Saturday, January 16, 2021

Weekly Indicators for January 11 - 15 at Seeking Alpha


 - by New Deal democrat

My Weekly Indicators post is up at Seeking Alpha.

While initial jobless claims have continued to worsen, there is no sign of a generalized downturn in the coincident data, and the upturn in long term interest rates isn’t enough to change their fundamental positive impact on the economy.

As usual, clicking over and reading is rewarding just a little bit for me, and brings you up to the virtual moment as to what is happening in the economy.

Friday, January 15, 2021

Good news (industrial production) and bad news (retail sales)


 - by New Deal democrat

This morning’s two reports on industrial production and retail sales for December were a case of good news and bad news.

Let’s do the good news first. Industrial production, the King of Coincident Indicators, rose 1.6% in December. The manufacturing component rose 1.0%. Needless to say, these are a strongly positive numbers. As a result, overall production is only -3.3% below its February level, while manufacturing is only down -2.4% since February:

Manufacturing has consistently been one of the biggest bright spots in the economy ever since April. 

Now on to the bad news.

Nominal retail sales declined -1.0% in December. Excluding the food services sector which has been especially hard hit by the pandemic and lockdowns, sales were nevertheless down -0.3%. Total sales are down -2.7% since September; excluding food sales they are still down -1.5%:

In real terms, retail sales declined -1.0%, and are down -2.7% since September:

The silver lining is that this type of decline doesn’t necessarily mean a steep decline back into recession lows. Similar declines happened several times during the last expansion. And they are still up 1.5% since February.

As I have pointed out many times, consumption leads employment. It’s even a better match for aggregate hours worked in the economy, as shown in the long term graph below:

And here’s what it looks like YoY compared with aggregate payrolls:

Last month I wrote that “I continue to expect employment to continue to rise - with a lag, and quite possibly a pause during this winter as the pandemic continues to rage - to match the level of sales.” Well, we certainly got the “pause” in December’s employment report, and if we get more initial jobless claims reports like yesterday’s, a further downturn. But I still expect employment to rise to meet sales once the winter surge in cases and shutdowns of outdoor activities including dining both abate.

Thursday, January 14, 2021

Jobless claims highest in three months - but seasonality still playing a huge role


 - by New Deal democrat

On a unadjusted basis, new jobless claims rose by 231,335 to 1,151,015. Seasonally adjusted claims also rose by 181,000 to 965,000. The 4 week moving average rose by 18,250 to 834,250.

Here is the close up since the end of July (these numbers were in the range of 5 to 7 million at their worst in early April): 

There is now a 2+ month trend of YoY% increases in initial claims. Further, by rising to over 900,000, seasonally adjusted claims hit one of my two markers for a fundamental change of trend. But the 4 week average, which is still under 850,000, did not.

For the last couple of months, I have been cautioning that the holiday season plays havoc with seasonality even in normal years, let alone a year when the pandemic is causing changes in weekly numbers by an order of magnitude. That was especially so this week. Typically in the weeks after Christmas and New Years’, claims go up by 25,000 to 50,000 on an unadjusted basis. This year claims went up by up to 10x as much.

But when we look at the YoY% change in all of the above metrics, which washes out those outsized seasonal affects, the picture is not nearly so negative:

While weekly seasonally adjusted claims have risen 366% - a YoY comparison last seen in August, non-seasonally adjusted weekly claims, and the 4 week average are still in the YoY range they were in October, at +240% and +290% respectively. In fact the non-seasonal trend looks more sideways than increasing.

In other words, this was one bad week of data, but we are clearly above November lows, but nothing awful (or at least, not materially *more* awful than before) has happened yet.

Turning to seasonally and non-seasonally adjusted continuing claims, which historically lag initial claims typically by a few weeks to several months, the former rose by 199,000 to 5,271,000, and the latter rose by 474,180 to 5,856,230:

Because these lag initial claims, I have suspected that we would see an upward reversal, and it appears to have arrived.

Finally, as I usually note, a reminder that both initial and continued claims remain at or above their worst levels from the Great Recession.

