Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Locked
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

https://investresolve.com/podcasts/larr ... -long-run/

Adam Butler of Resolve Asset Management interviews Larry in this podcast. It starts with Larry reviewing his career. The investing discussion covers a lot of ground: Fama-French, factor zoo, low beta, momentum, value, significance of multiple value screens, how to implement a multi factor approach, alternatives.

The most significant part of the interview is the discussion of “what do I do if I don’t have a long time to wait for a factor to provide it’s premium?” Larry points out that most investors view this issue backwardly. When any factor can underperform for a long period, the correct strategy is to diversify across factors, not avoid them. Larry drives this point home with a review of market beta performance. There have been 3 periods of 13 years or longer in the 93 year history of stock market data when the market has underperformed TBills: 1966-1982, 1929-1943, 2000-2012. That totals 45 of the 93 years for which we have stock market data. During each of those periods size and value performed well with premiums of 4-5%.

Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.

Dave
UpperNwGuy
Posts: 9479
Joined: Sun Oct 08, 2017 7:16 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by UpperNwGuy »

Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
What is that factor?
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

UpperNwGuy wrote: Mon May 27, 2019 3:05 pm
Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
What is that factor?
The equity market factor, market beta. Larry has explained it about as follows:
Stocks have volatility of about 20%: so 60% stocks x 20% SD = .12 “risk points”
Bonds have volatility of about 5%: so 40% bonds x 5% SD = 0.02 “risk points”

So in a 60/40 portfolio, stocks contribute 0.12/(0.12+0.02) = 86%

Dave
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Miriam2 »

Random Walker wrote: Mon May 27, 2019 3:38 pm
UpperNwGuy wrote: Mon May 27, 2019 3:05 pm
Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
What is that factor?
The equity market factor, market beta. Larry has explained it about as follows:
Stocks have volatility of about 20%: so 60% stocks x 20% SD = .12 “risk points”
Bonds have volatility of about 5%: so 40% bonds x 5% SD = 0.02 “risk points”

So in a 60/40 portfolio, stocks contribute 0.12/(0.12+0.02) = 86%

Dave
Why does Larry complicate investing so much?
User avatar
9-5 Suited
Posts: 1307
Joined: Thu Jun 23, 2016 12:14 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by 9-5 Suited »

Miriam2 wrote: Mon May 27, 2019 4:27 pm Why does Larry complicate investing so much?
He means well and genuinely believes his factor models are helping people manage risk, IMO. It's certainly an interesting different way of conceptualizing risk and has slightly influenced some decisions I've made.

As always, the debate arises between the camp who believe the research is useful going forward vs. useful only in the backtested environments it has been discovered. If I invest heavily behind companies that contain qualities described by the factors, do I have a reasonable expectation of a lower risk portfolio? I'm marginally sold, enough to tilt a bit, but have enough doubts to prevent an all-in strategy.

@ Random Walker ... I am curious, are there any areas of investing where you have significant difference of opinion with Larry Swedroe? I am a little weary reading your material because you come across as an acolyte following a guru, essentially doing virtually everything Larry says (including investing in new Larry recommendations like reinsurance markets). Don't mean that as a sleight, but I think hearing a few meaningful areas of difference would give me more confidence in your writings, which generally I consider very good and interesting.
User avatar
Miriam2
Posts: 4387
Joined: Fri Nov 14, 2014 10:51 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Miriam2 »

9-5 Suited wrote:
Miriam2 wrote: Why does Larry complicate investing so much?
He means well and genuinely believes his factor models are helping people manage risk, IMO. It's certainly an interesting different way of conceptualizing risk and has slightly influenced some decisions I've made.

As always, the debate arises between the camp who believe the research is useful going forward vs. useful only in the backtested environments it has been discovered. If I invest heavily behind companies that contain qualities described by the factors, do I have a reasonable expectation of a lower risk portfolio? I'm marginally sold, enough to tilt a bit, but have enough doubts to prevent an all-in strategy.

@ Random Walker ... I am curious, are there any areas of investing where you have significant difference of opinion with Larry Swedroe? I am a little weary reading your material because you come across as an acolyte following a guru, essentially doing virtually everything Larry says (including investing in new Larry recommendations like reinsurance markets). Don't mean that as a sleight, but I think hearing a few meaningful areas of difference would give me more confidence in your writings, which generally I consider very good and interesting.
Yes, I can see this.

However, the Larry threads over-populate the forum and I am concerned that newcomers to our forum will think that Larry-investing is part of the Boglehead approach and that they - as investors - should complicate their portfolios in order to manage risk and make more money.
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

Miriam2 wrote: Mon May 27, 2019 4:27 pm
Random Walker wrote: Mon May 27, 2019 3:38 pm
UpperNwGuy wrote: Mon May 27, 2019 3:05 pm
Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
What is that factor?
The equity market factor, market beta. Larry has explained it about as follows:
Stocks have volatility of about 20%: so 60% stocks x 20% SD = .12 “risk points”
Bonds have volatility of about 5%: so 40% bonds x 5% SD = 0.02 “risk points”

So in a 60/40 portfolio, stocks contribute 0.12/(0.12+0.02) = 86%

Dave
Why does Larry complicate investing so much?
I take the opposite view. I think he simplifies academic modern academic finance for amateurs like myself.

Dave
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by nedsaid »

9-5 Suited wrote: Mon May 27, 2019 4:35 pm
@ Random Walker ... I am curious, are there any areas of investing where you have significant difference of opinion with Larry Swedroe? I am a little weary reading your material because you come across as an acolyte following a guru, essentially doing virtually everything Larry says (including investing in new Larry recommendations like reinsurance markets). Don't mean that as a sleight, but I think hearing a few meaningful areas of difference would give me more confidence in your writings, which generally I consider very good and interesting.
I can't speak for Dave, obviously, but he is a Buckingham client. Dave has posted his portfolio and a few of us took some whacks at it. Pretty much, if you go through an advisory service, you get with the program. No sense in hiring an advisor just to engage in hand to hand combat over every decision. As I recall, he had some input but his advisor at Buckingham put it all together. So seeing that Larry Swedroe is Director of Research at Buckingham, it wouldn't be surprising that Dave's posts would follow along with what Larry would have to say.

