Instead of bonds; why not 100% Low Vol?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Forester
Posts: 135
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Instead of bonds; why not 100% Low Vol?

Post by Forester » Mon Jun 10, 2019 1:31 pm

MSCI US Minimum Volatility https://www.msci.com/documents/10199/f5 ... 761d009094
2008 -26.20%
2011 +11.94%
2015 +4.92%
2018 +0.18%

US 70/30 portfolio https://www.portfoliovisualizer.com/bac ... 0&total3=0
2008 -23.54%
2011 +3.76%
2015 +1.03%
2018 -2.94%

---

Since 2012, iShares USMV vs 60/40: https://www.portfoliovisualizer.com/bac ... 0&total3=0

mhalley
Posts: 7012
Joined: Tue Nov 20, 2007 6:02 am

Re: Instead of bonds; why not 100% Low Vol?

Post by mhalley » Mon Jun 10, 2019 2:46 pm

Utility stocks are not bonds. Low volatility stocks are not bonds. High dividend yield stocks are not bonds. Preferred stocks are not bonds.
In summary, Stocks are not bonds.
Unless you can do some backtesting to 1900, then I might rethink it.
Last edited by mhalley on Mon Jun 10, 2019 2:47 pm, edited 1 time in total.

ThrustVectoring
Posts: 740
Joined: Wed Jul 12, 2017 2:51 pm

Re: Instead of bonds; why not 100% Low Vol?

Post by ThrustVectoring » Mon Jun 10, 2019 2:46 pm

The lack of correlation between low volatility or bet-against-beta strategies tends to disappear during times of crisis (eg, 2008). The actual amount of allocation or leverage you can use is less than what the data from normal periods suggests, and then you're left holding a bunch of lower-yielding assets.

This is fundamentally the same reason why selling volatility doesn't help you as much as the naive math suggests. You're taking on more downside risk in precisely those situations where downside risk can ruin you, which reduces how aggressively you can chase returns. Meanwhile, the funds you're using have more turnover and higher management fees.

(There's also a high level signal-processing argument that you want to amplify (have a higher allocation with possible leverage) the signals (assets) with the best signal-to-noise ratio (least volatility compared to expected return). This means that the theoretical arguments for having a higher allocation to safer but lower-yielding assets are a better argument for levering up on short-term bonds, rather than algorithmically picking "boring" stocks to have more of. But that all is a pretty big sidetrack that I'm honestly not competent enough to explain fully.)
Current portfolio: 60% VTI / 40% VXUS

User avatar
nisiprius
Advisory Board
Posts: 38309
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: Instead of bonds; why not 100% Low Vol?

Post by nisiprius » Mon Jun 10, 2019 3:26 pm

Because the low volatility of bonds is a direct result of their fundamental nature, and can be expected to be robust and reliable going forward, while the low volatility of stocks is just an empirical observation--according to the S&P 500 methodology the stocks are weighted simply according to what their daily standard deviation was over the last year.

Bonds are business contracts, enforceable in a court of law, to pay specified numbers of dollars on specified days. Low-volatility stocks are just stocks that seem to have had low volatility. They don't represent any kind of promise or expectation set by the stock-issuing company. "Low volatility" doesn't even have the implied expectation of "regular dividend-paying stocks."

It is noteworthy that you couldn't use any real-world examples of low volatility mutual funds and ETFs. I think that's because there weren't any in existence during 2008-2009. So, yes, you have numbers and backtested data, but you don't have the real-world discipline of doing it out in the open, transparently, running real money, with published lists of holdings, and real-world expenses.

So, even assuming that 100% low volatility has had about the same risk and return as 70/30, what is the point? Is it just some kind of tour-de-force to show that you can do without bonds?

I see that there are already articles saying things like "low-volatility [is/is not] becoming a crowded trade." That would give me pause, as there have been so many examples of things attracting attention through calculated backtested indexes, and then having sharply disappointing results once mutual funds and ETFs based on them began to attracted real-world retail investors.
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: 4215
Joined: Fri Apr 10, 2015 12:29 am
Location: Taking a break from Bogleheads

Re: Instead of bonds; why not 100% Low Vol?

Post by Northern Flicker » Mon Jun 10, 2019 3:53 pm

It is reasonable to hold a low volatility equity portfolio and increase equity exposure some. Low-vol incorporates some term exposure compensating for the reduced term exposure from reducing one’s bond exposure some.

I think that there are two challenges to overcome with this approach: 1) such an equity portfolio may require enough industry sector tilting that it becomes exposed to some uncompensated sector risk; 2) the most diversified min-vol products have enough portfolio turnover that they may not be very tax-efficient in a taxable account. Perhaps an ETF structure will be sufficient to manage the capital gains.

The sector risk should not be underestimated. A min-vol portfolio typically will rise less than the broad market during a bull market, but any overweighted sector(s) in the min-vol portfolio could crash harder in a bear market if the drivers of the bear market are essential to those sectors (e.g. banks and home builders in 2008/2009, utilities in the 1970’s). Moreover, portfolio turnover means that the sector diversification is not static.
Last edited by Northern Flicker on Wed Jun 12, 2019 4:51 am, edited 2 times in total.

pdavi21
Posts: 1061
Joined: Sat Jan 30, 2016 4:04 pm

Re: Instead of bonds; why not 100% Low Vol?

