Tilting vs Total Market: Does it really matter that much?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Tilting vs Total Market: Does it really matter that much?

Post by 555 » Wed Mar 21, 2012 11:19 am

There's been a lot of debate lately about the relative merits of Tilting (e.g. towards small and value and maybe other things) versus Total Market (e.g. a simple Three Fund Portfolio).

But the question is, does it really matter that much, considering expected returns on a risk adjusted basis, once all considerations are taken into account?

If there's so much debate, it can't be that clear cut which is better.

pkcrafter
Posts: 12106
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Tilting vs Total Market: Does it really matter that much

Post by pkcrafter » Wed Mar 21, 2012 11:34 am

I suppose you should be able to generate somewhat higher returns by tilting over a long period, however, I also think the choice is based more on personality trait than anything else. Some investors have a need to be doing something more than just using a TR fund or Taylor's thrifty three. In fact they could not be content holding just those funds whether they outperform or not. Other investors simply aren't motivated by all the academic stuff.

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.

The Wizard
Posts: 11008
Joined: Tue Mar 23, 2010 1:45 pm
Location: Reading, MA

Re: Tilting vs Total Market: Does it really matter that much

Post by The Wizard » Wed Mar 21, 2012 11:37 am

It varies from year to year; check out that multi-colored bingo chart sometime.
If you're clever enough to rock & roll at the right times, you can easily clear a couple % extra each year.
It beats going to Vegas...
Attempted new signature...

User avatar
nisiprius
Advisory Board
Posts: 34151
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Tilting vs Total Market: Does it really matter that much

Post by nisiprius » Wed Mar 21, 2012 11:37 am

In the illustration of slice-and-dice portfolio construction that Larry Swedroe provides in The Quest For Alpha, at the end of it all, what has been achieved, over the period 1975-2009, is to increase the annualized return from 10.6% to 11.6% while reducing the standard deviation very slightly from 11.3% to 11.2%.

I don't think this illustration is supposed to be an example of the state-of-the-art in portfolio construction, but I think it's a fair illustration of how much it is supposed to matter. So then the arguments go, predictably:

a) That's backtesting./Yes, but it's not just empirical, the assets were not just selected arbitrarily to get those results, and there's somewhat convincing theory to suggest that it should continue to work, going forward.

b) It doesn't convince me./Yes, but the people who produced the theory are both better informed and smarter than you.

c) 1% isn't a big difference./Are you kidding? It's a huge difference. Compound that sucker out for forty years and see what you get.

d) Yes, but will you really get it?/No, not necessarily, but them's the odds.

The big thing for me is that buying low-cost index funds with like 0.20% ER gets you about a 1% edge over buying average index funds with like 1.20% ERs, and that's an absolutely sure thing. The claimed benefits of slice-and-dice are in roughly the same ballpark as the benefits of low-cost investing. But it is far less cut-and-dried that they actually exist, and far less cut-and-dried that you'll actually see them within your investing lifetime.

Plus of course the funds that are the precision tools for the slice-and-dice job have somewhat higher costs than total market index funds; and if they're DFA funds you need to pay an advisor. Let's say the claims of 1% higher returns are a little exaggerated, and it's really only 0.75%. Then let's say you replace some of your Vanguard Total Stock Market Index (Admiral shares) with 0.05% expense ratio with DFA US Small Cap Value Fund with 0.52% ER. Then let's say you pay an advisor 0.25% to 1.4% of your portfolio every year to get access to DFA funds. Well, little by little that theoretical extra 1% does get eroded away. Maybe not completely. But by the time you translate theory into practice, that 1% is starting to look more like 0.25%-0.50%, isn't it?

But anyway: how much does it really matter? According to the advocates, no more than about 1% in annual return.
Last edited by nisiprius on Wed Mar 21, 2012 11:50 am, edited 2 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
Sammy_M
Posts: 1863
Joined: Sun Nov 25, 2007 8:30 am

Re: Tilting vs Total Market: Does it really matter that much

Post by Sammy_M » Wed Mar 21, 2012 11:47 am

Excellent summary, nisi. Let me just add:

e) Most people will abandon S/V tilts as soon as they under-perform for a few years. / You're right, but I'm not one of those people (I think :? ).

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: Tilting vs Total Market: Does it really matter that much

Post by yobria » Wed Mar 21, 2012 11:50 am

Tilting vs Total Market: Does it really matter that much?


