LEADING ECON. INDIC.

Note: writing from Asia (currently in the Philippines), I am 13 hours ahead of N.A. so the date above may not match your date.

Some of the most important data last week:

U.S. Industrial Production Rebounds Following Hurricanes

The Federal Reserve indicated that industrial production recovered 0.3% (1.6% y/y) during September following a 0.7% August decline, last month estimated as -0.9%. The Fed indicated that storm damage reduced September’s increase in production by one-quarter of a percentage point. (…) Factory output edged 0.1% higher (1.0% y/y) following a 0.2% decline, initially estimated as -0.3%.

Production of consumer products increased 0.5% (0.2% y/y) and reversed August’s decline. Motor vehicle & parts production improved 0.1% (-3.2% y/y), after a 3.6% jump. (…)

Capacity utilization rose to 76.0% from 75.8%, but remained down versus the 76.6% June high. Factory sector utilization held steady at 75.1%, but also remained lower than the 76.0% April high. Total factory sector capacity rose 0.7% y/y.

Apart from Construction Supplies, not much momentum:

However:

U.S. Housing Starts Weaken to 12-Month Low

U.S. Existing Home Sales Improve While Prices Decline

Weak pending should keep existing sales weak next month as this Bloomberg Briefs chart suggests:

 

U.S. Leading Economic Indicators Slip

The Conference Board’s Composite Index of Leading Economic Indicators eased 0.2% (+4.0% y/y) during September following an unrevised 0.4% August increase. Three-month growth deteriorated to 2.2% (AR) from 5.5%.

Performance amongst several of the components of the leading indicator series deteriorated. The average workweek, building permits, initial unemployment insurance claims, orders for nondefense capital goods excluding aircraft and building permits each registered negative effects on the index. Consumer expectations for business/economic conditions and new orders for consumer goods had lessened positive effects.

A steeper interest rate yield curve had a steady positive impact on the index, though it was reduced substantially from earlier in the recovery.

To the upside, the positive effects from the ISM orders series and the stock price series improved.

The Index of Coincident Economic Indicators ticked 0.1% higher (1.7% y/y) following an unrevised zero change in August. Gains in industrial production, personal income less transfer payments and business sales were countered by a decline in payroll employment. Three-month growth in the index held steady m/m at 0.7% (AR), down from 2.5% in May.

The Index of Lagging Economic Indicators slipped 0.1% (+2.4% y/y) following an upwardly revised 0.4% increase. A longer average duration of unemployment had a sharp negative effect on the index. Commercial & industrial loans outstanding, the consumer credit-to-personal income ratio and the change in the services CPI had positive effects. Three-month growth in the lagging index fell sharply to 1.6%

The ratio of coincident-to-lagging indicators is often considered to be a leading indicator of economic activity. If economic growth is moderating, both sets of indicators will be rising, but the gains in the coincident index will be slower than the lagging; thus, the ratio will fall. This ratio improved slightly last month from its record low.

Hard eco data were negative while financials and survey-related had positive contributions. The diffusion index declined to a 9-month low 60.0. Not really what we want but not alarming.

Doug Short has the best charts on LEI:

Looked at from an annualized rate of change, the LEI is +4.0% on a 12m basis, +3.5% on a 6m basis and +2.2% on a 3m basis. Slow and slowing. And the Fed is tightening.

EARNINGS WATCH

Still early but the market seems to like what was shown so far.

From Factset:

Overall, 17% of the companies in the S&P 500 have reported earnings to date for the third quarter. Of these companies, 76% have reported actual EPS above the mean EPS estimate, 10% have reported actual EPS equal to the mean EPS estimate, and 14% have reported actual EPS below the mean EPS estimate. The percentage of companies reporting EPS above the mean EPS estimate is above the 1-year (71%) average and above the 5-year (69%) average.

In aggregate, companies are reporting earnings that are 0.6% above expectations. This surprise percentage is below the 1-year (+5.1%) average and below the 5-year (+4.2%) average.

The blended earnings growth rate for the S&P 500 for the third quarter is 1.7% today, which is lower than the earnings growth rate of 2.0% last week. The downside earnings surprise reported by General Electric ($0.29 vs. $0.49) was mainly responsible for the decrease in the overall earnings growth rate for the index during the past week.

If the Energy sector were excluded, the blended earnings growth rate for the remaining ten sectors would fall to -0.2% from 1.7%.

The Financials sector is reporting the highest (year-over-year) earnings decline of all eleven sectors at -10.1%. At the industry level, two of the five industries in this sector are reporting or are predicted to report a decline in earnings, led by the Insurance industry (-66%). This industry is also the largest contributor to the earnings decline for the sector. If the Insurance industry were excluded, the blended earnings growth rate for the Financials sector would increase to 4.7% from -10.1%.

This sector is also the top detractor to earnings growth for the entire S&P 500. If the Financials sector were excluded, the blended earnings growth rate for the remaining ten sectors would rise to 4.4% from 1.7%.

