Author Topic: Rate of Return on Everything: A 150-year History  (Read 5693 times)

SeattleCPA

  • Handlebar Stache
  • *****
  • Posts: 1226
  • Age: 58
  • Location: Redmond, WA
    • Evergreen Small Business
Re: Rate of Return on Everything: A 150-year History
« Reply #50 on: February 08, 2018, 10:20:27 AM »
DNA2RNA my post was not specifically directed to you. I was merely trying to show that REITs are correlated to equities. From the OP:

The other thing, really interesting to anybody who tries to employ modern portfolio theory to their investing, is that it looks like housing and equities are not very correlated.

It has been suggested that REITs could take the place of bonds in a portfolio and I was trying to point out that this is not the case.

Sorry for lack of clarity or poor writing... but just so there's not confusion about what I meant to say, I never thought or intended to say that REITs are uncorrelated to something like US stocks or international stocks. (I think the probably are less correlated than other stuff because Portfolio Visualizers says so... but not that much.)

It's direct residential real estate that displays low correlation and has in most countries for decades. That's the statistic that piques interest if one wants to apply MPT to your portfolio.

Kalergie

  • Stubble
  • **
  • Posts: 203
  • Location: European expat living almost everywhere
Re: Rate of Return on Everything: A 150-year History
« Reply #51 on: February 14, 2018, 04:12:13 AM »
I'd like to pick this thread up as this is a really interesting topic for me. I personally have never invested in any type of real estate at all.

I believe real estate is not for everyone (not for me anyway). I don't like to be a landlord, I don't like to leverage at all, and I don't like the "hey landlord, this is your tenant. My toilette is broken, I don't care it's christmas eve! Come now and fix it or I withhold rent!"-kind of engagement that real estate usually entails.

However, I like to have a piece of the cake somehow. Currently my portfolio consists only of US and Int'l. index funds which do have REITs in them (about 3-4%). But the whole reason I hold the entire world in index funds is that I wanted to own the world. My understanding of this 150 year study is that real estate should be a heavy part of a world-portfolio to represent real weighing of wealth. Did I get this right?

Now how can I achieve this reasonably without holding physical real estate as commodity? Is being heavy on international REITs the answer?

Somehow isn't this the same as saying that you believe in Gold as a holding therefore buy mining companies? My gut feeling says this is the same fallacy.

What do you guys think?


 
« Last Edit: February 14, 2018, 04:14:35 AM by Kalergie »

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2201
  • Age: 38
Re: Rate of Return on Everything: A 150-year History
« Reply #52 on: February 14, 2018, 01:37:44 PM »
But the whole reason I hold the entire world in index funds is that I wanted to own the world. My understanding of this 150 year study is that real estate should be a heavy part of a world-portfolio to represent real weighing of wealth. Did I get this right?

If your goal is truly to own a portfolio of assets that proportionately mirrors the actual allocation of worldwide "wealth," then yes, real estate should be a significant fraction of your portfolio.  But query whether that is in fact your goal.  As you noted, if it were, then you would also need to diversify your portfolio into a multitude of other asset classes presumably underrepresented in your current portfolio, potentially including (depending on how you set the parameters for "wealth") not only gold and other precious metals but also livestock, artwork, timber, pork bellies and various other asset categories too numerous to count.

Occasionally, someone will make the argument that the logic underlying a passive investment approach in the selection of individual assets within a specific asset category should be extended to the selection of asset categories at higher levels of generality, but that logic does not necessarily apply in the context of the latter.  I personally employ a passive investment approach in the selection of individual stocks within a geographic market (via purchasing broad stock market index funds in lieu of individual stocks), but I employ an active investment approach in the selection of general asset classes and subclasses in which to invest in the first place.  Compared to the actual global allocation of investable assets, my portfolio's allocation deliberately overweights U.S. stocks relative to non-U.S. stocks, and deliberately overweights, by an extreme margin, stocks in general relative to all other asset classes.

Kalergie

  • Stubble
  • **
  • Posts: 203
  • Location: European expat living almost everywhere
Re: Rate of Return on Everything: A 150-year History
« Reply #53 on: February 17, 2018, 02:43:40 AM »
Thanks Brooklunguy.
Would you mind elaborating on this? How exactly do you do this?
I personally employ a passive investment approach in the selection of individual stocks within a geographic market (via purchasing broad stock market index funds in lieu of individual stocks), but I employ an active investment approach in the selection of general asset classes and subclasses in which to invest in the first place.  Compared to the actual global allocation of investable assets, my portfolio's allocation deliberately overweights U.S. stocks relative to non-U.S. stocks, and deliberately overweights, by an extreme margin, stocks in general relative to all other asset classes.

