Bill Sharpe's preferred portfolio

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Quark
Posts: 1045
Joined: Sun Nov 01, 2015 4:32 pm

Bill Sharpe's preferred portfolio

Post by Quark »

Global stocks and bonds, cap weighted, free float, plus TIPS to taste. See https://www.youtube.com/watch?v=XsXOLZ9U7jI at about 39:00.

He uses Vanguard funds.
User avatar
saltycaper
Posts: 2650
Joined: Thu Apr 24, 2014 8:47 pm
Location: The Tower

Re: Bill Sharpe's preferred portfolio

Post by saltycaper »

Had to laugh when he said he wasn't sure how he felt about the currency hedging in the int'l bond fund.

Also, the bonsai really stole the show.
Quod vitae sectabor iter?
pkcrafter
Posts: 15461
Joined: Sun Mar 04, 2007 11:19 am
Location: CA
Contact:

Re: Bill Sharpe's preferred portfolio

Post by pkcrafter »

Thanks, Quark, good interview.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.
User avatar
billyv
Posts: 57
Joined: Fri Aug 05, 2016 10:25 am

Re: Bill Sharpe's preferred portfolio

Post by billyv »

Thanks for the link. Sounds like Mr. Sharpe would be happy with one of Vanguard's target-date or LifeStrategy funds as the main course, with a helping of TIPS on the side.
pascalwager
Posts: 2312
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

billyv wrote:Thanks for the link. Sounds like Mr. Sharpe would be happy with one of Vanguard's target-date or LifeStrategy funds as the main course, with a helping of TIPS on the side.
He wants VG to create a fund-of-funds using the four US/int'l total market stock/bond funds.

Sharpe personally used CRSP and FTSE market cap data for the stock funds and CITI market cap data for the two bond funds using a small portion of his own money to create such a fund-of-funds.

But some might decide, as you suggested, to use, say, the 40/60 VG LS fund as an approximation. Or even better, in this case, the VG TWS fund (plus the two VG bond funds) which is already at stock market cap proportions.
VT 60% / VFSUX 20% / TIPS 20%
User avatar
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Bill Sharpe's preferred portfolio

Post by Kevin M »

Thanks for the link. That led me to a video of the same guy interviewing Fama: https://youtu.be/dj-RO4mh-wA.

Fama basically says most people should own the market portfolio, which is always efficient, but some may choose to tilt to small and/or value based on "taste".

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Bill Sharpe's preferred portfolio

Post by Kevin M »

pascalwager wrote:
billyv wrote:Thanks for the link. Sounds like Mr. Sharpe would be happy with one of Vanguard's target-date or LifeStrategy funds as the main course, with a helping of TIPS on the side.
He wants VG to create a fund-of-funds using the four US/int'l total market stock/bond funds.

Sharpe personally used CRSP and FTSE market cap data for the stock funds and CITI market cap data for the two bond funds using a small portion of his own money to create such a fund-of-funds.

But some might decide, as you suggested, to use, say, the 40/60 VG LS fund as an approximation. Or even better, in this case, the VG TWS fund (plus the two VG bond funds) which is already at stock market cap proportions.
Yes, the Vanguard Target Retirement (TR) and LifeStrategy fund families are based on the four funds Sharpe seems to be recommending, but the stock/bond and US-bond/international-bond ratios are not based on market weights. So apparently the latter is the missing piece that Sharpe has been working on Vanguard to provide in a fund.

The lower-dated TR funds even have TIPS (short term), with the Target Retirement Income fund having about 17% of portfolio in TIPS.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
User avatar
badbreath
Posts: 987
Joined: Mon Jul 18, 2016 7:50 pm

Re: Bill Sharpe's preferred portfolio

Post by badbreath »

I liked it but more reason to "stay the course"
“While money can’t buy happiness, it certainly lets you choose your own form of misery.” Groucho Marx
rj49
Posts: 629
Joined: Tue Feb 22, 2011 11:22 pm

Re: Bill Sharpe's preferred portfolio

Post by rj49 »

Apart from his investments, Sharpe had interesting ideas years ago on how to allocate retirement funds, in 'lockboxes', a combination of a total market risk portfolio combined with a risk-free asset to be spent in each year of retirement. He said it's the optimal way to draw down funds and maximize growth and spending value, while protecting against cognitive decline that could result in unwise investment decisions later in life. If you google his academic page, or his name and 'lockbox', he has some interesting papers and presentations on the theory, although he was light on specifics on how to set up such investments. The closest I could think of would be to set up a ladder of Target Retirement funds in 5-year buckets, reaching out to age 90, and draw down each bucket over 5 years, before moving on to the next one, which would have been drawing down its stock allocation at the same time.
pascalwager
Posts: 2312
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

The last I heard, the world stock/bond ratio was about 40/60. This would constitute Sharpe's risky portfolio. Alongside would be his riskless portfolio, TIPS.

