Bill Sharpe's preferred portfolio

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
guppyguy
Posts: 465
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

I’ve got a basic question which may put the previous timing/rebalancing question in regard to WBS in context.

When the equity market collapses, bonds end up representing a greater percentage of the WBS. However, my understanding is that the combined market value of the WBS is lower. Right?

So rather then looking to market time by adjusting AA within WBS a better approach is to have enough safe assets and income outside of the WBS in which to make new purchases of the WBS at whatever AA is present at the time. This could be either spare savings from monthly income or any liquid funds above a basic emergency fund. This way you don’t have to make a decision as to whether stocks or bonds are over/under valued, the decision is only how much extra human capital and liquid reserves can you put into the WBS, perhaps by shifting short term priorities. Admittedly there are limits to this but it scratches the itch we all seem to have to want to buy something on sale.


(I’ve been reading thru Roth’s Depression Diary and the reoccurring theme seems to be to have enough safe assets (non WBS) to get thru the crisis and to have cash available to continue buying. The number of false bottoms and rallies during the Depression should convince anybody otherwise. )
rcmoormt
Posts: 19
Joined: Thu Jul 01, 2021 6:02 am

Re: Bill Sharpe's preferred portfolio

Post by rcmoormt »

Thank you for the replies, that makes sense. It was always my assumption that investors buy more equity when lower, I was wrong with that. It makes sense that the free fall could continue.

I stated in my initial question that the preexisting investments will roll with the global market. Am I correct about this? Let's say the global market is exactly 60/40 (vt/bndw) currently and I reallocate my 401k exactly to that. If in time stocks lower and it is a 50/50 split globally, will my current allocations also have moved to that same split with the market (given i didn't touch a thing in my portfolio)??

I think I explained it pretty well.... example:
Global market is 60/40. I put $60 in VT, $40 in BNDW. I leave it alone. Market eventually falls to 50/50. Will I look in my account and see a 50/50 split in vt/bndw?

Thank you so much everyone.
guppyguy
Posts: 465
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

rcmoormt wrote: Thu Jul 01, 2021 5:10 pm I stated in my initial question that the preexisting investments will roll with the global market. Am I correct about this? Let's say the global market is exactly 60/40 (vt/bndw) currently and I reallocate my 401k exactly to that. If in time stocks lower and it is a 50/50 split globally, will my current allocations also have moved to that same split with the market (given i didn't touch a thing in my portfolio)??
Yes, the allocation floats so it will vary over time by itself. You make contributions and withdrawals at the same proportion of whatever the current WBS AA shows. I agree, this at first feels counter to what a lot of others might advocate, but you're just owning the market and letting the market be the market.

The only time you might need to rebalance is when market cap values are updated on the spreadsheet. This is not due to valuation or timing, its because of things like share creation, etc etc. Its pretty minor but keeps everything tuned up.

:sharebeer
Gup
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

Hi guppyguy,

Part of what you're saying is consistent with classic Modern Portfolio Theory (MPT) -- that, given a bunch of (admittedly unrealistic) assumptions, every portfolio should be on the efficient frontier located on the Capital Market Line (CML), which consists of all linear combinations of the risk-free asset and the market portfolio of risky assets. I.e., according to classic MPT, all investors would hold just two components -- some amount of the risky market portfolio and some amount (perhaps negative, i.e., borrowing) of the risk-free asset. As Sharpe suggests, WBS is our proxy for the risky market portfolio, and TIPS is our proxy for a risk-free asset or what you called the "safe asset". It's very important to note that what Sharpe does not explicitly state (or maybe I didn't read RISMAT 7 closely enough) is that in order to be truly risk-free, particularly w.r.t. reinvestment / interest-rate risk, the TIPS need to be duration-matched to a predictable schedule of the investor's expenses/liabilities.

In the MPT framework, the typical narrative is that investors consciously pick a risk-vs-return tradeoff, which implicitly selects a point on the CML -- i.e., what combo of risk-free asset and risky market portfolio to hold. And this is where your proposal deviates from the usual MPT story. You're proposing to move around on the CML line, based on whether the total market cap of the market portfolio is "expensive" or "on sale" (in inflation-adjusted terms, I guess?).

Here are a couple of thoughts / questions about your approach:
  • How would you decide when WBS is "on sale"? Compare it to yesterday? Last year? Note that the market cap of the global market portfolio is generally increasing over the long run as we turn natural resources and innovation into future profit. So perhaps we could say that WBS is on sale when its market cap is below its all time high. This is similar to ERN's "active strategy" glidepath approach. But note that ERN also found that the active strategy (buy when "underwater", i.e., less than all-time high) doesn't make a big difference over the long run (60-year period). It's also tricky because we're measuring the market cap in terms of a currency that may be inflated or deflated relative to past measurements.
  • As mentioned above, in order for TIPS to be truly "risk-free" (e.g., safe from reinvestment/interest-rate risk), you'd need to know when you're going to use the money upfront, and buy an appropriate amount of duration-matched TIPS. This would require you to predict when you are going to sell the TIPS to buy WBS. Thus, we can't really use this "risk-free asset" to take advantage of unexpected opportunities that arise (even if we could recognize them).
  • Is it even possible to find a risk-free asset / safe asset that is safe from inflation risk and reinvestment/interest-rate risk and can be opportunistically consumed on an unpredictable schedule?
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Northern Flicker
Posts: 15367
Joined: Fri Apr 10, 2015 12:29 am

Re: Bill Sharpe's preferred portfolio

Post by Northern Flicker »

If your asset allocation is to hold stocks and bonds at market weight, once the allocation was implemented, you would never rebalance. This is the problem I have with using market cap to drive allocation weights across asset classes. Doing so gives up control over portfolio risk, and that risk varies with movements in the asset classes.

I believe the reason for saying TIPS to taste is that the a market cap allocation to TIPS would be a weighting of (I believe) about 2.5%, not enough to matter with respect to inflation protection. Once you say that an asset class can be over- or under-weighted to manage risk, you don't have a market cap portfolio. I could just as easily say I want to under-weight nominal bonds for inflation protection, and over-weighting TIPS would in fact just be the same thing if the TIPS allocation cannibalizes the nominal bond allocation. Thus, I don't even think that Dr. Sharpe is fully on board with the model.
guppyguy
Posts: 465
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

djm2001 wrote: Thu Jul 01, 2021 6:21 pm Hi guppyguy,

Part of what you're saying is consistent with classic Modern Portfolio Theory (MPT) -- that, given a bunch of (admittedly unrealistic) assumptions, every portfolio should be on the efficient frontier located on the Capital Market Line (CML), which consists of all linear combinations of the risk-free asset and the market portfolio of risky assets. I.e., according to classic MPT, all investors would hold just two components -- some amount of the risky market portfolio and some amount (perhaps negative, i.e., borrowing) of the risk-free asset. As Sharpe suggests, WBS is our proxy for the risky market portfolio, and TIPS is our proxy for a risk-free asset or what you called the "safe asset". It's very important to note that what Sharpe does not explicitly state (or maybe I didn't read RISMAT 7 closely enough) is that in order to be truly risk-free, particularly w.r.t. reinvestment / interest-rate risk, the TIPS need to be duration-matched to a predictable schedule of the investor's expenses/liabilities.

In the MPT framework, the typical narrative is that investors consciously pick a risk-vs-return tradeoff, which implicitly selects a point on the CML -- i.e., what combo of risk-free asset and risky market portfolio to hold. And this is where your proposal deviates from the usual MPT story. You're proposing to move around on the CML line, based on whether the total market cap of the market portfolio is "expensive" or "on sale" (in inflation-adjusted terms, I guess?).

Here are a couple of thoughts / questions about your approach:
  • How would you decide when WBS is "on sale"? Compare it to yesterday? Last year? Note that the market cap of the global market portfolio is generally increasing over the long run as we turn natural resources and innovation into future profit. So perhaps we could say that WBS is on sale when its market cap is below its all time high. This is similar to ERN's "active strategy" glidepath approach. But note that ERN also found that the active strategy (buy when "underwater", i.e., less than all-time high) doesn't make a big difference over the long run (60-year period). It's also tricky because we're measuring the market cap in terms of a currency that may be inflated or deflated relative to past measurements.
  • As mentioned above, in order for TIPS to be truly "risk-free" (e.g., safe from reinvestment/interest-rate risk), you'd need to know when you're going to use the money upfront, and buy an appropriate amount of duration-matched TIPS. This would require you to predict when you are going to sell the TIPS to buy WBS. Thus, we can't really use this "risk-free asset" to take advantage of unexpected opportunities that arise (even if we could recognize them).
  • Is it even possible to find a risk-free asset / safe asset that is safe from inflation risk and reinvestment/interest-rate risk and can be opportunistically consumed on an unpredictable schedule?
Hi djm2001-
I really wasn't thinking of it strictly from a MPT perspective. It's more personal finance and behavioral. Take spring of 2020 for example. If one has extra slack in their budget they could allocate cash or sell I-Bonds (no duration risk) or increase their paycheck withholdings, etc etc. (TIPS is also possibility but you bring up good objections). Discretionary spending can be converted as well as savings from lower prices of goods and services. We did all of this (minus the I-bond sale) during COVID. Of course the risk is in the adequacy of remaining liquidity to possible fund expenses unsupported by normal income ESPECIALLY over an unknown horizon, but I contend that if one has a healthy EF and high natural savings rate it isn't hard to individually quantify.

