Bill Sharpe's preferred portfolio

[longinvest]

This is probably a question for longinvest, but is there a way to get the historic stock/bond %? Looking at the spreadsheet you've compiled (and thank you & others for this!), it's not clear how to get these numbers.

Would be great for backtesting the allocation changes.

C17d6dcb, you'll find the calculated monthly weightings since December 31, 2020 in my posts. It's easiest to start with my most recent post to find all monthly posts. There's a link from each monthly post to the preceding one.

If you want to go further back than December 2020 with the specific GEISLMS (global stocks) and WBIG (global bonds) indices, I think that the FTSE Russell site publishes fact sheets for these indices since August 2019. You'll have to calculate the weightings from stock and bond market capitalization, as I do in my posts. Note that there's a three months gap in bond fact sheets in 2020, in August, September and October, more precisely. If I remember correctly, bonds weren't sufficiently liquid to calculate WBIG, so its publication was temporarily suspended during these months.

[watchnerd]

Opinions are divided on government bonds with reasonable arguments both ways. But the case for including corporate bonds in WBS is more clear cut. Some portion of earnings goes toward paying corporate bond interest. So you need to be a bondholder as well as a shareholder in order to receive the full public claim on corporate earnings. (In fact, as a shareholder you effectively own part of the liability for the debt with none of the upside, and issuance of new corporate bonds correspondingly decreases share price.) More info in the Components of the Market Portfolio section of Sharpe's RISMAT Chapter 7.

I'm going to highlight this because it's so often overlooked.

That is nice in terms of “academic completeness”, but does it actually matter in practice? A while ago I read the “Little Book of Common Sense Investing” by John Bogle, and in the first chapter, there is a chart that plots total corporate earnings vs total stock return for the largest traded public companies. Over several decades, the total earnings and total return end up at exactly the same place (within 0.01% or something like that). I don’t believe corporate bonds were included in that chart.

The largest, most successful companies (that dominate a total market index) are typically flush with cash and do not need to take on debt to fund their operations. If they do take on any debt, it is only when it is very favorable to the company (and not the bond holders). Thus, my theory is that only smaller or struggling companies have debt payback as a significant use of corporate earnings.

Anyway, while I admit that bond interest are a part of corporate earnings, I’m guessing that it is negligible in the grand scheme of things, and that holding VT alone will match 99% of the total earnings over the long run. It is probably similar to holding VOO vs VTI? I would love to see some data on this point.

Global market cap of IG and HY corp credit to stocks is about 1/4th, the last time I tracked it down.

FWIW, AAPL has about $100B in debt, MSFT $67B.

Yield of IG funds is about 4.9%

So that's the practical implication.

[longinvest]

This post documents the monthly return and asset-class weights as of September 30, 2024 of the (free-float) Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.

The September 2024 return was:

• Global Stock-and-Bond Portfolio: ((62.27% X 2.23% (global stocks)) + (37.73% X 1.28% (global bonds))) = 1.87% (USD)
  
  • Global stocks: Vanguard Total World Stock ETF (VT) NAV return
  • Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return

As of September 30, 2024 the weights were:

• Global stocks: $87,893,959 million USD Market cap -- 62.24%
  
  • FTSE Global All Cap Index (GEISLMS) -- Net MCap
• Global bonds: $53,318,630 million USD Market cap -- 37.76%
  
  • FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value

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 September 2024 return was 1.90%.

Here's a growth chart of the Global Stock-and-Bond Portfolio (and Vanguard's LifeStrategy Moderate Growth Fund (VSMGX)) from December 31, 2020 to September 30, 2024:

[spdoublebass]

I sort of have been going down a rabbit hole in regard to market weight in bonds. Mainly just out of curiosity.

BNDW tracks the FTSE WorldBig index. So I started looking at whats not inside of that index. Basically (as others have pointed out) HY corporate, TIPS, and Muni's.

What I was surprised at is there are relatively cheap ETFs available to invest in these markets including ExUS. (Except Muni's).

All of these market values can be found in the yield book FTSE indices.

I was surprised that while BNDW (or BNDX) includes EM bonds, some EM bonds don't make it into WorldBig. WorldBig includes the index EMGBI, but EMGBI excludes some countries from the smaller indexes it tracks. The individual indexes are AGBI, CEEMEAGBI, and LATAMGBI.
https://www.yieldbook.com/f/m/pdf/ftse_ ... 180730.pdf

I only bring that up because if you look at the smaller indexes and use that for market weight you end up adding some Emerging Market Gov Bond (VWOB) to BNDW.

I know others add HY bonds and TIPS, but why not ExUS as well? ETFs are available PGHY and WIP.