In general, No. "Tilting", or modestly personalizing your portfolio in any other way, won't dramatically effect returns. There are no free lunches. Extreme tilting, however, is dangerous - if you find 50% of your stock portfolio is concentrated in 5% of the market, for example, you're gambling, not investing, and exposing yourself to grave risks that could be diversified away.

When it doubt, keep it simple.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Tilting vs Total Market: Does it really matter that much

Post by 555 » Wed Mar 21, 2012 12:13 pm

yobria wrote:
Tilting vs Total Market: Does it really matter that much?

In general, No. "Tilting", or modestly personalizing your portfolio in any other way, won't dramatically effect returns. There are no free lunches. Extreme tilting, however, is dangerous - if you find 50% of your stock portfolio is concentrated in 5% of the market, for example, you're gambling, not investing, and exposing yourself to grave risks that could be diversified away.
When it doubt, keep it simple.

That's why I said "expected returns on a risk adjusted basis". If one put all stocks into small value then presumably they could increase the fixed income allocation to have the same risk.

KyleAAA
Posts: 6325
Joined: Wed Jul 01, 2009 5:35 pm
Contact:

Re: Tilting vs Total Market: Does it really matter that much

Post by KyleAAA » Wed Mar 21, 2012 12:20 pm

Realistically, I expect the returns from my tilted portfolio to range from -0.5% to +1% relative to a total market portfolio. I personally believe there is a bit more upside than downside based on the evidence I've seen over the years, but there are no guarantees. Add to that the fact that tilting will probably modestly increase your costs and I don't think it's difficult to argue the pro-total-market side. It's a bet. An educated bet, but a bet nonetheless.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Tilting vs Total Market: Does it really matter that much

Post by 555 » Wed Mar 21, 2012 12:30 pm

nisiprius wrote:...c) 1% isn't a big difference./Are you kidding? It's a huge difference. Compound that sucker out for forty years and see what you get.
d) Yes, but will you really get it?/No, not necessarily, but them's the odds.
The big thing for me is that buying low-cost index funds with like 0.20% ER gets you about a 1% edge over buying average index funds with like 1.20% ERs, and that's an absolutely sure thing. The claimed benefits of slice-and-dice are in roughly the same ballpark as the benefits of low-cost investing. But it is far less cut-and-dried that they actually exist, and far less cut-and-dried that you'll actually see them within your investing lifetime....But anyway: how much does it really matter? According to the advocates, no more than about 1% in annual return.

Certainly everyone here agrees that 1%/yr is huge, especially when it's a sure thing, and that's why lowering costs, tax efficiency etc. are things that are easily agreed upon here. But comparing two portfolios, it seems much harder to draw definitive conclusions.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: Tilting vs Total Market: Does it really matter that much

Post by yobria » Wed Mar 21, 2012 12:34 pm

555 wrote:
yobria wrote:
Tilting vs Total Market: Does it really matter that much?

In general, No. "Tilting", or modestly personalizing your portfolio in any other way, won't dramatically effect returns. There are no free lunches. Extreme tilting, however, is dangerous - if you find 50% of your stock portfolio is concentrated in 5% of the market, for example, you're gambling, not investing, and exposing yourself to grave risks that could be diversified away.
When it doubt, keep it simple.

That's why I said "expected returns on a risk adjusted basis". If one put all stocks into small value then presumably they could increase the fixed income allocation to have the same risk.


We've shown in recent threads that value is no riskier than growth in actual fund results, have no higher returns, and funds can't actually capture the theoretical Fama/French "value premium" anyway. So will limit my response to small/micro caps, which are in fact somewhat riskier than large caps, so we can expect slightly higher (maybe 1%/year) long term returns, as you say.

Question: Could you increase both your small cap and bond exposure and have the same risk profile as before? Sure, maybe, modestly, but going from stocks to bonds at current yields is swapping, say, an 8% expected return for a 3% return, so you're going to have to add 5% more small caps for every 1% increase in bonds.

And you certainly wouldn't want a 100% small cap allocation! They can underperform dramatically for long periods of time. You're going to accept a large amount of uncompensated risk. That's just a step away from buying individual stocks. Beware recency bias.

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by bob90245 » Wed Mar 21, 2012 12:35 pm

555 wrote:If there's so much debate, it can't be that clear cut which is better.

