As expected, March consumer inflation packed a (possibly recessionary) wallop

 - by New Deal democrat

As anticipated, the March CPI packed a wallop, up 0.9% for the month and causing the YoY% gain to increase to 3.3%, while core CPI was a tame 0.2% with a YoY% gain of 2.6%. Because of the impact of that big number, I am departing from my usual format to focus on energy and shelter, but also the impact on real wages and incomes.

But first, and as I’ve written in the past few months, here is an IMPORTANT CAUTIONARY NOTE: Because the October-November kludge in shelter prices of a mere 0.1% increase for two months is still present in the YoY calculations, and will be until this coming November, this is probably continuing to lower those comparisons by roughly -0.2%. In other words, take out that kludge and YoY headline CPI would probably be 3.5%, and core at 2.8%.

First, as per my usual practice for the past several years, let’s start with the YoY numbers for headline inflation (blue), core inflation (red), and inflation ex shelter (gold), which because of the big impact of oil, surged to 3.4% YoY:

Before I go further, let’s note some good news: the very large shelter component of CPI has continued to undergo disinflation. Actual rent was only up 0.1% for the month, the lowest monthly increase since the end of 2020, and up 2.6% YoY, the lowest number since June 2021. Similarly, Owners Equivalent Rent increased only 0.2%, and was up 3.1% YoY. These are also the lowest numbers since late 2020 (except for last September) and late 2021, respectively (Note this is also substantially true even if we apply the counter to the government shutdown kludge):

Further, as I have been writing since way back in 2021 is that the YoY% changes in the repeat home sales indexes lead shelter CPI by about 12-18 months. It did that on the way up, and it has been doing that on the way down. YoY home price increases continue near or at multi-year lows, the FHFA at 1.6%, and Case Shiller’s national index at 0.9%. And shelter inflation has followed, as shown in the graph below:

Shelter inflation has declined to *below* its pre-pandemic YoY range. Needless to say, because of the leading/lagging relationship of house prices to shelter inflation, we can expect even *further* deceleration in the shelter component of inflation during this year.

Let me put this succinctly: if nobody had started a war with Iran, this morning’s report would have been very good news, with the biggest problem child of the past 5 years, shelter, going into hibernation.

But, alas, that is not our world. Here is an update of a graph I put up several weeks ago, showing my K.I.S.S. estimate based on the increase in gas prices (blue), vs. the actual CPI number (red):

Because gas prices started out so low as a share of consumer spending, the increase was not nearly as bad as it could have been; but also, as I pointed out in that analysis several weeks ago, it is common for the effects of an oil shock to show up in the subsequent month (in this case, April) as well. 

But now let’s turn to the effect of this oil shock on household finances.

In last week’s jobs report, average hourly wages for nonsupervisory workers increased 0.2%. Now that we know what March inflation was, that means that real wages declined -0.7% in March:

This puts them at their lowest level since last April, and they are now only up 0.1% YoY. As the long term view shows, with the exception of the 1980s and early 1990s, when the massive entry of Boomers and women into the labor force acted to depress wage gains, this often is a harbinger of near term recession:

And the bad news doesn’t stop there. Remember that one of my big forecasting tools is real aggregate nonsupervisory payrolls, showing the amount of $$$ in total that average households have to spend. In March, nominally aggregate nonsupervisory payrolls rose 0.3%. With the inflation adjustment, real aggregate nonsupervisory payrolls declined -0.6%, and are -0.7% below their January peak:

This puts us back to just above where we were last September. Further, as I wrote in the past several weeks, a -0.7% decline from peak in this metric is about the median decline at the onset of past recessions.

Finally, on a YoY% basis, real aggregate nonsupervisory payrolls are only up 0.8%, the lowest since the pandemic:

Historically, outside of the 2002 near-“double-dip”, such a low YoY gain has only a few times outside of recessions, particularly in the case of 1979-80, where it occurred just a few months before, and also several isolated one month downturns (note: graph below adds 0.7% to value so that a 0.8% increase shows at the 0 line):

As suspected before the official numbers, needless to say this was a very poor report. The good news of disinflating shelter costs has been squandered, at least for March. Further, this is a very real loss in average consumers’ finances. By at least one important measure, it is at very least near-recessionary if not outright recessionary. The only silver lining is that this is just one month. If we get lucky and there is not another big increase in April, and further the situation in the Middle East is allowed to settle down sufficiently so that gas prices actually declined somewhat in the next several months, we might escape without more seriously negative consequences.

