If the economy is so good, why aren’t consumers happy?

The always-opinionated former US Treasury Secretary and current Harvard Professor Larry Summers coauthored a paper in February titled “The cost of money is part of the cost of living: new evidence on the consumer sentiment anomaly.” Despite a good labor market and falling inflation, consumer sentiment has remained below what would be expected in the current economic environment. Should we believe the inflation and employment data or consumers’ lying eyes?

Summers et al finds that low sentiment is due to including higher borrowing costs in inflation statistics and shows in the paper that much of the “sentiment gap” can be explained once borrowing costs are included — another win for trusting our lying eyes.

One way to visualize the sentiment gap is to plot the so-called Misery Index, which adds together unemployment and inflation against sentiment (Fig. 4). On the bright side, the gap seems to be closing, but the other side of the coin is that sentiment and stock market returns tend to be negatively correlated (low sentiment leads to high returns).

Headwinds are for bonds, tailwinds are for equities

The 10-year Treasury is currently priced nearly perfectly at a 2.5% Fed funds rate (Fig. 5). If the Fed funds rate is at 2.5%, the 10-year Treasury should yield 4% based on the historical relationship between the two. The easiest way to see this is to picture the salmon line in Fig. 5 slowly migrating to the light blue line as the Fed cuts rates.

If the economy remains resilient and the Fed can only cut to 4.25%, what does that mean for the rest of the yield curve? When using historical data on the “normal” shape of the yield curve, a 4.25% Fed funds rate implies a 5.7% yield on the 10-year Treasury. We’ve labeled that line “upper bound.” What does the market expect? That’s the darker blue line. The Fed cuts to 4.25%, and the 10-year Treasury gravitates upwards to 4.6%. Either way, our lying eyes are telling us that there’s still a headwind for bonds, even if the Fed’s forecasts won’t admit it (yet). Our asset allocation targets, which we use as a starting point for customization, are underweight bonds in favor of low duration, high quality private debt.

Unlike bonds, equities have a tailwind – particularly in the US (we’re overweight US equities in a globally diversified equity portfolio). The factors driving above-trend economic growth will also support corporate profit growth. In our opinion, US equity valuations remain at the higher end of reasonable based on current interest rates, but the positives (above-trend income growth, accelerating capital expenditure spending for AI and digital infrastructure, swelled household balance sheets for baby boomers, prime household formation years for millennials, etc.) outweigh the negatives. Even so, we see no reason to try to time equity markets. All-time-highs are usually great times to invest; historically, they are no more likely to lead to a meaningful pull back than investing at any other time.

Google & AI & Gemini, oh my! [2]

We wrote about artificial intelligence last week, so we won’t reiterate those thoughts here, but it’s worth noting the disastrous launch of Google’s rebranded AI, now called Gemini. Users quickly reported fairly odd results – For example, Gemini was uncertain as to whether Hitler or Elon Musk had caused more harm to humanity, it proclaimed that determining whether pedophilia is wrong required a nuanced answer and turned out to be incapable of drawing white people even in the appropriate historical context (e.g., Gemini decided medieval knights were Asian women, Nazis were black, etc.). There are dozens of more examples in a similar vein that you can find online. Overall, it’s a reminder that there will be winners and losers along the AI path, that human influence (politics, regulation, etc.) will be heavily embedded in AI models, and that large, successful firms can somehow completely mess up their most important projects without anyone realizing it. For his part, Google CEO Sundar Pichai called the issues “completely unacceptable, and we got it wrong.” Alphabet’s (Google’s parent company) share prices declined 7% since Gemini’s update on February 15th, versus an increase of 1.5% for the S&P 500 over the same period.

A very different AI story comes from the Swedish online shopping firm Klarna. Klarna integrated AI-powered customer assistance (AI handled the customers instead of human agents) and reported that it handled 2.3 million customer conversations in 35 languages, did the work of 700 full-time agents, was more accurate in error resolution than human agents, received similar customer satisfaction scores as human agents, resolved inquiries more quickly than humans, and will drive a $40mn profit improvement for Klarna in 2024. That’s the potential of AI.

[2] This is a reference to “Lions & Tigers & Bears, oh my!” from The Wizard of Oz.