He recommends this mainly for older investors. Younger investors might want more stocks and less TIPS.
VT 60% / VFSUX 20% / TIPS 20%
User avatar
CyberBob
Posts: 3387
Joined: Tue Feb 20, 2007 1:53 pm

Re: Bill Sharpe's preferred portfolio

Post by CyberBob »

Sharpe's "World Bond/Stock Index" that he is wishing for actually exists. It's the MSCI Global Capital Markets Index. There were even rumors about ten years ago that someone might start a fund based on the index. Still waiting. Unfortunately, MSCI took the index in-house in 2006 and so the breakdown is no longer publicly available. The last available breakdown from 2006 is below.


Image
User avatar
CyberBob
Posts: 3387
Joined: Tue Feb 20, 2007 1:53 pm

Re: Bill Sharpe's preferred portfolio

Post by CyberBob »

I looked up the indexes mentioned by Sharpe in the video.

CRSP US Total Market Index
Market cap: $24,541,947 million

FTSE Global All-Cap minus US
Market cap: $21,356,625 million

Note: The FTSE Global All-Cap index (the index followed by Vanguard's Total World Stock Index fund) actually shows the global total, as well as breaking out the number for the US. So, one could use that index for both US and ex-US numbers. I'm assuming Sharpe mentioned the CRSP index for the US number since that's the index that Vanguard actually tracks. Either way, the numbers are similar.

Citigroup US Broad Investment-Grade Index
Market cap: $18,348,310 million

Citigroup World Broad Investment-Grade minus US
Market cap: $16,993,990 million

Note: For the bond indexes, I used the market-value numbers rather than the par-value numbers.

Total Market-Cap
$81,240,872 million

So, that would give you a four-fund Vanguard portfolio of:
30% Total US Stock
26% Total International (ex-US) Stock
23% Total US Bond
21% Total International (ex-US) Bond

Sharpe only mentioned the four big Vanguard Total funds. But to tweak it even further, I noticed that the Vanguard Total International Bond fund doesn't actually include all of the emerging markets bonds. So, adding that in:

Citigroup Emerging Markets Broad Bond
Market cap: $1,690,300 million

Gives you a five-fund Vanguard Portfolio of:
30% Total US Stock
26% Total International (ex-US) Stock
22% Total US Bond
20% Total International (ex-US) Bond
2% Emerging Markets Government Bond

Not too much difference. Probably just easier to stick with the four big 'Total' funds.
TBillT
Posts: 1001
Joined: Sat Sep 17, 2011 1:43 pm

Re: Bill Sharpe's preferred portfolio

Post by TBillT »

Interesting I have half a mind to try his approach.
Would this approach be appropriate inside an tIRA vs. rIRA?
Does it make any difference?
Fclevz
Posts: 651
Joined: Fri Mar 30, 2007 11:28 am

Re: Bill Sharpe's preferred portfolio

Post by Fclevz »

Phil DeMuth had an article on Forbes.com where he also advocated for the global market portfolio. However, his breakdown looks different from the market cap numbers of the indexes suggested by Sharpe and presented by CyberBob above.
User avatar
CyberBob
Posts: 3387
Joined: Tue Feb 20, 2007 1:53 pm

Re: Bill Sharpe's preferred portfolio

Post by CyberBob »

CyberBob wrote:Note: For the bond indexes, I used the market-value numbers rather than the par-value numbers.
Hmmm, thinking about it, I wonder if Par Value wouldn't be a better number to use. Especially since the bonds aren't going to be sold now at their market values, but rather held long-term until maturity and redeemed at par.

In that case, using Par Values for the bond indexes, it changes things a bit.

Using Market Value:
30% Total US Stock
26% Total International (ex-US) Stock
23% Total US Bond
21% Total International (ex-US) Bond


Using Par Value:
31% Total US Stock
27% Total International (ex-US) Stock
23% Total US Bond
19% Total International (ex-US) Bond
User avatar
Kevin M
Posts: 15750
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Bill Sharpe's preferred portfolio

Post by Kevin M »

CyberBob wrote:
CyberBob wrote:Note: For the bond indexes, I used the market-value numbers rather than the par-value numbers.
Hmmm, thinking about it, I wonder if Par Value wouldn't be a better number to use. Especially since the bonds aren't going to be sold now at their market values, but rather held long-term until maturity and redeemed at par.
I think using market value makes more sense. The reason market value is higher than par value is because of the present value of the higher coupon payment's you'll receive than if they bonds were at par value. That's real value you'll receive, in addition to the par received at maturity, so I don't know why you'd ignore that.

Capital flows are influenced by market values, not par values, and the Sharpe idea is related to capital flows adjusting market values to some sort of equilibrium state.

