Investors became more sensitive to banks' financial weaknesses after their first discount window loans during the March 2023 bank run, research published by the Federal Reserve Bank of New York found.
The authors found no statistically detectable change after banks first borrowed from a new emergency program, although the comparison between facilities was not statistically decisive. The Nov. 17 Liberty Street Economics publication, “Does the Stock Market Stigmatize Banks that Borrow from the Fed?”, examined investors' responses to borrowing during the crisis.
The authors compared borrowers with 207 banks that used neither facility, examining seven trading days before and after each first loan, the central bank said. Between March 8 and March 31, 2023, 34 banks in the sample first borrowed from the Bank Term Funding Program and 33 from the discount window. Ten used both facilities.
The authors tracked how closely daily bank stock returns moved with portfolios reflecting four financial weaknesses: uninsured deposits, losses on securities not yet sold, cash holdings and capital strength. Their measure showed little evidence that investors focused on those risks before the run, but sensitivity rose sharply around its start.
Among banks using only the discount window, investor sensitivity to uninsured deposits and capital rose by a statistically detectable amount four to seven trading days after borrowing, the authors found. Sensitivity to capital was also higher across the full seven-day period after the first loan. For banks using only the emergency program, there was no statistically detectable change, the authors said.
The authors said different loan terms could explain part of the gap. The discount window offered short-term loans against securities valued at market prices. The emergency program offered loans lasting up to a year against eligible securities valued at face value, with borrowing rates typically below the window's.
A bank accepting the less favorable window terms might have exhausted better options or needed to pledge loans that only the window accepted, the authors said. They said investors could interpret that choice as evidence of weakness.
The authors also linked the difference to the window's longstanding reputation as a sign of financial trouble. The emergency program, announced on March 12, 2023, had no such history and addressed the specific problems banks faced that month. Individual borrowers from either facility were disclosed only after two years.
The study covered small groups over a few trading days in one rapidly changing episode, the authors said. They could not fully distinguish the message borrowing sent investors from the reasons banks needed loans. They said the findings suggest crisis-specific programs offering more generous terms and carrying no historical stigma may work better.
The authors are Natalia Fischl-Lanzoni, an MIT FutureTech research assistant and NYU Courant master's student, Martin Hiti, an MIT Sloan doctoral student, and Asani Sarkar, a New York Fed financial research advisor. The publication said their views do not necessarily reflect those of the central bank or the Federal Reserve System.



