Monday

NAHB/Wells Fargo Housing Market Index

Homebuilders are sad. Very sad. The saddest they’ve been since early 2012, according to this confidence measure. More than 4 in 10 builders cut prices in October. It’s a pretty good time to buy a house (if you don’t have to finance it).

Tuesday

Pending Home Sales

Pending home sales, which hint at what actual home sales will look like in a month or two, fell to the lowest level ever recorded in September (the record goes back 25 years). This isn’t a mystery; nobody wants to finance a home with rates in the mid-7% range.

New Residential Construction

Housing starts fell for the third straight month in September and are now down about 7.5% over the past year. Permitting also slowed a bit. Yes, there’s been a sharp drop in new housing construction, but we’re still building at roughly the same pace as in 2019.

The number of housing units under construction held steady in September and appears to have stopped its rapid descent.

Oil Stuff

Gas prices fell slightly during the week ending October 19^th, while diesel prices rose modestly, but next week will be an ugly one on the fuel price front. Brent crude futures surged to a five-month high due to simmering tensions in the Middle East.

Electricity Prices Shockingly High

Electricity prices have risen more since the start of 2020 than they did over the 30 years before that.

We’re working on a longer post about what this means specifically as we head into winter and broadly as U.S. electricity demand rises for the first time in decades.

Wednesday

Treasury Yields

Read more

Why is this economics newsletter leading off with links to a flying car? Because it’s incredible, and that’s not a word we’d use to describe any of this week’s economic data.

Monday

Nobel Prize Goes to Hall and Woodford

Robert Hall and Michael Woodford won the 2026 Nobel Prize in economics for their research on the “microeconomic foundations of macroeconomic stabilization.” Put as simply as possible, they used models based on household and business decision-making to show how monetary and fiscal policy can stabilize the economy.

If you want to read more, see Marginal Revolution’s mid-size write-up or Noahpinion’s long form one.

Tuesday

Existing Home Sales

Existing home sales (i.e., not newly built) increased 1.5% in September, but that’s not saying much because August was such an abysmal month for the housing market.

One takeaway is that this is a great buyers market (provided you aren’t financing or can stomach rates). “Higher inventory and lengthening days on market give buyers improved negotiating power. Home sellers need to be realistic about local market conditions, which could require price discounts and other incentives to attract buyers.”

NFIB Small Business Optimism Index

This measure of small business owner confidence fell slightly in September. For the third straight month, owners reported labor quality and availability as their biggest issue, followed by inflation. Overall, not much of note here.

Treasury Yields

Treasury yields fell slightly this week, though they remain extremely elevated. This is a little surprising given the hotter-than-expected inflation releases (discussed at length below), which you’d expect (or we’d expect, anyway) to push yields higher.

So what happened? Anthropic, the maker of Claude, announced that their revenues only increased 45% in the third quarter, which was less than investors expected and at least slightly tempered optimism about the AI build out.

Oil Stuff

Both gas and diesel prices fell slightly this week but remain really high relative to pre-Iran conflict levels.

Wednesday

Consumer Price Index (Inflation)

Read more

Monday

ISM Services Index

The services side of the economy grew for the 27^th straight month in September, according to this survey of purchasing managers. Thirteen of the 17 sectors tracked by the survey expanded. Four didn’t, including construction.

As always, the most interesting part of this is the comments from respondents; tariff complaints are out, fuel price complaints are in.

Treasury Yields Rise Even Higher

Treasury yields bounced around this week and remain extremely elevated, and we’re now living in an economic environment that most Americans have never experienced. This is the first time anyone under 45 has lived through a period of rising interest rates.

Oil Stuff

Oil prices bounced around this week but are, all things considered, still pretty high. Even so, both gas and diesel prices retreated for the week, though they remain extremely elevated.

Tuesday

The Trade Balance

The U.S. trade deficit expanded about 14% in August as imports grew (almost entirely due to data center equipment) significantly faster than exports. Aside from early 2025, when importers were racing to beat the April 2025 tariffs, this is the largest the trade deficit has ever been.

Wednesday

Fed Meeting Minutes

Read more

Monday

Treasury Yields Rise Even Higher

Treasury yields kept surging this week, with the 10-year (a good proxy for borrowing costs) at least briefly surpassing 5.3% for the first time in over 20 years. Yields have edged back from that high, but that doesn’t really change the takeaway here: this is extremely bad news for rate-sensitive segments.

Oil Stuff

Oil prices fell this past week, and that’s provided a slight bit of relief at the pump. Diesel prices fell back to $6.38/gallon this week, which is still $2.63 more expensive than during the same week last year. Gas prices were virtually unchanged at $4.60 per gallon.

Tuesday

Conference Board Consumer Confidence Index

Consumer confidence plunged to a 12-year low in September. This one isn’t hard to understand: consumers really hate high gas prices.

Job Opening & Labor Turnover Survey (JOLTS)

Hiring sped up slightly in August, while the pace of layoffs slowed and the number of open, unfilled jobs declined. It remains a great time to have a job and a somewhat difficult time to need one.

S&P Cotality Case-Shiller Home Price Indices

Home prices increased pretty slowly in July and are up just 1.9% over the past year. While that’s pretty tepid growth, affordability is still awful due to mortgage rates (at least for anyone who has to finance).

Wednesday

PCE Price Index (Inflation)

Read more

U.S. employers added just 29,000 jobs in September, well short of the expected 90,000, and revisions brought the prior two months down by 60,000.

Despite the bad monthly report, we’re still averaging 51,000 new jobs over the past three months. That’s neither spectacular nor particularly worrisome.

The unemployment rate rose for the second straight month, but that’s hardly a cause for concern. Unemployment is still way down from the November 2025 cyclical peak.

Wage Growth Slows

Average hourly earnings increased just $0.05 in September and are now up just 3.0% over the past year. That’s the smallest year-over-year increase since early 2021. That’s good news on the inflation front but bad news on the consumer-health front.

The Best News in Today’s Report

The prime age (25-54) employment to population ratio—a pretty important indicator—rose to a healthy 80.7% in September. That’s higher than at any point from May 2001 to March 2023 and shows that a large share of working age people are, in fact, working.

Job Growth Remains Highly Concentrated

Healthcare led job growth again in September. No surprise there. A few other industries tacked on jobs for the month, including trade/transportation, leisure/hospitality, construction (thank data centers), and manufacturing, but it’s still all about healthcare.

