Ultimate Buy and Hold - 8 slices vs 4

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
tetractys
Posts: 6194
Joined: Sat Mar 17, 2007 3:30 pm
Location: Along the Salish Sea

Post by tetractys »

Curious about two things Trev,

How would the 25% Each portfolio do with TSM as the LB for the tax efficiency minded folks here? I imagine it would make little or no difference, but what would you say?

Also concerning the limitations of running numbers with yearly rebalancing only. What about running numbers with 5/25 or other rebalancing methods? That could be difficult; but might it show that the 8 fund portfolio actually wins by a landslide?

Thanks, Tet
User avatar
BlueEars
Posts: 3968
Joined: Fri Mar 09, 2007 11:15 pm
Location: West Coast

Post by BlueEars »

tarnation wrote: I am personally not concerned about lack of fund history. 2100 stocks from ~44 Countries should be pretty representative of the international small cap asset class.

Also, since VINEX is active, I would be more concerned about it replicating any past performance in the future. Also, as active expect less tax efficiency.

In addition, I imagine the VINEX holdings are a subset of VFSVX, what is it about the other 1,891 stocks you don't like? :)

I am not jazzed about the 0.75% purchase fee, but we bought in during the subscription period.
VFSVX probably is a pretty good fund. But I don't know much about it's makeup or how that particular index has performed in the past. In particular, how volatile has this index been? My guess is there is some performance history for this index but VG and FTSE have not supplied the data. The FTSE site required a password (if I recall correctly) to access more info on the site. There was a document describing the makeup of the indexes but in very general terms.

With a combo of VGTSX (Total International) and VINEX one can get a pretty good distribution of of large/mid/small caps. One can also see the M* X-ray to get some sense of the distribution. Yes, VINEX is active and that may bother some. I'm just not convinced that VFSVX is a clear index substitute. I guess more data will be available some day.

Just want to be clear that I don't own VINEX at this time but would like to get more mid/small international exposure. I've got nothing against VFSVX except the lack of info.
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

Les,

I understand your concern but still think it is all about the asset class.

I do think that FTSE X-US Intl Small Cap will be a great way to get exposure to that. 2000+ International Small Stocks including around 20% EM Small.

If you look at past performance of Intl Explorer or a few other Intl Small Cap funds that have been around a few years (TRP has one with more actual history than Vanguards Intl Explorer) then you can see that the asset class is quite different than the others / US Large, Small, Intl Large.

I would have bought Intl Explorer years ago, if it had not been Closed.

I did buy FTSE X-US during the subscription period, and they were saying the 0.75 purchase fee would not apply if you did that. I have not checked into that, but even if it did apply, I would have still bought in.

Hopefully as with most new Vanguard fund offerings, the ER will eventually be reduced and that 0.75 PF will be reduced or perhaps even eliminated.

Cast your bread upon the waters, in time, it will return to you.
Diversify 7 or possibly 8 ways

Summarized version of what King Solomon said...

LB, SV, ILV, IS, REIT, IT Treasury, TIPS

8 ways :-)
livesoft
Posts: 85971
Joined: Thu Mar 01, 2007 7:00 pm

Post by livesoft »

Les et al., one has to get past the fund being an unknown and think strictly about the asset class. This will be true of many funds. Small/mid foreign has been around a long, long time ... it's just that Vanguard only had VINEX in this area. DFA, TRowePrice, Fidelity, MFS, Driehaus and other fund companies have had funds in this asset class for many years.

Years ago, I'm embarassed to write that the one load fund I owned was MIDAX and I bought it just to get into this asset class.

Nowadays there are many excellent ETFs that can be use for this asset class. The only track record that VSS needs is trading volume and it has that. You don't really have to look anything up in Morningstar. M* didn't have any info on VINEX when it started either. If you had waited for M* to get good info on VINEX, it would've been closed before you acted. In any event GWX is a good proxy for VSS. The differences between VSS, GWX, DLS and SCZ are gonna be small.
User avatar
BlueEars
Posts: 3968
Joined: Fri Mar 09, 2007 11:15 pm
Location: West Coast

Post by BlueEars »

livesoft wrote:Les et al., one has to get past the fund being an unknown and think strictly about the asset class. This will be true of many funds. Small/mid foreign has been around a long, long time ... it's just that Vanguard only had VINEX in this area. DFA, TRowePrice, Fidelity, MFS, Driehaus and other fund companies have had funds in this asset class for many years.

