Prior research acknowledges that the determinants, timeliness, and economic implications of banks' provisions for loan losses (PLL) vary across loan types. However, the lack of machine-readable data on PLL by loan type has precluded researchers from incorporating loan type into the evaluation of PLL beyond either controlling for or partitioning the sample on crude proxies for loan portfolio composition. We calculate PLL by loan type as the change in the allowance for loan losses by loan type, which we hand collect from Form 10-K filings, plus net charge-offs by loan type, which we obtain from regulatory filings. Using these data, we show that prior findings that banks exercise discretion over PLL to smooth earnings and increase regulatory capital are driven by commercial loans, a thin slice of banks' loan portfolios, and that commonly used measures of PLL timeliness vary substantially across loan types.

JEL Classifications: G21; G28; M41; M48.

You do not currently have access to this content.