**What doesn't have to be done is to change the AA of stocks and bonds against WBS or to make specific bets away from MCW. **

The timing of WBS being categorized as on sale should be obvious, I hope. I'm not talking about minor corrections or pullbacks, but obvious cataclysmic world events that has everybody glued to a television. My personal willingness to do all of the above in 2000/2001 was a lot different than 2008 which was still different than last year. The main difference was more assets, much higher savings rate, much higher EF, and also older and much more secure in our human capital and willingness to put a little more into the overall market.

Gup
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

Hi Northern Flicker,

Sharpe says, "allocate by market weight, and adjust for personal circumstance". In other words, an investor should start with market portfolio (which is the right choice for the average investor), and then make departures from that based on their personal risk exposures that makes them different from average. Or more accurately, an investor should start with their own individual risks, and then try to attain the efficient risk allocation picked out by the market portfolio by buying the market portfolio as well as other hedges to zero out their individual / atypical / idiosyncratic risks.

When framed this way, the reason for "overweighting" TIPS is clear. In theory, an investor would use a TIPS ladder (over and above holding the market portfolio) to neutralize some of their predicted expenses (e.g., as watchnerd does) in order to bring them back to average (or some guess at what average looks like). These future expenses thus neutralized (with duration-matched TIPS), the investor then holds their remaining money in the market portfolio.

But of course, this is all theory. In practice, it's hard for an individual investor to maintain an accurately duration-matched TIPS ladder, it's hard to predict future expenses and earning power, it's hard to know what the average investor's risks look like, let alone quantify how one differs from that. In practice, what I personally do is use a rough heuristic to pick a stock / bond ratio appropriate to my life situation (e.g., if I expect to have lots of high-earning years remaining, I reduce my bond allocation) and to hold no extra TIPS beyond market caps -- I just hold enough cash for near-term (3-6 month) expenses in a savings account.

That said, I don't buy the need to control portfolio risks (by playing with the knobs of asset class weights) for any reason other than to neutralize one's own atypical / idiosyncratic risks to bring oneself back to the (hopefully) somewhat efficient risk allocation picked out by market portfolio. Not unless an investor can pick a more efficient allocation that the market can... Maybe some investors can; I can't.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
guppyguy
Posts: 465
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

Northern Flicker wrote: Thu Jul 01, 2021 6:30 pm If your asset allocation is to hold stocks and bonds at market weight, once the allocation was implemented, you would never rebalance. This is the problem I have with using market cap to drive allocation weights across asset classes. Doing so gives up control over portfolio risk, and that risk varies with movements in the asset classes.

I believe the reason for saying TIPS to taste is that the a market cap allocation to TIPS would be a weighting of (I believe) about 2.5%, not enough to matter with respect to inflation protection. Once you say that an asset class can be over- or under-weighted to manage risk, you don't have a market cap portfolio. I could just as easily say I want to under-weight nominal bonds for inflation protection, and over-weighting TIPS would in fact just be the same thing if the TIPS allocation cannibalizes the nominal bond allocation. Thus, I don't even think that Dr. Sharpe is fully on board with the model.
Think of WBS as an AA within your overall household AA. WBS is strictly as set forth earlier and your household AA is really, anything that can ever be converted to value. Moving assets into/out of WBS IS what controls portfolio risk, not arbitrarily changing the average market portfolio twice (away then back to WBS). So if the AA of WBS goes from 60/40 to 50/50, the overall market value of both probably decreased which makes your additional household funded capital purchases that much more valuable.

Some might say this is just semantics or mental accounting but I think it is actually how we really act as humans. I'd love to make every penny fit in our household fit into a model but its just far simpler this way.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

guppyguy wrote: Thu Jul 01, 2021 7:19 pm Hi djm2001-
I really wasn't thinking of it strictly from a MPT perspective. It's more personal finance and behavioral. Take spring of 2020 for example. If one has extra slack in their budget they could allocate cash or sell I-Bonds (no duration risk) or increase their paycheck withholdings, etc etc. (TIPS is also possibility but you bring up good objections). Discretionary spending can be converted as well as savings from lower prices of goods and services. We did all of this (minus the I-bond sale) during COVID. Of course the risk is in the adequacy of remaining liquidity to possible fund expenses unsupported by normal income ESPECIALLY over an unknown horizon, but I contend that if one has a healthy EF and high natural savings rate it isn't hard to individually quantify.

**What doesn't have to be done is to change the AA of stocks and bonds against WBS or to make specific bets away from MCW. **

The timing of WBS being categorized as on sale should be obvious, I hope. I'm not talking about minor corrections or pullbacks, but obvious cataclysmic world events that has everybody glued to a television. My personal willingness to do all of the above in 2000/2001 was a lot different than 2008 which was still different than last year. The main difference was more assets, much higher savings rate, much higher EF, and also older and much more secure in our human capital and willingness to put a little more into the overall market.

Gup
"On sale" would mean that the market is inefficient for whatever reason (e.g., behavioral), right? I certainly would never claim that the market is perfectly efficient. I wonder though... is it more inefficient at the macro level or at the individual stock level? If you decided to market time at all... then would it be better to market time WBS as a whole? Or would it be better to time targeted stocks / sectors? In other words, if one believes in market inefficiency enough to exploit it as an investment strategy, why not specifically target the weak spots rather than targeting both the efficient and inefficient parts together?

Personally, during big drops I would (and twice have -- once successfully and once unsuccessfully) employ tax-loss harvesting (sell fund X and buy similar-but-not-identical fund Y) to maintain my WBS allocation while saving on taxes.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

guppyguy wrote: Thu Jul 01, 2021 5:49 pm
rcmoormt wrote: Thu Jul 01, 2021 5:10 pm I stated in my initial question that the preexisting investments will roll with the global market. Am I correct about this? Let's say the global market is exactly 60/40 (vt/bndw) currently and I reallocate my 401k exactly to that. If in time stocks lower and it is a 50/50 split globally, will my current allocations also have moved to that same split with the market (given i didn't touch a thing in my portfolio)??
Yes, the allocation floats so it will vary over time by itself. You make contributions and withdrawals at the same proportion of whatever the current WBS AA shows. I agree, this at first feels counter to what a lot of others might advocate, but you're just owning the market and letting the market be the market.

The only time you might need to rebalance is when market cap values are updated on the spreadsheet. This is not due to valuation or timing, its because of things like share creation, etc etc. Its pretty minor but keeps everything tuned up.

:sharebeer
Gup
I could be wrong, but I believe this story is complicated somewhat by the different amount of (net) issuances of new stocks and bonds that happen over time. This explains why the WBS allocation can still be at 60/40 despite the fact that stocks have dramatically outperformed bonds over the last many decades. For this reason, I think maintaining the allocation would require rebalancing.
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
rcmoormt
Posts: 19
Joined: Thu Jul 01, 2021 6:02 am

Re: Bill Sharpe's preferred portfolio

Post by rcmoormt »

HootingSloth wrote: Thu Jul 01, 2021 8:02 pm
guppyguy wrote: Thu Jul 01, 2021 5:49 pm
rcmoormt wrote: Thu Jul 01, 2021 5:10 pm I stated in my initial question that the preexisting investments will roll with the global market. Am I correct about this? Let's say the global market is exactly 60/40 (vt/bndw) currently and I reallocate my 401k exactly to that. If in time stocks lower and it is a 50/50 split globally, will my current allocations also have moved to that same split with the market (given i didn't touch a thing in my portfolio)??
Yes, the allocation floats so it will vary over time by itself. You make contributions and withdrawals at the same proportion of whatever the current WBS AA shows. I agree, this at first feels counter to what a lot of others might advocate, but you're just owning the market and letting the market be the market.

The only time you might need to rebalance is when market cap values are updated on the spreadsheet. This is not due to valuation or timing, its because of things like share creation, etc etc. Its pretty minor but keeps everything tuned up.

:sharebeer
Gup
I could be wrong, but I believe this story is complicated somewhat by the different amount of (net) issuances of new stocks and bonds that happen over time. This explains why the WBS allocation can still be at 60/40 despite the fact that stocks have dramatically outperformed bonds over the last many decades. For this reason, I think maintaining the allocation would require rebalancing.

This is exactly why I asked the question. I can't grasp how the allocation could remain as stable in a portfolio as it is in the market with stocks performing vastly better.

Guppyguy, do you have any insight?
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

rcmoormt wrote: Thu Jul 01, 2021 8:19 pm
HootingSloth wrote: Thu Jul 01, 2021 8:02 pm
guppyguy wrote: Thu Jul 01, 2021 5:49 pm
rcmoormt wrote: Thu Jul 01, 2021 5:10 pm I stated in my initial question that the preexisting investments will roll with the global market. Am I correct about this? Let's say the global market is exactly 60/40 (vt/bndw) currently and I reallocate my 401k exactly to that. If in time stocks lower and it is a 50/50 split globally, will my current allocations also have moved to that same split with the market (given i didn't touch a thing in my portfolio)??
Yes, the allocation floats so it will vary over time by itself. You make contributions and withdrawals at the same proportion of whatever the current WBS AA shows. I agree, this at first feels counter to what a lot of others might advocate, but you're just owning the market and letting the market be the market.

The only time you might need to rebalance is when market cap values are updated on the spreadsheet. This is not due to valuation or timing, its because of things like share creation, etc etc. Its pretty minor but keeps everything tuned up.

