Attribution analysis reveals the stocks and sectors that drive (or hold back) mutual funds' returns.
Bill Miller (Legg Mason Value LMVTX) got hurt owning stocks now resting in the financial company graveyard, Bruce Berkowitz (Fairholme FAIRX) cashed in on energy, and Will Danoff (Fidelity Contrafund FCNTX) scored big with Google GOOG. These are some anecdotal explanations for the performance of those funds, but a handy tool called attribution analysis allows investors to systematically dissect a fund's portfolio to help explain performance. Combining attribution with qualitative knowledge of managers' portfolios and strategies uncovers a complete picture of the underlying sources of their returns.
Attribution analysis is commonly used by institutions to help explain which sectors and stocks contributed to a fund's performance. Morningstar recently added this capability to its institutional software tool, Morningstar Direct, and Morningstar's analysts plan to start harnessing its insights as we dig into fund performance. For our first foray into attribution analysis, we decided to look at large-cap funds that are Morningstar Analyst Picks and that have the S&P 500 as their benchmark. The large-cap universe is a notoriously competitive and difficult place for funds to gain an edge, making it a prime candidate for deeper study of what factors explain performance. To tune out the noise that comes with short periods, we looked back over the past five years ended June 30, 2009, and compared each fund with the S&P 500 to see which stocks and sectors helped and hurt its performance. We dropped the index and enhanced-indexed picks from the study as well as active funds that use a benchmark other than the S&P. Finally, we excluded Clipper Fund CFIMX because it changed advisors in the middle of the period under study. Our short list focused on 18 actively managed funds.
We started off our analysis by comparing each fund's sector weightings with the index and then drilled down to see which stocks had the biggest effect on the fund's results. We then measured how much value (in percentage terms) over the index the managers added in their selections of stocks and sectors.
There are a few limitations worth noting. Morningstar's attribution analysis is based on public quarterly portfolio data and does not consider every trade a fund made, making it an imperfect tool for precise accounting of a fund's successes and failures. Therefore, the reliability of our analysis declines for higher-turnover funds such as Brandywine Blue BLUEX and Mainstay ICAP Select Equity ICSLX. In addition, we don't know which came first, the stock choice or the sector theme, so we can't be too precise about where one ends and the other begins. (View the related graphic here.)
Stocks, Not Sectors, Rule
Most managers of our large-cap Analyst Picks focus on picking stocks as opposed to making big sector bets or market-timing calls.
Fourteen out of the 18 funds we examined succeeded over the past five years by holding large positions in stocks that outperformed the S&P 500.
Fairholme bet big on Canadian Natural Resources CNQ, for example, and locked in gains as the stock shot higher while sitting in the portfolio's top position. Berkowitz drastically trimmed the fund's stake in Canadian Natural Resources at an opportune time. It ended up tanking in the summer of 2008 not long after Berkowitz trimmed the position to less than 5% from more than 15%. Although not as concentrated as Fairholme, Vanguard Primecap VPMCX struck it big with Potash POT and Monsanto MON. Just as important as what they did own was what they avoided. Sequoia SEQUX, for example, came up big for its stock selection. It didn't have any huge wins on individual stocks but managed to avoid the big disasters.
The funds that score poorly for their stock picks were sunk by some of their highest-conviction holdings. For example, Oakmark Select's OAKLX Washington Mutual position overwhelmed all else over the past five years. Its losses on WaMu mounted as the fund kept more than 10% and sometimes more than 15% in that one stock, which went on to lose almost all of its value by the time the fund sold it. The danger of its concentration is especially clear when you compare Oakmark Select's performance and attribution results to Oakmark OAKMX, which is also managed by Bill Nygren and follows a similar albeit less-concentrated approach. Select's WaMu losses far outweighed gains in other stocks such as Yum Brands YUM and McDonald's MCD. Other Analyst Picks were hurt by big holdings. General Motors GM and Dell DELL weighed heavily on Longleaf Partners LLPFX. Selected American Shares SLADX and Davis New York Venture NYVTX were stung by flailing financials such as American International Group AIG and Citigroup C.