Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Random Walker
Posts: 3961
Joined: Fri Feb 23, 2007 8:21 pm

Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by Random Walker » Wed May 22, 2019 4:49 pm

https://alphaarchitect.com/2019/05/22/v ... d-returns/

In this article for Alpha Architect Larry reviews a couple of studies that show benefits to volatility scaling: increasing leverage when volatility is low and decreasing leverage when volatility is higher. Volatility tends to cluster, and the best predictor of volatility in the current period is the volatility in the prior period. This is true across multiple asset classes. Volatility scaling improves Sharpe ratios, decreases kurtosis, improves skew, decreases maximum drawdowns, improves compounded returns for both individual assets and portfolios of assets.
Volatility scaling is used by some of the alternatives Larry recommends.

Dave

User avatar
vineviz
Posts: 4411
Joined: Tue May 15, 2018 1:55 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by vineviz » Wed May 22, 2019 6:03 pm

Random Walker wrote:
Wed May 22, 2019 4:49 pm
https://alphaarchitect.com/2019/05/22/v ... d-returns/

In this article for Alpha Architect Larry reviews a couple of studies that show benefits to volatility scaling: increasing leverage when volatility is low and decreasing leverage when volatility is higher.
Larry was primarily writing about managed funds, but even DIY investors (disciplined ones, anyway) can use "target volatility" in place of the more conventional target allocation approach and can do so with plain vanilla Vanguard mutual funds and without the use of leverage. Such a strategy can be INCREDIBLY helpful for retirees who are operating near the limit of conventional SWRs.

Basically, instead of setting a target asset allocation for their (e.g. 60% stocks & 40% bonds) they set a target amount of volatility based on their risk tolerance. This has been discussed here before, as in this thread.

For instance, Vanguard LifeStrategy Moderate Growth Fund (VSMGX) has a 60/40 allocation and over its life has averaged a annualized standard deviation of about 9.5%.

A DIY investor could easily set a target portfolio of 9.5% using a simple combination of two Vanguard mutual funds. For instance, let's say they chose Vanguard LifeStrategy Growth Fund (VASGX) as their "risky" asset and Vanguard Total Bond Market Index Fund (VBMFX) as their safe" asset. They can simply adjust the proportion of those two funds to keep their expected volatility at 9.5%, which is the same risk they'd be taking with a buy-and-hold purchase of VSMGX.

If such an investor started out 1995 with $1,000,000 invested in VSMGX and withdrew $5,500/month the first year and then adjusted the withdrawal for inflation annually they'd still have a balance of $1,292,402. (This is the orange line in the following graph)

Let's say a different investor also started out 1995 with $1,000,000 but invested in VASGX and altered that allocation by adding and subtracting VBMFX to keep close to their 9.5% volatility target. With the same $5,500/month inflation-adjusted withdrawal they'd still have a balance of $1,986,601, or over 50% more. (This is the blue line in the following graph)

Image

As you can see in the table below, the target volatility model (aka "timing model") wasn't perfect: the investor ended up with only about 8.41% volatility instead their targeted 9.5%, or about 100 bps less than VSMGX. In part because of this lower volatility, they also ended up with returns that were about 100 bps higher than VSMGX.

Image

The curious Bogleheads can take a look at the metrics tab in the backtest to see that the target volatility approach in this example did achieve better performance on all of the first four statistical moments to which Larry alluded in his article: higher mean, lower variance, more positive (well, less negative) skew, and less kurtosis.

All without the use of actual leverage and with a simple dedication to monthly rebalancing.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Thu Jun 20, 2019 11:24 pm

I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.

Here's an example of how this can be effective over a long-term period, 1987-2018. Maintaining vineviz's targeted volatility of 9.5% and using VFINX (S&P 500) and VUSTX (Long-term Treasuries), the portfolio's (S&P 500's) CAGR was 11.17% (9.87%), std. dev. was 9.61% (14.79%), worst year was -6.03% (-37.02%), and maximum drawdown was -16.95% (-50.97%). That's not a trivial improvement; it's 1.3% higher returns than the S&P 500 but with the downside risk (as measured by max. drawdown) of a 35/65 fixed AA.
“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: 11857
Joined: Fri Nov 23, 2012 12:33 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by nedsaid » Fri Jun 21, 2019 10:35 pm

Question, are there real life Bogleheads who are successfully implementing this strategy? If you are one of them, could you post and tell us what you are doing and what your results have been?

Just have to admit this looks great in theory but how has it worked in real life? I am curious but skeptical at the same time.
A fool and his money are good for business.

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Fri Jun 21, 2019 10:51 pm

nedsaid wrote:
Fri Jun 21, 2019 10:35 pm
Just have to admit this looks great in theory but how has it worked in real life? I am curious but skeptical at the same time.
Well, my opinion is that the results of volatility targeting have generally looked even better in practice than they do in theory.
Since their 2011 inception, these four iShares min vol ETFs have delivered between 15% to 20% less risk than their broader market indexes on an annualized basis. They have significantly captured less downside during volatile markets, but also, less upside during market rallies—just as these ETFs were designed. The long-term result has been market-like returns, but with less risk. For example, USMV has posted an annualized net gain of 14.6% versus the 14.2% gain of the S&P 500 Index, their volatility (as measured by standard deviation) has been 8.4% and 10.3%, respectively.
Many investors are driven by emotion, piling into the market at their highs and selling off near their lows. Because iShares min vol ETFs have captured less downside during market volatility, they have the potential to help investors weather market turbulence and stay invested long term.
https://www.blackrockblog.com/2016/10/2 ... ive-years/
“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

ohai
Posts: 807
Joined: Wed Dec 27, 2017 2:10 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by ohai » Sat Jun 22, 2019 12:27 am

Hi. There are a ton of financial products that express this sort of volatility controlled view, and they are becoming more popular over time. Although I am skeptical of many financial products, the very basic leverage controlled portfolio, as described here, has provided decent risk adjusted performance historically, based on what I have seen.

