banks dividend US.pdf

8/16/2019 banks dividend US.pdf

1/12

Dividend payouts: Evidence from U.S. bank holding companiesin the context of the nancial crisis

José Filipe Abreu a,⁎, Mohamed Azzim Gulamhussen b,1

a Banking Supervision Department, Banco de Portugal, Av. Almirante Reis, 71-5°, 1150-012 Lisbon, Portugalb ISCTE Business School, Instituto Universitário de Lisboa, Av Forças Armadas, 1649-026 Lisbon, Portugal

a r t i c l e i n f o a b s t r a c t

Article history:

Received 31 July 2012

Received in revised form 30 March 2013

Accepted 1 April 2013

Available online 8 April 2013

We study dividend payouts of 462 U.S. bank holding companies before and during the 2007–

09 financial crisis. Fama and French (2001) characteristics (size, profitability and growth

opportunities) explain dividend payouts before and during the financial crisis. The agency cost

hypothesis explains dividend payouts before and during (more pronouncedly) the finan-

cial crisis. The signaling hypothesis explains dividend payouts during the financial crisis.

Regulatory pressure was ineffective in limiting dividend payouts by undercapitalized banks

before the financial crisis. Our findings have implications for corporate finance and governance

theories, and also for the regulatory reforms that are being discussed among policymakers.

© 2013 Elsevier B.V. All rights reserved.

JEL classification:

G21

G28

G35

Keywords:

BanksBank regulation

Dividends

Financial crisis

1. Introduction

Researchers apply corporate finance and governance theories to financial firms on the grounds of the inherent interplay of interests of a wider set of stakeholders (depositors and regulators, as well as shareholders and managers), which make theiragency and governance problems more complex, and the relevance of financial firms for the good functioning and soundness of modern financial systems (see, among others, Anderson and Campbell, 2004; Brook et al., 2000 ). The financial crisis has furtherenhanced the interest in the application of corporate finance and governance theories due to the unique macroeconomic contextand the regulatory shift which is believed to have hit financial firms the most (see, for example, Erkens et al., 2012). We

contribute to this emerging strand in the literature by studying banks' dividend payout decisions before and during the financialcrisis.

Of the several corporate finance and governance issues that are attracting the attention of scholars, dividend policy is receivingsignificant attention, particularly from regulators and investors. The recent proposals to increase oversight of the dividendpayouts by the Federal Reserve Board (FRB, 2011) and the Basel Committee on Banking Supervision (BCBS, 2011) point towardsthe increasing regulatory relevance of banks' dividend payout policy. Forcing banks to plowback their earnings may however havethe unintended consequence of reducing their ability to both signal their future growth prospects to suppliers of debt and equity,and reduce agency conflicts of their managers with dispersed shareholders. Our paper sheds critical light on the tension between

Journal of Corporate Finance 22 (201 3) 54–65

⁎ Corresponding author. Tel.: + 351 213130511; fax: +351 213532591.

E-mail addresses: [email protected] (J.F. Abreu), [email protected] (M.A. Gulamhussen).1 Tel.: +351 210464131; fax: + 351 217964710.

0929-1199/$ –

see front matter © 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.jcorpn.2013.04.001

Contents lists available at SciVerse ScienceDirect

Journal of Corporate Finance

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j c o r p f i n

http://-/?-http://dx.doi.org/10.1016/j.jcorpfin.2013.04.001http://dx.doi.org/10.1016/j.jcorpfin.2013.04.001http://dx.doi.org/10.1016/j.jcorpfin.2013.04.001mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.jcorpfin.2013.04.001http://dx.doi.org/10.1016/j.jcorpfin.2013.04.001http://dx.doi.org/10.1016/j.jcorpfin.2013.04.001http://www.sciencedirect.com/science/journal/09291199http://crossmark.crossref.org/dialog/?doi=10.1016/j.jcorpfin.2013.04.001&domain=pdfhttp://www.sciencedirect.com/science/journal/09291199http://dx.doi.org/10.1016/j.jcorpfin.2013.04.001mailto:[email protected]:[email protected]://dx.doi.org/10.1016/j.jcorpfin.2013.04.001http://-/?-http://-/?-


2/12

the dividend payout decisions (before and) during the financial crisis by explicitly considering the major regulatory shifts thatoccurred during this period.

Although not new,2 the regulatory focus on dividend payouts by banks contrasts with the minimal attention paid to the issuein the literature, especially because empirical tests on dividends commonly exclude financial firms due to the specificity of theirleverage and reporting norms, thus hindering direct comparisons with non-financial firms (Foerster and Sapp, 2005). However,the fact that banks are regulated and supervised raises questions about the extent to which theories developed for non-financialfirms are applicable to financial firms.

Dividend policy in the context of financial firms has been addressed to some extent previously. We summarize the mainstudies in Table 1. For example, Filbeck and Mullineaux (1993), Collins et al. (1994) and Boldin and Leggett (1995) test thesignaling hypothesis and the evidence largely indicates that dividends are used as a signaling mechanism by banks. These studiesdo not account for the influence of regulatory pressure. An exception though is Theis and Dutta (2009) who in their study on theinfluence of the inside ownership on dividend payout control for the level of capitalization of banks. We extend these studies byconsidering the Fama and French (2001) characteristics of dividend payers, and the agency cost hypothesis alongside thepreviously tested signaling hypothesis. In addition, and so that the agency context in which financial firms operate is explicitlyfactored in, we deploy several measures of regulatory pressure based on the minimum capital requirements.

Regulators impose a minimum level of capital and recommend that banks operate with an adequate level of capital above thatminimum (i.e., a capital buffer that protects debtors against losses and, hence, against the possibility of failure).3 In the wake of the 2007–09 financial crisis, regulators were heavily criticized for the inadequate amount of minimum capital required by theirframeworks (see, among others, Allen and Carletti, 2010; Goodhart and Persaud, 2008). To address this inadequacy, the Basel IIIproposal incorporates a more challenging definition of capital and strengthens the capital requirements. Furthermore, the

proposal explicitly requires banks to conserve a certain amount of capital above the regulatory minimum to build buffers in “

goodtimes” that can be used to absorb losses during “bad times”. The capital conservation buffer is specifically defined in the proposaland should be met with common equity. Additionally, national regulators have the discretion to demand larger buffers in periodsof excessive credit growth (the countercyclical buffer). Although banks are allowed to draw on the buffer during periods of stress,plowback of earnings will be imposed if banks fall into the buffer range, and these constraints will increase as their actual capitalratios approach the minimum requirement (BCBS, 2011).

Broadening the scope of the restrictions placed on dividend payout, the Federal Reserve Board also requires large bank holdingcompanies to submit their capital plans to the Federal Reserve on an annual basis. Furthermore, the Board has asked these banksto provide prior notice to the Federal Reserve under certain circumstances before making a capital distribution, which can beoverruled by the Federal Reserve (FRB, 2011). These circumstances include the existence of unresolved supervisory issues, theinability to maintain capital above the minimum requirements, the inappropriateness of the capital plans and distributions, andthe presumption of unsound or illegal practices.