Very low jobless claims continue

 

 - by New Deal democrat

Jobless claims, along with stock market prices and upscale consumer spending (and recently, manufacturing orders) are one of the few important metrics holding up the economy. And the short summary of this morning’s data is: the new regime of very low claims continued.

Initial claims did rise 16,000 to a still very low 219,000, and the four week moving average increased 1,500 to 209,500. With the typical one week delay, continuing claims fell sharply to 1.794 million, the lowest number in two years:

On the YoY% basis more important for forecasting purposes, initial claims were down -1.8%, the four week moving average down -6.1%, and continuing claims down -3.1%:

This is very positive - in fact, jobless claims are probably the single most positive data point in the entire economy right now. If you have a job, there is simply very little chance you are getting laid off.

Finally, let’s take an updated view of how that played into the unemployment rate in last week’s jobs report for March. As anticipated, the unemployment rate declined, by -0.1%:

Jobless claims are forecasting further declines in the unemployment rate in the next several months.

Recessionary signals in February personal income and spending, but some bright spots as well

 

 - by New Deal democrat

Personal income and spending are among the most important monthly indicators of all, because they give us a detailed look at consumption by the broad range of American households. And since consumption leads employment, they also give us an idea of what is likely to happen with regard to jobs in the near future. This morning’s data was for February, so it is still several weeks later than usual. Keep in mind that this represents activity before the Iran war and its oil price shock. 

In February, nominally personal spending rose 0.5%, but personal income declined even nominally, by -0.1%. Since the PCE deflator increased 0.4%, real spending was only higher by 0.1%, while real income declined, after rounding, by -0.4%. Here is what they look like since the pandemic:

Note that real personal income has been flattening for almost a year, and February’s number was the lowest since last June.

Further, on a YoY% basis, both real income and spending have been decelerating since late 2024, income by more than spending:

Once we exclude government transfers — one of the important coincident metrics used by the NBER to date recessions, real personal income also declined -0.4% to the lowest level since last July:

On a YoY basis, this was up 0.5%. With only three exceptions —  2013 (which was an artifact of a change in Social Security withholding), 2022, and one month in 1995 — such a low rate has only happened during recessions:

In 2022, the economy was saved by the hurricane strength tailwind of deflating producer prices, which enabled a continued Boom in consumer spending. Needless to say, with the war with Iran that is not going to happen now. But despite the 0.4% increase in PCE prices for the month, the YoY% change remained at 2.8%:

Whether this suggests that the slowly increasing YoY trend over the pat 12 months will continue, or is flattening, is not at all clear.

Another important component of the data is spending on goods, and in particular durable goods, which is a leading indicator. Historically, the pattern has been that real spending on goods (red in the graph below) turns down in advance of recessions, and in particular spending on durable goods (orange), which tends to turn down first. Real spending on services (blue) has tended to rise even during all but the most prolonged or deep recessions. These have been flashing red warning signals. 

In February, there was something of a rebound. Real spending on services rose only 0.1%, but real spending on goods rose 0.2%, and on durable goods rose 09%. But neither of the latter two measures fully reversed their January declines. The below graph shows the post-pandemic record, normed to 100 as of January of last year:

The trend in goods spending was flat all last year. To smooth out some of the noise, I have been tracking the three month average. That average was almost completely flat in the period from July through December, peaking in November, and as of February is at a seven month low.

The updating of the PCE deflator also allows for an update to another important coincident indicator used by the NBER to consider whether the economy is in recession or not; namely, real manufacturing and trade sales, which is delayed by one additional month. These increased 0.4% in January to a new all-time high:

This is of a piece with the recent rebound in manufacturing data we have seen in things like durable and capital goods orders as well as manufacturing production. 

Finally, the personal saving rate - i.e., the portion of income left over after spending, declined in February back to 4.0% from its 4.5% reading in January:

Paradoxically, this is relatively good news, because a sharp retrenching by consumers is also something that typically happens just before the start of a recession. It looked like one might have been beginning in January, so February’s reversal means that consumers were a little more confident.

