How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Fri Sep 13, 2019 1:34 pm

For those of you who have decided to adopt factor investing because of a desire to diversify across multiple distinct sources of risk, how do you evaluate your ability and willingness to take on risks associated with factors other than market beta? This question comes from a perspective that factor models have at least some validity and are based on risk-based explanations rather than behavior-based explanations. If you think factor investing is bunk, you can safely ignore the post, and if you think tilting towards factors is a risk-free boost to your expected returns because of irrational behaviors of market participants, you can skip it as well.

For me at least, it is easiest to start off thinking about things in the Capital Asset Pricing Model (CAPM) without any factors other than market beta. In this model, each asset's expected return is equal to the risk-free return plus the asset's exposure to market beta (i.e., the sensitivity of the expected excess asset returns to the expected excess market returns) times the market risk premium. In this model, it is very easy to understand the nature of the risk that justifies higher investment returns in certain assets: An asset's exposure to market beta is essentially a measure of the asset's volatility, and more volatile assets have higher expected returns.

The classic result of the CAPM model is that efficient portfolios consist of a combination of the market portfolio plus some amount of risk-free assets. In the simplified world of this model, this produces a very neat picture where investors who want less volatility than the market portfolio (at the cost of lower expected returns) add in some risk-free assets and investors who want more volatility than the market portfolio (but with higher expected returns) leverage up the market portfolio by borrowing at the risk-free rate (if only!).

In the real world, most investors use their stock-bond asset allocation to achieve a similar effect. Rather than holding stocks and bonds in the same ratio as the market portfolio, investors who want higher expected return with higher volatility tilt their portfolio by overweighting stocks relative to the market portfolio. Similarly, investors who want lower volatility with a lower expected return tilt their portfolio by underweighting stocks relative to the market portfolio.

Achieving an appropriate stock-bond asset allocation for your personal risk tolerance can be pretty difficult, but there is--at least in principle--a well-understood framework for doing so. You want to pick a mix of stocks and bonds that is appropriate for your need for returns judged relative to your willingness and ability to be exposed to the risk of volatility.

In Fama-French style factor models, an asset's expected return depends not just on its exposure to market beta but also on its sensitivity to other risk factors, such as the risks characteristically associated with small versus large stocks or the risks characteristically associated with value versus growth stocks. Many factor investors (including Fama) view these models as consistent with some version of the Efficient Market Hypothesis because rational investors may want more or less exposure to these characteristic risks than the market portfolio provides. Because of his very strong commitment to the EMH, Fama himself would say that for every (dollar-weighted) investor that rationally should tilt towards small stocks there should be another (dollar-weighted) investor who rationally should tilt towards large stocks. Just as your stock-bond asset allocation rationally depends on your willingness and ability to be exposed to the risk of volatility in more standard models, in factor models your exposure to different factors (relative to the market portfolio) should be based on your willingness and ability to be exposed to the risks associated with that factor.

I am pretty sympathetic to all of this, but how is an investor supposed to judge their willingness and ability to take on the risks associated with factors other than market beta? I can understand the risk of volatility in a pretty granular way. We do this all the time when thinking about asset allocation decisions: how bad would it feel if the stock market dropped 50%, etc. But I rarely see a specific or quantitative explanation of the nature of the risks associated with other factors.

Sometimes you hear some explanations about these risks. Some might say, for example, that you need to be willing to face the risk of decades of underperformance of value in order to benefit from the rapid but large bursts of overperformance. But you can also find totally distinct explanations for the risk associated with value. For example, others might say that value stocks are characteristically more risky because they typically have higher financial leverage and so are more exposed to the risk of decline in overall credit conditions, such as those that occur during a financial crisis. From what I have seen after quite a bit of looking, there does not seem to be a coherent way to understand the risks associated with factors other than market beta in a clear and quantifiable way.

If you cannot clearly understand the nature of the risks associated with a factor, how can you decide whether you want more or less exposure to these risks than the market portfolio provides? How do all of you factor investors out there decide how deeply to tilt? How do you decide that you want more SMB or HML risk than the market rather than less? What is the process you use that is the analogue of the exercises we all go through to find the right stock-bond allocation?

psteinx
Posts: 3307
Joined: Tue Mar 13, 2007 2:24 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by psteinx » Fri Sep 13, 2019 2:34 pm

HootingSloth wrote:
Fri Sep 13, 2019 1:34 pm
Because of his very strong commitment to the EMH, Fama himself would say that for every (dollar-weighted) investor that rationally should tilt towards small stocks there should be another (dollar-weighted) investor who rationally should tilt towards large stocks.
Count me among the factor skeptics, so I am going against your recommendation for folks such as I to ignore this thread.

That said, as a cautiously factor-positive individual, I think you've found one of (but far from the only), flaws in factor investing.

Factor-heads frequently try to justify their position by saying that factors *SHOULD* continue to outperform, because of yada yada risk. "Just like beta, and you believe in beta, right?"

But whereas it's fairly easy for 3 fund proponents to identify those who should be over or under exposed to beta, based on their personal situation (and generally using bonds to fill out the lower-risk portion of their portfolio, as you've mentioned), it's MUCH harder to identify those who both:

1) Believe in factor models (either SV, or more complex models)
and
2) Who themselves have UNDERWEIGHTED SV or the other factors, attempting to mitigate their overall risk (and/or recommend this commonly to others).
Last edited by psteinx on Fri Sep 13, 2019 5:42 pm, edited 1 time in total.

User avatar
vineviz
Posts: 5129
Joined: Tue May 15, 2018 1:55 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by vineviz » Fri Sep 13, 2019 3:06 pm

HootingSloth wrote:
Fri Sep 13, 2019 1:34 pm
For me at least, it is easiest to start off thinking about things in the Capital Asset Pricing Model (CAPM) without any factors other than market beta. In this model, each asset's expected return is equal to the risk-free return plus the asset's exposure to market beta (i.e., the sensitivity of the expected excess asset returns to the expected excess market returns) times the market risk premium. In this model, it is very easy to understand the nature of the risk that justifies higher investment returns in certain assets: An asset's exposure to market beta is essentially a measure of the asset's volatility, and more volatile assets have higher expected returns.
I want to start my answer here, but because I think this assumption (" it is very easy to understand the nature of the risk" in a single-factor CAPM) glosses over the fact that it would be virtually impossible for an individual investor to objectively choose a portfolio even in such an environment.

With just one source of risk, i.e. variance, the investor would still need a very precise understanding of their utility function in order to choose a portfolio purely on objective grounds: Is your risk aversion absolute or relative? Constant, rising, or declining across payouts? If rising/declining is the change monotonic or exponential? Is your risk aversion intertemporally stable or time-varying? And so forth.

I bring this up mainly to bolster the argument that, even for purposes of theoretical discussions, we should be careful to apply similar standards of proof to investors who profess adherence to single-factor models as we apply to investors who profess adherence to multi-factor models.

HootingSloth wrote:
Fri Sep 13, 2019 1:34 pm
I am pretty sympathetic to all of this, but how is an investor supposed to judge their willingness and ability to take on the risks associated with factors other than market beta?
The factors we normally discuss are asset-pricing factors (market beta, size, value, etc.) and these may or may not directly map to the risks that investors accept or avoid in constructing their portfolios.

Some factors may line up (e.g. liquidity risk, inflation risk) and I suspect many investors at least implicitly account for that in choosing their investment portfolios. Other investors account for the factor exposure in their human capital (e.g. if your employment is in a sector like technology that typically has strong negative loads on the value factor, you should probably invest your equity portfolio in strong positive value factor investments).

But to be honest I suspect the most important thing that a potential multi-factor investor needs to be aware of about their utility function is how sensitive they are to things like confirmation bias, regret, and/or groupthink. Because constructing a portfolio that looks different from the market is, from a psychological perspective, itself a source of risk for the investor.

