Beating the Market One Wide Moat at a Time

Detailed analysis of competitive advantage and a steadfast focus on the long-term cash-generating ability of businesses are at the core of Morningstar's approach to equity investing, and so we spend a great deal of time thinking about economic moats and forecasting long-run cash flows. (Click here to learn more about how Morningstar analysts assign economic moat ratings to companies.)

But how can we have confidence that our approach adds value? After all, elegant theories are one thing, but generating excess returns in a very competitive marketplace is quite another, especially given the increased efficiency with which information flows between companies and investors.

To answer this question, we created an index several years ago called the Wide Moat Focus that holds the 20 wide-moat stocks trading at the largest discounts to our estimates of intrinsic value. If we have some skill at identifying businesses with competitive advantages, and if we have some skill at valuing the shares of those businesses, then a portfolio of cheap wide-moat stocks should outperform the market generally, and our wide-moat universe specifically.

I am happy to report that, so far, this has indeed been the case. A portfolio of the wide-moat stocks we view as being the most mispriced has generated substantially higher returns than both the S&P 500 and our overall wide-moat universe. I've summarized the performance data in the table below.

Trailing Performance of Wide Moat Focus ( % )
	2011 YTD	Trailing 1 Year	Trailing 3 Year	Trailing 5 Year	Since Inception (10/1/02)
Morningstar Wide Moat Focus Index*	7.1	20.7	11.7	9.4	15.5
Morningstar Wide Moat Index **	5.2	24.0	5.7	5.2	10.6
S&P 500 Index	5.9	25.9	1.8	3.0	8.1
Data as of 2/15/11 * The wide moat focus index has been "live" as an ETN (ticker: WMW) since 11/01/07. It has an expense ratio of 75 basis points, so returns are somewhat lower. The three-year trailing return for WMW is 10.9%, and the one-year trailing return is 19.9%. ** The Wide Moat Index contains all U.S.-domiciled wide-moat stocks. It is equal-weighted and rebalanced quarterly.

While these results are obviously gratifying to those of us on Morningstar's Equity Research team, they also raise some very interesting questions. How is it that a group of well-known, (mostly) large-cap companies with substantial Wall Street analyst coverage can become mispriced with such frequency? How can a mechanical portfolio-construction process using well-known, (mostly) large-cap companies generate substantially higher returns than both passive indexes and the vast majority of active managers? (After lopping off a 1% hypothetical management fee, the Wide Moat Focus Index bests 95% of large-cap funds--and 90% of mid-cap funds--over the trailing three- and five-year periods, as well as since its late-2002 inception.)

I recently performed a comprehensive analysis of the Wide Moat Focus Index to examine these questions, and I'd like to share the results with you. Did the Wide Moat Focus simply take on more risk? Were excess returns generated from sector allocation or security selection? Is there a size or style bias to the index that would explain the performance data in the tables above?

The Wide Moat Focus Index: Background
The Wide Moat Focus Index is drawn from a larger universe of domestic wide-moat companies with reasonable liquidity that constitute the Wide Moat Index. (Non-U.S. companies and master limited partnerships, in particular, are not included in the Wide Moat Index and are thus not eligible for inclusion in Wide Moat Focus.)

To construct the Wide Moat Focus Index, we sort on market price relative to Morningstar fair value estimates and include the 20 securities trading at the largest discount to fair value. The holdings are equal-weighted, and the index is rebalanced and reconstituted quarterly.

And that's it. No sector constraints, no top-down subjective input to portfolio construction. The Wide Moat Focus is simply the 20 wide-moat stocks that trade at the deepest discounts to Morningstar analysts' estimates of intrinsic value.

So, what has the portfolio looked like over time? Let's start with size and style, as shown in the "Style Trail" chart below, which plots each one of the index's quarterly portfolios on the Morningstar Style Box. (The current portfolio is circled in black.)

Given that our wide-moat stocks tend to be larger caps, the Wide Moat Focus portfolio has generally landed in one of the large-cap style boxes during the past eight years, though it has dipped into the mid-cap area a few times. In terms of style--about which we are completely agnostic when valuing equities--the index's portfolio has shown more variation, moving from value to growth and back again depending on which securities look cheapest to us at any point in time.

Wide Moat Focus: Size + Style, Since Inception

In terms of sector exposure, the Wide Moat Focus has been all over the place--which is exactly what one would expect from an unconstrained, purely bottom-up construction process. For example, almost 40% of the portfolio was in tech during the first half of 2007. Tech fell to a zero weighting at the end of 2007 as other sectors became cheaper. Likewise, health care has gone from a minimal weighting a couple of years ago to more than 30% of the current portfolio.