With that being said, if too many people invest in this strategy, the results could be catastrophic. These strategies force significant deleveraging if volatility increases suddenly. This means they will all sell the market even lower if the market drops suddenly. Managed strategies like this are believed to have exacerbated the two market drops in 2018 - in February and in December. February was more related to the unwind of short VIX positions. December seems to have been a combination of bad news (Fed and Trade War), poor liquidity, and managed strategies.

I should also mention that optimal Sharpe Ratio should not be the goal of most investors. A portfolio with 1% excess return and 0.0001% volatility has a huge Sharpe Ratio, but you would not want your money in such a portfolio unless you were extremely risk averse. For many people, higher leverage might make more sense, despite a decline in risk efficiency.

User avatar
nedsaid
Posts: 11857
Joined: Fri Nov 23, 2012 12:33 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by nedsaid » Sat Jun 22, 2019 9:25 am

ohai wrote:
Sat Jun 22, 2019 12:27 am
Hi. There are a ton of financial products that express this sort of volatility controlled view, and they are becoming more popular over time. Although I am skeptical of many financial products, the very basic leverage controlled portfolio, as described here, has provided decent risk adjusted performance historically, based on what I have seen.

With that being said, if too many people invest in this strategy, the results could be catastrophic. These strategies force significant deleveraging if volatility increases suddenly. This means they will all sell the market even lower if the market drops suddenly. Managed strategies like this are believed to have exacerbated the two market drops in 2018 - in February and in December. February was more related to the unwind of short VIX positions. December seems to have been a combination of bad news (Fed and Trade War), poor liquidity, and managed strategies.

I should also mention that optimal Sharpe Ratio should not be the goal of most investors. A portfolio with 1% excess return and 0.0001% volatility has a huge Sharpe Ratio, but you would not want your money in such a portfolio unless you were extremely risk averse. For many people, higher leverage might make more sense, despite a decline in risk efficiency.
This is my concern. I know from Larry's article the volatility tends to cluster but I have seen volatility turn on a dime, placid one moment and tempestuous the next. Not sure it is so easy to deleverage on a dime, things move so quickly these days. Sort of like a duel, the one with the quickest trigger finger wins. Investing by trigger finger makes me real nervous. So far, I am been patient enough to wait out volatility.
A fool and his money are good for business.

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Sat Jun 22, 2019 9:32 am

nedsaid wrote:
Sat Jun 22, 2019 9:25 am
ohai wrote:
Sat Jun 22, 2019 12:27 am
Hi. There are a ton of financial products that express this sort of volatility controlled view, and they are becoming more popular over time. Although I am skeptical of many financial products, the very basic leverage controlled portfolio, as described here, has provided decent risk adjusted performance historically, based on what I have seen.

With that being said, if too many people invest in this strategy, the results could be catastrophic. These strategies force significant deleveraging if volatility increases suddenly. This means they will all sell the market even lower if the market drops suddenly. Managed strategies like this are believed to have exacerbated the two market drops in 2018 - in February and in December. February was more related to the unwind of short VIX positions. December seems to have been a combination of bad news (Fed and Trade War), poor liquidity, and managed strategies.

I should also mention that optimal Sharpe Ratio should not be the goal of most investors. A portfolio with 1% excess return and 0.0001% volatility has a huge Sharpe Ratio, but you would not want your money in such a portfolio unless you were extremely risk averse. For many people, higher leverage might make more sense, despite a decline in risk efficiency.
This is my concern. I know from Larry's article the volatility tends to cluster but I have seen volatility turn on a dime, placid one moment and tempestuous the next. Not sure it is so easy to deleverage on a dime, things move so quickly these days. Sort of like a duel, the one with the quickest trigger finger wins. Investing by trigger finger makes me real nervous. So far, I am been patient enough to wait out volatility.
I don't believe that concern is well founded. The target volatility approaches that vineviz and I reported on only traded once per month and were very effective, even more so than the ETFs referenced above.
“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
nisiprius
Advisory Board
Posts: 38480
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by nisiprius » Sat Jun 22, 2019 9:36 am

How hard would it have been to write "Volatility targeting would have improved risk adjusted returns," rather than "Volatility targeting improves risk adjusted returns?"

How hard would it have been for academics Dreyer and Hubrich, in their paper, to write "we show that in the past, these strategies would have offered an improvement in risk-adjusted return" instead of "these strategies offer an improvement in risk-adjusted return?" They weren't writing for a lay audience, and they can't blame their words on any headline writer.

Note that it is important to say "would have" if you are referring to a hypothetical strategy, rather than studying the actual results actually obtained by actual investors who were following that strategy in the past over the time periods being studied.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
nedsaid
Posts: 11857
Joined: Fri Nov 23, 2012 12:33 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by nedsaid » Sat Jun 22, 2019 11:15 am

willthrill81 wrote:
Sat Jun 22, 2019 9:32 am
nedsaid wrote:
Sat Jun 22, 2019 9:25 am
ohai wrote:
Sat Jun 22, 2019 12:27 am
Hi. There are a ton of financial products that express this sort of volatility controlled view, and they are becoming more popular over time. Although I am skeptical of many financial products, the very basic leverage controlled portfolio, as described here, has provided decent risk adjusted performance historically, based on what I have seen.

