shares vs property (not a forgone conclusion)

[petronius]

An interesting academic paper was discussed here recently (The Rate of Return on Everything, 1870–2015)

http://conference.nber.org/confer//2017 ... Taylor.pdf

The most surprising conclusion is that both in USA and UK, the total return of investments in residential property (housing) is similar to that of investments in shares, but with far less volatility.

at page 21 they write:

This country-level evidence reinforces one of our main findings: housing has been as good a long-run investment as equities, and possibly better. Housing has offered a similar return to equity in the majority of countries and time periods. In the long-run, housing outperformed equities in absolute terms in 6 countries, and equities outperformed housing in 2. Returns on the two assets were about the same (i.e. less than 1 percentage point difference) in the remaining 8 countries. After WW2, housing was the best-performing asset class in 5 countries, and equities in 9.

Now, I feel that there are some important methodological issues in the paper. Historical rents are reduced to account for repairs, but how this was done is not too clear. Also, if I am not wrong, they do not take into account that housing requires that the investor either spends some active time on it or employs a letting agent that will reduce yield considerably (they do have a "management cost" but this seems to refer to repairs rather than agents' fees).

Still. I was very surprised by these results. There is a conventional wisdom that shares are more risky but significantly more profitable than houses. These data raise some serious doubts on this. As the authors say at page 36:

Arguably the most surprising result of our study is that long run returns on housing and equity look remarkably similar. Yet while returns are comparable, residential real estate is less volatile on a national level opening up new and interesting risk premium puzzles

It is also very interesting that the correlation for 10 years total returns on housing and shares seem to be very low after World War II (figure 5).

This may have serious implications for portfolio allocation. I am going through the bonds part of the article now...

[Hariseldon58]

Thank you for the link, I had seen references to the findings in the press and was surprised that there was sufficient data to draw any meaningful conclusions and I will study with interest.

The lack of volatility in prices of housing at first glance may tend to confuse the prices of houses sold in a thin market to prices overall where there is no transactions at all in most houses.

I had until recently a small sideline in property development and indirectly I undertook on an estate agent like role in the local area. I probably only visited a 100 or so properties but it was very interesting, people's perceptions of the value of a property and the disconnect between the perception and the reality.... the well presented and priced property sells well, the property with real potential sells well, the great location sells well but the rest can really struggle and probably half the sellers give up altogether, this I think will distort the prices reported.

[Lootman]

Arguably the most surprising result of our study is that long run returns on housing and equity look remarkably similar. Yet while returns are comparable, residential real estate is less volatile on a national level opening up new and interesting risk premium puzzles

They may have similar results but they get there in different ways, and at different times, so a combination of both should give you a better risk/reward profile that investing in only one. If property holds up better in downturns, it may be because people can sell shares more readily than they cans ell their home.

As an example, I bought a large investment property in 1998. At the time property had been in the doldrums for quite a few years, but the stock market was booming because of the dot.com euphoria. The chap who sold the building to me was an estate agent who made a conscious decision to get out of property. He said he thought the returns would be better in the stock market. And they were for the following two years, before the crash that started in 2000. Houses carried on going up in value for another 7 years. I think I got the better deal out of that.

For as long as I can remember, I've invested both in shares and property. The former is easier - you can do it sitting in an armchair.The latter is more effort and less diversified.

Property is also much easier to make geared bets on, because it is typically based on borrowing. Aside from margin accounts and derivatives, you need capital to play the stock market.

In terms of tax, shares are taxed more benignly than property, especially with recent changes to the way BTLs are taxed. And shares can be sheltered in an ISA where property cannot. A primary home is very tax-efficient but not other properties.

I haven't done detailed calculations on which has given me better returns, and even if I had there are also taxes to consider, as well as the cost of my time managing property. But intuitively I'd say they have both done well and contributed about equally to whatever financial security I now enjoy.

