Saturday, August 25, 2012

08/25/2012 a few rants and a trade


Small Portfolio
XLF & IAU
12.33%
Position
Date
Return
Days
VG
10/27/2011
-35.87%
303
BT
1/4/2012
11.68%
234
XEC
6/5/2012
15.89%
81
DECK
6/15/2012
3.23%
71
CVX
7/5/2012
5.15%
51
RIMM
7/16/2012
-4.28%
40
UEIC
7/30/2012
18.15%
26
QSII
8/6/2012
3.11%
19
CECO
8/9/2012
-6.22%
16
SWM
8/23/2012
1.70%
2
S&P
Annualized
3.96%
Small Portfolio
Annualized
9.97%
Large Portfolio
Annualized
15.27%

 

Scheduled rotation: selling VG; buying FCX
First a short term (current events) rant, then a long term rant…
 

Short term rant:

This week the Republican party stands poised to pay the ultimate price for its stupidity.

The Akin “legitimate rape” comment?

Nope.

The Romney nomination?

Nope.

The stupidity happened way back during the primary debates when all of the candidates agreed to fire Ben Bernanke.

Regardless of whether you actually agree with that sentiment, they should have never bragged about it.

Imagine for a moment that you are Ben Bernanke.  Yes, you are a card carrying Republican, but you also like your position, because you believe that you best know how to cushion the decline of Western Civilization from its collective sovereign debt train wreck.

We are just over two months from the election.  Bernanke is facing low inflation and high unemployment, and so he is ABLE to announce some form of new quantitative easing.

He is ABLE, but is he WILLING?

Yes, I think he’s willing – because that new round of QE will likely tilt the election toward Obama and away from the man who promised to fire him.

In other words, the one man on the planet with the MOST influence on the election is the man Romney promised to fire.

Uh, yeah.  I think we’ll do QE.

The market also seems to be preparing for Bernanke to pull the trigger.  Gold is rising again, and the most attractive industry in terms of money-flow and fundamentals is the mining stocks.

The Mousetrap is scheduled to sell VG and buy FCX.  As always, if they gap away from each other it will cancel the trade.
 

 

Now for the long term rant:

On Monday, SPY hit its all-time highs.

I don’t remember that hitting the news, but that’s the way it is.  SPY is not the S&P 500.  SPY is an S&P 500 ETF, which includes dividends.  If you want to beat the S&P 500, holding SPY is the simplest way to do it.

The difference in the returns is about 0.007401% a day.  Although miniscule on a daily basis, it adds up over time through compounded returns.  To get an idea of this, the S&P 500 index was 16.66 on January 3, 1950.  On Friday it was 1411.13.  A pretty good return!

But if you were to include reinvested dividends, the effective value of the S&P 500 would be 4530.13, or about three times the value we currently see for the index.

Dividends are a good thing, but getting a basket of high dividend stocks will NOT give us an outsized return.

Why?

Simple – returns are SPLIT between dividends and price appreciation.  If you have a basket of stocks that increases in company earnings at 10% a year and hold it for twenty years, you will get the same 10% return whether 7% is price appreciation and 3% dividends, or 7% dividends and 3% price.  In other words, you earn value off of earnings, regardless of how much of that is given through dividends or price appreciation.

So much for dividends.  There is another problem we investors face: taxes.  If you made an average SPYish return from 1950 but paid 30% capital gains (our short term rate), your portfolio would not be worth 4530.13, but instead 988.68.  Long term capital gains, if you rotated at just over once a year, would net you 2153.03.

We won’t even get into trading costs.

The good news for us little folk is that Romney wants to eliminate capital gains taxes for people who aren’t already rich.  The bad news is that Obama wants to raise them higher than they already are.  Didn’t know you were a rich fat cat, did you?  I realize that there are other things people are interested in than being able to retire.  We want to punish rich folk and forget that we ourselves want to BECOME rich enough to retire.  In any case, you might have very good reasons to vote for the current President, but your retirement plans is not one of them.

For its part, the Mousetrap model doesn’t CARE what the capital gains rates should or shouldn’t be.  The model simply tracks what they currently are in order to calculate the correct holding period.  At less than a year, capital gains is calculated at 30%, and at more than a year, 15%.  You can see the jump on the following graph.
 
On the right side of the graph you can see the jump between short and long term capital gains at 366 days, and on the left side of the graph you can see the point of maximum return at 112 days.

