Long-term strategic effects of mergers and acquisitions in Asia-Pacific banks

Accepted Manuscript

Long-Term Strategic Effects of Mergers and Acquisitions in Asia-Pacific Banks Yoko Shirasu PII: DOI: Reference:

S1544-6123(17)30237-4 10.1016/j.frl.2017.07.003 FRL 737

To appear in:

Finance Research Letters

Received date: Revised date: Accepted date:

27 April 2017 16 June 2017 5 July 2017

Please cite this article as: Yoko Shirasu , Long-Term Strategic Effects of Mergers and Acquisitions in Asia-Pacific Banks, Finance Research Letters (2017), doi: 10.1016/j.frl.2017.07.003

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

AC

CE

PT

ED

M

AN US

CR IP T

ACCEPTED MANUSCRIPT Highlights  Asian banks’ M&A enable banks to increase new loans and capital adequacy.  Acquirer banks fail to make profits due to non-performing loans.  Among the different countries’ systems, in case of cross border deals, target bank countries have strong legal systems and stringent regulations rules, such that acquirer banks can enjoy higher equity at a lower cost.  The legal system with strong investor protection and more stringent financial regulations plays an important role in resolving the problems faced by the banks.

1

ACCEPTED MANUSCRIPT Long-Term Strategic Effects of Mergers and Acquisitions in Asia-Pacific Banks

Yoko Shirasu a a

Department of Economics, Aoyama Gakuin University, 4-4-25 Shibuya, Shibuya-ku, Tokyo 150-8366,

CR IP T

Japan; email:[email protected]; tel: +81-3-3409-6431

Abstract

This study empirically examines the effects of the Asian banks’ M&A, focusing on the long-term changes

AN US

in banking management strategies for the acquirer banks. Target countries have tighter/more stringent legal and regulatory rules to ensure that the acquirer banks enjoy higher equity at lower cost. We find that Asian banks’ M&As contribute toward increasing new loans and enhancing capital adequacy. However, banks fail to make profits because of the non-performing loans. Most importantly, as part of cross-border

M

deals, strong legal systems and stringent regulations could enable Asian banks to operate effectively by

ED

undertaking M&A between countries with different economic systems.

PT

Keywords: Asian Bank M&A, Strategy, Investor Protection, Regulation, Capital

AC

CE

JEL classification: G15, G21, G34

2

1.

ACCEPTED MANUSCRIPT

Introduction

Since the 1990s, Asian banks have aggressively promoted mergers and acquisitions (M&A). Asian acquirer banks promote strategic banking businesses for their clients, but promote bank strategies for themselves using M&A. This study empirically examines the long-term effects of banking management strategies for acquirer banks. We examine the banking strategic management factor as performed in Altunbas and

CR IP T

Marques (2008), and explain a country’s characteristics that relate to a bank’s financial outcomes. When considering the differences between the legal systems or regulatory rules of different countries, the study shows that the English legal system, regulatory scope rule, and regulatory entry rule play a significant role

AN US

in the creation of sound Asian banks. Among the countries with different systems, target bank countries, in particular, have stronger legal systems and more stringent regulations rules, while acquirer banks are enjoy a higher equity at a lower cost. In Asia, the legal systems and financial regulations play an important role.

M

The remainder of this paper is structured as follows. Section 2 presents the relevant literature and

ED

discussion issues. Section 3 describes the study’s data and empirical methods. Section 4 describes the data

2.

PT

on Asian banks. Section 5 provides the study’s empirical results. Finally, Section 6 concludes the paper.

Literature review and discussion issues

CE

Many studies have conducted on financial conglomerates . Laeven and Levine (2007) find the

AC

diversification discount in a financial conglomerate, while Baele et al. (2007) and Artikis et al. (2008) recognize a significant relationship between the degree of functional diversification and franchise value. Moreover, Caiazza et al. (2012) present the “acquire to restructure” hypothesis, which posits that targets banks are typically less powerful banks that are acquired for restructuring, with the intention of boosting sales. The focus of most studies is limited to defensive M&A analyses of problems related to defensive non-performing loans (NPLs) (Sakai et al., 2009), business restructuring, and efficiency. Concerning the topics for discussion in this study, the first one pertains to the strategic 3

ACCEPTED MANUSCRIPT

management change of acquire banks; do the factors of banking strategic management have an impact on acquisitions, following one year and three years of an acquisition? Essentially, we examine four strategic management factors: earning diversification, risk, cost control, and capital adequacy level strategies. To ensure economic benefits, we test the effects of not only loan business growth and cost efficiency, but also return on assets (ROA) and market-to-book ratios. Rossi and Volpin (2004), Moeller and Schlingemann (2005), and Fauver et al. (2003) empirically