In terms of revenues, 72% of companies have reported actual sales above estimated sales and 28% have reported actual sales below estimated sales. The percentage of companies reporting sales above estimates is well above the 1-year average (61%) and well above the 5-year average (55%).

In aggregate, companies are reporting sales that are 1.0% above expectations. This surprise percentage is above the 1-year (+0.6%) average and above the 5-year (+0.5%) average.

The blended sales growth rate for the third quarter is 5.1% today, which is higher than the sales growth rate of 5.0% last week.

If the Energy sector were excluded, the blended revenue growth rate for the index would fall to 4.1% from 5.1%.

At this point in time, 18 companies in the index have issued EPS guidance for Q4 2017. Of these 18 companies, 12 have issued negative EPS guidance and 6 have issued positive EPS guidance. The percentage of companies issuing negative EPS guidance is 67% (12 out of 18), which is below the 5-year average of 75%.

Estimate revisions on S&P 500 companies improved last week.

Equity markets last week saluted the better odds of a tax deal, totally neglecting the potential $1.5T increase in the U.S. budget deficit that could come with it. The reality is that there has been no deleveraging of the U.S. economy during this long expansion period and the U.S. is about to take its debt/GDP ratio to a whole new level.

Just when the Fed is fully intent to “normalize” interest rates, another way of saying that the cost of this higher indebtedness will go up, hitting all sectors of the economy and taking much needed resources away from the real economy.

Even if Mnuching et al say that tax reform will more than pay for itself through higher GDP growth, there is no historical precedent to that claim which was also made by the Reagan team in the 1980s. In fact, the opposite is true as Moody’s demonstrates:

When the average ratio of nonfinancial-sector debt to GDP rose from the 135% of 1960-1984 to the 175% of 1985-2001, the average annualized rate of real GDP growth slowed from 3.6% to 3.3%, respectively. In conjunction with the 229% average ratio of debt to GDP since 2001, real GDP growth has slowed to 1.9% annualized, on average.

While leverage risk is reaching a new bubble level, masked by Fed-induced low interest rates, market risk is declining as measured by the VIX.

But Doug Kass rightly asserts that market risk is actually increasing dangerously:

Though large daily drops in the markets are rare, the factors that could contribute to a quick drop have increased. Investors have been concerned about the VIX for years, but the positioning has now moved to an extreme. Such positioning could accelerate a market drop as the chances of a flash crash have escalated.

But, Dr. Minsky has warned about the risks of becoming numb to the risks associated with a period of stability amid rising asset prices; it is not only inevitably followed by instability, it inevitably creates it.

In a world in which the chances of an external market shock are rising and at a time when volatility is cratering and stock prices never decline, the risks of a flash crash caused by the one-sided market positioning in VIX futures is increasing and are at a higher probability of occurring than at any time in history.

From Morgan Stanley’s Chris Metli:

It’s easy to become numb to the low volatility environment and the risks it presents. While trying to pick a trough in vol has been a fool’s errand, focusing on the risks resulting from vol being so low is not. Low volatility has produced a regime where the risks are asymmetric and negatively convex, so being prepared for an unwind is critical. This is not a call that vol is almost to spike, but you need a plan if it does.

OTHER THAN THAT

I was in Japan in 2014 when Obama was President. Clearly, Trump is a much bigger deal here. From the North Korean threat to America’s isolationist policies (economic and foreign affairs), let alone his character, Japanese media are full of Trump stuff.

   

These 2 op-eds from NY Times columnists Thomas Friedman and David Brooks were prominently featured in the Japan Times.

• The Trump Doctrine The president’s approach is: Rip things apart without thought and leave the problem to someone else.

(…) On nearly every major issue, President Trump’s position is: “Obama built it. I broke it. You fix it.”

And that cuts right to the core of what is the most frightening thing about the Trump presidency. It’s not the president’s juvenile tweeting or all the aides who’ve been pushed out of his clown car at high speed or his industrial-strength lying.

It’s Trump’s willingness to unravel so many longstanding policies and institutions at once — from Nafta to Obamacare to the global climate accord to the domestic clean power initiative to the Pacific trade deal to the Iran nuclear deal — without any real preparation either on the day before or for the morning after. (…)

In short, we’re simply supposed to take the president’s word that this or that deal “is the worst deal ever” — backed up by no serious argument or plan about how he will produce a better one. (…)

• The Essential John McCain

This piece by David Brooks is very à propos:

(…) McCain still believes that paideia is essential for democracy. Paideia is the process by which we educate one another for citizenship. Paideia is based on the idea that a healthy democracy requires a certain sort of honorable citizen — that if we’re not willing to tell one another the truth, devote our lives to common purposes or defer to a shared moral order, then we’ll succumb to the shallowness of a purely commercial civilization, we’ll be torn asunder by the centrifugal forces of extreme individualism, we’ll rip one another to shreds in the naked struggle for power. (…)

Public figures are the primary teachers in this mutual education. Our leaders have outsize influence in either weaving the moral order by their good example or ripping it to shreds by their bad example. (…)

We’ve reached a point in which the tasks of paideia have been abandoned and neglected. “One could say,” Gomá writes in his book “Public Exemplarity,” “that we are looking for the ideal of a virtuous republic composed of citizens relieved of the burden of citizenship.”

