Sunshine affects mood and mood can shape behavior. It is then plausible to test if weather is related to economic outcomes, such as market return. The ‘weather effect’ is documented by some, Saunders (1993) and Hirshleifer and Shumway (2003), and claimed an exercise in data mining by others, Kramer and Runde (1997). In the paper that follows, I study the relationship between weather and equity prices over time using various measures of the change in weather and market return.
To test the hypothesis that sunshine affects stock returns, I use simple regressions to examine the relationship between daily cloudiness, the inverse of sunshine, of New York City and the return on the Dow Jones Industrial Average index. For the time period of 1948 to 2010, there is a negative relationship between average cloudiness and DJI gross simple return. After controlling for market anomalies, such as the January and the weekend effects, a change in weather from sunny to overcast skies is associated with an additional .79 percentage point decline in gross return (t-statistic = -2.81). So the weather effect seems to exist for this time period in New York City. On sunny days, investors feel more optimistic and more willing to invest in risky assets; this change in behavior leads to higher stock prices. However, if the same simple regression is used to study 30 year sub-samples at a time, although the estimated coefficient on average cloudiness is negative for all periods, the relationship between cloudiness and market return is statistically insignificant for some sub-samples. To analyze whether the sunshine effect is robust over time, I measure the return difference between good and bad weather days for each year and study its evolution over time. Once the volatility of this difference in returns variable is reduced by computing its moving average, it is positive for almost all years and slightly increasing over the past 50 years. So the market return is higher on good weather days, defined as exceptionally sunny days, than on bad weather days, defined as exceptionally cloudy days. This difference in returns is strongly persistent and increasing for periods such as 1975 to 1980 or the late 1990s; for other sub-samples, namely 2000 to 2008, the difference in returns of sunny vs. cloudy days is sharply decreasing.
One possible explanation for the rise and fall of the sunshine effect over time is the entry of small investors into the market during periods in which equity investment attracts popular attention. These non-professionals’ misattribution of good mood on sunny days extends to their investment decision-making process more so than professional investors allow for such a psychological bias. Hence the weather effect is more pronounced for certain years, specifically those periods for which the market is not dominated by perfectly rational investors. This finding supports the theoretical argument of Mehra and Sah (2002) that investors’ feelings have a significant effect on equity prices. Furthermore, the increase of the weather effect for certain time periods provides empirical evidence for the ‘limits to arbitrage’ argument made by Barberis and Thaler (2002): equities can remain mispriced, due to the actions of a small subset of investors, even if arbitrageurs suspect mispricing.
Joel Greenblatt Owned Hedge Fund On Why Value Investing Isn’t Working Now
Acacia Capital was up 12.27% for the second quarter, although it remains in the red for the year because of how difficult the first quarter was. The fund is down 14.25% for the first half of the year. Q2 2020 hedge fund letters, conferences and more Top five holdings Acacia's top five holdings accounted for Read More
The overall implication of these results is that there is a significant relationship between weather and stock prices; this relationship exhibits a cyclical pattern over the past half-century. Thus, depending on the years under study, a researcher may find a significant relationship between weather and stock prices or may find insignificant results and label the weather effect “an exercise in data mining.” However, I conclude that extreme and intermediate weather changes in New York City are strongly correlated with within day DJI return.