Updated figures and results on slice and dice vs TSM

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Fri Apr 15, 2011 7:31 am

What does the historical data show?
Unfortunately I did not and do not have the data. Would be good to see something similar using international data.

My understanding from others here is that academics have looked at this and see similar results for the three factor model on international stocks, but I have not looked up the papers to confirm. To the extent the three factor model holds I would expect similarly that 4x25, TSM/SCV etc would have similar performance. But I don't have the data to check.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

cb474
Posts: 801
Joined: Tue Jan 19, 2010 6:32 am

Post by cb474 » Fri Apr 15, 2011 7:14 pm

Yes, I realize you didn't have the data. I was just trying to explain to living2notWork the question I had been asking further up in the thread, that he quoted. He seemed to think I was asking a question about specific funds and not about the more general theory/history of it all.

My reading in the end also suggested that the three factor model holds up on the international side. And so I made the same assumption you did, although in the end it is an assumption.

Anyway, I don't mean to detract from how valuable the original post in this thread is. I still refer to it, when I'm thinking about these matters. Thanks Rodc.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Sat Apr 16, 2011 5:10 pm

cb474 wrote:Yes, I realize you didn't have the data. I was just trying to explain to living2notWork the question I had been asking further up in the thread, that he quoted. He seemed to think I was asking a question about specific funds and not about the more general theory/history of it all.

My reading in the end also suggested that the three factor model holds up on the international side. And so I made the same assumption you did, although in the end it is an assumption.

Anyway, I don't mean to detract from how valuable the original post in this thread is. I still refer to it, when I'm thinking about these matters. Thanks Rodc.
Sorry, I misunderstood.

Rod
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Bongleur
Posts: 2062
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur » Wed Apr 20, 2011 6:44 am

From the OP in 2007:
"Figure 1: 80 years of asset class returns on top of all possible portfolios formed from the F-F benchmark portfolios and 90-day T-Bills (more on how things change with different types of bonds shortly in another post)."

Got a link to the post talking about how it changes with other than 90 day T bills ? Did you ever do it?

Is it possible to run the numbers with something subtracted from the historical bond returns, to correspond with what bond yields are currently expected to yield going forwards?

Although you would have to subtract something from the equity yields also; and that might be more problematic than recognizing the end of a secular phase in bonds.
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Wed Apr 20, 2011 7:57 am

Bongleur wrote:From the OP in 2007:
"Figure 1: 80 years of asset class returns on top of all possible portfolios formed from the F-F benchmark portfolios and 90-day T-Bills (more on how things change with different types of bonds shortly in another post)."

Got a link to the post talking about how it changes with other than 90 day T bills ? Did you ever do it?

Is it possible to run the numbers with something subtracted from the historical bond returns, to correspond with what bond yields are currently expected to yield going forwards?

Although you would have to subtract something from the equity yields also; and that might be more problematic than recognizing the end of a secular phase in bonds.
Hi Bongler,

Here is what I posted:

http://www.bogleheads.org/forum/viewtopic.php?t=9759

I have not tried running something adjusted for current yields.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Bongleur
Posts: 2062
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur » Thu Apr 21, 2011 6:18 am

What about what you could call the Relative Rebalancing Costs (RRC) of more slices?

It seems to me that more slices leads to more frequent rebalancing. It takes longer to exceed the rebalancing band on one large quantity of money than on, say, four uncorrelated funds each containing 1/4 the quantity of money. ???

If you have 2 funds, then your RRC is 2, because you must rebalance only from stock to bond and from bond to stock.

If you have 3 funds, then your RRC is 6.

If you have 6 funds... its 6! (six factorial), I think ???

So I'd like to know when the number of slices starts to produce a significant reduction in the theoretical returns because of the tax loss cost.

And every fund beyond that point probably adds a more than linear increase in reduction.

So I think that the slice & dice bonus might be significantly less than the calculations shown in this thread indicate ???
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

Bongleur
Posts: 2062
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur » Thu Apr 21, 2011 6:52 am

The returns you have calculated are strictly "accumulation" phase.

If a person is going to shift to a different AA for the retirement phase, then I think your calculated average returns might probably be overstated or need more caveats.

And if the slice & dice actually does perform to the point where the retiree does in fact have "enough" money, then he certainly will be changing his AA to have less risk & return.

There are 3 ways to shift from accumulation AA to retirement AA:

1) gradually shift during the 5-10 years prior to retirement.

The greatest gains in portfolio value occur in those last years because you are compounding a larger sum. But your averages do not account for not actually using the stated slice & dice AA during those last 5-10 years. The investor is moving to lower & lower risk & yield.

2) shift the AA all at once at retirement and pay a tax loss.

This will result in a smaller starting portfolio.

3) gradually shift during the 5-10 years after retirement.

