Is Low Volatility the best factor?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
MaxLeo
Posts: 22
Joined: Wed Jul 10, 2019 9:46 am

Is Low Volatility the best factor?

Post by MaxLeo »

If you have a look at the Sharpe-Ratio of the MSCI USA Factors, you will notice that Low Volatility has the best Sharpe-Ratio over all three listed periods (3 year, 5 year and 10 year)

Do you think that it is the best of all different factors? And do you now some interesting articles / websites which compare different factors and especially the Low Volatility factor?
User avatar
nedsaid
Posts: 19275
Joined: Fri Nov 23, 2012 11:33 am

Re: Is Low Volatility the best factor?

Post by nedsaid »

MaxLeo wrote: Sat Oct 19, 2019 12:29 pm If you have a look at the Sharpe-Ratio of the MSCI USA Factors, you will notice that Low Volatility has the best Sharpe-Ratio over all three listed periods (3 year, 5 year and 10 year)

Do you think that it is the best of all different factors? And do you now some interesting articles / websites which compare different factors and especially the Low Volatility factor?
So far, so good. A lot of these stocks, particularly Consumer Staples, are pretty expensive. Value type of stocks at Growth stock prices. My concern is that this strategy is too well known and overgrazed.
A fool and his money are good for business.
hdas
Posts: 1395
Joined: Thu Jun 11, 2015 8:24 am

[Deleted]

Post by hdas »

[Deleted]
Last edited by hdas on Wed Jan 29, 2020 8:24 pm, edited 1 time in total.
....
Elysium
Posts: 4120
Joined: Mon Apr 02, 2007 6:22 pm

Re: Is Low Volatility the best factor?

Post by Elysium »

Sector bet. Are Consumer staples, Consumer defensive, Utilities, and REITs better than other sectors? do you feel lucky to make bets on sectors that has performed well of late.
Jebediah
Posts: 805
Joined: Tue Aug 28, 2012 9:19 pm
Location: Austin TX

Re: Is Low Volatility the best factor?

Post by Jebediah »

According to PV, in the period 1964 - Present, BAB (low vol) returned 9.84% vs Rm-Rf (beta) which returned 5.01%.

So yes, by a long shot, low vol was the best factor.
User avatar
Forester
Posts: 2400
Joined: Sat Jan 19, 2019 1:50 pm
Location: UK

Re: Is Low Volatility the best factor?

Post by Forester »

Elysium wrote: Sat Oct 19, 2019 5:08 pm Sector bet. Are Consumer staples, Consumer defensive, Utilities, and REITs better than other sectors? do you feel lucky to make bets on sectors that has performed well of late.

Low vol = sector bets, simple sort of S&P 500. The biggest low vol fund in the world is actually Min Vol, USMV. No heavy sector bets there.

S&P methodology vs MSCI methodology.
Amateur Self-Taught Senior Macro Strategist
Elysium
Posts: 4120
Joined: Mon Apr 02, 2007 6:22 pm

Re: Is Low Volatility the best factor?

Post by Elysium »

Forester wrote: Sat Oct 19, 2019 5:43 pm
Elysium wrote: Sat Oct 19, 2019 5:08 pm Sector bet. Are Consumer staples, Consumer defensive, Utilities, and REITs better than other sectors? do you feel lucky to make bets on sectors that has performed well of late.

Low vol = sector bets, simple sort of S&P 500. The biggest low vol fund in the world is actually Min Vol, USMV. No heavy sector bets there.

S&P methodology vs MSCI methodology.
USMV, overweight REIT, Utilities, and underweight Technology. Meaningful enough to show a deviation from S&P 500.
stlutz
Posts: 5585
Joined: Fri Jan 02, 2009 12:08 am

Re: Is Low Volatility the best factor?

Post by stlutz »

Note that the MSCI factor "indices" don't allow for big sector bets.
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Is Low Volatility the best factor?

Post by Robert T »

.
FWIW, here’s some historical data from MSCI. Global portfolios are used throughout (All Cap World Index [ACWI])

A stock:bond mix that matched the volatility of returns (standard deviation) of the MSCI ACWI Minimum Volatility series was used. The bond portfolio used is equivalent to 5 yr T-notes. The portfolios are list by Sharpe Ratio – highest to lowest.

1999-2018 [longest time series of annual data available for the multifactor diversified series]
  • Annualized return (%) / SD / 2008 return / Sharpe Ratio / Stock:bond* ratio

    7.8 / 12.7 / -22.9 / 0.53 / 64:36 = Multi-factor diversified*
    7.5 / 12.7 / -23.3 / 0.51 / 71:29 = Prime Value
    7.4 / 12.7 / -24.2 / 0.50 / 64:36 = Momentum
    7.2 / 12.7 / -25.6 / 0.49 / 100:0 = Minimum Volatility
    6.1 / 12.7 / -24.6 / 0.40 /76:24 = Quality

    * Multi-factors = value, size, momentum, quality
From the above, over this period, minimum volatility did not perform better (based on Sharpe Ratio, or downside loss [2008 return]) than the multifactor diversified portfolio, prime value portfolio, or momentum portfolio paired with bonds.

