Why Hold Cash?

[randomep]

 

YES YES I am aware......

 

I addressed at in my post yesterday......

[Happy]

Some points:

- He might be able to meet his higher return requirement even in years where the market doesn't drop. It seems very tough to make realistic assumptions how often the cash is (not) invested.

- If you have some cash left, you might work harder to find something. Before this he just looked for 2x in 2-3 years and that was it. By patiently looking for the really fat pitches whenever he has some cash the quality of his positions might improve considerably improve over time.

- As the fund manager, his approach might make sense long-term because Pabrai's drop would have been far less severe and he would have made it back much quicker. If that lowers the volatility, it makes his fund more attractive and an increase in AUM could make it up for Pabrai even IF his returns should turn out slightly lower because of the cash. For clients it may make sense because they are less likely to leave at the wrong time (protecting them from themselves).

- Assuming equal returns per year (just +15% and -15%) might not be fair because the cash is less likely to be deployed after the market has shot up and more likely when it went down. So the less volatile your model, the worse cash does I would assume.

[racemize]

Some points:

- He might be able to meet his higher return requirement even in years where the market doesn't drop. It seems very tough to make realistic assumptions how often the cash is (not) invested.

- If you have some cash left, you might work harder to find something. Before this he just looked for 2x in 2-3 years and that was it. By patiently looking for the really fat pitches whenever he has some cash the quality of his positions might improve considerably improve over time.

- As the fund manager, his approach might make sense long-term because Pabrai's drop would have been far less severe and he would have made it back much quicker. If that lowers the volatility, it makes his fund more attractive and an increase in AUM could make it up for Pabrai even IF his returns should turn out slightly lower because of the cash. For clients it may make sense because they are less likely to leave at the wrong time (protecting them from themselves).

- Assuming equal returns per year (just +15% and -15%) might not be fair because the cash is less likely to be deployed after the market has shot up and more likely when it went down. So the less volatile your model, the worse cash does I would assume.

 

I agree, I think I've mostly come to a conclusion on this one, so I'll probably write up an essay on it.  If people are interested, I'll post it here.

[Happy]

I'd like to read it.

[Liberty]

I agree, I think I've mostly come to a conclusion on this one, so I'll probably write up an essay on it.  If people are interested, I'll post it here.

 

I'd love to read it. Thank you in advance for sharing your thoughts.

[racemize]

I'm in final stages of data collection at this point.  If any investors feel like posting or pm'ing me their 5+ year annual performance numbers, I can enter it in and see how much cash would have been ideal, given the returns.  Mostly, I'm doing this to see if it matches my thesis and compare it to real world data, so the more the merrier!  Perhaps you would like to know the answer as well?  (although probably you already know it).

 

In any event, I'll put it all together in an essay and get it posted, but those usually take me a while, so don't hold your breath!

 

 

[randomep]

Alice Schroeder talked about how Buffett thinks of cash as a call option with no expiration, and no strike on any asset class.

The cost is the opportunity cost of holding cash.  I am still trying to fully understand it.

 

http://www.portfoliomanagement.ca/warren-buffett-article.html

 

A couple of times when Buffett went largely to cash:

1.  1969 - he closed his partnership because he could not find anything.  I think the Dow was down about 30% after that.

2.  1987 - He sold a lot of stocks into the bull market before the crash.

 

 

I think Bernard Baruch and Joseph Kennedy had a lot of cash going into the Great Depression. 

Howard Marks had a fund ready with cash for the 2008 crash.

 

I don't think I can time the market, but if almost everything is overvalued I will just wait it out in cash.

 

I have usually been able to find new ideas on average within 2 years so base my calculations off that time frame. 

I would rather sit in cash earning 0% for 1 year, then deploy it in an idea with an expected IRR of 20%+ for a few years rather than buying something with an expected return of 10%.

 

 

 

I think to read stuff that brings new ideas... and this is a great one!

