At the same time, the income the typical worker earns from a job has stalled and slipped below the trend of the last expansion.

The average worker is falling behind while corporations are doing better than ever.

These are not separate stories.

They are actually two sides of the same coin.

For thirty years, a rising share of the economy’s total income has gone to capital and a shrinking share has gone to labor.

The stock market has responded as you’d expect, with record gains, but the average American has only felt worse off.

The reason runs deeper than corporate greed, Big Tech, or the surface-level talking points.

In this post, we’ll show how corporate profits are better than ever, why it’s happening, where this trend ultimately goes, and what you can do about it.

Record Profits, And Not Just Big Tech

The easy explanation for the record increase and all-time high in corporate profit margins is that a few trillion-dollar tech companies are dragging up the average.

This is true, but it’s not the answer.

The gains in corporate profitability are everywhere, not just technology.

Pre-tax margins in manufacturing hit 15% in 2022, cooled a bit, but still sit at nearly 13%, higher than any point in the last 25 years.

Corporate profit margins hitting all-time highs in every major industry is not a one-off or a cyclical dynamic; it’s a feature of how the economy has been structured and changed over time.

The Worker’s Shrinking Slice

Corporate profits and paychecks come from the same pie.

As one share climbs, the other has to fall.

In the following charts, you can think of “Gross Value Added” or GVA as business revenue.

Worker compensation averaged 64% of gross value added from 1970 through the turn of the century before falling sharply to 54%.

This is a 10-point swing from revenue going to wages vs. profits.

Companies also borrow money, so interest expense historically has been about 2% to 4% of revenue, but during the zero-interest rate environment, big corporations were able to take advantage of capital markets and issue long-term debt at low fixed rates, making them immune to the rate rising environment that followed.

Interest expense as a % of GVA is currently 0.1%.

Taken together, employee compensation plus interest fell from 66% in 1970 to 55% today, an 11-point boost to the bottom line.

For the worker who doesn’t have a large net worth in assets or isn’t old enough to receive Social Security and Medicare, this trend shows up as a paycheck that has stalled, hasn't kept up with inflation, and is slipping below the long-term trend.

This chart shows real private sector income compared to the long-term trend line, and the gap is widening at an alarming pace, directly correlated to the poor sentiment that’s widespread in economic and political surveys today.

The real question is “how did this happen?”

The Real Driver: “Superstar” Firms

The answer has a name. Economists call them superstar firms, which is a polite term.

A National Bureau of Economic Research working paper titled “The Fall of the Labor Share and the Rise of Superstar Firms” discussed this dynamic in detail.

Of course, there is not one single cause, as many politicians looking for a scapegoat would want you to believe, but here are the main causes:

In industry after industry, consolidation and size allowed a small number of firms to capture a huge share of sales.

These larger firms with higher market concentration have more efficiencies and can run larger margins by design. As they outcompete the smaller firms, their share grows.

But here is the nuance. The labor share didn’t fall because companies cut pay.

It fell because the work moved to the giants. Real income did increase over time, but at a slower and slower pace and at a speed well below the gains in revenue.

So wages as a share of revenue fell, leading to more robust profits.

You can see this consolidation everywhere.

In homebuilding, the ten largest builders went from under 9% of the market to nearly 45% in three decades.

In home insurance, the five biggest carriers now close almost half of all claims with zero payout, a dynamic fueled by less competition and more market concentration.

Globalization and job offshoring are factors, but they’re not the whole story.

If offshoring were the main reason, every company that imported goods would have seen margins improve, but the same NBER paper shows this is not true. The median firm has not seen an improvement in margins. It’s the gains accruing to the “superstar” firms, as the paper calls them.

Also, the improvement in margins is seen in sectors like services, where globalization doesn’t touch directly.

There are many factors that have contributed to this multi-decade trend. It’s not correct to say it’s all globalization, or it’s all corporate greed.

Other factors include weak antitrust, compliance costs that intentionally crush small firms, a tax code friendly to intangible assets, a patent system built for incumbents, and cheap capital for scalable winners.

The point is that industry concentration has happened everywhere; it’s causing corporate profit margins to rise to ever higher levels, it’s good for shareholders, but bad for the average worker.

Where This Goes, And What You Can Do

Left alone, this compounds in two directions.

For the worker, it gets worse. Wages capture less of the pie every year, income growth slows, and the wealth divide accelerates.

For the capital owner, it gets better. The ever more profitable industry-leading firms become more durable, more dominant, and command higher share prices.

Both are true at the same time, and that is the part that’s hard to hold.

Here’s the uncomfortable trade everyone is making. It’s been correct for decades, and it will stay correct until something structural breaks.

If you own stocks, you own the winning side. Pricing power keeps flowing to the dominant firms in every industry, which is exactly why large-cap indexes have stayed so well supported. Record margins are the foundation under the entire bull case.

But that foundation is built on a consumer who gets squeezed a little thinner every year. And a foundation like that doesn’t crack on a schedule you’ll read about in the headlines.

It cracks in a surprisingly narrow window, usually before the data confirms anything is wrong.

That’s the part this post doesn’t cover.

The structural story tells you what to own and why it’s been working, but it doesn’t tell you when you’re exposed.

Knowing where the economy sits in its sequence, and how to be positioned so a structural winner doesn’t turn into a cyclical loser at the worst possible moment, is a different question entirely.

That’s what I write about every Sunday: a free newsletter that provides the framework behind where we are in the cycle, and what the next 1–2 years likely look like.

Join For Free Here

New subscribers also get our free guide to the housing cycle, which teaches the framework from the ground up, focusing on residential construction - the most important sector in the economy.

Two months ago, the housing market was showing fragile signs of stabilization. New home sales had recovered from the August 2025 lows, building permits had bounced off the cycle bottom, and homebuilder sentiment was beginning to stabilize. The most interest-rate-sensitive indicators in the economy were starting to suggest the worst of the housing downturn might be behind us.

However, that stabilization was based on one key assumption: that monetary policy would continue to move towards lower interest rates. Before the Iran War started, there were 2-3 interest rate cuts priced in, and mortgage rates were starting to fall into the 5% range.

Then the spike in oil prices and the fear of more widespread shortages reignited inflation concerns. Mortgage rates moved back to the mid-6% range as markets are now pricing zero rate cuts.

The fragile stabilization emerging in the leading housing indicators is now at serious risk of reversal. A continued slowdown in housing construction comes at a very bad time, as new completed inventory is piling up and homebuilders face an ever harsher squeeze in profit margins.

In this post, we’ll review the sequence of the housing cycle, where it stands today, and how the surge in interest rates disrupted what was an emerging recovery in residential construction.

The Housing Sequence

Housing follows a repeatable sequence. Every cycle follows the same pattern. New home sales peak first, and building permits follow. Then the number of housing units under construction declines, and as a result, residential construction employment drops. Home prices are the last domino, often moving 18-36 months after the sequence began.

The full sequence, including infographics and the 2008 case study, is mapped in our Free Residential Construction Cycle Guide, available for download at https://www.epbresearch.com/housing.

The Leading Layer

Mortgage rates are the input variable for the entire housing sequence. The expected path of interest rates shifted from one suggesting mortgage rates in the 5% range by year-end to one holding in the mid-6% range, with no relief in sight. That is a massive change in the early part of the housing sequence and another rug pull in a series of head fakes for the housing cycle recovery.

Building permits and new home sales were responding to the prior easier path. Permits bottomed at 1.3 million in August 2025 and recovered to 1.45 million by December. New home sales went from averaging low 600,000 units to 764,000 in November. The data was starting to suggest the leading indicators had found a floor. The strength of the acceleration was in question, but the worst was over.

Homebuilder sentiment had stabilized over the same window, holding above the readings registered in the fall of 2025.

When all three variables move or stabilize together right after a big shift in monetary policy inputs, like mortgage rates, it is the type of cluster behavior our framework looks for. The cluster was stabilizing and turning up.

The Cyclical Layer

While the leading layer was showing tentative recovery, the cyclical layer never participated.

The number of housing units under construction continued to decline through the entire stabilization window. This is normal because there is a lag between the most leading variables and the actual construction activity. Given the stabilization we had started to see, there was optimism that a bottom in housing construction was coming soon.

The problem with this now-disrupted stabilization is that the cyclical variables never stopped declining. So if the leading variables turn lower again due to higher interest rates and inflation fears, the downturn in actual construction activity will not let up.

This also comes at a bad time for homebuilders because their inventory situation is much less healthy.

As we covered in a previous article, it’s not just the level of inventory that matters for builders, but more importantly, it's the level of completed inventory that has the biggest impact on profit margins.

The share of new homes for sale that are completed sits above 25% and is moving closer to 30%.

In 2022, when months supply last spiked above 10, only 8% of that inventory was completed, so there was a huge backlog of construction despite looking like an inventory glut on the surface. Today, that inventory situation is worse despite a lower months supply.

This difference is the entire ballgame for builder profit margins.

In 2022, builders could slow starts and let the pipeline drain. The homes were not built yet. Over 90% of the listed inventory was not started or was under construction.

There was nothing to discount.

Today, a much larger share of inventory consists of finished homes. Completed homes have more carrying costs with no revenue. Builders have more pressure to move them. That means incentives, discounts, and margin compression.

From Margin Compression to Job Losses

The Cyclical Economy (housing and manufacturing) is where the recession really occurs. Not because construction and manufacturing are large shares of GDP. Because their amplitude is so large that they account for the majority of job losses in every modern recession.

The transmission mechanism from slowing demand to layoffs runs through profit margins. When margins compress, layoffs follow. When layoffs hit, workers stop discretionary spending, credit defaults rise, and only then does services spending start to slow. This is the recessionary sequence, and it must flow through every step. Every domino must fall.

The buffer that has prevented this sequence from completing for three years has been the extraordinarily high level of starting profit margins. Homebuilders entered this period with margins near 20%. That cushion is why units under construction have declined materially without triggering large construction layoffs.

But buffers are finite. A completed inventory share nearing 30%, combined with months supply already elevated and now facing another potential downshift in sales volume, is the combination that compresses margins toward the threshold where layoffs become unavoidable.

Where the Sequence Stands Today

The leading housing indicators have lost the variable that was supporting their stabilization. The cyclical indicators never stopped declining and are now at risk of falling further. Builder profit margins have absorbed three years of cyclical weakness because they started so high, but that buffer is closer to the layoff threshold than at any point in this cycle.

Despite a slowdown in this part of the economy for three years, the housing sequence has not completed. There have not been large residential construction layoffs.

No layoffs in housing and manufacturing, no recession.

But the important framework condition, namely profit margins, that prevented this cycle from completing, is now under even more serious pressure.

This is a huge shift because the housing cycle was starting to stabilize, which would have removed the only vulnerability in the economy.

The AI and technology Capex cycle is keeping many sectors strong, including manufacturing. Housing was the one leg that could cause the economy to fold, and it was stabilizing.

This spike in interest rates puts housing back in a vulnerable position, and it's absolutely the most critical area of the economy to watch.

Since the housing cycle is the most important area of the economy to watch, we put together a Residential Construction Cycle Guide that maps the full five-step housing sequence. It tells you exactly what indicators to follow and in what order.

You can grab the free PDF guide below.

The EPB Guide To Housing

The same Federal Reserve, same mortgage rates, and same monetary policy, but two completely different housing markets.

The national home price index averages everything into one number, showing 0.7% price gains in the last twelve months, but that figure doesn’t describe any specific market.

Eight major US cities are making fresh all-time highs in home prices. Seven have been stuck below their 2022 peaks for almost four years. And in the middle of that, half the country had a real housing correction with declines up to 13%.

A 5%-10% nominal decline in a period of high inflation is a meaningful reset of values, and that happened in roughly half of the major US metros.

In this article, you’ll see the major split in how housing markets performed since 2022, and the one variable that determined the path each city took.

Different Housing Markets, All Running at Once

If we rank the 20 major US cities by the number of months since each one hit an all-time high in home prices, three distinct groups emerge.

Chicago, New York, Cleveland, Detroit, Boston, Charlotte, Minneapolis, and San Diego are at all-time highs in home prices right now, or within one to two months of an all-time high.

Phoenix, Tampa, Dallas, Denver, Seattle, Portland, and San Francisco have been stuck below their 2022 home price peak for 25 to 46 months. Almost four years in some cases.

In the middle sit five cities. Atlanta, Miami, Washington DC, Las Vegas, and Los Angeles. These have been drifting sideways for about a year.

That’s three completely different housing markets running at the same time, blended at the national level into one story that doesn’t fit any single market.

And the regimes are still moving apart.