Kyle Schuant

  • Bristles
  • ***
  • Posts: 491
  • Location: Melbourne, Australia
Re: Rate of Return on Everything: A 150-year History
« Reply #54 on: February 17, 2018, 04:00:11 AM »
Well, people always need somewhere to live, so it makes sense. Just don't buy in a one-industry town, since if that industry leaves, the town dies.

Kalergie

  • Stubble
  • **
  • Posts: 203
  • Location: European expat living almost everywhere
Re: Rate of Return on Everything: A 150-year History
« Reply #55 on: February 17, 2018, 04:25:19 AM »
Well, people always need somewhere to live, so it makes sense. Just don't buy in a one-industry town, since if that industry leaves, the town dies.

Yes, but also people always have to eat. That does not make me want to own a grocery or a farm. I am pretty real estate/brick and mortar averse though!


Kyle Schuant

  • Bristles
  • ***
  • Posts: 491
  • Location: Melbourne, Australia
Re: Rate of Return on Everything: A 150-year History
« Reply #56 on: February 17, 2018, 01:43:22 PM »
The price of food is relatively volatile, because so much more is produced than is needed, and each food has many substitutes. The same is not generally true of housing. If I decide that the $30/kg steak is too expensive, I can just get the $5/kg chicken instead. Over enough iterations, this gives the beef seller incentive to lower the price, and the chicken seller incentive to raise it, and so they become 20 and 10 instead, and so on.


As well, the lower price of food gives it some volatility. It's no big deal for a loaf of bread to go from $2 to $1 literally overnight, or vice versa; but a house and land won't go from $1,000,000 to $500,000 overnight unless arson is involved, or vice versa unless zoning laws change, or something. Real estate's more stable than food.


This is less true in countries with no real banking and credit regulation, like the US. A lack of regulation of the goods and services in a market lead to price volatility; more regulation, though often onerous, leads to price stability. It might be driven higher, or lower by the regulations - but it's more stable.

DNA2RNA

  • 5 O'Clock Shadow
  • *
  • Posts: 7
Re: Rate of Return on Everything: A 150-year History
« Reply #57 on: February 19, 2018, 07:19:32 AM »
I'd like to pick this thread up as this is a really interesting topic for me. I personally have never invested in any type of real estate at all.

I believe real estate is not for everyone (not for me anyway). I don't like to be a landlord, I don't like to leverage at all, and I don't like the "hey landlord, this is your tenant. My toilette is broken, I don't care it's christmas eve! Come now and fix it or I withhold rent!"-kind of engagement that real estate usually entails.

However, I like to have a piece of the cake somehow. Currently my portfolio consists only of US and Int'l. index funds which do have REITs in them (about 3-4%). But the whole reason I hold the entire world in index funds is that I wanted to own the world. My understanding of this 150 year study is that real estate should be a heavy part of a world-portfolio to represent real weighing of wealth. Did I get this right?

Now how can I achieve this reasonably without holding physical real estate as commodity? Is being heavy on international REITs the answer?

Somehow isn't this the same as saying that you believe in Gold as a holding therefore buy mining companies? My gut feeling says this is the same fallacy.

What do you guys think?

Goldminers are to gold as homebuilders are to homes.  Homebuilders in the US are not REITs,  internationally it is more muddled as real estate companies take a more holistic view, so a company may own a bunch of residential and office towers and also rent out the bottom floors as retail, while developing their next tower.

Yes if you want a true worldwide "index" of income producing assets it would be roughly 60% real estate, 20% bonds, 20% equities/stocks.

Kalergie

  • Stubble
  • **
  • Posts: 203
  • Location: European expat living almost everywhere
Re: Rate of Return on Everything: A 150-year History
« Reply #58 on: February 20, 2018, 02:47:29 PM »
Goldminers are to gold as homebuilders are to homes.  Homebuilders in the US are not REITs
Excellent point made. I didn't think of it like that.

,internationally it is more muddled as real estate companies take a more holistic view, so a company may own a bunch of residential and office towers and also rent out the bottom floors as retail, while developing their next tower.