It’s not working out. [Spanish philosopher Javier Gomá Lanzón] continues, “In a time of freedom such as ours marked by subjectivism and vulgarity, a tolerance not tempered by virtue will lead inevitably toward barbarism.”

Barbarism and vulgarity we have in profusion. Through his daily utterances, Trump is influencing the nation in powerful ways, but none would call it paideia. Few would say he is spreading a contagion that we’d like our children to catch.

The moral fabric of society is invisible but essential. Some use their public position to dissolve it so they can have an open space for their selfishness. McCain is one of the strongest reweavers we have, and one of our best and most stubborn teachers.

While humans are not getting much smarter, machines are. If you missed this, please read through it. Amazing. Scary.

Newest version of self-taught, ‘superhuman’ AI now even smarter: makers

(…) the new version of AlphaGo is entirely self-taught — a major step toward the rise of machines that achieve superhuman abilities “with no human input,” they reported in the science journal Nature.

Dubbed AlphaGo Zero, the artificial intelligence system learned by itself, within days, to master the ancient Chinese board game known as go — said to be the most complex two-person game ever invented.

It came up with its own, novel moves to eclipse all the go acumen that humans have acquired over thousands of years.

After just three days of self-training, it was put to the ultimate test against its forerunner AlphaGo, which had dethroned the top human champs.

AlphaGo Zero won by 100 games to zero.

“AlphaGo Zero not only rediscovered the common patterns and openings that humans tend to play … it ultimately discarded them in preference for its own variants, which humans don’t even know about or play at the moment,” said AlphaGo lead researcher David Silver.

The 3,000-year-old game, played with black and white stones on a board, has more possible configurations of moves than there are atoms in the universe.

AlphaGo made world headlines with its shock 4-1 victory in March 2016 over 18-time go champion Lee Sedol, one of the game’s all-time masters. (…)

Unlike its predecessors, which trained on data from thousands of human games before practicing by playing against itself, AlphaGo Zero did not learn from humans or by playing against them, according to researchers at DeepMind, the British artificial intelligence company developing the system.

“All previous versions of AlphaGo … were told: ‘Well, in this position the human expert played this particular move, and in this other position the human expert played here,’ ” Silver said in a video explaining the advance.

AlphaGo Zero skipped this step.

Instead, it was programmed to respond to reward — a positive point for a win versus a negative point for a loss.

Starting with just the rules of go and no instructions, the system learned the game, devised strategies and improved as it competed against itself, starting with “completely random play.” This is a trial-and-error process known as “reinforcement learning.”

Unlike its predecessors, AlphaGo Zero “is no longer constrained by the limits of human knowledge,” Silver and DeepMind CEO Demis Hassabis wrote in a blog.

AlphaGo Zero used a single machine — a human brain-mimicking “neural network” — compared to the multiple-machine “brain” that beat Lee.

It had four data processing units, compared to AlphaGo’s 48, and played 4.9 million training games over three days, compared to 30 million over several months.

“People tend to assume that machine learning is all about big data and massive amounts of computation, but actually what we saw with AlphaGo Zero is that algorithms matter much more,” said Silver.

The findings suggested that AI based on reinforcement learning performed better than those that rely on human expertise, Satinder Singh of the University of Michigan wrote in a commentary also carried by Nature.

“However, this is not the beginning of any end because AlphaGo Zero, like all other successful AI so far, is extremely limited in what it knows and in what it can do compared with humans and even other animals,” he said.

AlphaGo Zero’s ability to learn on its own “might appear creepily autonomous,” added Anders Sandberg of the Future of Humanity Institute at Oxford University.

But there is an important difference, he said, “between the general-purpose smarts humans have and the specialized smarts” of computer software.

“What DeepMind has demonstrated over the past years is that one can make software that can be turned into experts in different domains … but it does not become generally intelligent.”

It was also worth noting that AlphaGo was not programming itself, said Sandberg.

“The clever insights making Zero better was due to humans, not any piece of software suggesting that this approach would be good. I would start to get worried when that happens.”

AMAZON BAR

Amazon opened a “pop-up bar” last week in Tokyo. It will be opened for 10 days only. Suzanne and I tried to get in on the second day but we arrived at 5:30, half an hour after opening, and we were told we would have to wait 3 hours to get in. I found that totally unacceptable from a company that can ship almost anything within 2 hours of ordering on line. So we left and had dinner at Shiseido, yes, the cosmetics company. Only in Tokyo! What a great city!

Anyway, some pics of this unique Amazon Bar: