Bill Sharpe's preferred portfolio

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

pascalwager wrote: Fri Apr 30, 2021 10:33 pm
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.

Image

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)
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.

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.
Yes, I definitely recall him writing about something like you're describing, and I remember thinking how much I disliked it as it smacked of buckets. ;)

But I could swear he also talked about a LMP ladder in something else I either read or saw a video about.

Or maybe I'm just cherry picking subconsciously between what I like about Bernstein and Sharpe and mashing them together. ;)
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
HootingSloth
Posts: 1041
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

watchnerd wrote: Fri Apr 30, 2021 10:50 pm
pascalwager wrote: Fri Apr 30, 2021 10:33 pm 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.

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.
Yes, I definitely recall him writing about something like you're describing, and I remember thinking how much I disliked it as it smacked of buckets. ;)

But I could swear he also talked about a LMP ladder in something else I either read or saw a video about.

Or maybe I'm just cherry picking subconsciously between what I like about Bernstein and Sharpe and mashing them together. ;)
I thought Bill Sharpe had a (seemingly idiosyncratic) "lockbox" approach that more or less works by dividing the portfolio into different accounts for each year of consumption in the future. Into each lockbox, you place a certain amount of TIPS plus a certain amount of stocks/bonds in the risk portfolio (plus, in his ideal world, a certain amount of theoretical assets that you can't actually buy that he calls "m-shares" and that I haven't really tried to understand). You leave each lockbox alone until you get to the year it is assigned, and then you spend whatever ends up in the lockbox. It is described in chapter 15, here: https://web.stanford.edu/~wfsharpe/RISMAT/
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

HootingSloth wrote: Sat May 01, 2021 8:18 am

I thought Bill Sharpe had a (seemingly idiosyncratic) "lockbox" approach that more or less works by dividing the portfolio into different accounts for each year of consumption in the future. Into each lockbox, you place a certain amount of TIPS plus a certain amount of stocks/bonds in the risk portfolio (plus, in his ideal world, a certain amount of theoretical assets that you can't actually buy that he calls "m-shares" and that I haven't really tried to understand). You leave each lockbox alone until you get to the year it is assigned, and then you spend whatever ends up in the lockbox. It is described in chapter 15, here: https://web.stanford.edu/~wfsharpe/RISMAT/
Thanks -- I'm re-reading the whole RISMAT series so will get to chapter 15 soon.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

HootingSloth wrote: Sat May 01, 2021 8:18 am I thought Bill Sharpe had a (seemingly idiosyncratic) "lockbox" approach that more or less works by dividing the portfolio into different accounts for each year of consumption in the future. Into each lockbox, you place a certain amount of TIPS plus a certain amount of stocks/bonds in the risk portfolio (plus, in his ideal world, a certain amount of theoretical assets that you can't actually buy that he calls "m-shares" and that I haven't really tried to understand). You leave each lockbox alone until you get to the year it is assigned, and then you spend whatever ends up in the lockbox. It is described in chapter 15, here: https://web.stanford.edu/~wfsharpe/RISMAT/


To me, the lockbox sounds like a weird LMP portfolio where each lockbox is similar to a ladder, but has the following ingredients:
1. Zero-coupon TIPS maturing in 9 years
2. World Bond/Stock Mutual Fund or ETF Shares to be sold in 9 years
3. m-Shares maturing in 9 years
#1 doesn't actually exist; there are no zero coupon TIPS.

I half fake it, though, by using alternating rungs of TIPS and STRIPS in a bond ladder.

#2 Exists

#3 Does not exist


Chapter 15 is probably the only chapter in the RISMAT series where I feel Sharpe jumped the shark between applicable theory and just pure mental masturbation.

I think he sort of admits this, as he says:
The conclusion is that absent the availability of suitable m-shares, it is impossible to obtain a portfolio optimal for an investor with a constant relative risk aversion marginal utility function with risk aversion that differs from that priced in the market portfolio. That said, it is possible
to choose a combination of TIPS and the market portfolio for one year, find the implied marginal utility function, then find combinations for subsequent years that are optimal for the same marginal utility.
He then gives some utility functions for approximating the lockbox contents with TIPS and WSB shares.

At this point, I don't see the point in trying make a model to determine how many WSB shares to allocate a 'lockbox', as opposed to just withdrawing from the Risk Portfolio at a future date according to needs.

It feels tortured, and not solving a genuine problem, given that m-shares don't exist.

He then ends the chapter with:
Absent the availability of m-shares, retirees can either need to choose retirement income strategies that provide payments with significantly different probability distributions of income at future dates, adopt an approach consistent with marginal utility functions of future income
that differ substantially, or select some combination of the two approaches. Moreover, given the fact that most retirees will receive fixed real payments from Social Security or some other sort of defined benefit plan, it will be important take into account all sources of income. The
remaining chapters explore some of these implications in detail for the construction of strategies for providing income with and without insuring against longevity risk. In both contexts, lockboxes can play a prominent role.
I my case, I will have an income strategy that provide payments with significantly different probability distributions at future dates:

LMP TIPS/STRIPS ladder (mixture of fixed real and nominal payments)
Social Security (fixed real payments)
Risk Portfolio (variable distribution)

One could also swap in an annuity en lieu of the LMP ladder.

Given that m-shares don't exist, without which lockboxes can't really be implemented as envisioned, I'm going to consider LMP + WSB RP as 'close enough'. ;)
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
HootingSloth
Posts: 1041
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

watchnerd wrote: Sat May 01, 2021 6:47 pm I my case, I will have an income strategy that provide payments with significantly different probability distributions at future dates:

LMP TIPS/STRIPS ladder (mixture of fixed real and nominal payments)
Social Security (fixed real payments)
Risk Portfolio (variable distribution)

One could also swap in an annuity en lieu of the LMP ladder.

