QSPIX - thoughts on interesting fund

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Locked
User avatar
nisiprius
Advisory Board
Posts: 52105
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: QSPIX - thoughts on interesting fund

Post by nisiprius »

lack_ey wrote:
nisiprius wrote:How liquid are the assets held by leveraged long-short funds? What are those assets, exactly, and can they readily be sold quickly to meet redemptions, should the need arise?
Here, it's basically a huge pile of cash and little actual assets.... So the bulk of the fund is a large cash pile with money getting added or taken away from the pile every day based on whatever the performance of the underlying securities was... The fund just needs to go to its huge cash pile to meet redemptions.
Got it. Convincing. And borne out by Morningstar, whose portfolio analyses tend to go crazy on this kind of thing...
Image
And then close some contracts at the market price to adjust the leverage back to the target when convenient. If futures prices diverge considerably from the underlying securities, that's a big risk-free arbitrage opportunity for somebody. Maybe it happens. But the fund isn't obligated to open and close positions ASAP and could ride out some funny behavior unless all its positions were blowing up and its hand was forced.
I'll take that on faith. :)

So, the risks of the long-short leveraged portfolio are whatever they are, but the risks of "run on the fund" (the thing that happened to Third Avenue Focused Credit and that regulators worry about in less-liquid bond funds)--the fund would be solvent if only everyone wasn't demanding redemption at the same time--are minimal.
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.
grok87
Posts: 10512
Joined: Tue Feb 27, 2007 8:00 pm

Re: QSPIX - thoughts on interesting fund

Post by grok87 »

larryswedroe wrote:grok
Just as addendum, momentum crashes are always on the short side (and this fund does go short). But remember only a percentage of the fund is MOM and only a percentage of the MOM is in stocks, the rest are in commodities, currencies and bonds. So even if you get caught with the equity MOM it's still not a large percent of the entire fund position.

MOM in total is about 1/3 of the risk allocation and stocks about 40%. So perhaps about 12% of the fund would be exposed to the momentum crash in stocks. An additional 20% of the portfolio is in equity indexes, don't know how they would be impacted by MOM crash.
Larry
Thanks Larry.

Personally i find AQR's communicating the size of its positions in terms of "risk" not extremely helpful. It begs the question of how one is measuring risk. Fortunately the detailed holdings are available from the AQR website.
https://funds.aqr.com/data-downloads

as of 10/31 it lists the NAV as $1,423 M. And for the equity holdings we see the following:

long $2,274 M
short $1,806 M

so its interesting that they are not really market-netural here. My understanding is that they are more trying to be "beta" neutral. Part of their strategy is to go long "low beta stocks" and short "high beta stocks". the above positions would work out to beta neutral, if for example the longs had an average beta of 0.89 and the shorts had an average beta of 1.11. This is probably all ok much of the time, but in a market downturn its less clear what could happen with this net market exposure which amounts to 33% of NAV.

But let's put that to one side for a moment and use the average of the 2 positions as $2,046 M. Then we might say the fund is 143% long and 143% short of NAV (since 2046/1423 is about 143%). Using the 1/3 exposure to momentum that you cite, we might say the fund is 48% long/short momentum. So again in 2009 French's data shows that long/short momentum lost 83%. so 83% of 48% is 40%. So that is one guess at what the fund could lose from its stock positions in a 2009 scenario.

As you point out they have additional long-short equity exposure in the form of equity market futures. I agree its unclear how that would be affected in a momentum crash. It's harder to figure out the size of those positions from the holdings since they involve derivatives. But using your figures it might be half as big as the direct stocks. If we cut that in half due to the less clear exposure to a "momentum crash" scenario we might multiply the 40% up by say a factor of 1.25 which would take the estimated loss for a 2009 scenario to 50%.

Now its tempting to hope that the other factors might bail one out in this scenario. For example the idea that value and momentum might be negatively correlated. Perhaps. But there is of course no guarantee of this. In fact, in 2009, HML (value long-short factor) lost 5% when momentum lost 83%. So didn't seem to pan out that time.

cheers,
grok

ps thanks for the pointer to Superforecasting- I'm really enjoying it.
RIP Mr. Bogle.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

I'm pretty sure "risk" just means projected standard deviation here. This is probably reasonable enough for the purpose of sizing the long and short sides unless you expect them to have significantly different distributions and in some way you don't like.

The paper Investing with Style doesn't quite describe exactly what the fund does but I suspect the definitions and much of the premise should be consistent. You can try reading section 3.3:
https://www.aqr.com/~/media/files/paper ... -style.pdf

For what it's worth, if beta neutral you would have to say it's market neutral. Zero market beta in a four-factor model means no market exposure and no expectation for that to move with the market when it goes up or down. Net stock exposure by dollar is something else. It's an interesting point to bring up but this is as expected. I mean, if the fund were equal dollar long and short in stocks while loading on low beta, then the fund would have negative beta. Maybe that's what some want, and it might be a better hedge in an equity downturn, but this is counter to the idea of being neutral in asset classes and traditional exposures.
grok87
Posts: 10512
Joined: Tue Feb 27, 2007 8:00 pm

Re: QSPIX - thoughts on interesting fund

Post by grok87 »

lack_ey wrote:I'm pretty sure "risk" just means projected standard deviation here. This is probably reasonable enough for the purpose of sizing the long and short sides unless you expect them to have significantly different distributions and in some way you don't like.

