Relevant and even prescient commentary on news, politics and the economy.

“Hurricane adjusting” initial claims has proven its value

“Hurricane adjusting” initial claims has proven its value

For the last month, I deduced a “hurricane adjusted” number for initial claims, which showed that the previous underlying positive trend was intact, with the four week average remaining in the 230,000’s.

That approach was borne out by this week’s report, which, at 222,000, was the lowest since 1973.

Although I haven’t gone through the entire formal exercise, here’s how the numbers from the three affected jurisdictions compared in last week’s report compared with one year previously:

FL 13,861 (+6508 from 2016)
TX 16,656 (-225 from 2016)
PR 250 (-2409 from 2016) (DoL estimate)

Net change: +3904 from 2016

Since the seasonal adjustment last week was only ~6%, (244,000 vs. 229,289 NSA), this means last week’s “hurricane adjusted” number was on the order of 239,000 or 240,000.

Natural disasters will continue to strike. I am confident that the method I used in 2012 after Sandy, and again this past month, is a good way to distill the underlying trend from the disaster disturbance.

Comments (0) | |

Underlying industrial production trend ex-hurricanes remains positive

Underlying industrial production trend ex-hurricanes remains positive

A few weeks ago, I suggested a hurricane workaround for industrial production. That approach was to average the four regional Fed indexes excluding Dallas, and add the Chicago PMI, and finally discount for the unusual strength this year in these regional indexes vs. production.

Here was my conclusion:

The average of the 5 is 22.9.
Dividing that by 5 gives us +.5.
Subtracting .3 gives us +.2.

We can be reasonably confident that underlying trend in industrial production in September, despite the hurricanes, has been positive.

That approach was borne out yesterday when overall September Industrial Production was reported at +0.3%, with manufacturing production up +0.1% as shown in the graphs below.:

First, here’s the longer term view,. Note that the decline in 2015 was due to weakness confined to the Oil Patch:
Here is the close-up of this year:
That’s the good news.  The bad news, of course, is that even with this improvement, the big (revised) August decline of -0.7% in production, and -0.2% in manufacturing has not been overcome, and production is still below where it was this spring.

Comments (0) | |

A housing teaser

A housing teaser

Here is something I have been working on for the last month.  As it happens, last week Kevin Drum posted some aspects of the same data.
House prices have exceeded by a substantial margin median household income:
But the monthly mortgage payments have not:
This is because, while the prices of houses have increased, mortgage interest rates have decreased over the same period.
So, saving for the down payment is considerably more difficult (unless, e.g., parents are helping out), but once the house is bought, the monthly carrying cost for living in the house really hasn’t gone up at all.
 What’s missing in this discussion is comparing both household income and mortgage payments to the alternative (leaving aside living in mom and dad’s house) of paying rent.
I still have some number crunching to do, but once the three way comparison is finished, it will be a really illuminating look into how much the alternatives for shelter really cost.  Stay tuned.

Comments (6) | |

A thought for Sunday: the Rule of Gerontocracy

A thought for Sunday: the Rule of Gerontocracy

The US looks like government of, by, and for senior citizens.
President Donald Trump just had his 72nd birthday. He assumed office at age 71, the oldest person ever to do so.
In Congress, Senate Majority Leader Mitch McConnell is 75 years old.  His Democratic counterpart, Charles Schumer, is a relatively spry 66. The median age of US Senators is 63. A full 30 Senators are age 70 or older. Sixteen of them are over 75. Nine are over 80!
The oldest, Diane Feinstein of California, is 84 years old and just announced that she intends to run for re-election. Should she win, by the end of her term, she will be 91 years old — if she survives. The average life expectancy for an 85 year old woman is 6.9 years. In other words, she will have nearly a 50% chance of dying in office before she completes her term.
In the House of Representatives, Speaker Ryan is the baby of the group at age 47.  Democratic Minority Leader Nancy Pelosi is 77. The average House member is 57 years old, the oldest average ever. Over 30% of the Members are age 65 or older. Over 15% are over 70. Twelve Members are over 80!
The median age of Justices of the Supreme Court is 67. Two Justices are over 80.  One is 79. In the 19th Century, the average Justice served about 10 years. Now they sit on average close to 25 years.
In short, the majority of the leadership of all three branches of the US government are old enough to collect Social Security and Medicare.
Forget Boomers, most of the US leadership belongs to the Silent Generation, and formed their basic political opinions in the 1950s during the days of Ike and Senator Joseph McCarthy, and when court-ordered racial integration was just beginning.  And it shows.

