REITs as "pseudo diversification"

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
ERMD
Posts: 291
Joined: Thu Dec 26, 2013 11:26 am

REITs as "pseudo diversification"

Post by ERMD » Wed Sep 27, 2017 1:27 pm

i'm sure this has been covered before, but i couldn't find any threads on a forum search.

i'm interested in hearing from people who have specifically eschewed REITs in favor of individual property ownership. i've heard the argument that because of the correlation between REITs and equities, it doesn't serve to provide actual diversification, as opposed to traditional real estate investing (this is a position i've heard argued by several people, including Dennis Bethel of NestEggRx). i'm not sure about the evidence behind this assertion. others, like Jussi Askola at Seeking Alpha, argue that they are in fact sufficiently diversified and not highly correlated with equities, and provide significantly better risk adjusted return.
between scotch and nothing, i'll take scotch. -- faulkner

betablocker
Posts: 400
Joined: Mon Jan 11, 2016 1:26 pm

Re: REITs as "pseudo diversification"

Post by betablocker » Wed Sep 27, 2017 1:48 pm

All things even private property is no more or less correlated with equities than public REITs. Private property simply isn't "marked to market" every second. So the benefit is you can't look up the value on Yahoo Finance and that's it. The down side is all the idiosyncratic risk you get from owning one property in one area. Say there's a regional downturn or zoning changes, etc. You don't have that risk with diversified REITs. That individual property risk (just like that for an individual stock) is usually uncompensated. The exception is if you know something about the particular property being undervalued. That's akin to having insider information but legal in this case and insider information is rewarded in equities (see the show Billions) and properties. Or as was the case with private equity if you have special access to leverage or an ability to fix up the property (or company), you can get compensated for that. As far as REITs and their correlation with public equity, I believe that depends. In some downturns they correlate (aka 2008 when there is a deflationary flight to quality assets like treasuries and people dump REITs along with stocks) and in other inflationary downturns or shocks they should be uncorrelated and provide a hedge like commodities do. That's the theory at least.

User avatar
White Coat Investor
Posts: 13444
Joined: Fri Mar 02, 2007 9:11 pm
Location: Greatest Snow On Earth

Re: REITs as "pseudo diversification"

Post by White Coat Investor » Wed Sep 27, 2017 2:14 pm

REITs do provide diversification because the correlation between them and the overall market isn't perfect. It varies over time, but is low enough that I think there can still be a place for them for diversification reasons.

That said, I think that moderate correlation between REITs and TSM is a good reason to also consider other real estate investments. Another reason is potentially higher returns.

Many roads to Dublin. Moderation in all things.
1) Invest you must 2) Time is your friend 3) Impulse is your enemy | 4) Basic arithmetic works 5) Stick to simplicity 6) Stay the course

User avatar
galeno
Posts: 1351
Joined: Fri Dec 21, 2007 12:06 pm

Re: REITs as "pseudo diversification"

Post by galeno » Wed Sep 27, 2017 2:46 pm

I don't use REITs but I agree with WCI. They can and do diversify against equities.
AA = 40/55/5. Expected CAGR = 3.8%. GSD (5y) = 6.2%. USD inflation (10 y) = 1.8%. AWR = 4.0%. TER = 0.4%. Port Yield = 2.13%. Term = 34 yr. FI Duration = 6.2 yr. Portfolio survival probability = 95%.

betablocker
Posts: 400
Joined: Mon Jan 11, 2016 1:26 pm

Re: REITs as "pseudo diversification"

Post by betablocker » Wed Sep 27, 2017 2:50 pm

They also have terrible expected returns because of the run up in recent years and despite the recent losses. Still I have them as a diversifier and hedge against inflation.

hightower
Posts: 510
Joined: Mon Dec 12, 2016 2:28 am

Re: REITs as "pseudo diversification"

Post by hightower » Wed Sep 27, 2017 2:55 pm

I don't think it's "pseudo" at all. It's diversification for sure. I wouldn't have exposure to all of these various real estate companies if it weren't for the REIT funds. I own my home and about 10% of my portfolio as Vanguard REIT (VNQ). That's my exposure to real estate. If I had more money to invest and the time to do it, I would consider an investment property like a vacation rental somewhere that we could also use for a few weeks a year, but right now we can't afford that kind of investing. Hawaii would be my choice though:) :sharebeer 8-)

User avatar
ERMD
Posts: 291
Joined: Thu Dec 26, 2013 11:26 am

Re: REITs as "pseudo diversification"

Post by ERMD » Wed Sep 27, 2017 3:10 pm

hightower wrote:
Wed Sep 27, 2017 2:55 pm
If I had more money to invest and the time to do it, I would consider an investment property like a vacation rental somewhere that we could also use for a few weeks a year, but right now we can't afford that kind of investing. Hawaii would be my choice though:) :sharebeer 8-)
the rentable beach house dream is always in the back of our mind too, although hawaii is a bit far/expensive :wink:
between scotch and nothing, i'll take scotch. -- faulkner

not4me
Posts: 417
Joined: Thu May 25, 2017 3:08 pm

Re: REITs as "pseudo diversification"

