Bank Dividends and Signaling to Information-Sensitive Depositors (original) (raw)

Abstract

This study investigates whether banks use dividends to signal asset quality and liquidity to their debtholders. We exploit an exogenous shock to the asset opaqueness and perception of risks of Brazilian banks caused by the global financial turmoil of 2008. Our empirical identification takes advantage of the cross-sectional heterogeneity of types of depositors in Brazilian banks and the existence of several owner-managed banks (for which shareholder-targeted signaling is implausible) to identify that information-sensitive depositors (institutional investors) are targets of dividend signaling by banks. These costly signaling efforts are particularly strong during financial crises when asset opaqueness, informational asymmetry and depositors' concerns regarding bank liquidity are exacerbated. From a policy perspective, our results favor the imposition of limits on bank dividends during financial crises, because the banks' need to signal their financial health through dividends during crises intensifies the pro-cyclical effects of bank capital on lending. Ó 2015 Elsevier B.V. All rights reserved. banks (e.g., Bessler and Nohel, 2000) and show that stock prices react to dividend information. This study investigates whether dividends are used by managers to convey information to debtholders. More specifically, we use the exogenous shock to the opaqueness of assets and perception of risk of Brazilian banks arising from the financial turmoil that followed Lehman Brother's demise to investigate whether banks use dividends as a signal to information-sensitive depositors. Kauko's (2012) model relates bank dividends to funding stability. In this model, dividends are an important source of information for depositors because they signal both profitability and liquidity (i.e., liquid and profitable banks can pay larger dividends than illiquid and unprofitable banks). Depositors are particularly sensitive to bank liquidity during financial crises because of the potential negative effects of bank runs and fire sales. Therefore, banks may decide to increase their dividends to keep depositors calm and to prevent bank runs during periods of financial turmoil. Although Kauko's (2012) model assumes uniform depositors (i.e., all depositors are equally sensitive to information), recent theoretical models and empirical evidence (Huang and Ratnovski, 2011; Oliveira et al., 2015) suggest that wholesale financiers, such as institutional depositors, are both more prone to engaging in runs during periods with high informational asymmetry (asset opaqueness) and more sensitive to information (such as dividend payments) than retail depositors. Ben-David et al. (2012) argue that institutional investors are more reactive to information than other investors because they have internal risk management systems and funding

Figures (8)

This table presents descriptive statistics of the dependent variables. Panel A shows the average and standard deviations (in parentheses) of payout ratios by bank type over the years in the sample and the proportions of observations in which the payout ratio is equal to 0, between 0% and 40% or greater than 40% by bank type. Panel B shows the total amount of each form of payout (dividends, interest on equity and share repurchases) across the years. differential effect of institutional investors during the crisis). The 2008 crisis is an exogenous shock to perceptions of risk and asset opaqueness of Brazilian banks. Therefore we argue that the addi- tional partial effect of reliance on institutional investors on divi- dend payouts during the crisis (83) has a causal interpretation and is not subject to endogeneity or selection bias issues.'° In other words, the informational content of dividends during the crisis is lar- ger than in normal times because of an exogenous shock. This com- bination is stronger than any instrument. A comprehensive set of robustness checks also ensures the causal interpretation of our findings. deposits of institutional investors. In this case, a positive ~3 may reflect shareholder-targeted signaling during the crisis. To further investigate the debtholder-signaling hypothesis and avoid any con ban founding effects, we examine closely held banks, because such ks have no reason to engage into a costly signaling effort to shareholders. Therefore, if such banks are engaging in signaling, it is directed toward debtholders. Second, closely held Brazilian ban ks can easily circumvent the legal minimum dividend require- ments, whereas publicly traded banks cannot. These combined fea- tures allow us to conclude that dividends that are paid by closely held banks are not due to legal requirements nor they are a signal tos hareholders.'!

