Saturday, September 29, 2018

Weekly Indicators for September 24 - 28 at Seeking Alpha


 -by New Deal democrat

My Weekly Indicators post is up at Seeking Alpha.

No surprise that the Fed rate hike took center stage this week.

As usual, not only is this a good up to the moment look at the economy, but clicking through and reading puts a few pennies in my pocket.

Friday, September 28, 2018

Housing: comprehensive review of August reports


 - by New Deal democrat

I pay particular attention to housing because it is ian important long leading indicator for the economy.

And more and more evidence is accumulating  -- although it is not universal -- that housing may have passed its peak in this cycle. Last year housing was resued by an autumn surge. I don't think that will happen this year, but there are conflicting signals.

My comprehensive review of that evidence is up at Seeking Alpha.

As usual, your clicking over and reading hopefully will be educational. It will also reward me, ever so slightly, for the work I do putting this stuff together.

Thursday, September 27, 2018

Contracts for existing homes declined in August for the fourth time in the last five months


 - by New Deal democrat

This is something that I don't normally bother to cover, because it deals with existing homes, but since Bill McBride a/k/a Calculated Risk is off hiking, let's take a look.

The NAR reported this morning that pending home sales, i.e., contracts to buy existing homes, declined by 1.8% in August. This was the fourth decline in the last five months. This metric has now been negative YoY for the last eight months.

Since pending contracts become existing home sales one or two months later, this strongly suggests that existing home sales will continue their recent decline for the next several months.

According to the NAR's spokesman, Lawrence Yun:
The greatest decline occurred in the West region where prices have shot up significantly, which clearly indicates that affordability is hindering buyers.
So far, that makes perfect sense. But then he added,
and those affordability issues come from lack of inventory, particularly in moderate price points
which is somewhat misleading. Inventory has increased YoY for the last several months, which means that if we could seasonally adjust, it would probably have bottomed earlier this year.  And yet prices have continued to rise.

Elsewhere, the report suggests that inventory will continue to rise, and that potential sellers haven't gotten the message that prices are too high, because Yun also said:
According to the third quarter Housing Opportunities and Market Experience (HOME) survey, a record high number of Americans believe now is a good time to sell. Just a couple of years ago about 55 percent of consumers indicated it was a good time to sell; that figure has climbed close to 77 percent today.
That 77% number isn't because they expect to have to sell at lower prices.

The takeaway from today's Pending Home Sales report is that the decline in existing home sales from its peak a year and a half ago is going to continue.

Wednesday, September 26, 2018

New home sales show further signs of rolling over


 - by New Deal democrat

New home sales, despite a month over month increase, continued their slowdown in August. Here's the graph supplied by the Census Bureau:



Since new home sales are extremely volatile month over month, the best way to look at them is on a three month average basis, and on that basis new home sales made an 11 month low.

Note that the YoY comparison is with the worst levels of last year. Last autumn there was a surge in sales. The question is, will a similar surge come to the rescue of the housing market this year? My belief is, it will not.

Last week existing home sales also continued unchanged at 12 month lows. Existing home sales have not made a new high since March 2017.

Further, inventory of new homes continues to climb slowly. Combined with the slowdown in sales, this is at very least another yellow flag that the housing market may be rolling over.

I'll have a more detailed look at both sales and prices tomorrow at Seeking Alpha, and will link to it once it is up.

Tuesday, September 25, 2018

How close are we to full employment? September update


 -  by New Deal democrat

A couple of months ago, I estimated that we were about 0.8% away from "full employment." Theoretically, that should be when everyone who wants a job has one, but more realistically, the benchmark would be peak employment from the last few cycles.

We've had a couple more good employment reports since then, so where are we? In the below graphs, I'm going to compare our situation with the best situation in the last 50 years: the tech boom of 1999-2000.

First off, the primary difference between the U3 unemployment rate and the U6 underemployment rate is people who are working part time but would like a full time job.  At its best in 1999, 2.2% of the labor force were in this condition, so I've subtracted that in the graph below:



Notice that despite the big drop in the last couple of months, at the far right, we are still slightly above 2007 levels, and a little more above 1999 levels. Here is the close-up:



Currently the level of involuntary part-times is about 0.5% above its 1999 bottom.

Next, above the underemployment level are those who aren't even looking for a job, and so aren't counted in the labor force at all, but say they want a job now. This is roughly 5 million people. Even at its best level in 1999, when we include these people as well, and add them into the labor force, it totaled 9.8%, so that level is subtracted in the below graph:



Again, currently we are about 0.8% above the 1999 "full employment" level.

Put these two together, and we can calculate that if about 750,000 involuntary part-timers were to get full time jobs, and another 1.2 million people who aren't even in the labor force got jobs now, we would equal the best level of the "enhanced underemployment rate" 1999. That's one good estimate of how close we are to "full employment."

Notice that is the same level I caluclated two months ago.that's because, despite the progress on *under*employment, there's been no progress at all on those who are not in the labor force but want a job now (in fact that number has risen slightly since March). 

But now, let's look at the prime age employment population ratio:



Even if we added the 0.8% of people who want a job now, and the underemployed 0.5% got full time jobs, that would just put us at the levels of the 1989 and 2007 peaks. The level would still be roughly 1.5% below the 1999 peak.

This 1.5% comes from the population who aren't in the labor force, and indicate that they *don't* want a job now. Mainly these people are disabled, or raising children at home, or younger people who are in school. But, especially among the first two groups, if wages were higher, some would decide it was financially advantageous to join the labor force.

As I said yesterday, the signature of our current economy is the dual extremes of very low unemployment, and very low wage growth, and I did not think that was a coincidence. If wages were higher, some of those 1.5% out of the labor force would enter it, and not all would find jobs, meaning that the unemployment rate would be higher.

Monday, September 24, 2018

The *rate* of new jobless claims, at all-time lows, forecasts even lower unemployment


 - by New Deal democrat

It's a slow news week, so I thought I'd start out with something I haven't looked at in awhile: initial jobless claims as a share of the population and as a leading indicator for the unemployment rate.

This economic expansion has featured two contrary extremes in the labor market: low wage growth and increasingly vanishing layoffs (I don't think that's a coincidence, but that's a subject for another post!).  Let's take a look.

The first graph below is of weekly jobless claims as a share of the entire labor force. To generate this, I also take into account "covered employment." That is, not all jobs are eligible for unemployment insurance, and because of part time work and the gig economy, that share is less than it had been previously. So, first I divide the average number of jobless claims over a month by the percent of covered employment. The lower the share of covered employment, the higher the adjusted layoffs number. Then I divide that in turn by the number of people, both employed and unemployed, in the labor force.  Here's what I get, through August:



We are now experiencing the lowest percentage of people getting laid off as a share of the labor force, for the entire 50 year plus history of the data. As of August, only 138 out of 100,000 people were getting laid off each week (obviously, given that employment is growing, a bigger number were getting new jobs). Put another way, you almost have to work at getting laid off these days!

Secondly, initial jobless claims tend to lead the unemployment rate by a few months. Here's the 50 year+ graph:



If jobless claims decline, then over the next 2-3 months it's a good bet that the monthly unemployment rate will decline too.

Here's a close-up of the last nine years:



Not every zig and zag is followed, but the general relationship is clear. Note that initial jobless claims have continued to make new 48 year lows this month (September).

The unemployment rate made its most recent low of 3.8% in May. Based on the relationship with initial jobless claims, we should expect it to go even lower by the end of this year.