Oil shocks and real aggregate nonsupervisory payrolls

 

 - by New Deal democrat

As readers know well, one of my favorite “real life” indicators is real aggregate nonsupervisory payrolls, which measures how much in wages average American workers have to spend each month. When it is growing, economic expansions almost always continue; when it declines by any significant amount, recessions almost always ensue shortly.

With gas prices going from $3 to $4 a gallon in March, how is it likely to be impacted? Let’s take a look at a current estimate as well as some history.

Although I’ve done my own K.I.S.S. estimate, the folks at the Cleveland Fed take a much more detailed approach, and publish nowcasts monthly. As of last Friday, they were estimating that March headline CPI would increase 0.8%:

Although Friday is a religious holiday, and markets will be closed, the jobs report for March is scheduled to be released as usual. Although we obviously don’t have the actual figure yet, what we can say is that for the last three years, nominally aggregate payrolls have increased an average of a little under 0.4%; for the year 2025, it slowed to 0.3%:

In other words, if payrolls increased in March by the same percent they have averaged over the past year, real aggregate payrolls are likely to decline about -0.5%. As the below graph, which norms real aggregate payrolls to their peak in January, shows, that would take us back down to just above August and September levels, since February already saw a decline of -0.2%:

Per my previous analysis, that wouldn’t necessarily be enough to flag recession on its own, but it would be in the ballpark of the average such decline until the onset of recessions — and remember that the shelter kludge of CPI during fall’s shutdown suggests that CPI should have been about 0.2% higher, meaning that in *real* real terms average Americans might have a little less to spend than they did last summer.

Is that supported by the historical data? Well, let’s take a look at what has happened to real aggregate nonsupervisory payrolls in past oil shocks. Note that because the official gas price data didn’t begin until late 1991, I’m using spot oil prices for West Texas crude oil (/10 for scale) in the below graphs.

In the 1974 oil embargo, in January oil prices increased 125% from $4 to $10 a barrel (blue). Real payrolls (red) declined -1.1%:

In the second OPEC oil shock of 1979, in August oil prices increased 21.8% from $21.80/barrel to $26.50. Real payrolls declined -0.3%:

OPEC’s pricing power collapsed in 1986, but with Iraq’s invasion of Kuwait in August 1990, oil prices increased 45.8% from $18.60/barrel to $27.20. Real payrolls declined -0.5%:

The final graph below covers 3 separate events. Gas prices bottomed at the end of 1998. In March 1999, oil prices rose 22.1% from $12/gallon to $14.70. Although not shown, gas prices rose 19.3% from $0.90/gallon to $1.08. Real aggregate payrolls declined -0.3%.

When Katrina hit at the end of August 2005, over the two month period till the end of September, oil prices increased 10.7% from $58.70/barrel to $65,00. Gas prices rose from $2.29/gallon to $2.80. Real aggregate payrolls declined -0.3% in August and another -0.8% in September.

Finally, oil prices rose in March 2009 from their Great Recession bottom by 22.5% from $39.20/barrel to $48.00. Gas prices rose 7.2% from $1.91/gallon to $2.44 by the end of April. Real aggregate payrolls declined -0.9%:

This month oil prices started out at about $64.50/barrel. They are likely to end the month at about $100, a 55% increase. Gas prices are likely to be higher by about 35%.  This is about equivalent to the Kuwait invasion oil shock, and second only to the 1974 embargo. In the former, real aggregate payrolls declined -0.5%, and in the latter -1.1%. So the estimation of -0.5% in real aggregate payrolls based on the Cleveland Fed’s nowcast for March headline inflation appears likely, and if anything somewhat conservative.

Weekly Indicators for March 23 - 27 at Seeking Alpha

 

 - by New Deal democrat

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

Unsurprisingly, the main issue is the spreading disruption from the spike in oil prices as well as the bond market’s selloff due to heightened inflation concerns. As I wrote here yesterday, the latest victim is housing, which is being strangled by increased mortgage rates.

As usual, clicking over and reading will bring you up to date on the incipient carnage, and reward me a little bit for gathering and collating the information for you.

“Trump take housing:” how the Iran war is killing the housing sector’s “green shoots”

 

 - by New Deal democrat

“Trump take egg” was a social media meme popularized by MTSW at Bluesky, highlighting prices that T—-p had promised would come down, but increased instead.

The spike in egg prices was due to avian flu, but when it comes to internet memes, nevermind.

