Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
nisiprius
Advisory Board
Posts: 37101
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: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nisiprius » Mon Apr 16, 2018 11:16 am

With regard to "this is just how you'd expect these things to perform during a bull market," a couple of points. First, I did not compare them to stocks, nor bonds, nor 60/40. The only reason I showed stock and bonds was to show what happens when you look at category averages instead of known-in-hindsight-good funds.

I compared them to money market mutual funds. I haven't heard anyone say that since 2008 we've been in a bull market for money market mutual funds!

Second, I didn't pick the start point, or at least I didn't try to. I tried to find the oldest mutual fund (I have not yet found out a way to search on inception date with an unpaid Morningstar subscription). The starting point was picked by the companies that decided to launch mutual funds. I was trying to approximate "all available data for mutual funds." Similarly, whatever can be said about Morningstar's categorizations, they were made by an independent, disinterested third party.

After considerable work, I can show the full range of data for a) the Credit Suisse Managed Futures Liquid Index, together with b) Morningstar's managed futures category average. What I find really striking about this is how, just before the start of the category average, the no-cost index shows a lovely, stable 11% annual return for more than a decade--and, just at the time they decided to launch mutual funds the index itself turns a corner, bends down sharply, and falls off the 11% line--while the category average is basically horizontal.

Madoff-like performance, that stops just as the mutual funds start.

Rorschach test. Does this mean this is actually a terrific asset class after all? Or does it suggest a mediocre asset class getting attention because of recent chance outperformance? Do we count on future mean reversion bringing it back to that 11%/year line? Or is it the leveling off itself that was the mean reversion?

Image

P.S. There is a "Credit Suisse Managed Futures Strategy Fund," CSIAX... but its inception date was 2012. Funny, that. It did a respectable job of following the index, impressive given its 1.30% expense ratio.
Last edited by nisiprius on Mon Apr 16, 2018 1:07 pm, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

garlandwhizzer
Posts: 2070
Joined: Fri Aug 06, 2010 3:42 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by garlandwhizzer » Mon Apr 16, 2018 12:51 pm

Nisi's post and its dramatic graph give excellent illustrations of "skating where the puck was." Our financial markets are very competitive and dominated by full time professionals who are constantly data mining in search of alpha. When past outperformance shows up, products to capture it proliferate and the alpha tends to diminish or disappear.

Garland Whizzer

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by lack_ey » Mon Apr 16, 2018 2:30 pm

For managed futures in particular, I think crowding is a concern, and if you're looking at mutual funds then you're missing the bulk of AUM.

This article is over two years old but I think is informative:

https://www.rcmalternatives.com/2016/02 ... m-exactly/

The current figure from the source, BarclayHedge, shows $347 billion in managed futures strategies:

https://www.barclayhedge.com/research/i ... ement.html

These are among the many considerations you have to make looking at alternatives investments.

Anybody have a link to the Credit Suisse index methodology? Is the series backfilled or did it actually start in 1995? Obviously we have to be skeptical of backfilled results as strategies tend to be fit to the past data to look impressive, at least in part. That said, I think even plain-vanilla trend approaches should have done pretty well over that period without hindsight-bias cheating in terms of parameter and strategy tuning.

User avatar
HomerJ
Posts: 11964
Joined: Fri Jun 06, 2008 12:50 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by HomerJ » Mon Apr 16, 2018 2:38 pm

lack_ey wrote:
Mon Apr 16, 2018 2:30 pm
These are among the many considerations you have to make looking at alternatives investments.
Is having many considerations to look at before investing in alt investments a pro or a con?

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by lack_ey » Mon Apr 16, 2018 2:51 pm

HomerJ wrote:
Mon Apr 16, 2018 2:38 pm
lack_ey wrote:
Mon Apr 16, 2018 2:30 pm
These are among the many considerations you have to make looking at alternatives investments.
Is having many considerations to look at before investing in alt investments a pro or a con?
More work for the investor is generally bad, and juggling more things usually means more mistakes.

There are a lot of downsides to various types of alternatives even before you get to the usually significantly higher fees. We don't know that there are actually any benefits, but even supposing there are, they may not improve realized outcomes for any period we care about because returns are random. It can be a tough road, and a lot of the time you will look stupid following it.

Outside of perhaps some on the fringe, nobody wants to get rid of traditional asset classes for the long run. We all wish stock and bond returns will continue to the great, even remotely like past returns.

People focus too much on volatility and downside protection. The reality that long-term investors face is the possibility of a lot of assets not returning much of anything, and this needs to be weighted against the potential upsides and downsides of allocating to other baskets that may quite possibly be worse but might have a better shot under some futures, or other investment approaches.

All investing is ultimately speculative and uncertain. Prioritizing controlling on costs or complexity is reasonable enough, perhaps better. Some may rightly or wrongly look for other things.

User avatar
HomerJ
Posts: 11964
Joined: Fri Jun 06, 2008 12:50 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by HomerJ » Mon Apr 16, 2018 3:15 pm

lack_ey wrote:
Mon Apr 16, 2018 2:51 pm
HomerJ wrote:
Mon Apr 16, 2018 2:38 pm
lack_ey wrote:
Mon Apr 16, 2018 2:30 pm
These are among the many considerations you have to make looking at alternatives investments.
Is having many considerations to look at before investing in alt investments a pro or a con?
More work for the investor is generally bad, and juggling more things usually means more mistakes.

There are a lot of downsides to various types of alternatives even before you get to the usually significantly higher fees. We don't know that there are actually any benefits, but even supposing there are, they may not improve realized outcomes for any period we care about because returns are random. It can be a tough road, and a lot of the time you will look stupid following it.