 

I think what Buffett means when he says cash is like call options is:

 

- a call option with strike price $10 will expire if it doesn't move, and will net $10 if it goes to $20

- if you hold cash and a stock is $20, it will do nothing if it stays at $20, but you will net $10 in VALUE if it drops to $10

 

 

[Ham Hockers]

Alice Schroeder talked about how Buffett thinks of cash as a call option with no expiration, and no strike on any asset class.

The cost is the opportunity cost of holding cash.  I am still trying to fully understand it.

 

http://www.portfoliomanagement.ca/warren-buffett-article.html

 

A couple of times when Buffett went largely to cash:

1.  1969 - he closed his partnership because he could not find anything.  I think the Dow was down about 30% after that.

2.  1987 - He sold a lot of stocks into the bull market before the crash.

 

 

I think Bernard Baruch and Joseph Kennedy had a lot of cash going into the Great Depression. 

Howard Marks had a fund ready with cash for the 2008 crash.

 

I don't think I can time the market, but if almost everything is overvalued I will just wait it out in cash.

 

I have usually been able to find new ideas on average within 2 years so base my calculations off that time frame. 

I would rather sit in cash earning 0% for 1 year, then deploy it in an idea with an expected IRR of 20%+ for a few years rather than buying something with an expected return of 10%.

 

 

 

I think to read stuff that brings new ideas... and this is a great one!

 

I think what Buffett means when he says cash is like call options is:

 

- a call option with strike price $10 will expire if it doesn't move, and will net $10 if it goes to $20

- if you hold cash and a stock is $20, it will do nothing if it stays at $20, but you will net $10 in VALUE if it drops to $10

 

 

 

i think schroeder explained it pretty well. option price of cash is opportunitiy cost less interest earned on cash holdings. underlying security is EVERY sercurity. strike price is whatever you want it to be. term is forever.

 

cash has been a very expensive option for a couple years now.

[BG2008]

I think it is very hard to adhere to this principal.  In reality, most people tend to allocate their best ideas to their first 20-30% cash.  After that, the quality of the ideas tend to decline.  It's like

 

1st 30% - 3x in 2-3 years, 20% downside

2nd 30% - 2x in 2-3 years, 40% downside

3rd 20% - 1.5X in 2-3 years, 50% downside

Last 20% - Marginal, coin flip, 50/50 upside vs downside

 

The fact that this is the reality of how most people allocate capital is the reason why it's a good idea to hold cash.  If you can adhere to Pabrai's allocation method, then it probably makes sense.  The reality is that most people's portfolio are built the way that I illustrated. 

 

For example, after 2008, Pabrai has indicated that he thinks about cash allocation in the following manner:

1st 75% cash - 2x in 2-3 years

next 10% cash - 3x in 2-3 years

next 5% cash - 4x in 2-3 years

next 5% cash - 5x in 2-3 years

last 5% cash - >5x in 2-3 years

 

 

[Aurelius]

I've been reading this thread with great interest. It's a subject I've been having real trouble with. I've been going back and forth with regards to how to position my portfolio.

 

Since CAPE and Tobin Q have been the most reliable indicators on overvaluation and subsequent future returns I was wondering if you had tried to model if it makes sense to hold higher cash level when CAPE/Tobin Q are high assuming you got market returns.

 

It could be modeled something like this (using CAPE):

Less than 20; cash 0

20; cash 5%

21; cash 7%

22; cash 9%

23; cash 11%

24; cash 13%

25; cash 15 %

Etc. Or something of the sort.

 

I hope you post your essay as it seems like an interesting read.

[constructive]

Alice Schroeder talked about how Buffett thinks of cash as a call option with no expiration, and no strike on any asset class.

The cost is the opportunity cost of holding cash.  I am still trying to fully understand it.

 

http://www.portfoliomanagement.ca/warren-buffett-article.html

 

A couple of times when Buffett went largely to cash:

1.  1969 - he closed his partnership because he could not find anything.  I think the Dow was down about 30% after that.

2.  1987 - He sold a lot of stocks into the bull market before the crash.

 

 

I think Bernard Baruch and Joseph Kennedy had a lot of cash going into the Great Depression. 

Howard Marks had a fund ready with cash for the 2008 crash.

 

I don't think I can time the market, but if almost everything is overvalued I will just wait it out in cash.

