Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
If I wanted to measure a specific fund against an index, how do I decide which index to measure against? Do you ever measure a mutual fund against an ETF? How do you decide which ETF best fits?
It really depends on what is the appropriate benchmark for the fund. What is in the portfolio? Which fund are you interested in? Likely in the prospectus they will tell you which benchmark they measure themselves against. It is unlikely that an ETF would be an appropriate benchmark, but we'd have to know which fund you're thinking about. Please enlighten us.
It depends on the fund and also what you want to do with it.
As randy pointed out above, the usual way is to see what benchmark is stated in the fund prospectus/website. Presumably that is what the fund managers are using to measure their own performance.
But sometimes a fund will advertise itself as compared to one benchmark, even though it is using a completely different strategy. If you go to Morningstar and get a fund's profile, click on the "Ratings & Risk" tab and look under MPT Statistics. You'll see a listing for "Best Fit Index." That will tell you which index is statistically the most similar to this fund.
Sometimes I use an index based on the fund's intended role in my portfolio. For example, it's worth comparing every stock fund with the S&P 500 or total stock market index. If I am not doing better than these indexes, or if I am not getting the expected diversification benefits, then that's a good reason to rethink my strategy going forward.
There is often no one right answer to this question. We often read the most recent semi-annual report for a fund to see what the management is using for comparison. Often there are two indexes, and the S&P 500 is usually one of them. But keep in mind that this is included mostly because folks use that bogey as a general comparison tool. It's usually the other index included in the report that is more meaningful for comparison purposes. When you have sector or targeted global thematic funds, however, it may be difficult to find information on the index the management uses. Quite often the index does have a comparable ETF, and that can be helpful, since the index itself may be hard to track. For example, if you own an actively-managed global infrastructure fund, you can campare it against the S&P Global Infrastructure Index in the guise of IGF pretty quickly. Sometimes the most obvious index may not be the correct one, and the fact that M* often puts funds in incorrect asset classes does not help.
All these categorizations are out-of-date by definition, since the portfolio holdings are always based on old info. (obviously index funds & ETF's are different.) It has been said that the only way that an active fund manager can beat his peers is by "style drift". Schwab categories are terrible since they only have large cap and small cap and don't even consider mid cap. M* is somewhat better, but I prefer Lipper's methodology. But all this tends to be a mute issue for me since I prefer multi-cap blend/core funds that can go anywhere. I feel this gives the active manager the most flexibility to beat his peers. So I have decided that I am happy if all my funds beat just VFINX. That is my baseline standard. If any "domestic equity fund" does not beat VFINX over any time period longer than 3 years, then it is gone.
Building on claimui's suggestion of using the benchmark that the fund uses - I think this is often the best approach. Especially for funds that adjust their fees based on a benchmark. They're not going to pick a benchmark that doesn't reflect their fund, unless they change the benchmark as the market shifts (to keep picking a low bar to exceed); I don't think funds are that dishonest (at least not so blatantly).
Regarding style drift (Greg and claimui) - this drives some people and some companies mad. S&P tries to compare a fund's performance with whatever its behavior most closely resembles. Their fund graphs show a hypothetical breakdown (large cap, bond, etc.) that would sort of match how the fund's behaving. This seems like an attempt to eliminate asset allocation (style) as a factor and focus exclusively on individual security selection.
I don't agree. Asset allocation (or sector, or cap, or whatever) is part of the management skill, and you don't take that out of the equation when evaluating performance. Then the trick is how to correctly define its universe (I want a benchmark for that universe, regardless of which corner of that universe the fund is currently occupying by choice). For example, if you look at the Mutual Series Funds prospectuses, you'll see something that looks like a wild west, full abandon fund. And they do invest heavily in things like distressed companies. But they don't go wandering off through all possible investment, despite the prospectus giving them broad discretion. So their real universe is narrower than stated, but still much broader than what they happen to be invested in today.
In this respect, I think Lipper categories help identify a benchmark - these categories give a better, finer picture of where the fund generally wanders. But I'm not as enthusiastic about using those categories for peer comparisons - some of the categories are so narrow that there are not enough peers for a meaningful comparison.
I guess I'm giving a description of how one can use a seat-of-the-pants approach to identify a reasonable benchmark. (And I get to define reasonable
I am one of those who does not get excited about style drift, since I am hiring managers, not buying funds. If the managers I hire stick to their overall investment philosophy and strategies, I really don't care whether they drift from small to mid, or mid to large within reason. Like Greg, I actually like great managers that have the ability to go anywhere. Unfortunately fund companies often put restraints on the funds themselves that limit that option for their managers. That's probably why we started using alternative strategy funds years ago. For this reason, I think it's usually best to go with the index the manager uses to compare the fund's performance.
A digression of thread - Speaking of Mutual Series Funds - they seemed to be alot less volatile when Max Heine was at the helm. I think it got progressively worse through each manger change. What do you think of MUTHX Mutual Shares and MDISX Mutual Discovery these days. (not worrying about share class or load )
It would be interesting to see what David thought about the mutual series "family"
Comments
As randy pointed out above, the usual way is to see what benchmark is stated in the fund prospectus/website. Presumably that is what the fund managers are using to measure their own performance.
But sometimes a fund will advertise itself as compared to one benchmark, even though it is using a completely different strategy. If you go to Morningstar and get a fund's profile, click on the "Ratings & Risk" tab and look under MPT Statistics. You'll see a listing for "Best Fit Index." That will tell you which index is statistically the most similar to this fund.
Sometimes I use an index based on the fund's intended role in my portfolio. For example, it's worth comparing every stock fund with the S&P 500 or total stock market index. If I am not doing better than these indexes, or if I am not getting the expected diversification benefits, then that's a good reason to rethink my strategy going forward.
Regarding style drift (Greg and claimui) - this drives some people and some companies mad. S&P tries to compare a fund's performance with whatever its behavior most closely resembles. Their fund graphs show a hypothetical breakdown (large cap, bond, etc.) that would sort of match how the fund's behaving. This seems like an attempt to eliminate asset allocation (style) as a factor and focus exclusively on individual security selection.
I don't agree. Asset allocation (or sector, or cap, or whatever) is part of the management skill, and you don't take that out of the equation when evaluating performance. Then the trick is how to correctly define its universe (I want a benchmark for that universe, regardless of which corner of that universe the fund is currently occupying by choice). For example, if you look at the Mutual Series Funds prospectuses, you'll see something that looks like a wild west, full abandon fund. And they do invest heavily in things like distressed companies. But they don't go wandering off through all possible investment, despite the prospectus giving them broad discretion. So their real universe is narrower than stated, but still much broader than what they happen to be invested in today.
In this respect, I think Lipper categories help identify a benchmark - these categories give a better, finer picture of where the fund generally wanders. But I'm not as enthusiastic about using those categories for peer comparisons - some of the categories are so narrow that there are not enough peers for a meaningful comparison.
I guess I'm giving a description of how one can use a seat-of-the-pants approach to identify a reasonable benchmark. (And I get to define reasonable
A digression of thread - Speaking of Mutual Series Funds - they seemed to be alot less volatile when Max Heine was at the helm.
I think it got progressively worse through each manger change. What do you think of MUTHX Mutual Shares and MDISX Mutual Discovery these days. (not worrying about share class or load )
It would be interesting to see what David thought about the mutual series "family"