What it means
Monthly jobs reports are built from surveys of existing employers, but a survey cannot catch businesses that have just opened or just closed, because they are not yet in the sample or have already left it. Yet business births and deaths are a major source of job creation and destruction.
To fill the gap, the Bureau of Labour Statistics applies a birth-death model: it estimates the net employment effect of openings and closings from historical patterns and adds that estimate to the survey results. The model matters most at turning points.
In a recovery, new firms form faster than history suggests, and the estimate can understate job growth; in a sharp downturn, deaths outpace the model, and it can overstate employment. This is why benchmark revisions later restate the figures against fuller data from tax records, sometimes moving the totals by large amounts.
The birth-death adjustment is not guesswork inserted for effect, it is a necessary correction with known limits. For managers and investors, the lesson is to read monthly jobs data with humility.
Headline payroll numbers include a modelled component that will be revised, so single-month surprises should not drive big decisions. Watching the trend over several months, along with the annual benchmark revisions, gives a truer picture of how many businesses and jobs the economy is actually creating.
The mechanics have improved over time. The Bureau of Labour Statistics now updates its birth-death forecasts quarterly rather than annually, so the model reacts faster when business formation shifts.
Once a year the whole series is benchmarked to near-universe counts from unemployment-insurance tax filings, and the gap between the survey-plus-model estimate and that benchmark is published, giving users a direct measure of how large the modelled component's error turned out to be. Analysts who follow the series learn to expect the largest revisions around recessions and recoveries.
In practice
Real-world examples.
Example
In a typical expansion month, the birth-death model adds a positive adjustment because new firms historically create more jobs than closing firms destroy. The adjustment is published separately, so analysts can see how much of a headline figure came from the model.
Example
During a sudden recession, business closures outrun the model's assumptions, and later benchmark revisions show employment was weaker than initially reported. Economists who relied on the first release had to reverse their earlier readings of the labour market.
Example
An economist waits for the annual revision against unemployment-insurance tax records before declaring a change in the labour market trend. Those tax records cover nearly all employers, which is why they are treated as the benchmark the monthly survey is judged against.
Formula
Calculation
Net birth-death adjustment = estimated jobs from business births - estimated jobs from business deaths. Reported employment change = survey-measured change + net birth-death adjustment.
Worked example with invented figures.
- The monthly survey shows employment up by 120,000 jobs.
- The model estimates 90,000 jobs from new businesses and 60,000 lost from closures, a net adjustment of 90,000 - 60,000 = +30,000.
- Reported change = 120,000 + 30,000 = 150,000 jobs.
- If the annual benchmark against tax records later shows the true gain was 100,000, the revision is 100,000 - 150,000 = -50,000, so the first release overstated the gain by 50,000 / 150,000, or about 33%.Case study
Seen in the real world.
This fictional, illustrative example follows Alder & Finch, an invented staffing firm that built its sales forecast on monthly payroll headlines. During a recovery year the figures looked weak, and the firm cut recruiter hiring, only to watch benchmark revisions later show that job growth had been far stronger, with new-business births running ahead of the model. Alder & Finch had under-hired into a boom. Afterward, its analysts began tracking twelve-month trends and revision history instead of single prints, and their capacity planning errors shrank noticeably.
It also added a rule that no capacity decision rests on fewer than three consecutive months of confirming data. Smoothing helped immediately. Monthly gains of 90,000, 150,000 and 60,000 jobs average (90,000 + 150,000 + 60,000) / 3 = 100,000, a steadier signal than any single month. The firm and its figures are invented.
Watch out
Common mistakes.
- Treating the birth-death adjustment as manipulation, when it is a documented statistical correction for a real gap in survey coverage.
- Trading aggressively on a single monthly jobs print while ignoring that the modelled component will be revised against fuller records.
- Assuming the model fails only in downturns, when recoveries can produce equally large misses in the opposite direction as business formation surges.
Questions
People also ask.
Why is a birth-death model needed at all?
Monthly employer surveys cannot capture firms that just opened or closed, and those births and deaths are a large share of job change, so a model estimates their net effect.
Does the adjustment make jobs data unreliable?
No, but it makes single months noisy. Trends over several months, plus later benchmark revisions against tax records, give a dependable picture.
When is the model most likely to miss?
At economic turning points, when business formation or closure runs far from historical patterns, which is why early-recession and early-recovery readings get revised most. Watch the annual benchmark revision as the scoreboard: it shows exactly how far the monthly estimates drifted and in which direction.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%