Bill Sharpe's preferred portfolio
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Is there any update on VT/BNDW weightings? It would be nice if this thread was regularly updated with these weightings (without having to look into changing external spreadsheet) for preserving a historical record.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
longinvest wrote: ↑Thu Feb 18, 2021 8:19 pm Is there any update on VT/BNDW weightings? It would be nice if this thread was regularly updated with these weightings (without having to look into changing external spreadsheet) for preserving a historical record.
Apparently this spreadsheet is updated in real time. As of right now, it's reporting 59.88% VT and 40.12% BNDW.djm2001 wrote: ↑Fri Sep 11, 2020 10:14 am Current snapshot for Sharpe's World Bond/Stock portfolio:
4-Fund:2-Fund (ETF):Code: Select all
INDEX MKT CAP ($M) % VTSAX/VTI $ 32,979,959.19M 31.70% VTIAX/VXUS $ 24,855,297.83M 23.89% VBTLX/BND $ 23,728,201.65M 22.81% VTABX/BNDX $ 22,472,081.44M 21.60%
Source (updated in real time): https://docs.google.com/spreadsheets/d/ ... g_/pubhtml#Code: Select all
INDEX MKT CAP ($M) % VT/VTWAX $ 57,835,257.03M 55.59% BNDW $ 46,200,283.10M 44.41%
Looks like the world's holding good ol' 60/40!
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Thanks.cos wrote: ↑Fri Feb 19, 2021 12:33 amApparently this spreadsheet is updated in real time. As of right now, it's reporting 59.88% VT and 40.12% BNDW.djm2001 wrote: ↑Fri Sep 11, 2020 10:14 am Source (updated in real time): https://docs.google.com/spreadsheets/d/ ... g_/pubhtml#
Looks like the world's holding good ol' 60/40!
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
So the GLD allocation is 0.2%?djm2001 wrote: ↑Wed Dec 23, 2020 9:50 am I went through the exercise of adding gold to Sharpe's World Bond Stock (WBS) portfolio. (This was a thought experiment... I don't currently hold any gold in my own portfolio.)
TL;DR: Publicly investable gold allocation is ~0.2% when added to Sharpe's WBS portfolio. (See the "4-Fund + Gold" tab here.)
Hopefully this answers the "Do I need gold?" question for those who think that Sharpe's portfolio is a good idea. You probably don't need gold unless:Okay, on to the details of how to find the right market cap for gold...
- you have already addressed larger gaps in Sharpe's WBS portfolio (e.g., high-yield corporate bonds), AND
- your portfolio is large enough that 0.2% of it is a non-negligible enough amount to bother about. (I'd only start to bother about it when 0.2% of my portfolio is a meaningful fraction -- say 5% or more -- of my annual expense.)
To start with, I know that we get some exposure to REITs just by holding VTI / VTSAX. I wondered if the same was true for gold. Turns out not to be true -- i.e., VTI's holdings do not contain GLD, IAU, AAAU, etc. I verified this by searching VTI's holdings here. (A possible reason for VTI not covering these gold investments is that the gold ETFs are structured as "grantor trusts" which are not considered investment companies, and therefore(?) not tracked by the CRSP US Total Stock Market index.)
Having double-checked now that gold is indeed a gap in the WBS portfolio, we now move onto the problem of determining the size of the gap by finding the right market cap to use for gold...
(Aside: I should pause here to say that I wholeheartedly agree with the argument that gold is of limited industrial use and that stored gold produces nothing. But I disagree that this completely eliminates gold's validity as an investment. If the global market ascribes value to something, then it is worth considering, however mad it might appear at first glance... it's just a question of making sure that we don't over-allocate to those bets. As we shall see, it is rather easy to over-allocate to gold if we aren't careful about what we measure. But if we measure the right thing, the result is reasonable -- 0.2% is a fairly conservative amount to allocate for a speculative bet.)
Simply googling for "gold market cap" gives an answer of ~$10T. Sharpe World Bond Stock market cap is currently ~$113T (see the "4-Fund" tab here), so adding in ~$10T gold would result in a ~8% allocation to it. But... $10T is in fact not the market cap we're after if we use Sharpe's methodology. We're actually only interested in the publicly investable portion of all the world's gold. It turns out that only ~2% of the $10T of the world's gold is held in publicly investable securities -- gold ETFs, mutual funds, and the like (e.g., GLD, IAU, AAAU) -- while the rest of the gold is held in jewelry, industrial use (fabrication), physical bullion / coin, and official holdings (e.g., central bank reserves) (source). When determining the allocation for gold ETFs in one's portfolio, I believe Sharpe's methodology would count only the market cap of gold held in ETF-investable trusts which 1) is more precisely representative of what we are buying (e.g., GLD), and 2) has a stronger case for market efficiency. (Please correct me if I'm wrong on this crucial point!)
The World Gold Council releases monthly gold ETF market caps here. I had to make a (free) account to access the data, and will track it going forward in my WBS spreadsheet. (Sharpe's original "2-Fund" and "4-Fund" WBS portfolios are still preserved in the sheet; gold is added only to the "N-Fund" variant and the "4-Fund + Gold" variant.)
Btw, the nuance here about using only the publicly investable portion of gold is very reminiscent of searching for "bond market cap", and finding an answer of $100T or more (from SIFMA data), while Sharpe's methodology (i.e., using the market cap of the indexes actually being invested in) works out to less than half of that (~$47T) for bonds. The reason is that the SIFMA market caps include things like private placement bonds that are inaccessible to the retail investor, just as the total gold market cap contains private jewelry, central bank reserves, etc.
This story is also similar to how WBS's real estate exposure is provided by (only) market-weighted public REITs, as opposed to over-weighting public REITs to try to match the market cap of private REITs and privately-owned real estate.
My main takeaway from Sharpe's original WBS portfolio as well as this extension to gold is: buy what you measure, and measure what you buy.
P.S., I wonder how to extend this methodology to crypto. I found data sources for the overall crypto market caps here, but have not yet figured out how to measure the "investable" portion -- is it the whole market cap, or are there investor-inaccessible portions blocked off by certain entities? Any suggestions on how to figure this out? (edit: Found a recent attempt at measuring "free float crypto" here. My original market cap source claims to use "circulating supply" to measure market cap, but its numbers for Bitcoin total supply was equal to the circulating supply, so I don't think any free float correction was actually done.) But note that, even with the whole market cap, Bitcoin would be no more than a 0.4% allocation of WBS + Bitcoin portfolio. So again, it's ignorable for most investors until their portfolio grows very large and they have addressed the larger gaps in the WBS portfolio, particularly in junk bonds.
P.P.S., This post was a thought exercise to see whether Sharpe's market cap methodology extends reasonably to other asset classes. It seems to do a reasonable job that aligned well with my own previously-held beliefs and preferences (e.g., I would balk at an 8% allocation to gold), but I suppose that your reaction to the results and methodology will depend on your own beliefs about things like gold and crypto. I am not invested in either gold or crypto myself. (I won't invest in crypto until I am confident that some of the practical details (e.g., tax reporting and beneficiaries/inheritence) work smoothly and conveniently.)
edit: fixed a link, grammar
So $2k out of every $1M?
That wouldn’t seem to move the needle one way or the other.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
Yeah, gold investments (i.e., GLD + AAAU + IAU + ...) market cap is still about 0.2% (currently 0.18%) relative to world stocks/bonds. (Note: If you want to buy physical bullion / jewelry, use the appropriate bullion/jewelry market caps. But don't buy gold ETFs at jewelry market caps, or jewelry at ETF market caps.)
Bitcoin has a higher market cap (around 1%), but that market cap is not free-float corrected. So the free-float market cap (which is the theoretically correct thing to use) is lower than what we'd find on something like coinmarketcap.com. One source (coinmetrics.io) said that the correction factor for BTC was around 0.78. But who knows how accurate or current their estimate is. Personally, I'd play it safe, and use a 0.5 correction factor if I were to decide to invest in Bitcoin.
