Has Larry Swedroe been fair on American Funds?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
amarone
Posts: 459
Joined: Sun Feb 15, 2009 6:39 am
Location: Atlanta

Has Larry Swedroe been fair on American Funds?

Post by amarone » Fri Apr 15, 2011 7:19 am

Larry presents one of his analyses of a fund company to see if they are adding value, and today's victim is American Funds: http://moneywatch.bnet.com/investing/bl ... alue/2264/. His conclusion:
In neither case do we have evidence of overall outperformance.
My assessment of the data is: American has outperformed in all asset types except emerging markets, and Larry creates a portfolio in which the EM shortcomings swamp the rest.

Of the funds presented, AF outperforms Vanguard indexes in 7 of 8 funds and outperforms DFA in 6 of 9.

There are no small-cap funds presented, which I would like to see in order to construct a portfolio. Larry uses the funds to construct a portfolio of 25% US large growth, 25% US large value, 25% International large cap, and 25% emerging markets. With this high exposure to EM (compared to what most investors would put together - for example, Larry himself has only 6.7% of the equity portion in EM in the sample portfolio here: http://www.kiplinger.com/tools/investme ... ct=Swedroe), the underperformance in EM cancels out AF's outperformance in the other three asset classes.

No VG fund is presented for international large cap, but if you include Total Int Stock Index as the closest thing (even though it has some small cap which should help VG) with a 10-year return of 7% and then construct what I would regard as a more reasonable portfolio (if we can't have any small cap) of 30% US LG, 30% US LV, 30% Int large, 10% EM, then what we get is:

American Funds: 6.6%
DFA: 6.2%
Vanguard: 5.7%

That tells a different story.

btw, I have no American Funds, nor do I have any actively-managed funds.

User avatar
Bruce
Posts: 355
Joined: Fri Mar 02, 2007 8:02 am
Location: Alaska

Post by Bruce » Fri Apr 15, 2011 7:28 am

American funds outperform in collecting ongoing 12B-1 fees enriching the sales person who sold them as long as they are held in someone else's portfolio.

They may not be as bad as some load funds sold with an ongoing marketing load, but the drain on investing return continues each year you own them.

With multiple share classes and different break points in fund load they are worthwhile for brokers to sell, but few would seek them out and buy them on their own.

However, they are not as bad as some funds with even higher loads and marketing expenses.
Bruce | | Winner of the 2017 Bogleheads Contest | | "Simplicity is the master key to financial success."

User avatar
Kenkat
Posts: 5514
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Post by Kenkat » Fri Apr 15, 2011 7:45 am

American Funds have relatively low expense ratios and they tend to be relatively low turnover as well. So - they should be expected to perform in line with index funds with similar attributes - and - surprise - they do.

The article is too focused on the management style (active vs. index), which is relatively unimportant and not focused enough on the attributes of what makes a fund good - expenses, turnover, adherence to style or philosophy.

They often carry loads, but the American Funds advisors must earn a living just the same as the DFA guys.

They are decent choices in my opinion.

ScottW
Posts: 156
Joined: Tue Mar 25, 2008 8:13 am

Post by ScottW » Fri Apr 15, 2011 8:05 am

I'm not sure the results show that American Funds have added no value, either--it looks like more of a stalemate. As Amarone mentioned, they beat both Vanguard and DFA in two out of four categories. As for emerging markets, New World isn't a "pure" emerging markets fund, and invests in companies that derive a "significant portion of the company’s assets or revenues (generally 20% or more)" from developing countries. Last year the fund only invested 44% in emerging markets stock and 9% in emerging debt. The remaining 37% was in developed countries.

This seems a case of "how can I interpret these results to make my point?"

MP173
Posts: 2028
Joined: Fri Dec 07, 2007 6:03 pm

Post by MP173 » Fri Apr 15, 2011 8:26 am

For years, my employer's 401k had American Funds. I used Europacific and Washington Mutual. Both worked out well for me, particularly Europacfic.

A few years ago, we had a new advisor and she eliminated Washington Mutual and moved it to another fund. I did the research and screamed loud...it didnt work.

