Even a markets-based estimate of US recession risk has faded lately. This real-time approximation of economic conditions via a probit model analyzes four data sets on a daily basis: the US stock market (S&P 500); the Treasury yield curve (10-year yield less the 3-month T-bill yield); the credit spread (BAA-rated bond yield less AAA yield); and spot crude oil prices (based on the US benchmark, West Texas Intermediate). The current estimate has fallen lately, dipping to around 5% as of yesterday (Nov. 9). Even during the height of the recent market correction this measure of recession never topped much above the low-20% range—well below the levels in the early stages of the last two NBER-defined recessions.
Meanwhile, the Atlanta Fed’s GDPNow model is currently (Nov. 4) projecting that fourth-quarter GDP is headed for a modest rebound of 2.3% after Q3’s sluggish 1.5% rise.
Surprising? Not really. The US economic trend has wobbled lately, but the case for expecting the worst was less about the data vs. letting emotion and a selective view of indicators dictate the analysis. A superior methodology, and one that’s been proven valuable once again, is applying a relatively objective analytical lens to a diversified data set. The drama factor is low, compared with the usual suspects. But what you lose in terms of theatrical excitement you more than make up via a higher level of reliability.