Outside of perhaps some on the fringe, nobody wants to get rid of traditional asset classes for the long run. We all wish stock and bond returns will continue to the great, even remotely like past returns.

People focus too much on volatility and downside protection. The reality that long-term investors face is the possibility of a lot of assets not returning much of anything, and this needs to be weighted against the potential upsides and downsides of allocating to other baskets that may quite possibly be worse but might have a better shot under some futures, or other investment approaches.

All investing is ultimately speculative and uncertain. Prioritizing controlling on costs or complexity is reasonable enough, perhaps better. Some may rightly or wrongly look for other things.
That is a excellent post lack_ey.

NoRegret
Posts: 43
Joined: Sat Dec 16, 2017 2:00 am
Location: California
Contact:

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by NoRegret » Wed Apr 18, 2018 12:29 am

nisiprius wrote:
Wed Apr 11, 2018 7:15 pm
Low correlation? First things first. If your diversifier has less return than a money market mutual fund, it is going to drag down your return so much that it is almost impossible for any low correlation effect to offset it.
Excellent thread. I'm late to the conversation. You may find these two links interesting:

1. https://s3.amazonaws.com/static.content ... 0f17a7.pdf
2. https://www.rcmalternatives.com/2018/04 ... nto-a-bar/

The first paper from ThinkNewFound shows there're some nuances in asserting the lack of rebalance benefit simply because the return of the diversifying asset is low. The second just came to my attention today and I haven't thought through their calculations. But I found the results intriguing.

On a personal level, I'm a DIY investor and an unabashed market timer. I do my own hedging/leverage adjustment with minimal cost. For a static allocation, among the asset classes, I only have faith that gold will have near zero forward correlation with both stocks and bonds, and it doesn't have manager risk.

The raison d'etre of CTAs (commodity trading advisors, i.e. managed futures) has always been crisis alpha. In a large portfolio the portfolio manager farms pieces to various fund managers, one of which may be a CTA. I have always felt that hedging is best done at the portfolio level rather than the fund level. At a minimum the portfolio manager needs to make active adjustments, whereas a manager pursuing a static allocation is stuck. As a DIY market timer I face no such logical inconsistency.

NR
Market timer targeting long term cycles -- aiming for several key decisions per asset class per decade

User avatar
nisiprius
Advisory Board
Posts: 37101
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: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nisiprius » Wed Apr 18, 2018 7:07 am

NoRegret wrote:
Wed Apr 18, 2018 12:29 am
Excellent thread. I'm late to the conversation. You may find these two links interesting:

1. https://s3.amazonaws.com/static.content ... 0f17a7.pdf
I don't find it interesting, I find it frustrating because I am 99.9% certain that it is just plain wrong. However, they don't give enough details to spot the problem, and anyway life is too short to spend it trying to winkle out the flaw in every perpetual motion machine. This stuff is as difficult to deal with as the Monty Hall puzzle, where no matter how hard you try, people flatly disbelief the correct answer.

I believe that anyone who thinks adding coin-flips to a portfolio can improve it is delusional.

I will state this as a mathematical fact--I actually have done the algebra on it and confirmed it by experiment because I don't necessarily trust my own algebra.

Over some specific time period, if you have
--a "main portfolio" with a Sharpe ratio of A, and
--a "diversifying asset" with a Sharpe ratio of B, and
--the two have a correlation of R,

then adding the diversifier to the portfolio will improve it if and only if R < min(A, B) / max (A, B).

In particular, a diversifier with zero correlation and zero return cannot improve the portfolio.

(A problem with trying to show this by simulations is that if you in fact simulate their COINX--"at the end of each month, a coin is flipped by the portfolio team. If the coin lands on heads, the investors earn a 3% return for the day. If it lands on tails, the investors receive a -3% return," when you really do it, over any stated time period, there will usually be quite a large positive or negative return. Furthermore, there will be a positive or negative chance correlation with other assets. That means that in an actual case, it will improve or hurt the portfolio depending on chance. They don't say how they handled this).
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
nisiprius
Advisory Board
Posts: 37101
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: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nisiprius » Wed Apr 18, 2018 7:16 am

NoRegret wrote:
Wed Apr 18, 2018 12:29 am
...The raison d'etre of CTAs (commodity trading advisors, i.e. managed futures) has always been crisis alpha...
Well, look at the chart. Both the index and the category average existed in 2008-2009. What do you mean by "crisis alpha?" If you mean that they stayed level instead of plunging in 2008-2009, if that is "crisis alpha," then they showed it--but in that case bonds and cash also showed "crisis alpha" and they did it with less volatility and higher return, so why is the "crisis alpha" from managed futures preferable to the "crisis alpha" from bonds and cash?

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.

Theoretical
Posts: 1440
Joined: Tue Aug 19, 2014 10:09 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by Theoretical » Wed Apr 18, 2018 7:45 am

nisiprius wrote:
Wed Apr 18, 2018 7:16 am
NoRegret wrote:
Wed Apr 18, 2018 12:29 am
...The raison d'etre of CTAs (commodity trading advisors, i.e. managed futures) has always been crisis alpha...
Well, look at the chart. Both the index and the category average existed in 2008-2009. What do you mean by "crisis alpha?" If you mean that they stayed level instead of plunging in 2008-2009, if that is "crisis alpha," then they showed it--but in that case bonds and cash also showed "crisis alpha" and they did it with less volatility and higher return, so why is the "crisis alpha" from managed futures preferable to the "crisis alpha" from bonds and cash?