:sharebeer
Gup
I could be wrong, but I believe this story is complicated somewhat by the different amount of (net) issuances of new stocks and bonds that happen over time. This explains why the WBS allocation can still be at 60/40 despite the fact that stocks have dramatically outperformed bonds over the last many decades. For this reason, I think maintaining the allocation would require rebalancing.

This is exactly why I asked the question. I can't grasp how the allocation could remain as stable in a portfolio as it is in the market with stocks performing vastly better.

Guppyguy, do you have any insight?
It turns out Yardeni tracks this information, and it appears to be quite a significant effect: https://www.yardeni.com/pub/fofisspurstbnd.pdf. As you can see, net issuances of stock have often been negative over the past twenty years and have never been more than about $400bn in a single quarter (or maybe the figures are annualized, I can't quite tell). In contrast, net issuances of bonds are consistently positive, and have hit over $6trn.

Edited to add: Thinking about it, a rough and long-term stability would seem to make sense. As equities grow, aggregate enterprise value grows at a similar rate, and so the aggregate capacity of corporations to issue new debt grows at a similar rate. At the same time, stock-market-cap-to-gdp and public-debt-to-gdp both seem to stay within a (fairly wide) range over time (although both are unusually high in the US right at the moment), and so the amount of outstanding public debt should roughly keep pace with stock market capitalization through new issuances of government debt. Obviously there is room for fluctuation, but it shouldn't be surprising that bonds don't get diluted away in the WBS allocation.
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
Northern Flicker
Posts: 15367
Joined: Fri Apr 10, 2015 12:29 am

Re: Bill Sharpe's preferred portfolio

Post by Northern Flicker »

djm2001 wrote: Thu Jul 01, 2021 7:26 pm Hi Northern Flicker,

Sharpe says, "allocate by market weight, and adjust for personal circumstance". In other words, an investor should start with market portfolio (which is the right choice for the average investor), and then make departures from that based on their personal risk exposures that makes them different from average. Or more accurately, an investor should start with their own individual risks, and then try to attain the efficient risk allocation picked out by the market portfolio by buying the market portfolio as well as other hedges to zero out their individual / atypical / idiosyncratic risks.

When framed this way, the reason for "overweighting" TIPS is clear. In theory, an investor would use a TIPS ladder (over and above holding the market portfolio) to neutralize some of their predicted expenses (e.g., as watchnerd does) in order to bring them back to average (or some guess at what average looks like). These future expenses thus neutralized (with duration-matched TIPS), the investor then holds their remaining money in the market portfolio.

But of course, this is all theory. In practice, it's hard for an individual investor to maintain an accurately duration-matched TIPS ladder, it's hard to predict future expenses and earning power, it's hard to know what the average investor's risks look like, let alone quantify how one differs from that. In practice, what I personally do is use a rough heuristic to pick a stock / bond ratio appropriate to my life situation (e.g., if I expect to have lots of high-earning years remaining, I reduce my bond allocation) and to hold no extra TIPS beyond market caps -- I just hold enough cash for near-term (3-6 month) expenses in a savings account.

That said, I don't buy the need to control portfolio risks (by playing with the knobs of asset class weights) for any reason other than to neutralize one's own atypical / idiosyncratic risks to bring oneself back to the (hopefully) somewhat efficient risk allocation picked out by market portfolio. Not unless an investor can pick a more efficient allocation that the market can... Maybe some investors can; I can't.
Although the tone of your article suggests a different position from mine, after a careful reading, I realize we are saying the same thing in different ways. I would point out that whether or not you start from the market portfolio doesn't really matter. I could just as easily start with 60% total US stocks, 20% treasuries, and 20% TIPS, and modify to accommodate personal circumstance and risk. Where the portfolio lands ultimately is what matters.
Northern Flicker
Posts: 15367
Joined: Fri Apr 10, 2015 12:29 am

Re: Bill Sharpe's preferred portfolio

Post by Northern Flicker »

djm2001 wrote: Thu Jul 01, 2021 7:26 pm Hi Northern Flicker,

Sharpe says, "allocate by market weight, and adjust for personal circumstance". In other words, an investor should start with market portfolio (which is the right choice for the average investor), and then make departures from that based on their personal risk exposures that makes them different from average. Or more accurately, an investor should start with their own individual risks, and then try to attain the efficient risk allocation picked out by the market portfolio by buying the market portfolio as well as other hedges to zero out their individual / atypical / idiosyncratic risks.

When framed this way, the reason for "overweighting" TIPS is clear. In theory, an investor would use a TIPS ladder (over and above holding the market portfolio) to neutralize some of their predicted expenses (e.g., as watchnerd does) in order to bring them back to average (or some guess at what average looks like). These future expenses thus neutralized (with duration-matched TIPS), the investor then holds their remaining money in the market portfolio.

But of course, this is all theory. In practice, it's hard for an individual investor to maintain an accurately duration-matched TIPS ladder, it's hard to predict future expenses and earning power, it's hard to know what the average investor's risks look like, let alone quantify how one differs from that. In practice, what I personally do is use a rough heuristic to pick a stock / bond ratio appropriate to my life situation (e.g., if I expect to have lots of high-earning years remaining, I reduce my bond allocation) and to hold no extra TIPS beyond market caps -- I just hold enough cash for near-term (3-6 month) expenses in a savings account.

That said, I don't buy the need to control portfolio risks (by playing with the knobs of asset class weights) for any reason other than to neutralize one's own atypical / idiosyncratic risks to bring oneself back to the (hopefully) somewhat efficient risk allocation picked out by market portfolio. Not unless an investor can pick a more efficient allocation that the market can... Maybe some investors can; I can't.
But the market does not fully control the allocation of the global market portfolio. Public policy has a much bigger influence on the market cap of sovereign bonds than the market.

I believe allocation weights for asset classes should be decided by risk management. Cap-weighted market portfolios are excellent implementations of asset classes. Once you adjust from the global market weightings of individual asset classes to adjust for personal circumstance, it amounts to the same thing. The fact that global market weight of asset classes was the starting point is not relevant. It is where you end up that matters.
guppyguy
Posts: 465
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

djm2001 wrote: Thu Jul 01, 2021 7:45 pm
"On sale" would mean that the market is inefficient for whatever reason (e.g., behavioral), right? I certainly would never claim that the market is perfectly efficient. I wonder though... is it more inefficient at the macro level or at the individual stock level? If you decided to market time at all... then would it be better to market time WBS as a whole? Or would it be better to time targeted stocks / sectors? In other words, if one believes in market inefficiency enough to exploit it as an investment strategy, why not specifically target the weak spots rather than targeting both the efficient and inefficient parts together?

Personally, during big drops I would (and twice have -- once successfully and once unsuccessfully) employ tax-loss harvesting (sell fund X and buy similar-but-not-identical fund Y) to maintain my WBS allocation while saving on taxes.
You are placing two bets with that approach, first in allocating opposite of what macro forces are valuing the overall market and then second in trying to overweight a section of equities that you believe will mean revert. You would then need to place two more bets to return to WBS. Are you that different from the average investor? You may be or have individual circumstances that warrant this deviation but make no mistake you are deviating from the market and at some point, if desired, reconcile back. Furthermore, how do you know whether this behavior will actually be beneficial in the long run? Most of us do zero post trade analysis (at least I rarely did), we just act as though we are improving our position as compared to the market.
guppyguy
Posts: 465
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

rcmoormt wrote: Thu Jul 01, 2021 8:19 pm
HootingSloth wrote: Thu Jul 01, 2021 8:02 pm
guppyguy wrote: Thu Jul 01, 2021 5:49 pm
rcmoormt wrote: Thu Jul 01, 2021 5:10 pm I stated in my initial question that the preexisting investments will roll with the global market. Am I correct about this? Let's say the global market is exactly 60/40 (vt/bndw) currently and I reallocate my 401k exactly to that. If in time stocks lower and it is a 50/50 split globally, will my current allocations also have moved to that same split with the market (given i didn't touch a thing in my portfolio)??
Yes, the allocation floats so it will vary over time by itself. You make contributions and withdrawals at the same proportion of whatever the current WBS AA shows. I agree, this at first feels counter to what a lot of others might advocate, but you're just owning the market and letting the market be the market.

The only time you might need to rebalance is when market cap values are updated on the spreadsheet. This is not due to valuation or timing, its because of things like share creation, etc etc. Its pretty minor but keeps everything tuned up.

:sharebeer
Gup
I could be wrong, but I believe this story is complicated somewhat by the different amount of (net) issuances of new stocks and bonds that happen over time. This explains why the WBS allocation can still be at 60/40 despite the fact that stocks have dramatically outperformed bonds over the last many decades. For this reason, I think maintaining the allocation would require rebalancing.

This is exactly why I asked the question. I can't grasp how the allocation could remain as stable in a portfolio as it is in the market with stocks performing vastly better.

Guppyguy, do you have any insight?
The ratio of global stocks to bonds is just comparing two pools of capital, valued by number units available in the public market times what the average investors thinks each unit is worth. These two pools can increase/decrease for many reasons but rebalancing as Sharpe puts forth is only meant to capture the overall capitalization of each pool, not the valuation of each unit within that pool (at least not directly).
So it's true, it's possible for the AA to remain the same because price and supply move in opposite directions. Yet the average investor doesn't care about supply, so long as the market can function - which is also why he only focuses on the largest, meatiest parts of the economy and not things such as real estate and private equity of which market efficiency might not always as apparent.

The average investor is only trying to decide whether they want to be a contrarian investor as compared to all of the other investors in the world. 50/50 may be the new 60/40, for reasons completely unrelated to valuation, since the world is off the gold standard, who know. WBS isn't perfect but I can't think of anything as close.