With that being said, if too many people invest in this strategy, the results could be catastrophic. These strategies force significant deleveraging if volatility increases suddenly. This means they will all sell the market even lower if the market drops suddenly. Managed strategies like this are believed to have exacerbated the two market drops in 2018 - in February and in December. February was more related to the unwind of short VIX positions. December seems to have been a combination of bad news (Fed and Trade War), poor liquidity, and managed strategies.

I should also mention that optimal Sharpe Ratio should not be the goal of most investors. A portfolio with 1% excess return and 0.0001% volatility has a huge Sharpe Ratio, but you would not want your money in such a portfolio unless you were extremely risk averse. For many people, higher leverage might make more sense, despite a decline in risk efficiency.
This is my concern. I know from Larry's article the volatility tends to cluster but I have seen volatility turn on a dime, placid one moment and tempestuous the next. Not sure it is so easy to deleverage on a dime, things move so quickly these days. Sort of like a duel, the one with the quickest trigger finger wins. Investing by trigger finger makes me real nervous. So far, I am been patient enough to wait out volatility.
I don't believe that concern is well founded. The target volatility approaches that vineviz and I reported on only traded once per month and were very effective, even more so than the ETFs referenced above.
If you can do this and achieve superior results, more power to you. You have opened my eyes to this. Just hard to believe the hedge funds haven't arbitraged this away. If Larry is writing about this, lots of professional investors know about this phenomenon.
A fool and his money are good for business.

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Sat Jun 22, 2019 11:23 am

nedsaid wrote:
Sat Jun 22, 2019 11:15 am
willthrill81 wrote:
Sat Jun 22, 2019 9:32 am
nedsaid wrote:
Sat Jun 22, 2019 9:25 am
ohai wrote:
Sat Jun 22, 2019 12:27 am
Hi. There are a ton of financial products that express this sort of volatility controlled view, and they are becoming more popular over time. Although I am skeptical of many financial products, the very basic leverage controlled portfolio, as described here, has provided decent risk adjusted performance historically, based on what I have seen.

With that being said, if too many people invest in this strategy, the results could be catastrophic. These strategies force significant deleveraging if volatility increases suddenly. This means they will all sell the market even lower if the market drops suddenly. Managed strategies like this are believed to have exacerbated the two market drops in 2018 - in February and in December. February was more related to the unwind of short VIX positions. December seems to have been a combination of bad news (Fed and Trade War), poor liquidity, and managed strategies.

I should also mention that optimal Sharpe Ratio should not be the goal of most investors. A portfolio with 1% excess return and 0.0001% volatility has a huge Sharpe Ratio, but you would not want your money in such a portfolio unless you were extremely risk averse. For many people, higher leverage might make more sense, despite a decline in risk efficiency.
This is my concern. I know from Larry's article the volatility tends to cluster but I have seen volatility turn on a dime, placid one moment and tempestuous the next. Not sure it is so easy to deleverage on a dime, things move so quickly these days. Sort of like a duel, the one with the quickest trigger finger wins. Investing by trigger finger makes me real nervous. So far, I am been patient enough to wait out volatility.
I don't believe that concern is well founded. The target volatility approaches that vineviz and I reported on only traded once per month and were very effective, even more so than the ETFs referenced above.
If you can do this and achieve superior results, more power to you. You have opened my eyes to this. Just hard to believe the hedge funds haven't arbitraged this away. If Larry is writing about this, lots of professional investors know about this phenomenon.
I'm a trend follower, so I'm not using this approach, but a target volatility strategy seems to be something of a middle-ground between buy-and-hold and what I do since you're always invested in the market, just with varying allocations.

It isn't clear to me how the volatility targeting 'effect' could easily be arbitraged away, especially considering that the amount of total funds currently being used in this strategy seems to be very low, relatively speaking. The total assets of the four ETFs referenced above is just $47 billion vs. the total market's ~$30 trillion. And on top of that, the targeted volatility of each fund/investor can vary significantly.
Last edited by willthrill81 on Sat Jun 22, 2019 11:26 am, 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

ohai
Posts: 807
Joined: Wed Dec 27, 2017 2:10 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by ohai » Sat Jun 22, 2019 11:25 am

willthrill81 wrote:
Sat Jun 22, 2019 9:32 am
nedsaid wrote:
Sat Jun 22, 2019 9:25 am
ohai wrote:
Sat Jun 22, 2019 12:27 am
Hi. There are a ton of financial products that express this sort of volatility controlled view, and they are becoming more popular over time. Although I am skeptical of many financial products, the very basic leverage controlled portfolio, as described here, has provided decent risk adjusted performance historically, based on what I have seen.

With that being said, if too many people invest in this strategy, the results could be catastrophic. These strategies force significant deleveraging if volatility increases suddenly. This means they will all sell the market even lower if the market drops suddenly. Managed strategies like this are believed to have exacerbated the two market drops in 2018 - in February and in December. February was more related to the unwind of short VIX positions. December seems to have been a combination of bad news (Fed and Trade War), poor liquidity, and managed strategies.

I should also mention that optimal Sharpe Ratio should not be the goal of most investors. A portfolio with 1% excess return and 0.0001% volatility has a huge Sharpe Ratio, but you would not want your money in such a portfolio unless you were extremely risk averse. For many people, higher leverage might make more sense, despite a decline in risk efficiency.
This is my concern. I know from Larry's article the volatility tends to cluster but I have seen volatility turn on a dime, placid one moment and tempestuous the next. Not sure it is so easy to deleverage on a dime, things move so quickly these days. Sort of like a duel, the one with the quickest trigger finger wins. Investing by trigger finger makes me real nervous. So far, I am been patient enough to wait out volatility.
I don't believe that concern is well founded. The target volatility approaches that vineviz and I reported on only traded once per month and were very effective, even more so than the ETFs referenced above.
There are dozens if not hundreds of funds like this and their assets are growing. The fund you mentioned might trade once a month, but it is more common for funds to adjust leverage on a 2 day basis or so. I have personally sold hundreds of millions of dollars in December 2018 in response to sudden forced deleveraging; 5 years ago, I did zero. The effect is becoming real, although still small. If such funds grow 10x, compounded with 2008 kind of crisis, it will likely be noticeable to even non professional traders.