CR IP T

show the differences in nationalities, legal and market systems, and regulatory systems, while Stiegner and Sutton (2011) show that greater cultural differences have a positive influence on the long-term performance of banks. Wank et al. (2017) show bank type or original effect the productive scope in bank

AN US

M&A. Wank et al. (2016) find that one of the drivers of virtual efficiency is sharing the same accounting principles. Additionally, Barth et al. (2001, 2004, 2008) empirically show the differences among a broad array of bank regulations and supervisory practices as well as bank development, performance, and stability. Moreover, some literature present evidence that regulatory and cultural barriers limit the

M

international expansion of banks (De Haas and Van Lelyveld, 2010) and more profitable and larger banks

ED

find it easier to overcome such barriers (Calzolari and Loranth, 2011), and propose policy measures to increase the supervision of banks’ international activities (Ongena et al., 2013).

PT

The second topic concerns the fact that the available evidence on the differences in target countries’ characteristics could help us understand some factors in bank acquisition. Thus, the differences

3.

AC

CE

in legal systems, degree of economic freedom, and financial regulatory systems should be considered.

3.1.

Data and methodologies Data

Data on alliance and M&A announcements, as well as financial data, are drawn from Thomson ONE Investment Banking and Datastream, 2000–2011. We collect data on all the transactions of Asian 1 listed banks that have at least acquirer or target to be an Asian bank; we employ completed transactions of bank 1

The investigation uses Asian data from all the Asia-Pacific countries. 4

ACCEPTED MANUSCRIPT

acquisitions.2 We present four strategic financial variables along with Altunbas and Marques (2008), as seen in the Appendix. We use the data on legal systems obtained from the works of La Porta et al. (1997), Fauver et al. (2003), and Becka et al. (2003). Additionally, we employ a country’s Economic Freedom of the World (EFW) index, obtained from the study by Moeller and Schlingemann (2005). We use the available dataset of the bank regulatory environment derived by Barth et al. (2008) from the World Bank website. Further, the macroeconomic environment is likely to affect bank activities and investment

3.2.

CR IP T

decisions (Pana et al., 2010), measured as the annual growth rate of gross domestic product (GDP).

Difference-in-difference methods

AN US

For the long-term analysis, we utilize regression analysis using difference-in-difference (DID) methods, dependent variables are strategic variables. We set all the M&A deals as the treatment group and all the data on the non-M&A Asian listed banks as the control group.

In this study, following the econometric methods of Inui et al. (2013), we construct two models: it

  it , (1)

StrategicVariableit 3  StrategicVariableit1   0  1 Trend it   2

it

  it , (2)

ED

M

StrategicVariableit 1  StrategicVariableit1   0  1 Trend it   2

where Strategic Variable (SV) is used, as in Altunbas and Marques (2008); Trend is the dummy variable, it the vector of the control variables. We

PT

which is 1 for acquisitions data and 0 for non-acquisitions data;

employ the control variables of ln (asset), the GDP growth rate of target and acquirer countries,

CE

cross-border dummy, alliance dummy, diversification dummy, effective year dummy, and both acquirer

AC

and target country dummies. Equation (1) estimates the change in the M&A effects of the Strategic Variable from t−1 to 1+1, while Equation (2) estimates this change from t−1 to 1+3. Further, we assess the significance of the coefficient of Trend. Now, let us explain the country characteristics. In order to investigate how acquisitions affect

2

We select sample transactions that have a dollar value. The number of initial announced transactions is 1907, and of those, the completed transactions total 1137. Moreover, we construct our sample using the following procedures: (1) we select observations in which the acquirer company is a bank or financial holding company (800 observations), and (2) we select observations that have total asset data (563 observations). However, not all observations have all types of financial data; there are many missing data. This leaves us with less than 563 observations. 5

ACCEPTED MANUSCRIPT

strategic variables depending on the target country’s characteristics differently, the acquirer countries are divided as follows: 1) by the differences in legal systems (e.g., English vs. French); 2) by the differences in EFW values; and 3) by the differences in the strength of financial regulations (e.g., bank-activities-scope regulations (Regulation scope), foreign-bank-entry regulations (Regulation entry), and bank-self-monitoring regulations as information disclosure (Regulation monitoring). To investigate the differences in countries’ characteristics, following the methods of Nguyen and

differences, as follows: 3

SVit 1  SVit 1   0  1  SameLawTrend i

CR IP T

Wilson (2016), we set another econometric model using DID methods and the example of the legal system

  2  Different[ English]LawTrend i  3  Different[ French]LawTrend i , (3) it

  it

AN US

  4  Different[Other ]LawTrend i   4

where we split the LawTrend variable into four law trend dummy variables. If acquirers and targets have the same legal system, Same Law Trend is 1; if acquirers and targets have different legal systems and the

M

targets have the English legal system, Different[English]LawTrend 4 is 1 and 0 otherwise. Moreover, the French legal system case and other cases are similar. In another sample of EFW and financial regulation

ED

data, we split the Trend variable into three dummy variables. If acquirers and targets have the same categorized score, then the SameTrend is 1; the other data, including those of non-acquisitions, are 0. If

PT

acquirers have different categorized score and the target score is under the mean, UnderMean5 is 1, and

CE

the other data are 0. Further, if acquirers have different categorized score and the target score is above the mean, AboveMean is 1, and the other data are 0 as follows:

AC

SVit 1  SVit 1  0  1  SameTrend i

  2  Different[UnderMean]Trend i  3  Different[ AboveMean]Trend i   4

3.3.

