Mean reversion is a way for hedge funds to earn money on trades, but it’s also a reason not to panic when the stock market suddenly drops. The idea is that over longer periods, stock prices follow steady trends. If prices rise or fall sharply, they usually return to the long-term price trend eventually. There are different mean reversion strategies for different time frames, and hedge funds like to use a short-term mean reversion strategy to earn money.
Short-Term Mean Reversion
One way to set up a mean reversion strategy is to automatically sell the stocks that rose the most yesterday and buy the stocks that fell the most yesterday. The idea is that the stock price has already moved because of recent news. Some investors may have overreacted to the news and pushed the price too far in one direction. Additionally, investors with profitable trades will exit their positions to take profit, and investors who were wrong will use stop losses to close their positions.
Because of these factors, a stock that rose sharply is likely to drop afterward and a stock that fell sharply is likely to rise a bit. This type of mean reversion strategy is a short-term trading strategy where you only hold a position for a few days. It’s not a long-term investment strategy. But because you’re holding the stock for a few days, you can use this strategy without automation. Automation just makes it easier to find the stocks that went up and down the most.
Long-Term Mean Reversion
The Dogs of the Dow strategy is a longer-term mean reversion strategy. It means buying the blue-chip stocks with the highest yields, which often means buying the blue-chip stocks that dropped the most last year. And this strategy has longer holding periods, so dividends are part of the strategy. When a stock’s price falls, its dividend payout frequently remains the same, so its dividend yield will be higher afterward. For example, if a stock costs $100 and pays out a $3 dividend, that’s a 3 percent yield. But if the stock falls to $50 per share and its dividend is still $3, its yield is now 6 percent.
The idea with Dogs of the Dow is that you buy beaten down blue-chip stocks. These are stocks of strong and reputable companies because they’re members of the Dow. So their share prices are likely to recover eventually, and in the meantime you’ll receive high dividends.
The Dow part of this strategy is important. I’ve seen penny stocks that pay out 20 percent dividends because their share prices fell sharply. And I’ve seen companies in struggling business sectors that also pay out very high dividends. For example, businesses like video rental stores might pay out very high dividends because investors don’t want to buy stock in a company with bad long-term prospects.
If a publicly traded company is experiencing severe difficulties, it may stop paying out dividends, so it’s a good idea to pick blue-chip companies that are less likely to do that. If you use this strategy, you might want to screen for stocks that have consistently paid out dividends.
This strategy originally got my attention because mean reversion backtests on the algorithmic trading platform Quantopian showed very high gains. For example, if you were running a short-term mean reversion strategy since 2008, backtests often show returns of millions of percent. The catch is that many traders didn’t know about mean reversion back then, and now they do. So a basic mean reversion strategy won’t work as well now. Be careful about trading strategies that have earned very high profits in the past but show smaller gains in more recent years.
But the theory behind mean reversion is still valid. Traders will continue to exit profitable trades. Many investors will overreact to good news and bad news as well. And over the long term, prices are likely to return to the long-term trend because of these factors. So I’d predict that modified mean reversion strategies are still likely to generate alpha.
Hedge funds often check that there’s a valid theory behind any profitable investment strategy. Just showing them a backtest with very high returns is unlikely to impress them. They want to know why those returns occurred and whether the strategy is repeatable.
There are several reasons why mean reversion is a valid trading strategy and there are ways to apply it over several different time frames. It’s important to consider additional factors if you’re using mean reversion over longer periods, though. Algorithmic hedge funds that only hold a stock for a few days might not consider the long-term prospects of the stocks they select, but if you’re holding a stock for a year it’s more important to pick stronger companies.