Years ago, I'm embarassed to write that the one load fund I owned was MIDAX and I bought it just to get into this asset class.
Livesoft and Trev, I'm no stranger to small cap international having had substantial holdings in the early 90's. Did quite well with Price NewAsia and International Discovery, and Harbor International for ILV. Index funds (and forums) were not a feature back then.

I guess we'll just have to agree to disagree on the amount of past performance information one should have before investing in a given index fund. I hope we can agree that not all indexes are created equally even within an asset class.
User avatar
BlueEars
Posts: 3968
Joined: Fri Mar 09, 2007 11:15 pm
Location: West Coast

Post by BlueEars »

Just a further note, found some chart data on the VFSVX underlying index at:
http://www.findata.co.nz/Markets/StockQ ... E/SS03.htm
User avatar
tarnation
Posts: 2544
Joined: Thu Apr 26, 2007 12:36 am

Post by tarnation »

Les wrote:Just a further note, found some chart data on the VFSVX underlying index at:
http://www.findata.co.nz/Markets/StockQ ... E/SS03.htm
Looks like that goes back abut 3.5 years. What are your conclusions from this data.?
Image
User avatar
BlueEars
Posts: 3968
Joined: Fri Mar 09, 2007 11:15 pm
Location: West Coast

Post by BlueEars »

tarnation wrote:
Les wrote:Just a further note, found some chart data on the VFSVX underlying index at:
http://www.findata.co.nz/Markets/StockQ ... E/SS03.htm
Looks like that goes back abut 3.5 years. What are your conclusions from this data.?
I just thought those holding the fund might like to see this data. Admittedly, the data is pretty skimpy but it does show how the fund did in the market crash. I have not compared the decline to VINEX (probably more midcap then VFSVX) or other fund's I might think of as substitutes. But just eyeballing this it looks like a -53% return for 2008. VINEX was -47% for 2008.

I called a rep at VG and she was unable to point me to more data. She was very patient but VG is a big company and my request for FTSE index data will probably get buried in higher priority issues.
gvernon
Posts: 124
Joined: Fri Apr 20, 2007 7:28 pm
Location: Houston, TX

Post by gvernon »

Trev H,

Any chance you could also graph the remaining 4 slices on their own? I'm curious to see how a combination of LV, SB, ILB, ISV performs compared to LB, SV, ILV, ISB.

Thanks!
User avatar
big_tail
Posts: 42
Joined: Mon Jul 28, 2008 11:59 am

Post by big_tail »

Thanks to Trev H for this thread and analyses. I've read the original Fundadvice article and analyzed studies on slicing & dicing & various index & value premium articles. The final conclusion I had came to about the same.

One difference I have is that instead of straight 50% U.S/Intl split, I like a more market based 43%/57% (as of June 09). Having a 50/50 split further than a couple of decades ago was a risky thing and very unconventional. We just don't know how the U.S. market cap share of the total world market cap will behave going forward, so I'd rather just replicate it in the world market (In past history, it has oscilated around 50%). If the US dollar loses its value against other currencies (due to inflation or treasury printing presses), I would predict the U.S. market cap will also go down relatively. And if you remain 50/50, you're overweighting U.S. markets relative to the world and may miss out on growth outside the U.S. So I prefer to split it 50/50 large blend to small value first, then market ratio of U.S. to Intl exposure. So for now something like:

21.5% US Large Cap Blend (43% of 50%)
28.5% Intl Large Cap Blend (57% of 50%)
21.5% US Small Cap Value
28.5% Intl Small cap (hopefully to be replaced by intl small cap value if Vanguard comes out with such a product in the future).

The U.S. to world market cap ratio will automatically adjust, I just have to maintain the Large Blend vs Small Value ratio to 50/50.
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

3 - 4x25 options shown below.

LB,SV,ILV,IS

LB,SV,ILB,ISV

LB,SV,ILB,IS

I started to include the 8 slice in there with the others but it gets too cluttered if you do that. 8 slice performs very much like the others (you can see it in the first chart posted).


Image


===
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

Tet...

If you used TSM instead of 500 index for the US Large Market allocation.


