Swedroe new book, question on international REITS

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Swedroe new book, question on international REITS

Post by LH » Mon Aug 24, 2009 10:51 pm

International REITS I have been considering for some time.

Since REITS basically have to be in tax sheltered (unless on is withdrawal stage I guess), one loses the FTC with the foriegn reits, and likely have higher other expenses than US reits (ER and such).

Swedroe cited data in his recent book that showed international reits have low correlation with us stocks and bonds and international stocks. Which is good.

In deciding whether or not to include international REITS, I thought it may be important to know if the international REITS have low correlation with US REITS.

Because basically, its a question to me of having X amount all in US REITS as I do currently, versus having x/2 US REITS, x/2 International REITS.

So, given that international REITS have some downside to US REITS, loss of FTC (foriegn tax credit), and higher ER, if there is not low correlation with US REITS, then why bother was my gestalt??????

Dunno if a proper feeling or not on the matter, but a US to Foriegn historical correlation would seem to be something to consider in my situation? If US REITS have high correlation with foriegn REITS, why take the expense and FTC hit?

Thanks for help,

LH

PS I also have the question of where in the world did I put swedroes new book, but I hope no one can answer that for me. If someone said, its behind you in the blue chair, that would freak me out : P

User avatar
csoren
Posts: 267
Joined: Fri Feb 29, 2008 12:22 pm
Location: DC 'burbs

Post by csoren » Tue Aug 25, 2009 6:31 am

I was wondering recently whether IS/ISV funds hold a substantial chunk of international REITs. (US SV includes 12% ?? or thereabouts). Wisdom Tree provides the % for their smaller funds:

DLS (Dev Small) 5.3%
DGS (EM Small) 5.1%
DIM (Dev Midcap) 7.4%

But I can't find it for SCZ, GWX or VSS.

DLS is probably the most "value-y" of the IS funds (not knowing about VSS yet). I wonder whether the other IS funds hold a similar REIT allocation.

I have DLS VSS and DGS at the moment; I think that wee bit of int'l REIT is enough for me.
LH wrote:PS I also have the question of where in the world did I put swedroes new book, but I hope no one can answer that for me. If someone said, its behind you in the blue chair, that would freak me out : P
Check under the couch. My cat usually puts things there.

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

Post by larryswedroe » Tue Aug 25, 2009 7:44 am

Think about it this way.

International REITS are about the perfect equity asset class as they have low correlation with every other equity asset class (both domestic REITS and their domestic equity counterparts). Of course like all equities they have beta exposure so in times of crises they will suffer like all others---and if a financially driven crises (liquidity risks) then they will likely get hit even worse.

So you have a high diversification benefit. But you lose the FTC. I would estimate it at 9% of dividend yield. But that depends on country allocations and treaties. So you have to weigh the diversification benefit (assuming you are a disciplined rebalancer or there is none) versus the loss of the FTC. Some will find it worth the price and others not.

mikep
Posts: 3723
Joined: Wed Apr 22, 2009 9:27 pm

Post by mikep » Tue Aug 25, 2009 10:54 am

Any word on if Vanguard will get an international REIT index fund in the near future?

hafis50
Posts: 332
Joined: Fri Jun 22, 2007 1:55 am

Re: Swedroe new book, question on international REITS

Post by hafis50 » Tue Aug 25, 2009 11:48 am

LH wrote:I

In deciding whether or not to include international REITS, I thought it may be important to know if the international REITS have low correlation with US REITS.
I have no exact data and I believe it is difficult to perform such a study because you would have to separate currency returns from property returns.

The only data I have:

EPRA- Correlations of Property Stocks with other Asset Classes

They calculate the rolling five-year correlations (1990-2006) of US, Australian and European (4 countries) property stocks with US stocks.
I believe the foreign currencies were hedged (see Methodology in the first chapter).

Australian property stocks had a higher man of correlation with Us stocks than US property stocks. with US stocks.
All European indices had a lower mean of correlation.

Correlations were lower before 2000.

I wonder if the improved investability of this asset clalss will increase correlations.

If many investors invest in regional or global REITs portfolios every purchase or sale will indiscriminately buy and sell all countries at the same time.
OTOH, market prices should track the fair value over the long haul.

User avatar
Opponent Process
Posts: 5157
Joined: Tue Sep 18, 2007 9:19 pm

Post by Opponent Process » Tue Aug 25, 2009 12:26 pm

mikep wrote:Any word on if Vanguard will get an international REIT index fund in the near future?
probably not in the near future but you may be interested in the Northern Global Real Estate Index Fund. it's pretty expensive at 0.65 ER:

http://quote.morningstar.com/fund/f.aspx?t=NGREX
30/30/20/20 | US/International/Bonds/TIPS | Average Age=37

xerty24
Posts: 4827
Joined: Tue May 15, 2007 3:43 pm

Post by xerty24 » Tue Aug 25, 2009 1:46 pm

Opponent Process wrote:you may be interested in the Northern Global Real Estate Index Fund. it's pretty expensive at 0.65 ER:

http://quote.morningstar.com/fund/f.aspx?t=NGREX
Pretty expensive? You should see how much the other international RE funds cost :shock:.

