Search This Blog

Friday, April 27, 2007

When the Cat’s away….the replacement ruins the column…. And Presents Some Nonsense from Morningstar

Jonathan Clements’ Getting Going column in the WSJ is generally a good source of advice on personal finance, but it seems the minute he takes some time off, the gremlins sneak in and fill his column with nonsense. In this week’s paper we got this column. I’m still trying to stop my head from spinning from the mass of illogic that is presented. One thing is for sure, one can hardly conclude from the article that there really is any time when it makes sense to purchase and actively managed mutual fund.

Here we go…

When It Makes Sense to Buy
An Actively Managed Fund

April 25, 2007; Page D1

Fund managers who aim to beat the broad stock market are having a tough time these days.

Thanks to the proliferation of exchange-traded funds, low-cost index-tracking vehicles now cover nearly every sector of the stock and bond market. And the strong performance of broad-market benchmarks last year made many managers look like laggards. …. Now, one of investors' most-cherished mantras about active managers is being challenged by new research.

… two recent studies suggest the notion is based on flawed comparisons between active managers and indexes.

The author then presents Morningstar’s convoluted system for measuring the performance of actively managed vs. index funds. A close examination of this system shows it is basically useless as an aid to constructing portfolios.

Here’s the new Morningstar “system”

A large-cap value index, for example, is composed mostly of large-cap value stocks -- those that look cheap based on measures such as price-to-earnings ratios. But the average large-cap-value fund is composed of only about 40% large-cap-value stocks, according to investment-research firm Morningstar Inc. The rest is a hodgepodge of other large, midsize and small companies -- including growth stocks.

Fair enough, but why is that a good thing?

Basically what the folks at Morningstar are telling us is that the label on an actively managed mutual fund is meaningless. The average large cap value fund is only about 40% large cap value In other words one has absolutely no idea what a particular fund will own relative to its label. So one could hold a portfolio evenly divided between a large value and a small value actively managed mutual fund in the quest for a diversified portfolio but in fact be holding say only 25% large value stocks.

Furthermore, whatever the average allocation of large value stocks is at a given time, it gives no indication of what it will be like in the future. In fact since almost all mutual funds report holdings on only a quarterly basis, the Morningstar analysts are driving with their eyes firmly locked on the rear view mirror. They are analyzing a firm in the present based on fund holdings in the past in the quest to say something about performance in the future. Doesn’t sound too useful to me.

By contrast, an index fund or etf makes public the criteria for including stocks in its portfolio and seldom if ever changes those criteria (and if it does publicizes it beforehand) the fund will not “drift “ away from the index specified in the prospectus to some ever changing “custom benchmark”.

And in the bizarre logic of Morningstar the following is a positive:

When compared against a standard benchmark, the large-value managers look sluggish -- only 15% of them beat the Morningstar Large Value Index over the five years ended Dec. 31. But those results might tell us more about the fact that value trounced growth during that period than about active managers' skill.

But when constructing an investment policy the above is a good, not bad outcome. The worst thing that can happen is that value beats growth and your value fund underperforms because the manager of your actively managed value fund decided it was time to load up on growth stocks. This is called “style drift” and it is a key reason why actively managed funds are so bad for portfolio construction. There is no truth in labelling you do not know what you own.

In Morningstar’s endless quest to create new products and support the myth of active management we get the following:

To better compare indexes against active managers, Morningstar built benchmarks that more accurately reflect how stock pickers invest. Its custom large-cap-value index, for example, is composed of about 40% large-cap-value stocks.

I think the Morningstar folks must have spent some time working for a government agency and probably one in the old Soviet Union to come up with this oxymoron: a custom large value index that is 40% large value stocks. It strikes me as similar to a student designing his own GPA that only includes the grades from 40% of his classes.

And even after all these convoluted techniques to improve the odds for the active funds the results hardly justify the article’s headline:

Still, in a study by Morningstar director of fund research Russell Kinnel, stock pickers proved decidedly so-so against the custom benchmarks in nine domestic stock-fund categories, beating the indexes just 35% to 47% of the time. Those large-cap-value managers beat the tailor-made index 38% of the time -- a more-respectable but hardly awe-inspiring record.

Great news! (not) Even when creating an index based on what the fund actually owns rather than its label (information the potential buyer would only have in retrospect, not at the time of buying the fund) most actively managed funds fail to beat this “custom” (imo meaningless) index.