Post by not4me » Wed Sep 27, 2017 3:13 pm

To me, this is a somewhat broad, general question & the relevance of answers will reflect that. I would suggest a more targeted view when it comes to really putting this into action. I think most on this forum will interpret "individual property ownership" as being a rental -- house, vacation place, etc. REITs are likely thought of as vanguard mutual fund &/or etf. But, if your question was applied to buying a specific REIT, the answer changes. If a retail outlet or commercial office reit? Then maybe a higher correlation to broad stock market index than a senior living facility reit. As for the individual property, that brings in other factors but specifically location. Are the prospective tenants likely to be driven by general economic factors? for example, vacation condo rental vs apartment near college campus vs rental house in small town

If you also extend your view to include tax treatment, that may vary by investor. I think more investors will use leverage for individual property than when buying a reit.

aristotelian
Posts: 4694
Joined: Wed Jan 11, 2017 8:05 pm

Re: REITs as "pseudo diversification"

Post by aristotelian » Wed Sep 27, 2017 3:15 pm

It's not diversification if the REIT's are already included in Total Stock Market. It may be good or bad, but it's not diversification. Volatility-reducing?

avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: REITs as "pseudo diversification"

Post by avalpert » Wed Sep 27, 2017 3:41 pm

hightower wrote:
Wed Sep 27, 2017 2:55 pm
I wouldn't have exposure to all of these various real estate companies if it weren't for the REIT funds.
Except of course through the broad market indexes you own that already include them at their market weights...

JBTX
Posts: 4045
Joined: Wed Jul 26, 2017 12:46 pm

Re: REITs as "pseudo diversification"

Post by JBTX » Wed Sep 27, 2017 4:36 pm

betablocker wrote:
Wed Sep 27, 2017 2:50 pm
They also have terrible expected returns because of the run up in recent years and despite the recent losses. Still I have them as a diversifier and hedge against inflation.
This is me. I have a nominal amount as an inflation hedge. They outperformed during our last period of inflation by a large margin. But i agree with you that hedge will be muted given they yield is so low now.

JBTX
Posts: 4045
Joined: Wed Jul 26, 2017 12:46 pm

Re: REITs as "pseudo diversification"

Post by JBTX » Wed Sep 27, 2017 4:40 pm

aristotelian wrote:
Wed Sep 27, 2017 3:15 pm
It's not diversification if the REIT's are already included in Total Stock Market. It may be good or bad, but it's not diversification. Volatility-reducing?
Interesting point but I don't know that market weighting of the TSM necessarily by definition provides the lowest level of risk or the optimal risk return relationship does it?

User avatar
siamond
Posts: 4189
Joined: Mon May 28, 2012 5:50 am

Re: REITs as "pseudo diversification"

Post by siamond » Wed Sep 27, 2017 5:04 pm

ERMD wrote:
Wed Sep 27, 2017 1:27 pm
i've heard the argument that because of the correlation between REITs and equities, it doesn't serve to provide actual diversification, as opposed to traditional real estate investing (this is a position i've heard argued by several people, including Dennis Bethel of NestEggRx). i'm not sure about the evidence behind this assertion. others, like Jussi Askola at Seeking Alpha, argue that they are in fact sufficiently diversified and not highly correlated with equities, and provide significantly better risk adjusted return.
You can check recent stats (correlation, volatility, portfolio effect, etc) about REITs (US and International), in this Bogleheads blog post:
https://finpage.blog/2017/09/11/portfol ... ing-bonds/

Overall, this isn't black & white. The case for diversification is there (or at least was there in the past) but not terribly strong. Which is why different people reach different conclusions...

Bud
Posts: 139
Joined: Sun Jul 22, 2007 6:59 am

Re: REITs as "pseudo diversification"

Post by Bud » Wed Sep 27, 2017 5:04 pm

Just a few thoughts from having owned real estate, small businesses, REITs, and stocks for a few decades...

It is difficult to compare REITs and individual property ownership just as it is difficult to compare a TSM index and individual business ownership.

They are different beasts - do you own single family rentals, multi-family rentals, farm land, recreational use land, commercial rental property? REITs are liquid and constantly marked to market, real estate can be illiquid and market value can be ambiguous. REITs are usually purchased unleveraged while real estate is normally leveraged for purchasing. REITs can be bought and sold instantly, real estate often takes more than 30 days to transact. REITS do not require either a realtor or lawyer to purchase, real estate often requires both. REITs require no contracts to provide income, rental property should have a contractual agreement. REITs need no physical maintenance, real estate does.

So, in my thinking, real estate is a diversifier to a portfolio but the comparison to REITs is not a good one. A more accurate comparison is to consider starting a business to diversify from TSM.

User avatar
nisiprius
Advisory Board
Posts: 36682
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: REITs as "pseudo diversification"

Post by nisiprius » Wed Sep 27, 2017 5:23 pm

Although I am personally not convinced about REITs, a quick reality check does show that over the time period included by PortfolioVisualizer, they have done just what their advocates have claimed they would do. Here's what I did.