Random Walker wrote:
Mon Apr 22, 2019 4:49 pm
Not sure about knowledgeable investor. Seem to remember Buffett saying something to the effect that when dumb money recognizes it's dumb, it suddenly becomes very smart. This is my AA. Im 56 years old, uncertain length of time remaining in work force. This is a 40/36/24 portfolio. I was 80/20 until about 3 years ago, and started to take risk off the table at that time. With retirement somewhere on the horizon, I've moved to this portfolio to manage sequence of returns risk. The move involved increase tilt to size and value, move international to 50% equity, increase bonds, add alternatives.

fund symbol category target

DFA TA US Core Equity 2 DFTCX US Market 4.0%
DFA Tax Managed Marketwitde Value II DFMVX US Large Value 2.5%
DFA Tax Managed US Targeted Value DTMVX US Small Value 13.5%

DFA TA World ex US Core Equity 1 DFTWX International Market 3.0%
DFA Tax Managed International Value DTMIX International Value 4.5%
DFA International Small Cap Value DISVX International Small Value 7.5%
DFA Emerging Markets Core Equity I DFCEX Emerging Markets 5.0%

AQR multi style QRPRX Multi Style 10.5%
Stone Ridge Reinsurance SSRIX Reinsurance 4.5%
Stone Ridge Alternative Lending LENDX Alternative Lending 4.5%
Stone Ridge Risk Variance AVRPX Variance Risk Premia 4.5%

Municipal Bond Ladder Fixed Income 36.0%

100.0%

sorry, it didn't format the way I intended. but the info is there

Dave
My reaction to Dave's portfolio:
Nedsaid wrote: Mon Apr 22, 2019 7:11 pm
I did look at Random Walker's portfolio which is 36% bonds, 24% alternatives, and 40% stocks with half of the stocks being Small Value. I was surprised at such a large position in AQR Multi Style and that the Alts were more than 20% of the portfolio. I am sure there are good reasons for all of this but it did raise an eyebrow. I was expecting the Alts to be capped at 20% and split evenly between the four Alt Funds. But I can see the rationale. I guess that I don't like large bets, RW is making substantial bets on AQR Multi Style and on Small Value. That is just me though. Buckingham must have huge confidence in both.
A fool and his money are good for business.
Quaestner
Posts: 327
Joined: Tue Jul 18, 2017 6:39 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Quaestner »

Random Walker - I always appreciate your summaries and sharing of Larry's writings. Even if I don't take action, I like to learn. Please continue!
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

9-5 Suited wrote: Mon May 27, 2019 4:35 pm @ Random Walker ... I am curious, are there any areas of investing where you have significant difference of opinion with Larry Swedroe? I am a little weary reading your material because you come across as an acolyte following a guru, essentially doing virtually everything Larry says (including investing in new Larry recommendations like reinsurance markets). Don't mean that as a sleight, but I think hearing a few meaningful areas of difference would give me more confidence in your writings, which generally I consider very good and interesting.

If there are areas of difference of opinion, it would be like comparing the opinions of a seasoned cardiologist and a medical student about someone’s coronary artery disease. He’s a formally trained pro with academic training and a ton of practical experience. I read books for laymen, have struggled through a couple of academic texts, participate on Bogleheads, and think a lot (too much) about my own portfolio.

I can say that I have a very strong (nearly idealistic) concept of market efficiency. When I first read Malkiel and Bogle, it resonated strongly with me. I view markets like the intersecting lines of Econ 1 or chemical equilibrium. So after learning the hard way about individual stocks 1996-2001, I became a hardcore do it myself TSM Boglehead. I was, like most here, exceptionally focused on costs. I still think about costs a lot. So when I look as critically as I am capable of at Larry’s recommendations, it is from a cost point of view. This is especially true since we know costs are certain and benefits only potential. As I’ve written before, I think Larry’s recommendations are all internally consistent and fit within the framework of striving for a more efficient portfolio. As one goes further down the path of improved portfolio efficiency, the marginal costs increase and the marginal benefits decrease. The marginal benefits are still benefits nonetheless. Each investor needs to decide where to draw the line on those marginal costs and marginal benefits. Since I took the plunge and became a client of his firm, I decided to take full advantage of the potential improvements to portfolio efficiency provided by his firm. I have equities highly tilted to size and value both in US and internationally, I have a bigger dose of bonds than I’d have on my own in the form of an individual bond ladder, and I’ve taken on alternatives. So perhaps I am a bit of an acolyte following a guru, but it is with eyes pretty wide open, a fair amount of common sense, a strong layman’s knowledge of the material, an understanding that I probably can’t see my own behavioral errors, the scars of 200-2003 and 2007-2008 bear markets, and a big concern about sequence of returns risk.

So the closest I can come to a “difference of opinion”, would be my concern over costs. But in the end I became a client and have taken on the associated expense ratios and management fees. I definitely know (and this forum frequently reminds me) that TSM / LG has dominated over the last decade. :-) Am I avoiding the mistake of confusing strategy with outcome or am I blindly paying fees to underperform the TSM approach? Am I paying to blindly follow a guru, or am I looking at cost per unit value added as opposed to cost alone? So I’m sorry that I can’t really give you any areas where I disagree with Larry. If I could, there’s a high likelihood I’d be wrong. Also if there were significant areas of disagreement, I perhaps shouldn’t be a client.

Many have questioned Larry’s apparent reversals on CCFs and REITs. Even those changes in recommendations seem completely internally consistent to me when looked at from the point of view of modern portfolio theory and cost per unit value added to a portfolio. CCFs are like insurance. We shouldn’t be upset that we didn’t have to make a claim. REITs behavior can be explained by underlying factors that can be more efficiently accessed. Things change, and more efficient portfolio diversifiers have become available that can likely supplant those in precious tax advantaged space.

Dave
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by nedsaid »

Random Walker wrote: Mon May 27, 2019 6:03 pm
Many have questioned Larry’s apparent reversals on CCFs and REITs. Even those changes in recommendations seem completely internally consistent to me when looked at from the point of view of modern portfolio theory and cost per unit value added to a portfolio. CCFs are like insurance. We shouldn’t be upset that we didn’t have to make a claim. REITs behavior can be explained by underlying factors that can be more efficiently accessed. Things change, and more efficient portfolio diversifiers have become available that can likely supplant those in precious tax advantaged space.
Going from memory here, Larry gave up on Collateralized Commodity Futures because institutions flooded so much money into this market that this strategy no longer made sense. This is called financialization. It was a matter of valuations, the CCFs were just too expensive. My understanding also is that all that money took the commodity markets from backwardization (that is where futures prices were lower than spot prices, thus you could get a "roll return") to contango (where futures prices were higher than spot prices). Going from backwardization to contango made it much more difficult to make money doing a CCF strategy. Perhaps someone more knowledgeable about commodities and the futures markets could comment here but that is my understanding.

