Sunday, April 28, 2013

04/28/2013 Margin and Risk


Sector Model
XLI
1.98%
Large Portfolio
Date
Return
Days
BBRY
7/16/2012
107.17%
285
SEAC
9/25/2012
31.33%
214
CAJ
9/25/2012
5.44%
214
BOKF
2/4/2013
10.20%
82
SWM
2/12/2013
9.76%
74
TTM
4/1/2013
10.61%
26
MWW
4/11/2013
-2.01%
16
ABX
4/11/2013
-24.16%
16
TPX
4/22/2013
6.04%
5
NYCB
4/24/2013
-1.40%
3
S&P
Annualized
8.88%
Sector Model
Annualized
24.66%
Large Portfolio
Annualized
31.54%

 

No rotation for Monday.

Note also that the sector model is listing XLI, rather than XLB as posted before the close on Friday.  In the last few minutes of trading the technical configuration of XLI just barely passed XLB by 2 tenths of a percent.

Don’t sweat the sector model whipsaws.  The average holding period is a month, but that’s built off of times like this week averaged against months with no change at all – such as last year’s extremely long hold of XLF.

In any case, the opportunity loss of missing a day or so is offset by the benefit of saving on trading costs in a whipsaw week.

Now, on to the next missive (I’m still planning a post on time frames, but today I’ll write more detail on risk)…

Last week we looked at risk in a cash account, and this week we’ll look at risk in a margin account.

Most folks are charged around 8% borrowing costs for use of margin.  High net worth folks and hedge funds can do better, but I’ll stick with 8% for us small fry.

Now, to put this into perspective, since 1950 the S&P has averaged 7.46% per year.

The amount you are charged for margin is greater than the amount the S&P usually gains.

You’re already starting from behind.

The limit on your use of margin, then, should be based on your average return WITHOUT margin – as calculated through an ENTIRE business cycle.  If you average 10%, and the cost of margin is 8%, then you shouldn’t be more than 100 + 10 – 8 = 102% invested.

In THEORY, then, the Mousetrap COULD be 100 + 31.5 – 8 = 123.5% invested.

But it isn’t.  That 31.5% annualized performance has not yet seen a bear market, and it is not safe to assume that outperformance in a bull translates to outperformance in a bear.  Back-tests of the Sector Model indicate that the Mousetrap will also outperform, but a back-test isn’t as good as the real thing.

So, if you can’t go more than 100% net long, what about hedging?

Ah – hedging is another subject entirely.

You hedge when you are long one thing and short something else.  A typical hedge fund will be 100% long and 100% short.  Right now I’m reading a series of white papers put together by www.gargoylegroup.com about selling S&P calls, rather than using a simple short.  The idea is to reduce that 8% borrowing cost, and they have an extremely impressive track record with the strategy.

It’s probably the safest, most inspired way to hedge, by a value oriented investment model.  They INVEST; they don’t GAMBLE like Paulson.

If you’re rich, check them out!

If you’re not, keep reading…

First, let’s stick with that 8% borrowing cost to keep the math easy for us poor folks.

We saw above that your margin exposure should never exceed your average annual return, less 8%.

But HEDGING is a different animal altogether, because you have something ELSE working against you: the long term bullish bias of the S&P itself.  If you were 100% long the Mousetrap AND 100% short the S&P, your expected return would be 31.5% (Mousetrap) – 8% (borrowing cost) – 7.5% (S&P bias) = 16% per year.  Add in our lovely 43.8% tax rate and you NET 9%.

Think about this for a minute.  You’re pumped up on full use of margin, having to monitor the account on a daily basis to make sure you don’t EXCEED that margin, and you NET 9%?

Why not pop everything into SPY and have a life instead?

You’d do still better if you were 123.5% long the Mousetrap and 76.5% short the S&P, which would give you an expected return of 25%.

Here’s the breakdown:

Long Mousetrap 31.5% * 123.5% = 38.90%.