Yes if you want a true worldwide "index" of income producing assets it would be roughly 60% real estate, 20% bonds, 20% equities/stocks.
I think it is difficult to replicate the 60% real estate only with REITs though, am I right?   

brooklynguy

  • Handlebar Stache
  • *****
  • Posts: 2201
  • Age: 38
Re: Rate of Return on Everything: A 150-year History
« Reply #59 on: February 24, 2018, 02:15:10 PM »
Thanks Brooklunguy.
Would you mind elaborating on this? How exactly do you do this?
I personally employ a passive investment approach in the selection of individual stocks within a geographic market (via purchasing broad stock market index funds in lieu of individual stocks), but I employ an active investment approach in the selection of general asset classes and subclasses in which to invest in the first place.  Compared to the actual global allocation of investable assets, my portfolio's allocation deliberately overweights U.S. stocks relative to non-U.S. stocks, and deliberately overweights, by an extreme margin, stocks in general relative to all other asset classes.

I own one real estate rental unit (the second unit in the duplex in which I live) and otherwise invest essentially all of my money in stocks, with an 80/20 US/foreign allocation.

maizeman

  • Magnum Stache
  • ******
  • Posts: 2840
Re: Rate of Return on Everything: A 150-year History
« Reply #60 on: February 24, 2018, 04:33:56 PM »
So one important thing to note about volatility when people quote the finding that real estate has about the same return as equity with less volatility:

This paper is defining volatility as the standard deviation of average returns across different years. And by that metric, unleveraged real estate is, indeed, less volatile than stocks. However, while this metric makes sense for stocks because you can easily buy a diversified basket of stocks and be confident you're going to receive approximately the average return for a given year, there is not a great equivalent way to buy a diversified basket of real estate holdings across the entire country or world.*

So while the volatility of average real estate returns across all the different real estate markets in a whole country is relatively low from year to year, in practice it is also important to consider the volatility of real estate returns within a given year across different markets since most unleveraged real estate investors will be holding property in only one or a couple of markets. And I believe the variance across markets -- as opposed to across time -- tends to be quite large for real estate investments.

Anyway, it's a fascinating read, thanks for posting it SeattleCPA.

*As others have discussed, REITs are a very different animal, since they have significant additional management costs, employ a lot of leverage, and as a result have lots of exposure to interest rate risk that is not present in unleveraged real estate and that is at least mitigated for small scale investors buying houses using 30 year fixed rate mortgages.

SeattleCPA

  • Handlebar Stache
  • *****
  • Posts: 1226
  • Age: 58
  • Location: Redmond, WA
    • Evergreen Small Business
Re: Rate of Return on Everything: A 150-year History
« Reply #61 on: February 26, 2018, 06:53:02 AM »
At the start of this thread, @coopdog asked about applying modern portfolio theory thinking to the "Rate of Return on Everything" research results. I've finally had time to develop a first cut at how to do this, think about it, etc.

My thoughts appear here: Lessons from the Rate of Return on Everything paper.

But I think the general conclusion is housing represents a really useful component of one's portfolio. Maybe that means you own your own home, condo, apartment... or maybe it means you go beyond that.

Again, I would encourage anyone who's half interested to read the working paper. If you want just a teaser, you can also skim the first few paragraphs of the above blog post.

Prairie Stash

  • Handlebar Stache
  • *****
  • Posts: 1549
Re: Rate of Return on Everything: A 150-year History
« Reply #62 on: February 27, 2018, 07:29:44 AM »
Before getting excited, please note that property taxes aren't compared. You need to subtract property tax from your own calculations. According to the report, that's usually 1%. As an investment property vs. capital gains on stocks, the tax on selling also varies. The results are different for primary homes vs. secondary homes, as the authors point out.

"Taxes When calculating equity and housing returns, we do not account for taxes. From an
investor’s perspective accounting for taxes is clearly important...Additionally, housing tends to be subject to further asset-specific levies in the form of property taxes.
As a ballpark estimate, the impact of property taxes would lower the real estate returns by less
than one percentage point per year relative to equity


Prairie Stash

  • Handlebar Stache
  • *****
  • Posts: 1549
Re: Rate of Return on Everything: A 150-year History
« Reply #63 on: February 27, 2018, 07:57:33 AM »
Now if we exclude wartime, which is a major risk, the returns also change. In WW1 housing in Germany dropped 26%, as an average. In the USA, it increased 0.06% compared to equities 0.96%, the point being in War all the scenarios are irrelevant, it can't realistically be planned for. We see this in current Syria, housing and equity valuations are gone. I propose to ignore war effects as the burden is localized to places where damage happens, as the authors suggest.

