Data through: September 2026 · updated Oct 2 · Updates: Monthly · Next: Nov 6
MThe three bars show the last 3 months for this indicator, oldest left to newest right. Bar height reflects each reading relative to the other two — the tallest is the highest of the three, the shortest the lowest.
Historical context: Currently in the 30th percentile of all historical readings.
What this means right now: Growth between −3.3% and 1.7% is in the second quintile — builders holding or slowly trimming staff, as in about 1 month in 5 since 1986 by today's bands.
Residential Construction Employment (YoY) · Monthly · 1986–2026
NBER recessions
Scroll to read · zoom and pan in Expand
Monthly values
Residential Construction Employment (YoY) — the last 12 months
Every month's reading, with the band the heatmap gives it. The chart above shows the same series in full.
Month
Value
Band
Sep 2026
−0.4% YoY
Concerning reading
Aug 2026
−0.3% YoY
Concerning reading
Jul 2026
−1.5% YoY
Concerning reading
Jun 2026
−1.6% YoY
Concerning reading
May 2026
−1.5% YoY
Concerning reading
Apr 2026
−0.9% YoY
Concerning reading
Month-end readings of the same series the chart shows, from res_construction_employment. Values are the current record, not point-in-time: a month that has since been revised shows its revised value.
What is the Res. Construction Jobs?
This indicator shows the year-over-year percent change in employment at residential building construction establishments. That is the BLS Current Employment Statistics count of workers at firms whose primary business is building new houses and apartments (NAICS 2361), in thousands, seasonally adjusted. The heatmap cell and the vote on the Housing Cycle page use the year-over-year rate; the underlying headcount is stored alongside it.
Builders hire after work is under way and cut staff after it dries up, so residential construction jobs turn late. On the Housing Cycle's record the series' 12-month change correlates most with single-family starts 9 months earlier (correlation 0.7). That places it in the Housing Cycle's lagging stage, with house prices and mortgage delinquency. The charter settled the question by data after reviewers disagreed on whether construction jobs were coincident or lagging.
The series is narrower than the total construction employment already on the heatmap (FRED: USCONS, shown as construction_employment), which includes nonresidential and heavy construction. Residential specialty trade contractors (electricians, plumbers, roofers on residential work) are a separate BLS series that begins only in 2001; it duplicates this mechanism and is not shown.
How We Color-Code the Res. Construction Jobs
Our heatmap colors this indicator against historically significant thresholds:
Below −3.3%
Bottom quintile — residential builders shedding jobs, by today's bands about 1 month in 5 (98 of 489); 34 of those months fell inside a recession
−3.3% – 1.7%
Second quintile — flat to slipping
1.7% – 4.1%
Middle quintile — around the series' median growth
4.1% – 6.5%
Fourth quintile — strong hiring
Above 6.5%
Top quintile of the history since 1986 — hiring at boom rates, by today's bands about 1 month in 5 (99 of 489)
Historical Extremes — What Happened Next?
When this indicator reaches extreme levels, history shows consistent patterns:
Apr 2021Pandemic rebound
+20.4%
A base effect on the April 2020 layoffs more than a boom. The level had already regained its pre-pandemic peak in December 2020, four months before this reading.
May 2009Financial-crisis layoffs
−23.4%
The deepest year-over-year loss on record. Residential building employment fell by nearly half from its 2006 peak (1,022K in April 2006 to 557K in January 2011) before recovering.
Apr 2020Pandemic lockdown
−12.8%
A two-month collapse that reversed within the year as building resumed under record-low rates.
Frequently Asked Questions
How is this different from the construction employment indicator already on the heatmap?
That series (FRED: USCONS) counts all construction workers — residential, nonresidential, and heavy and civil engineering. This one counts only residential building construction (NAICS 2361), the part that follows the housing cycle. Data centers, highways and office towers do not move it.
Why is it a lagging member of the Housing Cycle?
By measurement: on the Housing Cycle's record its 12-month change correlates most with single-family starts 9 months earlier, beyond the charter's six-month coincident boundary. Builders staff up after work is under way and cut back after it dries up. The cross-correlation table is published on the Housing Cycle methodology page.
How are the color thresholds set?
By a five-band quintile rule: the 20th, 40th, 60th and 80th percentiles of the series' own year-over-year history, rounded to a tenth of a percent. Today's color reads against the whole history, 1986 to today: the edges now are −3.3%, 1.7%, 4.1% and 6.5%. A past month's color — in the Time Machine and in the monthly table — reads against the history up to that month, as it would have read then, so the edges move a little as data arrives. A color exists once the series has ten years of history (here from 1996). Nothing is tuned to the current reading.
Why show the year-over-year rate instead of the headcount?
The headcount has grown with the housing stock and the workforce, by about two-fifths since 1985, so its level mixes that long rise with the cycle. For example, the level sat in the top fifth of its history in every month from January 2022 to August 2026. The rate is the cyclical reading; the level is stored for context, and the Housing Cycle page works from it.