Timing the market

BritishInvestor

MarkCarnage said:

 If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

I suggest that it's a bit more nuanced than explaining it away by tilts to factors. The active fund managers I have invested with both institutionally and personally didn't think about tilting to some factor or another which ex post might 'explain' their outperformance. Nor would they decide to sell a company where they were still comfortable with it's long term prospects and valuation just because it didn't fit into a particular factor category. If they underperformed a market index for a year or two as a result of several such decisions, it didn't particularly bother them or me. 

I spent a lot more time being comfortable with a manager's philosophy and process, and the robustness of that process than the previous 12 months performance or whether a factor tilt might explain it. I've terminated mandates/sold holdings for various reasons, but I struggle to recall short/medium term performance being one. 

"The active fund managers I have invested with both institutionally and personally didn't think about tilting to some factor or another which ex post might 'explain' their outperformance. "

Maybe, but as an investor, I'd want to know what I was paying for. 

MK62

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

Hmmm, not what I meant.......
If you give a number of people the same piece of information, it's unlikely they'd all interpret it the same way, and act on it in the same way........fund managers are no different. 

Prism

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

The fund managers I use don't use any clever software, timing or sell side research. They just try and do something different to the masses.

BritishInvestor

MK62 said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

Hmmm, not what I meant.......
If you give a number of people the same piece of information, it's unlikely they'd all interpret it the same way, and act on it in the same way........fund managers are no different. 

But it doesn't change my point - fund managers will be unlikely to react to news quickly enough to profit from short term inefficiencies. And making money from long term inefficiencies is extraordinarily difficult (some would say impossible).

There's a chance it's a dying industry, much like other parts of the markets have been over the years. 

[Deleted User]

BritishInvestor said:

MK62 said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

Hmmm, not what I meant.......
If you give a number of people the same piece of information, it's unlikely they'd all interpret it the same way, and act on it in the same way........fund managers are no different. 

But it doesn't change my point - fund managers will be unlikely to react to news quickly enough to profit from short term inefficiencies. And making money from long term inefficiencies is extraordinarily difficult (some would say impossible).

There's a chance it's a dying industry, much like other parts of the markets have been over the years. 

I would say “changing”. Being an old-fashioned fund manager is very hard.  Lots of people received the same training so competition is tough. Snake oil is harder to sell. Then you have passive and robos.  Of course “passive” also requires managers.  Technology is changing what fund managers do.  Like they may be able to offer DIY ETFs with clients adding or removing certain types of companies to them. 

BritishInvestor

Deleted_User said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

Hmmm, not what I meant.......
If you give a number of people the same piece of information, it's unlikely they'd all interpret it the same way, and act on it in the same way........fund managers are no different. 

But it doesn't change my point - fund managers will be unlikely to react to news quickly enough to profit from short term inefficiencies. And making money from long term inefficiencies is extraordinarily difficult (some would say impossible).

There's a chance it's a dying industry, much like other parts of the markets have been over the years. 

I would say “changing”. Being an old-fashioned fund manager is very hard.  Lots of people received the same training so competition is tough. Snake oil is harder to sell. Then you have passive and robos.  Of course “passive” also requires managers.  Technology is changing what fund managers do.  Like they may be able to offer DIY ETFs with clients adding or removing certain types of companies to them. 

Yes, agreed on changing. Re your last point - direct indexing? But that surely can be automated to a large degree and offered for a very low cost? 

Linton

BritishInvestor said:

MK62 said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

Hmmm, not what I meant.......
If you give a number of people the same piece of information, it's unlikely they'd all interpret it the same way, and act on it in the same way........fund managers are no different. 

But it doesn't change my point - fund managers will be unlikely to react to news quickly enough to profit from short term inefficiencies. And making money from long term inefficiencies is extraordinarily difficult (some would say impossible).

There's a chance it's a dying industry, much like other parts of the markets have been over the years. 

Why should fund managers try to react to the latest news or profit from long term inefficiencies? The only reason to do this is chasing the last %s of possible return.  Maximum return generally should not be the objective of a serious investor.   Far more important is the objective and  management of the risk in achieving it.  This is getting increasingly problematic as online access for inexperienced investors is leading to waves of high risk investment driven more by social media than proper financial analysis.  This is reflected in the indexes which follow the herd into lala land as they are limited by the straight-jacket of capitalisation weighting .  We saw the consequences in the .com boom/bust and I fear it will happen again.

