In a 3-factor world, why wouldn't you go full tilt?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Mortgasm
Posts: 263
Joined: Mon Feb 06, 2012 7:54 pm

In a 3-factor world, why wouldn't you go full tilt?

Post by Mortgasm » Sun Feb 26, 2012 12:04 pm

First: If you don't accept a 3-factor model, that's fine. This isn't a debate about that. It's been done a lot already on this board.

However, If you do accept a 3-factor (or 4 if you like) model: Wouldn't you go with a full-tilt portfolio that is just almost entirely US/EM/I small cap value as the core equity components?

For reference: This has been discussed a lot in these threads as "Larry Swedroe's 's portfolio".
http://www.bogleheads.org/forum/viewtopic.php?t=31166
http://www.bogleheads.org/forum/viewtopic.php?t=66677

My questions are:

'Why don't more 3-factor people go full tilt? " None of the recommended portfolios in books, websites or advisors I have seen are 'full tilt', but they all reference Fama/French. (Except Larry.)
Or, expressed another way, why would a 3-factor investor ever want to buy just 'pure' beta like VTSMX or even DFA Core Equity 1 & 2? (Yes, In know there are some loads in those, but not much.)
'Do they do it just to avoid tracking error/regret?'
'Is it a hedge, in case the 3-factor model isn't right?'

User avatar
Rick Ferri
Posts: 8201
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Rick Ferri » Sun Feb 26, 2012 12:15 pm

Because there have been long periods of time when small-value tilts underperformed to market. And like everything else in investing, people forget why they did things when they start under-performing. That leads to trend-following trouble.

In my article, Expect Years of Pain before Market Gain, I calculated the longest periods since 1950 when small value stocks underperformed the rest of the market. The record was 18 years and the second longest was 15 years.

Rick Ferri


.
Last edited by Rick Ferri on Sun Feb 26, 2012 12:17 pm, edited 2 times in total.
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

livesoft
Posts: 55514
Joined: Thu Mar 01, 2007 8:00 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by livesoft » Sun Feb 26, 2012 12:16 pm

Because of years just like 2011.
This signature message sponsored by sscritic: Learn to fish.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by yobria » Sun Feb 26, 2012 12:27 pm

Are you sure tech stocks (Google, Apple, Amazon, etc) must do poorly over your time horizon, while banks prosper?

Are you sure?

Mortgasm
Posts: 263
Joined: Mon Feb 06, 2012 7:54 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Mortgasm » Sun Feb 26, 2012 12:41 pm

Rick Ferri wrote:Because there have been long periods of time when small-value tilts underperformed to market. And like everything else in investing, people forget why they did things when they start under-performing. That leads to trend-following trouble.

In my article, Expect Years of Pain before Market Gain, I calculated the longest periods since 1950 when small value stocks underperformed the rest of the market. The record was 18 years and the second longest was 15 years.
.


Thanks, Rick. I had read your article already. Time horizon is important. In my case it's 25 years before we expect to be in a withdrawal phase. But even then, shouldn't you mitigate the downside risk with bonds and not with TSM?

yobria wrote:Are you sure tech stocks (Google, Apple, Amazon, etc) must do poorly over your time


This is just an argument against 3F, which you've made in every post I have ever read. I commend you on your persistence! But, I'm not trying to convince you, or me or anyone that 3f is right. I'm asking 'if you accept 3F, wouldn't you go full tilt?'

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by yobria » Sun Feb 26, 2012 12:47 pm

Mortgasm wrote:
yobria wrote:Are you sure tech stocks (Google, Apple, Amazon, etc) must do poorly over your time


This is just an argument against 3F, which you've made in every post I have ever read. I commend you on your persistence! But, I'm not trying to convince you, or me or anyone that 3f is right. I'm asking 'if you accept 3F, wouldn't you go full tilt?'


Not as persistent as my friends at DFA. I commend them on their persistence! If, my "accept", you mean, "follow blindly and with complete certainty", you'd probably want to use options (long/short positions) and leverege to maximize your exposure. You could also borrow from friends and family.