Larry says it is okay to hold a market weighting of REITs but he no longer holds them as a separate asset class or overweights them in a portfolio. Two reasons: first the diversification benefit of REITs were explained by factors and could be achieved in different ways, second REITs got to be very expensive. At one point, Larry estimated that real returns from REITs going forward would be just 0.20% a year. Lots of folks chased REITs really hard and drove prices up sky high and yields very low. We were in an environment where investors were yield chasing in an environment of very low interest rates.

Again, going from memory. If I mangled what Larry might have said, I suppose he can straighten me out with a post or personal message. Random Walker could probably expand upon what I have said.
A fool and his money are good for business.
stlutz
Posts: 5585
Joined: Fri Jan 02, 2009 12:08 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by stlutz »

Perhaps it would be worth discussing some of the papers that Swedroe summarizes in greater detail? Just a suggestion for how to move onto something that is a little bit new?

FWIW.
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Rick Ferri »

Your allocation between stocks and bonds is the cake, everything else is the icing on the cake, and avoid an infatuation with brightly light candles because they are quickly extinguished and thrown away.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
azanon
Posts: 3142
Joined: Mon Nov 07, 2011 9:34 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by azanon »

Rick Ferri wrote: Tue May 28, 2019 7:28 am Your allocation between stocks and bonds is the cake, everything else is the icing on the cake, and avoid an infatuation with brightly light candles because they are quickly extinguished and thrown away.

Rick Ferri
Only if the factor claims are untrue, or much less true than they claim. RW/Larry know this better than me, but don't these factors in theory (combined) add some 150-300 basis points of expected return long-term? You add that much extra return vs. vanilla beta compounded in an IRA for 30 years, and you're talking 100s of thousands of dollars potential difference. Those candles would put off blinding light!
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Rick Ferri »

azanon wrote: Tue May 28, 2019 7:36 am
Rick Ferri wrote: Tue May 28, 2019 7:28 am Your allocation between stocks and bonds is the cake, everything else is the icing on the cake, and avoid an infatuation with brightly light candles because they are quickly extinguished and thrown away.

Rick Ferri
Only if the factor claims are untrue, or much less true than they claim. RW/Larry know this better than me, but don't these factors in theory (combined) add some 150-300 basis points of expected return long-term? You add that much extra return vs. vanilla beta compounded in an IRA for 30 years, and you're talking 100s of thousands of dollars potential difference. Those candles would put off blinding light!
The 150-300 bps factors estimates are pie in the sky. They're based on theoretical long-short portfolios that have no trading costs or taxes, and they were measured over a period when factor investing wasn't understood and there were not millions of "factor renters" in the marketplace as opposed to long-term factor owners. Will there be a premium in the future? Probably, but the weak money has to leave first and then don't count on earning 150-300 bps.

How much factor fund exposures is enough? I don't recommend any more than 25% of your equity exposure to be funds that allocate to additional factors.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
azanon
Posts: 3142
Joined: Mon Nov 07, 2011 9:34 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by azanon »

The weak money is pretty much already gone. Granted, not the most scientific way of measuring it, but I compared the amount of money invested in each of Vanguard's 9 specific index funds arranged by small/mid/large, and growth/blend/value, and there was no appreciable bias to value. Heck, just compare VG Large-cap growth index total net assets (90.1 Billion) to their Small-cap value index (31.0B).

Again, I know that's not scientific at all, but it certainly confirms that the herd is definitely not hiding out in SC value.
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Rick Ferri »

azanon wrote: Tue May 28, 2019 8:05 am The weak money is pretty much already gone.
People who manage factor funds would disagree. Wes Gray was my last guest on Bogleheads on Investing. Wes earned a Ph.D. in finance under Gene Fama at the University of Chicago's Booth School of Business. Toward the end of the interview, we discuss factor cost, the crowding out of factor premiums and potential returns.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
DaufuskieNate
Posts: 605
Joined: Wed May 28, 2014 11:53 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by DaufuskieNate »

Small cap value relative valuations show no evidence of being over-valued. The recent performance of SCV has lagged. Where's the evidence of "factor renters" making this a crowded trade? It's easy to point to a proliferation in the number of factor funds and ETF's. But the number of funds and ETF's of all types are growing significantly. Meanwhile, passive indexes are growing AUM at warp speed and active investors still like growth stories. It's easy to say that factors are a crowded trade or that short term renters are piling into factor funds, but where's the data to support it? Why aren't we seeing it in valuations?
azanon
Posts: 3142
Joined: Mon Nov 07, 2011 9:34 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by azanon »

Rick Ferri wrote: Tue May 28, 2019 8:17 am
azanon wrote: Tue May 28, 2019 8:05 am The weak money is pretty much already gone.
People who manage factor funds would disagree. Wes Gray was my last guest on Bogleheads on Investing. Wes earned a Ph.D. in finance under Gene Fama at the University of Chicago's Booth School of Business. Toward the end of the interview, we discuss factor cost, the crowding out of factor premiums and potential returns.

Rick Ferri
Factor funds are probably slightly more popular than Risk Parity right now, given the beatdown they've taken vs. vanilla beta and (worse) growth tilts. In a year or two, they'll be changing their fund names too (like risk parity), if things don't work out.

I have no problem with disagreement. If i have to choose between what "Wes" says, and actual holdings I can look up at a 5 trillion dollar institution, i bet you can guess which I'm going to pick. If i want popular, Large cap or mid cap US growth is what I want.

You mentioned costs twice. Vanguard's factor index funds are an extra 5-8 basis points or so. I guess I'm not understanding the issue. VFVA (the one i hold) is 0.13% ER. Ok, i get there are some expensive ones out there too, but I guess I'd suggest, don't buy those? :)
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Rick Ferri »

Cheaper fees are not better fees when investing to beat the market.

