というNBER論文をGorodnichenko＝Coibionらが上げている（ungated(BIS)版［H/T Mostly Ecoomics］）。原題は「Keep Calm and Bank On: Panic-Driven Bank Runs and the Role of Public Communication*1」で、著者はDamiano Sandri（BIS）、Francesco Grigoli（ジョージタウン大）、Yuriy Gorodnichenko（UCバークレー）、Olivier Coibion（テキサス大オースティン校）。
以下はその要旨。
Using a survey with information treatments conducted in the aftermath of SVB’s collapse, we study households’ perspectives on bank stability, the potential for panic-driven bank runs, and the role of public communication. When informed about SVB’s collapse, households become more likely to withdraw deposits, due to both a higher perceived risk of bank failure and higher expected losses on deposits in case of bank failure. Leveraging hypothetical questions and the exogenous variation in beliefs generated by the information treatments, we show that households reallocate deposit withdrawals primarily into other banks and cash, with little passthrough into spending. Information about FDIC insurance and communication about bank stability by the Federal Reserve can reassure depositors, while communication from political leaders only influences their electoral base.
（拙訳）
シリコンバレー銀の破綻後に実施された、情報の処置を伴うサーベイを用いて我々は、銀行の安定性に関する家計の観点、パニックによる銀行取り付け騒ぎの可能性、および公的コミュニケーションの役割を調べた。シリコンバレー銀の破綻を知らされた時の家計は、認識する銀行破綻リスクが高まったことと銀行破綻時の預金の予想損失が高まったことにより、預金を引き出す可能性が高まった。仮想的な質問*2、および、情報の処置*3により生じた考えの外生的な変動を利用して我々は、引き出した預金を家計は主に他の銀行と現金に再配分し、支出にはほとんど振り向けないことを示す。FDICの保証に関する情報と、銀行の安定性に関するFRBのコミュニケーションは預金者を安心させることができる半面、政治指導者のコミュニケーションは有権者の支持基盤にのみ影響する。
*1:タイトルの元ネタはおそらく「Keep Calm and Carry On - Wikipedia」。
*2:本文では「Before receiving the information treatments, survey participants were presented with a hypothetical scenario asking how they would react if their bank faced an imminent risk of failing. More precisely, participants were asked regarding their inclination to take deposits out and the share of deposits they would withdraw if their bank faced a certain probability of failing within 3 months. Additionally, participants were asked whether they would start using a new bank and how they would allocate the funds withdrawn. The probability of failure was randomized across participants, taking values between 1, 5, 10, 15, 20, 25, and 50 percent. 」と説明されている。
*3:本文では「Participants were then provided with the information treatments on standalone online screens. The four treatment groups were provided with one of these statements:
A. Considering that a few weeks ago, Silicon Valley Bank (SVB), a U.S.$200bn bank, failed after experiencing a sudden bank run,
B. The FDIC (Federal Deposit Insurance Corporation) is an independent agency of the United States government that protects bank depositors if a bank fails. Considering that the FDIC insures individually owned deposits up to $250,000,
C. Considering that a few weeks ago, President Biden declared that “Americans can have confidence that the banking system is safe,”
D. Considering that a few weeks ago, the Federal Reserve (Fed) declared that “the U.S. banking system is sound and resilient,”
followed by this sentence to alert people that they would be asked again about their views on bank and deposit risk:
we would like to ask you again about your perceptions that your bank may fail and your propensity to take out your bank deposits.
To keep the structure of the survey fully symmetric across the treated and control groups, people in the control group were also presented with a standalone online screen which displayed only the last sentence above.」と説明されている。
