Manager Monday: Legg Mason
Every Monday we dive into a manager's portfolio using the Novus platform. Learn the source of Legg Mason's outperformance in our new post.
Everything mentioned in this post is sourced exclusively from public data, including the manager’s profile, simulated performance and all other analysis and commentary. The data used here omits the short side, non-equity securities, many non-US securities and all non-public information such as actual fund performance.
For this Manager Monday we decided to depart from the world of hedge funds and look at a mutual fund while applying the same Novus lens we use with alternative investors. For our case study we use the Legg Mason Opportunity Trust managed by Bill Miller and Samantha McLemore. Their public fact sheet states that the $2B+ portfolio is concentrated (high percent of portfolio in relatively low number of securities), has a high active share (differs from the benchmark) and invests in “assets with large gaps between price and intrinsic value” Many refer to this as value investing, in the tradition of legendary investor Ben Graham. The fund is classified as Mid-Cap value by Morningstar and uses the S&P 500 as the benchmark, the same one we use in this article.
Having a high active share means the portfolio can deviate significantly from the holdings and performance of its benchmark. As we can see from the sector weights of the portfolio relative to the S&P 500, it is overweight Consumer Discretionary and underweight Energy, Utilities, Telecom and Staples (sectors that the fund does not have significant exposure to).
But what do these sector tilts mean? Is it a good or a bad thing for the funds investors?
We have simulated the fund’s performance based on quarterly filings with the SEC since the year 2000. Because our performance is simulated, there will be an expected difference between the actual fund’s performance on their fact sheet and the performance we examine here. Much of the difference should be attributable to intra-quarter trading. The funds simulated returns have greatly outperformed the benchmark, albeit with more volatility. This is also true of the funds actual returns that were higher than our simulation. The cumulative difference between the simulation and actual fund returns is around 30 percentage points over the last 15 years, most likely attributable to trading. The simulated performance lags behind the actual performance, implying that trading has added value since inception. Regardless of which return stream you analyze, the fund has outperformed the market by a wide margin. The very next question any Novus client would ask is how they did that and what the outperformance is attributable is to.
The portfolio has always been fairly concentrated and held between 20 and 65 securities at any point in time. The top ten securities represent roughly 40% of the total portfolio value. This is obviously much higher than the concentration of the benchmark, so security selection (inclusion of certain stocks and not others) should be very important to Legg Mason’s returns. Since inception, we see 281 unique stocks in the historical portfolio with 154 stocks (55%) contributing positively and 124 (45%) detracting. While 55% is not an exceedingly impressive batting average, their win/loss ratio paints a different picture. Legg Mason has a 1.8x win/loss (avg. contribution of winner / avg. contribution of detractor, 271/153) meaning their contribution from winners is roughly double their loss on detractors.
Digging deeper, we can examine the same concepts on a relative basis to the benchmark. As we can see from the table below, their performance is really driven by the stocks they pick that beat the S&P500.
Those relative winners contribute over 3x of what the relative losers detract. This is important for two reasons. First, it demonstrates that the PM’s show skill in sizing their outperforming names, also evidenced by higher win/loss ratios in large position size buckets, below.
Second, it demonstrates the PMs ability to “cut their weeds and water their plants” or ride winners and cut detractors. We talk more about this skill and why it important to managers in an earlier post.
Areas of Alpha Generation
Across sectors, as we mentioned before, Legg has been overweight Consumer Discretionary and has been growing that exposure in recent years. Using the Novus Framework, a proprietary tool to decompose returns into its components, we can segregate the portion of P&L attributable to sector allocation (lighter bars) and security selection (darker bars). Besides materials, which has never been a significant portion of their portfolio, the best area of selection skill is in Consumer Discretionary. This is what you want to see from an active manager and indicates that they know what they are good at and allocate capital accordingly.
Other areas of positive security selection skill (or stock picking alpha) is in the Healthcare and Industrials sectors. Materials also stands out but has not been a meaningful exposure for the portfolio recently. One potential area of opportunity for improvement is their Financials sector exposure. It has been a consistent overweight and at times reached as high as 42% of the portfolio. However, financials show negative selection skill historically, meaning they would have fared better in a Financials ETF. Information Technology is another area of potential improvement, showing negative contribution from stock selection while still representing meaningful exposure in the portfolio.
From the analysis of public ownership data on Legg Mason Opportunity Trust, we can see evidence of certain skills possessed by the fund managers and how they’ve contributed to the fund’s outperformance over their benchmark. We have also identified some areas of opportunity for further improvement of returns, especially in certain sectors. Having this information arms the investor with higher caliber tools for assessing a manager than simply looking at their Sharpe ratio of monthly returns.
When analyzing a manager’s performance it’s always important to keep in mind the skill dependent portion versus market dependent. Schedule a demo to learn more.