When I did the look at the final numbers, Emerging market bonds and ExUS Inflation bonds were over 1% at market weight and everything else was under 1%. So of course I then wonder is any of this worth it.

What I find most interesting is if you use VT and BNDW you get a market weight of 62/38. If you use all the other bonds you get about 5% less of stock, it's comes out to 57/43.

[exodusing]

But with equities, it does not appear to be the case that XYZ corporation owns its own shares. Rather it is outsiders or employees that have shares tied up.

What about share buy-backs?

When a company buys back its own shares the shares are retired (no longer outstanding) rather than being outstanding and owned by the company.

[watchnerd]

I sort of have been going down a rabbit hole in regard to market weight in bonds. Mainly just out of curiosity.

BNDW tracks the FTSE WorldBig index. So I started looking at whats not inside of that index. Basically (as others have pointed out) HY corporate, TIPS, and Muni's.

What I was surprised at is there are relatively cheap ETFs available to invest in these markets including ExUS. (Except Muni's).

All of these market values can be found in the yield book FTSE indices.

I was surprised that while BNDW (or BNDX) includes EM bonds, some EM bonds don't make it into WorldBig. WorldBig includes the index EMGBI, but EMGBI excludes some countries from the smaller indexes it tracks. The individual indexes are AGBI, CEEMEAGBI, and LATAMGBI.
https://www.yieldbook.com/f/m/pdf/ftse_ ... 180730.pdf

I only bring that up because if you look at the smaller indexes and use that for market weight you end up adding some Emerging Market Gov Bond (VWOB) to BNDW.

I know others add HY bonds and TIPS, but why not ExUS as well? ETFs are available PGHY and WIP.

When I did the look at the final numbers, Emerging market bonds and ExUS Inflation bonds were over 1% at market weight and everything else was under 1%. So of course I then wonder is any of this worth it.

What I find most interesting is if you use VT and BNDW you get a market weight of 62/38. If you use all the other bonds you get about 5% less of stock, it's comes out to 57/43.

I actually exclude Treasuries from my risk port because I have so many TIPS that I'd be massively overweight in US Treasuries if I also included them in my risk port.

So that means my risk port has everything else: corporates, high yield, etc.

To calculate that, I have to manually combine:

https://research.ftserussell.com/Analyt ... anual=True

https://research.ftserussell.com/Analyt ... anual=True

[spdoublebass]

I sort of have been going down a rabbit hole in regard to market weight in bonds. Mainly just out of curiosity.

BNDW tracks the FTSE WorldBig index. So I started looking at whats not inside of that index. Basically (as others have pointed out) HY corporate, TIPS, and Muni's.

What I was surprised at is there are relatively cheap ETFs available to invest in these markets including ExUS. (Except Muni's).

All of these market values can be found in the yield book FTSE indices.

I was surprised that while BNDW (or BNDX) includes EM bonds, some EM bonds don't make it into WorldBig. WorldBig includes the index EMGBI, but EMGBI excludes some countries from the smaller indexes it tracks. The individual indexes are AGBI, CEEMEAGBI, and LATAMGBI.
https://www.yieldbook.com/f/m/pdf/ftse_ ... 180730.pdf

I only bring that up because if you look at the smaller indexes and use that for market weight you end up adding some Emerging Market Gov Bond (VWOB) to BNDW.

I know others add HY bonds and TIPS, but why not ExUS as well? ETFs are available PGHY and WIP.

When I did the look at the final numbers, Emerging market bonds and ExUS Inflation bonds were over 1% at market weight and everything else was under 1%. So of course I then wonder is any of this worth it.

What I find most interesting is if you use VT and BNDW you get a market weight of 62/38. If you use all the other bonds you get about 5% less of stock, it's comes out to 57/43.

I actually exclude Treasuries from my risk port because I have so many TIPS that I'd be massively overweight in US Treasuries if I also included them in my risk port.

So that means my risk port has everything else: corporates, high yield, etc.

To calculate that, I have to manually combine:

https://research.ftserussell.com/Analyt ... anual=True

https://research.ftserussell.com/Analyt ... anual=True

Nice.
Do you invest in both US and ExUs HY?
Do you hold any ExUS Inflation bonds?

It would be really nice if there was an accurate overlap tool for Bond ETFs.

Again, much of this doesn't really matter, but I just find it interesting. I never was one to toy or tinker with my portfolio, but I always followed the market portfolio and updated a spreadsheet to get the current market value, maybe I like it cause it gives me something to do.

