Why do some releases move markets and others don't?

Test enough economic releases against enough markets and a puzzle appears. The jobs report makes the 2-year Treasury move at 2.35× a normal day. GDP — the broadest measure of the entire economy — barely registers at 1.12×. PCE, the Federal Reserve's officially preferred inflation gauge, comes in below a normal day. If importance drove market reactions, this ranking would make no sense. Something else is going on.

This page pulls together what our full results grid actually shows, and the pattern that explains most of it.

The ranking, in one view

Release-day move versus a normal day, at each release's most reactive market (robust results in bold):

The variable that explains it: surprise content

The pattern is not about how important a number is. It is about how much of it the market doesn't already know when it prints.

• The jobs report and CPI top the list because they are genuinely hard to forecast. Payrolls swing month to month; inflation turns have repeatedly caught forecasters out. When the number lands, it routinely differs from what was priced in — and that gap is what moves yields.
• PCE sits at the bottom for the same reason in reverse. It arrives about two weeks after CPI has already told the market that month's inflation story. By release day, there is almost nothing left to learn — so the dollar reaction is quieter than a normal day. The Fed's favorite gauge is the market's most predictable one.
• GDP describes a quarter that already ended, assembled largely from monthly data the market has seen. Treasuries barely react — old news, however comprehensive, is still old news.
• Jobless claims are frequent but tiny. Each weekly print updates the labor picture by so little that, even with 150 events of statistical power, no meaningful reaction shows up. Frequency dilutes surprise.

One sentence captures the whole grid: markets price expectations continuously, so a release moves markets only to the extent that it differs from those expectations — and some releases almost never do.

The second pattern: where on the curve it lands

The same grid shows where reactions concentrate. The releases that matter move the 2-to-5-year belly of the Treasury curve hardest — the maturities most sensitive to the Fed's path over the next couple of years, which is precisely what jobs and inflation surprises re-price. Meanwhile the 3-month bill barely moves on any release day (every ratio below 0.62): the front is pinned to the current policy rate, and no single data print changes that. The long end reacts progressively less as maturity grows — by the 30-year, no release survives correction. The full picture is on the Treasury curve map.

And still: direction stays unpredictable

Here is the discipline that keeps this from becoming a trading pitch. Across every one of these releases and every market — including the big movers — the direction of the release-day move is statistically indistinguishable from a coin flip. Volatility is forecastable in the sense that "jobs day will be a big day" is a safe statement. Direction is not, because the surprise that drives the move is, by definition, unknown in advance. A release can be reliably loud and reliably unpredictable at the same time — that combination is the single most consistent finding on this site.

What to take away

• Rank releases by surprise content, not headline status. Jobs and CPI are the genuine events; GDP and PCE are mostly confirmations.
• Watch the belly of the curve if you want to see the reaction — not the 3-month bill, not (usually) the 30-year.
• Distrust anyone who tells you the direction in advance. The data does not support it, for any release we have tested.

Historical statistics for informational purposes only, not financial advice. Results may vary with sample, period, and baseline definition. Sources: Federal Reserve Board and U.S. EIA series via FRED; release dates from BLS, BEA, Census, DOL, and the Federal Reserve — all public domain.