The paper Investing with Style doesn't quite describe exactly what the fund does but I suspect the definitions and much of the premise should be consistent. You can try reading section 3.3:
https://www.aqr.com/~/media/files/paper ... -style.pdf

For what it's worth, if beta neutral you would have to say it's market neutral. Zero market beta in a four-factor model means no market exposure and no expectation for that to move with the market when it goes up or down. Net stock exposure by dollar is something else. It's an interesting point to bring up but this is as expected. I mean, if the fund were equal dollar long and short in stocks while loading on low beta, then the fund would have negative beta. Maybe that's what some want, and it might be a better hedge in an equity downturn, but this is counter to the idea of being neutral in asset classes and traditional exposures.
thanks for the response and the pointer to the paper.
I'm pretty sure "risk" just means projected standard deviation here.
"projected" huh. well again that does beg the question of what model they are using to project.
For what it's worth, if beta neutral you would have to say it's market neutral.
i'm not so sure. betas can change after all. it's not like every stock gets issued with a stamp saying: you are a 1.3 beta stock, or a 0.5 beta stock.

i wonder what the Vanguard market-neutral fund does....my best guess is that they are aiming to be truly market neutral. i.e. the dollar size of their longs equals the dollar size of their shorts.
RIP Mr. Bogle.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

grok
Yes it's beta neutral which is IMO the RIGHT WAY to look at it. Otherwise you would be short stock RISK as you tend to be long low beta and short high beta. It's stock risk that matters not the net long and short.

Now when you get reversals it's the high beta stocks do tend to do the best (having been the ones that did the worst).

But again remember that before the reversal occurs (when volatility is rising) they would be DELEVERAGING.

Also note that it's almost natural that when value does well MOM does poorly and vice versa--MOM correlates with growth for obvious reasons, at least in stocks.

Larry
User avatar
grap0013
Posts: 1892
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 »

raywax wrote:
matjen wrote:Up .66% today which I am sure Grap enjoyed. I am still at 5%.
He is not the only one! Right now to me it looks like a good place to hide a substantial share of one's portfolio.

Ray
Asset allocation "picking" can be just like stock picking. I'm sure I had a little extra dopamine released in my brain last week from my AA change. Now I just have to make sure I do not give myself too much credit!

When I added QSPIX last Jan. 2015 it sunk for a while with equities flourishing. This time around with the extra 10% QSPNX I'm already seeing the benefit in the first couple of days. Global equities ~-5% last week while QSPIX/QSPNX were up +0.37%. Nice! Also, with a 20% stake it won't take as much to hit rebalancing bands as it would with smaller allocations.
There are no guarantees, only probabilities.
User avatar
Topic Author
Yesterdaysnews
Posts: 964
Joined: Sun Sep 14, 2014 1:25 pm
Location: Sugar Land, TX

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews »

Up 0.5% today!

QLENX down only 0.9%

Performing as advertised imo.
User avatar
grap0013
Posts: 1892
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 »

^Completely agree. YTD QSPIX +1.18% and QSPNX +1.08%. These funds make part of a sweet mix. I'm very impressed with them.

I'm currently:

40% SCV
20% iSCV
20% EMV
20% QSPIX/QSPNX

but one could move the dial further down the volatility spectrum and do:

30% SCV
15% iSCV
15% EMV
20% QSPIX/QSPNX
20% intermediate treasuries

OR

20% SCV
10% iSCV
10% EMV
20% QSPIX/QSPNX
40% intermediate treasuries

The last one is darn near bullet proof. One could use that as a 20 or 80 year old indefinitely without adjustments and be happy in all environments minus the usual tracking error.
There are no guarantees, only probabilities.
User avatar
matjen
Posts: 2189
Joined: Sun Nov 20, 2011 10:30 pm

Re: QSPIX - thoughts on interesting fund

Post by matjen »

Piling on with another short term look at QSPIX. Although total returns are important, how it interacts with the portfolio is what is fantastic (so far). Here is what it has looked like during this recent volatility/correction. I have compared it to Total World, Total Bond, and the 60/40 LifeStrategy. :beer

Image
A man is rich in proportion to the number of things he can afford to let alone.
User avatar
Topic Author
Yesterdaysnews
Posts: 964
Joined: Sun Sep 14, 2014 1:25 pm
Location: Sugar Land, TX

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews »

I'm not sure what secret sauce AQR has cooked up, but these two funds continue to deliver even on a day like today:

QLENX - down 0.67%
QSPNX - up 0.48%
gtwhitegold
Posts: 673
Joined: Fri Sep 21, 2012 1:55 pm

Re: QSPIX - thoughts on interesting fund

Post by gtwhitegold »

I personally won't invest in a long/short fund, I had rather specifically choose my longs, but AQR is the fund company that I will use whenever I can. I am going to invest in QMHNX and increase my holdings in QSPIX/QSPNX over the course of the year.
Yesterdaysnews wrote:I'm not sure what secret sauce AQR has cooked up, but these two funds continue to deliver even on a day like today:

QLENX - down 0.67%
QSPNX - up 0.48%
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

gtwhitegold wrote:I personally won't invest in a long/short fund, I had rather specifically choose my longs, but AQR is the fund company that I will use whenever I can. I am going to invest in QMHNX and increase my holdings in QSPIX/QSPNX over the course of the year.
I don't follow. Maybe I misread. Care to elaborate?

QMHNX - AQR Managed Futures Strategy HV [High Variance] Class N
QSPIX/QSPNX - AQR Style Premia Alternative Class I/Class N

Both those funds are long/short. Maybe you mean long/short equity, though both use equity futures and the style premia fund also goes long and short individual companies too.
gtwhitegold
Posts: 673
Joined: Fri Sep 21, 2012 1:55 pm

Re: QSPIX - thoughts on interesting fund

Post by gtwhitegold »

I apologize. What I was trying to say is that an equity long/short fund is not market neutral and does carry some market beta. Market beta can be obtained less expensively elsewhere, even through other AQR funds.

I intend to use style premia and managed futures because they are traditionally less correlated with equities and should provide a rebalancing bonus due to that.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

Keep in mind that this fund has NO correlation with stocks, not negative. That means there will be times when stocks do poorly and the fund doesn't do so well. Very hard to find negatively correlating assets.

Larry
User avatar
Topic Author
Yesterdaysnews
Posts: 964
Joined: Sun Sep 14, 2014 1:25 pm
Location: Sugar Land, TX

Re: QSPIX - thoughts on interesting fund

Post by Yesterdaysnews »

The negative part of these funds, however, is that they have massive turnover and thus can't be help in taxable accounts.