[aj76er]

That is nice in terms of “academic completeness”, but does it actually matter in practice? A while ago I read the “Little Book of Common Sense Investing” by John Bogle, and in the first chapter, there is a chart that plots total corporate earnings vs total stock return for the largest traded public companies. Over several decades, the total earnings and total return end up at exactly the same place (within 0.01% or something like that). I don’t believe corporate bonds were included in that chart.

The largest, most successful companies (that dominate a total market index) are typically flush with cash and do not need to take on debt to fund their operations. If they do take on any debt, it is only when it is very favorable to the company (and not the bond holders). Thus, my theory is that only smaller or struggling companies have debt payback as a significant use of corporate earnings.

Anyway, while I admit that bond interest are a part of corporate earnings, I’m guessing that it is negligible in the grand scheme of things, and that holding VT alone will match 99% of the total earnings over the long run. It is probably similar to holding VOO vs VTI? I would love to see some data on this point.

Global market cap of IG and HY corp credit to stocks is about 1/4th, the last time I tracked it down.

FWIW, AAPL has about $100B in debt, MSFT $67B.

Yield of IG funds is about 4.9%

So that's the practical implication.

Just to circle back to this, AAPL and MSFT have a combined market cap close to $7T. So $167B of combined debt is very much a rounding error. For example AAPL could probably pay off its debt with 1 quarter of earnings.

So my guess is that over long periods (e.g. decades), investing only in VT (at >80% market cap relative to corporate bonds) gives you the majority of cumulative corporate earnings (like in the high 90%’s).

Having said all that, I see the appeal of inclusion of global credit to capture all of the world’s corporate earnings, assuming you have enough tax sheltered space for it (as the high dividends are not very tax efficient). FWIW, I recently discovered that DFA has an ETF that tracks global credit so that would be a viable option for those that can’t reasonably access the Vanguard fund.

[watchnerd]

I actually exclude Treasuries from my risk port because I have so many TIPS that I'd be massively overweight in US Treasuries if I also included them in my risk port.

So that means my risk port has everything else: corporates, high yield, etc.

To calculate that, I have to manually combine:

https://research.ftserussell.com/Analyt ... anual=True

https://research.ftserussell.com/Analyt ... anual=True

Nice.
Do you invest in both US and ExUs HY?
Do you hold any ExUS Inflation bonds?

It would be really nice if there was an accurate overlap tool for Bond ETFs.

Again, much of this doesn't really matter, but I just find it interesting. I never was one to toy or tinker with my portfolio, but I always followed the market portfolio and updated a spreadsheet to get the current market value, maybe I like it cause it gives me something to do.

I hold both US and Ex-US HY.

I don't hold any ex-US inflation indexed bonds. I'm not even sure if there is a fund that is available to US investors that covers that.

[watchnerd]

Global market cap of IG and HY corp credit to stocks is about 1/4th, the last time I tracked it down.

FWIW, AAPL has about $100B in debt, MSFT $67B.

Yield of IG funds is about 4.9%

So that's the practical implication.

Just to circle back to this, AAPL and MSFT have a combined market cap close to $7T. So $167B of combined debt is very much a rounding error. For example AAPL could probably pay off its debt with 1 quarter of earnings.

So my guess is that over long periods (e.g. decades), investing only in VT (at >80% market cap relative to corporate bonds) gives you the majority of cumulative corporate earnings (like in the high 90%’s).

Having said all that, I see the appeal of inclusion of global credit to capture all of the world’s corporate earnings, assuming you have enough tax sheltered space for it (as the high dividends are not very tax efficient). FWIW, I recently discovered that DFA has an ETF that tracks global credit so that would be a viable option for those that can’t reasonably access the Vanguard fund.

If you look at the FTSE data linked above for global IG and HY credit, it's about 1/4th the market cap of VT.

So it's more than a rounding error, as a whole, but still definitely getting into the realm of pedantic perfectionism / intellectual appeal.

[spdoublebass]

I hold both US and Ex-US HY.

I don't hold any ex-US inflation indexed bonds. I'm not even sure if there is a fund that is available to US investors that covers that.

What fund do you use for ExUS HY? The ETF PGHY?

There is a fund for ExUS Inflation linked bonds with the ticker WIP. But it does have an er of .5%.
https://www.ssga.com/us/en/individual/e ... nd-etf-wip

[watchnerd]

I hold both US and Ex-US HY.

I don't hold any ex-US inflation indexed bonds. I'm not even sure if there is a fund that is available to US investors that covers that.

What fund do you use for ExUS HY? The ETF PGHY?