If they can resist the forces of regret and groupthink, significant tracking error might be tolerated. If they are susceptible to those forces? Probably not so tolerant. A wise investor will intuitively grasp this about themselves, but experience is the only true test.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Fri Sep 13, 2019 3:18 pm

psteinx wrote:
Fri Sep 13, 2019 2:34 pm
Count me among the factor skeptics, so I am going against your recommendation for folks such as I to ignore this thread.
The input is still very much appreciated, thanks. :sharebeer

For what it is worth, I do not personally have any factor tilts in my portfolio, and would only be willing to switch if I had a strong conviction that it would make sense for my personal situation to stick with a tilt for the very long term.
psteinx wrote:
Fri Sep 13, 2019 2:34 pm

But whereas it's fairly easy for 3 fund portfolios to identify those who should be over or under exposed to beta, based on their personal situation (and generally using bonds to fill out the lower-risk portion of their portfolio, as you've mentioned), it's MUCH harder to identify those who both:

1) Believe in factor models (either SV, or more complex models)
and
2) Who themselves have UNDERWEIGHTED SV or the other factors, attempting to mitigate their overall risk (and/or recommend this commonly to others).
I agree that the inability to identify the kinds of investors who rationally should underweight things like small value stocks is an important cause for concern. It seems relatively easier to find the kinds of investors who rationally should underweight market beta by holding a low proportion of stocks.

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Fri Sep 13, 2019 3:46 pm

vineviz wrote:
Fri Sep 13, 2019 3:06 pm
I want to start my answer here, but because I think this assumption (" it is very easy to understand the nature of the risk" in a single-factor CAPM) glosses over the fact that it would be virtually impossible for an individual investor to objectively choose a portfolio even in such an environment.

With just one source of risk, i.e. variance, the investor would still need a very precise understanding of their utility function in order to choose a portfolio purely on objective grounds: Is your risk aversion absolute or relative? Constant, rising, or declining across payouts? If rising/declining is the change monotonic or exponential? Is your risk aversion intertemporally stable or time-varying? And so forth.

I bring this up mainly to bolster the argument that, even for purposes of theoretical discussions, we should be careful to apply similar standards of proof to investors who profess adherence to single-factor models as we apply to investors who profess adherence to multi-factor models.
Thanks, vineviz. I think we are already mostly in agreement on this point. The nature of the risk (volatility) is relatively easy to understand but understanding how much exposure to that risk you want over your entire investing lifetime is still difficult. That is why I said in my original post that "Achieving an appropriate stock-bond asset allocation for your personal risk tolerance can be pretty difficult" despite the relatively simple nature of the risk.

I certainly do not have any illusion that you can achieve some sort of objective optimum stock-bond allocation. I only attempt to reach "an appropriate" allocation, which probably falls within a range of reason and depends as much on my ability to avoid behavioral mistakes as it does on trying to make a detailed specification of my personal utility function and how it changes over time. I think you and I would agree that it would be a futile endeavor to try to do that.

My point here is that, even with a risk that is relatively easy to understand, we still have a lot of work to do to figure out how much exposure to that risk we want. There are countless posts on Bogleheads doing just that, often within the framework of one's ability, willingness and need to take risk. This task seems even more difficult if we cannot get a real grasp on the nature of the risk. But there seems to be very little discussion on how to carry out this even more difficult task, and I was hoping to see how other's have thought about it.
vineviz wrote:
Fri Sep 13, 2019 3:06 pm
The factors we normally discuss are asset-pricing factors (market beta, size, value, etc.) and these may or may not directly map to the risks that investors accept or avoid in constructing their portfolios.

Some factors may line up (e.g. liquidity risk, inflation risk) and I suspect many investors at least implicitly account for that in choosing their investment portfolios. Other investors account for the factor exposure in their human capital (e.g. if your employment is in a sector like technology that typically has strong negative loads on the value factor, you should probably invest your equity portfolio in strong positive value factor investments).

But to be honest I suspect the most important thing that a potential multi-factor investor needs to be aware of about their utility function is how sensitive they are to things like confirmation bias, regret, and/or groupthink. Because constructing a portfolio that looks different from the market is, from a psychological perspective, itself a source of risk for the investor.

If they can resist the forces of regret and groupthink, significant tracking error might be tolerated. If they are susceptible to those forces? Probably not so tolerant. A wise investor will intuitively grasp this about themselves, but experience is the only true test.
It seems your answer is largely that the kinds of investors who should tilt towards factors like small and value are just those who have the wisdom to avoid confirmation bias, regret, and groupthink. This is, I think, essentially a behavior-based explanation of factor returns. Completely fine if this is your conviction. But I also may be misunderstanding you.

It seems to me that, instead, gaining an understanding of why you personally have a better ability to bear a particular kind of risk than other market participants is an important part of determining whether you will, in practice, be able to tolerate whatever tracking error shows up.

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Fri Sep 13, 2019 4:12 pm

My recommendation for factor tilters would be to tilt enough to make a difference but not to take extreme tilts. I would also limit the Alternative investments to 10% or at most 20% of a portfolio. I just don't think you can ignore the Large Growth stocks.

Just eyeballing, a Nedsaid 60/40 tilted portfolio might look something like this:

15% Total Stock Market Index
15% Total International Stock Index
20% US Small Value
5% Factor Style Premia fund (Long/short and leveraged)
5% Liquid Alt Income fund
35% US Bond Index

Edit: Whoops, I noted the percentages didn't add up to 100%, I was 5% short. I suppose I should have had 25% US Small Value instead. I didn't have International Small Value because good products in that area are hard to find unless you have access to Dimensional Funds through an advisor. You could also have a percentage in an International Value fund. This should have been a 55% stock, 35% bond, 10% Alt portfolio. Since 5% is Style Premia and equity-like, 25% could be allocated to equity factor products. Lots of folks here would put that 25% into a US Small Value fund.

Pretty much, factor funds would be no more than 50% of a stock portfolio, the rest would be in the Broad Indexes. Alternatives, whether the Liquid Alts and/or Interval Funds would be 10% to 20% of a portfolio if you used them at all. Your bond investments would be pretty plain vanilla, I know Larry would recommend Treasuries. You could also use factor ETFs like Momentum and Quality instead of the Factor Style Premia fund and some of the Small Value. Many ways to slice the pizza, the point is that I wouldn't go over 50% of a stock portfolio for factor tilts.

The so-called Larry portfolio is 30% Small Value both US and International and 70% Short Term US Treasuries. An extreme Small/Value tilt with the stocks, the risk taken by the extreme tilt offset by an ultra-conservative allocation of bonds and choice of bonds. I just don't go for extremes and I think it is okay to take a little risk on the bond side. I guess I was born to be mild.

My real-life portfolio has 5-6% in US Small Value, less than 1% in International Small Value, about 1.4% in Liquid Alts, and pretty standard investments for the rest. My own asset allocation is 62% stocks/38% bonds and cash. So my tilts are pretty mild and cautious.

Edit: My stock portfolio is 6% Small Value, most of that is US Small Value but I do own an International Small Value ETF. Small Value is about 4% of my portfolio as a whole. The Liquid Alts are 1.4% of my portfolio as a whole and this just over 1% of the equity part of my portfolio.
Last edited by nedsaid on Sat Sep 14, 2019 1:09 am, edited 3 times in total.
A fool and his money are good for business.

User avatar
Horton
Posts: 318
Joined: Mon Jan 21, 2008 3:53 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by Horton » Fri Sep 13, 2019 4:13 pm

Apologies if I am derailing the discussion, but perhaps it's better to first agree upon the definition of "risk". For many, risk seems to be mostly (entirely?) focused entirely on volatility. Personally, I define risk in more simple terms - "will I be able to achieve my personal goals?" When viewed from this lens, things like human capital*, health insurance, life insurance, disability insurance, savings rate, stock/bond allocation, retirement income, emergency fund, Social Security, etc. are paramount. Factor investing is pretty far down the list. It may work and it may not. It's a zero sum game in the end; if value is winning, then conversely growth is losing. I find Taleb's Antifragile idea to be an important mental model. Focus first on making your goals robust/antifragile. I'm not sure factor investing moves the needle much from that perspective.

Another random aside...and I'm not sure whether this is true or not...but I view factor investing as potentially more beneficial in the decumulation phase when volatility can pose a significant risk via sequence of returns risk if you are using a fixed or variable withdrawal strategy (as opposed to a SPIA, liability matching, or floor/upside).