Over the past year, healthcare has added more than 500,000 jobs, while all other industries have lost a combined 20,000 (though that does have a lot to do with federal government job losses).

What’s Going on with Information?

The information industry (think tech and media) continued to hemorrhage jobs in September, with employment now back to December 2020 levels.

Part of this is the ongoing, longstanding decline in traditional media. This century could be going a lot better for the publishing and broadcasting industries.

Since late 2025, though, we’ve seen a sudden and fairly sharp drop in tech employment. Is this AI replacing programming jobs? Ehhhh, it’s possible, but at least part of this—and maybe a lot of it—is tech companies trimming back their bloated staffs after the insane, zero-interest-rate fueled hiring binge of the early 2020s.

The Big Picture

This wasn’t a particularly strong jobs report, but it’s also not a cause for concern. Unemployment is low, we’re tacking on 50,000 new jobs a month, and the prime-age EPOP is just fine.

Before today’s jobs report the general consensus was that the labor market was running hot, driving up inflation. That’s maybe a little less of a concern now and, all else equal, takes a tiny bit of pressure off the Fed.

What’s Next

Anirban is currently working on Week in Review. That will be out later today and is just for paying subscribers. If that’s not you and you want it to be, just click the button below:

Subscribe now

Bad news: most of the data came from R-rated (bordering on X-rated) releases about the housing market, which just cannot catch a break.

Monday

Oil Stuff

Diesel prices jumped to another all-time high and, at $6.53/gallon, are just heinously elevated. Gas prices also continued to rise, up to a bruising $4.61/gallon. That’s over $1.30/gallon more expensive than at the same time last year.

Oil prices fell this past week, so we might see a little relief at the pumps next week, but fuel prices will remain pretty bleak.

TSA Checkpoint Travel Numbers

Air travel levels have improved slightly, according to TSA gate check numbers, but are still down about 2% compared to the same week last year.

Tuesday

Treasury Yields Rise Even Higher

Treasury yields kept surging this week, with the 10-year now closing in on 5.2%. The 10-year is a good proxy for borrowing costs, which is to say: it’s been a horrific week for rate-sensitive segments like housing.

Wednesday

Read more

Anirban’s Answers

With TSA numbers lower in 2026 than in 2025 and higher gas prices, how do you see domestic travel developing over the next 12 months?

I think air travel stands to be weak. Airfares have climbed rapidly, in part due to jet fuel prices, but also because of trimmed capacity. United Airlines slashed nearly 5 percent of its planned domestic seat capacity for a significant segment of the summer. That removed more than 23,500 domestic flights from July through early September.

My guess is that many families are shifting away from fly-to destinations to drive-to destinations. They save on airfare and by staying closer to home probably save on some other costs as well. We understand why this is happening, don’t we? Inflation remains high, job growth has become more erratic, and wage growth has softened.

“Aging in place” and a gray “tsunami” coupled with fast rising home prices are a big problem. What are the economics of ADUs and other similar measures? How can we provide for individuals, young couples, etc. to be aligned with an aging population for housing needs?

ADUs are accessory dwelling units. The number of units is rising across much of the nation (zoning/codes can frustrate development). Los Angeles County, for instance, reached a milestone with 10,000 ADUs constructed in 2025. According to Mesocore, the ADU market will expand from around $20 billion today to $43 billion by 2034.

With housing affordability shrinking and elderly population rising, ADUs represent a GREAT way for families to save money, provide support and companionship for the aging, and to enhance property value. These units can also be used to support young people not yet ready to leave the nest. Expect more ADUs, but also expect more community opposition to them since they also can be used as rentals.

What do you think of the statement: “Mercantilistic economies like China & South Korea, export unemployment?”

Look, I get it. We have been running large trade deficits with a number of societies. That may or may not be an indication of unfair trade. I’ll admit to liking the concept of free trade, but fair trade needs to be part of that bargain. I think America has made its point. If we are to open our markets to others, they must in turn open their markets to our manufacturers, service providers, etc.

Without question, unfair trade has hurt American manufacturing in recent decades. That has produced unemployment in industrial communities across the nation. Tariffs are viewed as a mechanism by which to reindustrialize America. But I think there are other ways that are more productive, including training of a 21^st century industrial workforce, better zoning to allow industrial development, much lower electricity rates (c’mon America), and stepped up documented immigration of those in other nations that are critical to manufacturing, including engineers.

The consensus GDP growth forecast for 2026 remains in the 2% area. Industrial Production has been improving recently. Is it fair to consider this trend as supporting or confirming the sturdy 2% growth expectation?

Statistically, industrial production encompasses the real physical output of manufacturing, mining, and utilities. With America reindustrializing, expansion in industrial production stands to be a support for overall GDP growth for many years to come. If one looks at construction employment growth over the past two years, the region that has grown the most over the past two years among the 25 most populous metropolitan areas in the U.S. is Houston. That says a lot.

At the same time, the manufacturing industry has lost 265,000 jobs since the start of 2023, and ask any manufacturer: it’s painstakingly difficult to fill industrial jobs these days.

More broadly, industrial production simply isn’t the economic driver it once was. Instead, it’s the U.S. consumer who has taken the wheel.

So while reindustrialization will certainly help support growth, it’s far from the more important factor influencing forecasts.

Subscribe now

Could you talk about the impact of demographics on the sturdiness of consumer spending, specifically as it relates to the “Graying” of America.

According to Federal Reserve data, Baby Boomers (born 1946-64) represent less than 20 percent of U.S. population, but hold a bit more than half the wealth. This group is no longer spending on college tuitions (probably – maybe as grandparents I suppose), and has seen their property values explode and their financial portfolios bulge. A few decades ago, we could not have imagined the asset price appreciation that has transpired. These folks are collectively far wealthier than anticipated.

It has also been a very fortunate generation, with many enjoying their peak earnings during the fabulous eighties and nineties. So they spend. But that also means that the places they’re leaving (New York, California, Maryland, etc.) are losing a lot of spending power while the places they’re going to (Florida, Arizona, South Carolina, etc.) are experiencing benefits of the bulge.

Homes are too expensive and that is unlikely to change since demand exceeds supply of affordable homes. Could commercial to residential conversions be the answer? Do you know of examples or stories of where this is happening that could help affordable living?