We test four hypotheses in the present study: (i) the applicability of Fama and French's (2001) characteristics of dividend

payers (size, profitability and growth opportunities); (ii) the signaling hypothesis, which states that dividends are used as anindicator of future prospects; (iii) the agency cost hypothesis, which states that dividends counterbalance the increased need formonitoring associated with independent banks; and (iv) the regulatory pressure hypothesis, which states that undercapitalizedbanks tend to retain earnings rather than pay dividends. The analysis covers two distinct macroeconomic environments: beforeand during the financial crisis. The sample includes 462 U.S. bank holding companies and contains 435 observations made beforethe financial crisis (i.e., from 2004 to 2006) and 441 observations made during the financial crisis (i.e., from 2007 to 2009), for atotal of 876 observations. Focusing on the U.S. provides a large dataset, while restricting the sample to bank holding companiesreduces the problems associated with unobserved heterogeneity.

A major finding of our study is that dividend policy depends on the macroeconomic conditions (i.e., before and during thefinancial crisis). The findings indicate that Fama and French's (2001) characteristics of dividend payers can be applied to banks inboth periods. That is, the larger, more profitable and low growth banks pay more dividends in both periods. Evidence alsosupports the agency hypothesis in both periods, although it is stronger during the financial crisis. The signaling hypothesis on theother hand only applies to the period during the financial crisis. Regulatory pressure was ineffective at limiting undercapitalized

banks' dividend payouts before the financial crisis, justifying the implementation of regulatory reforms.We contribute to the evolving body of literature examining corporate financial decisions in the banking industry in severalways. First, the Fama and French (2001) characteristics of dividend payers are extended to the case of banks. Second, previousstudies focus primarily on the Fama and French (2001) characteristics, the agency and signaling determinants of the dividendpayers without controlling for regulatory pressure. Third, the present study assesses the dividend policy during the 2007–09financial crisis, which as an exogenous shock to the economy, provides an interesting setting for the study of corporate financeand governance decisions. Finally, this paper contributes to the regulatory reforms that intend to constrain dividend paymentsunder certain limiting conditions.

2 For example, the 1933 Home Owners' Loan Act required the thrift subsidiaries of holding companies to give notice of dividend distributions 30 days in

advance (Kroszner and Strahan, 1996). More recently, the Federal Deposit Insurance Corporation Improvement Act (FDICIA) of 1991 restricted capital

distributions for banks classied as undercapitalized under the PCA thresholds.3 The Basel Accords promoted the international harmonization of minimum capital requirements, and both the Basel I and II frameworks provide national

supervisors with discretion to dene higher levels of minimum capital. The design of Basel II is clearer on this aspect: Pillar 1 de nes the minimum capitalrequirements, while Pillar 2 explicitly addresses the need for supervisors to assess the adequacy of internal capital while considering all the material risks.

55 J.F. Abreu, M.A. Gulamhussen / Journal of Corporate Finance 22 (2013) 54–65

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


3/12


4/12

The remainder of the paper is organized into four sections. We present the literature review in Section 2. We describe the sampleand the variables in Section 3. We discuss the findings in Section 4, and the main conclusions and policy implications in Section 5.

2. Related literature

The literature on dividend behavior dates back to Lintner's (1956) seminal study in which managers were interviewed abouttheir dividend payout policies. His findings indicated the irrelevance of capital spending, as formalized later by Modigliani and

Miller (1961). The study also found that managers prefer to base their decisions on earning potential rather than on target payoutratios, and that managers avoid making changes in payout ratios if these changes are not sustainable. His study introduced thepartial adjustment hypothesis (the gradual adjustment of dividends to earnings increases), and the permanent earninghypothesis (dividends will increase only when the permanent earnings increase), concluding that temporary earnings will notinfluence dividend payouts (see also Daniels et al., 1997).

A natural corollary of the early evidence from the field is that dividend payouts can reliably signal the quality of the permanentearnings while concurrently decreasing any uncertainty surrounding firm value (see, among others, Bhattacharya, 1979; Dong etal., 2005; Miller and Rock, 1985). If managers misleadingly raise dividends despite low permanent earnings, they will have toreduce dividend payouts in the future, as these payouts may not be sustainable in the long run. Unlike share repurchases, whichare used by managers to signal temporary changes in their firms' earnings or to indicate that their firms are undervalued, changesin dividend payouts provide a reliable signal of the permanent earning potential or quality ( Jagannathan et al., 2000) or the valueof the firms' growth options ( Jensen et al. , 2010).

The discretion to determine the dividend payout can exacerbate the agency conflict between shareholders and managers

( Jensen, 1986), a problem that is particularly severe for banks due to their highly levered capital structures ( John et al., 2010,among others). Managers of firms with dispersed owners may be more difficult to discipline than those of firms with moreconcentrated owners. Large dividend payouts reduce the amount of cash available to managers and, consequently, act as asubstitute for monitoring while reducing agency costs (see, among others, Easterbrook, 1984; Rozeff, 1982). The agencyhypothesis posits that the managers of banks with dispersed owners will pay more dividends to alleviate agency problems.

The fact that banks are regulated and monitored by supervisors raises questions about the extent to which regulatoryoversight or pressure influences dividend payouts (in the presence of signaling and agency). Highly levered capital structures maymean that the banks hold small capital buffers and are therefore, not allowed to distribute dividends. This constraint may inducebanks to retain their earnings to strengthen their capital base and guarantee their compliance with the regulatory minimumcapital requirements. The regulatory pressure hypothesis posits that the pressure associated with holding capital levels near orbelow the minimum requirement will lead banks to plowback earnings to recapitalize themselves.

In Table 1 we summarize the main papers related to dividend policy of bank holding companies. The Table shows that previousstudies either fail to explicitly address the regulatory pressure hypothesis or consider it using only the equity-to-total asset ratio

instead of the regulatory definition of capital. Thus, it cannot be asserted that corporate finance and governance theories areapplicable to financial firms until such an empirical test is conducted. Our study also fills this gap in the literature.

The evidence regarding the Fama and French (2001) characteristics of dividend payers (size, profitability and historicalgrowth opportunities) in the banking sector is mixed. Theis and Dutta (2009) did not find support for the positive relationshipbetween size and dividend payouts in a sample of 99 U.S. bank holding companies.4 Collins et al. (1994) found a statisticallysignificant inverse relationship between growth opportunities and dividend payouts in a sample of 104 U.S. bank holdingcompanies, while Theis and Dutta (2009) did not find a statistically significant relationship between these two elements.5 Theeasy access to alternative sources of capital and the greater stability in earnings, generally also associated with larger banks,should be positively related with dividend payouts, whereas the availability of investment opportunities should be negativelyrelated with dividend payouts. Therefore, our hypothesis is that the Fama and French (2001) characteristics of dividend payersapply to banks (i.e., larger, profitable and low growth banks should exhibit higher payout ratios).

