I wonder what a “soft landing” means and whether we are moving toward that goal. Despite the historical correlation findings, almost any regression based on term spreads will predict a recession by around mid-2024.
Booker and Wessel Discuss the soft landing problem and cite this table adapted from Blinder.
source: Booker and Wessel (2023).
I’m only familiar with the 1993-95 and 1999-2000 episodes. I think it’s interesting to compare the situation in 1993-95 with current real interest rates and GDP. First, 1993-95:
figure 1: 2017 Monthly GDP (billions) $SAAR (blue, left log scale), fed funds adjusted for Cleveland Fed 1-year expected inflation, % (tan, right scale), and by SPF 10-year Median expected inflation-adjusted 10-year Treasury yield, % (blue, right scale). Booker-Wessel-Blinder Dashed line at start/end point of soft soil. Source: SPGMI, Federal Reserve and Treasury Department through FRED, Cleveland Fed, Philadelphia Fed, and author’s calculations.
This is some sort of landing plot currently, not yet confirmed…
figure 2: 2017 monthly GDP in billions SAAR (blue, left log scale), federal funds adjusted for Cleveland Fed 1-year expected inflation, % (tan, right scale) and ten-year TIPS, % (blue, right scale). Dashed soft land start/end is determined by the author. Source: SPGMI, Federal Reserve and Treasury Department through FRED, Federal Reserve Bank of Cleveland, and author’s calculations.
Recent events have seen a sharp increase in the real fed funds rate (over 4.5% vs. 3%), a similar increase in the real 10-year Treasury note, and when fed funds rose (in the second quarter of 2022), output actually fell , while the federal funds rate continues to fall. Rising in 1994. Also conversely, 10-year real rates were falling by the end of what Blinder defined as a soft landing, and in the current case, they continue to rise.
Due to the large increase in interest rates, the debt service ratio has increased significantly.
image 3: Current private non-financial sector debt servicing ratio in % (black), 1993-95 debt servicing ratio in % (light blue). The 3-month interest rate was used for first-difference inference during 2023M04-10. Source: Bank for International Settlements and author’s calculations.
My guess is that the debt service ratio trajectory has flattened (actual data only extends to March 2023), but remains well above where it was before tightening began.
As calls for a soft landing continue, it is worth recalling that if historical correlations hold, the probability of a recession in spread-based probabilistic models will breach the 50% threshold around November-December this year, and reach a peak in May-June 2024 reaches its peak around the month.
Figure 4: Recession probabilities for the Probit model with only 10yr-3mo spreads (green), spreads, NFCI, foreign term spreads (tan), and spreads, NFCI, debt service ratios, foreign term spreads (blue). NBER-defined recession peak-to-trough dates appear gray. Source: Author’s calculations, NBER.
The important caveat is that historical correlations hold.