What I was pointing out was the potential for such strategies to exacerbate crises, even if that is not happening yet to a noticeable extent.

I think a lot of market paranoia effects are not well founded: passive investing destroying fundamentals is one example of something I believe is unlikely to happen. However, from my perspective as someone who trades stocks institutionally for the whole day, every day, derivatives or algorithmic selling has real effects. Maybe I am just another fake expert. That’s for you to consider and decide I guess.
Last edited by ohai on Sat Jun 22, 2019 11:38 am, edited 1 time in total.

User avatar
mrspock
Posts: 401
Joined: Tue Feb 13, 2018 2:49 am
Location: Vulcan

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by mrspock » Sat Jun 22, 2019 11:32 am

willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.

ohai
Posts: 807
Joined: Wed Dec 27, 2017 2:10 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by ohai » Sat Jun 22, 2019 11:37 am

nedsaid wrote:
Sat Jun 22, 2019 9:25 am
ohai wrote:
Sat Jun 22, 2019 12:27 am
Hi. There are a ton of financial products that express this sort of volatility controlled view, and they are becoming more popular over time. Although I am skeptical of many financial products, the very basic leverage controlled portfolio, as described here, has provided decent risk adjusted performance historically, based on what I have seen.

With that being said, if too many people invest in this strategy, the results could be catastrophic. These strategies force significant deleveraging if volatility increases suddenly. This means they will all sell the market even lower if the market drops suddenly. Managed strategies like this are believed to have exacerbated the two market drops in 2018 - in February and in December. February was more related to the unwind of short VIX positions. December seems to have been a combination of bad news (Fed and Trade War), poor liquidity, and managed strategies.

I should also mention that optimal Sharpe Ratio should not be the goal of most investors. A portfolio with 1% excess return and 0.0001% volatility has a huge Sharpe Ratio, but you would not want your money in such a portfolio unless you were extremely risk averse. For many people, higher leverage might make more sense, despite a decline in risk efficiency.
This is my concern. I know from Larry's article the volatility tends to cluster but I have seen volatility turn on a dime, placid one moment and tempestuous the next. Not sure it is so easy to deleverage on a dime, things move so quickly these days. Sort of like a duel, the one with the quickest trigger finger wins. Investing by trigger finger makes me real nervous. So far, I am been patient enough to wait out volatility.
Most of these funds trade on market close. Managers don’t always have discretion to choose when to adjust leverage. Most just systematically execute at a certain time.

Now, of course there is the potential for other traders to anticipate this market close interest, but it’s obviously pretty risky as you are subject to volatility from other sources.

Maybe if these algos become too big in the future, stock exchanges will have to institute a modified MOC and publish more information to help absorb these trades. Who knows.

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Sat Jun 22, 2019 11:39 am

mrspock wrote:
Sat Jun 22, 2019 11:32 am
willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
“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

MotoTrojan
Posts: 4977
Joined: Wed Feb 01, 2017 8:39 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by MotoTrojan » Sat Jun 22, 2019 11:49 am

In another couple of threads I’ve been discussing a similar but different approach that adjusts to the recent volatility in several assets and adjusts the allocation for risk parity or inverse volatility (which are equivalent when using 2 assets). This backtests extremely well for my particular funds of interest, long treasuries and S&P500.

Example of the volatility managed portfolio:
https://www.portfoliovisualizer.com/tes ... 0&total1=0

Comparison using long run risk parity allocation and rebalancing:
https://www.portfoliovisualizer.com/bac ... 0&total3=0

User avatar
mrspock
Posts: 401
Joined: Tue Feb 13, 2018 2:49 am
Location: Vulcan

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by mrspock » Sat Jun 22, 2019 12:56 pm

willthrill81 wrote:
Sat Jun 22, 2019 11:39 am
mrspock wrote:
Sat Jun 22, 2019 11:32 am
willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
Fair enough for my tax sheltered, but that still isn’t good enough for taxable: What happens if it stops working? My taxable account has things which I need high confidence I can get the expected returns until I’m dead and buried. I can’t risk a factor fund which might stop working and then I have to either sit in an under performing/high MER fund, or sell it and take a giant capitals tax hit.

Too risky. But you do make a good point for tax sheltered accounts.

MotoTrojan
Posts: 4977
Joined: Wed Feb 01, 2017 8:39 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by MotoTrojan » Sat Jun 22, 2019 1:00 pm

mrspock wrote:
Sat Jun 22, 2019 12:56 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:39 am
mrspock wrote:
Sat Jun 22, 2019 11:32 am
willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
Fair enough for my tax sheltered, but that still isn’t good enough for taxable: What happens if it stops working? My taxable account has things which I need high confidence I can get the expected returns until I’m dead and buried. I can’t risk a factor fund which might stop working and then I have to either sit in an under performing/high MER fund, or sell it and take a giant capitals tax hit.

Too risky. But you do make a good point for tax sheltered accounts.
I would never buy something I’m afraid could “stop working”. Factors need to be held for the long haul. Yes it’s a bet but selling if it’s underperforming is a surefire way to lose.