Now, we have moved on to Fidelity and have a larger selection.

Look, not everyone has access to Vanguard, Fidelity, or DFA funds. Sometimes we are limited. I love the Vanguard and Fidelity funds and use them extensively (along with others), but American Funds has served me well in the past.

Ed

User avatar
ddb
Posts: 5509
Joined: Mon Feb 26, 2007 12:37 pm
Location: American Gardens Building, West 81st St.

Post by ddb » Fri Apr 15, 2011 8:30 am

kenschmidt wrote:The article is too focused on the management style (active vs. index), which is relatively unimportant and not focused enough on the attributes of what makes a fund good - expenses, turnover, adherence to style or philosophy.
Bingo. Indexing isn't special. "Low costs" is special.

- DDB
"We have to encourage a return to traditional moral values. Most importantly, we have to promote general social concern, and less materialism in young people." - PB

Epaminondas
Posts: 178
Joined: Tue Feb 01, 2011 4:52 pm
Location: Ohio

Post by Epaminondas » Fri Apr 15, 2011 11:07 am

ddb wrote:
kenschmidt wrote:The article is too focused on the management style (active vs. index), which is relatively unimportant and not focused enough on the attributes of what makes a fund good - expenses, turnover, adherence to style or philosophy.
Bingo. Indexing isn't special. "Low costs" is special.

- DDB
I thought indexing was supposed to be special as well. You don't have to worry about manager risk and it provides you with a broadly diversified fund. Isn't trying to be average part of the idea? And isn't active management antithetical to that idea?

User avatar
ddb
Posts: 5509
Joined: Mon Feb 26, 2007 12:37 pm
Location: American Gardens Building, West 81st St.

Post by ddb » Fri Apr 15, 2011 11:13 am

Epaminondas wrote:
ddb wrote:
kenschmidt wrote:The article is too focused on the management style (active vs. index), which is relatively unimportant and not focused enough on the attributes of what makes a fund good - expenses, turnover, adherence to style or philosophy.
Bingo. Indexing isn't special. "Low costs" is special.

- DDB
I thought indexing was supposed to be special as well. You don't have to worry about manager risk and it provides you with a broadly diversified fund. Isn't trying to be average part of the idea? And isn't active management antithetical to that idea?
I should have been more specific, by writing "Assuming sufficient diversification, indexing isn't special."

Take two funds, both from the same asset class - fund A is an index fund with 0.5% expense ratio, 1,000 holdings, and 15% turnover. Fund B is an active fund with 0.5% expense ratio, 300 holdings, and 15% turnover.

Which would you rather own?

Well, Fund A is preferable, but only by the slimmest margin. And it's not preferable because it's an index fund, but because it has more diversification. But we're really splitting hairs, because at 300 holdings, you've already effectively eliminated non-systematic risk.

Now, pretend that Fund B has a 0.35% expense ratio. Now which is preferable? The rational investor should go for the lower-cost option, and not the one that is badged with an "index" label.

The general failure of active management has to do with costs, and not with poor manager decisions.

- DDB
"We have to encourage a return to traditional moral values. Most importantly, we have to promote general social concern, and less materialism in young people." - PB

exeunt
Posts: 907
Joined: Tue Jan 08, 2008 6:54 pm

Post by exeunt » Fri Apr 15, 2011 11:13 am

American Funds is a family of egregious closet indexers, for the most part. The two that stand out are Washington Mutual and Growth Fund of America. Many of their funds had exceptional early 2000 performances, but turned into closet indexers to protect their records. AGTHX's 3-year R^2 versus its benchmark, the Russell 1000 Growth, is 99%, and it underperformed by nearly 3%. Washington Mutual's 3-year R^2 versus the S&P 500 is 98%, and its alpha -0.93%.

User avatar
Cosmo
Posts: 1244
Joined: Mon Mar 05, 2007 9:46 pm

Post by Cosmo » Fri Apr 15, 2011 11:29 am

ddb wrote:
The general failure of active management has to do with costs, and not with poor manager decisions.

