Bogleheads

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Les,

Below the price listed by the MeasuringWorth sites (New York Market Price) vs the site you pointed me too which says the source is Global Financial Data.

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Using that data for Gold Returns that list the source as Global Financial Data below is what I get.

Left = my calc for annual return %
Middle = my calc for annual return % -.40 bp annual cost
Right = Simba's data

Mitchell777 · Joined: 21 May 2007 Posts: 151

I have been reading though this post with interest. I may be missing the info I'm looking for. Is there anywhere I can get the annual returns for each of the PP categories, 25/25/25/25 (TSM/LT Bonds/MMF/Gold), going back to the 1970's perhaps. Not the composite return for each year. For ex, something that would show the 1970 return for TSM and LT Terms and MMF and Gold (the return for each of the investments), then 1971, 1972 etc . Thanks

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

With my new asset class data set for Gold... did the crunch below.

I could not find where you folks settled on the exact allocation for the PP.

What I did was take a simple 3F globally diversified equity mix of US Large Blend, US Small Value, International Large Value, International Small Blend mixed with InterTerm Tresury Bonds (at 60/40 AA).

vs..

The same equity mix at 40% + Gold at 20%, and again 40% IT Treasury Bonds.

Will be glad to take request on charting different mixes as long as I have the asset class return data, will be glad to do so.

PS - may be tomorrow morning before I can post any more charts on different mixes.

Thanks

===
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

craigr · Joined: 13 Mar 2007 Posts: 1973

MediumTex · Joined: 01 Mar 2009 Posts: 447

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Here is one that shows the individual components 500 index, Long Term Treasury, Short Term Treasury and Gold - - - and then a simple 4x25 mix of each.

_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

MediumTex · Joined: 01 Mar 2009 Posts: 447

Thanks for the chart.

So if I am reading the chart above correctly and the one that you posted before, is the following correct:

If you had invested $10,000--40% in stock, 40% in intermediate treasuries and 20% in gold in 1970 (rebalanced annually), today you would have $640,000 or so.

If you had invested $10,000--25% in stock, 25% in ST treasuries, 25% in LT treasuries and 25% in gold in 1970 (rebalanced annually), today you would have $320,000 or so.

That's a little surprising to me. I wouldn't have thought that such a seemingly small change in allocation would result in such a different outcome.
_________________
"A Permanent Portfolio should let you watch the evening news or read investment publications in total serenity."
-Harry Browne

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Looks to me like what makes this work is mostly the low/negative correlation of 500 index to Gold.

Below is a look, excluding bonds.

_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

craigr · Joined: 13 Mar 2007 Posts: 1973

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Below is from my 1970-forward data set source document.

Craigr...

If you know of a better source than IFA.com for ILV, ISV, IS returns, let me know. That is where I got that data but they do list sources for the International returns from 1970 forward that seem to be legit.

The data prior to that, they are using US Equity Returns to simulate International Asset classes (I did not get any of that data from them). I started with 1970 and went forward, which is all I really wanted anyway.

The only asset classes that I am using IFA data for = ILV, ISV, IS.

And where possible (history wise) I use actual Vanguard Fund returns (investor share class).

Again - if you or anyone knows of a better source that will get us back to 1970 - speak up.

In the ILV, ISV and IS (below) I list the sources that IFA listed.

===

TSM = US Cap Weighted Market:
===============================
CRSP Market Decile 1-10 1970-1992
Vanguards Total Stock Market Index Fund 1993-2008

LB = US Large Blend:
========================
S&P 500: Standard & Poors 1970-1976
Vanguards 500 Index Fund 1977-2008

LV = US Large Value:
=====================
Fama and French 1970–1978
Russell 1000 Value Index 1979–1992
Vanguards Value Index Fund 1993-2008

LG = US Large Growth:
======================
Fama and French 1970–1978
Russell 1000 Growth Index 1979–1992
Vanguards Growth Index Fund 1993-2008

MB = US Mid Blend:
===================
CRSP Decile 3-5 1970-1978
Russell Mid Cap Index 1979-1998
Vanguards Mid Cap Index Fund 1999-2008

MV = US Mid Value:
===================
Ibbotson 1970-1985
Russell Mid Cap Value Index 1986-2006
Vanguards Mid Cap Value Index Fund 2007-2008

