Business cycles, stock returns and the transmission channels of conventional and unconventional monetary policy

ABSTRACT

The zero-lower bound (ZLB) may constrain the ability of Federal Reserve (Fed) to influence financial markets and the economy. This note examines the effectiveness of the interest rate channel and the credit channel of conventional and unconventional monetary policy while accounting for business cycle fluctuations. We use intraday industry returns and a number of industry-specific and firm-specific indicators to capture the sensitivity of firms’ demand to interest rates (interest rate channel) and firms’ financial constraints (credit channel). Our results indicate a dramatic change in the effectiveness of the transmission channels across business cycles and across periods. We find that the interest rate channel operates equally well during recessions and expansions pre-ZLB, but that this channel has ceased to function during the ZLB regardless of the stage of the business cycle. In contrast, the credit channel operates only during recessions in the conventional period, while it has been remarkably effective during both recessions and expansions in the ZLB era.¹

¹ We thank Stefania D’Amico, Emmanuel Lartey, Lin Yan, and conference participants of the Econometric Society Asia Meeting 2021, WEAI 2021, World Finance Conference 2022, and International Finance and Banking Society 2018 for helpful comments and suggestions. We gratefully acknowledge the support of the Faculty Research Grant of the College of Business and Economics, California State University, Fullerton.

KEYWORDS:

• Interest rate channel
• credit channel
• zero lower bound
• LSAP
• forward guidance

JEL CLASSIFICATION:

• G14
• E44
• E52
• E58

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ We thank Stefania D’Amico, Emmanuel Lartey, Lin Yan, and conference participants of the Econometric Society Asia Meeting 2021, WEAI 2021, World Finance Conference 2022, and International Finance and Banking Society 2018 for helpful comments and suggestions. We gratefully acknowledge the support of the Faculty Research Grant of the College of Business and Economics, California State University, Fullerton.

² See, for example, Kontonikas, MacDonald, and Saggu (2013), Wu (2018) and Farka (2022).

³ The conventional period runs from May 1999 to October 2008 and from December 2015 to December 2019. The unconventional period is from November 2008 to October 2015.:

⁴ We use intraday changes 15 minutes before and 1 hour and 45 minutes after a policy announcement in the first and third federal funds futures contracts ($\mathrm{\Delta }\mathit{FF}{R}_1^{fut}$, $\mathrm{\Delta }\mathit{FF}{R}_3^{fut}$), the second, third and fourth Eurodollar futures ($\mathrm{\Delta }\mathit{EU}{R}_2^{fut}$, $\mathrm{\Delta }\mathit{EU}{R}_3^{fut}$, $\mathrm{\Delta }\mathit{EU}{R}_4^{fut}$), and the 2-, 5-,10-, and 30-year Treasury yields. X has 178 observations (T = 178, corresponding to policy announcements) and 9 columns (n = 9, corresponding to interest rate data). Unlike Swanson (2021), X does not contain the S&P500 index and exchange rates for two reasons. First, it is closer to other measures in the literature that are based strictly on interest rate changes (Gertler and Karadi 2015). Second, the yield curve adequately captures factor loadings, as shown in Table 1. Our policy shocks are highly correlated with Swanson’s (2021): 99.4% (for the target rate), 94.5% (for forward guidance), and 96.5% (for LSAPs).

⁵ As in Kuttner (2001), m is the number of days in the month, d is the day of the monetary policy announcement and $\mathit{FF}{R}_t^{fut}\left(\mathit{FF}{R}_{t-1t}^{fut}\right)$are the futures rate at time t (t-1).

⁶ For durability, we rank industries by regressing industry returns on industrial production. Capital intensity is the ratio of investments over total capital; cyclicality is based on industry sensitivity to the Chicago Fed National Activity Index. Firm size is measured by market capitalization; financial leverage is the ratio of debt to total capital; cash flow is the ratio of cash flow to income. All data are obtained from Compustat.

⁷ Sector returns are computed by taking the log difference of average future prices 15 minutes before and 1 hour and 45 minutes after a policy announcement. Excess returns are computed by subtracting the 1-month Treasury rate from intraday returns.

⁸ We the Chicago Fed National Activity Index as a proxy for the state of the economy (Basistha and Kurov (2008).

⁹ For example, the interest rate channel is in effect during recessions in the conventional period if the spread differential $({\alpha }_1^{\mathit{high}}+{\beta }_1^{\mathit{high}})-({\alpha }_1^{low}+{\beta }_1^{low})$ is statistically significant for industries ranked based on durability, capital intensity or cyclicality indicators.