Saturday, May 13, 2023

Weekly Indicators for May 8 - 12 at Seeking Alpha


 - by New Deal democrat

My “Weekly Indicators” post is up at Seeking Alpha.

There was more slow deterioration in the coincident indicators, but interestingly a bounce in several of the short leading indicators, particularly in the weakness of the US$ (which paradoxically is a positive, because it helps exports and crimps imports).

But the biggest news was in one of the twice a month indicators, consumer confidence from the University of Michigan, which did this:

As I wrote yesterday, in 2011 all of the indicators cratered at once, and in particular consumer confidence, so the sudden downturn in consumer expectations in the past couple of weeks is a bad sign.

Friday, May 12, 2023

Real hourly and aggregate wages update; plus further comments on consumer and producer inflation


 - by New Deal democrat

Let’s update some inflation-related information.

First of all, real hourly wages for non-managerial personnel increased less than 0.1% in April. They are up about 3% from just before the pandemic, and also up a little over 1% since their June low last year:

Note the graph above is normed to 100 as of the long-time previous high for wages in January 1973.

Next, real aggregate payrolls for non-supervisory workers tell us how much money the middle/working class is earning in total, adjusted for inflation. This declined -0.1% in April, but is 4.4% above its pre-pandemic level:

But the overall trend remains a very slow increase. Here’s what the YoY% change looks like historically:

Note that when real aggregate payrolls go negative, that has *always* been a coincident indicator of recession. At 1.7% higher YoY, this is about average for previous expansions, and is an important positive sign.

Digging a little deeper in the CPI report, aside from shelter, the big YoY increases have been in food (13.5% weighting of the total, unchanged in the last 2 months but up 7.7% YoY, down from a peak of 11.3% last August); new cars (4.2% weighting, down -0.2% in April, but up 5.4% YoY vs. a peak of 13.2% in April last year); and something called “transportation services,” (5.9% weighting, down -0.2% in April, but up 11.1% YoY, down from 15.3% last October).

So, exactly what are “transportation services”? Here’s the detailed breakdown from the CPI report:

The two big items are motor vehicle insurance and repairs. This is telling me that we still have a bottleneck in vehicle parts, which are impacting both the manufacture of new cars and the repair of older ones - especially since so many people are holding on to their older cars, given the prices of the new ones.

Finally, the producer price index for raw commodities rose 0.1% in April, after declines in both February and March. Producer prices for final demand goods rose 0.2%. If this sounds well-contained to you, that’s because it is. YoY commodity prices are down -3.0% and finished goods prices are up only 0.8%:

An important difference between the PPI and the CPI is that there is no “shelter” component in the former. Below are two long term graphs of both producer and consumer prices, ever since the end of WW2:

With the notable exception of the 1980s and 1990s, when the work force was swelling both due to the entry of the Boomers and women, a downturn in YoY producer prices has always resulted in a deceleration of consumer prices as well. Since our present period most resembles the immediate post-WW2 Booms through the Korean War, as well as 1981-82, when the Fed continued hiking rates into a decreasing inflationary environment, this argues strongly that consumer prices will follow producer prices this time as well, despite producers’ efforts to maintain their recent price hikes in consumer products.

Thursday, May 11, 2023

Yellow flag from initial jobless claims turns a little more orangey


 - by New Deal democrat

Initial jobless claims rose 22,000 to 264,000 last week, while the 4 week average rose 6,000 to 245,250. Continuing claims, with a one week lag, rose 12,000 to 1.813 million:

Note that both measures of initial claims are at their highest levels since late 2021. Continuing claims are also at those levels, although slightly down from three weeks ago.

On a YoY basis, initial claims are up 25.7%, the 4 week average up 15.1%, and continuing claims up 24.4%:

If these YoY comparisons persist for another month, that would be sufficient to hoist the “red flag” recession warning. So this is a good time to reiterate that weekly data can be noisy, and this week’s spike could be the start of a trend - or it could just be an outlier. Many times in the past there have been brief crossings of the 12.5% YoY threshold that reversed quickly and did not signal a recession.

Here’s what the historical record of YoY readings in the three metrics have looked like (all normed to 0 as of this week’s reading, 1 week’s claims averaged by month):

In 4 of the 7 recessions before the pandemic, claims did not hit this level YoY until after the recession actually started. In 2 they hit this level 6 months or less before the recession, and in 1 (1990) 9 months before. But there were also 7 false positives: August 1971, February 1977, March 1979, February 1985, October 1995-June 1996, March 2005, and February 2007. Note that only one of these lasted longer than a month.