Ignoring WW1 and WW2 - Stocks yield 9.2%, housing 6.14 (ignoring property tax), Bonds 3.54%.

Upon selling all the above are subject to taxes, on a primary residence its nothing, on the secondary, it's a lot more. The Authors point out that taxes are deductible in the states so the typical 1% tax is likely 0.67%, so housing returns drop to 5.47% on the primary residence. What would a typical gains bill look like on a $300,000 property? On long term holdings, the taxes on stocks/bonds are quite a bit lower and can be sold in smaller chunks to avoid the high tax brackets.

The authors very clearly point out that the returns on primary and secondary properties in the USA will vary precisely because of tax treatment. The results should be tailored to an individual use. Personal tax rates vary,

SeattleCPA

  • Handlebar Stache
  • *****
  • Posts: 1226
  • Age: 58
  • Location: Redmond, WA
    • Evergreen Small Business
Re: Rate of Return on Everything: A 150-year History
« Reply #64 on: February 28, 2018, 07:47:10 AM »
Now if we exclude wartime, which is a major risk, the returns also change. In WW1 housing in Germany dropped 26%, as an average. In the USA, it increased 0.06% compared to equities 0.96%, the point being in War all the scenarios are irrelevant, it can't realistically be planned for. We see this in current Syria, housing and equity valuations are gone. I propose to ignore war effects as the burden is localized to places where damage happens, as the authors suggest.

Ignoring WW1 and WW2 - Stocks yield 9.2%, housing 6.14 (ignoring property tax), Bonds 3.54%.

Upon selling all the above are subject to taxes, on a primary residence its nothing, on the secondary, it's a lot more. The Authors point out that taxes are deductible in the states so the typical 1% tax is likely 0.67%, so housing returns drop to 5.47% on the primary residence. What would a typical gains bill look like on a $300,000 property? On long term holdings, the taxes on stocks/bonds are quite a bit lower and can be sold in smaller chunks to avoid the high tax brackets.

The authors very clearly point out that the returns on primary and secondary properties in the USA will vary precisely because of tax treatment. The results should be tailored to an individual use. Personal tax rates vary,

Prairie, I'm not sure we disagree. Or that we strongly disagree. Lots of details to consider if one wants to de-emphasize equities and then allocate a chunk to housing.

The two things that make me think housing deserves serious consideration are the potentially lower standard deviations (minor factor) and then the low correlation (major factor).

And then this comment too: The thing that makes me want to dial down my allocation to equities (especially if I'm probably dialing down my portfolio's standard deviation too) is they appear so expensive .

A final clarifying remark so I don't unintentionally mislead some reader of this comment... for the reasons outline in my blog post, I don't think most people should go hog wild about the real estate. Rather, I'm saying that  equity-centric investors (like I've been for three decades) should probably take a look at the study and then another look at direct real estate investment.


Mr. Boh

  • Stubble
  • **
  • Posts: 101
  • Location: The Land of Pleasant Living
Re: Rate of Return on Everything: A 150-year History
« Reply #65 on: February 28, 2018, 08:53:38 AM »
At the start of this thread, @coopdog asked about applying modern portfolio theory thinking to the "Rate of Return on Everything" research results. I've finally had time to develop a first cut at how to do this, think about it, etc.

My thoughts appear here: Lessons from the Rate of Return on Everything paper.

But I think the general conclusion is housing represents a really useful component of one's portfolio. Maybe that means you own your own home, condo, apartment... or maybe it means you go beyond that.

Again, I would encourage anyone who's half interested to read the working paper. If you want just a teaser, you can also skim the first few paragraphs of the above blog post.

Excellent blog post. I agree with your conclusions.


Prairie Stash

  • Handlebar Stache
  • *****
  • Posts: 1549
Re: Rate of Return on Everything: A 150-year History
« Reply #66 on: February 28, 2018, 09:17:57 AM »
Now if we exclude wartime, which is a major risk, the returns also change. In WW1 housing in Germany dropped 26%, as an average. In the USA, it increased 0.06% compared to equities 0.96%, the point being in War all the scenarios are irrelevant, it can't realistically be planned for. We see this in current Syria, housing and equity valuations are gone. I propose to ignore war effects as the burden is localized to places where damage happens, as the authors suggest.

