Ask an industrial economist how they help their company or trade group predict future demand, and the indicators they tick off will invariably fall into 2 buckets: a customer’s ability to afford a purchase and their desire to make that outlay. The two may intertwine, but they are actually distinct issues and never more so than at the top and bottom of an economic cycle.
Today we’ll focus on the American consumer’s “desire” to spend, as measured by the University of Michigan Consumer Sentiment Survey. Let’s start with where we stand today:
- Sentiment had been good over the past few years, running between 90-100 from the start of 2017 to February 2020.
- It then plummeted to a low of 71.8 in April 2020, bounced back to 78.1 in July but then dropped to 72 – 73 in July/August.
The upshot here is that the US consumer is not feeling great, which is totally understandable given both COVID and macroeconomic concerns.
Now, let’s look at the history of this indicator to make some broader observations. Here is a chart that shows the entire history of the U of M Consumer Sentiment survey results back to 1978:
Three important issues from this data:
#1: Once consumer sentiment “breaks” it remains fragile for a year or longer. Going from left to right on the graph above:
- The late 1970s – early 1980s (5 years): sentiment drops from 80 to 52 (a record low to this day), then bounces around 62 to 77 for 3 years, all due to the recessions caused by Paul Volcker’s fight with inflation.
- 1990 – 1993 (3 years): the Iraq invasion of Kuwait creates an oil shock, and even after Gulf War I concluded successfully sentiment was choppy until early 1994.
- 2001 – 2003 (3 years): the collapse of the dot com bubble, then the 9-11 terror attacks and finally Gulf War II kept consumer sentiment far from late 1990’s levels (the best the US has ever seen)
- 2008 – 2012 (4 years): the Financial Crisis, the Great Recession and the 2011 Greek debt crisis dampened sentiment for years, even after the stock market bottomed in March 2009.
Bottom line: consumer sentiment never “V-bottoms”. That doesn’t mean the US economy is stuck in a never-ending recession, however, as the chart above clearly shows.
#2: The difference between the COVID Crisis and the Financial Crisis/Great Recession in terms of its effect on consumer sentiment is pretty remarkable:
- Even at current trough levels of 72, sentiment is still better than the sub-60 levels of months in 2008 and early 2009 and again in 2011.
- The year-over-year percentage change in consumer sentiment for April – May 2020 was negative 26% – 28%, where April – August 2008 showed negative 28% – 32% comps.
Bottom line: while the US economy basically went into a statistical mini-Depression in the first half of 2020, consumer sentiment held up better than 2008 (and 1980, and 1990) because of aggressive fiscal and monetary stimulus.
#3: You probably noticed that US consumer sentiment had its all-time peak in early 2000 and perhaps wondered “what’s happened in the last 20 years that the American economy has never been able to replicate that level of confidence?” The January 2000 sentiment peak was 112.0. The next cycle peak was in January 2004 at 104. The peak for the last cycle was 101 in March 2018. It’s not a good trend.
Our answer: labor force participation, which also peaked in January 2000 at 67.3%. Here is the relationship between the percentage of American adults either working or looking for work (red line, left axis) and the Michigan sentiment data (blue line, right axis).
Bottom line: this is one reason we’ve been so focused on labor force participation, for it seems to set a cap on just how confident the American consumer can be long after a recession has passed.
Summing up: we don’t expect to see much good news on American consumer confidence for the remainder of 2020 because history says it takes time to improve after a shock; this does not, however, mean that the equity market is wrong in predicting its eventual recovery.