This table presents descriptive statistics of the dependent variables. Panel A shows the average and standard deviations (in parentheses) of payout ratios by bank type over the years in the sample and the proportions of observations in which the payout ratio is equal to 0, between 0% and 40% or greater than 40% by bank type. Panel B shows the total amount of each form of payout (dividends, interest on equity and share repurchases) across the years. differential effect of institutional investors during the crisis). The 2008 crisis is an exogenous shock to perceptions of risk and asset opaqueness of Brazilian banks. Therefore we argue that the addi- tional partial effect of reliance on institutional investors on divi- dend payouts during the crisis (83) has a causal interpretation and is not subject to endogeneity or selection bias issues.'° In other words, the informational content of dividends during the crisis is lar- ger than in normal times because of an exogenous shock. This com- bination is stronger than any instrument. A comprehensive set of robustness checks also ensures the causal interpretation of our findings. deposits of institutional investors. In this case, a positive ~3 may reflect shareholder-targeted signaling during the crisis. To further investigate the debtholder-signaling hypothesis and avoid any con ban founding effects, we examine closely held banks, because such ks have no reason to engage into a costly signaling effort to shareholders. Therefore, if such banks are engaging in signaling, it is directed toward debtholders. Second, closely held Brazilian ban ks can easily circumvent the legal minimum dividend require- ments, whereas publicly traded banks cannot. These combined fea- tures allow us to conclude that dividends that are paid by closely held banks are not due to legal requirements nor they are a signal tos hareholders.'!

Fig. 1. Evolution of the payout ratio. The dashed line shows the evolution of the average payout ratio for banks that have not issued any certificates of deposits (CDs) to institutional investors as of December 2007, and the solid line shows the average payout ratio for banks that have issued any amount of CDs to institutional investors as of the same date. Source: Authors.

Fig. 1. Evolution of the payout ratio. The dashed line shows the evolution of the average payout ratio for banks that have not issued any certificates of deposits (CDs) to institutional investors as of December 2007, and the solid line shows the average payout ratio for banks that have issued any amount of CDs to institutional investors as of the same date. Source: Authors.

issued to institutional investors as a percentage of total CDs issued, ROA is return on assets, size is total assets in BRL millions, leverage is liabilities divided by equity, capital adequacy ratio is regulatory capital divided by risk-weighted assets (as provided by the Central bank of Brazil), credit risk is nonperforming loans divided by total loans, and credit growth is the growth rate of the loan portfolio.

issued to institutional investors as a percentage of total CDs issued, ROA is return on assets, size is total assets in BRL millions, leverage is liabilities divided by equity, capital adequacy ratio is regulatory capital divided by risk-weighted assets (as provided by the Central bank of Brazil), credit risk is nonperforming loans divided by total loans, and credit growth is the growth rate of the loan portfolio.

This table presents the regression results for the estimation of the main model described in Section 3. The dependent variable is the payout ratio, defined as total payouts divided by earnings, and the covariates are defined in Table 1. Column (1) uses a pooled Tobit estimation, and column (2) uses instrumental variables to address the potential endogeneity of institutional investors. The instruments are the variable big bank and the lagged values of institutional investors. In columns (3) and (4), we use pooled Tobit estimations, and the sample is restricted to closely held banks and other bank types, respectively. Columns (5) and (6) are identical to columns (1) and (2), respectively, except that they use a placebo crisis dummy. All models have year dummies. The absolute values of the t- statistics of the coefficients of the covariates are shown in parentheses. The symbols *, ** and *** indicate statistical significance at the 10%, 5% and 1% levels, respectively. Table 4

This table presents the regression results for the estimation of the main model described in Section 3. The dependent variable is the payout ratio, defined as total payouts divided by earnings, and the covariates are defined in Table 1. Column (1) uses a pooled Tobit estimation, and column (2) uses instrumental variables to address the potential endogeneity of institutional investors. The instruments are the variable big bank and the lagged values of institutional investors. In columns (3) and (4), we use pooled Tobit estimations, and the sample is restricted to closely held banks and other bank types, respectively. Columns (5) and (6) are identical to columns (1) and (2), respectively, except that they use a placebo crisis dummy. All models have year dummies. The absolute values of the t- statistics of the coefficients of the covariates are shown in parentheses. The symbols *, ** and *** indicate statistical significance at the 10%, 5% and 1% levels, respectively. Table 4

The dependent variable is the payout ratio, and the covariates are defined in Table 1. Columns (3)-(5) present the results of Probit regressions. The dependent variable, which is a dummy indicating a large payout ratio, is equal to 1 when the payout ratio is greater than or equal to 40% and 0 otherwise. Column (3) uses a standard Probit regression, and column (4) uses instrumental variables to address the potential endogeneity of institutional investors. The instruments are the vari- able big bank and the lagged values of institutional investors. Column (5) presents results for the same estimation from column (3) but with the sample restricted to closely held banks. All models have year dummies. The absolute values of the t- statistics of the coefficients of the covariates are shown in parentheses. The symbols *, and *** indicate statistical significance at 10%, 5% and 1% levels, respectively. Robustness checks. Columns (1) and (2) present the results of re-estimations of the main model, excluding the interaction term crisis x institutional investors, using Tobit and IV-Tobit specifications, respectively.

