ADP's Dovish Signal Flies, Will Friday's Nonfarm Payrolls Confirm the Cooling?

ADP released August private-sector employment data: only 38,000 jobs were added, significantly below market expectations of 47,000 to 48,000, and also below July's revised 46,000, marking the lowest level since January this year. The market reacted swiftly after the data landed. CME Fed funds futures showed the probability of a 25 basis point rate hike in September pulled back from earlier levels, and all three major indices closed higher. The Russell 2000, which had fallen more than 1.2% intraday, reversed to close up 1.13% at 2,953. Small-cap stocks, the most sensitive to interest rates, drew a big V-shaped reversal within a single day.

In a U.S. job market of over 100 million people, why can a small number of tens of thousands repeatedly stir U.S. stocks, bonds, and gold, and even influence the Fed's rate hike pace? How will the market react after today? And what impact will it have on next week's FOMC meeting decision?

What is the ADP "mini nonfarm" employment data?

ADP is one of the world's largest payroll service providers, processing payroll for over 500,000 U.S. businesses and approximately 26 million private-sector employees. This coverage equals about one-fifth of all U.S. private employment—roughly one in every six American workers has their paycheck processed through ADP's system. ADP consistently releases its report one to two trading days before the official nonfarm payrolls data, typically on Wednesday morning New York time. It is the first systematic employment signal the market receives each month.

In a country of 300 million, how can tens of thousands move the whole market?

In a fast-growing economy, job gains often reach hundreds of thousands or even millions, driven by incremental growth, so fluctuations of tens of thousands can indeed be ignored. But the U.S. is a mature, saturated economy where the job market has long entered a steady state, and monthly net additions are inherently small numbers. In such a market, marginal changes of tens of thousands are enough to judge whether employment is heating up or cooling down. Its value never lies in the absolute number, but in the marginal direction.

The details are more alarming than the headline: How big is the U.S. employment risk?

The total of 38,000 actually masks severe internal divergence.

The entire private sector is being propped up by services. Services added a net 48,000 jobs, with education and health services alone contributing 45,000; but manufacturing lost 17,000, and professional and business services—often seen as an economic barometer—shed 16,000. In other words, growth is highly concentrated in a few defensive sectors, while most investment and development pillar industries are actually contracting or stagnating.

The divergence by company size is even more striking. Large enterprises contributed 34,000 new jobs, but mid-sized companies added almost none, and small businesses added only 3,000—a performance that exposes the awkward reality of U.S. employment earlier than the headline data: small and mid-sized businesses are most sensitive to interest rates and financing costs, and their hiring intentions are the first to stall.

But the wage component sends a signal in the opposite direction. Base pay for job stayers rose 3.0% year-over-year, while total pay for job switchers surged as much as 7.3%. Switching jobs still pays significantly more than staying, indicating that companies are still willing to pay a premium to poach talent.

This unabated "job-switching" heat shows that nominal wage stickiness persists. Hiring is cooling, but wages are not cooling with it—exactly the combination inflation least wants to see, and the hardest card in the hawks' hand.

Can ADP's dovish cooling save the market?

The "dove" brought by the ADP data flew solo for most of the day but couldn't escape the shadow of geopolitics. After the data release, the 10-year Treasury yield was indeed knocked down a notch, but pricing power throughout the morning remained in the hands of the Strait of Hormuz—Trump threatened to retaliate "more forcefully," and Brent crude climbed above $95. The inflation risk from supply shocks travels farther than artillery shells. As long as the oil price shock persists, the Fed has no room to retract its "inflation-first" stance. That was the logic of the first half.

The turning point came after New York Fed President Williams spoke. His remark that he "does not see an unusual broadening of the impact from high energy prices" implies that if the energy shock does not spill over into core inflation, it is merely a one-time shock, not a trend. He added that inflation expectations remain "well anchored," wage growth is "subdued," and he is "not convinced" that another rate hike is needed.

Thus the same data was read twice in one day: in the morning as "inflation risk dominates everything," and in the afternoon as "employment is cooling while inflation spillover is limited." The market consequently staged a textbook intraday reversal, with the Russell 2000 swinging from a 1.2% intraday drop to a 1.13% close gain—a daily range of over 2.3 percentage points. Small caps are the most rate-sensitive end of the entire market, and their reversal magnitude is the true reading of the day's shift in rate expectations.

What to verify next: How will employment data steer the Fed?

Warsh's statement at Jackson Hole was blunt—"must be convinced inflation is clearly returning to target," otherwise "there is more work to do." On the inflation side there is no room for debate; employment should be the only policy variable. If employment weakens, tightening should slow.