There is a fund for ExUS Inflation linked bonds with the ticker WIP. But it does have an er of .5%.
https://www.ssga.com/us/en/individual/e ... nd-etf-wip

I could micromanage it by picking a dedicated ExUS HY, but at 1.62% of total risk port weight, I just consider it "covered" via the lower credit BB-grade holdings of VGCAX.

[Trance]

This post documents the monthly return and asset-class weights as of September 30, 2024 of the (free-float) Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.

The September 2024 return was:

• Global Stock-and-Bond Portfolio: ((62.27% X 2.23% (global stocks)) + (37.73% X 1.28% (global bonds))) = 1.87% (USD)
  
  • Global stocks: Vanguard Total World Stock ETF (VT) NAV return
  • Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return

As of September 30, 2024 the weights were:

• Global stocks: $87,893,959 million USD Market cap -- 62.24%
  
  • FTSE Global All Cap Index (GEISLMS) -- Net MCap
• Global bonds: $53,318,630 million USD Market cap -- 37.76%
  
  • FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value

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 September 2024 return was 1.90%.

Here's a growth chart of the Global Stock-and-Bond Portfolio (and Vanguard's LifeStrategy Moderate Growth Fund (VSMGX)) from December 31, 2020 to September 30, 2024:

With each year the emerging market becomes a bigger share of the global pie for bonds. I wonder if Vanguard will ever add VWOB to their total etf funds or make a more comprehensive etf

[guppyguy]

Just curious if anybody is including gold and/or btc with WBS and what methodology they are using to determine market weights? Or if your not using market weights, what is your justification of whatever weighting you decided upon?

[watchnerd]

Just curious if anybody is including gold and/or btc with WBS and what methodology they are using to determine market weights? Or if your not using market weights, what is your justification of whatever weighting you decided upon?

See my signature, I use market weight.

That's all I'll say about it due to forum policies.

[guppyguy]

Watchnerd, would you share the funds you use for HY/IG exposure and how you estimate their mkt cap share?

djm, one area where it seems you and watchnerd disagree is whether to include government bonds in the WBS. Looks like you have them. I am wondering if there is a definitive argument either way.

thanks

The source for IG credit is the same as above, the FTSE World BIG index:

https://research.ftserussell.com/Analyt ... anual=True

I use the corporate and covered line items.

High yield is FTSE World High-Yield index:

https://research.ftserussell.com/Analyt ... anual=True

The fund I use is VGCAX, which isn't a perfect proxy for either of these indexes, but I haven't yet found anything better, and the weightings are "directionally correct" (mostly corporate, mostly IG, some HY, roughly half US, USD hedged):

https://investor.vanguard.com/investmen ... file/vgcax

Just curious if you have considered DGCB in leiu of VGCAX.

I also already have way too much in TIPS/T bills to justify BNDW inclusion of treasuries.

[watchnerd]

The source for IG credit is the same as above, the FTSE World BIG index:

https://research.ftserussell.com/Analyt ... anual=True

I use the corporate and covered line items.

High yield is FTSE World High-Yield index:

https://research.ftserussell.com/Analyt ... anual=True

The fund I use is VGCAX, which isn't a perfect proxy for either of these indexes, but I haven't yet found anything better, and the weightings are "directionally correct" (mostly corporate, mostly IG, some HY, roughly half US, USD hedged):

https://investor.vanguard.com/investmen ... file/vgcax

Just curious if you have considered DGCB in leiu of VGCAX.

I also already have way too much in TIPS/T bills to justify BNDW inclusion of treasuries.

DGCB didn't exist when I bought VGCAX.

It's only a year old (inception 11/07/2023).

They're pretty similar, so it's not obvious why DGCB gave 3.68% over the last year vs VGCAX giving 4.11%.

[guppyguy]

I actually exclude Treasuries from my risk port because I have so many TIPS that I'd be massively overweight in US Treasuries if I also included them in my risk port.

So that means my risk port has everything else: corporates, high yield, etc.

To calculate that, I have to manually combine:

https://research.ftserussell.com/Analyt ... anual=True

https://research.ftserussell.com/Analyt ... anual=True

Maybe an obvious question, are you stripping out everything except Corporate in the WorldBIG and adding in the World HY market value?

I'm just trying to figure out how some here are getting to the 75/25 approximate ratio of world equity/IGHY rather than the full WBS.

(Unless I'm missing the calculation in the GMP spreadsheet somewhere).

[watchnerd]

I actually exclude Treasuries from my risk port because I have so many TIPS that I'd be massively overweight in US Treasuries if I also included them in my risk port.

So that means my risk port has everything else: corporates, high yield, etc.

To calculate that, I have to manually combine:

https://research.ftserussell.com/Analyt ... anual=True

https://research.ftserussell.com/Analyt ... anual=True

Maybe an obvious question, are you stripping out everything except Corporate in the WorldBIG and adding in the World HY market value?