There are of course many examples of commercial to residential conversions, but often those conversions have been associated with higher rents. After all, these conversions are often wickedly expensive and these buildings are often in very prestigious parts of cities (e.g., historic central business districts). Given the cost of labor, construction materials, and elevated interest rates, I think it will be rare for commercial conversion to translate into affordable housing. An exception to this might be vacant shopping centers. Often, redeveloping these into living spaces is a bit less expensive, there is surface parking all around, and residential units are often quite small. That seems like a decent recipe.

I thought the DMV economy would crater as a result of DOGE’s DRP resignations, presumed migration out of the area, and some government contracts that were terminated abruptly in 2025. But it hasn’t, or at least it doesn’t feel like it has. Why not?

Actually, I think the DMV economy has taken a major wallop. Among the 25 most populous metropolitan areas, none has lost jobs at a faster clip than the Washington metro. Somehow, there is still a lot of traffic, but we’ve suffered massive job loss and the local federal contracting community has been severely impacted. I wouldn’t say the economy has cratered, but it has taken a major step backwards.

What would YOU recommend to solve Social Security and what do you think is the most politically viable solution?

Obviously, we can continue to raise retirement ages as one partial solution. But that’s hardly enough. Even with that, the Social Security Trust Fund is set to become insolvent by the end of 2032.

So we have to do more. The most obvious thing would be means testing. We probably need to treat Social Security as genuine social insurance. For those who made a lot of money over their careers (those who won capitalism), we may need to harshly limit payouts. Oh, I know that’s not popular. I’m not running for office. I’m just trying to answer your question.

We could also (more controversy forthcoming) step up documented immigration. We need more taxpayers. Rather than taking more out of the paychecks of the young, it might be nice to have more people paying into the system.

I believe that we are creating a caste society. The tax reform that significantly increased the standard tax exemption reduced the tax incentive for home ownership - the wealthy will own property and build generational wealth, the others will be renters for life. And, expanding 529 accounts down to primary education (not just secondary) provides the wealthy (the better off?) with tax incentives/advantages to send their children to private school - further separating the have’s from the have not’s. Thoughts?

I agree that America is bifurcating. When I was growing up in a farm town in Illinois that also happened to have a regional medical center, there was a large middle class. There weren’t that many really rich people and there weren’t that many really poor people (more poor than rich, though). That community, Kankakee, Illinois, also had a sizable manufacturing base (I played little league baseball for the Massey Ferguson Tractors). Between agriculture, healthcare, and manufacturing, there were plenty of middle income jobs.

But during the intervening decades, asset values have surged. Those who owned those assets have become far wealthier. Various technology cycles have further concentrated wealth. There was no Google when I was growing up, no IPhone, no Internet, no commercialized artificial intelligence, etc. While stockbrokers made plenty back then, it was nowhere near what leading financiers make today.

So wealth has become hyper concentrated. Many of the young, in particular, are falling behind. That is why they are opting for democratic socialism in many instances—they don’t think the system is fair and they don’t think they can make it. And it’s going to get worse with AI standing to dislocate many young people while enriching others. Bottom line: at some point, politics in America are going to change, and not in a way that favors free enterprise.

What are some items/policies that the governor could quickly change to encourage businesses to invest in Maryland?

I don’t know if you’ve noticed, but Maryland has suffered a spate of layoffs recently. Among them are Amentum, a defense contractor in Anne Arundel County that filed notices to lay off 382 workers; Crosby Marketing Communications in Annapolis, which is cutting 20 workers; WBAL NewsRadio; WMAR-TV; the University of Maryland, College Park (84 State-funded employees); Stanley Black & Decker, which is closing its Carroll County manufacturing facility eliminating 55 jobs; Thriftbooks, which is closing its Baltimore County warehouse and laying off 136 workers; and Savista, a global company that assists hospitals and medical practices manage billing, insurance, etc. – the company plans to lay off 39 employees in Bethesda.

In my judgment, Maryland’s standard of living is in sharp decline. The state was hammered last year by federal government and spending cutbacks that took the form of steep agency layoffs and diminished funding for research and nonprofits. Many of our most successful citizens are leaving, whether for Florida, Delaware, South Carolina, etc. So what can a governor do? Two things come to mind.

1. Cut the corporate tax rate—at 8.25%, it’s not competitive. We need to attract much more private investment to offset diminished federal support;

2. Get that bridge up already—I figured the State of Maryland would botch this. The Key Bridge collapsed in March 2024. Two and a half years have passed and there simply has not been enough progress. Cost estimates are spiking, and of course they are. Of course, Maryland isn’t alone here. The federal government has actively hindered efforts, and the White House’s trade policy is the primary factor behind those rising cost estimates.

Zack’s Answers

If you were advising a local government that was considering a moratorium on new data centers, what would you tell them?

I’d tell them:

First: It’s their job to convince local residents that data centers are a good idea. These debates have shown me that people simply do not care if their county is going to reap massive tax benefits several years down the road. Maybe they don’t trust the officials to spend that money wisely. Maybe they don’t even plan to live their a few years later.

It’s really great if a county can pay for a new school in cash instead of financing it, but does that matter to the average voter? Probably not.

So what would move the needle with constituents? Well, the idea that data centers will allow tax cuts—like we’ve seen in Loudon County—is pretty appealing. And while that won’t be feasible in many cases, specifically enumerating what the county plans to do with the new revenues would be a good first step.

Second: they need to stress that data center-supported tax revenues really are transformative. Look at Quincy, Washington, a farming community of about 8,000 people. About 60% of the town’s property tax revenues now come from data centers, and that has paid for a new high school, hospital, library, police and fire station, and other infrastructure upgrades.

Or look at the rural Louisiana school district whose teachers will get $50,000+ bonuses (!!!) because of data center-related tax revenues.

Fourth: while a lot of the arguments against data centers are overstated, acknowledge that they do need to be situated correctly. There are a few that most definitely are not (the xAI Colossus data center in Tennessee comes to mind…), and this sucks both for the people near them and broader efforts to build more data centers.

Big picture, data centers have to go somewhere, and local governments might regret it in a decade or two if they don’t capitalize on capturing those tax revenues.

How are “temporary” data center moratoriums impacting local economies?