Kevin
If I make a calculation error, #Cruncher probably will let me know.
lazyday
Posts: 3849
Joined: Wed Mar 14, 2007 10:27 pm

Re: Bill Sharpe's preferred portfolio

Post by lazyday »

CyberBob wrote:26% Total International (ex-US) Stock
21% Total International (ex-US) Bond
If you live in the US, doesn't this give you more than the ideal amount of currency risk? Won't that hurt your portfolio's Sharpe ratio?
Dirghatamas
Posts: 581
Joined: Fri Jan 01, 2016 5:18 pm

Re: Bill Sharpe's preferred portfolio

Post by Dirghatamas »

CyberBob, good work but I have concerns on the methodology.
For stocks, I think this is reasonable. The index providers use free float part of public stocks. FTSE index shows ~54% in US and 46% in rest of the world so your numbers look sort of reasonable.

My main concern is that the size of bond market vs. stock market seems way off. My understanding is that the total bond market is MUCH larger than stock market. As such, in your total portfolio, majority should be bonds but your index shows ~50/50 or even slightly heavier in stocks.

As an example of source data, here is the US total bond market as of 4/01/2017 from probably the best source: Security Industry and Fiinancial Markets Association (SIFMA). According to their latest data
http://www.sifma.org/research/statistics.aspx
Total US bonds as of this month were at 39.36 trillion dollars. Similarly, my understanding (though I can't find the source right now) is that the foreign bond market is much larger than your post shows.

What is the reason for discrepancy between your and SIFMA source data ?
User avatar
CyberBob
Posts: 3387
Joined: Tue Feb 20, 2007 1:53 pm

Re: Bill Sharpe's preferred portfolio

Post by CyberBob »

Dirghatamas wrote:What is the reason for discrepancy between your and SIFMA source data ?
Sharpe talks in the video about the indexes that the Vanguard funds follow. Essentially, they simply aren't quite as total as the SIFMA numbers.
For example, the funds don't include municipal bonds or money markets. SIFMA does, and so looks a lot bigger.

It also appears that SIFMA is counting all treasuries. The free-float indexes don't count treasuries held by the Fed or the ECB, and so are multiple trillions less in that department.

There's a bit of an allowance on the stock side too, as those indexes/funds are free-float as well, and also don't include micro-caps and frontier markets.
afan
Posts: 8169
Joined: Sun Jul 25, 2010 4:01 pm

Re: Bill Sharpe's preferred portfolio

Post by afan »

In addition to the government bonds held by central banks, it is not clear how to deal with another huge batch of government bonds: those held by commercial banks because their regulators force them to do so. In the US and all over Europe part of the reaction to the financial crisis has been to mandate more capital and the best and sometimes the only acceptable evidence of this is government bonds. Why would a bank buy government bonds at negative interest rates? Because they do not have a choice.

But how does an investor who does have a choice translate that to the market price of the bonds? Part of what the bank is getting is the ability to remain in business. For an individual who is not forced to buy government bonds, are they really efficiently priced? Or are they too expensive for people who do not have guns to their heads? Whatever your answer, how do you translate that to the market cap of bonds that is relevant to an individual?

In traditional theory, one takes the entire market portfolio and adjusts risk by varying the allocation to risk free investments. But if the market portfolio is too high in bonds, even 100% market might be too low risk. Theory assumes one can borrow at the risk free rate. In reality individuals cannot. That plus the mechanics of buying on margin lead to adjusting portfolio risk by varying the stock to bond ratio, not just the risky to risk free ratio.

And how do you take into account investable assets besides stocks and bonds? REITs just reproduce what you already have in stocks. How do you get at real estate not available as REITs?
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama
bjr89
Posts: 151
Joined: Sat Apr 23, 2016 8:15 am

Re: Bill Sharpe's preferred portfolio

Post by bjr89 »

-----
Last edited by bjr89 on Fri Nov 16, 2018 6:54 pm, edited 1 time in total.
harvestbook
Posts: 871
Joined: Sat Mar 18, 2017 7:12 pm

Re: Bill Sharpe's preferred portfolio

Post by harvestbook »

Thanks for the link. I wasn't clear if Sharpe believes currency-hedging the international bonds was better or not. In my mind, the currency risk is another form of diversification rather than merely a risk (after all, it's not like the dollar doesn't carry risk). But I could see where it would seem more volatile.

Also interesting that he put "some" of his money into his idea. I'm always amazed how few of the experts don't eat their own cooking, although I imagine his overall holdings are somewhat similar to this. Paul Merriman is the same way, as well as the guy who runs his portfolio in Motif.