(I'm probably off on some of the details as I have little formal economic training, but I did stay in a Holiday Inn once :happy )
buyer
Posts: 52
Joined: Tue Jul 20, 2021 1:25 am

Re: Bill Sharpe's preferred portfolio

Post by buyer »

I've been experimenting recently to determine the precise world ratio specific to ITOT and IXUS, instead of VTI/VXUS. I think I've got it, but I wanted to run it by you all to make sure I'm doing it right.

Per the wiki article, I need to get the market caps of the indexes tracked by ITOT and IXUS. As of 2021-06-30 they are:

ITOT: S&P Total Market Index (TMI) = 47,829,435.70
IXUS: MSCI ACWI ex USA IMI Index = 32,093,118.55

Then I need the most recent prices and the prices from 2021-06-30 to get the price ratio:

Price Ratio = Ending Price / Starting Price
ITOT: 97.33 / 98.76 = ~0.9855
IXUS: 70.98 / 73.30 = ~0.9683

Then I do the following to get the current adjusted percentage:

Adjusted % = (Market Cap) x (Price Ratio) / ( Sum ( (Market Cap) x (Price Ratio) ) )

ITOT: (47,829,435.70) x (~0.9855) / 78,214,234.43
IXUS: (32,093,118.55) x (~0.9683) / 78,214,234.43

My final results (as of 2021-07-19) are:

ITOT: 60.27%
IXUS: 39.73%


I was surprised to find these percentages are off by 1.68% from the currently calculated VTI/VXUS ratio (58.59% / 41.41% as of 07-19), a larger difference than I would have expected.

Do my results seem right? Does the nearly 2% discrepancy make sense given the different indexes used by ITOT/IXUS versus VTI/VXUS? Or did my math go wrong somewhere? Thanks!
Last edited by buyer on Tue Jul 20, 2021 4:53 am, edited 1 time in total.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

I checked your sources (factsheet data and ETF prices) and your math, and it is correct. And you're doing the right thing by calculating your allocation based on the indexes that are tracked by the funds that you actually own.

I don't know the cause of the 1.68% discrepancy between the two sets of indexes, but it's definitely an interesting puzzle to dig into. Perhaps it has to do with how the different companies adjust for "free float" or set some other threshold.

That said, a 1.68% discrepancy in US vs ex-US stock allocations when computed from two different index sources is not a total shocker. I recently compared US vs ex-US bonds from Bloomberg Barclays indexes against FTSE indexes, and found a ~5% difference between Bloomberg-Barclay-based allocations vs FTSE-based allocations.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
buyer
Posts: 52
Joined: Tue Jul 20, 2021 1:25 am

Re: Bill Sharpe's preferred portfolio

Post by buyer »

djm2001, thank you for double-checking my work, much appreciated. Also thanks for your GMP sheet, a very handy reference.

Yes, I just spent some time reading the BH wiki article on indexing, and realize now that there are a lot different methodologies between the various index providers, so it makes sense that they'll produce different ratios.

Anyway, I'm looking forward to achieving the proper ITOT/IXUS world ratio in my portfolio with my next contribution.
Joey Jo Jo Jr
Posts: 553
Joined: Mon Jun 21, 2021 11:38 pm

Re: Bill Sharpe's preferred portfolio

Post by Joey Jo Jo Jr »

I find market weights intriguing, but does the issuance of massive amounts of public debt, a good bit of which is not actually owned by the public (such as the Fed) distort the picture? Would buying sovereign debt just because it’s issued encourage fiscal irresponsibility?

I like treasuries for the negative correlation to equities and the resulting better risk adjusted return, but I can’t get my head around owning them just because they are out there.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

If the Federal Reserve is distorting the market in a way that is detrimental to your own financial goals, you could always do your own additional float adjustment (over and above the bond index's float adjustment) by subtracting out the portion of market cap owned by the Fed.

In an ideal world, float adjustment would personalized. It would go beyond limiting to "publicly available" portion of assets, and focus more narrowly on the market cap of assets owned by individuals who are similar to you (same country, similar life situation, similar job, similar expenses). This would give a more personalized asset allocation that better fits your own circumstances An index's public float adjustment is a one-size-fits-all adjustment. You can either be lazy and use it, or you can customize it further with additional effort.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
HootingSloth
Posts: 1050
Joined: Mon Jan 28, 2019 2:38 pm

Re: Bill Sharpe's preferred portfolio

Post by HootingSloth »

I believe that the Bloomberg Barclays index (which is used, e.g., by BND) already takes out Treasurys that are purchased by the Federal Reserve at auction. From their methodology document:
Float Adjustments to Amount Outstanding

US Treasuries

Federal Reserve purchases and sales of nominal and inflation-linked US Treasuries in open market operations are adjusted in the Bloomberg Barclays flagship US Aggregate, Global Treasury and Series-L US TIPS Indices using data made publicly available on the Federal Reserve Bank of New York website. These reports are released weekly by the Fed, usually Thursdays around 4:30pm New York time.

Adjustments to each security’s amount outstanding are made in the Projected Universe for government purchases and sales for the Federal Reserve SOMA (System Open Market Account) on a weekly basis, typically on Fridays. When US Treasury nominal or inflation-linked bonds are issued or reopened, the initial par amount outstanding that enters the Projected Universe is reduced for any issuance bought by the Federal Reserve at auction.
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
longinvest
Posts: 5682
Joined: Sat Aug 11, 2012 8:44 am

Re: Bill Sharpe's preferred portfolio

Post by longinvest »

This post documents the monthly return and asset class weights as of June 30, 2021 of the Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.

The June 2021 return was:
  • Global Stock-and-Bond Portfolio: ((60.54% X 1.19% (global stocks)) + (39.46% X 0.61% (global bonds))) = 0.96% (USD)
    • Global stocks: Vanguard Total World Stock ETF (VT) NAV return
    • Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
As of June 30, 2021 the weights were:
  • Global stocks: $75,163,286 million USD Market cap -- 60.92%
    • FTSE Global All Cap Index (GEISLMS) -- Net MCap
  • Global bonds: $48,217,210 million USD Market cap -- 39.08%
    • 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 June 2021 return was 1.07%.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
longinvest
Posts: 5682
Joined: Sat Aug 11, 2012 8:44 am

Re: Bill Sharpe's preferred portfolio

Post by longinvest »

This post documents the monthly return and asset class weights as of July 31, 2021 of the Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.

The July 2021 return was:
  • Global Stock-and-Bond Portfolio: ((60.92% X 0.55% (global stocks)) + (39.08% X 1.34% (global bonds))) = 0.86% (USD)
    • Global stocks: Vanguard Total World Stock ETF (VT) NAV return
    • Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
As of July 31, 2021 the weights were:
  • Global stocks: $75,447,002 million USD Market cap -- 60.65%
    • FTSE Global All Cap Index (GEISLMS) -- Net MCap
  • Global bonds: $48,960,430 million USD Market cap -- 39.35%
    • 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 July 2021 return was 0.83%.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
buyer
Posts: 52
Joined: Tue Jul 20, 2021 1:25 am

Re: Bill Sharpe's preferred portfolio

Post by buyer »

djm2001: I see you have added ITOT/IXUS to your GMP sheet, which is great.

I am curious why you use a calculation (Num Constituents * Mean Mkt Cap) for the snapshot Market Cap of the S&P Total Market Index, rather than their stated Total Market Cap. (The former is 49,786,783.27 while the latter is 49,786,777.03.)
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

Thanks for pointing that out! When I set this up originally, I couldn't find the total market cap in the "Index Characteristics" table where I'm used to finding it for other S&P factsheets, and so I used the data available there (Num Constituents x Mean Mkt Cap) to derive it. But now that you made me look, I see that total market cap listed in the "Country/Region Breakdown" table. I have updated the spreadsheet to use the total market cap directly. The tiny difference in the two numbers is due to the factsheet rounding Mean Mkt Cap to the nearest cent.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
buyer
Posts: 52
Joined: Tue Jul 20, 2021 1:25 am

Re: Bill Sharpe's preferred portfolio

Post by buyer »

Makes sense!

On a side note, I didn't realize until yesterday that some of the indexes publish their fact sheets more often than quarterly. I was still using the June market caps in my own spreadsheet when I made some contributions this week.

With the updated July/August market cap data, I see I've overweighted ITOT by about one share. Not a big deal of course (only 0.08% off), but from now on I'll be checking for the updated market caps every month. :thumbsup
redbarn
Posts: 83
Joined: Mon Feb 10, 2020 4:07 pm

Re: Bill Sharpe's preferred portfolio

Post by redbarn »

In this thread, a common piece of advice is to treat the world portfolio or some approximation to it as a suitable portfolio for everyone. The actual financial theory though would suggest accounting for risk aversion by holding a positive or negative position in a riskless asset in addition to the world market portfolio. There is some debate about what the riskless asset should be but TIPS would be the natural option. Based on financial theory, for long-term investors, longer duration TIPS would be the natural option.

For someone who wants a portfolio that is more conservative than the world portfolio, it would be fairly easy to approximate this by using a TIPS fund for example along with the world portfolio. If someone wants a more aggressive portfolio than the world portfolio, however, that would require holding a negative position in TIPS, which seems impractical.

What I have always wondered about is whether a negative position in TIPS in practice is very different from just reducing the bond allocation accordingly. I tried to see this in portfolio visualizer using conveniently available funds. I did this with Vanguard funds but would be open for suggestions that would give us something more precise and/or longer backtest.

