evidence from these results makes the case that bank capital ratios are reliable early warning mechanisms to indicate potential failure.


This study reviewed sixteen ratios to evaluate their predictive capacity for bank receivership, focusing in on already troubled entities. The results indicate that capital ratios and loan risk are the most successful ratios in predicting bank failure in earlier periods. This supports the regulatory focus and requirements for capital to be held within the banks.

The question as to why some banks with poor capital ratios fail while others survive might be due to active bank regulation. Unlike more commercial industries, banking is subject to active oversight, with the Federal Reserve and the Office of the Controller of the Currency having programs to help struggling banks back to financial health. This study hopefully provides regulators with additional tools and insight for considering which banks might be headed toward failure.

Community banks are an important part of our communities. With this, the ability to predict bank failure is very important for communities and the real economy due to the loss of liquidity for local investment. Further, bank failure can cascade affecting not only a local economy, but much farther afield. The capital ratios in the United States for banks have very strict regulatory requirements. While many authors have challenged bank capital and the reporting requirements, transparent reporting provides useful indicators to regulators for rehabilitation of struggling banks prior to collapse.


Altman, E.I. (1968). Financial ratios, discriminant analysis and the prediction of corporate bankruptcy. The Journal of Finance. 23(4). 589-609. Retrieved from

Altman, E.I. (1977). Predicting performance in the Savings and Loan Association industry. Journal of Monetary Economics. 3(4). 443-466. Retrieved from

Altman, E.I. (1983). Multidimensional graphics and bankruptcy prediction: A comment. Journal of Accounting Research. 21(1). 297-299.

Baek, S., Balasubramanian, S., & Lee, K. (2015), Capital structure and monitoring bank failure. Journal of Accounting and Finance, 15(4), 95-107. Retrieved from

Badea, I. & Matei, G. (2015). The Z-score model for predicting periods of financial instability. Z-score estimation for the banks listed on Bucharest Stock Exchange. Finances. 16(18). 24-35. Retrieved from

Barber, J.R., Chang, C.H., & Thurston, D.C. (1996). Bank failure, risk and capital regulation. Journal of Economics and Finance. 20(3). 13-20. Retrieved from

Cifuentes, R., Ferrucci, G., Shin, H.S. (2005). Liquidity Risk and Contagion, Journal of the European Economic Association, 3 (2-3), 1. 556-566

Diamond, D.W. & Rajan, R.G. (2000). A theory of bank capital. The Journal of Finance. 55(6). 2431-2465. Retrieved from

Diamon, D.W., & Rajan, R.G. (2005). Liquidity Shortages and Banking Crises. The Journal of Finance, 60(2). 615-647

Fama, E.F. (1980). Banking in the Theory of Finance. Journal of Monetary Economics. 39-57

Federal Deposit Insurance Corporation. (2012). FDIC Community Banking Study.

Federal Deposit Insurance Corporation. (2015). RMS Risk Management Manual of Examination Policies: Basic Examination Concepts and Guidelines. Retrieved from

Furlong, F. T., & Keeley, M. C. (1989). Capital regulation and bank risk-taking: A note. Journal of Banking & Finance, 13(4), 883-891.

Hair, J.F., Black, W.C., Babin, B.J., & Anderson, R.E. (2010). Multivariate data analysis. 7th ed. Upper Saddle River, NJ: Prentice Hall

Hart, O. D., & Jaffee, D. M. (1974). On the application of portfolio theory to depository financial intermediaries. Review of Economic Studies, 41(1), 129. Retrieved from

Kahane, Y. (1977). Capital adequacy and the regulation of financial intermediaries. Journal of Banking and Finance, 1(2). 207-218. Retrieved from

Kareken, J. H., & Wallace, N. (1978). Deposit insurance and bank regulation: A partial-equilibrium exposition. Journal of Business, 51(3), 413. Retrieved from

Keeley, M., & Furlong, F. (1990). A reexamination of mean-variance analysis of bank capital regulation. Journal of Banking & Finance, 14(1). 69-84. Retrieved from

Kim, D., & Santomero, A.M. (1988, December). Risk in banking and capital regulation. The Journal of Finance 43(5). 1219-1233. Retrieved from

Koehn, M., & Santomero, A.M. (1980, December). Regulation of bank capital and portfolio risk. The Journal of Finance, 35(5). 1235-1244. Retrieved from

Orgler, Y.E., & Taggart Jr., R.A. (1983). Implications of corporate capital structure theory for banking institutions. Journal of Money, Credit and Banking, 15(2). 212-221. DOI:10.1007/s11142-014-9281-z.

Peltzman, S. (1970). Capital investment in commercial banking and its relationship to portfolio regulation. Journal of Political Economy, 78(1). 1-26. Retrieved from

Prescott, E.S. (2001, Summer). Regulating bank capital structure to control risk. Federal Reserve Bank of Richmond Economic Quarterly, 87(3). 35-52. Retrieved from

Pyle, D. H. (1981). Bankruptcy risk in financial depository intermediaries. Journal of Monetary Economics, 8(1), 142-144. Retrieved from

Rochet, J. (1992). Capital requirements and the behavior of commercial banks. European Economic Review, 36(5). 1137-1170. Retrieved from

Santomero, A.M., & Watson, R.D. (1977). Determining an optimal capital standard for the banking industry. The Journal of Finance, 32(4). 1267-1282. Retrieved from

Sharpe, W. F. (1978). Bank capital adequacy, deposit insurance and security values. Journal of Financial & Quantitative Analysis, 13(4), 701-718. Retrieved from

Stein, J.C. (2012). Monetary policy as financial stability regulation. The Quarterly Journal of Economics, 127(1). 57-95. Retrieved from

Nancy Seelye

First Financial Bank

Paul Ziegler

University of Mary Hardin-Baylor

About the Authors:

Nancy Seelye in an Accountant for First Financial Bank headquartered in Cincinnati, Ohio. She also is an Adjunct Professor for Indiana Wesleyan University in Marian, Indiana. Additional research areas of interest are derivatives, banking, stakeholder theory, regulatory theory, and bankruptcy theory. She earned her B.S. in Accounting from Indiana University, a M.B.A. from Indiana Wesleyan University, and a Doctor of Business Administration from Anderson University of Indiana.