User avatar
MossySF
Posts: 2361
Joined: Thu Apr 19, 2007 9:51 pm
Contact:

Post by MossySF » Tue Aug 25, 2009 2:20 pm

Northern Rock is 25% Domestic which means you are actually paying even more for the international portion. If we plug in Vanguard REIT's 0.20%, we get: 0.20%. (X*3+0.20)/4=0.65% --> X = 0.80%.

The best options seem to be ETFs at this point. WisdomTree, iShares and State Street all have offerings in the 0.50%-0.60% range.

User avatar
Topic Author
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Post by LH » Tue Aug 25, 2009 6:33 pm

larryswedroe wrote:Think about it this way.

International REITS are about the perfect equity asset class as they have low correlation with every other equity asset class (both domestic REITS and their domestic equity counterparts). Of course like all equities they have beta exposure so in times of crises they will suffer like all others---and if a financially driven crises (liquidity risks) then they will likely get hit even worse.

So you have a high diversification benefit. But you lose the FTC. I would estimate it at 9% of dividend yield. But that depends on country allocations and treaties. So you have to weigh the diversification benefit (assuming you are a disciplined rebalancer or there is none) versus the loss of the FTC. Some will find it worth the price and others not.
Ok, so international reits have low correlation with US REITS.

That to me makes much more of a case for adding in the international REITS. In fact, with my one country risk of near 50 percent of my portfolio being US, and being where I work too (US), that will likely swing me over into going half US, half international REIT get correlation benefits at the expense of 9 percent of dividend yield cost, get 7 percent more out of US, and basically maintain current allocation...

thanks for help, I am enjoying your new book.

LH
Last edited by LH on Tue Aug 25, 2009 6:50 pm, edited 1 time in total.

User avatar
Topic Author
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Re: Swedroe new book, question on international REITS

Post by LH » Tue Aug 25, 2009 6:40 pm

hafis50 wrote:
LH wrote:I

In deciding whether or not to include international REITS, I thought it may be important to know if the international REITS have low correlation with US REITS.
I have no exact data and I believe it is difficult to perform such a study because you would have to separate currency returns from property returns.

The only data I have:

EPRA- Correlations of Property Stocks with other Asset Classes

They calculate the rolling five-year correlations (1990-2006) of US, Australian and European (4 countries) property stocks with US stocks.
I believe the foreign currencies were hedged (see Methodology in the first chapter).

Australian property stocks had a higher man of correlation with Us stocks than US property stocks. with US stocks.
All European indices had a lower mean of correlation.

Correlations were lower before 2000.

I wonder if the improved investability of this asset clalss will increase correlations.

If many investors invest in regional or global REITs portfolios every purchase or sale will indiscriminately buy and sell all countries at the same time.
OTOH, market prices should track the fair value over the long haul.
Thanks for reply.

In terms of the part I bolded above, from my portfolios view, what functional difference does it make if the low correlation is from the property, or from the currency? I mean, to really isolate it out, yeah I am with you, and appreciate it, but where the pedal hits the medal in constructing portfolio, I do not see that it matters????? I may well be wrong here, and its interesting to talk about.

I would posit that low correlation, is low correlation, as long as it can be reasonably expected to continue going forward?

So if international RE is low correlated with: US stocks, International stocks, US REITS, US Bonds, then it would seem to have an expected benefit regardless of what part came from property returns, and what part came from currency returns????

I mean to me, to get a bit silly, I do not care if the expected low correlation and return came from martians controlling it and providing it, as long as I expected that the martians would continue to do so going forward?

Now, as a matter of theory, separating the currency from the RE component is interesting regardless, I am just uncertain if functionally that matters. I have not read a breakdown per se of correlation due to currency versus correlation due to business return for say international stocks either that I remember. Does it functionally matter for international stocks? Thanks for bringing that up. Thanks for the link.

Thanks for help,

LH

huskerblue
Posts: 12
Joined: Thu Jan 08, 2009 10:04 pm

Post by huskerblue » Tue Aug 25, 2009 7:01 pm

Have I missed something? A thread about int'l REIT's and no mention of RWX?

the fixer
Posts: 75
Joined: Sat Jul 25, 2009 6:19 pm

Post by the fixer » Tue Aug 25, 2009 7:50 pm

larryswedroe wrote:Think about it this way.

International REITS are about the perfect equity asset class as they have low correlation with every other equity asset class (both domestic REITS and their domestic equity counterparts). Of course like all equities they have beta exposure so in times of crises they will suffer like all others---and if a financially driven crises (liquidity risks) then they will likely get hit even worse.
I don't know if I'd call international REITS "the perfect equity asset class". International small value stocks maybe. International REITS? not really.

We only have data on international REITS dating back to 1989 (and even there, we are only talking about 1 or 2 countries). The correlation between iREITS and the S&P 500 has been +0.37. Between international small value and the S&P 500? Only +0.18. Furthermore, the historical return of international REITS has been about 1.6% less per year than international small value. Over longer periods of time (using Wilshire REIT data), we have seen that REITS perform about in-line with market indexes (1/78 through 7/09 shows Wilshire REIT ahead of the CRSP 1-10 total stock index by 0.1% per year). Small value stocks, on the other hand, come in at 4% to 5% per year. Since 1981, int'l small value has indeed outpaced the EAFE by about 5.4% per year. Thats been higher than the +3.3% figure in the US. But averaged together, the global (developed) premium for small value stocks has been just what you'd expect: 4.5% or so.