Image


Got to look close to see there are actually 2 data sets charted..
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

Showing some different combo's below including Bernsteins Cowards Achademic Portfolio (LV,SV,ILV,ISV) and a Larry Swedroe type (SV,ISV) equity mix - included the Simplified Ultimate Buy and Hold (or what ever you want to call it) and your standard Lumper (all at 50/50 US/Intl).


Image

===
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

In this one I included 3 examples all at 50/50 US/Intl equity.

Starting with..

Lumper at 80 equity / 20% bonds

Simplified Ultimate B&H at 70 equity / 30% bonds

Larry at 60% equity / 40% bonds

As you tilt away from the "Markets" increasing bond allocation by 10%.


Image

===
WheatThin
Posts: 135
Joined: Wed Dec 10, 2008 10:21 am
Location: Orlando, FL

Post by WheatThin »

I think it makes the most sense to view LB/LV and ILB/ILV as interchangeable based on location. I'd use value if tax-deferred and blend if taxable. No point in picking one exclusively over the other.
Image
EyeDee
Posts: 1388
Joined: Mon Feb 19, 2007 11:15 pm

Post by EyeDee »

Trev H wrote:In this one I included 3 examples all at 50/50 US/Intl equity.

Starting with..

Lumper at 80 equity / 20% bonds

Simplified Ultimate B&H at 70 equity / 30% bonds

Larry at 60% equity / 40% bonds

As you tilt away from the "Markets" increasing bond allocation by 10%.
.
Trev,

Thank you for the neat graphs. If you are still taking requests, could you run this last one with Bonds being something like a fixed 4% per year (or 3%, etc. whatever you prefer), instead of ITT. This would remove any bias because of the interest rates during the period. Obviously, there is still period dependency for the stocks.

Thank you.
Randy | SCA - Build Savings early by living below one's means, minimize Costs including taxes, and maintain a diverse Allocation.
WheatThin
Posts: 135
Joined: Wed Dec 10, 2008 10:21 am
Location: Orlando, FL

Post by WheatThin »

Trev H wrote:
Image

===
From this, you can't tell if the difference in the two 4-way portfolios is attributable to the presence LV or ISV. Can you run:

LB/SV/ILV/IS versus
LV/SV/ILV/IS versus
LB/SV/ILV/ISV

?

Thanks for these Trev.
Image
conundrum
Posts: 871
Joined: Sat May 09, 2009 7:00 pm

Post by conundrum »

Trev, thanks so much for all your hard work and data.

The question I have is related to the fact that I have limited tax deferred space(15%) and it is filled with TIPS. I have a longtime holding of tax managed small cap (VTMSX) rather than small value for tax reasons. On review of yahoo charts and morningstar it seems that at least for the last 10 years it has done as well as VISVX and IJS.

In a taxable account what is the effect of substituting VTMSX, a small cap blend for small value.

Thanks

Drum
conundrum
Posts: 871
Joined: Sat May 09, 2009 7:00 pm

Post by conundrum »

Sorry for additional post

I guess the question I am asking is that you have graphs and data substituting LB,ILB and ISB for LV, ILV and ISV, what would happen if you substituted SB for SV on domestic side?

Thanks again

Drum
jimtramont
Posts: 31
Joined: Tue Jun 17, 2008 4:02 pm

Re: Funds...

Post by jimtramont »

Trev H wrote:The exact Vanguard Funds that could be used to setup the Simplified Ultimate Buy and Hold LB,SV,ILV,IS combo..

LB = Large Cap Index or 500 Index
SV = Small Cap Value Index
ILV = International Value
ISB = FTSE X-US Intl Small Cap (or International Explorer).

If you wanted to stick with all Index funds then could sub FTSE X-US Intl Large Cap fund for International Value.

On the Bond allocation side, I would suggest InterTerm Treasury Bonds and TIPS.

==
Thanks also!
Except for bonds and REIT, are the above OK for taxable accounts?
What effect does excluding REIT have?
User avatar
spam
Posts: 907
Joined: Tue Jun 10, 2008 9:47 am

Post by spam »

6/5/09 401k allocation

15% US Large Balanced
15% US Small Cap Index
15% Itl Large Balanced
15% Itl Small Cap
10% REIT
10% Global Bond
10% Pimco Junk
10% Pimpco TIPs

Thank you Trev H. You have made me a believer. These are not exactly like yours, but were the closest offerings I had.