No, there could be other interpretations. There is so much debate because people are very passionate about their side and put forward pursuasive arguments to a) justify their position b) convert the other side to their position.
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Tilting vs Total Market: Does it really matter that much

Post by 555 » Wed Mar 21, 2012 12:56 pm

bob90245 wrote:
555 wrote:If there's so much debate, it can't be that clear cut which is better.

No, there could be other interpretations. There is so much debate because people are very passionate about their side and put forward pursuasive arguments to a) justify their position b) convert the other side to their position.

Yes, I should have clarified that. Obviously the mere existence of a vigorous debate absolutely does not imply equal validity to both sides, and I can easily think of debates where they are not equal (e.g. round earth vs flat earth).

But in a Tilting vs Total Market comparison, genuine experts can differ. That's why I wonder if it really matters that much?

dbr
Posts: 23722
Joined: Sun Mar 04, 2007 9:50 am

Re: Tilting vs Total Market: Does it really matter that much

Post by dbr » Wed Mar 21, 2012 1:08 pm

555 wrote:
bob90245 wrote:
555 wrote:If there's so much debate, it can't be that clear cut which is better.

No, there could be other interpretations. There is so much debate because people are very passionate about their side and put forward pursuasive arguments to a) justify their position b) convert the other side to their position.

Yes, I should have clarified that. Obviously the mere existence of a vigorous debate absolutely does not imply equal validity to both sides, and I can easily think of debates where they are not equal (e.g. round earth vs flat earth).

But in a Tilting vs Total Market comparison, genuine experts can differ. That's why I wonder if it really matters that much?


Still another possibility is lack of clarity about how a tilted portfolio actually behaves and lack of clarity about why those results would or would not actually be better for anyone in particular. I'm not sure that genuine experts do differ, though if they do that could be because they are judging the benefits relative to different criteria. I wonder, by the way, what names we would name as being genuine experts. I think it is true that Fama and/or French, who do understand the three factor model for returns, have not ventured beyond saying index investing in TSM is the best choice for most investors to actually implement.

User avatar
wjo
Posts: 466
Joined: Fri Jan 22, 2010 12:17 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by wjo » Wed Mar 21, 2012 1:14 pm

yobria wrote:We've shown in recent threads that value is no riskier than growth in actual fund results, have no higher returns, and funds can't actually capture the theoretical Fama/French "value premium" anyway. So will limit my response to small/micro caps, which are in fact somewhat riskier than large caps, so we can expect slightly higher (maybe 1%/year) long term returns, as you say.


Just to be clear, when Yobria says "we've" he means he says and refutes all evidence to the contrary claiming the respondent is cherry picking the selected data series while putting up his own data that by his standards can also be said to be cherry picked.

That said, I believe that for a host of reasons most folks shouldn't slice and dice and should stick with the simple 3-4 fund portfolio.

dbr
Posts: 23722
Joined: Sun Mar 04, 2007 9:50 am

Re: Tilting vs Total Market: Does it really matter that much

Post by dbr » Wed Mar 21, 2012 1:23 pm

wjo wrote:That said, I believe that for a host of reasons most folks shouldn't slice and dice and should stick with the simple 3-4 fund portfolio.


And so typically when people are offering opinions about "what one should do," it would stand to reason that there would not be agreement.

I also believe that for a variety of reasons most folks should not slice and dice and I suggest that without disputing established facts about the behavior of such portfolios.

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by bob90245 » Wed Mar 21, 2012 1:48 pm

555 wrote:
bob90245 wrote:
555 wrote:If there's so much debate, it can't be that clear cut which is better.

No, there could be other interpretations. There is so much debate because people are very passionate about their side and put forward pursuasive arguments to a) justify their position b) convert the other side to their position.

Yes, I should have clarified that. Obviously the mere existence of a vigorous debate absolutely does not imply equal validity to both sides, and I can easily think of debates where they are not equal (e.g. round earth vs flat earth).

But in a Tilting vs Total Market comparison, genuine experts can differ. That's why I wonder if it really matters that much?