And talk about a convoluted means of burying the real info, take a look at this:

A recent study of active small-cap managers by index-fund giant Vanguard Group found a similar pattern. When compared against the Russell small-cap indexes, the long-term performance of the active managers looks pretty good. But the results change substantially against some other widely used benchmarks, Vanguard found.

The managers' performance -- even before fees -- is close to Morgan Stanley Capital International's small-cap indexes, for example. But after accounting for fees, the managers lag the indexes.

And when all else fails we get the “even a broken clock is right twice a day” argument for buying actively managed funds.

Yet strong active managers can be found in almost any market segment. The Morningstar study compared mutual funds' five-year annualized returns against their "best fit index," or the index that most closely correlates with their returns. Sure, the winners included small-cap funds like Buffalo Small Cap and Baron Small Cap. But they also included TCW Dividend Focused, a large-cap-value fund, and Fairholme, a midcap blend portfolio.

But again this “best fit index" methodology is useless. Let’s say you bought the Baron Small Cap fund for the small cap portion of your portfolio. Small cap stocks did well but the baron fund underperformed the small cap index. Not to worry, say the Morningstar folks, Baron Small Cap was 50% large cap/50% small cap and compared to a 50% large cap 50% small cap “custom index” the Baron fund outperformed. The problem is that your intended allocation was 50% large cap and 50% small cap which you established by buying 50% of a large cap index etf and 50% of the Baron Small Cap fund. But due to the strategy of the Baron Small Cap fund your portfolio was 75% large cap 25% small cap.

Furthermore as we have pointed out many times here on this blog, there is very little evidence that past performance of a mutual fund is a predictor of future performance. Using the” custom index” doesn’t perform any more of a usable benchmark. In fact it makes things more complicated:

In order for the information on funds that have beaten their custom benchmark as a successful tool for constructing a portfolio that would perform well going forward the following would have to occur:

  1. The active fund manager continues to be a good stock picker


  1. The active fund manager’s choice of sectors is better than the strategy indicated by the fund’s name (i.e. even though it is a small cap value fund, the manager bets correctly against small value by owning lots of large cap growth stocks)

If both of the above don’t turn out to be the case, the investor will likely be an unhappy investor in the active fund.

And of course constructing a portfolio allocation diversified across style and market cap is impossible to do with these funds, you simply don’t know what you own.

Here is one point on which I will agree with those that have faith in active funds. If for some reason (personally I can’t think of any) you are going to buy an actively managed fund, you might as well buy one that explicitly says that it can buy any kind of stock it wants. At least that way you know that you have no idea what the fund will actually own and you are just betting that you have given your money to a “genius” to invest.

Then the writer trots out the usual personal finance platitudes:

Instead of picking active managers based on sector, you should look for low costs and a long-term, consistent track record. Since fund expenses reduce returns, lower fees mean managers must clear a smaller hurdle to beat the index.

I’ll leave it to John Bogle in his latest book to explain why the long term past track record of a fund is a not very good predictor of past performance. As for picking a lower cost active fund vs. a higher cost one, no doubt that is true but it still is unlikely to be the best choice.

The article closes with this bit of nonsense:

As the editor of a newsletter about Vanguard funds, investment adviser Daniel Wiener knows a thing or two about indexing (actually as the editor of an investment newsletter he knows he won’t sell too many newsletters if he just tells investors to set up an indexed allocation and stick to it). But he doesn't use index funds as core holdings for clients because he's confident he can find active managers who will beat the market.

So the clients of Mr. Wiener’s newsletter take a triple leap of faith: that there are actively managed funds that reliably outperform the market, that Mr. Wiener will find them, and that when the chosen active fund managers lose their geniuses status he will know when to pull the plug and move on to another fund. Let’s just say this doesn’t seem like a strategy with a high likelihood of consistently beating a disciplined diversified indexed investor.

Here’s some more sage advice from Morningstar:

Consider buying the active funds that you have "the greatest confidence in," suggests Mr. Kinnel, "and use passive index funds in the slots where you can't find anything super-compelling."

And when your confidence gets shaky (based on what exactly: a bad quarter? a bad year?) do you move on to a new “super compelling” fund (and on what basis exactly does he recommend we chose that one?)