Given that m-shares don't exist, without which lockboxes can't really be implemented as envisioned, I'm going to consider LMP + WSB RP as 'close enough'. ;)
Thanks for the summary, watchnerd. I remember that chapter being a real slog. I definitely agree with your conclusion!
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Fri Apr 30, 2021 3:25 pm Methodology: Same as before.
I'm curious if you're implementing the N-fund or 4 Fund with gold?
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
trueblueky
Posts: 2365
Joined: Tue May 27, 2014 3:50 pm

Re: Bill Sharpe's preferred portfolio

Post by trueblueky »

watchnerd wrote: Sat May 01, 2021 6:47 pm
HootingSloth wrote: Sat May 01, 2021 8:18 am I thought Bill Sharpe had a (seemingly idiosyncratic) "lockbox" approach that more or less works by dividing the portfolio into different accounts for each year of consumption in the future. Into each lockbox, you place a certain amount of TIPS plus a certain amount of stocks/bonds in the risk portfolio (plus, in his ideal world, a certain amount of theoretical assets that you can't actually buy that he calls "m-shares" and that I haven't really tried to understand). You leave each lockbox alone until you get to the year it is assigned, and then you spend whatever ends up in the lockbox. It is described in chapter 15, here: https://web.stanford.edu/~wfsharpe/RISMAT/


To me, the lockbox sounds like a weird LMP portfolio where each lockbox is similar to a ladder, but has the following ingredients:
1. Zero-coupon TIPS maturing in 9 years
2. World Bond/Stock Mutual Fund or ETF Shares to be sold in 9 years
3. m-Shares maturing in 9 years
#1 doesn't actually exist; there are no zero coupon TIPS.

I half fake it, though, by using alternating rungs of TIPS and STRIPS in a bond ladder.

#2 Exists

#3 Does not exist


Chapter 15 is probably the only chapter in the RISMAT series where I feel Sharpe jumped the shark between applicable theory and just pure mental masturbation.

I think he sort of admits this, as he says:
The conclusion is that absent the availability of suitable m-shares, it is impossible to obtain a portfolio optimal for an investor with a constant relative risk aversion marginal utility function with risk aversion that differs from that priced in the market portfolio. That said, it is possible
to choose a combination of TIPS and the market portfolio for one year, find the implied marginal utility function, then find combinations for subsequent years that are optimal for the same marginal utility.
He then gives some utility functions for approximating the lockbox contents with TIPS and WSB shares.

At this point, I don't see the point in trying make a model to determine how many WSB shares to allocate a 'lockbox', as opposed to just withdrawing from the Risk Portfolio at a future date according to needs.

It feels tortured, and not solving a genuine problem, given that m-shares don't exist.

He then ends the chapter with:
Absent the availability of m-shares, retirees can either need to choose retirement income strategies that provide payments with significantly different probability distributions of income at future dates, adopt an approach consistent with marginal utility functions of future income
that differ substantially, or select some combination of the two approaches. Moreover, given the fact that most retirees will receive fixed real payments from Social Security or some other sort of defined benefit plan, it will be important take into account all sources of income. The
remaining chapters explore some of these implications in detail for the construction of strategies for providing income with and without insuring against longevity risk. In both contexts, lockboxes can play a prominent role.
I my case, I will have an income strategy that provide payments with significantly different probability distributions at future dates:

LMP TIPS/STRIPS ladder (mixture of fixed real and nominal payments)
Social Security (fixed real payments)
Risk Portfolio (variable distribution)

One could also swap in an annuity en lieu of the LMP ladder.

Given that m-shares don't exist, without which lockboxes can't really be implemented as envisioned, I'm going to consider LMP + WSB RP as 'close enough'. ;)
I enjoyed parts of the book. It seems longevity is not solved by lockboxes (you don't know how many boxes you will need), which is one of two major quandaries. The other is you don't know how much money you need the last year. If those two were known, the rest would become much easier.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

trueblueky wrote: Sun May 02, 2021 8:57 am
I enjoyed parts of the book. It seems longevity is not solved by lockboxes (you don't know how many boxes you will need), which is one of two major quandaries. The other is you don't know how much money you need the last year. If those two were known, the rest would become much easier.
The other issue is the inability to revise the lockboxes if something changes, e.g. Social Security benefits get reduced.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

watchnerd wrote: Sun May 02, 2021 8:25 am I'm curious if you're implementing the N-fund or 4 Fund with gold?
I currently have a version of N-fund without gold, adjusted slightly for personal circumstance. I hold TSM, TISM, TBM, TIBM, high-yield corporate bond, TIPS, and muni bonds.

Some wrinkles:
  1. For high-yield corporate bonds, I use VWHEX/VWEAX (in my Roth IRA for tax efficiency) which is not an index fund, but performs very closely to JNK and HYG which are index funds. Vanguard doesn't currently offer a high-yield corporate bond index fund.
  2. Because I used to lived in CA until recently and had a high tax rate, I hold only CA muni bonds, and in fact I apportioned some of my TBM allocation to CA muni bonds instead for greater tax efficiency. Now that I've moved to FL (no income tax state), I'm slowly adjusting the allocation back toward TBM, and I plan to switch my CA muni bonds to the broad US muni bond fund (VTEAX) as soon as I can sell them for a loss (or at least lower gain) when interest rates inevitably rise again.
  3. Unlike Sharpe, I treat TIPS as just another opaque asset class whose allocation I let the market caps decide for me. (Zero-coupon TIPS would be ideal for LMP... but as you noted, they don't exist.)
Things I am considering in the future:
  1. Considering adding gold in my Roth IRA. But the gold allocation is so low that I haven't been motivated to make the final call between GLD, SGOL, IAU, and AAAU yet. (Leaning toward IAU based on mixture of assets-under-management, expense ratio, and bid-ask spread considerations, but I haven't finished researching.)
  2. Considering buying BTC and ETH. But their float-adjusted market cap is not easy to find. Most sources list their non-float-adjusted market cap, which is not the right thing to use for determining the efficient market allocation. I am also unsure of wrinkles around security, taxation, and estate planning around holding these.
Since you were talking about LMP earlier... I'm not directly attempting LMP per se. Rather, I eyeball my intrinsic exposure to various risks (e.g., term risk based on my retirement horizon, credit risk based on my job security, sector risk based on my employment industry, etc.), and try to hand-wavily compensate for them in my RP so that my overall risk exposure (i.e., the combination of my intrinsic risk exposures and my investment portfolio's risk exposures) match the global market portfolio's (GMP's) risk exposure. For example, if I had a stable job in tech, I'd reduce my bond allocation (because I myself would function like a bond that has a fixed income via my salary) and maybe short the tech sector somehow (because my job compensation would already already exposed to tech sector risk). I think LMP + RP is just a very specific case of this general strategy of "getting back to the GMP's risk allocation". LMP zeroes out the risk (mostly term risk) of future cash outflows by hedging them against (ideally, zero-coupon) duration-matched TIPS. But you can do the same for other risks in your life as well. LMP + RP is no longer a "bucket / mental accounting" approach if viewed through this lens of targeting your risk to the GMP's risk.

edit: I'd summarize the above point by saying that we should be less focused on attaining the global market's allocation to asset classes in our investment portfolio, and instead more focused on attaining the global market's allocation to compensated risks in our overall risk exposure (intrinsic + portfolio). I.e., we should be trying to achieve the following equation:

Code: Select all

Global Market Portfolio's risks = My intrinsic risks + My portfolio's risks
... because the GMP gives an efficient allocation of compensated risks (via the Lego blocks of available asset classes).