Comments (13) | |

They are monsters

They are monsters

The President and his GOP majorities in Congress are monsters. As one commentator on NPR put it yesterday afternoon, the President’s default mode is to toss an armed hand grenade into a room in order to create chaos.  He can then pick out the most vulnerable, and use that leverage to enter into a win-lose deal.
Meanwhile, having been emboldened by the 2011 Debt Ceiling Debacle, the Congressional GOP majorities, who haven’t been able to legislate affirmatively, have become specialists in taking hostages and threatening to shoot them unless their agenda is enacted.
Trump and the GOP Congress combined have, as of this morning taken at least four hostages:
DREAMers – the DACA program is being terminated. After an initial claim that a deal had been made to protect young people who had been brought to the US as children and know of no other home, the malAdministration is now taking a hard line, refusing to protect the nearly 1 million enrollees from deportation unless it gets its entire immigration policy enacted.

SChip recipients – this program, which provides medical insurance coverage for over 8 million  lower income children, was allowed to expire on September 30.  Despite assurances from the Congressional GOP that it would be re instituted promptly, nothing has been done.

Puerto Ricans – Unlike Texas, Louisiana, and Florida, which are GOP majority states, the malAdministration never provided prompt aid to the over 3 million Puerto Ricans, and is threatening to withdraw the aid before basic services are restored.

Recipients of Obamacare subsidies – the malAdministration is refusing to make subsidy payments under the ACA to insurers who enroll those who have less than 2.5 times the income of the Federal poverty level, which includes about 7.5 million people who have enrolled under Obamacare.

That’s a total of close to 19,000,000 hostages.

Comments (4) | |

One more scene from the September jobs report: late cycle deceleration continues

One more scene from the September jobs report: late cycle deceleration continues

The rate of year over year job growth is probably the single best mid-expansion indicator, in part because there is very little noise in the Establishment survey jobs data YoY. But, as the below graph shows, going back all the way to 1948, while it is noisier the Household survey YoY jobs data also traces out the same pattern with very few exceptions (notable the early 1950s and the mid 1960s):

Even a cursory glance at the graph shows that we are on the decelerating side of that indicator. Here’s a close-up of the last 10 years:

Although the Establishment and Household numbers moved in very different directions this month, viewed in context both show a significant downshift in 2017 from the last few years.

Yesterday I noted that if leisure and hospitality jobs had grown by their 12 month average of +27,000 in September vs. their actual -111,000, the September Establishment survey would have grown by a relatively weak 105,000 (yes I know it is more complicated than that, but it is a good K.I.S.S. estimate). Since in September 2016 jobs grew by +249,000, even with that hurricane adjusted estimate, YoY job growth would have decelerated to 1.3%.

In the past-WW2 era, typically late-cycle deceleration was accompanied by (and generally caused by) an increase in inflation and an increase in Fed interest rates to chase after it. The few times there were multiple YoY peaks in job growth (the 1960s, 1980s, and 1990s), the Fed engineered “soft landings” where it lowered rates after initial raises.

Per Tim Duy, who has a good record of Fed-watching, even in the absence of rising inflation, they seem bound and determined to raise rates again in December. A December rate hike shouldn’t be enough to push the economy into a later recession, but it should put further downward pressure on job growth in 2018.

Comments (1) | |

Scenes from the September jobs report

Scenes from the September jobs report

On Friday I highlighted the difference between the results of the establishment survey and the household survey.  A 2006 paper from the BLS (pdf) explaining the differences in how jobs are counted in the two surveys shows us why:

Interviewers from the Census Bureau contact households and ask questions regarding the labor force status of members of the household during the calendar week that includes the 12th day of the month. The broad coverage of the CPS encompasses … workers temporarily absent from work without pay.

….