This exposes the retiree to Sequence of Returns risk.
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

User avatar
rustymutt
Posts: 3700
Joined: Sat Mar 07, 2009 12:03 pm
Location: Oklahoma

Post by rustymutt » Thu Apr 21, 2011 6:53 am

Bongleur wrote:What about what you could call the Relative Rebalancing Costs (RRC) of more slices?

It seems to me that more slices leads to more frequent rebalancing. It takes longer to exceed the rebalancing band on one large quantity of money than on, say, four uncorrelated funds each containing 1/4 the quantity of money. ???

If you have 2 funds, then your RRC is 2, because you must rebalance only from stock to bond and from bond to stock.

If you have 3 funds, then your RRC is 6.

If you have 6 funds... its 6! (six factorial), I think ???

So I'd like to know when the number of slices starts to produce a significant reduction in the theoretical returns because of the tax loss cost.

And every fund beyond that point probably adds a more than linear increase in reduction.

So I think that the slice & dice bonus might be significantly less than the calculations shown in this thread indicate ???
I have sliced and diced for 15 years now, and can tell you flat out the my cost for trading is not out of line with a 3 fund system. I don't rebalance but once a year, and sometimes not that often. I use ETF rather than funds. This help keep my ER low and the trades for my Vanguard ETF's are now free. My last rebalance, had $8 of trading fees involved.
I'm amazed at the wealth of Knowledge others gather, and share over a lifetime of learning. The mind is truly unique. It's nice when we use it!

Bongleur
Posts: 2062
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur » Thu Apr 21, 2011 8:18 am

No, I'm talking about the tax loss when you sell appreciated NAV that has risen above the band.

Not everyone has all his money in an IRA...

EDIT: someone who has charting software can find out how often the TSM/TBM portfolio went "out of band" in the last 30 years.

Then compare to a many-fund slice & dice.
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

User avatar
ruralavalon
Posts: 12744
Joined: Sat Feb 02, 2008 10:29 am
Location: Illinois

Post by ruralavalon » Thu Apr 21, 2011 8:23 am

Thanks RodC for the analysis and graphs.
RodC wrote: The point I am try to make is you don't have to be careful in order to do well. I often see people agonize over decision of exactly how to tilt, and my point is just pick one then go off and do something more interesting with your life. (or course I don't always take my own advice. )
We are domestically 80% TSM, 10% SCV and 10% REIT, so will stop thinking about this :) .
"Everything should be as simple as it is, but not simpler." - Albert Einstein | Wiki article link:Getting Started

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Fri Apr 22, 2011 7:55 am

Bongleur wrote:No, I'm talking about the tax loss when you sell appreciated NAV that has risen above the band.

Not everyone has all his money in an IRA...

EDIT: someone who has charting software can find out how often the TSM/TBM portfolio went "out of band" in the last 30 years.

Then compare to a many-fund slice & dice.
Tax loss is complicated and very much tied to your personal situation. If you have significant taxable investments it certainly would pay to carefully analyze your own situation. That is well beyond the intention of this thread, which as you note is mostly germane to someone investing primarily in a tax sheltered account.

There have been many threads on that topic that presumably one can find using the search function.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Fri Apr 22, 2011 8:12 am

Bongleur wrote:The returns you have calculated are strictly "accumulation" phase.

If a person is going to shift to a different AA for the retirement phase, then I think your calculated average returns might probably be overstated or need more caveats.

And if the slice & dice actually does perform to the point where the retiree does in fact have "enough" money, then he certainly will be changing his AA to have less risk & return.

There are 3 ways to shift from accumulation AA to retirement AA:

1) gradually shift during the 5-10 years prior to retirement.

The greatest gains in portfolio value occur in those last years because you are compounding a larger sum. But your averages do not account for not actually using the stated slice & dice AA during those last 5-10 years. The investor is moving to lower & lower risk & yield.

2) shift the AA all at once at retirement and pay a tax loss.

This will result in a smaller starting portfolio.

3) gradually shift during the 5-10 years after retirement.

This exposes the retiree to Sequence of Returns risk.
Well, if you wish it is worse than that because it is a lump sum analysis.

I am not inclined to go back and repeat with ever finer detail, in part because I'm not a fan of false precision and in part because this analysis makes the point I set out to make. Certainly there are other questions one can attempt to answer and I encourage anyone so interested to do so.

I think what you will find is that adding periodic investments will make differences even smaller between portfolios.

As you get into retirement and add a heavy dose of bonds/cash the differences again get smaller.

So, I don't think adding detail is going to change the basic conclusions.

But again, anyone interested in doing such an analysis should do so.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Bongleur
Posts: 2062
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur » Fri Apr 22, 2011 2:13 pm

>Tax loss is complicated and very much tied to your personal situation.
>

The amount is an individual thing; but the number of taxable transactions is different for each slice & dice. Knowing that one flavor has twice the number of potentially taxable transactions than another flavor would be useful qualitative data.