I am using different stratagies in my other accounts. I have two set up in what I feel will be a 10 year allocation plan. The others are under review.
User avatar
tarnation
Posts: 2544
Joined: Thu Apr 26, 2007 12:36 am

Post by tarnation »

spam wrote:
Thank you Trev H. You have made me a believer.
Have you read The Telltale Chart http://www.vanguard.com/bogle_site/sp20020626.html
Image
james22
Posts: 1992
Joined: Tue Aug 21, 2007 2:22 pm

Post by james22 »

Trev H wrote:LB, SV, ILV, IS, REIT, IT Treasury, TIPS
I don't mean to sidetrack this, Trev, but why the 10% REIT allocation when VG SV already so REIT heavy? Have you considered 5% Energy and 5% PM&M instead?
User avatar
tetractys
Posts: 6194
Joined: Sat Mar 17, 2007 3:30 pm
Location: Along the Salish Sea

Post by tetractys »

Trev H wrote:Tet...

If you used TSM instead of 500 index for the US Large Market allocation.

Got to look close to see there are actually 2 data sets charted..
Thank you kindly Trev, Tet
Note: The chart's above; I took it here out for brevity.
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

Sorry Folks... been out of pocket a bit. Niece got married (out of town) and had to spend some time helping my Mom out (health issues).

Back home for a bit now.

===

Here is a question that someone asked in a PM.

They wanted to see how it worked if you eliminated the Intl Small and used only US Small Value for tilting.

Below is a look at starting with Lumper at 50/50 and then adding in 20% slice weights of US SV and the combo of LB,SV,ILV,IS is shown as well.

Image

You would have to go to extreme weightings on US SV to get closer to the performance of the combo that includes both US and Intl Small. The extremely low correlation of SV, IS is a serious boost to the combo that includes both over any combo that only holds one of the small components.

If I had large taxable holdings and wanted to stick with TSM and Total Intl there, but had room in IRA/Roth type areas for (more than bond allocation) I would sure consider holding SV & IS there or some combo of US Small and Intl Small rather than just holding US SV.

===
User avatar
spam
Posts: 907
Joined: Tue Jun 10, 2008 9:47 am

Post by spam »

tarnation wrote:
spam wrote:
Thank you Trev H. You have made me a believer.
Have you read The Telltale Chart http://www.vanguard.com/bogle_site/sp20020626.html
Tarnation,

I had not read it. Thanks for the link. I will have to re-read it a few times to get a fuller understanding.

Using past data is always controversial when interpreted in a forward-looking way. While I am not certian, it seems that Jack might be predicting a correction or RTM any time one asset group gains an advantage. This idea is intuitively true when considering the attraction sector outperformance has for fresh dollars, or in broader terms, the underlying mechanics which make a rebalancing benifit possible.

I am attracted to Trev's approach for several reasons.

1) I rebalance so SD and low correlation is my friend
2) Simplicity. This makes 401k contribution management more nimble and efficient. The effect of constants 1/4 vs 1/8 on an increasing value of 1.
3) It fits my broader philosophy of Domestic, International, REIT, and Commodities.
4) Most importantly, the difference between a 401k and an IRA. My IRA's usually get funded in lumps and are more like a warehouse. My 401k gets more traffic and is more like the showroom.
5) Lastly, large cap (S&P) will work in a taxable account if necessary.

Time to hit the river. I have a date with a canoe and a trout.

Thank you for the great read.
spam
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

.
Tarnation,

Yes - I have read that bit from Bogle a few times - The Telltale Chart.
As well as others that do their best to convince the investor that TSM or US Large Blend is all you really need.

Below is a good one from the old "Ask the Experts post" on the original M* forum (that I saved many years ago).

====

22. *JUST* Total Stock Market?
M*_JennyB| 08-24-99 | 09:13 AM
Emayer:
Dear Mr. Bogle,

My deep appreciation to you as a Vanguard customer and longtime admirer.

There are thousands of mutual funds out there by now. It can drive you crazy trying to choose among them. You have advocated Vanguard's Total Stock Market Index as the best choice as the foundation of one's stock holdings. For a long-term investor looking for simplicity, would you advocate holding exclusively that fund (along with bonds and some cash)?

Thanks for your thoughts,
emayer@ymail.yu.edu

Jack Bogle:
Yes, I would. I just don't see the need to complicate investing unnecessarily.