For bank holding companies, the existing evidence largely supports the signaling hypothesis. Boldin and Leggett (1995) foundthat dividend payouts positively impacted external ratings of 207 listed bank holding companies.6 Along a similar line, Filbeck

and Mullineaux (1993) also found that unexpected dividend announcements had positive impacts on valuations of 177 publiclytraded U.S. bank holding companies. Collins et al. (1994) did not f ind a positive relationship between the market-to-book equityratio and dividend payouts in 104 U.S. bank holding companies.7 Our signaling hypothesis is that banks pay out dividends toconvey information to the market about their future growth opportunities.

4 Carow et al. (2004) found a similar non-signicant relationship in 372 U.S. mutual thrifts whereas Dickens et al. (2002) found a positive and signicant

association between size and dividend payouts in 677 U.S. commercial banks.5 In addition to bank holding companies, Casey and Dickens (2000) and Dickens et al. (2002) found similar negative relationship between historical growth

opportunities and dividend payouts in 41 commercial banks and 372 mutual thrifts in the U.S. However, Kroszner and Strahan (1996) did not nd a similar

statistically signicant relationship in 1285 thrifts.6 In a study of 56 listed commercial banks, Bessler and Nohel (1996) found that 81 different dividend reductions negatively inuenced equity valuations. These

reductions had stronger effects on larger banks. In a study of 406 U.S. banking rms' IPOs (117 banks and 289 thrifts), Cornett et al. (2011) used dividend

distributions to determine the underlying motivation driving banks to go public. The researchers concluded that the post-IPO dividend initiation or increase is

predictive of a subsequent acquisition by another bank.7

The combination of these ndings indicates that dividend levels are less important than the signal given by dividend changes. We acknowledge oneanonymous referee for this comment.


http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


5/12

The evidence on the agency hypothesis is also mixed. Filbeck and Mullineaux (1999) reported that dividend announcementsare not related with past financing announcements for U.S. bank holding companies, contradicting the agency argument thatdividends stimulate monitoring by keeping firms in the market for new capital ( Easterbrook, 1984). Filbeck and Mullineaux(1999) attributed this finding to the presence of regulators that reduce the need for market monitoring. To test the agencyhypothesis, Theis and Dutta (2009) examined the percentage of common stock held by insiders, assuming a negative relationshipbetween insider ownership and dividend payouts due to the reduced monitoring costs associated with higher insider ownership.Their findings provide evidence of a non-linear relationship between insider ownership and dividend payouts.8 Our hypothesis isthat banks that are more difficult to monitor will pay out higher dividend to counterbalance the increased need for monitoring.

Previous studies do not explicitly address the regulatory pressure hypothesis (see, for example, Boldin and Leggett, 1995;Carow et al., 2004) or consider it using the equity-to-total asset ratio instead of the regulatory definition of capital (Bessler andNohel, 1996; Casey and Dickens, 2000; Dickens et al., 2002, in the context of commercial banks). The exception is Theis and Dutta(2009), who considered the regulatory definition of capital and found a positive relationship between the level of capital anddividend payouts (i.e., highly levered banks plowed their earnings back to strengthen their capital levels) .9 Our hypothesis is thatbanks facing regulatory pressure (i.e., undercapitalized banks) plowback their earnings and, therefore, curtail their dividend payouts.

3. Sample, variables and descriptive statistics

We collected firm-level data from Bankscope for U.S. listed bank holding companies with minimum total assets of 100 millionUSD, which yielded 462 institutions. Banks that entered bankruptcy during the financial crisis were not considered. Our samplespans two distinct macroeconomic environments: the period before the financial crisis, from 2004 to 2006 (435 observations),

and the period during the financial crisis, from 2007 to 2009 (441 observations). The final sample is an unbalanced panel with atotal of 876 observations.

We used the dividend payout (dividend payout ) as the dependent variable and constructed it by averaging thedividend-to-total asset ratio for each reference period. We used total assets to scale dividends to ensure that the results werenot driven by stock price and earning volatility associated with the financial crisis. We focus on the characteristics of regulardividend payouts rather than on the prediction of the next year's dividend. Therefore, we use an averaging period that is longerthan one year.10 We opt a three-year averaging period for two main reasons. First, this choice avoids the impact of the 2003 taxcut on dividend payouts in the U.S. (see, among others, Brown et al., 2007). Second, a three-year period covers the time span of the entire financial crisis (2007 to 2009).11

Fama and French (2001) identified three common characteristics of dividend payers, which we test in our study: size,profitability and growth opportunities. Large banks are expected to be more difficult to monitor, and more prone to raising capitalin equity markets; therefore a positive relationship between size and dividend payout is expected. We measured bank size (size)through the natural log of the average of total assets for the reference period. Profitable banks are expected to pay out higher

dividends; therefore, a positive relationship between profitability and dividend payout is expected. We measured profitability( profitability) by the average of the net-income-to-total-assets (return on assets) ratio. Banks with high growth opportunities areexpected to plowback their earnings to avoid costly equity and debt financing. We captured this effect through the annualizedrate of growth of total assets throughout the reference period (historical growth).

The signaling hypothesis states that banks with positive future growth opportunities (expected growth) are expected to payout higher dividends to signal their banks' prospects and increase their potential to attract debt and equity financing whenrequired; therefore a positive relationship between expected growth and dividend payout is expected. Conversely, just likehistorical growth, banks with positive future growth opportunities (expected growth) will plowback their earnings to avoid costlydebt and equity financing; therefore a negative relationship is expected between expected growth and dividend payout . Thus, therelationship between expected growth and dividend payout can be positive or negative. We measured expected growth through theratio of market-to-book value of equity at the end of the observation period. The signaling hypothesis cannot be rejected if thecoefficient associated with expected growth is positive and statistically significant.

The agency cost hypothesis states that dividends counterbalance the increased need for monitoring associated with banks

with dispersed shareholders. A high degree of independence is associated with severe conflicts of interest between shareholdersand managers, which exacerbate the agency costs and the consequent need for tighter monitoring; therefore a positiverelationship is expected between independence and dividend payouts. We used the commonly deployed Independence Indicator(independence) developed by Bankscope to capture the effect of agency costs. This indicator classifies the degree of independence

8 Casey and Dickens (2000) and Dickens et al. (2002) also used the percentage of common stock held by insiders. Contrary to Casey and Dickens (2000),

Dickens et al. (2002) found support for the agency cost hypothesis.9 In addition to the context of bank holding companies, Kroszner and Strahan (1996) and Carow et al. (2004) found a positive relationship between the level of

capital and dividend payouts, whereas Bessler and Nohel (1996) and Casey and Dickens (2000) found that the level of capital has no signicant impact on

dividend payouts.10 There is no consensus in the literature on the averaging period. Previously, authors have used seven (Rozeff, 1982), three (Casey and Dickens, 2000) or even

two years (Dempsey and Laber, 1992). Depending on the information available for each bank, Kroszner and Strahan (1996) used averaging periods ranging from

two to ve years.11

We used two years of information in the case of banks with incomplete information for the reference period, a procedure previously used by Kroszner andStrahan (1996). We excluded observations with less than two years of information.


http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


6/12

of firms from their shareholders.12 Based on the data for the end of the period under analysis, the dummy independence assumes avalue of unity for the most independent banks and zero for all the others (banks with block shareholders). The agency costhypothesis cannot be rejected if the coefficient associated with independence is positive and statistically significant.