It would mean that if both showed the same returns from the usual sort of calculation, then the one with fewer transactions actually yields more.

>That is well beyond the intention of this thread, which as you note is mostly germane to someone investing primarily in a tax sheltered account.
>

EVERY thread that analyzes returns is biased towards tax sheltered accounts.
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

Rodc
Posts: 13601
Joined: Tue Jun 26, 2007 9:46 am

Post by Rodc » Fri Apr 22, 2011 3:12 pm

Bongleur wrote:>Tax loss is complicated and very much tied to your personal situation.
>

The amount is an individual thing; but the number of taxable transactions is different for each slice & dice. Knowing that one flavor has twice the number of potentially taxable transactions than another flavor would be useful qualitative data.

It would mean that if both showed the same returns from the usual sort of calculation, then the one with fewer transactions actually yields more.

>That is well beyond the intention of this thread, which as you note is mostly germane to someone investing primarily in a tax sheltered account.
>

EVERY thread that analyzes returns is biased towards tax sheltered accounts.
I agree. If taxable fewer taxable events all else equal is better.

I personally am heavily biased towards tax sheltered analysis as nearly all my investments are in tax sheltered accounts. For better or worse.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

Bongleur
Posts: 2062
Joined: Fri Dec 03, 2010 10:36 am

Post by Bongleur » Fri Apr 22, 2011 5:25 pm

We only have about 1/3 tax advantaged, so I am always affected by the bias. I need to dig deeper to find out if conclusions apply to my situation.
Seeking Iso-Elasticity. | Tax Loss Harvesting is an Asset Class. | A well-planned presentation creates a sense of urgency. If the prospect fails to act now, he will risk a loss of some sort.

Gotta Tri
Posts: 8
Joined: Tue Mar 22, 2011 9:36 pm

Post by Gotta Tri » Fri Apr 22, 2011 9:05 pm

This thread is great, thanks for the dialogue and information. Awesome to be just a fly on the wall observing this.

cb474
Posts: 801
Joined: Tue Jan 19, 2010 6:32 am

Post by cb474 » Sat Apr 23, 2011 1:54 am

Bongleur wrote:The amount is an individual thing; but the number of taxable transactions is different for each slice & dice. Knowing that one flavor has twice the number of potentially taxable transactions than another flavor would be useful qualitative data.
I think it's more complicated than this and a simple generalization does not apply equally to all possible asset allocations. The more slices you have, the more specific taxable events you will have. But each slice will be smaller, so each time you rebalance, you will be rebalancing a smaller amount and hence the taxes will be proportionally smaller for each "event."

Depending on what sort of asset allocations you're comparing, the effect of more taxable events, but for smaller amounts, could be a wash. Or not. But I don't think you can generalize as simply as you'd like to.
Bongleur wrote:It would mean that if both showed the same returns from the usual sort of calculation, then the one with fewer transactions actually yields more.
Assuming that both show the same returns is, again, a simplification that really begs the question.

Since those who accept the arguments for the value and small-cap premiums don't believe that a TSM lumper equity allocation and a value and small-cap tilted equity allocation have the same potential return, they would not assume that there is an apples to apples comparison that you could make between a TSM and a slice and dice portfolio. So the question is, does the potential higher return of the slice and dice equity allocation more than compensate for the potential tax disadvantages?

It's nice to say, "if both showed the same returns," but that essentially abstract idea may not apply in any practical reality.

Also, keep in mind that when you take money out of a TSM allocation, either as a withdrawal or to rebalance to your bond allocation, you are going to (one way or another) pay the capital gains taxes on whatever return you've earned. If you pay capital gains taxes sooner, because you're reblancing in a slice and dice portfolio, to some extent you are simply incurring your taxable event on a portion of your returns sooner, rather than later. But you are not avoiding it with a single slice TSM equity allocation.

In other words, again, all taxable "events" are not equal. More but smaller events may not add up to more taxes than fewer but larger events. Or, another way to think about it is that in a slice and dice portfolio you are potentially incurring more taxable events earlier, but you are also raising your cost basis and lowering the taxes you will pay when you eventually withdraw funds (or rebalance to the bond allocation).

Lastly, the potential for tax loss harvesting, in a slice and dice portfolio, is greater. This will also affect the tax consequences of the slice and dice portfolio versus TSM or fewer slices. Of course, no one can predict how this will play out, so, once again, it's hard to generalize about the tax consequences of different more or less sliced and diced equity allocations.

It is also perhaps worth noting that many of the financial writers, who are highly respected in this forum (Rick Ferri, Allan Roth, Larry Swedroe, William Bernstein, etc.), do not seem to consider the tax consequences of a slice and dice portfolio in a taxable account to pose a prohibitive cost, outweighing the potential benefits of a higher potential return.