I'm just trying to figure out how some here are getting to the 75/25 approximate ratio of world equity/IGHY rather than the full WBS.

(Unless I'm missing the calculation in the GMP spreadsheet somewhere).

Close, yes. I add in Covered line item, too.

Corporate (10T) + Covered (8T) + World HY total (1.7T)

Why include Covered?

They're (sort of) the flip side of REITs that are in the stock index.

[guppyguy]

Maybe an obvious question, are you stripping out everything except Corporate in the WorldBIG and adding in the World HY market value?

I'm just trying to figure out how some here are getting to the 75/25 approximate ratio of world equity/IGHY rather than the full WBS.

(Unless I'm missing the calculation in the GMP spreadsheet somewhere).

Close, yes. I add in Covered line item, too.

Corporate (10T) + Covered (8T) + World HY total (1.7T)

Why include Covered?

They're (sort of) the flip side of REITs that are in the stock index.

Ah, got it....learned something new. That jives more with your rounded signature proportions as well. Thanks.

[guppyguy]

What do most people assume for future real returns of the GMP and how do you come up with that number?

I'm trying to project out contributions/balances.

[watchnerd]

What do most people assume for future real returns of the GMP and how do you come up with that number?

I'm trying to project out contributions/balances.

I've used the Vanguard 10 YR forecasts

[djm2001]

Here's a summary of ways I considered estimating returns: https://bogleheads.org/forum/viewtopic. ... 1#p7645761

For my own planning, I have converged on using 10-year TIPS real yield minus its liquidity premium. My assumption is that expected 10-year yields for everything else should be higher since TIPS have less risk, and so this must be a conservative estimate.

10-year TIPS real yield:
https://fred.stlouisfed.org/graph/fredg ... 9999-01-17

10-year TIPS liquidity premium (see Column O):
https://www.federalreserve.gov/econres/ ... pdates.csv

For most planning you want a lower estimate, but for planning Roth conversions you don't want to underestimate growth in tax deferred accounts, so I use a small boost for that.

[aj76er]

Here's a summary of ways I considered estimating returns: https://bogleheads.org/forum/viewtopic. ... 1#p7645761

For my own planning, I have converged on using 10-year TIPS real yield minus its liquidity premium. My assumption is that expected 10-year yields for everything else should be higher since TIPS have less risk, and so this must be a conservative estimate.

10-year TIPS real yield:
https://fred.stlouisfed.org/graph/fredg ... 9999-01-17

10-year TIPS liquidity premium (see Column O):
https://www.federalreserve.gov/econres/ ... pdates.csv

For most planning you want a lower estimate, but for planning Roth conversions you don't want to underestimate growth in tax deferred accounts, so I use a small boost for that.

What do you do if the 10yr TIPS has a negative real yield? Do you set a floor at 0% or 1%?

[watchnerd]

My assumption is that expected 10-year yields for everything else should be higher since TIPS have less risk, and so this must be a conservative estimate.

This seems like an odd assumption, given we have had negative equity risk premiums in the past and also negative real returns for stocks in the past.

[djm2001]

What do you do if the 10yr TIPS has a negative real yield? Do you set a floor at 0% or 1%?

I'd just use the negative number. It's possible for expected growth to be negative. (My hope would be that current price would be overinflated enough to offset that.)

[djm2001]

This seems like an odd assumption, given we have had negative equity risk premiums in the past and also negative real returns for stocks in the past.

What's the contradiction with negative real returns? As for negative equity risk premiums, there is a thread where several posters question the definition and measurement of the claimed negative equity premium.

[watchnerd]

This seems like an odd assumption, given we have had negative equity risk premiums in the past and also negative real returns for stocks in the past.

What's the contradiction with negative real returns? As for negative equity risk premiums, there is a thread where several posters question the definition and measurement of the claimed negative equity premium.

The challenge is that we've had periods where bonds had positive real returns and stocks had negative real returns.

So assuming that stocks must have a positive real return above bonds seems problematic, as there must be (and have been in the past) periods where stocks can lose to bonds else they wouldn't have a risk premium.

[djm2001]

What's the contradiction with negative real returns? As for negative equity risk premiums, there is a thread where several posters question the definition and measurement of the claimed negative equity premium.

The challenge is that we've had periods where bonds had positive real returns and stocks had negative real returns.

So assuming that stocks must have a positive real return above bonds seems problematic, as there must be (and have been in the past) periods where stocks can lose to bonds else they wouldn't have a risk premium.