Chicago, New York, and Cleveland are all up more than 4% in the last twelve months.

Cities below their 2022 peaks are still declining 1% to 2%. Tampa, Seattle, and Denver are all down on the year.

When you hear someone say “the housing market” is up 0.7% in the past year, that number is averaging Chicago at +5%, with Denver at -2%.

National housing data is important, and we use it often in our research at EPB.

But home price changes are notoriously regional. Half the country had a legitimate housing correction after the summer of 2022, while the other half experienced a straight-line increase since 2020.

The Recession That Never Made the Headlines

The peak-to-trough drawdown numbers from the 2022 peak make the regional split even clearer.

The 2022 peak marked the end of the stimulus-fueled pandemic boom, and what happened after looked very different depending on where you lived.

San Francisco fell 12.7%, Seattle fell 12.2%, Las Vegas fell 8.7%, and Phoenix fell 8.4%.

These are not small numbers. A 12% drawdown in a major US housing market is a real housing recession, particularly in an environment of high inflation. A 10% nominal price correction during a period of high inflation is a significant reset of values.

The national price index had a minor correction but more or less continued to rise through the 2022 regional reset, which is why this housing downturn never turned into a national story.

In 2022 and 2023, half the cities fell and the other half kept climbing.

New York barely moved, Chicago barely moved, and Cleveland barely moved. Those cities averaged into the national index alongside the cities that were dropping double digits, and the average came out virtually flat.

So, Why Did Half the Country Fall

The answer lies in construction.

In some areas of the country, a booming housing market ushers in extreme building. Building permits in Austin and Phoenix increased by over 100% from 2017.

Building permits in Detroit and Chicago never increased by more than 20%. There was a 5x difference in the amount of building activity between boom cities and cities that make new construction very difficult.

So how did that contribute to differences in home price trends?

Detroit home prices increased 54% from 2017 to 2022, then another 17% from 2022 to today.

Chicago tells a similar story. A 34% increase from 2017 to 2022, and another 23% from 2022 to today.

The boom towns had much different performance. Phoenix, for example, increased 105% from 2017 to 2022, then fell 3.3% from 2022 to today.

Cities that had a lot of construction saw huge booms and then bigger price corrections. Cities with limited construction saw smaller initial gains, but the gains continued after 2022, with double digits in many cases.

The chart below plots the relationship across every major city, and the pattern is clear.

The price gain from 2017 to 2022 is on the vertical axis. On the horizontal axis, is the price gain from 2022 to today. There’s a clear relationship. The bigger the boom from 2017 to 2022, the smaller the gain from 2022 to today.

Phoenix sits in the top-left corner. Over 100% price gain from 2017 to 2022, then a decline over the last four years. If you bought in 2017, you’re still up huge. If you bought in 2022, you’re down over the last four years, and that’s before fees and costs. This is a housing recession in the most concrete sense. Someone could have bought a home and five years later be underwater before fees and transaction costs, and with interest rates so much higher than in 2022, they can’t refinance their way out of the problem either.

The other side of the spectrum is highlighted by New York, which sits in the bottom-right corner. More limited gains in the five years from 2017 to 2022, about 47%, but then another 25% gain from 2022 to today.

Two Paths to the Same Destination

Averaging boom-bust and low-supply markets makes the two distinct housing paths unmistakable.

The first composite is the average of New York, Chicago, Cleveland, and Detroit, four low-supply markets. There’s been a smooth rise with barely a pause in 2022.

The second composite is the average of Phoenix, Tampa, San Francisco, and Seattle. Four boom-bust markets. There was an aggressive run-up into 2022, then a fairly sharp drawdown into 2023, and then a recovery attempt that’s still ongoing.

Here’s what’s interesting.

Over nine years, both paths produced almost identical cumulative gains. The Low Supply Path produced an average 77% gain. The Boom-Bust Path produced an 81% average gain. The same nine years, almost an identical total return.

The boom-bust markets didn’t get rewarded for the volatility.

A homeowner in New York earned roughly the same total appreciation as a homeowner in Phoenix over the last nine years. But the New York homeowner never had to watch their home value drop 13% and stay underwater for almost four years.

The Phoenix homeowner experienced the entire emotional cycle. The euphoria of the 2021 boom, the dread of the 2022 reset, the false hope of the 2023 recovery, and now the slow grind of being stuck.

The Sequence Underneath It All

The cities that boomed the hardest from 2020 to 2022 are the cities where building permits, new home sales, and residential construction activity surged the most when rates fell to 3%.

When rates moved from 3% to 7%, the construction activity that drove the boom collapsed.

New home sales fell, building permits fell, and builders started cutting prices.

This is the exact housing cycle sequence we reference often in our research on the residential construction cycle.

Monetary policy moves first. Then permits and construction lead. Employment follows, and prices come last.

Reading Your Own Market

The headline national number is hiding different cycles running at once. Our residential housing framework gives you a way to read your own local market.

Are you in a boom-bust market with huge swings in construction, or are you in a market with limited construction and low inventory forever?

If you are in a boom-bust market, you need to follow the sequence and wait for monetary policy to ease enough that permitting and construction activity starts to pick up again before the housing market gets out of the slump it has been in since 2022. You also need to be far more careful with real estate investing decisions.

If you’re in a market like New York, the housing market is always tight since supply is so low, and the only thing that matters is a national recession with sustained job losses.

We put the full national housing cycle sequence into a free guide, with graphics, visuals, and timelines, using the 2008 period as a case study.

It’s the framework behind how to interpret incoming housing data, what the construction cycle means for the broader economy, and exactly which indicators to watch.

From there, regional differences take over. But it’s the national sequence that sets the tone and puts the whole thing in motion.

You can grab the free PDF guide below.

The EPB Guide To Housing & Home Prices

At the exact same moment, the eight largest public homebuilders posted operating margins of 20.4%, the highest level in their recorded history.

High inventory and high profit margins for homebuilders…two things that are not supposed to happen at the same time. In fact, this is the first time in history this occurred, leading to what is now known as the “housing recession that never came.”

In this post, we’ll review the extremely reliable track record of the months supply metric, how it failed in 2022, what we’ve done to correct it, and what the improved metric says about the housing cycle today.

Before diving into the post, I want to make sure you download our free housing cycle guide. It’s a highly detailed PDF that, step by step, walks through the exact sequence of the housing cycle, with updated graphics and visuals to help you properly interpret incoming housing data.

Download The Free Housing Cycle PDF

The Track Record of New Housing Months Supply

Months supply of newly constructed homes has been one of the best leading indicators of the residential construction cycle, and thus, the overall business cycle.

When the months supply of new home inventory increased, that signaled an oversupply of inventory relative to sales, a precursor to companies pulling back on new construction, reducing employment, and creating a downturn in associated industries.

A months supply above 7.0 was almost always associated with a recession, and certainly when months supply rose above 8.0.

The 2001 recession was a unique case, and the only recession in the last 60+ years that did not have a housing downturn associated with it.

But in every other case, housing was a primary driver of recessionary conditions, and months supply was the best metric.

However, that measure failed in 2022 as the months supply of new homes exploded to nearly 11.0, but construction continued, job losses didn’t come, and the economy never entered a recession.

Months supply has remained elevated, in historical recession territory since 2022.

Why has this measure failed?

Why Months Supply Failed In 2022

New homes can be listed as “inventory” when a housing unit is:

• Not started

• Under construction

• Completed

In 2022, when months supply surged above 10.0, almost none of that inventory was completed.

In fact, in 2022, the percentage of new home inventory listed as “completed” was below 8%, the lowest level on record.

This meant that 92% of the “inventory” was either “not started” or “under construction.”

A months supply of 10.0 with 50% of that inventory completed is far different than a months supply of 10.0 with 10% of that inventory completed.

Historically, builders carried a relatively stable portion of inventory as completed, around 20% - 30%, but due to the pandemic and supply disruptions, this number fell out of balance to the point where builders were carrying less than 10% of their inventory as completed.

This situation never happened before, thus never requiring an improved metric, but clearly, months supply lacked the detail to be an effective indicator in all environments and completed months supply is better.

A Better Metric: “Completed Months Supply”

Completed months supply, or the months supply level * the percentage of that inventory marked as “completed” performed just as well as the regular months supply measure in all previous recessions and was not subject to the false signal in 2022.

In fact, completed months supply fell to the lowest level on record in 2022, indicating a huge backlog of construction, despite what may have been declining sales volumes.

A gap between these two measures was historically non-existent, but when the gap opened in 2022, it was the completed months supply measure that was correct.

Where The Housing Cycle Is Headed

The residential construction cycle is the most important driver of the overall economy. A tightening of monetary policy leads to less affordable housing, a decline in new home sales, a pullback in construction, a fall in employment, and the classic economic stress that manifests from higher unemployment in the early industries.

This cycle, there was a tightening of monetary policy in 2022, which led to a pullback in home sales, but construction activity continued and stayed elevated until 2024 due to the historic backlog of completed homes.

As that backlog cleared, eventually, the number of housing units under construction started to decline, and normally, that comes with a decline in residential construction employment.

This cycle, however, a nearly 30% decline in the number of housing units under construction has so far only come with a 4.5% decline in residential builder employment.

The only way for companies to have a massive drop in production volumes but a limited drop in employment is for the difference to come out of profit margins.

If a business sees activity levels decline without reducing costs, profit margins must compress, and that’s exactly what happened.

Profit margins for the largest homebuilders declined from 20.4% in 2022 to less than 12% today, an 800bps decline in margins.

What’s been unique is that because margins started at 20%, builders were able to stomach the 800bps hit to margins without panicking.

At any other point in history, that magnitude of margin compression would have been too much to handle.

Before the pandemic, profit margins were highly stable at 10%. An 8% decline would have put profit margins at 2%, leading to a massive reduction in jobs. But this cycle, an 800bps decline in margins has left builders still in a more profitable position than they were from 2012 to 2020.

Historically, the standard months supply measure, inverted in this chart, has been a leading indicator of profit margins, but it decoupled in 2022.

The completed months supply measure, inverted, has been equally as good in the past, and did not decouple in 2022.

Builders are increasingly carrying more of their inventory as completed, which is a more expensive situation than carrying raw land, putting more pressure on the already declining profit margins.

And with the affordability problem in the housing market still extreme, the pace of new home sales will remain sluggish, forcing companies to continue reducing construction activity or carry even more completed inventory.

In either case, the residential construction cycle still hasn’t turned higher. It’s been the same cycle since 2022, just massively drawn out by major construction backlogs and extreme profit margins in the aftermath of the pandemic.

The broader economy will be relatively unimpacted until we see the cycle play out further in residential construction, with material job losses, because it’s the early-stage job losses in sectors like construction that create the problems for the rest of the economy.

Our free Residential Construction Cycle guide walks through the full housing sequence - step by step.

It is the deeper framework behind the completed months supply metric and the signals that come before and after it.

Download The Free Housing Sequence PDF

When you download the PDF, you also join our free weekly newsletter, which provides updates on our entire process and framework, along with some more business cycle teachings.

If you found this report helpful, please consider sharing it with your network!

Share

Across the last three recessions, manufacturing jobs peaked an average of 29 months before the downturn officially started. Construction jobs had an average lead time of 9 months.

The headline number gives virtually zero warning.

The economy has 158 million payrolls, but only 13% of them really matter for recessions.

Construction and manufacturing are the 21 million jobs in the economy responsible for nearly all the booms and busts. They account for virtually 100% of private-sector job losses in recessions and have the longest average warning signal. They move first, and they move the most.

In this post, we will highlight the importance of construction and manufacturing jobs as the driver of recessions, and your best warning signal, and why focusing on total payrolls or even the larger service sector jobs is a horrible idea.

Breaking Down Nonfarm Payrolls

Total nonfarm payrolls can be split between government and private-sector jobs.

Within the private sector, there are 9 major sectors. This chart shows a flowchart of nonfarm payrolls, peeling back the layers to arrive at the most important areas of the labor market.

After the split between private jobs and government jobs, there are cyclical jobs (construction & manufacturing) and everything else, or non-cyclical jobs.

There’s another layer below both construction and manufacturing - the split between residential and non-residential, and the split between durable goods and non-durable goods.

Before getting too specific, we have to establish why construction & manufacturing jobs are the only sectors that truly matter for recessionary developments.

This chart shows the 9 main private-sector categories and their respective average growth rates during expansionary and recessionary periods.

The difference between the average expansionary growth rate and recessionary growth rate is what we call the expansion-recession spread.