Otherwise, although pricey, they clearly are good products.
User avatar
grap0013
Posts: 1892
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 »

larryswedroe wrote:Keep in mind that this fund has NO correlation with stocks, not negative. That means there will be times when stocks do poorly and the fund doesn't do so well. Very hard to find negatively correlating assets.

Larry
Agreed. I have seen 3 separate periods with this fund so far. Equities up double digits in early 2015 and QSPIX was down. Then equities went down and QSPIX went up. Now equities are going down again and QSPIX is up again.

Here's how I look at it. If you take half a dozen factors that have positive returns about 2/3rds of the time and they are not correlated then most years you will have positive returns. Therefore, I'd casually wager this fund will have positive returns 80% of years. Since QSPIX is not correlated to equities, when equities are down 30% of years there is a pretty good chance QSPIX will be up which reduces your downside. Also, in the event both QSPIX and equities are down the former clearly has lower downside potential from what I've seen even on RBDs. Therefore, this fund reduces your downside paper losses either way when taken from the equity side of one's portfolio. Like all investments, I can envision some tracking error with QSPIX if equity funds have a couple years of double digit gains and QSPIX returns 5 or 6% or less.

"QSPIX is what we thought it was! Now if you want to crown them..."
There are no guarantees, only probabilities.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

grap agreed completely
here's what we think. If you take the allocation from equity side you'll get about same returns but lower SD. That's surely good
If you take from bond side you get higher returns but also higher SD but returns up more than SD, so more efficient portfolio, but bigger tails
Since I have a very low equity allocation and given bond yields so low, I decided to take from the bond side.

Larry
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T »

.
The larger the value, momentum, and quality tilt you have on the equity side of your portfolio the higher the correlation of your equity holdings will be with QSPIX.

As about 60% of QSPIX is essentially comprised of a positive load to value, momentum, and quality within equities, this should correlate fairly strongly with these tilts in your equity allocation. These additional loads to value, momentum, and quality could arguably be achieved simply with a larger tilt to these factors in you equity allocation. The additional advantage of QSPIX is it has much lower beta exposure (about a 0.25 load according Portfolio Visualizer so far). But for those targeting a moderate factor load tilt, the same equity portfolio factor loads (beta, value, momentum, and quality) could arguably be obtained by a more extreme factor tilt in equities and a larger bond allocation. The value added of QSPIX seems to be for those wanting and even greater equity tilt for value, momentum, quality relative to beta in their portfolio. This portion of QSPIX seems similar to QMNIX (long/short equity). While on a stand alone basis and without leverage, these long/short equity type funds/portions may provide relatively 'stable returns', I would caution against the apparent negative left tail in a QMNIX type investment. For example, Novy-Marx's long/short equity with a value/momentum/quality tilt shows relatively stable returns but with a significant outlier during momentum crashes, as in 2009, where the series declined 23% to 37% depending on how extreme the tilt was e.g. from long/short top/bottom 40% deciles to 10% deciles (these negative returns exclude leverage). In addition to a QMNIX type fund, QSPIX provides further diversification benefits with 40% of its allocation to other factors not readily accessible elsewhere - i.e. value, momentum, carry in bonds, currencies, commodities. These likely have lower correlations with other assets and will likely 'smoothen' returns relative to QMNIX.

Robert
.
Last edited by Robert T on Fri Jan 15, 2016 7:20 pm, edited 1 time in total.
grok87
Posts: 10512
Joined: Tue Feb 27, 2007 8:00 pm

Re: QSPIX - thoughts on interesting fund

Post by grok87 »

Robert T wrote:.
The larger the value, momentum, and quality tilt you have on the equity side of your portfolio the higher the correlation of your equity holdings will be with QSPIX.

As about 60% of QSPIX is essentially comprised of a positive load to value, momentum, and quality within equities, this should correlate fairly strongly with these tilts in your equity allocation. These additional loads to value, momentum, and quality could arguably be achieved simply with a larger tilt to these factors in you equity allocation. The additional advantage of QSPIX is it has much lower beta exposure (about a 0.25 load according Portfolio Visualizer so far). But for those targeting a moderate factor load tilt, the same equity portfolio factor loads (beta, value, momentum, and quality) could arguably be obtained by a more extreme factor tilt in equities and a larger bond allocation. The value added of QSPIX seems to be for those wanting and even greater equity tilt for value, momentum, quality relative to beta in their portfolio. This portion of QSPIX seems similar to QLENX (long/short equity). While on a stand alone basis and without leverage, these long/short equity type funds/portions may provide relatively 'stable returns', I would caution against the apparent negative left tail in a QLENX type investment. For example, Novy-Marx's long/short equity with a value/momentum/quality tilt shows relatively stable returns but with a significant outlier during momentum crashes, as in 2009, where the series declined 23% to 37% depending on how extreme the tilt was e.g. from long/short top/bottom 40% deciles to 10% deciles (these negative returns exclude leverage). In addition to a QLENX type fund, QSPIX provides further diversification benefits with 40% of its allocation to other factors not readily accessible elsewhere - i.e. value, momentum, carry in bonds, currencies, commodities. These likely have lower correlations with other assets and will likely 'smoothen' returns relative to QLENX.

Robert
.
yep
negative left tail plus leverage.
not a good risk story.
RIP Mr. Bogle.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

btw there's a long/short equity fund (QLEIX, QLENX) and a separate equity market neutral fund (QMNIX, QMNNX). The long/short equity fund targets a beta to global stocks between 0.3 and 0.7 with around 0.5 on average. The market neutral fund of course has offsetting beta exposure on the long and short sides, targeting about 0.
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T »

lack_ey wrote:btw there's a long/short equity fund (QLEIX, QLENX) and a separate equity market neutral fund (QMNIX, QMNNX). The long/short equity fund targets a beta to global stocks between 0.3 and 0.7 with around 0.5 on average. The market neutral fund of course has offsetting beta exposure on the long and short sides, targeting about 0.
Thanks - my reference was to QMNIX - made the change.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

Also, as a small point:
Robert T wrote:In addition to a QMNIX type fund, QSPIX provides further diversification benefits with 40% of its allocation to other factors not readily accessible elsewhere - i.e. value, momentum, carry in bonds, currencies, commodities. These likely have lower correlations with other assets and will likely 'smoothen' returns relative to QMNIX.
The style premia fund targets annualized volatility of 10% on average with a range of 8% to 12% listed in the prospectus which apparently depends on the level of agreement between the styles. On the other hand, the prospectus of the equity market neutral fund says it will target an volatility around 6% with a range of 4% to 9% (this presumably also depends on style agreement).