As for whether these will move the needle or not... gold and crypto have high reward in certain low-probability events. In theory, one should try to buy them... once it becomes practical to do so (e.g., once 0.2% of your portfolio equals $10k or so). Just as one should not turn one's nose up at junk bonds (1.5% allocation) once it becomes practical for you to buy them. But if you're missing a much huger asset class (e.g., international bonds), certainly makes more sense to tackle that larger gap in your portfolio first.
Last edited by djm2001 on Fri Feb 19, 2021 5:03 pm, edited 1 time in total.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Re: Bill Sharpe's preferred portfolio
At 0.2%, I don't think it's even worth the bother.djm2001 wrote: ↑Fri Feb 19, 2021 4:20 pmYeah, gold investments (i.e., GLD + AAAU + IAU + ...) market cap is still about 0.2% (currently 0.18%) relative to world stocks/bonds. (Note: If you want to buy physical bullion / jewelry, use the appropriate bullion/jewelry market caps. But don't buy gold ETFs at jewelry market caps, or jewelry at ETF market caps.)
Crypto (BTC + ETH + ...) has a higher market cap (around 1%), but that market cap is not free-float corrected. So the free-float market cap (which is the theoretically correct thing to use) is lower than what we'd find on something like coinmarketcap.com. One source (coinmetrics.io) said that the correction factor for BTC was around 0.78. But who knows how accurate or current their estimate is. Personally, I'd play it safe, and use a 0.5 correction factor if I were to decide to invest in Bitcoin.
As for whether these will move the needle or not... gold and crypto have high reward in certain low-probability events. In theory, one should try to buy them... once it becomes practical to do so (e.g., once 0.2% of your portfolio equals $10k or so). Just as one should not turn one's nose up at junk bonds (1.5% allocation) once it becomes practical for you to buy them. But if you're missing a much huger asset class (e.g., international bonds), certainly makes more sense to tackle that larger gap in your portfolio first.
Even at the 7 figure portfolio size level, it's just not going to make a difference.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
It is indeed updated in realtime (from realtime ETF prices), and I have been updating the factsheet snapshots within a day or so of the factsheets being released.
I was holding off on posting monthly snapshots because I didn't want to spam the thread, but I do see the value in having a historical view. I can update this thread each month (just like CyberBob kindly used to do) once all the appropriate factsheets have been released for that month.
Here's the latest snapshot as of end of day Feb 19, 2021.
4-fund:
Code: Select all
INDEX MKT CAP ($M) %
VTSAX/VTI $ 40,384,067.21 34.10%
VTIAX/VXUS $ 30,714,169.62 25.94%
VBTLX/BND $ 24,206,025.94 20.44%
VTABX/BNDX $ 23,114,141.15 19.52%
Code: Select all
INDEX MKT CAP ($M) %
VT $ 71,098,236.83 60.04%
BNDW $ 47,320,167.08 39.96%
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Effectively. That's $2,000 for a $1,000,000 portfolio. If global stocks and bonds lost -55% (this probably requires that stocks lose more than -80%) and gold tripled at the same time, gaining +200%, two investors, one holding a portfolio without gold and the other holding a portfolio with gold at market weight end up with ($1,000,000 - 55%) = $450,000 (without gold) and (($998,000 - 55%) + ($2,000 + 200%)) = $455,100 (with gold). That's (($455,100 / $450,000) - 1) = 1% more after the loss for holding a market weight of soaring gold during the crash. Differences in the timing of rebalancing (to market weight) events is likely to have a bigger impact on returns!
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
You could make this same argument for each individual stock, and eliminate them from your portfolio one by one. Just a thought.longinvest wrote: ↑Fri Feb 19, 2021 4:43 pmEffectively. That's $2,000 for a $1,000,000 portfolio. If global stocks and bonds lost -55% (this probably requires that stocks lose more than -80%) and gold tripled at the same time, gaining +200%, two investors, one holding a portfolio without gold and the other holding a portfolio with gold at market weight end up with ($1,000,000 - 55%) = $450,000 (without gold) and (($998,000 - 55%) + ($2,000 + 200%)) = $455,100 (with gold). That's (($455,100 / $450,000) - 1) = 1% more after the loss for holding a market weight of soaring gold during the crash. Differences in the timing of rebalancing (to market weight) events is likely to have a bigger impact on returns!
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
If I was buying individual stocks, that might be a valid argument.djm2001 wrote: ↑Fri Feb 19, 2021 4:50 pmYou could make this same argument for each individual stock, and eliminate them from your portfolio one by one. Just a thought.longinvest wrote: ↑Fri Feb 19, 2021 4:43 pmEffectively. That's $2,000 for a $1,000,000 portfolio. If global stocks and bonds lost -55% (this probably requires that stocks lose more than -80%) and gold tripled at the same time, gaining +200%, two investors, one holding a portfolio without gold and the other holding a portfolio with gold at market weight end up with ($1,000,000 - 55%) = $450,000 (without gold) and (($998,000 - 55%) + ($2,000 + 200%)) = $455,100 (with gold). That's (($455,100 / $450,000) - 1) = 1% more after the loss for holding a market weight of soaring gold during the crash. Differences in the timing of rebalancing (to market weight) events is likely to have a bigger impact on returns!
But I hold indexes.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
Fair enough. So the rationale is one of practicality, not just because the math says it's "negligible". If so, I'm on the same page. I don't (yet) hold gold or crypto because of the practicality concerns (combined with the negligibleness).
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Some indices contain micro-cap stocks, others don't. The minimal grade and other rules for including bonds (and stocks) and evaluating their free-float weight also differ between indices. The market weight of gold is within noise level, where it doesn't make much difference whether it's included or not.
There are also other considerations. It's perfectly reasonable for an investor not to wish to include, within the portfolio, assets which don't have an internal driver of returns. Both stocks and bonds have an internal driver of returns. Stocks are certificates of partial ownership of companies which strive to make a profit (leading to growth or dividend payments), along with voting rights to select different managers if the company is mismanaged. Bonds are contracts to pay coupons on specific dates and a face value at maturity. Most broad market indices tend to include stocks and bonds which meet certain investability criteria. Pink sheet stocks and junk bonds are usually excluded. High-yield bonds with a not-too-junky rating could be included or not. But, the main idea, here, is that getting money out of the investment is possible without finding a buyer. Share owners (as a group) of a company could elect a new management team to force the company to pay dividends. Bond holders can simply hold their (non-defaulting) bond to maturity to get all coupons and a face value. Theoretically, no buyer is absolutely necessary to get money back.
Gold and other commodities, as well as Beanie Babies, require a buyer to get money back. They don't have an internal driver of returns. On the contrary, they usually have storage (and maintenance) costs. It's perfectly rational for an investor not to want to include such purely speculative assets into the portfolio.
There are also derivatives, but they're a zero-sum game for players, before fees, and a negative-sum game, after fees. They can be useful to hedge currency, as done by BNDX and BNDW, but they're not held at market weight by these funds; they're used as tools to mostly eliminate the short-term impact of currency fluctuations.
Cash in bank accounts and bank CDs aren't marketable.
So, it comes down to the fact that the two main marketable investment-grade asset classes are stocks and bonds, with some differences in the inclusion filter of various capitalization-weighted indices.
I have no trouble skipping on a market-weight allocation to commodities, derivatives, cash, and Beanie Babies in my portfolio.*
* I explained in another thread that it can be reasonable for a global stock-and-bond market index investor to keep a moderate home bias due to the slightly higher risk, in general, of foreign investing.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Here are two posts that I wrote on a zombie thread (that was awoken earlier today) thinking it was this thread. They belong here.
FIRST POST
A Canadian bond returns 3% over a period of one year. A Canadian investor holding this bond experiences a 3% return (while Canadian inflation was 2%). During this same year, the Canadian dollar loses 20% relative to the U.S. dollar. A U.S. investor, holding the same Canadian bond, experiences a (((1 + 3%) X (1 - 20%)) - 1) = -17.6% return (while U.S. inflation was also 2%).