REITS do have some modest risk reducing characteristics, but the idea of taking money away from US or international small value stocks to fund a REIT allocation is a pretty bad idea. You get more "bang" for your diversification buck with small value stocks from a return standpoint. And if the risk is too high, bonds are always a more reliable reducer of risk.

Fixed

hafis50
Posts: 332
Joined: Fri Jun 22, 2007 1:55 am

Re: Swedroe new book, question on international REITS

Post by hafis50 » Wed Aug 26, 2009 7:23 am

LH wrote:In terms of the part I bolded above, from my portfolios view, what functional difference does it make if the low correlation is from the property, or from the currency? I mean, to really isolate it out, yeah I am with you, and appreciate it, but where the pedal hits the medal in constructing portfolio, I do not see that it matters????? I may well be wrong here, and its interesting to talk about.

I would posit that low correlation, is low correlation, as long as it can be reasonably expected to continue going forward?

So if international RE is low correlated with: US stocks, International stocks, US REITS, US Bonds, then it would seem to have an expected benefit regardless of what part came from property returns, and what part came from currency returns????
LH
I think they say that random currency movements (that can go in one direction for years) can mask the "true" underlying correlations.
Therefore, I wouldn't be sure that the historical correlations of the last (20) years are quite close to the "true long-term" averages.
OTOH, what if returns on foreign currencies were related with the returns on foerign REITs for fundamental economic reasons?

Furthermore, I believe correlations can change for various reasons (e.g., companies could start foreign investments, changing risk or diversification prefernences, investments of foreign investors).

Active investors need to estimate future correlations and returns. I am agnostic because I don't know more than the market.
the fixer wrote:REITS do have some modest risk reducing characteristics, but the idea of taking money away from US or international small value stocks to fund a REIT allocation is a pretty bad idea. You get more "bang" for your diversification buck with small value stocks from a return standpoint. And if the risk is too high, bonds are always a more reliable reducer of risk
Shouldn't higher diversification benefits be related with lower returns?

From a theoretical standpoint, investors like assets that reduce portfolio risk and they bid up their prices to the point that they have lower expected returns.
If all (domestic and foreign) investors diversify internationally assets that have the same risk will have the same expected returns.

REITs and small value stocks could have very different risks.
And would foreign REITs, foreign small value stocks and domestic bonds move in the same direction in the case of a global inflation?

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

Post by larryswedroe » Wed Aug 26, 2009 8:01 am

hafis, international small value stocks have lower correlation to US stocks and HIGHER returns.

Here is some data on int'l REITS from my book on alternative investments

= For the 15-year period ending November 2006, the correlation of Australian REITs to the public equity market in Australia was 0.76. In the Netherlands, the correlation of the national REIT market to the national equity market was 0.51. The correlation of international REITs to international equities was 0.6, and the correlation of U.S. REITs to U.S. equities was just 0.36.
= By limiting the data to the 10-year period ending in November 2006, we add two more countries. The correlations of REITs in Australia, the Netherlands, Belgium and Canada to their local equity markets ranged from 0.48 to 0.78. The correlation of international REITs to international equities was just 0.62 and the correlation of U.S. REITs to U.S. equities was just 0.33.
= When we limit the data to the five-year period ending November 2006, our universe expands to six countries, adding Japan and New Zealand. Correlations of REITs to their local equity markets ranged from 0.43 to 0.71. The correlation of international REITs to international equities was 0.61 and the correlation of U.S. REITs to U.S. equities was just 0.44.

The low correlations of REITs to their local markets, international equities and U.S. equities suggest that the addition of this asset class to a portfolio should provide significant diversification benefits.

As I noted earlier though the issue is one of location due to the FTC, and costs of the funds. And given that div yields now very high (say 7%) that means you have another cost of about 0.6 percent to add to the expenses of the funds, assuming you hold in tax advantaged accounts. In DFAs case the ER of the fund is 65 bp. So you have to decide if the diversification benefits, and the expected returns are worth the cost of 1.25%.

Tramper Al
Posts: 3629
Joined: Thu Oct 18, 2007 11:42 am

Post by Tramper Al » Wed Aug 26, 2009 8:16 am

Can we buy an International REITs-only fund or ETF yet? Or do they all still have a mix of REITs and those REOCs - publicly traded stocks which have some business activity or another related to real estate?

Just curious. I am as excited as the next guy about any "new" asset class area, particularly a "real" one. You can sign me up for a direct ownership TIAA RE Euro/Pac version right now. I'd be interested in international REITs in a pure form, with presumably better diversification properties vs. an already equities-dominated portfolio. I'm just not sure we'd be so happy having all those non-REIT components in VNQ and the like, you know?

User avatar
Rick Ferri
Posts: 9113
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Wed Aug 26, 2009 8:35 am

Theoretically, international real estate has its merits. There is at times low correlation within their own country as Larry point out and at times low correlation with other asset class.