User avatar
tc101
Posts: 2968
Joined: Tue Feb 20, 2007 3:18 pm
Location: Atlanta - Retired in 2004 at age 54

Re: Updated figures and results on slice and dice vs TSM

Post by tc101 » Sat Mar 09, 2013 12:12 pm

I'm reading over this fascinating discussion that took place from 2007-2011.

Is there any new research that would disagree with these results in the OP
The primary results are that:

1) Slice and dice has fairly consistently provided a risk/volatility adjusted benefit.
2) Any two vaguely similar slice and dice portfolios have about the same long term returns and volatility; how you get your tilt does not matter very much, and to the extent that it does, it is not easy to predict ahead of time which tilt is best.
In the wiki article on slice and dice, John Bogle says the thinks it is just a temporary phenomena from the time periods studied, and can not be counted on in the future.
. | The most important thing you should know about me is that I am not an expert.

dkturner
Posts: 1335
Joined: Sun Feb 25, 2007 7:58 pm

Re: Updated figures and results on slice and dice vs TSM

Post by dkturner » Sat Mar 09, 2013 1:32 pm

tc101 wrote:I'm reading over this fascinating discussion that took place from 2007-2011.

Is there any new research that would disagree with these results in the OP
The primary results are that:

1) Slice and dice has fairly consistently provided a risk/volatility adjusted benefit.
2) Any two vaguely similar slice and dice portfolios have about the same long term returns and volatility; how you get your tilt does not matter very much, and to the extent that it does, it is not easy to predict ahead of time which tilt is best.
In the wiki article on slice and dice, John Bogle says the thinks it is just a temporary phenomena from the time periods studied, and can not be counted on in the future.

I believe John Bogle's comment is based on a speech he delivered called "The Telltale Chart". Being kind to Mr. Bogle, let's just say the research he cites in the speech is a little light on statistical vigor.

User avatar
tc101
Posts: 2968
Joined: Tue Feb 20, 2007 3:18 pm
Location: Atlanta - Retired in 2004 at age 54

Re: Updated figures and results on slice and dice vs TSM

Post by tc101 » Sun Mar 10, 2013 9:15 am

OK, in reading over this thread I am convinced of the value of slice and dice in spite of what Mr. Bogle says about it.

I am about to get a windfall from the sale of a house and plan to invest that new money according to slice and dice principles. I will leave money that is already invested as it is.

Early in the thread someone said not to agonize over the details, but just pick one formula and stick with it. Here is my formula

25% S&P 500 Admiral
25% Small Value Admiral
25% FTSE All-World ex-US Index Admiral
25% FTSE All-World ex-US Small-Cap Index ETF

The Small Value will go in my IRA, the rest will be in taxable.
The only ETF wil be the small cap international. I don't see any reason to use ETFs for the rest because they are no better than admiral funds.

I am not 50% international so I will make adjustments by adding to my already existing TSM. I think rebalancing will be simplier if I just keep equal amounts in these 4 slice and dice funds.

Do this seem like a good plan?
. | The most important thing you should know about me is that I am not an expert.

dbr
Posts: 26139
Joined: Sun Mar 04, 2007 9:50 am

Re: Updated figures and results on slice and dice vs TSM

Post by dbr » Sun Mar 10, 2013 9:20 am

Sure it's a good plan except the idea that you want to manage two portfolios seems odd. You can't cover up the fact that all your money is all your money and the allocation is what you get from the whole thing. In reality you are tilting less than indicated in the above plan and more than you were before. That's fine if that's what you want to do, but why make the mental separation?

YDNAL
Posts: 13774
Joined: Tue Apr 10, 2007 4:04 pm
Location: Biscayne Bay

Re: Updated figures and results on slice and dice vs TSM

Post by YDNAL » Sun Mar 10, 2013 9:47 am

tc101 wrote:OK, in reading over this thread I am convinced of the value of slice and dice in spite of what Mr. Bogle says about it.

I am about to get a windfall from the sale of a house and plan to invest that new money according to slice and dice principles. I will leave money that is already invested as it is.
You are not very "convinced" when old money is invested differently than windfall from sale of residence. There is a certain amount of minutae in most every thing we read - proceed accordingly.

I would ask myself, tc101, what Asset Allocation is expectantly going to help meet my goals and necessities in retirement (or now if retired). Go with that!

2¢

ps. what made you go looking for a 2007 thread?... aren't you concerned with a "bond bubble" today (6 years later)? :D
Landy | Be yourself, everyone else is already taken -- Oscar Wilde

User avatar
tc101
Posts: 2968
Joined: Tue Feb 20, 2007 3:18 pm
Location: Atlanta - Retired in 2004 at age 54

Re: Updated figures and results on slice and dice vs TSM

Post by tc101 » Sun Mar 10, 2013 9:50 am

Sure it's a good plan except the idea that you want to manage two portfolios seems odd
There would be a lot of taxes if I sold the pre existing funds to create the new plan. I have been investing at Vanguard since 1985, and there are large cap gains in TSM and tax managed international.