In the short term, the effects are likely limited to less construction activity (there’s a reason unions are extremely pro data center).

To the extent these moratoria are indeed temporary, longer-term impacts could be pretty small. But that’s assuming that developers don’t abandon the potential data center projects that are delayed by the moratoria.

Subscribe now

What sector of labor is most under threat to lose jobs to AI over the next year to 3 years?

First of all, we really don’t know. If you look back at what jobs people thought the internet would disrupt, they were pretty wrong. I think the same thing will happen with AI, and we’ve definitely seen that so far. AI is great at writing code, and yet software developer job posting on Indeed have risen 25% since May 2025.

But if I had to pick, I’ll say administrative office workers. To be clear, I think these jobs will still exist, but there will be fewer people doing them because of how much of it can be augmented by AI.

The fields I think are safest are construction and healthcare. For the former, it’s notoriously difficult to make a robot that can lay bricks, and for the latter, regulations and the nightmarish idea of a robot cutting you open are probably going to maintain the status quo for a while (someone else can test the first catheter-placing robot…)

But really, who knows? Maybe AI will come for the economic consultants.

How do you believe artificial intelligence will affect the economy over the next three to five years, particularly in industries that rely on technical expertise and professional services?

The Financial Times published this graph showing the range of outcomes from human extinction to the end of scarcity. So there’s a pretty wide range of outcomes from here.

I think AI will make people more productive to some degree, and it’s going to increase the quality of output in certain sectors.

That said, I don’t think the effects will be as transformative as some expect. Can Claude or Chat GPT do a lot of what a business consultant does? Sure, and yet Anthropic (the makers of Claude) just hired Accenture.

I just don’t think AI is coming for technical experts in the near term (hopefully those aren’t the famous last words of this technical expert), though it may hurt entry-level hiring.

More broadly for the economy, I think people have very different definitions of “transformative.” I’ve seen many predictions of double-digit percentage GDP growth, which just isn’t going to happen if you ask me.

A more reasonable—and still transformative—outcome is an acceleration of 1 to 2 percentage points of GDP per year over the next half-decade. Sounds like small potatoes, but consider that if Panama had 1.5 pp faster growth since about 1980, it would have higher GDP per capita than the U.S. right now.

How does the potential AI “bubble” compare in size to the dot-com bubble? The numbers at play today seem almost too large to reasonably comprehend.

In some ways—like tech’s share of major stock indices—it’s already larger, and has also spread to more segments of the economy. The dot-com boom didn’t materially impact the construction industry, for instance, but data centers currently account for about 8% of all private nonresidential activity.

But this is way less bubbly than the dot-com boom. In 1999, the valuations on unprofitable companies were just silly. Forward price-to-earnings ratios were like 50x earnings (and higher for a lot of companies). There was just a ton of money flowing into companies that hadn’t proved they could make money. The AI darlings, on the other hand, are huge companies with strong cash flow.

This could change. Ask again next year and my answer could very well be different.

On a macro level, how does climate change affect our economy?

The longer-term effects of climate change are really uncertain, and that means the longer-term economic impacts are also hard to predict. A 2024 Congressional Budget Office report’s central estimate has GDP 3% lower in 2100 than it would have been if temperatures remained unchanged. So, not great, but tolerable.

The higher end estimates from that study are a lot worse, and other studies’ predictions are even more pessimistic.

In the shorter term, a warming climate is going to reduce labor productivity, especially in areas without air conditioning and where a large share of jobs are done outdoors (neither of which really applies to the U.S.). Just 8% of U.S. workers spend at least one third of their day outside, and 65% don’t have to do any work outside at all. The global share is significantly higher.

But the big answer here is that there’s not even close to an expert consensus.

What’s Next

Next up is Week in Review, our every Friday post that gives you everything you need to know about the economy in a breezy, five-minute read. Week in Review is just for paying subscribers. If that’s not you and you want it to be, just click the button below.

Subscribe now

In addition to a rate hike, this week brought us surging treasury yields, a slew of bad housing stats, signs of ongoing consumer strength, and more.

Monday

Oil Stuff

Oil prices remain above $100/gallon, so it was another bad week on the fuel-price front. Diesel prices keep surging, up to another all-time high at $6.29/gallon, and gas prices shot up to $4.46/gallon, which is by far the highest they’ve ever been this late in a year.

TSA Checkpoint Travel Numbers

After a strong labor day, the number of people flying has fallen back to about 3% below year-ago levels, according to TSA gate check numbers. This could be a warning sign about consumer health, or it could be the isolated effect of significantly higher airfares, which are up about 28% since November according to the Consumer Price Index.

Tuesday

Treasury Yields Keep Climbing

Treasury yields surged again at the start of this week, with the 10-year climbing to levels not seen since 2007. Notably, yields were largely indifferent to the Fed rate hike (more on that below).

The takeaway here is that borrowing costs are going to keep rising.

Wednesday

Fed Interest Rate Decision

Read more

Of course, this sword cuts both ways. Yes, slower labor cost escalation relieves pressure on inflation, but it also usually means smaller pay increases for workers.

Right now, we’re getting sliced by both edges. Employment costs—which despite cooling are still up a relatively large 3.3% year-over-year—are rising almost entirely due to surging health insurance prices. Wage and salary growth has slowed to roughly pre-pandemic levels.

The prices received by health insurers (a slightly different measure than costs incurred by employers) jumped 7.25% from January 2025 to January 2026. That’s the largest annual increase in at least 15 years.

About 4 in 10 employers have cut spending on other benefits in response to rising health insurance costs, according to this Mercer survey. As a result, total employer benefit costs have risen at a far slower pace than health insurance costs.

Two-thirds of employers are planning to raise the percentage of health insurance premiums paid by their employees in 2027, further reducing net compensation.

Which is to say, both employers and employees are taking higher health insurance costs right on the chin.

This isn’t getting better anytime soon

Employers expect medical plan costs to increase another 10% in 2027, according to IFEBP. What’s driving the increase?

There are a few smaller factors at play—like increased mental health utilization, an aging workforce, and AI’s ability to help providers more efficiently submit claims—but those are fairly minor compared to the two big cost drivers.

First: prescription drug costs are a substantial cause of cost increases, with none bigger (or newer) than GLP-1s (e.g., Ozempic). Just 3% of U.S. adults used a GLP-1 for weight loss in 2024, according to Gallup, but that share shot up to 11% in 2026.