When we put all this together, we get a mixed picture. Real income has been stalling out, and declined in February, both before and after taking government transfer payments into account. So have real spending on goods and in particular durable goods. All of these are recessionary or near-recessionary. But consumers did not retrench, and the rebound in manufacturing sales continued in January.

But to reiterate, all of this predated the oil price spike in March. We will find out just how much that impacted consumers with tomorrow’s CPI report.

The consequences of mafia style bust-outs and military belligerence for interest rates

 

 - by New Deal democrat

While there isn’t any big economic news today, there certainly was action overnight in response to the latest TACO. As I type this, oil is back down to $90/barrel, and stock futures are soaring. This for something (correctly I think) framed as a “fragile cease fire” by J.D. Vance.

But let’s look at some of the economic damage that is likely to persist.

In the first place, the mafia-style bust out that is the ballooning US budget deficit has definitely put an end to the 40 year downdraft in Treasury yields. The below graph shows yields on the 30 year (dark blue) and 10 year (light blue) Treasurys as well as the Fed Funds rate (red):

Notice that the 10 year bond is about equal in yield to what it was during 2023-24 when the Fed funds rate was at its peak. And it did not react at all to the last two Fed rate cuts. The record of the 30 year is even worse, as yields have acutally trended higher even as the Fed funds rate has been cut. This is all about the “bond vigilantes” waking up and demanding more interest to hold on to bonds from a government that at the moment appears to think it can issue infinite amounts of paper. This can be laid squarely at things like the “Big Beautiful Bill” as well as the astronomical military build-up.

The increase in yields has also hit mortgage rates, which typically follow longer dated Treasurys. As of one week ago, they had risen about .5% to about 6.5%:

And with a several week delay, mortgage applications responded. They have been trending down for several weeks, and this morning’s update showed both purchase mortgage applications (blue) and refinance applications (gray) lower YoY:

Here is a five year view of the same data:

showing that, while the increase in mortgage rates has not knocked either type of application down to their 2023 nadirs, but has effectively halted the rebound.

Tomorrow we will get personal income and spending for February, and on Friday we will get the March CPI. Both will be important, and the latter is likely to be absolutely lit!

March ISM reports show stagflationary expansion — light on the “stag-,” heavy on the “-flation”

 

 - by New Deal democrat

 

As I’ve previously noted a number of times, one of the more surprising developments in the past few months has been the resilience of manufacturing. After taking a beating following “Liberation Day” one year ago, companies adapted and resumed production if anything at an even more brisk pace.

That was apparent as recently as the preliminary data on new factory orders released this morning for February. While overall new orders for durable goods declined -1.4% for the month (blue, right scale), core capital goods orders rose 0.6% to a new post-pandemic record (red, left scale):

On a YoY basis, headline new orders were up 7.3%, while capital goods orders were up 5.1%, continuing the last six months’ trend of the best YoY growth since the beginning of 2023:

A similar, and more complex, story was told by the ISM manufacturing and services indexes for March. The headline number for services declined to 53.9, still a good showing (recall that any number above 50 indicates expansion), and for manufacturing came in at 52.7, the best number since the summer of 2022 (in the graphs below, the services number is in blue, the manufacturing number in gray):

And the more leading new orders subindexes showed even more strength, with services coming in at a very strong 60.6, the highest reading in three years, while manufacturing new orders declined to a still expansionary 53.5:

For forecasting purposes, I use the three month average of the series, with a 25% weighting to manufacturing and 75% to services. The weighted average of both the headline and new orders components are the strongest in three years.

If the present and leading conditions are without doubt positive, what about the stagflationary scenario?

Well, the prices paid components both came in sharply strong, with services at 78.2, and manufacturing even slightly higher at 78.3, both the highest since June 2022:

If both the goods producing and services providing sectors of the economy were being clobbered by inflation in March, the picture for employment was considerably weaker. While the “less bad” trend in manufacturing employment continued, with a slightly contractionary 48.7, still its second best reading in the past 12 months after January’s, the employment subindex in services declined sharply to 45.2, its worst reading since the pandemic except for December 2023:

This is somewhat foreboding for the official employment metrics for the next several months. According to Jill Coronado of the University of Texas at Austin, “the ISM non-manufacturing employment index, particularly the three month average has some significant predictive power.” Here is her accompanying graph:

The three month average of 49.1 isn’t as low as it was last summer, but nevertheless predicts slight contraction, particularly of services providing employment.