If you invest in factor funds, you're buying tracking error, so you want the greatest tracking error for the cost. This is determined using a cost per unit of risk equation. Same thing with active management. You want the highest active share for the money.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by larryswedroe »

Few quick things here to help
have written pieces on papers showing there is NO CROWDING in at least value though definitely in Low beta and low vol, and that is easy to show by looking at spreads in valuations. Value now cheaper relative to growth than was when FF published their paper, so how can their be overcrowding? . On other hand, while there has been a low beta premium, it's ONLY when in value regime, negative when in growth, and it's popularity has pushed it heavily into growth regime, hence negative premium should be expected. Note I have written on this as well. So crowding or not is dependent on which factor we are talking about. Valuations matter.
Hope that is helpful
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Rick Ferri »

...and the past always predicts the future.

No. Looking backward has never helped investors predict if premiums will occur or when they will occur. We can only pay the rent and hope they show up in enough quantity to a least cover that rent.

I'm not saying this is what you should or shouldn't do, I'm saying there are much bigger things to do as an investor than chase factors, like save more, keep fees low (especially any adviser fees), and be diligent on tax reduction strategies.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
DaufuskieNate
Posts: 605
Joined: Wed May 28, 2014 11:53 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by DaufuskieNate »

Rick Ferri wrote: Tue May 28, 2019 1:26 pm ...and the past always predicts the future.

No. Looking backward has never helped investors predict if premiums will occur or when they will occur. We can only pay the rent and hope they show up in enough quantity to a least cover that rent.

I'm not saying this is what you should or shouldn't do, I'm saying there are much bigger things to do as an investor than chase factors, like save more, keep fees low (especially any adviser fees), and be diligent on tax reduction strategies.

Rick Ferri
Certainly agree with your comments on fees, saving and paying attention to taxes. Not sure how looking at CURRENT valuations is looking backward and using the past to predict the future. There have been a lot of statements made here about overcrowding and short-term money going into factor strategies. In theory, this can of course be an issue. But other than low beta, which Larry mentioned, where's the evidence that overcrowding is driving valuations higher in factors like value? It makes good sense to take a haircut to historical factor premiums-no question. But I'd sure like to see some actual evidence to back up some of the comments being made here.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by larryswedroe »

Nate and Rick
There is no evidence on value being crowded as I noted. AQR has data on valuation spreads and for US and developed spreads now in 85th cheapest percentile, international is 91st cheapest and EM is cheapest ever! So clearly value not overcrowded. That's just another media myth. Again, other factors like low beta show clear crowding
Larry
BigJohn
Posts: 2627
Joined: Wed Apr 02, 2014 11:27 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by BigJohn »

Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
Dave, I’m still not sure that I see factors as diversifying assets with higher risk adjusted returns. But I’m curious as to how much diversification the factor gurus think you can really achieve since the lions share of the return is always market beta anyway. So a simple example to illustrate my thought process. Let’s say that market beta will give us 5% and a good low cost SCV will give us 6% (5% market beta plus 1% factor premium). If I tilt my stock portfolio 20% to SCV, my return is now 5.2%. It looks like 95% of the return is still coming from market beta so I’m not seeing how this provides significant risk diversification. What am I missing?
"The greatest enemy of a good plan is the dream of a perfect plan" - Carl Von Clausewitz
garlandwhizzer
Posts: 3565
Joined: Fri Aug 06, 2010 3:42 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by garlandwhizzer »

I'll skip over the question about whether value is currently overcrowded or not. I will also skip over the question of whether a single factor, multi-factor, or cap weight equity portfolio will outperform in a risk adjusted manner long term. I don't know with certainty the answer to these questions, nor do I believe that anyone does with certainty over any time frame. How well do academic fact studies of cost-free long/short portfolios without any trading frictions translate into the real world which typically is long only, after real costs and with trading frictions especially in the SC space?

What I do know is that clearly the results over the last 15 years have not been great for value investors in the large cap space. JKD the iShares Large Cap Index Fund (LCB) and JKF the iShares Large Cap Value Index Fund (LCV) have been in existence for almost exactly 15 years. During that decade and a half time the Value Index Fund has produced a 75% return while the cap weighted fund in the same LC space has produced a return of 167%. That's called tracking error and those true believers who held on to the loser value fund have thus far paid an opportunity cost of 92% for that belief over 15 years. The question now becomes how long will it take for the value fund to make up that enormous difference? 5 yrs., 10yrs, another 15, or will it ever do that before the investor holding it dies?

Those who are factor adherents believe that LCV will eventually outperform LCB) and that this long period of underperformance is just a setup for the coming outperformance. Those who are skeptical about factors are happy they didn't tilt and held on to a market portfolio. Also since 1992, the year the value factor was discovered by Fama, 27 years ago, VISVX the Vanguard Value Index Fund has underperformed both VFINX, the S&P 500 LCB fund, and VTSMX, their TSM fund. In other words in the large cap space there has been no evidence of a harvestable value factor since it was first described at least as Vanguard has defined value. In fact it lost money relative to beta. How can a real factor completely disappear and actually underperform for 27 years? Until 1992 according to Fama's original work the value factor had produced robust outperformance in both LC and SC spaces. Then all of a sudden it dies in the LC space. Over the last 15 years the same may have happened in the SC space. DFSVX has underperformed NAESX Vanguard's SCB fund over that span.

Those who are factor skeptical find their reassurance in the spotty record of real factor funds especially in recent years. Those who are true believers in factors get their reassurance from academic models created and brilliant and knowledgeable academics, managers, and advisors. Personally I hold both, but the lion's share (75%) is in good old TSM whose rock bottom cost, wide diversification, lack of capacity restraints, and minimal trading are all big pluses for me.

Garland Whizzer
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

BigJohn wrote: Tue May 28, 2019 6:30 pm
Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
Dave, I’m still not sure that I see factors as diversifying assets with higher risk adjusted returns. But I’m curious as to how much diversification the factor gurus think you can really achieve since the lions share of the return is always market beta anyway. So a simple example to illustrate my thought process. Let’s say that market beta will give us 5% and a good low cost SCV will give us 6% (5% market beta plus 1% factor premium). If I tilt my stock portfolio 20% to SCV, my return is now 5.2%. It looks like 95% of the return is still coming from market beta so I’m not seeing how this provides significant risk diversification. What am I missing?
Hi BigJohn,

Sounds like your understanding is about the same level as mine. Here’s what I can say though. First, I wouldn’t take credit for increased risk adjusted return at the individual asset level. Especially with size and value, I’d say increased expected return for increased risk. Any behavioral component is a bonus from my point of view. Second, to make an apples to apples comparison, I think it’s only fair to try to maintain the portfolio expected return constant. So when adding SV, I think one should decrease overall equity exposure. That decreases exposure to market beta and increases exposure to term. So you’re more evenly spreading your risks between market beta, size, value, term. Most long only SV funds I think still have market beta of about 1, and perhaps size and value loads of maybe 0.4. So if you really look to spread the low correlated risks around, more in the direction of risk parity, you can use only SV funds, decrease equity exposure a lot yet still have a lot of market beta exposure, increase safe bond exposure a lot. I don’t think you’re missing anything, just showing that a little tilt potentially does a little. Maybe need to go big? Increase SV exposure lots and decrease overall equity exposure lots.

