Quant investing

A different approach to managing your portfolio

By Andrew Hood

On Bay Street these days, it seems that most portfolio managers favour a “value” approach to investing, with the remainder describing themselves as “growth” managers.

Value managers base their decisions on some factor, or combinations of factors (e.g. low price-to-earnings; low price-to-book; high yields; etc.), in an effort to identify under-valued companies. They believe the true value of the company will eventually be reflected in the share price. Growth managers, on the other hand, use various indicators to identify trends in the market. They often invest in stocks that have appreciated recently with the assumption that share prices will continue to rise.

While both value and growth investing approaches have their merits, using either method in isolation carries too much (although different types of) risk for my taste. The value approach may mean investing in companies whose share price doesn’t appreciate for a very long time (i.e. “dead money”), while a growth style can expose an investor to sharp volatility swings. There is, however, an alternative combining elements of both styles that is accessible for the average investor: a “quant” method.

Key steps

Quant managers’ investment decisions are primarily based on statistical data. Different quant managers weight their criteria towards either value or growth factors, but the essential thing is that they rely much more heavily on quantifiable information than on subjective judgments. This style merges the value and growth approaches, and I favour a quant method that focuses on fundamentally sound companies with strong intermediate growth prospects. What follows is a step-by-step method for constructing a sound quant-style portfolio tailored to the individual investor.

The first step in designing your portfolio is to define your individual objectives considering: your risk profile, age, need for income, tax considerations, etc. Your objectives help determine the strategic asset allocation among different asset classes. What percentage should you invest in bonds versus equities? What proportion of the equity component should be invested in Canadian versus foreign securities? Those strategic decisions are very important, regardless of the investment style employed.

Next, you must formulate a strategic trading plan based on a macro analysis of the current market. Determining the primary trend — i.e. is it a “bull” or a “bear” market — is crucial. Examples of the indicators considered include: Dow Theory; liquidity measures; momentum for key fixed-income, equity and commodity markets; and, sentiment gauges. The direction of the primary trend, along with a good understanding of where we are in the financial cycle, allows you to determine which sectors are likely to outperform relative to the rest of the market. Overweighting or underweighting specific sectors may enhance portfolio performance. A word of caution: beware of the loose talk on television or in newspapers. Many media commentators provide free advice — and it’s often worth exactly what you pay.

Applications

Now you’re ready to apply quant analysis to your universe of stocks. That universe may be large — e.g. all TSX listings for your Canadian exposure — or much smaller, perhaps 20 stocks you want to compare. I prefer to use a relatively large group of companies, but managing the analysis may be cumbersome if you don’t have access to the appropriate databases. I employ six basic “screens” for the stocks under review. Three of the screens are “value” metrics, while the other three are “momentum” metrics.

The value screens employed are:

1. High dividend yield, calculated by dividing annual dividend rate by the current price of the stock, expressed as a percentage. Historically, dividends have accounted for more than half of the total return for stocks, and the decision to pay, increase, or cut a dividend is a serious, Board-level decision. A relatively high, consistent dividend yield is often taken as one characteristic of a “blue-chip” stock;
2. Low price-to-earnings (PE) ratio, arrived at by dividing the stock price by the company’s current operating earnings per share. The lower the ratio, the less an investor is paying for a dollar of earnings. PE ratio is probably the most commonly used value measure; and,
3. Low price-to-book value (PBV), found by dividing the current price of the stock by the book value per share. Book value consists of the proceeds to the company from all stock issues, less any stock repurchased, plus accumulated earnings, less any dividends over the company’s history. Investors focused on low PBV feel they avoid overpaying for shares by paying close to the liquidating value for a stock.

