Simple-minded inspection of actual low-vol fund

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39459
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

Simple-minded inspection of actual low-vol fund

Post by nisiprius » Thu Mar 29, 2018 8:50 am

Before I post my usual simple-minded comparison of an actual "low volatility stock" mutual fund to Total Stock--with particular emphasis on whether "low volatility" stocks had low volatility in 2008-2009--I want to ask for suggestions of any other funds to look at besides the Invesco Low Volatility Equity Yield fund, SCAUX. Which, of course, I have already looked at. And it was unimpressive, to put it mildly. However, whenever I test a strategy by looking at the real-world results of a real-world fund that embodied that strategy, I'm apt to get told that I picked The Wrong Fund.

I am looking for

a) a mutual fund (or ETF) with an inception date of 12/31/2007 or earlier, that

b) can be considered a "fair representative specimen" of low-volatility stocks, i.e. a competently managed fund, by managers who seem to be doing what the name of the fund implies.

SCAUX has an inception date of 03/31/2006, which is fine for my purposes. On the other hand, Vanguard Global Minimum Volatility had inception on 12/13/2013 and therefore did not even exist during the 2011 correction, let alone 2008-2009.

It may be clear to connoisseurs that one shouldn't expect "low volatility" funds to provide downside protection in crashes or corrections, but merely an incremental improvement in risk-adjusted return due to exploitation of a claimed "low-volatility anomaly" or "low-volatility premium"--that is, it is similar to a claim often made for dividend stocks--but given the name "low volatility" I think it is interesting to see what they really do in downturns.
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.

bloom2708
Posts: 6968
Joined: Wed Apr 02, 2014 2:08 pm
Location: Fargo, ND

Re: Simple-minded inspection of actual low-vol fund

Post by bloom2708 » Thu Mar 29, 2018 1:19 pm

Would Vanguard Dividend Growth (VDIGX) fall into the category?

PowerShares S&P 500 Low Volatility Portfolio (SPLV)
Russell 1000 Low Volatility ETF (LVOL)
iShares MSCI USA Minimum Volatility Index Fund (USMV)

Low volatility stock fund seems to be a misnomer or an oxy-moron. Selectively, some will move more than others. Picking which ones, that it hard.

Not sure if any of these funds have been around long enough. I here what you are saying though.
"People want confirmation, not advice" Unknown | "We are here to provoke thoughtfulness, not agree with you" Unknown | Four words. Whole food, plant based. Bing it.

User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39459
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: Simple-minded inspection of actual low-vol fund

Post by nisiprius » Thu Mar 29, 2018 1:27 pm

bloom2708 wrote:
Thu Mar 29, 2018 1:19 pm
Would Vanguard Dividend Growth (VDIGX) fall into the category?
I don't think so, "dividend growth" is a traditional category of its own. As for the others:

PowerShares S&P 500 Low Volatility Portfolio (SPLV), 5/5/2011
Russell 1000 Low Volatility ETF (LVOL), 2/29/2012
iShares MSCI USA Minimum Volatility Index Fund (USMV) 10/18/2011
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.

alex_686
Posts: 5097
Joined: Mon Feb 09, 2015 2:39 pm

Re: Simple-minded inspection of actual low-vol fund

Post by alex_686 » Thu Mar 29, 2018 1:41 pm

bloom2708 wrote:
Thu Mar 29, 2018 1:19 pm
Would Vanguard Dividend Growth (VDIGX) fall into the category?
No. Dividend Growth is not a good expression of any of the 4 factors - low vol, momentum, small, or value. If I had to pick one I would say that Dividend Growth was more closely associated with value, which is on the opposite extreme of low vol.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Simple-minded inspection of actual low-vol fund

Post by lack_ey » Thu Mar 29, 2018 4:50 pm

Well, there's a bit of a range of low volatility, defensive, min volatility, and related approaches.
nisiprius wrote:
Thu Mar 29, 2018 8:50 am
I am looking for

a) a mutual fund (or ETF) with an inception date of 12/31/2007 or earlier, that

b) can be considered a "fair representative specimen" of low-volatility stocks, i.e. a competently managed fund, by managers who seem to be doing what the name of the fund implies.
I think the best match for what you're looking for is Guggenheim Defensive Equity ETF (DEF). It's had a mid-high 0.60%+ expense ratio all this time.