Edit to add: My opinions on Rick's answer - it's true that S/V stocks can unperform for long periods of time. It's untrue that they must, at some point, leap forward and make up for all this underperformance (as that answer implies). And I disagree that people's lack of ability to stay the course is the reason one shouldn't go all in (which implies S/V must outperform, it's only the investor that's at fault). The reason is simply that this is a bet on a free lunch, and that bet may not pay off, whether you stay the course or not.
Last edited by yobria on Sun Feb 26, 2012 1:00 pm, edited 1 time in total.

dbr
Posts: 22515
Joined: Sun Mar 04, 2007 9:50 am

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by dbr » Sun Feb 26, 2012 12:53 pm

An option, suggested by Larry Swedroe, is to go full tilt and dilute the risk by reducing equity exposure altogether. If the data Larry is citing for this is valid, a particular benefit would be reduction of fat-tail risk on the low side.

User avatar
Rick Ferri
Posts: 8201
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Rick Ferri » Sun Feb 26, 2012 12:55 pm

Thanks, Rick. I had read your article already. Time horizon is important. In my case it's 25 years before we expect to be in a withdrawal phase. But even then, shouldn't you mitigate the downside risk with bonds and not with TSM?


I'm not saying don't do it. All I'm saying is know the risk, which is how you may react during long periods of under-peformance.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

User avatar
baw703916
Posts: 6605
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by baw703916 » Sun Feb 26, 2012 12:56 pm

livesoft wrote:Because of years just like 2011.



Actually, the portfolio that I proposed in the first linked thread of the OP outperformed a Total Market porfolio of comparable risk by 10% in 2011(9% gain vs. 1% loss). (I updated the yearly performance through 2011 on the last page). It's true that HmL, SmB, and international in general all gave negative "returns" last year--and the equities portion performed quite poorly, as one would expect. But, there's 30% TIPS and 30% long Treasuries which more than made up the difference. REITs, which are also included as a diversifier, did pretty well, too.

Having shorter duration Treasuries, as Larry advises, would have provided less of a boost, but the investor would still have been quite happy with the result.

But I agree with the point that I think you are making, that this strategy (or any strategy which deviates significantly from a total market portfolio) will inevitably have periods of significant underperformance. If an investor can't tolerate negative tracking error without jumping ship, then this is not a good portfolio for that investor.

Brad
Last edited by baw703916 on Sun Feb 26, 2012 1:20 pm, edited 1 time in total.
Most of my posts assume no behavioral errors.

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by larryswedroe » Sun Feb 26, 2012 1:00 pm

First
Two ways to use higher tilts, or full tilt, what we call RT6--because it loads about .6 on the two factors. Either higher returns or same returns accomplished by lowering BETA exposure.
Second, if use the latter historically you have much more efficient portfolio, substantially cutting tails.
Third, as Rick says the price you pay is same as price you pay once you diversify at all, tracking error, just have more of it. Should be purely psychological issue, but it is a real one for sure for most investors. Rick and I know that from experience.
Fourth remember that though in years like 2011 when SV does poorly (not relatively poorly but negative returns) that is when bonds likely do very well and if you do it my way (lowering beta and increasing bonds) you lose more on your full tilt but make more on your larger bond portfolio!!!! Never think of these things in isolation. So last year the full tilt really hurt the equity side, but full tilt allows for very low equity exposure and bonds did very well, so portfolio did not do so poorly

Best wishes
Larry

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by GregLee » Sun Feb 26, 2012 1:01 pm

Why? Two words: loss aversion. It's a fallacy of thought that leads many to reckon their gains and losses not with respect to what their retirement portfolio will be worth when they need the money, in retirement, but with respect to what they have right now, or at intermediate times before retirement. Consequently, they seek investments which, they think, will never lead to a loss of what they have.
In economics and decision theory, loss aversion refers to people's tendency to strongly prefer avoiding losses to acquiring gains. Some studies suggest that losses are twice as powerful, psychologically, as gains.
http://en.wikipedia.org/wiki/Loss_aversion
Greg, retired 8/10.

User avatar
G-Money
Posts: 2852
Joined: Sun Dec 09, 2007 7:12 am

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by G-Money » Sun Feb 26, 2012 1:06 pm

The 3-factor model simply says that small and value are risk factors for which an investor could expect higher returns. This is just a variation of the equity risk premium, which says that stocks are riskier than bonds, and investors could expect higher returns by increasing their exposure to stocks. Neither of these models say how much to tilt toward stocks or small and value.

