Saturday, April 29, 2023

Weekly Indicators for April 24 - 28 at Seeking Alpha


 - by New Deal democrat

My Weekly Indicators post is up at Seeking Alpha.

Several important short leading indicators rallied this week. In particular, the stock market seems to think the worst is over (for the moment!).

At the same time, several important coincident indicators of taxation and employment are on the cusp of rolling over again.

As usual, clicking over and reading will bring you up to the virtual moment on all of the crosscurrents in the economy, and reward me a little bit for collecting and organizing all of the metrics.

Friday, April 28, 2023

A mixed picture on real personal income, savings, and spending in March, and real total sales in February


 - by New Deal democrat

As I’ve indicated a number of times recently, right now I consider the report on personal income and spending co-equal to the employment report as the most important monthly data. For March, it was a mixed bag.

Nominally, personal income rose 0.3%, and personal spending was unchanged. Because the applicable deflator rose 0.1%, real personal income rose 0.2%, and real personal spending declined less than -0.1% (also rounding to unchanged) for the month.

Since the pandemic began, real income is up 4.0%, and real spending is up 7.6%. Because much of this was distorted by several rounds of stimulus, here’s the view normed to 100 as of July 2021:

Real personal spending has risen fairly consistently, while real personal income fell and then rose again with the rise and fall of gas prices last year:

Additionally, the pesonal savings rate rose slightly again to 5.1%, which is good for individuals, but due to the “paradox of saving,” bad for the economy as a whole.

Digging in to some further details, there was much dancing around the maypole yesterday that real spending in the Q1 GDP report was up 3.4%, a very healthy number. But I noticed that the quarterly increase was well below both the January and February monthly increases, so I suspected we would see either a big decline or some significant downward revisions today - and we did, especially for February, as shown below:

Basically, extra seasonal distortions around the post-pandemic Holidays gave us a big downdraft in November and December, and a big updraft in January. Compared with September and October, February and March were only up +0.6%.

Further decomposing real personal spending by types of purchase, we see that real spending on non-durable goods since July 2021 has actually declined, while total spending on goods is only up 1%. The big increases since July 2021 have been on services, and on durable goods (mainly cars), which declined sharply in November and December and then rose sharply in January:

In other words, the lion’s share of the big quarterly jump in consumer spending in yesterday’s GDP report was a spending spree on cars in January, driven by seasonal distortions.

Finally, let’s turn to the indicators that the NBER uses to determine the onset of and end of recessions, two of which were updated this morning.

The good news is that real personal income less government transfers (red in the graph below) rose 0.3% in March to a new high. The bad news is that real manufacturing and trade sales (blue) for February declined -0.4% from their recent high in January:

Note that industrial production, perhaps the most important coincident indicator, remains down about -0.5% from its September peak. On a YoY basis, real personal income less government transfers is up 2.1%, real manufacturing and trade sales are up 0.1%, and industrial production is up 0.5%:

The historical record going back over half a century shows that when all three of these coincident indicators have been at the YoY levels they are now, with one exception we have already started a recession:

The sole exception was 1989, when we were 6 months away.

To sum up: there was good news on real personal spending on services, and on real personal income less government transfers. Depending on further revisions, it is unlikely that the NBER will ignore growing nonfarm payrolls and declare that there was a cyclical peak in January.

But the news of real personal spending on goods was negative, as were real manufacturing and trade sales for February. Personal savings increased, consistent with consumers becoming more cautious in advance of a recession. And yesterday’s good Q1 GDP news on consumer spending turns out mainly to have been a car-buying spree in January.

Thursday, April 27, 2023

Leading components of Q1 GDP paint a mixed picture


 - by New Deal democrat

As you probably already know, real GDP increased 1.1% at a seasonally adjusted annualized rate in Q1. This doesn’t necessarily mean that the economy improved throughout the period. The median GDP for the quarter where post WW2 recessions have begun was +2.4%, and there are reasons to believe that the reason for the positive number in Q1 was big January gains. We’ll find out tomorrow with personal income and spending, and real business sales, whether these were reversed in the following two months.