Fund page:
https://www.guggenheiminvestments.com/e ... equity-etf
Effective October 24, 2016, Guggenheim Defensive Equity ETF (DEF) changed its benchmark index to the Guggenheim Defensive Equity Index from the Sabrient Defensive Equity Index. The ETF ticker and name did not change. Read Frequently Asked Questions (FAQs) for more information.
But it's had a similar concept through time.

The 2007 annual report says this:
The Claymore/Sabrient Defender ETF (the “Fund”) seeks investment results that correspond generally to the performance, before the Fund’s fees and expenses, of an equity index called the Sabrient Defensive Equity Index (the “Defensive Equity Index” or “Index”). The Fund will normally invest at least 90% of its total assets in common stock and American depositary receipts (“ADRs”) that comprise the Index. Claymore Advisors, LLC, the Fund’s investment adviser (the “Investment Adviser”), seeks a correlation over time of 95% or better between the Fund’s performance and the performance of the Index.

The Defensive Equity Index selection methodology is designed to identify companies with potentially superior risk/return profiles, as determined by Sabrient Systems, LLC (“Sabrient” or the “Index Provider”), during periods of weakness in the markets and/or the American economy overall. The Index is designed to actively represent a group of stocks that reflect occurrences such as low relative valuations, conservative accounting, dividend payments and a history of out-performance during bearish market periods. The Index constituents represent a “defensive” portfolio with the potential to outperform the S&P 500 Index and other broad market benchmark indices on a risk-adjusted basis during periods of market weakness, while still providing the potential for positive returns during strong market periods.
So it should work, at least with the understanding that one fund does not generalize to the entire category perfectly, and there are various different ways to define and select a low-volatility (or min vol, or defensive, etc.) portfolio.

nisiprius wrote:
Thu Mar 29, 2018 8:50 am
SCAUX has an inception date of 03/31/2006, which is fine for my purposes.
It shouldn't be fine in the sense that you could say that its "inception" as a low volatility fund was something like 2013. In the 2012 annual report, SCAUX was known as "Invesco U.S. Quantitative Core Fund" and was intentionally trying to target similar beta as the market.

If you look at more recent factor exposures, it's something of a low vol fund in name and behavior; prior to that, it was not. It's of no use to us here.

nisiprius wrote:
Thu Mar 29, 2018 8:50 am
It may be clear to connoisseurs that one shouldn't expect "low volatility" funds to provide downside protection in crashes or corrections, but merely an incremental improvement in risk-adjusted return due to exploitation of a claimed "low-volatility anomaly" or "low-volatility premium"--that is, it is similar to a claim often made for dividend stocks--but given the name "low volatility" I think it is interesting to see what they really do in downturns.
You think?

Low volatility constructions almost always overall should have lower beta and should pick stocks more likely to have lower future volatility. This creates a portfolio that should lose less on average during crashes or corrections, relative to the market. Not in every market crash, depending on circumstances and particularly depending on if there are constraints on sector weightings. In theory you could have low-vol stocks but a market crash that is especially bad for certain stocks and sectors. But reasonably, average behavior should be less downside and upside. The low-volatility anomaly is just an observation that the return for these has in the long run been higher than you'd expect based on the market beta (i.e. you get CAPM alpha once you account for the market beta).

User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39459
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: Simple-minded inspection of actual low-vol fund

Post by nisiprius » Thu Mar 29, 2018 8:37 pm

So, using DEF as my fair representative of "low vol" stocks, I find that (unlike SCAUX!) compared to VTSMX, it indeed had meaningfully lower volatility measured by standard deviation, lower drawdown during 2008-2009, and a higher Sharpe ratio. The stocks selected by the managers really did have lower volatility than the total market.

So the next question is: does it do anything that couldn't have been done just as well by adding a bond allocation? To test this, I used a portfolio of Total Stock (VTSAX) and Total Bond (VBMFX) and hand-adjusted the allocation by trial and error until the standard deviation matched that of DEF to within roundoff error. If the "low volatility anomaly" manifested itself during the lifetime of DEF, we would hope to see a higher return from DEF than from the equivalent-risk stock-bond portfolio.

Source
Image

The return from DEF was indeed a hair higher--14 basis points--and the Sharpe ratio a hair higher, but I don't view it as meaningful. In three other respects--best year, worst year, and drawdown--DEF was a hair worst.In reality, for practical purposes, "low volatility" achieved by stock choice do anything different from "low volatility" achieved by addition of bonds.