Construction has the largest spread, over 12%, closely followed by manufacturing. Together, these two sectors are only 15.5% of private payrolls, but the volatility or the difference in behavior during expansions and recessions is way larger than any other sector in the economy.

60% of the private sector labor market is what we’d call mildly cyclical - showing a modest spread in performance between expansions and recessions.

The remaining 25% of the private sector labor market is completely stable, showing virtually no difference in performance between expansions and recessions, with the largest component of this group being education and health care services.

Construction and manufacturing are clearly in a category of their own, driving all the cyclicality or all the booms and busts, while the rest of the labor market, over 100 million jobs, experiences very little decline during recessions.

Hiring conditions may change for these non-cyclical jobs. There may be times when it's harder or easier to get hired, but massive and sustained job losses in these categories are a byproduct of the damage in the cyclical sectors, not a warning signal.

15% of Private Sector Payrolls. 100% of the Job Losses.

This chart plots the decline in construction and manufacturing jobs and in all jobs.

In every recession from 1970 through 2001, construction and manufacturing lost more jobs than the rest of the economy. That means all jobs, excluding construction and manufacturing, actually added jobs through the recession.

This table shows the total job losses in construction & manufacturing and the total private-sector job losses in each recession.

In four of the last seven recessions, cyclical jobs accounted for more than 100% of private sector job losses.

In two of the last seven recessions, cyclical jobs were more than 90% of total job losses.

The 2008 financial crisis showed that cyclical jobs accounted for 52.9% of private-sector job losses, but only because the recession became so severe and morphed into a banking crisis, which led to job losses everywhere.

Even so, cyclical jobs accounted for 53% of all job losses, even though they represented just 18% of private jobs at the end of 2007.

If the importance of construction and manufacturing jobs over everything else is not already clear, this chart shows cyclical job losses and job losses for all private jobs excluding construction and manufacturing.

The importance of a sector, particularly for the labor market, is not tied to its size. Just because we are a services economy by size, does not mean those jobs matter more for the business cycle.

Just because construction and manufacturing account for only 13% of private payrolls today doesn’t mean they won’t account for more than 50% of all job losses in the next recession.

Big Doesn’t Mean Important. Most Sectors Never Decline

This chart shows the longest streak each sector has had without year-over-year job losses on the left, and the percent of time each sector spends in contraction on the right.

Important in this context doesn’t mean socially important; it means important for business cycle dynamics.

Education and healthcare jobs had a streak of 687 months without year-over-year job losses, which is over 57 years! The streak would have been even longer had it not been for the pandemic.

Education and healthcare jobs have only contracted in 1.6% of months since 1963.

Compare this to construction, which had a max streak of 106 months and contracts 26.5% of the time, or manufacturing, which has a max streak of 67 months and contracts almost 50% of the time.

Since we know there are sectors that virtually never contract, why include them in the analysis of cyclical or recessionary dynamics? Their stability dilutes the signal, under the hood, that develops in the more volatile, economy-sensitive sectors.

Construction and manufacturing move the most, and they move first. This is the signal.

Construction & Manufacturing: Your Best Warning Signal

From peak to trough, across seven recessions, construction jobs declined an average of 12.7%. Manufacturing declined 11.3%.

The next-closest sector in terms of percentage decline during recessions is information technology, with an average of 5.4%.

Financial activities, leisure and hospitality, education, healthcare, and other services all show an average decline of less than 2%.

Construction and manufacturing clearly decline the most.

They also decline the earliest. Manufacturing jobs peaked ahead of the last 3 recessions with an average lead time of 29 months. Some of this is due to the secular decline in manufacturing jobs, but even normalizing for that factor places manufacturing at the top of the list in terms of lead time.

Construction jobs peaked an average of 9 months ahead of the last three recessions, with a 20-month lead time ahead of the 2008 recession.

Technology ranks second on this list, but it only declined ahead of 1 recession, 2008, and this was still an artifact of the 2001 recession, when the tech sector shed jobs for over a decade straight, so this is not a reliable signal.

The point is that construction and manufacturing jobs decline the most, decline earliest, and provide the only true cyclical or recessionary signal in the labor market.

Key Takeaway & How To Monitor The Labor Market

Construction and manufacturing account for just 13% of private payrolls today, but those 21 million jobs are the most important in the entire economy in terms of business-cycle or recessionary signals. The rest is noise.

Within construction and manufacturing, we can narrow it down even further.

Durable goods have a much larger expansion-recession spread than non-durables, 12.2% versus 5.8%. Durables also carry 7.8 million jobs compared to 4.7 million for non-durables.

Durable goods are more cyclical and larger than non-durable goods and thus, the more important component within the manufacturing sector.

Residential building and non-residential building each hold a bit less than 950,000 jobs - almost the same size.

But residential jobs are far more cyclical, carrying a much larger expansion-recession spread.

Residential construction is more interest-rate-sensitive and faster-moving than non-residential construction, making it a more important focus within the construction sector.

The economy has 158 million payrolls, and the jobs report the media scrutinizes each month treats all these jobs as equal.

The business cycle runs on the narrow slice of construction and manufacturing jobs that account for up to 100% of all job losses during recessions, with residential construction and durable goods manufacturing driving the biggest swings.

Focusing on the headline numbers will never provide an early warning sign, as the important cyclical signal is buried by non-cyclical jobs like education and health care, which can go 50 years without contracting.

At EPB Research, we focus on the construction and manufacturing portions of the labor market in detail, uncovering signals within residential construction and durable goods manufacturing that provide the earliest warning signs the business cycle has to offer.

To see this framework applied to current data every week, join the free EPB Research newsletter.

Free EPB Newsletter & Housing Guide

When you sign up, you also get our free Housing Cycle Guide, which shows you the exact sequence of how job losses develop in the residential construction sector - the most important sector in the economy.

Whenever someone discusses the “housing market,” the first reaction is to think about home prices. But just like the broader economy, the housing market has a sequence that involves many parts, which all move in a particular order.

There are certain early warning signs that always turn first, years before home prices move. Think of it like a row of dominoes. One falling piece knocks over another, and it’s a chain reaction.

If your process for understanding the housing market starts with what home prices are doing in your local area, you are watching the last domino in the sequence and using that to predict when the next one will fall. This plan will never help forecast what the months and quarters ahead will look like.

In this post, we will walk through the exact sequence of the housing market from building permits to construction activity to supply to employment, all the way to home prices. The exact sequence that occurred in the 2008 housing crash and is repeated throughout history. Then we’ll review what the data says about the housing market today.

To help you understand this housing sequence, we created a free step-by-step housing guide with graphs and timelines like the one above, so you can always have the resource and the sequence handy. This will massively help you interpret incoming housing data and place it in the correct order. You can download the free guide below.

Download The Free Housing Sequence Guide

The Housing Sequence

The residential housing sequence is the single most important part of the entire economy. The housing market is extremely interest rate sensitive, so it reacts strongly to changes in monetary policy, and it has an extremely high multiplier effect, meaning booms or busts in the housing market can be felt in many industries like trucking, plumbing, the manufacturing of building materials and appliances, and the list goes on and on.

The housing sequence sets off a chain reaction that is quite literally the cause of many booms and busts in the economy throughout history. Let’s walk through this sequence using the 2008 housing cycle as our reference. At the end, we’ll see what the housing sequence says about the economy today.

When the Fed tightens monetary policy and interest rates rise, the housing market feels it almost immediately. But not through prices. The first response shows up in the volume of new home sales.

New home sales measure the volume of transactions, not the price. When mortgage rates rise, fewer buyers qualify or choose to transact, and the volume of sales declines. This response happens within weeks of a rate move. It is mechanical. Higher rates reduce affordability, fewer buyers show up, and the pace of sales drops. New home sales peaked in the summer of 2005, more than two years before the broader economy entered recession and nearly two years before home prices peaked.

Building permits follow. A permit is a forward commitment to build. When builders see sales slowing, they pull back on new projects. Building permits peaked in September 2005, confirming the signal from new home sales.

When both indicators decline together after a tightening cycle begins, it forms a cluster of leading indicators. One indicator declining can be noise. A cluster of indicators in the same bucket declining together is the sequence turning.

This is the critical concept. We are not looking at any single data point and drawing a conclusion. We are watching for a sequence of indicators all moving in the correct order, because that pattern has preceded every major housing downturn in the modern era.

From Leading Indicators To Real Economic Impact

As building permits decline, actual construction activity follows. The number of housing units under construction peaked in May 2006, several months after the volume of new home sales and building permits.

A new home sale or a building permit can change in an instant. If a home is under construction, it says under construction until it’s completed. In the terminology we use at EPB Research, this is the transition from the Leading Economy into the Cyclical Economy, from forward-looking commitments to actual production and employment.

Fewer homes under construction means fewer construction workers are needed. Residential construction employment peaked in mid-2006 and began declining. This is where the sequence becomes dangerous, because job losses in construction create downstream effects on the rest of the economy.

Workers who lose construction jobs immediately cut discretionary spending. They may default on auto loans, credit cards, and mortgages. Banks experience rising delinquencies and tighten lending standards. Credit contracts further, which means even fewer buyers can qualify for home purchases, which reinforces the decline in new home sales that started the sequence in the first place.

This feedback loop is the transmission mechanism. It is how weakness in residential construction spreads from the Cyclical Economy into the Aggregate Economy. And it is why the statement “housing is the business cycle” is true, but only when you are referring to the entire sequence, including construction activity and employment. Not building permits and not home prices in isolation.

Each metric moves at a different point in the sequence, and reading them in order is what gives you foresight. Reading any one of them in isolation is what gets you in trouble.

Transition to Home Prices

Home prices peaked in early 2007, almost two full years after new home sales peaked. Sellers resist cutting prices even as every other signal has deteriorated. Inventory builds. Days on market stretch. And only after sustained pressure do prices adjust, usually related to the labor market or job losses, which is why it is a critical piece of the puzzle.

This is not unique to 2007. It is how the sequence works in every cycle. Home prices are a lagging indicator. They confirm what the earlier parts of the sequence already showed. They do not predict what comes next.

Most people watched home prices through 2005 and 2006, saw them still rising, and concluded the housing market was healthy. Meanwhile, the leading and cyclical indicators had already turned. The sequence had been running for over a year. By the time prices confirmed the downturn, the opportunity to position ahead of it had long passed.

Where We Stand Today

So now that we have this sequence in mind, at least at a high level, we can run it through today’s data to see where we stand.

Download The Free Housing Sequence Guide

We outlined the sequence as a process that flows from monetary tightening to new home sales and building permits, then into housing units under construction and employment, and then into home prices. Home prices acting as the lagging indicator that responds to oversupply and weakening demand through labor market stress.

New home sales peaked in 2020, but this was strongly related to the initial reaction during the pandemic. New home sales had a secondary peak in late 2021.

Building permits peaked at the start of 2022, a few months later. Both new home sales and building permits have not staged significant upturns yet.

The number of housing units under construction had an official peak in October 2022, but stayed elevated until the start of 2024. This one year period where units under construction were stalling but not declining was a big reason why this cycle has taken a long time to play out.

Historically, housing construction peaks were more abrupt than gradual, sweeping ones.

Housing units under construction remain in a forceful downturn. With less construction activity comes less demand for residential housing employment.

This is the most critical step of the sequence today. We noted that housing units under construction didn’t really start to slide until the start of 2024 and this is when residential building employees peaked. However, we simply have not seen a huge decline in residential building employment.

Certainly not the level of decline that occurred in past cycles when construction activity fell, and this is the absolute most critical variable in the entire economy today. Without a decline in housing-related employment, the rest of the cycle cannot complete. The next domino won’t fall.

There’s not enough downstream stress created from early stage unemployment in cyclical areas.

The average sales price of a new home hasn’t declined much from peak. It’s a volatile series, and if we smooth the data, average new home prices haven’t turned down, which makes perfect sense with the sequence.

Prices have stalled, and housing is absolutely regional.

There have been some strong declines in new home prices in parts of Texas, Florida and areas with an oversupply of building.

The regional discussion is important and a topic for an entire article, but the national data is what we’re focusing on in this breakdown - and how it relates to the broader economy.

The Case-Shiller National Home Price Index, which tracks existing home prices rather than new home prices, is also at a cycle high, consistent with the sequence outlined and the fact that the home price domino cannot completely fall without the employment domino falling.

We will dive into the reasons why employment hasn’t moved as much as you would have expected, given the magnitude of the construction decline, in future posts.

The main takeaway from this analysis is the sequence or the order of events in which the housing cycle moves and how that plays into the broader business cycle.

