THE April WSJ survey was released (responses from April 5 to 9):
First of all, the level of GDP forecast over the last 3 surveys (7 months):
Figure 1: GDP (bold black), average implied level forecast by the WSJ for April 2024 (blue), January 2024 (red), October 2023 (light green), all in billions of SAAR 2017 dollars. Source: BEA 2023Q4 3rd version, surveys from the WSJ (various issues) and calculations by the author.
Currently, the average forecast for the first quarter is exceeded by two of the most recent nowcasts (Atlanta, NY Feds):
Figure 2 (updated): GDP (bold black), February SPF (light blue), PIBNow (4/15) (red square), NY Fed (light green square), St. Louis Fed (blue inverted triangle), all in bn.Ch.2017$ SAAR. Source BEA via FRED, Philadelphia Fed, Atlanta Fed, New York Fed, St. Louis Fed via FRED and author’s calculations.
Figure 2: GDP (bold black), February SPF (light blue), GDPNow (4/10) (red square), New York Fed (light green square), St. Louis Fed (blue inverted triangle), all in billions .Ch.2017$ SARRE. Source BEA via FRED, Philadelphia Fed, Atlanta Fed, New York Fed, St. Louis Fed via FRED and author’s calculations.
The average and median forecasts do not predict negative growth quarters. Even the reduced lower bound (removing the last 6 forecasts for 2024) does not show two consecutive negative quarters.
Figure 3: GDP (bold black), level implied from April 2024 WSJ average forecast (blue), median (beige), truncated 20% high/low for 2024 (gray), all in billions of dollars Ch. 2017 SAAR. Source: BEA 2023Q4 3rd version, WSJ surveys (various issues) and author’s calculations.
The lowest level is Mike Cosgrove/Econoclast, the highest level is Song Won Sohn/SS Economics. The median is Satyam Panday/S&P Global Ratings.
The highest growth rate forecast is that of eternal optimist James Smith/EconForecaster (3.3% if 2024 q4/q4). Andrew Hollenhorst and Veronica Clark/VSitigroup and Amy Crew Cutts/AC Cutts both forecast negative growth in the second and third quarters.
As for the recession (remember, The NBER does not define a recession using the rule of thumb of two consecutive quarters of negative GDP growth.), the views of economists diverge considerably from a purely statistical forecast (probit) based on the forward spread of 10 years to 3 months and from the WSJ survey.
Figure 4: WSJ survey on likelihood of recession In recession over one year (in blue) and over 10 years at 3 months based on probits In one year (tan), both in %. Probit estimates based on 1986-2018 (pre-pandemic). The NBER has defined the peak to trough dates of the recession in gray. Source: WSJ, NBER, author’s calculations.
It is worth noting that heading into the 2007-2009 recession, the probit model led the survey measure, while probit and survey grew in tandem through the end of 2022 , to diverge later.