===

What I think it really reveals about Bogle - is that even very smart folks, some would call "Experts" can be extremely wrong and even give very poor advice that is driven (IMO) by BIAS.

BIAS is (IMO) the main reason that so many investors lack proper equity diversification in the portfolio.

Whether it's Home Country BIAS, or Large Cap (Magical Market Cap Weighting) BIAS - in the end it adds up to a serious lack of diversification.

He mentions data in that TellTale Chart article pointing out the higher returns of Small and Value but then disses that by pointing out the higher volitility (risk) of those components.

If you look at components "stand alone" you could easily draw such conclusions, but if you look at how they work in the portfolio mix other factors come into play like correlations which figure into the end result.

Also as Larry always mentions, when taking on the higher expected return/risk of Small/Value should reduce equity risk. Bogle does not show that, or how even a small adjustment in the Equity/Bond AA completely eliminates that risk measure (volatility).

The chart below shows Bogle, vs Simplified B&H, and then a reduced equity risk mix.

===

Image


Yes - TSM was less volatile than the 4x25 S&D mix, but does that really matter at the portfolio level ?

Not if you adjust the Equity/Bond AA by a small amount.

Looking at longer term Fama French Data (the kind of volatility you might get with DFA funds) increasing the bond allocation by 20% or so does the job. With Vanguard/Russell index data (which is what is used in my set) a 10% shift to the bond allocation more than eliminates the volatility difference. Showing a 20% shift in the chart above and well you can see the volatility difference compared to 100% TSM.
Last edited by Trev H on Sat Jun 06, 2009 7:08 am, edited 3 times in total.
User avatar
SquawkIdent
Posts: 927
Joined: Tue Dec 23, 2008 6:14 pm
Location: Planet Earth

Post by SquawkIdent »

Trev H - have you ever looked at the permanent portfolio (either the mutual fund PRPFX or Harry Browne's original 4 x 25 - 25% TSM, 25% long term treasurys, 25% short term treasurys and 25% gold ETF)? How would that stack up on your chart? Thanks.
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

.
SquawkIdent,

I have been asked that before, on the permanent portfolio mix.

Unfortunately I do not have annual return data for Gold.

I checked the Simba Sheet (got a copy from the wikki) and did not find Gold returns there.

If you or anyone else can list the returns for Gold 1970-2008 - or point me to where I can get them online, I will gladly add that asset class to my sheet and post some portfolio results.

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

Post by Rodc »

Not surprising, though good to see. This is consistent with what I showed on the Domestic side ( http://www.bogleheads.org/forum/viewtopic.php?t=9445 ) which in turn was also not surprising.

All this really boils down to is that over rather long time periods in the past the Fama-French three factor model fits the past data, which in turn is no surprise as it was built on past data.

More interesting is what has happened in more recent years, post the original F-F paper, and while I have not specifically looked, Small-Hi (and others) have posted results showing that the F-F model continues to work well.

So, the number of slices is not material. What is material is targeting Market, Small, and Value. (or I should add, what was important)

If you wanted to make this analysis more useful, I would suggest breaking it into smaller periods to look at how consistent the results are. After all most of us don't lump sum invest and hold for 40 some odd years. Say break it into 4 periods and see if the results hold in each.
Last edited by Rodc on Sat Jun 06, 2009 8:33 am, edited 2 times in total.
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.
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

Spam,

You mentioned equally weighting LB,SB,ILB,ISB (for general equities).

Tilting to Small Market (only).

That is a mix we have not looked at so far in this thread.

Did that below.


Image

Small Market & Large Market have higher correlation than Large Market & Small Value, or Large Value and Small Market.

Also with Small Market you get more of that Small Growth exposure.

It shows up in the volatility.

But still - could adjust that out by increasing bond allocation a bit.

And Small Market(s) should be more tax efficient than Small Value or possibly Large Value, ( could use Tax Managed Small Cap, or more tax efficient ETFs) to hold those if investing in taxable area.
User avatar
Offshore
Posts: 510
Joined: Wed Apr 09, 2008 9:43 am
Location: Michigan's West Coast

Post by Offshore »

Trev H, fantastic job putting the data to chart form and starting this discussion.

Oh how I wish I was 20 years old again :wink:

For investors in thier 50's who are adjusting down their equity position, is this discussion mostly academic for us? Meaning, the 39 year graph history presented will apply to a small equity % of the portfolio.