I'm not saying TIPS haven't beaten stocks in the past. And I'm not saying that TIPS won't beat stock in the future. I'm saying that a priori, expected yield of TIPS should be lower than the expected yield of anything riskier (including stocks). (The actual outcome can differ from the expect value, of course, and that's why a plan should be conservative.)

I'm not sure if it's possible to conclusively demonstrate this empirically unless ex ante expected real returns of stocks can be accurately measured, and I don't know of a way to do that. (Perhaps with some kind of derivative instrument?)

The first thing I'd question when presented with negative equity premium claims before believing them is whether the equity premium measurement methodology was ex ante or ex post.

[watchnerd]

The challenge is that we've had periods where bonds had positive real returns and stocks had negative real returns.

So assuming that stocks must have a positive real return above bonds seems problematic, as there must be (and have been in the past) periods where stocks can lose to bonds else they wouldn't have a risk premium.

I'm not saying TIPS haven't beaten stocks in the past. And I'm not saying that TIPS won't beat stock in the future. I'm saying that a priori, expected yield of TIPS should be lower than the expected yield of anything riskier (including stocks). (The actual outcome can differ from the expect value, of course, and that's why a plan should be conservative.)

I'm not sure if it's possible to conclusively demonstrate this empirically unless ex ante expected real returns of stocks can be accurately measured, and I don't know of a way to do that. (Perhaps with some kind of derivative instrument?)

The first thing I'd question when presented with negative equity premium claims before believing them is whether the equity premium measurement methodology was ex ante or ex post.

How are you deciding how much of a spread to give to equities vs TIPS?

Because right now, the Vanguard's 10 YR forecast of US equities vs TIPS (nominal) is:

U.S. equities 3.2%–5.2%
U.S. Treasury Inflation-Protected Securities 3.5%–4.5%

https://advisors.vanguard.com/insights/ ... ed-returns

[djm2001]

The original question that I was replying to was about expected return, not variance. In my own plan, I don't incorporate variance per se. Instead, I use a conservative estimate for expected real yield over a 10-year horizon (updated continuously) and hope that the plan's withdrawal recommendation course corrects gradually over my (hopefully) much longer retirement. Obviously some flexibility is needed, but I'm comfortable with this since my actual spend is lower than my plan dictates.

To illustrate with concrete numbers... Current 10yr TIPS real yield is about 2%. After haircuts (trailing 90d liquidity premium of 10yr TIPS), I get a real yield of 1.5%. I am 45 and early retired. So my simplistic recommended withdrawal rate (see my signature) is PMT(1.5%, 100 - 45, 1) = 2.6% roughly. This is more conservative than VPW or the 4% rule or even Ben Felix's 2.7% rule. Moreover, it doesn't factor in Social Security. So I feel like with such a conservative plan, I don't need to worry about variance.

(I'll caveat that my actual withdrawals in real life are governed more by dividends and emergency spending needs than by my plan's recommended withdrawal. So far, my withdrawals have been less than my plan's recommendation.)

[watchnerd]

The original question that I was replying to was about expected return, not variance. In my own plan, I don't incorporate variance per se. Instead, I use a conservative estimate for expected real yield over a 10-year horizon (updated continuously) and hope that the plan's withdrawal recommendation course corrects gradually over my (hopefully) much longer retirement. Obviously some flexibility is needed, but I'm comfortable with this since my actual spend is lower than my plan dictates.

To illustrate with concrete numbers... Current 10yr TIPS real yield is about 2%. After haircuts (trailing 90d liquidity premium of 10yr TIPS), I get a real yield of 1.5%. I am 45 and early retired. So my simplistic recommended withdrawal rate (see my signature) is PMT(1.5%, 100 - 45, 1) = 2.6% roughly. This is more conservative than VPW or the 4% rule or even Ben Felix's 2.7% rule. Moreover, it doesn't factor in Social Security. So I feel like with such a conservative plan, I don't need to worry about variance.

(I'll caveat that my actual withdrawals in real life are governed more by dividends and emergency spending needs than by my plan's recommended withdrawal. So far, my withdrawals have been less than my plan's recommendation.)

Gotcha (I think).

Although this is getting waaaay off topic for this thread, so I won't ask further questions.

[longinvest]

This post documents the monthly return and asset-class weights as of October 31, 2024 of the (free-float) Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.

The October 2024 return was:

• Global Stock-and-Bond Portfolio: ((62.24% X -2.26% (global stocks)) + (37.76% X -1.60% (global bonds))) = -2.01% (USD)
  
  • Global stocks: Vanguard Total World Stock ETF (VT) NAV return
  • Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return

As of October 31, 2024 the weights were:

• Global stocks: $85,775,590 million USD Market cap -- 62.28%
  
  • FTSE Global All Cap Index (GEISLMS) -- Net MCap
• Global bonds: $51,954,710 million USD Market cap -- 37.72%
  
  • FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value

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 October 2024 return was -2.19%.

