There is Inflation, and then there is INFLATION!

How convenient: Just as a “top economist” for Moody’s says that GDP may not be as reliable as it once was for gauging when we go into recession, Trump’s new appointee to replace the director of the Bureau of Labor Statistics, whom he fired, says the best solution for the government’s poor reliability in reporting on the labor market is to stop doing monthly jobs report. That should help hide the mess, as it is just now starting to unfold.

I have to say something to the credit of both people, though. Mark Zandi, the Moody’s economist, did, at least, have the guts to say we are likely sliding into a recession soon—a limb economists always seem too cowardly to go out on. He is, of course, late since we already entered one in the first quarter of this year (and have been sliding into it for some time before that), but at least he is ahead of his colleagues. Had we seen a second quarter of declining GDP, he might have ventured a bolder statement, but tariff front-running may be jiggling us up into better numbers, which is, by nature, a briefly temporary anomaly.

[Economist Mark Zandi] delivered his dire warning about the state of the economy last week, and followed it up over the weekend with a series of posts on X about what he's eyeing to know if the economy has tipped into a downturn.

Notice the perfect tense used at the end there, meaning he sees recession as something that may have already become a fact, even though GDP didn’t show it in the second quarter. In other words, he’s not asking, IF it is tipping into recession, but if it HAS tipped already. How will we figure that out if GDP cannot be trusted in the present tariff environment?

Since Zandi suspects the downturn has already happened, he is looking to something other than GDP to figure out if that is the case. He looks to labor, which he says is a lagging indicator, but not as lagging as GDP, which could now even be called a “lacking indicator” due to the massive front-running of tariffs that hammered the throttle on the nation’s economic engines.

While the US is not yet in a technical recession—defined as two straight quarters of negative growth—other areas like the labor market have been sounding a warning.

Zandi … laid out a few key factors:

• Payroll employment

• Employment levels

• Job decline consistency

…"Given that the recent revisions to the jobs numbers have been consistently lower, much lower, it wouldn't be surprising if we learn with the coming revisions that employment is already declining," he said.

Unemployment usually makes its parabolic upturn exactly as the economy goes into recession, but it sometimes doesn’t make that turn until a few months after we enter recession, so it is a slightly lagging indicator. Therefore, if unemployment is already rising, we may already be in recession. Since Zandi even questions the current validity of the government’s unemployment rate, he looks to previous recessions to find the best employment metric to correlate with the onset of a recession and finds this:

For the second factor, Zandi pointed out that employment is falling throughout many industries. He said that previously, if half of the 400 industries being surveyed reported declining employment, an economic recession had begun.

Again, the perfect tense, meaning that, at that point, the recession has already become an established fact.

He cited recent data that revealed that in July 2025, more than 53% of industries reported job cuts, with only the healthcare sector showing employment growth.

By those measures, then, yes, the recession has already begun. With Zandi’s backing, I’m holding to all my earlier statements about that, particularly since I established almost two months in advance of the second-quarter GDP report that the tariffs may very likely create an anomaly that would make GDP an unreliable indicator. So, the rise in GDP back out of the negative came as no surprise.

I also noted in my Deeper Dive on the state of the economy that the one sector holding out in the labor market was the health-care industry and said that it is the last and least-likely to fall because, even when people hunker down on their purchases, they are likely to hunker down on health care last because it is essential (and, if they are still employed, because it is largely covered by the insurance they are getting)—sometimes more essential than food. So, jobs there fade slowly, as they also do in education and government.

"But, unemployment is a lagging indicator and given that the labor force has gone sideways this year as the number of foreign-born workers is declining, unemployment will be a particularly poor barometer of recession," he added.