Especially now that bonds cannot provide effective risk mitigation with a useful diversifying return, management of risk requires control of asset allocation and tilts.  This is much easier with active funds since in most fund sectors there is a range of different funds with different industry sector allocations and tilt preferences.  One can choose the ones that together create the portfolio one believes will achieve objectives at sufficiently low risk. It is impossible to do this with the range of passive funds that are available in the UK at the moment.

The second and simpler risk reduction role is to provide due diligence.  Choosing next years winner is impossible but trying to avoid the companies that wont be is much easier.

[Deleted User]

BritishInvestor said:

Deleted_User said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

Hmmm, not what I meant.......
If you give a number of people the same piece of information, it's unlikely they'd all interpret it the same way, and act on it in the same way........fund managers are no different. 

But it doesn't change my point - fund managers will be unlikely to react to news quickly enough to profit from short term inefficiencies. And making money from long term inefficiencies is extraordinarily difficult (some would say impossible).

There's a chance it's a dying industry, much like other parts of the markets have been over the years. 

I would say “changing”. Being an old-fashioned fund manager is very hard.  Lots of people received the same training so competition is tough. Snake oil is harder to sell. Then you have passive and robos.  Of course “passive” also requires managers.  Technology is changing what fund managers do.  Like they may be able to offer DIY ETFs with clients adding or removing certain types of companies to them. 

Yes, agreed on changing. Re your last point - direct indexing? But that surely can be automated to a large degree and offered for a very low cost? 

Yes, exactly.  Right now we can either do stock picking or fund/ETF picking.  But the former is hard to do well and the latter is never quite what you want.  Like all the proliferating "ethical" funds never provide the kind of ethics that the customers actually like. 

Or let's say, I have a hate on... Tesla.  Or rubber manufacturers.  Or that I work for Apple and don't want any APL in my ETFs to avoid extra risk.  Or that I like Emerging markets but not communist China or President Erdogan.   Would be nice to pick an all-world ETF and then subtract a few bits and pieces one wants to avoid.  

Basically, I am talking about customizable ETFs.  There will be some cost to it (compared to basically zero right now) but technology should be able to allow for it to be done very cheaply.  

These are the kinds of innovation that good fund providers might be offering next while the good old "trust in me" type of manager struggles. 

NedS

Deleted_User said:

Yes, exactly.  Right now we can either do stock picking or fund/ETF picking.  But the former is hard to do well and the latter is never quite what you want.  Like all the proliferating "ethical" funds never provide the kind of ethics that the customers actually like. 

Or let's say, I have a hate on... Tesla.  Or rubber manufacturers.  Or that I work for Apple and don't want any APL in my ETFs to avoid extra risk.  Or that I like Emerging markets but not communist China or President Erdogan.   Would be nice to pick an all-world ETF and then subtract a few bits and pieces one wants to avoid.  

Basically, I am talking about customizable ETFs.  There will be some cost to it (compared to basically zero right now) but technology should be able to allow for it to be done very cheaply.  

These are the kinds of innovation that good fund providers might be offering next while the good old "trust in me" type of manager struggles. 

I've often thought exactly the same. Why would I want to buy the market knowing full well there are 10, or 20, or 50 companies I definitely don't want to own in that basket of stocks. I may not be able (or feel confident) to pick a portfolio of winners, but I'd love the opportunity to buy an ETF style basket of stocks and then have the ability to customise it by deselecting the stuff I know I don't want, and I'd be happy to pay a higher management fee for that ability to customise.

BritishInvestor

Linton said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MK62 said:

BritishInvestor said:

MarkCarnage said:

I have always been quite sceptical about the robustness of outcomes from factor, or so called 'smart beta' investing. The theory of it sounds fine up to a point. How do you know 'ex ante' what factors are going to perform well and when? It smacks of the market timing argument to me.
It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money. Of course good fund managers will analyse their portfolios for factor tilts, but very often to check that there are no unanticipated ones there. However, it is not always easy to pin stocks down neatly into factors. They will and do change over time. 
I think that either passive or fundamental active are preferable. 

"How do you know 'ex ante' what factors are going to perform well and when?"