Here's a growth chart of the Global Stock-and-Bond Portfolio (and Vanguard's LifeStrategy Moderate Growth Fund (VSMGX)) from December 31, 2020 to October 31, 2024:

[guppyguy]

The original question that I was replying to was about expected return, not variance. In my own plan, I don't incorporate variance per se. Instead, I use a conservative estimate for expected real yield over a 10-year horizon (updated continuously) and hope that the plan's withdrawal recommendation course corrects gradually over my (hopefully) much longer retirement. Obviously some flexibility is needed, but I'm comfortable with this since my actual spend is lower than my plan dictates.

To illustrate with concrete numbers... Current 10yr TIPS real yield is about 2%. After haircuts (trailing 90d liquidity premium of 10yr TIPS), I get a real yield of 1.5%. I am 45 and early retired. So my simplistic recommended withdrawal rate (see my signature) is PMT(1.5%, 100 - 45, 1) = 2.6% roughly. This is more conservative than VPW or the 4% rule or even Ben Felix's 2.7% rule. Moreover, it doesn't factor in Social Security. So I feel like with such a conservative plan, I don't need to worry about variance.

(I'll caveat that my actual withdrawals in real life are governed more by dividends and emergency spending needs than by my plan's recommended withdrawal. So far, my withdrawals have been less than my plan's recommendation.)

Gotcha (I think).

Although this is getting waaaay off topic for this thread, so I won't ask further questions.

I should have given context to my question. We utilize a Risk Portfolio (GMP) + LMP (TIPS ladder). We've won the game and the LMP is fully funded, but have 10 more years of working. As we don't maintain a fixed allocation between the RP and LMP, the question is where to put additional savings.

I hadn't seen the Vanguard estimates, thanks! I didn't realize the expected returns between the two were so close.

[watchnerd]

Gotcha (I think).

Although this is getting waaaay off topic for this thread, so I won't ask further questions.

I should have given context to my question. We utilize a Risk Portfolio (GMP) + LMP (TIPS ladder). We've won the game and the LMP is fully funded, but have 10 more years of working. As we don't maintain a fixed allocation between the RP and LMP, the question is where to put additional savings.

I hadn't seen the Vanguard estimates, thanks! I didn't realize the expected returns between the two were so close.

I use TPAW to model where to put additional savings.

[aj76er]

So my simplistic recommended withdrawal rate (see my signature) is PMT(1.5%, 100 - 45, 1) = 2.6% roughly.

2.6% is approximately the current dividend yield of the WBS (i.e. looking at yields from 60% VT / 40% BNDW). Of course, yields can go down so no guarantees.

But anyway, having a plan to only spend dividends without considering any capital depletion and/or social security is quite conservative.

[djm2001]

So my simplistic recommended withdrawal rate (see my signature) is PMT(1.5%, 100 - 45, 1) = 2.6% roughly.

2.6% is approximately the current dividend yield of the WBS.

That is interesting to know and gels with my experience -- I pretty much live off the dividends of WBS in practice. But note that the PMT withdrawal rate formula above is age-dependent. E g., it's roughly 3.6% for a 65 year old.

[bmstrong]

This post documents the monthly return and asset-class weights as of March 31, 2024 of the (free-float) Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.

The March 2024 return was:

• Global Stock-and-Bond Portfolio: ((62.08% X 3.15% (global stocks)) + (37.92% X 0.98% (global bonds))) = 2.33% (USD)
  
  • Global stocks: Vanguard Total World Stock ETF (VT) NAV return
  • Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return

As of March 31, 2024 the weights were:

• Global stocks: $80,791,615 million USD Market cap -- 62.46%
  
  • FTSE Global All Cap Index (GEISLMS) -- Net MCap
• Global bonds: $48,554,930 million USD Market cap -- 37.54%
  
  • FTSE World Broad Investment-Grade Bond Index (WBIG) -- Market Value

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 2024 return was 2.25%.

Here's a growth chart of the Global Stock-and-Bond Portfolio (and Vanguard's LifeStrategy Moderate Growth Fund (VSMGX)) from December 31, 2020 to March 31, 2024:

Appreciate these update posts. I've been running this portfolio for a year or so now. I'm trying to settle on the TIPS ETF for the next 20-30 years to hold. Hopefully, we'll see a year end post as well?

