CAMEL, CAMEL ., ,,,,. 75.4% 76.1%,. :, CAMEL, 1972 ( ) * ( ** (

Transcription

1 CAMEL CAMEL % 761% : CAMEL 1972 ( ) * ( chang446@skkuackr) ** ( ykk9209@fssorkr)

2 IMF (Capital adequacy) (Asset quality) (Management) (Earnings) (Liquidity) CAMEL 2000 (Early Warning System)

3 (8 3 ) 3 1) 2 (Prompt Corrective Action) 5 < 1> (115 ) 1) 3 < 1>

4 < 2> 4% 2% 1% ( ) ) (= + - ) ( ) 2) % 3% CAMEL 4% 2% 1% 3

5 CAMEL (FRB) (OCC) (FDIC) 1979 (Federal Financial Institutions Examination Council) (UFIRS) 5 5 CAMEL 5 (Capital adequacy) (Asset quality) (Management) (Earnings) (Liquidity) 1997 (Sensitivity to market risk) CAMELS CAMEL (Sensitivity) CAMEL ) 5 1 ( ) 2 ( ) 3 ( ) 4 ( ) 5 ( ) 5 2) CAMEL CAMEL

6 CAMEL < 2> Altman(1968) ZETA (1977) Ohlson(1980) Bathory(1987) K-score (FDIC) (FRB) (1998) (1997)

7 CAMEL Sinkey(1975) (FDIC) Martin(1977) CAMEL 25 West(1985) 1900 (factor analysis) CAMEL CAMEL CAMEL Thomson(1991) 1980 CAMEL Espahbodi(1991)

8 CAMEL 13 Kolari Glennon & Caputo(2000) 1990 (non-parametric) (trait recognition approach) CAMEL Jagtiani Kolari Lemieux & Shin(2003) Laitinen & Laitinen(2000) 2 32 (10 ) 3

9 (WH Beaver) (EB Deakin) (G Foster) (financial distress) (real insolvency) (bankruptcy) ) 6 (6 ) (1 2 ) 3) t-

10 ( 6 ) < 3> (2002 ) ( ) (validation)

11 1/2 ( 1 ) ( 2 ) < 4> < 5> /

12 BIS BIS

13 (too big to fail) 1 16 ( ) Estrella (2000) 10% M&A < 6> 52 20

14 ) ( : ) ) (FDIC 1999) 25 < 7> 4) ( ) ( ) ( ) M&A 5)

15 (C) (A) (E) (L) x1 ( ) / ( ) x2 ( )** / ( ) x3 ** ( - ) /( - ) x4 * / x5 / x6 / x7 ** / x8 * / x9 * / x10 / ( x11 ** )/ ( ) x12 * / x13 / x14 ( ) ln( ) ( x15 16 ) 1 x16 / x17 10% D1 ( 1 0) ( 1 3 D2 1 0) x18 x19 x20 x21 (%p) x22 (%p) x23 (%p) * ** ( )

16 Altman(1968) ) ( ( 1997 ) ( ) (Logistic Regression Model Analysis) 6) k 1 ( ) (Maximum Likelihood Estimation) 6) (x14)

17 최소한하나의 (Likelihood Ratio) k k < 1> Pi( ) W i

18 (cutoff point) 05 (Maddala(1986) Thomson(1991)) (1) (1) (1) (2) (2) ( )

19 t- < 8> 8 Wald- 2 p x2( ) x3( ) x4( ) x7( ) x8( ) x9( ) x11( ) <00001 x12( ) * (LR 2 ) : ** : 744%( 1 329% 2 225%)

20 (x2) (x8) (x11) 1% (x3) (x12) 10% (x4) (x7) (x9) 744% 1 7) CAEL 8) 4 9) < 9> 1% 10% 4 (< 4> ) 732% 8 (744%) 7) < 3> 8) (x3) (x4) 0776 (x7) (x8) (x9) ) 2001 BIS

21 Wald- 2 p x2( ) x3( ) x7( ) x11( ) <00001 * (LR 2 ) : ** : 732%( 1 342% 2 237%) 2745% 10) ( ) < 2> 약 약 약 약 10)

22 (< 9>) 50% ( ) ( ) t- t- < 5> (x11) (x22) (x23) CAEL

23 (x8) (x12) t- (stepwise selection method) < 10> 3 ( ) (152) (170) (166) ** *** x2( ) *** *** x7( ) *** x8( ) *** x11( ) *** ** x12( ) 49133* x13( ) 39724*** x18( ) 25917*** x19( ) ** ( 2 LR) 9509*** 3722*** 4554*** (%) (%) (%) Concordant (%) Discordant (%) Tied (%) (cutoff point) *** 1% ** 5% * 10%

24 (robust) 3 1/2 < 11> (x2) (x7) (x8) (x11) ( =12483) 813% 185% Sommer s D Gamma Tau-a c (+) ( ) ( ) ( )

25 ( ) Wald- 2 p x2( ) x7( ) x8( ) x11( ) * (LR 2 ) : 6070 ** Percent Concordant 813 Somers' D 0627 Percent Discordant 185 Gamma 0628 Percent Tied 02 Tau-a 0264 Pairs c % < 12> (classification table) : 754% (=(132+52)/244) ( =030) 1 : 288% (=21/73) ( (Sensitivity)=712%) 2 : 228% (=39/171) ( (Specificity)=772%) 713% < 13>

26 : 713% (=(124+50)/244) ( : 030) 1 : 315% (=23/73) ( (Sensitivity)=685%) 2 : 275% (=47/171) ( (Specificity)=725%) < 14> (x2) (x3) (x11) (x22) 1 < 15> 761% 1 (754%) < 16> 526% ( 2 474%) 700% 1

