Macro Consensus Challenger: Early US Labour Market Softening, Fed to Cut in September

Summary

• May NFPs were higher than expected but employment growth is slowing once revisions are considered.
• Unemployment was unchanged based on poor quality household (HH) survey data.
• Other data from a variety of sources show the labour market has started loosening.
• This aligns with my expectations, so I still expect two-three Fed cuts in 2025.
• However, labour market loosening is happening slower than I expected. Therefore, I move my call for a first Fed cut to September from previously July.

Market Implications

• I expect the Fed to cut two-three times in 2025 starting in September, against markets expecting 1.8 cuts.

Labour Demand Slows

I am refocusing my data reviews on how they relate to the consensus macro narrative, starting with Friday’s employment report. Financial and betting markets as well as government and private economists have converged on a soft-landing narrative. Unemployment and inflation are expected to peak in Q4/Q1 and recession risks are seen at only around 30%.

I think this is implausible given the extreme disruption and policy uncertainties the Trump administration’s policies have created. I expect growth to be weaker than consensus, allowing inflation to remain on the benign trajectory consensus expects.

Friday’s employment report superficially fits the consensus narrative. Unemployment remains low, and wage inflation is benign. However, signs already exist of labour market softening.

At 139k, NFPs were higher than the 126k consensus. But negative two-month revisions at 95k more than offset this (see our event monitor for market impact). YoY growth is the better trend indicator, which has been slowing and stands at 1.1%, below 2019’s average of 1.4% (a bigger difference than it looks because total NFP are 160mn; Chart 1).

The private sector accounted for the NFP increase as a rise in state and local employment offset a 20k decrease in federal employment (Chart 2).

The breadth of job gains was limited to two sectors, health and education and leisure and hospitality, accounting for all the increase (Chart 3). This is the third successive month when these two sectors account for most of the NFP change.

Lastly, employment in the HH survey contracted by 696k and adjusted for payrolls definitions, by 791k. These numbers seem implausible and indicate continued issues with HH survey data. Despite a 2.2mn positive revision to January, HH survey employment remains unaligned with payrolls (Chart 4).

Chart 1: NFP Growth Trending Down	Chart 2: Private Sector Driven NFP
Chart 3: Job Gains Concentrated	Chart 4: HH Survey Employment Implausible

Labour Supply Contracts

The labour force contracted by 625k, partly reflecting HH survey data issues (Chart 5). Nevertheless, other elements of the survey signal unfavourable labour supply trends. For instance, average participation has been on a downward trend for the past two years and even prime age participation, which demographics do not impact, has been falling for the past year (Chart 6).

Also, undocumented immigration has stopped (Chart 7). As a result, the share of migrants in the labour force is stabilising (Chart 8).

Chart 5: Labour Force Fell	Chart 6: Participation Is Falling
Chart 7: Illegal Inflows Have Stopped	Chart 8: Slower Migrant Labour Supply Growth

Early Labour Market Softening

With a contraction in employment roughly equal to the labour force contraction, the unemployment rate was unchanged at 4.2% (Chart 9). Given HH data issues, I looked for corroboration within the survey as well as from other data sources and found early signs of labour market weakening.

Minorities unemployment and average wage growth also indicated labour market stabilisation (average hourly earnings). The spread of minorities over white unemployment remained at levels consistent with expansion (Chart 10). Wage growth was above expectations at 40bp MoM against the 30bp consensus, but YoY wage growth was unchanged from April’s revision at 3.9% (Chart 11)

Other indicators signalled early labour market softening. Unemployment changes increased over the past three months (Chart 12). In addition, a breakdown of these changes between layoffs, quits and those first looking for a job (entrants) shows a greater role for layoffs and a smaller role for quits and entrants, a pattern indicating a looser labour market.

Similarly, the broader unemployment rate U6, which includes involuntary part-timers and persons wanting work but are not actively looking, has been rising by more than the narrower U3 (Chart 9). JOLTS ratios of unemployed per opening and of actual hires per openings have also been rising for the past three months, signalling of a looser labour market (Chart 13).

Equally important are early signs of a labour market regime change. Labour market flows fell post pandemic. Following the acute labour shortages during the pandemic, employers were hiring and firing less as well as providing incentives for employees to quit less (Chart 14). This was shown with lower temp work and a sharp decline in the wage premium for job switchers.

Traditionally, a smaller share of temp work had been associated with greater unemployment (Chart 15). That relationship broke down in 2022-23 as temp work fell but unemployment fell or remained extremely low. Also, there was anecdotal evidence of employers converting temp workers into permanent staff, likely fearing renewed workers shortages.

However, while the decline in temp work has continued, unemployment has been increasing over the past few months. In addition, labour market flows have accelerated, which could signal employers are more willing to let employees go because labour shortages are less concerning. The negative temp work-unemployment correlation could be returning.

In this context, the zero wage premium for job switchers may have become driven more by labour market softening than by employers’ fear of being unable to find workers (Chart 16).

Chart 9: Broad Unemployment Is Rising	Chart 10: Minorities Unemployment Still Low
Chart 11: Stable Wage Growth	Chart 12: Layoffs Bigger Driver of UR Change
Chart 13: JOLTS Ratios Show Softening	Chart 14: Labour Market Flows Accelerating
Chart 15: Lower Temp Work Implies Softening	Chart 16: Less Incentive to Switch Jobs

Market Consequences

This early labour market weakening is consistent with the growth slowdown (Chart 17; due to measurement issues, the GDP ex net exports and inventories change is a more reliable trend indicator than headline GDP).

Labour market and GDP growth weakening are set to accelerate because tariff implementation has only started. Average effective tariffs increased to about 6% in April, up from 2% in February, but still far from the 18% or so the administration is targeting.

May’s manufacturing employment shows the limited tariff impact so far (Chart 18). Tariffs should hit manufacturing hardest because it is the most import-intensive part of the economy and the most exposed to trading partners retaliation. Yet May’s contraction was historically small. The gist of the tariff impact is yet to come.

I think a slowdown at such an early stage in tariff implementation implies a higher recession risk than the 30% consensus assumes. My distribution of economic outcomes is unchanged. I still expect 40% risk each for a self-correcting economic weakness and for a recession, and a 20% chance of a US Treasury crisis. Therefore, my view of two-three Fed cuts this year is unchanged, against markets pricing 1.8.

However, the labour market weakening is happening more slowly than I expected. Therefore, I shift my expectation for a first cut to September’s FOMC, from previously July. Markets are currently pricing about 85% chance of a cut by September’s FOMC.

Chart 17: GDP Growth Is Slowing	Chart 18: Manufacturing Jobs Yet to Receive Full Tariff Hit

(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.)