For most people, typically the portfolio is the free variable in the above equation. So we should skew our portfolio based on our intrinsic risk exposure. IMO that would be a principled way to do factor investing (as opposed to just saying, "I have high risk tolerance (whatever that means...), and so I can bet on small/value stocks").
Last edited by djm2001 on Sun May 02, 2021 12:10 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)
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Sun May 02, 2021 11:34 am For example, if I had a stable job in tech, I'd reduce my bond allocation (because I myself would function like a bond that has a fixed income via my salary) and maybe short the tech sector somehow (because my job compensation would already already exposed to tech sector risk). I think LMP + RP is just a very specific case of this general strategy of "getting back to the GMP's risk allocation". LMP zeroes out the risk (mostly term risk) of future cash outflows by hedging them against (ideally, zero-coupon) TIPS. But you can do the same for other risks in your life as well.
That would be my job.

I've thought about shorting, either broadly tech, or specifically my megacorp employer stock. But the cost of insurance adds up after a while.

Instead I just reduce my US exposure by an amount equal to my ESPP plan holdings, which is pretty straight forward. And low cost.

As I get closer to retirement (planned for 2025), I've thought of buying puts on my outstanding RSUs. If CAPE gets high enough, maybe I will. Although it may be too expensive and/or against company policy.

Muni bonds are tricky for me.

On the one hand, I should hold them, given our tax bracket and the fact we've run out tax-sheltered space for bonds.

On the other hand, holding market weight of VTEAX at 0.67% is too little to solve the 'bonds in taxable' problem.

Although, really, in WA state (no state income tax) the difference in tax equivalent yield is pretty small, even in our bracket (32-35%%).
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Sun May 02, 2021 11:34 am [

Things I am considering in the future:
  1. Considering adding gold in my Roth IRA. But the gold allocation is so low that I haven't been motivated to make the final call between GLD, SGOL, IAU, and AAAU yet. (Leaning toward IAU based on mixture of assets-under-management, expense ratio, and bid-ask spread considerations, but I haven't finished researching.)
I'm sure you've done this already, but I thought it was amusing how little of an impact the tradable market weight in gold has:

https://www.portfoliovisualizer.com/bac ... on5_2=0.16
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Blue456
Posts: 2151
Joined: Tue Jun 04, 2019 5:46 am

Re: Bill Sharpe's preferred portfolio

Post by Blue456 »

watchnerd wrote: Sun May 02, 2021 5:10 pm
djm2001 wrote: Sun May 02, 2021 11:34 am [

Things I am considering in the future:
  1. Considering adding gold in my Roth IRA. But the gold allocation is so low that I haven't been motivated to make the final call between GLD, SGOL, IAU, and AAAU yet. (Leaning toward IAU based on mixture of assets-under-management, expense ratio, and bid-ask spread considerations, but I haven't finished researching.)
I'm sure you've done this already, but I thought it was amusing how little of an impact the tradable market weight in gold has:

https://www.portfoliovisualizer.com/bac ... on5_2=0.16
Hey you made $39 :)
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

Blue456 wrote: Sun May 02, 2021 5:23 pm
Hey you made $39 :)
With senior discounts, that will be like $78!

Don't spend it all in one place, kids.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
aj76er
Posts: 1178
Joined: Tue Dec 01, 2015 10:34 pm
Location: Austin, TX

Re: Bill Sharpe's preferred portfolio

Post by aj76er »

djm2001 wrote: Sun May 02, 2021 11:34 am
  1. Considering buying BTC and ETH. But their float-adjusted market cap is not easy to find. Most sources list their non-float-adjusted market cap, which is not the right thing to use for determining the efficient market allocation. I am also unsure of wrinkles around security, taxation, and estate planning around holding these.
Look into the ETF BITW, which holds a market cap weighted index of crypto currencies.
"Buy-and-hold, long-term, all-market-index strategies, implemented at rock-bottom cost, are the surest of all routes to the accumulation of wealth" - John C. Bogle
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

aj76er wrote: Sun May 02, 2021 8:20 pm
djm2001 wrote: Sun May 02, 2021 11:34 am
  1. Considering buying BTC and ETH. But their float-adjusted market cap is not easy to find. Most sources list their non-float-adjusted market cap, which is not the right thing to use for determining the efficient market allocation. I am also unsure of wrinkles around security, taxation, and estate planning around holding these.
Look into the ETF BITW, which holds a market cap weighted index of crypto currencies.
EXPENSE RATIO 2.5%*

Dang!

Although if it's 1% of a port, maybe not such a big deal.

But it looks to be super thinly traded....volume of 18?
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
HootingSloth
Posts: 1041
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

aj76er wrote: Sun May 02, 2021 8:20 pm
djm2001 wrote: Sun May 02, 2021 11:34 am
  1. Considering buying BTC and ETH. But their float-adjusted market cap is not easy to find. Most sources list their non-float-adjusted market cap, which is not the right thing to use for determining the efficient market allocation. I am also unsure of wrinkles around security, taxation, and estate planning around holding these.
Look into the ETF BITW, which holds a market cap weighted index of crypto currencies.
That's interesting. They also list the float-adjusted market capitalizations for each cryptocurrency that is used in constructing their index, e.g. here: https://www.bitwiseinvestments.com/indexes/Bitwise-10

Along with this explanation:
About Market Capitalization
The Bitwise Crypto Indexes use a free-float measure of market capitalization to better reflect the true investable opportunity in the crypto space. The free float adjustment excludes issuance that is withheld from the public market, such as coins held by insiders or project foundations; the inflation adjustment captures the forecasted additional issuance of coins or tokens over the next five years, reflecting the fact that cryptoasset issuance is not fixed, but constantly evolving. One result of this institutional viewpoint is that Bitwise’s market capitalization statistics may not agree with other publicly available statistics that use naïve and/or static calculations of market capitalization.
It seems to me that it would likely be quite difficult to make these determinations for cryptocurrencies. For example, you may not be able to tell when insiders divide up holdings of cryptocurrency across a large number of different wallet addresses or have their holdings combined with the holdings of others in omnibus addresses. It would also seem impossible to know how much cryptocurrency has been "burned," i.e. transferred into a wallet address where no one continues to hold the private key. I'm sure there are other difficulties that I am also not thinking of.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

I'm skeptical that the BITW published allocations are actually float-adjusted. (Maybe the fund uses float-adjusted allocations internally, but the published weights do not look float-adjusted.)

