Monday, February 13, 2017

Back to basics

I joked a couple of weeks ago that the new Monday morning trading strategy seems to be -

1. Did we start World War III over the weekend?

2. If no, then buy every stock......

The bar is set incredibly low right now for the market's expectations for leadership in Washington so every day that the world doesn't explode seems to be a positive.  I'll spend a little time today reviewing just how silly the markets have become.  As a reminder, the argument that "well, it was worse in 2000" is crazy because those markets were so overstretched that comparison to them is laughable.  However, while this recent run to new highs has been accompanied by none of the normal hysteria, it is becoming nearly as extreme.

1. Apple:  This will be a consistent theme - notice how the stock has soared almost 50% (the green line), while earnings (you know the reason why you allegedly buy a stock) have tumbled by more than 15%.
Photo published for Apple Stock Soars Above Record Closing High  

2. This is sort of wonky but as the stock indexes are rising, fewer and fewer individual stocks are actually trading above their 50 day moving averages.  This means the advances are concentrated in a few names and the "new highs" are fragile. 

Another warning sign is the lack of trading.  Many traders are complaining that they could be sleeping in until 3:30pm because that's when the trading day actually seems to start. The number of shares traded on the NASDAQ is at its second lowest level in the past two years and is a sign of complacency that often proceeds a market top.

3. IRS receipts are signaling something.  While companies increasingly report "proforma" adjusted earnings, one area where they do not make up the data is on their tax returns.  Over the past 12 months corporate tax receipts are down almost 12%.  When tax receipts fall broadly it is an indication of weakness in the economy and when they fall on a 12 month rolling basis (as they did in January) it has signaled a recession every time going back to 1970.  But this time is definitely different, right?

While we are on the subject of recessions consider the following indicators -
a) Gross private domestic investment indicates we are in a recession right now.
b) Lending standards for small and medium sized businesses have tightened for 6 straight quarters.  that typically only occurs in recessions.
c) there has also been a tightening of lending standards for consumers
d) consumer bankruptcies rose year-over-year for consecutive months for the first time since 2010.
e) Oh, and gasoline demand is implying a 6% decline in consumer spending.  That's recessionary levels.

But it's definitely different this time.

4. Last February when the stock market was saved from collapse by a miraculous Central Bank meeting there were about 27% of investing gurus who were bullish.  Now, that stocks have jumped 28% in the past year, 63% of gurus are bullish.  Hmm, I'll let you draw your own conclusions.

One final observation on the speed of the markets today.  In 2000, Goldman Sachs employed roughly 600 equity traders who were buying and selling stocks for the firm's clients.  This human intervention slowed the decline of stocks while the dotcom bubble burst.  Today Goldman employs 2, yes 2, traders and a team of 200 computer engineers.  With 50% of trades today coming from computers, the speed of the next move, up or down, will be unlike anything we've experienced in the past.


* One final note - this isn't a political commentary.  I predicted back in the fall that no matter who won the Presidential election the US had a 40% chance of slipping into a recession in 2017.  I think those numbers still hold true today but I might increase the odds of recession slightly.

Friday, February 03, 2017

Jobs day

I'll offer up a bit of analysis that I've yet to hear anywhere re: the 227k jobs created in January.  A large portion of the "jobs" created in the report are "modeled jobs" based on samples.  Basically, they are a creation of someone working on an excel spreadsheet. 

Included in those models are seasonal adjustments which do things like add more jobs in the winter months even though the jobs don't exist because you want to smooth out the overall growth.

So while the real numbers might look like this:
Jan:     50
March: 75
June:     100
Sept:      80

When you multiply the data by your seasonal adjustments

Jan =     50 x 2      =    100
March =75 x 1.35 =    101.5
June =   100 x 1.03 =   103
Sept =    80 x 1.3 =      104

Ah, isn't that beautiful? Again, these aren't real numbers but they show the effect of seasonal smoothing. So, the seasonal factors have been built over many years where we have observed the impact on job growth change with the seasons.  However, we just had one of the warmest January's in 50 years and very little disruption to travel due to weather.  This probably made our January jobs data look a lot more like a March or April jobs number.  However, when you apply the January seasonal factor you get - boom!!! 227,000 jobs.  If and when we have numerous storms in a month, the jobs data is impacted and the first line out of everyone's mouth is "Well, the weather impacted the report...".  The truth is the weather can have a positive impact as well, but no one ever seems to complain about that.

I expect you'll see that number revised in the coming months but no one will notice if and when the revised numbers are released.