Link 1: Comparing 40% US stock/40% international stock/53% US bond/-33% Vanguard TIPS fund to 40% US stock/40% international stock/20% US bond. (This version doesn't use international bonds but has a longer backtest.)

Link 2: Comparing 40% US stock/40% international stock/26% US bond/27% international bond/-33% Vanguard TIPS fund to 40% US stock/40% international stock/10% US bond/10% international bond. (This version uses international bonds but has a shorter backtest.)

Link 3: Comparing 50% US stock/50% international stock/67% US bond/-67% Vanguard TIPS fund to 50% US stock/50% international stock.

Link 4: Same as Link 3 but including international bonds. (Again, short backtest.)

Based on all these comparisons, it seems that just subtracting bond holdings leads to a pretty good approximation of the world market portfolio with negative position in the riskless asset. This isn't too surprising to me because the riskless asset is basically a type of bond fund so a negative position in one bond fund plus more of another bond fund ends up being a bit of a wash. The TIPS fund in this case is intermediate duration so that is probably even more the case than with a more theoretically ideal long-term TIPS fund. Of course this only works up to 100% stocks. If this is true, investing in VT and BNDW with allocations chosen according to risk aversion appears to be a very good approximation of the intent of the world bond/stock portfolio.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

Using TIPS as the risk-free asset (for a given duration) sounds reasonable. So then, on the long side (i.e., for risk-averse investors), I was puzzled that the market cap of TIPS is so low -- only 1% compared to the market cap of World Bond Stock, and only 2% compared to the market cap of US stocks and bonds. Is this because the supply of TIPS is low? Or are there very few risk-averse (US) investors demanding a risk-free asset? Or do most risk-averse US investors use something other than TIPS (e.g., annuities, pensions, Social Security) as the risk-free asset instead?

My guess is the latter since Social Security and pensions address both inflation- and longevity risk, while TIPS address only inflation. So perhaps most investors use Social Security and pensions as their risk-free asset, and it's mainly the relatively few early retirees who most need TIPS (to build an inflation-indexed income stream for the years between early retirement and qualifying for pension / Social Security).
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

djm2001 wrote: Mon Sep 13, 2021 8:25 am Using TIPS as the risk-free asset (for a given duration) sounds reasonable. So then, on the long side (i.e., for risk-averse investors), I was puzzled that the market cap of TIPS is so low -- only 1% compared to the market cap of World Bond Stock, and only 2% compared to the market cap of US stocks and bonds. Is this because the supply of TIPS is low? Or are there very few risk-averse (US) investors demanding a risk-free asset? Or do most risk-averse US investors use something other than TIPS (e.g., annuities, pensions, Social Security) as the risk-free asset instead?

My guess is the latter since Social Security and pensions address both inflation- and longevity risk, while TIPS address only inflation. So perhaps most investors use Social Security and pensions as their risk-free asset, and it's mainly the relatively few early retirees who most need TIPS (to build an inflation-indexed income stream for the years between early retirement and qualifying for pension / Social Security).
Sharpe seems to consider TIPS the basic risk-free asset. I think he discusses SS and annuities in detail so the retiree investor thoroughly understands their risks when evaluating the TIPS % allocation. I don't think he considers them risk-free assets.

In my own case, I don't have SS and my pension only has a 2% COLA; so, my bonds are mainly TIPS and short-term investment-grade (in an account where TIPS are not available).

When I asked him once about possible poor matching between the typical investor and some of the intl debt, he said the TIPS also compensate for that.
redbarn
Posts: 83
Joined: Mon Feb 10, 2020 4:07 pm

Re: Bill Sharpe's preferred portfolio

Post by redbarn »

I guess another option is that treasuries (or other very safe sovereigns) effectively act as the risk-free asset. There is inflation risk but if investors see meaningful inflation as relatively improbable, it could be regarded as a reasonable approximation. Significant inflation in developed countries did seem like a low risk in the inflation targeting era, at least until recently. Under this interpretation, the market portfolio would be stocks plus ex-government bonds.
djm2001 wrote: Mon Sep 13, 2021 8:25 am Using TIPS as the risk-free asset (for a given duration) sounds reasonable. So then, on the long side (i.e., for risk-averse investors), I was puzzled that the market cap of TIPS is so low -- only 1% compared to the market cap of World Bond Stock, and only 2% compared to the market cap of US stocks and bonds. Is this because the supply of TIPS is low? Or are there very few risk-averse (US) investors demanding a risk-free asset? Or do most risk-averse US investors use something other than TIPS (e.g., annuities, pensions, Social Security) as the risk-free asset instead?

My guess is the latter since Social Security and pensions address both inflation- and longevity risk, while TIPS address only inflation. So perhaps most investors use Social Security and pensions as their risk-free asset, and it's mainly the relatively few early retirees who most need TIPS (to build an inflation-indexed income stream for the years between early retirement and qualifying for pension / Social Security).
redbarn
Posts: 83
Joined: Mon Feb 10, 2020 4:07 pm

Re: Bill Sharpe's preferred portfolio

Post by redbarn »

pascalwager wrote: Mon Sep 13, 2021 10:54 pm When I asked him once about possible poor matching between the typical investor and some of the intl debt, he said the TIPS also compensate for that.
Could you elaborate on what you mean by the possible poor matching here?
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

redbarn wrote: Mon Sep 13, 2021 11:09 pm
pascalwager wrote: Mon Sep 13, 2021 10:54 pm When I asked him once about possible poor matching between the typical investor and some of the intl debt, he said the TIPS also compensate for that.
Could you elaborate on what you mean by the possible poor matching here?
I can't recall, but an issue raised here by others, so I asked Sharpe. I mentioned it earlier in this thread but can't find it now.

Also, is it sensible to hold intermediate-term bonds just because the market does if your investment horizon is long-term?
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

pascalwager wrote: Tue Sep 14, 2021 11:57 am Also, is it sensible to hold intermediate-term bonds just because the market does if your investment horizon is long-term?
Apologies if I am misinterpreting, but I think you are suggesting that if you have a stable salary (esp., one that is more stable and higher than average income) then you should reduce your allocation to bonds whose duration precedes your retirement date. This would be consistent with vineviz's advice. Adjusting for expected future salary flow is probably the most common and significant "... and then adjust for personal circumstances" case for pre-retirees.

In practice, given the ongoing headache of either maintaining a non-rolling ladder of individual bonds or else maintaining an ever-changing mix of constant-duration bond index funds, I'd imagine that many individual investors might opt instead to simply adjust their ratio of global stocks to global bonds and call it a day. It's analogous to asking, "is it sensible to hold stock in the company you work for?" Ideally you would reduce your allocation to your company stock (or even short it, if that were allowed), but it's so much more convenient to just hold a total stock index fund even though it does contain your company's stock.

P.S., Do you still have an open line of communication with Sharpe? I would like to ask him whether he would agree with the computation of Bloomberg Barclays bond index market caps outlined in the Methodology section here (as opposed to directly using the FTSE market caps as he suggested). I emailed him the question on a long shot, but he didn't reply.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

djm2001 wrote: Tue Sep 14, 2021 4:44 pm
pascalwager wrote: Tue Sep 14, 2021 11:57 am Also, is it sensible to hold intermediate-term bonds just because the market does if your investment horizon is long-term?
Apologies if I am misinterpreting, but I think you are suggesting that if you have a stable salary (esp., one that is more stable and higher than average income) then you should reduce your allocation to bonds whose duration precedes your retirement date. This would be consistent with vineviz's advice. Adjusting for expected future salary flow is probably the most common and significant "... and then adjust for personal circumstances" case for pre-retirees.

In practice, given the ongoing headache of either maintaining a non-rolling ladder of individual bonds or else maintaining an ever-changing mix of constant-duration bond index funds, I'd imagine that many individual investors might opt instead to simply adjust their ratio of global stocks to global bonds and call it a day. It's analogous to asking, "is it sensible to hold stock in the company you work for?" Ideally you would reduce your allocation to your company stock (or even short it, if that were allowed), but it's so much more convenient to just hold a total stock index fund even though it does contain your company's stock.

P.S., Do you still have an open line of communication with Sharpe? I would like to ask him whether he would agree with the computation of Bloomberg Barclays bond index market caps outlined in the Methodology section here (as opposed to directly using the FTSE market caps as he suggested). I emailed him the question on a long shot, but he didn't reply.
No, I just used his public email address.
JimmyJammy
Posts: 322
Joined: Sat Oct 26, 2013 1:08 pm

Re: Bill Sharpe's preferred portfolio

Post by JimmyJammy »

By the way, for bonds besides municipals, do you put them all in your tax-sheltered account?

I'd like to lower the risk of my portfolio by buying bonds. Should I do all of that in my ROTH IRA / 401 (k)?
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

JimmyJammy wrote: Sat Sep 18, 2021 12:16 am By the way, for bonds besides municipals, do you put them all in your tax-sheltered account?
Yes, I do... to the extent that they fit. Lots of good info on this topic at the Bogleheads wiki article on tax-efficient fund placement.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
longinvest
Posts: 5682
Joined: Sat Aug 11, 2012 8:44 am

Re: Bill Sharpe's preferred portfolio

Post by longinvest »

This post documents the monthly return and asset class weights as of August 31, 2021 of the Global Stock-and-Bond Portfolio or, if you prefer, William Sharpe's Market Portfolio. Here's a link to the previous entry.

