Home › Learn › How to Read a Red-Folder Release
How to Read a Red-Folder Release: CPI, NFP, PPI and Jobless Claims
Everyone knows a red-folder release moves gold. Far fewer people can say what the three numbers on the calendar actually mean, why the market sometimes rallies on a bad print, or why the first minute is the worst possible time to have an opinion. Here is the mechanism, without the mythology.
The three numbers, and which one matters
Every economic calendar shows the same three fields. They are not equally important.
| Field | What it is | How much it matters |
|---|---|---|
| Previous | Last month's reading for the same series | Mostly context — but watch for revisions, see below |
| Forecast | The consensus of economists surveyed before the release | Critical. This is what is already in the price |
| Actual | The number that prints | Only meaningful relative to forecast |
The single most common beginner error is reading the actual on its own. A CPI print of 3.1% is not good or bad. If the forecast was 3.4%, it is a large downside surprise. If the forecast was 2.8%, it is a large upside surprise. Identical number, opposite market reaction.
What the market trades is the surprise: actual minus forecast. Everything the consensus expected was priced into gold before the release, in the days of positioning beforehand. Only the difference is new information.
Why gold cares about US data at all
Gold pays no interest. That one fact drives almost all of its reaction to US macro data.
If you hold gold, you give up whatever you could have earned holding an interest-bearing dollar asset instead. That giveaway is the opportunity cost of gold, and it rises and falls with real interest rates — nominal rates minus expected inflation. When real yields rise, holding gold gets more expensive relative to holding Treasuries, and gold tends to fall. When real yields fall, the opposite.
So the chain from a data release to a gold move runs like this:
Data surprise → changes the market's expectation of Federal Reserve policy → changes nominal yields and the dollar → changes the opportunity cost of holding gold → gold moves.
Every link in that chain is where the textbook reaction breaks down. The data can surprise without changing rate expectations, because the market had already discounted it. Rate expectations can change without moving real yields much, if inflation expectations move alongside. And gold can ignore all of it if something geopolitical is driving flow that day.
The four releases that matter most, and their direction
Almost every US release that moves gold prints at 08:30 US Eastern time — 18:00 IST in summer, 19:00 IST in winter.
CPI — consumer price inflation
Monthly, around mid-month. The textbook reading: hot CPI is bearish gold, because higher inflation raises the expected path of interest rates, which lifts nominal yields and the dollar. This reverses when the market believes the Fed will not respond — then hot inflation becomes a real-yield story and can be bullish. Watch core CPI, which strips food and energy, at least as closely as the headline; the Fed does.
Non-farm payrolls — the jobs report
First Friday of most months. A strong labour market supports tighter policy, so a strong payrolls number is typically bearish gold. This release is unusual in that it contains three numbers that can disagree: the headline job count, the unemployment rate, and average hourly earnings. It is entirely normal for jobs to beat while wages miss, and for the market to spend ten minutes deciding which one it cares about this month. That indecision is visible on the chart as a violent move in both directions.
PPI — producer prices
Inflation further up the supply chain, treated as a leading indicator of CPI. Same directional logic, usually a smaller reaction — except when it prints a day or two before CPI and the market treats it as a preview.
Initial jobless claims
Weekly, every Thursday. Individually small, but it is the highest-frequency read on the labour market, and in periods when the market is watching for cracks it can move gold more than its usual weight suggests. Remember the sign flips: higher claims means a weaker labour market, which is typically bullish gold.
Above all of these sits the FOMC decision itself — statement at 14:00 ET, press conference at 14:30 ET. The press conference frequently moves gold more than the decision, because the decision is usually already priced and the tone is not.
Why the textbook reaction fails so often
If you trade news for any length of time you will see gold rally on hot inflation and sell off on weak jobs, and it will feel like the market is broken. It is not. These are the usual reasons:
- It was already priced. If the whisper number differed from the published consensus — and it often does after a related release earlier in the week — the market positioned for the whisper, not the survey. A print in line with the survey is then a surprise in the other direction.
- Revisions outweighed the headline. Payrolls in particular routinely revises the previous two months, sometimes by more than the current month's print. A strong headline alongside a large downward revision to prior months is, on net, weak.
- The components disagreed. Headline beat, core missed. Jobs beat, wages missed. The market has to pick, and it sometimes takes several minutes.
- Two things released at once. At 08:30 ET several series often print simultaneously. If they point in opposite directions, the first move is close to noise.
- Positioning was extreme. If speculative positioning in gold was already heavily one-sided, a small surprise against that position triggers an unwind far larger than the data justifies.
- Something else was driving. Gold is a haven asset. On a day with a live geopolitical story, macro data is a footnote.
None of this means data is unreadable. It means the reading is probabilistic, and anyone who tells you a beat means down is describing a tendency as a rule.
What actually happens in the first three minutes
This is the part that costs people money, and it has nothing to do with direction.
- Spreads widen, often enormously. A gold spread that sits at 15–25 points can go to several hundred for a few seconds around a major release. Your entry is not where you think it is.
- Slippage is guaranteed, not possible. A market order at 08:30:00 fills at the next available price, which in a fast market can be a long way from the screen. A stop-loss is the same: it is an instruction to exit at the next price, not a promise of your level.
- Price frequently goes both ways first. A common sequence is a sharp move in one direction, a full reversal through the starting point, and only then the actual direction. If you are positioned before the release, both your stop and your target can be hit inside ninety seconds.
- Some brokers restrict execution around news. Check your own terms. Pending order rules, minimum stop distances and execution type can all differ in the minutes around a release.
The consequence: having the direction right is not sufficient. Plenty of traders read a release correctly and still lose on it, because the entry was filled 40 points worse than the screen and the stop was hit on the spike before the move they predicted actually happened.
Three defensible ways to handle a release
Stand aside. Close before it, reopen after. Unglamorous, and for most retail accounts it is the highest-expectancy choice available. You cannot lose on a release you were flat for.
Wait for the dust and trade the settle. Do nothing for the first five to fifteen minutes, let the spike and the reversal happen, and then trade the direction that holds once the spread is normal again. You give up the first move, which is the one you could not have captured cleanly anyway.
Be positioned in advance, sized for a gap. Legitimate, but only if the position is small enough that the worst-case fill is survivable, and only if you accept that your stop may not hold at its level. This is a size decision, not a prediction decision.
What is not defensible is the fourth option everybody actually takes: normal size, market order in the first seconds, stop where it would have gone on a quiet day. That is not a news strategy; it is a lottery with a wide spread.
Using the surprise mechanically
The news impact tool on this site applies exactly the logic above — it compares actual against forecast for CPI, PPI, payrolls and claims and tells you which way the textbook says gold should go, with the reasoning written out. It is worth being blunt about what that is: it is the null hypothesis, not a signal. It has no idea what was priced in, what positioning looked like, or what happened in the news overnight.
The useful way to use any such reading is as a starting point you then argue against. If the mechanical reading says bearish and gold rallies hard, the interesting question is not whether the tool was wrong. It is what the market knew that the arithmetic did not — and the answer to that is usually worth more than the trade.
I trade gold, forex and index CFDs, sell short-dated option volatility, and build the MT5 and TradingView tools published on this site. Everything here is written from my own screen time and my own losses — not from a content brief. More about how I trade and write.