MG = US Mid Growth:
===================
Ibbotson 1970-1985
Russell Mid Cap Growth Index 1986-2006
Vanguards Mid Cap Growth Index Fund 2007-2008

SB = US Small Blend:
=====================
Ibbotson 1970–1978
Russell 2000 Index 1979-1991
Vanguards Small Cap Index Fund 1992-2008

SV = US Small Value:
=====================
Ibbotson 1970-1978
Russell 2000 Value Index 1979-1998
Vanguards Small Cap Value Index Fund 1999-2008

SG = US Small Growth:
======================
Ibbotson 1970–1978
Russell 2000 Growth Index 1979-1998
Vanguards Small Cap Growth Index Fund 1999-2008

D10 = US MicroCap Blend:
==========================
CRSP Decile 10 1970-1997
BRSIX Bridgeway Ultra Small Market 1998-2008
REIT = Real Estate:
===================
Nat. Assn. of Real Estate Inv Trusts 1972-1996
Vanguards REIT Index Fund – 1997-2008

ILCD – International Developed:
===============================
MSCI EAFE Index 1970-2000
Vanguards Intl Developed Mkts Index Fund 2001-2008

ILC - Total International:
============================
MSCI EAFE Index 1970-1987 (Developed Only)
85% EAFE Index, 15% Emerg Mkt Index 1988-1996
Vanguards Total Intel Index Fund 1997-2008

ILV - International Large Value
===============================
85% IV, 15% EV (ifa.com) 1970-1990
IV – 1970-1974 MSCI EAFE Gross Div -3.67 bp/month
---- 1975-1990 MSCI EAFE Value Gross Div -3.67 bp/month
EV - 1970-1988 IFA EM Index
---- 1989-1990 Fama/French EM Value Index -5.0 bp/month
1991-2008 Vanguards International Value Fund

IS - International Small Cap including EM
============================================
1970-1996 85% IS, 15% ES (ifa.com)
IS - Dimensional Intl Small Cap Index -4.58bp/month
ES - 1970-1988 IFA Emerging Markets Index,
---- 1989-1996 Fama/French EM Small -6.5bp/month
1997-2008 Vanguards International Explorer Fund
2009-Forward Vanguards FTSE X-US Intl Small Cap Index

ISV – International Small Value including EM
============================================
1970-2008 85% ISV, 15% ESV (data from ifa.com + DFA Funds)
ISV 1970-1981 Dimensional Intl Small Cap Index -4.58 bp/month
--- 1982-1994 Dimensional Intl Small Value Index -5.75 bp/month
--- 1995-Present – DFA Intl Small Cap Value Fund DISVX
ESV = 50% EV, 50% ES *calculated* (ifa.com)
--- EV 1970-1988 IFA EM Index
------ 1989-1998 Fama/French EM Value Index -5.0 bp/month
------ 1999-Present DFA EM Value Fund DFEVX
--- ES 1970-1988 IFA EM Index
------ 1989-1998 Fama/French EM Small Index -6.5 bp/month
------ 1999-Present DFA EM Small Cap Fund DEMSX

COMM – Commodities / Natural Resources:
======================================
Standard & Poor's Commodity Index 1970-2002
Pimco PCRIX 2003-2008

GOLD = Gold
===========
1970-2004 www.finfacts.ie/Private/curenc....tprice.htm -.40 bp annual cost
2005-Present GLD ETF

LTT = Long Term Treasury:
==================================
Tamasset Spreadsheet 1970-1987
Vanguards Long Term Treasury 1988-2008

ITT – InterTerm Treasury:
=========================
Tamasset Spreadsheet 1970-1991
Vanguards InterTerm Treasury Fund 1992-2008

STT = Short Term Treasury
============================================
1970-1977 5 Yr T Notes – 1.5bp/month
1978-1991 US Treasury 1-3 Yr – 1.5bp/month
1992-2008 Vanguards Short Term Treasury Fund

TBILLS = Risk Free Benchmark for Sharpe:
========================================
T-Bills 1970-1983
Vanguard Treasury Money Market Fund 1984-2008

ITB = InterTerm Bonds:
======================
Ibbotson 1970-1972
Lehman Brothers 1973-1986
Vanguards Total Bond Index Fund 1987-2008

TIPS = Treasury Inflation Protected Securities:
===============================================
Synthetic TIPS 1970-2000
Vanguard Inflation Protected Securities Fund 2001-2008
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Tex,

===
That's a little surprising to me. I wouldn't have thought that such a seemingly small change in allocation would result in such a different outcome.
===

Over 40 years a change in AA can make a HUGE difference in outcome.