And adding the minimum volatility portfolio to the multi-factor diversified portfolio did not add diversification benefits as below:
  • 7.5 / 12.7 / -24.7 / 0.51 / 83:17 = Multifactor diversified: Minimum Volatility*

    * The stock:bond allocation in the multifactor portion was adjusted so that when combined 50:50 with the minimum volatility portfolio it resulted in the same volatility of returns (standard deviation) as the minimum volatility portfolio on its own.
Similar returns were derived for longer time series comparisons with other global multifactor portfolios.

Obviously no guarantees.

Robert
.
GaryA505
Posts: 2909
Joined: Wed Feb 08, 2017 1:59 pm
Location: New Mexico

Re: Is Low Volatility the best factor?

Post by GaryA505 »

Robert T wrote: Sat Oct 19, 2019 6:27 pm .
FWIW, here’s some historical data from MSCI. Global portfolios are used throughout (All Cap World Index [ACWI])

A stock:bond mix that matched the volatility of returns (standard deviation) of the MSCI ACWI Minimum Volatility series was used. The bond portfolio used is equivalent to 5 yr T-notes. The portfolios are list by Sharpe Ratio – highest to lowest.

1999-2018 [longest time series of annual data available for the multifactor diversified series]
  • Annualized return (%) / SD / 2008 return / Sharpe Ratio / Stock:bond* ratio

    7.8 / 12.7 / -22.9 / 0.53 / 64:36 = Multi-factor diversified*
    7.5 / 12.7 / -23.3 / 0.51 / 71:29 = Prime Value
    7.4 / 12.7 / -24.2 / 0.50 / 64:36 = Momentum
    7.2 / 12.7 / -25.6 / 0.49 / 100:0 = Minimum Volatility
    6.1 / 12.7 / -24.6 / 0.40 /76:24 = Quality

    * Multi-factors = value, size, momentum, quality
From the above, over this period, minimum volatility did not perform better (based on Sharpe Ratio, or downside loss [2008 return]) than the multifactor diversified portfolio, prime value portfolio, or momentum portfolio paired with bonds.

And adding the minimum volatility portfolio to the multi-factor diversified portfolio did not add diversification benefits as below:
  • 7.5 / 12.7 / -24.7 / 0.51 / 83:17 = Multifactor diversified: Minimum Volatility*

    * The stock:bond allocation in the multifactor portion was adjusted so that when combined 50:50 with the minimum volatility portfolio it resulted in the same volatility of returns (standard deviation) as the minimum volatility portfolio on its own.
Similar returns were derived for longer time series comparisons with other global multifactor portfolios.

Obviously no guarantees.

Robert
.
This looks bad for min vol until you realize that it uses 5-yr treasuries for all the ports except min vol. During this period treasuries went from something like 5% down to what they are today. If this used cash, or even short term treasuries, for the fixed income side that would be different.
Get most of it right and don't make any big mistakes. All else being equal, simpler is better. Simple is as simple does.
User avatar
Uncorrelated
Posts: 1237
Joined: Sun Oct 13, 2019 3:16 pm

Re: Is Low Volatility the best factor?

Post by Uncorrelated »

Here is a table of historical monthly factor premiums from Fama & French (1926~1992):

Code: Select all

name		mean		std
RM-RF		0.52		5.33
SMB		0.15		3.18
HML		0.29		3.48
And here with the term factor included (1963~1991), also from Fama & French.

Code: Select all

name		mean		std
RM-RF		0.43		4.45
SMB		0.27		3.89
HML		0.40		2.54
TERM 		0.06		3.02
DEF 		0.02		1.60
And here from Andrea Frazzini & Lasse Heje Pedersen (1926~2012)
https://www.sciencedirect.com/science/a ... 5X13002675

Code: Select all

name		mean		std
BAB US		0.70		3.11
BAB int		0.64		2.32
BAB US*		0.55		3.11
BAB int*	0.28		2.32
BAB = bet against beta. *corrected for 5-factor model (fame & french 3 factor + momentum + liquidity)

It so appears that BAB is indeed the best factor.



However, if we look at a factor regression of USMV (iShares Edge MSCI Min Vol USA ETF), we find the following factor exposure:
https://www.portfoliovisualizer.com/fac ... e&total1=0

Code: Select all

MKT		0.82
SMB		-0.25
BAB		0.24
TERM		0.15
CDT		-0.19
It appears that USMV is a mixture of 85% very large cap and 15% long term treasuries, with some BAB tilt.

If we multiply this by the expected factor premium on the longest time period available (except for TERM and CDT, where I use 0.12 and 0), we find an monthly performance of 0.5389 (6.47% annualized). For reference, plain old total stock market has a MKT exposure of 1 and expected performance of .52 (6.24% annualized) and IJS (small cap value) has an expected performance of .7344 (8.81% annualized).

Is USMV the best fund? Not from the perspective of highest return. Backtests of the last 20 years certainly make USMV shine, but that is not suprising: TERM performance the last 20 years was significantly above average and small cap performance was significantly under average.

My point is: don't buy USMV with the expectancy of outperforming total stock market, but you might get lower volatility due to better diversification across factors.
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: Is Low Volatility the best factor?

Post by Robert T »

GaryA505 wrote: Sat Oct 19, 2019 9:19 pm This looks bad for min vol until you realize that it uses 5-yr treasuries for all the ports except min vol. During this period treasuries went from something like 5% down to what they are today. If this used cash, or even short term treasuries, for the fixed income side that would be different.
If anything, I should have used longer-term treasuries to match the term exposure of minimum volatility (rather that shorter term treasures or cash). Lots of material on this - just do google search on interest rate risk and minimum volatility.
.
User avatar
Forester
Posts: 2400
Joined: Sat Jan 19, 2019 1:50 pm
Location: UK

Re: Is Low Volatility the best factor?