Let’s give a SV fund more credit than you did, say perhaps a total of 3% over the 5% market that you assumed. Perhaps it can get that from deeper exposures to size and value, add on screens for momentum and profitability. So total expected return 8%. Say we expect 2% from bonds.
The expected return of a portfolio is the weighted mean of portfolio component expected returns. The expected SD of a portfolio is less than the weighted mean of the component volatilities due to less than perfect correlations.

60/40 TSM portfolio
(0.6X0.05)+(0.4X0.02)=expected return 3.8%
(0.6X0.18)+(0.4X0.05)= expected SD <12.8%
Sharpe Ratio in neighborhood of 3.8%/12.8%>=0.30

40/60 SV portfolio
(0.4X0.08)+(0.6X.02)=expected return 4.4%
(0.4X0.22)+(0.6X0.05)=expected SD<11.8%
Sharpe Ratio in neighborhood of 4.4%/11.8%>=0.37

Increasing Sharpe Ratio from 0.30 to 0.37 would be significant.

So I decided to put the above thinking to a real world back test.
I created two portfolios:
#1 60/40 TSM
30% Vanguard Total Stock Market VGTSX
30% Vanguard Total International VITSX
40% DFA Global Bond Int term

#2 40/60 SV
20% DFA US Small Value DFSVX
20% DFA International Small Value DISVX
60% DFA Global Bond Int term

Took longest time period Portfolio Visualizer would give depending on when one of the funds was created 8/97-4/19

https://www.portfoliovisualizer.com/bac ... 0&total3=0

#1 60/40 VGTSX/VITSX/DFGBX
CAGR 6.00%, SD 9.11%, Max Draw Down -34%, Sharpe Ratio 0.48

#2 40/60 DFSVX/DISVX/DFGBX
CAGR 6.80%, SD 6.99%, Max Draw Down -24%, Sharpe Ratio 0.71

I know everyone has all sorts of complaints about back tested results. I do too! But I do think historic volatilities are reasonable to use. My guesstimates at the start were based on reasonable forward looking expected returns. Only after that did I go and look back at two similar real life portfolios.

Dave
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by packer16 »

larryswedroe wrote: Tue May 28, 2019 10:50 am Few quick things here to help
have written pieces on papers showing there is NO CROWDING in at least value though definitely in Low beta and low vol, and that is easy to show by looking at spreads in valuations. Value now cheaper relative to growth than was when FF published their paper, so how can their be overcrowding? . On other hand, while there has been a low beta premium, it's ONLY when in value regime, negative when in growth, and it's popularity has pushed it heavily into growth regime, hence negative premium should be expected. Note I have written on this as well. So crowding or not is dependent on which factor we are talking about. Valuations matter.
Hope that is helpful
I think the biggest issue with your assertion is that the spread between value and growth is dependent upon the growth rates of both value and growth. How do you know value is cheaper if you do not know the relative growth rates? If the market is more efficient, then the cheaper stocks should have lower growth over time & the growth stocks higher growth. There is a quantifiable way to incorporate growth into multiples via the Graham formula (8.5 + 2g) so the real measure of when value is cheap can be estimated but only if you have a growth rate. This is a case where factors are not tied to valuation which drives security values but to a pricing mechanism. Factors are loosely tied to the market pricing mechanism which is constantly changing thus the sporadic behavior.

Packer
Buy cheap and something good might happen
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by larryswedroe »

Packer, of course we cannot know the future, but we have powerful evidence from several studies showing the spread has significant predictive power as to future return of the premium--the wider the spread the wider the future premium----see 1999 for great example. IMO this looks very much like repeat as most of the underperformance of value is due to FANG stocks and amazingly the returns to stocks with negative cash flow and negative earnings, which historically have been very poor but in 2018, just like as in 1999 they dramatically outperformed!!! That cannot last. At least IMO.
As always my crystal ball remains cloudy but the historical evidence is very strong, and those who ignore history tend to pay for doing so.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by packer16 »

Larry,

I agree with your conclusion if the growth rates of the value & growth groups are constant over time but we do not know this. Do these referenced studies account for growth or at least deal with the issue?

The value factor is missing a key part of the valuation equation for stocks, growth. Without this information, the metric could be misleading or just wrong when looked at in context of growth rates. If for example, the growth rates of value group have declined & the growth rates of the growth group have increased (as we would expect in a more efficient market) then we would expect a larger spread. The Graham formula provides a quantitative way to incorporate growth into valuation multiples.

Packer
Buy cheap and something good might happen
azanon
Posts: 3142
Joined: Mon Nov 07, 2011 9:34 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by azanon »

Rick Ferri wrote: Tue May 28, 2019 1:26 pmI'm not saying this is what you should or shouldn't do, I'm saying there are much bigger things to do as an investor than chase factors, like save more, keep fees low (especially any adviser fees), and be diligent on tax reduction strategies.

Rick Ferri
False dichotomy. Does anyone that participated in this thread disagree with keeping fees low, saving more, and being diligent on tax reduction? I don't.

At least consider moving back to a position I thought was going to be a cornerstone for you, which is not to pick at someone's strategy, as long as the philosophy and discipline are there. As long as factor loading isn't a detriment, then I agree with the Rick that used to not have an issue with someone using a different low-cost, diversified strategy. If you're only going to make room for market-cap weighted strategies, then you're really not open to alternative strategies, from my point-of-view.
BigJohn
Posts: 2627
Joined: Wed Apr 02, 2014 11:27 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by BigJohn »

Random Walker wrote: Tue May 28, 2019 9:04 pm Let’s give a SV fund more credit than you did, say perhaps a total of 3% over the 5% market that you assumed.
Dave, thanks for the detailed response. I think I understand your points but I guess the assumption of a 3% SV premium seems pretty optimistic based on what I've read for even the smallest most valuey funds. If I'm more optimistic than my original post and assume a 1.5% premium, the return of your two examples becomes the same at 3.8% and the Sharpe increases to 0.32. As a result, you've gotten the same return with somewhat less volatility but, if that 1.5% premium doesn't show up, your returns will be lower. So, it appears to me that you've traded lower volatility for a higher risk of under performing a plain vanilla 60/40 market beta portfolio.

