Joint Meeting of the Mid-Atlantic and Northeast Regions

Short-Window Fluctuations in Consumer Demand: Information or Noise?

• In Event: Doctoral/Student Faculty Interchange II

Abstract

Managers receive a continual flow of data about consumer demand for their offerings, including its volatility. A growing number of investors are also using various types of alternative data to monitor consumer demand on a continual basis and are, therefore, exposed to the volatility in consumer demand. Consumer demand is ultimately reflected in periodic reported sales, which smooth out this volatility. The effect of this smoothing depends on whether the volatility in consumer demand represents incremental information about the riskiness of consumer demand or irrelevant noise and whether investors correctly infer its information content. To shed light on these issues, we use daily consumer perception data from a well-known market research firm to capture ongoing changes in consumer demand. We find that volatility in consumer perceptions is associated with more volatile stock returns and more frequent analyst forecast revisions, indicating that at least a subset of capital market participants track fluctuations in consumer demand. However, these fluctuations do not provide incremental information about either future sales or earnings, which suggests that they may represent irrelevant noise. Nevertheless, the market discounts consumer perceptions when consumer perceptions are more volatile. This finding is attributable to stocks owned by a higher proportion of institutional investors (particularly, transient institutional investors), who are most likely to have access to and to trade on alternative data and for firms for which alternative data is available. Finally, stocks with more volatile consumer perceptions (which are most likely to have trading based on alternative data) have higher bid ask-spreads and lower trading volume. This study demonstrates that periodic reporting enhances the reliability of information by smoothing out the noise in fluctuations in higher frequency data.

Authors

• Michael Kimbrough, University of Maryland-College Park

• Xu Wang, Saint Louis University

• Sijing Wei, Creighton University

• Lu Yan, University of Maryland