I wouldn't go full tilt even if I completely "accepted" the 3-factor model, just like almost no-one here recommends 100% stocks. Diversification helps. Plus, as others have pointed out, why stop at 100%? A true "accepter" might aim for 200% or 500% exposure to small and value. It all boils down to risk tolerance.
Don't assume I know what I'm talking about.

User avatar
baw703916
Posts: 6605
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by baw703916 » Sun Feb 26, 2012 1:17 pm

One reason why an investor might not employ this strategy is that tax efficiency and available funds can get in the way.

If you have your equities in deliberately risky asset classes, then you necessarily need to hold more bonds to keep your overall portfolio risk at a reasonable level. So you either need lots of tax-advantaged space, or have to compromise and use munis in your taxable account. Munis, even high quality ones, don't have the same flight-to-quality payoff as Treasuries in a full-blown financial panic, nor does a tax-exempt equivalent to TIPS exist.

For many people, there are limitations on available funds in tax-advantaged accounts which may restrict investors from using this approach. One can't do this in the TSP, nor using one of the many 401k lineups we've seen with an S&P 500 index fund, a bond index fund, and a bunch of overpriced active funds. DFA access is also very helpful, as they still have the only "real" passive ISV fund.

Brad
Most of my posts assume no behavioral errors.

User avatar
Whiggish Boffin
Posts: 242
Joined: Sun Dec 09, 2007 10:20 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Whiggish Boffin » Sun Feb 26, 2012 1:24 pm

This sounds like a kind of false syllogism that salesmen use. Posit a counterfactual premise: "If you accept the Fama-French 3-factor model" and draw an actionable conclusion: "Go full-tilt into small & value stocks."

The premise makes a subtle implication: the Fama-French model is metaphysically certain to be true. There is no possibility that it's wrong, that it stops working once somebody figures out how to profit from it, that it is a long-running lucky streak. But if it's a metaphysical certainty, then everyone with any sense knows it, too, and goes full-tilt small & value. Nobody invests in megacorps. Ooh looky! Megacorps just became undervalued and profitable, while SCV gets bid up to worthlessness!

But that isn't happening, so some people must doubt the Fama-French model. Some of them are as smart, and as informed, as you. So, maybe, there's less than metaphysical certainty.

I believe that the Fama-French model is probably correct. Larry Swedroe makes a pretty good case that it has a sound risk-vs-reward explanation. I do not have enough faith in my own cleverness, or even in Swedroe's, to risk my fortune on it. So, I tilt a little bit, and maybe it even matters.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by 555 » Sun Feb 26, 2012 1:49 pm

With typical fund lineups in 401k type plan it's easier (and cheaper) to just use broad index funds with no tilt. Most people just don't have the option of tilting.

User avatar
Aptenodytes
Posts: 3740
Joined: Tue Feb 08, 2011 8:39 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Aptenodytes » Sun Feb 26, 2012 2:45 pm

The question is posed oddly.

If one knew with certainty that the returns associated with the premiums associated with the Fama-French 3 factors were going to be repeated indefinitely into the future, then of course there's no reason not to pursue them fully. But of course, you could never know that. If you did know it, you could become very rich by fashioning custom instruments betting against the alternative.

Since you cannot predict the future with certainty, the premise to your question ("In a 3-factor world") becomes mythical, and therefore the question that follows is of no practical relevance. It is a bit like saying "If I knew that interest rates were going to double this year, what should I do?" You could answer that question in a logical manner, but it wouldn't mean anything because you can never know whether interest rates will double.

If you are trying to suss out why people don't "go full tilt," then in my case the answer is four-fold:

1) I have a modest attraction to REITs, which are not part of the FF equation, so they are about 6% of my portfolio. This is probably the least empirically grounded part of my AA, but at 6% I don't think it can do too much damage if it turns out to be wrong.

2) I find it impractical to replicate the three factors in my international holdings, which are about 43% of my equities. I loosely approximate and am satisfied.

3) My non-REIT U.S. equities are split as 20% growth, 25% small-growth, 55% small-value. I came up with those numbers about a year ago after a lot of reading and pondering. Today I couldn't tell you how close they come to a "full-tilt" FF, but they felt close enough given a number of issues, including some of the money locked up in accounts with limited options, and I'm sticking with it as my plan.

4) A little hedging seems prudent to me, because I don't know if the FF premiums will hold over the next few decades or not. I have most confidence in small-value, based on my reading of the research. But not enough to drop my growth holdings to zero.