[In the Establishment survey, b]usinesses report the number of persons on their payrolls who received pay during the pay period that includes the 12th day of the month. Workers who did not receive pay during the pay period are not counted.

Thus an employee at, say, Barnacle Bill’s Seashore Restaurant, who wasn’t paid during the week of Hurricane Irma because the restaurant was closed, doesn’t get counted in the Establishment survey, but *does* get counted in the Household survey.

Thus, although the household survey is the smaller sample, and thus subject to much more noise, it probably gave us a much truer picture of the labor market for the whole of September.  While the employment gain itself (906 thousand!) was insane, and surely not accurate, the ratios of unemployment, underemployment, and participation in the survey probably picked up the true  trend of improvement.

So let’s look at those.  First of all, here is the U6 underemployment rate.  I’ve subtracted 8.3% from the result, to better show how the present situation compares to the last two expansions:

In the last expansion, the underemployment rate newer got below 7.9%. The late 1990s was a genuine boom.

Comments (0) | |

Scenes from the September jobs report

Scenes from the September jobs report

On Friday I highlighted the difference between the results of the establishment survey and the household survey.  A 2006 paper from the BLS (pdf) explaining the differences in how jobs are counted in the two surveys shows us why:

Interviewers from the Census Bureau contact households and ask questions regarding the labor force status of members of the household during the calendar week that includes the 12th day of the month. The broad coverage of the CPS encompasses … workers temporarily absent from work without pay.

….

[In the Establishment survey, b]usinesses report the number of persons on their payrolls who received pay during the pay period that includes the 12th day of the month. Workers who did not receive pay during the pay period are not counted.

Thus an employee at, say, Barnacle Bill’s Seashore Restaurant, who wasn’t paid during the week of Hurricane Irma because the restaurant was closed, doesn’t get counted in the Establishment survey, but *does* get counted in the Household survey.

Thus, although the household survey is the smaller sample, and thus subject to much more noise, it probably gave us a much truer picture of the labor market for the whole of September.  While the employment gain itself (906 thousand!) was insane, and surely not accurate, the ratios of unemployment, underemployment, and participation in the survey probably picked up the true  trend of improvement.

Comments (1) | |

September jobs report: establishment survey stinks, but household survey rocks!

September jobs report: establishment survey stinks, but household survey rocks!

HEADLINES:
  • -33,000 jobs lost
  • U3 unemployment rate down -0.2% from 4.4% to 4.2% (new low)
  • U6 underemployment rate down -0.3% from 8.6% to 8.3% (new low)
Here are the headlines on wages and the chronic heightened underemployment:
Wages and participation rates
  • Not in Labor Force, but Want a Job Now:  down -216,000 from 5.844 million to 5.628 million
  • Part time for economic reasons: down -133,000 from 5.255 million to 5.122 million (new low)
  • Employment/population ratio ages 25-54: up +0.5% from 78.4% to 78.9% (new high) 
  • Average Weekly Earnings for Production and Nonsupervisory Personnel: up $.0.09 from $22.14,  to $22.23, up +2.5% YoY.  (Note: you may be reading different information about wages elsewhere. They are citing average wages for all private workers. I use wages for nonsupervisory personnel, to come closer to the situation for ordinary workers.)
Holding Trump accountable on manufacturing and mining jobs
 Trump specifically campaigned on bringing back manufacturing and mining jobs.  Is he keeping this promise? 
  • Manufacturing jobs fell by -1,000 for an average of  +9,800 a month vs. the last seven years of Obama’s presidency in which an average of 10,300 manufacturing jobs were added each month.
  • Coal mining jobs rose by 500 for an average of +133 a month vs. the last seven years of Obama’s presidency in which an average of -300 jobs were lost each month

July was revised downward by -51,000. August was revised upward by +13,000, for a net change of -38,000.

The more leading numbers in the report tell us about where the economy is likely to be a few months from now. These were mainly flat.
  • the average manufacturing workweek was unchanged at 40.7 hours.  This is one of the 10 components of the LEI.
  • construction jobs increased by +8,000. YoY construction jobs are up 184,000.
    • temporary jobs increased by +5,900.

Comments (1) | |