27 11) Wald- 2 p x2( ) x3( ) <00001 x11( ) <00001 x22( ) * (LR 2 ) : 9954 ** Percent Concordant 825 Somers' D 0652 Percent Discordant 173 Gamma 0653 Percent Tied 02 Tau-a 0274 Pairs c : 761% (=(180+65)/322) ( =030) 1 : 323% (=31/96) ( (Sensitivity)=677%) 2 : 204% (=46/226) ( (Specificity)=796%) 2 ( 10) ( ) 11) 1 ( ) 2 ( )

28 ( ) 12) (robustness) : 578%(=(61+35)/166) ( : 030) 1 : 300%(=15/50) ( (Sensitivity) = 700%) 2 : 474%(=55/116) ( (Specificity) = 526%) 12) (FRS) SEER (System to Estimate Examination Ratings) (pooled data) (Jagtiani 2003) (FDIC) SCOR(Statistical CAMELS Off-site Rating system) 12 1

29 BIS 13) CAMEL CAMEL 8 t- 4 (CAEL) 4 13) < 6>

30 / (713%) 2 (578%)

31 CAMEL CAMEL CAMEL (off-site surveillance) (control power)

32 ( ) : Couto Rodrigo Luis Rosa Framework for the Assessment of Bank Earnings BIS Financial Stability Institute Deakin E A Discriminant Analysis of Predictors of Business Failure Journal of Accounting Research 10(1) Espahbodi Pouran Identification of Problem Banks and Binary Choice Models Journal of Banking and Finance Estrella Arturo Sangkyun Park and Stavros Peristiani Capital Ratios as Predictors of Bank Failure Economic Policy Review(FRB of New York) July Jagtiani Julapa James Koloari Catharine Lemieux and Hwan Shin Early Warning Models for Bank Supervision: Simpler could be better Economic Respectives(FRB of Chicago) 3rd Quarter Kolari James Dennis Glennon Hwan Shin and Michele Caputo Predicting Large US Commercial Bank Failures Economic and Policy Analysis Working Paper Office of Comptroller of the Currency January Laitinen Erkki K and Teija Laitinen Bankruptcy Prediction Application of the Taylor s Expansion in Logistic Regression International Review of Financial Analysis

33 14 Maddala G S Econometric Issues in the Empirical Analysis of Thrift Institutions Insolvency and Failure Federal Home Loan Bank Board Invited Research Working Paper 56 October Martin Daniel Early Warning of Bank Failure A logit regression approach Journal of Banking and Finance Ohlson J S Financial Ratios and Probabilistic Prediction of Bankruptcy Journal of Accounting Research 1980 Spring Sinkey Joseph F Jr A Multivariate Statistical Analysis of the Characteristics of Problem Banks Journal of Finance 30 March Shleifer A and R W Vishny A Survey of Corporate Governance Journal of Finance Thomson James B Predicting Bank Failures in the 1980s Economic Review (FRB of Cleveland) 1st Quarter Issue 1 20 West Robert Craig A Factor-Analytic Approach to Bank Condition Journal of Banking and Finance 9 June History of the Eighties : Lessons for the Future - Vol 1 An Examination of the Banking Crises of the 1980s and Early 1990s FDIC Uniform Financial Institutions Rating System Supervisory Letter Board of Governors of the Federal Reserve System Dec

34 ` `94 `96 24 ` ( )

35 BIS (Capital adequacy) BIS (Asset quality) (Management) (Earnings) (Liquidity) 1) ) : < 6>

36 ) : 744% (=(265+98)/488) 1 : 329% (=48/146) ( (Sensitivity)=671%) 2 : 225% (=77/342) ( (Specificity)=775%) 2) Maddala(1986) Thomson(1991) : 732% (=(261+96) / 488) ( =030) 1 : 342% (=50/146) ( (Sensitivity)=658%) 2 : 237% (=81/342) ( (Specificity)=763%)

37 x1( ) * 176* x2( ) 344** 464** 385** x3( ) -564** -329** -101 x4( ) -501** -362** -272** x5( ) ** -245** x6( ) -565** -333** -198* x7( ) 709** 295** 357** x8( ) x9( ) -486** -312** -342** x10( ) -410** x11( ) 428** 378** 319** x12( ) ** x13( ) * x14(ln( )) x15( ) x16( ) x17( ) -262** x18( ) * 023 x19( ) x20( ) -177* x21( ) 373** x22( ) -400** -212** -194* x23( ) -384** -278** -239** t- t ** 5% * 10%

38 ( ) CAEL (8 ) CAEL (4 ) 1 1) ** *** *** ** *** 22665* 24086*** 69124*** * *** *** ** *** *** *** *** 23243* 744% ( ) 732% ( ) 754% (713%) 2) ** 761% (578%) 1) 1 ( 1/2 ) 2) 2 ( ) 3) *** 1% ** 5% * 10%

39 < Abstract > This study attempts to examine existing financial distress prediction models so as to determine whether these models could be applied to mutual savings banks (MSBs) Using the logistic regression analysis method failure prediction models have been estimated and their validities have been tested with holdout samples The result of an empirical analysis shows that CAMEL indicators - capital adequacy asset quality management status earnings and liquidity - bear some significance together with equity ratio ROA the ratio of cost to total asset current ratio and net non-performing loan ratio in predicting MSB failures But the variables related to corporate governance turned out to be insignificant in discerning survival MSBs from failure ones The in-sample classification accuracy of the estimated model ranges from 75% to 76% and out-of-sample classification accuracy ranges from 58% to 71% The prediction model estimated using four main CAEL indicators has a classification accuracy of 73% The main contributions of this study can be summarized in0to two points First it extends the previous studies on financial distress prediction to the area of financial institution failure prediction about which no previous studies had been done Secondly this study provides a set of test results on the validity of the CAMEL rating system