Here is one source of float-adjusted crypto market caps: https://coinmetrics.io/crypto-prices/

May 3 snapshot:
Bitcoin: $1,098.4B (non-float-adjusted); $856.2B (float-adjusted)
Ethereum: $366B (non-float-adjusted); $352.5B (float-adjusted)

As a sanity check, the non-float-adjusted mkt caps from coinmetrics.io match closely to those at https://coinmarketcap.com/.

May 3 snapshot:
Bitcoin: $1,100B
Ethereum: $366.2B

And here are the market caps used by the BITW fund for its allocations: https://www.bitwiseinvestments.com/funds/Bitwise-10

May 2 snapshot:
Bitcoin: $1,064B
Ethereum: $343B

The Bitcoin market caps match the non-float-adjusted numbers from coinmetrics.io and coinmarketcap.com, and do not match the float-adjusted market caps from coinmetrics.io. Maybe Bitwise vastly overestimates the free-float portion relative to Coin Metrics' methodology. I'm not confident in either methodology at this point.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
HootingSloth
Posts: 1041
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

djm2001 wrote: Mon May 03, 2021 7:23 am Maybe Bitwise vastly overestimates the free-float portion relative to Coin Metrics' methodology. I'm not confident in either methodology at this point.
I think this is very likely the case. If you look at, for example, the float adjustment methodology that S&P uses in the construction of its indexes (https://www.spglobal.com/spdji/en/docum ... nload=true), it is hard to imagine how you would try to translate that from the context of publicly traded stocks into cryptocurrencies. In many cases, the analogous categories of holdings would be difficult to define at best. Even when there are appropriate analogies, the necessary information typically would not be publicly available. It seems like large amounts of guesswork would be unavoidable in determining float adjustments for cryptocurrencies, so it should not be surprising that different sources come to very different conclusions.
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

HootingSloth wrote: Mon May 03, 2021 8:28 pm
djm2001 wrote: Mon May 03, 2021 7:23 am Maybe Bitwise vastly overestimates the free-float portion relative to Coin Metrics' methodology. I'm not confident in either methodology at this point.
I think this is very likely the case. If you look at, for example, the float adjustment methodology that S&P uses in the construction of its indexes (https://www.spglobal.com/spdji/en/docum ... nload=true), it is hard to imagine how you would try to translate that from the context of publicly traded stocks into cryptocurrencies. In many cases, the analogous categories of holdings would be difficult to define at best. Even when there are appropriate analogies, the necessary information typically would not be publicly available. It seems like large amounts of guesswork would be unavoidable in determining float adjustments for cryptocurrencies, so it should not be surprising that different sources come to very different conclusions.
S&P Crypto Indexes announced

https://www.coindesk.com/sp-goes-live-w ... to-indexes
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
HootingSloth
Posts: 1041
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

watchnerd wrote: Tue May 04, 2021 12:47 pm
HootingSloth wrote: Mon May 03, 2021 8:28 pm
djm2001 wrote: Mon May 03, 2021 7:23 am Maybe Bitwise vastly overestimates the free-float portion relative to Coin Metrics' methodology. I'm not confident in either methodology at this point.
I think this is very likely the case. If you look at, for example, the float adjustment methodology that S&P uses in the construction of its indexes (https://www.spglobal.com/spdji/en/docum ... nload=true), it is hard to imagine how you would try to translate that from the context of publicly traded stocks into cryptocurrencies. In many cases, the analogous categories of holdings would be difficult to define at best. Even when there are appropriate analogies, the necessary information typically would not be publicly available. It seems like large amounts of guesswork would be unavoidable in determining float adjustments for cryptocurrencies, so it should not be surprising that different sources come to very different conclusions.
S&P Crypto Indexes announced

https://www.coindesk.com/sp-goes-live-w ... to-indexes
That's interesting. Thanks for sharing.

Note, however, that S&P does not claim that the combined index is weighted by "float adjusted" market capitalization. The methodology includes a concept that plays a similar role to a float adjustment, called the "effective coin supply." However, there is little that I could find describing how the effective coin supply will be determined. The policy document only says that it will be determined once per quarter (on the rebalancing date of the index) and that "coin burning will adjust the effective coin supply at rebalancing." I don't know how they plan to take burning into account, as anyone can privately burn coins without any public disclosure that it has been done. https://www.spglobal.com/spdji/en/docum ... x-math.pdf
Global Market Portfolio + modest tilt towards volatility (80/20->60/40 as approach FI) + modest tilt away from exchange rate risk (80% global+20% U.S. stocks; currency-hedge bonds) + tax optimization
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

HootingSloth wrote: Tue May 04, 2021 1:03 pm

That's interesting. Thanks for sharing.

Note, however, that S&P does not claim that the combined index is weighted by "float adjusted" market capitalization. The methodology includes a concept that plays a similar role to a float adjustment, called the "effective coin supply." However, there is little that I could find describing how the effective coin supply will be determined. The policy document only says that it will be determined once per quarter (on the rebalancing date of the index) and that "coin burning will adjust the effective coin supply at rebalancing." I don't know how they plan to take burning into account, as anyone can privately burn coins without any public disclosure that it has been done. https://www.spglobal.com/spdji/en/docum ... x-math.pdf
Yes, it's some kind of 'points' system like the Dow uses, I guess?