Monetary policy has impacted the pace of new home sales and the level of new building permits. This led to a sharp decline in housing units under construction and, so far, just a small decline in the associated residential construction employment. This is where we are today.

Home prices on a national level will remain elevated until that employment domino fully declines.

There will continue to be huge regional differences in home prices, and that’s an important conversation, but for now, you know exactly where we stand in the national housing cycle.

We track this entire sequence and provide updates every month in the EPB Business Cycle Trends Report and our specific Residential Construction Report.

If you want the condensed version of this housing sequence in a quick reference guide, so you can track it yourself, download the free housing sequence guide using the button below.

Download The Free Housing Sequence Guide

If you found this breakdown helpful, please consider sharing it with your network.

Share

This statement is repeated nearly every day across financial media as proof that “the consumer” drives the economy.

The logic seems correct.

If consumer spending is almost 3/4 of the entire economy, and consumer spending slows, the economy will decline…

So the best practice must be to watch the consumer, track sentiment, and focus on anecdotal data points like “is the airplane full?” or “is the restaurant packed?”, all things we hear discussed frequently.

The problem is that these comments about the consumer and the focus on economic data like personal consumption treat the consumer as a single, uniform bucket.

The consumer is either still spending or not.

But when you dig beneath the headline and look across all categories of consumer spending, it becomes clear that the 70% figure is dangerously misleading. In reality, less than 20% of all consumer spending really matters for the economy.

In this post, we will highlight which categories of consumer spending actually impact the business cycle and why focusing on “the consumer” and waiting for a slowdown in total consumption is a horrible idea.

Breaking Down Consumption

Personal consumption can be split into three primary categories: durable goods, non-durable goods, and services.

Durable goods include items such as motor vehicles, furniture, and household appliances.

Non-durable goods are items such as food, clothing, and gasoline.

Service consumption is by far the largest category, including housing, healthcare, insurance, restaurants, and utilities.

These three primary categories of consumer spending can be split into 15 more specific buckets, which offer a lot of detail on how certain categories behave in recessions and expansions.

Going back over 50 years, we looked at the average growth rate for each of these 15 categories during recessionary periods and during expansionary periods.

The difference in growth rates during recessions and expansions is extremely informative and tells you exactly how “cyclical” a given bucket of consumer spending really is.

In other words, is this a category of consumer spending that really changes during recessions or not?

We can call this change in growth the expansion-recession spread.

Housing and utilities are 16% of consumer spending.

The average growth rate for this category of consumer spending during expansions is 2.9%. In recessions, the average growth rate is 2.0%. The spread is less than one percentage point. Rents get paid. Utilities get used. This category is essentially unaffected by the business cycle.

Health care is 18% of consumer spending, the single largest category in the entire basket. In expansions, average growth is 3.8%. In recession, 3.5%. The spread is 0.3 percentage points. Health care spending is as close to recession-proof as any category in the economy.

Financial services and insurance account for 7% of spending. In recessions, this category actually shows higher growth - part of this is because insurance is a very lagging indicator in terms of when prices increase. In expansions, average growth is 3.6%. In recession, 5.2%.

Food and beverages consumed at home, or groceries, are about 7% of spending. The expansion-recession spread is effectively zero.

We just listed more than 40% of the consumer spending basket that does not change at all, whether the economy is in expansion or recession. In recessions, these categories don’t even show negative growth. They are essential items that always grow.

This is why, when you look at total consumer spending, you are capturing huge buckets of the economy that don’t ever contract, giving a false sense of stability and completely overshadowing the parts of the economy that are more cyclical and offer insights into the financial health of the consumer.

Another 30% of consumer spending is what we’d call “mildly” cyclical, which leaves less than 20% of the entire consumer spending basket that really swings in recessions compared to expansions - this is the only area of consumer spending worth watching.

The 16% That Matters - Focus Here

Almost all the items in the less important 80% are either services or non-durable goods. That leaves the durable goods as the important swing factor.

The highly cyclical consumer spending categories are durable goods, which are big-ticket and generally purchased with financing, so they are sensitive to interest rates.

Motor vehicles and parts is 3.5% of consumer spending. Its cycle behavior is completely different from the stable categories. Average growth in expansions is 8.4%. In recessions, -4.8%. The expansion-recession spread is over 13 percentage points. This is the single most important factor in the consumer spending basket as it relates to the business cycle, even though it accounts for only 3.5% of the total.

Compare that to health care, the largest consumer category. The expansion-recession spread is 0.4 percentage points.

We also see very large swings in growth between expansions and recessions in recreational goods, furniture, household appliances, and transportation services.

When the labor market starts to cool and consumers start to feel the pinch, these are the categories where it shows up first. They are small, and they don’t move the total right away, so if you are watching the consumer as one big bucket, you’ll never see the early signs.

Putting It All Together

The consumer is 70% of GDP. This is true. But more than half of consumer spending belongs to categories that have essentially the same growth rate in expansions and recessions. They do not signal where the cycle is heading, and they can even mask the trends that are really developing beneath the surface.

A very narrow slice of consumer spending is actually volatile and provides a true signal of the underlying economy's momentum. Broadly speaking, these are durable goods.

The business cycle is not driven by the 70% of GDP. It is driven by a narrow slice within that 70% that most commentary never isolates.

At EPB Research, we focus on the three primary pressure points of the economy that drive all the booms and busts, and durable goods consumption is the only slice of consumer spending worth watching, along with residential housing investment and business equipment investment.

If you want this kind of analysis in your inbox every week, sign up for this free blog.

Also, check out our free videos on YouTube.

EPB Research YouTube

An accelerating economy can absorb nearly any geopolitical shock without falling into recession. An economy that is already deteriorating is vulnerable to even modest disruptions, because those disruptions only need to add pressure to a cycle that has already turned.

That distinction is critical for evaluating what the US-Iran conflict means for the economy today.

What History Shows

Geopolitical events are most frequently cited as recession triggers in public commentary, but often, they are misread.

The September 11 attacks did not cause the 2001 recession. Manufacturing employment had already peaked in 1998, more than two years before the attacks occurred.

The devastating attacks caused a legitimate pullback in corporate investment and financial market chaos, but it's clear that a major business cycle deterioration was already well underway.

The 2008 oil price spike did not cause the financial crisis. This is often described as the final piece that really broke the consumer.

But again, the cyclical economy had been contracting since 2006, two full years before oil reached its peak. Residential construction had already collapsed. Credit markets were already tightening. The energy shock added pressure to an economy already in serious trouble, but the recession itself had been set in motion years earlier by the housing cycle and the credit conditions that followed.

The pattern across both events is the same. The cyclical sectors of the economy turned first. The geopolitical shock arrived in an economy that was already weakening substantially. The shock accelerated the deterioration rather than causing it.

Now, there are also cases where shocks arrived before the cyclical economy had peaked. During the Iranian Revolution in January 1979 and the oil embargo in October 1973, construction and manufacturing employment had not yet reached its cyclical peak at the moment the shocks occurred.

In both cases, the shocks did not immediately produce recessions in the way that later oil shocks did. Let’s explore further.

The Data

The table below maps two critically important variables at the moment each major geopolitical or oil-related shock occurred: the number of construction and manufacturing jobs already lost from the cyclical employment peak, and the direction/level of corporate profit margins at the time of the shock.

The 2008 oil spike, the September 11 attacks, and the Gulf War, all shared two characteristics. The cyclical economy was already in decline, with hundreds of thousands of jobs already lost, and corporate profit margins were falling at the time of the shock. The two shocks from the 1970s were slightly different.

The economy had not lost jobs, but profit margins averaged about 9%.

Today’s reading sits in a middle position. The cyclical economy is 20 months into decline with 132,000 jobs lost, which is a real deterioration but nowhere near the scale of 2001 or 2008. At the same time, corporate profit margins are near 15.8%, which is substantially higher than they were heading into any of the recessionary shocks in this dataset.

Is There a Shock Buffer?

Profit margins deserve particular attention because they directly shape how businesses respond to deteriorating conditions. When margins are high, companies have room to absorb revenue pressure by accepting lower profitability rather than cutting payroll, which is a costly and difficult decision to quickly reverse.

When margins are thin, any revenue softness or cost increases force major headcount adjustments for survival.

Heading into the Gulf War, corporate margins were at 7% and falling. Heading into the September 11 attacks, they were at 7.5% and falling. Heading into the 2008 oil spike, they had dropped to 9% and were falling sharply as the housing collapse was already destroying profitability across the financial sector.

Today, margins are at 15.8% and so far, stable. That gap is meaningful. It represents the difference between an economy where businesses are already under pressure and will respond quickly to new shocks, and an economy where businesses have the financial cushion to absorb a period of disruption without immediately eliminating jobs.

Now, it is worth mentioning that profit margins are not evenly distributed across the economy, and the household buffer is extremely thin compared to the corporate buffer.

The Oil Transmission Mechanism

A sustained increase in energy prices can act as a tax on consumer spending or as a profit-margin compressor. When households spend more on gasoline and utilities, they have less to spend on discretionary goods. The first place that reduction shows up in the economic data is durable goods consumption, which are primarily big-ticket and more discretionary-type items.

Durable goods consumption is part of the 20% of the economy that drives all the major fluctuations in economic activity.

If energy prices stay elevated for multiple quarters rather than spiking briefly and reversing, the cyclical economy faces pressure from two directions simultaneously: the existing weakness that has been building since 2023, and a new demand-side headwind from higher energy costs. That combination would be more concerning than either factor in isolation.

The key distinction is between a brief spike and a sustained elevation. Oil markets have moved in response to the conflict, but as of now, the move does not yet represent the kind of prolonged elevation that would register as a meaningful cyclical drag, particularly considering the buffer that is present - at least on the corporate side.

Bottom Line

The US-Iran conflict deserves serious attention. It is a real event with potential economic consequences, particularly through its effect on oil prices and energy costs. But serious analysis requires separating what is knowable from what is speculative, and what is signal from what is noise.

The historical record is clear: geopolitical shocks do not write their own recessions. They accelerate deterioration that is already in motion, or they pass through economies that have the underlying strength to absorb them.

It goes without saying that a shock can be large enough to overcome any set of initial conditions, like economic lockdowns. Shocks definitely come with varying magnitudes, and that matters.

Today, the important cyclical engine of the economy has been weakening for 20 months, which is a real vulnerability. The weakening has been long in duration but mild in magnitude.

On the other side, corporate profit margins are at historically elevated levels.

The economy does not enter this shock from a position of strength in the cyclical sectors, but it also does not enter it with completely depleted buffers that were present in 2001 or 2008.

EPB Research subscribers receive regular updates on the full Business Cycle Sequence, including granular breakdowns of each cyclical component and what the latest data means for monetary policy, corporate profits, and the labor market.

As real income gains continued to fall short of society’s expectations, calls for government solutions grew louder, but ultimately, these efforts worsened the problem.

I produced a full animated breakdown on YouTube, walking through exactly why the rate of private investment has been declining, how this has impacted our population's standard of living, how we can fix it, and where the economy is heading if we don’t.

Watch it here:

The best way to help this research reach more people is to share it.

If you know someone who would find this valuable, send it their way!

Share

On the surface, this looks like a healthy economy.

A 6% pace of nominal growth is slightly above the pre-pandemic average, and there are no obvious signs of distress in the headline number.

If you stopped your analysis here, you would conclude that the economy is fine and that monetary policy is working.

In fact, the case for rate hikes could even come to the table. But lately, interest rates have been falling, exactly the opposite of what you’d expect from these stronger headline figures.

Stopping your analysis at the headline is exactly how you end up being wrong at the major turning points.

Stripping Away the Noise

At EPB Research, we don’t like to analyze total or headline GDP.

We break economic growth into layers, removing the components that carry no cyclical signal, until we’re left with the parts of the economy that actually drive the business cycle.

The first cut starts with total GDP and removes inventories, government expenditures, and net exports, which we call “Irregular GDP.”

These components are volatile, policy-driven, and provide almost no useful information about where the private economy is headed. What remains is Core GDP, or simply: consumption + fixed investment.

Core GDP is running at a 5.4% annualzied pace in nominal terms. This is still a reasonably strong pace of growth with no glaring signs of stress. But we’re not done.

Core GDP breaks into two distinct components: Cyclical GDP and Non-Cyclical GDP.

The non-cyclical portion, which is mostly services consumption and investment in nonresidential structures, is highly stable and not materially impacted by changes in monetary policy.

Non-Cyclical GDP is stable and currently growing at 5.9% in nominal terms. Watching the non-cyclical portion of GDP tells you almost nothing about where the economy is going next because the impacts of monetary policy are not felt in these areas.