I created a "traditional Boglehead-style three-fund portfolio" of 42% Vanguard Total [U. S.] Stock Market Index fund (VTSMX), 18% Vanguard Total International Stock Index Fund (VGTSX), and 40% Vanguard Total Bond Index Fund (VBMFX). That's portfolio 1, blue.

I then overweighted REITS by adding 15% Vanguard REIT Index Fund, and cutting down VTSMX and VGTSX proportionately (to 31.50% and 13.50%, respectively). That is, I replaced 15% of the portfolio devoted to stocks with VGSIX. PortfolioVisualizer has data for these four funds back to 1997, or a respectable 20 years of data. That's portfolio 2, red.

That had the effect of increasing return from 6.94% to 7.33% while simultaneously reducing the standard deviation, a measure of volatility or risk, from 9.16% to 9.06%. Accordingly, the Sharpe ratio, a measure of risk-adjusted return, was increased from 0.55 to 0.60.

Now I can give a bunch of reasons/rationalizations for saying that's not a very large effect and I don't have much confidence in it being reliable or persistent. The biggest issue is one that I'm not very proud of, but it's true: during 2008-2009, when stocks were plummeting and lost over half their value, VGSIX, which I had a very small allocation to, lost nearly 2/3rds of its value and I got so spooked I said the heck with it. In other words, regardless of statistics, VGSIX is not for me.

Nevertheless, over that 20-year period, REITs did exactly what it was claimed they would do. I have another demurral, which is that most of the effect was due to one fairly short time period, 2000-2002, during which stocks in general went down, but REITS (as well as international and small value) went up. Nevertheless there was still some benefit over other time periods.

Source
Image
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

aristotelian
Posts: 4694
Joined: Wed Jan 11, 2017 8:05 pm

Re: REITs as "pseudo diversification"

Post by aristotelian » Wed Sep 27, 2017 5:31 pm

JBTX wrote:
Wed Sep 27, 2017 4:40 pm
aristotelian wrote:
Wed Sep 27, 2017 3:15 pm
It's not diversification if the REIT's are already included in Total Stock Market. It may be good or bad, but it's not diversification. Volatility-reducing?
Interesting point but I don't know that market weighting of the TSM necessarily by definition provides the lowest level of risk or the optimal risk return relationship does it?
Probably not. It is generally acknowledged that something like 70 stocks/30 bonds provides better return relative to risk than 100 stocks. It is possible that an allocation to REIT would also do even better than that. (You could also try throwing Utilities in there). But risk-adjusted return is a different concept than diversification.

staythecourse
Posts: 6018
Joined: Mon Jan 03, 2011 9:40 am

Re: REITs as "pseudo diversification"

Post by staythecourse » Wed Sep 27, 2017 5:44 pm

I thought I read a paper by Vanguard in regards to REITS, direct real estate investments, and other modes in relation to equities. I don't think there is any real argument to any serious investor that REITS are NOT a substitute for direct real estate as a diversifier to a traditional capital asset portfolio (stocks/ bonds/ cash). The argument really lies in does the extra risk and extra work justify the increased diversification benefits? There are significant risk of traditional real estate just like any alternative investments, i.e. higher barriers of entry, illiquidity, non transparency, opacity, higher management fees, geographical risk, lack of diversification, etc...

For me, I would LOVE to invest directly in real estate, but don't have the knowledge of doing it on my work, will to want to do it on my own, and trust to give my money to someone else to do it on my own. So that means REITS for me. Each investor has to do what they are comfortable for them and fits in with the overall portfolio plan.

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle

User avatar
patrick013
Posts: 2372
Joined: Mon Jul 13, 2015 7:49 pm

Re: REITs as "pseudo diversification"

Post by patrick013 » Wed Sep 27, 2017 5:57 pm

VNQ REIT has a 3 yr. beta of .56 so it doesn't go as the market goes
all the time.

I suspect most RE investors expect equity REIT's in an index to be
high occupancy REIT's. Most would be happy when low occupancy
individual real estate investments are considered or possible.
Right now VNQ yields 4.39%. So it could be worth a 10% portfolio
allocation for additional RE investment and portfolio diversification.
age in bonds, buy-and-hold, 10 year business cycle

User avatar
NoRoboGuy
Posts: 732
Joined: Fri Apr 01, 2011 11:07 pm
Location: Just west of Austin TX

Re: REITs as "pseudo diversification"

Post by NoRoboGuy » Wed Sep 27, 2017 6:18 pm

The correlation between REITs and the TSM varies a great deal. However, over long periods of time, overweighting REITs (in tax-advantaged accounts, of course) has improved returns and reduced volatility of the portfolio.

I have always argued that if you are a long-term investor willing to tilt, REITs and small value is where you will get the best bang for the buck. Also, I tilt a bit towards emerging in the total international stock portion.

I agree it is "pseudo-diversification" from the standpoint that when you tilt, you are not actually adding any new holdings. Rather, you are overweighting segments within the total index.
There is no free lunch.