Here's Charlie Billelo's view on market cap weighting, drilling into the more obscure assets (but skipping that all important Beanie Baby-baseball card-Hummel figurine base):

http://pensionpartners.com/searching-fo ... portfolio/

Image
I'm not smart enough to know, and I can't afford to guess.
bjr89
Posts: 151
Joined: Sat Apr 23, 2016 8:15 am

Re: Bill Sharpe's preferred portfolio

Post by bjr89 »

-----
Last edited by bjr89 on Fri Nov 16, 2018 6:40 pm, edited 1 time in total.
zeugmite
Posts: 1176
Joined: Tue Jul 22, 2008 11:48 pm

Re: Bill Sharpe's preferred portfolio

Post by zeugmite »

CyberBob wrote:Sharpe only mentioned the four big Vanguard Total funds. But to tweak it even further, I noticed that the Vanguard Total International Bond fund doesn't actually include all of the emerging markets bonds. So, adding that in:

Citigroup Emerging Markets Broad Bond
Market cap: $1,690,300 million

Gives you a five-fund Vanguard Portfolio of:
30% Total US Stock
26% Total International (ex-US) Stock
22% Total US Bond
20% Total International (ex-US) Bond
2% Emerging Markets Government Bond

Not too much difference. Probably just easier to stick with the four big 'Total' funds.
It seems these indexes are just under-representing emerging market bonds. Citi themselves claims they've just recently started to add some of those countries. I am highly suspicious of the "coverage" of these market caps.
User avatar
aj76er
Posts: 1178
Joined: Tue Dec 01, 2015 10:34 pm
Location: Austin, TX

Re: Bill Sharpe's preferred portfolio

Post by aj76er »

Why would an individual want to hold their bond percentage equal to the global market weight? Risk it n a portfolio should be based on one's ability, willingness, and need to take risk (typically proportional to one's remaining working capital).

I could get behind global weighting of the stock component and the bond components individually; but the overall stock/bond ratio needs to be set according to individual circumstances.
"Buy-and-hold, long-term, all-market-index strategies, implemented at rock-bottom cost, are the surest of all routes to the accumulation of wealth" - John C. Bogle
lazyday
Posts: 3849
Joined: Wed Mar 14, 2007 10:27 pm

Re: Bill Sharpe's preferred portfolio

Post by lazyday »

aj76er wrote:Why would an individual want to hold their bond percentage equal to the global market weight?
I don't see any reason. All investors in total hold the market, but unless all investors are alike I don't see why an individual should market weight. I'm not like a pension, insurance company, foundation, hedge fund, state, or many individual investors.
I could get behind global weighting of the stock component and the bond components individually
Even here I'm not convinced. I might choose to market weight stocks, but then I'll want to use bonds of my home currency to avoid excessive currency risk. Some argue for overweighting stocks of the home currency.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Bill Sharpe's preferred portfolio

Post by NiceUnparticularMan »

aj76er wrote:I could get behind global weighting of the stock component and the bond components individually; but the overall stock/bond ratio needs to be set according to individual circumstances.
If you few bonds as risk-management tools, it doesn't make much sense to me to global weight your bond mix. That's just reflecting a mish-mash of what a whole bunch of different investors have done to manage their individual risks.
bjr89
Posts: 151
Joined: Sat Apr 23, 2016 8:15 am

Re: Bill Sharpe's preferred portfolio

Post by bjr89 »

-----
Last edited by bjr89 on Fri Nov 16, 2018 6:37 pm, edited 1 time in total.
Sconie
Posts: 970
Joined: Sun Feb 07, 2010 9:23 am
Location: Arizona

Re: Bill Sharpe's preferred portfolio

Post by Sconie »

Thanks OP!
I know you think you understand what you thought I said but I'm not sure you realize that what you heard is not what I meant. - Alan Greenspan
afan
Posts: 8169
Joined: Sun Jul 25, 2010 4:01 pm

Re: Bill Sharpe's preferred portfolio

Post by afan »

bjr,

I think the picking and choosing bonds follows from trying to customize risk. Just as many people cannot efficiently use leverage to get their desired risk level and so use a reduced bond allocation, they may also adjust their bonds away from the market.

Since the market is an average of the needs of all investors, dominated by huge institutional firm's, their optimal mix.could be different from an individual. Many banks are forced by their regulators to buy large amounts of low yielding government bonds. Since these banks have no choice, they can be made to buy negative interest bonds. It is hard to see why anyone would.do this if they had a choice,. Nonetheless, this huge volume of low or negative yielding bonds is part of the market. That does not mean an investor free of these contraints should want any of these bonds.