* I'm using human capital here in a very broad sense to include education, credentialing, pay, and enjoyment.
"You must know that there is nothing higher and stronger and more wholesome and good for life in the future than some good memory, especially a memory of childhood, of home." - Dostoyevsky

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Fri Sep 13, 2019 4:23 pm

Horton, good points raised. Lots of pillow fights here over the definition of diversification and indeed over the definition of risk. We probably come to agreement on 70-80% on the definitions and the fights are over the other 20-30%. So lots of common ground but still some disagreement.

To me the biggest risk is the loss of purchasing power of the US Dollar over time. Portfolio volatility is considered but to me is a secondary priority to inflation risk. To me, inflation is public enemy number one. The idea of factor investing is to boost returns a bit while reducing volatility at the same time addressing the inflation risk and the volatility risk at the same time. But we know that what looks so persuasive in a book or article somehow has an implementation problem, sort of an Ivory Tower vs. Real Life argument. Not so nice and neat in real life.

No real final answers or a way to resolve the debate as we don't know the future. I come down on the side of factor tilting but I say not to overdo it. You have to sort of hedge against the possibility of being wrong.
A fool and his money are good for business.

stlutz
Posts: 5348
Joined: Fri Jan 02, 2009 1:08 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by stlutz » Fri Sep 13, 2019 4:32 pm

Back in the late 90s when I dropped individual stocks picking and switched to index investing, I thought in terms of what percent of my portfolio I wanted to make active bets with.

I adopted a 75/25 stock bond allocation because that was generally what a "moderately aggressive" investor would do. I next made a pretty arbitrary decision to use 1/3 global stocks. I then fairly arbitrarily decided that I'd put 1/2 of my US in index funds that were buying the types of stocks I was selecting when I was buying individual stocks (i.e. value and smallcap stocks).

This was all before "factor investing" became a big marketing term. But I notice the word "arbitrary" coming up multiple times. "Ability and willingness to take risks" is a phrase with a lot syllables in it, but you're always taking a pretty rough cut at it.

(My investing philosophy has since changed so while I hold onto my small/value funds I'm not adding to them)

User avatar
Tyler Aspect
Posts: 1447
Joined: Mon Mar 20, 2017 10:27 pm
Location: California
Contact:

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by Tyler Aspect » Fri Sep 13, 2019 5:19 pm

To me the practice of involving in factor investing is an exercise in discipline. Obviously some people participate in factor investing because certain investment styles had a historical risk adjusted better performance compared to market average. There are many factors out there, and there should be some factors that out-performs as opposed to under-perform. The problem becomes if over a period of a few years you found out your invested factors were under-performers during the period you were invested in, would you switch your style tilt to a different set of more popular factors? That would threaten to devolve into garden variety performance chasing, and the all too-common effect of buy-high, sell-low.

"Small value is where the value is - pick me!"

"Large growth is where the money is - pick me!"

"Minimal volatility is the least risky stuff - pick me!"

"Join the momentum while it is hot - pick me!"

Can you really stick to your chosen factor?

Factor risks for me is the degree of tilt away from market weightings, and some additional risks for wavering investment behavior. These are objective measures of performance variance, and if these variances could be turned into realized losses.
Past result does not predict future performance. Mentioned investments may lose money. Contents are presented "AS IS" and any implied suitability for a particular purpose are disclaimed.

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Fri Sep 13, 2019 5:29 pm

nedsaid wrote:
Fri Sep 13, 2019 4:12 pm
Pretty much, factor funds would be no more than 50% of a stock portfolio, the rest would be in the Broad Indexes. Alternatives, whether the Liquid Alts and/or Interval Funds would be 10% to 20% of a portfolio if you used them at all. Your bond investments would be pretty plain vanilla,

The so-called Larry portfolio is 30% Small Value both US and International and 70% Short Term US Treasuries. An extreme Small/Value tilt with the stocks, the risk taken by the extreme tilt offset by an ultra-conservative allocation of bonds and choice of bonds. I just don't go for extremes and I think it is okay to take a little risk on the bond side. I guess I was born to be mild.

My real-life portfolio has 5-6% in US Small Value, less than 1% in International Small Value, about 1.4% in Liquid Alts, and pretty standard investments for the rest. My own asset allocation is 62% stocks/38% bonds and cash. So my tilts are pretty mild and cautious.
Thanks, nedsaid. This was the kind of thinking I was hoping people would be willing to share. It seems like you know yourself well.

User avatar
willthrill81
Posts: 12778
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by willthrill81 » Fri Sep 13, 2019 5:31 pm

vineviz wrote:
Fri Sep 13, 2019 3:06 pm
But to be honest I suspect the most important thing that a potential multi-factor investor needs to be aware of about their utility function is how sensitive they are to things like confirmation bias, regret, and/or groupthink. Because constructing a portfolio that looks different from the market is, from a psychological perspective, itself a source of risk for the investor.

If they can resist the forces of regret and groupthink, significant tracking error might be tolerated. If they are susceptible to those forces? Probably not so tolerant. A wise investor will intuitively grasp this about themselves, but experience is the only true test.
:thumbsup

I entirely agree.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Fri Sep 13, 2019 5:33 pm

Horton wrote:
Fri Sep 13, 2019 4:13 pm
Apologies if I am derailing the discussion, but perhaps it's better to first agree upon the definition of "risk". For many, risk seems to be mostly (entirely?) focused entirely on volatility. Personally, I define risk in more simple terms - "will I be able to achieve my personal goals?" When viewed from this lens, things like human capital*, health insurance, life insurance, disability insurance, savings rate, stock/bond allocation, retirement income, emergency fund, Social Security, etc. are paramount. Factor investing is pretty far down the list. It may work and it may not. It's a zero sum game in the end; if value is winning, then conversely growth is losing. I find Taleb's Antifragile idea to be an important mental model. Focus first on making your goals robust/antifragile. I'm not sure factor investing moves the needle much from that perspective.

Another random aside...and I'm not sure whether this is true or not...but I view factor investing as potentially more beneficial in the decumulation phase when volatility can pose a significant risk via sequence of returns risk if you are using a fixed or variable withdrawal strategy (as opposed to a SPIA, liability matching, or floor/upside).

* I'm using human capital here in a very broad sense to include education, credentialing, pay, and enjoyment.
Not derailing at all; good points. I'm getting to the point where I am feeling good about achieving my personal goals in terms of all of the other elements you listed, so that may be why I have started thinking more about whether or not any factor tilts make sense.

Maybe I should just wait until I am closer to decumulation before I really consider it. I think it's pretty interesting stuff that's enjoyable to read about for now, at least.
Last edited by HootingSloth on Fri Sep 13, 2019 5:45 pm, edited 1 time in total.

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Fri Sep 13, 2019 5:39 pm

HootingSloth wrote:
Fri Sep 13, 2019 5:29 pm
nedsaid wrote:
Fri Sep 13, 2019 4:12 pm
Pretty much, factor funds would be no more than 50% of a stock portfolio, the rest would be in the Broad Indexes. Alternatives, whether the Liquid Alts and/or Interval Funds would be 10% to 20% of a portfolio if you used them at all. Your bond investments would be pretty plain vanilla,

The so-called Larry portfolio is 30% Small Value both US and International and 70% Short Term US Treasuries. An extreme Small/Value tilt with the stocks, the risk taken by the extreme tilt offset by an ultra-conservative allocation of bonds and choice of bonds. I just don't go for extremes and I think it is okay to take a little risk on the bond side. I guess I was born to be mild.

My real-life portfolio has 5-6% in US Small Value, less than 1% in International Small Value, about 1.4% in Liquid Alts, and pretty standard investments for the rest. My own asset allocation is 62% stocks/38% bonds and cash. So my tilts are pretty mild and cautious.
Thanks, nedsaid. This was the kind of thinking I was hoping people would be willing to share. It seems like you know yourself well.
My bark is worse than my bite, I wax eloquently about factor investing but when you look at what I actually do, my tilts are actually pretty mild and cautious. I recently dipped my toe into the Liquid Alts after posting quite a bit about them. I have kept legacy investments, like my individual stocks, because they have done well enough and I pretty much know my investments will zoom right after I sell them. So I am also guarding against myself, and the possibility that I might be wrong. So I have chosen to take more small bets than a few big bets.

Owning a rather eclectic portfolio has given me a lot of investment experience, I know a lot about a lot of things as a result, but not sure all that has added to my results.