I need to re-read it, as it seemed kind of fuzzy to me on first read.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

Now that my LMP ladder pretty much built, here is the Global Market Portfolio allocation I'm using for my Risk Portfolio:

==Global Stocks== 59.7%
US Stocks: 34.67%
Ex-US Stocks: 25.03%
==Global Bonds== 38.2%
US Bonds: 19.96%
Ex-US Bonds: 18.23%
==Alts== 2.1%
Crypto: 1.94%
Gold Backed ETFs: 0.17%
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
dml130
Posts: 166
Joined: Wed Sep 02, 2020 11:40 am

Re: Bill Sharpe's preferred portfolio

Post by dml130 »

watchnerd wrote: Fri May 07, 2021 7:19 am Now that my LMP ladder pretty much built, here is the Global Market Portfolio allocation I'm using for my Risk Portfolio:

==Global Stocks== 59.7%
US Stocks: 34.67%
Ex-US Stocks: 25.03%
==Global Bonds== 38.2%
US Bonds: 19.96%
Ex-US Bonds: 18.23%
==Alts== 2.1%
Crypto: 1.94%
Gold Backed ETFs: 0.17%
If I recall correctly, you were considering EM local currency bonds as a sort of stagflation hedge a little while back before deciding against and instead going with a currency basket - does your newer "Ex-US bonds" portion specifically include those bonds?
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

dml130 wrote: Fri May 07, 2021 5:52 pm If I recall correctly, you were considering EM local currency bonds as a sort of stagflation hedge a little while back before deciding against and instead going with a currency basket - does your newer "Ex-US bonds" portion specifically include those bonds?
It's an annoying sticking point.

At the moment, I'm using Vanguard's international bond fund.

I would prefer a local currency international bond fund, but none of them have an ER as cheap as Vanguard.

Which is annoying at the current low yields.

BWX has an ER of 0.35, which is too high for my liking for a core holding that makes up a large chunk of my portfolio:

https://www.ssga.com/us/en/intermediary ... nd-etf-bwx
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
dml130
Posts: 166
Joined: Wed Sep 02, 2020 11:40 am

Re: Bill Sharpe's preferred portfolio

Post by dml130 »

watchnerd wrote: Fri May 07, 2021 6:06 pm
It's an annoying sticking point.

At the moment, I'm using Vanguard's international bond fund.

I would prefer a local currency international bond fund, but none of them have an ER as cheap as Vanguard.

Which is annoying at the current low yields.

BWX has an ER of 0.35, which is too high for my liking for a core holding that makes up a large chunk of my portfolio:

https://www.ssga.com/us/en/intermediary ... nd-etf-bwx
I see what you're saying. I suppose if low yields vs. expense ratio are the issue with BWX, you could keep Vanguard's fund and consider adding a separate EM local currency bond fund like LEMB - 30 day SEC yield is over 4.5% vs. ER of 0.30. Although maybe that would complicate the portfolio beyond what you'd want.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

I was unpleasantly surprised to discover that the ratio of US vs. international bond market caps derived from FTSE indexes varies significantly from BNDW's ratio of BND vs. BNDX. In particular, FTSE USBIG vs WorldBIG - USBIG currently gives a roughly 52:48 split, while BNDW is currently split roughly 47:53.  This means that the FTSE indexes do not approximate the Bloomberg Barclays indexes as accurately as we had hoped.

So I decided to take a different route to deriving the Bloomberg Barclays bond index market caps (used by BND, BNDX, and BNDW). The basic idea is to assume that the Bloomberg and FTSE indexes agree on the US Treasury + Agency portion of the index. The US Treasury + Agency market cap is mentioned as a percentage on Vanguard's BND and BNDW "Portfolio & Management" pages, and as an easily inferrable dollar value from FTSE USBIG's factsheet. We can solve for the Bloomberg total index market cap dollar values using these equations:

Code: Select all

Bloomberg_US_Bond_Mkt_Cap = FTSE_USBIG_Treasury_Agency_Mkt_Cap / US_Treasury_Agency_Fraction_of_BND

Bloomberg_Global_Bond_Mkt_Cap = FTSE_USBIG_Treasury_Agency_Mkt_Cap / US_Treasury_Agency_Fraction_of_BNDW

Bloomberg_Foreign_Bond_Mkt_Cap = Bloomberg_Global_Bond_Mkt_Cap - Bloomberg_US_Bond_Mkt_Cap
US_Treasury_Agency_Fraction_of_BND and US_Treasury_Agency_Fraction_of_BNDW are read off the "Treasury / Agency" row of the "Distribution by issuer (% of fund)" table on the "Portfolio & Management" pages of BND and BNDW, respectively.

FTSE_USBIG_Treasury_Agency_Mkt_Cap is computed as follows:

Code: Select all

FTSE_USBIG_Treasury_Agency_Mkt_Cap = FTSE_USBIG_Treasury_Mkt_Cap * (1 + FTSE_USBIG_Agency_Percent / FTSE_USBIG_Treasury_Percent)
  • FTSE_USBIG_Treasury_Mkt_Cap is read off the "US Treasury" row and "Market Value" column in the table in the FTSE USBIG factsheet.
  • FTSE_USBIG_Treasury_Percent and FTSE_USBIG_Agency_Percent are read off the "Geographical and Quality Composition (Market Weight %)" graph (under the labels of "US Treasury" and "Agency", respectively) of the same factsheet.
Then we forward project these market caps from end-of-last-month to current time using the then- and now prices of BND, BNDW, and BNDX, respectively.

Code: Select all

Current_Mkt_Cap = Snapshot_Mkt_Cap * Current_Tracker_Price / Snapshot_Tracker_Price
Some sanity checks:
  • The forward-projected US vs international bond ratio now almost exactly tracks the values shown on BNDW's Portfolio and Management page.
  • The sum of the US and international bond market caps are very close to the values derived from FTSE's bond indexes.  This results in the stock vs bond ratio changing very little (60.84 : 39.16 for FTSE vs. 60.68 : 39.32 for the new way).  TBH, this is a surprising coincidence, but the good news is that it results in minimal change to the stock vs. bond allocation in people's portfolios.
I plan to update the portfolios in the market-cap spreadsheet with this change... assuming that no one points out a fatal flaw in the methodology.
Last edited by djm2001 on Mon May 17, 2021 3:46 pm, edited 2 times in total.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

dml130 wrote: Fri May 07, 2021 8:34 pm

I see what you're saying. I suppose if low yields vs. expense ratio are the issue with BWX, you could keep Vanguard's fund and consider adding a separate EM local currency bond fund like LEMB - 30 day SEC yield is over 4.5% vs. ER of 0.30. Although maybe that would complicate the portfolio beyond what you'd want.
I looked at that.

Aside from it being a PITA, the backtesting doesn't show a massive difference in outcomes between USD-hedged and foreign currency bonds over decade long cycles.

Which, makes sense, theoretically, as we should expect hedges to zero out over time.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Mon May 17, 2021 12:50 pm

I plan to update the portfolios in the market-cap spreadsheet with this change... assuming that no one points out a fatal flaw in the methodology.
I actually can't follow from this written example.

