Poor shopper sentiment remains to be simply discuss | USAEMALL.com

Good morning. Yesterday afternoon US President Donald Trump announced 25 per cent tariffs on imported cars, to go into effect April 2 and be “100 per cent” permanent. Trump also indicated that tariffs on pharmaceuticals and lumber were imminent. The president has backtracked on tariffs before, but the tone of this announcement struck us as particularly firm. Today’s market reaction will be interesting, and judging only by late trading yesterday, it will not be good. Email us: robert.armstrong@ft.com and aiden.reiter@ft.com. 

Vibe-cession revisited

On Tuesday, the Conference Board’s consumer sentiment survey told a familiar tale: people are worried about the economy. The survey’s index fell to a four-year low of 93 in March, below consensus estimates and well below the 110 reading from when Trump was re-elected in November:

Expectations of future business conditions, the labour market and income prospects fell hardest, from 75 in February to 65 in March — the lowest aggregate expectations reading since March 2013. This suggests broad pessimism about US growth, inspired by concerns over tariffs. Rosenberg Research notes that “the household views on the future are more downbeat now than at the depths of the 2020 pandemic recession”.

Does this — and all the other recent indicators of poor consumer, business and investor sentiment — matter to the economy?

So far, for all the bad “soft” opinion data, the “hard” activity data shows at most a very gradual slowdown. One might even call it a “normalisation” after the above-trend growth of the past few years. Since the equity correction that ended two weeks ago, poor sentiment is not showing up in investor’s choices, either. As we pointed out yesterday, even though Bank of America’s latest Global Fund Manager Survey showed a very bearish mood, capital flows into US equity funds are strong. Expressed preference is one thing, revealed preference another.

And, we should note, some of the soft data is improving. On Monday, we got flash estimates for the March US PMI survey. The composite rose two points to a three-month high, mostly reflecting an uptick in the services economy. Flash PMIs are not perfectly reliable. Last month’s PMI turned out to be much better than the flash estimate. But still: even the soft data is not all pointing down. 

Will the bad vibes turn into bad hard data? And if so, when? Consumer confidence is a leading indicator, in theory. When consumers see hard times ahead, eventually they will start spending less. But that day might never come. According to Michael Weber at the University of Chicago Booth School of Business, “people read too much into the survey readings”:

Oftentimes these big changes have ramifications, but month to month changes are benign. What really makes a difference is a big drop in consumer sentiment, like we got in March 2020. The recent readings are not nearly as bad. 

Looking at the University of Michigan consumer confidence index, the recent drops have not been as big as the big shock of Covid-19 and have not been sustained as long as the drops that preceded the 2001 and 2007-09 recessions: 

We have to be cautious with this data. Most recent recessions (‘00, ‘08 and ‘20) were caused by semi-exogenous shocks, and the drops in consumer sentiment that led up to the recession occurred after the market had already begun to react. And a big drop does not always lead to a recession — we saw a huge one-month drop in 2011 over fears about the US debt ceiling, which did translate to a slowdown, but not a recession. 

It seems there is a closer relationship between the stock market and consumer sentiment. Earlier in the 2000s, it looked like the market was leading, which makes sense: people feel worse about the economy when their 401Ks aren’t performing. But that relationship is imperfect. In recent years — and especially in the past quarter — it seems that sentiment has occasionally presaged jumps and dips in the market. But the direction of causality is hard to parse:

According to Joanna Hsu, who oversees the Michigan survey, absolute levels of consumer confidence are not as important as trends:

Direction is the most important . . . The sharp decline in consumer sentiment leading into mid-2022, attributable to the surge in inflation, was not accompanied by a sharp pullback in consumer spending. Since mid-2022, sentiment was on the rise — which translated to strong consumer spending, even though the level of sentiment remained below its historical average.

This is all the more true when one considers America’s growing partisan divide. As more and more Americans live in different media bubbles and identify more strongly with a political “team”, they know different facts and interpret the same fact very differently. “There is always a partisan switch at the start of any administration,” says Stephanie Guichard at the Conference Board — with the losing party, in this case Democrats, seeing a drop in consumer confidence at the start of the term (though the divergence is more stark this year than in 2021). Growing partisanship may make the readings less predictive. In 2022-24, for example, Republicans were grim about the economy but continued to spend merrily.

Partisanship may have also made sentiment sub-indicators less reliable. Dominic White at Absolute Strategy argued to us that there was never much of a reliable leading relationship between consumer aggregate expectations and aggregate growth. There was, however, a decent relationship between the individual expectation components with current economic trends, specifically the employment expectations component from the Conference Board survey. But even that has fallen apart in recent years. Here is the proportion of Conference Board respondents’ expecting fewer jobs in six months’ time, plotted against US jobless claims. That relationship no longer holds (chart from Absolute Strategy):

Note, however, that the March responses in the Michigan survey showed declining expectations across age groups, wealth levels, geographic locations, and — crucially — political affiliations. 

While the absolute level of Republican discontent is low, it has turned over. And independents have seen a notable drop in confidence. Other trends are concerning, too. Wealthy consumers contribute the lion’s share to consumption and have grown pessimistic over the past two months. 

In sum: consumer sentiment does not appear to have a stable relationship with economic growth — but both common sense and the available evidence suggest there is a relationship. We don’t see reason for panic about the economy in the weak sentiment surveys. But we’d feel a lot better if they were trending the other way. 

(Reiter)

One good read

You don’t know what the term premium is.

FT Unhedged podcast

Can’t get enough of Unhedged? Listen to our new podcast, for a 15-minute dive into the latest markets news and financial headlines, twice a week. Catch up on past editions of the newsletter here.