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Sat Jun 22, 2019 2:55 pm

MotoTrojan wrote:
Sat Jun 22, 2019 1:00 pm
mrspock wrote:
Sat Jun 22, 2019 12:56 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:39 am
mrspock wrote:
Sat Jun 22, 2019 11:32 am
willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
Fair enough for my tax sheltered, but that still isn’t good enough for taxable: What happens if it stops working? My taxable account has things which I need high confidence I can get the expected returns until I’m dead and buried. I can’t risk a factor fund which might stop working and then I have to either sit in an under performing/high MER fund, or sell it and take a giant capitals tax hit.

Too risky. But you do make a good point for tax sheltered accounts.
I would never buy something I’m afraid could “stop working”. Factors need to be held for the long haul. Yes it’s a bet but selling if it’s underperforming is a surefire way to lose.
Absolutely. If you're unwilling to risk underperforming the market, then you shouldn't deviate from the market. Even the popularized value and small premiums have been negative for long periods. The value premium has been negative now for about 13 years, and the small premium once was negative for over 30 years.

What I find interesting, however, is how many here assume that the market's returns will be adequate for them to reach their financial objectives. There is obviously no guarantee of that, and the market's history has demonstrated that such an assumption may be very flawed unless you're fine with zero or negative returns when it matters the most to you.
“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

klaus14
Posts: 206
Joined: Sun Nov 25, 2018 7:43 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by klaus14 » Sat Jun 22, 2019 6:10 pm

willthrill81 wrote:
Sat Jun 22, 2019 11:39 am
mrspock wrote:
Sat Jun 22, 2019 11:32 am
willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
Those ETFs will also distribute lots of capital gains to you, right?

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Sat Jun 22, 2019 6:14 pm

klaus14 wrote:
Sat Jun 22, 2019 6:10 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:39 am
mrspock wrote:
Sat Jun 22, 2019 11:32 am
willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
Those ETFs will also distribute lots of capital gains to you, right?
Not necessarily. According to iShares, as of the end of May, 2019, the pre-tax returns since inception of USMV were 14.01%, and the after-tax returns with no sale were 13.45%. I believe this small difference to be due to ETFs' superior tax structure over mutual funds.
“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

klaus14
Posts: 206
Joined: Sun Nov 25, 2018 7:43 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by klaus14 » Sat Jun 22, 2019 6:24 pm

willthrill81 wrote:
Sat Jun 22, 2019 6:14 pm
klaus14 wrote:
Sat Jun 22, 2019 6:10 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:39 am
mrspock wrote:
Sat Jun 22, 2019 11:32 am
willthrill81 wrote:
Thu Jun 20, 2019 11:24 pm
I'm a bit surprised that this thread didn't get more attention. Volatility targeting has been demonstrated to be a very effective means of controlling downside risk with minimal negative impact on returns and potentially even positive impact. Building on vineviz's excellent post above, potentially the most effective counterweight to a stock or stock-heavy fund is long-term Treasuries.
The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
Those ETFs will also distribute lots of capital gains to you, right?
Not necessarily. According to iShares, as of the end of May, 2019, the pre-tax returns since inception of USMV were 14.01%, and the after-tax returns with no sale were 13.45%. I believe this small difference to be due to ETFs' superior tax structure over mutual funds.
USMV is not doing volatility targeting. it minimizes volatility by picking uncorrelated stocks. It's turnover is just 22%.
Whereas, DMRL does volatility targeting. It's turnover is 430%.
Still not as bad as i thought, 1-year numbers from fidelity: return 6.13% -> 5.34% (after tax)

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Sat Jun 22, 2019 6:29 pm

klaus14 wrote:
Sat Jun 22, 2019 6:24 pm
willthrill81 wrote:
Sat Jun 22, 2019 6:14 pm
klaus14 wrote:
Sat Jun 22, 2019 6:10 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:39 am
mrspock wrote:
Sat Jun 22, 2019 11:32 am


The problem is taxes. How do you implement such a strategy in anything other than a tax sheltered account? I just don’t have enough in mine to make this worth doing.
You could buy any one of the many ETFs that employ it.
Those ETFs will also distribute lots of capital gains to you, right?
Not necessarily. According to iShares, as of the end of May, 2019, the pre-tax returns since inception of USMV were 14.01%, and the after-tax returns with no sale were 13.45%. I believe this small difference to be due to ETFs' superior tax structure over mutual funds.
USMV is not doing volatility targeting. it minimizes volatility by picking uncorrelated stocks. It's turnover is just 22%.
Fair point, but the concept is very similar.
klaus14 wrote:
Sat Jun 22, 2019 6:24 pm
Whereas, DMRL does volatility targeting. It's turnover is 430%.
Still not as bad as i thought, 1-year numbers from fidelity: return 6.13% -> 5.34% (after tax)
And that's probably assuming the highest marginal tax rates. For the majority in lower tax brackets, the difference would be less.
“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

klaus14
Posts: 206
Joined: Sun Nov 25, 2018 7:43 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by klaus14 » Sat Jun 22, 2019 7:03 pm

willthrill81 wrote:
Sat Jun 22, 2019 6:29 pm
Fair point, but the concept is very similar.
Not really. DMRL (managed risk/volatility targeting) doesn't pick stocks. They invest in SP500 (or another index), when volatility gets high they move some money to safe assets (cash or treasuries). When volatility gets low again, they move back. This causes lots of turnover which may cause capital gains (i don't know if they can optimize this part or not).