To the contrary, if it doesn't really matter that much, would you see so much passion on both sides? No! Of course, not. No one expends all their energy if, at the end of the day, it really doesn't matter that much.[1]



[1] With the possible except of those who debate just for arguments sake. LOL
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Tilting vs Total Market: Does it really matter that much

Post by 555 » Wed Mar 21, 2012 1:55 pm

bob90245 wrote:To the contrary, if it doesn't really matter that much, would you see so much passion on both sides? No! Of course, not. No one expends all their energy if, at the end of the day, it really doesn't matter that much.[1]

[1] With the possible except of those who debate just for arguments sake. LOL

Actually there is a well-known penomenon of smaller differences, often yielding more intense debates.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: Tilting vs Total Market: Does it really matter that much

Post by yobria » Wed Mar 21, 2012 2:07 pm

wjo wrote:
yobria wrote:We've shown in recent threads that value is no riskier than growth in actual fund results, have no higher returns, and funds can't actually capture the theoretical Fama/French "value premium" anyway. So will limit my response to small/micro caps, which are in fact somewhat riskier than large caps, so we can expect slightly higher (maybe 1%/year) long term returns, as you say.


Just to be clear, when Yobria says "we've" he means he says and refutes all evidence to the contrary claiming the respondent is cherry picking the selected data series while putting up his own data that by his standards can also be said to be cherry picked.


If you've got any evidence (a simple regression is all it takes, as here: http://research-repository.st-andrews.a ... thesis.pdf ) that any index fund has ever captured the FF value premium, please link to it. I'm open to any new information, there just doesn't appear to be any.

There

vesalius
Posts: 775
Joined: Tue Jul 13, 2010 7:00 pm
Location: Texas

Re: Tilting vs Total Market: Does it really matter that much

Post by vesalius » Wed Mar 21, 2012 2:21 pm

wjo wrote:
yobria wrote:We've shown in recent threads that value is no riskier than growth in actual fund results, have no higher returns, and funds can't actually capture the theoretical Fama/French "value premium" anyway. So will limit my response to small/micro caps, which are in fact somewhat riskier than large caps, so we can expect slightly higher (maybe 1%/year) long term returns, as you say.


Just to be clear, when Yobria says "we've" he means he says and refutes all evidence to the contrary claiming the respondent is cherry picking the selected data series while putting up his own data that by his standards can also be said to be cherry picked.

That said, I believe that for a host of reasons most folks shouldn't slice and dice and should stick with the simple 3-4 fund portfolio.
Agreed on all accounts.

User avatar
wjo
Posts: 466
Joined: Fri Jan 22, 2010 12:17 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by wjo » Wed Mar 21, 2012 2:21 pm

yobria wrote:
If you've got any evidence (a simple regression is all it takes, as here: http://research-repository.st-andrews.a ... thesis.pdf ) that any index fund has ever captured the FF value premium, please link to it. I'm open to any new information, there just doesn't appear to be any.

There


For all those reading - do a search on the recent threads and you'll see lots of people posting evidence. In fact, Robert T just posted his own analysis of the data from the referenced paper, disagreeing with it!

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by bob90245 » Wed Mar 21, 2012 2:26 pm

555 wrote:
bob90245 wrote:To the contrary, if it doesn't really matter that much, would you see so much passion on both sides? No! Of course, not. No one expends all their energy if, at the end of the day, it really doesn't matter that much.[1]

[1] With the possible except of those who debate just for arguments sake. LOL

Actually there is a well-known penomenon of smaller differences, often yielding more intense debates.

Looks like my footnote [1] above was not a joke after all! LOL
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by Clearly_Irrational » Wed Mar 21, 2012 2:27 pm

555 wrote:That's why I said "expected returns on a risk adjusted basis". If one put all stocks into small value then presumably they could increase the fixed income allocation to have the same risk.


That's what I use it for. I prefer lower beta, higher skew, and lower kurtosis than normal. Small/Value is one piece of the puzzle that helps me to achieve that.

All investing is based on odds. I like the odds, but your milage may vary.

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by Beagler » Wed Mar 21, 2012 2:30 pm

yobria wrote:If you've got any evidence (a simple regression is all it takes, as here: http://research-repository.st-andrews.a ... thesis.pdf )....


That link takes the reader to a grad student thesis, not a paper published in a peer-reviewed journal. RobertT has found some of the thesis' data suspect.
viewtopic.php?f=10&t=92793&start=150
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

jginseattle
Posts: 686
Joined: Fri Jul 01, 2011 7:33 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by jginseattle » Wed Mar 21, 2012 4:45 pm

Tilting can also add diversification. You want to have sufficient exposure to all equity asset classes (market, value and small) to guard against periods of underperformance.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: Tilting vs Total Market: Does it really matter that much

Post by yobria » Wed Mar 21, 2012 5:15 pm

Beagler wrote:
yobria wrote:If you've got any evidence (a simple regression is all it takes, as here: http://research-repository.st-andrews.a ... thesis.pdf )....