User avatar
camontgo
Posts: 320
Joined: Tue Oct 27, 2009 1:46 pm
Location: Chicago, IL
Contact:

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by camontgo » Sun Feb 26, 2012 3:01 pm

Mortgasm wrote:Wouldn't you go with a full-tilt portfolio that is just almost entirely US/EM/I small cap value as the core equity components?


Full tilt means higher risk. If you believe there is an equity risk premium, it doesn't necessarily follow that you go 100% stocks...since to earn the premium you must take on risk. Same goes for tilting towards small and value.

You could argue that you go full tilt with small and value, and then reduce portfolio risk by adding bonds. In that case, you may indeed get a higher return for the same level of portfolio variance...i.e. you might improve your Sharpe ratio. This has at least been true historically. However, if you really believe in the 3-factor model then you shouldn't believe that risk = variance. Covariance with bad times also matters.

There is a common inconsistency among tilting advocates who say you can increase your Sharpe ratio through tilting (true!), and it follows that the portfolio is more optimal (not necessarily). In a multi-factor world the Sharpe ratio is not the right metric for evaluating your total risk.

If you believe that Sharpe ratio is all that should matter for the average investor, then you should believe that the average investor should tilt, and in that scenario the HML and SMB premiums will not persist into the future.

Now, you might also believe that every investor should tilt, but won't because they have an irrational fear of tracking error...in that case, you think the factors premiums are the result of investor psychology rather than rational risk.
"Essentially, all models are wrong, but some are useful." - George E. P Box

afan
Posts: 2799
Joined: Sun Jul 25, 2010 4:01 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by afan » Sun Feb 26, 2012 4:13 pm

The oft-cited Tower and Yang paper tried to ask whether a strong tilt, through DFA funds, beat straight indexing, and whether the DFA approach beat the closest match that Vanguard had. It suggested that the value weighted mix of DFA funds did well vs a similar mix of Vanguard index funds, which would argue that the tilting worked. Since these were real-world results for DFA, they took account of transaction costs within the funds. The outperformance persisted, at a lower level, when the authors tried to account for the extra investor expense of a DFA advisor to participate.

They did not address after tax returns. In fact, other than noting that the advisor expense would likely be all after tax money, they ignored the tax implications of the tilting approach. Instead, they suggested using individual stocks in a taxable account, holding the winners and tax loss harvesting the losers.

So, before taxes this suggested that the heavy style tilts of the DFA funds "worked" vs similar funds from Vanguard. The authors did not address the extent to which DFA results came from more aggressive securities lending, rather than superior performance of the style tilts.

If one were to tilt heavily, this might be a good predictor of the results, esp if one actually used DFA. Still, it is unclear whether doing this in a taxable account would be better than TSM.

http://public.econ.duke.edu/Papers//PDF ... y_2007.pdf
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by larryswedroe » Sun Feb 26, 2012 5:38 pm

GregLee
If loss aversion is the issue then one SHOULD greatly prefer a very highly tilted portfolio with low equity allocation, using the tilts to reduce beta exosure
It it TRACKING ERROR risk, purely psychological one, though real, that is the reason one should consider not doing so
Larry

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by larryswedroe » Sun Feb 26, 2012 5:40 pm

gmoney
Cannot tilt more than about 60% to value or much above 1 for size without going short, so your point is incorrect. Shorting has costs and also unlimited potential for losses
Best wishes
Larry

User avatar
alec
Posts: 2820
Joined: Fri Mar 02, 2007 2:15 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by alec » Sun Feb 26, 2012 5:41 pm

For a good explanation of reasons to tilt or not, read Cochrane's "portfolio advice for a multi factor world."
"It is difficult to get a man to understand something, when his salary depends upon his not understanding it!" - Upton Sinclair

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by GregLee » Sun Feb 26, 2012 5:56 pm

larryswedroe wrote:GregLee
If loss aversion is the issue then one SHOULD greatly prefer a very highly tilted portfolio with low equity allocation, using the tilts to reduce beta exosure

I don't understand that. Does "if loss aversion is the issue" mean "if loss aversion were not an issue"? And what does it have to do with preferring low equity allocation?
Greg, retired 8/10.

larryswedroe
Posts: 15597
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by larryswedroe » Sun Feb 26, 2012 6:33 pm

Greg
Again, see my post above on two ways to use the risk premiums. You I am guessing are only looking at the first, not the second,
The first seeks the higher returns, the second lowers beta risk while also tilting more, to get same expected return. If you do that you have gotten lower dispersion of returns, cutting the tail risk which is what loss aversion is about.
Larry

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by yobria » Sun Feb 26, 2012 6:41 pm

G-Money wrote:The 3-factor model simply says that small and value are risk factors for which an investor could expect higher returns. This is just a variation of the equity risk premium, which says that stocks are riskier than bonds, and investors could expect higher returns by increasing their exposure to stocks.