About half of all large employers covered GLP-1s in 2025, according to the aforementioned Mercer survey, but 6% are dropping it in 2026 and another 5% plan to do so in 2027.

Second: catastrophic claims are an even bigger driver than prescription drugs. While the cause of these massive claims (cancer, pre-term birth, rare genetic conditions, etc.) aren’t increasing in frequency, the costs they generate are.

Million-dollar claims rose 46% from 2022 to 2025, according to Sun Life. That has a little bit to do with inflation and a lot to do with new treatments.

Consider Elevidys, a gene therapy used to treat Duchenne muscular dystrophy that costs $3.2 million for a single infusion. It doesn’t take many $3.2 million treatments to drive up insurance costs across the board.

To be clear, it’s good that we’re coming up with new treatments and therapies. Good, but expensive.

Final thoughts

Health insurance costs are going to keep increasing over the next few years. Employment costs will continue to rise at a faster-than-ideal pace, and employers will keep trimming benefits, shifting costs to employees, giving smaller raises, and bumping prices as a result.

What’s next

We’re sending out our Q&A post on Thursday. Next up after that is Week in Review, our every-Friday post that covers all the economic news and data in a breezy, five-minute read.

Subscribe now

Week in Review is only for paying subscribers. If that’s not you and you want it to be, just click the subscribe button.

But then September comes around, SweeTangos hit the shelves, and Honey Crisps suddenly seem bloated, mundane, and—let’s be honest—overpriced. A SweeTango is simply divine. You can’t capture its splendor in words, though Applerankings.com takes a good run at it. Really, do yourself a favor and get one while they’re in season.

Why are we opening this economics newsletter with apples? To lighten the mood before diving into what was a fairly dismal week of news and data.

Inflation? High. Treasury yields? Highest in decades. Diesel prices? Highest ever. With no end in sight to the trade war with Canada and the technically-not-a-war with Iran, a rate hike now looks all but certain at the Fed’s meeting next week.

Tuesday

Oil Stuff

Diesel prices surged to an all-time high this past week, with real-time data showing the national average now above $6.00/gallon for the first time ever. Diesel hit $9.99/gallon at some pumps in California and presumably would have gone higher if the signs could accommodate another digit.

Gas prices also rose and, at $4.30 per gallon, have never been this high this late in the year.

Given that oil prices have surged back to around $100/barrel, gas and diesel prices will likely rise again next week.

Needless to say, this is bad news on the inflation front.

NFIB Small Business Optimism Index

Small business owner optimism worsened in August but is still in relatively decent shape. The comments on these surveys tend to be more interesting than the reading itself, and this one was no exception. Big issues are difficulty finding good workers, high interest rates, and uncertainty.

TSA Checkpoint Travel Numbers

It was a fairly strong Labor Day weekend for air travel, according to TSA gate check numbers, but the total number of flyers is still down about 3.5% since the middle of August.

Wednesday

Mortgage Applications

Mortgage applications fell again, continuing their downward trajectory. Given rates, don’t expect this to change anytime soon.

Thursday

President Trump Promises $5,000 Stimulus Checks

President Trump vowed to send every U.S. adult a $5,000 check, just as long as the GOP holds the House in November’s election. He suggested the checks would be paid for with tariff revenues, but given the $1.3 trillion estimated cost, that’s not going to work out.

Aside from tariff revenues not covering the cost of such a stimulus package, this would 1) require congressional approval, 2) drive up inflation, 3) massively add to the national debt, and 4) be as unlikely as all the previously promised stimulus checks that didn’t materialize.

Producer Price Index

Prices received by domestic producers rose pretty sharply in August, with the increase concentrated in goods rather than services. This is less important than the Consumer Price Index but still suggests that inflation is picking up.

Existing Home Sales

Sales of existing homes (i.e., not newly built) fell 2.0% in August and are down slightly over the past year. At the risk of sounding repetitive: blame rates.

Mortgage Rates

Mortgage rates rose again. The average 30-year fixed is now up to 6.76% and will be even higher next week.

Jobless Claims

Initial jobless claims stayed low. If you’re ever in need of good economic news, this is the place to start (at least over the past year or so).

Friday

Consumer Price Index

Consumer prices rose faster than expected in August. The relatively large increase had a lot to do with gas prices, but core prices (excludes food and energy) also jumped, rising at the fastest pace since May. That, combined with the fact that gas prices are now higher than they were in August, is bad news on the inflation front.

The big takeaway: there’s now a roughly 90% chance of a rate hike at the Fed’s meeting next week, according to markets.

Treasury Yields Keep Ripping

Treasury yields continued to soar this week, with 10-year yields—which heavily dictate mortgage rates—ending Thursday at levels not seen since 2007 (with the exception of a single day in October 2023). While yields have fallen somewhat this morning, they remain well above 4.9%.

Which is to say, borrowing costs are not falling in the near future.

University of Michigan Consumer Sentiment Index

Consumer sentiment fell pretty sharply in early September, and inflation expectations jumped to the highest level since June. Unsurprisingly, this has a lot to do with tariffs and oil prices.

Links of the Week

Insuring Against Success? Are Universities Using Prediction Markets to Fund Their Athletic Compensation Schemes? (Economic Forces)

Will AI soon lead to double-digit growth? (Ghosts of Electricity)

Liberalism needs a new philosophy of immigration (Noahpinion)

Final Thoughts

After this week, Anirban’s outlook for the economy is: Worse

The biggest thing that happened this week is diesel prices rising above $6/gallon for the first time ever. That’s going to create cost pressures throughout much of the economy. As a result, the Fed must hike.

After this week, Zack’s outlook for the economy is: Worse

Between rising rates and record high diesel prices, there’s just not much to like.

Looking Ahead

Next week is all about retail sales data, some housing stats, and the Fed’s next meeting.

Construction labor cost escalation cooled considerably from late-2022 to late-2024, but that trend is now firmly behind us. Average hourly earnings for non-managerial construction workers are now up 5.0% year over year.

That—stop us if you’ve heard this one before—is largely due to data centers. Specialty trade contractors are leading the labor-cost-escalation charge, with electricians (+7.3% year-over-year) at the vanguard.

