Friday, March 28, 2008

Madame X: U.S. Mid-Cap Funds

Hopefully this series isn't too much like water-torture :) I just don't feel like looking up more than 3 or 4 funds at a time and anyway comparing like funds with like is more instructive I think.

(Yahoo treats these funds as being in various peer categories like mid-cap blend, mid-cap growth etc - so the ranks are not strictly comparable across these funds)

These mid-cap funds are similar to the small caps in that the mid cap sector has declined in performance over the period relative to the S&P 500 which Yahoo uses as a benchmark. Over the full ten years all the funds outperform this index with the margin of outperformance generally declining over time. FSEMX is an index fund for the Wilshire 4500 and if we add back on the expense ratio we should have a good idea of the performance of the index. However, in a departure from the small-cap case, three of the managed funds are inferior to the index fund at least in the last 5 years as shown by both crude and risk-adjusted performance measures. In the last three years all the non-Fidelity funds are in the lower half of the performance ranks, with ARGFX being particularly poor. VSEQX did OK over the last three years, but based on beta appears to be taking more risk than CAAPX and has underperformed in the last year. On the other hand FMCSX looks to be a fairly decent fund, though Yahoo rank it worse than 71% of funds in its category in the last year, which is hard to believe.

I'd certainly look at getting rid of ARGFX. Unlike the other categories I've already examined you have more money in the poorer funds (ARGFX and CAAPX) here and less in the better ones (FSEMX, FMCSX). I'd look at rebalancing this at least to have equal amounts in CAAPX, FMCSX, and FSEMX.

Though VSEQX has lost 13% in the last year, assuming you put money in before this year too, it can't have lost 16% as reported. The cost basis must be at fault. You have a lot in ARGFX which has lost net for the last three years - it's quite likely you've lost money here. It's possible in CAAPX too, depending on how long you've been investing. But unless you recently switched in, your absolute loss shouldn't be high.

4 comments: said...

There seems to be considerable debate over whether or not your typical actively managed fund add more performance than the fees deduct. eg. less than half the actively managed funds perform better than the relevant index after adjusting for fees.

This then leads into the question of whether it's possible to "pick" the active fund managers who actually have exhibited skill rather than luck to achieve outperformance. There seems to be a LOT of luck involved in performance results. I don't recall the exact stats, but some studies reported along the lines of 'for any period (1, 3, 5 years or whatever) out of the top 1/3 of funds in the past period, in the next period 1/3 of them will continue to outperform, 1/3 will be within the middle tertial, and 1/3 will drop to the bottom 1/3 of the performance table).' Same sort of thing applies to the mid or bottom 1/3 of fund managers. ie. most performance is due to chance rather than skill. And it's very hard to pick the handful of outperformers than actually have skill.

The next problem is that if you do identify funds that have exhibited (in the past period) some actual skill, you need to establish that the same conditions will apply to the fund going forward eg. similar economic conditions that suit their style, that they haven't altered their investment style, that they haven't had major changes in key staff etc. etc.

Overall, my point is that even if you establish which funds have caused poor portfolio performance in the past, I'm not sure you can improve things going forward by trying to identify and eliminate the losers. Even the more modest proposal of "rebalancing" away from losers and towards the winners seems counter to usual "rebalancing" theory. You normally expect performance to revert to the mean, so you sell winning assets and shift funds to the underperforming assets to get back to your initial asset allocation. You seem to be proposing the exact opposite - selling the losers and chasing winners by adding extra weight to those that have outperformed in past periods.

mOOm said...

Well, there is also the concept of cutting losses :). I doubt there was any rationale to the amounts invested in these various funds originally. So I'm suggesting to start off again from an equal allocation and then rebalancing from there. Let's say that Madame X had $200k of new money to invest - would you advise her to invest larger shares in funds which had performed poorly recently? There might be an argument to invest more in asset classes that had performed poorly recently. I'm not suggesting here to chase asset class performance. With a 10 year record it seems not so hard to identify some particularly poorly performing managers. Maybe they'll improve, or maybe they won't. I doubt they'll suddenly become stars.

Obviously there is more than one way to look at this. said...

I've moved away from trying to pick actively managed funds and towards index funds over the years, due to generally lacklustre performance (ie. I don't have a good track record of picking skilled fund managers) and an annoying tendency for funds to get closed when they've had a couple of bad years (Fund families use survivor biasing to make their products performance histories look better than it is). When that happens to a fund that you've held for a long period it forces you to crystallise capital gains at times that may not suit for tax planning.

Perhaps a strategy would be to start out with a new selection actively managed funds, and then shift funds out of poorly performing managers into index funds that cover that asset sector during an annual rebalance/review. That way you'd sell off losers and replace them with guaranteed 'average' performance, while letting the winning fund investments run (until they lose their shine).

An alternative strategy is what the index funds like to call "core and satellite" investing, where you put a large chunk of your investment into index funds, and then add in selected actively managed funds (or direct stock picks) to try to add some outperformance. That way you shouldn't do significantly worse than average over the long run, and have some chance of boosting returns if you managed to pick winning stocks or fund managers.

mOOm said...

yes, this alpha + beta strategy makes a lot of sense.