3

it

  it .

(4)

Average Treatment Effect from Propensity Score Matching

Additionally, we estimate the change in the M&A effects of the Strategic Variable from t−1 to 1+3 and from t−1 to 1+3. 4 Different refers to cross-border M&A deals. 5 UnderMean and AboveMean are cross-border M&A deals. 6

ACCEPTED MANUSCRIPT

For robustness, we compute the average treatment effects (ATE) using the propensity score matching (PSM) method.6 To our knowledge, PSM is a relatively new subject in econometric papers, one of which has been used in relation to M&A (Behr and Heid, 2011). We estimate the propensity using the set of variables with year dummy, acquirer and target country dummies. After PSM, we check the balanced box charts and test for balance using the standardized difference.

Sample description

CR IP T

4.

Table 1 presents the basic statistics in our regression analysis. Panel A of Table 1 shows the one-year change in effectiveness, “Treatment Banks” data, and non-acquisition data that is called “Control Banks.”

Panel A) one-year change

AN US

Panel B of Table 1 shows the three-year change.

Panel B) three-year change

Variable

Obs

Std. Dev. Treatment Banks

Obs

The other operational income ratio NPL loans ratio

540

0.000

0.009

3,490

473

-0.010

0.077

2,846

Total loan ratio Deposit-loans ratio

529 522

-0.004 -0.047

0.049 0.390

3,377 3,351

Mean

Std. Dev.

Obs

Mean

-0.001

0.048

492

-0.001

0.012

3,215

-0.002

0.066

-0.064

2.704

430

-0.015

0.082

2,592

-0.161

3.732

-0.001 -0.260

0.126 11.724

481 475

-0.006 -0.083

0.068 0.490

3,100 3,077

-0.002 -0.514

0.219 14.614

3,264

-0.612

52.741

503

0.283

34.455

2,998

-1.841

50.025

4,255

-0.001

0.074

521

-0.002

0.078

4,046

-0.001

0.106

Control Banks

M

Mean

Std. Dev. Treatment Banks

Obs

Mean

Std. Dev. Control Banks

545

0.776

29.375

563

-0.002

0.061

Tier 1 capital ratio

402

-0.005

0.082

2,145

-0.001

0.103

361

0.003

0.089

1,803

-0.071

2.792

ROA

563

0.000

0.030

4,258

0.000

0.189

521

-0.002

0.027

4,050

0.001

0.206

Size Market-to-book

563 548

0.146 -0.017

0.234 0.260

4,262 3,781

0.103 -0.020

0.228 0.534

521 509

0.391 -0.022

0.338 0.282

4,054 3,605

0.312 -0.020

0.438 0.947

GDP growth(Acquirer)

806

3.781

4.171

6,632

3.790

4.239

806

3.781

4.171

6,632

3.790

4.239

GDP growth(Target)