So it may not be accurate to say that the other stuff will smoothen out the returns of the style premia fund if the overall risk level is higher there. Maybe smoother per level of risk from the equities.
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T »

lack_ey wrote:Also, as a small point:
Robert T wrote:In addition to a QMNIX type fund, QSPIX provides further diversification benefits with 40% of its allocation to other factors not readily accessible elsewhere - i.e. value, momentum, carry in bonds, currencies, commodities. These likely have lower correlations with other assets and will likely 'smoothen' returns relative to QMNIX.
The style premia fund targets annualized volatility of 10% on average with a range of 8% to 12% listed in the prospectus which apparently depends on the level of agreement between the styles. On the other hand, the prospectus of the equity market neutral fund says it will target an volatility around 6% with a range of 4% to 9% (this presumably also depends on style agreement).

So it may not be accurate to say that the other stuff will smoothen out the returns of the style premia fund if the overall risk level is higher there. Maybe smoother per level of risk from the equities.
My reference to 'smoothen' was related to the previous sentence i.e. referring to a thinner left tail (less negative skewness), than standard deviation. Re:
I would caution against the apparent negative left tail in a QMNIX type investment. For example, Novy-Marx's long/short equity with a value/momentum/quality tilt shows relatively stable returns but with a significant outlier during momentum crashes, as in 2009, where the series declined 23% to 37% depending on how extreme the tilt was e.g. from long/short top/bottom 40% deciles to 10% deciles (these negative returns exclude leverage).
Using Novy-Marx's long/short data - using value-momentum-quality long/short equity top/bottom 5 deciles from 1964 to 2013. Annualized return = 7.0%, Standard deviation = 7.6%. 2009 returns = -20.1%, skewness = -0.6. The 40% addition of other factors in QSPIX likely reduces the negative skewness (the Investing in Style paper still shows negative skewness for the style premia series [similar to QSPIX], but it is about half the size at around -0.3, or a quarter of the size if the same 1990-2013 time period is used for the Novy-Marx data). Obviously no guarantees. Now they may have found a way to partly avoid the severity of momentum crashes e.g. http://www.kentdaniel.net/papers/published/mom12.pdf . Time will tell.

Robert

And may also be worth a read: http://www.nowandfutures.com/download/Q ... sAug07.pdf . Negative tails do show up. If I recall, their Absolute return fund lost about 45 percent in 2008. They have likely built in all lessons learned to new products, stricter risk controls etc., but still likely some negative skewness present.
.
Last edited by Robert T on Fri Jan 15, 2016 8:59 pm, edited 1 time in total.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

Okay, that clears things up, thanks.

On an absolute fund-vs.-fund level I'm unconvinced the left tail is smaller. Or in investment terms, how bad a drawdown—let's say, a 2 percentile outcome—might look like.* I'd need to look into it more carefully, anyhow. If you say that the style premia fund has a higher (less negative) skewness number, that is a different matter. The addition of the extra asset classes should probably make the composite more normal, despite the decently high correlation between asset class returns within each style.

In any case, as you point out some adjustments to momentum might ameliorate some or most crashes. Any kind of volatility targeting/scaling, anything that might reduce investment or leverage when things are wild—raw momentum gets killed by shorting assets that are hitting a rebound. AQR's long/short funds do volatility targeting for a reason. It's not going to stop everything but it's probably better than not doing it.

*simplified illustration, not to scale (black is negatively skewed but blue has more mass out to the left because of higher variance despite being about symmetrical):
Image
User avatar
grap0013
Posts: 1892
Joined: Thu Mar 18, 2010 1:24 pm

Re: QSPIX - thoughts on interesting fund

Post by grap0013 »

Robert T wrote:.
The larger the value, momentum, and quality tilt you have on the equity side of your portfolio the higher the correlation of your equity holdings will be with QSPIX.

As about 60% of QSPIX is essentially comprised of a positive load to value, momentum, and quality within equities, this should correlate fairly strongly with these tilts in your equity allocation. These additional loads to value, momentum, and quality could arguably be achieved simply with a larger tilt to these factors in you equity allocation. The additional advantage of QSPIX is it has much lower beta exposure (about a 0.25 load according Portfolio Visualizer so far). But for those targeting a moderate factor load tilt, the same equity portfolio factor loads (beta, value, momentum, and quality) could arguably be obtained by a more extreme factor tilt in equities and a larger bond allocation. The value added of QSPIX seems to be for those wanting and even greater equity tilt for value, momentum, quality relative to beta in their portfolio. This portion of QSPIX seems similar to QMNIX (long/short equity). While on a stand alone basis and without leverage, these long/short equity type funds/portions may provide relatively 'stable returns', I would caution against the apparent negative left tail in a QMNIX type investment. For example, Novy-Marx's long/short equity with a value/momentum/quality tilt shows relatively stable returns but with a significant outlier during momentum crashes, as in 2009, where the series declined 23% to 37% depending on how extreme the tilt was e.g. from long/short top/bottom 40% deciles to 10% deciles (these negative returns exclude leverage). In addition to a QMNIX type fund, QSPIX provides further diversification benefits with 40% of its allocation to other factors not readily accessible elsewhere - i.e. value, momentum, carry in bonds, currencies, commodities. These likely have lower correlations with other assets and will likely 'smoothen' returns relative to QMNIX.