Unfortunately, there are some strong disadvantages. First, like US equity REITS, the correlation with local common stocks and other risky asset classes can become highly postitive just when you want it negative (an example is 2008). Second, international real estate is expensive to buy. Most funds are 4 times or more the cost of a US REIT index funds (VNQ is only 0.12%). Also, many funds are global, meaning they us US REITS as well, yet they charge just as much if not more than a foreign real estate fund so you are are paying an exceptionally high fee to by the US portion. Third, there are tax issues as talked about earlier. You will lose some of your return to foreign taxes.

At this time I like the asset class but there are no funds that meet the strict criteria for us to buy funds. So, I don't recommend international real estate because of the limited product availability and high fees rather than the asset class itself.

Rick Ferri

xerty24
Posts: 4827
Joined: Tue May 15, 2007 3:43 pm

Post by xerty24 » Wed Aug 26, 2009 9:24 am

larryswedroe wrote:international small value stocks have had lower correlation to US stocks and HIGHER returns.
A small but important caveat added above. Past performance and all...

gizzsdad
Posts: 129
Joined: Fri Feb 23, 2007 2:23 pm
Location: midwest

Post by gizzsdad » Wed Aug 26, 2009 9:59 am

When the family I use added Global RE nearly two years ago, I changed from the domestic to the Global right away, figuring that if I am using REITS as a diversifier, then global REITS have to be an even better diversifier - not to mention that the manager has more flexibility to find value.

The global fund costs 18 basis points more than the domestic fund.

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

Post by larryswedroe » Wed Aug 26, 2009 10:54 am

xerty
That is true but there is no logical reason to assume that won't continue--remember all real estate is local, and unless you expect all economies to move in sink all the time, the correlations should be relatively low

Now of course what is important to point out, but should not be necessary, is that the correlations of all risky assets tend to go toward one when we have financial and/or geopolitical crises. That is why one should limit their fixed income to ONLY US government or very high investment grade bonds if buying municipals.

What I find interesting is that the same people that say you should avoid an asset class because it CAN have high correlation at the wrong time, also recommend high yield bonds which have the same characteristic because they are hybrid instruments--returns explained partly by equities and partly by bonds

User avatar
Rick Ferri
Posts: 9113
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Wed Aug 26, 2009 11:45 am

larryswedroe wrote:What I find interesting is that the same people that say you should avoid an asset class because it CAN have high correlation at the wrong time, also recommend high yield bonds which have the same characteristic because they are hybrid instruments--returns explained partly by equities and partly by bonds
HY is not a hybrid security. A hybrid security is made up of two asset classes such as a bond and an option. For example, a convertible bond is composed of an underlying interest bearing bond and an attached equity option. The price of the interest bearing bond portion and the price of the underlying option are the ONLY factors that change the price of the hybrid convertible bond.

HY is not a hybrid security, nor does it act like one. There is no equity 'option' that underlines the price of a HY bond. While at times the price of the equities and HY bond credit spreads do have positive correlation, that correlation is not consistent and it is not always positive. Thus there is unique risk in high yield bonds that is not found in equity risk. This is what makes HY a unique asset class.

Rick Ferri

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

Post by larryswedroe » Wed Aug 26, 2009 12:27 pm

First a hybrid is not determined by what it is called, like a high yield bond. But what explains its returns. Preferred stocks are hybrids, as are convertible bonds and high yields.

Second as all the academic evidence shows there is almost nothing unique about high yield bonds as their returns are well explained by the various risk factors--a five factor model.

Reichenstein among others has written on the subject.

And of course we have seen high yield experience high correlations at the same time equities do, because they have equity type risks.

User avatar
drbagel
Posts: 186
Joined: Mon Aug 04, 2008 12:07 pm
Location: Massachusetts

Post by drbagel » Wed Aug 26, 2009 1:38 pm

I love it when you guys argue. It is both educational and entertaining. :)

hafis50
Posts: 332
Joined: Fri Jun 22, 2007 1:55 am

Post by hafis50 » Wed Aug 26, 2009 2:50 pm

larryswedroe wrote:hafis, international small value stocks have lower correlation to US stocks and HIGHER returns.
...
The low correlations of Intl. REITs to their local markets, international equities and U.S. equities suggest that the addition of this asset class to a portfolio should provide significant diversification benefits.
It was suggested to use Intl SV instead of REITs because of "higher expected returns and better or equal portfolio diversification".
SV may have higher reurns than REITs because of higher value risk.

I believe that in a CAPM model higher diversification is related to lower expected returns because investors bid up the prices of assets with low correlations.

And I believe that higher international diversification MIGHT change the risks and returns that were observed in the past.

To assess the benefits of an additional asset class we need to estimate its impact on the volatility of the total portfolio.

But if domestic and international investors diversify internationally the market prices are set by domestic AND international investors.

The market portfolio would be global for everybody and the risk and return of securities would depend on the covariance with this global market portfolio and not with the domestic market.

International investors could increase prices and decrease expected reurns although the returns could still be higher than the market return.

Likewise, correlations could increase if the economies become more interdependent and the businesses increase their international activities.

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

Post by larryswedroe » Wed Aug 26, 2009 2:58 pm

hafis,
But ISV is not international for the investors in their own home countries in the same manner that US SV is international for them, and a better diversifier than their domestic SV

Yes it is possible that correlations will rise but there is also the possibility that they will fall, and IMO RE is likely to be a better diversifier because it is truly local.