Similarly, many individuals are better off with muni bonds, which have no role in the portfolios of huge pension funds. Individuals and mutual funds sold to them are a segment of the market. Not clear who really should have a market weight of the total bond market. That may be why even Vanguard does not include munis in it's TBM.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama
bjr89
Posts: 151
Joined: Sat Apr 23, 2016 8:15 am

Re: Bill Sharpe's preferred portfolio

Post by bjr89 »

-----
Last edited by bjr89 on Fri Nov 16, 2018 6:36 pm, edited 1 time in total.
User avatar
aj76er
Posts: 1178
Joined: Tue Dec 01, 2015 10:34 pm
Location: Austin, TX

Re: Bill Sharpe's preferred portfolio

Post by aj76er »

By constantly adjusting your holdings based on market cap for US Stocks and exUS Stocks, it seems like you would always be buying and selling at the wrong time. In essence, you'd be chasing where capital is being allocated in the world. How would rebalancing in this way allow you to mechanically buy low and sell high, like holding fixed percentages does?

Taking a different spin on this, imagine doing the same thing with a subset of the US market. Take the market weighting of your favorite SCV index, and constantly change the percent of your portfolio based on it's market cap related to the S&P500. When SCV is doing well and popular, you'd hold more of it; when it was doing poorly and unpopular, you'd hold less. Seems completely backwards from what you should be doing.

I think fixed percentages of asset classes based on individual tastes and circumstances are the way to go.

I'd be interested in the back-testing data that either confirms or denies this
"Buy-and-hold, long-term, all-market-index strategies, implemented at rock-bottom cost, are the surest of all routes to the accumulation of wealth" - John C. Bogle
User avatar
aj76er
Posts: 1178
Joined: Tue Dec 01, 2015 10:34 pm
Location: Austin, TX

Re: Bill Sharpe's preferred portfolio

Post by aj76er »

bjr89 wrote:I'm just saying that same logic should probably then be applied to stocks, but often times it is not. That is, if the cap weighted bond market doesn't reflect the needs of an individual's circumstance, why should the cap weighted stock market... Why stop with bonds
Because when you are buying a cap weighted Market fund, you are creating an asset class. For example the US Stock Market is an asset class that tracks that us local economy. Developed foreign and emerging markets are also distinct asset classes. By holding all of these asset classes you are attempting to construct a higher risk-adjusted portfolio through the use of diversification. Using market capitalization to track the economies of specific regions makes perfect sense to me.

A total world market cap weighted fund creates another asset class that tracks global beta. It will likely not do as well as splitting up different geographical regions and two separate asset classes, but it's a lot simpler and many people would say it's good enough.
"Buy-and-hold, long-term, all-market-index strategies, implemented at rock-bottom cost, are the surest of all routes to the accumulation of wealth" - John C. Bogle
zeugmite
Posts: 1176
Joined: Tue Jul 22, 2008 11:48 pm

Re: Bill Sharpe's preferred portfolio

Post by zeugmite »

aj76er wrote:Why would an individual want to hold their bond percentage equal to the global market weight? Risk it n a portfolio should be based on one's ability, willingness, and need to take risk (typically proportional to one's remaining working capital).

I could get behind global weighting of the stock component and the bond components individually; but the overall stock/bond ratio needs to be set according to individual circumstances.
He uses TIPS as cash. Bonds and stocks are all risk instruments.
zeugmite
Posts: 1176
Joined: Tue Jul 22, 2008 11:48 pm

Re: Bill Sharpe's preferred portfolio

Post by zeugmite »

Sharpe is offering an overly academic solution. Really there are many more ways to construct efficient portfolios than to hold market proportions. Restricting to cash + market portfolio is only optimal in some equilibrium that never exists -- and at least in the original construction, if you don't care about returns distribution beyond second-order statistics.
User avatar
backpacker
Posts: 1620
Joined: Mon Sep 22, 2014 2:17 pm

Re: Bill Sharpe's preferred portfolio

Post by backpacker »

bjr89 wrote:I'm just saying that same logic should probably then be applied to stocks, but often times it is not. That is, if the cap weighted bond market doesn't reflect the needs of an individual's circumstance, why should the cap weighted stock market... Why stop with bonds?
Different investors clearly have liabilities with different terms. Someone buying a house in five years needs the money at a different time than someone buying a house in two weeks. So there is clearly differences when it comes to our ability to take on term risk.

I wrote up a post about this a few days back, but identifying similar pairs of investors for stock factors is much harder. What would my life have to be like for me to be someone who should heavily tilt towards large growth stocks? I have no idea.
bjr89
Posts: 151
Joined: Sat Apr 23, 2016 8:15 am

Re: Bill Sharpe's preferred portfolio

Post by bjr89 »

-----
Last edited by bjr89 on Fri Nov 16, 2018 6:35 pm, edited 1 time in total.
Dirghatamas
Posts: 581
Joined: Fri Jan 01, 2016 5:18 pm

Re: Bill Sharpe's preferred portfolio

Post by Dirghatamas »

bjr89 wrote:I'm just saying that same logic should probably then be applied to stocks, but often times it is not. That is, if the cap weighted bond market doesn't reflect the needs of an individual's circumstance, why should the cap weighted stock market... Why stop with bonds
I would disagree with this and give some reasons. For the record, I have done this (market cap global stocks 100% with 0% bonds) for 25 years so have had some time to think about it.