Which is the right return, for a U.S. investor? A return close to 3%, or a -17.6% return?
There's probably no right answer. One view is that, in U.S. dollars, the Canadian bond effectively lost -17.6%. Another view is that the bond's return was 3%, so investors holding it should get something close to it in their domestic currency.
The advantage of currency-hedged international bonds is that they behave like bonds in the investor's domestic currency.
Vanguard has written an interesting paper about it: "The portfolio currency-hedging decision, by objective and block by block":
At some point, it's futile to look for the perfect portfolio. It doesn't exist. But, some portfolios are good enough. Sharpe's portfolio (VT/BNDW at market weight) is good enough. There are other good enough portfolios such as Vanguard's global 60/40 stocks/bonds index fund, but with a moderate home bias (VSMGX) which is close enough to the principles of Sharpe's portfolio.
----
SECOND POST
Here's how Vanguard puts it:
Don't let negative international yields get in the way of a 'free lunch'
Let me add that there's a lot of value to the current thread, documenting Sharpe's World Stock/Bond portfolio, as it is a very good starting point for thinking about asset allocation.
FIRST POST
Currencies are significantly more volatile than bonds.watchnerd wrote:Ideally, I would think so.Blue456 wrote:Shouldn't that involve holding unhedged bonds?watchnerd wrote: In essence, it's a philosophical extension of the 'own the total market' philosophy.
After all, it holds unhedged stocks.
A Canadian bond returns 3% over a period of one year. A Canadian investor holding this bond experiences a 3% return (while Canadian inflation was 2%). During this same year, the Canadian dollar loses 20% relative to the U.S. dollar. A U.S. investor, holding the same Canadian bond, experiences a (((1 + 3%) X (1 - 20%)) - 1) = -17.6% return (while U.S. inflation was also 2%).
Which is the right return, for a U.S. investor? A return close to 3%, or a -17.6% return?
There's probably no right answer. One view is that, in U.S. dollars, the Canadian bond effectively lost -17.6%. Another view is that the bond's return was 3%, so investors holding it should get something close to it in their domestic currency.
The advantage of currency-hedged international bonds is that they behave like bonds in the investor's domestic currency.
Vanguard has written an interesting paper about it: "The portfolio currency-hedging decision, by objective and block by block":
It probably explains Vanguard's choice to hedge currencies for foreign bonds, and not to do it for foreign stocks. William Sharpe doesn't seem to disagree with Vanguard about it.[H]edging decisions that preserve the risk-and-return characteristics of the underlying assets lead to a portfolio-level hedging strategy that is consistent with the portfolio’s objectives.
At some point, it's futile to look for the perfect portfolio. It doesn't exist. But, some portfolios are good enough. Sharpe's portfolio (VT/BNDW at market weight) is good enough. There are other good enough portfolios such as Vanguard's global 60/40 stocks/bonds index fund, but with a moderate home bias (VSMGX) which is close enough to the principles of Sharpe's portfolio.
----
SECOND POST
High inflation will probably lead to higher U.S. short-term rates than Canadian short-term rates (assuming that the Bank of Canada sticks to its monetary policy which consists of "keeping inflation low, stable and predictable", e.g. keeping Canadian inflation at 2%). This will deliver a "hedge return" that will probably compensate enough for the difference in inflation rates, while preserving diversification of returns due to distinct Canadian and U.S. yield curves. The hedge return isn't perfect, but generally good enough.Blue456 wrote: Maybe so but what happens to your hedged bonds when US hits high inflation?
Here's how Vanguard puts it:
Don't let negative international yields get in the way of a 'free lunch'
----The decision to hedge also impacts the return experience of a U.S.-based investor. This is because an investment in an international bond by a U.S.-based investor consists of three return components: The price return, the income return, and the currency return. By hedging, you replace the volatile currency return component with a return component approximately equal to the short-term interest rate differential between the USD and the currency of the bond.
For example, when short-term U.S. interest rates are higher than the short-term interest rates of the country from which you are purchasing the bond, hedging the currency exposure can be additive to the overall total return. While the return from hedging the currency exposure can be either positive or negative, it will almost always create a total return on the bond that is different than the one implied by its yield to maturity. For this reason, we believe investors should focus less on yield comparisons across markets and instead maintain their focus on the diversification benefits that hedged international bonds can bring[...]
Let me add that there's a lot of value to the current thread, documenting Sharpe's World Stock/Bond portfolio, as it is a very good starting point for thinking about asset allocation.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 1
- Joined: Wed Mar 03, 2021 12:59 pm
Re: Bill Sharpe's preferred portfolio
Does anyone know where I can find historical data for the past weights of Sharpe's portfolio?
I have this spreadsheet: https://docs.google.com/spreadsheets/d/ ... g_/pubhtml#
But it only shows the current allocations.
I have this spreadsheet: https://docs.google.com/spreadsheets/d/ ... g_/pubhtml#
But it only shows the current allocations.
Re: Bill Sharpe's preferred portfolio
Sharpe's World Bond/Stock portfolio snapshot as of end of day, Mar 5, 2021:
4-fund:
Source: real-time tracker.
Methodology:
The forward-projection formula is simply:
where snapshot_market_cap is taken from the index's latest factsheet, and current_tracker_price and snapshot_tracker_price are the prices of an appopriate ETF tracker (e.g., VTI, VXUS, BND, BNDX) as of the current time and the date of the market cap reported on the factsheet.
edit: added Methodology section for posterity at longinvest's request
4-fund:
Code: Select all
INDEX MKT CAP ($M) %
VTSAX/VTI $ 39,360,593.74 33.95%
VTIAX/VXUS $ 29,672,138.46 25.59%
VBTLX/BND $ 23,983,295.09 20.69%
VTABX/BNDX $ 22,916,825.92 19.77%
Methodology:
- The VTSAX/VTI allocation was calculated from the "Index Market Cap" listed in the CRSP US Total Market Index factsheet for Dec 31, 2021; and then forward-projected using the price of VTI then and now.
- The VTIAX/VXUS allocation was calculated from the "Net MCap" listed in the FTSE Global All Cap ex-US factsheet for Feb 26, 2021; and then forward-projected using the price of VXUS then and now.
- The VBTLX/BND allocation was calculated from the "Market Value" listed in the FTSE US Broad Investment-Grade Bond Index (FTSE USBIG) factsheet for Feb 28, 2021, and forward-projected using the price of BND then and now.
- The VTABX/BNDX allocation was calculated from the difference of the "Market Value" fields listed in the FTSE World Broad Investment-Grade Bond Index (FTSE WorldBIG) factsheet and the FTSE USBIG index factsheet, both for Feb 28, 2021; and then forward-projected using the price of BNDX then and now.
- The VT allocation is the sum of the VTSAX/VTI and VTIAX/VXUS allocations.
- The BNDW allocation is the sum of the VBTLX/BND and VTABX/BNDX allocations.
The forward-projection formula is simply:
Code: Select all
current_market_cap := snapshot_market_cap * current_tracker_price / snapshot_tracker_price
edit: added Methodology section for posterity at longinvest's request
Last edited by djm2001 on Wed Mar 31, 2021 3:55 pm, edited 5 times in total.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
-
- Posts: 1911
- Joined: Wed Jan 29, 2020 9:29 am
Re: Bill Sharpe's preferred portfolio
Thanks for updating!djm2001 wrote: ↑Sat Mar 06, 2021 4:20 am Sharpe's World Bond/Stock portfolio snapshot as of end of day, Mar 5, 2021:
4-fund:2-fund:Code: Select all
INDEX MKT CAP ($M) % VTSAX/VTI $ 39,360,593.74 33.95% VTIAX/VXUS $ 29,672,138.46 25.59% VBTLX/BND $ 23,983,295.09 20.69% VTABX/BNDX $ 22,916,825.92 19.77%
Source: real-time tracker.Code: Select all
INDEX MKT CAP ($M) % VT $ 69,032,732.20 59.55% BNDW $ 46,900,121.01 40.45%
How in the world can/does one backtest this thing more than a decade or so?