So this week the yellow flag caution turned a little more orangey.

Wednesday, May 10, 2023

Inflation ex-shelter increasing at 1.0% annualized rate since last June; core inflation with actual house prices only up 3.0% YoY


 - by New Deal democrat

Two months ago, I “officially” took the position that inflation had been conquered, and that, properly measured, the economy had actually been experiencing deflation since last June. With revisions, the “actual deflation” is no longer the case; but for the second month in a row since then, this morning’s CPI report indicates that it is only because of the lagging and fictitious nature of the measure of shelter prices that inflation is considered elevated at all.

The primary reason, as I have been pounding on for almost 18 months, is that the shelter component of official inflation, which is 1/3rd of the total, and 40% of the “core” measure, badly lags the real data - as in, by a year or more.

Before we get into all that, let’s look at the headlines, with the monthly and YoY rates of change:

Total CPI up 0.4% m/m and 5.0% YoY (tied for lowest since May 2021)
Core CPI up 0.4% m/m and 5.6% YoY (lowest since January 2022)
CPI less shelter up +0.3% and 3.4% YoY (lowest since March 2021)
Core CPI less shelter up +0.4% and 3.8% YoY
Energy up 0.6% m/m and down -4.9% YoY
Food unchanged m/m and up 7.6% YoY (lowest since January 2022)
New cars down -0.2% m/m and 5.4% YoY (lowest since June 2021)
Owners Equivalent Rent up 0.5% m/m and 8.1% YoY (all time YoY high)

Notice the pattern with the above? “Lowest since…” for almost everything *except* shelter, which is at an all-time YoY high.

Simply put, at this point both core and headline inflation are being driven almost exclusively by Owners’ Equivalent rent (shelter), with a secondary assist by food and new cars - and even those two are decelerating substantially. It is only because shelter is such a large component of the aggregate that inflation is an issue. Even including new cars and food leaves consumer inflation at a YoY rate that ought to be in the Fed’s comfort zone. Only core inflation ex-shelter remains somewhat elevated, at 3.8% YoY.

Let’s start with the YoY% changes in headline inflation (blue), core inflation (red), and inflation ex-shelter (gold):

All 3 have been in declerating trends since last June (headline and ex-shelter) or last September (core).

Next, because of the importance of shelter to my analysis, here is an updated long term YoY graph of the big culprit, Owner’s Equivalent Rent (blue), which increased another 0.5% in April, with the FHFA house price index (red, /2.5 for scale), which has been declining since last June and was up 4.0% as of its last reading for February:

Exactly as I have been saying for the past 18 months, house prices dragged OER higher and with it the CPI indexes, with about a 12 month delay. House prices on a YoY basis plateaued for a year between late spring 2021 and mid year 2022, and now it appears OER is finally plateauing as well.

Here’s what core inflation ex-shelter would look like:

This measure is significant, because in the last several months it has stalled at roughly 4.0% YoY. But it only goes partway, because we really ought to include the true measure of shelter inflation - house prices - in the calculation.

As noted above, the FHFA index is only up 4.0% YoY as of February. If it has continued to decline in the 2 months since then at the same rate, it is only up about 1.7% YoY currently, vs. 8.1% for OER. If the FHFA index were substituted for OER, then total YoY CPI for April would only be 2.8%. Core inflation, which ex-shelter is up 3.5%, would only be up 3.0%. Neither of these warrants restrictive interest rate policy.

Additionally, because of the importance of gas prices, which peaked last June at over $5/gallon, to headline inflation, if we measure since last June, then headline inflation ex-shelter has only increased 0.8%, or at a 1.0% annualized rate:

Which means, as I said above, food and new car prices are also worth looking at.

Here are food prices both m/m (blue, left scale) and YoY (red, right scale):

Inflation in food prices has rapidly decelerated. In the past 6 months, inflation has been 2.0%, or a 4.0% annualized rate.

