Weekly Indicators for June 11 - 15 at XE.com

 - by New Deal democrat

My Weekly Indicators post is up at XE.com.

There was a little whipsawing among some data, but the overall picture remains consistent.

May industrial production: meh

 - by New Deal democrat

Industrial production is the ultimate coincident indicator. It is almost invariably the number that determines economic peaks and troughs.

In May it declined -0.1%. While that obviously isn't a positive, it does nothing to suggest any sort of change of trend:

and is in line with any number of similar monthly numbers during the expansion.

In this second graph I've broken it down into manufacturing (blue, left scale) and mining (red, right scale):

Again, there's nothing here to suggest any change of trend. In particular, the energy production sector of the economy which has done so much to undergird the overall number for the last 9 years, continued to improved.

So: meh.

May retail sales come in strong

 - by New Deal democrat

Real retail sales for May came in strong, up +0.6% just in the month:

As the graph shows, this is on trend for the entirety of this expansion, and is also a new high, surpassing that of last winter.

Per capita real retail sales also made a new high, an indicator that the expansion is likely to continue at least one more year:

Finally, the YoY% growth in real retail sales has also been increasing:

Since this is a short leading indicator (3 to 6 months) of the trend in YoY% growth in employment, it suggests that the recent run of strong monthly jobs numbers should continue, averaging in about the 175,000 to 200,000 range over the next few months.

Earlier this week I wrote that real retail sales would show how well the average household is doing holding up, considering the stagnation of real wages over the last several years.  Needless to say, the answer is "fine, so far."

I hope to say much more on this at some point in the next week or two.

The Fed's actions say "So much for symmetry"

 - by New Deal democrat

Interest rates and money supply are the two long leading indicators under the control of the Fed. With yesterday's indication that they will hike 4 rather than 3 times this year, they are belying their statements that they will treat 2% core inflation as a target rather than a ceiling.

This post is up at XE.com.

Chewing over the message of online job postings

 - by New Deal democrat

Here's an interesting graph I came across yesterday. It's from the Conference Board. What it does is track the number of job postings online, and breaks them down between first postings and repeat postings:

Let me say first of all that it is of limited use. The data only goes back to 2005, so there isn't much history -- heck, online job postings didn't even *exist* until the end of the 1990s! Further, online job postings have undergone a secular increase, as more and more companies have come to use it -- so I would expect the historical trend to be positive.  But has it reached maturity? Or is it still in its growing stage?  I dunno .... and probably neither does anybody else.

But with all those cautions, it struck me that the big gap between first time and repeat postings in of a piece with the gap between actual hires and job openings in the JOLTS series:

In the case of online job postings, the repeat postings are either trolling for resumes on an ongoing basis, or else evidence of inability to find a suitable candidate (at the wage the employer wants to pay) the first time around.

What is also interesting is how the two series converged beginning in 2006 as the 2007 recession approached. Which is what I would expect.  As the job situation softens, I would expect companies to pull back on repostings more than original postings, as they become hesitant to go through with a hire if a good candidate does not appear promptly.

I'm not sure what to make of the decline in both during 2016-17. It certainly was not borne out by other employment data.  But for now, the two series continue to diverge. That does not strike me as a sign that companies are skittish about hiring.

Gas- and housing-powered inflation mean real wages are going nowhere

 - by New Deal democrat

This morning consumer price inflation for May was reported at +0.2%. YoY inflation was 2.8%. This is tied for the highest in six years (blue):

The cause of the increase was primarily twofold -- and neither one reflective of wage inflation. First, gas prices have increased by over 20% in the past year (red, right scale above).

Second, the costs for shelter (housing) are picking up steam again, up 3.5% YoY(red):

Note that, courtesy of gas prices, inflation for everything else except for shelter has also been rising (blue).

Most everyone assumes the Fed will raise rates again later this week. Let me point out, as others have as well (e.g., Dean Baker) that this will only make housing costs *more* rather than less expensive, and so will not serve to bring inflation down, short of causing a recession.

Meanwhile, while *nominal* wage growth for ordinary workers has finally been rising in the last few months, up 2.8% YoY:

today's 2.8% YoY consumer inflation means that *real* wages haven't grown at all, up less than 0.1% YoY:

Further, since gas prices bottomed in winter 2016, *real* hourly wages have barely risen at all, as shown in the below graph normed to 100 as of February 2016:

Since that time, real wages have increased only 0.3%.

We'll get our best clue as to whether the ordinary American household is simply holding its own, or is in actually undergoing increased stress, when retail sales are reported later this week.

Update: wholesalers' sales and inventories -- it's all good

 - by New Deal democrat

Another slow start to the data this week, so let's take a look at relationship I haven't updated in awhile.

Total sales in the economy are broken up into three categories: manufacturers', wholesalers', and retailers'. We'll get retail sales, the biggest component of the three, later this week. 

But wholesalers' sales and inventories were released last week, and are a useful coincident barometer. They are a better measure than manufacturers' sales, since those have been very much secularly affected by the adoption of just-in-time inventory controls.

The important thing to remember is that sales (blue, left scale) lead inventories (red, right scale).  Here's both for the last 20 years:

Note than in addition to the two last recessions, sales also plateaued first in 2012 slightly before inventory growth did, and again during the "shallow industrial recession" of 2016. As of April, both sales and inventory were both rising, a very typical result during an expansion.

Also, frequently commentators will write about gyrations in the inventory to sales ratio. But a comparison of the ratio (green, right scale) to the more leading measure of sales (blue, left scale) reveals that there the inventory to sales ratio only started to rise significantly coincident (literally, as of the same month) with the downturn in sales for each of the last two recessions as well as the shallow industrial recession of 2016. 

Before that, any such upturn in the ratio was within the range of noise and so gave no meaningful signal. In other words, that inventories might be rising at a faster rate than sales doesn't give us information that is useful in terms of analyzing the business cycle.  And, if anything, the trend in the ratio recently is downward.

So the overall message is that, for now, where sales are concerned, it's all good.

Sent from my iPad