Finally re hybrids, there is an old saying, if it quacks like a duck it is a duck. Doesn't matter what you call something, it matters how the security acts.

User avatar
stratton
Posts: 11083
Joined: Sun Mar 04, 2007 5:05 pm
Location: Puget Sound

Post by stratton » Wed Aug 26, 2009 4:31 pm

Rick Ferri wrote:Theoretically, international real estate has its merits. There is at times low correlation within their own country as Larry point out and at times low correlation with other asset class.

Unfortunately, there are some strong disadvantages. First, like US equity REITS, the correlation with local common stocks and other risky asset classes can become highly postitive just when you want it negative (an example is 2008). Second, international real estate is expensive to buy. Most funds are 4 times or more the cost of a US REIT index funds (VNQ is only 0.12%). Also, many funds are global, meaning they us US REITS as well, yet they charge just as much if not more than a foreign real estate fund so you are are paying an exceptionally high fee to by the US portion. Third, there are tax issues as talked about earlier. You will lose some of your return to foreign taxes.

At this time I like the asset class but there are no funds that meet the strict criteria for us to buy funds. So, I don't recommend international real estate because of the limited product availability and high fees rather than the asset class itself.
SPDRs Intl RE etf (RWX) has a 0.6% ER. SPDRs US RE etf (???) has a 0.25% ER. Their global RE etf (RWO) has a 0.5% ER. With a Intl/US ratio of 60/40 it comes in with a 0.46% ER. So it's roughly 4 basis points for the privilege of one fund with a few less stocks than the two separate ETFs. The trading costs alone if you have less than ~$20K in RE might make it worth while to hold RWO if you can live with its limitations.

However, if use Vanguard's VNQ etf the weighted RE is 0.41% and thats with a global Intl/US 60/40 distribution. If you are like most investors and are 50/50 or 70/30 in this area then the weighted ER is 0.36% or 0.27%.

Paul

User avatar
Rick Ferri
Posts: 9113
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Wed Aug 26, 2009 5:39 pm

[personal attack removed by admin alex]

the fixer
Posts: 75
Joined: Sat Jul 25, 2009 6:19 pm

Post by the fixer » Wed Aug 26, 2009 6:56 pm

This debate reminds me of whether eating more cookies or ice cream is a better approach to losing weight!

The real gas is that neither high yield bonds or REITS fit into the Fama/French 5 Factor model, so portfolio inclusion forces you to deviate from the model as defined by the FF research (and both Fama and French have said that neither owns REITS or HY bonds, so they've voted with their minds and pocketbooks).

Proceed at your own caution, for sure.

Fixed

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

Post by larryswedroe » Wed Aug 26, 2009 7:07 pm

the fixer
The research, see Reichenstein for example, does show that high yield even fits a three factor model

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

Post by larryswedroe » Wed Aug 26, 2009 7:18 pm

For those interested in what the research in a peer reviewed journal says, as opposed to opinions.

Journal of Fixed Income
Spring 2008 'Returns-Based Style Analysis of High-Yield Bonds'


DALE L. DOMIAN AND WILLIAM REICHENSTEIN

The returns on BB-rated bonds are similar to those achievable with a portfolio of 22.3% stocks, 73.2% bonds, and 4.5% Treasury bills. To be more precise, the stock portion consists of 2.9% large-cap value, 7.3% largecap growth, 11.5% small-cap value, and 0.6% small-cap growth stocks, while the bond portion consists of 63.3% corporate, 6.2% long-term government, and 3.7% intermediate- term government bonds.6 As Sharpe [1992] stated, “style analysis provides measures that reflect how returns act, rather than a simplistic concept of what the portfolios include” (p. 13).

Based on the embedded put option in HY bonds, we would expect the sum of bond components to decrease and the sum of stock components to increase as the credit rating decreases from BB to B to CCC. The results in Exhibit 5 confirm this expectation. The sum of bond components decreases from 73.2% for BB, to 68.0% for B, and to 48.4% for CCC-rated bonds. The sum of stock components increases from 22.3% for BB, to 32.0% for B, and to 51.6% for CCCrated bonds.

All 60 HY bond funds’ returns exhibit negative skewness and positive kurtosis. The average fund has a skewness of -0.71 and kurtosis of 2.98. These statistics suggest that HY bond funds tend to be prone to more large losses than would occur if returns were normally distributed. (like stocks)

This conclusion is consistent with Reilly and Wright [2001] who found high-yield bond returns were more highly correlated with small-cap than large-cap stock returns.

In short, as predicted by theory, HY bond returns are sensitive to high-grade bond and stock returns. Furthermore, as the credit rating decreases from BB to CCC the sum of bond components decreases from about 73% to 48%, while the sum of stock components increases from about 22% to 52%. There is also a strong small-cap tilt in HY bond index returns.