"I am better off than he is – for he knows nothing and thinks that he knows. I neither know nor think that I know." - Socrates. "Nobody knows nothing." - Jack Bogle
Re: Bill Sharpe's preferred portfolio
I don't know where to get the historical data, unfortunately. But just don't be tempted into using the data from the FTSE AAAP tool that Sharpe advocates. The AAAP tool uses rather different indexes from what the WBS funds track, and consequently significantly overweights bonds (see asset_chaos's explanation). In a market portfolio approach, we should avoid these unintentional distortions caused by measuring one thing and then buying something else.CarnageCard21 wrote: ↑Wed Mar 03, 2021 1:06 pm Does anyone know where I can find historical data for the past weights of Sharpe's portfolio?
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Re: Bill Sharpe's preferred portfolio
That is probably what CarnageCard21 is also after. Personally, I backtest a few fixed allocations, and look at the spread in the results. The "central" fixed allocation that I use is 31% VTSAX, 24% VTIAX, 23% VBTLX, 22% VTABX. Example: https://bit.ly/3bZSUIH.whereskyle wrote: ↑Sat Mar 06, 2021 5:19 am How in the world can/does one backtest this thing more than a decade or so?
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Re: Bill Sharpe's preferred portfolio
1999 is as far back as I can go with Portfolio Visualizer using the asset class tool.whereskyle wrote: ↑Sat Mar 06, 2021 5:19 amThanks for updating!djm2001 wrote: ↑Sat Mar 06, 2021 4:20 am Sharpe's World Bond/Stock portfolio snapshot as of end of day, Mar 5, 2021:
4-fund:2-fund:Code: Select all
INDEX MKT CAP ($M) % VTSAX/VTI $ 39,360,593.74 33.95% VTIAX/VXUS $ 29,672,138.46 25.59% VBTLX/BND $ 23,983,295.09 20.69% VTABX/BNDX $ 22,916,825.92 19.77%
Source: real-time tracker.Code: Select all
INDEX MKT CAP ($M) % VT $ 69,032,732.20 59.55% BNDW $ 46,900,121.01 40.45%
How in the world can/does one backtest this thing more than a decade or so?
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Djm2001,
I have two small additional wishes.
1) I would prefer for snapshots to be reported as of the last day of each month. For example, the last snapshot would report the weightings as of February 28, 2021 (which would in reality be as of the last trading day of that month, February 26, but reported as of the 28th as is usually done).
This would make it possible to eventually calculate benchmark returns using available monthly return data for VT and BNDW. (I might start doing that, once we get at least 12 reports of consistent end of month weighting data).
2) It would be nice if the monthly post reported the actual data sources that were used (by the spreadsheet) to calculate the weightings.
This is actually important, so that others can check the validity of reported weightings. Also, in 10 years, when someone comes back reading the post, the spreadsheet might have changed its data sources. Reporting the sources along with data reduces the dependency on an external spreadsheet.
In other words, readers shouldn't have to trust a random stranger on the internet about the weightings of Sharpe's World Bond/Stock Portfolio. Allowing readers to easily recover the same data and reproduce calculations (without using an external spreadsheet) is critical to establish confidence in the reported numbers.
Thanks,
longinvest
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
Good idea. Edited my post above to include a Methodology section with calculation details and links to the underlying factsheets. Going forward, I'll post the numbers as of end of day of the last day of the month.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Thanks Djm2001! It's really nice to see the methodology along with data.
Last edited by longinvest on Wed Mar 10, 2021 10:43 pm, edited 1 time in total.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
I suggest to simplify tracking the Global Stock-and-Bond Portfolio (or, if you prefer, William Sharpe's Market Portfolio) to:
The global stock index is "free-float weighted to ensure that only the investable opportunity set is included". As for the global bond index, the FTSE Fixed Income Index Guide says that it has "market accessibility level criteria" such as "(excludes Federal Reserve holdings)". So, it would appear that the bond index is somewhat free-float weighted, too.
As of December 31, 2020 the weights were:
The January 2021 return of William Sharpe's Market Portfolio was:
The February 2021 return of William Sharpe's Market Portfolio was:
I'll wait until we have 12 months of returns to provide a growth chart.
For practical investing purpose I think that the Vanguard LifeStrategy Moderate Growth Fund (VSMGX) is a close-enough approximation of the Global Stock-and-Bond Portfolio with a moderate home bias justified by the slightly higher risks (political, confiscation, etc.) and costs of foreign investing.
- Global stocks: FTSE Global All Cap Index (GEISLMS) -- Market cap: Net MCap
- Global bonds: FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market cap: Market Value
The global stock index is "free-float weighted to ensure that only the investable opportunity set is included". As for the global bond index, the FTSE Fixed Income Index Guide says that it has "market accessibility level criteria" such as "(excludes Federal Reserve holdings)". So, it would appear that the bond index is somewhat free-float weighted, too.
As of December 31, 2020 the weights were:
- Global stocks: $66,377,472 million USD Market cap -- 57.94%
- FTSE Global All Cap Index (GEISLMS) -- Net MCap
- Global bonds: $48,180,600 million USD Market cap -- 42.06%
- FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value
The January 2021 return of William Sharpe's Market Portfolio was:
- Global Stock-and-Bond Portfolio: ((57.94% X -0.39% (global stocks)) + (42.06% X -0.73% (global bonds))) = -0.53%
- Global stocks: Vanguard Total World Stock ETF (VT) NAV return
- Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
- Global stocks: $66,059,233 million USD Market cap -- 57.95%
- FTSE Global All Cap Index (GEISLMS) -- Net MCap
- Global bonds: $47,943,910 million USD Market cap -- 42.05%
- FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value
The February 2021 return of William Sharpe's Market Portfolio was:
- Global Stock-and-Bond Portfolio: ((57.95% X 2.74% (global stocks)) + (42.05% X -1.62% (global bonds))) = 0.91%
- Global stocks: Vanguard Total World Stock ETF (VT) NAV return
- Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
- Global stocks: $67,770,296 million USD Market cap -- 58.97%
- FTSE Global All Cap Index (GEISLMS) -- Net MCap
- Global bonds: $47,153,410 million USD Market cap -- 41.03%
- FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value
I'll wait until we have 12 months of returns to provide a growth chart.
For practical investing purpose I think that the Vanguard LifeStrategy Moderate Growth Fund (VSMGX) is a close-enough approximation of the Global Stock-and-Bond Portfolio with a moderate home bias justified by the slightly higher risks (political, confiscation, etc.) and costs of foreign investing.
Last edited by longinvest on Sun May 23, 2021 7:01 pm, edited 1 time in total.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
Sharpe's World Bond/Stock portfolio snapshot as of end of day, Mar 31, 2021:
4-fund:
2-fund:
Source: Real-time tracker.
Methodology:
(Note: I improved the methodology for the 2-fund portfolio since last time based on longinvest's suggestion to use FTSE Global All Cap Index for the VT allocation, and so I'm restating the methodology anew. In subsequent posts, I will simply link to this post to reference the methodology, so long as the methodology remains unchanged.)
The forward-projection formula is simply:
where snapshot_market_cap is taken from the index's latest factsheet, and current_tracker_price and snapshot_tracker_price are the prices of an appopriate ETF tracker (e.g., VTI, VXUS, BND, BNDX, VT, BNDW) as of the current time and the date of the market cap reported on the factsheet.
edit: added mention that the market caps for the 2-fund portfolio are projected forward using VT and BNDW prices.