That link takes the reader to a grad student thesis, not a paper published in a peer-reviewed journal. RobertT has found some of the thesis' data suspect.
viewtopic.php?f=10&t=92793&start=150


If a paper in a full peer reviewed journal will satisfy you, (now) Professor Kenneth Scislaw, Ph.D., has co-authored the following article:

Scislaw, Kenneth and McMillan, David G. . "The Search for an Exploitable Value Premium in Market Indexes" Journal of Asset Management . (forthcoming)

http://www.lebow.drexel.edu/Faculty/KennethScislaw.html

But, since that paper will likely conflict with your need to believe in a capturable value effect, I'm sure that data will be suspect too. Why, I have a feeling this man's whole career is going to be one big fraud. :) Otherwise the whole DFA value effect song and dance was just a marketing ploy. And we know big financial companies are only looking out for our best interests, so that can't be true!

User avatar
Clearly_Irrational
Posts: 3087
Joined: Thu Oct 13, 2011 3:43 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by Clearly_Irrational » Wed Mar 21, 2012 5:29 pm

yobria wrote:If you've got any evidence (a simple regression is all it takes, as here: http://research-repository.st-andrews.a ... thesis.pdf ) that any index fund has ever captured the FF value premium, please link to it. I'm open to any new information, there just doesn't appear to be any.

There


I haven't finished reading the entire paper yet, but halfway through it seems more like he's just having trouble explaining the effect rather than dismissing it.

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by Beagler » Wed Mar 21, 2012 6:15 pm

yobria wrote:
Beagler wrote:
yobria wrote:If you've got any evidence (a simple regression is all it takes, as here: http://research-repository.st-andrews.a ... thesis.pdf )....


That link takes the reader to a grad student thesis, not a paper published in a peer-reviewed journal. RobertT has found some of the thesis' data suspect.
viewtopic.php?f=10&t=92793&start=150


If a paper in a full peer reviewed journal will satisfy you, (now) Professor Kenneth Scislaw, Ph.D., has co-authored the following article:

Scislaw, Kenneth and McMillan, David G. . "The Search for an Exploitable Value Premium in Market Indexes" Journal of Asset Management . (forthcoming)

http://www.lebow.drexel.edu/Faculty/KennethScislaw.html


Citing an as-yet unpublished work as "proof" is as close to grasping at straws as I can imagine. You have no idea what are the data, or the conclusions, do you?

yobria wrote:Why, I have a feeling this man's whole career is going to be one big fraud. :)

Why do you say that? He spent two decades working "with retail, institutional, buy-side, sell-side, and academic and investment research segments of the finance profession," including the likes of Merrill Lynch and Templeton.

Dr. Scislaw entered the academic profession after two decades of work in institutional investments, including Merrill Lynch trading in New York, Templeton Investment Counsel in both Florida and the Bahamas, and his own institutional investment consulting firm. He has worked with retail, institutional, buy-side, sell-side, and academic and investment research segments of the finance profession. http://tinyurl.com/77qzho7

yobria wrote:And we know big financial companies are only looking out for our best interests, so that can't be true!

Are you saying that Merill Lynch and Templeton Investments aren't looking out for the best interests of their customer?
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

bschultheis
Posts: 200
Joined: Mon Jul 23, 2007 10:15 am

Re: Tilting vs Total Market: Does it really matter that much

Post by bschultheis » Wed Mar 21, 2012 6:35 pm

jginseattle wrote:Tilting can also add diversification. You want to have sufficient exposure to all equity asset classes (market, value and small) to guard against periods of underperformance.


JG, you sum up nicely a significant reason to tilt. The period 1995 to 1999, and then 2000 to 2010 is a prime example of periods where it is nice to not have all eggs in the underperfoming equity class.

Whether one tilts or not is not nearly as important as choosing a strategy, and then sticking with it when things aren't going your way.

Bill

Muchtolearn
Posts: 1563
Joined: Sun Dec 25, 2011 10:41 am

Re: Tilting vs Total Market: Does it really matter that much

Post by Muchtolearn » Wed Mar 21, 2012 6:39 pm

KISS.

User avatar
bob90245
Posts: 6511
Joined: Mon Feb 19, 2007 8:51 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by bob90245 » Wed Mar 21, 2012 7:56 pm

bschultheis wrote:
jginseattle wrote:Tilting can also add diversification. You want to have sufficient exposure to all equity asset classes (market, value and small) to guard against periods of underperformance.