There are other potential factors at play, like stricter immigration policy. The share of construction jobs unfilled is near a two-year high and right in line with 2019 (when construction workers were pretty hard to find). That’s what you’d expect to see if immigration policy was shrinking the industry’s workforce.

But you’d also expect to see faster wage gains outside of specialty trades, and we really haven’t (at least not yet). It’s possible that immigration policy is shrinking the workforce but activity is weak enough that it’s not dramatically impacting labor costs beyond AI-related segments.

What’s Next

Next up is Week in Review, our every-Friday post that covers all the economic news and data in a breezy, five minute read.

Subscribe now

Week in Review is only for paying subscribers. If that’s not you and you want it to be, just click the subscribe button.

Hot growth can cause overheating, though, and this week also gave us a new year+ high in mortgage rates, rising treasury yields, and a few reasons to think that inflation is going to pick up over the coming months.

Missed our posts earlier this week? You can read about why this economy makes everyone sad, the data center boom, and today’s jobs report.

Monday

Oil Stuff

Both gas and diesel prices fell slightly during the week ending Monday 8/31, but oil prices rose pretty sharply starting on Tuesday, and these prices will keep rising next week. According to GasBuddy, diesel prices hit an all-time high on Thursday.

Big picture, the Strait of Hormuz is still effectively closed, we’re still bombing Iran, and the outlook for oil prices remains less than great.

Nationalizing (Another Country’s) Oil

The U.S. and Venezuela reached an agreement that gives the U.S. a 35% equity stake in a company that now has 100-year concessions to 65 billion barrels of proven oil reserves. Put simply, the U.S. gets extensive governance rights and guaranteed access to at least 20% of production at cost.

Note that proven reserves—an estimate with reasonable certainty of how much oil can be extracted from a field—are not the same as strategic reserves, which are existing, tucked away oil supplies.

The White House has framed this as a way to refill the Strategic Petroleum Reserves, with the replenishment starting as soon as November. Unfortunately, Venezuelan heavy crude really isn’t compatible with the SPR, so this could go one of three ways:

1. It won’t be used to refill the SPR.

2. We can technically put the heavy crude in there, but it will cause problems down the road.

3. We can trade heavy crude barrels for the medium sour that typically goes in the SPR.

Big picture, this won’t really affect oil prices in the short term—note that oil prices have risen since this announcement.

Tuesday

Job Opening & Labor Turnover Survey (JOLTS)

Hiring slowed in July, but so did layoff activity and quits. Job openings held pretty steady. It remains a nice time to have a job and a bad time to need one.

ISM Manufacturing PMI

Manufacturing activity expanded for the 8^th straight month in August, according to this survey of industry purchasing managers. That’s the good news.

The bad news is that the survey suggests goods prices are going up, and they’re going to start going up fast. As always, the respondent comments are the most informative part of this, and they’re not particularly upbeat this month.

Construction Spending

Read more

U.S. employers added an impressive 162,000 jobs in August, and the prior two months’ estimates were revised up by a combined 55,000 jobs. As you can see below, this is a nice rebound from the slower growth that occurred from May to July.

While job growth only just bounced back in August, unemployment has been improving since it reached a cyclical high of 4.5% last November. It ticked up slightly in August (from 4.09% to 4.14%) but is still low by historical standards. Nothing to worry about here.

Employment gains were ever-so-slightly broader than in previous months, though healthcare and leisure and hospitality still lead the way.

Government tacked on another 35,000 jobs for the month, and that was entirely due local governments adding 41,900 education jobs. Both federal (-5,000) and state (-10,000) government lost jobs for the month.

The information industry (think tech and media) lost another 23,000 jobs in August. It’s tempting to chalk the industry’s ongoing struggles up to AI, but I’m skeptical. Traditional media has been shrinking for years, and the tech sector is still pulling back from its early 2020s low-interest-rate fueled hiring binge.

Over the past year, job growth is still incredibly concentrated in the healthcare sector.

Concentrated job growth aside, there’s just not much to dislike in today’s report. If you really want to quibble, the prime age (25-54) employment-to-population ratio hasn’t bounced back after dropping sharply earlier this summer, but that would be some pretty serious quibbling.

The Big Picture

The labor market is in pretty good shape. That allows the Fed to focus on inflation. Which is to say, the odds that the Fed hikes rates at its next meeting are higher than they were before this release.

What’s Next

Anirban is currently working on Week in Review. That will be out later today and is just for paying subscribers. If that’s not you and you want it to be, just click the button below:

Subscribe now

[The graph’s title, “Data Center Construction to the Moon,” could become a real thing. There are at least semi-serious efforts to put data centers in space/on the moon.]

The segment has grown at an unfathomable 116% annual rate since March. If that pace continues, it will expand to $162.6 billion by this time next summer, making it 29% larger than computer/electrical manufacturing construction was at its peak in June 2024.

This boom is absorbing a lot of the capital that would typically flow to other construction segments. Over the past year, data center spending has risen by 57% (+$27.3 billion), while all other nonresidential construction spending is down 3.5% (-$43.7b).

This dynamic has benefited a narrow subset of—mostly large—contractors. Our latest ABC CBI data shows that the 12% of contractors with data center work have way longer backlog (11.4 months) than the 88% who aren’t (7.5 months).

What’s Next

Next week, we’re going to look into some of the broader dynamics around data centers (like increasingly intense community opposition). We also have a Q&A post in the works, and Friday’s BLS jobs report gets its own write up.

Then it’s Week in Review, our every-Friday post that covers everything you need to know about the economy in a breezy, 5-minute read.

Subscribe now

Week in Review is just for paying subscribers. If that’s not you and you want it to be, just click the subscribe button.

Both of those times were objectively worse from an economic standpoint. Unemployment is just 4.1% right now. It was at least twice that high over a 42-month stretch from early 2009 to mid-2012 and soared to nearly 15% (!!!) in early 2020.

This discontent is broadly shared across demographics. Democrats? Not happy. Republicans? Also sad. Old? Young? College educated? GED? Doesn’t matter. Everyone’s upset.

[Okay, rich people are slightly less upset, but that’s to be expected.]

The vibes are bad, as the kids would say, but they’re not inexplicable.

Part 1: Inflation

People hate inflation. Can’t stand it. Look at that consumer sentiment chart above. The only other time it’s been this bad was 1980, when prices were rising at a nearly 15% annual rate.