4-fund:
Code: Select all
INDEX MKT CAP ($M) %
VTSAX/VTI $ 40,525,235.96 34.57%
VTIAX/VXUS $ 29,929,908.00 25.53%
VBTLX/BND $ 23,876,201.64 20.37%
VTABX/BNDX $ 22,900,788.96 19.53%
Code: Select all
INDEX MKT CAP ($M) %
VT $ 69,521,004.83 59.72%
BNDW $ 46,900,182.00 40.28%
Methodology:
(Note: I improved the methodology for the 2-fund portfolio since last time based on longinvest's suggestion to use FTSE Global All Cap Index for the VT allocation, and so I'm restating the methodology anew. In subsequent posts, I will simply link to this post to reference the methodology, so long as the methodology remains unchanged.)
- The VTSAX/VTI allocation was calculated from the "Index Market Cap" listed in the CRSP US Total Market Index factsheet for Dec 31, 2021; and then forward-projected using the price of VTI then and now.
- The VTIAX/VXUS allocation was calculated from the "Net MCap" listed in the FTSE Global All Cap ex-US factsheet for Feb 26, 2021; and then forward-projected using the price of VXUS then and now.
- The VBTLX/BND allocation was calculated from the "Market Value" listed in the FTSE US Broad Investment-Grade Bond Index (FTSE USBIG) factsheet for Feb 28, 2021, and forward-projected using the price of BND then and now.
- The VTABX/BNDX allocation was calculated from the difference of the "Market Value" fields listed in the FTSE World Broad Investment-Grade Bond Index (FTSE WorldBIG) factsheet and the FTSE USBIG index factsheet, both for Feb 28, 2021; and then forward-projected using the price of BNDX then and now.
- The VT allocation was calculated from the "Net MCap" listed in the FTSE Global All Cap Index factsheet for Feb 26, 2021; and then forward-projected using the price of VT then and now. Note that VT directly tracks the FTSE Global All Cap Index, and so it is more appropriate to use this index's market cap, rather than adding together the market caps for the indexes used by VTSAX/VTI and VTIAX/VXUS.
- The BNDW allocation was calculated from the "Market Value" listed in the FTSE World Broad Investment-Grade Bond Index (FTSE WorldBIG) factsheet for Feb 28, 2021; and then forward-projected using the price of BNDW then and now.
The forward-projection formula is simply:
Code: Select all
current_market_cap := snapshot_market_cap * current_tracker_price / snapshot_tracker_price
edit: added mention that the market caps for the 2-fund portfolio are projected forward using VT and BNDW prices.
Last edited by djm2001 on Sun Apr 04, 2021 6:08 am, edited 1 time in total.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
-
- Posts: 2312
- Joined: Mon Oct 31, 2011 8:36 pm
Re: Bill Sharpe's preferred portfolio
Sharpe recommended the world market portfolio for "many retirees", but with TIPS on the side.
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
This post documents the return and weights as of March 31, 2021 of the Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.
The March 2021 return of William Sharpe's Market Portfolio was:
The March 2021 return of William Sharpe's Market Portfolio was:
- Global Stock-and-Bond Portfolio: ((58.97% X 2.73% (global stocks)) + (41.03% X -0.59% (global bonds))) = 1.37%
- Global stocks: Vanguard Total World Stock ETF (VT) NAV return
- Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
- Global stocks: $69,827,178 million USD Market cap -- 59.90%
- FTSE Global All Cap Index (GEISLMS) -- Net MCap
- Global bonds: $46,745,860 million USD Market cap -- 40.10%
- FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value
Last edited by longinvest on Sun May 23, 2021 7:00 pm, edited 1 time in total.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
To quote the FTSE World Broad Investment-Grade Bond Index (WorldBIG®):longinvest wrote: ↑Sat Apr 17, 2021 12:39 pm This post documents the return and weights as of March 31, 2021 of the Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.
The March 2021 return of William Sharpe's Market Portfolio was:As of March 31, 2020 the weights were:
- Global Stock-and-Bond Portfolio: ((58.97% X 2.73% (global stocks)) + (41.03% X -0.59% (global bonds))) = 1.37%
- Global stocks: Vanguard Total World Stock ETF (VT) NAV return
- Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
For practical investing purpose I think that the Vanguard LifeStrategy Moderate Growth Fund (VSMGX) is a close-enough approximation of the Global Stock-and-Bond Portfolio with a moderate home bias justified by the slightly higher risks (political, etc.) and costs of foreign investing. Conveniently, it can be used as a One-Fund Portfolio. Its March 2021 return was 1.29%.
- Global stocks: $69,827,178 million USD Market cap -- 59.90%
- FTSE Global All Cap Index (GEISLMS) -- Net MCap
- Global bonds: $46,745,860 million USD Market cap -- 40.10%
- FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value
Multi-currency bonds, to me, means local currency.The FTSE World Broad Investment-Grade Bond Index (WorldBIG) is a multi-asset, multi-currency benchmark, which provides a broad-based measure of the global fixed income markets.
So wouldn't a local currency bond fund be better?
BWX is local currency, but it's only sovereign debt.
Although that's probably 70%+ of the non-USD debt market if US TBM is any kind of benchmark.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Watchnerd, here's the link to a Vanguard document discussing currency hedging: The portfolio currency-hedging decision, by objective and block by block. In summary, hedging the currency exposure of international bonds is sensible; it preserves "the risk-and-return characteristics of the underlying assets". In other words, currency hedging allows international bonds to have bond-like volatility (instead of stock-like volatility when exposed of currency fluctuations).watchnerd wrote: ↑Fri Apr 23, 2021 11:49 pmTo quote the FTSE World Broad Investment-Grade Bond Index (WorldBIG®):longinvest wrote: ↑Sat Apr 17, 2021 12:39 pm This post documents the return and weights as of March 31, 2021 of the Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.
The March 2021 return of William Sharpe's Market Portfolio was:As of March 31, 2020 the weights were:
- Global Stock-and-Bond Portfolio: ((58.97% X 2.73% (global stocks)) + (41.03% X -0.59% (global bonds))) = 1.37%
- Global stocks: Vanguard Total World Stock ETF (VT) NAV return
- Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
For practical investing purpose I think that the Vanguard LifeStrategy Moderate Growth Fund (VSMGX) is a close-enough approximation of the Global Stock-and-Bond Portfolio with a moderate home bias justified by the slightly higher risks (political, etc.) and costs of foreign investing. Conveniently, it can be used as a One-Fund Portfolio. Its March 2021 return was 1.29%.
- Global stocks: $69,827,178 million USD Market cap -- 59.90%
- FTSE Global All Cap Index (GEISLMS) -- Net MCap
- Global bonds: $46,745,860 million USD Market cap -- 40.10%
- FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value
Multi-currency bonds, to me, means local currency.The FTSE World Broad Investment-Grade Bond Index (WorldBIG) is a multi-asset, multi-currency benchmark, which provides a broad-based measure of the global fixed income markets.
So wouldn't a local currency bond fund be better?
BWX is local currency, but it's only sovereign debt.
Although that's probably 70%+ of the non-USD debt market if US TBM is any kind of benchmark.
Here's William Sharpe's opinion about it (source):
One aspect of the Barclays Non-U.S. bond index, and of the Vanguard fund that tracks it, merits mention. Both are “currency hedged”. In effect, VTABX provides investment in non-U.S. Bonds plus monthly side-bets on changes in exchange rates between the U.S. Dollar and the home currencies of the bonds in the portfolio. This is accomplished by committing to exchange foreign currencies and dollars at a pre-specified exchange rate a month hence. There will of course be traders or investors on the other side of each of these transactions (undoubtedly including some Europeans, Japanese, and British), who may well consume goods produced in other countries. To be sure, if exchange rates always reflected the relative costs of buying goods and services with different currencies so that purchasing power parity held, real returns on unhedged positions might be little affected by changes in such rates. But at best, this is likely to be true only in the long run.