- DDB
If the manager was able to make even marginally good decisions on a CONSISTENT basis over time, then they would be able to easily overcome these expenses.

Roy
Posts: 970
Joined: Wed Sep 10, 2008 9:34 am

Post by Roy » Fri Apr 15, 2011 11:45 am

Epaminondas wrote: I thought indexing was supposed to be special as well. You don't have to worry about manager risk and it provides you with a broadly diversified fund. Isn't trying to be average part of the idea? And isn't active management antithetical to that idea?
It is primarily about the comparative costs. And one can certainly question whether they wish to take manager risk at all, even assuming comparable costs and reasonable diversification.

User avatar
CrankyManager
Posts: 243
Joined: Wed May 12, 2010 10:12 am
Location: Nebraska

Post by CrankyManager » Fri Apr 15, 2011 11:54 am

I would prefer my employer's 401k held American Funds rather than the mish-mash of high ER novelty funds that we have now.

I probably wouldn't have any problem with holding select American Funds, even in my own accounts (but I don't).

My fiancee's TIAA-CREF has access to AF EuroPac, and we use it to provide international exposure. It has a great performance track record compared to the other fund that's available, and the EuroPac ER for that share class is actually sane.

There's quite a few things to admire about American.

They don't chase fads, they don't have a star manager that's always on the verge of leaving, they're easily accessable ($250 minimum investment), funds with solid, long-term returns, and reasonable (for load funds) fees.

Would I open up an account today? If I had to go through an advisor, probably. But since I don't, no.
"Does not Dionysius seem to have made it sufficiently clear that there can be nothing happy for the person over whom some fear always looms?" -- Cicero

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

helps to have facts right

Post by larryswedroe » Fri Apr 15, 2011 12:01 pm

American funds underperformed in 2 of the 4 asset classes, not just one. US LV was the other besides EM.

And in one the international large the outperformance may all have come from their exposure to EM, not in the developed market asset class,. So it might have been three out of four--don't know. But we do know the comparison as shown unfairly rewards American. So I was being more than fair to them

If there were ten year periods for comparable funds would have shown them for other asset classes

Relative to other active funds American funds are good, but that is as they say damning with faint praise.

And in addition to the failure to show outperformance we have the risks of style drift.

Epaminondas
Posts: 178
Joined: Tue Feb 01, 2011 4:52 pm
Location: Ohio

Post by Epaminondas » Fri Apr 15, 2011 12:03 pm

ddb wrote:The general failure of active management has to do with costs, and not with poor manager decisions.

- DDB
Thanks for your response. I thought that it was both though... I thought that one of the things you have to watch out for with active is not just the costs or the lack of diversification but the fact that at least half the managers are bound to fall below the average. Isn't that part of it? Personally, for my equity investments, I just want to own as many different companies as possible of all different sectors and sizes. I don't want the risk of anyone else thinking about what "might" be better to own at this particular time.

User avatar
ddb
Posts: 5509
Joined: Mon Feb 26, 2007 12:37 pm
Location: American Gardens Building, West 81st St.

Post by ddb » Fri Apr 15, 2011 12:27 pm

Epaminondas wrote:
ddb wrote:The general failure of active management has to do with costs, and not with poor manager decisions.

- DDB
Thanks for your response. I thought that it was both though... I thought that one of the things you have to watch out for with active is not just the costs or the lack of diversification but the fact that at least half the managers are bound to fall below the average. Isn't that part of it? Personally, for my equity investments, I just want to own as many different companies as possible of all different sectors and sizes. I don't want the risk of anyone else thinking about what "might" be better to own at this particular time.
Active managers fall below the average because of the costs of implementation, not because of any inherent property of active management. Costs matter, security selection does not.

- DDB
"We have to encourage a return to traditional moral values. Most importantly, we have to promote general social concern, and less materialism in young people." - PB

User avatar
Kenkat
Posts: 5514
Joined: Thu Mar 01, 2007 11:18 am
Location: Cincinnati, OH

Re: helps to have facts right

Post by Kenkat » Fri Apr 15, 2011 12:37 pm

larryswedroe wrote:American funds underperformed in 2 of the 4 asset classes, not just one. US LV was the other besides EM.