 

I have usually been able to find new ideas on average within 2 years so base my calculations off that time frame. 

I would rather sit in cash earning 0% for 1 year, then deploy it in an idea with an expected IRR of 20%+ for a few years rather than buying something with an expected return of 10%.

 

 

 

I think to read stuff that brings new ideas... and this is a great one!

 

I think what Buffett means when he says cash is like call options is:

 

- a call option with strike price $10 will expire if it doesn't move, and will net $10 if it goes to $20

- if you hold cash and a stock is $20, it will do nothing if it stays at $20, but you will net $10 in VALUE if it drops to $10

 

 

 

i think schroeder explained it pretty well. option price of cash is opportunitiy cost less interest earned on cash holdings. underlying security is EVERY sercurity. strike price is whatever you want it to be. term is forever.

 

cash has been a very expensive option for a couple years now.

 

Schroeder completely mangled Buffett's point. Cash is like a put, not like a call.

[racemize]

I've been reading this thread with great interest. It's a subject I've been having real trouble with. I've been going back and forth with regards to how to position my portfolio.

 

Since CAPE and Tobin Q have been the most reliable indicators on overvaluation and subsequent future returns I was wondering if you had tried to model if it makes sense to hold higher cash level when CAPE/Tobin Q are high assuming you got market returns.

 

It could be modeled something like this (using CAPE):

Less than 20; cash 0

20; cash 5%

21; cash 7%

22; cash 9%

23; cash 11%

24; cash 13%

25; cash 15 %

Etc. Or something of the sort.

 

I hope you post your essay as it seems like an interesting read.

 

Still working on this in general.  Here's a good paper on this topic:

https://www.dropbox.com/s/q11vsobd8ubr2bx/market%20timing%20at%20home%20and%20abroad-031505%20%28Fisher%20Statman%29.pdf

 

Probably a month or so before I'll finish it.

[Liberty]

Thanks for the update, looking forward to your findings!

[frommi]

Still working on this in general.  Here's a good paper on this topic:

https://www.dropbox.com/s/q11vsobd8ubr2bx/market%20timing%20at%20home%20and%20abroad-031505%20%28Fisher%20Statman%29.pdf

 

Probably a month or so before I'll finish it.

 

Interesting paper, but they probably should have compared the earnings yield with the t-bill or 10 year treasury yield instead of a fixed yield. It should be obvious that the switching points depend on current inflation.

[Aurelius]

Interesting paper, but it would have been even more interesting if they focused more on the extreme spectrums, i.e. what happens when we get high CAPE overvaluation.

 

They did state:

The 10-year averaging of earnings makes a great difference in the market timing success of trading rules. The median P/E ratio where earnings are averaged over the preceding 10 years is 16.4 and market timers who were to use it as the critical P/E ratio would have accumulated $72,750 by the end of 2002, more than the $67,672 of buy and hold investors.

Not sure if this means that market timers profited going completely out of equities when CAPE was over 16.4. I'm quite surprised by this but I'm not sure if the sample is significant. It's a little disappointing that they spent so much time talking about regular PEs vs CAPE.

 

I wonder how market timers would do if they only went out of equities when CAPE was highly overvalued? For example: 

CAPE 20; 20% cash

CAPE 25; 40% cash

CAPE 30; 60% cash

Etc.

[frommi]

Thats a classical example of curve fitting. You won`t get the same results doing this going forward. The best example is where they do it with the dividend yield. When you switched out when the dividend yield goes below 1.5% you make more than the buy and hold investor. But hey, there was just one point in history where they would have switched out and that was 2000. So does anybody think that will repeat exactly? Surely not :).

[racemize]

The 16.4 number has extreme hindsight bias issues, which I think they acknowledge (or at least did in another paper).  Using rolling medians, instead of the future median, they underperform.  Additionally, they show the number is very sensitive, so it is almost impossible to replicate.

[racemize]

Still working on this in general.  Here's a good paper on this topic:

https://www.dropbox.com/s/q11vsobd8ubr2bx/market%20timing%20at%20home%20and%20abroad-031505%20%28Fisher%20Statman%29.pdf

 

Probably a month or so before I'll finish it.