User avatar
nisiprius
Advisory Board
Posts: 36682
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: REITs as "pseudo diversification"

Post by nisiprius » Wed Sep 27, 2017 6:25 pm

Very tangential but I noticed in the proxy vote that for shareholders of VGSIX, the Vanguard REIT Index Fund, there is "Proposal to reclassify the diversification status to nondiversified." Apparently the composition of the index, together with the size of the fund, means that in order to track the index by replication, it is running afoul of the requirement that a fund is not supposed to hold more than 10% of the outstanding stock in any individual issue.

Just to be clear, the fact that the VGSIX REIT index fund may be classified as "nondiversified" under the provisions of the Investment Company Act of 1940 does not mean that it could not have a diversifying effect in a portfolio, but it does make you think just a bit about what's going on when you commit a medium-sized chunk of your assets to a little-sized bit of the market.

If Vanguard continues to grow we will probably start seeing more of this. There aren't enough small value, REITs, etc. to go around if everybody wants to hold them in much more than their cap-weighted market share.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

JBTX
Posts: 4045
Joined: Wed Jul 26, 2017 12:46 pm

Re: REITs as "pseudo diversification"

Post by JBTX » Wed Sep 27, 2017 7:19 pm

aristotelian wrote:
Wed Sep 27, 2017 5:31 pm
JBTX wrote:
Wed Sep 27, 2017 4:40 pm
aristotelian wrote:
Wed Sep 27, 2017 3:15 pm
It's not diversification if the REIT's are already included in Total Stock Market. It may be good or bad, but it's not diversification. Volatility-reducing?
Interesting point but I don't know that market weighting of the TSM necessarily by definition provides the lowest level of risk or the optimal risk return relationship does it?
Probably not. It is generally acknowledged that something like 70 stocks/30 bonds provides better return relative to risk than 100 stocks. It is possible that an allocation to REIT would also do even better than that. (You could also try throwing Utilities in there). But risk-adjusted return is a different concept than diversification.
Interesting thread on the subject here.

viewtopic.php?t=153259

The crux of the question is if you have a minuscule presence in an index, is that really diversification?

An valid counter argument presented is you should just mirror he market, whatever that is. A different interesting argument is if trying to mirror market, then why don't we overweight bonds which have bigger market than stocks ?

User avatar
patrick013
Posts: 2372
Joined: Mon Jul 13, 2015 7:49 pm

Re: REITs as "pseudo diversification"

Post by patrick013 » Wed Sep 27, 2017 8:04 pm

nisiprius wrote:
Wed Sep 27, 2017 6:25 pm
If Vanguard continues to grow we will probably start seeing more of this. There aren't enough small value, REITs, etc. to go around if everybody wants to hold them in much more than their cap-weighted market share.
The newer S&P Equity REIT index has approx. $847 billion market cap.
I wonder how they approach that problem in their funds or are just concerned
with too many investors in general ? Not many details noted regarding that.
Some of those funds are smallish so they may not have that problem at all.
age in bonds, buy-and-hold, 10 year business cycle

jalbert
Posts: 3709
Joined: Fri Apr 10, 2015 12:29 am

Re: REITs as "pseudo diversification"

Post by jalbert » Wed Sep 27, 2017 8:28 pm

REITs are a sector of the market. The utility sector has been less correlated to the market, as has been precious metal equity, although the fund vgpmx has non-US exposure.
Risk is not a guarantor of return.

hightower
Posts: 510
Joined: Mon Dec 12, 2016 2:28 am

Re: REITs as "pseudo diversification"

Post by hightower » Thu Sep 28, 2017 9:37 am

avalpert wrote:
Wed Sep 27, 2017 3:41 pm
hightower wrote:
Wed Sep 27, 2017 2:55 pm
I wouldn't have exposure to all of these various real estate companies if it weren't for the REIT funds.
Except of course through the broad market indexes you own that already include them at their market weights...
Right, but that's like 3% of those index funds holdings. And considering that my total stock market allocation is only like 30% of my portfolio, it's not very much exposure. That's why I like to add more with a separate allocation of REIT funds.

alex_686
Posts: 3855
Joined: Mon Feb 09, 2015 2:39 pm

Re: REITs as "pseudo diversification"

Post by alex_686 » Thu Sep 28, 2017 10:02 am

jalbert wrote:
Wed Sep 27, 2017 8:28 pm
REITs are a sector of the market. The utility sector has been less correlated to the market, as has been precious metal equity, although the fund vgpmx has non-US exposure.
This does not show this. Correlations assume constant volatility, which has not been the case since 2004. A reason why REITs are showing a higher correlation is because REIT's volatility has been more volatile than Utilities. (I know the last sentence is going to require you to think through it) Either you need to cut your time period down to something under 36 months or you need to dig out a different set of statistical tools.

avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: REITs as "pseudo diversification"