I'd have to see it in a spreadsheet to get the difference.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

watchnerd wrote: Mon May 17, 2021 12:59 pm I'd have to see it in a spreadsheet to get the difference.
The underlying spreadsheet is here so you can see all the formulas. (I had resisted sharing it until now because it reveals my Google account :annoyed , but I guess that's fine.) The newly derived data is on the Snapshots, FwdProj, and MktCaps tab. See the lines starting with "Bloomberg ...". I also added a temporary "Updated 4-Fund" tab which you can compare with the original "4-Fund" tab.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Mon May 17, 2021 1:10 pm
watchnerd wrote: Mon May 17, 2021 12:59 pm I'd have to see it in a spreadsheet to get the difference.
The underlying spreadsheet is here so you can see all the formulas. (I had resisted sharing it until now because it reveals my Google account :annoyed , but I guess that's fine.) The newly derived data is on the Snapshots, FwdProj, and MktCaps tab. See the lines starting with "Bloomberg ...". I also added a temporary "Updated 4-Fund" tab which you can compare with the original "4-Fund" tab.
Thanks!
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

A little late to the thread....but isn't an adaptive asset allocation predicated on a completely efficient market?

Suppose there is a market shock (like March 2020), and one has new monies available to invest. Should one not contribute more to the equity side or does Sharpe advocate making all new contributions per current global AA (stock/bond) on ANY market day, regardless of any apparent irrational market behavior.

I can accept the floating global AA concept but not the idea that there are never any deep "sales" in the market.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

guppyguy wrote: Wed May 19, 2021 10:56 am A little late to the thread....but isn't an adaptive asset allocation predicated on a completely efficient market?

Suppose there is a market shock (like March 2020), and one has new monies available to invest. Should one not contribute more to the equity side or does Sharpe advocate making all new contributions per current global AA (stock/bond) on ANY market day, regardless of any apparent irrational market behavior.

I can accept the floating global AA concept but not the idea that there are never any deep "sales" in the market.
I'm not using adaptive AA.

I read Sharpe's paper on it and didn't find it something I wanted to try implementing.

I'm just using a floating global AA.

As for how to deal with new contributions (or withdrawals), Sharpe discusses this in RISMAT 7.

https://web.stanford.edu/~wfsharpe/RISMAT/RISMAT-7.pdf
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

watchnerd wrote: Wed May 19, 2021 12:50 pm
guppyguy wrote: Wed May 19, 2021 10:56 am A little late to the thread....but isn't an adaptive asset allocation predicated on a completely efficient market?

Suppose there is a market shock (like March 2020), and one has new monies available to invest. Should one not contribute more to the equity side or does Sharpe advocate making all new contributions per current global AA (stock/bond) on ANY market day, regardless of any apparent irrational market behavior.

I can accept the floating global AA concept but not the idea that there are never any deep "sales" in the market.
I'm not using adaptive AA.

I read Sharpe's paper on it and didn't find it something I wanted to try implementing.

I'm just using a floating global AA.

As for how to deal with new contributions (or withdrawals), Sharpe discusses this in RISMAT 7.

https://web.stanford.edu/~wfsharpe/RISMAT/RISMAT-7.pdf
Just so we're on the same page, by floating global AA your mean within global equities and global bonds, not between/across global equities/bonds? I presume its (stock/bond) 60/40 based on your signature.

Did you end up using BNDW or still splitting between BND/BNDX?
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

guppyguy wrote: Wed May 19, 2021 1:07 pm
Just so we're on the same page, by floating global AA your mean within global equities and global bonds, not between/across global equities/bonds? I presume its (stock/bond) 60/40 based on your signature.

Did you end up using BNDW or still splitting between BND/BNDX?
Floating across everything.

The floating AA just coincidentally happens to round out to about 60% stocks at the moment.

The actual two significant digit allocation at the quarterly last rebalancing was:

Stocks: 59.67%
Bonds: 38.16%
Crypto & securitized gold: 2.17%

I use BNDW in IRA.

In taxable, I use VBTLX and VTABX
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

Yes, the efficiency of the market portfolio rests on the assumption that the market is efficient.  In practice, nobody could argue with a straight face that the global market is perfectly efficient, but the hope here is that it's efficient enough that lazy investors can get something close to an efficient portfolio with minimal analysis / speculation / effort. The global market portfolio is a logical starting point.  From there, you can tilt and time based on your personal circumstances (or "preferences", if you prefer).

The other part of your question raises an interesting question that I've often wondered about: How should I allocate my next dollar into an existing portfolio? Should I buy $1 of the market portfolio (regardless of what my existing portfolio is)? Or should I allocate the $1 to bring my existing portfolio as close to the market portfolio as possible? The former will buy a small basket of efficiently allocated assets, while the latter tries to make the existing allocation more efficient as quickly as possible (even though the new transaction would be contrarian). My intuition is that the latter (i.e., getting close to the target portfolio as quickly as possible) is the better strategy, assuming the portfolio is going to be held for a long time. Of course, if you already hold the market portfolio, then both approaches are equivalent.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

watchnerd wrote: Wed May 19, 2021 1:28 pm
guppyguy wrote: Wed May 19, 2021 1:07 pm
Just so we're on the same page, by floating global AA your mean within global equities and global bonds, not between/across global equities/bonds? I presume its (stock/bond) 60/40 based on your signature.

Did you end up using BNDW or still splitting between BND/BNDX?
Floating across everything.

The floating AA just coincidentally happens to round out to about 60% stocks at the moment.

The actual two significant digit allocation at the quarterly last rebalancing was:

Stocks: 59.67%
Bonds: 38.16%
Crypto & securitized gold: 2.17%

I use BNDW in IRA.

In taxable, I use VBTLX and VTABX
I'm confused then...your using a floating AA across everything but not using Sharpe's adaptive asset allocation? I thought those were one and the same.

So is it that you are not rebalancing quarterly? How are you employing new monies then?

What fund/ETF did you end up using for crypto and gold?
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Wed May 19, 2021 3:36 pm Yes, the efficiency of the market portfolio rests on the assumption that the market is efficient.  In practice, nobody could argue with a straight face that the global market is perfectly efficient, but the hope here is that it's efficient enough that lazy investors can get something close to an efficient portfolio with minimal analysis / speculation / effort. The global market portfolio is a logical starting point.  From there, you can tilt and time based on your personal circumstances (or "preferences", if you prefer).