People who are literally abducted from the fields by ICE don’t show up on unemployment numbers, though their removal may actually boost the number of new jobs being offered, offsetting other jobs that are going away due to the bad economy. So, this kind of event, which coincides with tariffs, also makes the usual unemployment rate a poor indicator right now. I think Zandi’s point is that, because the government’s unemployment rate may be seriously undercounting lost jobs, that is why he narrows in on the metric he gave for the percentage of businesses reducing employment as being a better metric than the government’s stated unemployment rate, that is not count anything lost to deportations.

In his previous post, Zandi said that Trump's policies, mainly tariffs and immigration, are the big drivers of his view that the economy is teetering at the edge of a downturn.

Just when we need to rely on jobs as the gauge for the economy, the job metric is being messed up, too, meaning we are really flying almost blind. As if that is not bad enough, it is so messed up that Trump’s new BLS-nominated replacement head is thinking we need to scrap the monthly measure for now. He did say some smart things, though, most notably gently disagreeing with his boss on the direction of the errors, even while agreeing with his boss's assessment that the Bureau of Labor Statistics does a poor job:

EJ Antoni, nominated by President Donald Trump Monday to become the next commissioner for the Bureau of Labor Statistics, suggested that the agency should suspend its monthly jobs report — one of the most crucial and historic measurements of US economic activity — claiming it is unreliable and frequently overstated.

And that was the qualification I had made, too, when Trump said the department head needed to be fired because of errors: yes, the old head should be fired for running such an unreliable operation, but it was not unreliable in the way Trump imagined (of understating job numbers to make him look bad); it was unreliable in frequently overstating job numbers and then having to revise them down (to make the bosses in each regime look good).

“Until it is corrected, the BLS should suspend issuing the monthly job reports but keep publishing the more accurate, though less timely, quarterly data,” he told Fox. “Major decision-makers from Wall Street to D.C. rely on these numbers, and a lack of confidence in the data has far-reaching consequences.”

It does, but we’ll be flying blind right when we most need to see the numbers in order, as Zandi just said, to ascertain if we are already in a recession, given that GDP right now cannot be relied on either!

Press Secretary Karoline Leavitt, however, said she believes the Trump admin plans to keep publishing the reports and that they “will be data that the American people can trust.”

Trump claimed, without evidence, the revisions constituted a “scam” and a vendetta against his presidency.

Antoni told Fox he disagrees with Trump, saying he doesn’t believe the BLS intentionally manipulated the jobs data. But Antoni has criticized the BLS’ approach to data collection, noting that revisions to monthly jobs reports have been significantly larger since the pandemic.

OK. He doesn’t sound too far off, but we’ll see how he is if he actually gets the job and has to kowtow to his boss.

“When we look at the jobs report in context; including the revisions, when we look at the industry mix; when we look at what’s going on with average hourly earnings and weekly hours and so forth, all of that is very helpful in gauging where the economy is going,” Faucher told CNN Tuesday in a phone interview. “I don’t think the fact that we’ve had revisions, and even if the ones we got in July were larger than average, I don’t think that’s justification for not releasing the data on a monthly basis.”

Apparently it is if you don’t like what the monthly data are showing.

Measuring economic activity, has grown more challenging in recent years, because of the pandemic’s seismic impact on global supply chains, US businesses and households and the lingering effects that still continue to this day.

Yes, everything is still jarred and working wonky because of those dumb lockdowns Trump authorized in 2020. We’re still a bit dizzy from it all.

Every year, the Bureau of Labor Statistics conducts an annual benchmark revision by making a thorough review of the survey-based employment estimates from the monthly jobs report and reconciling those estimates with fuller employment counts measured by the QCEW program.

And those, since the pandemic, have typically been massive downward jolts to the tune of half a million to a million fewer jobs than originally reported.

Economists, statisticians and policymakers have sounded the alarm bells for years, stating that federal data is in a precarious state, because of decreased funding, response rates and public trust.

Also, measuring economic activity, has grown more challenging in recent years, because of the pandemic’s seismic impact on global supply chains, US businesses and households and the lingering effects that still continue to this day. For example, the BLS had to implement several methodological changes to help construct its seasonal adjustment models, which smooth out cyclical fluctuations in the economy.