[dspp]

In case you are interested this study (Long Term Returns ....) has been discussed nearby on TLF here

viewtopic.php?f=8&t=6972

and that discussion includes links to other places where you may find the discussion of interest. Personally I have been pondering the summary carefully, and remain unsure, but unsure if I am right to be unsure. Anyway, here it is:

Over the long run of nearly 150 years, we find that advanced economy risky assets have performed strongly. The average total real rate of return is approximately 7% per year for equities and 8% for housing. The average total real rate of return for safe assets has been much lower, 2.5% for bonds and 1% for bills. These average rates of return are strikingly consistent over different subsamples, and they hold true whether or not one calculates these averages using GDP-weighted portfolios. Housing returns exceed or match equity returns, but with considerably lower volatility—a challenge to the conventional wisdom of investing in equities for the long-run.

regards, dspp

[Alaric]

Housing returns exceed or match equity returns, but with considerably lower volatility—a challenge to the conventional wisdom of investing in equities for the long-run.[/i]

I would have thought housing returns only show lower volatility because houses and property are not continuously traded or at least not in the volumes that shares are traded in. You don't therefore have reliable short term values, so volatility isn't really measurable.

What is known that for sufficiently liquid stocks, a sizeable holding can be liquidated without undue impact on prices. That's not really the same for property.

[colin]

I agree with Alaric.
In fact I am quite puzzled that anyone should be surprised that house prices are less volatile than shares when liquid shares can be sold instantly, it would be a fairer comparison to use less liquid traded investments such as pibs and preference shares or private equity investments.

[petronius]

The surprise here is not that residential property is less volatile than shares, but that the total return over a large number of 10 year periods is similar for the two investments.

Usually it is assumed that lower volatility comes with lower returns. Hence, the unexpected result.

[Alaric]

Usually it is assumed that lower volatility comes with lower returns. Hence, the unexpected result.

I don't think there's enough trading to measure volatility.

It's like when advisers claimed that with profits funds were low risk on account of low volatility. The price stability is a function of how the market operates rather than anything more intrinsic.

Whether an asset is readily realisable is also a good test. You may have a share portfolio worth £ 200,000 that might only be worth £ 195,000. If you press the sale button, you get the quoted value minus only a limited sale commission and you get the value within a couple of days. Now compare that with trying to realise to cash the same value in property.

[petronius]

In UK house prices are monitored quite closely by several organizations. The Halifax house price index is a popular one. As with everything, there are methodological limitations, but I think the data are overall reliable. The index is capable of capturing quick, large movements in both directions, as shown for instance by the late eighties crash.

Arguably, historical rental values are more difficult to assess, as (as far as I know) there is no systematic record of rental contracts vs type of property. Even more difficult is to assess repair costs, insolvencies, etc.

[colin]

A house bought after world war two as an investment to be rented out may well be worth £250,000-£300,000 pounds today. Almost all of its value will be capital gain on which tax will need to be paid when the house is sold and the gain realized. An equal capital gain from a portfolio of shares can be realized with out triggering a tax liability by selling a measured portion of the portfolio in different tax years.
Of course if one buys a house just to live in then that's another matter but who really wants to sell their home?

[UncleEbenezer]

Bear in mind the politics. UK governments have for generations poured ever-increasing taxpayer billions into pushing up property prices, through measures to subsidise both sales and rent to big chunks of the population - and then ever more extraordinary measures to prevent a correction in 2004/5 and then limit the one in 2008/9. When, for instance, housing benefit accounts for 40% of the rental market, it gets completely decoupled from the behaviour of a market economy. And of course house prices are driven by supply and demand not primarily of property but of credit (and of course sentiment).

A generation's worth of political memory tells us there are far fewer lost votes in a stockmarket crash than a housing correction. How long that will hold is anyone's guess.

[dspp]

The study is not just a UK study. We should not just interpret it through a UK lense. Unless the UK lense is representative of the many other datasets in the study - are they ?

regards, dspp

[colin]

I think the study focused on the UK and the USA, in the states banks have been much more inclined to foreclose on unpaid mortgages and repossess peoples homes, we were never going to see the scale of evictions here that occurred in 2008/9 in the States.