To take a more realistic example, let's ask how much if any benefit we would see substituting DEF for Total Stock in a three-fund portfolio with a moderate stock allocation. Portfolio 1, blue, will be a 60/40 stock allocation with 30% international within the stocks, using DEF; i.e. 42% DEF, 18% Total International (VTIAX); and 40% Total Bond (VBTLX). For portfolio 2, red, we will substitute Total Stock, VTSAX in place of DEF. Since it has higher volatility, we will again hand-adjust the allocation of VTSAX and VBTLX until the portfolio 2 closely matches Portfolio 1 in standard deviation.

Source
Image

As in the previous example, there are hair-thin differences in return and Sharpe ratio, favoring the portfolio with DEF, and hair-thin differences in best year, worst year, and drawdown, favoring the portfolio with Total Stock. But I would say the differences are not meaningful.

In short, in this second case I saw no obvious visible benefit to using stocks selected for low volatility, compared to using a total stock market portfolio and simply increasing the bond allocation.
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
Topic Author
nisiprius
Advisory Board
Posts: 39459
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: Simple-minded inspection of actual low-vol fund

Post by nisiprius » Thu Mar 29, 2018 8:48 pm

lack_ey, what's a good place to check for leopards that have changed their spots, i.e. SCAUX being rechristened "low volatility" and changing strategy and aims? I thought Morningstar tried to draw vertical dotted lines at such discontinuities but I guess they don't always do it.
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.

AlohaJoe
Posts: 4935
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Simple-minded inspection of actual low-vol fund

Post by AlohaJoe » Thu Mar 29, 2018 8:54 pm

Robeco was one of the first to champion minimum volatility. Most of their funds launched in 2006. They're a European firm, though, so not sure if doing a comparison would be easy. MSCI set up a minimum volatility index in 2008.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Simple-minded inspection of actual low-vol fund

Post by lack_ey » Thu Mar 29, 2018 9:27 pm

nisiprius wrote:
Thu Mar 29, 2018 8:37 pm
So the next question is: does it do anything that couldn't have been done just as well by adding a bond allocation?
It did, as a strategy, just not as much when saddled with an ER of around 0.65% over the period. If let's say Vanguard or iShares were running it at 0.15% through the last ten years (USMV is 0.15%, VFMV is 0.13%), I think the conclusion would look a little bit different.

Overall this was a pretty good period for low-volatility stocks, to the point that in the last few years they have looked a bit overbought if anything, with valuations pushed up.

But it should also be understood that term risk was an exceptionally good diversifier also with high Sharpe ratio (and then significantly negative correlation with stocks)—any stock/bond combination looks relatively good from a risk/return perspective evaluated over this period. The diversification benefit will generally not be as large.

That said, of the low-vol "anomaly" can be attributed to some effective loading on term risk, but it doesn't explain all of it or the results in the totality over this period.
nisiprius wrote:
Thu Mar 29, 2018 8:48 pm
lack_ey, what's a good place to check for leopards that have changed their spots, i.e. SCAUX being rechristened "low volatility" and changing strategy and aims? I thought Morningstar tried to draw vertical dotted lines at such discontinuities but I guess they don't always do it.
I check the old fund filings via the SEC site, though I've never had much luck searching the site itself.

Just run something like this through Google:

Code: Select all

site:sec.gov edgar guggenheim defensive def
First link is the page we want -> click on Series S000014618 corresponding to the ETF -> search for "Certified Shareholder Report" and click on "Documents" (hit next 40 as necessary to look for older docs) -> click on the largest html file in the list

CedarWaxWing
Posts: 739
Joined: Sun Nov 02, 2014 12:24 pm

Re: Simple-minded inspection of actual low-vol fund

Post by CedarWaxWing » Thu Mar 29, 2018 10:49 pm

YACKX is portfolio 3.

Portfolio Returns
Portfolio Initial Balance Final Balance CAGR Stdev Best Year Worst Year Max. Drawdown Sharpe Ratio Sortino Ratio US Mkt Correlation
Portfolio 1 $10,000 $23,595 7.93% 12.16% 22.41% -30.34% -43.36% 0.63 0.87 0.88
Portfolio 2 $10,000 $23,121 7.73% 12.18% 26.96% -29.24% -41.97% 0.62 0.90 1.00
Portfolio 3 $10,000 $28,675 9.82% 14.43% 59.31% -26.05% -41.01% 0.67 1.11 0.89

I don't think this satisfies your search criteria exactly, but it is still interesting. Although StDEV was higher, that in spite of the fact that its Worst year was a little better than the other two, but its best year was much better. No bonds. I suspect one of the ways they did so well was that they held cash instead of bonds, and invested it after things dropped so much in 2007 and 2008.