To make this graph I used the average return RATE for each day, which is the return, multiplied by 365.25, and then divided by the number of days.  A 5% return in six months would be a 10% return base.  That number is then multiplied by what the government lets you keep – which in short term capital gains would turn that 10% return rate into an annualized return of 7%.

I also include typical trading costs for a 100,000 portfolio.  If I were doing this for clients I’d calculate the rotation rate based on the capital gains taxes in their home country and the impact of trading costs for the size of their actual portfolio.  But this is just a blog.  And this isn’t professional advice.  This is just one trader sharing ideas.

In any case, as you can see, the model is self-adapting.  I feed it data after each trading day and it tracks the returns of stocks the model has selected, whether I’ve actually sold those stocks or not – so that I can see how often I SHOULD trade in the future.  The blue line is the annualized return: counting price, dividends, taxes, and trading costs.  The red line is SPY for the same holding periods.  The green line is the difference between the two.  The correct holding period is the maximum point on that green line.

You’ll note that after an initial surge the Mousetrap stocks tend to stagnate while the S&P catches up.  There’s a reason for that.  The Mousetrap doesn’t pick stocks that will have the greatest earnings growth in the course of the next year (i.e. “growth stocks”).  Instead, the model picks stocks that have been slammed in price a bit more than they should have (i.e. “value stocks”).  Value stocks will outperform in the next 1 to 2 quarters, and also outperform in the next 5 years.  But growth stocks will dominate at other times.

Basically what happens is that people dump stocks that have had earnings hits, but they dump them a bit too aggressively.  At the same time, people jump into stocks that are having great earnings prospects, also too aggressively.  There’s a dynamic see-saw between them, and whatever strategy you use, you have to track how long to actually hold your positions.

It turns out that people are right: growth stocks do indeed grow in earnings more than value stocks for the next few years… but it is rarely to the degree that they are expecting.

The same strategy that outperforms in one holding period will ALSO underperform in another holding period.  Two people can pick the exact same stocks for the exact same reasons, and one will make money while the other loses money BECAUSE THEY HELD THEM FOR DIFFERENT LENGTHS OF TIME.

Investing isn’t just about buying the right stocks; it’s also about holding them for the right period of time, as calculated against capital gains taxes, trading costs, investor behavior, earnings growth rates, and probably a dozen other things I can’t even think of.  If you have a consistent strategy, and track the best holding period, you can outperform.

But if you aren’t able to do that, then park the money in SPY with automated dividend reinvestments setup, and call it a day.

But for goodness sake, do NOT use your instinct to determine when to trade and when not to trade.  The average return rate for all stocks in all time periods that my model has selected is 19.44%.  But the actual return rate I’ve realized is 15.27%.  For the first 9 months of the model I had no way to determine the optimal holding periods, and human instinct cost me almost 25% of the returns I would have experienced had I just flipped a coin once a week to determine if I was going to trade or not.  Your gut is NOT your friend.  The people who get rich in the stock market, get rich off of the money YOU lose.

Don’t guess.  Track your return periods, and trade accordingly.

Tim

Thursday, August 23, 2012

08/23/2012 change of trade - SWM for SAI

Yesterday's substantial gap at the open between CECO and FCX cancelled the trade.

The trade is now set to sell SAI and to buy SWM.

As always, a negative gap would cancel this trade as well.  The model holds each stock from 1 to 2 quarters on average.  No need to chase anything.  Let it come to us.

Wednesday, August 22, 2012

08/22/2012 selling CECO; buying FCX (again)


Small Portfolio
XLF & IAU
11.65%
Position
Date
Return
Days
VG
10/27/2011
-35.87%
300
BT
1/4/2012
13.06%
231
SAI
5/30/2012
7.84%
84
XEC
6/5/2012
20.36%
78
DECK
6/15/2012
6.71%
68
CVX
7/5/2012
5.13%
48
RIMM
7/16/2012
-1.24%
37
UEIC
7/30/2012
22.92%
23
QSII
8/6/2012
7.21%
16
CECO
8/9/2012
1.08%
13
S&P
Annualized
4.11%
Small Portfolio
Annualized
9.48%
Large Portfolio
Annualized
17.93%



Rotation: selling CECO; buying FCX

The industry rotations are about as chaotic as anything I’ve seen since I began this model 15 months ago.  FCX was a buy, then 10 days later a sell.  CECO was a buy, then 10 days later a sell.  This is not normal at all.