Post by Forester »

https://www.msci.com/www/blog-posts/wha ... 0608475213

What do rising interest rates mean for minimum volatility strategies?
We found that the minimum volatility strategy’s exposure to the equity factor was generally stable: It remained around 0.77-0.78 across varying interest-rate regimes. Interestingly, its exposure to the bond factor when rates rose was less than half of the exposure when rates declined (0.16 vs. 0.37). This implies that the minimum volatility strategy was less sensitive to interest-rate changes in a rising rate environment.

Attributing returns to the various factors further illustrates that minimum volatility strategy responded differently during periods of declining and rising interest rates. We found that the bond factor contributed about 1.1% of positive performance to the minimum volatility strategy during a declining rate regime but caused only a negligible performance drag of -16 basis points per annum during a rising-rate regime.
Amateur Self-Taught Senior Macro Strategist
snailderby
Posts: 1421
Joined: Thu Jul 26, 2018 11:30 am

Re: Is Low Volatility the best factor?

Post by snailderby »

From Larry Swedroe at https://alphaarchitect.com/2017/02/21/s ... sk-effect/:
Research from a 2012 white paper by Pim van Vliet, “Enhancing a Low Volatility Strategy is Particularly Helpful When Generic Low Volatility is Expensive,” sheds some light on this question. Using data from 1929 through 2010, he found that while on average low-volatility strategies tend to have exposure to the value factor, that exposure is time-varying. The low-volatility factor spends about 62 percent of the time in a value regime and 38 percent of the time in a growth regime.

This regime-shifting behavior impacts the performance of low-volatility strategies. When low-volatility stocks have value exposure, they have outperformed the market, returning an average of 9.5 percent annually versus the market’s 7.5 percent. The low-volatility factor has also exhibited lower volatility, with an annual standard deviation of 13.5 percent versus the market’s 16.5 percent. However, when low-volatility stocks have growth exposure, they have underperformed, returning an average of 10.8 percent annually versus the market’s 12.2 percent. The low-volatility factor did continue to have lower annual volatility, at 15.3 percent versus 20.3 percent for the market. The result has been a higher risk-adjusted return in either regime. The bottom line is that in either regime, low volatility predicts future low volatility. However, when low volatility has negative exposure to the value factor (as it did in mid-2016), it also forecasts below-market returns.
I think it's also worth noting that low volatility (e.g., SPLV) is not the same thing as minimum volatility (e g., USMV). See https://www.indexologyblog.com/2019/05/ ... -the-same/.
Topic Author
MaxLeo
Posts: 22
Joined: Wed Jul 10, 2019 9:46 am

Re: Is Low Volatility the best factor?

Post by MaxLeo »

stlutz wrote: Sat Oct 19, 2019 6:17 pm Note that the MSCI factor "indices" don't allow for big sector bets.
And nevertheless, return/risk of low vol is impressive.
InvestInPasta
Posts: 213
Joined: Sat Sep 16, 2017 12:42 pm
Location: Italy

Re: Is Low Volatility the best factor?

Post by InvestInPasta »

The Low Vol does not seem that great.
It even had a worse MaxDD than the S&P500.

Image
https://www.portfoliovisualizer.com/bac ... bol4=VTSAX

This kind of stuff reminds me when Bogle talking about portfolio strategies said "don't waste your time looking for the Holy Grail, it doesn't exist". :wink:
When I study English I am lazier than my portfolio. Feel free to fix my English and investing mistakes.
User avatar
nisiprius
Advisory Board
Posts: 52219
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: Is Low Volatility the best factor?

Post by nisiprius »

InvestInPasta wrote: Wed May 25, 2022 6:04 am The Low Vol does not seem that great.
It even had a worse MaxDD than the S&P500...
This kind of stuff reminds me when Bogle talking about portfolio strategies said "don't waste your time looking for the Holy Grail, it doesn't exist". :wink:
Indeed. Cue "but you are looking at the Wrong ETF," "but they are using the Wrong Index," "but that is the Wrong Time Period," or "but that is the Wrong Strategy" (i.e. you need a long-short portfolio to extract enough of the factor to matter).

Also, it seems odd but as nearly as I can tell, cribbing from the models provided in PortfolioVisualizer,

"low volatility" does not emerge as a named factor in

--the Fama-French three-factor model (market, size, value)
--the Fama-French five-factor model (market, size, value, investment, profitability)
--the AQR four-factor model (market, size, value, momentum)
--the Alpha Architect five-factor model (market, size, value, momentum, quality)

Berkin and Swedroe's Your Complete Guide to Factor-Based Investing relegates it to an appendix, commenting that a study found that
the excess return associated with forming the long low-volatility/short high-volatility portfolios are basically present only in the first month after formation, and that they are largely subsumed by high transaction costs associated with low-liquidity stocks (such as low-priced/high-volatility stocks). They also found that the anomalous returns within value-weighted portfolios are largely eliminated when omitting low-priced (less than $5) stocks and are not at all present within equal-weighted portfolios. In fact, the average price of stocks in the highest-volatility quintile was just more than $7, indicating that many, if not most, would be considered “penny stocks.” And finally, they found that the low-risk effect has been noticeably weaker since 1990. New regulations were passed in that year, aimed at reducing fraud connected to trading in penny stocks. (We would add that many high-beta stocks simply disappeared after the dot-com crash, and the number of stocks on public U.S. exchanges has decreased dramatically since then.) The authors concluded: “Our findings cast some doubt on the practical profitability of a low risk trading strategy.”
Last edited by nisiprius on Wed May 25, 2022 7:02 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Booglie
Posts: 152
Joined: Sun Aug 15, 2021 2:17 am

Re: Is Low Volatility the best factor?

