The credentials of the yield curve as a ‘super-forecaster’ of recessions should be treated with scepticism. At best, the yield curve is a mirror that reflects what we already know, or think we know, about the economy.
The slope of the yield curve and whatever message it sends about recessions risks should therefore reflect nothing more than the market’s considered assessment of public information on economic fundamentals or the conduct of economic policy, for example; and potentially much less.
This mirror image is likely distorted, offering an inaccurate view of the true state of the macro outlook because the slope of the yield curve is also driven by other factors, unrelated to the risk of recession.
There is no plausible, stable mapping between a percentage point increase in the perceived (or actual) probability of a recession and the slope of the yield curve; be it the corresponding shift in either expectations of short rates or the term premium.
The only reason to ignore fundamentals and focus instead on their reflection is if you believe that bond investors are better at economics than economists – that the slope can partially reveal the superior forecasts of those investors.
In any case, it seems unlikely that a market where prices are increasingly driven by global investors and global factors can speak authoritatively on local developments. Moreover, the very shift in the shape of the yield curve that supposedly sends a signal of looming recession reduces its likelihood by stimulating demand.
The real risk of recession may arise in ‘tantrums’: when the yield curve decouples from fundamentals and there is an unwarranted increase in medium to long-term yields that will depress activity.
We accept there are reasons to be concerned about the outlook for demand at the global level, but we maintain that investors are better off forecasting recessions by thinking about fundamentals and discussing with economists than excessively analysing the distorted reflection of fundamentals in the yield curve mirror.