Chart #1 —
Double-edged job market statistics
The past week saw the publication of several highly anticipated statistics on the US job market. At the start of the week, jobless claims for April came in at 8.06 million, against expectations of 8.4 million. The gap between the published and anticipated figures is relatively wide. A drop in job creation has only been observed 3 times since the year 2000, suggesting that the US job market is indeed slowing down.
But on Friday, another statistic blurred the message. The BLS survey of government entities and private sector companies (yellow bars on the chart below) reported 272,000 jobs created for May, exceeding consensus estimates (+185,000). These job creations are well above those of April (+165,000).
The labour market remains resilient despite the Fed's restrictive monetary policy, reducing the likelihood of a rate cut in September. However, the labour force has shrunk, pushing the US unemployment rate from 3.9% in April to 4.0%, the first time above 4.0% since February 2022.
More bad news for the Fed (and bond markets): wage growth surprised on the upside, perhaps due to reduced labour supply, which could cause bottlenecks given still-steady job creation. Wage growth remains a hard-to-control source of inflation, rising at a rate of 4.1%, which is still far too high for the Fed.
It should be noted that the household employment survey shows the opposite result to the above statistics: it reveals a sharp drop in the number of people employed, with 408,000 fewer jobs (see white bars below). Full-time jobs fell by 625,000, the biggest drop in full-time employment since December 2023, while part-time jobs rose by 286,000.
Between household and establishment surveys, the figures seem to be out of sync. This makes analysis of economic data quite difficult.
Source : Bloomberg
NB: The household survey includes agricultural workers, unincorporated self-employed workers, unpaid family workers and private household workers among the employed. These groups are excluded from the establishment survey. The household survey includes people on unpaid leave among the employed. This is not the case for the establishment survey.