Bolstering Economic Growth with AI

If you haven’t already, please check out part 1 and part 2 of this series before reading this article.

Why AI Will Increase GDP

If AI can improve productivity and reduce factor prices, production will increase, decreasing prices and increasing aggregate demand (Acemoglu & Restrepo, 2018; Capital Economics). This potential to increase economic output is reaffirmed by Damioli, Roy, and Vertesy (2021) who found that AI patent applications have significant positive effects on labor productivity in the service industry, particularly among SMEs.

Prerequisite Conditions

However, as an economy, labor productivity growth has been slow since the 1970s even among industries that have invested heavily into digital technologies (Brynjolfsson 1993). Brynjolfsson, Mitchell, and Rock (2018) offer a potential explanation for this in the age of AI, being that complementary technologies and labor skills must catch up to AI developments for productivity gains to be realized. This is supported by the finding that AI-adopting firms, which are more likely to possess such complementary skills and technologies, do have higher labor productivity than similar establishments that have not adopted AI (Acemoglue et al., 2022).

AI can certainly promote economic growth, depending on the nature of automation and its implementation (Aghion et al., 2017). Aghion, Jones, and Jones found several conditions under which balanced growth is possible with capital shares under 100%, (see Figure 2). Still, they do note that slow productivity growth in essential service industries may restrict the extent of such growth due to Baumol’s “cost disease.”