Bottom line: renewed partial lockdowns and increased consumer caution due to the out of control pandemic have caused increased layoffs. Should we get another week like this one in the next couple of weeks, that will probably mark the decisive break of trend. At the same time, it isn’t quite as bad as I would have thought several months ago. Further, since I expect Biden to tackle COVID as his very first priority (even before prosecutions for the January 6 putsch attempt), hopefully we will see the worst within 4 to 8 weeks.

Wednesday, January 13, 2021

December inflation: real aggregate wages fall as real average wages rise due to compositional effects


 - by New Deal democrat

Consumer inflation typically rises as expansions continue, and declines once recessions start. Once a recovery begins, inflation typically steadies again. While the pandemic has affected both consumer demand and the supply chain, overall the paradigm should still apply. With that in mind, let’s take a look at December’s report.

For the past 40 years, recessions had typically happened when CPI less energy costs (red) had risen to close to or over 3%/year. As of this month that number is exactly 2%: 

In other words, there is no troublesome price pressure outside of the pandemic. Because pandemic affects are probably influencing seasonality, I also show non-seasonally adjusted inflation below (red):

Typically inflation increases in the first few months of the year, and abates in the latter part. The unusual 0.4% seasonally adjusted increase for December appears to have been primarily driven by a 0.3% increase in food prices. I expect that price pressure to pass in the next few months.

Now let’s take a look at how inflation has affected real wages. Because wages are “stickier” than prices, typically as recessions beat down prices (or at least price increases), in real terms wages rise, either during or just after a recession. That has been the case for the coronavirus recession as well. It is the “real” buying power of wages among those still securely employed during a recession that is one of the engines that usually restarts growth. 

Also as a result, real hourly wages for non-supervisory workers have exceeded their previous 1973 peak by 1.6%:

Finally, one of the most telling metrics of the overall health of the middle/working class is that of real aggregate wages. After declining -13.8% from February through April, they recovered to -3.1% below their peak in November, before declining -0.2% last month. Here is comparison of the YoY changes in average (blue) vs. aggregate (red) wages, showing the outsized upward skew in average wages vs. the huge hit to aggregate wages - even bigger than during the Great Recession:

While average wages have risen by over 5% in the past year - the best showing since the inflationary 1970s - this is due to compositional effects, as the pandemic has resulted in outsized layoffs to lower paid service workers. 

What we want to see is the red line above zero, and the blue line at least not declining further. All of this is at the mercy of the course of the pandemic. 

Tuesday, January 12, 2021

November JOLTS report shows renewed impact of pandemic, partial lockdowns


 - by New Deal democrat

This morning’s JOLTS report for November (remember - a month in which there were total job gains) showed a jobs market recovery that at least paused due to the increasing effects of the out of control pandemic. Hires were up (good), while quits were unchanged, openings declined (bad) and layoffs and discharges rose (bad).

While the JOLTS data is a deep dive into the dynamics of the labor market, since it only dates from 2001, there are only 2 previous recoveries with which to compare the present. Nevertheless it is worthwhile to make the comparison.

In the two past recoveries:
  • first, layoffs declined
  • second, hiring rose
  • third, job openings rose and voluntary quits increased, close to simultaneously

Let’s examine each of those in turn. In each case, I break out 2001-19 in a first graph and then this year in a second.

What appears below is that, although there has been some variation, the year 2020 through November recapitulated the pattern from the last two early recoveries: the first two data series to turn - layoffs and hires - have indeed turned, while the last two - job openings and voluntary quits - have appeared to bottom but have had a much less dramatic rise. With increased pandemic restrictions and consumer caution, several renewed negative readings in November, but not enough to significantly change the trend.

This first graph compares layoffs and discharges (blue) with the 4 week average of initial jobless claims (red) prior to this recession, for reasons of scale since March and April would be “off the charts”:

You can see that, by the end of the recessions, layoffs were already declining, and continued to decline steeply over the next 3-8 months before reaching a “normal” expansion level. The turning point coincides exactly with the much less volatile, but more slowly declining, level of initial jobless claims.

The same had been the case this year up until October. Layoffs and discharges already declined to their “normal” level in May, while initial jobless claims peaked one to two months later, and continued to decline (slowly) through November. We already know, however, that the rise in JOLTS layoffs in November and December absolutely showed up  in initial jobless claims in December: 

The continued rise in layoffs in December suggests that initial claims will continue elevated over their November lows through this month.