A few Buckingham clients have posted their portfolios here and also have followed Paul Merriman's thoughts on these topics as well. They seem to allocate 50% of their equity portfolios to tilts. I have seen a couple of places that an Alt allocation ought to come equally from the equity and fixed income parts of the portfolio. I have read a lot about what Larry Swedroe has said on these topics, so this doesn't just come out of thin air.

So good to know this has helped someone.
A fool and his money are good for business.

User avatar
willthrill81
Posts: 12778
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by willthrill81 » Fri Sep 13, 2019 5:40 pm

nedsaid wrote:
Fri Sep 13, 2019 4:23 pm
To me the biggest risk is the loss of purchasing power of the US Dollar over time. Portfolio volatility is considered but to me is a secondary priority to inflation risk. To me, inflation is public enemy number one.
I certainly agree that inflation is a big risk, but if it's the biggest risk, then why aren't we telling people to load up on TIPS and I bonds when they have a 0% yield or better?

In my view, not having the funds we will need to pay for future liabilities is the biggest risk we face.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Fri Sep 13, 2019 5:44 pm

stlutz wrote:
Fri Sep 13, 2019 4:32 pm
Back in the late 90s when I dropped individual stocks picking and switched to index investing, I thought in terms of what percent of my portfolio I wanted to make active bets with.

I adopted a 75/25 stock bond allocation because that was generally what a "moderately aggressive" investor would do. I next made a pretty arbitrary decision to use 1/3 global stocks. I then fairly arbitrarily decided that I'd put 1/2 of my US in index funds that were buying the types of stocks I was selecting when I was buying individual stocks (i.e. value and smallcap stocks).

This was all before "factor investing" became a big marketing term. But I notice the word "arbitrary" coming up multiple times. "Ability and willingness to take risks" is a phrase with a lot syllables in it, but you're always taking a pretty rough cut at it.

(My investing philosophy has since changed so while I hold onto my small/value funds I'm not adding to them)
I have undergone similar thought processes. For one thing, I was doing a form of factor investing before I knew what factors were, back then they were called investment styles. Second, I did eyeballing and what you call arbitrary decisions. In my case, I looked at the information and made a decision not trying to be too precise about it. I knew it was all well informed and educated guesses to start with, it was a waste of time to make the imprecise precise. A lot of what I called eyeballing and what you called arbitrary. We think a lot alike and that is why I have enjoyed your posts so much.
A fool and his money are good for business.

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Fri Sep 13, 2019 5:48 pm

willthrill81 wrote:
Fri Sep 13, 2019 5:40 pm
nedsaid wrote:
Fri Sep 13, 2019 4:23 pm
To me the biggest risk is the loss of purchasing power of the US Dollar over time. Portfolio volatility is considered but to me is a secondary priority to inflation risk. To me, inflation is public enemy number one.
I certainly agree that inflation is a big risk, but if it's the biggest risk, then why aren't we telling people to load up on TIPS and I bonds when they have a 0% yield or better?

In my view, not having the funds we will need to pay for future liabilities is the biggest risk we face.
Again, another case of my bark being worse than my bite. I definitely own TIPS but probably not in large enough proportions of my fixed income portfolio. I suppose my bonds should be 50% TIPS but I am far short of that. Fortunately, inflation expectations are built into nominal bond prices and inflation has been very subdued, a bond investor would have done well even without TIPS. Pretty much TIPS are to guard against unexpected inflation. I actually am concerned about deflation.
A fool and his money are good for business.

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Fri Sep 13, 2019 5:52 pm

willthrill81 wrote:
Fri Sep 13, 2019 5:31 pm
vineviz wrote:
Fri Sep 13, 2019 3:06 pm
But to be honest I suspect the most important thing that a potential multi-factor investor needs to be aware of about their utility function is how sensitive they are to things like confirmation bias, regret, and/or groupthink. Because constructing a portfolio that looks different from the market is, from a psychological perspective, itself a source of risk for the investor.

If they can resist the forces of regret and groupthink, significant tracking error might be tolerated. If they are susceptible to those forces? Probably not so tolerant. A wise investor will intuitively grasp this about themselves, but experience is the only true test.
:thumbsup

I entirely agree.
Willthrill, have you written anything comparing your choice to follow a trend following strategy with a multi-factor fund type strategy? There seem to be at least some parallels in terms of the goal of these strategies to potentially avoid the worst of a big market drawdown/ underperformance of beta, and the need to have conviction to stay the course with an "alternative" strategy. Just curious.

User avatar
willthrill81
Posts: 12778
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by willthrill81 » Fri Sep 13, 2019 6:07 pm

HootingSloth wrote:
Fri Sep 13, 2019 5:52 pm
willthrill81 wrote:
Fri Sep 13, 2019 5:31 pm
vineviz wrote:
Fri Sep 13, 2019 3:06 pm
But to be honest I suspect the most important thing that a potential multi-factor investor needs to be aware of about their utility function is how sensitive they are to things like confirmation bias, regret, and/or groupthink. Because constructing a portfolio that looks different from the market is, from a psychological perspective, itself a source of risk for the investor.

If they can resist the forces of regret and groupthink, significant tracking error might be tolerated. If they are susceptible to those forces? Probably not so tolerant. A wise investor will intuitively grasp this about themselves, but experience is the only true test.
:thumbsup

I entirely agree.
Willthrill, have you written anything comparing your choice to follow a trend following strategy with a multi-factor fund type strategy? There seem to be at least some parallels in terms of the goal of these strategies to potentially avoid the worst of a big market drawdown/ underperformance of beta, and the need to have conviction to stay the course with an "alternative" strategy. Just curious.
I haven't written anything in-depth about it, but you're right that there certainly are some parallels. I do believe and said many times that if one deviates from owning 'the market' (e.g. TSM, TBM) in any way, whether it be factor investing, trend following, or something other strategy, they should be prepared to underperform the market for lengthy periods of time, potentially a decade or longer. As vineviz pointed out, some investors can tolerate this tracking 'error' (I don't think that it's really an error, but that's the terminology widely used), and some cannot.

I do believe that both factor investing and trend following may help to reduce an investor's downside risk. For instance, SCV performed very well during the 1970s and 2000s when TSM went nowhere (at least in real dollars). And while many here have derided SCV of late because it's underperformed TSM over the last decade, its actual returns have been still been very good during this period. Trend following has had somewhat similar performance in that it has tended to lag the market when the market is doing well but far outperforms the market when the market is doing poorly. The potential for both of these strategies to 'smooth out' stock returns is undervalued by far too many investors, IMHO, especially retirees who can be greatly exposed to sequence of returns risk or its twin, sequence of income risk. I would certainly prefer to 'trade' some of the returns from the very good times into the very bad times if I could, especially if doing so doesn't erode long-term returns, which fixed income has tended to do and is expected to continue to do going forward.

Circling back to the OP, I don't think that we can objectively say that "ability" has much to do with factor investing (or trend following) unless, like many seem to do here, you assume that instruments like total market funds will enable you to achieve your investment goals. I do not believe that assumption to have any basis in reality, but many here seem to believe that factor investing is inherently riskier, which I believe the data strongly suggest is not the case.

However, I do believe that "willingness" is an absolute must for an investor contemplating factor investing (or trend following), just as it is with any investment strategy. Sticking with one's chosen strategy is very likely to be more important than which strategy they adopt.
Last edited by willthrill81 on Fri Sep 13, 2019 6:12 pm, edited 1 time in total.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Fri Sep 13, 2019 6:10 pm

Thanks, that also seems consistent with Horton's comment that these kinds of strategies may make the most sense when you are approaching and at the start of decumulation.

User avatar
Taylor Larimore
Advisory Board
Posts: 28594
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Factor Investing?

Post by Taylor Larimore » Fri Sep 13, 2019 6:19 pm

If you think factor investing is bunk, you can safely ignore the post.
HootingSloth:

I think "factor investing" is mostly a marketing scheme by the investment industry. You can read more here:

Factor Investing

Best wishes.
Taylor
Jack Bogle's Words of Wisdom: "By and large I do not approve of factor funds."
"Simplicity is the master key to financial success." -- Jack Bogle

AHTFY
Posts: 55
Joined: Mon Jul 23, 2018 2:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by AHTFY » Fri Sep 13, 2019 6:59 pm

All this talk of factor investing and being prepared to possible long periods of underperformance reminds me of a famous quote:

"The market can remain irrational longer than you can remain solvent."