For example if for Equity you use 100% 500 Index vs 50% each US Small Value / International Small Value.

01/01/1970-05/31/2009 (yearly rebalancing)

100% 500 Index
========
10K Growth
324,831.71
CAGR
9.09
StDev
17.95
Sharpe
0.28

50% US Small Value
50% Intl Small Value
========
10K Growth
2,680,253.78
CAGR
15.00
StDev
21.13
Sharpe
0.54

===
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

MediumTex · Joined: 01 Mar 2009 Posts: 447

That gold graph doesn't look right. It doesn't look like gold goes down enough after the spike in 1980 to where it bottomed in 2001, and then it doesn't look like it goes up enough from the bottom to where it is today.

Those are great charts, though. Thanks again.

***

On the subject of a small AA change resulting in a big difference, that is one reason that I like doing 85% VTSMX and 15% VGTSX in the stock portion. This mix is supposed to provide the best risk adjusted return (as opposed to all domestic equities), and the added volatility of VGTSX is not a big deal to me since it is such a small piece of the stock allocation.
_________________
"A Permanent Portfolio should let you watch the evening news or read investment publications in total serenity."
-Harry Browne

craigr · Joined: 13 Mar 2007 Posts: 1973

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Craigr,

I read the disclosures before I got the data, reads the same as the disclosures stated on just about any data set you can find online, and I also fully understand the value of backtesting.

On this statement..

"in the interests of presenting conservative results it is a better idea".

I am not interested in presenting conservative results, much more interested in presenting the results of diversified portfolio's (and IMO it's not diversified unless globally diversified including 3 factor exposure).

Given your restrictions, what could one show except 500 index and perhaps treasury bonds.

Or perhaps shorter data spans only (where actual fund returns exist) rather than using index data.

I can do that if you want, but you do get the same results.

Global 3F diversification, beats Home Country Bias'd Lumping (500 index or TSM) either way.

So now you have inspired me to show PP results for different general equity mixes including HCBL (Home Country Biased Lumper) vs GD3F (Global Diversified 3 Factor).

Comming soon...

==
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

MediumTex,

==
That gold graph doesn't look right.
==

When I first started posting the backtest results using excel graphs, it was pointed out to me by several individuals here that I needed to specify "logarithmic scale" when showing something like the 10K growth path for different portfolio examples.

It changed the look quite a bit to do so.

I bet that is the difference you are noticing.

when I get a few minutes I will post 500 index & GLD with log scale on & off to show the difference.

==
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

The results for 500 index & Gold with Logarithmic Scale enabled...

The results for 500 index & Gold with Logarithmic Scale disabled...

As you can see the results are quite different.

I remember someone mentioning that you could do charts/graphs in the sheet that simba (and others) did but that the results were not displayed with logarithmic scale enabled, and they really needed to be.

I am showing the results with logarithmic scale enabled since several suggested it needed to be, and no one suggested the other results were more appropriate for portfolio comparisons.

Hope this helps.
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

Tramper Al · Joined: 18 Oct 2007 Posts: 2374

Wonk · Joined: 11 Jul 2008 Posts: 204

I'm about to put fresh money into my PP and I'm contemplating the offshore gold portion. What does everyone like for gold holdings outside the U.S?

I already have unallocated @ The Perth Mint, but I feel like diversifying geographically in this category might be important. I was thinking of CEF, but read that it has been trading at a premium to it's NAV.

Anyone have other options? Goldmoney.com? Bullionvault? Safe deposit box with physical coin in Singapore? I'd love to hear all options.

Thanks...

Tramper Al · Joined: 18 Oct 2007 Posts: 2374

makalu · Joined: 05 Feb 2008 Posts: 44

MediumTex · Joined: 01 Mar 2009 Posts: 447

I've heard good things about goldmoney.com.

