Last week, we argued that most of the equity market's adjustment so far this year is due to multiple compression, given higher interest rates, and that we believed the forces pushing stock prices down could begin to ease: the war in Ukraine no longer seems likely to escalate into a wider conflict; the number of new Covid cases in China continues to decline, allowing for a greater opening of the economy; and global inflation may be peaking. See (https://www.fynsa.cl/newsletter/estrategia-3/).
However, risk markets have continued to sell off this week and government bonds have risen (rates have fallen). These market movements are consistent with market perceptions of a higher probability of recession.
The question that follows then is... How much of a recession are the markets pricing in?
Recognizing that there is no perfect way to infer recession probabilities from market prices, examining a variety of approaches across asset classes may be more helpful, as noted in a recent JP Morgan report.
Actions: The simplest way to assess how much is included in the price of a recession in the US is to use the average 26% decline in the S&P 500 index during the last 11 recessions (see Chart No. 1). So far, the S&P 500 has fallen 18.6% from its peak (as of May 19), so the stock markets are pricing in an 18.6/26 = 72% probability of recession. (see chart No. 2).
Source: Fynsa Estrategia; Bloomberg; JP Morgan. The blue bars correspond to recessions.Credit: The average level of IG spreads during recessions (the last six) is 250 bp, and the average level outside of recession is 100 bp (see Chart 3). In this case, the recommendation is to use averages rather than maximums and minimums, due to the extreme behavior of credit spreads post-Lehman, which skews the relatively limited data history, especially for high-yield credit.
IG US spreads have widened considerably in recent weeks and stand at 148 bp above Treasury bonds as of May 19 (around their historical averages). This suggests that the price of IG credit markets has a (148-100) / (250-100) = 32% probability of a recession in the US (see Chart 4).
Source: Fynsa Estrategia; Bloomberg; JP Morgan. The blue bars correspond to recessions.Meanwhile, US HY spreads have widened to 480 bp over Treasury bonds. This compares with an average recession level for HY spreads (over the last three recessions) of 1,000 bp and an average non-recession level of 350 bp. (see Chart 5). The same calculation suggests that US HY spreads are priced at (480-350) / (1000-350) = 20% probability of recession. In other words, US HY credit appears to have a much lower probability of recession relative to US IG credit (see Chart 6).
Source: Fynsa Estrategia; Bloomberg; JP Morgan. The blue bars correspond to recessions.Government bonds: If we use the five-year US Treasury yields as a reference, these fell by an average of 200 bp during the last three recessions (measured from the peak in the months prior to the start of the recession to the trough during the recession) (see Chart 7). This compares with a drop of around 24 bp in 5-year Treasury yields from the peak on May 6 to May 19. This suggests that the Treasury market is pricing in around 24 bp/200 bp = 12% probability of a recession in the US, a low probability to which the slope of the curve is still positive. (see charts 8 and 9).

Source: Fynsa Strategy; Bloomberg; JP Morgan.Overall, US equity markets are pricing in a very high probability of recession at 72%.%. The corresponding price for a recession by US IG credit markets exceeds 30%, while HY credit is pricing in a nearly 20% probability of recession. In contrast, rate markets are pricing in a fairly low probability of recession at 12%, creating a divergence with equity markets in particular. Either the equity markets are right and a recession occurs, leading to much larger declines in bond yields, or the rate markets are right and a recession is avoided, leading to a recovery in the equity markets.
Our view remains that equity markets are overpricing the risk of recession and that the probability of this occurring in the next 6 to 12 months, although increasing, is still low. We therefore maintain a risk-on stance.