One might ask why non-U.S. Bond exposures should be hedged but not non-U.S. Stock exposures. A partial justification would be the much greater uncertainty about the value of stock positions a month hence, making currency hedges based on the estimated value a month hence of the securities currently held far less than perfect. Happily, despite providing this additional feature, the expense ratio for VTABX is considerably lower than the costs for Non-U.S. bond funds that do not employ currency hedging.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
I've read the Vanguard paper on USD-hedging bonds half a dozen times; I find their logic tortured.longinvest wrote: ↑Sat Apr 24, 2021 2:25 pm
Watchnerd, here's the link to a Vanguard document discussing currency hedging: The portfolio currency-hedging decision, by objective and block by block. In summary, hedging the currency exposure of international bonds is sensible; it preserves "the risk-and-return characteristics of the underlying assets". In other words, currency hedging allows international bonds to have bond-like volatility (instead of stock-like volatility when exposed of currency fluctuations).
Here's William Sharpe's opinion about it (source):
One aspect of the Barclays Non-U.S. bond index, and of the Vanguard fund that tracks it, merits mention. Both are “currency hedged”. In effect, VTABX provides investment in non-U.S. Bonds plus monthly side-bets on changes in exchange rates between the U.S. Dollar and the home currencies of the bonds in the portfolio. This is accomplished by committing to exchange foreign currencies and dollars at a pre-specified exchange rate a month hence. There will of course be traders or investors on the other side of each of these transactions (undoubtedly including some Europeans, Japanese, and British), who may well consume goods produced in other countries. To be sure, if exchange rates always reflected the relative costs of buying goods and services with different currencies so that purchasing power parity held, real returns on unhedged positions might be little affected by changes in such rates. But at best, this is likely to be true only in the long run.
One might ask why non-U.S. Bond exposures should be hedged but not non-U.S. Stock exposures. A partial justification would be the much greater uncertainty about the value of stock positions a month hence, making currency hedges based on the estimated value a month hence of the securities currently held far less than perfect. Happily, despite providing this additional feature, the expense ratio for VTABX is considerably lower than the costs for Non-U.S. bond funds that do not employ currency hedging.
If I want to make side-bets on currency, I can do that without holding USD-hedged foreign bonds; in fact, I do exactly that, by holding currency futures.
I also find Sharpe's commentary to be inconsistent with using FTSE WBIG is at the target index.
I think he knows it, too, given he uses the word 'justification'.
The FTSE WBIG is *not* currency hedged -- ergo, a preferred fund that maps to it shouldn't be hedged, either.
Why monkey around with all of this rationalization for holding USD-hedged bonds when the more philosophically consistent, and simpler, action that fits with the market portfolio is to hold local currency bonds?
After all, we hold local currency stocks...
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Watchnerd,watchnerd wrote: ↑Sat Apr 24, 2021 3:49 pm I've read the Vanguard paper on USD-hedging bonds half a dozen times; I find their logic tortured.
If I want to make side-bets on currency, I can do that without holding USD-hedged foreign bonds; in fact, I do exactly that, by holding currency futures.
I also find Sharpe's commentary to be inconsistent with using FTSE WBIG is at the target index.
The FTSE WBIG is *not* currency hedged -- ergo, a preferred fund that maps to it shouldn't be hedge, either.
Why money around with all of this rationalization for holding USD-hedge bonds when the more philosophically consistent action that fits with the market portfolio is to hold local currency bonds?
After all, we hold local currency stocks...
In my reports about the Global Stock-and-Bond Portfolio (GSBP), I use the WBIG index fact sheet to estimate the (free-float) market capitalization of investment-grade global bonds. Currency hedging only affects returns; it doesn't affect market capitalization.
In other words, the GSBP's allocation was estimated at 59.90/40.10 global stocks/bonds at the end of March 2021. Currency hedging wasn't involved.
I also estimated the GSBP's February 2021 return to 1.37% using the returns of VT and BNDW, therefore assuming that the currency exposure of ex U.S. bonds was hedged. Had an investor not hedged the currency exposure of international bonds, the return of global bonds would most probably have been more volatile than the return of BNDW.
Last edited by longinvest on Sat Apr 24, 2021 4:09 pm, edited 1 time in total.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
It's a risk portfolio.longinvest wrote: ↑Sat Apr 24, 2021 4:07 pm
In my reports about the Global Stock-and-Bond Portfolio (GSBP), I use the WBIG index fact sheet to estimate the (free-float) market capitalization of investment-grade global bonds. Currency hedging only affects returns; it doesn't affect market capitalization.
In other words, the GSBP's allocation was estimated at 59.90/40.10 global stocks/bonds at the end of March 2021. Currency hedging wasn't involved.
I also estimated the GSBP's February 2021 return to 1.37%, assuming that the currency exposure of ex U.S. bonds was hedged. Had an investor not hedged the currency exposure of international bonds, the return of global bonds would most probably have been more volatile than the return of BNDW.
Why is volatility a problem if you couple it with an LMP (presumably for income), as Sharpe advocates?
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
-
- Posts: 5672
- Joined: Sat Aug 11, 2012 8:44 am
Re: Bill Sharpe's preferred portfolio
Dear Watchnerd,watchnerd wrote: ↑Sat Apr 24, 2021 4:09 pmIt's a risk portfolio.longinvest wrote: ↑Sat Apr 24, 2021 4:07 pm Watchnerd,
In my reports about the Global Stock-and-Bond Portfolio (GSBP), I use the WBIG index fact sheet to estimate the (free-float) market capitalization of investment-grade global bonds. Currency hedging only affects returns; it doesn't affect market capitalization.
In other words, the GSBP's allocation was estimated at 59.90/40.10 global stocks/bonds at the end of March 2021. Currency hedging wasn't involved.
I also estimated the GSBP's February 2021 return to 1.37% using the returns of VT and BNDW, therefore assuming that the currency exposure of ex U.S. bonds was hedged. Had an investor not hedged the currency exposure of international bonds, the return of global bonds would most probably have been more volatile than the return of BNDW.
Why is volatility a problem if you couple it with an LMP (presumably for income), as Sharpe advocates?
We've discussed this in January 2020 in a thread titled "Bill Sharpe World Stock/Bond portfolio, who holds it?". Here's the reply I wrote:
Best regards,longinvest wrote: ↑Sun Jan 12, 2020 5:29 pmYes, I'm (unfortunately) aware of this. I do not agree with the bucket method that William Sharpe is advocating as it's illogical.
I also think that it's illogical to consider TIPS as a "safe asset" and ignore that they're marketable securities with a market capitalization and fluctuating prices; this contradicts the very principles on which his (beautiful and logical) Market Portfolio is based.
But, we're getting off topic.
longinvest
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
Re: Bill Sharpe's preferred portfolio
If you're going to make the Market Portfolio your macro / main portfolio, I can see the need to reduce the risk.longinvest wrote: ↑Sat Apr 24, 2021 4:27 pm
Yes, I'm (unfortunately) aware of this. I do not agree with the bucket method that William Sharpe is advocating as it's illogical.
I also think that it's illogical to consider TIPS as a "safe asset" and ignore that they're marketable securities with a market capitalization and fluctuating prices; this contradicts the very principles on which his (beautiful and logical) Market Portfolio is based.
This is what Vanguard does for exactly that reason.
But if you have a riskless LMP (as the RISMAT series proscribes), I don't see any reason to de-risk the risky Market Portfolio by using currency hedging for the bonds.
This would also be a way to combine Bernstein's "safety first" LMP approach with a Risk Portfolio, in this case the Risk Portfolio being the Market Portfolio.
One the one hand, it could be considered bucketing or mental accounting.
One the other hand, if the LMP is an annuity, it manifests as income and not part of the portfolio in the normal sense, anyway.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
FWIW, here is a backtest comparing USD-hedged vs un-hedged international bonds.
The ratios used are the same as currently indicated by VSMGX.
https://www.portfoliovisualizer.com/bac ... on5_2=12.4
The difference is more minor than one might think.
6 bps difference in CAGR, miniscule differences in Std Dev and Drawdown.
The bigger challenge probably lies in fund selection.