Post by pdavi21 » Mon Jun 10, 2019 4:20 pm

1. A low volatility fund isn't going follow the low volatility factor predictably and the performance of the low volatility funds are likely to diverge over time based on Index/Method/Expenses.
2. I don't think the low volatility factor is statistically significant according to the factor zoo. I think only momentum (new mined factor), quality (new mined factor), and to a lesser extent, value (one of the originals) have a reasonable statistical significance.
3. You are unlikely to have the conviction to hold a stock tilt for a long period of time if you are having trouble settling on a bond percentage (arguably most important investment decision).
4. Low volatility loses out on the diversification benefit of holding the entire market. The goal is to lower risk and return, so the outcome is unlikely to be a higher risk-adjusted return than a 100% stock portfolio.
5. VFMV only holds 200 securities and is not market cap weighted. Tootsie Roll Industries has a higher weight than Johnson & Johnson, for example.
6. VFMV only has $36 million AUM, so it has low liquidity and will probably die within a decade if performance ends up being poor.
"We spend a great deal of time studying history, which, let's face it, is mostly the history of stupidity." -Stephen Hawking

User avatar
vineviz
Posts: 4088
Joined: Tue May 15, 2018 1:55 pm

Re: Instead of bonds; why not 100% Low Vol?

Post by vineviz » Mon Jun 10, 2019 5:20 pm

nisiprius wrote:
Mon Jun 10, 2019 3:26 pm
It is noteworthy that you couldn't use any real-world examples of low volatility mutual funds and ETFs. I think that's because there weren't any in existence during 2008-2009. So, yes, you have numbers and backtested data, but you don't have the real-world discipline of doing it out in the open, transparently, running real money, with published lists of holdings, and real-world expenses.
Several money managers implemented such strategies before 2008, including Acadian, Analytics Investors, Robeco, State Street, and Unigestion. MSCI launched its index in early 2008 (which seems prescient with the benefit of hindsight) as a response to "real-world" demand, not as a stimulant for it.

Acadian's Global Managed Volatility Strategy, launched in 2006, has outperformed Vanguard Global Equity Fund (VHGEX) by over 1.2% per year with 37% lower volatility for example.

While I wouldn't consider 100% minimum volatility to have the same expected volatility as an old-fashioned 60/40 portfolio I'd be very comfortable with investors including minimum volatility funds in a portfolio.

For instance, a 75/25 portfolio using USMV might replace a 60/40 portfolio using VTI.

Image
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

randomguy
Posts: 7698
Joined: Wed Sep 17, 2014 9:00 am

Re: Instead of bonds; why not 100% Low Vol?

Post by randomguy » Mon Jun 10, 2019 6:22 pm

nisiprius wrote:
Mon Jun 10, 2019 3:26 pm

So, even assuming that 100% low volatility has had about the same risk and return as 70/30, what is the point? Is it just some kind of tour-de-force to show that you can do without bonds?
Historically though you have had the same risk but higher returns. Larry did an article a while back on this (https://www.etf.com/sections/index-inve ... or-caveats). In then end, you are going to get into some variation of factor investing (low volatility is more a combo of a couple of factors than a pure factor) which requires the investor to have faith in it.

Personally I think you are off in crazy land if you 100% on something like this. Now maybe going 40% VTI, 30% Low vol, 30% bonds instead of 50/50 total/bonds might work out to having similar volatility and hopefully higher returns. Note I didn't actual run the match to see figure out what the exact split should be. I have faith in factor investing but my faith is a bit limited:)

ThrustVectoring
Posts: 740
Joined: Wed Jul 12, 2017 2:51 pm

Re: Instead of bonds; why not 100% Low Vol?

Post by ThrustVectoring » Tue Jun 11, 2019 12:50 pm

randomguy wrote:
Mon Jun 10, 2019 6:22 pm
nisiprius wrote:
Mon Jun 10, 2019 3:26 pm

So, even assuming that 100% low volatility has had about the same risk and return as 70/30, what is the point? Is it just some kind of tour-de-force to show that you can do without bonds?
Historically though you have had the same risk but higher returns. Larry did an article a while back on this (https://www.etf.com/sections/index-inve ... or-caveats). In then end, you are going to get into some variation of factor investing (low volatility is more a combo of a couple of factors than a pure factor) which requires the investor to have faith in it.

Personally I think you are off in crazy land if you 100% on something like this. Now maybe going 40% VTI, 30% Low vol, 30% bonds instead of 50/50 total/bonds might work out to having similar volatility and hopefully higher returns. Note I didn't actual run the match to see figure out what the exact split should be. I have faith in factor investing but my faith is a bit limited:)
Low-vol isn't a free lunch. You're taking on more extreme-downside risk and "unknown unknowns", much like someone replacing an index fund holding with a put-write or buy-write volatility selling strategy.
Current portfolio: 60% VTI / 40% VXUS

Post Reply