Question: how does tinkering around with fixed income holdings affect risk/return? Does adding international fixed income change risk/return much? What type of int'l. bond holdings?

What do you think? It's not discussed much by the indexing community so I'm guessing there's not much to be gained from diversifying in this way.
User avatar
BlueEars
Posts: 3968
Joined: Fri Mar 09, 2007 11:15 pm
Location: West Coast

Post by BlueEars »

Rodc wrote:...
If you wanted to make this analysis more useful, I would suggest breaking it into smaller periods to look at how consistent the results are. After all most of us don't lump sum invest and hold for 40 some odd years. Say break it into 4 periods and see if the results hold in each.
How many of us have held a portfolio with rebalanced components for even 10 years and not tinkered with it? Many of us were not indexing even a part of the portfolio a decade ago.
User avatar
SquawkIdent
Posts: 927
Joined: Tue Dec 23, 2008 6:14 pm
Location: Planet Earth

Post by SquawkIdent »

Trev H wrote:.
SquawkIdent,

I have been asked that before, on the permanent portfolio mix.

Unfortunately I do not have annual return data for Gold.

I checked the Simba Sheet (got a copy from the wikki) and did not find Gold returns there.

If you or anyone else can list the returns for Gold 1970-2008 - or point me to where I can get them online, I will gladly add that asset class to my sheet and post some portfolio results.

Thanks
Trev H - Here's the page from the crawling road blog where the returns are listed - http://crawlingroad.com/blog/2008/12/22 ... /#more-299
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

GLD

Post by Trev H »

Hey - I found rev8c of Simbas sheet and it had Gold listed 1972-2008 but he did not list the source.

Looking around online I found this site.

http://www.measuringworth.org/datasets/gold/result.php

Where you can specify to list the New York Market Price for Gold (and I did that for 1969-2007 (their data ends 2007).

Below is what I got.

Code: Select all

1969 41.51 
1970 36.41 
1971 41.25 
1972 58.60 
1973 97.81 
1974 159.74 
1975 161.49 
1976 125.32 
1977 148.31 
1978 193.55 
1979 307.50 
1980 612.56 
1981 459.64 
1982 375.91 
1983 424.00 
1984 360.66 
1985 317.66 
1986 368.24 
1987 447.95 
1988 438.31 
1989 382.58 
1990 384.93 
1991 363.29 
1992 344.97 
1993 360.91 
1994 385.42 
1995 385.50 
1996 389.09 
1997 332.39 
1998 295.24 
1999 279.91 
2000 280.10 
2001 272.22 
2002 311.33 
2003 364.80 
2004 410.52 
2005 446.00 
2006 610.00 
2007 702.57 

If I calc the yearly return and back out an ER of .40 annually (the ER of GLD ETF) then below is what I get for Gold Returns (- todays reasonable cost).

I can use the 2008 and YTD returns of the GLD ETF to complete the set.

Does that look about right ?

I am going to be out of town the rest of the day - will follow up on this and compare the Gold returns below to what Simbas sheet has.

Later !



Code: Select all

1970	4.44
1971	16.95
1972	38.81
1973	61.53
1974	1.35
1975	-36.57
1976	22.59
1977	44.84
1978	113.55
1979	304.66
1980	-153.32
1981	-84.13
1982	47.69
1983	-63.74
1984	-43.40
1985	50.18
1986	79.31
1987	-10.04
1988	-56.13
1989	1.95
1990	-22.04
1991	-18.72
1992	15.54
1993	24.11
1994	-0.32
1995	3.19
1996	-57.10
1997	-37.55
1998	-15.73
1999	-0.21
2000	-8.28
2001	38.71
2002	53.07
2003	45.32
2004	35.08
2005	163.60
2006	92.17

User avatar
BlueEars
Posts: 3968
Joined: Fri Mar 09, 2007 11:15 pm
Location: West Coast

Post by BlueEars »

What would I do for gold? VG has a precious metals fund. Does it track pure gold well? Or is it broader and less volatile? Whatever backtest data is used, one should have a good feeling that there is something like it going forward.

