The Top 10 ETF Model Portfolio

By Matthew Hougan

Think building an exchange-traded funds portfolio is complicated? It turns out you could do a lot worse than just buying the 10 biggest ETFs.

Sounds too easy, right? But take a look. Using data as of March 31, the 10 largest ETFs by assets were:

If you took an equal position in each of those ETFs, you'd have a nice, balanced portfolio: 60% stocks, 30% bonds and 10% gold.

Not only is that a reasonably well-diversified portfolio, it includes a number of twists and tilts that I find pretty appealing right now.

Starting in equities, both growth and large-caps have been performing relatively well, as they often do in difficult economic environments. Similarly, the tilt toward Technology via the QQQs captures one of the best-performing sectors in the market. And while the portfolio is light on international exposure, its strong relative weight in emerging markets is appealing.

In the fixed-income market, it is almost dead-on. If you split out the exposure in AGG, the fund is 16% corporate, 10% TIPS and 4% Treasuries. Considering that most experts think that Treasuries are at least fully valued, and corporate undervalued, that's just about right.

Finally, the addition of significant (20%) exposure to inflation-hedging assets like gold and TIPS is very appealing in the current environment, where many of us (even Jim) expect a major uptick in inflation.

Another approach would be to create a portfolio weighted by the size of assets held in each ETF. That portfolio is slightly more aggressive, with 68% equities, 16% bonds and 16% gold.

As good as these portfolios seem to be, I'm tempted to look a bit further down the tables to see what other ETFs are on the edge of being added to the mix.

Looking through that list, there are a few tempting funds. I love the Vanguard MSCI Total Market ETF (NYSE: VTI), the cheapest way to gain access to the total market. And I'd be very tempted to elevate the iShares FTSE/Xinhua China 25 ETF (NYSE: FXI) into the mix, as I think China is underweight in global benchmarks and is positioned well to perform.

Still, that's all happening on the edges. You don't really have to go much further than the top 10. As it turns out, ETF investors are a pretty smart lot. So much for the madness of the crowds.

Original post

Comments

[bricki:]

This portfolio is grossly overweighted in US large cap stocks. As a result returns will underperform the market as a whole. Not to mention that the only exposure to commodities is gold.

I wouldn't recommend this particular collection of ETFs.

Apr 10 10:28 AM

[WebWatcher:]

This approach is sound, and the author is not saying that you have to go exactly as the example shows. It's quite easy to turn the weights to one's like based on one's assumptions and profile.

If we consider the risk factor, this approach is better than most of the stock portfolio. Good article!

Apr 10 11:20 AM

[Geoffrey Lordi:]

Nice consideration of mainstream etfs, Matt. Your article provides an excellent foundation for those interested in building etf portfolios for the intermediate to long term.

I'm a little concerned about high correlations among the above selections, however. With few exceptions, most of those choices - regardless of money flows - are highly correlated with one another over 1 and 3 year periods. And while many folks might hold them for longer, a large chunk of their portfolios would still be locked up in pretty high r^2 choices in the 4+ year (and beyond) range.

(Remember, anyone interested may 1) download data (closing price) for an etf in a prescribed period from Yahoo Finance to a spreadsheet, 2) align that data column near one for SPY, DIA, or QQQQ (or whichever other average for which you seek correlation data), and 3) use the CORREL command to check comovement: as such "CORREL(A1:A100,B1:B10... Performing your own correlation checks often 1) yields more varied period options than you'll see in the various etfs' glossy literature and 2) allows you to build tables that better help you analyze the r^2 of your portfolio.)

May I recommend an alternative to FXI? Loaded with financials (49%), one could argue for alternatives unless one seeks that bias. CAF, since its inception as a Morgan Stanley CEF, has outperformed FXI. It also does so on the 1- and 2-year time scales (even with roughly double the yearly fee), and holds less of a financial disposition (34%) to boot. (Note: The fund currently has 14% in cash.)

I do like LQD, own GLD, and have owned VTI long enough in several accounts that it has performed as advertised.

QQQQ is a relatively diversified pick, as well, for a popular average (0.84 r^2, depending on the period) because of its relatively large health care allocation: 20% (Amgen, Gilead, Teva); and, consumer discretionary: 12% (Costco, Starbucks, and DirecTV).

