Abstract
Term auction facility (TAF) was created during the financial crisis as a substitute for the Federal Reserve’s discount window, the lender of last resort. We hypothesize if TAF borrowing is viewed as a bailout then publicly traded banks would borrow relatively fewer TAF funds to avoid a bailout stigma. We find publicly traded banks did borrow less (as a percent of total assets) in the TAF program than privately held banks. Further, too-big-to-fail banks and investment banks borrowed relatively less than other publicly traded banks indicating greater levels of public scrutiny reduces borrowing under emergency government liquidity programs. We also find that publicly traded banks pledged lower quality and less liquid collateral than private banks when borrowing under the program. Our results suggest TAF provided more benefit to traditional privately held banks with strong balance sheets that were able to borrow relatively greater amounts in anticipation of either future liquidity needs as suggested by Ivashina and Scharfstein (J Financ Econ 97:319–338, 2010) or increased lending as found by Berger et al. (The Federal Reserve’s discount window and TAF programs: “pushing on a string?” Working paper, University of South Carolina, 2014).
Similar content being viewed by others
Notes
We thank an anonymous referee for pointing out that while the reputation of publicly traded banks is easily measured in share prices; privately held banks can suffer from stigma as well. During a crisis, privately held banks have difficulty raising capital and stigma costs could significantly increase equity costs for such banks. That is, while it is easier for these banks to conceal borrowing, it may be more costly if revealed.
More details on the Fed’s financial crisis programs is available at: http://www.federalreserve.gov/monetarypolicy/bst_crisisresponse.htm.
The FAQ from the Fed’s webpage on TAF states TAF will have terms of 28 or 84 days with slight adjustment for holidays. The TAF data show other terms in days of: 13, 17, 35, 42, 70, 83, and 85 days. Clearly, 83 days and 85 days are holiday accommodations. The other maturities are outside the stated maturities. These maturities represent less than 6 % of the loans made under TAF.
Stigum (1990) states there are two situations in which the Fed can provide emergency aid to banks. The first is an act of God—floods, hurricanes etc.—which adversely affects a group of banks, their borrowers or their depositors. The second, is when in the judgement of the Fed, long-term financing is needed to offset risk to the banking system as a whole while a long-term solution is worked out. TAF is this latter type of program.
Blau et al. (2016) point out that despite the promise of anonymity by the Fed for those banks accessing the Fed, the Fed continued to release aggregate information on its lending programs by Federal Reserve District. Hence, knowledgeable market participants were able to identify the largest borrowers leading to statistically and economically significant negative returns for those banks.
Gorton and Metrick (2012) note that securitized banking has historically been the business of investment banks.
For a more complete discussion of PD see Adrian, Burke, and McAndrews (2009).
We do not analyze these events specifically. Instead, we use the events as reference points. While dozens of events may have influenced banks during the crisis, these three appeared to signal major turning points in the market in general. Afonso et al. (2009), Kapercyzk and Schnabl (2010) and Griffiths et al. (2011) all use similar events in their analyses of the money markets during the crisis.
Of the 19 institutions that are stress tested, American Express, GMAC, and MetLife are removed from the sample since they are neither primarily banks nor investment banks. We recognize that regulators stress tested many more banks than these, but these are the ones initially reported to the public and such disclosure implies special status and thus, more scrutiny.
Smaller privately held banks do not have to report the same interest rate risk data that larger banks do, thus many of the privately held banks had missing data when calculating GAP12. Accordingly, we removed this variable from the analysis.
We develop our model from the model Cyree, Griffiths and Winters (2013, 2016) use to examine bank returns relative to participation in the crisis program examined in this paper. These variables, in general, are bank performance and risk measures drawn from the extant literature in general and Cornett et al. (2011) and Delis et al. (2014) in particular.
The market value of some of these banks swings substantially in a matter of a few days. For example, Bank of America trades for $47.63 at the beginning of the sample period, rises to $52.71 on 10/5/2007, falls to $3.14 on 3/6/2009 before recovering to $17.83 by the end of the sample period. Our results are similar if we scale by deposits or other financial statement variables.
We thank an anonymous referee for pointing out that we cannot directly discern if it was a one-to-one substitute in TAF and the discount window.
A search of news releases during this period revealed 12 events related to capital infusions. The most significant of which were efforts by CitiGroup to raise an additional $3 billion, WaMu raising $7 billion and Wachovia planning to raise several billion dollars. Reports drawn from the Factiva Database.