Post by Booglie »

When talking about systemic volatility (as opposed to e.g, portfolio volatility), the lower volatility is, the higher the risk you can take, because the market as a whole is more predictable.

That's no surprise that many people defend that you should increase risk when systemic volatility is low, and reduce risk when systemic volatility is high. In fact, that's the core strategy of people who use a "target volatility" investment method.

The downside of this approach is that "volatility is volatile", so risk can materialize at any point. This means that a strategy that is safe in a low volatility market but is potentially risky could blow up in your face. I.e, you can take more risk than you imagined you could take, and it then backfires.

Of course, if risk spikes that much, then even regular strategies can suffer pretty badly. Just look at the Great Depression, where the stock market fell 90% on average, or how balanced portfolios (stocks + bond ETFs) are doing this year (which is less worse, but still devastating if you took more risk than you could bear).
Random Walker
Posts: 5561
Joined: Fri Feb 23, 2007 7:21 pm

Re: Is Low Volatility the best factor?

Post by Random Walker »

Robert T wrote: Sat Oct 19, 2019 6:27 pm .
FWIW, here’s some historical data from MSCI. Global portfolios are used throughout (All Cap World Index [ACWI])

A stock:bond mix that matched the volatility of returns (standard deviation) of the MSCI ACWI Minimum Volatility series was used. The bond portfolio used is equivalent to 5 yr T-notes. The portfolios are list by Sharpe Ratio – highest to lowest.

1999-2018 [longest time series of annual data available for the multifactor diversified series]
  • Annualized return (%) / SD / 2008 return / Sharpe Ratio / Stock:bond* ratio

    7.8 / 12.7 / -22.9 / 0.53 / 64:36 = Multi-factor diversified*
    7.5 / 12.7 / -23.3 / 0.51 / 71:29 = Prime Value
    7.4 / 12.7 / -24.2 / 0.50 / 64:36 = Momentum
    7.2 / 12.7 / -25.6 / 0.49 / 100:0 = Minimum Volatility
    6.1 / 12.7 / -24.6 / 0.40 /76:24 = Quality

    * Multi-factors = value, size, momentum, quality
From the above, over this period, minimum volatility did not perform better (based on Sharpe Ratio, or downside loss [2008 return]) than the multifactor diversified portfolio, prime value portfolio, or momentum portfolio paired with bonds.

And adding the minimum volatility portfolio to the multi-factor diversified portfolio did not add diversification benefits as below:
  • 7.5 / 12.7 / -24.7 / 0.51 / 83:17 = Multifactor diversified: Minimum Volatility*

    * The stock:bond allocation in the multifactor portion was adjusted so that when combined 50:50 with the minimum volatility portfolio it resulted in the same volatility of returns (standard deviation) as the minimum volatility portfolio on its own.
Similar returns were derived for longer time series comparisons with other global multifactor portfolios.

Obviously no guarantees.

Robert
.
I believe Robert’s data above supports Larry Swedroe’s thesis that in a world where any factor can underperform at any time, it makes most sense to diversify across factors. And I believe it’s fair to say that the shorter the timeframe, the more important is diversification. Robert, please correct me if you disagree.

Dave
User avatar
Anon9001
Posts: 1884
Joined: Fri Dec 20, 2019 8:28 am
Location: بھارت

Re: Is Low Volatility the best factor?

Post by Anon9001 »

MaxLeo wrote: Sat Oct 19, 2019 12:29 pm If you have a look at the Sharpe-Ratio of the MSCI USA Factors, you will notice that Low Volatility has the best Sharpe-Ratio over all three listed periods (3 year, 5 year and 10 year)
Sharpe Ratio tends to fluctuate quite heavily depending on start date and end date chosen so this means absolutely nothing. If you want to get useful information out of Sharpe Ratio its best to do a Rolling Sharpe Ratio comparison which can be done in Excel.
Land/Real Estate:89.4% (Land/RE is Inheritance which will be recieved in 10-20 years) Equities:7.6% Fixed Income:1.7% Gold:0.8% Cryptocurrency:0.5%
scout1
Posts: 234
Joined: Thu Oct 18, 2018 3:26 pm

Re: Is Low Volatility the best factor?

Post by scout1 »

As others have mentioned, it was the best factor because interest rates went down over the period that you are looking. People look at low Vol stocks as perpetuities and you can see that as the discount rate goes from 6% to 3%, the valuation on earnings goes from 16.6x to 33.3x. The value for other stocks doesn't skyrocket like that because they have a higher percentage of their required return coming from their beta and equity risk premium. Under CAPM, the discount rate is beta*ERP + risk free rate. If your beta is zero, your entire valuation is changes in the risk free rate. If your beta is high, changes in the risk free rate aren't the main determinate of your valuation.


https://russellinvestments.com/us/blog/ ... volatility
GaryA505
Posts: 2909
Joined: Wed Feb 08, 2017 1:59 pm
Location: New Mexico

Re: Is Low Volatility the best factor?

