Another Reason Not to Watch CNBC When the Market is Open

Stocks seem to earn their returns when the market is closed. When you compare the returns on stocks overnight to the returns intraday, overnight dominates. You really don’t want to know how bad the performance is during the day.

While you’re unlikely to profit by this finding, Eric Falkenstein recently documented the size of the effect for a handful of stocks and ETFs: it seems stocks earn their returns when the market is closed.

Falkenstein looked at daily log returns for a half-dozen stocks and a couple ETFs, comparing their returns from Open-to-Close (Intraday) to Close-to-Open (Overnight). The annualized results are in the table below.

Start Date	Ticker	Open-Close	Close-Open	stdev(O-C)	stdev(C-O)
1999	QQQ	-0.13	0.14	0.29	0.16
1993	SPY	-0.03	0.10	0.17	0.10
1984	AAPL	-0.13	0.30	0.41	0.29
2008	TNA	0.03	0.43	0.78	0.46
1986	YRCW	-0.25	-0.01	0.61	0.30
1986	MSFT	0.12	0.10	0.32	0.21
1962	IBM	0.02	0.06	0.22	0.12
	Avg.	-0.05	0.16	0.40	0.23

Source: Falkenblog

I verified the data myself. The sum of the Intraday and Overnight equals the annualized daily log return over the period indicated for each respective time series. Not sure what to make about Microsoft.

I wondered how the Overnight/Intraday spreads vary across sectors. Perhaps the most surprising result was that the spread seems greatest for the Utilities sector. The magnitudes seem too big to be explained by data errors associated with dividends.

Here’s a look at how the Overnight/Intraday spread plays out among Sector Select Spider ETFs:

Source: PortfolioWizards

Can you use this information to your advantage? Probably not so much for improving performance as for your peace of mind.

As Falkenstein wrote in the comments of his post, he doesn’t believe this is exploitable because if he did he wouldn’t have written about it.

This finding further supports Richard Thaler’s advice pertaining to investor myopia: stop watching the market!