Which is by way of introduction to saying that the economic damage done by the Iran war is spreading out.

At the end of February, mortgage rates hit a 3.5 year low at 5.98%. As of yesterday, according to Mortgage News Daily, they made a new 7 month high at 6.62%.

 

I have been writing for nearly a year that the housing market was recessionary, and that the last dominos were falling. But recessions do end and turn into recoveries, and in the last few months there have been signs of “green shoots” in things like mortgage applications that suggested that left to its own devices, any such recession would likely be short. Well, the Iran war is in the process of killing those green shoots.

First, let’s take a look at 10 and 30 year Treasury interest rates (dark and light blue, right scale) vs. mortgage rates (red, left scale) since the Fed first started raising rates 4 years ago:

Note that in the first year, mortgage rates reacted more strongly to the change in the interest rate climate than did long term Treasurys. In the past two years, they gave back that premium, as mortgage rates declined from a high of about 7.80% to the aforementioned 5.98%. Now here is the short term look to emphasize the reversal in the past 4 weeks:

Note that the mortgage rate in the above graph is weekly, and does not include the further increase in the last few days, which would take us all the way back to last August.

And the increase in mortgage rates has already had an effect on new purchase mortgage applications (blue, left scale) in the graphs below) as well as refinancing (gray, right scale). Here’s the longer term look, showing how that after almost completely drying up in 2023-24, mortgage applications generally rose during 2025 and into this year, in response to lower mortgage interest rates:

Now here is the close-up of the past year:

Note that both types of applications fell sharply in the past several weeks - and the two graphs above end as of one week ago.

Refinancing is particularly sensitive to mortgage rates. As the below graph shows, it is virtually a mirror image:

Again, this graph does not include the further increase in mortgage rates this week.

As a result, we can expect both purchase and refinancing mortgage rates to decline further. This is almost certainly going to put an end to the “green shoots” that were beginning to appear in the housing data. And if those higher rates persist, if there is a recession this year, it is only going to be made deeper and longer.

New and continuing jobless claims remain near historic lows

 

 - by New Deal democrat

Along with the weekly update of retail sals, new jobless claims continue to be the most positive economic data in the entire specturm.

Last week new cliams increased 5,000 to 210,000, which is about average for the entire post-pandemic period. The only other time they were this low was 2018-19, and before that, the 1960s! The four week moving average declined -250 to 210,500. With the typical one week delay, continuing claims declined -32,000 to 1.819 million, the lowest number since June 2024:

On the YoY basis more important for forecasting purposes, initial claims were down -6.2%, the four week average down -5.9%, and continuing claims down 1.8%:

This week let me include the long term historical look at how initial claims lead the *number* of unemployed (red in the graph below), and to a much lesser extent, so do continuing + initial claims:

The same is true with respect to the unemployment rate:

With the significant drop in new and now continuing jobless claims since last November, the forecast is very much that the number of unemployed in the next several jobs reports is likely to decline:

Note that the number of unemployed peaked in September and November. Jobless claims forecast that this number will remain below those peak months.

And the same is true of the unemployment rate, even though it ticked up in the last jobs report:

The unemployment rate is likely to tick down to 4.2% or even 4.1%. The only complicating factor is whether the number of *employed*, as well as the number of unemployed, also declines.

I continue to suspect that, not only is there residual post-pandemic seasonality in the jobless claims numbers, but that the drying up of immigration as well as the ramping up of deportations in the past year has had a great deal to do with both the stalling of employment as well as the relative persistence of the unemployment rate.

Updating the K.I.S.S. estimate of the coming shock in CPI

 

 - by New Deal democrat

There’s no big economic data today, so let me update something I posted last week, in which I warned readers to expect a shock in the next CPI report. 

I wrote that “based on past history and using conservative assumptions, the model forecasts a 1.8% increase in CPI between March and April. Using normal assumptions it would forecast a 2.1% increase in these two months. And if I were to plug in today’s $3.92/gallon average vs. $3.72, the model would forecast a 2.5% increase in consumer prices by the end of April.”

Well, as of today’s weekly update from the E.I.A., gas prices as of the 20th were $3.96/gallon:

That would translate to an increase of 2.6% in consumer prices using my K.I.S.S. method of estimating the ballpark increase.

And according to GasBuddy, as of today, gas prices are right at $3.99:

which would translate into a 2.8% increase.

There is no way on earth wages would be able to keep up with that kind of shock.