Stocks and bonds are truly different markets and shouldn't be clubbed together. Bonds in particular are weird because there are lots of laws which force their usage. First, there are big whales like central banks which can buy bonds (usually their own country and only modestly of other countries) which can distort market cap. Further large insurance companies have to buy bonds of certain durations for liability matching. Large sovereign funds, endowments, non profits, pension funds often have laws about what they can invest in. Often they are told by law to hold at least X% of a country's bonds regardless of whether they want to or not.

Consider the Norwegian sovereign fund which is almost a trillion dollars. Its was explicitly allowed to hold < a certain % in stocks and pushed to increase it recently. Similarly Japan Post pension, the largest pension fund in the world was till recently forced to only invest in Japanese bonds. More recently, under Abe, it was allowed to invest (a certain %) in foreign bonds and even more recently a certain % in stocks.

All this creates a lot of inefficiencies. I see absolutely no reason why a retail investor should try to mimic this market capitalization. It makes no sense whatsoever because it is NOT an efficient market. Whales hold what they hold because they are told to..while a retail investor has no such restriction.

Stocks are very different. First the main non mobile part of stocks are insiders like founders, states, Govts etc. These are already removed when indexes are created (free float capitalization). What is left is freely investable (with some minor exceptions like China A shares). As far as I know, whales have no restrictions once they are allowed to own stocks as to which stocks they should own. This creates a much more efficient market place compared to bonds as capital can flow freely.

As such, I think the world capitalization argument for stocks makes a hell of a lot more sense than for bonds.
Dirghatamas
Posts: 581
Joined: Fri Jan 01, 2016 5:18 pm

Re: Bill Sharpe's preferred portfolio

Post by Dirghatamas »

aj76er wrote:By constantly adjusting your holdings based on market cap for US Stocks and exUS Stocks, it seems like you would always be buying and selling at the wrong time. In essence, you'd be chasing where capital is being allocated in the world. How would rebalancing in this way allow you to mechanically buy low and sell high, like holding fixed percentages does?

I'd be interested in the back-testing data that either confirms or denies this
With due respect, I don't think you understand the mechanics of how this works. I have done this precisely (hold 100% stocks, global cap weighted with no bonds) for 25 years so can provide some info on the mechanics.

You are NEVER adjusting things like you say. That is in fact the beauty of the global market cap portfolio. The idea is to own a fixed constant % of every public company in the world, regardless of country. In this philosophy "country" doesn't mean anything and isn't a factor. You DON'T need to do this using just one fund (which didn't exist till recently). You can certainly do it using two funds but require no extensive rebalancing. If on day 1 you buy US total stock market and ex US total stock market in the right ratio, then you are set. As the markets move up and down, your holdings naturally move to be cap weighted without you doing anything.

Now for full info, there are second and third order effects like taxes, share buybacks and dividends which are different between US and ex US. As such once a year or once every few years, there is a very minor adjustment needed. That is NOT a first order effect.

So, holding world stock in market cap weighted certainly is a passive strategy that doesn't require much of rebalancing or attention to the portfolio.
bjr89
Posts: 151
Joined: Sat Apr 23, 2016 8:15 am

Re: Bill Sharpe's preferred portfolio

Post by bjr89 »

-----
Last edited by bjr89 on Fri Nov 16, 2018 6:35 pm, edited 1 time in total.
psteinx
Posts: 5785
Joined: Tue Mar 13, 2007 2:24 pm

Re: Bill Sharpe's preferred portfolio

Post by psteinx »

I would think that hedge funds by and large are NOT setting the worldwide prices on bonds, nor evening out inefficiencies between different classes of bonds. It's a bit of a guess, but:

1) I doubt hedge fund scale is all that large relative to the size of the bond market
2) Yes, hedge funds can leverage, to achieve larger effective scale than their net assets, but there are limits to leverage
3) In order for hedge funds to really "fix" pricing anomalies, they would basically need to be able to go short, cheaply and easily, as well as long. While I understand that some bond shorting is possible, both directly and via derivatives, I suspect that once you move outside of the biggest issues (various government bonds and other major issuers), that shorting quickly becomes relatively difficult to do - costly, inefficient, etc.
User avatar
aj76er
Posts: 1178
Joined: Tue Dec 01, 2015 10:34 pm
Location: Austin, TX

Re: Bill Sharpe's preferred portfolio

Post by aj76er »

Dirghatamas wrote:
aj76er wrote:By constantly adjusting your holdings based on market cap for US Stocks and exUS Stocks, it seems like you would always be buying and selling at the wrong time. In essence, you'd be chasing where capital is being allocated in the world. How would rebalancing in this way allow you to mechanically buy low and sell high, like holding fixed percentages does?