 

Interesting paper, but they probably should have compared the earnings yield with the t-bill or 10 year treasury yield instead of a fixed yield. It should be obvious that the switching points depend on current inflation.

 

Here's a paper on that.  It is the only one that I've found with actual outperformance.  However, it does not take into account transaction costs or taxes.  I think this may erase the outperformance, but I'm not too sure.

 

https://www.dropbox.com/s/cm788kdkxdwwp0r/Spread%20Timing.pdf

[frommi]

 

Here's a paper on that.  It is the only one that I've found with actual outperformance.  However, it does not take into account transaction costs or taxes.  I think this may erase the outperformance, but I'm not too sure.

 

https://www.dropbox.com/s/cm788kdkxdwwp0r/Spread%20Timing.pdf

 

Thanks, thats my evening lecture, that was exactly what i was searching for. :)

[frommi]

Here's a paper on that.  It is the only one that I've found with actual outperformance.  However, it does not take into account transaction costs or taxes.  I think this may erase the outperformance, but I'm not too sure.

 

https://www.dropbox.com/s/cm788kdkxdwwp0r/Spread%20Timing.pdf

 

They have done it with 1% spread costs and stil came out ahead. Today you should not have a lot of spread or transaction costs. Taxes are surely a concern, but i don`t intend to hold my value investments forever so its a good hint at when to switch to cash and you have a good timing method to hedge with puts to avoid taxes. It worked in 1974, 1987, 1999 and 2008 and it makes totally sense, when i would time the market i would do it exactly that way and i think WB did it that way either when he has done it.

 

Why invest in a risky asset when the riskfree asset gives a greater return?

 

The short spread (S&P Earnings yield - tbill yield) under -1% looks like a good barrier to act and at 0% we should start to be very careful. This implies that we are currently in safe waters and are at least 1-2 years away from a major downmove. :D

[ukvalueinvestment]

But the tbill yield is artificially low because of financial repression?

[frommi]

But the tbill yield is artificially low because of financial repression?

 

The beauty of numbers and statistics is that you don`t have to argue with them.  ;D

[racemize]

Here's a paper on that.  It is the only one that I've found with actual outperformance.  However, it does not take into account transaction costs or taxes.  I think this may erase the outperformance, but I'm not too sure.

 

https://www.dropbox.com/s/cm788kdkxdwwp0r/Spread%20Timing.pdf

 

They have done it with 1% spread costs and stil came out ahead. Today you should not have a lot of spread or transaction costs. Taxes are surely a concern, but i don`t intend to hold my value investments forever so its a good hint at when to switch to cash and you have a good timing method to hedge with puts to avoid taxes. It worked in 1974, 1987, 1999 and 2008 and it makes totally sense, when i would time the market i would do it exactly that way and i think WB did it that way either when he has done it.

 

Why invest in a risky asset when the riskfree asset gives a greater return?

 

The short spread (S&P Earnings yield - tbill yield) under -1% looks like a good barrier to act and at 0% we should start to be very careful. This implies that we are currently in safe waters and are at least 1-2 years away from a major downmove. :D

 

Taxes can have a large effect, but it depends on your time horizon.  I'm just finishing two essays on taxes, which should help figure out whether their outperformance would still have worked.  One issue, in particular, is how much short term gains they had to deal with. 

 

Overall though, this is the best timing-outperformance I have found so far.  I do wish it was used over the entire 1871 period.  I guess I could recreate that if there were 10 year and tbill rates that far back...

[racemize]

It worked in 1974, 1987, 1999 and 2008 and it makes totally sense, when i would time the market i would do it exactly that way and i think WB did it that way either when he has done it.

 

Where did you get the 2008 one from, another paper?

 

I'm working on recreating the model on a yearly basis from 1871 to see how it does. 

[frommi]

Where did you get the 2008 one from, another paper?

 

I looked it up myself on multpl.com.

 

I'm working on recreating the model on a yearly basis from 1871 to see how it does.

 

Very nice. I can`t wait to see it. :)