I think the bottom line remains that you really have to believe that these premiums will continue to exist at the same magnitude as the historical data. Plus, you absolutely must be willing to stay the course through potentially 10+ years of under performance. I just don't have the confidence that history is a good predictor at this level of granularity as no amount of statistical analysis can turn the very human nature of economies/stocks into a science. However, I do hope it works out as expected for you.

As always, I appreciate the education, thanks :beer
"The greatest enemy of a good plan is the dream of a perfect plan" - Carl Von Clausewitz
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

BigJohn wrote: Wed May 29, 2019 9:07 am
Random Walker wrote: Tue May 28, 2019 9:04 pm Let’s give a SV fund more credit than you did, say perhaps a total of 3% over the 5% market that you assumed.
Dave, thanks for the detailed response. I think I understand your points but I guess the assumption of a 3% SV premium seems pretty optimistic based on what I've read for even the smallest most valuey funds. If I'm more optimistic than my original post and assume a 1.5% premium, the return of your two examples becomes the same at 3.8% and the Sharpe increases to 0.32. As a result, you've gotten the same return with somewhat less volatility but, if that 1.5% premium doesn't show up, your returns will be lower. So, it appears to me that you've traded lower volatility for a higher risk of under performing a plain vanilla 60/40 market beta portfolio.

I think the bottom line remains that you really have to believe that these premiums will continue to exist at the same magnitude as the historical data. Plus, you absolutely must be willing to stay the course through potentially 10+ years of under performance. I just don't have the confidence that history is a good predictor at this level of granularity as no amount of statistical analysis can turn the very human nature of economies/stocks into a science. However, I do hope it works out as expected for you.

As always, I appreciate the education, thanks :beer
BigJohn,
All the above concerns are mine as well! As Bogle wrote “investing is an act of faith” :-). More important to stick to a plan than have the optimal one.

Dave
User avatar
HomerJ
Posts: 21282
Joined: Fri Jun 06, 2008 12:50 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by HomerJ »

Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
And a 30/70 stocks/bonds portfolio has like 95% of its risk wrapped up in a single factor. So your method of determining risk makes it sound like the 30/70 portfolio is even riskier than than 60/40 portfolio.

And of course, you gloss over the fact that the "single" factor actually includes almost ALL factors.

That's like saying Total Stock Market Index Fund is riskier than a bucket of stocks because investing in TSM has all its risk in a "single" fund.
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Rick Ferri »

azanon wrote: Wed May 29, 2019 8:48 am
Rick Ferri wrote: Tue May 28, 2019 1:26 pmI'm not saying this is what you should or shouldn't do, I'm saying there are much bigger things to do as an investor than chase factors, like save more, keep fees low (especially any adviser fees), and be diligent on tax reduction strategies.

Rick Ferri
False dichotomy. Does anyone that participated in this thread disagree with keeping fees low, saving more, and being diligent on tax reduction? I don't.

At least consider moving back to a position I thought was going to be a cornerstone for you, which is not to pick at someone's strategy, as long as the philosophy and discipline are there. As long as factor loading isn't a detriment, then I agree with Rick that used to not have an issue with someone using a different low-cost, diversified strategy. If you're only going to make room for market-cap weighted strategies, then you're really not open to alternative strategies, from my point-of-view.
Successful index investing requires the discipline to stay with a strategy for a very long time - a lifetime. If you're willing to stick with a high factor tilt for a lifetime, then go for it. Most people will not. So they shouldn't own them now.

Some people will say that's why you hire an adviser.

Prepare for a rant...

As a person who consults hourly with dozens of different investors each month, I've seen a tremendous overuse of alternative funds in client's accounts. Particularly bad is the overuse of tax-inefficient DFA funds in taxable accounts, put there by advisers. Many times I come across redundant coverage of factors using overlapping DFA funds, and DFA funds and ETFs that cover the same factors.

This is complexity for the sake of job security. The advisers think they're going to be employed by the client forever because they make things so darn complicated. But what those advisers don't understand is the internet is a tremendous knowledge leveling machine. It takes the mystique away from the adviser and exposes portfolio flaws and extra cost, which includes the adviser's fee. That's when the adviser's "value-added" con game falls apart, and a client fires their adviser.

And that's when I see these poorly managed portfolios. I'm asked to unwind the damage to the extent it can be unwound.

There is NO DOUBT that most people should not use factors, and definitely should not hire an adviser who touts the glory of a factor fund portfolio that most people WILL someday fall out of love with and seek an expensive (taxing) divorce.

Rick Ferri
Last edited by Rick Ferri on Wed May 29, 2019 9:37 am, edited 1 time in total.
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
azanon
Posts: 3142
Joined: Mon Nov 07, 2011 9:34 am

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by azanon »

Rick Ferri wrote: Wed May 29, 2019 9:26 am
azanon wrote: Wed May 29, 2019 8:48 am
Rick Ferri wrote: Tue May 28, 2019 1:26 pmI'm not saying this is what you should or shouldn't do, I'm saying there are much bigger things to do as an investor than chase factors, like save more, keep fees low (especially any adviser fees), and be diligent on tax reduction strategies.

Rick Ferri
False dichotomy. Does anyone that participated in this thread disagree with keeping fees low, saving more, and being diligent on tax reduction? I don't.

At least consider moving back to a position I thought was going to be a cornerstone for you, which is not to pick at someone's strategy, as long as the philosophy and discipline are there. As long as factor loading isn't a detriment, then I agree with Rick that used to not have an issue with someone using a different low-cost, diversified strategy. If you're only going to make room for market-cap weighted strategies, then you're really not open to alternative strategies, from my point-of-view.
Successful index investing requires the discipline to stay with a strategy for a very long time - a lifetime. If you're willing to stick with a high factor tilt for a lifetime, then go for it. Most people will not. So they shouldn't own them now.

