What prediction markets expect the next U.S. data releases to show, beside the Street consensus for the same print; and what markets, professional forecasters, the Fed and CBO expect for this year and next. The market numbers are market expectations — the mean of the distribution that Kalshi's prices imply, as of the time shown — not official data and not forecasts of ours.
Loading…
Each row is one upcoming print. "Markets expect" is the Kalshi market's expected value of that print; "Street consensus" is the median forecast of professional forecasters polled before the release, for the same print, where one has been published (it usually appears in the week before the release); "Past week" is how much the market's expectation moved over the last seven days. For GDP, the Econvitals Composite and the Atlanta Fed's GDPNow are shown beside the market, from the GDP Trackers page.
Annual forecasts from four sources, side by side. Each row names the measure it is on, and a cell is shown only where that source publishes that measure on that basis — a dash means it does not. A cell marked with a note is on a stated different basis from the row; nothing is converted from one basis to another.
Market expectations. Kalshi lists each release as a ladder of yes/no markets ("real GDP growth above 2.5%", "above 3.0%", …) or as a set of ranges. The prices across the ladder are a probability distribution over the print, and the number shown is its mean, computed from the midpoint of each market's bid and ask. A ladder whose middle has no tight two-sided market is not shown. The distribution view shows the probability the market puts on each possible print. These are prices at which people are willing to bet, not official statistics and not a forecast of ours.
What each market settles on. The quarterly GDP market settles on the advance estimate, at an annual rate; the calendar-year GDP market on annual-average growth as first published with the fourth-quarter advance estimate; CPI and core CPI on the seasonally adjusted monthly change rounded to one decimal; payrolls on the first-print monthly change; unemployment on the U-3 rate. The year-end federal funds market is a ladder on the upper bound of the target range after the December meeting; the number shown is the expected midpoint of the range, the basis the Fed's own projections use.
Comparability. The Fed's Summary of Economic Projections gives growth and inflation as fourth quarter over fourth quarter, unemployment as the fourth-quarter average and the funds rate as the target-range midpoint at year end. CBO publishes both fourth-quarter and calendar-year figures; its funds rate is the annual average of the effective rate. The Survey of Professional Forecasters publishes unemployment as an annual average and as quarterly readings, inflation fourth quarter over fourth quarter, and real GDP as annual-average levels, from which annual-average growth is computed the way the Philadelphia Fed does — growth of the median levels, with the prior year's level as it stood when the survey was taken. Where a cell is on a different basis from its row (a December unemployment print beside a fourth-quarter average; December-over-December CPI beside fourth quarter over fourth quarter; an annual-average funds rate beside a year-end one), it carries a note saying so.
Timing. The market numbers are as of the time stamped on the page and refresh several times a day; the surveys refresh when a new one is published — the SPF each mid-quarter, the SEP at four FOMC meetings a year, CBO once or twice a year.