Post by avalpert » Thu Sep 28, 2017 10:21 am

alex_686 wrote:
Thu Sep 28, 2017 10:02 am
jalbert wrote:
Wed Sep 27, 2017 8:28 pm
REITs are a sector of the market. The utility sector has been less correlated to the market, as has been precious metal equity, although the fund vgpmx has non-US exposure.
This does not show this. Correlations assume constant volatility, which has not been the case since 2004. A reason why REITs are showing a higher correlation is because REIT's volatility has been more volatile than Utilities. (I know the last sentence is going to require you to think through it) Either you need to cut your time period down to something under 36 months or you need to dig out a different set of statistical tools.
The story doesn't change over shorter time frames - utilities and precious metals (and energy I believe) are consistently less correlated to the overall market than REITs.

jalbert
Posts: 3709
Joined: Fri Apr 10, 2015 12:29 am

Re: REITs as "pseudo diversification"

Post by jalbert » Thu Sep 28, 2017 11:42 am

This does not show this. Correlations assume constant volatility, which has not been the case since 2004. A reason why REITs are showing a higher correlation is because REIT's volatility has been more volatile than Utilities. (I know the last sentence is going to require you to think through it) Either you need to cut your time period down to something under 36 months or you need to dig out a different set of statistical tools
I was careful to use past tense because I was reporting sample correlations from a historical sample. We don't know the covariances of the actual joint distributions of return because we don't know the actual distributions. Sample covariances are not necessarily very good estimates of the actual covariances, But a shorter time period would provide a less reliable estimator.

And increased volatility does not increase correlation. High volatility (i.e. variance) actually makes it harder for two random variables to be fully correlated (if both are always jumping all over the map, it is harder for them to stay in sync).
Last edited by jalbert on Thu Sep 28, 2017 11:46 am, edited 1 time in total.
Risk is not a guarantor of return.

alex_686
Posts: 3855
Joined: Mon Feb 09, 2015 2:39 pm

Re: REITs as "pseudo diversification"

Post by alex_686 » Thu Sep 28, 2017 11:44 am

jalbert wrote:
Thu Sep 28, 2017 11:42 am
And increased volatility does not increase correlation.
Why do you say this?

jalbert
Posts: 3709
Joined: Fri Apr 10, 2015 12:29 am

Re: REITs as "pseudo diversification"

Post by jalbert » Thu Sep 28, 2017 12:58 pm

alex_686 wrote:
Thu Sep 28, 2017 11:44 am
jalbert wrote:
Thu Sep 28, 2017 11:42 am
And increased volatility does not increase correlation.
Why do you say this?
Here is an example. Flip two coins in a repeated time series and measure sample correlation. Suppose they are fair coins (50% chance of heads or tails each). They will match values (2 heads or 2 tails) with probability 0.5 and differ with 0.5 probability. Now suppose both coins are unfair with a 0.9 probability of heads. Variance of outcome is much lower.The same time series has a 0.82 probability of matching values on a flip, which translates to much higher correlation.

Low variance does not guarantee high correlation-- suppose one coin is biased to heads and one biased to tails. Now the correlation is strongly negative.

But high variance is a hurdle to overcome for high correlation as the example shows.
Last edited by jalbert on Fri Sep 29, 2017 2:19 am, edited 1 time in total.
Risk is not a guarantor of return.

alex_686
Posts: 3855
Joined: Mon Feb 09, 2015 2:39 pm

Re: REITs as "pseudo diversification"

Post by alex_686 » Thu Sep 28, 2017 2:13 pm

jalbert wrote:
Thu Sep 28, 2017 12:58 pm
Low variance does not guarantee high correlation-- suppose one coin is biased to heads and one biased to tails. Now the correlation is strongly negative.

But high variance is a hurdle to overcome for high correlation as the example shows.
Let us try a slightly different game.

First, let us flip 2 fair coins 80 times. Heads = 1, Tails = 2. Run a correlation test. We should set 0. Agreed?

Second, let us roll 2 6-sided dice with values form 1 to 6. Run a correlation test. We should set 0. Agreed?

Now, let us add 80 flipped coins and the 20 rolled dice. Run a correlation test. We should get something between .5 and .6 Agreed?

If that last one has you confused try it yourself. It is pretty easy to do. Now, do you think that by adding these 2 sets of random numbers together we can predict future numbers? I don't think so and I don't think you do either.

I am sure you have heard the phrase "correlation does not equal causation". This is a good example. By increasing the variation within the set I can increase the correlation. No causation needed. Just pure randomness.

Going back to REITs. Over the time period you chose volatility changed and REITs volatility had more volatility than utility's volatility. So here is the question. Are utilities a better diversifer or are what we are seeing is a meaningless statistical artifact. Then we have to throw in questions about 2008. No correlation here. Real estate blew up, which blew up the economy and the market in general and REITs specifically. How much weight do we want to give that event? As a counterpoint I would point you to the Dot Com melt down. REITs where a excellent diversifer then.

User avatar
patrick013
Posts: 2372
Joined: Mon Jul 13, 2015 7:49 pm

Re: REITs as "pseudo diversification"

Post by patrick013 » Thu Sep 28, 2017 3:00 pm

I like beta and wonder why others don't use it. Too accustomed
to a correlation number I guess.