The dependent variable is the payout ratio, and the covariates are defined in Table 1. Columns (3)-(5) present the results of Probit regressions. The dependent variable, which is a dummy indicating a large payout ratio, is equal to 1 when the payout ratio is greater than or equal to 40% and 0 otherwise. Column (3) uses a standard Probit regression, and column (4) uses instrumental variables to address the potential endogeneity of institutional investors. The instruments are the vari- able big bank and the lagged values of institutional investors. Column (5) presents results for the same estimation from column (3) but with the sample restricted to closely held banks. All models have year dummies. The absolute values of the t- statistics of the coefficients of the covariates are shown in parentheses. The symbols *, and *** indicate statistical significance at 10%, 5% and 1% levels, respectively. Robustness checks. Columns (1) and (2) present the results of re-estimations of the main model, excluding the interaction term crisis x institutional investors, using Tobit and IV-Tobit specifications, respectively.

The dependent variable in columns (1) and (2) is the ratio of the value of CDs held by institutional investors to the total value of CDs issued by banks. Column (1) presents the results for a fixed effects static panel, whereas column (2) presents results for a fixed effects dynamic panel data. All models have dummies for years. The t-statistics of the coefficients of the independent variables are shown in parentheses. The symbols *, ** and *** indicate statistical significance at the 10%, 5% and 1% levels, respectively. This table shows the factors that affect the selection of banks by institutional investors.

The dependent variable in columns (1) and (2) is the ratio of the value of CDs held by institutional investors to the total value of CDs issued by banks. Column (1) presents the results for a fixed effects static panel, whereas column (2) presents results for a fixed effects dynamic panel data. All models have dummies for years. The t-statistics of the coefficients of the independent variables are shown in parentheses. The symbols *, ** and *** indicate statistical significance at the 10%, 5% and 1% levels, respectively. This table shows the factors that affect the selection of banks by institutional investors.

The covariates are defined in Table 1. Columns (1), (2) and (5) use the full sample, and columns (3), (4) and (6) use the subsample containing only closely held banks. Columns (1), (3), (5) and (6) use POLS estimations, and columns (2) and (4) use instrumental variables to address the potential endogeneity of institutional investors. The instruments are the variable big bank and the lagged values of institutional investors. Columns (5) and (6) use a placebo crisis dummy. All models have year dummies. The absolute values of the t-statistics of the coefficients of the covariates are shown in parentheses. The symbols *, ** and *™ indicate statistical significance at the 10%, 5% and 1% levels, respectively. Robustness checks. This table presents the regression results for the main model with the percent change in payout as the dependent variable.

The covariates are defined in Table 1. Columns (1), (2) and (5) use the full sample, and columns (3), (4) and (6) use the subsample containing only closely held banks. Columns (1), (3), (5) and (6) use POLS estimations, and columns (2) and (4) use instrumental variables to address the potential endogeneity of institutional investors. The instruments are the variable big bank and the lagged values of institutional investors. Columns (5) and (6) use a placebo crisis dummy. All models have year dummies. The absolute values of the t-statistics of the coefficients of the covariates are shown in parentheses. The symbols *, ** and *™ indicate statistical significance at the 10%, 5% and 1% levels, respectively. Robustness checks. This table presents the regression results for the main model with the percent change in payout as the dependent variable.

Loading...

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

References (30)