To summarize, on the bright side, left to their own devices the manufacturing and services data indicate not just continued expansion, but even more robust expansion. But it is a stagflationary expansion, with simultaneously moribund employment and widespread price increases.

And of course, neither have been left to their own devices. Even the March data only marginally reflects the impacts of the Iran war. Those are likely to show up much more drastically in the April and May reports. To put it another way, “Buckle your seatbelt, Dorothy, ‘cause Kansas is going bye-bye.”

The “real,” wage adjusted price of gas isn’t at privation levels yet

 

 - by New Deal democrat

Back in the “before” days, as in January, before the Iran war, I wrote about how low gas prices were actually a tailwind for the economy. Because since the start of the Millennium over 25 years ago, they had only been so low compared with average hourly wages on only 3 occasions: after the 2001 recession, late in the Great Recession, and during the COVID lockdowns. Put another way, it only took about 7 minutes of work to buy a gallon of gas. This leaves a lot left over for other consumption - just as it did at the end of the two non-COVID recessions.

Needless to say, that has changed. But by how much, really? On the one hand, as I’ve pointed out previously, on a percentage basis this is the biggest one-month spike in gas prices since the 1970s. We’ll find out just how badly that effected the CPI for March this coming Friday.

But how much of the “tailwind” has been taken away? That’s what the updated graph below, of the “real” cost of gas compared with average hourly nonsupervisory wages, shows:

The “size” of the spike is about equal to the 2005 Katrina spike, and less that the 2022 Ukraine invasion spike. But in relative terms, it has not come anywhere close to the 2008 spike that helped exacerbate the Great Recession, nor the Ukraine invasion spike. Nor, for that matter, what I used to call the “Oil Choke Collar” of the early 2010’s, when gas prices put a lid on the velocity of any expansion in the early years of the last recovery. In order to approach the level of those shocks, we would need to see gas prices of $5/gallon, at minimum.

The gas price information doesn’t go all the way back to the 1970’s, but the price of oil, specifically West Texas Crude, does. So here is the same graph, of oil prices relative to average hourly nonsupervisory wages, going all the way back to before the first oil shock:

Here you can see that just before the start of the Iran war, the “real” price of oil was equivalent to the levels it was at from 1986-99, when gas prices were not a consumer issue at all. The current spike has not taken us back up to the levels of either the first Gulf War spike of 1990 nor the second oil shock of 1979-80.

The bottom line here is that, although this price spike is enough to marginally change consumer behavior, it isn’t yet at the point where in the past it has created a sense of real privation (in 1974 the spike was accompanied by an embargo that resulted in gas rationing). That isn’t to say it couldn’t get there in another month or two. Although I won’t bother with a graph, according to GasBuddy the national average has risen as much as another $0.12 in April up to $4.11. To reiterate, my sense is that a real sense of privation isn’t likely to kick in unless gas prices reach $5/gallon.

Weekly Indicators for March 30 - April 3 at Seeking Alpha

 

 - by New Deal democrat

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

After zigging upward last week, interest rates zagged downward - but not as much - this week, enough to change the ratings on some interest rate sensitive indicators, like mortgages. And consumers continue to spend, despite all the shocks and sluggishness in things like the labor market in the past 15 months.

As usual, clicking over and reading will bring you up to the virtual moment as to the state of the economy, and give me a little extra jingle in my pocket next time I go to the local bookstore.

March jobs report: the birds that came home to roost play an April Fool’s joke, shrieking “Nevermind!”

 

 - by New Deal democrat

I described two months ago as “the month the birds came home to roost.” Last month, pace Edgar Allen Poe, I said the birds were screeching “recession!”

This month, Poe’s birds decided to play with us, screeching instead: “Nevermind!”

This was a good report with mainly good internals, with one large exception.

Below is my in depth synopsis.