Image
The main reason is that they'd show that crisis alpha in an inflationary and rising rates environment at least as well as they did in the deflationary panic. Why?

2 reasons:

1. The collateral is typically invested in tbills or other ultrashort debt that's moderately correlated to inflation. If anything, the collateral was a drag in 2008 compared to how it would act in inflationary times.

2. In a 1970s scenario, the trend followers would be short US stocks, bonds, and the dollar, while being long foreign currencies and commodities.

In such a scenario cash has positive but mild returns and long bonds have very negative returns, especially on a real basis.


Edit: As an aside, one of the things I found a bit disappointing in Larry's new book on alternatives is that Managed Futures get very little weighting in the portfolio relative to the high tail risk lending, premia, reinsurance, and volatility funds, being weighted at only 20% of 25% instead of a lot more. It struck me as performance chasing and taking on some significant deep risk, especially for those taking from the bond allocation.

not4me
Posts: 475
Joined: Thu May 25, 2017 3:08 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by not4me » Wed Apr 18, 2018 12:43 pm

nisiprius wrote:
Wed Apr 18, 2018 7:16 am
NoRegret wrote:
Wed Apr 18, 2018 12:29 am
...The raison d'etre of CTAs (commodity trading advisors, i.e. managed futures) has always been crisis alpha...
Well, look at the chart. Both the index and the category average existed in 2008-2009. What do you mean by "crisis alpha?" If you mean that they stayed level instead of plunging in 2008-2009, if that is "crisis alpha," then they showed it--but in that case bonds and cash also showed "crisis alpha" and they did it with less volatility and higher return, so why is the "crisis alpha" from managed futures preferable to the "crisis alpha" from bonds and cash?

Image
I find myself struggling to get the point. Soon after the thread began, I'd read over & felt there were way more holes than cheese to the methodology & didn't "get the point". Now coming back, there's more to read than I have time to, but it seems to get more & more obscure. So, let me walk thru an example of my confusion -- I'll stick with the managed futures.

I think you're saying "the no-cost index shows a lovely, stable 11% annual return for more than a decade" & then an industry arose to sell a new category of mutual funds to retail investors....should that surprise me? Then with something new, look at the average so that I include all those who would never make it....having seen other new products enter a market (financial or not), it isn't surprising to me that some didn't do that great & brought the average down, especially considering the next point....

I'm not an advocate for managed futures & perhaps my understanding is at fault. But, I would have expected them to be somewhat based on needing risk to "manage". Great timing to intro funds, but that is also a time when central banks around the world intervened & (imho) caused the pricing of risk to become distorted...thus, I would NOT expect managed futures to perform as "advertised"...that is, that would be a time when they'd be "out of favor".

So, what am I to conclude? course of action?

User avatar
nisiprius
Advisory Board
Posts: 37101
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: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nisiprius » Wed Apr 18, 2018 12:56 pm

not4me wrote:
Wed Apr 18, 2018 12:43 pm
...So, what am I to conclude? course of action?...
In my case, my chosen course of action is:
  • Ignore "innovative" mutual funds that, for the first time, finally bring to retail investors strategies and asset classes, formerly available only to sophisticated wealthy investors, hedge funds, and institutional investors.
  • Ignore assets that aren't available as mutual funds, for the same reasons I don't invest in individual stocks, only more so. Beyond questions of diversification, these non-traditional assets are often subject to problems of low liquidity, less transparency, asymmetrical information, high transaction costs, wide bid-asked spreads, etc. I am not a "qualified" investor and I know it. The Investment Company Act of 1940 is my friend.
This isn't just managed futures. I don't have enough data to prove it, and I may have unconscious bias, but four different non-traditional assets--managed futures, commodities, multialternatives, market neutral strategies--all somehow managed to launch mutual funds just the exact moment when they were about to go out of favor. It almost looks like a pattern.

And, point that some seemed to misunderstand, I began by looking at category averages of stock and bond funds to show that they did OK despite including all the bad funds. You can't object to category averages unless you are confident that ordinary investors with reasonable diligence can reliably identify the good funds. On that theory, of course, we should all be using actively managed stock and bond funds.

So, this leads to a third course of action:
  • Before investing in a mutual fund in an unfamiliar asset category, check out the category averages, and think twice about categories that have performed poorly for the entire life of the category--unless you are truly convinced you know how to pick the good funds.
Last edited by nisiprius on Wed Apr 18, 2018 1:14 pm, edited 3 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

GAAP
Posts: 695
Joined: Fri Apr 08, 2016 12:41 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by GAAP » Wed Apr 18, 2018 1:06 pm

not4me wrote:
Wed Apr 18, 2018 12:43 pm
But, I would have expected them to be somewhat based on needing risk to "manage".
Maybe the only risk is a lack of new things to sell to suckers...

User avatar
nedsaid
Posts: 10712
Joined: Fri Nov 23, 2012 12:33 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nedsaid » Wed Apr 18, 2018 1:06 pm

nisiprius wrote:
Wed Apr 18, 2018 7:16 am
NoRegret wrote:
Wed Apr 18, 2018 12:29 am
...The raison d'etre of CTAs (commodity trading advisors, i.e. managed futures) has always been crisis alpha...
Well, look at the chart. Both the index and the category average existed in 2008-2009. What do you mean by "crisis alpha?" If you mean that they stayed level instead of plunging in 2008-2009, if that is "crisis alpha," then they showed it--but in that case bonds and cash also showed "crisis alpha" and they did it with less volatility and higher return, so why is the "crisis alpha" from managed futures preferable to the "crisis alpha" from bonds and cash?

