Yves here. This is a tidy, persuasive analysis which is relevant to the debate among investors, economists, and the great unwashed public about the severity of the downside risk of the AI bubble. It looks at the features of the 1920s boom that could have produced the Depression. The authors argue that the robust consumption of the first part of the decade, produced in large measure by electrification plus consumer uptake of new products like cars, radios and washing machines, petered off. The economic driver became investment in “process innovation” as in greater production efficiency that investors believed would stoke yet more growth. But “build it and they will come” or here, “build more with less,” is not a good growth engine when markets are saturated.
By Harold Cole, James Joe-Jin Kim Professor of Economics University Of Pennsylvania, Stefano Cravero, PhD student in Economics University Of Pennsylvania, and Jeremy Greenwood, Professor of Economics University Of Pennsylvania.…