M

smesman
Posts: 97
Joined: Thu Apr 27, 2017 7:11 am

Re: Simple-minded inspection of actual low-vol fund

Post by smesman » Fri Mar 30, 2018 2:04 am

nisiprius wrote:
Thu Mar 29, 2018 8:37 pm
In short, in this second case I saw no obvious visible benefit to using stocks selected for low volatility, compared to using a total stock market portfolio and simply increasing the bond allocation.
Thank you nispirius, it is always enlightening to see such graphs.

The question is whether their performance was similar only by chance or not. One major advantage of bonds during this period were the declining interest rates. Do low volatility funds have the same sensitivity to interest rates? On the otherr gand, low vol also has had a giant influx of money the last years which have surely attributed to its returns.

User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39459
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: Simple-minded inspection of actual low-vol fund

Post by nisiprius » Fri Mar 30, 2018 6:43 am

CedarWaxWing wrote:
Thu Mar 29, 2018 10:49 pm
YACKX is portfolio 3.
According to the fund company, YACKX
Provides investors exposure to a portfolio of U.S. stocks which the manager believes are undervalued based on three primary characteristics; good business, shareholder-oriented management and low purchase price.
That, to me is not a "low volatility fund." It is an actively-managed fund based on a manager's stock-picking ability. "Volatility" isn't even mentioned as an important selection criterion, and the name of the fund does not reflect anything at all about the portfolio, it reflects the name of the star manager. The behavior of the fund might tell us something about Stephen A. Yacktman, but I don't think it tells us anything about low volatility funds or the real-life importance of the "low volatility anomaly."
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.

Valuethinker
Posts: 39065
Joined: Fri May 11, 2007 11:07 am

Re: Simple-minded inspection of actual low-vol fund

Post by Valuethinker » Fri Mar 30, 2018 6:54 am

alex_686 wrote:
Thu Mar 29, 2018 1:41 pm
bloom2708 wrote:
Thu Mar 29, 2018 1:19 pm
Would Vanguard Dividend Growth (VDIGX) fall into the category?
No. Dividend Growth is not a good expression of any of the 4 factors - low vol, momentum, small, or value. If I had to pick one I would say that Dividend Growth was more closely associated with value, which is on the opposite extreme of low vol.
I agree that High Yield funds are very similar to value funds.

Conversely, I suspect Dividend Growth are much closer to Growth or at least Morningstar's Blended category?

When I have looked at these funds I have tended to see companies with high dividend cover ratios, hence the long term track record of increasing the dividend (earnings growth is at least dividend growth). Thus Microsoft for example. J&J as the canonical example. These don't tend to have dividend yields much above the market average?

These are rated as growth stocks by the market, generally? They don't pay high yields (much above market average) but they have grown them consistently.

HY stocks tend to have low dividend cover, thus the share prices are depressed because the markets don't think the dividend and earnings growth will be there, or in fact in the very high yielding stocks, the market may be expecting a dividend cut. Think the German utilities which are all in varying degrees of financial trouble.

That's classic Value territory-- things are bad but the market overestimates how bad they are -- excess pessimism.
Last edited by Valuethinker on Fri Mar 30, 2018 7:45 am, edited 1 time in total.

User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39459
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: Simple-minded inspection of actual low-vol fund

Post by nisiprius » Fri Mar 30, 2018 7:01 am

Since the correlation of DEF and VTSMX is 0.80, which to me is not a "low" correlation at all but nevertheless is imperfect, I thought I ought to test one other possibility: the possibility that a combination of VTSMX and DEF might be better than either alone. The reasoning here is that it meets some of the criteria for low correlation to be useful: two assets with roughly similar risk and return, but imperfect correlation. I didn't actually expect there to be a benefit, because 0.80 isn't low enough, but I thought I should try.

Portfolio 1 is DEF by itself

Portfolio 2 is 81.60% VTSAX (total stock), 19.40% VBTLX (total bond), i.e. "don't fuss with special stocks, just use bonds." As before, the percentages were adjusted by trial and error so as to give the same standard deviation as DEF.

Portfolio 3 is 46.24% DEF, 46.24% VTSAX, 7.52% VBTLX, i.e. "did adding DEF to VTSMX produce a whole greater than the sum of its parts, due to low correlation?" I would say "no--not in any meaningful way."

Source

Image
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.