The Wide Moat Focus has had very different sector allocations than the S&P 500, since we tend to find more wide moats in some areas of the market than in others. Specifically, the Wide Moat Focus has had a minimal or zero weighting in energy, materials, telecom, and utilities, which have collectively made up about 15%-20% of the S&P 500.

In sum, the Wide Moat Focus Index has been mainly in large caps, and it has not had a very well-defined style bias. Relative to the sector weightings of the S&P 500, it has been overweight financials, health care, and tech, and underweight energy, industrials, materials, utilities, and telecom. Do any of these characteristics help explain its performance?

 

The Wide Moat Focus Index--Attribution
There are several reasons why a strategy or manager may outperform a benchmark. Perhaps a manager just took more risk, in which case the performance numbers would look great, but risk-adjusted metrics like the Sharpe and Sortino ratios would be unimpressive. Or, the manager may have biased the portfolio to certain styles of investing, certain market caps, or certain sectors without displaying stock-picking acumen within those areas. Finally, the strategy or manager may have simply selected securities that outperformed, even after controlling for style, size, and sector effects.

Risk and Return
Let's start with risk. The Wide Moat Focus portfolio has been somewhat more volatile than both the S&P 500 and the wide-moat universe, as measured by standard deviation and beta. However, the extra return achieved by the portfolio has more than compensated for the additional volatility. For example, the Wide Moat Focus portfolio has a Sharpe ratio--measuring excess return per unit of volatility risk assumed--of 0.64, which is substantially higher than the S&P 500's Sharpe ratio of 0.36.

We can also look at upside and downside capture ratios, which indicate how a strategy or manager has performed relative to a benchmark during up and down markets. If a strategy blows away the benchmark during good times, but gets crushed during bad times, the investor experience may not match the trailing returns, as investors pile in during good times but cash out at the bottom. On these metrics, the Wide Moat Focus holds up quite well. Relative to the S&P 500, the Wide Moat Focus has risen 20% more than the benchmark when stocks were generally rising, yet it has fallen about 10% less when the broad market was heading south.

Size and Style
Next, let's turn to potential size and style biases. During the past eight years, the Wide Moat Focus portfolio has been tilted a bit in the growth direction relative to the S&P 500 and meaningfully tilted away from large caps. (About 40% of the Wide Moat Focus' holdings have been small or mid-caps--mostly mid-caps--while the S&P 500 has only averaged 12% in small and mid-caps.) So, it's possible that the Wide Moat Focus portfolios have outperformed simply because they were "middier" and "growthier" than the broad market.

Certainly, the past several years have favored some sizes and styles more than others. Small- and mid-cap stocks have trounced large caps over pretty much any time period during the past several years. In terms of style, growth has beaten value recently, but the two are neck-and-neck since the inception of the Wide Moat Focus in late 2002. Given that the mid-cap "tilt" of the Wide Moat Focus relative to the S&P 500 was much stronger than its growth leaning, and that the outperformance of mid-caps relative to large caps has been much greater than the outperformance of growth relative to value, I'll focus my attention on the size question.

To analyze the question of whether the Wide Moat Focus' significantly higher weighting in mid-caps could have accounted for a substantial proportion of its outperformance relative to the S&P 500, I used Morningstar Direct to create a custom benchmark that replicates the size tilt of the Wide Moat Focus. The custom benchmark uses Morningstar's large-cap, mid-cap, and small-cap indexes and weights them according to the large/mid/small breakdown of the Wide Moat Focus portfolios. If the Wide Moat Focus did not add alpha via security selection, then it should show minimal or no outperformance relative to this size-matched benchmark.

The chart below shows three-year cumulative rolling returns for the Wide Moat Focus relative to this size-matched benchmark. As you can see, the Wide Moat Focus showed outperformance in a substantial majority of rolling periods, and the excess returns were frequently substantial in size. I think the conclusion from this analysis is that while the Wide Moat Focus did get a tailwind from its mid-cap tilt relative to the S&P 500, the effect of the size bias was small relative to the effect of security selection, as shown by the substantial excess returns of the Wide Moat Focus over a size-matched benchmark.

Three-Year Rolling Returns: Wide Moat Focus Relative to Size-Matched Benchmark

 
Data as of 01-31-11

 

Sector Weightings
Finally, let's look at sector weightings. The Wide Moat Focus portfolios have been very heavily weighted in consumer discretionary, financials, health care, and tech relative to the S&P 500. Wide Moat Focus has been somewhat underweight consumer staples and industrials, and it has had little to no exposure to energy, materials, telecom, or utilities.