HEADLINES:

• 178,000 jobs gained, the biggest number since December 2024. Private sector jobs increased 186,000, while government jobs declined -8,000. The three month average rose from a puny +6,000 to +68,000.
• The pattern of downward revisions to previous months did continue. While January was revised upward by +34,000, February was revised downward by -41,000, for a net decline of -7,000. 
• The alternate, and more volatile measure in the household report, declined by -64,000 jobs. On a YoY basis, this series DECLINED for the second month in a row, by -561,000 jobs, or an average of -47,000 monthly.
• The U3 unemployment rate fell -0.1% to 4.3%. 
• The U6 underemployment rate rose +0.1% to 8.0%.
• Further out on the spectrum, those who are not in the labor force but want a job now rose by +66,000.

Leading employment indicators of a slowdown or recession

These are leading sectors for the economy overall, and help us gauge how much the post-pandemic employment boom is shading towards a downturn. These were mainly positive:

• The average manufacturing workweek, one of the 10 components of the Index of Leading Indicators, fell -0.1 hour to 41.4hours, but still is now down only -0.2 hour from its 2021 peak of 41.6 hours.

• Manufacturing jobs rose +15,000, only the second increase in the last 12 months.
• Truck driving declined another -800.
• Construction jobs rose +26,000.
• Residential construction jobs, which are even more leading, rose +3,100, continuing the trend of stabilizing since last April.
• Goods producing jobs as a whole rose +43,000.. 

• Temporary jobs, which have declined by over -650,000 since late 2022, declined again this month, by -4,400, but remained above their post-pandemic low set last October.
• The number of people unemployed for 5 weeks or fewer declined -181,000.

Wages of non-managerial workers 

• Average Hourly Earnings for Production and Nonsupervisory Personnel increased $.05, or +0.2%, to $32.07, for a YoY gain of +3.4%, its lowest YoY% gain since the pandemic. While this remains higher than the YoY inflation rate through February, even that is among the lowest gains in the past three years — and it is very much likely to change once March’s CPI is reported.

Aggregate hours and wages: 

• The index of aggregate hours worked for non-managerial workers increased +0.2%, and is up only 0.7% YoY, below average for the past two years.

• The index of aggregate payrolls for non-managerial workers rose 0.3%, and is up 4.1% YoY, close to its post-pandemic low of 4.0% set last June.

Other significant data:

• Professional and business employment (for a change!) rose +2,000. These tend to be well-paying jobs. This remains above its October low, it remains lower YoY by -0.4%, which in the past 80+ years - until now - has almost *always* meant recession.
• The employment population ratio declined -0.1% to 59.2%, vs. 61.1% in February 2020, and its lowest since October 2020.
• The Labor Force Participation Rate declined -0.1% to 61.9% , vs. 63.4% in February 2020, and its lowest since November 2020.

SUMMARY

As I wrote at the opening above, this was a good report, but with a few significant negatives. 

Let’s start with the good, which obviously include both the headline number and the decline in the unemployment rate and short term unemployed, as has been telegraphed by extremely low initial jobless claims. Goods producing jobs increased, including manufacturing, construction, and residential construction jobs. Professional and business jobs had a positive month, for a change. 

There were some negatives, including a decline in the manufacturing work week, EPOP and LFPR. Truck driving jobs continued to decline. And the underemployment rate rose slightly. 

But the most significant negatives had to do with wages. The increase in hourly nonsupervisory wages was among the lowest since the pandemic, and the YoY% change was the lowest. Aggregate hours for nonsupervisory also had a relatively small gain. Which means that, even nominally, the gain in aggregate nonsupervisory payrolls was close to its post-pandemic low. Consumer prices last March were unchanged. If the Cleveland Fed’s estimate of a 0.8% gain this March is accurate, that will mean March CPI will come in a 3.2% YoY. The estimated *real* gain in YoY nonsupervisory payrolls would only be 0.9%, the lowest since the pandemic, and a major cause for concern.

So it is very possible that this rosy-looking outlook could change by the end of next week, but for today the birds that came home to roost have played an April Fool’s joke: “Nevermind!”

Jobless claims continue near historic lows; I expect the unemployment rate to decline

 - by New Deal democrat

 With the stock market flailing around trying to keep its head above water, jobless claims along with consumer spending are the only two metrics that solidly support a continued economic expansion (ok, maybe ISM manufacturing is trending in that way as well).