And in one the international large the outperformance may all have come from their exposure to EM, not in the developed market asset class,.
You could probably say a similar thing about the DFA Value Fund - its outperformance may have come from its greater value loading. Do we throw that result out also?
larryswedroe wrote:So it might have been three out of four--don't know. But we do know the comparison as shown unfairly rewards American. So I was being more than fair to them
If you don't know if it's three out of four, you don't know. You can't then draw the "more than fair" conclusion. Using the DFA Value example above, it could be one out of four. Not a valid conclusion either.
larryswedroe wrote:Relative to other active funds American funds are good, but that is as they say damning with faint praise.
I really don't think this was substantiated by the data. Overall, they performed right in line with what would be expected relative to their cost profile. Active management neither added to or subtracted from performance in any appreciable way.

User avatar
Cosmo
Posts: 1244
Joined: Mon Mar 05, 2007 9:46 pm

Post by Cosmo » Fri Apr 15, 2011 1:12 pm

ddb wrote:
Epaminondas wrote:
ddb wrote:The general failure of active management has to do with costs, and not with poor manager decisions.

- DDB
Thanks for your response. I thought that it was both though... I thought that one of the things you have to watch out for with active is not just the costs or the lack of diversification but the fact that at least half the managers are bound to fall below the average. Isn't that part of it? Personally, for my equity investments, I just want to own as many different companies as possible of all different sectors and sizes. I don't want the risk of anyone else thinking about what "might" be better to own at this particular time.
Active managers fall below the average because of the costs of implementation, not because of any inherent property of active management. Costs matter, security selection does not.

- DDB
If you are actively managing a portfolio, security selections absolutely do matter -not just in selection but also timing of sale/purchase etc.

User avatar
ddb
Posts: 5509
Joined: Mon Feb 26, 2007 12:37 pm
Location: American Gardens Building, West 81st St.

Post by ddb » Fri Apr 15, 2011 1:25 pm

Cosmo wrote:
ddb wrote:
Epaminondas wrote:
ddb wrote:The general failure of active management has to do with costs, and not with poor manager decisions.

- DDB
Thanks for your response. I thought that it was both though... I thought that one of the things you have to watch out for with active is not just the costs or the lack of diversification but the fact that at least half the managers are bound to fall below the average. Isn't that part of it? Personally, for my equity investments, I just want to own as many different companies as possible of all different sectors and sizes. I don't want the risk of anyone else thinking about what "might" be better to own at this particular time.
Active managers fall below the average because of the costs of implementation, not because of any inherent property of active management. Costs matter, security selection does not.

- DDB
If you are actively managing a portfolio, security selections absolutely do matter -not just in selection but also timing of sale/purchase etc.
For the typical actively-managed fund where non-systematic risk is virtually eliminated (say, 100 plus holdings, no position comprises more than 5% or so of the fund), security selection doesn't matter over the long-run. Long-term returns will be based on risk factor exposure and costs. The actual securities being purchased/sold/traded play an extremely minor role in performance.

- DDB
"We have to encourage a return to traditional moral values. Most importantly, we have to promote general social concern, and less materialism in young people." - PB

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

ken

Post by larryswedroe » Fri Apr 15, 2011 3:51 pm

First, if you want to see if fund added alpha then one would need to perform a three factor regression. While it is possible that DFA outperformed in LV due to higher loading, it is also possible that American funds in the large growth/blend category outperformed because they had more exposure to smaller stocks or less to growth stocks.

Second, investors have choice of either passive or active fund in each asset class one wants exposure to. I simply presented the data from each asset class.


The question for the typical investor is to decide which is the strategy that is MORE likely to get you the best result. And even active investors must admit that if they go active they run the risk of underperformance. And since almost all individuals are risk averse, one should really be highly confident of being right, not just more than 50%, because the cost of being wrong is more painful than the benefit of being right. Also the data shows that the losers tend to underperform by more than the winners outperform.

Like I said, American funds is about as good as one can get if you have to choose active managers but I don't see why one would conclude they should go that route from this sample of data.