793

6,632

3.790

4.239

793

3.989

4.315

6,632

3.790

4.239

PT

ED

Total cost ratio Total capital ratio

4.315

Table 1. Basic statistics

AC

CE

3.989

6

Our employed matching algorithm method is greedy matching. 7

ACCEPTED MANUSCRIPT Panel A) Treatment Banks Acquirer banks 2000 4 2

2001 14 5

3

2 2 9

1 3 1 1 2 9 26

2002 8 2

2003 14 4

2004 7 16 1 2

2005 5 11 3

2006 15 6 2

2007 12 6 1 1

2008 15 5 4 2

11

13

7

5

2 3 6 1 1 1

1 2 3 2 1

3 2 2 4 2

6 1 3 2 1 1

6 2 2 8 4 1

3 7 1 2

12 1 3 3 1 1 1

8 1 2 1 4

6 51

7 39

8 45

4 55

9 50

5 48

12 51

2001 13 5 1 2 2 9

2002 8 2

2003 14 4

2004 7 16 3 1

2005 4 10 4

2006 15 4 1 2

2007 10 5 4

2008 14 5 7 1

11

12

7

5

2 3 5 1 1 2

1 2 2 1 2 3

3 2 2 2 2 3

6 3 2 1 1 1

7 2 2 1 3 5

3

3 3 2 1 2

12 1 3 3 1 1 1

3 7 2

2 1 3 3 1

1

7

3 1

6

1

8

3

8

1

12 52

3

8

4 1 4 1 9

2010 7 6 1

2011 4 3 2

2012 3

7 1

3

2

2

3

2

1 1 3

3 3 6 37

3 5 24

2009 10 5 4 2 9

3 2

1

15

1

2010 5 6 1

2011 2 3 1

2012 2

8 1

3

1

2

1

2 1

2 1

3

1 2 4

2

3

1

3

4

5

1

1

1

1

1 3 4 2 1 1 8 2

2013

2013

1

1

1

M

1

1

1

4

2

1

1 50

39

3

8 1 250

2002 87 27 23 4 7 7 7 16 15 23 11 19 8

AC 207

55

50

48

51

1 68

52

37

1 23

15

2003 88 31 24 4 7 9 10 18 16 25 11 21 8

2004 92 31 17 3 8 8 13 18 12 24 7 23 10

2005 96 33 19 4 9 6 15 15 14 25 10 20 14

10 1 283

1 9 1 277

1 8

2006 95 38 27 3 11 6 17 15 11 27 8 20 12 5 3 7

2007 91 37 24 3 12 6 20 15 9 28 7 16 12 7 5 7

2008 94 39 26 3 12 7 19 19 8 27 7 20 15 9 5 7

2009 82 38 24 3 11 4 23 16 9 28 6 18 19 11 5 7

2010 93 40 26 3 11 9 19 21 11 25 8 20 15 25 5 9

2011 95 42 33 4 15 9 20 17 11 21 9 19 20 27 7 8

2012 101 41 42 4 15 10 21 19 11 19 9 19 21 27 5 12

2013 102 43 43 4 16 12 21 19 11 20 9 21 25 27 8 14

289

305

299

317

304

340

357

376

395

PT

2001 83 26 24 3 4 10 9 17 16 18 9 15 7

CE

2000 86 11 11 2 4 9 8 10 14 19 8 18 4

45

ED

26

Panel B) Control Banks Japan India Indonesia Singapore Sri Lanka Thailand Pakistan Philippines Malaysia Korea Hong Kong Taiwan China Bangladesh Vietnam Australia New Zealand Total

1 8

3 3 2

AN US

2000 4 2 1

Japan India Indonesia Singapore Sri Lanka Thailand Pakistan Philippines Malaysia Korea Hong Kong Taiwan China Kazakhstan Vietnam Macao Australia Tonga New Zealand Fiji Samoa Rus. U.S. Kenya The others Total

2 8 1 3

7 1 12 69

Target Entities

2009 10 5 3

CR IP T

Japan India Indonesia Singapore Sri Lanka Thailand Pakistan Philippines Malaysia Korea Hong Kong Taiwan China Vietnam Australia Total

8 1 263

Source: Thomson Reuter Data Base

1

Table 2. Number of max deals using analysis Table 2 presents the number of max deals (i.e., number of data observations that include financial data on total assets) from our analysis. The upper of Panel A of Table 2 shows the treatment banks, or the 8

Total 118 71 15 9 4 94 5 24 30 39 15 15 23 6 95 563 Total 108 67 27 6 5 95 4 26 20 33 18 16 23 3 14 2 73 1 5 1 1 1 10 1 3 563

Total 1285 477 363 47 142 112 222 235 168 329 119 269 190 138 45 117 4 4262

ACCEPTED MANUSCRIPT

acquirer Asian banks. In our sample, many acquisitions occurred in Japan (118), Australia (95), and Thailand (94). Moreover, the lower of Panel A7 shows the target countries, of which Japan, Thailand, and Australia have the highest (108 banks), second highest (95), and third highest (73) share, respectively. Panel B of Table 2 shows the control banks, which are all Asian banks without acquisitions. The number of control banks increased by double each passing year, from 207 banks in 2000 to 395 banks in 2013. The

CR IP T

panel further shows that the largest country is Japan, while the second largest is India.

Empirical results

5.

Change in strategies

5.1.