What’s happening here is that CECO is in an economically defensive industry and FCX in a bullish liquidity driven industry.  As such they are both extremely sensitive to money flow betting between a deflationary spiral (favoring education) or an inflationary surge (favoring mining).

Personally, I don’t think any rational person can claim to know the answer here – perhaps not even uncle Ben Bernanke himself.

All we can do is follow the money as it chases its own tail.

As always, if FCX and CECO gap away from each other, that would prevent the trade.

Tim


Monday, August 20, 2012

08/20/2012 timing note

The weakest sector is XLK again.  Since I'm not actively using the model to short, it's simply worth watching.  The XLU (theoretical) short was bullish, pointing to a breakout.  They'll probably whipsaw a bit for the next week or so.

Sunday, August 19, 2012

08/19/2012 up, up, and away...


Small Portfolio
XLF & IAU
10.60%
Position
Date
Return
Days
VG
10/27/2011
-34.65%
297
BT
1/4/2012
13.29%
228
SAI
5/30/2012
9.12%
81
XEC
6/5/2012
22.59%
75
DECK
6/15/2012
5.79%
65
CVX
7/5/2012
5.76%
45
RIMM
7/16/2012
3.03%
34
UEIC
7/30/2012
22.38%
20
QSII
8/6/2012
11.55%
13
CECO
8/9/2012
-4.86%
10
S&P
Annualized
4.44%
Small Portfolio
Annualized
8.68%
Large Portfolio
Annualized
18.71%



No rotation tomorrow, but there will be one by next Monday (at the latest).

Behind the scenes, mining and railroad stocks continue to accumulate money-flow, and the next rotation may pick one of those stocks.

The combination of those two industries points to another liquidity rally, but I seriously doubt that it will happen.  Sometimes the big players are wrong.  The only true protection for those times when technicals fail is a solid long term fundamental basis for the stocks in our portfolio.

Of the ten worst industries on my model, four of them are in the utilities sector, and indeed on my sector model XLU is now the weakest sector.  These also point to a break out to new highs on the S&P – PREDICATED on central bank liquidity.  The smart money is betting on QE3.  As I said, sometimes the smart money is wrong.

In the long run, the market goes up.  In the short run, well… we can enjoy the ride while it lasts.

But the day it becomes obvious uncle Ben isn’t pouring any more syrup on our flapjacks, watch out.

Tim


Wednesday, August 15, 2012

08/15/2012 timing note

The sector model is now confirming XLU (utilities) to be the weak sector, replacing XLK (technology) as the sector to short.

So, the model is now long XLF (financials) or short XLU (if you're shorting).

That doesn't mean I'll be shorting.  It only means that this is the weakest sector -- which is bullish for the broad market.

Monday, August 13, 2012

08/13/2012 model returns so far


These are the returns for the model calculated through August 9, 2012.

SPY (in Blue) is the S&P ETF.  It includes dividends and is a robust comparison for the model.

Beta (in Orange) is a representation of the actual returns for the model.  Since I gradually loaded the model over a six month period, it starts mostly in cash and is only fully loaded at the end of October 2011.

Mousetrap (in Red) is what the Beta returns would look like if cash is excluded.

Finally, Margin (in Green) is the way a Margin account would use the model.  It would invest 10% of the maximum previous account value into each stock, rather than the current account value.  This averages about 6% of margin over time, with a theoretical maximum of 27%.


During the live run of the model, SPY (S&P plus dividends) has experienced a 6.71% annualized return.  The Beta run has experienced an 18.20% annualized return.  Had the Beta been fully loaded with no cash reserves the return rate would be 22.78% annualized, and a margined strategy would be 27.11% annualized.


Finally, there is the question of maximum drawdown and a risk / reward ratio:


 
Drawdown
Return Rate
Strategy
Risk / Reward Ratio
22.50%
6.71%
SPY
335.11%
19.19%
18.20%
Beta
105.46%
24.51%
22.78%
Mousetrap
107.56%
25.32%
27.11%
Margin
93.39%

In terms of risk, simply holding SPY was riskier than any of the Mousetrap strategies, whether the actual Beta run, or fully loaded.  The most intriguing result was the fact that the Margined strategy would actually have greater returns than risk.