This doesn’t need too much explanation. Rapid inflation makes us poorer. Nobody likes being poorer.

It is worth noting, however, that people hate inflation even when it causes their personal earnings to rise faster than overall prices. To be clear, that was pretty common in 2022. Take job switchers. Inflation peaked at a 9.1% year-over-year pace in 2022, while the median job changers’ pay increased at a roughly 16% annual rate.

The psychology at play here: when you get a raise or a new higher-paying job, that’s all you. When the prices you pay rise, that’s on someone else (corporations, the president, Congress, etc.).

Never mind that the same forces that push prices higher often push wages higher too.

Part 2: Real average hourly earnings

The flow chart above is fun and true: inflation pushes down sentiment even when earnings are rising faster than overall inflation, which was broadly the case in 2022 and 2023.

That changed when gas prices took off in February, and falling real earnings are a legitimate reason for people to be upset about the economy.

Part 3: Housing

Won’t belabor this one: mortgage rates are high, and that makes it really expensive to buy a house. Prospective homebuyers are justified in hating the current economy.

Part 4: Supply-constrained goods & services

The average U.S. household is richer than ever and, in many ways, enjoys a lifestyle that would be unimaginable even a few decades ago. They watch a bigger, nicer TV and drive a bigger, nicer car. Their house is loaded with time-saving appliances. They take more vacations, and those vacations take them farther from home.

But all of those things aren’t particularly supply-constrained. Society can easily up the production of TVs and cars and appliances and airplanes.

It’s supply-constrained goods and services that have everyone pissed off. For example, consider these two facts:

1. The number of households with more than $200,000 in inflation-adjusted income has increased by 20x since 1967 and roughly doubled since 2013.

2. The capacity of Madison Square Garden (19,812) has not changed since 1967.

If you made (again, the inflation-adjusted equivalent of) $200,000 in 1967, you could definitely go to an NBA Finals game. You could go to all of them, maybe even the road ones too.

In 2026, it was borderline-irresponsible for a household making $200,000 to go to an NBA Finals game at Madison Square Garden; the lowest-priced tickets on resale markets were over $1,000.

This same dynamic applies to everything supply-constrained. The number of homes in nice neighborhoods has not doubled since 2013, nor has the capacity of Disney World or the enrollment at that nice private school.

This effect is present across the entire earnings distribution. The working-class family that used to rent a house at that nice local beach for a full week now gets a weekend in a condo (or a week at a cheaper destination).

An upper-middle-class family—maybe one earner is a doctor—that could be front row at a popular concert in the ’80s now sits in general admission at the Taylor Swift show.

The millionaire who would have had a Warhol hanging in their office in the ’80s now has to settle for a work by some artist you’ve never heard of.

Etc. etc.

Your ability to access these supply-constrained things/experiences depends on your income relative to everyone else rather than your income compared to overall prices, and the resulting lack of access makes people feel worse off than previous generations who attained the same level of financial success.

There’s no easy or quick solution for this. Madison Square Garden can’t double its capacity every decade, and good luck trying to get your favorite vacation town to build more housing.

What’s Next

We’ll (finally) have our Q&A out in the next few days, and maybe a construction piece too. After that, it’s Week in Review, our every-Friday post that gives you everything you need to know about the economy in a breezy five-minute read.

Subscribe now

Week in Review is just for paying subscribers. If that’s not you, just click this button.

We may never see another American achieve this level of favorability.

The economic news and data this week were a little less favorable, with fairly uninspiring updates on inflation, spending, income, saving, some housing stats, and more.

Monday

TSA Checkpoint Travel Numbers

Air travel remains in bad shape according to TSA gate check numbers.

Oil Stuff

Gas prices inched up to $4.22/gallon, which is $0.95/gallon higher than at the same time last year.

As we wrote last week, diesel prices, which are up to $5.65/gallon, are still the bigger story here. There’s a good chance diesel surges above its previous all-time high ($5.78/gallon in June 2022) over the next few weeks.

Tuesday

New Home Sales

Sales of newly built homes fell sharply in July, down more than 10% from June. While new home sales have slowed a little during 2026, they’ve also been remarkably stable—and unspectacular—over the past decade (with the exception of the pandemic boom).

Conference Board Consumer Confidence Index

Consumers lost economic confidence in August, though they felt much better about current conditions and much worse about the future. As you can see below, the expectations part of the index is remarkably low by historical standards.

S&P Cotality Case-Shiller Home Price Index

Home prices rose just 1.5% between June 2025 and June 2026. That’s good from a home-buying affordability standpoint, but virtually everyone would be happier if rates were lower and home prices were rising faster.

Wednesday

PCE Price Index (the Fed’s favorite measure of inflation)

Read more

And the Orioles have lost four in a row, so, not great.

Monday

NAHB Housing Market Index

Homebuilder confidence improved slightly in August, but that’s not saying much. Homebuying demand is historically low, and 35% of homebuilders cut prices during the month.

TSA Checkpoint Travel Numbers

Air travel continues to fall off, now about 5% below year-ago levels over the past week, according to TSA gate check numbers data. Not a good sign!

Oil Stuff

Gas prices rose modestly this week, up to an average of $4.18/gallon. That’s still extraordinarily high for this time of year, but also insignificant compared to this week’s bigger and badder oil-related story: the widening diesel crack spread. (Stop right there, Beavis.)

A crack spread is the gap separating crude oil prices and refined oil products. It’s at a record high for diesel. Diesel prices jumped about $0.20/gallon this week to $5.45/gallon. The record high is $5.82/gallon. It’s possible we break that over the next few weeks as the war in Iran continues right into the midterms.

This is bad news on the inflation front. Diesel is an input to virtually everything. It’s especially important for agriculture. Harvest season is approaching, and diesel prices are currently above $7.00/gallon in California, the largest agriculture-producing state.

A few thoughts:

1. There’s no end in sight for the Iran conflict, which is squeezing supply. This week we learned that various parties aren’t even talking.

2. Ukraine continues to attack Russian energy infrastructure, including refineries. That’s putting downward pressure on global supply.

3. We’ve largely exhausted the geopolitical shock absorbers (like the Strategic Petroleum Reserve) that were in place to soften the blow. Good news here though – China has plenty of oil in reserve. They hold the world’s largest reserves.