AN US

We empirically extract the change in strategies of acquirers after acquisitions. Panel A of Table 3 show the results of the DID regression on the change in strategies one year after the acquisitions. There are no economically favorable results and no significant results of returns on assets and market-to-book ratios. In the initial stage after the acquisition, acquirer banks become larger and grow more loans. However, after

M

three years, these banks may have been renewing many loan agreements and deposits, and hence , in Panel

of NPLs

regression

(1) 0.0006 (0.897) 4,815 0.0036

AC

n r2

ΔROA

Δ Δthe other market-to-book operational income ratio

CE

Dependent variable

PT

Panel A) after one year

ED

B, they issue larger loans after three years and finally qualify as adequate capital banks, despite the growth

(2) 0.0219 (0.117) 4,323 0.0166

(3) -0.0004 (0.581) 4,025 0.01

ΔNPL ratio

Δsize

(4) 0.0395 (0.407) 3,315 0.0079

(5) 0.0297 (0.058) 4,820 0.1425

Δtotal loans

*

(6) 0.0346 (0.065) 3,897 0.2086

*

ΔNPLs

Δtotal costs

Δtotal capital

(7) 0.0706 (0.251) 3,307 0.0437

(8) 0.0357 (0.100) 3,802 0.1807

(9) 0.0303 (0.147) 4,771 0.1047

*

Panel B) after three year regression n r2

(1) -0.0054 (0.443) 4,566 0.0045

(2) -0.007 (0.760) 4,109 0.0101

(3) 0.0006 (0.645) 3,703 0.013

(4) 0.1783 (0.015) 3,019 0.014

**

(5) 0.039 (0.239) 4,570 0.1721

(6) 0.0755 (0.032) 3,575 0.2548

**

(7) 0.2861 (0.003) 3,019 0.1097

***

(8) 0.0512 (0.342) 3,494 0.2862

(9) 0.0809 (0.074) 4520 0.1581

Table 3. DID results for acquirers

Notes: The results are of the one- or three-year DID of acquirers with some control variables. Heteroscedasticity-corrected p-values are in parenthesis. The symbols ***, **, and * denote statistical significance at the 1%, 5%, and 10% level, respectively

7

The target entities by industry distribution are banks (178/563), securities (91/563), consumer credit (83/563), and other financial industries (80/563). The number entities having industry code as No. 1000, No. 2000, No. 3000, No. 4000, No. 5000, No. 7000, and No. 8000 are 12, 22, 34, 10, 8, 34, and 10. 9

*

ACCEPTED MANUSCRIPT 5.2.

Characteristics of Asian countries: DID

The goal of this section is to examine the acquirer’s effects, by adding the differences in a country’s characteristics between the acquirers and target countries, including cross-border cases. First, we check the difference of the acquirers’ and targets’ legal systems. The English legal system, with its common law origin, provides investors with the strongest legal protection, while the French legal system, with its civilian law origin, provides the least protection. 8

CR IP T

Second, we check the difference between the countries’ degrees of economic freedom based on the EFW index of the targets. Third, we check the impacts of regulatory barriers on the targets. We focus on the following three regulation systems: barriers concerning bank scope restrictions, entry into banking

AN US

requirements for foreign banks, and private monitoring.

Moreover, in relation to the changes occurring after one year, we observe certain changes there are some differences in some systems, as shown in Panel A of Table 4. 9 Surprisingly, all market-to-book ratios, categorized as the “Same” are positively significant. The same systems of law, EFW values, and

M

financial regulations promote high quality in banks before acquisitions. Further, all of the total costs

ED

categorized as “Different”, “Under” and “Above”, are positively significant. In different social systems, acquirer banks incur more costs.

PT

Despite the completion of three years since acquisition (as shown in Panel B in Table 4), it is

CE

disappointing to observe that the market-to-book ratio is insignificant; however, all the NPLs that are categorized as the “Same” are positively significant.

AC

Here, contrarily, we argue the effects among the countries with “Different” systems. For example,

concerning the legal systems, in the English legal system, acquirer banks have the positively hi ghest coefficient of total capital. The strong investor protection found in the English common law, promote banks to be more adequately capitalized before acquisitions. This result is consistent with that of Burkart

8

Although Fauver et al. (2003) empirically show that the French legal system (civilian law system) has greater magnitude when compared to the English legal system (common law system), Suzuki (2012) proposes that M&A premiums in common law countries, such as Australia, India, Malaysia, and Singapore, are higher than in countries that do not use common law. 9 We do not report the results of all dependent variables because they are not significant to this study. 10

ACCEPTED MANUSCRIPT

et al. (2014), who find that strong investor protections increase the external funding capacity of acquirers. Subsequently, we compare the results among the three financial regulations. In the “ Different; Above Mean” category, strong banking regulations enable banks to decrease costs and become more

AC

CE

PT

ED

M

AN US

CR IP T

adequately capitalized, whereas weak regulations increase costs and cause banks to become unsound.