Readers wishing to add additional etfs for consideration to Matt's thoughtful presentation might want to study those in international energy or real estate, a currency hedge vehicle (DBV or ICI), or some commodities (but be mindful of the oils' waitings: unless you're content with over saturation, and equal-weight product such as Greenhaven's GCC might be more your style). A few other considerations might be to include an allocation to municipal and junk bonds, international or domestic utilities, or alternative style etf constructs (I'm thinking mainly of WisdomTree's dividend-paying based models as alternatives: for example, DEM instead of EEM or VWO).

Apr 10 12:46 PM

[ron3637:]

You left out tghe most important information: how did they perform over the last two years

Apr 10 01:17 PM

[Jake2:]

Nice work and much appreciated.

Apr 10 01:43 PM

[Paul H. M.:]

U.S. large caps may preform better since our system of business and law is still so much better than emerging markets.

Emerging markets have more growth, but are also more prone to a spectacular collapse (like the U.S. had when it was an emerging nation).


On Apr 10 10:28 AM bricki wrote:

> This portfolio is grossly overweighted in US large cap stocks. As
> a result returns will underperform the market as a whole. Not to
> mention that the only exposure to commodities is gold.
>
> I wouldn't recommend this particular collection of ETFs.

Apr 10 02:02 PM

[ThinkRight:]

US stocks are favored during increasing dollar, else european or emerging countries. During commodity favored and risk taking envirnment emerging countires and commodity rich countries favored. Some of the risks associated with US favor portfolio when using equal weight top 10 assets based etfs (10% each) can be mitigated by 5% equal weight in top 20 etfs which includes small cap, mid, cap, emerging, brazil (commodity, BRIC) and short term treasury.

If you are relative strength fanatics, you can rank top 20 by relative strength and invest in top 3 to 5 etfs! When SHY is at top, shy away from market !!

Apr 10 02:38 PM

[grh1212:]

Thank you, Geoffrey, for alerting us to the importance of minimizing highly correlated ETFs in our portfolios, showing us how to detect such, and finally for steering us to the place where we can obtain the data. Your spirit of helpfulness and generosity in sharing "tricks of the trade" are what the SA boards are about and many of us appreciate your post.


On Apr 10 12:46 PM Geoffrey Lordi wrote:

> Nice consideration of mainstream etfs, Matt. Your article provides
> an excellent foundation for those interested in building etf portfolios
> for the intermediate to long term.
>
> I'm a little concerned about high correlations among the above selections,
> however. With few exceptions, most of those choices - regardless
> of money flows - are highly correlated with one another over 1 and
> 3 year periods. And while many folks might hold them for longer,
> a large chunk of their portfolios would still be locked up in pretty
> high r^2 choices in the 4+ year (and beyond) range.
>
> (Remember, anyone interested may 1) download data (closing price)
> for an etf in a prescribed period from Yahoo Finance to a spreadsheet,
> 2) align that data column near one for SPY, DIA, or QQQQ (or whichever
> other average for which you seek correlation data), and 3) use the
> CORREL command to check comovement: as such "CORREL(A1:A100,B1:B10...
> Performing your own correlation checks often 1) yields more varied
> period options than you'll see in the various etfs' glossy literature
> and 2) allows you to build tables that better help you analyze the
> r^2 of your portfolio.)
>
> May I recommend an alternative to FXI? Loaded with financials (49%),
> one could argue for alternatives unless one seeks that bias. CAF,
> since its inception as a Morgan Stanley CEF, has outperformed FXI.
> It also does so on the 1- and 2-year time scales (even with roughly
> double the yearly fee), and holds less of a financial disposition
> (34%) to boot. (Note: The fund currently has 14% in cash.)
>
> I do like LQD, own GLD, and have owned VTI long enough in several
> accounts that it has performed as advertised.
>
> QQQQ is a relatively diversified pick, as well, for a popular average
> (0.84 r^2, depending on the period) because of its relatively large
> health care allocation: 20% (Amgen, Gilead, Teva); and, consumer
> discretionary: 12% (Costco, Starbucks, and DirecTV).
>
> Readers wishing to add additional etfs for consideration to Matt's
> thoughtful presentation might want to study those in international
> energy or real estate, a currency hedge vehicle (DBV or ICI), or
> some commodities (but be mindful of the oils' waitings: unless you're
> content with over saturation, and equal-weight product such as Greenhaven's
> GCC might be more your style). A few other considerations might be
> to include an allocation to municipal and junk bonds, international
> or domestic utilities, or alternative style etf constructs (I'm thinking
> mainly of WisdomTree's dividend-paying based models as alternatives:
> for example, DEM instead of EEM or VWO).