4. U.S. refineries have been running at >97% utilization rate. That makes sense; the crack spread is essentially the refiner’s margin.

5. With refiners essentially running at capacity, any disruption will push prices even higher.

6. Weather is the most typical cause of disruptions, and we’re heading into the heart of hurricane season.

7. There’s a super El Niño this year. Without going into the meteorological details, that makes hurricanes less likely to form in the Atlantic, so at least there’s that.

Big picture, it’s a nervy time for diesel prices, and that could be a big problem for the broader economy, including for contractors.

Tuesday

New Residential Construction

July was the best month for new housing permits since February. The increase was spread across all housing types. At the same time, housing starts plunged, down about 12% on both a monthly and yearly basis.

Housing completions also plunged to their lowest level since May 2020. We’re still finishing more units each month than before the pandemic, but that might not last too long given rising financing costs.

Pending Home Sales

Pending home sales, which are supposed to show us what actual home sales will look like in a month or two, fell in July. No surprise there. It’s a bad housing market that’s only going to get worse in coming weeks.

Industrial Production

U.S. manufacturing, mining, and utility production continues to increase, albeit at a pretty sluggish pace. Still, this is one of the indicators used to diagnose recession, so slow growth is still good.

Wednesday

The Bond Market Pops Off

Read more

The bad news is the reason why lettuce prices fell, and also most of the other economic data that came out this week.

Monday

TSA Checkpoint Travel Numbers

Air travel remained 2-3% below year-ago levels over the past week, according to TSA gate check numbers. This, like a few other indicators in today’s review, is a downbeat sign for the U.S. consumer.

Oil Stuff

Both gas and diesel prices fell this week, but they’re still pretty ugly. It’s particularly unusual for gas prices, which averaged $4.14/gallon over the past week, to be this high this late in the year.

Tuesday

NFIB Small Business Optimism Index

Small business owner confidence improved in July, rising to the highest level in almost a year. There was a large improvement in hiring plans, which may or may not translate into an actual bump in hiring.

Existing Home Sales

Sales of existing homes (i.e., not newly built) slowed in July and are just barely higher than one year ago. As the NAR chief economist put it in the release, “there’s no doubt that the housing market would be thriving if average mortgage rates were to return near 6%.” Unfortunately, it could be a while until that happens.

Wednesday

Consumer Price Index (Inflation)

Consumer prices, as expected, inched just barely higher in July, rising 0.1% for the month.

Read more

The Federal Government isn’t quite seizing the means of production, but it’s being handsier than at any point in recent history.

This week was loaded with economic data releases including construction spending, jobs, job openings, several industry surveys, and more.

Monday

Construction Spending

Construction spending inched lower in June and is down more than 3% over the past year. This is particularly bad because data center construction spending surged an almost unthinkable 7% for the month. Put another way, data center spending rose by $4.5 billion (seasonally adjusted annual rate) in June, and total construction spending still fell by $2.0 billion.

The only other real bright spot is the power category, which as of April surpassed manufacturing as the largest nonresidential subsegment. Even that momentum is a function of the data center boom, though.

There are lots of dark spots. Warehouse investment, which looked to be stabilizing at the start of 2026, is back in freefall. Manufacturing construction spending continues to fall like an anvil.

The residential segment continues to soften. No surprise there, given mortgage rates.

Big picture, this is a discouraging release.

ISM Manufacturing PMI

Manufacturing activity expanded again in July, according to this survey of the industry’s purchasing managers. That’s good news, though the manufacturing side of the economy is still far from strong. This comment from a paper product manufacturer stands out:

“Our customers in Asia continue to procure elsewhere to avoid paying a tariff. While the Iran war was paused, it was terrific to see fuel prices (and delivery costs) falling steadily. Now that skirmishes have resumed, we expect fuel to rise again.”

Senior Loan Officer Opinion Survey

Lending standards are still very tight, though banks noted that standards are looser than one year ago for all categories except consumer loans.

TSA Checkpoint Travel Numbers

Nothing good to report with air travel, with passenger counts remaining 2-3% below year-ago levels, according to TSA gate check numbers data.

Tuesday

The Trade Deficit

Read more

U.S. employment dropped by 23,000 in July. That decline, the first since February, fell comically short of the consensus expectation of about 80,000 new jobs (“comically” only in a dark, maybe-we-shouldn’t-laugh way). Even worse, downward revisions knocked about 100,000 jobs off the prior two months of growth.

You’re probably thinking this sounds bad. And it is. Losing jobs is never a good thing.

For context, employers added jobs in 113 straight months from late-2010 to the start of the pandemic. We’ve now lost jobs in 6 of the past 14 months. Three-month average job growth has slowed to an anemic 20,000.

But (unlike most cases, what came before the “but” actually does matter here), there are a few reasons that today’s report is ever-so-slightly less bad than it looks at a glance.

One: the unemployment rate fell to 4.1% in July, the lowest level in over a year. Part of the drop was due to a slight decline in labor force participation. Still, layoffs remain exceedingly low. There’s no reason to worry about unemployment at the moment.

Two: local government education employment fell by almost 50,000 jobs in July, and that’s likely a seasonal adjustment quirk. An upward revision is possible, even probable, next month.

Three: wage growth was weak. Average hourly earnings rose just $0.02 for the month and are now up a mere 3.2% year over year. That’s that smallest annual increase since 2018.

Three and a half: yes, this has bad implications for workers, but it’s good news on the inflation front.

Four: the prime age (25-54) employment-to-population ratio—a very important indicator—ticked back up in July after falling sharply in June.

Five: cyclical industries, like manufacturing and construction, added jobs for the month.

Six: a softening labor market is potentially good news on the borrowing cost front because it stops the Fed from raising rates. 10-year treasury yields fell this morning.

The Big Picture

If you really squint, you could maybe justify calling today’s report “mediocre.” Even that’s overselling it, in my opinion.

The labor market is weaker than we thought. Layoffs aren’t a problem, but job growth is. Even if this is a good sign for borrowing costs, it’s a bad sign for economic health.

What’s Next

This week was packed with data releases, and we’ll cover all of them in Week in Review, our every-Friday post that gives you everything you need to know about the economy in a breezy, five-minute read.

Subscribe now

That will be out later today and is just for paying subscribers. If that’s not you and you want it to be, just click the subscribe button.