The impact of monetary policy is felt in the Cyclical portion of the economy.

The cyclical economy, comprised of durable goods consumption, residential investment, and business equipment investment, is where all the meaningful fluctuation occurs.

This is the 20% of the economy that drives 100% of recessions.

Cyclical GDP is rising at a 3.5% annualized pace, which is actually not alarming on its own. The Cyclical portion of GDP always contracts into recessionary periods - even in nominal dollars - so a 3.5% growth rate is actually reassuring.

But it’s what’s inside that 3.5% that matters.

The AI Distortion

While it’s very difficult to isolate the specific parts of the GDP accounts directly affected by AI spending, we can roughly approximate the impact by examining the growth rates of data center investment, computer equipment investment, and software investment.

These three buckets of GDP, what we call “AI GDP,” are growing at a whopping 24% annualized pace, exploding since late 2023.

This is clearly an extraordinary growth rate, reflecting a massive wave of capital expenditure on AI infrastructure from the world's largest technology companies.

Isolating this category of growth is not to say it doesn’t matter. Quite the opposite. This is real investment and real economic activity.

The Impact on Cyclical GDP

Cyclical GDP measures durable goods consumption, business equipment investment, and residential investment.

Computer equipment investment is a very narrow slice of business equipment investment and a driver of our “AI GDP” bucket.

When we strip out computer equipment investment from Cyclical GDP, the growth rate falls from 3.5% to 0.6%. Computer equipment investment represents just 6% of Cyclical Core GDP, yet it accounts for nearly the entire difference between a cyclical economy that looks relatively stable and one that is essentially stagnant and flashing recession.

In the chart above, for most of the period from 2015 through 2019, Cyclical GDP and Cyclical GDP less computer equipment tracked each other closely. The gap between them was minimal, and that made sense since computer equipment was just 6% of the total.

Today, that gap is as wide as it has ever been, and it's masking real weakness within the traditional interest rate-sensitive sectors that are in need of policy support.

What the Rest of the Cyclical Economy Is Saying

Without the major contribution from computer equipment investment and Cyclical GDP is on the verge of contraction.

Historically, nominal contractions in Cyclical GDP are recessionary. There was a brief and minor contraction in 2022, but the labor market was strong, and this was largely driven by the exhaustion of pent-up demand during the COVID-spending spree that was quickly reversed.

The interest rate sensitive, big-ticket parts of the economy that respond most directly to monetary policy have been under sustained pressure.

Why This Matters for Monetary Policy

It is clear that the cyclical economy, stripped of AI investment, is not overheating. It is barely moving. These sectors are in need of monetary policy support as they are the traditional engine of economic growth.

The counterargument is that AI investment is still economic activity, still generating income, still supporting employment in technology and construction, and that is true.

The question is not whether AI investment is real today. The question is whether it can sustain its current pace indefinitely and whether it can continue to offset weakness in every other part of the cyclical economy.

The Bottom Line

Now, it’s not accurate to say that AI investment is keeping the economy out of recession, despite all the charts and data we have just reviewed. Cyclical GDP excluding computer equipment investment is still not contracting, but it’s decelerating and stagnating, and the risk is that it will continue to slide.

This would be a different conversation if Cyclical GDP was contracting at -5%, but the economy was still avoiding recession. That is likely what you’d think if you only listened to mainstream analysis, but that is not supported by the data.

The facts are that total GDP is running at a 6% nominal pace. The ever-important Cyclical portion of GDP is growing at a 3.5% pace, but if we remove a narrow 6% slice of computer equipment investment, we drop to a stagnant 0.6% (negative in real terms).

That is not a sign of broad economic strength. It is a sign of an economy that remains more dependent on monetary policy support than the aggregate data suggests.

Hopefully, the head-scratching decline in interest rates amongst seemingly strong headline figures should be less confusing after this growth breakdown.

EPB Research subscribers receive regular updates on the full Business Cycle Sequence, including granular breakdowns of each cyclical component and what the latest data means for monetary policy, corporate profits, and the labor market.

This wasn’t a random fluke. There were a couple of forces at work that distorted the traditional signals, and understanding what happened is one of the most important lessons of this business cycle.

I put together a full animated breakdown on YouTube, walking through exactly what drove the divergence between leading and coincident data, and what it means for how we use these tools going forward.

Watch it here:

The best way to help this research reach more people is a simple share.

If you know someone who would find this valuable, send it their way!

Share

All economic fluctuations come from a few narrow sectors, and if you want to anticipate major changes or downturns, you need to focus on these smaller sectors that actually drive them.

The question is: which sectors lead the most and which are the most reliable?

Before moving on, make sure you are subscribed to our email list. This will keep you notified of all our latest posts on all platforms, including YouTube and X.

In this post, we’ll rank the major components of GDP by two critical dimensions: how far in advance each indicator declines before recessions and how deeply each indicator contracts.

A good leading indicator, or a sector with predictive value, is one that contracts early and deeply.

The results of this study serve as the foundation for the EPB Research framework and our approach to analyzing business cycles.

What Makes A Good Leading Indicator

A good leading indicator needs to do two primary things:

1. Decline early

2. Decline deeply

An indicator that declines early but only has a very shallow contraction will never provide enough confidence or a clear enough signal for action. An indicator that declines deeply but late is not useful for any predictive analysis. An indicator like this may be useful as a confirmation tool, but not as a leading signal.

The chart below plots 15 major components of GDP across these two dimensions, using data from all recessions between 1955 and 2019.

The y-axis shows the duration of decline in quarters or how far in advance each component begins contracting before a recession.

The x-axis shows the depth of decline or how severe the contraction is relative to the component’s historical volatility. Negative Z-scores indicate contractions that are deeper than normal; positive Z-scores indicate components that barely contract or even grow into recessions.

The upper-left quadrant contains the best leading indicators: components that decline both early and deeply.

The lower-right quadrant contains the worst leading indicators: components that decline late, barely at all, or not until well into a recession.

The Best Leading Indicators

In the chart above, a few components of GDP stand out as the best leading indicators.

Residential investment: leads by roughly 7 quarters on average and also holds the deepest average contraction heading into recessionary periods. Housing investment is the clear standout as the best leading segment of the economy. It’s often said, “housing is the business cycle,” for this reason.

Transportation equipment: leads by roughly 5 quarters on average. The contraction ahead of recessionary periods is not as deep as residential, but the combination of depth and duration still ranks business investment in transportation equipment as the second-best component of the economy for recession forecasting.

Other equipment: this is another component of business investment, declining about 4.5 quarters on average ahead of recessions, with a similar depth of decline as transportation equipment.

Durable goods consumption: leads by about 3.5 quarters with a deep average contraction. Cars, appliances, and furniture are all durable goods items, and these big-ticket consumer purchases are financed with credit and are highly sensitive to interest rates and consumer confidence. Durable goods consumption doesn’t always contract early, but sharp declines are common at the onset of a recession.

These components, residential investment, business equipment investment, and durable goods consumption, are all interest rate sensitive, generally lean more discretionary, and are big-ticket items that are financed.

These three components of GDP are interconnected and empirically the best segments of the economy to watch for early cyclical movements. They are the earliest and deepest movers ahead of recessions. In a previous post, “The 20% of the Economy That Drives 100% of Recessions,” we highlight why this is where investors and economy-watchers should focus. This data provides even more supportive evidence.

The Worst Leading Indicators

Everything in the bottom right corner of the first chart is a poor leading indicator, and these are often the largest segments of the economy.

Non-residential investment in structures is an example of a poor indicator.

Investment in non-residential structures like hospitals, education buildings, and office buildings is a bad indicator because it generally peaks after a recession has already started, unlike residential investment, which offers a 7-quarter lead-time on average.

Non-residential investment is a volatile component of GDP and can ultimately make a recession mild or severe, but it is not an early warning sign.

Today’s boom in data center investment falls into this category.

Why This Matters for Economic Analysis

The implications of this ranking are extremely important for those watching the economy and looking to forecast changes in corporate profits or monetary policy.

Make sure you are subscribed to our email list. This will keep you notified of all our latest posts on all platforms, including YouTube and X.

When economic pundits say “the consumer is fine” or “we are a services economy,” they’re pointing to the exact components of GDP that never lead recessions.

These are very large segments of the economy, but when these sectors weaken, the recession is already well underway, and other categories like residential investment or business equipment investment have been contracting for over a year.

This is why conventional economic analysis is always reactive rather than anticipatory.

The important business cycle signal is concentrated in a narrow slice of the economy, the roughly 20% of GDP, that can decline up to 7 quarters in advance. Everything else is noise.

It’s Not All Gravy (Dealing With False Signals)

If residential investment is the best leading indicator with the longest lead time and deepest contraction, why not just watch housing and call it a day?

Because even the best leading indicator comes with false signals.

Residential investment has declined meaningfully outside of recession periods multiple times since 1955.

Take 1967 as an example. There was a big decline in residential investment, but a full recession never transpired. When these false signals occur, it is almost always because profit margins started from an elevated level, and the contraction did not spill over into employment.

In fact, this dynamic is exactly what has transpired over the last 2 years and something we documented in our latest article: Why Leading Indicators Failed This Cycle.

This is why it's never a good idea to use only one indicator. A basket approach helps eliminate these false signals, or taking the combined signal from residential investment, business equipment investment, and durable goods consumption.

This cycle also highlighted an important lesson about watching profit margins in cyclical industries like residential construction and in the broader economy.

High margins create resilience. Low margins create fragility. The same demand shock can have wildly different employment effects depending on where margins started.

The Bottom Line

Not all economic indicators are equal.

Residential investment, transportation equipment, and durable goods consumption lead recessions by quarters and contract sharply, offering strong, easier-to-spot signals.

Services consumption, investment in non-residential structures, and government consumption barely move until a recession is already underway, and sometimes, not at all.

Far and away the best sector of the economy to track is housing, but false signals are a genuine concern. Tracking a basket or a combination of the top leading components can help with this problem.

Also, having a good understanding of profit margin dynamics can inform you of the fragility in the economy and whether a recession can develop quickly or require longer bouts of monetary and credit contractions.

If your analysis is dominated by the 80% of the economy that doesn’t lead, you’ll always be late. Focus on The 20% of the Economy That Drives 100% of Recessions.

Track the Leading Indicators With EPB Research

This post provides the empirical foundation for the EPB Research framework. Our subscribers receive regular updates on the leading components of GDP and a process for contextualizing each data release within the broader Business Cycle Sequence.

Subscribe to EPB Research

If you found this blog post helpful, subscribe to our email list for notifications of more posts like this on all platforms, including YouTube and X.

See you in the next post!

Understanding why requires revisiting what leading indicators actually are, and why they work.

In this post, we’ll use one of the most popular leading indicators, building permits, as a prime example to understand what happened this time and what it means for the validity and usefulness of traditional leading indicators going forward.

What Leading Indicators Are & Why They Work

The economy moves in a sequence. There is a mechanical chain of events that always occurs, and that sequence cannot change. Leading indicators work because they sit at the front of that sequence, measuring the early changes or the early dominoes that fall before the economic changes occur in the later part of the sequence, in data like employment.

Recessions are synonymous with declines in employment. A recession cannot occur without a decline in employment. When leading indicators decline, they are essentially forecasting a decline in employment, although that is several steps away in the normal order of events.

Consider the logic or the normal chain reaction. Businesses don’t hire and fire randomly. They respond to changes in demand or changes in profit margins. But these demand or profit changes don’t appear instantly; they build through a series of steps.

For example, a customer places a new order before a good is produced. Production generally occurs before revenue, and changes in revenue affect corporate staffing decisions. So a decline in new orders forecasts a future decline in employment, but several steps happen along the way that can take an unknown amount of time.

This is why leading indicators have historically worked so well for recession forecasting. They capture the early weakness before it becomes visible in headline employment numbers. By the time unemployment rises, the leading indicators have been declining for months or even years.

But here’s the critical insight that most people miss: leading indicators don’t cause recessions. They measure the front of a sequence that, if completed, produces a recession. The indicator can work perfectly, measuring exactly what it’s supposed to measure, while the sequence itself fails to complete.

That’s precisely what happened this cycle, and we will use the sequence in residential construction to demonstrate this point.

The Incomplete Residential Construction Sequence

Building permits offer the clearest example of what happened during this economic cycle because the sequence or transmission mechanism is well documented, and the failure of the sequence to complete is unambiguous.

Building permits are issued before construction begins. Building permits signal what residential builders intend to do over the coming months. When building permits decline, fewer homes get started, which means fewer homes are under construction 6-12 months later.

