I frequently see suggestions that real estate [i.e., REITs, real-
estate stocks, etc] should account for no more than 5%-10% of one's
portfolio. Does anyone understand the basis for this guideline? Why
The reason I've seen is that most homeowners already have a
significant holding in real estate. For example, if you have $100,000
in stock/bond/mutual-fund investments and live in a $200,000 house,
2/3 of your "portfolio" already is in real estate. Adding more would
simply decrease your diversification.
If you are not a homeowner, then you could increase your real estate-
related investment holdings, but it might be better to become a
homeowner. This is a lifestyle choice, not an investment decision.
Thanks for your response. Owning a home means you have a stake in a
single piece of real estate. If you own a real estate mutual fund
[for example], you have an ownership stake in hundreds of properties,
spread across the country or the world - and across the commercial and
residential sectors. So it seems that a fund offers much more
diversification than owning a single home. I tend to think of my home
as a place to live, rather than an investment.
But I follow your thinking.
estatestocks, etc] should account for no more than 5%-10% of one's
Trade that $200,000 house from my previous example for $200,000 in a
real estate mutual fund and you have increased your diversification,
not across the economy, but within one sector of it. Keep your house
and replace a growth stock mutual fund with a real estate fund, and
you have decreased your diversification.
That is true if you paid cash for the $200,000 house. But if you made a
downpayment of $20,000 and took a mortgage for $180,000, the comparison with
stocks or mutual funds is not quite the same. Your $20,000 downpayment on
the house gives you control of an asset that could appreciate and eventually
return as much profit as the entire $100,000 in stocks.
Swenson (Yale endowment manager and author of several investment
books) recommends 20%+ real estate. There's a few other authors that
also recommend higher real estate allocations so it's not a totally
crazy idea. To be honest, it doesn't matter that much whether you
decide 5% or 20%. Whether 5% or 20% will do better in the future, who
knows? The most important thing is once you decide on some percentage,
you stick with it and don't flip-flop trying to performance chase.
I'll have to change the numbers to make my point, but from a portfolio
composition point of view, what is the difference between
1. having a $200,000 mutual fund portfolio and a house with a $180,000
loan against it, and
2. having a $200,000 house and a mutual fund portfolio with a $180,000
loan against it?
Either way, you are 50-50 with a large debt.
I agree with that statement.
Surely, real estate holds a special place in portfolio theory versus
most other specific industries - but I understand your point. But
that reasoning, why specifically invest in real estate? Isn't it
represented in the typical stock mutual fund - just like other
That is true. But everybody knows that, over long periods, stocks on average
increase in value despite temporary ups and downs. The same is true of real
estate. The main question is: Are your chances of eventual gain or loss with
$100,000 in stocks more or less than your chances with $20,000 in real
estate? And to make the comparison fair, you have to consider what you do
with the other $80,000 that you did not put into real estate.
Yes, that seems like a good comparison. But one point: For a small investor
with limited assets it is possible to obtain a $180,000 mortgage to buy a
house but not a $180,000 loan against mutual funds.
It's mainly a matter of diversification. Real estate stocks represent
only a small percentage of the available equities, so you should feel
well represented at that level investment.
Real estate does represent a high proportion of the equity of American
families, and a house probably is all you need for real estate
Owning rental properties is an attractive way to go if you have a high
income and can take advantage of some of the tax benefits. Others who
are willing to maintain their own units can also do well in the rental
market. I have heard that 8 rental units are needed to make a rental
operation start to pay off.
I think there's a difference between how much real-estate you "need"
versus how much might make sense in your portfolio. For example, a
real estate fund generally invests in commercial property. I think
owning commercial property around the country or the world can be a
significant diversifier above and beyond owning a single home in a
single location. In addition, you cannot sell a portion of your home
[or buy a little more] to rebalance your portfolio. Thus, your home
cannot particpate in the Efficient Frontier.
I can state with certainty that over the past 5 years, having some
real estate in my portfolio [20%] has significantly boosted returns
and reduced volatility. Going forward - one can never say?
