In my new book, The Guru Investor, I detail ten of the best-performing stock-picking strategies of all-time. These guru-based approaches are a great way to get a leg up on the market, but, as I’ve found over the years, having a proven stock-picking methodology is just one part of the path to market-beating returns. Stock investors also must deal with a number of portfolio management questions, such as, “Which strategy is best for me?”, “How many stocks should I own?”, and, “When should I sell?”
As I’ve worked with my guru-inspired models over the years, I’ve learned a great deal about how to implement them in the context of a practical portfolio, and have been able to answer many of those questions. These lessons, I believe, can be just as important as the strategies themselves, which is why my new book includes what I call the “Six Guiding Guru Principles”.
In the latest Validea.com Hot List, I examine the first of these key principles, which details how to combine individual strategies used by the gurus to bolster your portfolio. Here’s a look at what that entails:
Principle 1: Combining Strategies to Minimize Risk and Maximize Returns
Over the long haul, there is no better or safer, when you consider inflation, investment class than stocks. But as we’ve seen, in the short term, Mr. Market (as the great Benjamin Graham called the stock market) can be extremely fickle. Sometimes he likes growth stocks; sometimes he likes value. Often he likes stocks with low price-sales ratios; other times, he thumbs his nose at them. There’s no single strategy that will please him all of the time and if you try to predict his whims by jumping from strategy to strategy, as we’ve seen, you’ll far more often than not end up buying high and selling low. The conundrum is thus how to even out those rough patches while sticking to your guns.
One of the best ways to do this, we’ve found, is to use a strategy that blends together different guru-based models that have a lower degree of correlation. What do we mean by “lower degree of correlation” It’s simple, really. It means combining strategies that perform differently in the same kind of market conditions. The simplest example would be growth stocks and value stocks. When growth stocks are in favor, value stocks tend to be out of favor; when growth stocks are out of favor, value stocks tend to be in favor. If you use a two-pronged approach that includes a growth-focused strategy and a value-focused strategy, you’ll see less volatility in your portfolio. Your highs might not be quite as high but, more importantly, your lows won’t be as low during down times.
Why is it particularly important to smooth out those down times? The biggest reason is that downside volatility isn’t just unpleasant it also costs you money. Let’s see why. Consider a $10,000 portfolio (you could use any amount) that gains 25 percent one year, loses 30 percent the next, and gains 14 percent in the third. On the surface, it would seem like your average annual gain was 3 percent, because that’s what you get when you average 25, 30, and 14 and divide by 3 years. But let’s look at what you actually would gain.
The first year, your 25 percent gain on that $10,000 investment grows your portfolio to $12,500. The second year, you lose 30 percent on that new amount, dropping the $12,500 down to $8,750. Then, you gain 14 percent on that $8,750 in the third year. That leaves you with $9,975 after three years, $25 less than what you started with.
What happened to the 3 percent per year gain we were expecting? In a sense, it got washed away in the 30 percent drop in that second year, which significantly knocked down your capital. The 30 percent decline was 30 percent of $12,500 ($3,750); by comparison, the 14 percent gain the next year was 14 percent of just $8,750 (what you were left with after that bad second year). Because of that lower starting point, your 14 percent gain in the third year didn’t get you back about half of the 30 percent you lost in year two, as you might expect; it instead recouped just $1,225 less than a third of what you lost in the second year.
The bottom line here is that in the stock market, your gains are compounded; that is, after your first year, you start earning money not on your initial investment, but on whatever you had at the end of the previous year, be it more or less than your initial investment. A bad down year thus doesn’t just mean you lose a bunch of money one year; it’s also limiting the potential money you can make next year, because any percentage gain will now be made on a lower base.
On the other hand, compound interest works for you if you’re gaining ground, since you are generating returns off your principal and your gains from the previous year(s). The more you can smooth out the valleys and keep your portfolio growing in a steady upward trend, the better, even if it means you’re smoothing out some of the peaks in big years.
There’s another less tangible reason to limit your downside volatility: You’ll feel less of an urge to ditch your approach when times get tough. Ideally, the data we’ve presented showing that sticking to a strategy is imperative has been so moving that you won’t need this reassurance. But when investment dollars start disappearing in chunks, emotions can get so intense that even the best investors can lose their cool and jump ship. Anything that helps keep you stay the course and stick to your long-term strategy is thus a help, and combining strategies to limit losses during down times does just that.