Since banks are regulated, and major regulatory shifts occurred during our reference period 2004–09, it is necessary to controlfor their influence in our model. The degree of regulatory pressure should capture the differences in the dividend payouts acrossdistinct degrees of capitalization and risk appetites. The assumption is that banks with (risk-weighted) capital ratios below orclose to the minimum requirements will be subject to closer monitoring by their supervisors (see Section 1). Previous studiescaptured the effect of regulatory pressure by deploying the ratio of equity to total assets. However, because regulators closelyfollow the regulatory definition of capital, we measured regulatory pressure (capitalization) as the average of the tier 1 leverageratio (tier 1 capital to assets) during the reference period.13 Lower leverage (i.e., higher values for capitalization) signals strongerfinancial health and is expected to be associated with higher dividend payouts. Therefore, a positive relationship is expectedbetween capitalization and dividend payout .

Additionally, we included a dummy variable f or regulatory pressure based on the capital categories of the Federal DepositInsurance Corporation Improvement Act (FDICIA).14 Section 131 of the FDICIA establishes a system of prompt corrective actionsderived from a classification system that divides banks into five categories: “well capitalized”, “adequately capitalized”,

“undercapitalized”, “substantially undercapitalized” and “critically undercapitalized.” Banks are classified according to thresholdsusing risk-based capital and leverage ratios as the basis. The majority of the banks in our sample were classified as “well capitalized”.We considerthat regulators increase their pressure on banks when banks are approaching the minimum levelsof capital and not onlywhen those levels are breached. Therefore, the banks considered to be subjectto increased regulatory pressure are those not classifiedas “well capitalized” and those currently classified as “well capitalized” but which may be downgraded (i.e., banks that present

leverage or risk-weighted capital ratios close to the limits of adequate capitalization). For the purpose of this variable, the followingthresholds were considered:8% instead of 6% forthe tier 1 risk-weighted capital ratio,and 7% instead of 5% forthe tier 1 leverage ratio.To capture this effect, we included a dummy PCA in our model. This variable assumes a value of unity if a bank does not meet at leastone of these thresholds. In total, the variable assumes a value of unity for 44 observations before the crisis and 73 observationsduringthe crisis. Undercapitalized banks (banks subject to regulatory pressure) are expected to be associated with lower dividend payouts.Therefore, a negative relationship is expected between PCA and dividend payout .

In order to test the hypothesis that undercapitalized banks faced greater regulatory pressure to plowback their earnings thanwell-capitalized banks, we considered the interaction of PCA with profitability. With this variable, the relevance of profitability as adeterminant of banks' dividend payout becomes dependent on the level of capital. That is, the impact of profitability on dividendsis determined for well-capitalized banks by the coefficient associated with the variable profitability and for undercapitalized banksby the sum of the coefficients associated with the variable profitability and profitability ∗ PCA. Undercapitalized banks (bankssubject to regulatory pressure) areexpected to payout less of their earnings as dividends, instead using theearnings for recapitalization.For that reason, a negative relationship is expected between profitability ∗ PCA and dividend payout , and the sum of the coefficients

associated with profitability and profitability ∗

PCA is expected to be lower than the coefficient associated with profitability.As a result, the regulatory pressure hypothesis cannot be rejected if the coefficient associated with capitalization is statisticallysignificant and positive and/or the coefficient associated with PCA is statistically significant and negative and/or the coefficientassociated with profitability ∗ PCA is statistically significant and negative.

We present a summary of these variables in Table 2. Eq. (1) reflects all the considered variables with their respectivehypothesized signs and relationships with the dependent variable.

Dividendpayout ¼ δ0 þ δ1size þ δ2profitability−δ3historicalgrowth δ4expectedgrowth

þ δ5independence þ δ8capitalization−δ9regulatorypressure−δ10profitability

regulatorypressure þ εi;t

ð1Þ

We present the sample summary statistics in Table 3 and the correlation matrix in Table 4. Certain impacts of the financialcrisis are well reflected in the data: namely, the deterioration of profitability, historical growth and expected growth. Although theaverage capitalization level deteriorated only slightly (from a tier 1 leverage ratio of 8.9% to 8.7%), the number of banks facingincreased regulatory pressure rose sharply (from 44 to 73 banks). The dividend payouts also deteriorated during the financial crisis.That is, the payouts decreased from an average of 0.36% of total assets before the crisis to 0.30% of total assets during the crisis.

12 The indicator can assume ve levels: “A” if no shareholder has more than 25% of the direct or total ownership; “B” if one or more shareholders have an

ownership percentage higher than 25%, but none have an ownership percentage (direct or total) higher than 50%; “C” if there is a shareholder with more than 50%

of the total ownership; “D” if there is a shareholder with more than 50% of the direct ownership; and “U” if there is an unknown degree of independence from the

shareholders. The 25% threshold for ownership concentration is used in other studies of corporate governance (see, for example, Andres, 2008).13 We also performed a test with the ratio of equity to total assets. The ndings remained unaltered.14

Federal Deposit Insurance Corporation Improvement Act of 1991 (P.L. 102-242, 105 STAT. 2236). Accessed at http://thomas.loc.gov/cgi-bin/query/F?c102:3:./temp/~c102bLoNJT:e53844.


http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://thomas.loc.gov/cgi-bin/query/F?c102:3:./temp/~c102bLoNJT:e53844http://thomas.loc.gov/cgi-bin/query/F?c102:3:./temp/~c102bLoNJT:e53844http://thomas.loc.gov/cgi-bin/query/F?c102:3:./temp/~c102bLoNJT:e53844http://thomas.loc.gov/cgi-bin/query/F?c102:3:./temp/~c102bLoNJT:e53844http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


7/12

4. Findings and discussion

We report all findings with robust standard errors because White's test indicated the presence of heteroskedasticity in thedata.15 Because the dependent variable (dividend payout ) does not assume negative values, the distribution can be consideredcensored to the left, a situation in which OLS can produce inconsistent estimates ( Wooldridge, 2002). While the issue in thesample is not severe, Eq. (1) was estimated with a TOBIT regression.16

4.1. Baseline

We present the baseline findings of our model in Table 5, considering only capitalization to capture the effect of regulatorymonitoring (column 1), and additionally PCA and its interaction with profitability (column 2) to capture the effect of regulatorypressure on undercapitalized banks.

The findings indicate that the Fama and French (2001) characteristics of dividend payers can be applied to banks: size and profitability are positively and significantly related and historical growth is negatively related to dividend payout , i.e. larger andmore profitable banks paid larger dividend payouts, and banks with low historical growth opportunities also paid more dividends.Thesefindingsare consistent with the findings of previous studies (Casey and Dickens, 2000; Collins et al., 1994;Dickenset al., 2002).