While GDP by and large is a look in the rear view mirror, there are two leading components. First, private residential investment as a share of GDP is a long leading indicator popularized over 15 years ago by Prof. Edward Leamer. It tends to turn down 6-7 quarters before a recession hits. It is even slightly more leading when calculated in real inflation-adjusted terms. Unsurprisingly, in Q1 of this year this declined again, although less so than in the last several quarters, whether measured nominally or in real terms:

Secondly, proprietors’ income (light and dark blue in the graph below), a proxy for corporate profits (light and dark red), which won’t be reported for another month, were up +0.2% nominally, and +3.9% without inventory valuation:

But the “official” leading metric uses unit labor costs as a deflator, which we also don’t know yet, so I’ve substituted the implicit GDP deflator as a temporary fix: 

So adjusted, proprietors’ income declined -0.7% for the quarter with the inventory adjustment, but increased 3.2% without it.

Finally, last month I noted that real final sales at levels as low as it had been in Q4 were typically seen within 3 quarters of the onset of a recession. These did improve in Q1, but don’t undo the negative inference going forward:

Overall, this is a mixed picture, but still tending to the negative in terms of implications for the near future.

Four week average of initial claims drops below “yellow flag” level, for now


 - by New Deal democrat

While GDP will get the lion’s share of attention today (and I’ll post on it later on), by and large it is a look in the rear view mirror. The more forward-looking data is weekly jobless claims.

This week initial claims declined -16,000 last week to 230,000. The more important 4 week average declined -4,000 to 236,000. Continuing claims, with a one week lag, declined -3,000 to 1.858 million:

More importantly for forecasting purposes, YoY initial claims were up 11.1%, the 4 week average up 9.8%, and continuing claims up 22.1%:

That the 4 week average declined below 10% removes the “yellow flag” recession caution for the moment, although the increase in continuing claims is consistent with a recession in the immediate future.

Wednesday, April 26, 2023


 - by New Deal democrat

Aside from the monthly jobs report, at this time imo the most important economic data will be issued this Friday: namely, real personal income and spending, the deflators of which also figure in the calculation of real manufacturing and trade sales.  That’s because, while real *consumer* spending of goods has been flat to declining for a year, and manufacturing production has been flagging, real personal consumption of *services* has been running historically hot, while producer price deflation has helped buoy *producer* sales.

Let’s start with a follow-up on my note this morning on durable goods orders. I pointed out that in 2015-16 there was a “shallow industrial recession” which never brought down the economy as a whole. That’s because, while durable goods orders (bright red) and industrial production (dark red) both declined 10% or more during that period, consumer spending as measured by both real retail sales (light blue) and real personal consumption expenditures (dark blue) sailed right along:

Let’s compare that to our post-pandemic period. For roughly the past 18 months, both real retail sales have been flat or even slightly declining, both industrial production and durable goods spending have only in the past six months done the same. Meanwhile real personal consumption expenditures have continued to improve:

 Further dissection of personal consumption expenditures shows that *nominal* expenditures for goods has historically tracked very closely with *nominal* retail sales:

But the deflators for the two series are different, as a result of which *real* retail sales have not performed nearly as well as *real* personal consumption expenditures for goods:

Further, historically both real retail sales and real personal consumption expenditures for goods have turned down YoY both earlier and more deeply than real personal consumption expenditures for services (gold):

The same graph since the pandemic recession shows that YoY spending on goods is flat to declining in both series, while YoY real spending on services is still a historically robust 3%:

Finally, it’s worth pointing out again that personal saving tends to increase just before recessions, as consumers grow more cautious:

Note that this has already occurred in the past 6 months:

To return to my main theme, what has been so important about the reports on real personal spending and real manufacturing and trade sales is that (1) the deflators are more favorable to growth than in other “real” series; and (2) in particular, real spending on services has barely flagged at all.