It is my experience that when people don't like what the research finds they begin to use terms like academic poo poo or whatever. Kind of like the kid who says, "no you are." They fall back on such terms because they cannot come up with intellectual arguments. And failing that they resort to personal attacks, as Rick has done

Also note, my books are also filled with my own research, and my associates who worked for and with me have even had articles published in peer reviewed academic journals and I even have an article that I co-authored that will appear in the November issue of the Journal of Investing. Though I have no need to defend myself I thought I would point out the statements made by Rick are in fact without basis, and the question should be why he would make such claims when he has no clue about whether I do my own research or not, which must now be painfully obvious to anyone. He has also made other false statements which is why I stopped engaging him in any debates. I simply point out when I think he is making statements that IMO are wrong or false or not supported by the evidence so that readers can make an informed decision.
Last edited by larryswedroe on Wed Aug 26, 2009 7:57 pm, edited 1 time in total.

the fixer
Posts: 75
Joined: Sat Jul 25, 2009 6:19 pm

Post by the fixer » Wed Aug 26, 2009 7:37 pm

larryswedroe wrote:the fixer
The research, see Reichenstein for example, does show that high yield even fits a three factor model
Yes, backing into HY through attribution analysis will give you a series of stock/bond allocations. FF, however, said this in their intro of Common Risk Factors in the Returns on Stocks and Bonds:
This paper identifies five common risk factors in the returns on stocks and bonds. There are three stock-market factors: an overall market factor and factors related to firm size and book-to-market equity. There are two bond-market factors, related to maturity and default risks. Stock returns have shared variation due to the stock-market factors, and they are linked to bond returns through shared variation in the bond-market factors. Except for low-grade corporates, the bond-market factors capture the common variation in bond returns.
Fixed

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

Post by larryswedroe » Wed Aug 26, 2009 7:55 pm

Fixed
Low grade corporates are high yield
Except for low-grade corporates, the bond-market factors capture the common variation in bond returns.
AAA bonds have almost no equity component, but even DFA's short term extended credit fund, which is all investment grade, has some significant equity component, which one should account for in their asset allocation if you use the fund. As you go down in rating the equity component increases.

hafis50
Posts: 332
Joined: Fri Jun 22, 2007 1:55 am

Post by hafis50 » Thu Aug 27, 2009 6:19 am

the fixer wrote:so portfolio inclusion forces you to deviate from the model as defined by the FF research (and both Fama and French have said that neither owns REITS or HY bonds, so they've voted with their minds and pocketbooks).

Proceed at your own caution, for sure.

Fixed
I think this misrepresents Fama's ideas:

Every time I read him he said the total market is an excellent choice:

See this post written by John Norstad:
The most direct statement I've seen from Fama on this issue is from a question and answer video interview which is available at the DFA web site.

Q: Some people cite your research showing that value and small firms have higher average returns over time and they assume that you would recommend most investors have a big helping of small and value stocks in their portfolios. Is that a fair representation of your views?

A: ...no... this a risk story...in every asset pricing model, the market portfolio is always an efficient portfolio. It's always a relevant portfolio for an investor to hold. And investors can decide to tilt away from that based on their personal tastes... You can decide to tilt toward more value or smaller size based on your tastes for these dimensions of risk. But you needn't do it. You could also decide to go the other way. You could look at the premiums and say, no, I think I like the growth stocks better. Then, as long as you get a diversified portfolio of them, I can't argue with that either... And I have no recommendations about because I think it's totally a matter of taste...
Emphasis added.

Vanguard Diehards Thread 36850

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

Post by larryswedroe » Thu Aug 27, 2009 7:34 am

Hafis
I don't see how what I wrote disagrees with anything Fama said. In fact, in working with DFA on their new products they agree fully with my view that corporate bonds have equity exposure and the lower the grade the more equity risk there is. We even had them do attribution analysis showing how the new extended credit fund loaded on the equity risk factors.

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

Post by larryswedroe » Thu Aug 27, 2009 7:58 am

Hafis
Further
If you were referring to whether investors should invest in high yield bonds I offer two points.

First, in general just because someone creates a product doesn't mean investors should buy them. There are many products that I believe are meant to be sold and not bought.

Second, and this is more directly related to high yield, note I placed them in the flawed, not the bad or ugly category, in my alternative investment book. There are two main reasons. The first is that unlike non-taxable institutional investors, taxable investors with a choice of location should prefer to place equities (or equity risks) in taxable accounts for a variety of reasons. Among them are the ability to loss harvest. The more volatile the investment, the more the tax option is worth. And obviously high yield is a more volatile asset class than Treasuries. So since you can replicate the exposure to the risk factors separately those with choice should do so. They also save money because they can get the fixed income exposure cheaper--especially if you buy Treasuries directly or in wholesale markets. And you can diversify the risks more effectively on the equity side where a typical DFA type global portfolio owns about 15k stocks, eliminating the unsystematic risks.

Note institutional investors don't have those issues.


The other reason is that high yields contain call risk which I prefer to avoid--no evidence it has been rewarded (in fact the evidence on junk is that it has been penalized) and it doesn't mix well either with equity risks. (note it is interesting that those that are willing to take call risk never seem to write calls on the bonds they own that don't have calls, like Treasuries--if it is good for one instrument, why not others?)

Finally, I have also said that if you decide that investing in high yield is appropriate (you don't have the location issue) then you should account for the equity risks you are taking by adjusting your AA appropriately or you will be taking more risk. Perfect example was in 2008. Say you decide that the most equity risk you will take is 60%, but if you own high yield for the bond side your risks will be greater and thus in a year like 2008 you will EXCEED your tolerance for risk and that could cause your plan to blow up--either because of emotions, or the increased volatility in the withdrawal phase could cause the plan to fail.