Marketman
Posts: 196
Joined: Sun Feb 04, 2018 2:24 pm
Location: Texas

Re: Simple-minded inspection of actual low-vol fund

Post by Marketman » Fri Mar 30, 2018 9:44 am

Thanks for the fine analysis above! If low vol does not do anything that total stock market plus bonds cannot achieve, it seems to me the low vol strategies would give one excess risk because they are usually concentrated in sectors such as consumer staples, etc. Total stock market plus bonds would be WAY more diversified.

alex_686
Posts: 5097
Joined: Mon Feb 09, 2015 2:39 pm

Re: Simple-minded inspection of actual low-vol fund

Post by alex_686 » Fri Mar 30, 2018 3:16 pm

nisiprius wrote:
Fri Mar 30, 2018 7:01 am
Since the correlation of DEF and VTSMX is 0.80,
Actually, in my mind .8 is a fairly low correlation in the financial markets. Plus, the promise of low volatility funds is not they are uncorrelated with the stock market. They can have the same directional movements, just not the same magnitude.

Plus you are including the financial crisis. Correlations assume constants volatility. Change the volatility and you are going to get junk results.
nisiprius wrote:
Fri Mar 30, 2018 7:01 am
Portfolio 2 is 81.60% VTSAX (total stock), 19.40% VBTLX (total bond), i.e. "don't fuss with special stocks, just use bonds." As before, the percentages were adjusted by trial and error so as to give the same standard deviation as DEF.
This I think is an argument for low volatility stocks. Bonds have been blessed with falling interest rates and falling inflation rates since 2008. The chance of this happening again is low. Yields are historically low and risk is historically high. Bonds will be facing considerable headwinds in the future.

We won't be facing the same fact pattern in the future as we have in the past. This ties back to the back-testing thread. The underlying economic factors have changed.

So, if bonds are facing a headwinds, can we lower our risk by substituting some of our bond and equity AA to low volatility? The equity side should promise a decent hedge against inflation and we have lowered our exposure to the riskier bits of the equity market.

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: Simple-minded inspection of actual low-vol fund

Post by lack_ey » Sat Mar 31, 2018 11:46 am

While we're on the topic of asset allocation, one potential advantage of using low-vol stocks (or some other similar construction) instead of a combination of broader stocks and bonds would be the fact that stocks are real-valued assets at least in theory, whereas bonds are nominal-dollar assets, other than TIPS and other inflation-linked issues.

If meeting real liabilities with a relatively low-risk allocation heavy in bonds, this could matter in some rare cases.

GaryA505
Posts: 313
Joined: Wed Feb 08, 2017 2:59 pm
Location: New Mexico

Re: Simple-minded inspection of actual low-vol fund

Post by GaryA505 » Sat Mar 16, 2019 12:29 pm

Old thread, but I'm going to reply anyway because this is of great interest to me. I've done some of the same comparisons Nisiprius did, but to remove the effect of falling interest rates I used ST treasuries to get the StDev to match. When you do that, the low-vol funds look pretty good. Now, it's always possible that interest rates will be higher in the future, but that could be a long wait.

HEDGEFUNDIE
Posts: 3656
Joined: Sun Oct 22, 2017 2:06 pm

Re: Simple-minded inspection of actual low-vol fund

Post by HEDGEFUNDIE » Sat Mar 16, 2019 12:41 pm

Like vineviz, I have come to believe that the magic “low-vol” asset to pair with TSM is classic old utilities:

https://www.portfoliovisualizer.com/bac ... tion3_3=50

Yes, it does add benefit beyond just adding TBM

MotoTrojan
Posts: 6877
Joined: Wed Feb 01, 2017 8:39 pm

Re: Simple-minded inspection of actual low-vol fund

Post by MotoTrojan » Sat Mar 16, 2019 1:53 pm

HEDGEFUNDIE wrote:
Sat Mar 16, 2019 12:41 pm
Like vineviz, I have come to believe that the magic “low-vol” asset to pair with TSM is classic old utilities:

https://www.portfoliovisualizer.com/bac ... tion3_3=50

Yes, it does add benefit beyond just adding TBM
Max drawdown is still more inline with 100% equity relative to bonds.

andrew99999
Posts: 529
Joined: Fri Jul 13, 2018 8:14 pm

Re: Simple-minded inspection of actual low-vol fund

Post by andrew99999 » Sun Mar 17, 2019 8:27 am

nisiprius wrote:
Thu Mar 29, 2018 8:37 pm
In short, in this second case I saw no obvious visible benefit to using stocks selected for low volatility, compared to using a total stock market portfolio and simply increasing the bond allocation.
nisiprius,