When fewer homes are under construction, there is less demand for employment associated with home construction.

So building permits don’t forecast a recession directly; they forecast changes in residential construction employment, which, given its volatility, drives the vast majority of typical recessionary job losses.

Residential construction and the associated employment are so critical because the sector is highly interest rate-sensitive, meaning changes here occur before other sectors.

A laid-off construction worker cuts back on restaurants, retail purchases, and services. This multiplier effect connects the housing sector to broader consumer spending and, ultimately, to recessionary dynamics.

The sequence, when it completes, looks like this:

Permits fall → housing units under construction decline → residential employment falls → broader labor market weakens → consumer spending slows → recession risk rises.

This cycle, the front of the sequence worked exactly as expected. Building permits fell, and then housing units under construction declined. But residential construction employment barely budged. The indicator did its job. The sequence didn’t complete.

Understanding why is essential for forecasting the economy going forward. We need to know if the old relationships or sequence of events have permanently changed or if they have only been temporarily delayed.

The Two Changes That Broke The Sequence

There are two main factors that explain why (so far) the permit-to-employment sequence has failed to complete and thus, failed to produce a traditional recession despite three years of steady declines.

1. Profit Margins

2. Residential Remodeling

Profit margins in the residential construction sector exploded higher during the pandemic-era boom. Homebuilders, which typically operated with an average 10% operating margin in 2018 and 2019, saw margins explode to 20% as demand vastly exceeded supply.

In early 2022, building permits began to decline as the Federal Reserve raised interest rates. Typically, when permits decline, construction slows, and profit margins compress.

In fact, that’s exactly what happened!

However, profit margins started from such an extreme level, above 20%, that three years of margin compression left builders with an average margin of 12%, still above where they operated for the entire decade pre-2020.

Normally, when construction spending slows, profit margins compress, and that compression triggers layoffs. A builder with thin margins cannot absorb a revenue decline. Costs must be cut, and labor is typically the largest controllable cost.

In this cycle, the margin buffer absorbed the demand slowdown without forcing the employment adjustments that leading indicators had predicted.

Permits declined, construction activity cooled, but profit margins cushioned the blow and have prevented the sequence from fully playing out.

Could the full sequence still occur? Yes.

Margins have been compressing, and if they continue to fall, builders will eventually face the same cost-cutting pressures they always face during downturns.

The question is whether demand stabilizes before margins reach critical levels. If permits and construction spending bottom out while margins remain comfortable, the employment decline never occurs. If permits continue to fall or margins erode further, the classic transmission mechanism could still come into play.

The second factor, remodeling, is more structural in nature.

Building permits track new residential construction. They don't tell us much about remodeling and renovation activity.

Over the past several decades, remodeling has captured a larger share of overall residential construction spending.

This matters because residential remodelers are part of the overall residential building employment sector, which is a big swing factor in the labor market.

In the early 2000s, residential remodelers were only 30% of total residential building employment. Today, remodelers are nearly 50% of total employment.

If we revisit the function of leading indicators, they primarily predict changes in employment, which ultimately drive the recessionary process.

Historically, employment in the residential construction sector was dominated by new construction activity, which building permits accurately forecast. As employment shifted toward remodeling, building permits have less influence on forecasting these employment changes.

The Future of Leading Indicators

This business cycle contained several important lessons regarding leading indicators.

First, the leading indicators don’t predict the recession. The leading indicators are the first mover in a sequence of events that ultimately lead to changes in employment. Following every step of the sequence is extremely important, and jumping from the first to the last step without considering what happens in the middle can lead to mistakes.

Second, leading indicators often measure a subset of economic activity. When that subset's share of the total changes, the indicator's usefulness changes too.

Third, profit margins are an underappreciated variable in the sequence or the transmission mechanism from demand to employment. High margins create resilience. Low margins create fragility. The same demand shock can have wildly different employment effects depending on where margins started.

Where We Stand Now

As of today, the transmission mechanism from building permits to residential employment remains incomplete. The front end worked as expected. Permits fell, and construction activity followed. But the back end, the employment and labor market consequences, never materialized at scale.

The most likely explanations are the combination of remodeling strength and profit margin buffers. Both factors are still operative but potentially eroding. The question going forward is whether economic conditions stabilize before these buffers are exhausted or whether continued pressure eventually completes the sequence that leading indicators predicted three years ago.

This analysis was intended to provide a high-level breakdown of how to track the Business Cycle and its full sequence of events.

Subscribers to our research or consulting services receive updates on the entire Business Cycle Sequence and more granular breakdowns of key themes, such as employment.

Our goal is to help you understand the business cycle, so you are always in a position to make informed decisions with clarity and confidence.

Subscribe Today

As real income gains continued to fall short of society’s expectations, calls for government solutions grew louder, but ultimately, these efforts worsened the problem.

The main issue is that real income growth is tied to productivity gains, and sustained increases in productivity growth require what’s called “capital deepening,” or more (and better) machinery, equipment, and physical infrastructure for workers, allowing each person to produce more in the same amount of time.

This capital deepening process requires continued investment in physical goods and infrastructure, and over the last several decades, the US economy has invested less and less in these critical areas.

In this post, we’ll explore why the rate of private investment has been declining, how this has impacted our population's standard of living, how we can fix it, and where the economy is heading if we don’t.

How The Standard of Living Improves

The standard of living or the perception of prosperity improves when real income grows at a sufficient pace for the average person. Real income, or income after accounting for inflation, is tied to productivity growth.

An economy can only consume what it produces (not counting the trade deficit…we will get there).

If you want to consume more, you have to produce more. Producing more per hour is productivity, and that’s the only way real incomes rise.

If wages go up but productivity doesn’t, prices just rise to match the increase in wages. You earn more dollars, but it buys the same amount of stuff, and nothing changed.

To produce more, it's helpful to have more and better equipment, machinery, or infrastructure to work with.

This chart shows real personal income per capita excluding government transfer payments. Essentially, we are measuring all the income the private sector can generate on its own, adjusted for inflation, per person. At the end of 2025, this amounted to roughly $48,500 per person. We’ll refer to this measure simply as real income.

Real income has increased over time, rising from $20,000 in the early 1970s to almost $49,000 today. It’s essential that real income increases over time, but the rate of increase is what’s really critical and easy to lose sight of.

From the 1960s through 2007, real income increased at a 1.8% pace. After the 2008 recession, real income dropped and never regained the 1.8% trend line as it did after all recessionary periods in the prior four decades.

Had real income growth remained on the 1.8% trendline, the US economy would be generating closer to $53,500 per capita, $5,000 more per capita, or $20,000 more for a family of four.

For many decades, people grew to expect a certain level of real income growth or a certain pace of improvement in their standard of living.

Just as with financial investments in the stock market, small changes in the rate of compounding are hard to see year to year but can grow enormously over many decades.

Shifting the growth rate in real income from 1.8% to 1.3% can’t be felt over one year. But over 18 years, the average person is almost 10% worse off than they would have been had we stayed on the same trend.

What would the average family of four be able to do with an extra $20,000 per year? Reduce debt, pay for child care, increase savings? Or, looked at in reverse, this $20,000 gap is what many families expected in 2005 or 2006 as they projected future income, and this shortfall has been plugged with debt.

It should be clear why most people feel they are not keeping up or their standard of living is not improving as fast as they expected, because for the average person, it’s not. This becomes even worse when looking at the bottom percentage of the income spectrum, since the averages are actually improved by the very wealthy. But why is it happening?

We know that real income is tied to productivity and that real income growth has shifted down. Even a decline from a 1.8% trend to a 1.3% trend has created problems that have grown very large over the last two decades.

This problem has its roots in a lack of investment.

Net private investment measures how much the private sector is investing in structures, equipment, and intellectual property after accounting for depreciation. The economy has to spend a certain amount of money just to stand still because roads, machinery, and buildings get worn out over time, and money needs to be spent simply to offset this erosion.

Net investment measures how much the economy is investing in new structures and equipment after accounting for this wear and tear.

From 1950 to 1980, we invested about 8% of GDP. From 1980 to 2007, we invested about 6% of GDP, and from 2009 to 2025, about 4% of GDP.

If we maintained just a 6% net investment rate, that would be the equivalent of $600 billion of new investment each year - investment that we are not making today.

But why?

There are several reasons for this decline in investment, but they mainly stem from an increase in government size.

Government Size & Net Private Investment

There are certain mathematical identities that describe the economy, and one critical formula is the National Savings & Investment Identity.

The National Savings & Investment Identity explains the relationships among private-sector savings, private-sector investment, government budget deficits, and the trade deficit, with the opposite of the trade deficit being foreign capital inflows.

Domestic saving plus foreign saving must equal private investment plus government borrowing.

This can be written as:

Budget Deficit = Private Savings - Private Investment + Trade Deficit

In other words, if the US budget deficit increases, private savings must increase, private investment must fall, or the trade deficit must widen (or some combination of the three).

Since the US budget deficit has increased faster than the trade deficit, it’s been necessary for private savings to increase or private investment to fall.

Private saving is the opposite of private consumption, and in the United States, we are not keen on reducing consumption, so the burden has led to a decline in private investment.

Beyond the mathematical identity that has caused a decline in net private investment due to larger government budget deficits, there are also shifting incentives arising from a larger government spending firehose.

When government spending and deficits rise as a share of GDP, private investors face increasing uncertainty about future taxes, regulations, and the rules of the game. Long-term capital projects like factories, machinery, and infrastructure require committing resources today for returns that materialize over decades, exposing investors to years of unpredictable policy risk.

Rational actors respond by shortening investment horizons, maximizing liquidity, and favoring financial engineering over real engineering.

Talent flows toward rent-seeking behavior and navigating government rather than building productive capacity.

The result is a secular decline in net investment, a slower-growing capital stock, and compounding productivity erosion.

Fixing The Problem: Is It Even Possible?

Of course, the growing government size and budget deficits are not the only reasons the rate of net investment has declined in the United States, but they are certainly the largest factors.

Fixing the problem is extremely difficult if we agree that the deficit is the largest driver, because the United States is structurally locked into higher spending through Social Security, Medicare, and Medicaid, which make up the bulk of what are called transfer payments.

Transfer payments and interest are now 58% of all government spending, which is more than $11 trillion when factoring in federal, state, and local governments.

If reducing spending is not politically feasible, increasing taxes can close the budget deficit, but that is an equally unpopular political platform.

If we assume government budget deficits will grow in the future, then the trade deficit must increase to maintain our current rates of consumption and domestic investment.

Tariff policy is designed to close the trade deficit, which is why this is a dangerous path if reducing the budget deficit is not a simultaneous plan.

Even still, it’s unlikely the trade deficit can grow at the same speed as the budget deficits are growing, even if we wanted to. So the mathematical result is lower consumption or lower investment, or some combination of the two.

A continued decline in the rate of net investment is almost assured.

Where We Are Going & What You Can Do About It

As the rate of net investment continues to stair-step lower as it has over the last several decades, productivity growth will falter.

A lower and lower rate of net investment over time means worse and less reliable roads, airports, power grids, machinery, train stations, power plants, and factories.

Decaying infrastructure leads to lower productivity outcomes, which in turn lead to lower rates of real income growth.

This chart shows the percentage of time real income growth exceeded 3% over rolling 10-year periods.

So in the early 1970s, real income growth exceeded 3% more than 60% of the time. Today, real income growth has exceeded 3% just 23% of the time in the last 10 years.

This will continue to slide, reducing the average person's sense of prosperity and leaving out the people at the bottom of the income structure.

Savings equals investment, and the core issue is that the US doesn’t save.

Savings is essential, and one thing you can do is zig while everyone is zagging. Increasing your personal savings rate is likely the best antidote to a society with a savings deficiency.

Wealth and power will accumulate to the companies that are capital-light and monopoly-like in pricing power, which is why these well-known companies have been and likely will continue to be strong investments. But importantly, these gains have been built on an increasingly unstable foundation that will continue to erode if we remain on this same path, which many find unavoidable.

In this post, we’ll review the current state of the labor market using our sequential framework and highlight the key labor-market variables to watch in the quarters ahead.

The Headline

The December employment report showed the economy added 50,000 jobs, with the unemployment rate moving down to 4.4%.

The household survey ratios (where the unemployment rate comes from) are the gold standard labor market metrics because they come with minimal revisions.

At the end of each year, the BLS updates the seasonal adjustment factors, which results in small changes to the unemployment rate, generally 0.1% in either direction. The revisions for November shaved the unemployment rate down from 4.6% to 4.5% and then the December report showed a drop from 4.5% to 4.4%, making the month-over-month drop seem like it was from 4.6% to 4.4%.