Robert
.
Porfolio Visualizer (PV) may say there is some overlap with factor loads between QSPIX and equities, but I do not feel or see this in relation to my other funds. Take these nicely tilted factor based funds and their net correlation to QSPIX is still nil.

https://www.portfoliovisualizer.com/ass ... X%2C+QSPIX

Beta from the equity funds + non-equity alternatives in QSPIX drowns out any appreciable overlap in tilts.
There are no guarantees, only probabilities.
User avatar
Robert T
Posts: 2803
Joined: Tue Feb 27, 2007 8:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: QSPIX - thoughts on interesting fund

Post by Robert T »

grap0013 wrote:Porfolio Visualizer (PV) may say there is some overlap with factor loads between QSPIX and equities, but I do not feel or see this in relation to my other funds. Take these nicely tilted factor based funds and their net correlation to QSPIX is still nil.

https://www.portfoliovisualizer.com/ass ... X%2C+QSPIX

Beta from the equity funds + non-equity alternatives in QSPIX drowns out any appreciable overlap in tilts.
I agree correlations are low. The related points highlighted earlier were: (i) as the extent of a long-only equity tilt to value-momentum-quality goes up so do correlations with a long-short fund of these three factors, and (ii) as the long-short fund diversifies to other factors and assets (e.g. carry, bonds, currencies, commodities) then correlations go down. While the time period we have for analysis is very short we can see some of this in the data. e.g.

- Correlation of QSPIX and QMNIX was 0.64, approximately equivalent to the equity allocation in QSPIX.

- Low correlations of these long-short funds with long-only equities is primarily the result of diversification away from beta. For example QLEIX (long/short with some beta) has correlation with long-only funds of about 0.7.

- A purely beta fund with no tilt to other factors had lower (about half the) correlation with QSPIX than the AQR Large Cap Multi-style fund (0.07 vs. 0.14). And I note that the AQR multi-style funds, with a larger tilt to momentum than DFA funds, had larger correlations with QSPIX. And if we use a purer momentum fund such as MTUM, correlations go up to 0.21.

- Correlations of QSPIX with long-only tilted funds is lower than QMNIX (likely due to the diversification of QSPIX to other factors and assets).

https://www.portfoliovisualizer.com/ass ... +QLEIX+IEI

It will be interesting to see how this plays out over the longer-term.

Personally, I think it is important to look at overall factor exposure at portfolio level, aligned to targets, and decide how best to achieve them through various funds. For example, correlations of different factors within some of my funds likely have as low a correlation with beta as QMNIX/QSPIX - (as the low historical correlation between the market (Mkt-rf), size (SmB), value (Hml), momentum (UmD), and quality (QmJ) suggests), its just that I don't see this explicitly as its all embedded within the funds.

For those who want lower beta, and higher tilts to other factors than can be achieved with a combination of long-only factor tilted funds + bonds, then a long-short factor fund may make sense.

Robert
.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

Robert
There are good papers on avoiding momentum crashes, as they are all on the Short side, not long side. Long MOM was up in 2009 when MOM crashed. And the way to do that is to SCALE your exposure. You lower exposure as volatility increases. Crashes happen at peaks in volatility when you get reversals. This is exactly what AQR does, it lowers exposures as volatility rises

And for another poster fwiw portfolio visualizer is worthless for a fund like QSPIX given it's exposure to other asset classes

Larry
Avo
Posts: 1099
Joined: Wed Jun 11, 2008 2:21 am
Location: California

Re: QSPIX - thoughts on interesting fund

Post by Avo »

PortfolioVisualizer was being used to study correlations between QSPIX and various long-equity DFA funds, with the conclusion that correlations are low. That seems perfectly valid to me.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

Avo
yes if it is for correlations that's fine, but not for loading on factors
countmein
Posts: 653
Joined: Fri Dec 06, 2013 8:10 pm

Re: QSPIX - thoughts on interesting fund

Post by countmein »

Isn't the zero correlation between factors and beta tautological (and thus meaningless)? You can go long/short anything and it will have zero correlation. I'm not impressed that QSPIX is uncorrelated to beta. I'm impressed that it exposes one to past sources of return (that ought to endure) in a diverse collection of asset types.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

countmein wrote:Isn't the zero correlation between factors and beta tautological (and thus meaningless)? You can go long/short anything and it will have zero correlation. I'm not impressed that QSPIX is uncorrelated to beta. I'm impressed that it exposes one to past sources of return (that ought to endure) in a diverse collection of asset types.
Factors in general shouldn't be assumed to be uncorrelated with each other or with market beta because they're not explicitly constructed and defined to be that way. In fact, value and momentum are convincingly negatively correlated (depends on which test you run on what data, but there are reasonable tests showing a t-stat lower than -10 for that). Credit in bonds and market beta are positively correlated because economically some of the underlying driving forces frequently affect both, particularly on the downside.

There's also a difference between raw factors and what can be implemented in practice. It's not completely inconceivable that some difference of the actual fund implementation causes the correlation to be different from the frictionless raw data of the factors. It would be weird and I wouldn't expect it, but you never know.

However, based on the past history of the factors in asset classes covered by the fund and what is known about what drives them, it should not be surprising anyone that the fund has (so far) been largely uncorrelated with equities.
countmein
Posts: 653
Joined: Fri Dec 06, 2013 8:10 pm

Re: QSPIX - thoughts on interesting fund

Post by countmein »

lack_ey wrote: Factors in general shouldn't be assumed to be uncorrelated with each other or with market beta because they're not explicitly constructed and defined to be that way.
I think you are correct re "factors with each other" but not correct re "factors with market beta". I don't have the mathematical proof but it seems intuitive that long/short any equity parameter (factor) is by definition not correlated to the market (because you've explicitly cancelled the market out of the factor time series). That's why I think it's kind of bogus to get hyped about QSPIX's value as a "non-correlated asset" in your AA (at least the equity part of QSPIX). If you're thinking that way, why not dedicate 10% of your portfolio to a new DIY "asset" that goes long stocks that begin with 'A' and short those that begin with 'B'? It'll add a non-correlated asset to your portfolio, but so what?
grok87
Posts: 10512
Joined: Tue Feb 27, 2007 8:00 pm

Re: QSPIX - thoughts on interesting fund

Post by grok87 »

lack_ey wrote:
countmein wrote:Isn't the zero correlation between factors and beta tautological (and thus meaningless)? You can go long/short anything and it will have zero correlation. I'm not impressed that QSPIX is uncorrelated to beta. I'm impressed that it exposes one to past sources of return (that ought to endure) in a diverse collection of asset types.
In fact, value and momentum are convincingly negatively correlated (depends on which test you run on what data, but there are reasonable tests showing a t-stat lower than -10 for that). Credit in bonds and market beta are positively correlated because economically some of the underlying driving forces frequently affect both, particularly on the downside.
i would advise caution on these correlation assumptions, particularly on the idea that value and momentum are negatively correlated.
Fama and French have the data on that going back to 1929 and i have just looked at it (monthly series). I'm not very good at posting charts but would be happy to do so if someone can talk me through it.

