A New Stock-picking Formula

Debashis Basu 28 April 2017

For the patient do-it-yourself investor, High Returns from Low Risk by Pim Van Vliet is a breakthrough book, in the same league as The Little Book that Beats the Market. In that book, Joel Greenblatt explained how ranking stocks on the basis of high return on capital and low valuation does the job of picking long-term winners. This book, too, offers a quantitative formula. Van Vliet demonstrates that ranking stocks on three filters—low volatility, high dividend yield and rising momentum —yields terrific market-beating results. He shares the excellent back-tested results of this formula.

Van Vliet shows the impact of progressively adding one filter at a time on the same set of US stocks, going back all the way to 1929. He starts with volatility. He took the monthly closing prices of US traded stocks from January 1926 to December 2014, a period of more than 80 years—through the Great Depression of the 1930s, World War II, the prosperous 1950s, the swinging 1960s, multiple recessions, the oil crisis of the 1970s and market excesses like the Dotcom bubble of the 1990s. Any results obtained over such a long period are “unlikely to be the result of pure luck or temporary effects that might disappear at some point in the future.”

To keep things simple, the author picked the 1,000 largest companies by market-capitalisation and excluded penny stocks (less than $1 in price). He ranked these 1,000 stocks, by the volatility of monthly return over the previous three years. He then created two portfolios, one containing 100 stocks with the lowest volatility (low-risk portfolio) and the other with highest volatility (high-risk portfolio). But volatility changes, over time. To ensure that the portfolio retains the initial character, Van Vliet rebalanced the portfolio every quarter. If a stock in one of the portfolios no longer qualified to be there, he just exited that holding and replaced it with a new stock. What were the returns of the two portfolios?

Assuming that you put $100 into both of the portfolios on New Year’s Day 1929 and re-invested any capital gains for 86 years until New Year’s Day 2015, the low volatility group yielded a return of $395,000 (compounded annual growth rate of 10.1%), while the high volatility portfolio’s return was only $21,000 (6.3%). High risk does not equate with high returns. Low-risk stocks beat high-risk stocks by 18 times!

So why not just buy the top low-risk stocks? Because we can do even better with two more filters. As the author puts it, “If you could choose between two stocks with exactly the same low risk characteristics but one was very expensive and the other was very cheap, chances are you’d opt for the cheap one… like a lot of other things in life, it pays to focus not only on what you buy, but also on the price you are to pay for it. This idea of buying cheap stocks is a popular and well-known investment strategy, applied by many investors all over the world.”

This is popularly known as value investing. One of the ways value is expressed is dividend yield (dividing the dividend declared by market price). But some companies prefer not to pay dividends; they would rather buy back shares. So, Van Vliet ranked stocks by dividends plus buybacks, calling this filter ‘income’. If a stock becomes expensive, the income yield will fall and vice versa. At this stage in the process, we have low volatility stocks with high income yield. We need a third filter. Why? Some low-priced stocks may offer a high income yield but may face poor business prospects which is why they are inexpensive. If so, the stock will languish and even decline—value trap. How to filter these? Enter momentum, the third factor.

There are many ways of measuring momentum, the most common of which is to filter stocks on the basis of their returns over the previous 12 months. The assumption is that strong momentum usually continues. This list of high-momentum stocks was then combined with the other two filters. In practical terms, Van Vliet picked 500 lowest-volatility stocks. He then ranked them on income and 12-month momentum. Each stock got a score between 1 and 500, based on these two factors; these scores were simply added together. Then, he combined these two lists and picked the top-100 stocks. This shortlist now would have a stock with low volatility, high income, and good momentum. With this approach, the average compounded return since 1929 gets bumped up, from 10.1% to 15% per annum. Quite impressive.

Comments