Stocks must be riskier than bonds - they're far more volatile, stockholders get paid last in the case of bankruptcy, bond interest is paid before dividends, etc.

But merely doing a regression with market risk and some other factor (eg height of CEO) doesn't make market risk a variation of CEO height risk.

Can you finish the following sentence with an answer that must be true, as I did for market risk?: "Small value stocks will be far riskier than small growth stocks going forward because _________ "

User avatar
camontgo
Posts: 320
Joined: Tue Oct 27, 2009 1:46 pm
Location: Chicago, IL
Contact:

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by camontgo » Sun Feb 26, 2012 6:46 pm

alec wrote:For a good explanation of reasons to tilt or not, read Cochrane's "portfolio advice for a multi factor world."


That is a great paper, which addresses the OP's question about why you might choose not to tilt even if you do believe the 3-factor model.

Do not forget, the average investor holds the market. If you're pretty much average, all this thought will lead you right back to holding the market index. To rationalize anything but the market portfolio, you have to be different from the average investor in some identifiable way. The average investor sees some risk in value stocks that counteracts their attractive average returns. Maybe you should too!


Italics are Cochrane's

Of course, most investors are not "average". Different time horizons, total wealth, tax status, risk aversion, etc may lead to valid reasons for tilting....but if the premiums are based on risk then it can't be optimal for everyone.

Paper is here: http://www.chicagofed.org/digital_asset ... 3Q99_4.pdf
Last edited by camontgo on Sun Feb 26, 2012 7:18 pm, edited 1 time in total.
"Essentially, all models are wrong, but some are useful." - George E. P Box

Mortgasm
Posts: 263
Joined: Mon Feb 06, 2012 7:54 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Mortgasm » Sun Feb 26, 2012 6:50 pm

Maybe I could have done better, but I deliberately worded the question that way because I did not want to start (again) the discussion of why or whether one should support a multifactor model (as predictive.) (Yobria is gonna turn this thread into that if he can :D )

These are actually the kind of replies I was hoping to see. I really wanted to know how much you weighed the multifactor research in constructing your own model and why you did it that way.

It seems like a lot of people hedge - I'm probably in that camp too. And yes that paper is a good one.

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by GregLee » Sun Feb 26, 2012 7:03 pm

larryswedroe wrote: If you do that you have gotten lower dispersion of returns, cutting the tail risk which is what loss aversion is about.
Larry

But, you see, I was attacking loss aversion as a fallacy, while you're suggesting ways to pander to it. It made it hard for me to follow the thread. Now, I think I'm clearer what you're saying.
Greg, retired 8/10.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Clearly_Irrational » Sun Feb 26, 2012 7:20 pm

1) Just the fact that there is more than one kind of risk doesn't make you necessarily prefer a higher risk level overall.

2) Tax efficiency, fees and limited fund selection issues can prevent you from choosing funds with high small and value loadings. For example there is no Vanguard Emerging Market Small Cap Value Fund.

For me, I use my my higher small and value loadings to let me take less beta while keeping my expected returns high. I prefer a more positive skew and a more negative kurtosis than is the default.

KyleAAA
Posts: 6230
Joined: Wed Jul 01, 2009 5:35 pm
Contact:

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by KyleAAA » Sun Feb 26, 2012 7:40 pm

No, because in the real world I'm not 100% sure the 3 factor model is the best model. I think perhaps it is, at least so far, but I can't be sure. So I hedge my bets and only tilt halfway. It's the same reason I split 50/50 among domestic/international and nominal bond/TIPS lines.

