Macro Consensus Challenger: Jobs Report Points at September Fed Cut

Summary

• NFPs were lower than expected and previous months were massively revised down.
• Labour supply contracted further due to the decline in migrant workers and to native population aging.
• Unemployment aligned with expectations at 4.2%, up 0.1ppt relative to June and other indicators also indicated a weaker balance between labour demand and supply.
• Average hourly earnings growth aligned with expectations and accelerated by 0.1ppt to 3.9% YoY. Other wage indicators also pointed at stability.

Market Implications

• I still expect two-three Fed cuts in 2025 in line with market pricing but only one cut in 2026, below markets pricing about three cuts.

Labour Demand Much Weaker Than Initially Thought

The signs of emerging labour market weaknesses I noted in June’s employment report have become more pronounced in July’s report. They validate Chair Powell’s pre-NFP view that labour market downside risks are rising.

July payrolls surprised negatively at 73k versus 106k consensus (Chart 1). In addition, May-June was revised down 258k. As a result, the 6mma fell to 80k, down from April’s 180k.

Private payroll estimates at 83k were boosted by the birth/death model adjustments (Chart 2). Without the birth/death model adjustment private payrolls would have contracted by 174k (in recent years the birth/death model has been biased by COVID-related labour market reshuffling and by a surge in undocumented immigration).

All the increase in employment took place in the private sector. Total government employment contracted by 10k (Chart 3). Federal government employment continued contracting but this time the increase in state and local employment was not large enough to offset the federal contraction.

Most of the private sector employment increase was accounted for by healthcare, a defensive sector (Chart 4). The diffusion index was 51, indicating the number of expanding activities was only slightly above the number of contracting ones.

Manufacturing employment fell by 11k while private services increased by 96k (Chart 5). Post-pandemic, manufacturing employment has been contracting since Q1 2023.

Household survey employment fell by 260k and by 753k when adjusted for payroll survey definitions (Chart 4). Household survey data remains out of sync and noisier than payroll data.

Chart 1: NFPs Still Trending Down	Chart 2: Strong Boost From B/D Model
Chart 3: Total Government Employment Fell	Chart 4: Concentrated Private Job Growth
Chart 5: Manufacturing Jobs Shrunk Further	Chart 6: HH Survey Employment Noisy

Labour Supply Contracted Further

The labour force contracted by 38k following a 130k contraction in June, which partly reflects household survey data issues (Chart 7). Nevertheless, details show the negative impact of the end of the immigration surge (Chart 8).

The decline in participation accelerated (Chart 9). Average participation was 10bp below expectations at 62.2% and 50bp below its level of a year ago. This reflects population aging and the decline in immigration, as migrants have higher participation rates than native-born workers (Chart 10). Prime age (i.e., 25-54yrs) participation has also declined by 50bp relative to a year ago, even though it is not impacted by population aging.

Chart 7: Labour Force Fell Further	Chart 8: Immigration Surge Is Over
Chart 9: Participation Continued Falling	Chart 10: Higher Migrant Participation

Labour Demand-Supply Balance Weakened

This time signs of labour market weakening were stronger than June’s report. By contrast with June’s report, this time there were early signs of greater demand than supply weaknesses.

Unemployment (U3) rose 0.1ppt to 4.2%, aligning with expectations (Chart 11). The broader unemployment measure U6, which includes unvoluntary part-time workers and workers marginally attached to the workforce, increased by 0.2% to 7.9%. By contrast with U3, which has remained stable in a 4-4.2% range, U6 has been increasing since January. One of the drivers of the U6 increase has been the rising number of unvoluntary part-time workers (chart 12).

The growing imbalance between labour demand and supply was confirmed by the decline in the employment to population ratio (EPOP), which sidesteps unemployment definition issues (Chart 13). Even the prime age EPOP has been falling since end-2024.

The spread of African-American over white unemployment increased further, perhaps this release’s strongest sign of a weaker labour market (Chart 14).

The JOLTS ratios, unemployment and hires relative to openings, have been weakening on a 3mma basis (Chart 15).

While continuing unemployment claims have stabilised over the past month, the insured unemployment rate, a low volatility series, is up by the equivalent of one standard deviation since May (Chart 16). Also, the Richmond Fed recession indicator, which is based on the insured unemployment rate, has been rising steadily, though it remains far from the recession trigger.

Chart 11: Broad Unemployment U6 Rose	Chart 12: Unvoluntary Part Timers Up
Chart 13: Falling Employment/Population	Chart 14: Rising Minorities Unemployment
Chart 15: Weaker JOLTS Ratios	Chart 16: Rising Recession Indicator

Stable Wage Growth

Average hourly earnings growth accelerated by 10bp MoM to 0.3%, aligning with expectations (Chart 17). This aligns with Q2 ECI that was 10bp above expectations QoQ but unchanged YoY. Real wage growth was unchanged at 1.1% YoY (Chart 18).

Wage growth for job stayers and quitters was roughly equal (Chart 19). The wage premium of quitters over stayers has been falling since mid-2022, which could reflect, until recently, employers attempt to limit turnover and, lately, labour market easing.

Contrary to the Trump administration’s stated goals of lifting low wages, those continued underperforming higher wages (Chart 21). In addition, manufacturing wages continued losing ground to services (Chart 22). The relative decline of manufacturing wages since March 2025 is consistent with a loss of manufacturing productivity caused by US tariffs.

Aggregate wage income growth (i.e., wages times NFP and weekly hours) signalled personal disposable income stability though post-pandemic wage income growth from payroll data and national income have decoupled (Chart 20).

Market Consequences

Friday’s data is consistent with my big picture view of the US economy: exceptional policy uncertainty has caused a rise in business and consumers’ precautionary savings that has slowed growth, weakened business pricing power and contained the inflationary consequences of tariffs.

The revised employment numbers show businesses are much more cautious with employment decisions. And while the collapse of immigration has decreased labour supply, it has also decreased consumer demand, and ultimately, demand for labour.

Following Friday’s print, the market is back to pricing 2.3 cuts in 2025, with 85% chance of a September cut, from 1.3 cuts and 40% before the release.

I still expect two-three cuts in 2025 starting in September. My base case is for a 25bp cut based on unemployment remaining in its 4-4.2% range. This assumes wage growth remains benign, there are no signs of inflation contagion from tariffed goods to non-tariffed goods and services, and long-term BEs remain stable (also my base case). If these conditions hold, a pickup in core PCE from the current 2.7% would not deter the Fed.

Markets expect three cuts in 2026. I think this is too much as I expect the administration to stimulate the economy before the mid-term elections, which could worsen underlying cost pressures. I expect at most one cut next year.

(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs. You are not permitted to publish, transmit, or otherwise reproduce this information, in whole or in part, in any format to any third party without the express written consent of Macro Hive. This includes providing or reproducing this information, in whole or in part, as a prompt.)