Saving for the machines
From Jesús Fernández-Villaverde: When machines can perform a growing share of tasks, growth depends on two margins: how the task structure affects the returns to accumulating machines, and how strongly capital accumulation responds to those returns. I argue that, in a standard task model, capital supply governs the size of the growth boom associated with automation, […] The post Saving for the…
When machines begin to handle an increasing portion of tasks, the growth of an economy hinges on two key factors: how the structure of tasks influences the returns from investing in machines, and how responsive capital accumulation is to those returns. According to Jesús Fernández-Villaverde, capital supply determines the extent of the growth surge brought about by automation, while the nature of tasks dictates the length of this growth period and the subsequent changes in wages.
The same technological path that results in an annual growth rate of 11.7%, under the assumption that households optimize their savings, yields only 2.2% when the savings rate is set at the U.S. level of 2025, which is 23%. This suggests that slow growth could stem from insufficient capital accumulation rather than technological constraints. Moreover, the level of savings tends to increase primarily due to the rise in the capital share caused by automation, not out of frugality on the part of households.
The impact of automation on interest rates is significant. The ten-year real yield surges from 5.1% to 13.6% as a result. If households are in a state of constant consumption, their earnings would be reduced by the equivalent of two-fifths to two-thirds of their current consumption. Temporarily maintaining this consumption trend would only incur a cost of up to a fifth of the GDP.
This analysis is encapsulated in a thread discussing the paper, along with accompanying slides. The topic of potential bottlenecks in growth due to automation continues to be explored.
Written by urgent.news from Marginal Revolution's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
