ReportGem ReportGem 中文

REAL-TIME GLOBAL RESEARCH

How Much Will Vol Go Up Under Warsh?

Published: 2026-06-24Institution: Morgan StanleyPages: 14Original language: EnglishEvidence page: 3

Research evidence excerpt

How Much Will Vol Go Up Under Warsh?

IdeaMimportant question is whether market sensitivity to payroll surprises has changed over

time.

Exhibit 2 cuts through the noise by reporting the estimated sensitivity of 2-year yields to

NFP surprises, along with 90% confidence intervals. While the estimates are subject to

uncertainty given the limited number of observations within each Fed regime, the trend is

clear: market sensitivity to payroll surprises declined steadily from the Greenspan era

through the pre-COVID period.

During the pre-COVID period of Powell's first term, the response of the 2-year Treasury

yield to a given payroll surprise was roughly half of its Greenspan-era level. While

multiple factors likely contributed to this decline, the increase in policy transparency and

forward guidance was clearly part of the shift.

The pandemic reversed this trend. Elevated inflation, higher policy rates, and a series of

large macro shocks increased the market's response to economic data, bringing sensitivity

to NFP back to levels comparable to the Greenspan era.

As Warsh reduces the role of forward guidance, we expect some further increase in

sensitivity. While the exact magnitude is uncertain, a full reversal of the decline since the

Greenspan era would imply an increase of up to 0.5x daily volatility per unit of data

surprise.

Exhibit 1: Normalized market response to NFP Exhibit 2: Market sensitivity to NFP surprises

surprises since 1997 declined since Greenspan

Greenspan normalized

2y move Bernanke

6 Yellen

Powell (pre COVID)

Powell (post COVID) 4

-2

-4

-6

-3 -2 -1 0 1 2 3

NFP surprise z-score

Source: Bloomberg. Morgan Stanley Research.

Different pattern for inflation data

Exhibit 3 plots normalized surprises in monthly CPI against normalized changes in the 2-

The English excerpt is extracted automatically from the cited source page and may contain layout or recognition errors. It is never batch translated.

Open report viewer