Paul Ziegler is an Associate Professor of Finance at the University of Mary Hardin-Baylor, Belton Texas. Additional research areas of interest include financial short selling, portfolio management, and information systems. He also works as a consultant in the aviation industry, designing flight track management and analysis software. He earned a B.S. in Mathematics from the University of Texas at Austin, a M.S. in Information Systems at the University of Mary Hardin-Baylor, and a Doctor of Business Administration from Anderson University in Indiana.

The authors would like to thank the anonymous reviewers for their important comments in strengthening the final paper.
Table 1
Logistic Regression Results for 2010 to 2011

            Coeff   O.R.   p-value    Goodness  Coeff   O.R.
                                      of Fit

T1           0.129  1,128  0.13       0.165      0.248  1,281
T2           0.075  1.078  0.18       0.144      0.159  1.172
T3           0.165  1.179  0.00 (**)  0.060      0.076  1.079
LLL          0.003  1.003  0.56       0.366     -0.004  0.996
LR          -0.014  0.986  0.17       0.296     -0.030  0,970
CF           0.001  1.001  0.67       0.145      0.002   1002
ROA         -0.098  0.907  0.31       0.218     -0.019  0,931
ROE          0.002  1.002  0.69       0.129      0.005  1.005
NOP AT ROE  -0.004  0.996  0.70       0.281      0.002   1002
OP           0.006  1.006  0.18       0.399      0.002  1.002
ROEA        -0.067  0.935  0,43       0.296     -0.057  0,945
LM          -0.157  0.855  0.07       0.143     -0.059  0.943
ROFL        -0.004  0.996  0.75       0.141      0.010  1,010
LLR          0.128  1.137  0.56       0.128      0.392  1.480
LS          -0.006  0.994  0.95       0.060     -0.081  0.922
IDC CAMEL    0.002  1.002  0.55       0.538      0.001  1.001

            p-value    Goodness
                       of Fit

T1          0,00 (**)  0,057
T2          0.00 (**)  0.156
T3          0,17       0,371
LLL         0.15       0.270
LR          0.03 (*)   0,037
CF          0.07       0.128
ROA         0,84       0 187
ROE         0.34       0.214
NOP AT ROE  0.&3       0.322
OP          0.45       0.158
ROEA        0,61       0,176
LM          0.20       0.077
ROFL        0.38       0,156
LLR         0.22       0.155
LS          0,50       0104
IDC CAMEL   0.78       0.338

Table 2
Logistic Regression Results for 2012 and Mean Scores

                          2012                                Mean
           Coeff   O.R.   p-value    Goodness  Coeff   O.R    p-value
                                     of Fit

Tl          0.184  1.202  0.01 (**)  0.058      0.300  L350   0.00 (**)
T2          0.062   1064  0.11       0.0S2      0.136  1146   0.02 (*)
T3          0.091  1.095  0.05 (*)   0.056      0.032  1.033  0.47
LLL        -0.013  0.987  0.00 (**)  0.090     -0.009  0.991  0.02 (*)
LR          -0026  0.974  0-10       0.109     -0.029  0.971  0.10
CF          0.006  1.006  0.00 (**)  0.062      0.008  LOOS   0.00 (**)
ROA         0.282  1.326  0.01 (**)  0.239      0.037  1.038  0.75
ROE         0.020  1.020  0.00 (**)  0.288      0.016  1.016  0.02 (*)
NOPAT ROE   0.045   1046  0.00 (**)  0.359      0.023  1023   0.08
OP          0.014   1014  0.00 (**)  0.467      0.012  1012   0.01 (**)
ROEA        0.235  1.265  0.03 (*)   0.321      0.038  1.039  0.79
LM         -0.115  0-891  0.01 (**)  0155      -0.149  0862   0.03 (*)
ROFL        0.039   1040  0.00 (**)  0.357      0.027  1027   0.09
LLR        -0.481  0.618  0.62       0.129      0.243  1.275  0.55
LS          0.296  1.344  0.01 (**)  0.159      0.161  1.175  0.31
IDC CAMEL   0.002  1.002  0.59       0.074      0.003  1.003  0.53

           of Fit

Tl         0.184
T2         0.110
T3         0.317
LLL        0.128
LR         0.292
CF         0.528
ROA        0.150
ROE        0.338
NOPAT ROE  0.241
OP         0.265
ROEA       0.237
LM         0.083
ROFL       0.225
LLR        0.131
LS         0.094
IDC CAMEL  0.312
COPYRIGHT 2019 International Academy of Business and Public Administration Disciplines
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Seelye, Nancy; Ziegler, Paul
Publication:International Journal of Business and Economics Perspectives (IJBEP)
Date:Sep 22, 2019

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters