Bank manager's discretion over loan loss provision and determinants of signaling.

摘要

The loan loss provision is the expense which represents bank management's estimate of the year's net change in probable loan losses. Bank managers exercise discretion over the timing of recognition of certain loan losses by utilizing their judgement in assessing the probability of future loan losses. Prior studies address four major motivational factors which affect the loan loss provision (LLP) established by management: regulatory capital, income smoothing, signaling, and tax incentives. These studies have documented conflicting empirical evidence for all motivating factors except for tax incentives.;The objectives of this study are: (1) to reexamine three potential motivating factors in bank managers' discretion over their loan loss provisions (LLPs); and (2) to provide new evidence that the strength of LLP signaling depends upon various firm specific characteristics.;The results about managers' discretion over LLPs lead to three observations. First, significant results for the income smoothing hypothesis are robust to alternative research settings. Second, average signaling coefficients estimated from bank-specific regressions are systematically larger than corresponding coefficients of pooled time series cross-sectional regressions and are statistically significant. This supports the main hypothesis of this study that LLP signaling is associated with bank-specific characteristics that affect the degree of information asymmetry. Third, bank managers seem to use LLPs to manage their regulatory capital levels to meet minimum levels specified by regulators. This is consistent with the notion that the appropriate target for regulatory capital management is the minimum specified by bank regulators rather than the time series bank-specific mean capital ratio.;This study provides evidence, by utilizing a bank-specific time series regression approach, that bank managers' signaling through LLP is associated with various bank-specific economic characteristics such as business risk, future investment opportunity, and degree of income smoothing. This study also finds evidence that a bank manager signals simultaneously with discretionary LLP and dividend changes.;This study makes three major contributions to the extant literature. First, given that investors rely upon accounting information in assessing a bank's value, the findings help less informed investors to evaluate the observed discretionary LLP as a credible signal by relating it to firm-specific economic factors that affect the degree of information asymmetry. Second, the study reveals that researchers' conclusions regarding the coefficients on signaling and capital management are likely overstated (understated) depending on the research designs employed in their studies. Finally, this study confirms that a bank manager with an incentive to mitigate asymmetric information can select multiple signals to maximize the purported signaling effects.