Image
I might be wrong but I have two theories on what happened. First, was Peak China, my thought is that China's rates of economic were growth peaking right around 2008, just in time for the Beijing Olympics. As Chinese growth slowed, so did the formerly insatiable demand for commodities. Second, is the financialization of the Commodities Futures markets. People noticed the returns and a lot of institutional money flooded in, distorting, perhaps forever, these markets. I suppose you could say that the Bozos got in right at the time the Commodities market peaked. This happens too often to the retail investor.
A fool and his money are good for business.

NoRegret
Posts: 43
Joined: Sat Dec 16, 2017 2:00 am
Location: California
Contact:

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by NoRegret » Wed Apr 18, 2018 1:44 pm

nisiprius wrote:
Wed Apr 18, 2018 7:16 am
NoRegret wrote:
Wed Apr 18, 2018 12:29 am
...The raison d'etre of CTAs (commodity trading advisors, i.e. managed futures) has always been crisis alpha...
Well, look at the chart. Both the index and the category average existed in 2008-2009. What do you mean by "crisis alpha?" If you mean that they stayed level instead of plunging in 2008-2009, if that is "crisis alpha," then they showed it--but in that case bonds and cash also showed "crisis alpha" and they did it with less volatility and higher return, so why is the "crisis alpha" from managed futures preferable to the "crisis alpha" from bonds and cash?
If you're willing to grant that managed futures can provide the same "crisis alpha" as cash and bonds then why not use them all? The question then becomes whether managed futures can beat cash and bonds during non-crisis periods and it's a much lower bar to clear.

I'm not a fan of managed futures and probably will never use them myself. The real questions to holders of stock/bond two-asset portfolios are: Will the negative correlation between stocks and bonds continue to hold? What's the plan if they both drop as in a stagflationary environment, or a currency/bond crisis? I don't think these are mere conjectures.

NR
Market timer targeting long term cycles -- aiming for several key decisions per asset class per decade

User avatar
HomerJ
Posts: 11964
Joined: Fri Jun 06, 2008 12:50 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by HomerJ » Wed Apr 18, 2018 1:51 pm

NoRegret wrote:
Wed Apr 18, 2018 1:44 pm
The real questions to holders of stock/bond two-asset portfolios are: Will the negative correlation between stocks and bonds continue to hold? What's the plan if they both drop as in a stagflationary environment, or a currency/bond crisis?
Non-correlation isn't negative correlation.

Both could drop at the same time, but we'd expect bonds to drop less. In any case, the plan is to buy and hold, keep adding to our positions, and wait for the recovery.

NoRegret
Posts: 43
Joined: Sat Dec 16, 2017 2:00 am
Location: California
Contact:

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by NoRegret » Wed Apr 18, 2018 2:00 pm

HomerJ wrote:
Wed Apr 18, 2018 1:51 pm
NoRegret wrote:
Wed Apr 18, 2018 1:44 pm
The real questions to holders of stock/bond two-asset portfolios are: Will the negative correlation between stocks and bonds continue to hold? What's the plan if they both drop as in a stagflationary environment, or a currency/bond crisis?
Non-correlation isn't negative correlation.

Both could drop at the same time, but we'd expect bonds to drop less. In any case, the plan is to buy and hold, keep adding to our positions, and wait for the recovery.
What's the plan for someone in the withdrawal phase who thought bonds would protect them from a stock market decline and whose cash would be eroding in purchasing power? cf. the 1960's.
Market timer targeting long term cycles -- aiming for several key decisions per asset class per decade

User avatar
HomerJ
Posts: 11964
Joined: Fri Jun 06, 2008 12:50 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by HomerJ » Wed Apr 18, 2018 2:14 pm

NoRegret wrote:
Wed Apr 18, 2018 2:00 pm
HomerJ wrote:
Wed Apr 18, 2018 1:51 pm
NoRegret wrote:
Wed Apr 18, 2018 1:44 pm
The real questions to holders of stock/bond two-asset portfolios are: Will the negative correlation between stocks and bonds continue to hold? What's the plan if they both drop as in a stagflationary environment, or a currency/bond crisis?
Non-correlation isn't negative correlation.

Both could drop at the same time, but we'd expect bonds to drop less. In any case, the plan is to buy and hold, keep adding to our positions, and wait for the recovery.
What's the plan for someone in the withdrawal phase who thought bonds would protect them from a stock market decline and whose cash would be eroding in purchasing power? cf. the 1960's.
That's why we only pull 4%.

4% covers that worst case (so far worst case). 4% worked starting from most years in the 1960s (I think it was 3.8% for 1966, the worst year to retire in the past 100 years - luckily, more people had pensions back then).

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by lack_ey » Wed Apr 18, 2018 2:30 pm

nisiprius wrote:
Wed Apr 18, 2018 12:56 pm
This isn't just managed futures. I don't have enough data to prove it, and I may have unconscious bias, but four different non-traditional assets--managed futures, commodities, multialternatives, market neutral strategies--all somehow managed to launch mutual funds just the exact moment when they were about to go out of favor. It almost looks like a pattern.
I don't think that timeline is right, at least not for alts in general or for all of those four categories, mostly because I'm not sure about prior performance being any good (outside of well-known outperformers).