The other part of your question raises an interesting question that I've often wondered about: How should I allocate my next dollar into an existing portfolio? Should I buy $1 of the market portfolio (regardless of what my existing portfolio is)? Or should I allocate the $1 to bring my existing portfolio as close to the market portfolio as possible? The former will buy a small basket of efficiently allocated assets, while the latter tries to make the existing allocation more efficient as quickly as possible (even though the new transaction would be contrarian). My intuition is that the latter (i.e., getting close to the target portfolio as quickly as possible) is the better strategy, assuming the portfolio is going to be held for a long time. Of course, if you already hold the market portfolio, then both approaches are equivalent.
If your goal is to move towards holding the market portfolio, I would just look at your AA in total, compare it to the GMP, and add the new money where it is needed to bring it closer to the GMP.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

guppyguy wrote: Wed May 19, 2021 3:48 pm
I'm confused then...your using a floating AA across everything but not using Sharpe's adaptive asset allocation? I thought those were one and the same.

So is it that you are not rebalancing quarterly? How are you employing new monies then?

What fund/ETF did you end up using for crypto and gold?
If by Sharpe's adaptive allocation strategy you're referring to this:

https://web.stanford.edu/~wfsharpe/aaap/wfsaaap.pdf

No, I'm not doing that. That's for adapting specified target allocations adaptively.

That's from 2009 and I believe it predates his RISMAT series work.

The GMP is floating.

For crypto monies held in brokerage, I'm using GBTC and ETHE held in market weight relative to each other.

For gold, I'm using GLDM.

Stocks and bonds are contributed per pay cycle, as usual for accumulation.

If they're out of whack, I rebalance them quarterly as the new GMP data is published.

Our portfolio is 7 figures, so it doesn't get much out of whack from new payroll contributions.
Last edited by watchnerd on Wed May 19, 2021 4:12 pm, edited 1 time in total.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

djm2001 wrote: Wed May 19, 2021 3:36 pm The other part of your question raises an interesting question that I've often wondered about: How should I allocate my next dollar into an existing portfolio? Should I buy $1 of the market portfolio (regardless of what my existing portfolio is)? Or should I allocate the $1 to bring my existing portfolio as close to the market portfolio as possible? The former will buy a small basket of efficiently allocated assets, while the latter tries to make the existing allocation more efficient as quickly as possible (even though the new transaction would be contrarian). My intuition is that the latter (i.e., getting close to the target portfolio as quickly as possible) is the better strategy, assuming the portfolio is going to be held for a long time. Of course, if you already hold the market portfolio, then both approaches are equivalent.
Personally, if I were to implement this, I would add new monies only quarterly when the global AA is updated/rebalanced. This way the new contributions are employed to a portfolio most reconciled with the global portfolio.
(Slight cash drag as my contributions hit every two weeks, but they are a very small percentage of the overall portfolio.)
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

watchnerd wrote: Wed May 19, 2021 3:58 pm
guppyguy wrote: Wed May 19, 2021 3:48 pm
I'm confused then...your using a floating AA across everything but not using Sharpe's adaptive asset allocation? I thought those were one and the same.

So is it that you are not rebalancing quarterly? How are you employing new monies then?

What fund/ETF did you end up using for crypto and gold?
If by Sharpe's adaptive allocation strategy you're referring to this:

https://web.stanford.edu/~wfsharpe/aaap/wfsaaap.pdf

No, I'm not doing that. That's for adapting specified target allocations adaptively.
Gotcha, we are on the same page.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

watchnerd wrote: Wed May 19, 2021 3:58 pm For crypto monies held in brokerage, I'm using GBTC and ETHE held in market weight relative to each other.

For gold, I'm using GLDM.
Why use ETFs instead of holding BTC and ETH directly in something like Coinbase? Isn't the ER high? Is it idea to hold it within a Roth IRA to avoid capital gains tax?

Also, what criteria did you use to select GLDM over SGOL, IAU, AAAU, etc? Mainly just ER and bid-ask spread?
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Wed May 19, 2021 4:15 pm
watchnerd wrote: Wed May 19, 2021 3:58 pm For crypto monies held in brokerage, I'm using GBTC and ETHE held in market weight relative to each other.

For gold, I'm using GLDM.
Why use ETFs instead of holding BTC and ETH directly in something like Coinbase? Isn't the ER high? Is it idea to hold it within a Roth IRA to avoid capital gains tax?

Also, what criteria did you use to select GLDM over SGOL, IAU, AAAU, etc? Mainly just ER and bid-ask spread?
Yes, the idea is to hold the crypto in Roth to avoid capital gains.

Yes, the ER is high, but it beats paying capital gains tax by holding at a crypto exchange. Also, the Grayscale funds are actually trading at a discount to NAV.

I picked GLDM for the ER lower than GLD, but I also didn't research real hard. It's such a tiny slice it doesn't matter much, really. Pick whatever you like.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

djm2001 wrote: Sun May 02, 2021 11:34 am
For most people, typically the portfolio is the free variable in the above equation. So we should skew our portfolio based on our intrinsic risk exposure. IMO that would be a principled way to do factor investing (as opposed to just saying, "I have high risk tolerance (whatever that means...), and so I can bet on small/value stocks").
This is the problem I see with any factor tilt. How does one quantify an adequate amount of tilt in any direction? There seems to be no rational or consensus for what might constitute market based "optimum" tilt, just the behavioral consideration, which while real is exactly the thing one would want to remove from an investing decision. It's either all tilt or none at all perhaps.

I'm speaking from personal experience.... now how do I incorporate my 25% SCV tilt with a floating GMP??? :D They really aren't compatible, not unless I have split my portfolio into two distinct portfolios, maybe 401K (GMP) and Roth (Larry portfolio).....
Last edited by guppyguy on Wed May 19, 2021 4:33 pm, edited 1 time in total.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

guppyguy wrote: Wed May 19, 2021 4:31 pm
djm2001 wrote: Sun May 02, 2021 11:34 am
For most people, typically the portfolio is the free variable in the above equation. So we should skew our portfolio based on our intrinsic risk exposure. IMO that would be a principled way to do factor investing (as opposed to just saying, "I have high risk tolerance (whatever that means...), and so I can bet on small/value stocks").
This is the problem I see with any factor tilt. How does one quantify an adequate amount of tilt in any direction? There seems to be no rational or consensus for what might constitute market based "optimum" tilt, just the behavioral consideration, which while real is exactly the thing one would want to remove from an investing decision. It's either all tilt or none at all perhaps.