Some people will say that's why you hire an adviser.

Prepare for a rant...

As a person who consults hourly with dozens of different investors each month, I've seen a tremendous overuse of alternative funds in client's accounts. Particularly bad is the overuse of tax-inefficient DFA funds in taxable accounts, put there by advisers. Many times I come across redundant coverage of factors using overlapping DFA funds, and DFA funds and ETFs that cover the same factors.

This is complexity for the sake of job security. The advisers think they're going to be employed by the client forever because they make things so darn complicated. But what those advisers don't understand is the internet is a tremendous knowledge leveling machine. It takes the mystique away from the adviser and exposes portfolio flaws and extra cost. That's when the advisers "value added" con game falls apart, and a client fire their adviser.

And that's when I see these poorly managed portfolios. I'm asked to unwind the damage to the extent it can be unwound.

There is NO DOUBT that most people should not use factors, and definitely should not hire an adviser who touts the glory of a factor fund portfolio that you WILL someday fall out of love with and seek a divorce.

Rick Ferri
Thanks for the response and the clarification.

Yeah, I know I'm fortunate that 95% of my overall portfolio is tax-advantaged, so I'm afforded tremendous flexibility where that's concerned (I'm among the "mortals" who actually qualifies for Roth IRAs (+ spouses) and have a 401(k), and I can't afford to max all of those + save more).

My only (significantly sized) taxable holding is all in Vanguard Target Retirement Income Mutual fund (VTINX). I imagine it's not perfect tax-efficiency especially being a mutual fund, but it's good enough and I'm pretty comfortable with 30% equities for an emergency fund/shorter-term holdings.
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

HomerJ wrote: Wed May 29, 2019 9:14 am
Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
And a 30/70 stocks/bonds portfolio has like 95% of its risk wrapped up in a single factor. So your method of determining risk makes it sound like the 30/70 portfolio is even riskier than than 60/40 portfolio.

And of course, you gloss over the fact that the "single" factor actually includes almost ALL factors.

That's like saying Total Stock Market Index Fund is riskier than a bucket of stocks because investing in TSM has all its risk in a "single" fund.
If assume SD stocks about 20% and SD bonds 5%, actually about 63% of your 30/70 portfolio risk still wrapped up in equity market factor. That’s a lot less than the 85-90% risk in a 60/40 portfolio. As you know, TSM does contain all stocks with all their given factor exposures. There is just no net exposure to size, value, cs momentum, profitability, etc. By definition the positive exposures are cancelled out by the negative exposures.

Dave
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Seasonal »

If markets are more or less efficient, all a factor based portfolio and the market portfolio would be expected to have the same risk-adjusted return.There's no free lunch in this case. Furthermore, in this case, the market chooses the optimal portfolio for the representative investor, which is the cap-weighted market portfolio.

An investor may be different from the representative investor, but all investors can't be different in the same direction.

Markets may well be substantially inefficient. In this case, it's a bit odd to follow a strategy ultimately based on research by Gene Fama, who is a major advocate of efficient markets and won a Noble prize for his work in the area.

There are no doubt a lot of investors who don't have a clue. Whether they all don't have a clue in the same direction and have enough market power to move the market in a direction that would make factor investing better than market cap investing on a risk-adjusted basis is, shall we say, very far from clear. Fama's research also indicates that skill (the ability to beat the market consistently on a risk-adjusted basis) is remarkably rare. Most who can keep the excess profit for themselves (through fees).
User avatar
Rick Ferri
Posts: 9708
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Rick Ferri »

Every investment adviser thinks they're above average.

Every investment adviser charges what they believe is a bargain for their astute advice and services.

Every investment adviser says they add-value to a client's financial well-being.

Very few advisers do.
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Seasonal »

Random Walker wrote: Wed May 29, 2019 9:39 am
HomerJ wrote: Wed May 29, 2019 9:14 am
Random Walker wrote: Mon May 27, 2019 2:30 pm Most people don’t appreciate that a typical 60/40 portfolio has 85-90% of its risk wrapped up in a single factor.
And a 30/70 stocks/bonds portfolio has like 95% of its risk wrapped up in a single factor. So your method of determining risk makes it sound like the 30/70 portfolio is even riskier than than 60/40 portfolio.

And of course, you gloss over the fact that the "single" factor actually includes almost ALL factors.

That's like saying Total Stock Market Index Fund is riskier than a bucket of stocks because investing in TSM has all its risk in a "single" fund.
If assume SD stocks about 20% and SD bonds 5%, actually about 63% of your 30/70 portfolio risk still wrapped up in equity market factor. That’s a lot less than the 85-90% risk in a 60/40 portfolio. As you know, TSM does contain all stocks with all their given factor exposures. There is just no net exposure to size, value, cs momentum, profitability, etc. By definition the positive exposures are cancelled out by the negative exposures.

Dave
In a multi-factor world, SD is far from an adequate metric for risk.
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Seasonal »

Rick Ferri wrote: Wed May 29, 2019 9:44 am Every investment adviser thinks they're above average.

Every investment adviser charges what they believe is a bargain for their astute advice and services.

Every investment adviser says they add-value to a client's financial well-being.

Very few advisers do.
Absolutely. For those who do add value, encouraging investors to have a portfolio with an appropriate amount of risk and to stay the course rather than acting emotionally most likely far outweighs considerations such as factors v. cap weighting.
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

Seasonal wrote: Wed May 29, 2019 9:45 am In a multi-factor world, SD is far from an adequate metric for risk.
Even more reason to diversify as broadly as possible across factors and other sources of risk/return.

Dave
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by larryswedroe »

Just to add, in any world SD is only ONE measure of risk, others must be considered, like skewness and kurtosis, and liquidity, and when risks tend to show up.
Larry
User avatar
9-5 Suited
Posts: 1307
Joined: Thu Jun 23, 2016 12:14 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by 9-5 Suited »

Random Walker wrote: Mon May 27, 2019 6:03 pmIf there are areas of difference of opinion, ...
Dave
Thanks for taking the time for such a thoughtful reply. That perspective makes a lot of sense. I find Larry's material to be among the best of the pack, so I certainly understand and the point about not being able to recognize one's own behavioral errors and submitting fully to a plan is an interesting one. And it's obvious from the general quality of your posts that you have internalized the meaning of Larry's writings to a significant degree, and help others to understand the implications.