But to the point of this post: last week initial jobless claims declined -9.000 to 202,000 — again, near historic 50+ year lows. The four week moving average declilned -3,000 to 207,750. Meanwhile, with the typical one week delay, continuing claims rose 25,000 to 1.841 million, still significantly below the 1.900+ we were seeing for most of last year:

[NOTE: For some reason FRED has not gotten around to posting these this morning, so here is the equivalent graph from TradingEconomics.com]:

As per usual, the YoY% changes are more important for forecasting purposes. So measured, initial claims were lower by -9.4%, the four week average down -6.8%, and continuing claims down -1.9% [Since TradingEconomics doesn’t have the YoY comparisons, you’ll have to imagine this until FRED gets around to it].

These are very good comparisons. While most of the data has been very weak, it is just very hard to imagine an economic downturn occurring with for all intents and purposes no layoffs.

Finally, since tomorrow is jobs day, and jobless claims lead the unemployment rate, here is our final look for the month. First, here are the 4 week average of initial claims (blue) vs. the unemployment rate (noisier but more leading):

And here are continuing claims (blue) vs. the unemployment rate (much less noisy albeit less leading):

I expect the unemployment rate to decline, or at very least not increase tomorrow. We’ll see then.

March ISM manufacturing shows expansion, but at an inflationary price

 

 - by New Deal democrat

While much of the official government data is still delayed, months after the end of the shutdown, privately sourced data remains fully up to date.

And March data started out with the ISM manufacturing index, which was our second piece of (mainly) good news of the morning.

The headline ISM number (blue in the graph below) rose 0.3 to 52.7 (recall that any number above 50.0 indicates expansion). The more leading new orders subindex (gray) did decline 2.3 from 55.8 to 53.5, but obviously that is also still positive. The three month averages, which smooth out a little volatility, improved to 52.6 and 55.5:

One of the surprises since last autumn has been the rebound in manufacturing despite the tariff situation, even though much of it is likely due to AI data center construction. 

Goods production is only about 25% of the US economy, so normally I weight it against the comparable services numbers, but this month there is really no need, since services have been above 50, indicating expansion, consistently since late last summer.

The bottom line is, this number suggests continuing expansion in the next few months - although I feel compelled to add that I doubt much of the impact of higher fuel costs and associated interruptions from the Iran war debacle has made it through into the index numbers yet.

Several other components of the index are worth noting this month as well.

First, the “less bad” trend in employment continued in March, as it declined very slightly, -0.1, to 48.8. But all three months so far this year have been significantly better than the dismal readings that began last February:

There was one important negative in the report, however - prices paid. These shot up to 78.3, the highest number since June of 2022:

This continues the sharp inflationary pulse that started in February. 

So, while the ISM manufacturing index indicates expansion, it is an inflationary expansion, which is going to put continued upward pressure on interest rates, and tend to keep the Fed on the sidelines in terms of any hope of further rate cuts in the immediate future.

Some good news for a change: real retail sales rebounded in Febuary

 

 - by New Deal democrat

After all these months, we are still feeling the effects of the government shutdown last fall.  Normally construction spending is released on the first day of the month for the second previous month - in today’s case, that would be for February. But half a year after the shutdown began, February and March construction spending are both scheduled to be released on May 7. As I’ve said a number of times already, this is simply not the way a first world country should operate.

But in today’s case, we at least get a consolation prize in the form of retail sales, one of my favorite broad-economy indicators, for February - about three weeks later than normally scheduled. And for a change compared with most recent data, it was good news.

Nominally, retail sales rose 0.6% in February. After taking the monthly 0.3% increase in consumer prices into account, real sales up 0.3%. The below graph also shows the similar but more comprehensive measure of real personal spending on goods (gold, right scale):

Even so, real retail sales remain -0.4% below their peak last August, and indeed below most of their levels from last year. Further, if you believe, as I do, that the shutdown shelter kludge removed about 0.2% from consumer inflation during the September-November period, then the comparison becomes similarly worse. 

February’s good number also means that on a YoY% basis, after a one month flirtation with trending negative, real retail sales have rebounded to +1.3%:

Since consumption leads employment, this is also good news for the latter in the next few months, after deteriorating through most of 2025. Here is the update of YoY real sales and real personal spending on goods (/2 for scale) together with employment (red):

But most likely the deterioration in spending last year has not been fully absorbed by employment yet. This morning ADP reported that private payrolls only grew by 18,000 in March. We’ll find out on Friday if a similarly poor number is true in the official jobs report.