User avatar
willthrill81
Posts: 12778
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by willthrill81 » Fri Sep 13, 2019 7:45 pm

AHTFY wrote:
Fri Sep 13, 2019 6:59 pm
All this talk of factor investing and being prepared to possible long periods of underperformance reminds me of a famous quote:

"The market can remain irrational longer than you can remain solvent."
Except that statement doesn't apply here at all. The underperformance I referred to earlier was still generating double-digit positive returns. On the contrary, factor investing may be helpful in keeping you solvent.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

stlutz
Posts: 5348
Joined: Fri Jan 02, 2009 1:08 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by stlutz » Fri Sep 13, 2019 8:58 pm

vineviz wrote:
Fri Sep 13, 2019 3:06 pm

But to be honest I suspect the most important thing that a potential multi-factor investor needs to be aware of about their utility function is how sensitive they are to things like confirmation bias, regret, and/or groupthink. Because constructing a portfolio that looks different from the market is, from a psychological perspective, itself a source of risk for the investor.

If they can resist the forces of regret and groupthink, significant tracking error might be tolerated. If they are susceptible to those forces? Probably not so tolerant. A wise investor will intuitively grasp this about themselves, but experience is the only true test.
I kind of disagree with this. Everybody is always underperforming *something*. If stocks are flat and gold is up 50%, equity investors experience regret. If stocks are up and bonds are down, then one wonders why they hold any bonds. And so forth.

Unlike professional investors, individual investors don't have to benchmark against one particular index. The problem this creates for individuals is that we benchmark against what we are aware we could be making if we only had a better portfolio. This problem is often talked about as if it only impacts factor tilters. In reality, it impacts everyone.

User avatar
willthrill81
Posts: 12778
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by willthrill81 » Fri Sep 13, 2019 9:22 pm

stlutz wrote:
Fri Sep 13, 2019 8:58 pm
This problem is often talked about as if it only impacts factor tilters. In reality, it impacts everyone.
That's why it's important for everyone to stick with their strategy, whatever that strategy may be*.

*Assuming of course that it's at least halfway reasonable. Sticking with a strategy of all palladium futures probably doesn't meet that criterion.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
abuss368
Posts: 15132
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by abuss368 » Sat Sep 14, 2019 7:16 am

AHTFY wrote:
Fri Sep 13, 2019 6:59 pm
All this talk of factor investing and being prepared to possible long periods of underperformance reminds me of a famous quote:

"The market can remain irrational longer than you can remain solvent."
An excellent reminder that applies in any investment environment or situation.
John C. Bogle: "Simplicity is the master key to financial success."

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Sat Sep 14, 2019 3:50 pm

With all due respect to Taylor, John Bogle was one of the early providers of factor funds. The Value Index and Growth Index funds as well as the Small Value Index and Small Growth Index funds have been around for a long time. Also, Vanguard offers factor based ETFs: Low Volatility, Value, Momentum, Liquidity, Quality, and Multi-Factor. I have read good comments about the Vanguard factor products including from Larry Swedroe. I have been an investor in the Vanguard Small Value Index ETF for over a decade now.
A fool and his money are good for business.

User avatar
willthrill81
Posts: 12778
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by willthrill81 » Sat Sep 14, 2019 5:01 pm

nedsaid wrote:
Sat Sep 14, 2019 3:50 pm
With all due respect to Taylor, John Bogle was one of the early providers of factor funds. The Value Index and Growth Index funds as well as the Small Value Index and Small Growth Index funds have been around for a long time. Also, Vanguard offers factor based ETFs: Low Volatility, Value, Momentum, Liquidity, Quality, and Multi-Factor. I have read good comments about the Vanguard factor products including from Larry Swedroe. I have been an investor in the Vanguard Small Value Index ETF for over a decade now.
While many posit that Bogle was absolutely opposed to factor investing, active investing, market timing, and owning gold, there were times and/or conditions where he was fine with each. What he was most consistent about was his espousal of keeping costs low.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
abuss368
Posts: 15132
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by abuss368 » Sat Sep 14, 2019 5:05 pm

nedsaid wrote:
Sat Sep 14, 2019 3:50 pm
With all due respect to Taylor, John Bogle was one of the early providers of factor funds. The Value Index and Growth Index funds as well as the Small Value Index and Small Growth Index funds have been around for a long time. Also, Vanguard offers factor based ETFs: Low Volatility, Value, Momentum, Liquidity, Quality, and Multi-Factor. I have read good comments about the Vanguard factor products including from Larry Swedroe. I have been an investor in the Vanguard Small Value Index ETF for over a decade now.
One has to note that Mr. Bogle was also building a business and a brand. If Vanguard, in it's present form, did not offer these funds, the AUM would be much less.
John C. Bogle: "Simplicity is the master key to financial success."

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Sat Sep 14, 2019 5:29 pm

willthrill81 wrote:
Sat Sep 14, 2019 5:01 pm
nedsaid wrote:
Sat Sep 14, 2019 3:50 pm
With all due respect to Taylor, John Bogle was one of the early providers of factor funds. The Value Index and Growth Index funds as well as the Small Value Index and Small Growth Index funds have been around for a long time. Also, Vanguard offers factor based ETFs: Low Volatility, Value, Momentum, Liquidity, Quality, and Multi-Factor. I have read good comments about the Vanguard factor products including from Larry Swedroe. I have been an investor in the Vanguard Small Value Index ETF for over a decade now.
While many posit that Bogle was absolutely opposed to factor investing, active investing, market timing, and owning gold, there were times and/or conditions where he was fine with each. What he was most consistent about was his espousal of keeping costs low.
Bogle envisioned that people would invest in the Growth Index while working and switch to the Value Index when retired. He thought that retirees would like the higher dividend yield from the Value Index. It is fair to say that Bogle was not a proponent of factors, I suspect his thinking on this topic was more flexible than portrayed.
A fool and his money are good for business.

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Sat Sep 14, 2019 5:30 pm

abuss368 wrote:
Sat Sep 14, 2019 5:05 pm
nedsaid wrote:
Sat Sep 14, 2019 3:50 pm
With all due respect to Taylor, John Bogle was one of the early providers of factor funds. The Value Index and Growth Index funds as well as the Small Value Index and Small Growth Index funds have been around for a long time. Also, Vanguard offers factor based ETFs: Low Volatility, Value, Momentum, Liquidity, Quality, and Multi-Factor. I have read good comments about the Vanguard factor products including from Larry Swedroe. I have been an investor in the Vanguard Small Value Index ETF for over a decade now.
One has to note that Mr. Bogle was also building a business and a brand. If Vanguard, in it's present form, did not offer these funds, the AUM would be much less.
Mr. Bogle did recognize the realities of meeting investor demand, folks were probably asking for them. He also wasn't immune to marketing, he wanted his company to grow.
A fool and his money are good for business.

User avatar
vineviz
Posts: 5129
Joined: Tue May 15, 2018 1:55 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by vineviz » Sat Sep 14, 2019 5:32 pm

nedsaid wrote:
Sat Sep 14, 2019 5:29 pm
It is fair to say that Bogle was not a proponent of factors, I suspect his thinking on this topic was more flexible than portrayed.
I think it’d be more fair to say that Bogle, like many Bogleheads who have followed him, was absolutely okay with factor investing just as long as it wasn’t called factor investing.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
abuss368
Posts: 15132
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by abuss368 » Sat Sep 14, 2019 5:34 pm

nedsaid wrote:
Sat Sep 14, 2019 5:30 pm
abuss368 wrote:
Sat Sep 14, 2019 5:05 pm
nedsaid wrote:
Sat Sep 14, 2019 3:50 pm
With all due respect to Taylor, John Bogle was one of the early providers of factor funds. The Value Index and Growth Index funds as well as the Small Value Index and Small Growth Index funds have been around for a long time. Also, Vanguard offers factor based ETFs: Low Volatility, Value, Momentum, Liquidity, Quality, and Multi-Factor. I have read good comments about the Vanguard factor products including from Larry Swedroe. I have been an investor in the Vanguard Small Value Index ETF for over a decade now.
One has to note that Mr. Bogle was also building a business and a brand. If Vanguard, in it's present form, did not offer these funds, the AUM would be much less.
Mr. Bogle did recognize the realities of meeting investor demand, folks were probably asking for them. He also wasn't immune to marketing, he wanted his company to grow.
Agreed. Always love reading articles or books on the founding and building of Vanguard. “The Vanguard Experiment” is an excellent book.
John C. Bogle: "Simplicity is the master key to financial success."

