Lewis Braham: The Best Mutual Funds For Investors: Cheap And Boring

Ted

FYI: Asset managers are hawking free products and services as if they were carnival barkers. JPMorganoffers free stock trades; Fidelity boasts about its zero-cost index funds; Vanguard is waiving trading costs for exchange-traded funds. The appeal is simple: Costs—whether incurred by buying and selling a fund, or imposed through an annual expense ratio—eat into returns. But there’s another, surprising benefit: Cheap funds may save you from your worst impulses.
Regards,
Ted
https://www.barrons.com/articles/best-mutual-funds-for-investors-cheap-and-boring-1535145075

msf

(Curiously, I was able to read it directly; no tricks, no special software. Though I am on a new machine, and Barron's might be counting articles.)

A nit to pick: FBTCX and FBIOX are not different share classes of the same fund, but shares of two different, albeit it similar funds. (For example, they have a different #3 holding as of 6/30/18.)

I agree that which share class you look at likely does affect investor return data. As I commented in this thread, the selling point for C shares is that they are supposedly better for people who only want to own the shares for a year or two. (Or even less.)

So it might be interesting to look M (formerly T) shares or A shares of Fidelity Advisor Biotech. Since M* doesn't have 10 year investor data for A shares ($787million), I pulled up the investor data for M shares ($127million). Despite the small share class size.

Small funds (under $500K) were excluded in the article because they are more likely to have erratic cash flows making their performance too volatile. All well and good.But that's not a reason to exclude share classes of large funds. The small size of a share class doesn't affect the fund performance. If anything, because a share class is small, its investors will have a smaller impact on the fund performance.

Over the ten year period ending 6/30/18, M* reports:
FBIOX: 12.73% (investor return), 17.15% (fund return) - matches Barron's reported M* data
FBTTX: 11.88% (investor return), 15.86% (fund return)
FBTCX: 5.97% (investor return), 15.35% (fund return) - matches Barron's reported M* data

Similar gaps between investor returns and fund returns for FBIOX and FBTTX. So at least here, it doesn't look like the existence of a load or the popularity of a fund matter. What might matter more is whether the shares are designed/marketed for shorter term trading (such as C shares).

More generally there are various confounding factors that aren't sorted out. Do investors do well in cheaper funds because they have more patience when the fund costs less (as speculated)? Or perhaps it is because "high expense funds have much more volatile risk-adjusted returns", and it's just the volatility that affects investor behavior?

Livingston, Zhou, 2016, The Volatility of Mutual Fund Performance, http://www.fmaconferences.org/Vegas/Papers/QUANTILE-12-31-2015.pdf

hank

“Asset managers are hawking free products and services as if they were carnival barkers.”

One carnival barker is enough. Thank you.

LewisBraham

@MSF The sentence in the story is:

We included only fund share classes with over $500 million in assets, as the money flows in tiny funds are too volatile.

I am not saying that flows affect the fund's performance, but that in a smaller shareclass of a fund, flows can seem far more erratic because one or two large shareholders of that particular shareclass can have a disproportionate effect on flows. That means flows and thus investor returns may not reflect the experience of the average shareholder but that of just a handful of shareholders driving this small shareclass. These flows do not affect the overall performance of the fund--at least not in an easily measurable way--but have an impact on investor returns in that particular shareclass. It is possible that a small group of shareholders could dramatically affect flows in larger shareclasses of funds, but it is less likely the bigger the shareclass is.

msf

I'd thought you might have meant that, except for the portion: "the money flows into tiny funds are too volatile".

If the concern really were with small funds, then excluding small share classes would have been a quick and dirty way to ensure that you excluded all small funds. (A small fund could not have a large share class.) That interpretation is consistent with the literal words, though not with the intended meaning.

The similarity of investor gaps in FBTTX and FBIOX suggests that a few larger M class investors aren't excessively influencing average dollar returns of M class shares.

In any case, to own even 1% of class M shares would mean owning over $1M (of a $127million share class). At that level, loads for both M class and the cheaper (lower ER) A class are waived. So I don't think you're going to find many 1%+ owners of M class shares of this particular fund.

None of this says that for funds in general it isn't a good idea to exclude small share classes. Just that for this particular fund, looking at M class investors should be okay.