JG, you sum up nicely a significant reason to tilt. The period 1995 to 1999, and then 2000 to 2010 is a prime example of periods where it is nice to not have all eggs in the underperfoming equity class.

Whether one tilts or not is not nearly as important as choosing a strategy, and then sticking with it when things aren't going your way.

Bill

Or said another way, period 1995 to 1999 was easy to stick with the market-only strategy. But it was very hard to stick with it from 2000 to 2010. Reverse can be said for those who had sufficient exposure to all equity asset classes (market, value, small, REIT, international).
Ignore the market noise. Keep to your rebalancing schedule whether that is semi-annual, annual or trigger bands.

yobria
Posts: 5978
Joined: Mon Feb 19, 2007 11:58 pm
Location: SF CA USA

Re: Tilting vs Total Market: Does it really matter that much

Post by yobria » Wed Mar 21, 2012 9:15 pm

jginseattle wrote:Tilting can also add diversification. You want to have sufficient exposure to all equity asset classes (market, value and small) to guard against periods of underperformance.


But are you gaining diversification here, or losing it? Consider a portfolio that's 50% TSM, 50% Microsoft stock. MSFT and TSM surely won't be perfectly correlated, so you've gained diversification in that sense. You've guarded against periods when TSM underperforms MSFT.

The problem is, you've lost economic diverisification. By overweighting MSFT, you've made a bet that it will outperform all the other stocks you underweighted. And you've accepted uncompensated risk, unique stock risk that can be diversified away.

Overweighting this or that corner of the market has the same effect. This is why Jack Bogle says if you feel the urge to tilt, don't do it with more than 5-10% of your portfolio.

stlutz
Posts: 4015
Joined: Fri Jan 02, 2009 1:08 am

Re: Tilting vs Total Market: Does it really matter that much

Post by stlutz » Thu Mar 22, 2012 12:38 am

[1] With the possible except of those who debate just for arguments sake. LOL


Since our portfolios are all so simple and require so little time to mange, we all have lots of time to debate on bogleheads! (That said, am I the only one who is getting a little annoyed that disagreements with what yobria said "on that other thread" are being brought into 10 other ones--to the point of copying and pasting the same comment?--just sayin').

But the question is, does it really matter that much, considering expected returns on a risk adjusted basis, once all considerations are taken into account?


Keep in mind the big picture. Costs are guaranteed alpha--no hope involved. Staying the course (i.e. not going 100% stocks in 1999 and going 100% bonds in 2009) also makes a big difference. Diversification between stocks and bonds is also important. Doing those things are in the overall scheme much more important and are things that most everyone here agrees upon.

If there's so much debate, it can't be that clear cut which is better.


Another thing we can all agree on is that the future is uncertain. So, if it doesn't seem clear cut your conclusion is correct. Unknown future events will play the dominant role over your future returns. Which again means that focusing on what you can control (costs, diversification) is more important.

User avatar
Kevin M
Posts: 8868
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tilting vs Total Market: Does it really matter that much

Post by Kevin M » Thu Mar 22, 2012 3:05 am

I just posted this in another thread, but it applies just as much here. Here is a blog post I wrote almost a year ago, attempting to summarize the arguments for tilting vs. total market: Lumpers vs. Splitters. Here is the last paragraph of the post:

No one knows whether the lumpers or the splitters will do better over the next 20, 30 or 40 years. Many financial advisors and authors write and talk as if they are very confident that one or the other strategy is superior, but remember that there is academic theory and research to support both positions. I am not highly confident that tilting away from the market portfolio is a superior strategy, but that is the way I lean.


If you bother to read the post, feel free to let me know if I missed any of the major arguments, or if you think I misrepresented either point of view.

Kevin
||.......|| Suggested format for Asking Portfolio Questions (edit original post)

dbr
Posts: 23722
Joined: Sun Mar 04, 2007 9:50 am

Re: Tilting vs Total Market: Does it really matter that much

Post by dbr » Thu Mar 22, 2012 8:40 am

Kevin, I think the crux of the argument lies in the issue of what it means for an approach to be "superior." I will say right up front that all the evidence is that S&D is superior and also, relative to my needs and objectives, "So what?" The question, then, is what needs and objectives does a particular investor have that would motivate whatever additional costs and hazards are involved to elect a S&D portfolio. I think there is a tendency for the possible superiority to be regarded as so self evident that what it actually is gets lost in the discussion, if it exists at all.* The possible costs and downsides are freely admitted to exist, certainly. Your article alludes to the much mentioned problem of tolerance for "tracking error," as they call it. I think you do address the classic Pareto philosophy that total market investing achieves almost all the possible benefit at minimum cost and hazard. I think a powerful argument is needed to motivate going beyond that, which is really Mr. Bogle's point.