I am reading low/min volatility stocks are very crowded now. Thus I like DMRL approach more. This is one of the options that i am considering to reduce my risk (if valuations get even crazier)
willthrill81 wrote:
Sat Jun 22, 2019 6:29 pm
And that's probably assuming the highest marginal tax rates. For the majority in lower tax brackets, the difference would be less.
Or more with states taxes.

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Sat Jun 22, 2019 9:34 pm

klaus14 wrote:
Sat Jun 22, 2019 7:03 pm
willthrill81 wrote:
Sat Jun 22, 2019 6:29 pm
Fair point, but the concept is very similar.
Not really. DMRL (managed risk/volatility targeting) doesn't pick stocks. They invest in SP500 (or another index), when volatility gets high they move some money to safe assets (cash or treasuries). When volatility gets low again, they move back. This causes lots of turnover which may cause capital gains (i don't know if they can optimize this part or not).

I am reading low/min volatility stocks are very crowded now. Thus I like DMRL approach more. This is one of the options that i am considering to reduce my risk (if valuations get even crazier)
The implementation is very different, but the general concept is similar (i.e. pairing uncorrelated or weakly correlated assets), although min. vol. funds are doing so while maintaining a static AA, and volatility targeting is taking on a dynamic AA.

I agree that the latter approach is more appealing. But to be honest, if one has adequate tax-advantaged space, the simplest solution is to implement a target volatility approach yourself with your own funds and volatility target.
klaus14 wrote:
Sat Jun 22, 2019 7:03 pm
willthrill81 wrote:
Sat Jun 22, 2019 6:29 pm
And that's probably assuming the highest marginal tax rates. For the majority in lower tax brackets, the difference would be less.
Or more with states taxes.
After having lived in a no income tax state for a while, I almost forget that most other folks have to deal with them.
“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
patrick013
Posts: 2621
Joined: Mon Jul 13, 2015 7:49 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by patrick013 » Sun Jun 23, 2019 12:35 pm

klaus14 wrote:
Sat Jun 22, 2019 7:03 pm
willthrill81 wrote:
Sat Jun 22, 2019 6:29 pm
Fair point, but the concept is very similar.
Not really. DMRL (managed risk/volatility targeting) doesn't pick stocks. They invest in SP500 (or another index), when volatility gets high they move some money to safe assets (cash or treasuries). When volatility gets low again, they move back. This causes lots of turnover which may cause capital gains (i don't know if they can optimize this part or not).

I am reading low/min volatility stocks are very crowded now. Thus I like DMRL approach more. This is one of the options that i am considering to reduce my risk (if valuations get even crazier)
willthrill81 wrote:
Sat Jun 22, 2019 6:29 pm
And that's probably assuming the highest marginal tax rates. For the majority in lower tax brackets, the difference would be less.
Or more with states taxes.
Funds have been trying advanced math methods in hedge funds and quant funds for years. This one is a short term quant but looks better than watching daily spreads and day trading. Devotion req'd, a market that cooperates, and some luck also req'd. Most if not all of these funds that use these math methods don't beat the market portfolio and usually close in a few years. So more fund history needed or individual attempts and actual experience.

Watching funds move in and out of treasuries has been predictive for years and years.

Kinda reminds me of trading within a bollinger band or a moving average tunnel. Some LT assets to sell would be nice and a market that cooperates with some pricing annually and within the short term window.

Seems that DMRL is the one to watch then.
age in bonds, buy-and-hold, 10 year business cycle

MotoTrojan
Posts: 4977
Joined: Wed Feb 01, 2017 8:39 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by MotoTrojan » Wed Jul 17, 2019 12:06 pm

MotoTrojan wrote:
Sat Jun 22, 2019 11:49 am
In another couple of threads I’ve been discussing a similar but different approach that adjusts to the recent volatility in several assets and adjusts the allocation for risk parity or inverse volatility (which are equivalent when using 2 assets). This backtests extremely well for my particular funds of interest, long treasuries and S&P500.

Example of the volatility managed portfolio:
https://www.portfoliovisualizer.com/tes ... 0&total1=0

Comparison using long run risk parity allocation and rebalancing:
https://www.portfoliovisualizer.com/bac ... 0&total3=0
FWIW I thought I'd mention I am putting my money where my mouth is. Last month I started this risk-parity variation of targeting equity volatility with UPRO & TMF directly. After playing with things a bit I am looking to transition and lock things down for my August rebalance by using a 50/50 blend of TMF and EDV and then risk-parity allocating it monthly with UPRO. Diluting TMF with EDV will reduce the overall leverage slightly, and make the swings in monthly equity exposure a bit more mild as well (straight UPRO/TMF had allocations as high as 70% UPRO which is a bit scary).

petulant
Posts: 557
Joined: Thu Sep 22, 2016 1:09 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by petulant » Wed Jul 17, 2019 12:38 pm

willthrill81 wrote:
Sat Jun 22, 2019 11:23 am
*snip*
I'm a trend follower, so I'm not using this approach, but a target volatility strategy seems to be something of a middle-ground between buy-and-hold and what I do since you're always invested in the market, just with varying allocations.

It isn't clear to me how the volatility targeting 'effect' could easily be arbitraged away, especially considering that the amount of total funds currently being used in this strategy seems to be very low, relatively speaking. The total assets of the four ETFs referenced above is just $47 billion vs. the total market's ~$30 trillion. And on top of that, the targeted volatility of each fund/investor can vary significantly.
There can be a complicated interrelationship between short-term debt instruments, derivatives, stock prices, and volatility. I wouldn't doubt the power of the market to achieve arbitrage in the future just because a few ETFs seem small.

