Consumer confidence sinks to a 12-year low of 81.9 despite steady jobs data

Economists expected a rise. Add the 6.7-point fall back to September's reading and August comes out at 88.6. The consensus forecast of 89.2 therefore called for a gain of 0.6 points, and the index finished 7.3 points below it ^{[2][4][14][16][15]}.

The labour figures gave no sign of a drop coming. August had about 3.25 hires for every layoff, and the rates for openings, hires and layoffs each moved by only 0.1 percentage point from the month before ^[17][7]. "Employers are not as inclined to purge payrolls," Jeffrey Roach, chief economist at LPL Financial, wrote in comments shared with The Daily Upside ^[10].

Respondents were looking at prices. "References to prices, the high cost of goods and services, and oil and gas prices in particular, rose to new heights, reflecting September's surge in fuel costs," said Dana M. Peterson, the board's chief economist ^[11]. The Daily Upside's account is that a market with little hiring and little firing leaves workers stuck in jobs at relatively constant pay while prices rise around them ^[12]. JOLTS counts who keeps a job and who gets one, but not what the pay buys ^[5][12].

The evidence could resolve in different directions. Suppose fuel prices ease and confidence recovers while the jobs figures stay where they are. Then the survey caught a price shock, and the labour data was the better guide to spending all along. A turn from slow hiring to actual firing would mean the survey was early. Spending that holds up through a reading this low would mean the index recorded mood more than behaviour ^[1].

I think prices explain most of the gap. They are the cause respondents named, and the jobs figures do not track them ^[11][3]. The counter-case is that the layoff count records decisions employers have already made, while the index records what households say about decisions they have not yet made. The view is wrong if household spending holds up while confidence stays near 81.9 ^[2].