In any case, the unemployment rate fell in December, which is a positive sign for the labor market, and it fell below the September level, so it was a sizable improvement.

(Under)Employment

The two main unemployment rates are the U3 and U6. The U6 unemployment rate is a broader measure, which includes people who are working part-time but would prefer to work full-time, so it’s a measure of unemployment and underemployment.

While the U3 rate fell below the September level, the broader U6 rate declined from the November spike but remains 0.3% higher than in August, September, and October.

So the U6 unemployment rate only partially reversed the spike that was expected to be an anomalous reading, suggesting that there was a legitimate rise in the rate of underemployment.

These charts show full-time and part-time employment ratios, and again, we reversed some of the abnormal spike we saw in November, but not the full change.

Full-time employment as a share of the labor force bounced back to 78.8% which was a positive development in December.

The part-time for economic reasons ratio, a driver of the U6 unemployment rate, hardly moved lower and remains very close to a cycle high, so there is a definitive trend in the labor market of underemployment, which is a negative, juxtaposed with a healthy improvement in the U3 unemployment rate, which is a positive.

Sequential Sector Payroll Breakdown

Our sequential business cycle process involves peeling back the layers or sectors of payrolls and narrowing in on the cyclical areas that drive all the changes in the labor market and broader economy.

Total nonfarm payrolls can be split between private and government. Government payrolls have been declining at the Federal level, which is putting downward pressure on the overall numbers.

Private payrolls can be split between construction & manufacturing (cyclical payrolls) and all private jobs excluding construction & manufacturing (non-cyclical payrolls). Cyclical payrolls appeared to rise in November, marking a transition to job gains, but those gains were revised away.

The non-cyclical payrolls, as we’d expect, continue to march higher.

Through our sequential process, we know that virtually all recessionary job losses come from cyclical payrolls.

There are job losses in some non-cyclical areas, but they are never the driver of self-reinforcing downturns.

Splitting cyclical payrolls into construction and manufacturing shows that all the declines are coming from manufacturing. Total construction payrolls edged slightly lower in December after hitting a cycle high in November.

Manufacturing payrolls are in a relentless decline with an accelerating pace starting in April 2025.

To conclude 2025, the cyclical portion of the labor market remains weak, with most of the downward pressure coming from the manufacturing sector.

The major question for the economy in 2026 is whether the monetary easing we’ve seen so far is enough to arrest the labor market declines in these cyclical sectors before they accumulate to several hundred thousand and cause downstream stress in the less cyclical and larger areas of the economy.

No Hire, No Fire Is Hurting The Youth

The increase in the unemployment rate over the last few years has fallen primarily on the youth, with more than a 3-percentage point increase from early 2023. This fits perfectly with the “no hire, no fire” labor market.

As younger age groups graduate from college, the low hiring is preventing them from finding jobs, directly pushing them into the unemployment bucket. Older age groups generally already have jobs, so the low firing rate is preventing unemployment rates from rising as rapidly in these groups.

All the reversal in the U3 unemployment rate to close out 2025 came from the 25-54 age group, which is a positive, as this is the most concerning group to see a big rise in unemployment.

Looking Into 2026

Given the low hiring nature of the labor market, there is still plenty of risk that the unemployment rate creeps back above 4.5% and heads towards 4.7% which would require the Fed to adjust their SEP and deliver on more than the one rate cut they have priced in.

The labor market has not soured in a recessionary fashion, as the layoff rate has remained muted.

While layoffs have been muted, low hiring has caused the unemployment rate to rise from roughly 3.5% in early 2023 to nearly 4.5% at the end of 2025.

The biggest risk to the labor market in 2026 is a continued slide in cyclical payrolls, as this would put continued upward pressure on the unemployment rate.

However, if cyclical payrolls stabilize, or simply decline only through attrition rather than outright layoffs, then the labor market is unlikely to create broad recessionary stress in the near term.

Stay Ahead of the Cycle With EPB Research

EPB Research subscribers receive updates on the entire Business Cycle Sequence and granular breakdowns of key themes like employment and inflation, so you’re always in a position to anticipate the next turn, not react to it.

Subscribe to EPB Research

If you found this analysis valuable, please share it with others who might benefit from a more rigorous approach to understanding the Business Cycle.

Share

Investment includes residential structures, nonresidential structures, business equipment, and intellectual property.

The answer is increasingly troubling, but it receives little attention because the negative repercussions unfold over years and decades, not months and quarters.

From the 1950s through the early 1980s, net private investment averaged around 8% of GDP. The economy was genuinely building factories, equipment, housing, and infrastructure that would compound productivity for decades. Then it dropped to around 6% through the 1990s. Today, we’re hovering around 3.7%.

A smaller share of GDP is expanding productive capacity. More of what we call “investment” is simply replacing worn-out capital rather than adding new capacity. The decline in structures and equipment is particularly concerning.

Structures constitute the physical backbone of the economy, including factories, commercial buildings, warehouses, power grid infrastructure, mining facilities, and other facilities. These assets have a long life and can serve productivity for decades.

Equipment includes industrial machinery, transportation equipment (trucks, aircraft, rail cars), information processing hardware, medical equipment, and construction machinery.

When net investment in these categories declines, we’re not expanding industrial capacity. Fewer new factories, fewer new warehouses, less physical infrastructure. Workers have less capital and less reliable technology to work with. Machinery ages and becomes less efficient.

Why is this happening? Several factors can explain the secular decline…

The economy has shifted toward intellectual property and software, which depreciate far faster than physical assets. Globalization offshored a lot of capital-intensive manufacturing. An aging population naturally reduces investment intensity. A clear focus on asset prices has redirected capital toward financial engineering and dividend extraction from corporate profits rather than reinvestment in physical capital formation. And larger government budget deficits are crowding out private investment.

The long-term implications are serious. Net investment is the basis of future productivity. Lower investment today means a smaller capital stock tomorrow, and ultimately slower wage growth since real wage improvements are tied to productivity enhancements. In addition, supply constraints emerge more frequently because capacity wasn’t built when needed, and the economy becomes more fragile.

This is the kind of structural problem that doesn’t show up year to year but rather compounds over decades.

The ongoing and growing concern over living standards, affordability, and general economic advancement is all tied to an insufficiency of net investment and worse productivity growth.

Changing this trend will be extremely difficult with large federal budget deficits, a corporate sector that favors dividend extraction from profits over reinvestment, and a low household savings rate due to constraints of expenditure and aggregate profit accounting identities.

But if it doesn’t change, many of our current societal struggles will persist.

This is why the frustration over stagnant real wages isn’t going away. Real wages rise when productivity rises. Productivity rises when workers have more and better capital to work with. And that requires net investment—not just replacing what’s worn out, but genuinely expanding productive capacity.

At 3.7% of GDP, we’re not doing that.

Well…not nearly enough…

At EPB Research, we separate the economy or GDP by categories of cyclicality. Some parts of the economy experience large booms and busts, while others are extremely stable, showing little difference in growth between recessionary and non-recessionary periods.

In a recent post, we discussed what we call “Cyclical GDP” and showed how 20% of the economy drives 100% of recessions.

Using the same framework, we can examine the Q3 GDP report in greater detail to identify the main drivers of the strong headline number.

Nominal GDP clocked in at 8.2% in Q3, but it was powered almost entirely by the non-cyclical parts of the economy, which expanded at a 9% annualized pace.

Cyclical GDP grew at a 4.3% annualized rate in nominal terms, while the even narrower subcomponents of Cyclical GDP actually contracted at a 2.6% annualized rate.

The cyclical share of GDP declined in Q3 after two years of stability. That stability has offset near-term recession risk, but negative momentum in interest rate-sensitive sectors resumed in the back half of 2025.

The divergence becomes even more striking when you decompose cyclical GDP into its three segments. Durable goods consumption and residential investment continue to struggle under the weight of higher interest rates. But equipment investment has been surging.

Specifically, business investment in computer equipment tied to the AI boom has shown remarkable growth rates, upwards of 50% in real terms in recent quarters, with its share of total GDP nearly doubling.

This growth is real and cannot be ignored. However, it’s clear that the AI-driven investment surge in narrow computer equipment is masking what would otherwise be a broad contraction in the traditional cyclical and interest rate-sensitive sectors.

This creates a challenging environment for monetary policy. Some sectors clearly need support, while others are printing historic growth rates. The Fed can't set one policy rate for two separate economic situations.

The Federal Reserve has penciled in just one rate cut in 2026. The gravity of negative momentum in the cyclical sectors may pose a challenge to that optimistic economic forecast.

The charts and framework behind this analysis are part of what EPB Research members receive every month, along with detailed coverage of the labor market, housing, and monetary policy.

Learn More & Subscribe Today

While most economists, Fed watchers, and media pundits focus on headline GDP, which was growing over 6% in nominal terms in 2000, business equipment investment was beginning to collapse.

In December 2000, the unemployment rate was 3.8%, and in March 2001, the unemployment rate was 4.3%, lower than today’s level.

Just like the 2008 recession, the 2001 economic downturn was predictable by focusing on the correct sectors, the 20% of the economy that drives 100% of recessions.

In previous posts, we highlighted the importance of Cyclical GDP: three narrow sectors of the economy that drive all the economic booms and busts.

In this post, we highlight why monitoring cyclical GDP helped predict the 2001 downturn and how focusing on the headline statistics used by most economic watchers was a costly mistake.

What Everyone Watches (And Why It’s Wrong)

The two most popular metrics, frequently cited by the Federal Reserve and major financial outlets, are GDP and the unemployment rate. These two metrics are used as barometers of economic health, and to some extent, this is true.

These are not flawed metrics in the sense that they are incorrect, but they do not tell us where the economy is heading. In fact, they tell you how the economy was doing slightly in the past.

Throughout the year 2000, leading up to the March 2001 recession start date, nominal GDP growth averaged over 6%. Wall Street consensus confidently declared the economy was booming.

These calls were emboldened by the fact that the unemployment rate was 3.8% in December 2000, just four months before the recession officially started.

The Federal Reserve, misled by these measures into conducting forward-looking policy, raised interest rates through March 2000 and did not begin lowering them until January 2001, when the recession was already entrenched.

Even worse and more dangerous, something done frequently today, is conducting monetary policy or making economic forecasts based on the stock market.

In September 2000, the S&P 500 was at an all-time high, and the high-flying Nasdaq was only down about 13% - a drawdown level that we’ve seen three times in the last three years.

Waiting for these metrics to sour, as many often do, will lead to a very late recognition of economic changes, as shifts in the economy occur beneath the surface, in only a few interest-rate-sensitive parts of the economy that respond forcefully to prior changes in monetary policy.

The Data That Matters

Headline GDP is an aggregate measure that includes some highly stable areas of the economy, such as healthcare spending, education, and government services.

These stable parts of the economy are often the largest in size, so they can mask weaknesses that develop in the smaller yet more cyclical parts of the economy.

Cyclical GDP, as we define it at EPB Research, is the combination of:

• durable goods consumption

• business equipment investment

• residential fixed investment

Every downturn in the economy is driven by at least one of these areas - very often more than one - but at least one.

The 2001 recession was unique in that it was one of the few economic downturns in which residential fixed investment did not contribute.

The 2001 recession was driven by investment in business equipment.

Yes, there was a technology component as information processing equipment, a smaller component of business equipment investment, collapsed in 2000 and 2001.

But there was also a major downturn in the equipment investment category of transportation equipment, which is business investment in cars, trucks, airplanes, buses, and ships.

In fact, one of the earliest cyclical signs in the 2000s cycle was the downturn in transportation equipment seen as early as 1999.

When companies dramatically pull back on investment, they are preparing or responding to lower demand.

The decline in investment leads to lower levels of production, which lowers the need for employment, prompting layoffs and then recession. This sequence was already in motion throughout 2000.

Cyclical GDP Divergence

In the first quarter of 2000, Cyclical GDP growth began to underperform total GDP growth, indicating that economic momentum was faltering and that gravity would pull the broader economy lower.

The structural components of GDP, such as services, healthcare, education, and government, kept GDP strongly positive despite contraction in key sectors.

While the exact component of Cyclical GDP that drives the downturn is different each time, the overall pattern repeats every cycle.

In a previous article, we discussed how the 2008 recession was predictable using the same process. In fact, the Cyclical GDP divergence was even more pronounced.

Cyclical GDP growth started to contract in late 2006, while total nominal GDP growth was greater than 5%.

In the first quarter of 2008, residential investment growth was -27%, Cyclical GDP growth was -9% and total nominal growth was +2.6%.