If historically an allocation can be interchangeable between low-vol and a portion of stocks & bonds, then I am wondering if you (or others) have any thoughts on the idea of lowering a portion of bonds & stocks and replacing with an equivalent (based on risk/return) amount of low-volatility fund while in a low interest rate environment following a multi-decade bull run on bonds to maintain the same risk/return while lowering bonds?
PassiveInvestingAustralia.com

User avatar
Forester
Posts: 556
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Simple-minded inspection of actual low-vol fund

Post by Forester » Sun Aug 18, 2019 2:03 pm

Is there any reason to think low/min vol funds wouldn't have fared better in 08/09 ? We've had a couple of sizeable corrections recently and these products held up well. Low vol proponents say one should expect the same or slightly better broad equity returns + two thirds of the volatility and the actual performance bears that out. The burden of proof lies with the low vol sceptics.

https://www.portfoliovisualizer.com/bac ... total3=100

HEDGEFUNDIE
Posts: 3656
Joined: Sun Oct 22, 2017 2:06 pm

Re: Simple-minded inspection of actual low-vol fund

Post by HEDGEFUNDIE » Sun Aug 18, 2019 2:19 pm

Forester wrote:
Sun Aug 18, 2019 2:03 pm
Is there any reason to think low/min vol funds wouldn't have fared better in 08/09 ? We've had a couple of sizeable corrections recently and these products held up well. Low vol proponents say one should expect the same or slightly better broad equity returns + two thirds of the volatility and the actual performance bears that out. The burden of proof lies with the low vol sceptics.

https://www.portfoliovisualizer.com/bac ... total3=100
They fared slightly better. 47% total drawdown vs 55% for TSM. -26% vs -38% in 2008.

https://www.msci.com/documents/10199/f5 ... 761d009094

MoneyMarathon
Posts: 556
Joined: Sun Sep 30, 2012 3:38 am

Re: Simple-minded inspection of actual low-vol fund

Post by MoneyMarathon » Sun Aug 18, 2019 2:41 pm

Forester wrote:
Sun Aug 18, 2019 2:03 pm
Low vol proponents say one should expect the same or slightly better broad equity returns + two thirds of the volatility and the actual performance bears that out. The burden of proof lies with the low vol sceptics.
Short term backtest believers will love low vol until they are proven wrong. That's their prerogative.

When I see a slice that's done as well or better over the last 12 years at lower risk, I see "uh oh, look out below." At a certain point, all the low vol believers are bought in, to the maximum extent they can be on a net basis. After that, you can expect see a steady toilet bowl pattern of people dumping what was hyped to them as "slightly better" long term returns with lower volatility. Ain't no such thing.

If you want to expect slightly better returns, buy international. Why? It's been dumped for the last 12 years. I expect international to outperform US min vol over the next 12 years. I can't buy the last several years of performance of min vol, so I'm not really impressed by how they've done lately. More the opposite.

"If past history is all there was to the game, the richest people would be librarians." - Buffett

User avatar
Forester
Posts: 556
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Simple-minded inspection of actual low-vol fund

Post by Forester » Sun Aug 18, 2019 3:31 pm

MoneyMarathon wrote:
Sun Aug 18, 2019 2:41 pm
"slightly better" long term returns with lower volatility. Ain't no such thing.
Why not? There may be no compensation for holding lottery stocks. People say the same of US vs ex-US stocks, not being worth the risk.

MoneyMarathon
Posts: 556
Joined: Sun Sep 30, 2012 3:38 am

Re: Simple-minded inspection of actual low-vol fund

Post by MoneyMarathon » Sun Aug 18, 2019 5:14 pm

Forester wrote:
Sun Aug 18, 2019 3:31 pm
MoneyMarathon wrote:
Sun Aug 18, 2019 2:41 pm
"slightly better" long term returns with lower volatility. Ain't no such thing.
Why not? There may be no compensation for holding lottery stocks.
Because now everyone can buy low volatility, with a low expense ratio, and the market inefficiencies due to shorting constraints are diminishing as more and more people overweight low volatility in a long-only portfolio. The historical anomaly closes, the low volatility stocks become correctly priced, and people can expect only to get low volatility from low volatility stocks, with somewhat below-market long term returns because there is enough money out there looking for the opportunities for better risk-adjusted return (like, for example, all the money looking for it in min vol over the last several years).