Next, here is the entire historical relationship between hires (red) and job openings (blue) through 2020:

In the past two recoveries, actual hires started to increase one to two months before job openings.

Both made troughs in April, but hires have rebounded more sharply ever since May compared with job openings, although both have essentially leveled off at those levels, and openings slightly declined in November and may even have a slightly declining trend:

Next, here are quits (green) vs. job openings (blue): 

In the past two recoveries, openings started to rise slightly before quits made a bottom. After that, both rose more or less together (suggesting it is openings that leads to the increase in voluntary quits).

This year, both made a trough in April. Since then, openings appears to have continued to slightly led quits, with both flattening out in the past several months:

Because seasonal adjustments might not be giving us a true picture because of the enormous moves during this pandemic year, here are job openings (blue), hires (red), and voluntary quits (green), measured YoY without seasonal adjustments for the entirety of the series up through the present:

We can see that, even taking out the seasonal adjustments, hires rebounded first following the 2001 and 2008-09 recessions. Quits and openings moved generally in tandem with a slight lag. The same pattern generally appeared in 2020, with quits perhaps slightly lagging.

Finally, I have broken out layoffs and discharges separately below, because the their level in April and May of this year would obliterate all other variations (note: inverted so that fewer layoffs shows as positive):

This metric returned to normal almost immediately after both of the past two recessions, and did so again by July of this year. But its gain decelerated in October, and turned negative YoY in November (which, since the line is inverted, means higher total layoffs and discharges:

To sum up:

1. The JOLTS report continues to show a pattern generally consistent with the past 2 recoveries, with layoffs having returned to normal levels, then hiring having increased, and finally quits and openings increasing as well; but

2. We are seeing the results of the out of control pandemic in the increasing layoffs and separations in November, likely due to renewed partial lockdowns and increasingly cautious consumer behavior as well as consumers pulling back on “al fresco” activities in the cold winter weather.

Monday, January 11, 2021

Scenes from the December jobs report


 - by New Deal democrat

Friday’s December jobs report saw the first decline in employment since the lockdowns of March and April. Let’s take a closer look.

As I pointed out Friday, the losses were concentrated in the food and dining (restaurant) and amusement and recreation sectors, both of which are shown below normalized to 100 as of February:

The two sectors are down 20% and 30% from their February peaks.

By contrast, the leading job sectors of manufacturing, residential and overall construction, and temporary help positions all continued with gains, and are close to if not completely recovered from their pandemic losses. Below I show these YoY in two time periods for easier comparison (note two of the series did not begin until the 1980s).


1983 - present:

This is one of the many signs pointing to a strong rebound in the overall economy once the pandemic is brought under control.

Another important way to look at employment is via the differential impact of the pandemic on the goods-producing vs. service-providing sectors of the economy. Below are the YoY% changes also broken down into two time periods for easier comparison.


1983 - present:

Up until the Great Recession, the goods-producing sector always bore the brunt of layoffs. In fact the worst YoY losses in service jobs was 1.4% in 1949 and 1.3% in 1958, respectively. In 2009, by contrast, the worst losses were 3.2%. Last year, there were 13.8% losses in April, improving to their “best” reading of 6.3% losses in November.

Finally here are the breakdowns in YoY job gains and losses between men (blue) and women (red), divided into the same two time periods:

Because women have disproportionately been employed in the lower paying service sectors, they did not suffer layoffs *relatively* as bad as men during recessions prior to the turn of the Millennium. This got worse during the Great Recession, and in this pandemic recession they have been hit the worst of all historically.

Programming note


 - by New Deal democrat

Four year ago I wrote a valedictory piece about the Obama Administration, and separately wrote of my fears of what the Trump Administration would wreak.

Needless to say, especially in light of events of the past week, I intend to do the same retrospective as to Trump and the current state of the GOP and the Republic. Much of what I have to say is in agreement with disparate threads I have read on twitter, but I want to weave those strands together into one cohesive piece. Hint: I keep thinking about old episodes of Supernanny, where a toddler’s behavior was allowed to get worse and worse without consequence. The longer it went on, the more forceful and resolute the parents’ response ultimately had to be.

Hopefully this will be a long-form piece next Sunday.