I suppose for myself, that would feel too much like an appeal to Larry's authority (abandoning the capacity to disagree on the grounds that he is smarter or more knowledgeable than me) since there are certainly many experts who disagree with him on practical portfolio construction. So the cardiologist example fails a bit there, as it's rare in true scientific disciplines to have cardiologists disagreeing about the most foundational truths of how the heart works. There are areas of intense disagreement in science to be sure, but it's precisely in those areas where it pays to be a skeptic, at least IMO.
Seasonal
Posts: 3045
Joined: Sun May 21, 2017 1:49 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Seasonal »

Random Walker wrote: Wed May 29, 2019 10:19 am
Seasonal wrote: Wed May 29, 2019 9:45 am In a multi-factor world, SD is far from an adequate metric for risk.
Even more reason to diversify as broadly as possible across factors and other sources of risk/return.

Dave
To the extent that implies something other than cap weighting, that has never made any sense to me.

Just because there are many ways to measure or quantify something has no obvious implications for the ratio of those measurements. Objects have height, width and depth, but equalizing the three measures is silly. The food we eat has carbs, fats and proteins, but an ideal diet would not equalize the three measures.

See my post above about the market choosing the ideal mix, presuming a reasonable degree of efficiency.
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

9-5 Suited wrote: Wed May 29, 2019 10:35 am I suppose for myself, that would feel too much like an appeal to Larry's authority (abandoning the capacity to disagree on the grounds that he is smarter or more knowledgeable than me) since there are certainly many experts who disagree with him on practical portfolio construction. So the cardiologist example fails a bit there, as it's rare in true scientific disciplines to have cardiologists disagreeing about the most foundational truths of how the heart works. There are areas of intense disagreement in science to be sure, but it's precisely in those areas where it pays to be a skeptic, at least IMO.
I think one of the biggest tests when trying to evaluate someone else’s statements, logic, clarity of thought is internal consistency. When one reads Larry’s writings from the points of view of modern portfolio theory, improved portfolio efficiency, and cost per unit value added, I think he is completely internally consistent. He strongly believes in market efficiency with a big nod to behavioral anomalies that are unlikely to be arbitraged away. I am not sure there are many experts who would disagree with him on portfolio construction. The basics of how individual investments mix in a portfolio is effectively “done science”. (That statement says zero about how any individual investment will actually do in the future.) I think the only issues are costs, specifically marginal costs/marginal benefits, and that is where people need to make their own decisions.

Dave
User avatar
9-5 Suited
Posts: 1307
Joined: Thu Jun 23, 2016 12:14 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by 9-5 Suited »

Random Walker wrote: Wed May 29, 2019 11:41 am I think one of the biggest tests when trying to evaluate someone else’s statements, logic, clarity of thought is internal consistency. When one reads Larry’s writings from the points of view of modern portfolio theory, improved portfolio efficiency, and cost per unit value added, I think he is completely internally consistent. He strongly believes in market efficiency with a big nod to behavioral anomalies that are unlikely to be arbitraged away. I am not sure there are many experts who would disagree with him on portfolio construction. The basics of how individual investments mix in a portfolio is effectively “done science”. (That statement says zero about how any individual investment will actually do in the future.) I think the only issues are costs, specifically marginal costs/marginal benefits, and that is where people need to make their own decisions.

Dave
Internal consistency is a valuable trait, but it is susceptible to the bias of "begging the question". To use an extreme argument, if I take as presupposition crystals have healing powers, it is perfectly internally consistent to lay in a bath of crystals every night and advise others to do the same.

In this less extreme case, there are absolutely significant disagreements about the forward predictability of the "factor zoo". There's minimal disagreement for example that small and value have outperformed, but a lot of disagreement about whether that's likely to persist and/or will actually lower future portfolio risk. It isn't "settled science" as you are presenting it. Settled science - if there even is such a thing, and I'm not so sure there is!! - tells me that if I drop a rock off a building, it will fall to the ground.
Topic Author
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by Random Walker »

9-5,
We are in agreement. When I mentioned “settled science”, I was referring generically to how uncorrelated investments with similar expected returns mix in a portfolio. I agree there is no telling how specific factors will perform in the future.

Dave
User avatar
matjen
Posts: 2189
Joined: Sun Nov 20, 2011 10:30 pm

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by matjen »

Great interview. Thanks for posting Dave.
A man is rich in proportion to the number of things he can afford to let alone.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: Excellent Larry Swedroe Podcast Interview: Factors For The Long Run

Post by larryswedroe »

Seasonal, to be helpful re this comment
"To the extent that implies something other than cap weighting, that has never made any sense to me. "

Here is the simple logic that many even legendary investors like Bridgeway's Ray Dalio have adopted in one form or another (like the permanent portfolio or the all weather portfolio)

The basic premise of your belief system is markets are highly if not perfectly efficient.
If you believe that then it follows you should believe that all risky assets should have similar risk-adjusted returns, not similar returns, but similar risk-adjusted returns. If not then arbitrageurs would move capital to bring equilibrium.

If you believe that then you should diversify across as many unique (low correlated) sources of risk as you can identify that meet your established criteria. The ones I've laid out, and I believe most in the academic community accept and use are persistence, pervasiveness, robustness, intuitiveness and implementability.
In the whole factor zoo where there are more than 600 factors in the literature, as shown in my factor book I believe only a mere handful or so meet that criteria.

If all risk assets/factors have same risk adjusted return why would you want to load up a portfolio with 90% or so of risk in just one factor instead of diversifying across unique factors, putting more weight on the ones you have the most confidence in, or equal weighting (a risk parity portfolio). And as explained in my books the typical Boglehead portfolio or typical 60/40 TSM portfolio has 85%+ of risk in one single factor called market beta. That is inconsistent with the above beliefs. Which is why btw legendary investors like David Swensen and Ray Dalio and others use more of a risk parity type approach, diversifying across unique sources of risk. Diversification is for those that don't know what the future holds. Concentration is for those that think they do!

Of course you are free to disagree with any of the premises, but then must accept that 90% of risk is likely to be in one factor, which has underperformed dramatically over very long periods, even underperforming riskless tbills over 17 years from 66-82 during which the S&P underperformed small value, by get this, over 1100% in terms of total returns. And there are two other periods of at least 13 years when S&P underperformed tbills, totalling 45 of the 90 years since 1929!!!!

I hope that is helpful
Larry
Locked