And typically, those “seasonal adjustments” seem to provide plenty of headroom for improving the raw numbers in the initial release to the advantage of the incumbent regime. I’ve rarely seen them make the numbers worse to any sizable degree. As Antoni said, the revisions to the original reports based on better data that comes out later are almost always downward. So, the BLS’s initial numbers regularly overestimate the economy’s strength … probably by design.

Trump, who criticized economic data when not in office, has continued to question the legitimacy of economic metrics during his second term.

(At that point, also claiming the BLS was overestimating, not that it was trying to make Biden look bad. He plays their incompetence in whichever direction is to his advantage at the time.)

Additionally, his administration not only has shut down several data-heavy federal programs and websites, but also made funding and employment cuts at statistical agencies.

At the BLS, the most clear-cut example has been recent reductions in the price collections that feed into the Consumer Price Index, the most widely used inflation gauge that also is used to adjust Americans’ payments (such as Social Security), set wages, and deflate other economic series to adjust for inflation.

Those reductions are likely to make the monthly CPI data more volatile and subject to revisions, economists say.

And, while government is likely bloated, I’ve been saying those DOGE changes are not likely to make the data any better. Cutting wood trim with a broad axe doesn’t typically result in finer finish work.

CPI overview

Speaking of that other flawed report put out by the BLS, it just arrived, and it showed that core inflation has gotten worse. As could be expected, the market makers who have stocks to pitch focused on the headline number, and the algorithms focus on all the headlines, so stocks soared as CNBC and others reported such things as “S&P 500, Nasdaq both notch record close as inflation report gives Fed green light to cut rates.”

That headline was practically a blatant lie, but I included it in the headline list just to discuss it. First, and foremost, the Fed does not pay much attention to CPI. The government does as it sets wages and particularly benefits. Therefore, employers in the business sector go by those numbers; but the Fed goes by the PCE Index (the Personal Consumption Expenditures Index, which is put out by the Commerce Department Dept., not the BLS (which has been under the Department of Labor, which was separated from Commerce in 1913), and that report won’t be out until August 29th.

Moreover, the Fed goes by core inflation, not the headline number. So, there is nothing here that would cause the Fed to cut rates; BUT that was the claim throughout the mainstream media this past week because that is what sets stocks on a tear, especially when those attentive AI-driven algorithms listen in. They especially listen to big people with big audiences who handle billions in managed funds, such as BlackRock, which also reported, surely knowing it is total baloney, “CPI gives Fed justification for a half-point cut in September.”

“It looks like a bit of Goldilocks right now for the stock market,” said Tom Hainlin, national investment strategist at U.S. Bank Asset Management Group. “More and more people are expecting a rate cut in September. So, rates kind of on a downward bias, earnings on an upward bias — that’s a pretty good environment for the broad stock market.”

Uh-huh, yeah, whatever. I’m sure Donald Trump still knows of a bridge in Manhattan he’ll sell you that even comes with a bankrupt casino, too. Believe in Goldilocks children’s stories if you want. Otherwise, let’s get real.

While we won’t have the gauge the Fed relies on until August 29th, one could still try to argue that headline CPI gives, at least, a hint of what the PCE will show near the end of the month, justifying the market’s exuberance, but does it? CPI and PCE have been in variance before. HOWEVER, even if CPI does give a hint of what is to come, the hint is NOT something that will make the Fed inclined to lower rates!

Having shared the above with everyone, this is where we go below the paywall to the privileged, deeper content for subscribers who are willing to contribute to my work, so I can focus on all of this with the huge amount of time I give to it. (It is still, dollars-per-hour, not the most profitable venture, but those paying subscriptions, at least, say some people appreciate all the effort enough to make it worth soldiering on, and maybe someday it will grow to where it makes solid financial sense too.)

*********