Daylight saving effect – paper review

The daylight saving effect argues that sleep disruption caused by daylight saving time changes results in a negative impact on stock returns on the trading day following the changes.

Even if the effect does exist it’s unlikely to be economically significant, but it is interesting to academics in the context of behavioral finance and whether external factors (such as the weather) can influence the mood of investors sufficiently to affect share returns.

Compared to other market anomalies this is a pretty straight-forward one; the academic debate has largely centered around methodology.

This article presents a brief review and listing of academic papers on the daylight saving effect.

An academic paper in 1976 found that transitions to and from daylight saving time (DST) caused sleep desynchronosis (disruptions). One paper described it as having an effect similar to jet lag. In 1980 a paper claimed that DST changes led to an increase in traffic accidents; while another paper a little later claimed it led to a decrease in accidents. Oh well.

As far as we’re concerned the story starts in 2000 when Kamstra, Kramer and Levi found that in the US, UK, Canada and Germany DST had a negative impact on stock returns on the trading day following the changes. They called this the daylight saving anomaly. A new anomaly – this was quite exciting, nowadays there aren’t that many new anomalies found in stock markets (certainly not in developed markets).

Anyway, this got the ball rolling.

Unfortunately not many others agreed with them. First up was Pinegar (2002) who looked at the US market and only found significance for the autumn DST changes, and that that was attributable to two data outliers for the market in October 1987 and October 1997. Kamstra, Kramer and Levi (2002) replied quickly to this maintaining their claims by showing that the distribution of returns on days following DST changes shifted to the left.

Next up was Worthington (2003) who found no effect in the Australian market. Lamb, Zuber and Gandar (2004) replicated and agreed with the findings of Pinegar (2002) and aggressively concluded that the original findings of Kamstra, Kramer and Levi (2000) “did not survive serious scrutiny”.

The disagreements continued with Müller, Schiereck, Simpson and Voigt (2009) who found no daylight saving effect in European bond and equity markets.

Gerlach (2010) had a different slant, claiming that any correlation between stock returns and DST changes was not due to the daylight saving effect but rather to seasonal patterns in market-related information.

Gregory-Allen, Jacobsen and Marquering (2010) criticized the original paper for using too little data; and crunched the numbers on 22 markets and over a longer period. They found no evidence of an observable DST effect on stock returns.

Berument, Dogan and Onar (2010) widened the study to look at volatility as well as stock returns, and found no DST effect. This started a merry game of paper ping-pong with Kamstra, Kramer and Levi (2010) replying with a “comment” that criticized the analysis of Berument, Dogan and Onar (2010) and maintaining the effect was still in place. Berument and Dogan (2011) replied with a “reply”, effectively saying, “isn’t”. Which prompted Kamstra, Kramer and Levi (2013) to reply with a “rebuttal” to the “reply” to the “comment” effectively saying, “is”.

INDEX (of papers listed below)

[Papers listed in reverse date order; indicates major paper.]

  1. Effects of daylight-saving time changes on stock market returns and stock market volatility: rebuttal [2013]
  2. Does Mood Affect Trading Behavior? [2012]
  3. A reexamination of the effect of daylight saving time changes on U.S. stock returns [2012]
  4. Effects of daylight saving time changes on stock market volatility: a reply [2011]
  5. The Daylight Saving Time Anomaly in Stock Returns: Fact or Fiction? [2010]
  6. Effects of daylight-saving time changes on stock market volatility: a comment [2010]
  7. Daylight and investor sentiment: a second look at two stock market behavioral anomalies [2010]
  8. Effects of daylight savings time changes on stock market volatility [2010]
  9. Daylight saving effect [2009]
  10. Robust global mood influences in equity pricing [2008]
  11. Do Daylight-Saving Time Adjustments Really Impact Stock Returns? [2007]
  12. Weather, Biorhythms and Stock Returns – Some Preliminary Irish Evidence [2005]
  13. Don’t lose sleep on it: a re-examination of the daylight savings time anomaly [2004]
  14. Losing sleep at the market: an empirical note on the daylight saving anomaly in Australia [2003]
  15. Losing Sleep at the Market: The Daylight Saving Anomaly: Reply [2002]
  16. Losing Sleep at the Market: Comment [2002]
  17. Losing Sleep at the Market: The Daylight-Savings Anomaly [2000]