Post by GaryA505 »

InvestInPasta wrote: Wed May 25, 2022 6:04 am The Low Vol does not seem that great.
It even had a worse MaxDD than the S&P500.

Image
https://www.portfoliovisualizer.com/bac ... bol4=VTSAX

This kind of stuff reminds me when Bogle talking about portfolio strategies said "don't waste your time looking for the Holy Grail, it doesn't exist". :wink:

More troubling than the slightly greater drawdown is the relatively poor recovery AFTER the drawdown.
Get most of it right and don't make any big mistakes. All else being equal, simpler is better. Simple is as simple does.
Dry-Drink
Posts: 530
Joined: Fri Jan 14, 2022 9:50 am

Re: Is Low Volatility the best factor?

Post by Dry-Drink »

InvestInPasta wrote: Wed May 25, 2022 6:04 am The Low Vol does not seem that great.
It even had a worse MaxDD than the S&P500.

Image
https://www.portfoliovisualizer.com/bac ... bol4=VTSAX

This kind of stuff reminds me when Bogle talking about portfolio strategies said "don't waste your time looking for the Holy Grail, it doesn't exist". :wink:
Looks alive and well to me, not sure what you mean:
https://www.portfoliovisualizer.com/bac ... tion3_2=15

Higher returns with lower volatility than just straight beta.
nisiprius wrote: Wed May 25, 2022 6:58 am
InvestInPasta wrote: Wed May 25, 2022 6:04 am The Low Vol does not seem that great.
It even had a worse MaxDD than the S&P500...
This kind of stuff reminds me when Bogle talking about portfolio strategies said "don't waste your time looking for the Holy Grail, it doesn't exist". :wink:
Indeed. Cue "but you are looking at the Wrong ETF," "but they are using the Wrong Index," "but that is the Wrong Time Period," or "but that is the Wrong Strategy" (i.e. you need a long-short portfolio to extract enough of the factor to matter).
Well, but it is pretty misleading to compare a long-only low-vol ETF to an index fund when stock returns are solidly positive. You know low-vol doesn't mean "higher returns than regular index", let's not act like that's the claim.
It means "volatility is lower by a bigger chunk than returns are lower". That is, a leveraged, low-vol portfolio should outperform an index fund in both risk and return. OR A low-vol long-only ETF should outperform an index fund + cash in both risk and return.

So far that has been the case for a decade+ so what's the confusion?
User avatar
nisiprius
Advisory Board
Posts: 52219
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: Is Low Volatility the best factor?

Post by nisiprius »

Dry-Drink wrote: Wed May 25, 2022 12:47 pm...You know low-vol doesn't mean "higher returns than regular index", let's not act like that's the claim.
It means "volatility is lower by a bigger chunk than returns are lower". That is, a leveraged, low-vol portfolio should outperform an index fund in both risk and return. OR A low-vol long-only ETF should outperform an index fund + cash in both risk and return.

So far that has been the case for a decade+ so what's the confusion?...
The confusion is that in the decade exhibited by InvestInPasta, which includes a full decade, the low-vol ETF had lower Sharpe and Sortino ratios than an S&P 500 ETF. Which means that it would not have outperformed an index fund + cash.

So let's verify that.

Since drawdown is what was mentioned, I will equalize the drawdowns by adding cash to the S&P 500 index fund. Wait, I can't do that, because the S&P 500 fund already had a lower drawdown. So I will equalize another aspect of risk, the standard deviation.

Source

Image

A "straight" S&P fund, with its risk diluted with cash, had exactly the same standard deviation as the low-vol fund. But it had somewhat higher return, and meaningfully lower drawdown.

The differences aren't large, but the robust benefits of low-vol didn't show up. Reducing volatility through stock-picking didn't work any better than reducing volatility by reducing stock allocation.

There was nothing wrong with SPLV, but over the time period shown, it looks like complication without compelling evidence of any practical benefit.
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.
freyj6
Posts: 554
Joined: Thu Jun 19, 2014 2:09 am

Re: Is Low Volatility the best factor?

Post by freyj6 »

Larry Swedroe has several good articles on this as well as a good section in his book.

From what I remember:

-A lot of the premium of low vol comes from not owning the highest volatility stocks. Once you’ve excluded the top decile or two of volatile stocks, the premium declines a lot.

-The performance of these stocks depends in large part on how cheap or expensive they are relative to the market. For the past few years they’ve been quite expensive.

-Low volatility stocks tend to have exposure to duration risk. That means the last 40 years of falling bond prices has been very good for them, but we may not see a repeat over the next 40.
Dry-Drink
Posts: 530
Joined: Fri Jan 14, 2022 9:50 am

Re: Is Low Volatility the best factor?

Post by Dry-Drink »

nisiprius wrote: Wed May 25, 2022 1:13 pm
Dry-Drink wrote: Wed May 25, 2022 12:47 pm...You know low-vol doesn't mean "higher returns than regular index", let's not act like that's the claim.
It means "volatility is lower by a bigger chunk than returns are lower". That is, a leveraged, low-vol portfolio should outperform an index fund in both risk and return. OR A low-vol long-only ETF should outperform an index fund + cash in both risk and return.

