Chicago’s real estate market

I was surprised to see historical charts of Chicago’s real estate market in Crain’s.

Especially compared to how Austin, Denver, Seattle, etc. have fared in the last decade.

Chicago-area home prices have returned to their decade-ago peaks, but only in a few neighborhoods and suburbs–and our recovery has been remarkably slow compared to the rest of the nation. Single-family home prices are still at least 15 percent below the old peak in 82 of 163 Chicago-area ZIP codes.

Percent change in home prices:

screen-shot-2016-10-12-at-11-21-21-amscreen-shot-2016-10-12-at-11-21-14-am

US Home sales slump

  • Sales of existing homes fell 7.1 percent last month
  • Home-buying fell sharply in the Northeast and Midwest last month, while slipping less severely in the South and West.
  • Inventory is particularly low for starter homes, keeping first-time buyers out of the market, according to a separate report released Monday by the real estate firm Trulia.

US HOME SALES SLUMP IN FEBRUARY; SUPPLY SHORTAGE HITS MARKET

Bubble stock market

Apple, FANGS And Monetary Fools

Great article this weekend by David Stockman on the extreme valuations of tech stocks like Facebook, Amazon, Netflix and Google.

And that brings us to the lunatic valuation of the FANGs (Facebook, Amazon, Netflix and Google), which was also on display again this week. To wit, 100X+ PE multiples are always and everywhere a deformed artifact of central bank driven Bubble Finance, not the emission of an honest capital market.

The fact is, the greatest technology-based businesses of modern times accomplished its dramatic growth spurt in just over 20 quarters between 2011 and 2015. That was after the i-Phone incepted and the i-Pad worked up a serious head of steam.

Now Apple is pancaking or worse, and it is hard to believe that gimmick products like Apple Watch or Oculus can fill the hole from the fast fading i-Pad and the stalling i-Phone. No harm done, of course, and its entirely possible the APPL will have another modest growth run.

But here’s the thing. Apple essentially proves you can’t capitalize anything at 100X except in extremely rare cases because of the terminal growth rate barrier. That is, after a few years of red hot growth almost every large company’s organic growth rate bends toward the single digit path of GDP.

 

In this respect, Microsoft and Cisco surely prove the rule. After their blistering 1990’s growth, the were valued at 75X and 230X reported earnings, respectively, at the tech bubble peak in 2000. But the terminal year growth rates implied in those towering multiples were not to be——even for two of the most inventive companies of the digital age.

During the last fifteen years, Microsoft’s net income has grown at a mere 5.9% annual rate and Cisco’s at only 7.8%. Technology shifts, competition and global macroeconomic trends together conspired to bend their red hot growth trajectories toward the flat line.

As a consequence, they have spent the entirety of this century shedding market cap and enduring PE multiple compression.
MSFT Market Cap Chart

MSFT Market Cap data by YCharts

This is where it gets interesting: the stocks lose market cap because their P/E ratio must contract. Without massive growth there is no support for 100x P/E ratios. And the 50% growth can only last so long.

As can be seen above, between December 1999 and the end of 2014, Microsoft’s market cap declined from $600 billion to $380 billion, meaning it shed $220 billion of valuation, as its PE contracted from 75X to 18X (its market cap currently is slightly higher at $420 billion).

Likewise, Cisco’s market cap plunged from $530 billion to $140 billion. Thus, despite 15 years of respectable growth it was forced to shed $410 billion of market cap, while its PE multiple ended up resembling Ross Perot’s famous sucking sound to the south, plunging from 230X at the tech bubble peak to 19X at the end of 2014.

Since these two companies are among the best long term performers and profit generators of all time, they speak powerfully to the terminal year growth barrier. In effect, when “investors” capitalize a growth company’s hot years at a triple digit PE they are drastically overvaluing its terminal year potential.  And virtually without exception, they spend the post growth streak years in PE compression and market cap shrinkage.

At best, these high flyers turn out to be 10-20% long term growers, or at worst shooting stars like Lucent or Dell where 80-100% of the peak market cap gets wiped out by the passage of time and events.

Consider the 25-year growth rate of five important and surviving growth icons of modern times. Apples’ quarter-century net income CAGR ending in the year just reported is 20%. That for Cisco is 18%, Microsoft 17.5%, Intel15% and Berkshire Hathaway 15%.

Yet today every one of these companies trade at market caps well below their boom year peaks—–even after giving consideration to the balance sheet deterioration resulting from debt-financed stock buybacks. Specifically, relative to its peak market cap, Cisco is current down by 80%, Intel by 68%, Microsoft by 32%, Apple by 26% and  Berkshire Hathaway by15%.

So there’s the skunk in the woodpile. What is going on under the heading of “growth” company valuation is just pure speculation in the casino. There is absolutely nothing rational or economically efficient about the 100X PEs. The latter get attached to the casino’s trading sardines of the month or year which get temporarily designated as “growth”  stars.

History virtually proves that you can’t go wrong selling the high flyers if you have the patience and capital depth to wait out the serial bubble cycles of the world’s out-of-control central banks. In that respect, this week brought another raft of just that kind of opportunity.

Regarding Amazon:

Yes, Amazon disappointed the absurdly precise and ultra aggressive “growth” targets of Wall Street analysts and traders. But so what?

Its PE ratio was marked down from 900X to 475X, meaning it’s still a screaming sale. Back up the trucks!

After all, when you spend 25 years creating a monstrous machine that is resolutely and  maniacally focussed on spending every dime of revenue on global empire building, it will never generate a measureable profit or return on capital. Well, at least not more than the anemic 3.3% pre-tax return it posted for 2015.