I'll think about what you say though. Maybe there is a way to sell just some of the old funds and create a new plan.

I'm 63 and retired.
. | The most important thing you should know about me is that I am not an expert.

dbr
Posts: 26139
Joined: Sun Mar 04, 2007 9:50 am

Re: Updated figures and results on slice and dice vs TSM

Post by dbr » Sun Mar 10, 2013 10:03 am

If you have TSM and tax managed international already, why not just add as much small cap value from your windfall as you need to hit the tilt you want, up to all of it?

Also, I didn't look in detail before, but why S&P500 instead of TSM?

Default User BR
Posts: 7501
Joined: Mon Dec 17, 2007 7:32 pm

Re: Updated figures and results on slice and dice vs TSM

Post by Default User BR » Sun Mar 10, 2013 12:27 pm

dbr wrote:Also, I didn't look in detail before, but why S&P500 instead of TSM?
TSM is not very slicey! Why have those small and mids polluting the large blend category? :)

It, of course, doesn't matter too much. For the record, I use S&P 500[401(k)] + MGC[taxable]) for my LCB holdings.


Brian

User avatar
tc101
Posts: 2968
Joined: Tue Feb 20, 2007 3:18 pm
Location: Atlanta - Retired in 2004 at age 54

Re: Updated figures and results on slice and dice vs TSM

Post by tc101 » Sun Mar 10, 2013 2:33 pm

If you have TSM and tax managed international already, why not just add as much small cap value from your windfall as you need to hit the tilt you want, up to all of it?
Good idea. I'm going to look into that. Did you intentionally not mention international SC, or was that an oversight?
Also, I didn't look in detail before, but why S&P500 instead of TSM?
My original idea was to add 2 new funds so I could have the money very neatly divided into 25% each funds. Since I already have TSM and tax managed international, the new funds were going to be S&P500 and total international.

I see 1 major advantage to doing it he way suggested here:
1 - I could more easily get a bigger slice into SCV and international SC.

I see 2 advantages to doing it the way I originally planned:
1 - It would be better for taxes long term because I could rebalance out of the new funds which would have less cap gains than the original funds.
2 - It would be easier to see exactly what I was doing with rebalancing because there would be 4 funds that would always be rebalanced to be exactly equal.
. | The most important thing you should know about me is that I am not an expert.

Frank2012
Posts: 50
Joined: Fri Jan 25, 2013 7:16 pm

Re: Updated figures and results on slice and dice vs TSM

Post by Frank2012 » Mon Mar 11, 2013 8:02 am

Rodc,

Very impressive post and well thought out. Looks like you have definitively proven the superiority of slice & dice over the three-fund portfolio.

But for me, I can't handle a lot of funds, so I'm sticking with my three-fund portfolio (50% Total Bond, 35% Total US, 15% Total International) which I rebalance once a year. I may lose the opportunity of a percentage or two for not slicing and dicing, but it's way better than what I was doing before (stocks, trading, managed funds), and I think I'm much more likely to stay the course.

dkturner
Posts: 1335
Joined: Sun Feb 25, 2007 7:58 pm

Re: Updated figures and results on slice and dice vs TSM

Post by dkturner » Mon Mar 11, 2013 9:43 am

Frank2012 wrote:Rodc,

Very impressive post and well thought out. Looks like you have definitively proven the superiority of slice & dice over the three-fund portfolio.

But for me, I can't handle a lot of funds, so I'm sticking with my three-fund portfolio (50% Total Bond, 35% Total US, 15% Total International) which I rebalance once a year. I may lose the opportunity of a percentage or two for not slicing and dicing, but it's way better than what I was doing before (stocks, trading, managed funds), and I think I'm much more likely to stay the course.
Interesting. Your three-fund portfolio is exactly the benchmark I use to measure the performance of my 50/35/15 slice & dice portfolio. I use about 5 funds, on average. Annualized returns (as of 12/31/2012) over the last 3,5,10 and 15 years: Vanguard fund benchmark 7.8, 3.2, 7.2 and 5.7. My actual slice & dice portfolio of Vanguard funds 9.2, 4.3, 8.3 and 7.5.

Slicing & dicing certainly has worked for me.

cb474
Posts: 801
Joined: Tue Jan 19, 2010 6:32 am

Re: Updated figures and results on slice and dice vs TSM

Post by cb474 » Mon Mar 11, 2013 5:53 pm

tc101,

I agree with others that you need to think about all of your assets together as one portfolio and come up with an overall asset allocation that suits your needs. It makes little sense to do otherwise. And in the end, you will be subject to the performance of all of your assets as a whole, anyway, so there's really no way of avoiding this. By pretending otherwise, you are in a sense just letting arbitrary factors determine your asset allocation. I think this is possibly the biggest error people make when considering their investments. Failing to see the big picture and think about how every small choice within that pictures effects every other part of the portfolio. There is only one total asset allocation. There are no mini-asset allocations within your portfolio. Either you make a deliberate choice about what that asset allocation is going to be. Or you let incidental factors and rationales determine it for you.

That aside, I don't really see the benefit of S&P 500 over TSM. TSM will have tax advantages in a taxable account (because of less turnover). If you believe in slice and dice, then TSM is for all intents and purposes the same as S&P 500, because the amount of mid and small equities in TSM isn't not significant. So the only material difference is the tax advantage of TSM.

As far as tax advantages for rebalancing, I believe as long as you use the specific shares method of cost basis calculation, you can rebalance with whichever shares (i.e. the new ones) you want and have the same advantage of selling shares with less cap gains.

I think the convenience of having the four equal funds for rebalancing purposes is 1) putting the cart before the horse (just make a spreadsheet to help yourself visualize things) 2) ignoring that those four funds are only part of your total assets and therefore don't really represent your asset allocation, per my first comments, and therefore entirely an illusion with regard to creating a simpler rebalancing situation.

With respect to the four funds you originally propose. You might also consider Vanguard's large value fund (VTRIX), which is unfortunately actively managed, or the iShares large value index fund EFV (popular amongst people in this forum). That might allow you to get a more even mix of value and small funds. This is sort of the archetypal thread on the four slice approach to doing this: http://www.bogleheads.org/forum/viewtop ... 10&t=38374

pkcrafter
Posts: 12568
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Updated figures and results on slice and dice vs TSM

Post by pkcrafter » Mon Mar 11, 2013 7:21 pm

TC wrote:
Do this seem like a good plan?
Hmm, I don't think so, TC. Continue with your boring moderated portfolio.


Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

sdrone
Posts: 108
Joined: Wed Jun 10, 2009 11:17 pm

Re: Updated figures and results on slice and dice vs TSM

Post by sdrone » Mon Mar 11, 2013 7:54 pm

I'm glad this post was replied to so that it popped up - I was not aware of it.

I've seen a couple of references to a thread on "TrevH's simplied slice and dice portfolio" using TSM and SV. My google foo seems to be failing - does anyone have a link handy to that thread?

mtwoy
Posts: 41
Joined: Tue Jun 19, 2012 10:39 am

Re: Updated figures and results on slice and dice vs TSM

Post by mtwoy » Mon Mar 11, 2013 7:59 pm

sdrone wrote:I'm glad this post was replied to so that it popped up - I was not aware of it.

I've seen a couple of references to a thread on "TrevH's simplied slice and dice portfolio" using TSM and SV. My google foo seems to be failing - does anyone have a link handy to that thread?
I believe this may be the one you are looking for.

http://www.bogleheads.org/forum/viewtop ... 10&t=38374

sdrone
Posts: 108
Joined: Wed Jun 10, 2009 11:17 pm

Re: Updated figures and results on slice and dice vs TSM

Post by sdrone » Mon Mar 11, 2013 8:20 pm

Hmmm - thanks, I'll go through there again. Maybe I'm misunderstanding what the thread topic was.

Default User BR
Posts: 7501
Joined: Mon Dec 17, 2007 7:32 pm

Re: Updated figures and results on slice and dice vs TSM

Post by Default User BR » Tue Mar 12, 2013 1:26 pm

sdrone wrote:Hmmm - thanks, I'll go through there again. Maybe I'm misunderstanding what the thread topic was.
The "four slices" referred to in the topic is Trev's simplified S/D portfolio.


Brian

User avatar
DueDiligence
Posts: 83
Joined: Mon Sep 09, 2013 3:57 pm

Re: Updated figures and results on slice and dice vs TSM

Post by DueDiligence » Fri Sep 13, 2013 3:02 pm

Here is an updated historical comparison of "Boglehead 3-Fund" and "Simplified Ultimate Buy and Hold" portfolio results through 2Q 2013.
The results are based on Yahoo finance historical data.
TBM/VBMFX is used for all cases at varying percentages to get the range of standard deviations shown in the figure.
All cases have 70%US equity and EOY rebalancing.