11

Panel A) after one year

Panel B) after three years

Others law origin EFW

Same Under Mean Above Mean

Scope regulation

Same Under Mean Above Mean

Entry regulation

Same Under Mean Above Mean

Self-monitoring regulation

Same Under Mean Above Mean

Δmarket-to-b ook

(1) 0.0392 (0.024) 0.1419 (0.038) -0.0619 (0.614) 0.0997 (0.507) (4) 0.0403 (0.021) -0.1161 (0.390) 0.1898 (0.165) (7) 0.0394 (0.023) 0.2407 (0.085) -0.0367 (0.780) (10) 0.0403 (0.021) 0.1914 (0.116) -0.1106 (0.506) (13) 0.0398 (0.022) -0.1762 (0.260) 0.1779 (0.173)

(2) 0.0084 (0.638) -0.0692 (0.505) 0.0672 (0.651) 0.3341 (0.001) (5) 0.0091 (0.610) 0.4377 (0.014) -0.2158 (0.177) (8) 0.009 (0.613) -0.1527 (0.371) 0.4396 (0.004) (11) 0.011 (0.535) -0.2052 (0.240) 0.3744 (0.024) (14) 0.0109 (0.540) 0.1583 (0.376) -0.2835 (0.080)

(3) 0.0089 (0.607) 0.0677 (0.423) 0.2339 (0.008) -0.0713 (0.378) (6) 0.01 (0.560) 0.2066 (0.058) -0.1796 (0.158) (9) 0.0111 (0.519) -0.2154 (0.066) 0.1487 (0.247) (12) 0.0106 (0.534) -0.2243 (0.059) 0.1254 (0.349) (15) 0.0103 (0.546) 0.2449 (0.038) -0.1726 (0.155)

(1) 0.0227 (0.423) 0.0925 (0.442) -0.2815 (0.119) -0.0343 (0.895) (6) 0.021 (0.457) 0.04 (0.820) -0.2166 (0.013) (11) 0.0201 (0.474) -0.0833 (0.707) 0.1691 (0.400) (17) 0.0204 (0.469) -0.1701 (0.421) 0.06 (0.777) (22) 0.0201 (0.474) 0.2184 (0.277) -0.1314 (0.418)

** **

**

** *

**

**

***

**

***

**

*

***

*

*

Δthe other operational income ratio -

Δtotal loans

ΔNPLs

(2) -0.0047 (0.700) 0.034 (0.648) 0.0345 (0.613) 0.1378 (0.046) (7) -0.0049 (0.688) 0.0974 (0.087) -0.0781 (0.095) (13) -0.0046 (0.704) 0.028 (0.721) -0.0486 (0.602) (18) -0.0046 (0.702) 0.0292 (0.719) -0.045 (0.616) (23) -0.0047 (0.696) 0.0167 (0.764) 0.012 (0.829)

(3) 0.2509 (0.008) -0.5754 (0.455) 0.2398 (0.792) 0.6813 (0.452) (8) 0.2602 (0.006) -1.0187 (0.081) 0.5241 (0.365) (14) 0.256 (0.006) 1.4834 (0.073) -1.2744 (0.143) (19) 0.2557 (0.007) 0.9677 (0.239) -2.1032 (0.010) (24) 0.2542 (0.007) 0.7852 (0.192) 0.2226 (0.725)

AN US

French law origin

Δtotal capital

M

English law origin

Δtotal costs

ED

Same

Δmarket-to-book

PT

Dependent variable Legal law

CR IP T

ACCEPTED MANUSCRIPT

*

**

-

**

-

(12) 0.0012 (0.384) -0.0092 (0.251) 0.0198 (0.041)

**

-

Table 4. DID results with country characteristics

**

* *

***

*** *

*** *

***

*** ***

Δtotal costs

Δtotal capital

(4) 0.0166 (0.657) -0.2277 (0.388) 0.2631 (0.311) 0.4213 (0.167) (9) 0.0005 (0.990) -0.3909 (0.348) 0.3009 (0.238) (15) 0.0007 (0.985) 1.0477 (0.055) -1.2293 (0.094) (20) 0.0013 (0.973) 0.9977 (0.066) -1.2781 (0.082) (25) -0.0005 (0.989) -0.0417 (0.810) 0.2509 (0.389)

(5) 0.0542 (0.099) 0.2943 (0.001) 0.1371 (0.310) -0.1918 (0.165) (10) 0.055 (0.091) 0.2636 (0.074) -0.001 (0.994) (16) 0.0561 (0.085) -0.3235 (0.047) 0.4334 (0.020) (21) 0.0556 (0.088) -0.2782 (0.089) 0.4848 (0.010) (26) 0.0561 (0.085) 0.2541 (0.071) -0.1062 (0.396)

* *

* *

AC

CE

Notes: The results are of the one- or three-year DID of acquirers with some control variables. Heteroscedasticity-corrected p-values are in parenthesis. The symbols ***, **, and * denote statistical significance at the 1%, 5%, and 10% level, respectively.

12

* ***

* *

* ** ** * * *** * *

ACCEPTED MANUSCRIPT Additionally, “Above Mean” regulation promotes banks’ business diversificatio n. Surprisingly, in the “Different; Above Mean” category, the strong regulations concerning entry into banking requirements enable banks to eliminate NPLs and become adequately capitalized. Conversely, the strong regulations concerning private monitoring a re not significant. Thus, we suggest that self-disclosure regulations are not effective in the Asian financial market.