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Wed Jul 17, 2019 12:43 pm

petulant wrote:
Wed Jul 17, 2019 12:38 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:23 am
*snip*
I'm a trend follower, so I'm not using this approach, but a target volatility strategy seems to be something of a middle-ground between buy-and-hold and what I do since you're always invested in the market, just with varying allocations.

It isn't clear to me how the volatility targeting 'effect' could easily be arbitraged away, especially considering that the amount of total funds currently being used in this strategy seems to be very low, relatively speaking. The total assets of the four ETFs referenced above is just $47 billion vs. the total market's ~$30 trillion. And on top of that, the targeted volatility of each fund/investor can vary significantly.
There can be a complicated interrelationship between short-term debt instruments, derivatives, stock prices, and volatility. I wouldn't doubt the power of the market to achieve arbitrage in the future just because a few ETFs seem small.
I'm far from convinced that anything and everything can be 'arbitraged away' by the market. A minimum volatility approach theoretically could be, but I've yet to see a convincing argument as to how that could happen to a target volatility approach. Target volatility seems to work at least in part due to the volatility clustering phenomenon, and that certainly seems to be alive and well. TV has also worked well across asset classes.
“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
whodidntante
Posts: 5606
Joined: Thu Jan 21, 2016 11:11 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by whodidntante » Wed Jul 17, 2019 1:03 pm

What leverage would do is open up more possibilities for exposure to sources of risk and return, and to improve upon volatility and expected returns. Which sounds oh so great and never seems to work, or if it does work, managed future fund managers are doing it wrong or have been the unluckiest group imaginable.

petulant
Posts: 557
Joined: Thu Sep 22, 2016 1:09 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by petulant » Wed Jul 17, 2019 1:05 pm

willthrill81 wrote:
Wed Jul 17, 2019 12:43 pm
petulant wrote:
Wed Jul 17, 2019 12:38 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:23 am
*snip*
I'm a trend follower, so I'm not using this approach, but a target volatility strategy seems to be something of a middle-ground between buy-and-hold and what I do since you're always invested in the market, just with varying allocations.

It isn't clear to me how the volatility targeting 'effect' could easily be arbitraged away, especially considering that the amount of total funds currently being used in this strategy seems to be very low, relatively speaking. The total assets of the four ETFs referenced above is just $47 billion vs. the total market's ~$30 trillion. And on top of that, the targeted volatility of each fund/investor can vary significantly.
There can be a complicated interrelationship between short-term debt instruments, derivatives, stock prices, and volatility. I wouldn't doubt the power of the market to achieve arbitrage in the future just because a few ETFs seem small.
I'm far from convinced that anything and everything can be 'arbitraged away' by the market. A minimum volatility approach theoretically could be, but I've yet to see a convincing argument as to how that could happen to a target volatility approach. Target volatility seems to work at least in part due to the volatility clustering phenomenon, and that certainly seems to be alive and well. TV has also worked well across asset classes.
The TV investor would be expected to exit equities during periods of high volatility and buy debt securities during periods of low volatility. Although any particular TV investor would have a specific TV, the aggregate of TV investors with this behavior would be selling equities and buying debt together at each change in volatility. There are no TV investors buying or selling from each other.

For example, imagine that in time 0 volatility is 1 (measured arbitrarily), while in time 1 volatility is 2 for a benchmark. TV investors as a group must all respond by selling equities and buying debt. There is no TV investor who responds by now buying equities since, if they wanted a position with more volatility when the volatility benchmark is 2, they would have already tried to obtain that volatility when the benchmark was at 1. That means each move in the volatility benchmark could be expected to have a set of TV transactions. To satisfy this demand, the TV investors in aggregate must attract new buyers of the stock. To attract these buyers, the TV investors would need to offer lower prices. Buying selling lower, TV investors give up future returns and new investors have better expected returns. Further, if volatility clustering was well-known (which it is in Wall Street), those new investors would only purchase at price levels that take into account the higher risk from recent volatility. Also, if these transactions could be expected to take 1-2 months to fully play out, it's entirely possible traders looking for a large TV effect would front-run these transactions by shorting or using puts until the price went so low that the market bottom is approached faster and/or expected future returns are higher as TV investors sell.

That is even before speculating how arbitrage between debt and stock prices might be achieved through derivatives. For example, if recent volatility resulted in TV investors selling stock, the counterparty buying the stock could create a synthetic debt instrument by holding the stock and using put/call parity at some future date to eliminate price risk. Thus, if TV investors crowd out of stocks and into debt, sophisticated entities could arbitrage through the options market (assuming the options market itself is large and liquid enough).

User avatar
willthrill81
Posts: 10937
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by willthrill81 » Wed Jul 17, 2019 1:12 pm

petulant wrote:
Wed Jul 17, 2019 1:05 pm
willthrill81 wrote:
Wed Jul 17, 2019 12:43 pm
petulant wrote:
Wed Jul 17, 2019 12:38 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:23 am
*snip*
I'm a trend follower, so I'm not using this approach, but a target volatility strategy seems to be something of a middle-ground between buy-and-hold and what I do since you're always invested in the market, just with varying allocations.

It isn't clear to me how the volatility targeting 'effect' could easily be arbitraged away, especially considering that the amount of total funds currently being used in this strategy seems to be very low, relatively speaking. The total assets of the four ETFs referenced above is just $47 billion vs. the total market's ~$30 trillion. And on top of that, the targeted volatility of each fund/investor can vary significantly.
There can be a complicated interrelationship between short-term debt instruments, derivatives, stock prices, and volatility. I wouldn't doubt the power of the market to achieve arbitrage in the future just because a few ETFs seem small.
I'm far from convinced that anything and everything can be 'arbitraged away' by the market. A minimum volatility approach theoretically could be, but I've yet to see a convincing argument as to how that could happen to a target volatility approach. Target volatility seems to work at least in part due to the volatility clustering phenomenon, and that certainly seems to be alive and well. TV has also worked well across asset classes.
The TV investor would be expected to exit equities during periods of high volatility and buy debt securities during periods of low volatility. Although any particular TV investor would have a specific TV, the aggregate of TV investors with this behavior would be selling equities and buying debt together at each change in volatility. There are no TV investors buying or selling from each other.