Waiting for total GDP growth to turn, or the unemployment rate to rise sharply, or the Fed to acknowledge a slowdown is always too late. The cyclical sectors of the economy often give 6-12 months warning and in some cases like 2008, up to 18 months warning.

The Key Takeaway

The 2001 recession was one of the milder downturns, but it wasn’t unpredictable, and it wasn’t caused by a spontaneous collapse of stock prices.

Growth in the cyclical sectors began to decline in late 1999 and continued to decline through 2000.

(Not to mention, corporate profit margins peaked in Q3 1997.)

Watching headline GDP, total nonfarm payrolls, the unemployment rate, and stock prices will lead to reactive, not predictive decisions virtually every time.

You will always receive advanced warning in the above metrics by focusing on cyclical sectors.

This is the framework, process, and sequence we use at EPB Research - we focus on the 20% of the economy that drives all the booms and busts.

You can (and should) track these indicators, as well as a similar process for employment, inflation, and more. Or, have us do it for you and save the time.

This is what we deliver at EPB Research. Detailed written and video analysis covering the business cycle, labor market, inflation, and Fed policy - plus real-time discussion when data drops.

Learn More About EPB Research

Throughout 2007, calls for economic improvement, a soft landing, and fears of inflation dominated. In fact, in mid 2008, the Federal Reserve discussed raising interest rates as some participants downplayed the odds of a recession (which was already underway).

Below are highlighted comments from a June 2008 Federal Reserve meeting.

The reality is that using a sequential framework and focusing on the right sectors of the economy revealed weakness as early as 2005 and crisis-level warnings in 2007, long before the official onset of the recession.

In this post, we’ll highlight several charts that predicted the 2008 recession up to 18 months in advance. To be fair, this process did not suggest a global financial crisis was around the corner, but that crisis stemmed from a Business Cycle downturn that was easy to spot with advanced warning by focusing on the correct sectors.

In a previous post, we discussed the importance of “Cyclical GDP” and how recessions always develop through durable goods consumption, residential investment and business equipment investment - highlighted in the GDP breakdown below.

At EPB Research, we focus intensely on the cyclical components of GDP and we even go layers deeper in our premium reports, emphasizing the importance of sectors like motor vehicle consumption, or investment in transportation equipment.

Cyclical GDP is the combination of durable goods consumption, residential investment and business equipment investment - three narrow slices of the economy that are interest rate sensitive and drive all the booms and busts.

The chart below shows Cyclical GDP as a share of total GDP. When the cyclical share of GDP is declining, that is generally a sign that monetary policy has restricted the availability of money & credit and caused a downturn in the interest rate sensitive part of the economy.

A declining share of Cyclical GDP is an early warning sign that the broader economy will cool, since all downturns (and booms) start in the cyclical sectors.

The cyclical share of GDP peaked in Q3 of 2005, long before any discussion of a recession.

Not every recession is seen in all three cyclical sectors, but there is always at least one that drives the downward momentum.

In the 2008 recession, it was, of course, the residential investment component that started declining in 2005 and collapsed through 2010. Equipment investment did not turn down until almost 2008.

In addition to looking at the cyclical share of GDP chart, the first of three charts that predicted the oncoming recession, it’s useful to look at the composition or breakdown of the three cyclical sectors to identify where the focal point of the downturn is.

By 2007, it was clear the residential investment share of GDP was in free fall and if you use a process or framework that emphasizes the importance of residential investment, recession watch was in full effect.

Part of the problem is that so many investors or economic pundits look at aggregate numbers like total GDP or total nonfarm payrolls and do not subscribe to the idea that it is the narrow cyclical sectors that drive the booms and busts.

For example, the third chart that helped predict the downturn perfectly highlights why most people miss critical turning points in the economy, even when they happen over a year in advance.

The first picture in this post showed comments from Fed member Lacker, in June of 2008, saying that the labor market has not shown the “accelerating declines” that he feared.

Similar to the cyclical components of GDP, it’s vital to focus on the cyclical sectors of the labor market, namely construction and manufacturing.

By June 2008, the May 2008 labor market report was in hand and it showed more than 1 million job losses for construction and manufacturing employment, declines that started in April 2006, two years before these misguided comments.

By focusing on the 20% of the economy that drives 100% of recessions, spotting economic downturns far in advance is possible.

The 2008 recession was spotted more than 18 months ahead of time by the cyclical share of GDP, the component breakdown highlighting the residential investment focal point, and cyclical employment.

Predicting a global banking crisis is not the purpose of this framework. Rather what a sequential Business Cycle process can do is highlight the brewing vulnerabilities in the parts of the economy that always respond first to changes in monetary policy or the availability of money & credit.

The origin of the 2008 recession in residential construction was as clear cut as it gets. Other recessions originated in durable goods consumption or equipment investment. But in all cases, the combined cyclical part of the economy suffered a downturn earlier and harder than the broader economy.

This framework didn’t stop working after 2008. It’s a consistent, repeatable process built on the foundation of credit-based economic systems.

We use this framework every month at EPB Research to track exactly where we are in the current cycle. The same methodology, applied to current data, plus deep dives on employment, inflation, housing, and more.

If you’d like to see what these indicators are showing right now, and where the cycle is headed next, click the link below to learn more about our premium research services.

Learn More About EPB Research

“The consumer is fine.”

“The consumer is holding up the economy.”

“We are a services economy.”

These comments are generally accepted as true, given the frequency that they are repeated, but the reality is that these statements are exactly the opposite of what you want to focus on if your goal is to track and get ahead of major changes in the Business Cycle.

Despite popular opinion, most segments of the economy do not contract and are generally extremely stable. That stability makes them useless for understanding where we are in the cycle—and where we’re headed next.

In this post, we’ll walk through the structure of the economy, explain why conventional analysis is always late to the major inflection points, and highlight the sequential Business Cycle framework we use at EPB Research to track the segments that actually drive every boom and bust.

Breaking Down GDP

The economy has four primary categories: personal consumption, private investment, net exports, and government expenditures.

It’s true that personal consumption, or “the consumer,” accounts for about 70% of total GDP.

But that fact leads to an incorrect conclusion: that the consumer drives the ebbs and flows of the Business Cycle.

Within personal consumption, we have durable goods, nondurable goods, and services.

Within private investment, we have residential investment and nonresidential investment, which further break down into business equipment investment, nonresidential structures investment, and intellectual property investment.

Most analysts spend their time watching total consumption or total GDP. But not all of these categories behave the same way across the Business Cycle.

The Three Sectors That Actually Matter

The Duncan Leading Index was made popular by economist Wallace Duncan, and it was designed around a simple but powerful concept: changes in the economy stem from a few select categories.

The Duncan Leading Index involves tracking durable goods consumption, residential investment, and nonresidential investment.

At EPB Research, we’ve modified the original Duncan Leading Index to make it more cyclical and timely. We track three specific categories:

1. Durable Goods Consumption

2. Residential Investment

3. Business Equipment Investment

We call this combination “Cyclical GDP,” and it’s truly the only part of the economy you need to watch if your goal is to understand the Business Cycle.

In our modified version of the Duncan Leading Index concept, we exclude nonresidential structures investment and intellectual property investment from our framework.

That’s intentional.

These categories are not leading indicators, and in most business cycles, they’re lagging.

Nonresidential structures (hospitals, office buildings, shopping centers) often peak during a recession and bottom well after a downturn is over. Including them would add noise, not signal. These structures are less sensitive to short-term fluctuations in demand, having longer construction timelines, and include many government or quasi-government projects that are less sensitive to pure economic conditions.

Why These Three Categories Are All That Matter. Really.

Most economic observers simply do not realize that the non-cyclical part of the economy, the part that comprises over 80% of GDP, almost never contracts.

Since 1960, Cyclical GDP has contracted in 21% of all quarters. Total GDP has contracted in just 4.2% of quarters. And the non-cyclical remainder? Just 3.8%.

All the signal is in Cyclical GDP. All the noise and the narrative is everywhere else.

By focusing on “the consumer,” you are really focused on an area of the economy that doesn’t often contract, other than the narrow slice of durable goods, which is the only segment of consumer spending worth devoting extra attention.

The difference becomes even more clear when we compare growth rates during recessions versus expansions.

During non-recessionary periods, Cyclical GDP grows at an average rate of 8.0% in nominal terms (including inflation). During recessions, it contracts at -5.2% on average.

Now look at Non-Cyclical GDP. During expansions, it grows at 6.6% on average. During recessions? It still grows—at 6.2%.

The consumer is almost always fine, right up until the moment they’re not. And by then, Cyclical GDP has already been flashing warning signs for quarters.

Cyclical GDP as a Leading Indicator

Importantly, Cyclical GDP, plotted as a % of total GDP, has been a historically reliable leading indicator.

As with all leading indicators, lead times vary. These are not stock market timing tools, but rather signals for when momentum in the economy shifts, helping to anticipate changes in monetary policy, corporate profits and trends in unemployment.

When the cyclical share of GDP starts declining, that is usually a signal that monetary policy tightness is slowing the interest rate sensitive parts of the economy.

Conversely, big increases in the cyclical share of GDP indicate an economy that has building momentum.

The chart below shows the three individual components of Cyclical GDP : durable goods consumption (orange), business equipment investment (green), and residential investment (blue) - each expressed as a % of GDP.

In many recessions, like the 1991 recession, all three components decline simultaneously.

In some recessions, like 2001, weakness is concentrated in one component, like equipment investment, while other areas such as residential investment remain strong. In these cases, recessions tend to be more mild and only occur when one component declines enough to offset the strength elsewhere in the Cyclical Economy.

The chart below is the same as above, but with the pre-recessionary periods highlighted for emphasis.

After the pandemic boom, the cyclical components experienced a hangover, mainly durable goods consumption, but over the last 2.5 years, there has not been much deceleration in the durable goods share or the residential share, and the equipment share has actually been rising.

While durable goods and residential investment have not been weak enough to create recessionary conditions (yet), the real story is the rise in equipment investment which is unusual in the face of a monetary tightening cycle.

If we look at the main categories of equipment investment, we can see that there’s been a large upswing in information processing equipment and transportation equipment.

Information processing equipment is the AI story. Large capital expenditures for AI investment have driven an increase in computer and processing equipment.

While the AI story is a major driver of the strength seen in the broader equipment investment category, the rise in transportation equipment is also important and perhaps even more unusual.

Transportation equipment is highly cyclical but the normal sequence in this sector was disrupted during the pandemic supply chain crisis. Transportation equipment, a cyclical sector, collapsed as the economy was booming in 2021 and 2022 - a very unusual occurrence.

In 2023 and 2024, transportation equipment investment had to catch up even as the broader economy started to cool.

If we zoom in on the main categories of cyclical GDP, we can see stability in recent quarters for durable goods and residential investment but a surge in equipment investment.

Equipment investment has powered ahead on the heels of AI capital expenditures but also an unusual dynamic in the transportation sector that’s been unfolding since the early days of the pandemic.

The motor vehicle manufacturing sector is now shedding jobs so the residual strength in transportation equipment is almost certainly set to fade in the quarters ahead.

That leaves the AI story as the lone driver of equipment investment and the open question as to whether equipment investment can continue to rise and offset the flat-to-down trends we are seeing the share of durable goods consumption and residential investment.

Through the second quarter of 2025, we can see the nominal growth rate of the cyclical components of GDP is 5.7%, a healthy pace and one that precludes the economy from entering recession.

The fate of the economy in 2026 rests with the developments in the cyclical sectors.

The Bottom Line: Focus on What Moves, Not Just What’s “Big”

Most economic analysis is highly lagging and is reduced to nothing more than trend following because it focuses on the wrong things.

Total GDP, total consumption, or the aggregate labor market show where the economy has been or at best, where it is today, but certainly not where it’s going.

The three sectors that make up Cyclical GDP - durable goods consumption, residential investment, and business equipment investment - generate all the fluctuations in the economy.

This sequential breakdown of Leading, Cyclical, and Aggregate components of the economy is the foundation of the EPB Research framework.

We never react to headlines or random stock market gyrations. We track the sequential progression of the Business Cycle through the sectors that actually drive it.

Track the Cycle With EPB Research

This post is a snapshot and a high-level breakdown of how to track the Business Cycle using a consistent and repeatable framework.

EPB Research subscribers receive updates on the entire Business Cycle Sequence and more granular breakdowns of key themes like employment and inflation, so you’re always in a position to anticipate the next turn, not react to it.

Subscribe to EPB Research

If you found this analysis valuable, please share it with others who might benefit from a more rigorous approach to understanding the Business Cycle.

Share