The inflection point at which low volatility is correctly priced, getting the same long term risk-adjusted return as the market itself (and less in the short run, since enthusiasm overshoots fair pricing and we'll have to correct to get there), will be a lot sooner than its fans expect. I don't know when peak min vol will be, but the hype train left the station a long time ago now, & whenever the next market crash happens & the following bull market sounds about right for the time when we'll play musical chairs and min vol will be left underperforming on a risk-adjusted basis.

https://www.wsj.com/amp/articles/money- ... 76?tesla=y
Net inflows this year for the iShares Edge MSCI Min Vol USA ETF (USMV) stood at $5.92 billion through June, according to data from Morningstar Direct—more than 10% of all U.S.-stock ETF inflows for that period, which totaled $53.49 billion. A second ETF devoted to low-volatility stocks, Invesco S&P 500 Low Volatility ETF (SPLV), took in almost $2 billion. Those results ranked the two funds second and eighth among all domestic ETFs at the end of June (the most recent month such aggregate data was available), behind Vanguard S&P 500 ETF (VOO) at No. 1, with $8.33 billion.
The fundamentals (higher P/E than the market itself despite lower growth in earnings) also indicate that min vol is either correctly priced or has already passed that point and it's now just a matter of how big the bubble will get before popping & investors lose interest because the promises of "slightly better" returns with "2/3" the volatility because "why not?" turn out to be slick little lies based on data mining in a time period when it was either (a) hard to invest in or (b) during this early phase of enthusiasm and increasing prices because of the Ponzi effect / valuation multiple expansion of people being excited about this magical asset class.

Yes, you can't see this in a price chart. You would have to have a viewpoint on this kind of new hot thing generally and skeptical that "this time is different." Without that broad viewpoint, everything checks out for min vol... at least, until it doesn't. Can't be proven wrong historically until burned financially.

User avatar
Forester
Posts: 556
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: Simple-minded inspection of actual low-vol fund

Post by Forester » Sun Aug 18, 2019 7:15 pm

MoneyMarathon wrote:
Sun Aug 18, 2019 5:14 pm
Because now everyone can buy low volatility, with a low expense ratio, and the market inefficiencies due to shorting constraints are diminishing as more and more people overweight low volatility in a long-only portfolio. The historical anomaly closes, the low volatility stocks become correctly priced, and people can expect only to get low volatility from low volatility stocks, with somewhat below-market long term returns because there is enough money out there looking for the opportunities for better risk-adjusted return (like, for example, all the money looking for it in min vol over the last several years).
MSCI US min vol is about 1/3 of the large cap universe, that may make it less susceptible to crowding. Presumably the stocks which aren't purchased are found in other factor strategies (momentum, value, quality etc).

The fundamentals (higher P/E than the market itself despite lower growth in earnings) also indicate that min vol is either correctly priced or has already passed that point and it's now just a matter of how big the bubble will get before popping
You can argue that low vol stocks are richly priced at the moment but this says nothing about the future as the index is continually replacing constituents. If a stock becomes high risk then the low vol strategy automatically rotates out of that name.

MoneyMarathon
Posts: 556
Joined: Sun Sep 30, 2012 3:38 am

Re: Simple-minded inspection of actual low-vol fund

Post by MoneyMarathon » Sun Aug 18, 2019 8:11 pm

Forester wrote:
Sun Aug 18, 2019 7:15 pm
The fundamentals (higher P/E than the market itself despite lower growth in earnings) also indicate that min vol is either correctly priced or has already passed that point and it's now just a matter of how big the bubble will get before popping
You can argue that low vol stocks are richly priced at the moment but this says nothing about the future as the index is continually replacing constituents. If a stock becomes high risk then the low vol strategy automatically rotates out of that name.
The index is committed to buying low volatility stocks. Just as those low volatility stocks can outperform on a risk-adjusted basis when there is relatively more money flowing into min vol, they can and will underperform on a risk-adjusted basis when relatively more money is flowing out of min vol. Given the easy access to min vol tickers and the large and growing amount of money parked there recently, in the near future this has become more likely than ever.

The short-term risk-adjusted performance of the min vol strategy drifts depending on the popularity of overweighting it with investors, and being low volatility does not mean that the stocks in the index can't relatively-gently glide down to a lower Sharpe (losing more per unit of volatility, or gaining less per unit of volatility, than the market) just as they've glided up to their higher Sharpe recently.

It goes without saying that, in an episode of having lower Sharpe and lower volatility, it can greatly underperform the market in the short term, before just getting along as expected with well-priced risk-adjusted returns (some underperformance to the higher-risk overall stock market in total return) in the long term.