I feel confident in finding offerings in the 9 style boxes for the US. Don't have the same confident feeling for finding ISV yet, but feel good about finding ILB, ILV, and a combo of midcap/smallcap international (International Explorer).
caklim00
Posts: 2420
Joined: Mon May 26, 2008 10:09 am

Post by caklim00 »

Les wrote:I feel confident in finding offerings in the 9 style boxes for the US. Don't have the same confident feeling for finding ISV yet, but feel good about finding ILB, ILV, and a combo of midcap/smallcap international (International Explorer).
DLS/DGS for ISCV, but you have to go outside of Vanguard for that.
conundrum
Posts: 871
Joined: Sat May 09, 2009 7:00 pm

Post by conundrum »

Trev

Thanks so much for looking at LB,SB,ILB,ISB combo. I have wondered about the combo for awhile as my portfolio is predominantly taxable. As I noted above we use tax managed small cap VTMSX as our small blend fund. Overall not as good as small value but not too bad.

Thanks again for your work.

Drum
jimtramont
Posts: 31
Joined: Tue Jun 17, 2008 4:02 pm

Taxes

Post by jimtramont »

If splitting between taxable and IRA, how would you place these funds:

US lg cap or S&P 500
SCV fund
Int value
Int small cap

Thanks
buynhold
Posts: 83
Joined: Tue Apr 21, 2009 11:22 am

Post by buynhold »

Are you only using Excel for this analysis? I just got out of engineering school and have the weird urge to crunch some numbers... :shock:
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

Jim asked...

===
If splitting between taxable and IRA, how would you place these funds:

US lg cap or S&P 500
SCV fund
Int value
Int small cap
===

PS - I am all IRA,ROTH,403b at this point so non-issue for me.

If Taxable, then you might want to consider Large Cap index, 500 index or TSM (for US Large) and then FTSE X-US or Total International for International Large.

Then in the IRA type areas if you have room for more than just bonds, then hold US Small Value and Intl Small Cap there.

Tax Managed Small Cap would be fine for taxable account, and I do think that several here hold US Small Value and Intl Small options (in taxable) utilizing hopefully more tax efficient ETF's - instead of mutual funds.

===

Buynhold - Yes I use Excel (only).
User avatar
spam
Posts: 907
Joined: Tue Jun 10, 2008 9:47 am

Post by spam »

Trev H wrote:Spam,

You mentioned equally weighting LB,SB,ILB,ISB (for general equities).

Tilting to Small Market (only).

That is a mix we have not looked at so far in this thread.

Did that below.


Image

Small Market & Large Market have higher correlation than Large Market & Small Value, or Large Value and Small Market.

Also with Small Market you get more of that Small Growth exposure.

It shows up in the volatility.

But still - could adjust that out by increasing bond allocation a bit.

And Small Market(s) should be more tax efficient than Small Value or possibly Large Value, ( could use Tax Managed Small Cap, or more tax efficient ETFs) to hold those if investing in taxable area.
Hi Trev,

Thank you for calculating my mix. I do think it is the best 401k strategy for me even though some of my correlations are stronger than in your model.

Not to sidetrack this discussion, but I will be easing more money into bonds as the dow approaches historic highs. I feel that market risk is not static, that prices tend to move from grossly oversold to insanely overbought. RTM means (to me) that the odds of a significant negative correction increase as the prices climb (dow 14,000), and the odds of a significant positive correction increase as prices drop (dow 6,500). Maybe Jack would agree, and maybe he wouldn't. Anyway, I will be rebalancing to different risk allocations (small changes) along the way.

I do feel that your model has real value. Especially in a big-picture sort of way. It certianly works for me.
caklim00
Posts: 2420
Joined: Mon May 26, 2008 10:09 am

Post by caklim00 »

Trev H wrote:Jim asked...

===
If splitting between taxable and IRA, how would you place these funds:

US lg cap or S&P 500
SCV fund
Int value
Int small cap
===

PS - I am all IRA,ROTH,403b at this point so non-issue for me.

If Taxable, then you might want to consider Large Cap index, 500 index or TSM (for US Large) and then FTSE X-US or Total International for International Large.

Then in the IRA type areas if you have room for more than just bonds, then hold US Small Value and Intl Small Cap there.

Tax Managed Small Cap would be fine for taxable account, and I do think that several here hold US Small Value and Intl Small options (in taxable) utilizing hopefully more tax efficient ETF's - instead of mutual funds.