I'd be interested in the back-testing data that either confirms or denies this
With due respect, I don't think you understand the mechanics of how this works. I have done this precisely (hold 100% stocks, global cap weighted with no bonds) for 25 years so can provide some info on the mechanics.

You are NEVER adjusting things like you say. That is in fact the beauty of the global market cap portfolio. The idea is to own a fixed constant % of every public company in the world, regardless of country. In this philosophy "country" doesn't mean anything and isn't a factor. You DON'T need to do this using just one fund (which didn't exist till recently). You can certainly do it using two funds but require no extensive rebalancing. If on day 1 you buy US total stock market and ex US total stock market in the right ratio, then you are set. As the markets move up and down, your holdings naturally move to be cap weighted without you doing anything.

Now for full info, there are second and third order effects like taxes, share buybacks and dividends which are different between US and ex US. As such once a year or once every few years, there is a very minor adjustment needed. That is NOT a first order effect.

So, holding world stock in market cap weighted certainly is a passive strategy that doesn't require much of rebalancing or attention to the portfolio.
Thanks for the info. I overestimated the effects of share-creation, destruction, etc...
So, based on what you are saying, the "growth" factor dominates and the stock portfolio becomes somewhat self-driven. Not a bad way to manage the split between U.S. Market + ex-U.S. market. I think that I would always want these as separate funds(rather than something like VT) because I can minimize tax complexity (and to a certain point, tax efficiency) by keeping much of the ex-U.S. market in tax-sheltered accounts.

I currently am doing fixed 70/30 US/exUS split (and have been for 12yrs); but holding the market weighting for these two is definitely food-for-thought.

I still contend that holding the market weighting for Bond-vs-Stock is nonsensical for an individual investor.
"Buy-and-hold, long-term, all-market-index strategies, implemented at rock-bottom cost, are the surest of all routes to the accumulation of wealth" - John C. Bogle
User avatar
backpacker
Posts: 1620
Joined: Mon Sep 22, 2014 2:17 pm

Re: Bill Sharpe's preferred portfolio

Post by backpacker »

bjr89 wrote:An answer I've heard is someone who is employed at a small distressed company may want to tilt to large growth. But I'm not sure that really makes sense. Do employment hedging properties really matter enough to affect portfolio decisions so much? Also that assumes no behavioral component to value of which there certainly is, and so why sign yourself up for negative systematic alpha...
I share your skepticism here. I've never seen anyone who thought that value stocks had higher expected returns conclude that they should nevertheless tilt towards growth stocks to mitigate risk. On the other hand, it's very common to see investors hold shorter bonds while happily admitting that longer bonds have higher expected returns.

Also might depend on what one means by "factors". Industries have strong explanatory power. The prospects for one energy company tend to move with of the others. Basically everyone thinks that energy is a factor in this sense. But almost no one thinks that generally speaking, a portfolio that goes long energy stocks and short non-energy stocks with make any money. So the factor has no positive expected returns despite being explanatory.
Dirghatamas
Posts: 581
Joined: Fri Jan 01, 2016 5:18 pm

Re: Bill Sharpe's preferred portfolio

Post by Dirghatamas »

aj76er wrote: Thanks for the info. I overestimated the effects of share-creation, destruction, etc...
So, based on what you are saying, the "growth" factor dominates and the stock portfolio becomes somewhat self-driven. Not a bad way to manage the split between U.S. Market + ex-U.S. market. I think that I would always want these as separate funds(rather than something like VT) because I can minimize tax complexity (and to a certain point, tax efficiency) by keeping much of the ex-U.S. market in tax-sheltered accounts.

I currently am doing fixed 70/30 US/exUS split (and have been for 12yrs); but holding the market weighting for these two is definitely food-for-thought.

I still contend that holding the market weighting for Bond-vs-Stock is nonsensical for an individual investor.
Yes, a passive global portfolio is mostly set it and forget it stuff. It doesn't need much hand-holding. Yes, holding two funds instead of one is fine just for ER reasons.

I would caution you on tax efficiency. I pay a lot of attention to it as I have for ever been in ~>50% marginal tax rate (thanks my wonderful state). There ISN'T a simple rule of thumb like International in tax deferred and US in taxable that works. It is complex enough that you need to actually work it out for your case in a spreadsheet. It can be highly non-intuitive. Taxable has advantage of foreign tax credit but disadvantage of higher dividends and lower QDI. Each state taxes things differently. Tax rules in US change every few years and you may get stuck in one type of accounts if you put all in one basket.