Image

The difference in historical CAGR is of order 0.2 to 0.4% for balanced portfolios at the longer historical times.
I'd be interested in reactions/criticisms etc.
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
cflannagan
Posts: 949
Joined: Sun Oct 21, 2007 11:44 am
Location: Palm Harbor, Florida

Re: Updated figures and results on slice and dice vs TSM

Post by cflannagan » Fri Sep 13, 2013 4:44 pm

DueDiligence wrote:Here is an updated historical comparison of "Boglehead 3-Fund" and "Simplified Ultimate Buy and Hold" portfolio results through 2Q 2013.
The results are based on Yahoo finance historical data.
TBM/VBMFX is used for all cases at varying percentages to get the range of standard deviations shown in the figure.
All cases have 70%US equity and EOY rebalancing.

Image

The difference in historical CAGR is of order 0.2 to 0.4% for balanced portfolios at the longer historical times.
I'd be interested in reactions/criticisms etc.
I was expecting bigger differences. Using Jon's backtester, even for only 5 years, I'm showing difference as large as .8. For 20 years, I had more than 1.0% difference.

What funds were you using?

I was using following setups:

2008-2012 (which Jon's Backtester came back saying 5 years. the backtester did not include 2013 data).

17.5% VTRIX, 17.5% VFSVX, 17.5% VFINX, 17.5% VISVX, 30% VBMFX for the SUBH
30% VBMFX, 35% VTSMX, 35% VGTSX

Let me know if anything is amissing.

Also, does Yahoo Finance normally include reinvested dividends?

User avatar
DueDiligence
Posts: 83
Joined: Mon Sep 09, 2013 3:57 pm

Re: Updated figures and results on slice and dice vs TSM

Post by DueDiligence » Fri Sep 13, 2013 5:35 pm

cflannagan: "I was expecting bigger differences. Using Jon's backtester, even for only 5 years, I'm showing difference as large as .8. For 20 years, I had more than 1.0% difference.
What funds were you using? I was using following setups:
2008-2012 (which Jon's Backtester came back saying 5 years. the backtester did not include 2013 data).
17.5% VTRIX, 17.5% VFSVX, 17.5% VFINX, 17.5% VISVX, 30% VBMFX for the SUBH
30% VBMFX, 35% VTSMX, 35% VGTSX
Let me know if anything is amissing.
Also, does Yahoo Finance normally include reinvested dividends?"

I used the same funds you referenced except:
(1) ETF VSS rather than VFSVX
(2) VTSMX rather than VFINX
However equities are 70% US rather than the the 50% you referenced.
But 70%US has better peformance than 50% for most timeframes in last 20 years.
Specifically I used 30% VBMFX, 24.5% VTSMX, 24.5% VISVX, 10.5% VTRIX and 10.5% VSS.
The Yahoo finance data that I used does include dividends and the results agree with Vanguard TR data for individual funds.
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

Trev H
Posts: 1871
Joined: Fri Mar 02, 2007 10:47 pm

Re: Updated figures and results on slice and dice vs TSM

Post by Trev H » Fri Sep 13, 2013 9:26 pm

17.5% VTRIX, 17.5% VFSVX, 17.5% VFINX, 17.5% VISVX...

My Equity allocation exactly... except I use Large Cap Index instead of 500 Index.

Get a few more of Mel"s unloved mid-caps that way :-)

Trev H

afan
Posts: 3354
Joined: Sun Jul 25, 2010 4:01 pm

What it you expect to be a net saver through retirement?

Post by afan » Sat Sep 14, 2013 2:21 pm

Tax deferred investments represent less that a quarter of my financial assets. I plan to live off required minimum distributions from tax deferred accounts and dividends from TSM, international total stock, and intermediate term munis in taxable accounts. That means that the choice is not "paying cap gains taxes now through slice and dice or later when I liquidate stocks in retirement". The choice is paying cap gain tax now due to the inherently higher turnover for small value portfolios, and the need to rebalance, or deferring cap gain taxes forever. My heirs would get the stepped up basis, and any gains in the portfolio at my death would escape taxation permanently.

For this reason, the tax drag on a slice and dice portfolio becomes highly significant. If I have to lose a large share of any gains to taxes it is not clear that I end up better off with the pretax advantage the small value tilt might produce. Given the unpredictable nature of the premium, and its tendancy to disappear for long periods of time, I wonder whether there is really any advantage.

It seems I get lower expenses, lower taxes, simplier taxes, and a larger share of my gains escape taxation altogether. Would I really be better off tilting and forsaking the tax advantages?
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

User avatar
cflannagan
Posts: 949
Joined: Sun Oct 21, 2007 11:44 am
Location: Palm Harbor, Florida

Re: Updated figures and results on slice and dice vs TSM

Post by cflannagan » Mon Sep 16, 2013 1:00 am

Trev H wrote:17.5% VTRIX, 17.5% VFSVX, 17.5% VFINX, 17.5% VISVX...

My Equity allocation exactly... except I use Large Cap Index instead of 500 Index.