Beta has positive, negative, and measurements over +1 and less than
-1, so it does a good job showing that different funds have different
market behaviors.

Some stats show total return correlations or price correlations so have
to read between the lines to grasp that at times.
age in bonds, buy-and-hold, 10 year business cycle

jalbert
Posts: 3709
Joined: Fri Apr 10, 2015 12:29 am

Re: REITs as "pseudo diversification"

Post by jalbert » Thu Sep 28, 2017 3:18 pm

You are conflating different concepts. Correlation is a property of joint probability distributions, which describe the probability of future joint outcomes. Sample correlations from historical samples may be used as estimators for the actual correlation of the actual joint distribution of future returns. Central limit theorems for sample covariances are complex, requiring some non-trivial preconditions on things like kurtosis integrals being finitely bounded if I recall correctly, but like all central limit theorems, they require a sufficiently large random sample of independent trials for the sample statistic to converge to the distribution parameter being estimated.

In plain language, if you exclude a period like 2008/2009 you are excluding an event that actually occurred, which means it has a non-zero probability of occurring, which means it has to be accounted for in the joint probability of future returns. That doesn't mean we should assume it will re-occur in the next few years-- the probability may be very low of it re-occurring, so yes, giving it a lot of weighting in a sample used to estimate correlation may overestimate actual correlation of REITs and equities, but excluding it is also likely to underestimate actual correlation.

REITs haven't been around long enough to have a sufficiently large random sample to get reliable estimators for return distribution parameters.

Investors may not care about the actual/overall correlation of assets across all possible outcomes. Some investors may in fact prefer to stress test correlation estimates using periods of market stress. I don't overly worry about sample correlations when the market is tame. I care about whether purported lower correlation held up in times of market duress. A biased sample from, say, all bear markets since WW2 is more interesting to me.
Risk is not a guarantor of return.

User avatar
LadyGeek
Site Admin
Posts: 48097
Joined: Sat Dec 20, 2008 5:34 pm
Location: Philadelphia
Contact:

Re: REITs as "pseudo diversification"

Post by LadyGeek » Thu Sep 28, 2017 3:32 pm

This thread is now in the Investing - Theory, News & General forum (theory).
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.

User avatar
patrick013
Posts: 2372
Joined: Mon Jul 13, 2015 7:49 pm

Re: REITs as "pseudo diversification"

Post by patrick013 » Thu Sep 28, 2017 3:51 pm

jalbert wrote:
Thu Sep 28, 2017 3:18 pm
You are conflating different concepts. Correlation is a property of joint probability distributions, which describe the probability of future joint outcomes.
You're probably right. Using return correlation is dissimilar to using
price correlation which is more similar to beta.

If I have a small cap with beta of 1.5 I can expect high volatility more so
than a dividend fund with beta of .75 in a bull market period. One to one
comparisons of return correlations seem less reliable than average return
estimates but I haven't used them much, except a few overall portfolio
estimates which ended up using average return estimates anyway. So if I have
negative return correlation I know I have less earnings to establish price in
the sample somewhere.

Many things can affect return but a revenue correlation could unlock some
relationships. With aggregate stats I get high correlations between things
like CPI and 500 market price. Other stats relate to a sector's business revenues.

I don't think I'd glorify negative correlation as much as low beta for portfolio
stability anyway you look at it.
age in bonds, buy-and-hold, 10 year business cycle

A440
Posts: 225
Joined: Mon Sep 02, 2013 7:46 am
Location: NJ

Re: REITs as "pseudo diversification"

Post by A440 » Sat Sep 30, 2017 6:36 am

We hold about 10% Vanguard REIT index as part of our diversified Roth portfolio.
We purchased a single MREIT (Mortgage REIT) stock, TWO, as an additional way to generate some quarterly dividends to help pay for family vacations. We reinvest capital gains but take dividends from taxable accounts. After paying the taxes, TWO and the dividends of our other taxable funds (mostly index) pay for our modest family vacations every year.

Valuethinker
Posts: 36046
Joined: Fri May 11, 2007 11:07 am

Re: REITs as "pseudo diversification"

Post by Valuethinker » Sat Sep 30, 2017 7:01 am

A440 wrote:
Sat Sep 30, 2017 6:36 am
We hold about 10% Vanguard REIT index as part of our diversified Roth portfolio.
We purchased a single MREIT (Mortgage REIT) stock, TWO, as an additional way to generate some quarterly dividends to help pay for family vacations. We reinvest capital gains but take dividends from taxable accounts. After paying the taxes, TWO and the dividends of our other taxable funds (mostly index) pay for our modest family vacations every year.
The long run returns for Mortgage REITs are absolutely dreadful. I believe VG does not include them in its VNQ fund? They really are a specialized form of financial, and during the financial crash the Mortgage REIT sub index dropped something like 70-80%?

You are using dividends as a way of budgeting. That's conceptually nice, but it can lead you to make some significant asset allocation errors.