So far that has been the case for a decade+ so what's the confusion?...
The confusion is that in the decade exhibited by InvestInPasta, which includes a full decade, the low-vol ETF had lower Sharpe and Sortino ratios than an S&P 500 ETF.
Why are you cutting the PV data range by choosing the yearly data? If you use all available PV (monthly) data, SPLV had higher Sharpe:
https://www.portfoliovisualizer.com/bac ... bol4=CASHX
nisiprius wrote: Wed May 25, 2022 1:13 pm Since drawdown is what was mentioned, I will equalize the drawdowns by adding cash to the S&P 500 index fund. Wait, I can't do that, because the S&P 500 fund already had a lower drawdown. So I will equalize another aspect of risk, the standard deviation.
You should use St Dev because St Dev is a measure of volatility and that's what low-vol's entire argument is. Lower volatility MIGHT mean lower drawdowns but that will be period-dependent of course.
nisiprius wrote: Wed May 25, 2022 1:13 pm A "straight" S&P fund, with its risk diluted with cash, had exactly the same standard deviation as the low-vol fund. But it had somewhat higher return, and meaningfully lower drawdown.
No, it had lower returns once you equalize for St Dev and you use all of the PV data.:
https://www.portfoliovisualizer.com/bac ... n4_2=17.33
nisiprius wrote: Wed May 25, 2022 1:13 pm There was nothing wrong with SPLV, but over the time period shown, it looks like complication without compelling evidence of any practical benefit.
I think there's plenty wrong with SPLV, it lags its own low-vol index by 0.3-0.5% each year. I just do not like Invesco funds in general (I'm a Vanguard and iShares gal). I prefer the lower fees of iShares (like USMV) but you'll complain that I'm just cherry-picking ETFs so we can stick to Invesco funds I suppose.
Dry-Drink
Posts: 530
Joined: Fri Jan 14, 2022 9:50 am

Re: Is Low Volatility the best factor?

Post by Dry-Drink »

MaxLeo wrote: Sat Oct 19, 2019 12:29 pm If you have a look at the Sharpe-Ratio of the MSCI USA Factors, you will notice that Low Volatility has the best Sharpe-Ratio over all three listed periods (3 year, 5 year and 10 year)

Do you think that it is the best of all different factors?
IMO, it's the best one. It is so intuitive to me, and so clear, especially in the bond market where a long-term bond could have 10x the volatility but barely 10% more return. However, the edge is relatively small, I really do think you have to harvest it in long-short fashion (ideally across multiple asset classes), with multiples of leverage (maybe 5-10x) to actually get it to an acceptable risk/return level to include. That is, the Sharpe is really high (like 1.0+), but the actual returns and risks so tiny that long-short, highly-leveraged, multi-asset is the only way it makes sense to me.

In long-only, unleveraged fashion, I just couldn't care less for it.
User avatar
nisiprius
Advisory Board
Posts: 52219
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: Is Low Volatility the best factor?

Post by nisiprius »

Dry-Drink wrote: Wed May 25, 2022 1:47 pm...Why are you cutting the PV data range by choosing the yearly data? If you use all available PV (monthly) data, SPLV had higher Sharpe:...
My bad. You are correct. I modified InvestInPasta's PortfolioVisualizer inputs without noticing they had chosen whole-calendar-year instead of monthly data. "All available data" is IMHO the appropriate default choice, and the extra six months of data is enough to tip the balance in favor of the low-vol fund.
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.
Northern Flicker
Posts: 15368
Joined: Fri Apr 10, 2015 12:29 am

Re: Is Low Volatility the best factor?

Post by Northern Flicker »

nisiprius wrote: Also, it seems odd but as nearly as I can tell, cribbing from the models provided in PortfolioVisualizer,

"low volatility" does not emerge as a named factor in

--the Fama-French three-factor model (market, size, value)
--the Fama-French five-factor model (market, size, value, investment, profitability)
--the AQR four-factor model (market, size, value, momentum)
The AQR model has the BAB (bet against beta) factor. There is controversy about whether low volatility is a legitimate equity risk factor, but it does exist in some models. Funds like USMV are based on portfolio optimization, not on a low vol factor model as far as I know, but VFMV seems to be built using a low vol factor.
InvestInPasta
Posts: 213
Joined: Sat Sep 16, 2017 12:42 pm
Location: Italy

Re: Is Low Volatility the best factor?

Post by InvestInPasta »

Dry-Drink wrote: Wed May 25, 2022 1:47 pm No, it had lower returns once you equalize for St Dev and you use all of the PV data.:
https://www.portfoliovisualizer.com/bac ... n4_2=17.33
Ok, thanks, I did not even know I could use portfoliovisualizer month to month.
I looked at the chart, couple of thoughts:
  • Even if the Stdev is the same it amazes me that the MaxDD is still a bit higher for the Low Vol, I understand it's called Low Vol and not Low DrawDown.
  • How can someone take advantage of the Low/Min Volatility? Let's say I have a portfolio that is 50% SP500 and 50% Intermediate Treasury, and I replace SP500 part with the Low/Min Volatility version. Basically it would be similar to reduce the SP500 to 44% and just keep 6% in cash.
    But a portfolio with 6% kept in cash is a better solution, I would not need anymore an emergency fund.
Image
Simulation here
Last edited by InvestInPasta on Thu May 26, 2022 4:22 am, edited 2 times in total.
When I study English I am lazier than my portfolio. Feel free to fix my English and investing mistakes.
InvestInPasta
Posts: 213
Joined: Sat Sep 16, 2017 12:42 pm
Location: Italy

Re: Is Low Volatility the best factor?

