Authors: James McAndrews, Asani Sarkar, and Zhenyu Wang
The Term Auction Facility (TAF), the first auction-based liquidity initiative by the Federal Reserve during the 2007–2009 financial crisis, was aimed at improving conditions in the overnight dollar money market and bringing down the significantly elevated London interbank offered rate (Libor). The effectiveness of this innovative policy tool is crucial for understanding the role of the central bank in financial stability, but academic studies disagree on the empirical evidence of the TAF effect on Libor. This paper shows that the disagreement arises from misspecifications of econometric models. Regressions using the daily level of the Libor-OIS spread as the dependent variable suffer from the unit-root problem and produce unreliable test statistics. Those regressions also miss either permanent or temporary TAF effects, depending on the choice of independent variables. In contrast, regressions using the daily change of the Libor-OIS spread are robust to the unit-root problem and the persistence of the TAF effect, consistently producing reliable evidence that the TAF was associated with downward shifts of the Libor-OIS spread. The evidence indicates the efficacy of TAF in helping the interbank market to relieve liquidity strains.