Weather-Induced Mood, Institutional Investors, and Stock Returns

Studies showing that weather patterns in major financial centers influence stock index returns provide suggestive evidence that investor mood influences asset prices (Saunders, 1993; Hirshleifer and Shumway, 2003). Individuals may misattribute mood induced by weather as information when making assessments about objects that should be otherwise unrelated (Schwarz and Clore, 1983), leading to mood-congruent judgments. For example, sunnier days may induce good moods amongst investors, generating overly optimistic beliefs regarding their investments and congruently influencing their trading decisions. Despite strong evidence of the weather effect on stock index returns, establishing plausibility in mood-based explanations relies in part on distinguishing which group of investors drives the weather effect, and directly confirming mood effects in their judgments.

In our paper, Weather-Induced Mood, Institutional Investors, and Stock Returns, forthcoming in the Review of Financial Studies, we evaluate economic mechanisms through which weather-induced mood can affect asset prices. In particular, we investigate how mood affects the beliefs and trading behavior of a group of investors that play a key role in the price formation process: institutional investors. Despite commonly held assumptions about investor sophistication, there is evidence that institutional investors and other market participants are also susceptible to behavioral biases (e.g., Coval and Shumway, 2005; Goetzmann and Zhu, 2005; Frazinni, 2006). We conjecture that weather patterns would affect mispricing perceptions of institutional investors through its influence on mood. Focusing on institutional investors allows us to isolate mood effects in trading behavior, and, in turn, analyze its impact on stock prices.

We use a variety of data sources to overcome limitations in the existing literature. Similar to previous studies, we use deseasonalized cloud cover measured from weather stations located nearby investors as our mood priming device. The weather data are merged with survey data from the Yale Investor Behavior Project, which provides institutional investors’ opinions on stock market investments, and allows us to construct a variety of measures that capture perceived investor mispricing. A similar merging procedure is used to link the weather data to disaggregated data on investor trade histories. The ANcerno database provides daily trades of a large number of institutional investors, allowing us to construct measures of buy-sell trade imbalances. Finally, using 13(f) institutional holdings data, we construct novel, stock-level measures of investor mood, which incorporate weather conditions of all institutional investors holding the stock as of the most recent reporting date.

Our main results can be summarized in three parts. First, we examine the impact of mood on investor beliefs. Using the survey data, we find that sunnier days are associated with optimistic responses related to stock mispricing. That is, investor optimism increases the likelihood of perceived underpricing in both individual stocks and the Dow Jones Industrial Average. The results are robust to a variety of specifications that account for factors related to investor characteristics, time period, and region, and are significant both statistically and economically. We also show that the effects are strongest when incorporating weather information from several weeks prior to the responses date, suggesting that the composite of affective reactions to information collected over time is meaningful. This result is consistent with similar findings on conditioning factors related to mood documented in the social psychology literature (Adaval 2001), as well as clinical research on the therapeutic effects of light therapy on seasonal (Rosenthal et al., 1984) and nonseasonal depression (Goel et al., 2005).

Second, we examine the impact of mood on investor trading behavior. Even if weather-induced mood affects beliefs, as we show, it is unclear whether it is strong enough to also affect decision-making. We show that sunnier days are associated with increased buying propensities of institutional investors. When restricting our tests to only comparing mood across investors for the same stock and date, we obtain similar results. Coupled with our findings on investor beliefs, the results suggest that the impact of investor mood on beliefs is sufficient to influence financial decisions with meaningful economic consequences. Moreover, these findings establish plausibility of price effects in individual stocks due to mood-induced weather.

Finally, we examine whether the mood of institutional investors holding the same stock can influence future stock returns. Using the stock-level, investor mood proxy that incorporates weather conditions of all investors currently holding the same stock, we evaluate its impact on daily stock returns in two ways. First, we show that the investor optimism proxy can have a positive effect on individual stock returns. In stocks where mispricing can be more costly to correct for (i.e. higher arbitrage costs), we posit that price effects due to weather-induced mood should be stronger. Consistently, we show that the results are concentrated in stocks that are the most costly to arbitrage. Second, we examine the return comovement of stocks based upon the investor mood proxy. We show that stocks associated with similar moods exhibit return comovement, while those associated with dissimilar moods do not. These effects are significant one month after portfolio formation and dissipate after three months, suggesting that the effects are not permanent.

The full paper is available for download here.