Post by InvestInPasta »

Dry-Drink wrote: Wed May 25, 2022 1:47 pm I think there's plenty wrong with SPLV, it lags its own low-vol index by 0.3-0.5% each year. I just do not like Invesco funds in general (I'm a Vanguard and iShares gal). I prefer the lower fees of iShares (like USMV)
USMV is based on "MSCI Min Vol" index, that is not the same of "Low Vol" index (SPLV)
For what is worth according to the paper below "Low Vol" strategy should achieve more volatility reduction than "Min Vol".
Forester wrote: Sun Jan 24, 2021 11:14 am Great recent white paper from State Street on low volatility vs minimum volatility; https://www.ssga.com/library-content/st ... arison.pdf
- here is the discussion: low vol vs min vol
When I study English I am lazier than my portfolio. Feel free to fix my English and investing mistakes.
User avatar
nisiprius
Advisory Board
Posts: 52219
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: Is Low Volatility the best factor?

Post by nisiprius »

MaxLeo wrote: Sat Oct 19, 2019 12:29 pm If you have a look at the Sharpe-Ratio of the MSCI USA Factors, you will notice that Low Volatility has the best Sharpe-Ratio over all three listed periods (3 year, 5 year and 10 year)...
Choosing an investment by running down a table of 3, 5, and 10-year returns is not a good way to pick investments.

And doing the same thing for Sharpe ratios isn't, either. It's nice that MSCI gives them and avoids at least one trap, that of ignoring risk.

But the back-and-forth within this very thread shows you just how fickle and endpoint-dependent these things are.

InvestInPasta noted that the SPLV ETF has had a lower Sharpe ratio than plain SPY (S&P 500 ETF) over the time period Jan 2012 - Apr 2022. InvestInPasta assumed that they were showing SPLV back to inception.

And then Dry-Drink pointed out that the default setting for PortfolioVisualizer only starts on exact calendar year beginnings, and if you set it to display "month-to-month" it will show a slightly longer period, six more months, Jun 2011 - Apr 2022.

Sharpe ratios:
From Jan 2012 to April 2022, SPLV 0.98, SPY 1.02. SPY was higher.
From Jun 2011 to April 2022, SPLV 0.97, SPY 0.91. SPLV was higher.

A six month change in starting date was enough to flip the comparison.
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.
Dry-Drink
Posts: 530
Joined: Fri Jan 14, 2022 9:50 am

Re: Is Low Volatility the best factor?

Post by Dry-Drink »

InvestInPasta wrote: Thu May 26, 2022 4:19 am
Dry-Drink wrote: Wed May 25, 2022 1:47 pm No, it had lower returns once you equalize for St Dev and you use all of the PV data.:
https://www.portfoliovisualizer.com/bac ... n4_2=17.33
Ok, thanks, I did not even know I could use portfoliovisualizer month to month.
I looked at the chart, couple of thoughts:
  • Even if the Stdev is the same it amazes me that the MaxDD is still a bit higher for the Low Vol, I understand it's called Low Vol and not Low DrawDown.
  • How can someone take advantage of the Low/Min Volatility? Let's say I have a portfolio that is 50% SP500 and 50% Intermediate Treasury, and I replace SP500 part with the Low/Min Volatility version. Basically it would be similar to reduce the SP500 to 44% and just keep 6% in cash.
    But a portfolio with 6% kept in cash is a better solution, I would not need anymore an emergency fund.
Image
Simulation here
The use case is a straight-replacement for your total stock index fund, the index+cash comparison isn’t meant to be the replacement, only a way to show the effect is there.

In that replacement, the investor is essentially saying “hey, I’m happy making a little less on really good market and, instead, I get a bigger reduction in volatility. That means less fluctuations and generally less drawdowns”. Lower drawdowns aren’t ASSURED, but they’re more likely.

I wouldn’t complicate and would stick with VTI but I get what others would do it. The effect is there and it’s not totally trivial.
Dry-Drink
Posts: 530
Joined: Fri Jan 14, 2022 9:50 am

Re: Is Low Volatility the best factor?

Post by Dry-Drink »

InvestInPasta wrote: Thu May 26, 2022 4:20 am
Dry-Drink wrote: Wed May 25, 2022 1:47 pm I think there's plenty wrong with SPLV, it lags its own low-vol index by 0.3-0.5% each year. I just do not like Invesco funds in general (I'm a Vanguard and iShares gal). I prefer the lower fees of iShares (like USMV)
USMV is based on "MSCI Min Vol" index, that is not the same of "Low Vol" index (SPLV)
For what is worth according to the paper below "Low Vol" strategy should achieve more volatility reduction than "Min Vol".
Forester wrote: Sun Jan 24, 2021 11:14 am Great recent white paper from State Street on low volatility vs minimum volatility; https://www.ssga.com/library-content/st ... arison.pdf
- here is the discussion: low vol vs min vol
I’m aware of the differences, I very much prefer a Min Vol approach than Low Vol approach because, in general, the former will be less volatile and less sector-concentrated. That means more exposure to the low vol phenomenon.