For example, imagine that in time 0 volatility is 1 (measured arbitrarily), while in time 1 volatility is 2 for a benchmark. TV investors as a group must all respond by selling equities and buying debt. There is no TV investor who responds by now buying equities since, if they wanted a position with more volatility when the volatility benchmark is 2, they would have already tried to obtain that volatility when the benchmark was at 1. That means each move in the volatility benchmark could be expected to have a set of TV transactions. To satisfy this demand, the TV investors in aggregate must attract new buyers of the stock. To attract these buyers, the TV investors would need to offer lower prices. Buying selling lower, TV investors give up future returns and new investors have better expected returns. Further, if volatility clustering was well-known (which it is in Wall Street), those new investors would only purchase at price levels that take into account the higher risk from recent volatility. Also, if these transactions could be expected to take 1-2 months to fully play out, it's entirely possible traders looking for a large TV effect would front-run these transactions by shorting or using puts until the price went so low that the market bottom is approached faster and/or expected future returns are higher as TV investors sell.

That is even before speculating how arbitrage between debt and stock prices might be achieved through derivatives. For example, if recent volatility resulted in TV investors selling stock, the counterparty buying the stock could create a synthetic debt instrument by holding the stock and using put/call parity at some future date to eliminate price risk. Thus, if TV investors crowd out of stocks and into debt, sophisticated entities could arbitrage through the options market (assuming the options market itself is large and liquid enough).
Fair enough, although that sounds more than a bit like the arguments we've seen like "if everyone indexed, it would be terrible," even though everyone isn't going to. I think that we're a long ways off from there being significant front-running in the face of relatively high volatility.
“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

petulant
Posts: 557
Joined: Thu Sep 22, 2016 1:09 pm

Re: Larry Swedroe: Volatility Targeting Improves Risk Adjusted Returns

Post by petulant » Wed Jul 17, 2019 1:29 pm

willthrill81 wrote:
Wed Jul 17, 2019 1:12 pm
petulant wrote:
Wed Jul 17, 2019 1:05 pm
willthrill81 wrote:
Wed Jul 17, 2019 12:43 pm
petulant wrote:
Wed Jul 17, 2019 12:38 pm
willthrill81 wrote:
Sat Jun 22, 2019 11:23 am

I'm a trend follower, so I'm not using this approach, but a target volatility strategy seems to be something of a middle-ground between buy-and-hold and what I do since you're always invested in the market, just with varying allocations.

It isn't clear to me how the volatility targeting 'effect' could easily be arbitraged away, especially considering that the amount of total funds currently being used in this strategy seems to be very low, relatively speaking. The total assets of the four ETFs referenced above is just $47 billion vs. the total market's ~$30 trillion. And on top of that, the targeted volatility of each fund/investor can vary significantly.
There can be a complicated interrelationship between short-term debt instruments, derivatives, stock prices, and volatility. I wouldn't doubt the power of the market to achieve arbitrage in the future just because a few ETFs seem small.
I'm far from convinced that anything and everything can be 'arbitraged away' by the market. A minimum volatility approach theoretically could be, but I've yet to see a convincing argument as to how that could happen to a target volatility approach. Target volatility seems to work at least in part due to the volatility clustering phenomenon, and that certainly seems to be alive and well. TV has also worked well across asset classes.
The TV investor would be expected to exit equities during periods of high volatility and buy debt securities during periods of low volatility. Although any particular TV investor would have a specific TV, the aggregate of TV investors with this behavior would be selling equities and buying debt together at each change in volatility. There are no TV investors buying or selling from each other.

For example, imagine that in time 0 volatility is 1 (measured arbitrarily), while in time 1 volatility is 2 for a benchmark. TV investors as a group must all respond by selling equities and buying debt. There is no TV investor who responds by now buying equities since, if they wanted a position with more volatility when the volatility benchmark is 2, they would have already tried to obtain that volatility when the benchmark was at 1. That means each move in the volatility benchmark could be expected to have a set of TV transactions. To satisfy this demand, the TV investors in aggregate must attract new buyers of the stock. To attract these buyers, the TV investors would need to offer lower prices. Buying selling lower, TV investors give up future returns and new investors have better expected returns. Further, if volatility clustering was well-known (which it is in Wall Street), those new investors would only purchase at price levels that take into account the higher risk from recent volatility. Also, if these transactions could be expected to take 1-2 months to fully play out, it's entirely possible traders looking for a large TV effect would front-run these transactions by shorting or using puts until the price went so low that the market bottom is approached faster and/or expected future returns are higher as TV investors sell.

That is even before speculating how arbitrage between debt and stock prices might be achieved through derivatives. For example, if recent volatility resulted in TV investors selling stock, the counterparty buying the stock could create a synthetic debt instrument by holding the stock and using put/call parity at some future date to eliminate price risk. Thus, if TV investors crowd out of stocks and into debt, sophisticated entities could arbitrage through the options market (assuming the options market itself is large and liquid enough).
Fair enough, although that sounds more than a bit like the arguments we've seen like "if everyone indexed, it would be terrible," even though everyone isn't going to. I think that we're a long ways off from there being significant front-running in the face of relatively high volatility.
That's fair.

Post Reply