The August 2021 return was:
  • Global Stock-and-Bond Portfolio: ((60.65% X 2.37% (global stocks)) + (39.35% X -0.31% (global bonds))) = 1.32% (USD)
    • Global stocks: Vanguard Total World Stock ETF (VT) NAV return
    • Global bonds: Vanguard Total World Bond ETF (BNDW) NAV return
As of August 31, 2021 the weights were:
  • Global stocks: $77,163,375 million USD Market cap -- 61.17%
    • FTSE Global All Cap Index (GEISLMS) -- Net MCap
  • Global bonds: $48,977,490 million USD Market cap -- 38.83%
    • 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 August 2021 return was 1.33%.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

In retirement, I'm now in the process of converting my own holdings into a WBS construction with about 27% TIPS.

I have two portfolios: one with total market funds and inflation bonds and the other with only factor-type and EM stock funds in a taxable account. The former (total market and TIPS) will be re-allocated as a WBS portfolio and inflation bonds. The latter (factor-type account) will be called my "Extraneous Stocks Portfolio" and spent-down first, or the distributions continuously used to augment my RISMAT holdings. (The "spend-down first" idea was suggested to me by vineviz in another portfolio re-construction situation, but unrelated to this thread.)

The extraneous stocks portfolio includes large value, micro-cap, small value, intl small value and emerging markets and is about 29% of my holdings.

Recently discovering the convenient real-time WBS allocator spreadsheet (by BH ajm2001) in the wiki was the final push needed to make this happen. I've already modified my spreadsheet for the new approach including all rebalance trades.
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

pascalwager wrote: Sun Oct 03, 2021 3:47 pm In retirement, I'm now in the process of converting my own holdings into a WBS construction with about 27% TIPS.

I have two portfolios: one with total market funds and inflation bonds and the other with only factor-type and EM stock funds in a taxable account. The former (total market and TIPS) will be re-allocated as a WBS portfolio and inflation bonds. The latter (factor-type account) will be called my "Extraneous Stocks Portfolio" and spent-down first, or the distributions continuously used to augment my RISMAT holdings. (The "spend-down first" idea was suggested to me by vineviz in another portfolio re-construction situation, but unrelated to this thread.)

The extraneous stocks portfolio includes large value, micro-cap, small value, intl small value and emerging markets and is about 29% of my holdings.

Recently discovering the convenient real-time WBS allocator spreadsheet (by BH ajm2001) in the wiki was the final push needed to make this happen. I've already modified my spreadsheet for the new approach including all rebalance trades.
Conversion to Sharpe WBS portfolio alongside 27% TIPS completed. It's probably not something for the spreadsheet-shy until Vanguard or Blackrock finally provide a world market-proxy etf.

The only problem was being forced to mainly use the Vanguard Global Bonds Index Fund for my bonds. It holds a fixed 70/30 US/intl bonds. In other accounts I used only BNDX for bonds, but could only improve to 66/34.

In retirement, at least, I personally like the idea that the market is determining my stock/bond ratio and I'm not depending on back-testing for basic investment decisions. And if value and EM do go wild, I'll still have a then-ballooning 29% in my Extraneous Stock Funds collection. Actually, that's similar to someone here who explicitly uses three different portfolios to meet their personal sense of diversification.

So, now, I guess I fall somewhere between Cochrane-Fama and Swedroe in taking advice for the average investor, but my real preference is for Sharpe and the WBS.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

Thank you for sharing the process behind converting a non-WBS portfolio to a WBS one. It was instructive to see a live example. I have a bunch of questions about the details. :D Feel free to ignore any of the questions that you are not comfortable answering.
pascalwager wrote: Tue Oct 05, 2021 10:04 pm Conversion to Sharpe WBS portfolio alongside 27% TIPS completed.
How did you arrive at 27% TIPS? Did you target a certain dollar amount which happened to land at 27% of your first portfolio?

In what form do you hold TIPS? I.e, do you hold a TIPS fund, or did you manually construct a non-rolling ladder of individual TIPS? (I believe the latter is more akin to a risk-free asset, but is trickier to set up.) Are you using the TIPS to set up a liability matching portfolio?

Do you hold any I-bonds in addition to the TIPS? Taxes aside, I-bonds seem to be a more flexible candidate for the risk-free asset than TIPS since I-bonds can fund unpredictable expense anytime between 5 and 30 years of purchase due to their flexible maturity, whereas TIPS would have to be carefully duration-matched upfront to a fixed/predetermined time in the future.

Was your existing "total market funds + inflation bonds" portfolio entirely in tax-exempt accounts? If so, how did you split your final WBS stocks and bonds between Roth vs traditional retirement accounts?
pascalwager wrote: Tue Oct 05, 2021 10:04 pm The only problem was being forced to mainly use the Vanguard Global Bonds Index Fund for my bonds. It holds a fixed 70/30 US/intl bonds. In other accounts I used only BNDX for bonds, but could only improve to 66/34.
Do you have a link to the Vanguard Global Bond Index fund that you selected? The only one I could find was this foreign-domiciled one. Btw, the Market Allocation chart in that link seems funny -- the pie chart shows US as being almost exactly 50%, but the table alongside says the US is 41.3%. :?

Did your previous "total market funds" portfolio already include (non-TIPS) bonds?
pascalwager wrote: Tue Oct 05, 2021 10:04 pm In retirement, at least, I personally like the idea that the market is determining my stock/bond ratio and I'm not depending on back-testing for basic investment decisions.
+1. I fancifully think of WBS as turning the world into a supercomputer that predicts the distribution of future outcomes and then outputs the allocation best poised to address that outcome distribution. WBS does seem like a great even-keeled, "everything considered" approach that smooths out some of the idiosyncratic biases that an individual may get hung up on.

re: WBS in retirement... I do have this vague but persistent worry in the back of my mind that the average retiree is probably not the average investor, specifically regarding the accounting of personal human capital and the general equilibrium story that would balance out the underweighting of bonds by accumulators. But I don't have enough mental clarity on this to make me deviate from WBS just yet. (If I would deviate, perhaps it would be to increasingly overweight global bonds relative to WBS allocation between retirement and death.)
pascalwager wrote: Tue Oct 05, 2021 10:04 pm And if value and EM do go wild, I'll still have a then-ballooning 29% in my Extraneous Stock Funds collection.
The 29% "extraneous stocks" portfolio already has some exposure to market risk. Did you decide to factor that into your WBS equity allocation, or ignore it?

Has your belief in the persistence of factor premia reduced over time?

Would you still spend down the "extraneous stocks" first if the market takes a prolonged dive (but not enough of a dive to enable tax-loss harvesting into WBS)?
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

djm2001 wrote: Wed Oct 06, 2021 9:44 am Thank you for sharing the process behind converting a non-WBS portfolio to a WBS one. It was instructive to see a live example. I have a bunch of questions about the details. :D Feel free to ignore any of the questions that you are not comfortable answering.
pascalwager wrote: Tue Oct 05, 2021 10:04 pm Conversion to Sharpe WBS portfolio alongside 27% TIPS completed.
How did you arrive at 27% TIPS? Did you target a certain dollar amount which happened to land at 27% of your first portfolio?

In what form do you hold TIPS? I.e, do you hold a TIPS fund, or did you manually construct a non-rolling ladder of individual TIPS? (I believe the latter is more akin to a risk-free asset, but is trickier to set up.) Are you using the TIPS to set up a liability matching portfolio?

Do you hold any I-bonds in addition to the TIPS? Taxes aside, I-bonds seem to be a more flexible candidate for the risk-free asset than TIPS since I-bonds can fund unpredictable expense anytime between 5 and 30 years of purchase due to their flexible maturity, whereas TIPS would have to be carefully duration-matched upfront to a fixed/predetermined time in the future.

Was your existing "total market funds + inflation bonds" portfolio entirely in tax-exempt accounts? If so, how did you split your final WBS stocks and bonds between Roth vs traditional retirement accounts?
pascalwager wrote: Tue Oct 05, 2021 10:04 pm The only problem was being forced to mainly use the Vanguard Global Bonds Index Fund for my bonds. It holds a fixed 70/30 US/intl bonds. In other accounts I used only BNDX for bonds, but could only improve to 66/34.
Do you have a link to the Vanguard Global Bond Index fund that you selected? The only one I could find was this foreign-domiciled one. Btw, the Market Allocation chart in that link seems funny -- the pie chart shows US as being almost exactly 50%, but the table alongside says the US is 41.3%. :?

Did your previous "total market funds" portfolio already include (non-TIPS) bonds?
pascalwager wrote: Tue Oct 05, 2021 10:04 pm In retirement, at least, I personally like the idea that the market is determining my stock/bond ratio and I'm not depending on back-testing for basic investment decisions.
+1. I fancifully think of WBS as turning the world into a supercomputer that predicts the distribution of future outcomes and then outputs the allocation best poised to address that outcome distribution. WBS does seem like a great even-keeled, "everything considered" approach that smooths out some of the idiosyncratic biases that an individual may get hung up on.

re: WBS in retirement... I do have this vague but persistent worry in the back of my mind that the average retiree is probably not the average investor, specifically regarding the accounting of personal human capital and the general equilibrium story that would balance out the underweighting of bonds by accumulators. But I don't have enough mental clarity on this to make me deviate from WBS just yet. (If I would deviate, perhaps it would be to increasingly overweight global bonds relative to WBS allocation between retirement and death.)
pascalwager wrote: Tue Oct 05, 2021 10:04 pm And if value and EM do go wild, I'll still have a then-ballooning 29% in my Extraneous Stock Funds collection.
The 29% "extraneous stocks" portfolio already has some exposure to market risk. Did you decide to factor that into your WBS equity allocation, or ignore it?

