Trading NFP: how the US jobs report works and why it moves forex

Non-farm payrolls (NFP) is the headline number in the US Employment Situation report, published by the Bureau of Labor Statistics (BLS), normally on the first Friday of the month at 08:30 New York time (13:30 London for most of the year). It estimates how many jobs US employers added or lost in the previous month, alongside the unemployment rate and average hourly earnings. Because it feeds directly into what the Federal Reserve does with interest rates, it tends to produce one of the sharpest bursts of volatility in the forex month, with wide spreads and fast, often two-way price moves in the first minutes.

The general mechanics of how data releases move currencies are covered in how economic data moves forex, and how to read the calendar itself in how to use an economic calendar. This guide goes into the one release that most retail traders have heard of, and that catches out more of them than any other.

When NFP is released

The rule of thumb is "first Friday of the month, 08:30 New York". It is a good rule of thumb and a bad rule. The BLS sets the date by its own calendar, which is built around the survey reference period (the pay period that includes the 12th of the month) and the time needed to process the data. Most months that lands on the first Friday. Sometimes it does not:

• When the first Friday comes too early. If the first Friday is the 1st or 2nd of the month, the report is sometimes pushed to the second Friday. In 2027 the January report is scheduled for Friday 8 January, not Friday 1 January, which is New Year's Day anyway.
• Holidays. A federal holiday on or near the usual Friday can move the release, occasionally to a Thursday.
• Government shutdowns. When the federal government shuts down, BLS stops publishing. In autumn 2025 the shutdown delayed the September report, and the BLS cancelled the October 2025 employment report outright because the household survey could not be collected; the October payroll figure was published with the November data instead.

The confirmed 2027 dates for the first half of the year, from the BLS schedule, are 8 January, 5 February, 5 March, 2 April, 7 May and 4 June. Our quarterly economic calendars list each quarter, but always check the BLS release schedule itself, because dates can change.

The time in your time zone

08:30 New York is 13:30 London and 14:30 Central European Time for most of the year. Twice a year the US and Europe change their clocks on different weekends (two to three weeks in March, one week around the start of November), and during those gaps the release is an hour earlier for Europeans: 12:30 London, 13:30 CET. The market hours tool shows the current offset for your location, and the Q4 calendar guide works through an example of a payrolls Friday that falls in the November gap week.

What is inside the Employment Situation report

The report is several numbers from two surveys. Knowing which number comes from where explains a lot of the confusing price action on the day.

The establishment survey (employers)

The establishment survey, also called the Current Employment Statistics survey, asks a large sample of businesses and government agencies how many people they have on the payroll. It produces:

• Non-farm payrolls: the monthly change in the number of jobs, excluding farm workers, private household employees and a few other categories. This is the headline.
• Private payrolls: the same figure excluding government jobs. Some traders watch this more closely, because government hiring can be lumpy.
• Average hourly earnings: wage growth, given as a monthly and an annual rate. For the inflation outlook, and therefore for the Fed, this can matter as much as the job count.
• Average weekly hours: a quieter number, but a fall in hours can signal that employers are cutting back before they cut staff.

The household survey (people)

The household survey, the Current Population Survey, asks a sample of households about their own employment status. It produces:

• The unemployment rate: the share of the labour force that is out of work and looking for work.
• The participation rate: the share of the working-age population that is in the labour force at all.
• Broader measures of underemployment, such as the U-6 rate, which includes people working part-time for economic reasons.

The two surveys have different samples, different definitions and different error margins. It is perfectly normal for payrolls to rise strongly while household employment falls, or for the unemployment rate to rise even though payrolls grew. The household survey is smaller and noisier month to month. Markets usually treat payrolls as the headline and the unemployment rate as the second most important number, but a sharp move in unemployment can dominate the reaction, because it ties more directly into the Fed's employment goal.

Revisions: the part beginners ignore

Each report revises the payroll figures for the two previous months as more employer responses come in. These revisions can be large relative to the headline, and the market adds them to the surprise.

Worked example: a strong headline that reads weak

Illustrative numbers only:

• Consensus for this month's payrolls: +150,000.
• Actual: +190,000. On the face of it, a beat by 40,000.
• Previous month: revised from +160,000 to +110,000 (down 50,000).
• Month before that: revised from +140,000 to +120,000 (down 20,000).

Add it up: the level of employment is 40,000 higher than expected for this month, but 70,000 lower across the two previous months. Net, the economy has about 30,000 fewer jobs than the market thought before the release. A trader who reacted to the headline alone might have bought the dollar in the first second; the market, once it read the revisions, could reasonably go the other way. This is one of the commonest reasons for the "spike then reversal" pattern on NFP charts.

