Table 9.

Decomposition of Credit Spread Changes on March 23 and April 9 Announcement.

				State Variables			Decomposition		Liquidity
		$\Delta s$		$\Delta r$	$\Delta s\tau$	$\Delta DD$	$\Delta s^r$	$\Delta s^l$	$\Delta s^r\left(LIQ\right)$
		(bps)		(bps)			(bps)	(bps)	(bps)
Panel A. Reactions to March 23 Announcement
All		−47		192	−260	1300	−24	−22	0.3
By rating	IG	−59		256	−364	1027	−55	−4	−0.1
	HY	17		−22	52	2794	−4	21	2.2
By maturity	6m−1y	−107		77	−90	898	−30	−77	−3.4
	1-2y	−80		103	−123	977	−35	−45	−1.6
	2-3y	−59		169	−148	1230	−46	−13	−0.8
	3-5y	−46		188	−181	1334	−31	−18	0.8
	5y−10y	−31		144	−184	1584	−16	−15	1.0
	10y-	−36		409	−349	1178	−25	−11	1.4
Panel B. Reactions to April 9 Announcement
All		−68		362	−446	−304	−35	−33	0.2
By rating	IG	−52		316	−362	−307	−48	−4	−0.1
	HY	−146		502	−635	−232	−63	−83	1.4
By maturity	6m−1y	−88		37	−57	−1426	−6	−66	−1.9
	1−2y	−102		80	−158	−256	−38	−69	0.3
	2-3y	−96		241	−244	390	−75	−26	0.0
	3-5y	−80		311	−299	−616	−50	−30	0.6
	5y-10y	−59		480	−374	−375	−29	−30	0.3
	10y-	−44		444	−569	−53	−30	−14	0.3

This table reports the two-day changes in the credit spreads, state variables, as well as the variance decomposition results around March 23 and April 9, 2020. The vector of state variables ($X_t$) include average bond excess returns ($r_t^e$), the product of the average credit spreads and average duration ($s_t{\tau }_t$), and the average distance to default ($D{D}_t$). We multiply $X_t$ by 10,000 so that the long-run expected returns ($s_t^r$), $e_{r}G{X}_t/{\tau }_t,$ and the long-run expected credit loss ($s_t^l$), $e_{l}G{X}_t/{\tau }_t,$ are expressed in basis points. The average changes in credit spreads and their components are reported using the whole sample as well as the subsamples grouped by rating and maturity. To compute changes over the two-day window, we restrict the bond sample to have valid price observations and stock returns on both March 20 and March 24, and on both April 8 and April 13, for the March 23 and April 9 announcements, respectively. $\Delta s^r\left(LIQ\right)$ is the fitted value of the regression of two-day risk premium changes on bid-ask spreads and trading volume.