User avatar
firebirdparts
Posts: 357
Joined: Thu Jun 13, 2019 4:21 pm

Re: Simple-minded inspection of actual low-vol fund

Post by firebirdparts » Mon Aug 19, 2019 6:21 am

Forester wrote:
Sun Aug 18, 2019 2:03 pm
Is there any reason to think low/min vol funds wouldn't have fared better in 08/09 ? We've had a couple of sizeable corrections recently and these products held up well. Low vol proponents say one should expect the same or slightly better broad equity returns + two thirds of the volatility and the actual performance bears that out. The burden of proof lies with the low vol sceptics.
Click on hedgefundies link and just plain XLU (existed in 1999) and plain VTSAX.

I would agree that the reason to think this won't happen again is that everybody noticed it now (I guess). It is peculiar.
A fool and your money are soon partners

garlandwhizzer
Posts: 2511
Joined: Fri Aug 06, 2010 3:42 pm

Re: Simple-minded inspection of actual low-vol fund

Post by garlandwhizzer » Mon Aug 19, 2019 2:43 pm

I agree with nisi on this point. The following is an article by Christine Benz of Morningstar on portfolio diversification. Her main point is summarized in the lead sentence:
Despite a proliferation of diversification alternatives, high-quality bonds have provided the best antidote to equities over the past decade.
The article is worth reading for those who seek to diversify their portfolios. Bonds work most potently and most consistently is this regard. She concludes that other alternatives are often not reliable diversifiers.

https://www.morningstar.com/articles/69 ... e-simplest

Garland Whizzer

User avatar
Portfolio7
Posts: 664
Joined: Tue Aug 02, 2016 3:53 am

Re: Simple-minded inspection of actual low-vol fund

Post by Portfolio7 » Mon Aug 19, 2019 6:44 pm

nisiprius wrote:
Fri Mar 30, 2018 7:01 am
Since the correlation of DEF and VTSMX is 0.80, which to me is not a "low" correlation at all but nevertheless is imperfect, I thought I ought to test one other possibility: the possibility that a combination of VTSMX and DEF might be better than either alone. The reasoning here is that it meets some of the criteria for low correlation to be useful: two assets with roughly similar risk and return, but imperfect correlation. I didn't actually expect there to be a benefit, because 0.80 isn't low enough, but I thought I should try.

Portfolio 1 is DEF by itself

Portfolio 2 is 81.60% VTSAX (total stock), 19.40% VBTLX (total bond), i.e. "don't fuss with special stocks, just use bonds." As before, the percentages were adjusted by trial and error so as to give the same standard deviation as DEF.

Portfolio 3 is 46.24% DEF, 46.24% VTSAX, 7.52% VBTLX, i.e. "did adding DEF to VTSMX produce a whole greater than the sum of its parts, due to low correlation?" I would say "no--not in any meaningful way."

Source

Image
I understand why you wanted to start in 2007 or before, but I think in doing so you may be placing an inappropriate burden on DEF.

USMV and SPLV have outperformed DEF since their inceptions by most risk/efficiency measures. I wonder if DEF may not really be an appropriate compare. That's not to say Min or Low Volatility will live up to any of it's promises, just that DEF seems to be slightly a different beast.

I'm not going to try to logic my way to a conclusion regarding whether the factor has disappeared, either - maybe, maybe not, but even if that logic is accurate, "the market can stay irrational longer than I can stay solvent..."

In any case, I think you make a good case that it is unwise to trust solely (and maybe at all) to Low/Min Vol for safety; it's far too new and untested to have confidence in it's behavior.

I'll note that I own a non-negligible amount of VMNVX in my portfolio, in conjunction with high quality bonds. I am not trying to capture any benefit from lack of correlation, merely hoping to see the reduced volatility during time periods when correlations move to '1' for most equities. I am also prepared for the idea that I'm entirely wrong, and that m/l vol funds could be a bust, in which case only the bond component of my AA is likely to limit losses. The choice to use such funds is complicated by the fact that you have to carry a good chunk of your AA in such funds for them to have a significant benefit.
"An investment in knowledge pays the best interest" - Benjamin Franklin

NMBob
Posts: 253
Joined: Thu Apr 23, 2015 8:13 pm

Re: Simple-minded inspection of actual low-vol fund

Post by NMBob » Mon Aug 19, 2019 7:07 pm

AKREX -- akre focus fund since 2009 not 2007 though. Has lower volatility than average despite being a growth fund , not a low volatility fund. he has a buy and hold style that is successful and tax efficient. chuck akre, one of the top mutual fund managers for over 2 decades at fbr and then his own company.

Post Reply