Expected growth is positively and significantly related with dividend payout , which does not allow us to reject the signalinghypothesis. Therefore, our findings support the signaling argument that dividends work as a signal of future growthopportunities, consistent with the findings from Filbeck and Mullineaux (1993) and Bessler and Nohel (1996). Independence ispositively and significantly related to dividend payout , i.e. banks with dispersed shareholders pay out more dividends to balance

15

χ 2

-test statistics of 221.58 (p-value = 0.000) rejects the null hypothesis of homoscedasticity.16 We also estimated our model with OLS and the ndings proved to be qualitatively similar. We do not report the tables for the sake of brevity.

Table 2

Variables.This table describes the selected dependent and explanatory variables.

Variable Used in Definition (as in the present study) Expected effect

Dividend payout (dependent variable) The average of the dividends-to-total asset ratio for the

reference period (%)

n.a.

Size Boldin and Leggett (1995),

Theis and Dutta (2009)

The natural logarithm of the average of total assets for

the reference period

+

Profitability The average of the net income-to-total asset (returns on assets)

ratio for the reference period (%)

+

Historical growth Boldin and Leggett (1995),


The annualized growth rate of total assets for the

reference period (%)

−

Expected growth Collins et al. (1994),


The ratio of the market-to-book value of equity at the end of

the reference period

+/−

Independence a Dummy that assumes the value of unity for the most

independent banks (classified as “A” in the Bankscope

Independent Indicator) and zero for all the others, using end

of the reference period data

+

Capitalization Theis and Dutta (2009) The average of the tier 1 leverage ratio (tier 1 capital/total assets)

for the reference period (%)

+

PCA b Dummy that assumes the value of unity if the bank is not well

capitalized, considering thresholds of 8% for the tier 1

risk-weighted capital ratio (tier 1 capital/risk-weighted assets)

and 7% for the tier 1 leverage ratio (tier 1 capital/total assets)

−

a Previous studies used the percentage of common stock held by insiders (Theis and Dutta, 2009) to test the agency cost hypothesis.b Previous studies considered the role of regulatory pressure only through the level of capitalization.

Table 3

Sample summary statistics.This table presents the summary statistics for the selected dependent and independent variables. The variables are the same as those defined in Table 2.

Variables Full sample: 2004–200 9 Before the financi al cri sis:

2004–2006

During the financial crisis:

2007–2009

# of obs. Mean Std dev # of obs. Mean Std dev # of obs. Mean Std dev

Dividend payout (dependent variable) 876 0.329 0.298 435 0.358 0.296 441 0.300 0.298

Size 876 7.570 1.444 435 7.425 1.436 441 7.713 1.440

Profitability 876 0.541 1.263 435 1.106 0.686 441 −0.016 1.443

Historical growth 876 9.768 14.073 435 12.996 14.670 441 6.584 12.694

Expected growth 876 1.305 0.927 435 1.904 0.780 441 0.714 0.635Independence 876 0.898 0.302 435 0.901 0.299 441 0.896 0.306

Capitalization 876 8.787 2.920 435 8.901 2.492 441 8.676 3.287

PCA 876 0.134 0.340 435 0.101 0.302 441 0.166 0.372


http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


8/12

the increased need for monitoring, which does not allow us to reject the agency cost hypothesis. As a result, and despite thepresence of external regulators in the financial industry, the findings still support the agency argument that dividendscompensate for the need for monitoring, as also observed by Casey and Dickens (2000) and Dickens et al. (2002).

In terms of the variables used to capture the effect of regulatory pressure, as expected capitalization is positively andsignificantly related with dividend payout , i.e. the more levered banks retained their earnings to rebuild their capital buffers, afinding that is consistent with the findings of Kroszner and Strahan (1996), Carow et al. (2004) and Theis and Dutta (2009); but incontradiction with the findings of Bessler and Nohel (1996) and Casey and Dickens (2000). The variable deployed to capture PCAis not significant when considered in isolation (column 2), but its interaction with profitability, profitability ∗ PCA, is negativelyand significantly related to dividend payout , i.e. undercapitalized banks (banks subject to regulatory pressure) are expected toplowback the earnings to build their capital buffers; a result also consistent with the regulatory pressure hypothesis.

4.2. Sample split

The “Chow test”17 for the periods before and during the financial crisis allows us to reject the null hypothesis that theregression coefficients were equal before and during the financial crisis. This finding supports the decision to analyze the twoperiods separately due to the distinct macroeconomic conditions and the major regulatory shifts in the industry (see also Erkenset al., 2012; Fuller and Goldstein, 2011). We present the findings in Table 6.

4.2.1. Evidence before the nancial crisisWe present the findings for the period before the financial crisis in columns 1 and 2 of Table 6. The evidence supports the

applicability of the Fama and French (2001) characteristics of dividend payers to banks: size and profitability are positively relatedto dividend payout , while historical growth is negatively related to dividend payout . The coefficient associated with independence ispositive and significant at the 10% level, which does not allow us to reject the agency cost hypothesis. Expected growth is not

17χ 2-test statistic of 65.76 (p-value = 0.000) rejects the null hypothesis of equal coef cients between the two regressions.

Table 4

Correlation matrix.This table presents the correlations between the selected variables. The variables are the same as those defined in Table 2.

Size Profitability Historical growth Expected growth Independence Capitalization PCA Dividend

payout

Panel A: Full sample (2004– 2009)

Size 1

Profitability 0.032 1

Historical growth −0.090 0.268 1

Expected growth 0.012 0.588 0.158 1

Independence 0.011 0.044 0.065 −0.025 1

Capitalization −0.138 0.501 0.113 0.118 −0.038 1

PCA 0.274 −0.269 −0.170 −0.094 −0.057 −0.358 1

Profitability ∗ regulatory pressure 0.163 0.534 0.216 0.293 0.086 0.101 −0.173

Dividend payout 0.192 0.570 −0.112 0.360 0.066 0.466 −0 .087 1

Panel B: Before the financial crisis (2004– 2006)

Size 1



Expected growth 0.131 0.395 −0.041 1

Independence −0.002 0.044 0.055 0.027 1

Capitalization −0.223 0.627 0.076 0.014 −0.034 1

PCA 0.409 −0.020 −0.064 0.060 0.009 −0.345 1

Profitability ∗ regulatory pressure 0.451 0.077 −0.034 0.182 0.040 −0.325 0.903

Dividend payout 0.247 0.609 −0.305 0.272 0.087 0.314 0.009 1

Panel C: During the financial crisis (2007 – 2009)

Size 1



Expected growth 0.064 0.541 0.093 1

Independence 0.026 0.049 0.076 −0.121 1

Capitalization −0.072 0.526 0.139 0.228 −0.042 1

PCA 0.157 −0.353 −0.239 −0.146 −0.108 −0.363 1

Profitability ∗ regulatory pressure 0.098 0.581 0.303 0.228 0.114 0.238 −0.54

Dividend payout 0.162 0.634 0.050 0.537 0.044 0.584 −0.15 1


http://-/?-http://-/?-http://-/?-


9/12

significantly related to dividend payout , i.e. we do not find supporting evidence for the signaling hypothesis. Capitalization is eithernot significantly (column 1) related, or it is significantly related at the 10% level (column 2) but with a sign contrary to the oneexpected. PCA and its interaction with profitability are not related to dividend payout at a statistically significant meaningful level.Therefore, we do not find supporting evidence for the regulatory pressure hypothesis.

