Educational content only. This is not financial advice or an investment recommendation. The author is not a registered financial adviser. Nothing here is a forecast or a trading signal.
In short: Four scheduled US statistical releases account for most of gold’s calendar-driven moves: the FOMC decision, the consumer price index, the monthly employment report and the core PCE price index. They matter for the same underlying reason, and it is not the one usually given. Each one changes what the market expects future interest rates to be, and that expectation is what sets the cost of holding an asset that pays nothing.
Why a statistic moves a metal
Gold pays no coupon. Hold an ounce for a year and you still have one ounce. So the cost of owning it is whatever a safe interest-bearing alternative would have paid you over the same period, after inflation. That is the real yield, and it is covered in detail in How Bond Yields Move Gold.
A Treasury yield is not a fixed quantity. It is a price, and like any price it reflects what buyers and sellers currently believe about the future — in this case, about the path of the policy rate over the life of the bond. Change that belief and the yield moves, which changes the real yield, which changes the cost of holding gold.
That is the whole chain, and it is worth writing out in order, because most explanations skip the middle:
- A release changes what the market expects the Federal Reserve to do.
- That expectation is embedded in Treasury yields and in inflation breakevens.
- The difference between those two is the real yield.
- The real yield is the opportunity cost of holding gold.
A release that does not change expectations does not move much of anything, however dramatic the headline it produces. This is why an inflation number can be alarming and the gold price can sit still.
The number is not what matters. The surprise is.
This is the single most common misreading of release-day price action, and it is worth being blunt about it.
Before any of these four releases, economists publish forecasts and news agencies aggregate them into a consensus. Markets trade ahead of the release, so by the time the number lands, the consensus is already in the price. What is not in the price is the gap between the consensus and the actual figure.
So the quantity that moves markets is the surprise, not the level. Inflation at 3% is not inherently bullish or bearish for anything. Inflation at 3% when 2.7% was expected is a different event from inflation at 3% when 3.3% was expected, even though the published number is identical.
Two consequences follow, and they explain most of the confusion around release days.
A good number can move the price against you. If the figure is strong but weaker than expected, the surprise is negative even though the level is high.
The consensus is not a fact. It is a survey of a self-selected group of forecasters, compiled differently by different agencies, and it is revised up to the release. Reconstructing after the event what “the market expected” is far less precise than it is usually made to sound.
The four releases
| Release | Published by | How often | Typical time (ET) |
|---|---|---|---|
| FOMC statement | Federal Reserve | 8 times a year | 2:00 pm |
| Consumer Price Index | Bureau of Labor Statistics | Monthly | 8:30 am |
| Employment Situation | Bureau of Labor Statistics | Monthly | 8:30 am |
| Personal Income and Outlays (PCE) | Bureau of Economic Analysis | Monthly | 8:30 am |
All four are free, published on a pre-announced calendar, and released to everyone at the same instant. None of them require a subscription to anything.
1. The FOMC decision
The Federal Open Market Committee sets the target range for the federal funds rate at eight scheduled meetings a year. The statement is released at 2:00 pm Eastern on the second day, followed shortly afterwards by the Chair’s press conference.
The rate decision itself is usually the least informative part of the day. By the time the committee meets, the decision has normally been signalled well in advance and is already priced. What is not priced is everything around it:
- The changed wording. The statement is edited rather than rewritten, so comparing it line by line against the previous one isolates what the committee chose to change.
- The Summary of Economic Projections. Published at four of the eight meetings, it includes the chart of individual rate projections commonly called the dot plot. It is a snapshot of where participants think rates are heading, not a commitment, and the Fed says so explicitly.
- Dissents. A vote that is not unanimous says something about how settled the committee is.
- The press conference. Price moves during the press conference are frequently larger than the move on the statement itself.
The minutes of each meeting follow about three weeks later and occasionally move markets on their own, when they reveal that the discussion was less settled than the statement suggested.
2. The consumer price index
The CPI measures the change in prices paid by urban consumers for a basket of goods and services. It arrives around the middle of the month and covers the month before.
Two versions are published together. Headline CPI includes everything. Core CPI strips out food and energy, not because those do not matter to households — they obviously do — but because they are volatile enough to obscure the underlying trend over short periods.
For gold, the CPI is the most double-edged of the four, and this is where intuition fails most often. An upside inflation surprise pushes inflation breakevens up, which on its own lowers the real yield. But the same surprise also pushes expectations of the policy rate up, which raises nominal yields. The real yield is the difference between the two, so which way it ends up moving depends on which effect is larger — and that is not knowable in advance.
The practical implication is the point of this whole section: “inflation is up, so gold goes up” is not a mechanism. It is an assumption, and it is wrong roughly as often as it is right. The Treasury yields tool shows both components separately, which is the only way to see which one actually moved.