Annual benchmark revisions

Once a year the BLS realigns the payroll series with more complete records drawn from unemployment insurance tax filings. A preliminary estimate of this benchmark revision is published in late summer, and the final revision is incorporated with the January report published in February. Benchmark revisions can change the picture of the past year by hundreds of thousands of jobs, and the preliminary estimate has become a market event in its own right. You do not need to model it, but you should know it exists, because it can make the past year's "strong" or "weak" story look different overnight.

Why NFP moves currencies

The chain is the same one that runs through every data release: data changes expectations for the central bank, expectations change interest rate expectations, and rate expectations change the currency. The central banks and interest rates guide explains the Fed's dual mandate of maximum employment and stable prices. The jobs report speaks to both halves of that mandate: employment directly, and inflation through wages.

What matters is the surprise relative to the consensus, not whether the number is "good". Markets price the expected outcome in advance, often through interest rate futures and short-dated Treasury yields. A strong report that was fully expected can leave the dollar unchanged, and a "decent" report can sink it if the market was positioned for something better. The interest rate differentials guide shows why a shift in expected US rates relative to other countries' rates is what actually moves a pair like EUR/USD or USD/JPY.

Which number the market cares about changes over time

There is no fixed ranking of payrolls, unemployment and wages. When inflation is the Fed's main concern, average hourly earnings can drive the reaction and a big payroll number with soft wages can be shrugged off. When the market is worried about a slowdown, the unemployment rate and revisions take over. Read the commentary in the week before the release to understand which part the market is focused on; the economic calendar guide suggests a weekly routine for doing that without drowning in opinion.

Other data in the same week

The jobs report rarely arrives alone. Earlier in the week, the JOLTS job openings report, the ADP private payrolls estimate, the ISM purchasing managers' surveys (which include employment sub-indices) and weekly jobless claims all feed into expectations. None of these is a reliable predictor of the BLS figure, and ADP in particular measures something different and has often diverged from it, but together they shape how surprised the market is on Friday.

What happens in the first minutes

This is where the release becomes a practical, rather than economic, problem.

Spreads widen

Liquidity providers pull or widen their quotes in the seconds before 08:30 because they do not want to be filled at a price that is stale a second later. A spread that is normally a fraction of a pip on EUR/USD can widen to several pips, and on crosses and less liquid pairs much more. The why spreads widen guide explains the mechanics; for NFP the practical point is that the widening is predictable in timing even though it is not predictable in size.

Orders slip

Stop orders become market orders when triggered, and in a fast move the next available price can be far from the stop level. A stop placed 20 pips away might fill 30 or 40 pips away if price jumps through it. Limit orders do not slip in price, but may not fill at all. Slippage explained covers order behaviour in detail. Pending orders placed either side of the price to "catch" the move are a classic way to be filled on both sides in a whipsaw, paying the widened spread twice.

Price whips both ways

The first move is often driven by the headline alone, by algorithms that read it in microseconds. The second and third moves come as people read the revisions, the unemployment rate and wages, and as the bond market reprices. It is common to see a sharp move in one direction, a full reversal within minutes, and a settled direction only later in the New York morning, if at all. Nothing about the first candle tells you which of these you are looking at.

It is not only EUR/USD

The dollar is on one side of most of the market, so the report moves dollar pairs, gold (XAU/USD), US index CFDs and Treasury yields at the same time. Positions that look unrelated can all be the same bet on the dollar, which the currency correlation guide explains.

Managing risk around NFP

There are three broad approaches, and none of them involves predicting the number.

1. Stay flat. Close or avoid new positions before the release and come back once spreads have normalised. Many experienced traders do exactly this. It costs nothing but some missed opportunities.
2. Hold through with a position sized for the worst case. If you are in a longer-term trade, you can keep it, but size it so that a gap or slippage through your stop is survivable.
3. Trade the aftermath, not the print. Wait until the first wave has passed, spreads have returned to normal and the market has digested the details, then apply your normal approach. You give up the first move in exchange for normal costs and a clearer picture.

Trying to trade the first second is the approach that retail traders are worst placed for: they are slower than algorithms, they pay the widest spreads of the day, and their stops are exposed to the worst fills. The news trading risks guide goes through this in more detail.