That idea about a Singapore safe deposit box sounds pretty good. I can visualize the Singapore government rounding up bureaucrats from other countries snooping around for peoples' gold and caning them.
_________________
"A Permanent Portfolio should let you watch the evening news or read investment publications in total serenity."
-Harry Browne

Wonk · Joined: 11 Jul 2008 Posts: 204

Tramper Al · Joined: 18 Oct 2007 Posts: 2374

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Below is a comparison of the PP strategy vs Larry Swedroe strategy.

Where the PP depends on the low/negative correlation of Equity/Gold to stabalize the ride, Larry leans on short term highest quality bonds and holds only equity components with highest expected return and very low correlation (US & Intl SV).

_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

meckaneck · Joined: 31 Jul 2008 Posts: 189

Trev,
Can you provide the yearly performance figures for the Swedroe portfolio as well as what specific instruments comprise the allocation?

dore · Joined: 08 Jan 2008 Posts: 37

Trev,

Another request. Could you re-run the PP numbers using something that would approximate Vanguard's Total World Stock index (VTWSX)?

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Dore,

In this example I showed the original with 500 index only and another with the equity split 50/50 between TSM and Total International.

_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

FYI - using a combination of Long Treasury and Cash seems silly to me.

Below shows what happens if you replace the LTT & Cash with STT.

_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

meckaneck,

Showing the Larry strategy again below at 25% Equity, 75% STT with the Equity split evenly between US and Intl Small Value.

There was another post recently (past 2 weeks I think) with Larrys portfolio as the subject. You might locate it and get more details on the exact components.

Blow are the annual returns by year thru 05/31/2009.

1970 12.49
1971 16.29
1972 12.32
1973 -1.70
1974 -1.71
1975 18.57
1976 17.54
1977 13.37
1978 12.94
1979 10.11
1980 13.79
1981 10.09
1982 19.00
1983 15.77
1984 11.82
1985 23.16
1986 15.68
1987 9.50
1988 12.52
1989 14.81
1990 1.73
1991 13.76
1992 6.49
1993 14.37
1994 1.78
1995 12.26
1996 6.19
1997 6.00
1998 5.18
1999 5.42
2000 8.41
2001 7.04
2002 4.84
2003 14.89
2004 8.06
2005 5.09
2006 8.96
2007 6.13
2008 -4.44
5/31 3.33

==
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

meckaneck · Joined: 31 Jul 2008 Posts: 189

Trev,
Thanks for the reply. One last question, could you please process the performance data for the permanent portfolio using short term treasuries versus cash as I am curious how this would compare to Swedroe's portfolio? I believe this is what CraigR compares to.

GA Ray · Joined: 16 May 2009 Posts: 8

Trev,

How do you define short term treasuries? 1-2 year duration like the ETF SHY?

Thanks,

Ray

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Ray,

A few post up from this one I listed the annual return sources from my 1970-forward data set.

STT = Short Term Treasury
============================================
1970-1977 5 Yr T Notes – 1.5bp/month
1978-1991 US Treasury 1-3 Yr – 1.5bp/month
1992-2008 Vanguards Short Term Treasury Fund

===
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Meckaneck - you asked...

===
One last question, could you please process the performance data for the permanent portfolio using short term treasuries versus cash as I am curious how this would compare to Swedroe's portfolio? I believe this is what CraigR compares to.
===

Do you mean

25% 500 index
25% Gold
25% Long Term Treasury
25% Short Term Treasury (instead of Cash - Tbills/Treas MM ?)

??

Let me know and I will be glad to crunch and post.

May be tomorrow morning though, gettin kids ready for bed here in a bit.

Later !
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

meckaneck · Joined: 31 Jul 2008 Posts: 189

Trev,
Please chart the following:
20% Total Stock Market
5% Int'l Total Stock Market
25% ST Treasuries
25% LT Treasuries
25% Gold

noah09 · Joined: 16 May 2009 Posts: 4

A question: if you're adding to the portfolio from a paycheck, do you put a quarter of the money you're investing in every pay cycle (weekly, biweekly, etc.)? Or do you wait and do it monthly or annually? And how does this affect rebalancing?

craigr · Joined: 13 Mar 2007 Posts: 1973

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Meckaneck,

Here you go..

I did one splitting the bonds between LTT and STT and another making all of the bond allocation ITT.