  1. Acharya, V.V., Le, H., Shin, H.S., 2013. Bank capital and dividend externalities, working paper. Available at: http://pages.stern.nyu.edu/\~sternfin/vacharya/.
  2. Aït-Sahalia, Y., Andritzky, J., Jobst, A., Nowak, S., Tamirisa, N., 2012. Market response to policy initiatives during the global financial crisis. Journal of International Economics 87, 162-177.
  3. Barclay, M.J., Smith Jr., C.W., Watts, R.L., 1995. The determinants of corporate leverage and dividend policies. Journal of Applied Corporate Finance 7 (4), 4-19.
  4. Ben-David, I., Franzoni, F., Moussawi, R., 2012. Hedge fund stock trading in the financial crisis of 2007-2009. Review of Financial Studies 25 (1), 1-54.
  5. Bessler, W., Nohel, T., 1996. The stock-market reaction to dividend cuts and omissions by commercial banks. Journal of Banking and Finance 20 (9), 1485- 1508.
  6. Bessler, W., Nohel, T., 2000. Asymmetric information, dividend reductions, and contagion effects in bank stock returns. Journal of Banking and Finance 24 (11), 1831-1848.
  7. Bhattacharya, S., 1979. Imperfect information, dividend policy, and 'the bird in the hand' fallacy. Bell Journal of Economics 10 (1), 259-270.
  8. Boldin, R., Leggett, K., 1995. Bank dividend policy as a signal of bank quality. Financial Services Review 4 (1), 1-8.
  9. Boulton, T.J., Braga-Alves, M.V., Shastri, K., 2012. Payout policy in Brazil: dividends versus interest on equity. Journal of Corporate Finance 18 (4), 968-979.
  10. Brav, A., Graham, J.L., Harvey, C.A., Michaely, R., 2005. Payout policy in the 21 st century. Journal of Financial Economics 77 (3), 483-527.
  11. Casey, K.M., Dickens, R.N., 2000. The effects of tax and regulatory changes on commercial bank dividend policy. The Quarterly Review of Economics and Finance 40, 279-293.
  12. Dickens, R.N., Casey, K.M., Newman, J.A., 2002. Bank dividend policy: explanatory factors. Quarterly Journal of Business and Economics 41, 3-12.
  13. Easterbrook, F.H., 1984. Two agency-cost explanations of dividends. The American Economic Review 7 (4), 650-659.
  14. Fatemi, A., Bildik, R., 2012. Yes, dividends are disappearing: worldwide evidence. Journal of Banking & Finance 36, 662-677.
  15. Huang, R., Ratnovski, L., 2011. The dark side of bank wholesale funding. Journal of Financial Intermediation 20 (2), 248-263.
  16. Kashyap, A., Stein, J., 2004. Cyclical implications of the Basel II capital standards. Federal Reserve Bank of Chicago Economic Perspectives 28, 18-31.
  17. Kauko, K., 2012. Why is equity capital expensive for banks? Bank of Finland working paper.
  18. Koch, A.S., Amy, X.S., 2004. Dividend changes and the persistence of past earnings changes. The Journal of Finance 59 (5), 2093-2116.
  19. Leary, M.T., Michaely, R., 2011. Determinants of dividend smoothing: empirical evidence. Review of Financial Studies 24 (10), 3197-3249.
  20. Martins, T.C., Novaes, W., 2012. Mandatory dividend rules: do they make it harder for firms to invest? Journal of Corporate Finance 18 (4), 953-967.
  21. Mayne, L.S., 1980. Bank holding company characteristics and the upstreaming of bank funds. Journal of Money, Credit and Banking 12 (2), 209-214, Part 1.
  22. Michaely, R., Roberts, M., 2012. Corporate dividend policies: lessons from private firms. Review of Financial Studies 25 (3), 711-746.
  23. Michaely, R., Thaler, R.H., Womack, K.L., 1995. Price reactions to dividend initiations and omissions: overreaction or drift? Journal of Finance 50 (2), 573-608.
  24. Miller, M., Modigliani, F., 1961. Dividend policy, growth and the valuation of shares. Journal of Business 34, 411-433.
  25. Miller, M., Rock, K., 1985. Dividend policy under asymmetric information. The Journal of Finance 40 (4), 1031-1051.
  26. Nissim, D., Ziv, A., 2001. Dividend changes and future profitability. The Journal of Finance 56 (6), 2111-2133.
  27. Oliveira, R.F., Schiozer, R.F., Barros, L.A.B.C., 2015. Depositors perception of ''too-big- to-fail''. Review of Finance 19 (1), 191-227.
  28. Renneboog, L., Trojanowski, G., 2011. Patterns in payout policy and payout channel choice. Journal of Banking & Finance 35 (6), 1477-1490.
  29. Rozeff, M.S., 1982. Growth, beta, and agency costs as determinants of dividend payout ratios. The Journal of Financial Research 5, 249-259.
  30. Sant, R., Cowan, A.R., 1994. Do dividends signal earnings? The case of omitted dividends. Journal of Banking and Finance 18 (6), 1113-1133.