I actually end up with almost mirror accounts in both taxable and tax deferred after much hand wringing. These tax efficiency calculations change so much over decades that it is no better than throwing darts. If you pick the wrong one and tax laws change, a decade later you can get stuck with huge capital gains and no way to move back. Hedging by having very similar accounts also diversifies you against future tax law changes :wink:
bjr89
Posts: 151
Joined: Sat Apr 23, 2016 8:15 am

Re: Bill Sharpe's preferred portfolio

Post by bjr89 »

-----
Last edited by bjr89 on Fri Nov 16, 2018 6:34 pm, edited 1 time in total.
User avatar
backpacker
Posts: 1620
Joined: Mon Sep 22, 2014 2:17 pm

Re: Bill Sharpe's preferred portfolio

Post by backpacker »

bjr89 wrote:
backpacker wrote:
bjr89 wrote:An answer I've heard is someone who is employed at a small distressed company may want to tilt to large growth. But I'm not sure that really makes sense. Do employment hedging properties really matter enough to affect portfolio decisions so much? Also that assumes no behavioral component to value of which there certainly is, and so why sign yourself up for negative systematic alpha...
I share your skepticism here. I've never seen anyone who thought that value stocks had higher expected returns conclude that they should nevertheless tilt towards growth stocks to mitigate risk. On the other hand, it's very common to see investors hold shorter bonds while happily admitting that longer bonds have higher expected returns.

Also might depend on what one means by "factors". Industries have strong explanatory power. The prospects for one energy company tend to move with of the others. Basically everyone thinks that energy is a factor in this sense. But almost no one thinks that generally speaking, a portfolio that goes long energy stocks and short non-energy stocks with make any money. So the factor has no positive expected returns despite being explanatory.
I'd say plenty of people are in fact tilting to growth stocks. They feel more comfortable with the business' prospects relative to the market and thus buy them over the market. That's why they have lower expected returns in the first place. But I hear ya. And it may not be such a conscious decision as it is when people tilt value
Definitely. It could be that no one is overweighting diversified growth funds but that lots of individual active investors have a bias towards growth stocks. The other side of diversified value is biased active investors, not diversified growth.

If that's right, an interesting corollary is that we should also expect the value premium to shrink over time. As more investors buy index funds, there will be fewer active investors and so fewer growth biased active investors. So the price of value will rise relative growth.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Bill Sharpe's preferred portfolio

Post by NiceUnparticularMan »

bjr89 wrote:I'm just saying that same logic should probably then be applied to stocks, but often times it is not. That is, if the cap weighted bond market doesn't reflect the needs of an individual's circumstance, why should the cap weighted stock market... Why stop with bonds
In a multi-factor equity-risk/return model, I would tend to agree--it doesn't make much sense to assume the aggregate appetite for such risk/returns is the same as my appetite. That said, a single-factor model may be good enough for many investors, in which case a TSM approach can make sense.
NiceUnparticularMan
Posts: 6103
Joined: Sat Mar 11, 2017 6:51 am

Re: Bill Sharpe's preferred portfolio

Post by NiceUnparticularMan »

bjr89 wrote:An answer I've heard is someone who is employed at a small distressed company may want to tilt to large growth. But I'm not sure that really makes sense. Do employment hedging properties really matter enough to affect portfolio decisions so much?
So I like to break this into two different questions. First, is this the right sort of risk, meaning with the right sort of correlation, to explain the relevant premium. Second, does the magnitude make sense.

When it comes to these theories about correlations with "human capital", I think they probably pass the first question. I have doubts about the second question, but that is equally true of the general equity premium (it seems too big if you only allow for rational amounts of risk aversion).

So maybe even that person with higher human capital risk should tilt to SV but then hedge in a more efficient way by, say, holding a larger reserve in short-term instruments. That would be a reasonable conclusion if the magnitude was badly off in terms of the SV premiums.
Fclevz
Posts: 651
Joined: Fri Mar 30, 2007 11:28 am

Re: Bill Sharpe's preferred portfolio

Post by Fclevz »

Quark wrote:Global stocks and bonds, cap weighted, free float, plus TIPS to taste...
Did I miss it where he mentions any kind of recommendation as to the amount of TIPS vs The Global Bond/Stock Portfolio?
Or is that one of those 'depends on the individual' things?
Or should I check out his 'lockbox' writings for guidance in that direction?
pascalwager
Posts: 2312
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

Fclevz wrote:
Quark wrote:Global stocks and bonds, cap weighted, free float, plus TIPS to taste...
Did I miss it where he mentions any kind of recommendation as to the amount of TIPS vs The Global Bond/Stock Portfolio?
Or is that one of those 'depends on the individual' things?
Or should I check out his 'lockbox' writings for guidance in that direction?
See the discussion and colored graph in Chapter 7: https://web.stanford.edu/~wfsharpe/RISMAT/RISMAT-7.pdf

Sharpe assumes the investor is working with an advisor, and the riskless portfolio proportion seems to be highly variable (from zero to 100%). I'm using the Sharpe world market portfolio, but with zero TIPS as I have a pension and large cash reserves.
VT 60% / VFSUX 20% / TIPS 20%
Post Reply