If this type of blending sounds familiar, it’s because you’ve seen it before. James O’Shaughnessy used such an approach in developing his United Cornerstone strategy, which we covered in Chapter 11. O’Shaughnessy’s United Cornerstone approach didn’t produce the best absolute returns in his study of more than four decades of stock market data; that distinction belonged to an approach in which he targeted stocks with price-sales ratios less than 1.0 and high relative strengths. But O’Shaughnessy settled on the United Cornerstone approach because it had the best risk-adjusted returns, as demonstrated by its Sharpe ratio, a risk-adjusted return measure developed by Nobel-laureate William Sharpe.
The Sharpe ratio takes into account not only returns, but also standard deviation. (If mathematical terms such as standard deviation make your head hurt, don’t worry; it’s basically just a measure of how volatile a strategy or a portfolio is). Notes O’Shaughnessy in What Works on Wall Street, “Generally, investors prefer a portfolio earning 15 percent a year with a standard deviation of 20 percent to one earning 16 percent a year with a standard deviation of 30 percent. A 1 percent absolute advantage doesn’t compensate for the terror of the wild ride.” The best combination of lower-risk, higher-return strategies, O’Shaughnessy found, was the blended United Cornerstone approach.
Our own findings support O’Shaughnessy’s research. A Total Blend portfolio of stocks that used all 10 of the strategies detailed in this book, weighting each one equally, would have produced a better Sharpe ratio than any of the individual strategies from July 2003 through April 2008. That is, it would have had a better combination of high returns and low risk than any of our individual models. Its annualized return of 19.89 percent was better than all but three of our individual models, and it posted those returns while having a lower standard deviation than all of the individual models except the Greenblatt approach (and it trailed that by only 0.08 percentage points).
Using blended strategies in a single portfolio isn’t just a way to smooth out returns. Done properly, it can also improve returns over the long haul. While the “Total Blend” approach we examined in the previous section puts the top picks of different individual strategies into a single, large portfolio, another type of blending involves looking for stocks with the most combined interest from different strategies. Most of our models examine a series of different variables when analyzing a stock, which means that using just one of these strategies ensures you’re getting a stock that is financially strong on a number of different levels. But when you focus on stocks with multiguru approval, you’re getting stocks that have really been put through the ringer.
A good case in point is the Validea Hot List portfolio that we track on our website. The Hot List looks for stocks that get the most combined interest from our strategies. It also gives greater weight to the strategies with the most historical success, meaning that the stocks it picks are fundamentally strong on a number of levels and get interest from strategies that have been very successful over the long term.
The results show what a multiguru blending approach can do. After five years of tracking, the Hot List had a higher annualized return than all but two of our guru-based models (our Kenneth Fisher- and Benjamin Graham-based approaches). From its July 15, 2003 inception through July 15, 2008, the portfolio gained 123.4 percent, more than five times the S&P 500s 21.4 percent gain during that time. What’s more, the Hot List posted those impressive gains while having a standard deviation (remember, that’s a measure of volatility) not much greater than most of our individual strategies. On a risk-adjusted basis (i.e. based on its Sharpe ratio), the only strategy that beat the Hot List by any significant amount was our Fisher model.
While an individual guru model may outperform the Hot List in a given period (as the Fisher model has done in recent years), we believe that over the longer term a blended approach will achieve the best results, because it limits downside risk when an individual strategy is going through a down period.
How does using a blended approach keep volatility in check and still beat out so many individual strategies in terms of absolute returns? A big part of it has to do with the thoroughness of our diverse group of individual models. For example, you can usually find a handful of stocks in the market that pass both our Peter Lynch-based fast-grower approach and our James O’Shaughnessy-based value model. These stocks must be growing earnings at a clip of at least 20 percent over a five-year span and have manageable debt to pass the Lynch fast-grower test, but they also must have the size and strong dividend yield and cash flow that the O’Shaughnessy value method requires. That combination of factors makes for a very, very complete stock, one that is growing earnings quickly, is conservatively financed, and is even paying a nice dividend. Over the long haul, it’s hard to imagine many stocks with this kind of multiguru approval not improving.