The residential construction cycle moves in the same order every time. By the time prices fall, the sequence has been running for years. This is the sequence to watch…
The sequencing framework here is the most useful thing you can hand someone who keeps asking why prices haven't crashed yet — because the honest answer is that the employment domino hasn't fallen, and without that, the price domino structurally can't. That's not a prediction, it's just how the cycle works, and EPB lays out the historical pattern clearly enough that you can track it yourself in real time rather than waiting for someone to tell you what already happened. The regional divergence point is worth sitting with too: national averages are masking some genuinely stressed submarkets in oversupplied Texas and Florida metros where the sequence has moved further along than the headline numbers suggest. The piece doesn't tell you when — and it's honest about that — but it tells you what to watch and in what order, which is a more durable tool than a price forecast. For anyone trying to time a purchase or position a portfolio around where housing goes next, the residential construction employment number is the one variable worth tracking monthly right now. What are you seeing in the regional data that suggests which markets are furthest along in the sequence?
I work in heavy civil construction (roads, water and sewer). Contractor bids have come down a lot and each project is receiving more bids. Based on our conversations with contractors the bid reduction is primarily coming from margin compression. So the past few years they were making a lot of money. The high number of bidders is the first sign there isn't enough work out there. Then prices come down and then eventually fewer bidders. I do agree that this cycle seems to be going in slow motion. Thanks for the great article.
The bid compression signal Bryan is describing from heavy civil is worth layering into this framework, because it's a parallel sequence running one step ahead of what shows up in residential construction employment data. When contractors start competing harder on margin to fill their backlog, that's the leading indicator of the leading indicator — it means the work pipeline is thinning before the workforce reduction shows up in payroll numbers. EPB's point about residential construction employment being the critical missing domino is exactly right, and the slow-motion character of this cycle probably has something to do with the fiscal infrastructure spending that has kept heavy civil and public works contractors busy enough to absorb labor that would otherwise have come off residential projects faster. The sequence is intact — it's just running with more friction than historical cycles because the employment buffer hasn't cleared. What Bryan is seeing on the bid side suggests that buffer may be starting to thin. The employment domino doesn't need to fall all at once to complete the sequence — it just needs to start moving. Are you seeing the margin compression show up yet in subcontractor pricing, or is it still primarily at the general contractor level?
This sequence framework is the cleanest way to think about housing as a leading indicator. New home sales and permits roll over long before headline prices admit anything is wrong. Watch the order, not the level.
Great analysis. I believe a likely explanation for the lack of decline in residential construction employment is that at the peak in 2022, there was a meaningful shortage of construction workers to build the houses. Timeframes for new home deliveries extended out months beyond norms. It was virtually impossible to get a plumber or HVAC person to handle a service call as they were swamped with new construction. By 2024, the residential construction labor maker was able to build new homes on time and handle service calls on installed systems. I think that dynamic likely explains the lag in layoffs at this point in the cycle.
The sequencing framework here is the most useful thing you can hand someone who keeps asking why prices haven't crashed yet — because the honest answer is that the employment domino hasn't fallen, and without that, the price domino structurally can't. That's not a prediction, it's just how the cycle works, and EPB lays out the historical pattern clearly enough that you can track it yourself in real time rather than waiting for someone to tell you what already happened. The regional divergence point is worth sitting with too: national averages are masking some genuinely stressed submarkets in oversupplied Texas and Florida metros where the sequence has moved further along than the headline numbers suggest. The piece doesn't tell you when — and it's honest about that — but it tells you what to watch and in what order, which is a more durable tool than a price forecast. For anyone trying to time a purchase or position a portfolio around where housing goes next, the residential construction employment number is the one variable worth tracking monthly right now. What are you seeing in the regional data that suggests which markets are furthest along in the sequence?
I work in heavy civil construction (roads, water and sewer). Contractor bids have come down a lot and each project is receiving more bids. Based on our conversations with contractors the bid reduction is primarily coming from margin compression. So the past few years they were making a lot of money. The high number of bidders is the first sign there isn't enough work out there. Then prices come down and then eventually fewer bidders. I do agree that this cycle seems to be going in slow motion. Thanks for the great article.
The bid compression signal Bryan is describing from heavy civil is worth layering into this framework, because it's a parallel sequence running one step ahead of what shows up in residential construction employment data. When contractors start competing harder on margin to fill their backlog, that's the leading indicator of the leading indicator — it means the work pipeline is thinning before the workforce reduction shows up in payroll numbers. EPB's point about residential construction employment being the critical missing domino is exactly right, and the slow-motion character of this cycle probably has something to do with the fiscal infrastructure spending that has kept heavy civil and public works contractors busy enough to absorb labor that would otherwise have come off residential projects faster. The sequence is intact — it's just running with more friction than historical cycles because the employment buffer hasn't cleared. What Bryan is seeing on the bid side suggests that buffer may be starting to thin. The employment domino doesn't need to fall all at once to complete the sequence — it just needs to start moving. Are you seeing the margin compression show up yet in subcontractor pricing, or is it still primarily at the general contractor level?
I found your article very insightful. Does your new home price graph showing no decline account for inflation?
This sequence framework is the cleanest way to think about housing as a leading indicator. New home sales and permits roll over long before headline prices admit anything is wrong. Watch the order, not the level.
We're right on time for the 18.6 year cycle for problems to arise
Great analysis. I believe a likely explanation for the lack of decline in residential construction employment is that at the peak in 2022, there was a meaningful shortage of construction workers to build the houses. Timeframes for new home deliveries extended out months beyond norms. It was virtually impossible to get a plumber or HVAC person to handle a service call as they were swamped with new construction. By 2024, the residential construction labor maker was able to build new homes on time and handle service calls on installed systems. I think that dynamic likely explains the lag in layoffs at this point in the cycle.
Hi Eric, I had one question did the boom in renovation of old houses has any role to play in construction employment not falling enough?