XacTactX
Posts: 11
Joined: Sat May 11, 2019 5:46 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by XacTactX » Sat Sep 14, 2019 5:45 pm

HootingSloth wrote:
Fri Sep 13, 2019 1:34 pm
If you cannot clearly understand the nature of the risks associated with a factor, how can you decide whether you want more or less exposure to these risks than the market portfolio provides? How do all of you factor investors out there decide how deeply to tilt? How do you decide that you want more SMB or HML risk than the market rather than less? What is the process you use that is the analogue of the exercises we all go through to find the right stock-bond allocation?
These are excellent questions and I think someone with a formal background in personal finance/investments might be able to give a textbook answer, but sadly I'm not knowledgeable enough to do this. Instead I can tell you what factor tilts I target and why.
  • Factors are the underlying drivers of risk and return in capital markets. Over very long periods of time (20+ years) a portfolio with positive factor loadings should outperform one that has negative loadings or is neutral.
  • Over relatively short periods (1-10 years) factors can have negative performance. In order to successfully use the factors, one has to stick with the strategy during these negative periods. Therefore, one must have knowledge and conviction to follow the strategy for the long haul.
  • According to Your Complete Guide to Factor-Based Investing by Swedroe/Berkin the legitimate equity factors are size, value, momentum, and profitability/quality.
  • According to Reducing the Risk of Black Swans by Swedroe/Grogan having unique sources of risk and return in a portfolio can be used to either 1. improve absolute performance or 2. improve risk-adjusted performance. Which one depends on the amount of Treasuries/total risk in the portfolio.
  • Having exposure to multiple factors with a bottom-up index is more effective than targeting single factors or using a top-down strategy. Targeting single factors leads to more negative tracking error, and top-down strategies water down their factor exposure as the number of factors increases. To be fair the bottom-up strategy has worse transparency and greater complexity than the other approaches, no index methodology is perfect.
  • Having said all of this, my personal goal is to get as much factor exposure as possible while avoiding uncompensated risk like stock-picking risk or sector risk. I also pay attention to practical issues that can damage factor performance, like high turnover, management fees, etc.
  • Over the long run (20+ years) I think this factor exposure might net me 2-3% in added returns over a TSM portfolio, but of course there is no guarantee. But I've never considered how this factor exposure will change the standard deviation of my portfolio. This is probably a weak link in my approach and I should start thinking critically about it.
Holdings: SMLF, ISCF, EMGF, LendingClub, HYSA

User avatar
nedsaid
Posts: 12417
Joined: Fri Nov 23, 2012 12:33 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by nedsaid » Sat Sep 14, 2019 6:17 pm

vineviz wrote:
Sat Sep 14, 2019 5:32 pm
nedsaid wrote:
Sat Sep 14, 2019 5:29 pm
It is fair to say that Bogle was not a proponent of factors, I suspect his thinking on this topic was more flexible than portrayed.
I think it’d be more fair to say that Bogle, like many Bogleheads who have followed him, was absolutely okay with factor investing just as long as it wasn’t called factor investing.
He did love that Wellington Fund which right now is 65% stocks/35% bonds and definitely has a Value tilt. But you are right, we can't CALL it a Value fund. :wink: I suppose we will have to talk in code around here.
A fool and his money are good for business.

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Sun Sep 15, 2019 8:17 am

Thanks for the thoughtful comments, XacTactX.

I get the sense from reading Larry Swedroe's writings that he emphatically disagrees with Fama's belief that for every investor that ought to rationally invest on one side of a factor there should be another investor that ought to rationally take the other side. He seems, instead, to view both risk-based and behavior-based explanations as important to factor returns. For example, I have seen him say on several occasions that small growth (what he sometimes calls "lotto") stocks are basically bad investments for everyone because their prices are irrationally bid up by investors who are just looking to hit home runs.

I have made a personal decision not to base my portfolio on a belief that I am smarter, wiser, more rational, less sensitive to groupthink, or more virtuous in whatever other way than the marginal investor, so behavior-based explanations of factor returns are not personally compelling to me. I understand others disagree with that approach, and I have no quarrel with that if it works for them. However, because of Larry's emphasis on behavior-based explanations, I have not found his writings to be particularly helpful in making decisions for my own portfolio.

dbr
Posts: 30137
Joined: Sun Mar 04, 2007 9:50 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by dbr » Sun Sep 15, 2019 8:30 am

XacTactX wrote:
Sat Sep 14, 2019 5:45 pm
HootingSloth wrote:
Fri Sep 13, 2019 1:34 pm
If you cannot clearly understand the nature of the risks associated with a factor, how can you decide whether you want more or less exposure to these risks than the market portfolio provides? How do all of you factor investors out there decide how deeply to tilt? How do you decide that you want more SMB or HML risk than the market rather than less? What is the process you use that is the analogue of the exercises we all go through to find the right stock-bond allocation?
These are excellent questions and I think someone with a formal background in personal finance/investments might be able to give a textbook answer, but sadly I'm not knowledgeable enough to do this. Instead I can tell you what factor tilts I target and why.
  • Factors are the underlying drivers of risk and return in capital markets. Over very long periods of time (20+ years) a portfolio with positive factor loadings should outperform one that has negative loadings or is neutral.
  • Over relatively short periods (1-10 years) factors can have negative performance. In order to successfully use the factors, one has to stick with the strategy during these negative periods. Therefore, one must have knowledge and conviction to follow the strategy for the long haul.
  • According to Your Complete Guide to Factor-Based Investing by Swedroe/Berkin the legitimate equity factors are size, value, momentum, and profitability/quality.
  • According to Reducing the Risk of Black Swans by Swedroe/Grogan having unique sources of risk and return in a portfolio can be used to either 1. improve absolute performance or 2. improve risk-adjusted performance. Which one depends on the amount of Treasuries/total risk in the portfolio.
  • Having exposure to multiple factors with a bottom-up index is more effective than targeting single factors or using a top-down strategy. Targeting single factors leads to more negative tracking error, and top-down strategies water down their factor exposure as the number of factors increases. To be fair the bottom-up strategy has worse transparency and greater complexity than the other approaches, no index methodology is perfect.
  • Having said all of this, my personal goal is to get as much factor exposure as possible while avoiding uncompensated risk like stock-picking risk or sector risk. I also pay attention to practical issues that can damage factor performance, like high turnover, management fees, etc.
  • Over the long run (20+ years) I think this factor exposure might net me 2-3% in added returns over a TSM portfolio, but of course there is no guarantee. But I've never considered how this factor exposure will change the standard deviation of my portfolio. This is probably a weak link in my approach and I should start thinking critically about it.
I pretty much like this description. At the end of the day risk in the context of asset allocation is still volatility (perhaps measured by SD of annual returns) and the answer is that after looking at the risk and return of your factor allocated portfolio you decide on how much return you need and how much volatility you have the ability and willingness to accept. Return and volatility dictate many other definitions of risk such as max drawdown, risk of running out of money in retirement, ability to predict end point wealth for a period, and so on.

If someone can specify and quantify exactly what hazard to a person's financial situation is associated with loading on one factor or another, then the conversation can go on from there.