Short S&P -7.5% * 76.5% = -5.74%

Borrowing cost -8%

38.9% - 5.74% - 8% = 25.17%

That’s fantastic!  Let’s go, right?

Er, no.

We forgot those taxes.  After taxes you’d NET 14%.

Almost double the S&P, and not too far behind an after tax Mousetrap NET of 18%, with much less volatility.

But ponder this:  Why is it that only rich people invest in HEDGE funds?

The purpose of a hedge fund isn’t to GET rich, but to AVOID becoming poor.  Those of us trying to GET rich should make full use of an IRA account – which does NOT allow hedging.  The NET for the Mousetrap in an IRA is 31.5%, rather than 18% for a taxable account.

I DO plan to create a second model for taxable accounts, but I haven’t yet found the sell point.  Once I reach it, I’ll let you know.  Right now it appears to be somewhere between 2 and 3 years per stock.

In any case, the point I’m trying to make is that margin is NOT a tool for maximizing returns, but instead for using a hedge to minimize risk.  If you aren’t already rich, you’re better off staying in a cash account and calling it a day.

So, do you have enough to live off of for the rest of your life if you quit working?  Great!  Hedge!

But if you don’t, your best bet for improving your reward / risk ratio is by good old fashioned value investing and fundamental stock picking.  Ben Graham called this the “margin of safety.”

And THAT kind of margin doesn’t charge you 8%, but instead PAYS you in dividends!

Tim

Friday, April 26, 2013

Tuesday, April 23, 2013

04/23/2013 rotation: selling OKE; buying NYCB


Sector Model
XLI
0.72%
Large Portfolio
Date
Return
Days
BBRY
7/16/2012
97.66%
281
SEAC
9/25/2012
33.17%
210
CAJ
9/25/2012
12.97%
210
BOKF
2/4/2013
10.83%
78
SWM
2/12/2013
9.84%
70
OKE
2/25/2013
9.32%
57
TTM
4/1/2013
9.18%
22
MWW
4/11/2013
-5.36%
12
ABX
4/11/2013
-28.09%
12
TPX
4/22/2013
5.13%
1
S&P
Annualized
8.80%
Sector Model
Annualized
24.00%
Large Portfolio
Annualized
31.21%

 

Rotation: selling OKE; buying NYCB.

As always, a negative gap will prevent the trade.

So, per my model, these are the best industries:

FURNITUR
TOBACCO
ENTTECH
BANKMID
ELECFGN
GOLDSILV
ADVERT
WIRELESS
AUTO
THRIFT

 

Furniture, Tobacco, Wireless, Auto – THINGS people need.

Elecfgn, Bankmid, GoldSilv, Thrift – point to more liquidity from the U.S. Fed.

Advertisment, Entertainment Technology – point to more time on people’s hands than they know what to do with.

My guess is that the Japanese liquidity has created some more buying opportunities for the next U.S. Treasury salvo in this currency war.

If my model is wrong, it’s going to be VERY wrong here…

Tim

 

 

 

Monday, April 22, 2013

04/22/2013 before the close

Sector model continues to whipsaw.  Selling XLB and buying XLI.

A cash account tracking this can only trade once every three days, so it would have to continue in XLB until at least Wednesday.


Sunday, April 21, 2013

04/21/2013 Managing Risk


Sector Model
XLB
0.88%
Large Portfolio
Date
Return
Days
BBRY
7/16/2012
90.90%
279
SEAC
9/25/2012
33.17%
208
CAJ
9/25/2012
10.32%
208
BOKF
2/4/2013
10.54%
76
SWM
2/12/2013
8.68%
68
GMCR
2/19/2013
23.92%
61
OKE
2/25/2013
8.31%
55
TTM
4/1/2013
6.55%
20
MWW
4/11/2013
-4.46%
10
ABX
4/11/2013
-25.72%
10
S&P
Annualized
7.97%
Sector Model
Annualized
22.99%
Large Portfolio
Annualized
30.14%


Rotation: selling GMCR; buying TPX.

