For as long as markets of any kind have been around, mankind has surely been looking for way to explain how prices move. Since Adam Smith, we know at a micro-level that prices reflect supply and demand for a good or, in the case of the stock market, an asset.
But can we actually predict how the forces of supply and demand will play out in the future? Academics have theorized no, believing that markets incorporate all known information into their prices already, and that future movements are therefore totally random. But evidence from recent studies as well as the experiences of practioners would suggest that this is not the whole story.
An idealized view of how security prices should move
In the conventional view presented by many intro-level Finance and Economics textbooks, stock prices follow what is called a "random walk." This means that prices have no "memory" of the past. In other words, the movement of the stock market today is a new "toss of the dice" and has no relation to how the stock market performed yesterday, last week, or over the last year.
If stock prices really do randomly "drift" upwards over time then the implication is that the past movement of a stock should not influence its future movements at all. In other words, buying stocks that have performed well in the past and selling those that have done poorly (or vice versa) should not be a profitable strategy.
How security prices actually move
When finance professors decided that it might be time to test their assumptions about the way security prices move (it took a few decades), they found something kind of interesting, and that is that prices don’t really follow their models at all (Being finance professors, this took another couple decades to convince them that maybe they should adjust their view of the world to reflect evidence rather than the assumptions of their theory, but that is a story for another article).
There have been a number of peer-reviewed academic papers that have shown that, in fact, past performance is a reliable and predictor of future performance at some time scales. Professors Jegadeesh and Tittman broke this news to the world in 1993 (though savvy traders had happily been following momentum strategies since at least the 1920s) with an often-cited paper showing that a strategy of buying the stocks that had performed the best over the past nine months and selling those that had performed the worst over the past nine months would have generated excess returns of 12% a year since 1965 (see http://papers.ssrn.com/sol3/papers.cfm?abstract_id=299107 for the 2001 update). This level of excess returns was highly statistically significant and could not be explained by other factors, like higher risk.
The momentum effect exists across all markets and geographies
In the last 20 years, the findings of Jegadeesh and Tittman have been found to hold across countless different time periods, asset classes, and countries, indicating that momentum may be a fundamental property of financial markets.
Some of the international findings include:
- In the United Kingdom, winner stocks (defined as stocks with returns in the highest twenty percentile) have outperformed loser stocks (those with returns in the lowest twenty percent) by 11% a year. $1 invested in a "winners" portfolio in 1900 would have grown to $4.3 million by 2008. $1 invested in a "losers" portfolio would have grown to only $111 (source: ABN Invesment Yearbook 2008).
- Of 17 country stock markets studies by ABN Amro, 16 had positive returns to momentum over the 1975-2000 period. When the study was updated, 16 out of 17 also had positive "out of sample" returns in the 2000-2008 period. The average excess return was nearly 10% a year, a huge level.
- US residential housing prices have strong momentum effects. While not investable, a strategy of buying houses in the local markets that have performed the best in the past year and selling houses in the markets that have performed the worst would have had excess returns of up to 8.9% annually (Source: http://link.springer.com/article/10.1007%2Fs11146-009-9210-2)
- Momentum strategies have also been shown to work across markets. Selling the stock markets that have performed the worst over the past six months and buying those that have performed the best has been a winning strategy.
In totality, this paints a strong picture that the momentum effect is not an artifact of “data mining.” It is a real effect that needs to be recognized and confronted by both academics trying to construct grand market theories, and by investors trying to rationally preserve and grow their wealth.
In the next article, we will look at whether momentum is something that is likely to continue to work in the future.
Get our next article delivered to your inbox.
Sign up below and be the first to know about our freshest data-driven thinking on the markets, and investing. We will send you no more than one email a week. This is free.
Ready to start putting this into action?
Take a free two-week trial to IvyVest premium -- our premium subscription service. You'll get access to our rules-based dynamic asset allocation model, tools that will show you exactly what you need to buy in your own discount brokerage account (and when to re-balance) to implement it for yourself, and an insightful monthly newsletter that will keep you on abreast of the most important things going on in the markets. There is no credit card required. Get Started Now!