With equity market neutral, what kind of returns do you expect? If you think long-only active funds are unlikely to generate alpha above costs as a group via stockpicking, how can equity market neutral funds operating on the same set of securities generate alpha via stockpicking when they're charging higher fees on average and have borrowing costs on the short side? This is only really possible if some stocks are identifiably so bad that most managers would want a weighting below 0, and the alpha generated from long-only funds underweighting these stocks to 0 pales in comparison with going short on these stocks (not coincidentally, to the extent that identifiably bad stocks exist, they tend to be very expensive to short). I guess it's possible, but I really wonder about returns on market neutral generally. I'm not convinced that they got worse after mutual funds launched; they probably just weren't that good to begin with, evaluated over many managers including the ones nobody knows about because reporting is spotty outside of mutual funds. I suppose they may have gotten worse along with long-only stockpicking getting worse.

It depends on the multialternatives and the strategies pursued but to some extent I think this applies there as well.

With managed futures I think you're overstating the case of performance differences by comparing (1) a (backfilled?) strategy index to (2) live fund returns and then not accounting for differences in cash return. Correct me if I'm wrong. For older live fund returns, check the BTOP50 (an average of a couple dozen funds in the space), which actually goes back to 1987. With the BTOP50, I see a return from 1998 through the end of 2007 as 3.4% annualized above cash. Cash return was 3.5% annualized.

Let me generate a quick graph.

Image

I guess it's kind of changed since 2007 or so? Maybe not? I don't think it's that obvious or drastic a change, anyhow, even if you look at the mutual fund category average rather than the BTOP50 over your evaluation window.

As for commodities futures, that has probably actually changed. I'll give you that one, though there's also bad luck involved with the past decade of returns (inflation below expected, etc. etc.)

nisiprius wrote:
Wed Apr 18, 2018 12:56 pm
And, point that some seemed to misunderstand, I began by looking at category averages of stock and bond funds to show that they did OK despite including all the bad funds. You can't object to category averages unless you are confident that ordinary investors with reasonable diligence can reliably identify the good funds. On that theory, of course, we should all be using actively managed stock and bond funds.
I still really object to evaluating fund category performance like that relative to cash rather than relative to cheaply available passive exposures. You should not give funds credit for market beta. As investments the stock and bond categories did terribly, worse than some of the alt categories you looked at.

As for category averages, we can almost assuredly do better over the long run than that, picking funds within a category. Screen on ER and you can already win for most categories, just not over a costless benchmark. What's unknown and up to debate is by how much, and that should vary by category. As I've been saying, in some categories, many funds are very identifiably different in exposures and characteristics, though of course realized outcomes will always be largely subject to luck. It's not like even stocks beat bonds reliably.

nedsaid wrote:
Wed Apr 18, 2018 1:06 pm
I might be wrong but I have two theories on what happened. First, was Peak China, my thought is that China's rates of economic were growth peaking right around 2008, just in time for the Beijing Olympics. As Chinese growth slowed, so did the formerly insatiable demand for commodities. Second, is the financialization of the Commodities Futures markets. People noticed the returns and a lot of institutional money flooded in, distorting, perhaps forever, these markets. I suppose you could say that the Bozos got in right at the time the Commodities market peaked. This happens too often to the retail investor.
I think that's relevant to understanding what happened to commodities (futures) funds. But I don't see how that directly relates to managed futures, which are not net long commodities futures. They're variously long and short a lot of different assets, including outside commodities, usually averaging about 0 exposure to each over a market cycle. To explain the performance you need to look at trend behaviors, as most primarily use trend-following signals.

Texanbybirth
Posts: 999
Joined: Tue Apr 14, 2015 12:07 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by Texanbybirth » Wed Apr 18, 2018 2:44 pm

nisiprius wrote:
Wed Apr 18, 2018 12:56 pm
not4me wrote:
Wed Apr 18, 2018 12:43 pm
...So, what am I to conclude? course of action?...
In my case, my chosen course of action is:
  • Ignore "innovative" mutual funds that, for the first time, finally bring to retail investors strategies and asset classes, formerly available only to sophisticated wealthy investors, hedge funds, and institutional investors.
  • Ignore assets that aren't available as mutual funds, for the same reasons I don't invest in individual stocks, only more so. Beyond questions of diversification, these non-traditional assets are often subject to problems of low liquidity, less transparency, asymmetrical information, high transaction costs, wide bid-asked spreads, etc. I am not a "qualified" investor and I know it. The Investment Company Act of 1940 is my friend.
This isn't just managed futures. I don't have enough data to prove it, and I may have unconscious bias, but four different non-traditional assets--managed futures, commodities, multialternatives, market neutral strategies--all somehow managed to launch mutual funds just the exact moment when they were about to go out of favor. It almost looks like a pattern.

And, point that some seemed to misunderstand, I began by looking at category averages of stock and bond funds to show that they did OK despite including all the bad funds. You can't object to category averages unless you are confident that ordinary investors with reasonable diligence can reliably identify the good funds. On that theory, of course, we should all be using actively managed stock and bond funds.

So, this leads to a third course of action:
  • Before investing in a mutual fund in an unfamiliar asset category, check out the category averages, and think twice about categories that have performed poorly for the entire life of the category--unless you are truly convinced you know how to pick the good funds.
This is a great post.

I know it wasn't the only intended point of the OP; but it helped summarize the OP's thoughts in this thread for me, and it's probably the most important point.

User avatar
nisiprius
Advisory Board
Posts: 37101
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: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nisiprius » Wed Apr 18, 2018 3:52 pm

Lack_ey, I agree that the shape of the BTOP50 index looks more like some kind of continuous, smooth slowdown, not a corner at 2007.