===

Buynhold - Yes I use Excel (only).
I would first hold FTSE X-US Large in Taxable, then TSM (or Large Cap) in Taxable. I thinks its a guess as to whether ISC or Domestic SCV is any better in taxable. You loss the ability to take the foregin tax credit when you put foreign in tax-deferred, but DLS, DGS, VSS, SCZ might be very tax efficient (I haven't done the analysis as I hold in tax-deferred)
conundrum
Posts: 871
Joined: Sat May 09, 2009 7:00 pm

Post by conundrum »

Great information and discussion.

Will review my portfolio and consider increasing my small blend component and adding international small blend.

Great charts!

Drum
User avatar
tarnation
Posts: 2544
Joined: Thu Apr 26, 2007 12:36 am

Post by tarnation »

Trev H wrote:Tarnation,

Below a look at what you suggested.

The difference between holding ILB instead of ILV for the International Large component - which you could do at Vanguard with an all index portfolio.
Thank you for that.
Image
RedTail
Posts: 81
Joined: Mon May 11, 2009 10:01 pm

Post by RedTail »

Question on this topic:

Why is it that LB and SV are chosen as the domestic portion of the asset allocation and LV and SB as the international portion? Why not the other way around? Would that make any difference?
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

RedTail asked...

==
Why is it that LB and SV are chosen as the domestic portion of the asset allocation and LV and SB as the international portion? Why not the other way around? Would that make any difference?
==

The Ultimate Buy and Hold Portfolio suggest holding US and Intl Equity in 8+ slices something like this

LB,LV,SB,SV,ILB,ILV,IS,ISV

Having looked at data a LOT over the years, simply noticed that on the US Equity Side

50% each LB/SV

Performs almost exactly the same as

25% each LB,LV,SB,SV

With 2 US Equity components you get almost exactly the same results as 4.

I simply carried that forward to the International Side and found that the combo of ILV,ISB performs very much the same as ILB,ILV,ISB,ISV.

Also at Vanguard you can get ILV and ISB, but no ISV.

And when you compare 4x25 split of LB,SV,ILV,ISB to the 8x12.5 split of LB,LV,SB,SV,ILB,ILV,IS,ISV you get almost exactly the same results.

4 components vs 8 and same results (see first charts shown for comparison).

Other good reasons for the selections of LB,SV,ILV,ISB

25% Slice of US Large Blend (instead of Large Value) would help out a bit in regards to Tracking Error (if you pay attention to that).

Also on the US equity side holding a Blend and Value component (lowers correlation), and on the International Side you are doing the same, ILV, ISB (lower correlation).

And also on the International Side the standard EAFE Developed Market stocks are poorest diversifier (of the International options) for US Equity.

I think those are some good reasons that specific combo works well.

===
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

Also - check out the post dated June 4, at 5:31 am where 3 - 4x25 options were compared.

LB,SV,ILV,IS

LB,SV,ILB,ISV

were very similar - but then again Vanguard does not offer ISV.

===
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

I checked out the difference in (I think all reasonable combo's)

1 - 8x12.5 (3 Factor) combo..

LB,LV,SB,SV,ILB,ILV,ISB,ISV cagr 12.62, stdev 19.56, sharpe .45


4 - 4x25 (3 Factor) combo's..

LB,SV,ILV,ISB cagr 12.74, stdev 19.36, sharpe .46
LB,SV,ILB,ISV cagr 12.50, stdev 19.05, sharpe .45
LV,SB,ILV,ISB cagr 12.71, stdev 20.22, sharpe .44
LV,SB,ILB,ISV cagr 12.48, stdev 19.89, sharpe .44

1 - 4x25 (2 Factor) combo

LB,SB,ILB,ISB cagr 11.87, stdev 20.31, sharpe .40

1 - 2x50 (1 Factor) combo

TSM,ILB cagr 9.70, stdev 18.86, sharpe .31

I started to chart em, but the 3F combo's don't vary all that much.

I have alredy shown the 1 Factor, and 2 Factor examples (compared to others) up earlier in this thread.

===
Topic Author
Trev H
Posts: 1896
Joined: Fri Mar 02, 2007 9:47 pm

Post by Trev H »

EyeDye / Randy asked...

==
you are still taking requests, could you run this last one with Bonds being something like a fixed 4% per year (or 3%, etc. whatever you prefer), instead of ITT. This would remove any bias because of the interest rates during the period. Obviously, there is still period dependency for the stocks.
==

I got to this request this morning.

Used a Fixed 4% in the charts below instead of IT Treasury.


Image
Post Reply