The Global Multi-Asset Market Portfolio 1959-2012

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Barry Barnitz
Wiki Admin
Posts: 2977
Joined: Mon Feb 19, 2007 10:42 pm
Contact:

The Global Multi-Asset Market Portfolio 1959-2012

Post by Barry Barnitz » Mon Dec 16, 2013 11:08 am

Hi:

Total market investing enthusiasts might find this recent study of interest. The authors provide the global market portfolio over the extended 1959 - 2012 period.

Doeswijk, Ronald Q. and Lam, Trevin W. and Swinkels, Laurens A. P., The Global Multi-Asset Market Portfolio 1959-2012 (November 11, 2013). Financial Analysts Journal, Forthcoming. Available at SSRN: http://ssrn.com/abstract=2352932

Abstract:
The global multi-asset market portfolio contains important information for strategic asset-allocation purposes. First, it shows the relative value of all asset classes according to the global financial investment community, which one could interpret as a natural benchmark for financial investors. Second, this portfolio may also serve as the starting point for investors who use a framework in the spirit of Black and Litterman (1992), or for investors who follow adaptive asset-allocation policies as advocated by Sharpe (2010). We estimate the invested global market portfolio for the period 1990-2012 by estimating the market capitalization for the eight asset classes: equities, private equity, real estate, high-yield bonds, emerging-market debt, investment-grade credits, government bonds and inflation-linked bonds. For the main asset categories - equities, real estate, non-government bonds and government bonds - we extend the period to 1959-2012. We provide these annual historical estimates in tabular form so that practitioners and academics can easily use these historical data going forward. Next, we compare the asset allocations of institutional global investors to the market portfolio. To our knowledge, we are the first to document the global multi-asset market portfolio at these levels of detail for such a long period of time.


regards,
Image | blb | December Birthday Celebration: Ludwig van Beethoven

asset_chaos
Posts: 1210
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by asset_chaos » Mon Dec 16, 2013 2:31 pm

Will read with interest. Thanks for bringing this paper to our attention.
Regards, | | Guy

DetroitRed
Posts: 177
Joined: Tue May 17, 2011 4:16 pm

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by DetroitRed » Mon Dec 16, 2013 3:38 pm

Will try to remember to look at this when it gets published.

Very interested to see what their sources are for the global asset class returns back to 1959.

Global real estate return data back to 1959? Definitely not familiar with that source.

User avatar
nisiprius
Advisory Board
Posts: 34340
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: The Global Multi-Asset Market Portfolio 1959-2012

Post by nisiprius » Mon Dec 16, 2013 3:55 pm

DetroitRed, I was able to download the whole paper, weren't you? With regard to their real estate source, they say:
This study focuses on commercial real estate only. The commercial real-estate market is valued by using data from RREEF Real Estate Research, see Hobbs and Chin (2007).
That reference in turn is
Hobbs, P., and H. Chin. 2007. “The Future Size of the Global Real Estate Market.” RREEF Research Paper.
After looking at the RREEF website, http://www.rreef.com/home/index.jsp , and saying I was domiciled in the United States, I was able to find the paper in question, and it can be downloaded at

http://www.rreef.com/research/research_5529.jsp

It is not a published academic paper, it is a research report from a private wealth advisory firm.

That paper begins "It is extremely difficult to gain accurate figures on the size of the global real estate market due to the fragmented and confidential nature of the industry across many countries."

On skimming the paper, I am darned if I see where it presents any sort of time course for past global real estate values, either as a table or as a chart. The purpose of the paper is to "[estimate] how the market is likely to change over the coming five years" so there are various charts showing their predictions of the future, and it would be interesting to know how close the same.

Doeswik, Lam and Swinkels go on to say
We use the market capitalization of the GPR General PSI Global Index (GPRGLES) to backfill the period 1984-2005, as well as to fill the period 2007-2012. Here we use the 2006 estimate of USD 4.0 trillion as a starting point. Subsequently we use percentage changes in the market capitalization series to arrive at estimates for all other years.
I'm not too clear on what they are doing or what the "backfilling" and "filling" procedures are.
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.

DetroitRed
Posts: 177
Joined: Tue May 17, 2011 4:16 pm

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by DetroitRed » Mon Dec 16, 2013 6:21 pm

nisiprius wrote:
Doeswik, Lam and Swinkels go on to say
We use the market capitalization of the GPR General PSI Global Index (GPRGLES) to backfill the period 1984-2005, as well as to fill the period 2007-2012. Here we use the 2006 estimate of USD 4.0 trillion as a starting point. Subsequently we use percentage changes in the market capitalization series to arrive at estimates for all other years.
I'm not too clear on what they are doing or what the "backfilling" and "filling" procedures are.