Worked example: sizing a position that stays open through NFP

Illustrative numbers. An account of 5,000 US dollars, risking 1% per trade, which is 50 dollars. The trade is long EUR/USD, where one standard lot is worth about 10 dollars per pip.

• On a normal day the stop would be 25 pips away. Size: 50 divided by (25 × 10) = 0.20 lots.
• Because the trade will be open through NFP, the trader widens the stop to 60 pips, beyond the range of a typical post-release whipsaw. Size: 50 divided by (60 × 10) = 0.083, rounded down to 0.08 lots. At 0.08 lots, each pip is worth 0.80 dollars, and a 60-pip stop costs 48 dollars.
• Now allow for slippage. If price gaps through the stop and it fills 15 pips worse, the loss is 75 pips × 0.80 = 60 dollars, or 1.2% of the account instead of 1%.

The adjustment is the point: a wider stop with a smaller position, plus a margin for slippage. Keeping the original 0.20 lots with a 25-pip stop would have meant a normal NFP whipsaw could take the stop out, and a 15-pip slip would have cost 80 dollars, 1.6% of the account. The position size calculator does the arithmetic for any pair, and the position sizing guide explains the method.

Leverage makes all of this worse

CFDs on currency pairs are leveraged, so a fast move against an open position can eat through margin quickly. Under ESMA rules, EU retail clients are limited to 30:1 on major pairs and have negative balance protection, but a gap through a stop can still produce a loss larger than planned. Most retail CFD accounts lose money; see the risk warning and the leverage and margin guide.

Common mistakes

• Trading the headline without reading the rest. Revisions, unemployment and wages can reverse the meaning of the headline within seconds.
• Assuming the first Friday rule. Check the BLS schedule; the January 2027 report, for example, is on the second Friday.
• Using the wrong time. In the daylight-saving gap weeks the release is an hour earlier for Europe. Missing that by an hour is a common way to be caught with an open position.
• Straddling the release with pending orders. Both sides can be triggered in a whipsaw, with slippage on each.
• Keeping normal size with a normal stop. A stop sized for a quiet afternoon is likely to be hit by noise during the release.
• Treating ADP as a preview. It is a separate measure from a private company, and its relationship with the BLS figure has been loose.
• Reading one month as a trend. Payrolls are noisy and heavily revised; economists and the Fed look at averages over several months.
• Forgetting the rest of the portfolio. Gold, dollar crosses and US indices can all move on the same release.

How to prepare in practice

A simple routine for payrolls week:

1. On Monday, confirm the date and time in your own time zone.
2. Note the consensus for payrolls, the unemployment rate and average hourly earnings, and the previous month's figures that might be revised.
3. Check which part of the report the market is focused on (wages, unemployment, the headline) from mainstream financial news.
4. Decide in advance what you will do with each open position: close, reduce, or hold with a resized stop. Write it in your trading journal.
5. After the release, record what happened to spreads and fills at your broker. Over a few months that gives you a real picture of execution at your broker during news, which is more useful than any general rule.

Frequently asked questions

What time is the NFP report released?

The BLS publishes the Employment Situation report at 08:30 New York time. That is normally 13:30 in London and 14:30 in Central Europe, but an hour earlier for Europeans during the short periods in March and around early November when US and European clocks are out of step.

Is NFP always on the first Friday of the month?

Usually, but not always. The BLS sets the date based on the survey reference week and processing time, so when the first Friday falls on the 1st or 2nd the report may move to the following Friday, and holidays or a government shutdown can also shift it. Check the BLS release schedule rather than assuming.

Why does the dollar sometimes fall on a strong payrolls number?

Because the market reacts to the surprise against expectations and to the whole report, not to the headline alone. Downward revisions to earlier months, a rise in unemployment or soft wage growth can outweigh a strong headline. Positioning matters too: if the market had already priced a strong number, there may be little left to buy.

What is the difference between NFP and the unemployment rate?

Payrolls come from a survey of employers and count jobs. The unemployment rate comes from a separate survey of households and measures the share of the labour force looking for work. The two surveys can point in different directions in the same month.

Should I close my trades before NFP?

That depends on your strategy and your risk tolerance, and only you can decide. What is true in general is that spreads widen and slippage is more likely around the release, so a position held through it needs a stop and size that allow for that. Many traders choose to be flat, and some only trade once the first reaction has settled.

Which pairs move most on NFP?

Pairs with the US dollar on one side, especially the most liquid ones such as EUR/USD, USD/JPY and GBP/USD, along with gold and US index CFDs. The size of the move varies a lot from month to month and depends mainly on the size of the surprise.