What an evil week. 

Steven Pinker wrote “The Better Angels of Our Nature” to show how violence has declined over the centuries, and even during our lifetimes.  He makes a compelling statistical case, but it’s no help in an age of instant communication.   Statistics don’t make you feel much safer, when you see a week of bombs, poison notes, and a plant explosion.

The most dangerous animal you’ll ever meet is a human being.

But… back to statistics (and the subject of this blog).

After Friday the model estimates a 100% bullish bias for the broad market over the next quarter.  For what it’s worth, every timing expert I respect is bearish.  So if my model is wrong, it’s as wrong as possible. 

And that brings me to the question of risk.  Most folks try to manage risk through Beta, Hedging, or (infinitely worse – letting their fear and greed “time” their portfolios into oblivion).

Beta creates risk by scaring investors.  By itself it does nothing.  Higher Beta is usually associated with smaller caps, which outperform large caps.  However, within the same capitalization levels lower Beta is associated with better returns.  If you’re going to try to do something with Beta, you have to measure it as a ratio of capitalization.  My own model sees little to no value in Beta, but that could just be a quirk of the model.  In the investing world, the Capital Asset Pricing Model tries to get the most return for the least risk, and hedge funds are often evaluated by a model developed by Fama.  Both of these want the greatest return for the least risk, and while there are reasons for them to do so, those reasons do not translate well for a personal investor.  The actual purpose of these models is to maximize the amount of margin that can be used and to lose the least amount of clients.  A hedge fund is ultimately in the business of managing assets, and scaring clients on the way to the best possible returns will not help their clients or themselves.

So, they hedge…

Hedging manages risk by lulling investors to sleep.  By itself it actually reduces returns because it works against a long term 8% annualized positive bias in the market, and charges traders shorting costs on top of it.  Even worse, if you’re right you can only gain up to 100%, but if you are wrong you could lose MORE than 100%.  A hedge fund can do it (most can’t, but some can), but practically no individual is equipped to do this for themselves, nor do they have the time to sit staring at their computer screen all day.

Ignoring Beta, Hedging, and Timing, what does that leave for a person trying to invest for himself in a personal IRA?

Two things, and only two things:

1) Time

2) Position size

I have a friend who was in the emergency room for one day, then in the hospital for another, and when he got out two days later he had lost 80% of his retirement – because he was using heavy margin while day trading.

Basically the larger your position size, the shorter amount of time you have.  And normally you’ll be wrong on occasion and will be forced to sell at a loss.

And you only have to be wrong once.

My own model invests no more than 10% in any single stock.  Any time I’ve violated that 10% limit in the past I’ve lost money because I was forced to sell at the wrong time.

And I mean that I sold at EXACTLY the wrong time.  I am uncanny in my ability to create the worst possible time decision when working from instinct.  And any position size greater than 10% will pull that instinct out of hiding and ruin my day.

Think of any position larger than 10% as a clear and present danger.  You can be right ninety-nine times in a row, and wrong the hundredth, and you could lose everything.  You’re like a gambler at a craps table, letting everything ride over and over and over again till you have an insane amount of chips on the table and you cannot make yourself stop.

But once you lose, the table makes you stop.

We do the same with the market, much more likely to sell for a 15% loss than a 15% gain.  Or worse, to hold for a 75% loss instead of a 75% gain.  If the position is only 10% of your portfolio, you can wait until you have the right sell point.  If it’s any larger, you’re trapped – glued to the screen – losing sleep over any potential tick of the market that goes the wrong way.  And then you’ll close the position and watch the market INSTANTLY reverse in your favor.  But you aren’t there any more to reap the benefit.

THAT’s risk.

In fact, it’s even worse than risk – it’s practically a guarantee that you’ll lose.

So, on the long side, time is your friend.  But if you are heavily margined or are too heavily invested in any given stock, you won’t be able to wait for the right time to take a profit.

10%.

No more, in any given stock.

Period.

Tim