I used Morningstar Direct to analyze whether the Wide Moat Focus was simply in the right sectors at the right times (sector allocation) or whether good security selection had a bigger impact. Overall, the sector weightings of the Wide Moat Focus portfolios added about 18 percentage points of cumulative return relative to the market, with the tech overweight adding the most relative return. On the flip side, having essentially no energy exposure cost Wide Moat Focus a material amount of relative return.

However, the approximately 18 percentage points of cumulative relative return added by sector weightings was far overshadowed by the more than 120 percentage points of relative return added by good security selection. In other words, Wide Moat Focus' sector choices were just OK relative to the market, but the strategy did an excellent job of picking specific companies within those sectors. Bottom line: The Wide Moat Focus' outperformance relative to the broad market was not the result of mainly being in the right sectors at the right times. Good stock picks made a much bigger contribution.

Attribution: Conclusion
The outperformance of the Wide Moat Focus Index was not due to excessive risk-taking, nor to size or style tilts, nor to sector weightings. The Wide Moat Focus strategy has produced excellent performance on both an absolute and a relative basis because we have done a good job at identifying undervalued wide-moat stocks.

We could just conclude the analysis here, but I think that the excess returns generated by the Wide Moat Focus strategy raise some larger issues that are quite intriguing. As I mentioned earlier, the Wide Moat Focus strategy has beaten 95% of large-cap fund managers during the past several years and 90% of mid-cap managers. Given that the Wide Moat Focus portfolios have generally owned larger, well-known companies that are well-followed by scores--if not hundreds--of analysts and portfolio managers around the world, it seems reasonable to examine possible sources for Morningstar's "edge."

The Wide Moat Focus Index--Reasons for Alpha
About a decade ago, Russell Fuller wrote a great paper that talked about three possible sources of excess returns. An investor can acquire superior information and create an informational advantage; he or she can develop a superior process to analyze commonly available information and create an analytical advantage; or the investor can exploit the behavioral biases exhibited by other investors.

Having an informational advantage on the larger companies that tend to populate the Wide Moat Focus seems improbable, so I think we can quickly rule out the first possibility. The possibility that Morningstar analysts have an analytical advantage has some merit, since our economic moat framework is unique to Morningstar, and competitive advantage is generally an underanalyzed area by most investors. Moreover, Morningstar has analyzed well more than 2,000 diverse businesses during the course of the past decade, which gives us a large collective mental database to strengthen our decisions about which companies have wide moats and which ones do not.

However, any analytical advantage we have would be far less powerful without our ability to take advantage of the market's rampant behavioral biases. In particular, our relentless focus on estimating the long-term business value of the companies we analyze--regardless of short-term movements in stock prices or generally accepted accounting principles earnings--has allowed us to exploit the market's preference for short-term returns over long-term returns. In a way, these two advantages are inextricably linked, since thinking about sustainable competitive advantages requires a long-term mind-set.

The average turnover for a plain-vanilla U.S. equity mutual fund is around 100% these days, which is a holding period of one year. I think it's safe to say that the bulk of hedge fund money turns over at a substantially higher velocity, to say nothing of the high-frequency trading crowd. As a result, a substantial proportion of the capital being invested in equities is seeking returns over a reasonably short (one year or less) time horizon. Less competition for long-term returns leads to less-efficient pricing of securities based on their long-term prospects, as recency bias causes the marginal market participant to extrapolate the most recent information about a security (good or bad) too far into the future.

At Morningstar, we work very hard to stay focused on the long-run ability of businesses to generate cash, while most market participants work very hard to predict the short-term path of share prices. In my opinion, the results of the Wide Moat Focus portfolios are clear evidence that the former is an endeavor with much better odds of success.

Discipline and Patience Pay Off
The Wide Moat Focus index started off as a "proof of concept" that would validate our thesis that myopic markets misprice moats, and it later evolved into an exchange-traded note. After more than eight years, I think our thesis stands on solid ground. Moats do get mispriced, and they get mispriced with great regularity.

Moreover, the results of the Wide Moat Focus Index show that Morningstar's equity research team has demonstrated some skill at identifying which businesses have wide economic moats, at forecasting what the long-run cash flows of those businesses are likely to be, and at standing their ground in the face of disconfirming market behavior.

It appears, then, that beating the market requires neither inside information, nor a black-box quantitative model, nor a co-located server. It merely requires discipline and patience. Thankfully for active investors, these are scarce commodities on Wall Street. As long as they continue to be scarce, I think that Wide Moat Focus and similar strategies have a good chance of generating alpha.