Apr 10 03:26 PM

[Terry330:]

US Stocks 20%
International stocks 20%
Commodities 20%
REITS 20%
TIPS 20%

Apr 10 06:42 PM

[bricki:]

US small caps are essentially completely absent from the top 10 portfolio, and are very likely to be a great source of growth going forward.

On Apr 10 02:02 PM Paul H. M. wrote:

> U.S. large caps may preform better since our system of business and
> law is still so much better than emerging markets.
>
> Emerging markets have more growth, but are also more prone to a spectacular
> collapse (like the U.S. had when it was an emerging nation).

Apr 10 06:45 PM

[JR Ewing:]

Does anyone have an opinion on the DBA food ETF? I got burned a little on the USO and little cautious playing ETF's. Is the USO just broken or do all commodity ETF's have this issue?

DBA could be a good hedge here if we do get a pull back and some inflation.

Apr 11 07:22 AM

[mkreisel:]

EEM, EWZ, FXI are approaching their 200 day moving average, if they can stand above it for more than a month, time to buy again.

But chances are, the 200 day moving average will prove to be a impenetrable ceiling that will knock these ETFs down to a new abyss.

The EWZ from 2006-4/9/2009 bears more resemblance to DIJA from 1927-4/9/1930 than DIJA from 1985-4/8/1988. The history is repeating again.

Apr 11 04:07 PM

[Diegojames:]

USO is not considered very appropriate or correlated. Maybe some experts will step in and comment. CNBC HD has even presented negative information about it.
I recently sold down CVX since it is showing projected negative revenue growth for next 2 years and schwab made it "D". I have XOM on watchlist and am waiting for natural gas to bottom.
Will go back into TCK as possible action by temasek is forecast.
Will buy more CNI on market dip and add to BHP, HON, RHT,CSCO,VLO,SLV,KO,SPG, and I am holding BTU.
The construction and industrials will improve over the next several months. Hard to know when to come in and which ones.
California is a economic disaster and more home and condos are being built around the LA county edges.Who will buy? Low priced condos are being snapped up with OBAMA money added to real estate broker discount by investors.
LA Unified building a school in Granada Hills with a solid steel frame . CSUN is still adding buildings and infastructure. I will be checking out the San Joaquin Valley next week. Homes are being built south of San Diego. It is time to check out ETF building supplies, construction companies and industrials .
Northridge, California

Apr 11 04:11 PM

[2HEADPIG:]

NO CASH, NO REALESTATE, NO SHORTS

Apr 12 09:10 AM

[2HEADPIG:]

NO CASH, NO REAL ESTATE, NO SHORTS. ????

Apr 12 09:12 AM

[Jaysan:]

USO does not accurately track the spot price of oil. It rolls over its futures when they expire, so contango distorts it.

On Apr 11 07:22 AM JR Ewing wrote:

> Does anyone have an opinion on the DBA food ETF? I got burned a
> little on the USO and little cautious playing ETF's. Is the USO
> just broken or do all commodity ETF's have this issue?
>
> DBA could be a good hedge here if we do get a pull back and some
> inflation.

Apr 12 12:01 PM

[Roowns:]

Well. If you had like 10 mil. dollars and all you want is to keep your wealth and have low exposure to risk I guess you could easily use a mix of those.
But since it's not really anyone case in here, I'll have to pass. Thank you.

Now can you tell me how to reach 10 mil. dollars so I can use this article ?

Apr 12 05:42 PM

[pockyclips ...:]

I agree. A chart of this portfolio over the last two years would be interesting. The last 16 months would be even better.


On Apr 10 01:17 PM ron3637 wrote:

> You left out tghe most important information: how did they perform
> over the last two years

Apr 13 08:22 AM

[proudgrampa:]

Reminds me of Steve Martin's joke: How to make a million dollars and not pay taxes. First, make a million dollars. Then when the IRS comes to find out why you didn't pay taxes say, "I forgot."


On Apr 12 05:42 PM Roowns wrote:

> Well. If you had like 10 mil. dollars and all you want is to keep
> your wealth and have low exposure to risk I guess you could easily
> use a mix of those.
> But since it's not really anyone case in here, I'll have to pass.
> Thank you.
>
> Now can you tell me how to reach 10 mil. dollars so I can use this
> article ?

Apr 13 01:01 PM