I think the best one can say is that value and momentum go through periods of negative correlation and periods of positive correlation. for example using 36 month rolling correlations, the correlation was negative from 1929-1943 and positive from 1943-53. most recently value and momentum were positively correlated between 2000-2007.

investing is not physics. these correlation relationships are not stable. thinking that these relationships are stable- being overconfident- is dangerous as it may lead one to leverage up ones portfolio based upon an assumed relationship. for example QSPIX is pretty leveraged. then when the relationship between value and momentum reverses (correlation turns positive) one may find that "leverage works both ways"

cheers,
RIP Mr. Bogle.
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

countmein wrote:
lack_ey wrote: Factors in general shouldn't be assumed to be uncorrelated with each other or with market beta because they're not explicitly constructed and defined to be that way.
I think you are correct re "factors with each other" but not correct re "factors with market beta". I don't have the mathematical proof but it seems intuitive that long/short any equity parameter (factor) is by definition not correlated to the market (because you've explicitly cancelled the market out of the factor time series). That's why I think it's kind of bogus to get hyped about QSPIX's value as a "non-correlated asset" in your AA (at least the equity part of QSPIX). If you're thinking that way, why not dedicate 10% of your portfolio to a new DIY "asset" that goes long stocks that begin with 'A' and short those that begin with 'B'? It'll add a non-correlated asset to your portfolio, but so what?
No, that's not how the factors are defined. If nothing else, check the "betting against beta" factor in an equity data series, how it's defined, and check the correlation with the market return:
https://www.aqr.com/library/data-sets/b ... rs-monthly

For those reading who don't want to check and aren't immediately familiar, BAB is long low beta assets, short high beta assets in the right proportion to get beta neutral. (That requires the low beta side to be leveraged relative to the short beta side.)

BAB is defined to be beta neutral. What is its correlation with the market? What might the correlation of a collection of long/short stocks be that had zero market beta but significant positive (or negative) BAB?

grok87 wrote:i would advise caution on these correlation assumptions, particularly on the idea that value and momentum are negatively correlated.
Fama and French have the data on that going back to 1929 and i have just looked at it (monthly series). I'm not very good at posting charts but would be happy to do so if someone can talk me through it.

I think the best one can say is that value and momentum go through periods of negative correlation and periods of positive correlation. for example using 36 month rolling correlations, the correlation was negative from 1929-1943 and positive from 1943-53. most recently value and momentum were positively correlated between 2000-2007.

investing is not physics. these correlation relationships are not stable. thinking that these relationships are stable- being overconfident- is dangerous as it may lead one to leverage up ones portfolio based upon an assumed relationship. for example QSPIX is pretty leveraged. then when the relationship between value and momentum reverses (correlation turns positive) one may find that "leverage works both ways"

cheers,
Of course the relationships aren't stable and are not consistent in any given period of time. Plus on top of that the realized results may deviate significantly from that.

But that doesn't mean that all equity long/short portfolios (down to zero beta) are uncorrelated with the market, like countmein is trying to argue, which is incorrect. If the non-beta factors have nonzero correlation with the market and you have exposures to those while keeping a zero market beta allocation, you may well end up with correlation not zero.

I had to give examples that show realized nonzero correlations. It doesn't matter whether these things are actually consistent over time or not for the sake of the argument. For example in this old article Bill Bernstein says value is negatively correlated with the market:
https://www.ifa.com/articles/Factor_Rot ... rrelation/

but that's not really been the case over longer periods of time. In any case, it's not consistent and of course rolling statistics will capture a lot of randomness and noise. But even that would prove the point.
countmein
Posts: 653
Joined: Fri Dec 06, 2013 8:10 pm

Re: QSPIX - thoughts on interesting fund

Post by countmein »

lacy_ey,

I think you're confusing volatility with beta. Zero beta = no correlation to beta.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

Size factor and market factor have pretty high correlation at about .4 if memory serves. Size and value and market and value are very low, small positive correlations
Larry
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

I just checked and Kenneth French yearly return data 1928-2014 (um, I didn't bother updating my dataset) shows 0.42 correlation between Mkt-RF and SMB, yeah.

Mkt-RF and SMB: 0.42
Mkt-RF and HML: 0.13
SMB and HML: 0.12
User avatar
oldcomputerguy
Moderator
Posts: 17878
Joined: Sun Nov 22, 2015 5:50 am
Location: Tennessee

Re: QSPIX - thoughts on interesting fund

Post by oldcomputerguy »

iceport wrote:AQR Style Premia Alternative (QSPIX) ER=1.50%?
Actually, according to the May 1, 2015 Prospectus, total expenses for Class I (QSPIX) weighs in at 2.61%. Yikes!!
There is only one success - to be able to spend your life in your own way. (Christopher Morley)
User avatar
matjen
Posts: 2189
Joined: Sun Nov 20, 2011 10:30 pm

Re: QSPIX - thoughts on interesting fund

Post by matjen »

smartinwate wrote:
iceport wrote:AQR Style Premia Alternative (QSPIX) ER=1.50%?
Actually, according to the May 1, 2015 Prospectus, total expenses for Class I (QSPIX) weighs in at 2.61%. Yikes!!
Nope. Has been covered many times before (though understandably would be hard to pick up in a search). They waive the fees to bring it to 1.5% and are keeping it there.
A man is rich in proportion to the number of things he can afford to let alone.
longinvest
Posts: 5672
Joined: Sat Aug 11, 2012 8:44 am

Re: QSPIX - thoughts on interesting fund

Post by longinvest »

Just in case new members were confused about it, QSPIX is NOT a recommended Bogleheads investment.