On Friday I will be paying particular attention to whether or not this pattern continues, or whether real sales and real consumption at last turn down, and whether real consumption on services in particular decelerates significantly or not.

Transportation orders increase, but core capital goods orders decline further in March


 - by New Deal democrat

Durable goods orders increased in March by 3.2%, which sounds great, except that it was primarily transportation orders (Boeing). Core durable goods excluding transportation and defense declined -0.4%:

While both core and total durable goods orders are down from their peaks last year, joining the recent decline in residential construction among the leading sectors, neither are off nearly as much as their 10%+ declines in 2015-16 that at the time I labeled the “shallow industrial recession,” or their declines before the 2001 recession:

By contrast, the 2008-09 recession started off as a consumer-led downturn, where durable goods orders gave no advance warning.

This is important, because as the US in the past 40 years offshored most of its base manufacturing, that sector has had increasingly less impact on an economy that is now 70% consumer-driven.

The best foretaste of consumption is real retail sales. But the broadest measure is real private consumption expenditures, particularly for services, which will be reported in two days. I intend to post a heads-up on what to look for in that report later today.

Tuesday, April 25, 2023


 - by New Deal democrat

For the past few months I have speculated that home sales were bottoming. This morning’s report on March new home sales put an exclamation mark on that idea.

New home sales increased 57,000 in March (from a February level downwardly revised by -17,000) to 683,000 annualized (blue in the graph below). The increasing trend in sales from the bottom of 543,000 last July at this point seems crystal clear. As I have said many times, new home sales are very noisy, and very heavily revised, but frequently turn first. For confirmation, I use single family permits, which have very little noise and usually clear trends (red). And they are almost certainly confirming the trend from new home sales:

Since mortgage interest rates peaked last October, this is not surprising.

Meanwhile, just as we saw with the house price indexes earlier this morning, the median price of new homes increased slightly YoY for the second month in a row, now up +3.2% (gold, compared with the YoY% changes in new home sales, blue):

As is usual, prices have followed sales with a significant lag.

This is good news for the economy in 2024, as it tends to put a floor under any downturn later this year, suggesting that if there is a recession, it will be relatively brief and shallow (Fed permitting).

House prices on track to go negative YoY by summer, despite monthly increase in February


 - by New Deal democrat

House prices through February as measured by both the FHFA (gold in the graphs below) and Case Shiller (red) Indexes rose, the former by 0.5% (after a downwardly revised 0.1% in January), and the latter by 0.2% (after a -0.2% decline in January). Here’s what the monthly changes look like for each, as compared with Owners’ Equivalent Rent in the CPI (blue):

[Note that both house prices indexes are /2.5 for scale]. Since a year ago, both house price indexes were rising at almost 2% a month, the YoY% changes have continued to decelerate sharply:

The FHFA index is only up 4.0% YoY through February, while the Case Shiller is only up 2.1% YoY. At the rate of decline since last summer, the FHFA Index will be negative YoY by about June, and the Case Shiller Index could go negative YoY by next month’s report for March.

The implications for CPI is that the Owners’s Equivalent Rent component is likely to stabilize at current YoY levels for several more months before declinining, and CPI ex-shelter, which actually was slightly in *deflation* since last June as of the March report, will continue to be flat or lower.

Finally, while the increases in house prices have been quite small compared with the recent past, I was expecting a bigger decline from both indexes once the tide turned last summer. Undoubtedly the reason has a lot to do with the below graph, showing that while the active listing count of houses for sale has increased by over 50% since one year ago (teal), in absolute terms it is much lower than before the pandemic (blue); and indeed the new listing count has continued its almost relentless decline beginning 2 years ago, now down about -20% from a year ago:

The very low number of houses for sale puts a low ceiling on supply, meaning even normal demand can still create bidding wars.

That increasing interest rates is causing fewer houses to be put on the market, as potential move-up buyers do not want to trade 3% and 4% mortgages for 6% and 7% mortgages, creates quite a conundrum for the Fed.

