If no recession is forthcoming, what can we conclude, given most term spread models were signaling a “sure bet”? Unlikely outcome (it’s a probabilistic world!), breakdown in historical correlations, omitted variable problem? In order to shed some light on this question, I examine probability estimates from (i) plain vanilla spread, (ii) debt-service-ratio and foreign term spread augmented, and (iii) term-premium adjusted spread specifications.

Figure 1: Estimated 12 month ahead recession probabilities, based on spread and short rate (blue), on spread, short rate, debt service ratio and foreign term spread (tan), and term premium adjusted term spread (green). NBER defined peak to trough recession dates shaded gray. Source: NBER and author’s calculations.

Estimated 12 month ahead recession probabilities are obtained using probit models. The first specification (blue line) is a plain vanilla term spread model estimated 1990-2023M04 assuming no recession in the US to 2024M04. The second (tan line) is

Keep reading this article on Econbrowser Blog - James Hamilton & Menzie Chinn.

Leave a Reply