The Economic Calendar: Which Releases Move Markets and When

For many years the arrangement at the Department of Labor was the same on the morning of a major release, and reporters were admitted to a locked room half an hour early, handed the document, and left to write against a clock, with communication lines physically controlled until the second the embargo lifted. At 8:30:00 the switch was thrown, the stories went out, and the Treasury market repriced faster than any human being could read a headline. The procedure has been revised more than once. The principle behind it has not: a small number of scheduled documents carry enough information to move the price of money, so the moment of their release is controlled to the second.

The schedule is worth knowing. The month repeats in roughly the same order, which means you learn it once and it serves for years.

The month in order

Release	Source	Timing	Tier
ISM manufacturing	Institute for Supply Management	First business day, 10:00 am ET	Second
ISM services	Institute for Supply Management	Third business day, 10:00 am ET	Second
Employment Situation	Bureau of Labor Statistics	Usually first Friday, 8:30 am ET	First
Consumer price index	Bureau of Labor Statistics	Mid-month, 8:30 am ET	First
Producer price index	Bureau of Labor Statistics	Around CPI, 8:30 am ET	Third
Retail sales	Census Bureau	Mid-month, 8:30 am ET	Second
Housing starts and permits	Census Bureau	Mid-month, 8:30 am ET	Third
FOMC decision	Federal Reserve	Eight times a year, 2:00 pm ET	First
GDP estimate	Bureau of Economic Analysis	Late month, 8:30 am ET	Third
PCE price index	Bureau of Economic Analysis	Late month, 8:30 am ET	Second
Initial jobless claims	Department of Labor	Every Thursday, 8:30 am ET	Second
JOLTS job openings	Bureau of Labor Statistics	Monthly, 10:00 am ET	Third
Beige Book	Federal Reserve	Two weeks before each FOMC meeting, 2:00 pm ET	Third

First tier means the release has regularly repriced the front of the Treasury curve and moved equity index futures visibly within the first minute, second tier moves rates when it surprises by a wide margin, and third tier is mostly read for context, though any of them can matter enormously when the market has fixed on the particular question it answers.

The distance from consensus

Prices already contain what is expected. A release moves markets through the distance between the actual figure and the consensus, and through what that distance implies about the path of policy.

Take an inflation example. Consensus looks for core CPI to rise 0.2 percent on the month, and annualized that pace is 1.002 ^ 12 - 1 = 0.0243, or 2.4 percent, close enough to a 2 percent target to be comfortable.

The print arrives at 0.4 percent. Annualized, 1.004 ^ 12 - 1 = 0.0491, or 4.9 percent.

Two tenths of a percentage point on a single monthly figure is the difference between a pace consistent with the target and a pace roughly twice it. That is why a number most people would describe as tiny reprices the entire curve, and why the reaction to the release is so much larger than the arithmetic of one month would suggest. The mechanics are in inflation and CPI explained.

Consensus is itself a median of forecasts collected from economists in the days before the release, a reasonable proxy for expectations and an imperfect one, since positioning can sit some distance from it. When a number lands exactly on consensus and the market moves sharply anyway, real expectations had already drifted away from the published figure, and the drift is the information.

What to read in the first minute

Read past the headline before forming any view.

For the jobs report, take the headline together with the revisions to the previous two months, then the participation rate, then average hourly earnings, all of which are explained in the jobs report explained, and for CPI the core figure and the shelter component come before the headline. For retail sales, the control group, which excludes autos, gasoline, building materials and food services, is the piece that feeds into the consumption estimate in GDP, and for the ISM surveys, new orders and employment carry more forward information than the headline index does.

For an FOMC meeting, the order is the statement read against the previous one, then the projections if it is a projection meeting, then the press conference, and the contents of each document are set out in how the Federal Reserve works.

Where the reaction shows up

The clearest read on what a release did to expectations is the short end of the Treasury curve, which prices the policy path more directly than anything else, and a hot inflation print lifts 2-year yields immediately while the rest of the curve reshapes around them. The yield curve tool shows the shape changing from week to week, and the interpretation is in the yield curve explained.

Equities respond second, through the discount rate, with the longest-duration companies moving most, a chain worked through in how interest rates affect stocks.

The Treasury auction schedule is a separate calendar and worth knowing, since a large auction that goes badly can move yields on its own with no economic data behind it at all.

Preparing for a release

Write down what you expect before the print. More usefully, write down what result would change your mind about something, because a number you cannot act on either way is a number you can read at lunch.

Know the period being covered. A mid-September CPI release describes August, and an August that ended before a move in oil prices tells you nothing whatever about September, and know whether the figure is seasonally adjusted, since the raw and adjusted series can differ enormously in January and September.

And expect revisions. Payrolls are revised twice and benchmarked annually, GDP is revised three times and then restated in annual and comprehensive updates, and anyone who built a conviction on a first print built it on a number with a real chance of changing.

One habit saves more confusion than any other. Read the release itself. Every agency publishes the tables alongside the headline, and the text names the distortions that affected the month, whether a strike, a hurricane, a change in collection method or an unusual seasonal factor. Two paragraphs of the source document take a minute and will regularly tell you that the figure everyone is reacting to describes something other than what the headline claims.

Where the calendar stops helping

Scheduled data is only one part. Central bank speeches between meetings, geopolitical events, oil supply decisions, credit incidents and earnings from the largest index members all arrive outside the calendar, and any of them can matter more than a monthly print. A calendar tells you when volatility is likely. It says nothing about when volatility is possible. That is always.

The other limit is that attention moves. A release that dominated one year can be ignored the next, and the only way to know which is which is to watch what has actually moved prices over the preceding few months, and to hold any tiering table loosely, including the one above.

See what all this data adds up to in recession indicators, and the Fed and the economy quiz covers the calendar and the releases on it.