Monday, April 24, 2023

Income tax withholding payments stumble again


 - by New Deal democrat

The important data this week will include new home sales tomorrow, Q1 GDP and initial jobless claims on Thursday, and most importantly of all (imo) real personal income and spending, along with real manufacturing and trade sales on Friday.

In the meantime, today let me take another look at a significant coincident indicator, income tax withholding payments, because the situation has changed in the past week.

Tax withholding payments have for years been employed as a proxy for jobs. Unfortunately, there’s no monthly or quarterly data published on FRED. The best representation is annual data from 1947 to 2020. Below I show the YoY% change in that annual data, adjusted for inflation, compared with the YoY% change (*3 for scale) in monthly nonfarm payrolls:

Because of a quirk in FRED graphing, it appears that jobs lag tax payments, but that’s just a byproduct of comparing monthly vs. annual data. Had I used annual payroll averages, the peaks and troughs would match exactly (but the jobs data would be less fine-grained). The bottom line is that, while the two haven’t matched exactly, especially in the 1980s, typically the increases and decreases move in tandem.

Turning to the present, last week I cited to the California Department of Revenue, showing that tax payments in that State had declined steeply compared with the prior fiscal year during the last four months of 2022, before stabilizing during the first three months of this year.

For the nation as a whole Matt Trivisonno has the YoY data, measuring the entire 365 day total of tax withholding vs. the entire previous 365 days, and has a public graph with a 3 month delay. Here’s his latest:

Like the California graph, it shows a steep deceleration during 2022, which had been as high as +21% YoY in March, down to only about +6% by the end of December. Thereafter through January, the YoY data stabilizes.

Indeed, by my own calculations, for the first three months of fiscal 2023 ending December 31, withholding tax payments were only up +1.2% YoY. But for Q2 they rebounded sharply, up +5.4% YoY. 

But in the last 10 days they have stumbled. For the first 14 withholding days in April, payments are down -3.4%, $189.7 Billion vs. $196.3 Billion one year ago. For the last 4 weeks as a whole, withholding payments are down -5.0%, $270.2 Billion vs. $284.5 Billion.

What is notable about that, in addition to including the April 18 deadline for payment of taxes this year, is that the CA Department of Revenue had suggested that the late 2022 stumble was due to stock market declines meaning that stock options hadn’t vested.

Well, since last October the stock market has rallied, and last week was very close to an 8 month high:

Only a short term shortfall at this point, and of course it could reverse by the end of the month, but if stock options are vesting and withholding payments are still down, even before accounting for inflation, that suggests renewed trouble in the jobs market.

Sunday, April 23, 2023

The last dissent of Thurgood Marshall: the Rule of Law vs. the transitory Edicts of 5-4 Court majorities


 - by New Deal democrat

Daniel Kiel at the TPM Cafe, on the supreme differences between Clarence Thomas and his predecessor, Thurgood Marshall, writes:

“Thurgood Marshall, … in his final opinion before retiring after a quarter century on the court, [ ]warned that his fellow justices’ growing appetite to revisit – and reverse – prior decisions would ultimately ‘squander the authority and legitimacy of this Court….’”

This criticism has never seemed more on point than in the aftermath of Dobbs, as Red State Legislatures and Trumpy lower court judges swing for the fences to invite the obliteration of existing precedents.

Marshall’s final dissent occurred in the case of Payne v. Tennessee, a case that involved the scope of victim impact statements and testimony in the sentencing portion of capital murder trials. A badly splintered Court in that case overruled two previous 5 to 4 rulings that were less than 10 years old to hold expansively in favor of the prosecution. The various plurality, concurring, and dissenting opinions all dealt extensively with the doctrine of stare decisis, which simply means that decisions that have already been made should be left in place.