Table 5

Baseline.This table presents the findings for the baseline specification. Column (1) considers only capitalization as a proxy for regulatory pressure and column (2)

additionally considers PCA and profitability ∗ PCA. The variables are the same as those defined in Table 2. The heteroskedasticity-consistent standard errors are

presented in brackets.

Variables (1) (2)

Intercept −0.394⁎⁎⁎ −0.406⁎⁎⁎

(0.006) (0.079)

Size 0.044⁎⁎⁎ 0.045⁎⁎⁎(0.015) (0.006)

Profitability 0.113⁎⁎⁎ 0.136⁎⁎⁎

(0.001) (0.018)

Historical growth −0.006⁎⁎⁎ −0.006⁎⁎⁎

(0.013) (0.001)

Expected growth 0.037⁎⁎⁎ 0.033⁎⁎⁎

(0.029) (0.013)

Independence 0.081⁎⁎⁎ 0.086⁎⁎⁎

(0.006) (0.029)

Capitalization 0.029⁎⁎⁎ 0.027⁎⁎⁎

(0.085) (0.005)

PCA 0.030

0.027

Profitability ∗ PCA −0.070⁎⁎⁎

(0.018)

Number of observations 876⁎ 876Number of banks 462⁎⁎ 462

F test 42.15⁎⁎⁎ 45.27⁎⁎⁎

⁎⁎⁎ Denotes significance at the 1% level.⁎⁎ Denotes significance at the 5% level.⁎ Denotes significance at the 10% level.

Table 6Sample split.This table presents the findings considering separately the period before the financial crisis and the period during the financial crisis. Columns (1) and (3)

consider only capitalization as a proxy for regulatory pressure and columns (2) and (4) additionally consider PCA and profitability ∗ PCA. The variables are the

same as those defined in Table 2. The heteroskedasticity-consistent standard errors are presented in brackets.

Variables Before the financial crisis: 2004–2006 During the financial crisis: 2007–2009

(1) (2) (3) (4)

Intercept −0.034 −0.042 −0.448⁎⁎⁎ −0.469⁎⁎⁎

(0.008) (0.104) (0.007) (0.105)

Size 0.022⁎⁎⁎ 0.028⁎⁎⁎ 0.034⁎⁎⁎ 0.035⁎⁎⁎

(0.029) (0.008) (0.017) (0.007)

Profitability 0.284⁎⁎⁎ 0.298⁎⁎⁎ 0.071⁎⁎⁎ 0.095⁎⁎⁎

(0.001) (0.030) (0.002) (0.021)

Historical growth −0.008⁎⁎⁎ −0.008⁎⁎⁎ −0.003⁎⁎ −0.003⁎

(0.019) (0.001) (0.023) (0.001)Expected growth 0.002 0.005 0.147⁎⁎⁎ 0.137⁎⁎⁎

(0.042) (0.018) (0.034) (0.025)

Independence 0.074⁎ 0.075⁎ 0.094⁎⁎⁎ 0.105⁎⁎⁎

(0.007) (0.043) (0.009) (0.034)

Capitalization −0.008 −0.014⁎ 0.035⁎⁎⁎ 0.034⁎⁎⁎

(0.107) (0.008) (0.116) (0.008)

PCA 0.026 0.035

(0.085) (0.036)

Profitability ⁎ PCA −0.102 −0.063⁎⁎⁎

(0.078) (0.024)

Number of observations 435 435 441 441

Number of banks 435 435 441 441

F test 48.12⁎⁎⁎ 42.54⁎⁎⁎ 23.16⁎⁎⁎ 25.43⁎⁎⁎

⁎⁎⁎ Denotes significance at the 1% level.⁎⁎

Denotes significance at the 5% level.⁎ Denotes significance at the 10% level.


http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


10/12

4.2.2. Evidence during the nancial crisisWe present the findings for the period before the financial crisis in columns 3 and 4 of Table 6. As in the period before the

crisis, the evidence supports the applicability of the Fama and French (2001) characteristics of dividend payers to banks: size and profitability are positively related to dividend payout , while historical growth is negatively related to dividend payout (although theevidence is weaker for historical growth). The coefficient associated with independence is also positive and now it is significant atthe 1% level which does not allow us to reject the agency cost hypothesis.

Unlike the period before the crisis, expected growth is positively related to dividend payout , a finding that does not allow us toreject the signaling hypothesis. Additionally, we also do not reject the regulatory pressure hypothesis: capitalization is positivelyrelated to dividend payout , and profitability ∗ PCA is negatively related to dividend payout .

4.3. Robustness checks

A key feature of our study is the consideration of the macroeconomic setting and the regulatory shifts which, according toErkens et al. (2012), occurred during the financial crisis. We consider that banks with levels of capital close to the minimumrequirement or below that minimum faced additional regulatory pressure, which we define using add-ons on the PCA thresholds.To take into account the subjectivity of this measure, we redefined the variable PCA considering both the actual PCA thresholdsand an add-on of 1 percentage point on those thresholds. In both cases and for the 3 scenarios (full sample, period before thefinancial crisis, period during the financial crisis), the findings remained similar to the baseline .18

In order to further test the robustness of our findings, we considered an alternative specification for capitalization. We used thetier 1 risk-weighted capital ratio (tier 1 capital ratio) instead of the tier 1 leverage ratio (capitalization). The tier 1 capital ratiokeeps the tier 1 capital in the numerator and considers the risk-adjusted assets (RWA) instead of the total assets in thedenominator. We measure tier 1 as the average of the tier 1 capital ratio throughout the reference period, with higher values fortier 1 reflecting a better position in terms of risk-adjusted capital. Therefore, a positive relationship between tier 1 and dividend

payout is expected. For the period during the financial crisis, we also considered another specification for regulatory pressure.Using the data relating to the participation of banks in the U.S. Troubled Assets Relief Program (TARP) ,19 we replaced the variablePCA with a dummy (TARP ) that assumes a value of unity if the bank received public funds from the TARP .20 As for PCA, we alsoconsidered the interaction of TARP with profitability, expecting a negative relationship between both TARP and profitability ∗ TARP ,and dividend payout . We present the findings in Table 7.

For the 3 scenarios considered (full sample, period before the financial crisis, period during the financial crisis), the findingsare consistent with the previous results when we consider tier 1 capital ratio to capture the effect of regulatory monitoring, andalso when we consider PCA and its interaction with profitability. Interestingly, the evidence for the regulatory pressure hypothesisis stronger when using the variable TARP (Table 7, column 7) relatively to the baseline. Both TARP and profitability ∗ TARP arenegatively and significantly related to dividend payout . In the previous definition of regulatory pressure, only the interactionvariable profitability ∗ PCA was significant. This finding indicates the robustness of the support for the regulatory pressurehypothesis during the financial crisis to differing specifications of regulatory pressure.

