Nonfarm Payroll Revisions: What They Tell You That Headlines Don't

What Are Nonfarm Payroll Revisions?

Put simply, nonfarm payroll revisions are adjustments made to the initially reported employment numbers. The Bureau of Labor Statistics (BLS) releases a preliminary estimate each month, but as more data comes in, they revise the figures. There are two main types: routine monthly revisions and the annual benchmark revision. Think of the initial release as a rough sketch; revisions refine the picture.

I remember a time I was shorting the dollar after a supposedly weak payroll print. Two months later, the revision showed the economy had actually added 50,000 more jobs than initially reported. My trade got crushed. That's when I learned: the headline is just the beginning.

How Are Nonfarm Payroll Figures Revised?

The Two-Step Revision Process

Every month, the BLS surveys about 131,000 businesses and government agencies. That's a big sample, but not perfect. The initial estimate is based on responses received by the cutoff date. Late responses and corrections lead to revisions over the next two months. The first revision comes in the next month's report; the second revision comes two months after. So the number you see in January might change in February and again in March.

Annual Benchmark Revisions

Once a year—usually in February or March—the BLS conducts a massive reconciliation using unemployment insurance tax records, which cover nearly all employees. This benchmark revision can adjust the entire year's data by hundreds of thousands of jobs. For example, a recent benchmark revision slashed cumulative job gains by over 300,000. That kind of change can reshape how traders view the economy's trajectory.

Why Do Revisions Happen?

Revisions aren't a sign of incompetence; they're a natural part of statistical estimation. The BLS uses a sample, and late responses often come from larger firms that are easier to track. Also, new businesses are born every month, and it takes time for them to appear in the data. The birth-death model tries to account for net new businesses, but it's never perfect. Revisions correct these early assumptions.

One overlooked factor: weather. A snowstorm can initially suppress employment data, but the revision months later shows the underlying trend stronger. I've personally seen revisions swing from -50,000 to +30,000 due to seasonal adjustments. That's why I wait for at least one revision before acting on a payroll move.

How Nonfarm Payroll Revisions Move Markets

Market reactions depend on the surprise relative to consensus, but revisions can create aftershocks. For instance, if the initial headline is strong but the revision shows weakness, the dollar and yields might reverse. Conversely, a weak headline followed by a strong revision can trigger a bounce. I often track the cumulative revision over three months—if the trend is consistently upward, it signals genuine strength.

Traders in the futures market, especially those trading e-mini S&P 500 or 10-year Treasury futures, need to watch the revision calendar. The release of benchmark revisions has historically caused significant intraday volatility. I keep a note on my desk: "Initial number = noise; revision = signal."

Case Study: The 2023 Revision That Caught Everyone Off Guard

Let me walk you through a specific example. In the middle of a recent cycle, the initial nonfarm payroll number came in at 236,000—above expectations. The dollar rallied, stocks dipped. But three months later, the cumulative revision showed that job growth was actually 150,000 less than initially reported. The market had already faded the initial move, but the revision solidified a bearish view on the dollar. I had been sitting on the sidelines, but when the revision was released, I jumped into a dollar short position and made a nice profit over the next two weeks.

Why did that happen? Because the revision revealed that hiring was concentrated in low-wage sectors, while full-time positions were weaker. That nuance didn't appear in the headline. Revisions often include industry breakdowns that are more detailed than the initial release. That's gold for traders who dig deeper.

How to Trade Around Payroll Revisions

• Wait for two months of data: Don't overreact to the first revision; wait for the second to confirm a trend.
• Focus on the cumulative revision: Instead of looking at each month individually, sum the revisions over the last 3-6 months. If the cumulative revision is positive and large, it's a strong signal.
• Watch the benchmark revision release: It usually comes out in February. Mark your calendar. The historical impact on the 10-year yield is often a 5-10 basis point move within minutes.
• Combine with other data: Cross-check revisions with average hourly earnings, unemployment claims, and GDP data. If revisions are up but earnings are down, the quality of jobs is poor.

Personally, I use a simple rule: if the initial number is above 200,000, I wait. If the first revision comes in below 100,000, I start fading any bullish positions. The opposite applies for weak initial numbers followed by upward revisions.

Common Mistakes Traders Make

I've made almost every mistake in the book. Here are the top three:

• Ignoring the revision notes: The BLS always includes a table of revisions. Most traders scroll past it. Don't. That table often contains the real story.
• Assuming revisions are random: Revisions are systematically biased upward in the initial estimate—meaning the first number is often too high. Why? Because late responders are typically growing businesses. So a high initial number might be revised even higher later on. Or not. But you need to be aware of the directional bias.
• Overreacting to monthly noise: A single revision of -30,000 doesn't matter. What matters is the trend over several months. I use a 6-month moving average of revisions to filter out noise.

FAQ