Market Reaction to Macroeconomic News: The Role of Investor Sentiment

• Author: Hung‐Kun Chen,Chen‐Ting Lien
• Published date: 01 December 2017
• Date: 01 December 2017
• DOI: http://doi.org/10.1111/ajfs.12198

Market Reaction to Macroeconomic News:

The Role of Investor Sentiment*

Hung-Kun Chen**

Department of Banking and Finance, Tamkang University, Taiwan

Chen-Ting Lien

Department of Finance, London School of Economics and Political Science, United Kingdom

Received 1 October 2016; Accepted 28 May 2017

Abstract

This paper demonstrates that risk is unlikely to be the only factor explaining the excess

return difference between macroeconomic announcement days and other trading days. We

argue that investor sentiment may affect the return on the prescheduled macroeconomic

news release. We find that the return on macroeconomic announcement days is much higher

than the return on other days, especially when changes in investor sentiment are higher. The

impact of investor sentiment on the macroeconomic announcement effect is greater when

investor sentiment is higher. Our results remain valid using different investor sentiment mea-

sures and testing for heteroscedasticity in stock returns.

Keywords Macroeconomic announcements; Investor sentiment

JEL Classification: G12, G14

1. Introduction

Macroeconomic news released in the form of prescheduled announcements, which

are usually published by public agencies, is important and influential in stock valua-

tion in a capital market. Savor and Wilson (2013, 2014) show that returns are quite

different for days on which macroeconomic news is scheduled for release (hereafter

“announcement day”) and other trading days (hereafter “non-announcement day”).

Their studies reveal that average daily index excess return is about 0.11% on an

*We acknowledge useful comments from Konan Chan, Sheng-Syan Chen, Yan-Shing Chen,

Chia-Wei Huang, Woan-lih Liang, Yanzhi Wang, and seminar participants at National

Chengchi University for helpful comments and suggestions.

**Corresponding author: Hung-Kun Chen, Department of Banking and Finance, Tamkang

University, No. 151, Yingzhuan Road, Tamsui District, New Taipei City 25137, Taiwan. Tel:

+886-2-2621-5656 ext. 3328, Fax: +886-2-2620-9653, email: hkchen@mail.tku.edu.tw.

Asia-Pacific Journal of Financial Studies (2017) 46, 853–875 doi:10.1111/ajfs.12198

©2017 Korean Securities Association 853

announcement day whereas the average daily index excess return is only 0.01% on

a non-announcement day.

1

Because the outcome of prescheduled economic news is

uncertain in advance, the findings of Savor and Wilson (2013, 2014) are consistent

with the notion of the risk–return trade-off story that investors demand a premium

when they are faced with stock market uncertainty.

The risk factor, however, is unlikely to be the only factor explaining the dif-

ference in excess returns between announcement days and non-announcement

days. We propose that investor sentiment may affect the return on the presched-

uled macroeconomic news release. On announcement days, both sentiment-dri-

ven noise traders and rational investors participate in the stock market, whereas

fewer noise traders tend to participate on non-announcement days (Nofsinger,

2001). The participation of noise traders influences stock returns (De Long et al.,

1990; Shleifer and Summers, 1990). It is reasonable to posit that stock returns

may be higher when noise traders become more optimistic (Lee et al., 1991).

Therefore, we argue that changes in aggregate investor sentiments, which imply

changes in investor composition, predict stock returns on announcement days.

2

Since it is difficult to directly observe the participation of sentiment driven by

noise traders, several investor sentiment indices are employed to track the effects

of noise traders. Moreover, the noise traders are also affected by status, weak or

strong, of the market anomalies (Stambaugh et al., 2012). Thus we conjecture

that during periods with higher sentiment, the impact of changes in sentiment

on the stock return on the macroeconomic announcement day should be

stronger.

We follow Savor and Wilson (2013, 2014) and collect US prescheduled news

about inflation, unemployment, and Federal Open Markets Committee (FOMC)

interest rate decisions during the period 1966–2010. CRSP value-weighted and

equal-weighted index excess returns are used to examine the market reaction on

economic announcement days and non-announcement days. To measure investor

sentiment, we employ the Baker and Wurgler (2006) sentiment index, which con-

sists of information from closed-end fund discounts, the NYSE share turnover, the

number of initial public offerings (IPOs), the average first-day returns of IPOs, the

equity share in new issues, and the dividend premium. The higher the sentiment

index, the more optimistic the investor sentiment.

Empirically, we find that, consistent with Savor and Wilson (2014), during

announcement days, the average CRSP value-weighted index excess return is 1.46%

(converted to a monthly basis), whereas the average index return on a month ly

1

Their results dovetail with those of recent studies showing that the interest rate, unemploy-

ment rate, and inflation rate strongly affect stock returns (Bernanke and Kuttner, 2005; Ioan-

nidis and Kontonikas, 2008; Bjørnland and Leitemo, 2009; Birz and Lott, 2011).

2

We use the changes in sentiment to examine the relation between investor sentiment and

stock return, as this approach to measurement has been comprehensively used in the litera-

ture (Lee et al., 1991; Baker and Wurgler, 2006, 2007; Lemmon and Portniaguina, 2006).

H.-K. Chen and C.-T. Lien

854 ©2017 Korean Securities Association