See: http://www.federalreserve.gov/newsevents/reform_taf.htm. Accessed 3/18/2014.
We also use Tier 1 capital-to-assets, Tier 1 risk-adjusted capital-to-assets, and Total risk-based capital-to-assets for robustness with little difference. Our main results and conclusions are not changed.
DWCHGA remained significant in all cases in subsequent two-way (cross-sectional and times-series) models.
The Hausman m-statistic provides information about the appropriateness of the random-effects specification assumes under the null hypothesis, there no correlation between the effects variables and the regressors. Hence, a test can be based on the result that the covariance of an efficient estimator with its difference from an inefficient (OLS) estimator is zero. Rejection of the null hypothesis suggests that the fixed-effects model is more appropriate.
The three most likely causes of endogeneity are omitted variables, simultaneity, and measurement errors. Of these, the most plausible cause of endogeneity for this study is simultaneity since borrowings are possibly jointly determined. The 3SLS model estimated here defines the change in borrowing programs for each program as endogenous and the other independent variables as exogenous.
References
Acharya V, Fleming M, Hrung W, Sarkar A (2012) Dealer financial conditions and lenders-of-last resort facilities. Federal Reserve Bank of New York staff report 673
Adrian T, Burke C, McAndrews J (2009) The Federal Reserve’s primary dealer credit facility. Fed Reserve Bank N Y Curr Issues Econ Financ 15:1–10
Afonso G, Kovner A, Schoar A (2009) Stressed, not frozen: the federal funds market in the financial crisis. J Financ 66:1109–1139
Akay O, Griffiths M, Kotomin V, Winters D (2013) A look inside AMLF: what traded and who benefited. J Bank Financ 37:1643–1657
Armantier O, Ghysels E, Sarkar A, Shrader J (2013) Discount window stigma during the 2007–2008 financial crisis. Federal Reserve Bank of New York staff report 483
Berger A, Black L, Bouwman C, Dlugosz J (2014) The Federal Reserve’s discount window and TAF programs: “pushing on a string?” Working paper, University of South Carolina
Bernanke B (2009) The Federal Reserve’s balance sheet, speech. Delivered at the Federal Reserve Bank of Richmond 2009 credit markets symposium on April 3
Blau B, Hein S, Whitby R (2016) The financial impact of lender-of-last-resort borrowing from the Federal Reserve during the financial crisis. J Financ Res 39:179–206
Boyson N, Helwege J, Jindra J (2014) Crises, liquidity shocks, and fire sales at commercial banks. Financ Manage 43:857–884
Cecchetti S (2009) Crisis and responses: the Federal Reserve in the early stages of the financial crisis. J Econ Perspect 23:51–76
Cornett M, McNutt J, Strahan P, Tehranian H (2011) Liquidity risk management and credit supply in the financial crisis. J Financ Econ 101:297–312
Cyree K, Griffiths M, Winters D (2013) Federal Reserve financial crisis lending programs and bank stock returns. J Bank Financ 37:3819–3829
Cyree K, Griffiths M, Winters D (2016) U.S. Bank lending activity in the post-crisis world. Working paper, Texas Tech University
Delis M, Kouretas G, Tsoumas C (2014) Anxious periods and bank lending. J Bank Financ 38:1–13
Furfine C (2003) Standing facilities and interbank borrowing: evidence from the Federal Reserve’s new discount window. Int Financ 6:329–347
Gorton G, Metrick A (2012) Securitized banking and the run on repo. J Financ Econ 104:425–451
Griffiths M, Kotomin V, Winters D (2011) The Federal Reserve and the 2007–2009 financial crisis: treating a virus with antibiotics? Evidence from the commercial paper market. Financ Rev 46:541–567
Ivashina V, Scharfstein D (2010) Bank lending during the financial crisis of 2008. J Financ Econ 97:319–338
Kapercyzk M, Schnabl P (2010) When safe proved risky: commercial paper during the financial crisis 2007–2009. J Econ Perspect 24:29–50
Shleifer A, Vishny R (2010) Unstable banking. J Financ Econ 97:306–318
Stigum M (1990) The money market. Dow Jones/Irwin, Homewood
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Cyree, K.B., Griffiths, M.D. & Winters, D.B. Implications of a TAF program stigma for lenders: the case of publicly traded banks versus privately held banks. Rev Quant Finan Acc 49, 545–567 (2017). https://doi.org/10.1007/s11156-016-0600-2
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11156-016-0600-2