I hope that is helpful

User avatar
Rick Ferri
Posts: 9113
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Thu Aug 27, 2009 9:35 am

larryswedroe wrote:For those interested in what the research in a peer reviewed journal says, as opposed to opinions.

Journal of Fixed Income
Spring 2008 'Returns-Based Style Analysis of High-Yield Bonds'


DALE L. DOMIAN AND WILLIAM REICHENSTEIN

The returns on BB-rated bonds are similar to those achievable with a portfolio of 22.3% stocks, 73.2% bonds, and 4.5% Treasury bills. To be more precise, the stock portion consists of 2.9% large-cap value, 7.3% largecap growth, 11.5% small-cap value, and 0.6% small-cap growth stocks, while the bond portion consists of 63.3% corporate, 6.2% long-term government, and 3.7% intermediate- term government bonds.
This regressive analysis study measured a mix of other asset classes that came the closest to the return of high yield over the period measured. That is all it does, and that does not do much. The mix that will be closest to HY in the future is different from period to period. Larry is trying to say that HY is a hybrid, and therefore it is always going to have the same regressive mix. He is mistaken. Neither Larry, Domian or Reichenstein could have predicted what the mix was before the study was done, nor can they predict what the mix will be in the future.

This conversation on international real estate was hijacked into a conversation on high yield because someone did not read or understand my comments about investing in international real estate. So, let's get back on point. As I said in an earlier post, I do believe that international real estate has merit, but I do not like the products that are on the market. Their costs are too high in relation to investing in US real estate and other equity asset classes. Investing in international real estate theoretically makes sense, but economically it does not.

Rick Ferri

User avatar
fluffyistaken
Posts: 1435
Joined: Fri Apr 04, 2008 1:32 pm

Post by fluffyistaken » Thu Aug 27, 2009 9:47 am

Most if not all int'l REIT funds/ETFs will have more than half the holdings in Japan, Australia, and Hong Kong. So it's a concentrated bet on those three countries, in addition to being a bet on a specific sector.

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

Post by larryswedroe » Thu Aug 27, 2009 10:43 am

Larry is trying to say that HY is a hybrid, and therefore it is always going to have the same regressive mix
Why do people make statements that are not true.

In the future it is certainly possible that the mix might be different. As with all data on investing, which is not a hard science, the longer the data we have the more confidence we can have in it, but it is not certain. But we can use the information to understand the nature of the risks and how to account for that in portfolios. This is no different than any correlation data. The fact that correlations can and do shift does not make the information of no value

User avatar
Rick Ferri
Posts: 9113
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Thu Aug 27, 2009 11:09 am

Larry,

Your contradicting yourself. You say high yield is a hybrid security, but then admit it does not act like a hybird security would act over different periods because the regression cannot be predicted as a hybrid would. Thus it cannot be a hybrid. It does not matter, this conversation is about international real estate. Lets stick to the topic.

Rick Ferri

User avatar
Topic Author
LH
Posts: 5490
Joined: Wed Mar 14, 2007 2:54 am

Post by LH » Thu Aug 27, 2009 1:06 pm

thanks for all the replies, great discussion : )

hafis50
Posts: 332
Joined: Fri Jun 22, 2007 1:55 am

Post by hafis50 » Thu Aug 27, 2009 1:30 pm

larryswedroe wrote:Hafis
If you were referring to whether investors should invest in high yield bonds I offer two points.
No. Not to this hot topic :wink:

I wrote about the question if the Fama-French-model recommends that nobody (or no retail investors) should invest in REITs.

pop77
Posts: 457
Joined: Wed Dec 12, 2007 7:57 pm

Post by pop77 » Thu Aug 27, 2009 3:08 pm

fluffyistaken wrote:Most if not all int'l REIT funds/ETFs will have more than half the holdings in Japan, Australia, and Hong Kong. So it's a concentrated bet on those three countries, in addition to being a bet on a specific sector.
I have the same concern. Moreover the Austalian allocation is typically on the Westfield group which owns a big chunk of Real Estate in US.
http://westfield.com/corporate//propert ... ed-states/

I found EGLRX focusses on real estate properties in some emerging economies like Brazil and is one of my holdings. Of course this is an actively managed fund with high expense ratio.

But till I get a true international RE index fund that is well diversified through out the world,represents RE in emerging markets AND is low cost I do not have a choice.

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

Post by larryswedroe » Thu Aug 27, 2009 6:27 pm

hafis
The FF model does not recommend that no one invest in REITS. But taxable investors should only invest in REITs (unless in lowest brackets) if can do so in tax advantaged accounts. Institutions that are tax exempt don't have this issue. So a taxable investor would not invest in REITs if in high brackets. There are reasons that some investors should consider some investments but others not

AS to this false statement
Your contradicting yourself. You say high yield is a hybrid security, but then admit it does not act like a hybird security would act over different periods because the regression cannot be predicted as a hybrid would. Thus it cannot be a hybrid.
I thought it would be impossible to come to this conclusion. But I guess I was wrong. All it says is we don't know the EXACT make up in the future but we do know that there is a significant equity component--whether in the future it will be 20% or 25% or 22.5% we cannot know. But this is really not different than looking at say the beta of RE. We don't know what it will be in future, only what it was. But we know it is equity risk.
Have to remember this is not a hard science, but doesn't mean the data is not meaningful.