Has your belief in the persistence of factor premia reduced over time?

Would you still spend down the "extraneous stocks" first if the market takes a prolonged dive (but not enough of a dive to enable tax-loss harvesting into WBS)?
Today, one of my extraneous stock funds went negative, so I was able to sell and immediately buy into my WBS + TIPS. So, right now, I'm at 26/54/20 | Extraneous/WSB/TIPS. My BND/BNDX ratio is now down to 63/37.

Vg Global Bond Index Fund has the same ratio as used in their life-cycle funds, but it's only available as an insurance sub-account. (You can find it by searching "Vg Insurance Funds".) Not ideal, but better conformance within the WSB than using only BND. As I reinvest into WSB BNDX out of Extraneous, the ratio will improve.

My TIPS are multiple funds (LTPZ, SCHP, VAIPX) with annually-adjusted duration-matching to my investment horizon. Sharpe doesn't mention duration-matching, but probably thought the LTPZ expense ratio was too high to consider, if it was even available when he was writing RISMAT-7--or maybe even thought to leave the issue to the recommended paid advisor. Funds do carry liquidity risk, but so do individual TIPS if you can't hold them to maturity during market turmoil.

No I-Bonds anymore. Used them all for some necessary large purchases (new home, cars) after retiring.

I'd probably continue to hold my extraneous funds during a bear market and sell across my entire holdings as necessary. I've basically had two portfolios for a long-time: factor/EM and 50/50 US/intl market. All of my former nominal and TIPS bonds were purchased out of the world market stocks as my factor funds were all in taxable with long-term gains.

The 3-factor model was my first investing tutorial in 1995; but no, I don't believe in factor premia or using back-testing research. So, I've been using global value and EM for diversification based on long-term correlations (anyway, that's my rationalization). I think factors are largely about marketing and the Sharpe WBS isn't going to excite anyone.

Vanguard uses 17% TIPS in their 30/70 retiree income fund with a (declared) assumption of an SS input. I don't have SS so I wanted to go higher up to the point of not needing to sell much, if any, in my existing total market stock portfolio. I just varied the TIPS target % on my spreadsheet until my criteria were met. (I do have a pension with a 2% max COLA and an 80% floor.) So, no liability matching is involved.

I have five accounts (two taxable). At Vanguard my TIPS RMDs are overflowing into taxable and repurchased as TIPS. (Grabiner said this is OK.) My ROTH is 100% VTI and VXUS.

I now have both stock and bond funds in all of my tax-deferred and taxable accounts.

The average investor has a PhD in finance or math, etc--or at least is an investment professional--and that's not me. By holding the market created by those investors, I'm riding on their coattails. I don't know how to be more efficient than that. And a TIPS allocation can mask some of the personal nominal bond discrepancies.

My portfolio is also relevant for a particular, younger dependent/heir, and not just myself. She won't have a pension and not much SS; so I was willing to reduce market risk when reinvesting Extraneous proceeds into WBS + TIPS.
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

When Sharpe discusses (in RISMAT) WBS re-balancing, he's addressing the fluctuations of the four-fund S/B proxy only. However, if you're also holding duration-matched TIPS, then the overall fluctuations could be greater, especially for a young retiree (without access to beneficial wage inflation) with a long investment horizon stretching out to beginning SS and using all LTPZ, for example.

I'm seeing this tendency with a moderate overall TIPS duration of only 12 years. So, monthly re-balancing might be needed, in some cases, rather than quarterly. Or maybe separating the WBS from the TIPS for some rebalance calculations and just rebalancing with included TIPS semi-annually, for example. But my own TIPS % deviation hasn't exceeded 1%, so far.

Actually, some more recent commentary by Sharpe did propose a monthly re-balance if needed.
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

djm2001 wrote: Wed Oct 06, 2021 9:44 am
pascalwager wrote: Tue Oct 05, 2021 10:04 pm The only problem was being forced to mainly use the Vanguard Global Bonds Index Fund for my bonds. It holds a fixed 70/30 US/intl bonds. In other accounts I used only BNDX for bonds, but could only improve to 66/34.

Vanguard was using a more investor-acceptable 30% intl bonds. It's not to meant to represent market cap and I wouldn't use it except by necessity.
Do you have a link to the Vanguard Global Bond Index fund that you selected? The only one I could find was this foreign-domiciled one. Btw, the Market Allocation chart in that link seems funny -- the pie chart shows US as being almost exactly 50%, but the table alongside says the US is 41.3%. :? Agreed. Also, see above.

Did your previous "total market funds" portfolio already include (non-TIPS) bonds? Yes, most previous was about 50% TIPS and the rest STIG (a stand-in for ST TIPS).
pascalwager wrote: Tue Oct 05, 2021 10:04 pm In retirement, at least, I personally like the idea that the market is determining my stock/bond ratio and I'm not depending on back-testing for basic investment decisions.
+1. I fancifully think of WBS as turning the world into a supercomputer that predicts the distribution of future outcomes and then outputs the allocation best poised to address that outcome distribution. WBS does seem like a great even-keeled, "everything considered" approach that smooths out some of the idiosyncratic biases that an individual may get hung up on. Sharpe would say it's actual investing as opposed to betting.

re: WBS in retirement... I do have this vague but persistent worry in the back of my mind that the average retiree is probably not the average investor, specifically regarding the accounting of personal human capital and the general equilibrium story that would balance out the underweighting of bonds by accumulators. But I don't have enough mental clarity on this to make me deviate from WBS just yet. (If I would deviate, perhaps it would be to increasingly overweight global bonds relative to WBS allocation between retirement and death.) I would say that individual investors are never the average investor since 90% of the market is traded by large entities.
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

djm2001 wrote: Wed Oct 06, 2021 9:44 am re: WBS in retirement... I do have this vague but persistent worry in the back of my mind that the average retiree is probably not the average investor, specifically regarding the accounting of personal human capital and the general equilibrium story that would balance out the underweighting of bonds by accumulators. But I don't have enough mental clarity on this to make me deviate from WBS just yet. (If I would deviate, perhaps it would be to increasingly overweight global bonds relative to WBS allocation between retirement and death.)

Sharpe dislikes glide-paths because they amount to rebalancing which implies inefficient markets. And the bonds are part of your market risk portfolio. Sharpe intends that the TIPS allocation be used to reduce portfolio risk in accordance with MPT.

The 29% "extraneous stocks" portfolio already has some exposure to market risk. Did you decide to factor that into your WBS equity allocation, or ignore it?

Mostly ignored it regarding the reason you specified, but only to the extent that $820k is actually ignorable.
guppyguy
Posts: 465
Joined: Tue Jan 30, 2018 3:24 pm

Re: Bill Sharpe's preferred portfolio

Post by guppyguy »

Refresh my memory, what does Sharps say about withdrawal rates/payout policy?
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

Thank you for your detailed replies.
pascalwager wrote: Thu Oct 07, 2021 12:36 pm When Sharpe discusses (in RISMAT) WBS re-balancing, he's addressing the fluctuations of the four-fund S/B proxy only. However, if you're also holding duration-matched TIPS, then the overall fluctuations could be greater, especially for a young retiree (without access to beneficial wage inflation) with a long investment horizon stretching out to beginning SS and using all LTPZ, for example.

I'm seeing this tendency with a moderate overall TIPS duration of only 12 years. So, monthly re-balancing might be needed, in some cases, rather than quarterly. Or maybe separating the WBS from the TIPS for some rebalance calculations and just rebalancing with included TIPS semi-annually, for example. But my own TIPS % deviation hasn't exceeded 1%, so far.

Actually, some more recent commentary by Sharpe did propose a monthly re-balance if needed.
Do you have a link or pointer to Sharpe's commentary that suggests monthly rebalancing might be necessary? The implication seems to be that one would rebalance between TIPS and WBS over time, which seems at odds with Sharpe's "lockbox strategy" advice.

Putting my Cochrane hat on (i.e., general equilibrium, payoff streams, "it's all a giant insurance market"), I would not rebalance between TIPS and WBS. In theory, I'd use the following strategy instead:

Portfolio construction: Buy I-bonds and duration-matched TIPS to neutralize the portion of my future net liability stream that is in excess of the average liability stream, and invest the remainder in WBS.

Withdrawal strategy: Pay for average expenses using WBS (by selling "units" of WBS, thus preserving the prevailing WBS allocation) and pay for excess-of-average expenses using TIPS (for expenses with a predictable schedule) and I-bonds (for expenses with an unpredictable schedule).

The idea is simple -- the TIPS/I-bonds would offset my consumption stream, transforming it into the average consumption stream... for which WBS is the optimal portfolio!

This strategy is just theoretical. I don't actually know what the global average liability stream looks like, nor do I necessarily even know for sure what all my own liabilities will be. In practice, my real-life portfolio weights WBS and TIPS by market cap proportions... which means that I use a broad TIPS fund as roughly 1% of my portfolio. My plan is to convert some of this market portfolio into a pool of I-bonds over the next few years to fund future unexpected or unusual expenses (i.e., a large-ish emergency fund).
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

djm2001 wrote: Fri Oct 08, 2021 8:45 am Thank you for your detailed replies.
pascalwager wrote: Thu Oct 07, 2021 12:36 pm When Sharpe discusses (in RISMAT) WBS re-balancing, he's addressing the fluctuations of the four-fund S/B proxy only. However, if you're also holding duration-matched TIPS, then the overall fluctuations could be greater, especially for a young retiree (without access to beneficial wage inflation) with a long investment horizon stretching out to beginning SS and using all LTPZ, for example.