Stare decisis was important to Hamilton’s rebuttal to Brutus in Federalist #78, the essay that famously claimed that the judiciary would be “the least dangerous branch.” He wrote that:

“To avoid an arbitrary discretion in the courts, it is indispensable that they should be bound down by strict rules and precedents, which serve to define and point out their duty in every particular case that comes before them; and it will readily be conceived from the variety of controversies which grow out of the folly and wickedness of mankind, that the records of those precedents must unavoidably swell to a very considerable bulk”

This was central to Hamilton’s argument. He believed that as time went on, the Supreme Court would be increasingly hemmed in by precedents, and thus unable to enact their ideological whims or prejudices. 

Well, we know how that has worked out, don’t we?

But back to Marshall’s last dissent. The crux of his argument is:

“the majority declares itself free to discard any principle of constitutional liberty which was recognized or reaffirmed over the dissenting votes of four Justices and with which five or more Justices now disagree. The implications of this radical new exception to the doctrine of stare decisis are staggering. The majority today sends a clear signal that scores of established constitutional liberties are now ripe for reconsideration, thereby inviting the very type of open defiance of our precedents that the majority rewards in this case….

“The overruling of one of this Court's precedents ought to be a matter of great moment and consequence. Although the doctrine of stare decisis is not an ‘inexorable command,’ [citation omitted] this Court has repeatedly stressed that fidelity to precedent is fundamental to ‘a society governed by the rule of law,’ [citations omitted] ‘[I]t is indisputable that stare decisis is a basic self-governing principle within the Judicial Branch ….’

“…. By limiting full protection of the doctrine of stare decisis to ‘cases involving property and contract rights,’ [ ] the majority sends a clear signal that essentially alldecisions implementing the personal liberties protected by the Bill of Rights and the Fourteenth Amendment are open to reexamination. Taking into account the majority's additional criterion for overruling -- that a case either was decided or reaffirmed by a 5-4 margin ’over spirited dissen[t],’ [ ] -- the continued vitality of literally scores of decisions must be understood to depend on nothing more than the proclivities of the individuals who now comprise a majority of this Court.”

To be fair, where the 5 to 4 rulings are less than a decade old, Scalia’s response in his concurrence seems a much more accurate point:

quite to the contrary, what would enshrine power as the governing principle of this Court is the notion that an important constitutional decision with plainly inadequate rational support must be left in place for the sole reason that it once attracted five votes.”

Point well taken. But then, Scalia goes completely off the rails:

“[S]tare decisis[ ], to the extent it rests upon anything more than administrative convenience, is merely the application to judicial precedents of a more general principle that the settled practices and expectations of a democratic society should generally not be disturbed by the courts.”

It strikes me that the expectations of a democratic society are a helluva lot bigger principle in play than mere “administrative convenience.”

But even worse, Marshall was exactly on point in his criticism of the plurality opinion by Rehnquist, for they did indeed say:

“Stare decisis is not an inexorable command; rather, it ’is a principle of policy and not a mechanical formula of adherence to the latest decision.’ [citation omitted] This is particularly true in constitutional cases, because in such cases ’correction through legislative action is practically impossible.’ [citation omitted]. Considerations in favor of stare decisis are at their acme in cases involving property and contract rights, where reliance interests are involved,”

Up until the last sentence, the majority is exactly correct. Constitutional decisions by the Supreme Court are almost impossible to reverse by democratic means. And as we have seen with the Fifteenth Amendment, even when those Herculean hurdles are cleared, a majority of the Court might simply elide them away, as Roberts did in Shelby County.

But seriously, the reliance of a democratic society on settled precedents of the Court is at its peak in *property or contract cases*??? How one drafts a contract or a title deed is more important than who one can marry, who one can be romantic with, what one can do with their own body??? This is simply breathtaking in its fundamental ignorance.

To wit: the American public should not have to draft new Constitutional Amendments and get them passed by 2/3’s of both Houses of Congress and 3/4’s of all States, in order to protect civil rights that have been upheld by Supreme Court decisions and been in effect for decades.

Simply put, the rule by an ever-shifting 5 to 4 majority on the Supreme Court is not by any means the Rule of Law. Marshall was spot on in his last dissent that Edicts by shifting majorities on the Supreme Court have indeed “squandered its authority and legitimacy.”