Get a few more of Mel"s unloved mid-caps that way :-)

Trev H
Thanks for input -I noticed that Jon's backtester doesn't even have VLACX (Large Cap fund) data, only VFINX! Trying to fix that manually, not sure where to find VLACX annual returns easily

Edit: It appears VFINX and VLACX has near-identical movement.. in fact, in many charts, I can't really see one line over the other. And also, the 500 index fund has lower ER.

Link to Morningstar growth charts

http://quote.morningstar.com/fund/chart ... %2C0%22%7D

User avatar
DueDiligence
Posts: 83
Joined: Mon Sep 09, 2013 3:57 pm

Re: Updated figures and results on slice and dice vs TSM

Post by DueDiligence » Mon Sep 16, 2013 6:31 pm

cflannagan: "I was expecting bigger differences. Using Jon's backtester, even for only 5 years, I'm showing difference as large as .8. For 20 years, I had more than 1.0% difference...."
Trev H: "17.5% VTRIX, 17.5% VFSVX, 17.5% VFINX, 17.5% VISVX..... "

Thanks for pointing me to the Backtester. I have updated the estimated performance for International Small prior to inception of VSS in order to be more consistent with the Backtester. I used Vang Intl Explorer VINEX, though there is ambiguity since there have been some large historical differences between ISG and ISB.
The modified results are shown below and are generally consistent with cflannagan's expectations of ~1% outperformance for 70% Equity portfolios.

I have also modified my terminology to reflect that fact that "Simplified Buy and Hold" SUBH has equal parts US and foreign equity. Sorry Trev H.

Image

Again I'd appreciate comments and criticisms.
"Personal preferences, circumstances, and abilities affect portfolio construction in a profound manner..." David Swensen

User avatar
cflannagan
Posts: 949
Joined: Sun Oct 21, 2007 11:44 am
Location: Palm Harbor, Florida

Re: Updated figures and results on slice and dice vs TSM

Post by cflannagan » Mon Sep 16, 2013 6:36 pm

DueDiligence wrote:cflannagan: "I was expecting bigger differences. Using Jon's backtester, even for only 5 years, I'm showing difference as large as .8. For 20 years, I had more than 1.0% difference...."
Trev H: "17.5% VTRIX, 17.5% VFSVX, 17.5% VFINX, 17.5% VISVX..... "

Thanks for pointing me to the Backtester. I have updated the estimated performance for International Small prior to inception of VSS in order to be more consistent with the Backtester. I used Vang Intl Explorer VINEX, though there is ambiguity since there have been some large historical differences between ISG and ISB.
The modified results are shown below and are generally consistent with cflannagan's expectations of ~1% outperformance for 70% Equity portfolios.

I have also modified my terminology to reflect that fact that "Simplified Buy and Hold" SUBH has equal parts US and foreign equity. Sorry Trev H.

Image

Again I'd appreciate comments and criticisms.
Looks correct now. Thanks, was getting worried for a minute.. because had the difference actually be lesser than 1%, then I would feel that there is no point for me to do SUBH.. I'd go ahead and simplify further to BH3 or something simpler than SUBH ;)

sharke
Posts: 59
Joined: Wed Apr 25, 2012 10:20 pm

Re: What it you expect to be a net saver through retirement?

Post by sharke » Fri Sep 27, 2013 7:56 pm

afan wrote:Tax deferred investments represent less that a quarter of my financial assets. I plan to live off required minimum distributions from tax deferred accounts and dividends from TSM, international total stock, and intermediate term munis in taxable accounts. That means that the choice is not "paying cap gains taxes now through slice and dice or later when I liquidate stocks in retirement". The choice is paying cap gain tax now due to the inherently higher turnover for small value portfolios, and the need to rebalance, or deferring cap gain taxes forever. My heirs would get the stepped up basis, and any gains in the portfolio at my death would escape taxation permanently.

For this reason, the tax drag on a slice and dice portfolio becomes highly significant. If I have to lose a large share of any gains to taxes it is not clear that I end up better off with the pretax advantage the small value tilt might produce. Given the unpredictable nature of the premium, and its tendancy to disappear for long periods of time, I wonder whether there is really any advantage.

It seems I get lower expenses, lower taxes, simplier taxes, and a larger share of my gains escape taxation altogether. Would I really be better off tilting and forsaking the tax advantages?
This is my feeling as well. I'm actually willing to gamble that the Small / Value Premium will persist, but implementing it doesn't appear to be practical for me. Like you, I hold the majority of my assets in taxable accounts. As for my 401(k), it doesn't offer a value fund in its core holdings. It does offer a small stock fund in its core holdings, but the expense ratio is over 1% (vs .10 or less for the index funds it offers). I do have access to other funds via a PCRA, but I've found that tinkering with that (e.g. making purchases manually, summing my portfolio manually) is not worth the hassle to me.
I survived the Great Bond Crash of 2013!

Post Reply