The preliminary May look at consumer sentiment from the University of Michigan stunk. The more you dig, the worse the smell.
The headline May Sentiment Index fell 9.8 points from the final April result; it’s at 67.4. The month-to-month decline is notably large. There have been 520 U of M sentiment surveys since 1980, the index has fallen by as much, or more than 9.8 points only 9 times prior to May 2024. Joanne Hsu, the Michigan Survey Director, wrote, “This 10 index-point decline is statistically significant and brings sentiment to its lowest reading in about six months. This month’s trend in sentiment is characterized by a broad consensus across consumers, with decreases across age, income, and education groups…While consumer had been reserving judgment for the past few months, they now perceive negative development on a number of dimensions. They expressed worries that inflation, unemployment and interest rates may all be moving in an unfavorable direction in the year ahead.” The steepest declines in the Michigan survey do not occur in a vacuum, these instances tend to happen when there is stuff going on. For instance, during the thick of the Great Recession in October 2008, or the opening months of the pandemic in March and April 2020. Additionally, there was sizeable stumble was in the aftermath of Hurricane Katrina in September 2005, and a couple of others, such as March 2011 or December 2012, that do not stand out as broad-based and dreadful milestones, but were indeed periods of economic weakness, with some degree of uncertainty. The interesting thing now is that the May dip in sentiment occurs at a time of sub-four percent unemployment, consistent payroll growth and above trend GDP. One could ask, why the dissonance in May? However, after a little digging through the details of the May Michigan report, one would realize the divergence between consumer sentiment and the headline economic data did not begin in May. For instance: Despite job growth and a very low jobless rate the expectations of US households for their finances over the next couple years is at a twelve-year low. The last time this survey was so low it was the result of several years of slack payroll increases, followed by a multi-year period of job losses.
The UM asks its survey participants if it’s a good time to buy a house. The corresponding index just set a life of survey low at 26; so, in a nut shell, no, not a great time to hunt a house. This attitude did not begin in May, just got a bit more adamant early this month.
So, to the extent that it’s true that the Michigan Survey does not have an epic stumble in a vacuum, it is probably a good idea to have your head on a swivel.
One more thing: Then there is the credit card bill; that’s what makes up most of what is known as Revolving Consumer Credit. The total bill passed a $1 trillion for the first time in December 2007, but retreated from there because that’s the month the Great Recession began and people stopped spending. Revolving Credit once again surpassed the trillion-dollar mark in late 2017, dipped below that level in the early pandemic period, but there is no looking back now. As of March, the price tag is up at $1.337 trillion.
High use of credit can be a sign of confidence, but the story evolves. At some point credit replaces cash flow and outstanding credit is burdensome and then the high balances become a sign of desperation. As you can see in the chart below the change from a year ago spikes in the good times and falls when the recession hits. Revolving credit has never spiked like this before, and it has turned.
But there’s more to the story. The Buy-Now-Pay-Later (BNPL)platforms, such as Affirm Holdings and Klarna are a growing source of consumer credit. But these companies don’t report on the size of their loans and these totals are not included in the Fed’s consumer debt reports. Their share has grown every year since at least 2020 and, according to a report from Bloomberg, this credit “is projected to reach almost $700 billion globally by 2028.” Wells Fargo analysis says there’s about $46 billion of BNLP “phantom debt” outstanding now. A Bloomberg report on a Harris Poll survey adds interesting color to this component of consumer credit, something that may be weighing heavily on consumer sentiment.