To mean, true diversification means never going ALL IN on ANYTHING.
Last edited by KyleAAA on Sun Feb 26, 2012 7:43 pm, edited 2 times in total.

edge
Posts: 3181
Joined: Mon Feb 19, 2007 7:44 pm
Location: Great Falls VA

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by edge » Sun Feb 26, 2012 7:41 pm

GregLee wrote:
larryswedroe wrote:GregLee
If loss aversion is the issue then one SHOULD greatly prefer a very highly tilted portfolio with low equity allocation, using the tilts to reduce beta exosure

I don't understand that. Does "if loss aversion is the issue" mean "if loss aversion were not an issue"? And what does it have to do with preferring low equity allocation?


Um, is that a serious question?

Obviously if you have a low equity exposure you are probably less subject to large losses.

User avatar
GregLee
Posts: 1748
Joined: Wed Oct 27, 2010 3:54 pm
Location: Waimanalo, HI

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by GregLee » Sun Feb 26, 2012 8:21 pm

edge wrote:Um, is that a serious question?

Obviously if you have a low equity exposure you are probably less subject to large losses.

Yes, it was a serious question. My previous comment contended that it was irrational to avoid loss of what you have now, when this is at the cost of failing to gain what you don't yet have. As you say, limiting equity exposure makes you less subject to losses, but it is also true that it makes you less subject to gains.
Greg, retired 8/10.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by yobria » Sun Feb 26, 2012 9:27 pm

Mortgasm wrote:Maybe I could have done better, but I deliberately worded the question that way because I did not want to start (again) the discussion of why or whether one should support a multifactor model (as predictive.) (Yobria is gonna turn this thread into that if he can :D )


In fact, I and others have answered you exact question. As poster G-Money pointed out:

why stop at 100%? A true "accepter" might aim for 200% or 500% exposure to small and value.


Leverage and use options to magnify effects if you're absolutely certain these market patterns must continue. If you're in over your head at this point, drift back to shallow waters and master the fundamentals: risk, return, efficient markets, etc. The Boglehead's Guide is a great place to start.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by 555 » Sun Feb 26, 2012 10:36 pm

I wonder how this idea of leverage works The effects of compounding would be pretty interesting if you could lose 300% every year. :wink:

User avatar
baw703916
Posts: 6605
Joined: Sun Apr 01, 2007 1:10 pm
Location: Seattle

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by baw703916 » Sun Feb 26, 2012 10:42 pm

555 wrote:I wonder how this idea of leverage works The effects of compounding would be pretty interesting if you could lose 300% every year. :wink:


That kind of performance tends to be unsustainable, though. :wink:
Most of my posts assume no behavioral errors.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by 555 » Sun Feb 26, 2012 10:53 pm

baw703916 wrote:
555 wrote:I wonder how this idea of leverage works The effects of compounding would be pretty interesting if you could lose 300% every year. :wink:

That kind of performance tends to be unsustainable, though. :wink:

I'd be happy to do it two years in a row. :wink:

User avatar
grap0013
Posts: 1857
Joined: Thu Mar 18, 2010 1:24 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by grap0013 » Mon Feb 27, 2012 8:21 am

livesoft wrote:Because of years just like 2011.


What about 2012? Or 1927-2011? :wink:
There are no guarantees, only probabilities.

User avatar
grap0013
Posts: 1857
Joined: Thu Mar 18, 2010 1:24 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by grap0013 » Mon Feb 27, 2012 8:29 am

Mortgasm wrote:
'Why don't more 3-factor people go full tilt? " right?'


I don't know. Even when SCV has periods of underperformance relative to TSM it isn't by much. 1984-1999 had some pretty large tracking error, but it helped keep you out of the ensuing bubble. It all comes down to conviction and discipline.
There are no guarantees, only probabilities.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Rodc » Mon Feb 27, 2012 9:49 am

Anyone who believes the 3-factor model full tilt has not learned any lessons of history.

There is always a non-trivial chance the model is wrong. It is also possible that the model is good but the funds won't capture the premium. So no thanks on full bore small value, even if you think the odds are good that the model is pretty good.

An example, most of us believe that over the long haul the expected return for stocks (as in "expected" as defined by statistics) is greater than the expected return for bonds. But very few of us are 100% stocks. Why? Same answer for small value tilt.
We live a world with knowledge of the future markets has less than one significant figure. And people will still and always demand answers to three significant digits.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Rodc » Mon Feb 27, 2012 9:54 am

alec wrote:For a good explanation of reasons to tilt or not, read Cochrane's "portfolio advice for a multi factor world."


