Why CEOs are bad economic forecasters

In his book, The Beginning of Infinity, Oxford University physicist David Deutsch outlines the common mistakes we make in trying to understand the universe. One of them is “parochialism”, which in his words is when people “mistake quirks of one’s own, familiar environment or perspective for objective features of what one is observing”.

This is why humans thought the earth was flat for so long. It is also why the philosophy of ‘finitism’ exists in mathematics, which believes only finite abstract entities (such as numbers) exist. This small-minded view isn’t good for understanding, with Deutsch criticising finitism as “nothing but a project for preventing progress”. 

But parochialism isn't just an issue in science. In January 2023 on an earnings call, Tesla chief executive Elon Musk made some macroeconomic predictions, forecasting the US economy would have a “pretty difficult recession this year” and that deflation would cause Tesla's input costs to drop. 

It is not hard to see where Musk got this from. As the graph shows, up to that point Tesla had been growing revenue rapidly. Its main issue was whether its production could keep up with demand. However, all of a sudden demand growth started to slow. From Musk’s parochial view, the world would have looked like it was in a recession.

Of course, we now know this was not true. In 2023, the US’s nominal GDP grew around 6 per cent and another 5 per cent last year. As expected, most large US companies benefited from this with the S&P 500 earnings per share increasing 6 per cent over the past two years. Tesla was the outlier, rather than the rule.

Even two years on from Musk’s forecast, Tesla still hasn’t come out of this slump. Last Thursday, it announced it delivered 495,570 vehicles last quarter, up 2.3 per cent year-on-year but was below market expectations. Meanwhile, for the full year, it delivered 1.79mn, down slightly from the 1.81mn in 2023 and the first annual drop since 2011.

Its market share in the US and Europe is slipping, falling below 4 per cent and 3 per cent respectively. Tesla is trying to win back customers by slashing prices and is also offering “free supercharging for life” for its most expensive model. Given its 10 per cent operating margin (more than double most other car manufacturers), it can squeeze its competitors with price cuts. It also wants to solve autonomous driving, but to do that it needs cars on the road collecting data. Trading current profits to win that technological race would be worth it, ideally though, it wouldn’t need to be trading off between profits and volume.

This failure of parochialism isn’t exclusive to Musk. On July’s earnings call, former Starbucks chief executive Laxman Narasimhan referred to some variation of “challenging” or “cautious” consumers six times. Whereas, at Walmart chief executive Doug McMillon said that “so far, we aren’t experiencing a weaker consumer overall”. The difference is, at Starbucks same-store sales were down 3 per cent year-on-year while at Walmart revenue was up 4.8 per cent.

It turned out that the Starbucks' board thought it might be more of a localised problem, and a few days after the earnings call, Narasimhan was replaced by then-Chipotle chief executive Brian Niccol. 

This isn’t to criticise any of these chief executives' intelligence. They should have a parochial view because there are only so many hours in the day and shareholders want them to be obsessed with their business. But, just be aware that being an expert in one thing doesn’t guarantee knowledge of any other domain. And when a chief executive blames the economy for their performance, remind yourself what small part of the universe they are looking at.