I'm not quite sure what the BTOP50 Index is; these don't seem to be mutual funds:

AlphaGen Capital Limited (Long Short Agriculture Fund
AlphaSimplex Group, LLC (Managed Futures Composite)
AQR Capital Mgmt. (AQR Managed Futures MV Strategy)

And I don't know if the index is a theoretical index or whether it could be realized in practice in the real world. I've never heard of a mutual fund investing in hedge funds and I doubt it's possible.

By the way, since there are results through "April, 2018" I assume these are the beginnings months, not the ends of months, i.e. Jan 2018 = the total return obtained during the month of December, 2017. A month's shift shouldn't make much difference, but... what do you think?

Anyway, with a little work, I uploaded this into PortfolioVisualizer and am now ready to examine the question: if there had been a way to get the same returns as BTOP50--a low-cost BTOP50 index fund, say--would it have improved a traditional portfolio?

Portfolio 1 is 40% US Total Stock Market, 20% Global Ex-US Stock Market, 40% Total US Bond Market. Portfolio 2 is 32, 16, 32, and 20% BTOP50. The result is a considerable increase in Sharpe and Sortino ratios from adding BTOP50. But I'm always interested in "hold risk constant, take the improvement in return."

For Portfolio 3, I kept 20% BTOP50 and hand-adjusted the stock-bond allocation to bring the standard deviation close to equal to Portfolio 1. I ended up with 41.4% Total US Stocks, 20.7% Global Ex-US Stocks, 17.9% Total US Bonds, and, again, 20% BTOP50. There is an improvement of about 33 basis points in return, 0.33%, with no increase in standard deviation.

So, adding managed futures didn't hurt, and did produce a small improvement. It's not a very dramatic improvement, and In real life, I seriously doubt that those 33 basis points could be realized. Also, I won't display the results of "the opposite of cherry-picking" but if you shorten the length of the history, BTOP50 no longer produces any improvement.

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.

Valuethinker
Posts: 36734
Joined: Fri May 11, 2007 11:07 am

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by Valuethinker » Wed Apr 18, 2018 4:02 pm

nisiprius wrote:
Wed Apr 18, 2018 3:52 pm
Lack_ey, I agree that the shape of the BTOP50 index looks more like some kind of continuous, smooth slowdown, not a corner at 2007.

I'm not quite sure what the BTOP50 Index is; these don't seem to be mutual funds:

AlphaGen Capital Limited (Long Short Agriculture Fund
AlphaSimplex Group, LLC (Managed Futures Composite)
AQR Capital Mgmt. (AQR Managed Futures MV Strategy)

And I don't know if the index is a theoretical index or whether it could be realized in practice in the real world. I've never heard of a mutual fund investing in hedge funds and I doubt it's possible.
https://www.barclayhedge.com/research/indices/btop/
Barclay BTOP50 Index
The BTOP50 Index seeks to replicate the overall composition of the managed futures industry with regard to trading style and overall market exposure. The BTOP50 employs a top-down approach in selecting its constituents. The largest investable trading advisor programs, as measured by assets under management, are selected for inclusion in the BTOP50. In each calendar year the selected trading advisors represent, in aggregate, no less than 50% of the investable assets of the Barclay CTA Universe. To be included in the BTOP50, the following criteria must be met:

Program must be open for investment
Manager must be willing to provide us daily returns
Program must have at least two years of trading activity
Program's advisor must have at least three years of operating history
The BTOP50's portfolio will be equally weighted among the selected programs at the beginning of each calendar year and will be rebalanced annually.
For 2018 there are 20 funds in the Barclay BTOP50 Index. To see the complete list of constituents, scroll down o
Nothing to do with Barclays Bank and the indices it produces.

User avatar
nisiprius
Advisory Board
Posts: 37101
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: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nisiprius » Wed Apr 18, 2018 4:05 pm

Valuethinker wrote:
Wed Apr 18, 2018 4:02 pm
Barclay BTOP50 Index
The BTOP50 Index seeks to replicate the overall composition of the managed futures industry with regard to trading style and overall market exposure. The BTOP50 employs a top-down approach in selecting its constituents. The largest investable trading advisor programs, as measured by assets under management, are selected for inclusion in the BTOP50. In each calendar year the selected trading advisors represent, in aggregate, no less than 50% of the investable assets of the Barclay CTA Universe. To be included in the BTOP50, the following criteria must be met:

Program must be open for investment
Manager must be willing to provide us daily returns
Program must have at least two years of trading activity
Program's advisor must have at least three years of operating history
The BTOP50's portfolio will be equally weighted among the selected programs at the beginning of each calendar year and will be rebalanced annually.
For 2018 there are 20 funds in the Barclay BTOP50 Index. To see the complete list of constituents, scroll down
Nothing to do with Barclays Bank and the indices it produces.
But what are those? What is an "advisor program?" Is that a fancy name for a hedge fund? In the real world could you created a mutual fund that invests equally in each of 20 hedge funds? How liquid are these "advisor programs" and how meaningful is it to tabulate an "index" if you can't actually liquidate your holdings in each of these twenty "advisor programs," at will, in any desired month? And are those returns net of expenses? You'd hope so, but why don't they say?
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.

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by lack_ey » Wed Apr 18, 2018 5:26 pm

nisiprius wrote:
Wed Apr 18, 2018 4:05 pm
But what are those? What is an "advisor program?" Is that a fancy name for a hedge fund? In the real world could you created a mutual fund that invests equally in each of 20 hedge funds? How liquid are these "advisor programs" and how meaningful is it to tabulate an "index" if you can't actually liquidate your holdings in each of these twenty "advisor programs," at will, in any desired month? And are those returns net of expenses? You'd hope so, but why don't they say?
The phrasing ("trading advisor programs") probably comes from the fact that historically these are known as commodity trading advisors (CTAs). Managed futures is maybe a more generic and broader term, probably more accurate as many trade more than just commodity futures. Many are still called CTAs but "managed futures" has increasingly become better known, I think.