Bogleheads investment philosophy:
1 Develop a workable plan
2 Invest early and often
3 Never bear too much or too little risk
4 Diversify
5 Never try to time the market

6 Use index funds when possible
7 Keep costs low
8 Minimize taxes
9 Invest with simplicity

10 Stay the course

QSPIX fails to meet four of our ten principles. It is not an index fund, it is very costly, it has incredibly high turnover causing increased taxes (forcing an investor to displace bonds from tax-sheltered accounts to taxable accounts to make place for it), and it is the anti-thesis of simplicity (it is an almost completely opaque* fund using complex derivatives and leverage).

Bogleheads are recommended to stay away from it.

Of course, I do know that not all Bogleheads members adhere to our ten principles. This is OK. Our forums are an open discussion space. This very thread is ample proof of it.

* Actually, it is so opaque that I can't say if it fails or not our fifth principle to "never try to time the market". There's a possibility that its managers have to time the market to move from deprecated factors to newly discovered factors (or to bet on the right momentum direction).
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
DaufuskieNate
Posts: 605
Joined: Wed May 28, 2014 11:53 am

Re: QSPIX - thoughts on interesting fund

Post by DaufuskieNate »

longinvest wrote: 6 Use index funds when possible
7 Keep costs low
8 Minimize taxes
9 Invest with simplicity
6 Key words here are when possible. It's not possible to use an index fund for this type of strategy. This applies to other areas as well. For example, many here would say that it is at times preferable to use something other than index funds for bonds. Also, this fund can be combined in a portfolio with equity index funds. The index funds provide inexpensive exposure to Beta while this fund provides efficient exposure to multiple factors.

7 The costs of this fund are quite low for its category. In the past, this type of strategy was only available in hedge funds which tend to be much more expensive. This is much like international equity funds. They are often more expensive than U.S.-only funds. But if you want to gain exposure to international equities, look for funds that have low costs in that category.

8 This fund should definitely be held in tax-sheltered accounts. This is much like REITs and taxable bond funds in this regard. Any Bogleheads out there using REITs or taxable bond funds?

9 This fund provides all-in-one exposure to a variety of factors in pure long-short form across multiple markets. Pretty simple way to gain diversified exposure to factors.
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

Just to add to what DaufuskieNate said

Unfortunately almost all indices other than TSM are "dumb" indices with negatives that can be minimized or eliminated and positives that can be maximized. The perfect example is the R2k which underperformed the similar CRSP 6-10 by something like 2% a year. The question then are the higher fees of an alternative, but still passive (no individual stock picking or market timing), strategy worth the extra costs. That should be the criteria. In other words, it's value added not costs.

In the case of this fund, the long short strategy means that in reality you are paying 0.75 for the long and 0.75 for the short side, not cheap, but far cheaper than the alternatives. And you are gaining exposure not only to four factors across markets as Nate said, but across ASSET CLASSES as well. So for example, with MOM, it's long short not only across global equities, but currencies, commodities and bonds, providing diversification benefits. In other words while MOM may not be working in stocks may not have anything to do with whether it is working in commodities or currencies. That's why the correlations among the factors are so low.

Is it highly complex, yes. Will you ever be able to guess how the fund did by looking at what markets did that day? No. Should you buy it if you don't understand these issues, and not confuse strategy and outcome, NO!

Larry
User avatar
matjen
Posts: 2189
Joined: Sun Nov 20, 2011 10:30 pm

Re: QSPIX - thoughts on interesting fund

Post by matjen »

^ +1

But let me just thank Longinvest for repeating what he originally wrote on the first page of this very same thread.
viewtopic.php?f=10&t=167241

Longinvest on Sunday, June 7th 2015 wrote:
Somehow, I'm really having a difficult time to see how QSPIX fits into our philosophy:

1 Develop a workable plan
2 Invest early and often
3 Never bear too much or too little risk
4 Diversify
5 Never try to time the market
6 Use index funds when possible
7 Keep costs low
8 Minimize taxes
9 Invest with simplicity
10 Stay the course

Of course, our philosophy is so boring* that we need to spend our time discussing active strategies to keep this forum alive. That must be it!

But, as a result, this forum ends up being noise that most of us should probably be tuning out... :annoyed
Although Larry is certainly correct about strategy vs. outcome, QSPIX is doing so well in a challenging environment for most asset classes I thought I would have some fun.* Here is QSPIX stacked up against the 60/40 and 40/60 Vanguard LifeStrategy Funds since Longinvest's words of warning.... 8-)

I would rather have the performance than the slavish devotion to certain principles.

Image

*Short term results and viewing a fund like this in isolation are not as worthwhile as viewing your portfolio as a whole over a longer time period.
A man is rich in proportion to the number of things he can afford to let alone.
longinvest
Posts: 5672
Joined: Sat Aug 11, 2012 8:44 am

Re: QSPIX - thoughts on interesting fund

Post by longinvest »

Matjen,

Nowhere did I write that QSPIX couldn't outperform.

I simply wrote that it does not fit the Bogleheads Investment Philosophy, and I stand by it.

The wiki page goes into the details of each principle. Nowhere do I see a recommendation to seek alternative investments with leverage, and that a 1.5% is cheap for that.

I am not saying that it is a bad investment, or that it can't outperform. I have no opinion about that. But, I am firm in my assessment that it is NOT a Bogleheads-type investment.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
longinvest
Posts: 5672
Joined: Sat Aug 11, 2012 8:44 am

Re: QSPIX - thoughts on interesting fund

Post by longinvest »

larryswedroe wrote: Unfortunately almost all indices other than TSM are "dumb" indices with negatives that can be minimized or eliminated and positives that can be maximized.
Larry,

It is really nice to know that you have found the Holy Grail of investments and that you don't need, anymore, to use "dumb" products introduced by Jack Bogle (except TSM, of course).

