The New Morningstar Style Box
Improved style box will mean better portfolios for investors.
Fund Spy
Starting this week, we will be featuring a new-and-improved version of the Morningstar style box for domestic-stock funds. (Our foreign-stock and fixed-income style-box calculations will remain the same.) The nine-box grid will remain, but we will use a new methodology that provides a more robust measurement. The results should prove to be more consistent, because they will better capture the factors that drive a manager's investment strategy. This, in turn, will make the style box a more useful tool for portfolio building. For more on why this will be helpful, clickhere.
For the first time, our style box will measure growth as well as valuations. (Those familiar with our old style-box calculation for U.S. stock funds know that it relied on just two factors--price/earnings ratios and price/book ratios--to slot funds in value, blend, or growth.) Valuations generally carry an implicit growth rate because investors will pay a higher-than-average price for current earnings if they expect rapid earnings growth in the future. But there are cases for which P/Es don't offer insight into a company's expected growth rate. For example, during a recession an industrial company's depressed earnings may give it a high P/E, even though investors expect it to grow slowly over the long haul. Measuring growth rates directly will lead to more-accurate classifications for such firms.
Instead of using two factors to gauge a fund's investment style, we'll now use five valuation measures and five growth measures--a total of 10--for each portfolio. This will make for a more robust measurement of a fund's current investment style. Different measures go in and out of fashion in the market, but by capturing the 10 most important factors, the style box will more accurately reflect what's driving a fund's performance. For example, cash flow was popular in the mid-1990s when investors were looking for attractive merger candidates. Later, projected earnings growth became the key gauge as growth became the dominant theme amid heated tech-stock speculation. Later, valuations and cash flow came back as investors sought the stability of profits.How It Works
As we did in the past, we begin our style-box calculation by examining each stock in the portfolio. Under the new system, each stock will be given a value score between zero and 100 and a growth score between zero and 100 based on how it stacks up on a variety of valuation and growth measures. We then subtract a stock's value score from its growth score. The higher the resulting number, the further a stock will map to the growth side. When it comes to determining the investment style for an entire portfolio, we will weight each stock's score based on the percentage of the portfolio that the stock represents. Then, we'll use the portfolio's weighted score to determine a fund's placement in the style box.
Our value measures are: price/projected earnings (50% of value score), price/book (12.5% of value score), price/sales (12.5%), price/cash flow (12.5%), and dividend yield (12.5%). In the past we used trailing net earnings in our P/E calculation, but now we use projected operating profits--the industry standard. Because the markets look ahead 12 months rather than behind, this is a better measure of what drives a stock's price.
Our growth measures are: long-term projected earnings growth (50%), historical earnings growth (12.5%), sales growth (12.5%), cash-flow growth (12.5%), and book-value growth (12.5%).
For an example of how our inclusion of growth factors in determining investment style marks a change for the better, consider the strange case of JDS Uniphase (JDSU). In early 2000, JDS had our analysts scratching their heads, because it was registering as a value stock under our old style-box system. Although the fiber-optic equipment maker appeared almost exclusively in growth portfolios, the stock was mapping to value because it had no P/E (no earnings) and its price/book ratio was fairly low. Our new methodology, however, would have captured the firm's growth rates and sky-high price/sales ratio, and JDS would have taken up its rightful home in the growth column of our style box.
A New Look at the S&P Challengers
Intriguingly, the new 10-factor style box has made some changes to some funds that we've held in large value but have long considered the S&P 500 their true competition. Legg Mason Value (LMVTX), for example, moves from large value to large blend. Although the name says value, manager Bill Miller's focus is on beating the S&P 500. He has been able to beat it even in growth-dominated years because he looks at a broad spectrum of industries and is willing to hold a stock long after its valuations have exceeded those of the market. Adding in growth scores captures the fact that Miller is in more-direct competition with the index. The fund's performance tracks the S&P 500 more closely than the large-value index.
Some of the quantitative funds that aim to tackle the S&P 500 head on, are also moving to large blend. Vanguard Growth & Income (VQNPX) aims to make incremental bets against the index while keeping sector exposure and other factors right in line with the bogy. However, because the fund favored stocks with slightly lower valuations than the market's, it mapped to value under our old system. Now, we've factored in growth and the fund seems a better fit in the middle of the style box.
Janus Is Back in Growth
Janus has always looked for above-average growth rates but lately its style boxes have shifted to blend and even value. Under the new methodology, most shift back to growth. On a valuation basis, funds like Janus Mercury (JAMRX) look like value funds because they've shifted into lower P/E financials, among other things. However, most of the companies are producing above-average growth, so the fund looks more like a growth fund when you factor in earnings growth and other growth factors. Likewise, Janus Orion (JORNX) and Olympus are moving back to the growth part of the style box.
Changes to Market Cap
Although our calculation of investment style is receiving the biggest makeover, we're also making some modest changes to how we determine whether a fund is small-, mid-, or large-cap. As is the case with investment style, we begin calculating a portfolio's market cap by examining each stock in the portfolio. We'll consider large caps to be stocks that fall within the largest 70% of the U.S. market's value, and mid-caps as those that fall within the next largest 20%. We'll call the rest small caps. Previously, we defined the cutoff levels by making the 250th largest stock the large/mid border and the 1,000th largest stock the small/mid border. Targeting a portion of the market's value makes it a little more clear how much you'd want in each bucket to build a portfolio that matches the market.
On the fund level, we'll also calculate a portfolio's market capitalization a little differently than we did in the past. We'll use the geometric mean--a weighted measurement of all of a fund's holdings. It's an improvement over our old measurement, in which we averaged the middle 10% of a fund's holdings to come up with a portfolio's median market cap. Under the old system, a fund's holdings might be tightly bunched around the median or widely dispersed. The geometric mean gives you a better feel for the fund's overall market-cap exposure. We like it better than a simple average because a straight average can be easily distorted by a few giant stocks. Beginning in July we will switch the market-cap figure displayed for a fund from median market cap to the new geometric mean.
The changes have raised our small/mid-cap cutoff point a bit. As a result, Baron Small Cap (BSCFX) and Columbia Small Cap (CMSCX), among others, dip back from mid-cap to small cap.
Categories
Note that the changes we're introducing this month affect a fund's current investment style box only--not its category placement. (Our style box serves as a snapshot of a fund's current investment style, you'll remember, whereas we rely on the past three years' worth of style boxes to determine a fund's category.) We'll apply the new style box to our category calculations in July.