They're looking at investment funds, whatever the structure. I think all those listed are not '40 Act mutual funds. There are various different institutional arrangements out there. You could call many of them hedge funds, though that's not a description of a legal structure either. Performance should be net of fees.

I think for many of these you could probably have monthly liquidity, if not better than that. After all, the underlying futures traded range from extremely to fairly liquid.These kinds of strategies are available as mutual funds, after all. For some, maybe not. Generally speaking, I think the performance over a given period is of interest (to understand the strategy and characteristics), whether you can liquidate on that timeframe or not.

The BTOP50 tracks investable (by institutions) funds, those accepting new money. If you were running a large pension fund, you could invest in all 20 of the funds covered.

There's not much particularly special about this index other than the fact that it goes back to 1987. I reference it for that reason. Well, it also may be a better representation of live-trading managed futures performance overall than the Morningstar category, based on the relative AUM covered.

User avatar
nedsaid
Posts: 10712
Joined: Fri Nov 23, 2012 12:33 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nedsaid » Wed Apr 18, 2018 6:09 pm

lack_ey wrote:
Wed Apr 18, 2018 2:30 pm
nedsaid wrote:
Wed Apr 18, 2018 1:06 pm
I might be wrong but I have two theories on what happened. First, was Peak China, my thought is that China's rates of economic were growth peaking right around 2008, just in time for the Beijing Olympics. As Chinese growth slowed, so did the formerly insatiable demand for commodities. Second, is the financialization of the Commodities Futures markets. People noticed the returns and a lot of institutional money flooded in, distorting, perhaps forever, these markets. I suppose you could say that the Bozos got in right at the time the Commodities market peaked. This happens too often to the retail investor.
I think that's relevant to understanding what happened to commodities (futures) funds. But I don't see how that directly relates to managed futures, which are not net long commodities futures. They're variously long and short a lot of different assets, including outside commodities, usually averaging about 0 exposure to each over a market cycle. To explain the performance you need to look at trend behaviors, as most primarily use trend-following signals.
I read in Rick Ferri's book on asset allocation that the real return on commodities over long periods of time is a negative one percent a year. I figured this would make any commodities strategy an ultimate loser. Then someone pointed out, well that is true, but commodity futures have more stock-like returns. It was further explained to me that trading in the commodities themselves was one thing, trading in the futures was another thing altogether. It was here I learned about Backwardization and Contango, the commodity futures markets were at one time in Backwardization so Collateralized Commodity Futures could get a so-called roll return as spot prices were higher than futures prices. Pretty much, the strategy had an underlying return over time similar to stocks and thus made the chances of a CCF strategy succeeding very likely.

My understanding is that so much money flooded into these futures markets that commodity futures went into Contango, that is futures prices are higher than spot prices. So instead of having a positive rate of return built into a CCF strategy, you have a negative rate of return built in on top of the negative real return of the commodities themselves. So yes, a managed futures strategy could still work but that assumes you will actually have success with shorting strategies. More things have to go right now for the Managed Futures strategy to achieve positive returns.

Perhaps you could increase my understanding of how all this works. But it seems to me that now you have two weak links in the chain rather than just one. The weak links being negative real returns for the commodities themselves and Contango. So you have headwinds now and not tailwinds. Different from 2000-2008 when commodities themselves were in a bull market, and the futures markets in backwardization. You had two links in the chain that were strong and not weak and thus giving the Managed Futures strategy headwinds.

It sounds like the Managed Futures strategy is a momentum strategy as you mentioned trend following. This is all very interesting.
A fool and his money are good for business.

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by lack_ey » Wed Apr 18, 2018 6:34 pm

nedsaid wrote:
Wed Apr 18, 2018 6:09 pm
I read in Rick Ferri's book on asset allocation that the real return on commodities over long periods of time is a negative one percent a year. I figured this would make any commodities strategy an ultimate loser. Then someone pointed out, well that is true, but commodity futures have more stock-like returns. It was further explained to me that trading in the commodities themselves was one thing, trading in the futures was another thing altogether. It was here I learned about Backwardization and Contango, the commodity futures markets were at one time in Backwardization so Collateralized Commodity Futures could get a so-called roll return as spot prices were higher than futures prices. Pretty much, the strategy had an underlying return over time similar to stocks and thus made the chances of a CCF strategy succeeding very likely.

My understanding is that so much money flooded into these futures markets that commodity futures went into Contango, that is futures prices are higher than spot prices. So instead of having a positive rate of return built into a CCF strategy, you have a negative rate of return built in on top of the negative real return of the commodities themselves. So yes, a managed futures strategy could still work but that assumes you will actually have success with shorting strategies. More things have to go right now for the Managed Futures strategy to achieve positive returns.

Perhaps you could increase my understanding of how all this works. But it seems to me that now you have two weak links in the chain rather than just one. The weak links being negative real returns for the commodities themselves and Contango. So you have headwinds now and not tailwinds. Different from 2000-2008 when commodities themselves were in a bull market, and the futures markets in backwardization. You had two links in the chain that were strong and not weak and thus giving the Managed Futures strategy headwinds.
That's a bit of an oversimplification as even 20+ years ago different futures would variously change from being in backwardation and contango, generally at different times. Even today, many are in backwardation. The amount of backwardation vs. contango changes all the time.

