Knowledge Problems with Discretionary Monetary Policy

Abstract

When questioned by US Representative Alexandria Ocasio-Cortez about the Federal Reserve's inaccurate forecasts of long-run unemployment, Chairman Powell replied, "We have learned thatyou can't identifythis is something you can't identify directly. I think we have learned its [the natural rate of unemployment] lower than we thoughtit's substantially lower than we thought" (C-Span 2019). This was a surprisingly candid admission of the limits of expert knowledge by the head of one of the most important central banks in the world. 1 The reason this matters is because the natural rate of unemployment (along with the natural rate of interest) is an important metric guiding both ordinary and extraordinary monetary policies. Chairman Powell acknowledged that we cannot directly observe or predict these natural rates in a changing economy. In more detailed remarks, Powell (2019a), referring to the tendency of economists to label these metrics as u* (natural rate of unemployment) and r* (natural rate of interest), explains, "Unlike celestial stars, these stars move unpredictably and cannot be directly observed. We must judge their locations as best we can based on incoming data and then add an element of risk management to be able to use them as guides." While both u* and r* were once thought to be roughly stable, recent changes in the economy have made it clear that they are far more 1 Chairman Powell's frankness earned him both criticism and praise. For example, Williamson (2019) writes in response, "There's a lot Powell doesn't know, and it shows." Cochrane (2019b), responding to Williamson, writes, "I would much rather have Powell's healthily acknowledged uncertainty than a PhD economist who thinks he or she 'knows' how the Phillips curve really works." 22 of use, available at https://www.cambridge.org/core/terms.