I've seen this rule of thumb many times but never seen it attributed
to existing home ownership (as it has been in some of the above
posts). I have to ask, if home ownership is the reason for the rule,
how much should a renter have invested in real estate? Still 5-10%?
And how do we vary our plan for a 0% equity home owner as opposed to a
100% equity homeowner. I have never seen that distinction made in
regards to asset allocation. With reverse mortgages, post-retirement
downsizing, etc, etc.. becoming more popular it seems that residences
may begin to play a larger part in portfolio development.
Traditionally they have not, nor do I believe this is the primary
reason "they" recommed 5-10% (but perhaps it will be some day or
should be now).
The essence of the "rule of thumb" is to reduce non-systematic risk
through diversification. Diversification relies upon the intermingling
of NON-CORRELATED assets, not just having a bunch of funds/stocks. As
PeterL said, the rule is not specific to real estate. Its a general
rule used for lots of sectors to help assure non-correlation. Its also
the reason many of us use index funds and ETFs. By getting broad
exposure to the entire market, non-correlation is most efficiently
achieved. Focusing a large % of assets in one sector, even if you have
a large number of funds, may still result in a high correlation and
little diversification. Imagine if an investor had "diversifed" by
only buying a bunch of different tech stocks in 2000-2001.
JJ, without knowing specifically what you're looking at it's hard to
say, but I'd be wary of any rules of thumb about asset allocation. Some
people don't even consider REITs to be a separate asset class, for
example, so they might say 0% is appropriate. Others might be relying on
historical models over time periods where REITs performed extremely
well, and weren't highly correlated with other asset classes, using that
as a justification for a high investment in them. Just looking at the
numbers you can find periods that would justify allocations higher than
20%...they've outperformed US stocks over quite a few time periods.
One rationale for keeping it small is that it is, as another poster
mentioned, a relatively tiny part of the public equities market, based
on total market capitalization. It's a small subset of the Russell 2000,
really. Commercial real estate may have a lot of value associated with
it, but a small percentage of properties are owned by REITs. So I
question the capacity of existing US REITs to absorb a 20% allocation by
the "typical investor"...that'd be too much money chasing too few
stocks. There was some commentary about this after REIT gains earlier
this year; some blamed institutional dollars that were tied to
investments in the 200-odd public REITs that are members of the major US
I don't know how relevant home ownership is to a REIT allocation -- as
you said, they're really different things. To some extent there may be
correlations in value but even that strikes me as tenuous. And as you
said, one's an investment, the other is the place you live.
Don't confuse the size of one's mortgage with the amount of real estate
one has. The 0% equity guy happens to be short a bond which happens to
have the same value as his house. But they are both subject to the rise
and fall of the real estate market, in term of impact to their wealth.
Having no intention to use my home's value to retire (e.g. no particular
plan to downsize and take money from the house to invest) I don't count
the house as part of my 'portfolio'. Not like I can withdraw 4% each
year at retirement. I think there's little to be gained in overweighting
real estate in one's portfolio. Chasing sectors to me is a form of
market timing, isn't it?
I think what I am asking is IF one were to consider their home as part
of their portfolio does the "rule of thumb" change based on equity
ownership? Renting or 0% equity means you actually have no real estate
position. You may in the future (by paying off your mortgage) but if
you liquidated all your holdings today and headed to beach, you'd get
nothing for your "holdings". 100% equity would be a different
scenario. It just so happens that I don't count my residence in my
portfolio so this is a moot point for me.
Agreed. I wonder if the practice of not including it will change as
underfunded retirement plans, unsustainable defined benefit plans, and
reverse mortgages become more common?
I also don't recommend my clients take a large RE position (usually
4-7%). I am a buy and hold investor so I don't care how the RE market
does. I'm gonna rebalance and keep'em in that range regardless. Using
ETFs also helps in that with 1 or 2 funds I can obtain well
diversified positions for my clients.
I'd think not.
is not much different than;
$200K margin loan
Of course the dynamic is different, a drop in house prices doesn't force
a margin call, just hinders one's ability to refinance. And the margin
loan tends to be at higher interest rates, and variable. Both people
above have $200K RE, $400K stock, -$200K cash.