Day vs Night trading

When I started my career in 1977 after-hours trading was an exotic specialty. Markets opened at their appointed time in the morning and closed in the afternoon. Commodities had their own special times. But, as far as I can remember, the NYSE opened at 9:30 AM and closed at 4:00 PM every trading day except for early closings before holidays like Christmas. 

I remember early 90s one night watching as crude oil exploded from $20 to $40 a barrel (on the Singapore exchange?) when the Gorbachev coup failed and then the market fell back to $20. The markets were never the same. Today crude oil and many other commodities trade nonstop from 6:00 PM Sunday to 5:00 PM Friday.

Some ETFs also trade after hours. The exact set of ETFs that trade 24 hours is set by your broker. Here's my broker's 24 hour ETF list today:

Source: Thinkorswim

It's a little bit of everything. The major stock indexes, SPY, QQQ, IWM, some bond ETFs, TLT, AGG, and an eclectic bunch of commodities and foreign ETFs apparently selected by popular demand. There are also a few hincky inverse and sector ETFs. Ok, so what does 24 hour trading give us? More profit or more loss?

I tested this using daily open, high, low, and closing prices with three measures: the portion of the daily return during day only trading hours (DO POR), the day only continuously compounded rate of return (DO CCROR) and the, usually quoted, daily CCROR. The first test looks at prices since ETF inception.

Data Source: Microsoft Stock History function.

How to read this? Look at SPY. The daily stock history starts with 1/29/1993. 

• DO POR = The Day Only Portion of Return = the "percentage of daily return from day only hours". Suppose on a given day SPY closed 35c higher than the morning's open and $1 higher than yesterday's close. The DO POR would equal .35/1.00 = 35%.SPY's AVERAGE DO POR since inception = 66.5%. So, yes, most of the action happened in the day session. 
• DO CCROR = the average continuously compounded rate of return for the day only hours since inception = 1.0%. This is what you get when you never hold overnight. 
• CCROR = the continuously compounded daily return since inception, i.e. for ALL hours = 2.1%

In this case, holding SPY overnight doubled the daily return. This is in contrast to the so-called risk averse investor who buys on the open and sells on the close, never taking a position overnight. These are stylized situations but you get the point.

For the QQQs and ILK, holding after hours is the a major difference between a gain and a loss in the position. Eight other ETFs had higher returns than the DO returns. Inverse ETFs, commodities and a few others have hinky results that may merit further investigation. 

One issue that may account for some of the hinkiness could be technology. In the early years few investors had access to technology the made after-hours trading easy. Since the widespread dawn of powerful and FREE trading platforms, such as TD Ameritrade'sThinkorsim, night trading by retail investors is enabled or even encouraged. So, lets look at the same table for this year only.

. 

Here we are looking at the measures for the most current data, this year's data. For instance, SPY's total return this year is 4%. IF we took day only positions our gain would have been 9%. in essence this is telling us that SPY was generally DOWN in the overnight. Look at UNG. Down 51%, your loss would only have been 16% in the day only positions.

Our conclusion here is what many market studies show, your mileage may vary. Choice of names and time periods create no clear trends. Buy and hold, over the long term, may be best for broad based stock indexes. 

Disclaimer: Posts are for education only, may be subject to change without notice, and, while prepared with care, may be subject to omissions and errors. Send request to gdrahal@outlook.com to follow this blog and for additional information. 

© 2023 George Rahal