Especially this should be the case when the funds have no grown so large that it is hard to avoid the closet index problem which increases the hurdle of generating alpha.

User avatar
Random Musings
Posts: 5581
Joined: Thu Feb 22, 2007 4:24 pm
Location: Pennsylvania

Post by Random Musings » Fri Apr 15, 2011 3:59 pm

Owning American Funds makes it more complex to know what asset classes you truly own - so you have to do a M* type analysis.

Makes it more of a pain to rebalance also, probably less effective than owning individual asset classes that are more defined.

Probably less efficient in taxable accounts as well.

But, there are far worse active funds to own.

RM

User avatar
pointyhairedboss
Posts: 420
Joined: Fri Mar 02, 2007 10:47 pm

Post by pointyhairedboss » Fri Apr 15, 2011 4:06 pm

Owning American Funds makes it more complex to know what asset classes you truly own - so you have to do a M* type analysis.

Makes it more of a pain to rebalance also, probably less effective than owning individual asset classes that are more defined.

Probably less efficient in taxable accounts as well.

But, there are far worse active funds to own.
Yes, the asset class impurity of active funds gives you a little asset allocation fuzziness. An international fund might dabble in emerging markets. A domestic fund might dabble in international markets. If you believe in active management, it is not necessarily a bad thing. You allow to active management to have some small sway in your allocation.

Not necessarily stating my opinion, just playing devils advocate.

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

other points

Post by larryswedroe » Fri Apr 15, 2011 4:33 pm

BTW-nothing prevented the manager's of the American Value Funds from tilting more to value if they thought value would outperform.

User avatar
rustymutt
Posts: 3781
Joined: Sat Mar 07, 2009 12:03 pm
Location: Oklahoma

Post by rustymutt » Fri Apr 15, 2011 5:22 pm

Has Larry been picking on American funds? Larry picks a lot fund families.
And I'm glad that someone does, cause they deserve to be picked on.
Larry would you stop picking on these monster sized companies?
I'm surprised they have not tried to buy you Larry.
I'm amazed at the wealth of Knowledge others gather, and share over a lifetime of learning. The mind is truly unique. It's nice when we use it!

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

not picking on anyone

Post by larryswedroe » Fri Apr 15, 2011 5:42 pm

Just presenting the data which IMO pretty much speaks for itself

As Rick Ferri has pointed out, the odds of a fund outperforming a benchmark are much greater than a portfolio of funds. By presenting the data set of comparable funds one can see a clearer picture of the odds of outperforming versus just data mining and selecting a particular fund (unknowable which fund would outperform ex ante).

Also when presenting the data it might even contain survivorship bias, since M* doesn't show returns of funds not alive, and I typically try to show ten years of data where it exists, settling for less if have to.

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Post by Beagler » Fri Apr 15, 2011 6:00 pm

Random Musings wrote:Owning American Funds makes it more complex to know what asset classes you truly own - so you have to do a M* type analysis.

Makes it more of a pain to rebalance also, probably less effective than owning individual asset classes that are more defined.

Probably less efficient in taxable accounts as well.

But, there are far worse active funds to own.

RM
Many AF are "go anywhere" funds and are not designed to fit into an indexing portfolio.
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Post by Beagler » Fri Apr 15, 2011 6:03 pm

Burton Malkiel wrote a glowing foreword for this book http://www.amazon.com/Capital-Story-Lon ... 483&sr=1-5
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

User avatar
ddb
Posts: 5509
Joined: Mon Feb 26, 2007 12:37 pm
Location: American Gardens Building, West 81st St.

Re: ken

Post by ddb » Fri Apr 15, 2011 6:48 pm

larryswedroe wrote:First, if you want to see if fund added alpha then one would need to perform a three factor regression.

<snip>

Especially this should be the case when the funds have no grown so large that it is hard to avoid the closet index problem which increases the hurdle of generating alpha.
What is your preferred method for measuring alpha in the case of a tactical allocation fund where factor exposure is highly dynamic? As an example, the Ivy Asset Strategy Fund.