As far as I can tell, there are:
--Lower ER unhedged / local currency sovereign bond funds
--Higher ER unhedged / local currency sovereign + corp bond funds
BNDX may not be as conceptually pure to global market exposure, due to currency hedging, as I would like, but at least it's pretty cheap at ER=.08
The ratios used are the same as currently indicated by VSMGX.
https://www.portfoliovisualizer.com/bac ... on5_2=12.4
The difference is more minor than one might think.
6 bps difference in CAGR, miniscule differences in Std Dev and Drawdown.
The bigger challenge probably lies in fund selection.
As far as I can tell, there are:
--Lower ER unhedged / local currency sovereign bond funds
--Higher ER unhedged / local currency sovereign + corp bond funds
BNDX may not be as conceptually pure to global market exposure, due to currency hedging, as I would like, but at least it's pretty cheap at ER=.08
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
I'm curious, why do the proportions add up a little differently if I'm holding VTI/VXUS/BND/BNDX vs VT/BNDW?djm2001 wrote: ↑Wed Mar 31, 2021 3:30 pm Sharpe's World Bond/Stock portfolio snapshot as of end of day, Mar 31, 2021:
4-fund:2-fund:Code: Select all
INDEX MKT CAP ($M) % VTSAX/VTI $ 40,525,235.96 34.57% VTIAX/VXUS $ 29,929,908.00 25.53% VBTLX/BND $ 23,876,201.64 20.37% VTABX/BNDX $ 22,900,788.96 19.53%
Source: Real-time tracker.Code: Select all
INDEX MKT CAP ($M) % VT $ 69,521,004.83 59.72% BNDW $ 46,900,182.00 40.28%
Methodology:
(Note: I improved the methodology for the 2-fund portfolio since last time based on longinvest's suggestion to use FTSE Global All Cap Index for the VT allocation, and so I'm restating the methodology anew. In subsequent posts, I will simply link to this post to reference the methodology, so long as the methodology remains unchanged.)
Note that VBTLX/BND, VTABX/BNDX, and BNDW actually track Bloomberg Barclay's US Aggregate Bond Index (AGG), Bloomberg Barclay's Global Aggregate ex-USD Float Adjusted RIC Capped Index (USD Hedged) Index, and Bloomberg Barclays Global Aggregate Float Adjusted Composite Index, respectively. However, these indexes do not release factsheets publicly, and so Sharpe suggests using FTSE USBIG, FTSE WorldBIG - USBIG, and FTSE WorldBIG as reasonable approximations.
- The VTSAX/VTI allocation was calculated from the "Index Market Cap" listed in the CRSP US Total Market Index factsheet for Dec 31, 2021; and then forward-projected using the price of VTI then and now.
- The VTIAX/VXUS allocation was calculated from the "Net MCap" listed in the FTSE Global All Cap ex-US factsheet for Feb 26, 2021; and then forward-projected using the price of VXUS then and now.
- The VBTLX/BND allocation was calculated from the "Market Value" listed in the FTSE US Broad Investment-Grade Bond Index (FTSE USBIG) factsheet for Feb 28, 2021, and forward-projected using the price of BND then and now.
- The VTABX/BNDX allocation was calculated from the difference of the "Market Value" fields listed in the FTSE World Broad Investment-Grade Bond Index (FTSE WorldBIG) factsheet and the FTSE USBIG index factsheet, both for Feb 28, 2021; and then forward-projected using the price of BNDX then and now.
- The VT allocation was calculated from the "Net MCap" listed in the FTSE Global All Cap Index factsheet for Feb 26, 2021; and then forward-projected using the price of VT then and now. Note that VT directly tracks the FTSE Global All Cap Index, and so it is more appropriate to use this index's market cap, rather than adding together the market caps for the indexes used by VTSAX/VTI and VTIAX/VXUS.
- The BNDW allocation was calculated from the "Market Value" listed in the FTSE World Broad Investment-Grade Bond Index (FTSE WorldBIG) factsheet for Feb 28, 2021; and then forward-projected using the price of BNDW then and now.
The forward-projection formula is simply:where snapshot_market_cap is taken from the index's latest factsheet, and current_tracker_price and snapshot_tracker_price are the prices of an appopriate ETF tracker (e.g., VTI, VXUS, BND, BNDX, VT, BNDW) as of the current time and the date of the market cap reported on the factsheet.Code: Select all
current_market_cap := snapshot_market_cap * current_tracker_price / snapshot_tracker_price
edit: added mention that the market caps for the 2-fund portfolio are projected forward using VT and BNDW prices.
It's not a big difference between 4 fund and 2 fund (like 20 bps), but I'm curious why they're not identical.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
Excellent question. Thanks for highlighting it. There are two reasons for the discrepancy:
- VT tracks a slightly different index than VTI + VXUS. VT tracks FTSE Global All Cap Index. VTI tracks CRSP US Total Market Index. VXUS tracks FTSE Global All Cap ex-US Index. If VTI tracked FTSE US All Cap Index instead of the CRSP one, then VT's market cap would equal VTI + VXUS market caps... at least, "as of" the factsheet publication dates.
- The forward projections for VT/BNDW portfolio use VT and BNDW prices which can vary slightly from the (weighted sum of) VTI + VXUS and BND + BNDX prices. The reason for those differences are perhaps down to BNDW's internal rebalancing policies.
IMO, the 4-fund portfolio is better for a couple of reasons. First, the CRSP US Total Market Index is more comprehensive that the FTSE US All Cap Index. Second, splitting VT into VTI + VXUS allows you to recoup the foreign tax credit. VT is not eligible for foreign tax credit because of its large US allocation. Also, the expense ratio of VT is (slightly) higher than the weighted combination of expense ratios of VTI + VXUS.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Re: Bill Sharpe's preferred portfolio
Thanks!djm2001 wrote: ↑Wed Apr 28, 2021 7:42 pmExcellent question. Thanks for highlighting it. There are two reasons for the discrepancy:I think the difference is legit per the "buy what you measure, and measure what you buy" approach I talked about earlier. I.e., it is undesirable to calculate the allocation of a fund based on an index different from the one that the fund tracks. (Note that we violate this for bonds because we have no choice -- the Vanguard bond index funds track Bloomberg Barclay indexes, but since those factsheets aren't public, we are forced to use FTSE indexes as an approximation.)
- VT tracks a slightly different index than VTI + VXUS. VT tracks FTSE Global All Cap Index. VTI tracks CRSP US Total Market Index. VXUS tracks FTSE Global All Cap ex-US Index. If VTI tracked FTSE US All Cap Index instead of the CRSP one, then VT's market cap would equal VTI + VXUS market caps... at least, "as of" the factsheet publication dates.
- The forward projections for VT/BNDW portfolio use VT and BNDW prices which can vary slightly from the (weighted sum of) VTI + VXUS and BND + BNDX prices. The reason for those differences are perhaps down to BNDW's internal rebalancing policies.
IMO, the 4-fund portfolio is better for a couple of reasons. First, the CRSP US Total Market Index is more comprehensive that the FTSE US All Cap Index. Second, splitting VT into VTI + VXUS allows you to recoup the foreign tax credit. VT is not eligible for foreign tax credit because of its large US allocation. Also, the expense ratio of VT is (slightly) higher than the weighted combination of expense ratios of VTI + VXUS.
To get even closer to the market portfolio, you can also add in the 0.4% to frontier markets that MSCI I uses.
(that's what I do, combined with VT)
Last edited by watchnerd on Wed Apr 28, 2021 9:52 pm, edited 1 time in total.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
- abuss368
- Posts: 27850
- Joined: Mon Aug 03, 2009 2:33 pm
- Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
- Contact:
Re: Bill Sharpe's preferred portfolio
Essentially a 50% US and International allocation between stocks and bonds. Those differences are rounding errors.CyberBob wrote: ↑Sat Apr 21, 2018 10:46 am The March 2018 end-of-quarter numbers are out for the indexes that Sharpe mentions in the video.