The growth screens employed are:

1. High return on equity (ROE), measured by the operating earnings as a percentage of the current book value of the company. Many believe that high ROE reflects a firm’s effectiveness in investing shareholders’ money;
2. Earnings momentum, calculated as the growth in operating earnings over the past 12 months as a percentage of the current stock price. Growth investors look for high earnings and sales growth, which they believe will fuel subsequent growth; and,
3. High price momentum, derived from a composite of one-year price momentum, combined with three-month and volume momentum. The calculation basically weights the long-term price momentum versus the short-term momentum measure. For many managers, price momentum is a key statistic pointing to future growth potential.

For each of the above screens, the stocks are ranked from best to worst with the highest score (e.g. 100 out of 100) going to the best and lowest (e.g. 1 out of 100) for the worst. After running each screen, the stock’s score for each screen is added together to produce its “composite score.” Only stocks in the top 25% in terms of composite scores will be considered for the portfolio.

Two final comments on the screens: first, the numbers used are all reported data, not projections for earnings, commodity prices, economic growth, etc. We’re focused on what is actually happening with the company and in the market. Second, each screen ranks all the stocks in the sample on a relative basis against the rest of the sample. No stock is excluded from consideration for the portfolio on the basis of a single factor (e.g. the PE is too high), but the value screens will generally filter out companies that have growth momentum without improving valuations (anyone who remembers the tech bubble of 2000 will appreciate the significance of this).

Making adjustments

Ranking the stocks analyzed by their composite scores produces our final focus list. Again, we concentrate on the top quartile and select stocks to provide diversification by sector. You could be successful investing on the basis of those results alone, rebalancing them quarterly, but before I go into a position I analyze each stock using technical indicators (chart action, relative strength, momentum, trend indicators, etc.) to select the best point of entry based on the balance of probabilities.

Once you’re invested in the portfolio, you must monitor and make adjustments as warranted. In most cases, quarterly rebalancing produces good results. I monitor portfolios daily and make changes according to certain buy/sell indicators. Practical experience shows that a portfolio constructed using the criteria outlined will tend to have a low to moderate annual turnover.

Of course, the proof is always in the results. As an illustration of what this methodology can achieve, BMO Nesbitt Burns’s research department has back-tested the performance of stock selections based on the screens described. From a sample of the 120 largest stocks on the TSX, ranked by their composite scores, the “portfolio” consists of the top half of the companies by ranking, and it’s rebalanced monthly. This (admittedly mechanical) approach has produced the following historical annual rates of return (as of May 31st, 2007), which is compared to the BMO Cap 10% Index (the TSX index with a cap on individual company’s index weight at a maximum of 10%):

                                  Composite Portfolio                    BMO Cap 10% Index

Past Year                              24.45%                                         22.73%
Past 3 Years                         24.19%                                         21.14%
Past 5 Years                         18.12%                                         15.20%
Past 10 Years                       18.65%                                         11.61%

While no one would build a portfolio of 60 stocks (over-diversification!), the point is that the method can produce results superior to the index return over both short and longer time periods.

The key to the quant approach described here is the ability to compare the relative merits of a large number of companies based on reported financial data (not forecasts) using screens that together have a track record of indicating strong stock selections for a portfolio. The method imposes a discipline on your investing and allows you to rank many companies against one another on a relative basis, using consistent criteria.

This helps an investor avoid the trap of becoming emotionally attached to a particular company. Using both value and growth screens, in effect, identifies relatively undervalued companies that are exhibiting strong growth characteristics. In my opinion, positioning yourself in the market by using a quant approach is the best way to select investments that are both   fairly valued and attracting the attention of the market. It’s neither pure value nor pure growth, but perhaps the best of both worlds.

Andrew Hood (andrew.hood@nbpcd.com), Ph.D., is a financial planner & investment adviser with BMO Nesbitt Burns Inc. Opinions are those of the author and may not reflect those of BMO Nesbitt Burns. The information and opinions contained herein have been compiled from sources believed reliable but no representation or warranty, express or implied, is made as to their accuracy or completeness. BMO Nesbitt Burns is a member of CIFP.

Top