- DDB
"We have to encourage a return to traditional moral values. Most importantly, we have to promote general social concern, and less materialism in young people." - PB

exeunt
Posts: 907
Joined: Tue Jan 08, 2008 6:54 pm

Post by exeunt » Fri Apr 15, 2011 7:07 pm

ddb, that's pretty straight forward. Just add more factors to the regression representing the various risk tilts the portfolio can assume. If the fund successfully time its exposures, alpha will show up.

Alternatively, a more precise method is the Brinson attribution, which requires portfolio holdings to calculate the value added or subtracted by portfolio weighting deviations from the benchmark. Here's a quick overview.

User avatar
ddb
Posts: 5509
Joined: Mon Feb 26, 2007 12:37 pm
Location: American Gardens Building, West 81st St.

Post by ddb » Fri Apr 15, 2011 8:02 pm

exeunt wrote:ddb, that's pretty straight forward. Just add more factors to the regression representing the various risk tilts the portfolio can assume. If the fund successfully time its exposures, alpha will show up.

Alternatively, a more precise method is the Brinson attribution, which requires portfolio holdings to calculate the value added or subtracted by portfolio weighting deviations from the benchmark. Here's a quick overview.
How is knowing Larry's preferred method "pretty straight forward"? I'm aware of various methods, I'm just curious how Larry determines value-added (or lack thereof) of a TAA fund.

- DDB
"We have to encourage a return to traditional moral values. Most importantly, we have to promote general social concern, and less materialism in young people." - PB

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

ddb

Post by larryswedroe » Fri Apr 15, 2011 9:40 pm

As the gentleman said, run a three factor regression, or now it seems you need to do more, a separate factor for small value and one for large value gives better results as value premium is not linear.

Can also run it for momentum, and can also add two momentum factors, one for large momentum and one for small (which is what the new FF paper does)

Larry

treypar
Posts: 92
Joined: Wed Feb 27, 2008 6:55 pm
Location: GA

Post by treypar » Sat Apr 16, 2011 5:21 am

Larry
Thanks for the article. My 401K unfortunately uses American Funds and allows only eight of their funds to select from. I have invested in the Growth and Europacific Funds and your analysis has confirmed that I made the correct selections.

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

treypar

Post by larryswedroe » Sat Apr 16, 2011 7:40 am

You are relatively lucky. There are plans with far worse choices. American funds are broadly diversified, minimizing the idiosyncratic risks and thus are unlikely to produce big tracking error and their costs are relatively good for active funds (though not Vanguard like). Those two issues are the big problems with active funds, so consider yourself somewhat lucky.
Shouldn't stop you though from lobbying for Vanguard's or DFAs funds as choices.

User avatar
anthau
Posts: 146
Joined: Wed Feb 21, 2007 5:29 pm
Location: Laramie, WY

Post by anthau » Sat Apr 16, 2011 9:45 am

ddb wrote:I should have been more specific, by writing "Assuming sufficient diversification, indexing isn't special."

Take two funds, both from the same asset class - fund A is an index fund with 0.5% expense ratio, 1,000 holdings, and 15% turnover. Fund B is an active fund with 0.5% expense ratio, 300 holdings, and 15% turnover.

Which would you rather own?

Well, Fund A is preferable, but only by the slimmest margin. And it's not preferable because it's an index fund, but because it has more diversification. But we're really splitting hairs, because at 300 holdings, you've already effectively eliminated non-systematic risk.

Now, pretend that Fund B has a 0.35% expense ratio. Now which is preferable? The rational investor should go for the lower-cost option, and not the one that is badged with an "index" label.

The general failure of active management has to do with costs, and not with poor manager decisions.

- DDB
A bit OT from the original post, but this got me thinking about expense vs. diversification. Assuming the same asset class: If Fund A has an ER of 0.25%, turnover of 12%, 105 holdings, with 42% in the top 10 holdings; Fund B has an ER of 0.65%, turnover of 22%, 370 holdings, with 21% in the top 10 holdings; which is preferable? Or, to put it more generally: If diversification costs more, how much more should be one willing to pay for it?
Best, | | Anth

Post Reply