I'm seeing this tendency with a moderate overall TIPS duration of only 12 years. So, monthly re-balancing might be needed, in some cases, rather than quarterly. Or maybe separating the WBS from the TIPS for some rebalance calculations and just rebalancing with included TIPS semi-annually, for example. But my own TIPS % deviation hasn't exceeded 1%, so far.

Actually, some more recent commentary by Sharpe did propose a monthly re-balance if needed.
Do you have a link or pointer to Sharpe's commentary that suggests monthly rebalancing might be necessary? (Can't remember where, but think it was in his blog).

The implication seems to be that one would rebalance between TIPS and WBS over time, which seems at odds with Sharpe's "lockbox strategy" advice.

See RISMAT-7, page 38.

The colored graph shows a percentage allocation to TIPS for different risk levels. (My personal risk level happens to be 100-27 = 73.) Someone with no TIPS would have a risk level of 100.

So, I don't see how there could be any question about rebalancing between TIPS and WBS. RISMAT-7 isn't at all about using TIPS for liability matching but rather as an overall-portfolio component. [I recently sold $94k in my Extraneous Stocks account and moved 27% into LTPZ (30%) and SCHP (70%) and the rest into WBS.]

In RISMAT-7, Sharpe even advocates hiring an advisor for the single purpose of determining the percentage of WBS + TIPS to be used for TIPS.

That being said, it doesn't mean that your personalized approach (below) isn't just as good or even better, for you.

Putting my Cochrane hat on (i.e., general equilibrium, payoff streams, "it's all a giant insurance market"), I would not rebalance between TIPS and WBS. In theory, I'd use the following strategy instead:

Portfolio construction: Buy I-bonds and duration-matched TIPS to neutralize the portion of my future net liability stream that is in excess of the average liability stream, and invest the remainder in WBS.

Withdrawal strategy: Pay for average expenses using WBS (by selling "units" of WBS, thus preserving the prevailing WBS allocation) and pay for excess-of-average expenses using TIPS (for expenses with a predictable schedule) and I-bonds (for expenses with an unpredictable schedule).

The idea is simple -- the TIPS/I-bonds would offset my consumption stream, transforming it into the average consumption stream... for which WBS is the optimal portfolio!

This strategy is just theoretical. I don't actually know what the global average liability stream looks like, nor do I necessarily even know for sure what all my own liabilities will be. In practice, my real-life portfolio weights WBS and TIPS by market cap proportions... which means that I use a broad TIPS fund as roughly 1% of my portfolio. My plan is to convert some of this market portfolio into a pool of I-bonds over the next few years to fund future unexpected or unusual expenses (i.e., a large-ish emergency fund).
VT 60% / VFSUX 20% / TIPS 20%
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Bill Sharpe's preferred portfolio

Post by pascalwager »

pascalwager wrote: Fri Oct 08, 2021 3:49 pm
djm2001 wrote: Fri Oct 08, 2021 8:45 am Thank you for your detailed replies.
pascalwager wrote: Thu Oct 07, 2021 12:36 pm When Sharpe discusses (in RISMAT) WBS re-balancing, he's addressing the fluctuations of the four-fund S/B proxy only. However, if you're also holding duration-matched TIPS, then the overall fluctuations could be greater, especially for a young retiree (without access to beneficial wage inflation) with a long investment horizon stretching out to beginning SS and using all LTPZ, for example.

I'm seeing this tendency with a moderate overall TIPS duration of only 12 years. So, monthly re-balancing might be needed, in some cases, rather than quarterly. Or maybe separating the WBS from the TIPS for some rebalance calculations and just rebalancing with included TIPS semi-annually, for example. But my own TIPS % deviation hasn't exceeded 1%, so far.

Actually, some more recent commentary by Sharpe did propose a monthly re-balance if needed.
Do you have a link or pointer to Sharpe's commentary that suggests monthly rebalancing might be necessary? (Can't remember where, but think it was in his blog).

The implication seems to be that one would rebalance between TIPS and WBS over time, which seems at odds with Sharpe's "lockbox strategy" advice.

See RISMAT-7, page 38.

The colored graph shows a percentage allocation to TIPS for different risk levels. (My personal risk level happens to be 100-27 = 73.) Someone with no TIPS would have a risk level of 100.

So, I don't see how there could be any question about rebalancing between TIPS and WBS. RISMAT-7 isn't at all about using TIPS for liability matching but rather as an overall-portfolio component. [I recently sold $94k in my Extraneous Stocks account and moved 27% into LTPZ (30%) and SCHP (70%) and the rest into WBS.]

In RISMAT-7, Sharpe even advocates hiring an advisor for the single purpose of determining the percentage of WBS + TIPS to be used for TIPS.

That being said, it doesn't mean that your personalized approach (below) isn't just as good or even better, for you.

Putting my Cochrane hat on (i.e., general equilibrium, payoff streams, "it's all a giant insurance market"), I would not rebalance between TIPS and WBS. In theory, I'd use the following strategy instead:

Portfolio construction: Buy I-bonds and duration-matched TIPS to neutralize the portion of my future net liability stream that is in excess of the average liability stream, and invest the remainder in WBS.

Withdrawal strategy: Pay for average expenses using WBS (by selling "units" of WBS, thus preserving the prevailing WBS allocation) and pay for excess-of-average expenses using TIPS (for expenses with a predictable schedule) and I-bonds (for expenses with an unpredictable schedule).

The idea is simple -- the TIPS/I-bonds would offset my consumption stream, transforming it into the average consumption stream... for which WBS is the optimal portfolio!

This strategy is just theoretical. I don't actually know what the global average liability stream looks like, nor do I necessarily even know for sure what all my own liabilities will be. In practice, my real-life portfolio weights WBS and TIPS by market cap proportions... which means that I use a broad TIPS fund as roughly 1% of my portfolio. My plan is to convert some of this market portfolio into a pool of I-bonds over the next few years to fund future unexpected or unusual expenses (i.e., a large-ish emergency fund).
I just remembered. Sharpe isn't on board with conventional re-balancing (selling high, buying low), so maybe I'm wrong. Maybe you start out with a particular % of TIPS and WBS and then do AAA rebalancing. I need to clarify with Bill.
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Bill Sharpe's preferred portfolio

Post by djm2001 »

pascalwager wrote: Fri Oct 08, 2021 6:10 pm I just remembered. Sharpe isn't on board with conventional re-balancing (selling high, buying low), so maybe I'm wrong. Maybe you start out with a particular % of TIPS and WBS and then do AAA rebalancing. I need to clarify with Bill.
Here's a relevant excerpt from Sharpe's RISMAT, Chapter 15:
William Sharpe, RISMAT, Chapter 15, Lockboxes wrote: Lockbox Contents

To begin, let's focus on the provision of income in a specific future year – for example, year 10 (9 years hence). To keep things simple, let's also assume that both Bob and Sue will be alive at that time. Today we create Lockbox10 to provide their income in year 10. And today we put into the box, chosen amounts of some or all of three types of investments:
  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
The box is to be sealed after the contents are put in, then opened at maturity (here, 9 years).
My interpretation is that he is recommending:
  1. some form of liability matching -- e.g., matching year 10's expenses to a portfolio designed for year 10, which includes TIPS of specific duration.
  2. no rebalancing between WBS and TIPS -- each lockbox is sealed and left undisturbed until it is opened (i.e., sold) and used.
I use duration and maturity interchangeably above since he's talking about zero-coupon TIPS, where duration equals maturity.

(I still don't understand m-shares...)

Another excerpt where Sharpe talks about TIPS and liability matching is in Chapter 6:
William Sharpe, RISMAT, Chapter 6, Inflation-Protected Investments wrote: TIPS Mutual Funds and ETFs

While it is possible to purchase TIPS directly, many investors choose instead to invest in a fund that holds TIPS of different maturities.

An advantage for direct investment is the ability to create a ladder of such securities, with different maturities that will pay desired amounts at each of a number of future dates. When yields were higher it was more difficult to accomplish this, but clever people on Wall Street (seeking a chance to make a profit) made it easier by purchasing TIPS, then using them to back new instruments, each of which paid a given real amount on a single date.

...

On the other hand, the timing of the cash flows from the portfolio held by a TIPS fund or ETF may not be ideal for a particular investor. TIPS mutual funds and ETFs typically hold all available securities, with maturities from less than one year to as much as 30 years. As indicated earlier, in bond parlance these funds provide a ladder of bond cash flows. A common measure of the timing of cash flows from a bond or bond fund is its duration – a weighted average of the future times at which cash would be received, using weights based on the present values of the cash flows.

...

For many, investment in one or two of funds with different mixes of maturities should suffice. But some may wish to purchase individual TIPS securities to better match likely horizons despite the increased effort, bookkeeping, tax accounting, etc.
The handling of TIPS -- duration-matched individual TIPS/STRIPS vs. mixing two TIPS funds of different duration (one long, one short) vs. single broad TIPS fund -- is up to an individual investor's sophistication. I currently use the simplest option (i.e., one broad TIPS fund) myself. But my interpretation of the parts of RISMAT that I have read is that, complexity concerns aside, the ideal use of TIPS is a duration-matched TIPS ladder that is not rebalanced against the WBS component.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Post Reply