Previously there was on average a bias to backwardation. Recently maybe more like the reverse. But that's overall, averaging. In any case, this is just one component of returns.

If investing in futures, you also get a return on all the collateral (all the otherwise uninvested cash, which you could put in short-term bonds or TIPS or whatever else). In theory the futures pricing reflects what the collateral return would be, but maybe not necessarily or all the time with these kinds of futures where storage is difficult and various arbitrage mechanisms that apply elsewhere don't really go into effect. In any case, this portion of the return used to be higher as well.
nedsaid wrote:
Wed Apr 18, 2018 6:09 pm
It sounds like the Managed Futures strategy is a momentum strategy as you mentioned trend following. This is all very interesting.
Yes, that's the actual point. Everything we've been discussing about being net long commodities futures is not really that relevant, as managed futures strategies are not generally net long (at least not by much) commodities when averaged over time. They just go short futures contracts of whatever is going down, and go long whatever's going up. Some also look at signals other than trend, but largely it's about trend following.

It's a bit like you're telling me that stock valuations are high and I'm managing an equity market neutral fund. Okay, that matters to those who are long stocks. But if I'm long/short, then that just means I'm long expensive stocks and short expensive stocks. It doesn't seem to be that relevant in this case.

longinvest
Posts: 3114
Joined: Sat Aug 11, 2012 8:44 am

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by longinvest » Wed Apr 18, 2018 6:55 pm

A Futures Contract is a derivative involving a buyer and a seller who both pay a fee for the contract. After costs, it's a negative-sum game for the players as a group.

Playing a negative-sum game on top of a negative-return* asset is highly speculative. I fail to see any relation between that and the our investment philosophy.

* A commodity has no internal return, but it has storage costs.
Bogleheads investment philosophy | Lifelong Portfolio: 25% each of (domestic / international) stocks / domestic (nominal / inflation-indexed) long-term bonds | VCN/VXC/VLB/ZRR

Waba
Posts: 376
Joined: Wed Feb 26, 2014 9:19 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by Waba » Wed Apr 18, 2018 7:27 pm

longinvest wrote:
Wed Apr 18, 2018 6:55 pm
A Futures Contract is a derivative involving a buyer and a seller who both pay a fee for the contract. After costs, it's a negative-sum game for the players as a group.
But the group is not homogeneous. For example Oil producing companies use future contracts to secure a fixed price for a part of their future production so they can make sure they won't go out of business when the oil price suddenly drops. To them it's an insurance product and not an investment, it's reasonable to expect that there is a premium to collect for the counter-party that provides them that insurance.

User avatar
nisiprius
Advisory Board
Posts: 37101
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: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by nisiprius » Thu Apr 19, 2018 8:17 am

With regard to the nature of the BTOP50 index, I emailed BarclayHedge with some questions and received a reply saying:

--returns are net of fees

--the April return as of April 18 reflects an estimate of the Index return from April 1 - 17 based on the daily returns of he underlying funds.

--The advisors do have monthly liquidity. And yes, it is possible to do if you set it up properly.

(My third question was whether these programs have monthly liquidity, so that, if you had the wherewithal to actually invest in all of them, you could actually obtain the returns of the index).
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.

VinhoVerde
Posts: 90
Joined: Fri Nov 21, 2014 8:39 am

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by VinhoVerde » Thu Apr 19, 2018 9:57 am

It is my contention that gold is a separate asset class from commodities or currencies. I communicate with an humble mini iPad and am unable to reproduce Nisiprius charts. Is anyone able to reproduce a 10 year chart with GLD versus VMMXX?
Thanks,
VinhoVerde

lack_ey
Posts: 6636
Joined: Wed Nov 19, 2014 11:55 pm

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by lack_ey » Thu Apr 19, 2018 10:20 am

VinhoVerde wrote:
Thu Apr 19, 2018 9:57 am
It is my contention that gold is a separate asset class from commodities or currencies. I communicate with an humble mini iPad and am unable to reproduce Nisiprius charts. Is anyone able to reproduce a 10 year chart with GLD versus VMMXX?
Thanks,
VinhoVerde
Image
http://quotes.morningstar.com/chart/fun ... A%5B%5D%7D

You only wanted 10 years?

Gold in some contexts is considered a hard commodity, in the same group as other mined materials. As an investment, given relative ease of storage, ability to own physical (or with a fund that owns physical) and avoid futures, etc. given the historical context and based on widespread usage across cultures, it's more of its own category, I suppose.

VinhoVerde
Posts: 90
Joined: Fri Nov 21, 2014 8:39 am

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by VinhoVerde » Thu Apr 19, 2018 12:00 pm

I am very appreciative, thanks.
VinhoVerde

VinhoVerde
Posts: 90
Joined: Fri Nov 21, 2014 8:39 am

Re: Unimpressive category averages of managed futures, multialternative, market neutral, etc.

Post by VinhoVerde » Thu Apr 19, 2018 3:21 pm

Looking at the GLD vs VMMXX chart above, GLD, a ETF proxy for physical gold, is indeed a good diversifier for a stock/ bond portfolio and does outperform a month money market fund. In 2010-2011 the price almost doubled when it was widely believed that congress would allow the treasury to default on debt payments and S&P downgraded US bonds. It has declined since as the stock market has continued to strengthen and default less likely.
However,,it has strengthened 3.8% YTD while Vanguard Total Stock is down .6% and Total Bond down 1.4%. It does zig while stocks and bonds zag. It's the little gold alt engine that could. I use it as 10% of my portfolio.
VinhoVerde

Post Reply