If you ask a day trader, he will tell you to close out all your positions by the end of the day. And this is not without good reasons. Because a day trader is focused only on the short term. For them to make money, they need to be able to monitor and respond quickly to the markets. Thus, holding overnight risk is, for the most part, a strict no for them.
But is an overnight risk something that terrible? Well, not if you are looking at the long term. In fact, if you don’t hold positions overnight, you are missing out big time.
Your Investments Grow The Most After Market Closed
I know this might sound incredulous but the bulk of your returns actually come from moves that happen overnight. To be clear about what I meant, let’s first talk about daily returns. Daily returns are measured from the market close to close. So if the price of Stock A close at 100 today and then 110 the following day, the return for that day is 10%. However, this daily return can be further split into 2 parts. The first part is the return from the prior day’s close to today’s open. The second is the return from today’s open to today’s close. The first part is your overnight returns during which markets are closed and the second is the return during regular active market hours.
Overnight and Active Hour Return
A look at US S&P 500
For a start, let’s track the overnight and active hour returns of SPY (S&P 500 ETF) since its inception in 1993. Let’s also assume there are no transaction costs. In order to capture the overnight returns, we invest at the market close each day and then close out the position when the market next opens. Similarly, to get the active hour returns, we invest at the opening of each day and liquidate the position when the market closes on the same day. With these returns, we can build a Net Asset Value (NAV) series that shows how each performed over time and we also compare them against a simple Buy and Hold strategy on SPY.
Performance of Buy & Hold, Overnight, Active Hour Strategy
The results are pretty evident. The Overnight strategy is the one driving the gains over time delivering a Compound Annual Growth Rate (CAGR) of 9.4%. It also outperforms in terms of risk as its profile shows lower volatility and drawdown. The Active Hour strategy, on the other hand, gave a flat performance over this close to 30-year period.
If there is a tradable spread on their returns, i.e. long overnight and short active hour returns, this is how its performance would look like.
Spread NAV Strategy - Long Overnight Short Active
How About Other Markets?
Now, is this something unique to the US, or does the same behavior extend to the stock markets of other countries? Let’s look at Hong Kong, Japan, and Germany. As a proxy for investment performance, I used their respective local indices, namely, Hang Seng, Nikkei 225, and DAX.
Hang Seng Buy & Hold, Overnight and Active Strategy Performance
Nikkei 225 Buy & Hold, Overnight and Active Strategy Performance
DAX Buy & Hold, Overnight and Active Strategy Performance
Every one of them exhibited the same pattern. Japan, in particular, was especially pronounced with a 9% CAGR for overnight returns and -7.5% for active hour returns. Does that mean all we have to do is buy at the close and sell at the following open each day for the best performance? Sounds pretty good, isn't it? But before you jump into it thinking you stumbled upon a golden goose, do note that these indices are not tradable products and their official opening print can differ from what you would be able to get when the actual market opens. On top of that, there might bad print data on these indices.
Even if all the data are clean, a tradable product exists, and you can execute your trades near the open and close prices, there is still an often overlooked and underestimated component.
An Often Underestimated Component – Transaction Costs
To implement such a strategy, the costs can be immense. Let’s take the case of SPY with about 252 trading days. That means 252-round trades. Even if each round trade were to cost you only a few basis points, it can still pack a pretty strong punch.
We do a simple illustration using SPY using a flat fee of $0.005/share for buying and selling. And for simplicity, let’s also assume that there are no minimum commissions. Then add to this charge another 1-cent slippage (smallest tick size) on execution each way. Hence, for a round trip, the total costs will come up to $0.03/share. $0.005 for buying, $0.005 for selling, $0.01 for bid-ask on buying, and $0.01 for bid-ask for selling. And this is the impact.
Adding a cost of $0.03 per share shaves a massive chunk off the returns. The CAGR drops from 9.4% to just 2.9%. So what is often too good to be true usually is too good to be true.
Then how about if my broker does not charge commissions? Then there is still the bid-ask spread and slippages. While I have not done thorough research on comparison across such brokers, one such broker I tested out tends to give me a less favorable executed price when done side by side against another broker that charges comms. And it is understandable. Because there is no such thing as a free lunch. These brokers typically earn their fees by routing their order flows to other market makers who pay for them. And in the process, your trades might not be executed at the best price available.
What Is Driving The Growth Overnight?
At the very best, we can attempt some intelligent guesses. Events or news don’t cease just because the regular trading hours are over. In fact, plenty of market-moving news comes after the market close e.g company earnings. When one market closes, another opens. With globalization, there is hardly any market that stands alone. And if you look in terms of time, regular trading covers only 6-7 hours of the entire day (US Market). Thus, whatever news or events happen “overnight” gets digested and priced in when the market opens the following day.
Read more about the markets at: https://allquant.co/blog
$SPDR S&P 500 ETF Trust(SPY)$ $iShares MSCI Japan ETF(EWJ)$ $iShares MSCI Germany ETF(EWG)$ $iShares MSCI Hong Kong ETF(EWH)$
Comments