5. Conclusions

Researchers in corporate finance have often found interest in studying financial firms due to their amplified agency andgovernance problems, and their critical importance for the good functioning of the modern financial system. The 2007–09financial crisis has further enhanced the interest as a result of the unique macroeconomic setting and the regulatory shifts thatoccurred during this period.

We construct a new dataset on 462 U.S. bank holding companies to study the dividend policy in the context of the 2007–09financial crisis. We test the signaling and agency hypotheses alongside the Fama and French (2001) characteristics of dividendpayers controlling for the regulatory shifts during the period under analysis.

Our main findings indicate that dividend payouts depend on the macroeconomic context (before and during the financialcrisis). The Fama and French (2001) characteristics – larger, more profitable and low growth banks tend to pay more dividends –hold for both periods. Interestingly, despite the presence of external regulators in the financial industry, the findings still supportthe agency argument that dividends compensate for the need for monitoring. Our findings also support the signaling argumentthat dividends work as a signal of future growth opportunities, although only during the financial crisis. A possible interpretationis that bank holding companies have little need to signal unless the whole industry is under strain and it is important to beidentified as a better than average bank.21 The controlling regulatory pressure hypothesis that undercapitalized banks plowbackearnings to recapitalize themselves only holds during the financial crisis, a period during in which regulators exerted morepressure on banks with low capital buffers.

18 We do not report the tables for the sake of brevity.19 We collected data from ProPublica, an independent non-prot organization. ProPublica maintains a list of TARP recipients on the following website: http://

projects.propublica.org/bailout/list .20

Because TARP was implemented in October 2008, the variable could only be constructed for the nancial crisis period.21 We acknowledge one anonymous referee for this comment.


http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://projects.propublica.org/bailout/listhttp://projects.propublica.org/bailout/listhttp://projects.propublica.org/bailout/listhttp://projects.propublica.org/bailout/listhttp://projects.propublica.org/bailout/listhttp://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


11/12

Governance problems are often considered to be more severe and complex in financial firms. The 2007–09 financial crisisfurther exposed the governance issues in financial firms creating the pathway for increased regulatory pressure, and theirconsequent implications for performance. Our findings shed light on the dividend payout of bank holding companies in a uniquemacroeconomic setting that spans a period before and during the 2007–09 financial crisis and in which the banking industryforesaw major shifts in the regulatory landscape.

The recent regulatory focus on dividend policy contrasts with the limited attention paid to the issue in past empirical studies. Theregulatory reforms currently being put into place impose a capital conservation mechanism by constraining dividend payouts forbanks whose capital buffers fall within a range close to the minimum requirements. The ineffectiveness of regulatory pressure inlimiting dividend payouts by undercapitalized banks before the financial crisis – a period during which banks were supposed to buildcapital buffers – supports the Federal Reserve and the Basel Committee's initiatives to limit dividend payouts by undercapitalizedbanks. Because our findings provide robust empirical support for the signaling and agency cost hypotheses, the reforms may have anunintended impact on the use of dividends as bothsignaling and agency costreduction mechanisms. Inability to use these governance

mechanisms may reduce the potential to attract external financing, both debt and equity. The level up to which regulators may wantto allow signaling and agency mechanisms to function is an issue that deserves serious attention from academics and regulators alike.

Acknowledgments

We are grateful to the Managing Editor, Jeffry Netter, and the anonymous reviewer, for their comments and suggestions. Weacknowledge the financial support from “Fundação para a Ciência e Tecnologia” (PTDC/EGE-ECO/114977/2009). The views andopinions expressed in this paper are those of the authors and do not necessarily represent those of the institutions with which theauthors are affiliated. Any errors are the responsibility of the authors.

References

Allen, F., Carletti, E., 2010. An overview of the crisis: causes, consequences, and solutions. Int. Rev. Financ. 10 (s1), 1–26.

Anderson, C., Campbell, T., 2004. Corporate governance of Japanese banks. J. Corp. Financ. 10 (3), 327–

354.Andres, C., 2008. Large shareholders and firm performance — an empirical examination of founding-family ownership. J. Corp. Financ. 14 (4), 431–445.

Table 7

Robustness tests.This table presents the findings for the robustness tests on the definition of regulatory pressure considering the full sample, the period before the financial crisis

and the period during the financial crisis. The tier 1 risk-weight capital ratio (tier 1 capital ratio) is used instead of the tier 1 leverage ratio (capitalization). In

column (7), regulatory pressure is considered through the dummy variable TARP , which assumes a value of unity if the bank has received public funds from the

U.S. Troubled Assets Relief Program (TARP). The variables are the same as those defined in Table 2. The heteroskedasticity-consistent standard errors are

presented in brackets.

Variables Full sample: 2004–2009 Before the financial crisis:

2004–2006

During the financial crisis: 2007–2009

(1) (2) (3) (4) (5) (6) (7)

Intercept −0.382⁎⁎⁎ −0.403⁎⁎⁎ −0.187⁎ −0.226⁎⁎ −0.430⁎⁎⁎ −0.445⁎⁎⁎ −0.407⁎⁎⁎

(0.075) (0.073) (0.100) (0.099) (0.092) (0.083) (0.091)

Size 0.041⁎⁎⁎ 0.045⁎⁎⁎ 0.029⁎⁎⁎ 0.035⁎⁎⁎ 0.031⁎⁎⁎ 0.033⁎⁎⁎ 0.034⁎⁎⁎

(0.006) (0.006) (0.008) (0.008) (0.007) (0.007) (0.007)

Profitability 0.094⁎⁎⁎ 0.115⁎⁎⁎ 0.239⁎⁎⁎ 0.242⁎⁎⁎ 0.055⁎⁎⁎ 0.078⁎⁎⁎ 0.073⁎⁎⁎

(0.014) (0.017) (0.029) (0.030) (0.014) (0.017) (0.018)

Historical growth −0.006⁎⁎⁎ −0.005⁎⁎⁎ −0.007⁎⁎⁎ −0.007⁎⁎⁎ −0.003⁎⁎ −0.002⁎ −0.002⁎

(0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)

Expected growth 0.041⁎⁎⁎ 0.038⁎⁎⁎ 0.015 0.018 0.127⁎⁎⁎ 0.119⁎⁎⁎ 0.118⁎⁎⁎

(0.012) (0.012) (0.019) (0.018) (0.024) (0.025) (0.024)

Independence 0.091⁎⁎⁎ 0.095⁎⁎⁎ 0.083⁎ 0.085⁎⁎ 0.100⁎⁎⁎ 0.107⁎⁎⁎ 0.110⁎⁎⁎

(0.030) (0.029) (0.043) (0.043) (0.035) (0.034) (0.036)

Tier 1 capital ratio 0.021⁎⁎⁎ 0.019⁎⁎⁎ 0.003 0.003 0.027⁎⁎⁎ 0.026⁎⁎⁎ 0.025⁎⁎⁎

(0.003) (0.003) (0.004) (0.004) (0.005) (0.004) (0.004)PCA 0.002 0.021 −0.001

(0.024) (0.083) (0.033)

Profitability * PCA −0.066⁎⁎⁎ −0.072 −0.064⁎⁎⁎

(0.017) (0.076) (0.021)

TARP −0.042⁎⁎

(0.021)

Profitability * TARP −0.042⁎⁎

(0.018)

Number of observations 876 876 435 435 441 441 441

Number of banks 462 462 435 435 441 441 441

F test 45.89⁎⁎⁎ 46.96⁎⁎⁎ 35.01⁎⁎⁎ 28.62⁎⁎⁎ 34.17⁎⁎⁎ 38.06⁎⁎⁎ 28.77⁎⁎⁎

⁎⁎⁎ Denotes significance at the 1% level.⁎⁎ Denotes significance at the 5% level.⁎ Denotes significance at the 10% level.


http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


12/12

BCBS, 2011. Basel III: a global regulatory framework for more resilient banks and banking systems. Basel Committee on Banking Supervision (June, Available at:http://www.bis.org/publ/bcbs189.pdf ).