This Friday's nonfarm payrolls release will be the next, harder test. Current market consensus is for an increase of 50,000 to 55,000 jobs, with the unemployment rate holding at 4.1%. Several major investment banks have also issued their own professional estimates: Goldman Sachs and Crédit Agricole both expect 65,000, while Wells Fargo is more optimistic at 80,000. When the data is released that evening, three scenarios may unfold:

Falling within the estimated range—unresolved. This is the most likely scenario and also the hardest to trade: it neither proves employment is collapsing nor proves it remains sufficiently resilient. Market focus may shift to the August CPI released in mid-September, and volatility may actually converge as the market waits for the next catalyst. In this scenario, sector rotation may be more telling than the broad index itself.

Stronger than the upper end of estimates—rate hike may be locked back in. This would challenge the narrative of rapidly weakening employment, and the probability of a September hike could climb further from 70%. The dollar and Treasury yields may strengthen in tandem. In this scenario, more vulnerable assets may include high-valuation growth stocks and small caps: rising discount rates hit valuations, not earnings themselves. Gold may come under pressure, though geopolitical risk premiums may partially offset the impact—so a grinding decline is more likely than a cliff-like drop.

Unemployment rate rises above 4.3%—doves finally get ammunition. This is an outcome that can materially suppress the probability of a rate hike, likely accompanied by a weaker dollar, a rebound in Treasuries, and lower real yields pushing gold higher. But beware of second-order risks—if nonfarm payrolls are negative for two consecutive months, the trading narrative may jump directly from "hike or hold" to "are we heading into recession?" At that point, even rising rate-cut expectations may not save the stock market, because what gets repriced will be earnings themselves.

How to guard against the war situation?

Last night's market action reminded us that once an exogenous shock of this magnitude—such as geopolitical events—intervenes, inflation risk becomes the dominant variable again, and the directional weight of employment data is temporarily suppressed.

If passage through the Strait of Hormuz remains obstructed and oil prices continue to rise, inflation expectations will step up again, and long-end yields and rate hike expectations will continue to climb together. At that point, the biggest concern may not only be a one-sided stock market decline, but a stagflationary pullback with both stocks and bonds falling—a replay of the late February to late March episode this year, when the S&P 500 drew down as much as 9.1% and the VIX spiked to 31. The 30-year Treasury yield will be a key threshold; once it breaks out decisively, all assets dependent on low discount rates may be repriced.

But if there is an unexpected de-escalation or ceasefire, the opposite direction is easier to miss. The second half of last night actually previewed this: once the judgment that "energy will not spill over" is confirmed, yields may fall, and Treasuries and growth stocks will have a repair window. But beware of gold: its current price contains two opposing forces—"geopolitical safe-haven premium" and "high real interest rate suppression." If the safe-haven premium disappears, gold may fall before stocks. Gold's deepest drawdown this year was 25% from its January peak of $5,318, and that lesson is not far away.

Data Sources

· ADP official press release: August private sector added 38,000 (including industry, company size, and wage breakdowns)

· CNBC: Private sector added 38,000 in August, below expectations

· Fox Business: Details of ADP August report and July upward revision to 46,000

· CNBC: New York Fed President Williams says yield surge stems from strong economic outlook

· Bond Buyer: Williams not yet convinced a rate hike is needed, advocates wait-and-see (including original quotes on energy non-spillover and anchored inflation expectations)

· Investrade: September 2 closing data (three major indices, Russell 2000, oil prices, gold, and sector performance)

· Barchart: Dollar retreats from two-and-a-half-week high, gold and silver close higher, September hike probability 65%

· Federal Reserve H.15: September 2 official closing Treasury yields (2-year 4.39%, 10-year 4.79%, 30-year 5.27%)

· FXStreet: Dollar remains strong initially after data, September hike probability briefly around 70%

· Blockonomi: Spot gold falls to three-week low intraday

· TheStreet: September 2 market action and Strait of Hormuz mine incident

· Forbes: August 31 CME FedWatch shows September hike probability 66%

· CNBC: After Jackson Hole, September decision becomes a "coin toss," hike probability rises

· CNBC: After August 7 nonfarm payrolls miss, hike probability declines

· TOPONE Markets: August nonfarm payrolls preview and three scenarios (consensus +50,000 to 55,000)

· Dallas Fed: Breakeven employment growth has fallen to near zero

· Pew Research: Comparison of ADP and BLS data coverage, methodology, and correlation

· Federal Reserve: FOMC meeting calendar (September 15–16 including dot plot)

Disclaimer: Market data in this article is as of the U.S. stock market close on September 2, 2026. This content is for informational and educational purposes only and does not constitute investment advice, an offer to invest, or a solicitation to buy or sell any financial product. The securities and market views mentioned are solely for illustrating index construction and passive investment mechanisms and do not represent BIT's recommendation or endorsement. Historical performance and market reactions to index adjustments do not indicate future performance. Investing involves risk; security prices may fluctuate, and investors may lose part or all of their principal. Please make prudent judgments based on your own circumstances. Relevant services and products are subject to laws, regulations, regulatory, and geographic restrictions in applicable jurisdictions.