Random Walker
Posts: 4106
Joined: Fri Feb 23, 2007 8:21 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by Random Walker » Sun Sep 15, 2019 8:58 am

psteinx wrote:
Fri Sep 13, 2019 2:34 pm
But whereas it's fairly easy for 3 fund proponents to identify those who should be over or under exposed to beta, based on their personal situation (and generally using bonds to fill out the lower-risk portion of their portfolio, as you've mentioned), it's MUCH harder to identify those who both:

1) Believe in factor models (either SV, or more complex models)
and
2) Who themselves have UNDERWEIGHTED SV or the other factors, attempting to mitigate their overall risk (and/or recommend this commonly to others).
There is good reason for this. You are saying that we should be able to identify people who want a less risky portfolio. If the person is a factor believer (he believes the factors are unique and independent), he will not decrease risk by decreasing exposure to the factors. He will continue to heavily tilt the equity portion of his portfolio and increase his exposure to safe bonds. If he believes in factors, at any risk tolerance, he will perceive potential increased portfolio efficiency by diversifying across factors on the equity side.

For example one could argue that an investor who wants to play it safer will tilt towards growth. But a true factor believer would more rationally continue the SV tilt and increase the bond exposure for same expected return portfolio.

Dave

dbr
Posts: 30137
Joined: Sun Mar 04, 2007 9:50 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by dbr » Sun Sep 15, 2019 9:09 am

Random Walker wrote:
Sun Sep 15, 2019 8:58 am


For example one could argue that an investor who wants to play it safer will tilt towards growth. But a true factor believer would more rationally continue the SV tilt and increase the bond exposure for same expected return portfolio.

Dave
Exactly so, assuming he can establish that he should thereby expect similar return at less risk. I don't have a personal position on whether or not this works out the way you would want it to, but would not be surprised if it does work out nor surprised if it doesn't. The result is also highly variable across different periods. I don't doubt there is data where it has worked out. This is, of course, the famous "Larry" portfolio which is offered in an attempt to reduce left tail risk, which also might be a portfolio with a lower risk of running out of money in retirement, as nuances of risk go.

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Sun Sep 15, 2019 9:10 am

dbr wrote:
Sun Sep 15, 2019 8:30 am
I pretty much like this description. At the end of the day risk in the context of asset allocation is still volatility (perhaps measured by SD of annual returns) and the answer is that after looking at the risk and return of your factor allocated portfolio you decide on how much return you need and how much volatility you have the ability and willingness to accept. Return and volatility dictate many other definitions of risk such as max drawdown, risk of running out of money in retirement, ability to predict end point wealth for a period, and so on.

If someone can specify and quantify exactly what hazard to a person's financial situation is associated with loading on one factor or another, then the conversation can go on from there.
dbr, I think that the concept behind factor-based asset pricing models is that there are sources of risk that the market will compensate for with higher expected returns *other than volatility*. I think the question is how much exposure you want to these other risks.

Let's say that, for the sake of argument, the risk associated with the value factor was additional sensitivity to deteriorating credit conditions, such as those that occur during a financial crisis. This is one explanation that is sometimes given (because value stocks typically have high leverage). If, for example, your job (and hence salary) was in an industry that is particularly exposed to deteriorating credit conditions (I suppose many people in finance and probably others), then you would want to tilt towards growth stocks. If, instead, your financial situation were more insulated from the risks of a financial crisis than most investors, then you would want to tilt towards value. After taking these tilts into account you could still adjust your stock-bond allocation to get whatever expected return-volatility trade-off you wanted.

At least, I think this is the theory of people like Fama who believe in risk-based explanations of factor returns. I think, as you say, the difficulty seems to be in specifying and quantifying, at least roughly and getting the direction right--I don't think we need to insist on being exact--what hazard to a person's financial situation is associated with loading on one factor or another.

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Sun Sep 15, 2019 9:15 am

Random Walker wrote:
Sun Sep 15, 2019 8:58 am
psteinx wrote:
Fri Sep 13, 2019 2:34 pm
But whereas it's fairly easy for 3 fund proponents to identify those who should be over or under exposed to beta, based on their personal situation (and generally using bonds to fill out the lower-risk portion of their portfolio, as you've mentioned), it's MUCH harder to identify those who both:

1) Believe in factor models (either SV, or more complex models)
and
2) Who themselves have UNDERWEIGHTED SV or the other factors, attempting to mitigate their overall risk (and/or recommend this commonly to others).
There is good reason for this. You are saying that we should be able to identify people who want a less risky portfolio. If the person is a factor believer (he believes the factors are unique and independent), he will not decrease risk by decreasing exposure to the factors. He will continue to heavily tilt the equity portion of his portfolio and increase his exposure to safe bonds. If he believes in factors, at any risk tolerance, he will perceive potential increased portfolio efficiency by diversifying across factors on the equity side.

For example one could argue that an investor who wants to play it safer will tilt towards growth. But a true factor believer would more rationally continue the SV tilt and increase the bond exposure for same expected return portfolio.

Dave
Dave, I think that your conclusion is only correct if a "true factor believer" means someone who believes behavior-based explanations of factor returns. You also seem to be expressing a view where there is one relevant dimension of portfolio risk--volatility--when you are referring to a "less risky portfolio." But I may be misunderstanding.

dbr
Posts: 30137
Joined: Sun Mar 04, 2007 9:50 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by dbr » Sun Sep 15, 2019 9:21 am

HootingSloth wrote:
Sun Sep 15, 2019 9:10 am

dbr, I think that the concept behind factor-based asset pricing models is that there are sources of risk that the market will compensate for with higher expected returns *other than volatility*. I think the question is how much exposure you want to these other risks.
Yes. The Fama-French model is about exactly that. But the question is what do these other risks mean to the objectives of the investor in any kind of a quantifiable way, including even what will the volatility of the resulting portfolio be.

We can't answer the question in the OP if we don't know how to relate the risks to our objectives in order to determine need, ability, and willingness. The only numbers I know of that we can obtain are return and variability of return (measured in the different varieties of ways we might imagine).

Random Walker
Posts: 4106
Joined: Fri Feb 23, 2007 8:21 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by Random Walker » Sun Sep 15, 2019 9:32 am

dbr wrote:
Sun Sep 15, 2019 9:09 am
Random Walker wrote:
Sun Sep 15, 2019 8:58 am


For example one could argue that an investor who wants to play it safer will tilt towards growth. But a true factor believer would more rationally continue the SV tilt and increase the bond exposure for same expected return portfolio.

Dave
Exactly so, assuming he can establish that he should thereby expect similar return at less risk. I don't have a personal position on whether or not this works out the way you would want it to, but would not be surprised if it does work out nor surprised if it doesn't. The result is also highly variable across different periods. I don't doubt there is data where it has worked out. This is, of course, the famous "Larry" portfolio which is offered in an attempt to reduce left tail risk, which also might be a portfolio with a lower risk of running out of money in retirement, as nuances of risk go.
I think the table showing 1/n portfolios in Chapter 9 of Larry’s factor book say it all. Any factor, including the market factor, can underperform for long periods. But diversifying across the factors minimizes the likelihood of negative outcome over all time periods. Seems most significant for investors with shorter time periods / most subject to sequence of returns risk. No guarantees.

Dave

dbr
Posts: 30137
Joined: Sun Mar 04, 2007 9:50 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by dbr » Sun Sep 15, 2019 9:38 am

Random Walker wrote:
Sun Sep 15, 2019 9:32 am
dbr wrote:
Sun Sep 15, 2019 9:09 am
Random Walker wrote:
Sun Sep 15, 2019 8:58 am


For example one could argue that an investor who wants to play it safer will tilt towards growth. But a true factor believer would more rationally continue the SV tilt and increase the bond exposure for same expected return portfolio.

Dave
Exactly so, assuming he can establish that he should thereby expect similar return at less risk. I don't have a personal position on whether or not this works out the way you would want it to, but would not be surprised if it does work out nor surprised if it doesn't. The result is also highly variable across different periods. I don't doubt there is data where it has worked out. This is, of course, the famous "Larry" portfolio which is offered in an attempt to reduce left tail risk, which also might be a portfolio with a lower risk of running out of money in retirement, as nuances of risk go.
I think the table showing 1/n portfolios in Chapter 9 of Larry’s factor book say it all. Any factor, including the market factor, can underperform for long periods. But diversifying across the factors minimizes the likelihood of negative outcome over all time periods. Seems most significant for investors with shorter time periods / most subject to sequence of returns risk. No guarantees.

