by Samuel Lee
Buy the dip or avoid falling knives? The two dicta represent opposing postures to falling prices. Both can't be right, yet many investors adopt one of the rules of thumb with little more evidence than personal experience or intuition. The question is particularly important to exchange-traded fund investors with a tactical bent, and even stodgy strategic investors who perceive a buying opportunity. We set out to find out which rule works with the help of Morningstar's trove of historical data.
Testing the Strategy
We tested a simple dip-buying strategy: At the beginning of each month, if an asset's three-month return is below the 30-day Treasury bill return over the same period, hold the asset; otherwise, hold T-bills. The goal is to simulate an investor who consistently tilts against the market's short-term gyrations. If buying the dip works, our hypothetical investor should reap excess profits.
To get the broadest, deepest sample of data, we selected indexes representing the major asset classes with the longest histories, such as the S&P 500 and the Goldman Sachs Commodity Index. We stuck with large, investable asset classes, so private equity and hedge funds are out.
In order to make sense of our results, we needed a fair benchmark. One way to assess a timing strategy is to see if luck can explain its returns. For each index, we calculated how often the buy-the-dip strategy held T-bills or the asset. We simulated a thousand different historical scenarios where we randomly selected which months to hold cash or the asset, in the same frequency as the buy-the-dip strategy. The simulations' average return was our benchmark. The difference between the benchmark's and the strategy's returns, along with the distribution of the simulations' returns, tells us how probable the strategy's returns were.
As you can see in the table below, buying the dips hurt returns in stocks, real estate, and currencies, but not long-term U.S. government bonds. The losses are big and exceedingly unlikely to be random events. An investor who bought the S&P 500 on dips and sold into rallies would have lagged the benchmark by 2.3% annualized since 1926. The chance of suffering such underperformance by randomly selecting months to buy the index is less than 1%.
Returns From Buying on Dips
To verify our results, we ran the tests on the indexes' components and foreign country stock indexes. (See the table below.) The same pattern repeated itself. Out of 73 different asset classes tested, only the Cyprus Pound and Kuwaiti Dinar enjoyed Sharpe ratio improvements and only the Cyprus Pound earned higher returns than the benchmark.
Returns From Buying on Dips - Component Returns
The bottom line: Buying on dips hurt risk-adjusted returns over one-month horizons for almost every asset class. The result may seem counterintuitive in light of value investors' historical outperformance. However, value hunting and dip-buying are distinct strategies. Value is pegged to fundamental measures such as book value or earnings, whereas dip-buying is a price-driven rule--technical trading, really. It's possible for an asset to shed dollars and still be overpriced. Value is also realized over years-long horizons, while our dip-buying strategy only held for a month (though there were extended periods where it rode markets down). There's evidence that the worst-performing assets over the previous years go on to outperform in the long run as mean reversion brings valuations back up to historical norms.
The Notorious M.O.M.
The pervasive underperformance of our short-term dip-buying strategy isn't surprising to those familiar with price momentum (sometimes abbreviated MOM in academia). The tendency for prices to trend can be tremendous. In 1993, Narasimhan Jegadeesh and Sheridan Titman published a study showing that holding the top 10% of U.S. stocks ranked by trailing 12-month returns and shorting the bottom 10%, rebalanced and reconstituted monthly, earned excess returns of 1% a month from 1965 to 1989, or more than 12% annually. The study spurred a frenzy of research that documented momentum in virtually all stock markets and asset classes, including the Victorian-era British stock market. Thanks to this rich body of research, we have a decent idea of why momentum exists and how it behaves.
Running With the Herd
The most convincing explanation scientists have come up to explain momentum goes something like this: In light of surprising or extreme news, investors may "anchor" new price estimates to old prices, preventing prices from fully reflecting new information. Investors are also loath to realize losses, preferring to keep dogs until they break even, and are too quick to sell winners to "lock in" profits. Both biases prevent prices from quickly reflecting new information; instead, prices slowly adjust to fair value, creating sustained movements, up or down. Other investors, observing these price trends, tend to hop on the bandwagon, creating a positive feedback loop where sentiment drives prices. In the ideal world envisioned by efficient-market theories, rational investors would prevent such irrational price movements by shorting overheated assets and buying bargains, but in the real world short-selling is an expensive, risky process, and managers buying falling knives risk losing their jobs.
Value Investing Isn't Dead
According to the behavioral theorists, dip-buyers may be succumbing to anchoring bias, where they've unfairly pegged their intrinsic value estimate to the recently prevailing price. This doesn't mean all dip-buying is irrational. If you've done the hard work of calculating an asset's intrinsic value, and dips bring the price below intrinsic value, the strategy is a rational exercise in value investing. However, without that legwork, dip-buying is a remarkably bad technical trading rule.