How would one analyze the claim "American Funds Outperform"

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
bnes
Posts: 404
Joined: Thu May 22, 2014 12:29 am
Contact:

How would one analyze the claim "American Funds Outperform"

Post by bnes »

As has been well discussed here, American Funds have issued a whitepaper claiming an "active advantage" (See http://www.bogleheads.org/forum/viewtop ... &p=1827688 ).

The paper essentially takes the thesis that while mutual funds overall underperform index investing, that it's possible that a select group of funds has bested the market. And naturally the group of American Funds are among those rare outperformers. Now: wow would one analyze that claim? Not critique the sloppy methods of the paper, but rather analyze the assertion from the ground up?

When I look at 5 and 10 year "growth of $10,000" type charts, the American Funds do indeed tend to come out ahead, even after fees. The American Funds seem to loose in tax efficiency, and sales load. But not everybody pays those, thus the analysis would have to allow plugging in:

* Load to be paid, or 0% if the load is a sunk cost
* Applicable expense ratio, which might be quite low in a IRA.
* Applicable tax rate, or 0%.

Personally my hunch is that American Funds won't continue any outperformance, because style based outperformance does not really exist in a world of arbs and computer trading. Analysis based outperformance might exist: looking forward to sectors that will do well, but it's also dangerous, tricky, and mostly not what AF does.
psteinx
Posts: 5801
Joined: Tue Mar 13, 2007 2:24 pm

Re: How would one analyze the claim "American Funds Outperfo

Post by psteinx »

Well, the claim has two slightly different flavors/nuances:

1) AF *HAS* outperformed
2) AF *WILL* outperform (going forward)

In turn there are many further variables and assumptions you'd need to really consider either claim, and they'd both be difficult for a typical home based prospective investor to consider thoroughly.

For claim 1:

* Gather a complete list and history of all AF funds (including any that have been closed), going back to their foundation, along with performance history.
* Determine what's the most appropriate benchmark for each fund individually and/or the funds as a group.
* Determine how to weight things (i.e. Weight the biggest funds most heavily? Weigh things evenly over time, or in some fashion tilt towards the more modern era?)
* Determine how to deal with issues of taxation, loads, risk, etc.
* Keep in mind that AF goes back to a time long before there WERE index funds. I doubt most individual investors in, say, the 1950s were approximating an index fund in their own investments. So how, if at all, should comparisons be adjusted for this?

For claim 2, there are many other complications that I won't get into...
Last edited by psteinx on Mon Aug 25, 2014 2:35 pm, edited 1 time in total.
User avatar
Taylor Larimore
Posts: 32842
Joined: Tue Feb 27, 2007 7:09 pm
Location: Miami FL

"A Tale of Two Fund Giants"

Post by Taylor Larimore »

How would one analyze the claim "American Funds Outperform?"
bnes:

This article in the Wall Street Journal will help answer your question:

A TALE OF TWO FUND GIANTS

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle
User avatar
Raybo
Posts: 2244
Joined: Tue Feb 20, 2007 10:02 am
Location: San Francisco
Contact:

Re: How would one analyze the claim "American Funds Outperfo

Post by Raybo »

Due to needing a broker to hold brokered CDs I had when I moved, I briefly had an Edward Jones account. When the sales person saw my portfolio, I got a lot of American Funds outperform propaganda. A cursory examination showed that they were comparing against the S&P500. I'm guessing that that was without dividends reinvested. Second, the funds they were comparing to the S&P500 were not just large cap funds, so the benchmark wasn't the propriety one.

Also, I'd look closely at the assumptions they make. How do they deal with the load? What about 12-b-1 fees?

Also, have funds been merged or split over the time period? Survivor bias might be at play.

Anytime a fund claims that their funds outperform, it is marketing hype or data mining, likely both. I would put it in the same category of the clams that car manufacturers make about their cars/trucks being number 1.
No matter how long the hill, if you keep pedaling you'll eventually get up to the top.
User avatar
ThePrune
Posts: 960
Joined: Wed Nov 10, 2010 8:38 am
Location: Midland, MI

Re: How would one analyze the claim "American Funds Outperfo

Post by ThePrune »

bnes wrote:Now: how would one analyze that claim? Not critique the sloppy methods of the paper, but rather analyze the assertion from the ground up?
If you are really interested in this topic, why not take a look at Bernd Fisher and Russ Wermer's book Performance Evaluation and Attribution of Security Portfolios.

If you'd like to take a quick look at the sorts of techniques that Wermers is advocating, read the article False Discoveries in Mutual Fund Performance: Measuring Luck in Estimated Alphas by L. Barras, O. Scaillet and Russ Wermers.
Investment skill is often just luck in sheep's clothing.
Post Reply