Ignoring WW1 and WW2 - Stocks yield 9.2%, housing 6.14 (ignoring property tax), Bonds 3.54%.

Upon selling all the above are subject to taxes, on a primary residence its nothing, on the secondary, it's a lot more. The Authors point out that taxes are deductible in the states so the typical 1% tax is likely 0.67%, so housing returns drop to 5.47% on the primary residence. What would a typical gains bill look like on a $300,000 property? On long term holdings, the taxes on stocks/bonds are quite a bit lower and can be sold in smaller chunks to avoid the high tax brackets.

The authors very clearly point out that the returns on primary and secondary properties in the USA will vary precisely because of tax treatment. The results should be tailored to an individual use. Personal tax rates vary,

Prairie, I'm not sure we disagree. Or that we strongly disagree. Lots of details to consider if one wants to de-emphasize equities and then allocate a chunk to housing.

The two things that make me think housing deserves serious consideration are the potentially lower standard deviations (minor factor) and then the low correlation (major factor).

And then this comment too: The thing that makes me want to dial down my allocation to equities (especially if I'm probably dialing down my portfolio's standard deviation too) is they appear so expensive .

A final clarifying remark so I don't unintentionally mislead some reader of this comment... for the reasons outline in my blog post, I don't think most people should go hog wild about the real estate. Rather, I'm saying that  equity-centric investors (like I've been for three decades) should probably take a look at the study and then another look at direct real estate investment.
I'm pretty sure we agree. There really wasn't much in my post beyond regurgitating the numbers.

What I found unsettling in the report is how easy it was to gloss over taxes. They mentioned property tax was omitted since it varies and its hard to address. They also mention capital gains, which is present on secondary properties but not primary property. So an individual will see two different returns, on neighbouring houses they own, based on residency status. I'm from Canada so I struggle a little with US tax rules, but I understand you could lose a good chunk of a house if you hold it for 20 years (a $100K house purchased in 2015 sold in 2035 could easily gain $400k if the numbers are accurate).

Lets make a case study using a  houses purchased in 2015. Assume property tax is 0.6% after deducting it (1% prior to tax deduction). Assume rate of growth on the house of 6.14%, non war growth. In 2035 if I sell, based on current gains tax, what would be my net after tax return? Assume a household income of $60k. Then compare purchasing bonds and selling tranches at 2 year intervals to avoid the gains tax and reinvesting. How much better will the house perform vs bonds after tax?

Feel free to adjust the scenario, lets make it generic for a broad swath of people. I'm looking to understand this better, what better way then run a case study? Perhaps I'm in the wrong and didn't understand the report, perhaps I'm right and we should investigate this farther.

When I run the scenario (for my own taxes, not applicable to US), using the 6.14% non wartime, I get an after tax return of 5.1%.

The biggest problem is that all the gains are paid in a single year, instead of bonds or stocks where you can spread out capital gains to hit lower tax brackets.

We're not disagreeing at all, we're helping each other understand this better. If anyone is going to do this, its important to understand it completely.

CorpRaider

  • Bristles
  • ***
  • Posts: 332
    • The Corpraider Blog
Re: Rate of Return on Everything: A 150-year History
« Reply #67 on: March 07, 2018, 09:31:38 AM »
Saw this paper.  Not sure how to reconcile with Shiller's work.  Guess this non Shiller paper includes more markets (several of which equities went to zero while housing at least muddled though...

http://www.nytimes.com/2013/04/14/business/why-home-prices-change-or-dont.html

maizeman

  • Magnum Stache
  • ******
  • Posts: 2840
Re: Rate of Return on Everything: A 150-year History
« Reply #68 on: March 07, 2018, 11:16:38 AM »
Saw this paper.  Not sure how to reconcile with Shiller's work.  Guess this non Shiller paper includes more markets (several of which equities went to zero while housing at least muddled though...

http://www.nytimes.com/2013/04/14/business/why-home-prices-change-or-dont.html

I think one big difference is that in the paper at the top of the thread they are looking at the total returns of housing (which would include rent), and the work being referenced in that nytimes article sounds like it is looking only at price appreciation of the house.

So the profit from real estate investments when looking at many markers over more than a century may come mostly from "dividends" (rent), rather than "capital gains" (price appreciation in excess of inflation).