Does the GDP release move Treasury yields?

Not measurably. Across 12 advance estimates the 10-year yield moved about 4.8 basis points on release day versus 4.3 on a normal day — about 1.12 times normal, nowhere near significant (p=0.30). GDP describes a quarter that already ended, and the monthly data has usually told the story first.

Why is the GDP sample so small?

The advance estimate comes out once a quarter, so a 2023-2025 window contains only 12 of them. That is too few to detect a modest effect, and we say so rather than reporting the null as if it were settled. Second and third estimates are revisions of a quarter already reported and are not included.

Does GDP move Treasury yields?

Gross Domestic Product is the broadest measure of the U.S. economy, and its quarterly release gets headline treatment. But GDP has a feature that sets it apart from monthly data: it describes a quarter that has already ended, and it arrives after most of the monthly indicators that feed into it are already known. That raises a natural question for the bond market — is there any new information left in GDP by the time it prints? We tested the 10-year Treasury yield's reaction across 12 quarterly advance estimates.

Verdict

Direction (release day): Not Significant

Volatility (release day): Not Significant

Volatility (5 days): Not Significant

GDP measures the previous quarter — and by the time it lands, the market already knows. On GDP-release days the 10-year Treasury yield moves about as much as a normal day (1.1×, not significant), with no consistent direction.

Bottom line: GDP is backward-looking — the quarter is already over and well-estimated, so Treasuries barely react.

Bar chart: the 10-year Treasury's average absolute move is 4.8 basis points on GDP release days versus 4.3 on normal days — essentially a normal day; 5-day windows are similar, 11.0 versus 10.5.

Average absolute 10-year yield move around GDP advance estimates versus normal days (N=12, low sample). Chart: Macro or Noise, from Federal Reserve Board data via FRED (public domain).

What the numbers say

On release day, the 10-year yield moved about 4.8 basis points versus 4.3 on a normal day — roughly 1.12× a normal day, which is not statistically significant (p=0.30). Direction is a wash as well (about +1.7 basis points, p=0.39). Over a five-day window there is no elevation either. On every dimension, GDP day looks like an ordinary day for the 10-year.

The rate-sensitive 2-year moves a little more (about 1.37×), but even that does not clear the significance bar in this sample.

Why a "top-tier" number barely registers

Two things work against GDP as a market mover, and both are informative:

It is backward-looking and well-anticipated. By the time the advance estimate lands, the market has already seen three months of jobs, inflation, retail sales and more. GDP largely stitches those together, so it rarely contains a surprise big enough to reprice the curve.
The market-moving version is the advance estimate. We test the first (advance) release, which is the one traders react to. The later second and third revisions move yields even less — they are updates to old news.

As with PCE, the lesson is that a release moves markets in proportion to the new information it carries, not its headline importance.

A note on sample size

Only 12 quarterly advance estimates fall in our 2023–2025 window, so this test has less statistical power than our monthly studies — a modest real effect could go undetected. We report the low sample honestly rather than overstating a "no effect" conclusion; what we can say is that no reaction large enough to detect showed up in these 12 releases.

What this means in practice

GDP day is, on the evidence so far, a non-event for the 10-year. It moves about like a normal day.
Backward-looking data carries less market punch. The story was mostly told by the monthly numbers already.
Read this one with its small sample in mind. Twelve observations is a real result but a modest one.

None of this is advice — it is a description of what 12 releases actually did, and results can change with a different sample, period, or definition.

The data

Dimension	Horizon	Value	Baseline	Test stat	p-value	Verdict
Direction	release day	1.7 bps	0.0 bps	0.90	0.388	Not Significant
Volatility	release day	4.8 bps	4.3 bps	1.12	0.295	Not Significant
Volatility	5 days	11.0 bps	10.5 bps	1.05	0.383	Not Significant

Methodology

Events (N): 12 GDP advance estimates.
Window: 2023-01-26 → 2025-12-23.
Measurement: change in the 10-year yield from the prior close to the close at the end of the holding window (look-ahead protected).
Baseline: the unconditional distribution of same-length moves across all trading days.
Tests: one-sample t-test of the signed move against zero (direction), and a bootstrap of the absolute move against the baseline (size).

Caveats

Only 12 quarterly advance estimates fall in this window (2023–2025), so statistical power is limited — a modest real effect could go undetected.
Uses advance (first) estimates only — the market-moving release; later revisions move yields even less.
Surprise-versus-forecast conditioning is not applied — only the release-day reaction is measured.
Historical statistics for informational purposes only, not financial advice. Results may vary with sample, period, and baseline definition.

Related tests

Do jobs reports (NFP) move Treasury yields? — the monthly data that GDP summarizes.
Does CPI move the 10-year Treasury yield? — a forward-looking release, for contrast.
Which U.S. data releases move the Treasury curve? — GDP against everything else.

Source

10-Year Treasury Constant Maturity, Federal Reserve Board via FRED (Tier A, U.S. public domain) — DGS10.
GDP (advance estimate) release dates, U.S. Bureau of Economic Analysis — Gross Domestic Product.