A440
Posts: 225
Joined: Mon Sep 02, 2013 7:46 am
Location: NJ

Re: REITs as "pseudo diversification"

Post by A440 » Sat Sep 30, 2017 10:00 am

Agreed.
I am holding it Short-term as our only individual stock, until I can find another investment that pays 10% or more. It has a very low allocation in our taxable holdings. BTW, the investment is up 15% YTD. We've held it for a little over 5 years. The price goes up and down, but the dividend is fairly consistent. If we sold this month, we would still have a capital gain. :D

stlutz
Posts: 4752
Joined: Fri Jan 02, 2009 1:08 am

Re: REITs as "pseudo diversification"

Post by stlutz » Sat Sep 30, 2017 11:39 am

REITs were popular in a bunch of investing books that were published about 15 years ago because over the 30 years prior to then, they had outperformed the market and were not all that correlated to the market. Thus by including them you would have had higher return and lower realized risk.

Of course the problem is that those books were published 30 years too late. One of the factors that drove that outperformance was the transition from being an esoteric to a mainstream asset class. That can only happen once. And once it does, multiples are higher. So, I don't think there is a structural reason to think that REITs will beat the market over the next 30 years.

That does not mean that that overweighting REITs is a bad idea. The lower-correlation aspect still has merit. However, as has been pointed out by others, there are also other sub-sectors of the overall market that offer lower correlation as well. If you're seeing correlation as a reason to overweight REITs but not utilities or precious metals equities (to use examples mentioned earlier), then something is wrong with your thinking about REITs. Either you should be overweighting multiple sectors or none of them, but just picking one and treating it as being super-special doesn't make sense to me.

The argument for TSM is not and never has been that it provides the optimal weight of various sectors of the market. It does not. Rather, it offers an allocation that is "good enough" and offers simplicity and low costs--both management and tax costs.

However, I could certainly use sector funds to come up with an allocation that I find to be more optimal than TSM is, whether I'm using mathematical portfolio optimization techniques or just logically trying to balance out various economic risks and opportunities. For me personally, I don't see enough of an advantage in doing so and it's fair from likely that all of this analysis will actually yield superior risk-adjusted returns in the future. However can rationally come to a different conclusion on this matter than I have.

stlutz
Posts: 4752
Joined: Fri Jan 02, 2009 1:08 am

Re: REITs as "pseudo diversification"

Post by stlutz » Sat Sep 30, 2017 11:43 am

I think individual property ownership should be thought of as starting and running your own business as opposed to being like passive equity investment. How profitable such a business will be depends a lot on how much time and effort you put into it--first in selecting good properties and second in doing most of the property management yourself.

User avatar
abuss368
Posts: 12844
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: REITs as "pseudo diversification"

Post by abuss368 » Sat Sep 30, 2017 8:37 pm

We have invested in the Vanguard U.S. and International REIT Index funds for many years and plan to stay the course.
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

jalbert
Posts: 3709
Joined: Fri Apr 10, 2015 12:29 am

Re: REITs as "pseudo diversification"

Post by jalbert » Sat Sep 30, 2017 9:11 pm

Vanguard's US REIT Index fund and ETF track the MSCI US REIT Index, which does not include mortgage REITs. I think the Dow Jones US Real Estate (REIT) index includes mortgage REITs, while the Cohen & Steers Realty Majors Index and the MSCI US REIT Index both exclude mortgage REITs.
Last edited by jalbert on Sun Oct 01, 2017 1:10 pm, edited 1 time in total.
Risk is not a guarantor of return.

User avatar
ERMD
Posts: 291
Joined: Thu Dec 26, 2013 11:26 am

Re: REITs as "pseudo diversification"

Post by ERMD » Sun Oct 01, 2017 7:09 am

nisiprius wrote:
Wed Sep 27, 2017 6:25 pm
Very tangential but I noticed in the proxy vote that for shareholders of VGSIX, the Vanguard REIT Index Fund, there is "Proposal to reclassify the diversification status to nondiversified."
just got the proxy in the mail today and noticed that, in addition to what you noted above, there is also a vote to "Change the investment objective of Vanguard REIT Index fund.." to include real estate management and development companies. so, ironically (?), they are asking to reclassify VGSIX as "nondiversified" for technical reasons outlined in the ICA, but are also simultaneously seeking to broaden the fund's investments (although on closer reading, the primary focus seems to be to continue to track the REIT spliced index without having to worry about being constrained by the fund's size running against the 5%/10% limitations in the ICA, and so i'm not sure how much more "diversified" it will become in practice).
between scotch and nothing, i'll take scotch. -- faulkner

User avatar
zonto
Posts: 119
Joined: Tue Aug 24, 2010 3:45 pm
Location: Boston, MA

Re: REITs as "pseudo diversification"

Post by zonto » Sun Oct 01, 2017 8:07 am

Nothing "pseudo" about the diversification here. Public REITs perform more like private real estate, and their correlation to the broad stock market drops, over longer periods.

See:
Historically, adding REITs to a stock portfolio comprised of broad market indices served to both reduce volatility and increase CAGR, when rebalanced regularly. This despite REITs being included in total market indices at market weight, which for the U.S. market means roughly 3%.