Effects of daylight-saving time changes on stock market returns and stock market volatility: rebuttal
Authors [Year]: Mark J. Kamstra and Lisa A. Kramer and Maurice D. Levi [2013]
Journal [Citations]: Psychological Reports, 112(1), pp89-99
Abstract: In a 2011 reply to our 2010 comment in this journal, Berument and Dogen maintained their challenge to the existence of the negative daylight-saving effect in stock returns reported by Kamstra, Kramer, and Levi in 2000. Unfortunately, in their reply, Berument and Dogen ignored all of the points raised in the comment, failing even to cite the Kamstra, et al. comment. Berument and Dogen continued to use inappropriate estimation techniques, over-parameterized models, and low-power tests and perhaps most surprisingly even failed to replicate results they themselves reported in their previous paper, written by Berument, Dogen, and Onar in 2010. The findings reported by Berument and Dogen, as well as by Berument, Dogen, and Onar, are neither well-supported nor well-reasoned. We maintain our original objections to their analysis, highlight new serious empirical and theoretical problems, and emphasize that there remains statistically significant evidence of an economically large negative daylight-saving effect in U.S. stock returns. The issues raised in this rebuttal extend beyond the daylight-saving effect itself, touching on methodological points that arise more generally when deciding how to model financial returns data.
Ref: AA766

Continue reading

Social Share Toolbar

Daylight saving effect

British Summer Time will start this Sunday, 30 March.

Although Benjamin Franklin first suggested the idea of a daylight saving time in 1784, it was William Willett, a builder living in Petts Wood, Kent, who first took the idea seriously enough to circulate a pamphlet to Members of Parliament in 1907. Willett’s argument was that advancing the clocks in spring and returning to GMT in the autumn would lead to an increase in health and happiness, and it would also save the country £2.5 million pounds (after taking account of the loss of earnings to the producers of artificial light).

British Summer Time (BST) was finally introduced in April 1916, as an economy measure during wartime – a move quickly followed by many other countries in the following few weeks. During the second world war double summer time (two hours advance on GMT) was introduced for the summer, while winter clocks stayed one hour ahead of GMT. For three years from 1968, Britain kept on BST throughout the whole year. But the experiment was abandoned in 1972, since when the Britain has kept GMT in winter and BST in summer.

In 1996 all clocks in Europe changed on the same day for the first time. The European Union has now adopted The Ninth European Parliament and Council Directive on Summer Time Arrangements in which summer (or daylight saving) time will be kept between the last Sunday in March to the last Sunday in October. The changes will take place at 01.00 GMT.

And the relevance to shares is…?

Research has shown that even minor sleep disruptions can cause profound changes in cognition, leading to anxiety, inattention, and impaired judgment.

An academic paper (Kamstra, Kramer, and Levi, 2000) looked at the potential effect the switches to and from daylight saving time might have on stock markets in the US, UK, Canada and Germany. The authors investigated  the  possibility  that  investors  prefer  safer  investments  and  shun  risk  on  the trading day following daylight saving time changes.

Some of the results they found were:

  1. In all cases, it was found that stock returns were significantly lower following a daylight saving time change, than they were on other trading days, even after controlling for other known seasonalities like the Monday effect. This is consistent with anxiety-prone investors selling risky assets on the trading day following the sleep disruption caused by daylight saving time changes.
  2. The daylight saving effect was found to be extremely significant for the UK and US markets, strongly significant in Canada, and relatively insignificant in Germany (likely due to the relatively shorter period of data available for Germany since there were no daylight saving time changes in that country between 1950 and 1979).
  3. The Autumn daylight saving change had a greater impact than that in Spring.
  4. In the US, the financial impact of this phenomenon amounts to roughly $30 billion on average, each time the clocks were shifted over the past 30 years.

The following chart shows the FTSE 100 returns for the days following the start of British Summer Time (BST) for the period 1985-2013.

FTSE 100 return on day after start of British Summer Time [1985-2013]Over the period the FTSE 100 average daily return was -0.07% on the days following the start of BST. In line with the paper’s findings, this was significantly lower than the average day returns for all days (+0.03%) and for Mondays (-0.01%).




Social Share Toolbar