So to keep the portfolio allocations up-to-date for anyone who is interested, here are the numbers (in millions of USD):
27,139,040 - US stocks - CRSP US Total Market Index
24,838,130 - ex US stocks - FTSE Global All-Cap Index minus US
18,890,430 - US bonds - Citigroup US Broad Investment-Grade Index
19,331,390 - ex US bonds - Citigroup World Broad Investment-Grade Index minus US
So that equates to allocation percentages of:
30.09% US stocks
27.54% ex US stocks
20.94% US bonds
21.43% ex US bonds
Tony
John C. Bogle: “Simplicity is the master key to financial success."
- spdoublebass
- Posts: 889
- Joined: Thu Apr 27, 2017 10:04 pm
- Location: NY
Re: Bill Sharpe's preferred portfolio
Just to be clear, the whole point of this thread has nothing to do with the US/Int breakdown currently being a rounding error. It is not the same as holding a 50/50 split because if the US market value grows or declines VT adjusts accordingly. You always own the haystack at market weight.
Yes, currently it is almost 50/50, but that isn't the point.
I'm trying to think, but nothing happens
- abuss368
- Posts: 27850
- Joined: Mon Aug 03, 2009 2:33 pm
- Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!
- Contact:
Re: Bill Sharpe's preferred portfolio
Thanks for the clarification!spdoublebass wrote: ↑Wed Apr 28, 2021 9:10 pmJust to be clear, the whole point of this thread has nothing to do with the US/Int breakdown currently being a rounding error. It is not the same as holding a 50/50 split because if the US market value grows or declines VT adjusts accordingly. You always own the haystack at market weight.
Yes, currently it is almost 50/50, but that isn't the point.
Tony
John C. Bogle: “Simplicity is the master key to financial success."
Re: Bill Sharpe's preferred portfolio
The tricky bit is adjusting the bond/stock ratio.spdoublebass wrote: ↑Wed Apr 28, 2021 9:10 pmJust to be clear, the whole point of this thread has nothing to do with the US/Int breakdown currently being a rounding error. It is not the same as holding a 50/50 split because if the US market value grows or declines VT adjusts accordingly. You always own the haystack at market weight.
Yes, currently it is almost 50/50, but that isn't the point.
I used to think VSMGX would be the perfect lazy solution, but I realized it sticks pretty close to 60/40 by design.
And its US bond / ex-US bond ratio is >2:1
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
You might be double-counting/overweighting over 30% of frontier market holdings. Use the ETF overlap tool to compare FM with VT.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Re: Bill Sharpe's preferred portfolio
Hmmm, I thought VT stopped at emerging markets?djm2001 wrote: ↑Thu Apr 29, 2021 9:22 amYou might be double-counting/overweighting over 30% of frontier market holdings. Use the ETF overlap tool to compare FM with VT.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
I think different indexes (and funds) do not necessarily use the same thresholds or policies to define things, especially across companies (e.g., FTSE vs MSCI), and they do not keep in sync as holdings move across the thresholds. There are similar issues on the bonds side: VWOB has some overlap with BNDX if I recall correctly. And it's much harder to quantify that overlap because the ETF overlap tool does not work for bond funds. Instead, you have to actually find the two funds' holdings and compare manually. (See this post for some other info about overlap between emerging market bond index and "total" bond index.)
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Re: Bill Sharpe's preferred portfolio
Actually, it looks like it's the 'Kuwait problem'.djm2001 wrote: ↑Thu Apr 29, 2021 10:00 amI think different indexes (and funds) do not necessarily use the same thresholds or policies to define things, especially across companies (e.g., FTSE vs MSCI), and they do not keep in sync as holdings move across the thresholds. There are similar issues on the bonds side: VWOB has some overlap with BNDX if I recall correctly. And it's much harder to quantify that overlap because the ETF overlap tool does not work for bond funds. Instead, you have to actually find the two funds' holdings and compare manually. (See this post for some other info about overlap between emerging market bond index and "total" bond index.)
Kuwait, in MSCI land, got promoted from FM to DM. Which would make it consistent with FTSE.
It should have been removed from the FM ETF by iShares, but they haven't done that yet and thus give FM the squishy name of 'Frontier and Select EM'
It's like 0.12% overlap with the rest of the stocks, though, so I'm not too worried about the impact.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
Sharpe's World Bond/Stock portfolio snapshot as of end of day, Apr 30, 2021:
4-fund:
2-fund:
Source: Real-time tracker.
Methodology: Same as before.
4-fund:
Code: Select all
INDEX MKT CAP ($M) %
VTSAX/VTI $ 42,788,112.85 35.52%
VTIAX/VXUS $ 30,846,854.51 25.60%
VBTLX/BND $ 24,492,867.24 20.33%
VTABX/BNDX $ 22,349,275.15 18.55%
Code: Select all
INDEX MKT CAP ($M) %
VT $ 72,712,420.63 60.84%
BNDW $ 46,798,686.81 39.16%
Methodology: Same as before.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Re: Bill Sharpe's preferred portfolio
Thanks!djm2001 wrote: ↑Fri Apr 30, 2021 3:25 pm Sharpe's World Bond/Stock portfolio snapshot as of end of day, Apr 30, 2021:
4-fund:2-fund:Code: Select all
INDEX MKT CAP ($M) % VTSAX/VTI $ 42,788,112.85 35.52% VTIAX/VXUS $ 30,846,854.51 25.60% VBTLX/BND $ 24,492,867.24 20.33% VTABX/BNDX $ 22,349,275.15 18.55%
Source: Real-time tracker.Code: Select all
INDEX MKT CAP ($M) % VT $ 72,712,420.63 60.84% BNDW $ 46,798,686.81 39.16%
Methodology: Same as before.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Re: Bill Sharpe's preferred portfolio
Posting this graph from page 38 of RISMAT 7:
https://web.stanford.edu/~wfsharpe/RISMAT/RISMAT-7.pdf
I hadn't see it here before.
I think the original intent is to show static allocations.
But it also happens to reflect the non-rolling LMP TIPS (and STRIPS) ladder we'll be building as our rising equity glide path.
As the TIPS get chewed down over time, and the equity portion of the glide path rises, the total portfolio of LMP + World Stock Bond will resemble the above graph.
(although, if I'm being picky, the relative risk in the graph should be non-linear given how much more of a risk contribution is made by equities vs bonds)
https://web.stanford.edu/~wfsharpe/RISMAT/RISMAT-7.pdf
I hadn't see it here before.
I think the original intent is to show static allocations.
But it also happens to reflect the non-rolling LMP TIPS (and STRIPS) ladder we'll be building as our rising equity glide path.
As the TIPS get chewed down over time, and the equity portion of the glide path rises, the total portfolio of LMP + World Stock Bond will resemble the above graph.
(although, if I'm being picky, the relative risk in the graph should be non-linear given how much more of a risk contribution is made by equities vs bonds)
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
-
- Posts: 2312
- Joined: Mon Oct 31, 2011 8:36 pm
Re: Bill Sharpe's preferred portfolio
In the past, I'd wondered if you'd seen this; but maybe we have differing interpretations. I've always thought Sharpe intended rebalancing between the market portfolio and TIPS--so the TIPS wouldn't be considered an LMP, but rather a proportioned riskless asset.watchnerd wrote: ↑Fri Apr 30, 2021 9:22 pm Posting this graph from page 38 of RISMAT 7:
https://web.stanford.edu/~wfsharpe/RISMAT/RISMAT-7.pdf
I hadn't see it here before.
I think the original intent is to show static allocations.
But it also happens to reflect the non-rolling LMP TIPS (and STRIPS) ladder we'll be building as our rising equity glide path.
As the TIPS get chewed down over time, and the equity portion of the glide path rises, the total portfolio of LMP + World Stock Bond will resemble the above graph.
(although, if I'm being picky, the relative risk in the graph should be non-linear given how much more of a risk contribution is made by equities vs bonds)
So Sharpe intended hiring a professional to determine the relative percentages, and then you took over from there and did the portfolio management during the remainder of your retirement.