Our "Four Readers Agree on 65,000" Broke in Eleven Hours. The Page That Broke It Is Older Than Ours.
Eleven hours ago we published that four independent readers carried the October payrolls consensus at 65,000 to the thousand.
A page published the day BEFORE ours says economists expect 59,000 — and names six houses running 50,000 to 95,000, a 45,000 spread against the 40,000 we reported. The unanimity we spent a section dismantling never existed, and the evidence was already on the web when we wrote it. Barclays 50,000 with unemployment falling to 4.1%, Deutsche Bank 55,000, Wells Fargo 65,000, RSM US 95,000 — and the two houses closest on jobs are furthest apart on the unemployment rate. The prior we built the argument on is also two numbers: September's consensus was 90,000 or 93,000, so the miss was 61,000 or 64,000. ADP is worse, because it is a revision question: September is 90,000 as published and 86,000 revised, so October's month-on-month change is zero or minus 4,000, both correctly derived. One reader files today's forecast under the label "September". The stack: 12:15 Pill, 13:30 US payrolls AND Canadian employment in the same minute, 15:00 four events at once — Michigan sentiment, Michigan inflation expectations, the Ivey PMI and the New York option cut — then Ueda at the IMF at 16:00. Options price about 1.49% for payrolls; one-week dollar-yen vol is 7.78% from 6.23% this morning. And a Canadian house forecasting 6.5% unemployment is really forecasting 6.485%: a 4.5bp fall invisible at one decimal.