I wish you and your customers the best of luck.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 10:55 pm

Re: QSPIX - thoughts on interesting fund

Post by lack_ey »

longinvest wrote:
larryswedroe wrote: Unfortunately almost all indices other than TSM are "dumb" indices with negatives that can be minimized or eliminated and positives that can be maximized.
Larry,

It is really nice to know that you have found the Holy Grail of investments and that you don't need, anymore, to use "dumb" products introduced by Jack Bogle (except TSM, of course).

I wish you and your customers the best of luck.
I recommend re-reading what was posted (and not just the part quoted here, as the rest explains what either you're missing or ignoring) and then your own response. If nothing else, if evaluating value added then sometimes or frequently the cost of the less-dumb product may be not worth it if the dumb thing is only very marginally dumber and significantly cheaper. Point is, it depends.

Also, the hyperbole doesn't really speak well for you.
Last edited by lack_ey on Sun Jan 17, 2016 1:59 pm, edited 1 time in total.
User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: QSPIX - thoughts on interesting fund

Post by Maynard F. Speer »

longinvest wrote:Matjen,

Nowhere did I write that QSPIX couldn't outperform.

I simply wrote that it does not fit the Bogleheads Investment Philosophy, and I stand by it.

The wiki page goes into the details of each principle. Nowhere do I see a recommendation to seek alternative investments with leverage, and that a 1.5% is cheap for that.

I am not saying that it is a bad investment, or that it can't outperform. I have no opinion about that. But, I am firm in my assessment that it is NOT a Bogleheads-type investment.
You know what they did to Galileo when he suggested it was the Earth that revolves around the Sun?

I don't think Bogle should change his advice - because it does take expertise to understand alternative investments, and if you haven't got it, you need to put quite a bit of trust in someone who does ... But, as a softener, Bogle does have this to say on Cliff Asness:

"I greatly admire Cliff Asness as well. These are guys who run large pools of capital with non-traditional investment strategies. They’re smart, savvy, can keep their perspective, and understand both sides of an issue. They’re far more knowledgeable from an investment standpoint than I am, and are great additions to my large group of heroes"
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes
longinvest
Posts: 5672
Joined: Sat Aug 11, 2012 8:44 am

Re: QSPIX - thoughts on interesting fund

Post by longinvest »

Maynard F. Speer wrote: You know what they did to Galileo when he suggested it was the Earth that revolves around the Sun?

I don't think Bogle should change his advice - because it does take expertise to understand alternative investments, and if you haven't got it, you need to put quite a bit of trust in someone who does ... But, as a softener, Bogle does have this to say on Cliff Asness:

"I greatly admire Cliff Asness as well. These are guys who run large pools of capital with non-traditional investment strategies. They’re smart, savvy, can keep their perspective, and understand both sides of an issue. They’re far more knowledgeable from an investment standpoint than I am, and are great additions to my large group of heroes"
Maynard,

If you think that our Bogleheads Investment Philosophy is "dumb", then you should propose to change it.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
longinvest
Posts: 5672
Joined: Sat Aug 11, 2012 8:44 am

Re: QSPIX - thoughts on interesting fund

Post by longinvest »

lack_ey wrote:
longinvest wrote:
larryswedroe wrote: Unfortunately almost all indices other than TSM are "dumb" indices with negatives that can be minimized or eliminated and positives that can be maximized.
Larry,

It is really nice to know that you have found the Holy Grail of investments and that you don't need, anymore, to use "dumb" products introduced by Jack Bogle (except TSM, of course).

I wish you and your customers the best of luck.
I recommend re-reading what was posted (and not just the part quoted here, as the rest explains what either you're missing or ignoring) and then your own response. If nothing else, if evaluating value added then sometimes or frequently the cost of the less-dumb product may be not worth it if the dumb thing is only very marginally dumber and significantly cheaper. Point is, it depends.

Also, the hyperbole doesn't really speak well for you.
The first index fund that Bogle introduced was an S&P 500 fund. To this day, the performance of this fund and of TSM are within basis points of each other.

Larry has just qualified this fund of "dumb" on a Bogleheads forum ("all indices except TSM").

Larry believes in factors and all the complexity of AQR funds. I don't.
Variable Percentage Withdrawal (bogleheads.org/wiki/VPW) | One-Fund Portfolio (bogleheads.org/forum/viewtopic.php?t=287967)
larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 7:28 am
Location: St Louis MO

Re: QSPIX - thoughts on interesting fund

Post by larryswedroe »

longinvest
You might just want to consider facts before lambasting.

For example, as I noted the R2k was a very dumb index to replicate. From 79-14 it underperformed the very similar CRSP 6-10 11.96 vs. 13.58. Now the Vanguard small cap index fund used to use that index and did so for a long time. Finally Sauter pointed out that it was a dumb index and replicating it cost investors. So he convinced Vanguard to switch indices. For years DFA used simple common sense in creating buy and hold ranges, instead of trying to replicate some index, which reduced forced trading, which created less turnover and more tax efficiency (while also picking up some benefits from momentum). Thus IMPLEMENTATION costs actually came DOWN. And guess what, the newer indices like MSCI now incorporate what DFA had been doing for years because it's common sense, even if it wasn't in the 10 principles of Bogleheads. That's just one simple example of a dumb index and how it can be improved on.

As to QSPIX being a holy grail, where did I ever say anything that could lead anyone to that conclusion? Please provide the quotation. What I have said is that it is based on the academic literature that in each case is persistent and pervasive and robust, with logical explanations either behavioral or risk based, not opinions, and it's run passively, not based on opinions. It convinced me that it justified a small allocation to my portfolio and for others that can understand the fund construction and the risks. Now if it was the so called Holy Grail I would recommend a very large allocation or even 100%.

Those who close their eyes to learning things that don't happen to agree with their preconceived ideas never learn anything new. Just because something is more complex or more expensive doesn't mean it's bad (nor does it mean it's good). And one should be skeptical, make sure the evidence is compelling and logical. But when you are skeptical of everything you learn nothing. Don't know about you but when I learn something new I change my opinions.

Best wishes
Larry
Locked