Again - I don't see the reason for complicating the bond side (by going Long and Short or Long and Cash) vs just picking something in the middle and it sure does not seem prudent given the minor differences in the results.

_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

dore · Joined: 08 Jan 2008 Posts: 37

Thanks Trev.

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Below is a comparison of the PP Strategy vs Larry Swedroe Strategy.

Both portfolio's hold equity components that are not the "Norm".

Equity/Gold - vs - SV,ISV

Larry always suggest that when tilt to equity components with higher expected risk/return (SV, ISV) that you also reduce equity risk by decreasing equity, increasing bonds.

Showing the PP at 50/50 equity/bonds

vs

Larry at 40/60 and 30/70

Annual Returns listed by year.

Roy · Joined: 10 Sep 2008 Posts: 341

Roy · Joined: 10 Sep 2008 Posts: 341

MediumTex · Joined: 01 Mar 2009 Posts: 447

The one thing we haven't seen since 1970 is a period of sustained deflation.

If Paul Krugman and others are correct, we may be looking at four to ten years of a deflationary environment where no matter how much money the government spends the velocity of money remains low enough that deflationary forces cannot be shaken (see Japan).

I wonder how the Larry approach would compare to the PP during such periods. We may be finding out soon.

As for 25% to ST and LT compared to 50% intermediate, I think that there is a strong case to be made for keeping 25% in cash or near cash so that no matter what sort of market dislocations are occurring, you will have a pot of liquid and stable cash you can spend if necessary.

During a tight money recession, without a cash stash, you could see the value of stocks, gold and intermediate bonds falling. If you needed some of your money during such a period you might be forced to liquidate holdings at a loss.

Another practical consideration is that by including a chunk of cash in the PP, it allows you to include ALL of your assets in the portfolio, including your checking and savings accounts, CDs and other highly liquid assets.

There is a segment of the population (especially those who lived through the Depression) that likes the idea of ZERO chance of a loss (my Mother is an example of such a person). For these people, the idea that 25% of the portfolio is absolutely safe and is available to spend if needed at any time is comforting.

The PP is woven together with strands of common sense, market wisdom and subtle insights into psychology and human nature. Even if a 50% allocation to intermediate treasuries might under some conditions provide a slightly higher return, I think that there are other reasons for sticking with the 25% x 4 allocation (including the need to sleep well no matter what the markets are doing).

As I have noted before, the creative deployment of the cash portion into things like I-Bonds when the fixed rate is above 0% can provide a better return than intermediate treasuries may be paying, with zero principal risk, so sitting on cash equivalents doesn't necessarily mean low returns.
_________________
"A Permanent Portfolio should let you watch the evening news or read investment publications in total serenity."
-Harry Browne

larryswedroe · Joined: 22 Feb 2007 Posts: 5368 Location: St Louis MO

Few thoughts
a) I include EMV in my portfolio for diversification and returns reasons (highest expected returning asset class)--say something like 40% US SV, 45% ISV and 15% EMV to have 60% int'l
b) a large advantage IMO is owning TIPS vs ST treasuries so you pick up the term premium without the inflation risk (and get more diversificaiton benefits due to negative correlation with equities and higher volatility

DP · Joined: 17 Apr 2008 Posts: 481

Hi,
Trev, if you're taking requests, I created a portfolio taking ideas from Larry, PP, and Scott Burns: equal parts US SCV, EM, Gold, LT Treasuries, TIPS, and Global Bonds. When data for Global bonds is not available I substitute ST Treasuries. It might be interesting to see how it compares to PP and Larry's portfolio's.

Don

newbie001 · Joined: 24 Nov 2008 Posts: 165

Trev H · Joined: 02 Mar 2007 Posts: 1456 Location: Tennessee

Just a FYI for Larry and others...

The ISV in my data set includes Intl Developed SV and EM SV at 85:15 ratio.

The EM SV is calculated 50/50 from EV and ES.

Details below.