Meanwhile, while used car prices increased 4.4% for the month, on a YoY basis inflation in both new cars (red) is also decelerating, albeit slowly, now at 5.4%, while YoY used car prices (blue) have actually turned over into deflation:

But to reiterate: properly measured inflation is no longer a significant issue. Inflation is only heightened because of the fictitious, and lagging, measure of Owner’s Equivalent Rent. If actual house prices were used, even core inflation would only be up about 3.0%. Even if OER is a valid way to measure inflation, because of the serious lag the Fed should be relying on house prices, and declare victory. It won’t, but it should.

Tuesday, May 9, 2023

Credit conditions worsen, and likely to worsen further due to Debt Ceiling Debacle II


 - by New Deal democrat

The Senior Loan Officer Survey, which measures credit on offer by banks, and the demand for credit by their customers, was released yesterday afternoon for Q1, and the news - unsurprisingly - was not good.

Credit conditions not only tightened, but they tightened at a higher rate than they had in previous quarters, as about half of all banks tightened terms for credit (in the below graph, a positive number means tightening, i.e., is worse for the economy):

Note that in the past, more often than not credit conditions were at their most tight even before recessions began. 

Not only that, but more than half of all banks reported that demand for new loans by commercial customers had declined:

The number of banks reporting reduced demand is on par with the worst of the last few recessions.

What this report tells us is that the economy has been continuing to grow based on the “catching up” on supply chain bottlenecks both on the producer and consumer sides. New demand for and offers of credit for capital projects is drying up.

There’s one other ominous sign, at least for the short term. Below is a graph of several weekly measure of financial conditions by the Chicago Fed. In these indexes, like the data above, a positive number is poor. Typically these reasonably approximate, and give advance notice of, the quarterly data in the Senior Loan Officer Survey.

The leverage subindex (red), touted by the Chicago Fed as the most leading, has done exactly that, as it is presently at levels that more often than not in the past have occurred shortly before or during recessions. The Adjusted Index, which leads but with a less variable record, is still below zero:

Here’s the ominous note: the two times that the leverage subindex has been as poor as it is now, but not associated with recessions, were:
(1) the stock market crash of 1987.
(2) the debt ceiling crisis of 2011.

That the leverage index is at those levels, as we appear headed for yet another debt ceiling train wreck, is as I said above, ominous.

I went back and checked my posts from that period in 2011. In August and September, all of the monthly data - long leading, short leading, coincident, you name it - tanked in unison, to the point where it appeared a recession may have already and suddenly started. It should not be surprising at all if the same occurs as we approach the brink of a debt default again.

Monday, May 8, 2023

Scenes from the April employment report: the Fed just can’t kill the employment beast


 - by New Deal democrat

There’s no economic news this morning, so let’s take a closer look at some important trends from last Friday’s April jobs report.

As I and many others wrote, an important theme was that the deceleration in job gains continued, as shown in this graph since January 2021 (note 222,000 is subtracted so that latest average is at zero level):

The last 3 months have averaged 222,000 jobs, the lowest since the pandemic recovery began in 2020.

But on an absolute scale, in the past 40 years, an average of 222,000 jobs per month over a 3 month period has been better than about 80% of all Quarters (graph subtracts 222,000*3 so that quarterly average equivalent to last 3 months shows at zero):

So, on an absolute scale, all that has happened is that the white hot jobs growth of 2021, which slowed to red hot jobs growth in the first half of 2022, has now cooled to simply hot jobs growth.

All of which has resulted in the highest employment to population ratio among the prime age population since the 1990s tech boom (current level of 80.8% is subtracted to show as zero):

And I won’t even bother to show the current unemployment rate, which is equivalent to the lowest in nearly the past 70 years.

As I also pointed out on Friday, there was some mixed data among the leading employment sectors: temporary (gold) and residential construction (red) did decline, but manufacturing (blue) increased to a new post pandemic high:

If all three have or are in the process of rolling over, in the past 30+ years that has typically occurred many months before the actual onset of ensuing recessions:

Why is employment holding up so well in the face of the turning down of so many other indicators - not just leading indicators, but also things like industrial production or real retail sales?

I believe it has to do with employment still having a ways to go to “catch up” with the huge increase in total consumption, including consumption of services, in real terms. Here is what growth in real consumption and employment look like since the end of the Great Recession. Since generally consumption of goods (but not services) increases faster than employment, I have normalized the trend line in consumption to best show the comparison during the last expansion:

There remains a large gap between the growth in consumption, and the growth in employment to service that consumption. To paraphrase “Hotel California,” the Fed keeps stabbing the economy with their steely knives, but they just can’t kill the employment beast.