I was able to download the paper.

I looked up the GPR General Index and it seems to be based on international publicly traded real estate stocks. Hardly representative of the world real estate market.

I don't have a lot of faith in their research data.

Did you look at their entry for the source of private equity data? It only goes back to 1990 yet they somehow pull the data back to 1959. Granted I doubt private equity was a very big sector back then, but as far as I can tell, they've taken a 1990 number and mysteriously projected it back to '59.

asset_chaos
Posts: 1210
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by asset_chaos » Mon Dec 16, 2013 10:58 pm

DetroitRed wrote:Did you look at their entry for the source of private equity data? It only goes back to 1990 yet they somehow pull the data back to 1959. Granted I doubt private equity was a very big sector back then, but as far as I can tell, they've taken a 1990 number and mysteriously projected it back to '59.

I may be misreading, but for the data they construct back to 1959 and use to make their figure 4, they say on p. 8 "Private equity is not included in this analysis". My reading is that their 1959-2012 data set does not include private equity in any of the years.
Regards, | | Guy

LaurensSwinkels
Posts: 2
Joined: Tue Dec 17, 2013 2:48 am

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by LaurensSwinkels » Tue Dec 17, 2013 6:56 am

Thank you for all the positive critical feedback on our paper, much appreciated!

Just to clarify:

1. I hope it is clear how we construct the series from the entire quote in the appendix of the paper. Since the GPRGLES index is much smaller than our estimate of the market, we plug in USD 4.0 trillion in 2006 and then use percentage change in market cap of the GPRGLES index to estimate the market cap of the global real estate sector:

We use the market capitalization of the GPR General PSI Global Index (GPRGLES) to backfill the
period 1984-2005, as well as to fill the period 2007-2012. Here we use the 2006 estimate of USD 4.0
trillion as a starting point. Subsequently we use percentage changes in the market capitalization series
to arrive at estimates for all other years.


2. We had tremendous difficulty finding finding a (set of) good series with accurate estimats of the global invested market capitalization of real estate. We welcome suggestions to improve our estimate, especially from those who have little faith in the estimates we use. We acknowledge in the paper that our historical estimate of real estate in the market portfolio might be downward biased, but with the lack of a good alternative it is hard to change that.

Footnote 9 It is possible that we underestimate the weight of real estate. We estimate the global weight of real estate at 2.2% in 1984,
while Ibbotson, Siegel and Love (1985) end up with an estimated weight of real estate in the US of 4.3% in 1984. While we
are aware that backfilling global data for real estate over the period 1959-1983 with US data might introduce a bias into our
data, the US market was then the largest real-estate market in the world. In addition, real estate is an asset class that has a
small weight in the market portfolio. So the impact of a bias would be limited for the total market portfolio.


We welcome any comments on improvements in our methodology and datasources.

DetroitRed
Posts: 177
Joined: Tue May 17, 2011 4:16 pm

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by DetroitRed » Tue Dec 17, 2013 11:37 am

Thanks for posting the clarifications, Laurens.

I don't have any good ideas for a historical real estate data source, but will let you know if something feasible comes to mind.

asset_chaos
Posts: 1210
Joined: Tue Feb 27, 2007 6:13 pm
Location: Melbourne

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by asset_chaos » Tue Dec 17, 2013 4:22 pm

Laurens,

Thanks for the paper and thanks for discussing it with us here. The thing I'd like to get a better understanding of is the size of the error estimates on the 1959 data series. For instance, in figure 4 for the percentage of the major asset classes, when you say government bonds are 31% of global investable assets one year, is the estimate 31 the midpoint of a range that could be 31.5 to 30.5 or is it 33 to 29 or 41 to 21? You've gone though some of this estimation above for real estate, which suggests a percentage point or so for the range on that asset class. Do you have similar estimates for the other classes? I haven't gone through your appendices in detail, so feel free to point me to the page where this is done if I've missed it. Thanks.
Regards, | | Guy

LaurensSwinkels
Posts: 2
Joined: Tue Dec 17, 2013 2:48 am

Re: The Global Multi-Asset Market Portfolio 1959-2012

Post by LaurensSwinkels » Thu Dec 19, 2013 11:25 am

Appendix III contains a robustness analyses. Unfortunately not exactly the one you are asking for, but you could at least use it to form your own opinion about the sensitivity to some choices.

Post Reply