[longinvest]

Appreciate these update posts. I've been running this portfolio for a year or so now. I'm trying to settle on the TIPS ETF for the next 20-30 years to hold. Hopefully, we'll see a year end post as well?

Bmstrong, I post monthly updates. Here's a link to last update (October 2024). The December 2024 update should be posted during the month of January 2025.

Instead of trying to perfectly replicate Sharpe's theoretical Market Portfolio, I invest into a globally-diversified all-in-one balanced index ETF similar to Vanguard's LifeStrategy Moderate Growth fund, but with a different home bias. I find this close enough. The simplicity has many advantages discussed in the One-Fund Portfolio thread.

[209south]

The original question that I was replying to was about expected return, not variance. In my own plan, I don't incorporate variance per se. Instead, I use a conservative estimate for expected real yield over a 10-year horizon (updated continuously) and hope that the plan's withdrawal recommendation course corrects gradually over my (hopefully) much longer retirement. Obviously some flexibility is needed, but I'm comfortable with this since my actual spend is lower than my plan dictates.

To illustrate with concrete numbers... Current 10yr TIPS real yield is about 2%. After haircuts (trailing 90d liquidity premium of 10yr TIPS), I get a real yield of 1.5%. I am 45 and early retired. So my simplistic recommended withdrawal rate (see my signature) is PMT(1.5%, 100 - 45, 1) = 2.6% roughly. This is more conservative than VPW or the 4% rule or even Ben Felix's 2.7% rule. Moreover, it doesn't factor in Social Security. So I feel like with such a conservative plan, I don't need to worry about variance.

(I'll caveat that my actual withdrawals in real life are governed more by dividends and emergency spending needs than by my plan's recommended withdrawal. So far, my withdrawals have been less than my plan's recommendation.)

I own a LOT of TIPs - can you please help me with this >50bp haircut for liquidity? I'm struggling trying to understand that. I understand TIPs are less liquid than nominal Treasuries, but that might explain 1bp!

[djm2001]

The original question that I was replying to was about expected return, not variance. In my own plan, I don't incorporate variance per se. Instead, I use a conservative estimate for expected real yield over a 10-year horizon (updated continuously) and hope that the plan's withdrawal recommendation course corrects gradually over my (hopefully) much longer retirement. Obviously some flexibility is needed, but I'm comfortable with this since my actual spend is lower than my plan dictates.

To illustrate with concrete numbers... Current 10yr TIPS real yield is about 2%. After haircuts (trailing 90d liquidity premium of 10yr TIPS), I get a real yield of 1.5%. I am 45 and early retired. So my simplistic recommended withdrawal rate (see my signature) is PMT(1.5%, 100 - 45, 1) = 2.6% roughly. This is more conservative than VPW or the 4% rule or even Ben Felix's 2.7% rule. Moreover, it doesn't factor in Social Security. So I feel like with such a conservative plan, I don't need to worry about variance.

(I'll caveat that my actual withdrawals in real life are governed more by dividends and emergency spending needs than by my plan's recommended withdrawal. So far, my withdrawals have been less than my plan's recommendation.)

I own a LOT of TIPs - can you please help me with this >50bp haircut for liquidity? I'm struggling trying to understand that. I understand TIPs are less liquid than nominal Treasuries, but that might explain 1bp!

I'm using the DKW inflation model from the Federal Reserve. The data is published monthly as a staff research project. There's a direct link to the data in one of my recent posts above, or you can get it from the Data Release section of an article published by the authors of the model here: https://www.federalreserve.gov/econres/ ... 90521.html (The 10 year TIPS liquidity premium is listed in column O if you import the csv into a spreadsheet.)

I am not confident in my understanding of how they calculate the liquidity premium in the DKW model, but perhaps forum member Kevin M has posted an explanation at some point. Here's one of his posts from a couple of years ago when the premium was 0.40%: viewtopic.php?p=6910489#p6910489

That said, you shouldn't worry if you hold TIPS to maturity. You're benefitting from a positive liquidity premium in that case. The reason I subtract it out is because I want to apply the real yield (see equation 3 of the DKW article I linked above, i.e., TIPS yield = real yield + TIPS liquidity premium) to non-TIPS assets that don't benefit from the TIPS liquidity premium, essentially as a lower bound on the expected yield of my entire (WBS) portfolio.

[Newt888]

This is probably a question for longinvest, but is there a way to get the historic stock/bond %? Looking at the spreadsheet you've compiled (and thank you & others for this!), it's not clear how to get these numbers.

Would be great for backtesting the allocation changes.

You may be looking for page 22 of this paper: https://papers.ssrn.com/sol3/papers.cfm ... id=2978509