American confidence with Chinese characteristics?

Ed Yardeni recently highlighted the surge in small business confidence, earnings and employment plans as part of a scenario of what could go right.


Beneath the surface, there were a number of contradictions that were evocative of official Chinese economic statistics.

A confidence bifurcation

The first problem is how consumer and business confidence is measured. Jim O’Sullivan observed a significant partisan divide in the Bloomberg consumer comfort figures.


There was a similar partisan divide in University of Michigan consumer expectations.


Opinions vs. hard data

The NFIB small business population sample is skewed heavily small-c conservative and Republican. It is therefore no surprise that small business confidence and investment plans are surging. But how reflective is that of reality?

Instead of focusing on survey opinion data of how respondents felt, I analyzed hard figures of actual small business employment from Paychex, which processes payroll data for a wide swath of American small businesses. To my surprise, the Paychex Small Jobs Index was actually declining. Wait, what?


The wage data was nothing to write home about. The pace of wage gains was decelerating, despite the widespread complaints from the NFIB survey about finding qualified applicants. Weekly hours have been rising, but pace was not significantly different from levels during the Obama years.


Are these American business confidence statistics with Chinese characteristics? You tell me.

3 thoughts on “American confidence with Chinese characteristics?

  1. In the last Paychex chart the weekly hours are up but the hourly pay is down. I would suspect there is something in the Paychex data survey that is askew.

    Of course, the other charts show a Dem non-belief of the current expansion. Without further analysis I would suspect this is, at least partially, a result of the Dem partisans being able to sell their message to their followers.

    1. I am not surprised to see a partisan divide in the sentiment data. I’ve seen various versions of:
      1) My guy is in the White House and it’s going to be great
      2) This guy is in the WH and he’s going to be a disaster
      I’ve seen that said about Bush II, Obama, and it’s no surprise that Trump is that divisive.

      Unless you can tell me there is something in the Paychex data set that is unrepresentative of either the small business community, e.g. skewed towards or away from certain industries, or the economy, this sounds like a case of “watch what they do, not what they say”.

Comments are closed.