3. The employment report
Formally the Employment Situation, released at 8:30 am Eastern and most often on the first Friday of the month, though the schedule varies: in 2026 three of the eleven releases fall on another day. It is built from two separate surveys, which is why it sometimes appears to contradict itself.
The establishment survey asks employers about their payrolls and produces the headline jobs number. The household survey asks households about their own employment and produces the unemployment rate. Different samples, different definitions, different revision patterns. A month in which payrolls look strong and the unemployment rate rises is not an error; it is two surveys measuring different things.
Three things inside the report deserve more attention than the headline:
- Revisions to the previous two months. Every report revises the two before it, and a large revision can more than offset the new month’s figure. The headline is reported; the revision frequently is not.
- Average hourly earnings. Wage growth feeds the inflation side of the Fed’s mandate, so it can matter more for rate expectations than the job count.
- Labour force participation. The unemployment rate can fall because people found work or because they stopped looking. Those are opposite events with the same headline.
4. The core PCE price index
The Personal Consumption Expenditures price index is published by the Bureau of Economic Analysis inside the monthly Personal Income and Outlays release, towards the end of the month, covering the month before.
It gets less press than CPI, and it is the one the Fed actually targets. The committee’s 2% objective is defined on the PCE price index, and the core measure is watched closely as a guide to the underlying trend. If you want to know what the Fed is looking at, this is it.
Because it lands two weeks after the CPI for the same month, much of its content is already inferable from CPI and PPI components by the time it is published, which is why it typically produces a smaller reaction. Typically is not always.
Why CPI and PCE give different answers
Two official inflation measures for the same month routinely differ, and PCE inflation has tended to run below CPI inflation. Three structural differences explain most of the gap.
- Scope. CPI counts what urban consumers pay out of pocket. PCE also counts spending made on households’ behalf — most significantly employer-paid and government-paid healthcare, which is a large category.
- Weights. The two use different expenditure weights, updated on different schedules. Housing carries a far larger share of CPI than of PCE, so housing costs dominate the CPI in a way they do not dominate the PCE.
- Formula. PCE allows for consumers substituting between goods as relative prices change; the CPI’s fixed-basket construction allows for less of it. Mechanically, this pushes PCE below CPI.
Neither index is the correct one. They answer different questions, and the headline comparison “inflation is X” is incomplete without saying which measure produced X. The inflation-adjusted gold price tool uses CPI, and says so on the page, for exactly this reason.
Reading the reaction afterwards
The useful work is not predicting the number. It is understanding, afterwards, what moved and why — which is something anyone can do with free public data, a day or two later, with no subscription and no hurry.
- Split the yield. Did the nominal 10-year move, the breakeven, or both? A gold move alongside a falling real yield has a straightforward explanation. A gold move with the real yield flat does not, and needs one from somewhere else.
- Check the positioning. The weekly Commitments of Traders report shows whether speculative positioning was already stretched before the release. A crowded position and a surprise in the other direction is a different event from a neutral position and the same surprise.
- Check the dollar. Gold is quoted in dollars, so part of any move is the currency rather than the metal.
- Allow for the release not being the cause. Markets move for reasons that have nothing to do with the calendar. Attributing every move on a release day to the release is the most common error in this kind of analysis, and it is unfalsifiable unless you check.
What this framework will not do
It will not tell you what any of these numbers will be, and knowing the number in advance would not tell you how the price would react anyway, because the reaction depends on the consensus and on positioning as much as on the figure.
These four also do not explain everything. Central bank purchases, physical demand, currency moves and geopolitical stress all push on gold and none of them arrive on a schedule. There are extended stretches where the relationship between the macro releases and the gold price is weak.
And the calendar itself is not guaranteed. Statistical releases depend on federal agencies being funded and operating; when that lapses, data is delayed or, occasionally, never published at all. There is no consumer price index for October 2025, which is a hole that no amount of analysis can fill in.
Where to get the calendar
Each agency publishes its own release schedule a year ahead, free, with exact dates and times. Those are the primary sources, and they are more reliable than any aggregated economic calendar, which are frequently wrong about times and occasionally wrong about dates.
Sources
- Board of Governors of the Federal Reserve System, FOMC calendars, statements and minutes, and the Summary of Economic Projections published with them.
- Board of Governors of the Federal Reserve System, Statement on Longer-Run Goals and Monetary Policy Strategy, for the 2% objective defined on the PCE price index.
- US Bureau of Labor Statistics, CPI release schedule and Employment Situation release schedule.
- US Bureau of Labor Statistics, CPI questions and answers, and Handbook of Methods, for the establishment and household surveys and for the differences between CPI and PCE.
- US Bureau of Economic Analysis, Personal Consumption Expenditures Price Index and the Personal Income and Outlays release schedule.
- Release dates and times are published by each agency and change from year to year. Check the agency’s own schedule rather than this page before relying on a time.