*One of the most convincing arguments I hear is Larry Swedroe's suggestion of a heavily tilted equity allocation offset by a higher overall allocation to bonds. The advantage is supposed to be reduction of low side fat tail risk. The hazard is that the equity allocation is significantly exposed to SV volatility, the tracking error hazard. I have not noticed a lot of enthusiasm to adopt that approach in discussions here.

A second analysis that isn't presented is how tilted portfolios actually perform in retirement ruin scenarios, which should be a more relevant evaluation than a more abstract risk adjusted return measure.

On another thread there has been discussion of what sort of asset allocation funds there are in the market. There is an interesting approach there by Ishares, which involves targeting risk rather than return, which is sort of upside down compared to the usual obsession with return.

dbr
Posts: 23722
Joined: Sun Mar 04, 2007 9:50 am

Re: Tilting vs Total Market: Does it really matter that much

Post by dbr » Thu Mar 22, 2012 8:50 am

I'm going to post here a point of view I posted in the other thread, which is this:

"I would say this underlines my point that slice and dice investing strategy is all about finding a way to pursue a particular investing objective or another where a simple total market approach is for some reason considered inadequate."

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

Re: Tilting vs Total Market: Does it really matter that much

Post by Rodc » Fri Mar 23, 2012 12:35 pm

yobria wrote:
Tilting vs Total Market: Does it really matter that much?


In general, No. "Tilting", or modestly personalizing your portfolio in any other way, won't dramatically effect returns. There are no free lunches. Extreme tilting, however, is dangerous - if you find 50% of your stock portfolio is concentrated in 5% of the market, for example, you're gambling, not investing, and exposing yourself to grave risks that could be diversified away.

When it doubt, keep it simple.


Well, something like vanguard's small value has 1000 stocks spread over all conceivable industries, and the top 10 companies are only 4.3% of the portfolio so not dominated by a small number of companies. It is heavier in financials, but some of those are REITS not just banks etc.

So hard to see the wild risk. It is not like that 5% is just a couple of industries.

Unless one believes there is a correct theory of small and value that is the polar opposite of FF, ie we live in an anti-matter sort of universe where not only are FF wrong, but the world is a mirror image of FF, at worst SV provides a very widely diversified low cost portfolio where risk is risk and the expected returns are simply proportional to risk, ie a simple 1D model of the world is pretty close to reality. The primary long term risk is simply that (1) you lose a small fraction of one percent due to higher costs and (2) you are disappointed that tilting failed to provide a benefit. The risk seems to be no more than the risk of not choosing the optimal split between US and non-US stocks.
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.

User avatar
Kevin M
Posts: 8868
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tilting vs Total Market: Does it really matter that much

Post by Kevin M » Fri Mar 23, 2012 1:51 pm

bschultheis wrote:
jginseattle wrote:Tilting can also add diversification. You want to have sufficient exposure to all equity asset classes (market, value and small) to guard against periods of underperformance.


JG, you sum up nicely a significant reason to tilt. The period 1995 to 1999, and then 2000 to 2010 is a prime example of periods where it is nice to not have all eggs in the underperfoming equity class.

Yes. This is the way I stated it in my Lumpers vs. Splitters blog post (referenced above):

In considering whether or not to further diversify, I think more about the possibility of a poor outcome than the probability of beating the total US stock market return. The total US stock market is dominated by large-cap US stocks, even though it includes mid-cap and small-cap stocks (for example, over many periods, you can barely see the difference if you compare a chart of the Vanguard Total Stock Market Index fund to the S&P 500, a US large-cap index). I would rather be somewhat more diversified among international, small-cap and value stocks in case large US growth stocks have an extended period of poor performance. I am willing to accept the possibility that large US stocks could be the best performing asset class over the next 20 years, and that by diversifying among other stock asset classes, I’ll underperform relative to the S&P 500.


bschultheis wrote:Whether one tilts or not is not nearly as important as choosing a strategy, and then sticking with it when things aren't going your way.