I'm speaking from personal experience.... now how do I incorporate my 25% SCV tilt with a floating GMP??? :D
I think tilting is philosophically incompatible.

Tilting is an active decision to pivot away from market allocations.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

watchnerd wrote: Wed May 19, 2021 4:32 pm
guppyguy wrote: Wed May 19, 2021 4:31 pm
djm2001 wrote: Sun May 02, 2021 11:34 am
For most people, typically the portfolio is the free variable in the above equation. So we should skew our portfolio based on our intrinsic risk exposure. IMO that would be a principled way to do factor investing (as opposed to just saying, "I have high risk tolerance (whatever that means...), and so I can bet on small/value stocks").
This is the problem I see with any factor tilt. How does one quantify an adequate amount of tilt in any direction? There seems to be no rational or consensus for what might constitute market based "optimum" tilt, just the behavioral consideration, which while real is exactly the thing one would want to remove from an investing decision. It's either all tilt or none at all perhaps.

I'm speaking from personal experience.... now how do I incorporate my 25% SCV tilt with a floating GMP??? :D
I think tilting is philosophically incompatible.

Tilting is an active decision to pivot away from market allocations.
I agree theoretically, but it's nonetheless difficult to shift from one camp to the other.
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

guppyguy wrote: Wed May 19, 2021 4:36 pm
I agree theoretically, but it's nonetheless difficult to shift from one camp to the other.
I can't help you with that one.

I gave up on tilting years ago for behavioral reasons. ;)
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

watchnerd wrote: Wed May 19, 2021 4:32 pm
guppyguy wrote: Wed May 19, 2021 4:31 pm
djm2001 wrote: Sun May 02, 2021 11:34 am
For most people, typically the portfolio is the free variable in the above equation. So we should skew our portfolio based on our intrinsic risk exposure. IMO that would be a principled way to do factor investing (as opposed to just saying, "I have high risk tolerance (whatever that means...), and so I can bet on small/value stocks").
This is the problem I see with any factor tilt. How does one quantify an adequate amount of tilt in any direction? There seems to be no rational or consensus for what might constitute market based "optimum" tilt, just the behavioral consideration, which while real is exactly the thing one would want to remove from an investing decision. It's either all tilt or none at all perhaps.

I'm speaking from personal experience.... now how do I incorporate my 25% SCV tilt with a floating GMP??? :D
I think tilting is philosophically incompatible.

Tilting is an active decision to pivot away from market allocations.
I would agree that tilting based on pure speculation (e.g., "I believe tech will outperform, so I'll tilt toward that") is philosophically incompatible. But tilting to balance out one's own personal/innate risks is logical and compatible -- and indeed to not do so would be illogical and incompatible. For example... Clearly not everyone's portfolio can be tilted toward SCV relative to the average (market portfolio). There has to be an equal amount of aggregate tilt away from SCV as there is toward SCV. A logical reason to bet against SCV would be if one already has intrinsic SCV exposure (e.g., as an employee of an SCV company) and need to balance out that risk by tilting against it in their portfolio in order to get back to the (allegedly efficient) market portfolio risk allocation.

Ignoring one's own mortgage and job compensation and job sector and retirement horizon, and not factoring those into one's portfolio allocation seems like a mental accounting / bucketing trap, just as much as it would be to ignore one's Roth IRA when considering one's brokerage allocation. Yes, I admit it gets fuzzy when quantifying one's own risk exposure, and finding assets that can balance out that risk.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

djm2001 wrote: Wed May 19, 2021 5:05 pm I would agree that tilting based on pure speculation (e.g., "I believe tech will outperform, so I'll tilt toward that") is philosophically incompatible. But tilting to balance out one's own personal/innate risks is logical and compatible -- and indeed to not do so would be illogical and incompatible. For example... Clearly not everyone's portfolio can be tilted toward SCV relative to the average (market portfolio). There has to be an equal amount of aggregate tilt away from SCV as there is toward SCV. A logical reason to bet against SCV would be if one already has intrinsic SCV exposure (e.g., as an employee of an SCV company) and need to balance out that risk by tilting against it in their portfolio in order to get back to the (allegedly efficient) market portfolio risk allocation.

Ignoring one's own mortgage and job compensation and job sector and retirement horizon, and not factoring those into one's portfolio allocation seems like a mental accounting / bucketing trap, just as much as it would be to ignore one's Roth IRA when considering one's brokerage allocation. Yes, I admit it gets fuzzy when quantifying one's own risk exposure, and finding assets that can balance out that risk.
Unfortunately, I'm not a good person to ask about how to balance most of those elements with the GMP.

I have no mortgage, no debt, am 4 years away from early retirement, have a LMP portfolio that covers cost of living until 2037 (full SS eligibility).

Basically, we're pretty de-risked from a life stage POV.

So we use the GMP entirely as a risk portfolio for general nest feathering and planned future discretionary spending.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
guppyguy
Posts: 463
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

Does the name “N-fund” have any other significance then just a variable number of components making up the GMP? First I’ve seen of it on BH.

I don’t completely agree with the asset inclusion past stocks and bonds. Seems like one is just cherry picking micro-asset classes based on what’s popular, which is contrary to Sharpe’s thesis. Just an observation, not a judgement. Also, if one wanted a small SWR of their GMP in retirement it would have to be very large (3M plus) to precisely account for a .17% allocation with non-fractional ETF shares. Still mulling over junk bonds, TIPS, and munis.

@djm2001 Thank you for google sheet, it’s great!

:sharebeer
Gup
User avatar
watchnerd
Posts: 13609
Joined: Sat Mar 03, 2007 10:18 am
Location: Gig Harbor, WA, USA

Re: Bill Sharpe's preferred portfolio

Post by watchnerd »

Klipinger's happened to have an article on using Black Litterman modeling to determine that amount of crypto to add to a global stock / bond portfolio:

https://www.kiplinger.com/investing/602 ... 0portfolio.

I'm not intending to dig into the crypto angle due to recent Bogleheads bands on crypto discussions, but I did find it interesting that the global stock / bond ratio they're using to represent the global market is a different weighting.
In early 2021, the global market for stocks totaled $95 trillion and the global bonds market reached $105 trillion.
The base portfolio in this article is:

Stocks: 47%
Bonds: 52.5%

Unforunately, the data source wasn't cited so I don't know why there is such a large discrepancy.
Global stocks, IG/HY bonds, gold & digital assets at market weights 75% / 19% / 6% || LMP: TIPS ladder
Post Reply