Since inception, USMV has had higher returns AND lower volatility than SPLV. Some of that is luck, but I do attribute some to a superior methodology.

Just my thoughts
Northern Flicker
Posts: 15368
Joined: Fri Apr 10, 2015 12:29 am

Re: Is Low Volatility the best factor?

Post by Northern Flicker »

Dry-Drink wrote: Thu May 26, 2022 11:30 am
InvestInPasta wrote: Thu May 26, 2022 4:20 am
Dry-Drink wrote: Wed May 25, 2022 1:47 pm I think there's plenty wrong with SPLV, it lags its own low-vol index by 0.3-0.5% each year. I just do not like Invesco funds in general (I'm a Vanguard and iShares gal). I prefer the lower fees of iShares (like USMV)
USMV is based on "MSCI Min Vol" index, that is not the same of "Low Vol" index (SPLV)
For what is worth according to the paper below "Low Vol" strategy should achieve more volatility reduction than "Min Vol".
Forester wrote: Sun Jan 24, 2021 11:14 am Great recent white paper from State Street on low volatility vs minimum volatility; https://www.ssga.com/library-content/st ... arison.pdf
- here is the discussion: low vol vs min vol
I’m aware of the differences, I very much prefer a Min Vol approach than Low Vol approach because, in general, the former will be less volatile and less sector-concentrated. That means more exposure to the low vol phenomenon.
Actually, SPLV has had a higher exposure to a low vol factor than USMV, and that likely is the source of its underperformance so far relative to USMV:

https://www.portfoliovisualizer.com/fac ... sion=false
InvestInPasta
Posts: 213
Joined: Sat Sep 16, 2017 12:42 pm
Location: Italy

Re: Is Low Volatility the best factor?

Post by InvestInPasta »

Dry-Drink wrote: Thu May 26, 2022 11:05 am The use case is a straight-replacement for your total stock index fund, the index+cash comparison isn’t meant to be the replacement, only a way to show the effect is there.

In that replacement, the investor is essentially saying “hey, I’m happy making a little less on really good market and, instead, I get a bigger reduction in volatility. That means less fluctuations and generally less drawdowns”. Lower drawdowns aren’t ASSURED, but they’re more likely.
Ok, thanks!
When I study English I am lazier than my portfolio. Feel free to fix my English and investing mistakes.
InvestInPasta
Posts: 213
Joined: Sat Sep 16, 2017 12:42 pm
Location: Italy

Re: Is Low Volatility the best factor?

Post by InvestInPasta »

Northern Flicker wrote: Fri May 27, 2022 10:11 pm Actually, SPLV has had a higher exposure to a low vol factor than USMV, and that likely is the source of its underperformance so far relative to USMV:

https://www.portfoliovisualizer.com/fac ... sion=false
Is it "Bet Against Beta (BAB)" the factor we are looking at?
When I study English I am lazier than my portfolio. Feel free to fix my English and investing mistakes.
MatthewLM
Posts: 47
Joined: Sun Jul 04, 2021 7:10 am

Re: Is Low Volatility the best factor?

Post by MatthewLM »

Why only look at the US market for low-vol? Isn't there any world low-vol/min-vol ETF available in the US? Here in the UK I use the iShares MSCI World Minimum Volatility ETF (MINV).

I like it for providing exposures (albeit time-varying) to profitability and investment factors which is hard to find otherwise. However low/min-vol is not completely explained by the FF5 factors and may provide some of its own benefits. By itself it has historically outperformed with lower risk. Not only with lower volatility but smaller drawdowns too. It wont always keep up in a strong bull market, but it more than compensates during bear markets.
Northern Flicker
Posts: 15368
Joined: Fri Apr 10, 2015 12:29 am

Re: Is Low Volatility the best factor?

Post by Northern Flicker »

InvestInPasta wrote: Sat May 28, 2022 6:23 am
Northern Flicker wrote: Fri May 27, 2022 10:11 pm Actually, SPLV has had a higher exposure to a low vol factor than USMV, and that likely is the source of its underperformance so far relative to USMV:

https://www.portfoliovisualizer.com/fac ... sion=false
Is it "Bet Against Beta (BAB)" the factor we are looking at?
BAB is a low vol factor (low beta). SPLV holds the 100 stocks in the S&P500 with the lowest volatility. This is not equivalent to the 100 stocks with the lowest beta, but there is considerable overlap. If the index tracked by SPLV has no constraints on sector balance, that also would be a source of performance deviation from USMV.

Min vol products minimize portfolio beta or volatility subject to a set of constraints. This allows the correlation or lack thereof of different stocks to be considered implicitly. Thus, min vol is based on modern portfolio theory and is a top-down model, while low vol is a factor-based, bottom-up model.

This article is interesting.

https://www.blackrock.com/us/individual ... volatility

Here is an article on interest-rate sensitivity of low vol stocks:

https://www.man.com/maninstitute/low-vol
Last edited by Northern Flicker on Sun May 29, 2022 1:06 pm, edited 3 times in total.
selters
Posts: 702
Joined: Thu Feb 27, 2014 8:26 am

Re: Is Low Volatility the best factor?

Post by selters »

If I were to deviate from straight indexing and invest in one factor for the rest of my investing lifetime, it would have to be some kind of quality factor. MSCIs quality indexes have been steady outperformers in pretty much any region over pretty much any Five year time horizon.
Post Reply