But skip the section about using market timing to double your return with no increase in risk. :roll:
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.

vesalius
Posts: 774
Joined: Tue Jul 13, 2010 7:00 pm
Location: Texas

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by vesalius » Mon Feb 27, 2012 9:56 am

yobria wrote:
Mortgasm wrote:Maybe I could have done better, but I deliberately worded the question that way because I did not want to start (again) the discussion of why or whether one should support a multifactor model (as predictive.) (Yobria is gonna turn this thread into that if he can :D )


In fact, I and others have answered you exact question. As poster G-Money pointed out:

why stop at 100%? A true "accepter" might aim for 200% or 500% exposure to small and value.


Leverage and use options to magnify effects if you're absolutely certain these market patterns must continue. If you're in over your head at this point, drift back to shallow waters and master the fundamentals: risk, return, efficient markets, etc. The Boglehead's Guide is a great place to start.
Straw man argument. Neither you nor any other boglehead that is sure of the equity premium would ever recommend leverage or options in that case, so not using it in regards to FF is a non sequitur.

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

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by yobria » Mon Feb 27, 2012 10:49 am

vesalius wrote:
yobria wrote:
Mortgasm wrote:Maybe I could have done better, but I deliberately worded the question that way because I did not want to start (again) the discussion of why or whether one should support a multifactor model (as predictive.) (Yobria is gonna turn this thread into that if he can :D )


In fact, I and others have answered you exact question. As poster G-Money pointed out:

why stop at 100%? A true "accepter" might aim for 200% or 500% exposure to small and value.


Leverage and use options to magnify effects if you're absolutely certain these market patterns must continue. If you're in over your head at this point, drift back to shallow waters and master the fundamentals: risk, return, efficient markets, etc. The Boglehead's Guide is a great place to start.
Straw man argument. Neither you nor any other boglehead that is sure of the equity premium would ever recommend leverage or options in that case, so not using it in regards to FF is a non sequitur.


a) Many people on this board in fact do leverage their equity investment - anyone with stocks and a mortgage does for example
b) The OP isn't asking what a rational boglehead would do, but about a naive investor that blindly projects the FF3F patterns going forward. Unfortunately, as people would rather cheat on a test than master the material, most would rather latch on to free lunch ideas than accept efficient markets. That is why they buy active funds, or "value stocks", or day trade, etc.

vesalius
Posts: 774
Joined: Tue Jul 13, 2010 7:00 pm
Location: Texas

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by vesalius » Mon Feb 27, 2012 11:12 am

yobria wrote:
vesalius wrote:
yobria wrote:
Mortgasm wrote:Maybe I could have done better, but I deliberately worded the question that way because I did not want to start (again) the discussion of why or whether one should support a multifactor model (as predictive.) (Yobria is gonna turn this thread into that if he can :D )


In fact, I and others have answered you exact question. As poster G-Money pointed out:

why stop at 100%? A true "accepter" might aim for 200% or 500% exposure to small and value.


Leverage and use options to magnify effects if you're absolutely certain these market patterns must continue. If you're in over your head at this point, drift back to shallow waters and master the fundamentals: risk, return, efficient markets, etc. The Boglehead's Guide is a great place to start.
Straw man argument. Neither you nor any other boglehead that is sure of the equity premium would ever recommend leverage or options in that case, so not using it in regards to FF is a non sequitur.


a) Many people on this board in fact do leverage their equity investment - anyone with stocks and a mortgage does for example
b) The OP isn't asking what a rational boglehead would do, but about a naive investor that blindly projects the FF3F patterns going forward. Unfortunately, as people would rather cheat on a test than master the material, most would rather latch on to free lunch ideas than accept efficient markets. That is why they buy active funds, or "value stocks", or day trade, etc.

a) not the same thing that was suggested, options and going for 200% to 500% SV exposure, and you obviously realize that.
b) have a nice day, going to respect the OP's wishes and not contribute to further tilting at windmills.

Mortgasm
Posts: 263
Joined: Mon Feb 06, 2012 7:54 pm

Re: In a 3-factor world, why wouldn't you go full tilt?

Post by Mortgasm » Tue Feb 28, 2012 4:23 am

Yep. The leverage argument against multifactor is a straw-man. It's slightly interesting, but it applies equally to anyone who believes in stocks for the long run, which is all of us.


But I do appreciate the constructive responses.

Post Reply