You don't, and that's exactly my point. If fund manager "outperformance" is explained away by tilts to factors which cannot be timed, why would you pay for this?

"It's also true that where there were perhaps informational advantages to be exploited that they have been eroded by volume of money"
Yep, I'd be surprised if an active manager had any form of informational advantage in modern markets - far, far too competitive for that. 

It's not just the information itself (though personally I suspect fund managers probably get quite a bit a little before it hits Google), but how it's interpreted and acted upon.

That will be these people

https://en.wikipedia.org/wiki/Renaissance_Technologies

The fund managers will be way behind the curve

https://ofdollarsanddata.com/medallion-fund/

"Unbeknownst to Mercer, Chrysler had been acquired by Daimler AG in years prior and no longer existed as a stock! "

Hmmm, not what I meant.......
If you give a number of people the same piece of information, it's unlikely they'd all interpret it the same way, and act on it in the same way........fund managers are no different. 

But it doesn't change my point - fund managers will be unlikely to react to news quickly enough to profit from short term inefficiencies. And making money from long term inefficiencies is extraordinarily difficult (some would say impossible).

There's a chance it's a dying industry, much like other parts of the markets have been over the years. 

Why should fund managers try to react to the latest news or profit from long term inefficiencies? The only reason to do this is chasing the last %s of possible return.  Maximum return generally should not be the objective of a serious investor.   Far more important is the objective and  management of the risk in achieving it.  This is getting increasingly problematic as online access for inexperienced investors is leading to waves of high risk investment driven more by social media than proper financial analysis.  This is reflected in the indexes which follow the herd into lala land as they are limited by the straight-jacket of capitalisation weighting .  We saw the consequences in the .com boom/bust and I fear it will happen again.

Especially now that bonds cannot provide effective risk mitigation with a useful diversifying return, management of risk requires control of asset allocation and tilts.  This is much easier with active funds since in most fund sectors there is a range of different funds with different industry sector allocations and tilt preferences.  One can choose the ones that together create the portfolio one believes will achieve objectives at sufficiently low risk. It is impossible to do this with the range of passive funds that are available in the UK at the moment.

The second and simpler risk reduction role is to provide due diligence.  Choosing next years winner is impossible but trying to avoid the companies that wont be is much easier.

"Why should fund managers try to react to the latest news or profit from long term inefficiencies? "

If you're not paying them for alpha, what are you paying for?

" Far more important is the objective and  management of the risk in achieving it"

Agreed, but if you accept that risk and return are related, and costs are a drag on returns, an active manager has to take more risk to generate the same return as a tracker, in theory.

"This is getting increasingly problematic as online access for inexperienced investors is leading to waves of high risk investment driven more by social media than proper financial analysis.  This is reflected in the indexes which follow the herd into lala land as they are limited by the straight-jacket of capitalisation weighting .  We saw the consequences in the .com boom/bust and I fear it will happen again."

I think it's worth separating the human decisions from the implementation. While index investing might be contributing to bubbles. a quick look at the top-selling funds indicate that they are dominated by large-cap growth active funds - I would suggest the problem is more the human and their chasing of recent winners.
There's nothing to stop you using a small cap passive fund (for example) if you want to tilt away from overall market cap.

"Especially now that bonds cannot provide effective risk mitigation with a useful diversifying return, management of risk requires control of asset allocation and tilts."

I don't really understand what an active manager would do here, and given that most star fund managers at the moment are in the large cap growth space, some would argue this is quite a risky space to be in given current valuations.

"This is much easier with active funds since in most fund sectors there is a range of different funds with different industry sector allocations and tilt preferences."

" One can choose the ones that together create the portfolio one believes will achieve objectives at sufficiently low risk. It is impossible to do this with the range of passive funds that are available in the UK at the moment."

Would be useful to see some examples of how this has worked historically (if we accept that tactical asset allocation doesn't work), and also what you cannot do with the available passive/factor funds.

"The second and simpler risk reduction role is to provide due diligence.  Choosing next years winner is impossible but trying to avoid the companies that wont be is much easier."

I'm not clear how you undertake successful due diligence. I once raised a Roger Moore style eyebrow when a portfolio building firm said that part of their due diligence was looking a fund manager in the eye! I'm not also not clear how you avoid the companies that won't be winners - that implies a market mispricing unless I'm misunderstanding what you are saying?