Bessler, W., Nohel, T., 1996. The stock-market reaction to dividend cuts and omissions by commercial banks. J. Bank. Financ. 20 (9), 1485–1508.Bhattacharya, S., 1979. Imperfect information, dividend policy, and ‘the bird in the hand’ fallacy. Bell J. Econ. 10 (1), 259–270.Boldin, R., Leggett, K., 1995. Bank dividend policy as a signal of bank quality. Financ. Serv. Rev. 4 (1), 1–8.Brook, Y., Hendershott, R., Lee, D., 2000. Corporate governance and recent consolidation in the banking industry. J. Corp. Financ. 6 (2), 141–164.Brown, J., Liang, N., Weisbenner, S., 2007. Executive financial incentives and p ayout policy: firm responses to the 2003 dividend tax cut. J. Financ. 62 (4), 1935–1965.Carow, K., Cox, S., Roden, D., 2004. Mutual holding companies: evidence of conflicts of interest through disparate dividends. J. Bank. Financ. 28 (2), 277–298.Casey, K., Dickens, R., 2000. The effects of tax and regulatory changes on commercial bank dividend policy. Q. Rev. Econ. Financ. 40, 279–293.Collins, M., Blackwell, D., Sinkey, J., 1994. Financial innovation, investment opportunities and corporate policy choices for large bank holding companies. Financ.

Rev. 29 (2), 223–247.Cornett, M., Fayman, A., Marcus, A., Tehranian, H., 2011. Dividends, maturity, and acquisitions: evidence from a sample of bank IPOs. Rev. Financ. Econ. 20 (1), 11–21.Daniels, K., Tai, S., Cheng, F., 1997. The information content of dividend hypothesis: a permanent income approach. Int. Rev. Econ. Financ. 6 (1), 77–86.Dempsey,S., Laber, G., 1992. Effectsof agency and transactioncosts on dividendpayoutratios:furtherevidence of the agency-transactioncost hypothesis. J. Financ. Res.

15 (4), 317–320.Dickens, R., Casey, K., Newman, J., 2002. Bank dividend policy: explanatory factors. Q. J. Bus. Econ. 41 (1/2), 3–12.Dong, M., Robinson, C., Veld, C., 2005. Why individual investors want dividends. J. Corp. Financ. 12 (1), 121–158.Easterbrook, F., 1984. Two agency-cost explanations of dividends. Am. Econ. Rev. 74 (4), 650–659.Erkens, D., Hung, M., Matos, P., 2012. Corporate governance in the 2007 – 2008 financial crisis: evidence from financial institutions worldwide. J. Corp. Financ. 18

(2), 389–411.Fama, E., French, K., 2001. Disappearing dividends: changing firm characteristics or lower propensity to pay? J. Financ. Econ. 60, 3–43.Federal Reserve Board (FRB), 2011. Federal Reserve System, 12 CFR Part 225, capital plans. Fed. Regist. 76 (231), 74631 –74648 (Available at http://www.

federalreserve.gov/reportforms/formsreview/RegY13_20111201_ffr.pdf ).Filbeck, G., Mullineaux, D., 1993. Regulatory monitoring and the impact of bank holding company dividend changes on equity returns. Financ. Rev. 28 (3), 403–415.Filbeck, G., Mullineaux, D., 1999. Agency costs and dividend payments: the case of bank holding companies. Q. Rev. Econ. Financ. 39 (3), 409–418.Foerster, S., Sapp, S., 2005. Valuation of financial versus non-financial firms: a global perspective. J. Int. Financ. Mark. Inst. Money 15 (1), 1–20.Fuller, K., Goldstein, M., 2011. Do dividends matter more in declining markets? J. Corp. Financ. 17 (3), 457–473.

Goodhart, C., Persaud, A., 2008. How to avoid the next crash. Financ. Times 30 (January). Jagannathan, M., Stephens, C., Weisbach, M., 2000. Financial flexibil ity and the choice between dividends and stock repurchases. J. Financ. Econ. 57 (3), 355–384. Jensen, M., 1986. Agency costs of free cash flow, corporate fi nance, and takeovers. Am. Econ. Rev. 76 (2), 323–329. Jensen, G., Lundstrum, L., Mil ler, R ., 20 10. What do di vidend reducti ons signal? J. Corp . F inanc. 16 (5), 73 6–747. John, K., Mehran, H., Qian, Y ., 201 0. Ou tside monitoring and CEO compensati on in the banking industry. J. Corp. Financ. 16 (4), 383–399.Kroszner, R., Strahan, P., 1996. Regulatory incentives and the thrift crisis: dividends, mutual-to-stock conversions, and financial distress. J. Financ. 51 (4), 1285–1319.Lintner, J., 1956. Distribution of incomes of corporations among dividends, retained earnings, and taxes. Am. Econ. Rev. 46 (2), 97–113.Miller, M., Rock, K., 1985. Dividend policy under asymmetric information. J. Financ. 40 (4), 1031–1051.Modigliani, F., Miller, M., 1961. Dividend policy, growth, and the valuation of shares. J. Bus. 34 (4), 411–433.Rozeff, M., 1982. Growth, beta and agency costs as determinants of dividend payout ratios. J. Financ. Res. 5 (3), 249–259.Theis, J., Dutta, A., 2009. Explanatory factors of bank dividend policy: revisited. Manag. Financ. 35 (6), 501–508.Wooldridge, J., 2002. Econometric Analysis of Cross Section and Panel Data. MIT Press, Cambridge MA.

http://www.bis.org/publ/bcbs189.pdfhttp://www.federalreserve.gov/reportforms/formsreview/RegY13_20111201_ffr.pdfhttp://www.federalreserve.gov/reportforms/formsreview/RegY13_20111201_ffr.pdfhttp://www.federalreserve.gov/reportforms/formsreview/RegY13_20111201_ffr.pdfhttp://www.federalreserve.gov/reportforms/formsreview/RegY13_20111201_ffr.pdfhttp://www.bis.org/publ/bcbs189.pdf

banks dividend US.pdf

Documents

Transcript of banks dividend US.pdf