See:
  • Portfolio Visualizer efficient frontier plot of the U.S. stock market and U.S. REIT indices since 1994 (through end of August 2017): https://tinyurl.com/ybud5edp. The maximum volatility reduction was achieved with 20% REITs.
  • The same held true (and at/near the same ratio) with the total international stock market and foreign real estate indices since 2011 (the year of Vanguard’s foreign real estate index fund’s inception). See: https://tinyurl.com/y8sr9n6a.
  • Portfolio Visualizer efficient frontier plot of Vanguard’s U.S. REIT index fund (VGSIX) and Vanguard’s global real estate index fund (VGXRX) since 2011 (again, year of the latter's inception): https://tinyurl.com/ycxullrx. The maximum volatility reduction was achieved with about 50% U.S. and 50% foreign real estate securities.
U.S. REITs as an asset class are relatively expensive now, especially compared to foreign real estate securities. Per Morningstar data as of this morning, VGSIX and VGXRX, respectively, compare as follows:

Price / prospective earnings: 31.45 vs. 12.37
Price / cash flow: 14.30 vs. 7.66
Dividend yield: 4.28% vs. 3.72%
2017 YTD returns: 3.41% vs. 19.82%

I personally use Schwab’s Fundamental Global Real Estate Index (SFREX) which has current geographic allocations of 45% U.S. and 55% international. My target for REITs is 20% of my equity, which for me means about 14% of the total portfolio.
“Diversification is about accepting good enough while missing out on great but avoiding terrible.” - Ben Carlson

User avatar
abuss368
Posts: 12844
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: REITs as "pseudo diversification"

Post by abuss368 » Sun Oct 01, 2017 9:01 am

ERMD wrote:
Wed Sep 27, 2017 3:10 pm
hightower wrote:
Wed Sep 27, 2017 2:55 pm
If I had more money to invest and the time to do it, I would consider an investment property like a vacation rental somewhere that we could also use for a few weeks a year, but right now we can't afford that kind of investing. Hawaii would be my choice though:) :sharebeer 8-)
the rentable beach house dream is always in the back of our mind too, although hawaii is a bit far/expensive :wink:
One option could be to build up a REIT fund until there is sufficient cash flows from dividends to pay the mortgage on the beach house!
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

Valuethinker
Posts: 36046
Joined: Fri May 11, 2007 11:07 am

Re: REITs as "pseudo diversification"

Post by Valuethinker » Sun Oct 01, 2017 9:21 am

stlutz wrote:
Sat Sep 30, 2017 11:39 am
REITs were popular in a bunch of investing books that were published about 15 years ago because over the 30 years prior to then, they had outperformed the market and were not all that correlated to the market. Thus by including them you would have had higher return and lower realized risk.

Of course the problem is that those books were published 30 years too late. One of the factors that drove that outperformance was the transition from being an esoteric to a mainstream asset class. That can only happen once. And once it does, multiples are higher. So, I don't think there is a structural reason to think that REITs will beat the market over the next 30 years.

That does not mean that that overweighting REITs is a bad idea. The lower-correlation aspect still has merit. However, as has been pointed out by others, there are also other sub-sectors of the overall market that offer lower correlation as well. If you're seeing correlation as a reason to overweight REITs but not utilities or precious metals equities (to use examples mentioned earlier), then something is wrong with your thinking about REITs. Either you should be overweighting multiple sectors or none of them, but just picking one and treating it as being super-special doesn't make sense to me.

The argument for TSM is not and never has been that it provides the optimal weight of various sectors of the market. It does not. Rather, it offers an allocation that is "good enough" and offers simplicity and low costs--both management and tax costs.

However, I could certainly use sector funds to come up with an allocation that I find to be more optimal than TSM is, whether I'm using mathematical portfolio optimization techniques or just logically trying to balance out various economic risks and opportunities. For me personally, I don't see enough of an advantage in doing so and it's fair from likely that all of this analysis will actually yield superior risk-adjusted returns in the future. However can rationally come to a different conclusion on this matter than I have.
I am OK with all of this logic

BUT

There's a reason REITs are different from all other equity sub sectors.

REITs are tied to the cash flow of properties. The 90% payout ratio makes that certain.

Management of REITs cannot retain earnings. Empirically, the market turns out to be a better deployer of free cash flow than management teams (on average, in aggregate).

In addition property is:

- bond like in that leases are fixed payments
- rank ahead of equity in terms of claims on tenant cash flow, thus have lower risk (companies need to keep paying the rent to keep operating, even in bankruptcy after a certain amount of time)
- leases tend to rise with inflation, so there is greater inflation correlation than with equities as a whole

For all these reasons, there are good theoretical and empirical reasons why REITs, and investments in commercial property, are not just like any other equity.

A closer analogy in the US would be MLPs, or in Canada Income Trusts- -again where the payout of cash flow rather than its retention is mandated. Also infrastructure assets (in institutional portfolios, post the construction phase).

Post Reply