CR IP T

5.3. Robustness using PSM: Characteristics of Asian countries

In this study, we compute the ATE using the PSM method for robustness. Table 5 shows the results of the ATE from PSM. In relation to a one-year change since acquisition, some of the ATE contribute

AN US

toward an increase in total loans, total capital, and total cost; additionally, after a certain period, these ATE contribute toward an increase in NPLs. Regrettably, although the significance level is 10%, the ATE contributes toward a significant decrease in ROA. We can attribute this to the non-profitability and negative ROA of acquirer banks in Asian bank M&A. Finally, after PSM is analyzed, we check

M

the results using some empirical tests. 10

CE

ATE from PSM: after three year N

PT

ATE from PSM: after one year N

ΔROA (1) -0.0085 (0.102) 2,963 (6) -0.0132 (0.073) 2,855

Δtotal loans (2) 0.0568 *** (0.010) 2,888 (7) 0.0891 *** (0.009) 2,758

ED

Outcome variable

*

ΔNPLs (3) 0.0673 (0.263) 2,519 (8 0.2015 ** (0.043) 2,399

Δtotal costs (4) 0.0533 ** (0.033) 2,564 (9) 0.1680 *** (0.000) 2,468

Δtotal capital (5) 0.0910 * (0.093) 2,960 (10) 0.1497 ** (0.025) 2,817

Table 5. ATE calculated using PSM for acquirers

AC

Notes: The results depict one- and three-year ATE calculated using PSM for acquirers with some control variables. P-values are in parentheses. The symbols ***, **, and * denote statistical significance at the 1%, 5%, and 10% level, respectively.

6.

Conclusion

This study empirically examines the effects of the Asian banks’ M&A, focusing on the long-term changes in banking management strategies for acquirer banks.

10

We check the balanced box charts between the treatment and con trol groups. We can see the overlap conditions after matching (results are not reported). Moreover, we examined the balance test results, compared with the raw data, and matched them using standardized difference and the variance ratios (results are not reported). If the variance ratios are near 1, they are considered good matches.

ACCEPTED MANUSCRIPT In the initial stage, acquirer banks expand, issuing higher grossing loans. after a certain period, acquirer banks accumulate loans, thereby enhancing their equity. However, simultaneously, acquisitions continue to lead banks toward amassing NPLs; these banks also end up incurring more costs and losing their profitability in the long run. Additionally, we consider the country characteristics. While acquirer profitability among the countries using the same legal/economic system increase in the initial stage, acquirer banks become

CR IP T

unprofitable after three years due to the growing burden of NPLs. Contrarily, among the countries using different systems, the countries of the target banks, in particular, have stronger/more stringent legal and financial regulations, and acquirer banks can enjoy higher capital adequacy and lower costs.

AN US

The legal system in Asia with its strong investor protections and stringent financial regulations, plays an important role in resolving these problems.

Most importantly, cross-border M&A between countries with different economic systems could enable Asian banks to operate effectively by undertaking M&A . The impacts of Islamic finance

M

in Asia must be explored further study. The recent tendency of Islamic banking increasingly seeking

ED

diversification into foreign investments might offer interesting opportunities in studying Asian

Acknowledgments

PT

banks’ acquisitions.

AC

CE

The author is grateful to Katsushi Suzuki, Kotaro Inoue, Vadym Volosovych, He Fan, Barbara Casu, Hubert de La Bruslerie, Sascha Kolaric, Serif Aziz Simsir, and Jeffrey Callen, as well as the seminar participants of the 2016 Paris Financial Management Conference, the Finance Seminar of Hitotsubashi University, 2016 Nippon Finance Association, 2016 Japan Finance Association, 2016 Southernwest Finance Association, 2015 Midwest Finance Association , 2015 European Financial Management Association, and 2014 European Financial Management Association . All remaining errors are mine. Funding

This work was supported by the Public Interest Foundation KAMPO 2016 . And the JSPS KAKEN Grant Number is [15K03547].