All the academic papers I have read agree that junk acts like a security with both equity and bond characteristics for very logical reasons, Ignore it at your peril. Those that ignore that learned a hard lesson in 2008 and it might have caused them to take more risk than they could handle, and the increased volatility for those in withdrawal phase can really be deadly

User avatar
Rick Ferri
Posts: 9113
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Thu Aug 27, 2009 7:00 pm

Larry,

I never said there was no equity 'beta' in high yield bonds returns. I have always said that credit spreads and equity risk are correlated at times. In All About Asset Allocation, I even include charts showing exactly what the rolling 36 month correlation between equity risk and high yield credit spreads have been. The chart also shows that there are many periods when the correlation was very low, meaning that something else is going on that cannot be explained by equity returns. Other charts show that high yield is influences by term risk at times, and at times it is not.

If high at times is not under the same risk influence as equities at times, and not under the same risk influence as term risk at times, then high yield exhibits its own unique characteristics at times. The static linear regression study over an entire 'flat' period does not pick up dynamic movements in correlation. It would be like saying the stock market crash of 1987 did not happen because for the entire year in 1987 the stock market was up 5.1%. You must go beyond the flat long-term regressions and look between periods to see how things really move in relation to each other.

Rick Ferri

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

Post by larryswedroe » Thu Aug 27, 2009 7:47 pm

deleted by admin alex - Could we keep this on topic please?

User avatar
Rick Ferri
Posts: 9113
Joined: Mon Feb 26, 2007 11:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Post by Rick Ferri » Thu Aug 27, 2009 9:07 pm

Ditto
Last edited by Rick Ferri on Fri Aug 28, 2009 7:00 am, edited 1 time in total.

hafis50
Posts: 332
Joined: Fri Jun 22, 2007 1:55 am

Post by hafis50 » Fri Aug 28, 2009 5:20 am

larryswedroe wrote:hafis
The FF model does not recommend that no one invest in REITS... There are reasons that some investors should consider some investments but others not
That's what I wrote in my previous posts.

Just to make it clear: I've never said thet the model recommends it. Somebody else asked this question.
fluffyistaken wrote:Most if not all int'l REIT funds/ETFs will have more than half the holdings in Japan, Australia, and Hong Kong. So it's a concentrated bet on those three countries, in addition to being a bet on a specific sector
If the REITs of some countries have higher market values than the REITs of other countries investors cannot invest in equal proportions.
In May 2007 the USA represented 38% of the global FTSE/EPRA Real Estate Index.
Factsheet and county breakdown, May 2007

Asian investors could argue the weight of US REITs is "too high".
But investors have to find an equilibrium between their intention to diversify and thei intention not to distort the fair market values that are not equally distributed.

User avatar
topper1296
Posts: 684
Joined: Fri Apr 03, 2009 10:50 pm
Location: Nashville TN

Post by topper1296 » Sat Aug 29, 2009 7:34 pm

T Rowe Price has a global REIT fund (PAGEX) that started in Oct 08. It currently has 65% in foreign REITS, 32% in US REITS, and the balance in cash. ER is 1.05% until the expense limitation ends in 4/2011. The gross ER is 4.73%!!, but it also only has about $12M in assets right now.

markcoop
Posts: 1159
Joined: Fri Mar 02, 2007 8:36 am

Post by markcoop » Sun Aug 30, 2009 9:54 am

For those who recommend int'l REITs, I'd be curious to hear how much they recommend?

I have 10% of my stock allocation in REITs. If I divide that among int'l REITs, then I'm left with 5%, or maybe something like a third (if I followed the rest of my domestic to int'l ratio). That works out to a very small percentage of my total portfolio - if I have 60/40 stock/bond ratio, then I would have as little as 2% in int'l REITs.
Mark

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Post by yobria » Sun Aug 30, 2009 10:02 am

In my opinion international "REITs" are poor investment choices:

a) Low internal diversification, high fee

b) Not REITs at all, but "real estate related companies". Simply a sector play. Why not buy international, say, airplane related stocks if you seek complexity?

c) Exchange rate movements, which you get with any intl fund, reduce the portfolio level diversification benefit further still vs domestic REITs.

d) Inferior tax treatment vs real US REITs, which are taxed once, not twice.

Nick

pop77
Posts: 457
Joined: Wed Dec 12, 2007 7:57 pm

Post by pop77 » Sun Aug 30, 2009 11:49 am

markcoop wrote:For those who recommend int'l REITs, I'd be curious to hear how much they recommend?

I have 10% of my stock allocation in REITs. If I divide that among int'l REITs, then I'm left with 5%, or maybe something like a third (if I followed the rest of my domestic to int'l ratio). That works out to a very small percentage of my total portfolio - if I have 60/40 stock/bond ratio, then I would have as little as 2% in int'l REITs.
For me Real Estate is a separate asset class and and gets its own allocation. I allocate 10% of total portfolio to Real estate. Within that it is a 60/40 split between US and International.

Overall I have 60% Stocks, 15% Bonds, 10% Real Estate, 5% Commodities, 5% Infrastructure, 5% Cash

Post Reply