===

ISV – International Small Value including EM
============================================
1970-2008 85% ISV, 15% ESV (data from ifa.com + DFA Funds)
ISV 1970-1981 Dimensional Intl Small Cap Index -4.58 bp/month
--- 1982-1994 Dimensional Intl Small Value Index -5.75 bp/month
--- 1995-Present – DFA Intl Small Cap Value Fund DISVX
ESV = 50% EV, 50% ES *calculated* (ifa.com)
--- EV 1970-1988 IFA EM Index
------ 1989-1998 Fama/French EM Value Index -5.0 bp/month
------ 1999-Present DFA EM Value Fund DFEVX
--- ES 1970-1988 IFA EM Index
------ 1989-1998 Fama/French EM Small Index -6.5 bp/month
------ 1999-Present DFA EM Small Cap Fund DEMSX

===

I have some other examples I want to show and will try to get to the other request perhaps later this evening (after work) or tomorrow morning.

==
_________________
22.5% US LCB, 22.5% US SCV, 10.0% US REIT, 22.5% Intl LCV, 22.5% Intl SCB

"As you can probably tell by now, my sympathies lie with the splitters"

Trev H

craigr · Joined: 13 Mar 2007 Posts: 1973

craigr · Joined: 13 Mar 2007 Posts: 1973

The performance edge you see in a vacuum on a spreadsheet does not transfer to actuality in the implementation. There are expenses involved for taxable and non-taxable investors.

For a taxable investor comparison we need to incorporate not only the annual expense ratio and advisor fees, but also the tax cost ratio that Morningstar reports on fund efficiencies. The tax cost ratio reflects the amount of returns you can expect to lose due to tax impacts each year from the fund's operations.

Then, if you want to be truly evil, you also need to deduct capital gains from your rebalancing each year. *

Vanguard TSM

Expense Ratio: 0.16% (even lower with the ETF 0.07%)
Tax Cost Ratio: 0.39%
Advisor Fee: 0.00%

DFA Intl. Small Value

Expense Ratio: 0.69%
Tax Cost Ratio: 1.25%
Advisor Fee: 1% (can range lower or higher)

DFA Small Cap Value

Expense Ratio: 0.52%
Tax Cost Ratio: 1.80% (!!!)
Advisor Fee: 1% (can range lower or higher)

Total Overhead Costs

Vanguard TSM: 0.55% a year (or lower)
DFA Intl Small Value: 2.94%
DFA Small Value: 3.32%

If you are non-taxable you can deduct the tax cost ratios. Or if you use a non-DFA fund you can get rid of the advisor fee. But the higher expenses are still there regardless. That's at least 0.50% a year reduction regardless under best conditions.

So now when you look at the return difference it's not so clear cut. Assuming the ISV or SV fund can beat the TSM fund going forward (and nobody knows that), you have to deduct real world expenses. In this case you are losing perhaps 3% of the returns per year you are seeing in these spreadsheet results.

Here's the best part: You are losing that money whether or not the fund beats TSM.

So it comes back to Harry. Not Harry Browne, but Dirty Harry: "Do you feel lucky?"

You can buy the TSM and get a guaranteed low expense investing vehicle or you can go for the specialty asset classes and pay much more. If your bet pays off that's great. But if it doesn't (and with a 3% overhead it very well may not), then you'll lose to TSM.

* I have a spreadsheet I call: "The Backtesting Spreadsheet from Hell" - The name is based on William Bernstein's "Retirement Calculator from Hell". It has similar goals of looking at what happens when the investment returns touted in marketing brochures meet reality.

This spreadsheet has not only historical data, but also includes typical expense ratios of funds, morningstar tax cost ratio for funds (with at least 10 years worth of data to capture bull and bear markets), and variable long term capital gains rates to deduct when you rebalance each year (so you can simulate a 15% rate vs. 28% for instance). Also the capital gains rates can use historical rates so you can actually see what the tax losses were for each year when you rebalanced (US avg long term capital gains rate is around 25%).

It is designed not to provide the best case rosy scenario of a portfolio, but the worse case cost and tax burden on a strategy. I don't care when things are going great as much as I care about when things are going poorly.

What I found with my spreadsheet are that more complicated portfolios with specialty asset classes largely do no better than simpler portfolios. Investors who load up on risky funds hoping to get higher returns generally give those extra returns to Uncle Sam, the mutual fund company, an advisor or all three. In essence, you are taking on the risk but others are getting the rewards.

I am not releasing the spreadsheet because I don't want to support it. However I think anyone doing backtests should incorporate these expenses at a minimum to get an idea how their strategy performs under real world conditions. The conclusions I drew from it were "Keep It Simple" and "Costs matter."