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN US

CR IP T

References Altunbas, Y., Marques, D., 2008. Mergers and acquisitions and bank performance in Europe: The role of strategic similarities. J. Econ. Bus. 60(3), 204–422. Artikis, P.G., Stanley, S., Staikouras, S., 2008. A practical approach to blend insurance in the banking network. J. Risk Finance. 9(2), 106–124. Baele, L., Oliver, D.J., Rudi, V.V., 2007. Does the stock market value bank diversification? J. Bank. Finance. 31(7), 1999–2023. Barth, J.R., Caprio, G., Levine, R., 2001. The regulation and supervision of banks around the world: A new database. World Bank Policy Research Working Paper 2588. Barth, J.R., Caprio, G., Levine, R., 2004. Bank regulation and supervision: What works best? J. Financial Intermed. 13(2), 205–248. Barth, J.R., Caprio, G., Levine, R., 2008. Bank regulations are changing: For better or worse? World Bank Policy Research Working Paper 4646. Becka, T., Demirguc-Kunt, A., Levine, R., 2003. Law, endowments, and finance. J. Financial Econ. 70 (1), 137–181. Behr, A., Heid, F., 2011. The success of bank mergers revisited: An assessment based on a matching strategy. J. Empir. Finance. 18(1), 117–135. Burkart, M., Gromb, M., Muller, H.M., Panumze, F., 2014. Legal investor protection and takeovers. J. Finance. 69(3), 1129–1165. Caiazza, S., Andrew, C., Alberto, F.P., 2012. What do bank acquirers want? Evidence from Worldwide Bank M&A targets. J. Bank. Finance. 36(9), 2641–2659. Calzolari, G., Loranth, G., 2011. Regulation of multinational banks: A theoretical inquiry. J. Financial Intermed. 20(2), 178–198. De Haas, R., Van Lelyveld, I., 2010. Internal capit al markets and lending by multinational bank subsidiaries. J. Financial Intermed. 19(1), 1–25. Fauver, L., Houston, J., Naranjo, A., 2003. Capital market development, international integration, legal systems, and the value of corporate diversification: A c ross-country analysis. J. Financial Quant. Anal. 38(1), 135–157. Inui, T., Edamura, I., Xiaofei, T., Todo, Y., Hada, S., 2013. Do Chinese firms success the cross border M&A? RIETI Policy Discussion Paper 12-P-005 (In Japanese). Laeven, K.L., Levine, R., 2007. Is there a diversification discount in financial c onglomerates? J. Financial Econ. 85(2), 331–367. La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R.W., 1997. Legal determinants of external finance. J. Finance. 52(3), 1131–1150. Moeller, S.B., Schlingemann, F.P., 2005. Global diversification and bidder gains: A comparison between cross-border and domestic acquisitions. J. Bank. Finance. 29(3), 533–564. Nguyen, L., Wilson, J., 2016. How does bank lending react to a catastrophic weather e vent? Proceeding of World Finance Conference 2016. Ongena, S., Popov, A., Udell, G., 2013. When the cat’s away the mice will play: Does regulation at home affect bank risk taking abroad? J. Financial Econ. 108(3), 727–750. Pana, E., Park, J., Query, T., 2010. The impact of bank mergers on liquidity creation. J. Risk Manag. Financial Inst. 4(1), 74–96. Rossi, S., Volpin, P., 2004. Cross-country determinants of mergers and acquisitions. J. Financial Econ. 74(2), 277–304. Sakai, K., Tsuru, K., Hosono, K., 2009. Merger of credit unions. Kinyu-keizai kenkyu 28, 47–63 (in Japanese). Steigner, T., Sutton N.K., 2011. How does national culture impact internalization benefits in cross-border mergers and acquisitions? Financial Rev. 46(1), 103–125. Suzuki,I., 2012.Features of Cross Border M&A in Asia-Pacific from Empirical Analysis. MARR 218. (in Japanese) Wanke, P., Barros, C,P., Azad, M,A,K., Constantino, D., 2016. The Development of Mozambican Banking Sector and Strategic Fit of Mergers and Acquisitions: A Two -Stage EDA Approach. African

ACCEPTED MANUSCRIPT

AC

CE

PT

ED

M

AN US

CR IP T

Devel. Rev. 28(4), 444–461. Wanke, P., Maredza, A., Gupta, R., 2017. Merger and Acquisitions is South Africa banking: A network DEA model. Res. Intl. Business and Finance 41, 362–376.

ACCEPTED MANUSCRIPT Appendix Strategy

Altunbas and Marques (2008)

Proxy variables used in this paper

1. Earning diversification strategy

(1) Diversity of earnings (2) Off-balance sheet activity

The other operational income ratio = other operational revenue／total assets

2. Risk strategy

(1) Credit risk

Loan loss provision ratio = loan loss provisions ／net interest revenue Non-Performing Loan ratio = non-performing loans／total loans Loan ratio = total loans／total assets Deposit-Loans ratio = total loans／total deposits Total cost ratio = total costs／operating income Total capital ratio= total capital／total asset Tier 1 Capital ratio = Tier 1 capital／risk asset ROA= net income/total asset Size = ln(asset) Market-to-Book = market value of capital/book value of capital

ROA Size Market-to-book

CR IP T

3. Cost controlling strategy 4. Capital adequacy level strategy The others

(2) Loan ratio (3) Deposit activity

AC

CE

PT

ED

M

AN US

Table A.1. Strategy variables for Asian banks