That is the part that many people miss – Amazon doesn’t try to make a profit. And at 475x profits, the growth rate needed to sustain the current market cap with a normal P/E of 10 or 20 is absurd.

Thus, it is only a matter of time before unforeseen external events, such as an ordinary business cycle downturn, to say nothing of the impending global deflation, trigger a drastic re-rating and PE compression. Then the fast money and robo machines will dump the stock hand over fist as usual—–and long before the inattentive home gamers wake up to find themselves the victim of another epic slaughter.

Likewise, Facebook purportedly “surprised” to the upside, and is now trading at close to 90X. But here’s the thing. If you wish to believe that FB really has 1.6 billion carbon unit-based monthly users, you might also want to check into some Florida swampland.

Otherwise recognize that the social media sphere is crawling with silicon-based fake users and mercenary bots, and that their contamination of Facebook’s preposterous user stats is undoubtedly beyond reckoning. Yes, FB occasionally purges the worst incursions, but surely Justin Beiber’s 3.5 million fake fans was only a drop in the bucket.

Its true, people heavily weight the number of users, and apply linear metrics across the user base. For example, $10 per user * 2 billion users, $20 billion! But theres a power law distribution here, not a normal distribution. Most of the revenue comes from a very, very small part of the user base.

Besides that, it doesn’t really matter what the mix of carbon vs. silicon based units is in FBs massive user stats. At the end of the day, Facebook is essentially a digital billboard. It’s a place where mostly millennials idle their time in or out of their parents’ basement. Whether they grow tired of Facebook or not remains to be seen, but one thing is certain.

Namely, advertising is slaved to GDP growth and always takes a hit when the business cycle turns south. Indeed, the pool of advertising dollars is relatively fixed at about $175 billion in the U.S. and $575 billion worldwide and grows in the low single digit zip code after you take account of the severe cyclical fluctuations which always slam the ad spend. For instance, during the Great Recession, the U.S. advertising spend declined by 15% and the worldwide spend dropped by 11%.

And therein lies another skunk in the woodpile. Due to its sharp cyclicality, the ad spend is a distinctly low-growth pool of money. And that pool of money is fiercely competed for by the various media venues. Thus, the U.S. ad spend of $177 billion in 2015 was only $5 billion greater than the $172 billion spend way back in 2008.

So when you average out the cyclical fluctuations, the trend growth in U.S. ad spending has been about 0.5% per annum. Likewise, the global ad spend increased from about $490 billion in 2008 to $575 billion in 2015, reflecting a growth rate of 2.3% annually.

Yes, there has been a rapid migration of dollars from TV, newspapers and other traditional media to the digital space in recent years. But the big shift there is already over.

The big shift is over.

The digital share of the U.S. ad pool, in fact, rose from 13.5% in 2008 to an estimated 32.5% last year. But even industry optimists do not expect the digital share to gain more than a point or so per year going forward. After all, television, newspapers, magazines and radio and highway billboards are not going to disappear entirely.

Moreover, roughly half of the digital share comes in the form of search advertising. FB does not even participate in that sector, where Google has upward of a 55% market share.

Consequently, there are not remotely enough advertising dollars in the world to permit the endless gaggle of social media space entrants to earn revenue and profits commensurate with their towering valuations and the sell side’s hockey stick growth projections.

So once again we have the delusion that the 52% revenue growth that FB booked in its most recent quarter can be continued indefinitely, when it will, in fact, by pulled-down to the GDP growth line in only a matter of time.

Accordingly, now would be an excellent time to sell the stock, and back up the trucks. You can’t  capitalize a one-time share shift in the quasi-static ad spending pie as if it represents a permanent rate of growth. Indeed, most especially not in a world heading into an extended period of deflation and static or even shrinking nominal GDP.

It is hard not to be bearish after this reality check. Market caps are too high. The growth rates are not sustainable. And the previous generation of tech spent the next decade and a half shedding market cap with P/E compression. Companies cannot be priced at 100x earnings and grow 50% a year.

 

Hiring developers, using contractors

Hiring programmers has been a challenge for me.

Building the machine learning models, coding in Python, delivering to clients have been challenging, but all within the realm of my current capabilities.

Evaluating the skill level of the developer is very difficult initially. Some write elegant code, some write spaghetti code.

Part of the difficulty in hiring on UpWork is setting the pricing too low. People were willing to take the offers, but it was inadvertently driving away the most talented programmers. It looks like there is a sea of inexpensive talent. But it can be hard to differentiate.

There is a balance between price and quality of code. The code working is not the value of the code. There is readability, maintainability, scalability. So when two developers write code, both passing test cases, the code is not necessarily equal. And what you pay might have longer term consequences.

So, I came up with a heuristic.

In the valley, good developer median salary is $125,000.

There is a cost to hiring. Lets say its .25x. Head hunters would charge that. There is interviewing time, potentially multiple rounds, with multiple people. Their hourly cost.

If I take the $125,000 plus the search costs of $31,250, we have a budget of $156,250 to spend on contractors in search of the best. If each one is paid $2,500 for a small project, we can effectively have 62.5 long productive interviews.

2014-07-11 Sunday

Workout: 20:00 on the stairmaster. HR intervals.  1 min full effort (setting=12), followed by low effort (setting=2) until HR < 135.  75 flights, 1110 steps.

Dinner: ribeye steak, seared and cooked on low in the oven.  brussel sprouts.

Goal: jits 4x this week.  circadian rhythms.