Loan Spreads and Unexpected Earnings

Abstract

This paper provides a more direct test of the superior information hypothesis of banks and informs a long standing policy debate about whether banks serve a special information role in the economy. I circumvent the self-selection bias that contaminated prior studies by obtaining bank loan contracts from mandatory SEC filings. I find that banks set loan spreads at the loan initiation as if they have anticipated (several months in advance) the sign and magnitude of borrowers’ future earnings news unexpected by the stock market. The sensitivity of loan spreads to unexpected earnings is larger where banks have greater incentives to collect private information, for example, for borrowers with negative earnings shocks, fewer analyst following and income-increasing abnormal accruals. Furthermore, a difference-in-differences design confirms that the sensitivity of loan spreads to unexpected earnings increases after Reg FD, when the disclosure regulation limits analysts’ access to private information while banks are exempted from the regulation. Finally, this paper exploits the timing difference in information availability to differentiate whether the documented association of loan spreads and unexpected earnings captures information advantage or correlated omitted risk factors. Consistent with the information argument but not correlated omitted risk factors, the association between loan spreads and unexpected earnings becomes weaker one quarter forward, when uncertainty gradually resolves; and there is no association between loan spreads and unexpected earnings measured two quarters forward, after more private information is revealed during quarterly earnings announcements and bank loan contracts are filed with the SEC for public access.

Keywords: Banks’ Information Advantage, Unexpected Earnings, Debt Contracting

JEL Classification: G21, M41