The Atlanta Fed produces occasional estimates for GDP, using the most up to date government data sources. They call this forecast “GDPNow“, and the New York Fed produces a similar forecast. Before proceeding further, let me emphasize that this post is not bashing these two Fed banks, indeed the latest Atlanta Fed forecast comes with this disclaimer:
GDPNow is not an official forecast of the Atlanta Fed. Rather, it is best viewed as a running estimate of real GDP growth based on available data for the current measured quarter. There are no subjective adjustments made to GDPNow—the estimate is based solely on the mathematical results of the model. In particular, it does not capture the impact of COVID-19 beyond its impact on GDP source data and relevant economic reports that have already been released. It does not anticipate the impact of COVID-19 on forthcoming economic reports beyond the standard internal dynamics of the model.
The Atlanta Fed issued an updated forecast today, showing 2.7% RGDP growth in Q1. The New York Fed forecast 1.7% RGDP growth in Q1 and 0.3% in Q2. Now you see why they put in the disclaimer.
It’s clear that these models are not actually giving us “GDPNow” in the sense of the optimal forecast given today’s information. They are giving us the optimal forecast given publicly available macro data released by the government. But the rational expectations model says that optimal forecasts are based on all publicly available information.
As of today, Hypermind forecasts 0.1% NGDP growth in Q1 and minus 3.6% in Q2. Presumably their implicit RGDP forecasts are even a bit lower, at least for Q1. Even so, I don’t believe Hypermind is deep and liquid enough to provide an optimal forecast, and for weeks I’ve viewed that market as being somewhat behind the curve in recognizing the severity of the oncoming slump.
Nonetheless, it would be nice for at least one of the Fed banks to produce a true NGDPNow and RGDPNow forecast, given all publicly available information. That’s not easy, but I imagine that if you look back throughout history you’ll find that periods with multiple back to back stock market crashes in the 6% to 12% range, a sudden collapse on T-bond yields, plunging oil prices, and (especially) enormous weakness in assets like junk bonds, are usually associated with a sudden and sharp drop in the broader economy.
[Someone correct me if I’m wrong–Perhaps 1987 was an exception, but how did other markets besides stocks do in 1987?]
So how about it? Which Fed will step up to the plate and produce a truly cutting edge forecast? The St Louis Fed used to be monetarist—how about trying market monetarism?
PS. I do understand that this crisis is unique, and no mathematical model would be expected to perform all that well. (Imagine trying to forecast yesterday’s weekly jobless claims). But I still believe that more reliance on market indicators would improve GDPNow.
PPS. David Levey sent me this video of a guy imitating Trump. Even if you like Trump, you got to admit he’s a talent mimic.
PPPS. A couple of years ago, I said I would not blame Trump for any recession, and I’m sticking with that. I’m not even blaming him for more than small portion of the surge in cases in the US, despite him being behind the curve on preparedness. I’m sticking with the “Presidents only explain 3% of outcomes” worldview. I try to avoid motivated reasoning, much as I’d enjoy blaming Trump for everything.
PPPPS. I see Boris is doing his part to build up the UK’s “herd immunity”. I don’t like his politics but I wish him well.