Thanks for reinforcing this Bill. This is what I was getting at in this post.

Kevin
||.......|| Suggested format for Asking Portfolio Questions (edit original post)

555
Posts: 4955
Joined: Thu Dec 24, 2009 7:21 am

Re: Tilting vs Total Market: Does it really matter that much

Post by 555 » Sat Mar 24, 2012 12:53 am

Kevin M wrote:I just posted this in another thread, but it applies just as much here. Here is a blog post I wrote almost a year ago, attempting to summarize the arguments for tilting vs. total market: Lumpers vs. Splitters. Here is the last paragraph of the post:
No one knows whether the lumpers or the splitters will do better over the next 20, 30 or 40 years. Many financial advisors and authors write and talk as if they are very confident that one or the other strategy is superior, but remember that there is academic theory and research to support both positions. I am not highly confident that tilting away from the market portfolio is a superior strategy, but that is the way I lean.

If you bother to read the post, feel free to let me know if I missed any of the major arguments, or if you think I misrepresented either point of view.
Kevin

One major factor is what people have in their 401k-type plans. You hope to be able cobble together some type of 3-fund portfolio with modest expenses. Making a tilted portfolio (at low cost) is not even an option, with a lot of typical 401k-type fund lineups, so the decision is moot.

User avatar
Kevin M
Posts: 8868
Joined: Mon Jun 29, 2009 3:24 pm
Contact:

Re: Tilting vs Total Market: Does it really matter that much

Post by Kevin M » Sat Mar 24, 2012 1:31 am

555 wrote:One major factor is what people have in their 401k-type plans. You hope to be able cobble together some type of 3-fund portfolio with modest expenses. Making a tilted portfolio (at low cost) is not even an option, with a lot of typical 401k-type fund lineups, so the decision is moot.

Not only that, but most people don't have the foggiest idea what tilting is or how to determine if it might make sense for them. So, I agree that for most people, a total market approach probably is the way to go.

However, if you want to tilt, you can do it in your IRA. That's what two of my adult children do. Large blend US and international in the 401k or 403b plans, and some of the other (sub) asset classes in their IRAs.

Oh, and even cobbling together a low-cost 3-fund portfolio in a 40X plan is sometimes impossible. One of my kids has no equity fund with an ER of less than about 1.5% in her 401k, and no match to boot.

Kevin
||.......|| Suggested format for Asking Portfolio Questions (edit original post)

Beagler
Posts: 3442
Joined: Sun Dec 21, 2008 7:39 pm

Re: Tilting vs Total Market: Does it really matter that much

Post by Beagler » Sat Mar 24, 2012 1:04 pm

bschultheis wrote:
jginseattle wrote:Tilting can also add diversification. You want to have sufficient exposure to all equity asset classes (market, value and small) to guard against periods of underperformance.


JG, you sum up nicely a significant reason to tilt. The period 1995 to 1999, and then 2000 to 2010 is a prime example of periods where it is nice to not have all eggs in the underperfoming equity class.

Bill


Bill, your book http://tinyurl.com/6myrcoe is a very accessible way for novice investors to learn about asset class diversification. (BTW, used your pumpkin pie recipe in the book for Thanksgiving years ago -- nice!) The model Coffeehouse Portfolio served as a starting point for many investors.

Image

The results speak for themselves. http://tinyurl.com/8xgp8d4
Image

IIRC these results were obtained with VG funds (open to correction).

I see from your site that you now employ mostly DFA funds.
Image

555 wrote:There's been a lot of debate lately about the relative merits of Tilting (e.g. towards small and value and maybe other things) versus Total Market (e.g. a simple Three Fund Portfolio).

But the question is, does it really matter that much, considering expected returns on a risk adjusted basis, once all considerations are taken into account?


Investors who are not comfortable with tracking error should not employ tilts IMHO. Compounded over time, the rewards for tilting have been economically significant, but not every investor is comfortable with the attendant risk. They should recognize that TSM will provide them an essentially LCB portfolio. They will harvest little, if any, contribution from smaller stocks or anything more valuey than held by a cap-weighted TSM fund. And they may fine with that.

Here's a comparison of TSM to VG's Large Cap Index Fund, from the inception date of the latter:
Image
“The only place where success come before work is in the dictionary.” Abraham Lincoln. This post does not provide advice for specific individual situations and should not be construed as doing so.

Post Reply