Dave
Curiously nowhere in this thread on SoR does the word "Larry" appear: viewtopic.php?f=2&t=289026&start=150

Random Walker
Posts: 4106
Joined: Fri Feb 23, 2007 8:21 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by Random Walker » Sun Sep 15, 2019 9:43 am

HootingSloth wrote:
Sun Sep 15, 2019 9:15 am
Dave, I think that your conclusion is only correct if a "true factor believer" means someone who believes behavior-based explanations of factor returns. You also seem to be expressing a view where there is one relevant dimension of portfolio risk--volatility--when you are referring to a "less risky portfolio." But I may be misunderstanding.
Speaking only for myself, no. I believe size is risk story, value is risk and behavioral, momentum is behavioral, and not sure on profitability. I appreciate there are many types of risk: skew, kurtosis, doing badly in bad times. I think the different dimensions of risk only reinforce the concept of diversifying across as many sources of risk / return as possible. That being said, I also think we can get pretty far in discussions by reducing risk to standard deviation and thinking in terms of narrowing distribution, cutting tails, moving to northwest of efficient frontier plot.

Dave

dbr
Posts: 30137
Joined: Sun Mar 04, 2007 9:50 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by dbr » Sun Sep 15, 2019 9:52 am

Random Walker wrote:
Sun Sep 15, 2019 9:43 am
HootingSloth wrote:
Sun Sep 15, 2019 9:15 am
Dave, I think that your conclusion is only correct if a "true factor believer" means someone who believes behavior-based explanations of factor returns. You also seem to be expressing a view where there is one relevant dimension of portfolio risk--volatility--when you are referring to a "less risky portfolio." But I may be misunderstanding.
Speaking only for myself, no. I believe size is risk story, value is risk and behavioral, momentum is behavioral, and not sure on profitability. I appreciate there are many types of risk: skew, kurtosis, doing badly in bad times. I think the different dimensions of risk only reinforce the concept of diversifying across as many sources of risk / return as possible. That being said, I also think we can get pretty far in discussions by reducing risk to standard deviation and thinking in terms of narrowing distribution, cutting tails, moving to northwest of efficient frontier plot.

Dave
Right, with some emphasis that a "risk story" does not contain anything about why an investor would take the risk or can assess need, ability, and willingness to take the risk until that risk story can be related to what the investor wants. The conversation of what is the distribution of annual returns is one we do relate to what does the investor want, and we do that for factor loaded portfolios all the time.

Topic Author
HootingSloth
Posts: 24
Joined: Mon Jan 28, 2019 3:38 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by HootingSloth » Sun Sep 15, 2019 9:56 am

Random Walker wrote:
Sun Sep 15, 2019 9:43 am
HootingSloth wrote:
Sun Sep 15, 2019 9:15 am
Dave, I think that your conclusion is only correct if a "true factor believer" means someone who believes behavior-based explanations of factor returns. You also seem to be expressing a view where there is one relevant dimension of portfolio risk--volatility--when you are referring to a "less risky portfolio." But I may be misunderstanding.
Speaking only for myself, no. I believe size is risk story, value is risk and behavioral, momentum is behavioral, and not sure on profitability. I appreciate there are many types of risk: skew, kurtosis, doing badly in bad times. I think the different dimensions of risk only reinforce the concept of diversifying across as many sources of risk / return as possible. That being said, I also think we can get pretty far in discussions by reducing risk to standard deviation and thinking in terms of narrowing distribution, cutting tails, moving to northwest of efficient frontier plot.

Dave
Thanks for the clarification. Would you mind sharing a bit about how you have figured out how deeply to tilt your portfolio towards different factors and whether/how thinking in terms of those general effects on your distribution of returns has influenced those decisions? Is it more or less a thought process along the lines of: "This combination of factor loadings has a very good Sharpe ratio historically, looks like it cuts down on the left tail of returns, and can be implemented with reasonable cost"?

Random Walker
Posts: 4106
Joined: Fri Feb 23, 2007 8:21 pm

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by Random Walker » Sun Sep 15, 2019 10:45 am

HootingSloth wrote:
Sun Sep 15, 2019 9:56 am
Random Walker wrote:
Sun Sep 15, 2019 9:43 am
HootingSloth wrote:
Sun Sep 15, 2019 9:15 am
Dave, I think that your conclusion is only correct if a "true factor believer" means someone who believes behavior-based explanations of factor returns. You also seem to be expressing a view where there is one relevant dimension of portfolio risk--volatility--when you are referring to a "less risky portfolio." But I may be misunderstanding.
Speaking only for myself, no. I believe size is risk story, value is risk and behavioral, momentum is behavioral, and not sure on profitability. I appreciate there are many types of risk: skew, kurtosis, doing badly in bad times. I think the different dimensions of risk only reinforce the concept of diversifying across as many sources of risk / return as possible. That being said, I also think we can get pretty far in discussions by reducing risk to standard deviation and thinking in terms of narrowing distribution, cutting tails, moving to northwest of efficient frontier plot.

Dave
Thanks for the clarification. Would you mind sharing a bit about how you have figured out how deeply to tilt your portfolio towards different factors and whether/how thinking in terms of those general effects on your distribution of returns has influenced those decisions? Is it more or less a thought process along the lines of: "This combination of factor loadings has a very good Sharpe ratio historically, looks like it cuts down on the left tail of returns, and can be implemented with reasonable cost"?
First of all, I’m subject to the same behavioral biases that any of us are. Secondly, I appreciate that the long only funds with even the most potent SV exposure still have a market beta of about 1. Thirdly, I’m just about idealistically a believer in efficient markets. Fourthly, I’m 56 years old and am a salaried employee, so I somewhat fear being forced into retirement sooner than I’d like. The more I have learned, the more I have concerned myself with sequence of returns risk, especially for someone in my position. So putting all the above together, I’ve evolved to taking on the Larry approach big time: big tilt to factors on the equity side, overall very substantial decrease in equity allocation, and have added alternatives. My current allocation is 40% equities / 36% bonds / 24% Alternatives. The equities are equally split US/Int and heavily tilted to size and value.

I started out as a hardcore TSM Boglehead very proud of my overall portfolio expense ratio. Over time I came to appreciate that lack of portfolio efficiency is a cost and I became cognizant of the potential costs of behavioral errors. So I’ve adopted this Larry Swedroe “hyperdiversification” approach. I could be wrong; the increased costs are certain.

When I think of risk, now I think of losing money. I think about high equity valuations, lower mean future expected returns, the whole potential distribution of future returns shifted left. Good outcomes will be more modest and bad outcomes worse. I think about a 50% market decline needing 100% return to break even, 33% market decline needing 50% return to break even, 25% market decline needing 33% to break even. I think about the findings of behavioral finance that the pain of a loss is twice as much as the happiness derived from an equal sized gain, and that ratio rises dramatically with net worth! So I have taken on a big risk of sorts. I have actively chosen to pay significantly higher fees for the sake of minimizing the potential costs of decreased portfolio efficiency and behavioral errors.

The decreased likelihood of underperformance over any timeframe represented by the 1/n portfolios in Chapter 9 of Larry’s Factor book have been a big influence. The concept of narrowing the SD and cutting tails from Larry’s Black Swan give another view of the same concept. Importantly, when equities do poorly, the correlation between equities and bonds tends to turn sharply negative. Thus the highly tilted equity / increased bond allocation portfolio tends to cut the bad left tail significantly more than the good right tail.

Dave

dbr
Posts: 30137
Joined: Sun Mar 04, 2007 9:50 am

Re: How do You Evaluate Ability and Willingness to Take Risks Associated with Factor Investing

Post by dbr » Sun Sep 15, 2019 10:55 am

HootingSloth wrote:
Sun Sep 15, 2019 9:56 am

Thanks for the clarification. Would you mind sharing a bit about how you have figured out how deeply to tilt your portfolio towards different factors and whether/how thinking in terms of those general effects on your distribution of returns has influenced those decisions? Is it more or less a thought process along the lines of: "This combination of factor loadings has a very good Sharpe ratio historically, looks like it cuts down on the left tail of returns, and can be implemented with reasonable cost"?
My answer is that I don't tilt my portfolio at all. In other words the risk and return properties of my mostly total market portfolio without addressing any factors seem to be sufficient to meet my needs and within my ability and willingness to take risk, as I see it.

I have no idea how many people need to diversify their portfolios to various factors to meet their needs and stay within ability and willingness.

Post Reply