With Q3 2019 starting to wind down, it’s time to look at what the New York and Atlanta Feds’ “nowcasting” models have to say about the current pace of US economic growth. While neither number is an official Fed estimate, these algorithms are both pretty good (certainly as accurate as the Blue Chip consensus) and also let you follow along as they move around with the release of new econ data.
To set the stage, here are the last 4 quarters of US GDP growth:
- Q3 2018: 2.9%
- Q4 2018: 1.1%
- Q1 2019: 3.1%
- Q2 2019: 2.0%
- Average: 2.8%
For Q3 2019, here is what the Fed models have to say and how they have trended through the quarter:
Atlanta Fed GDPNow: currently expecting 1.9%:
- This is slightly below the Blue Chip economist mean estimate of 2.0%.
- The model’s output has been quite volatile this quarter, estimating GDP growth as high as 2.3% and as low as 1.5% across just a few weeks from mid-August to early September.
- Positive revisions have come from US consumer data such as retail trade and housing.
- Negative revisions have come from the manufacturing sector and international trade.
The New York Fed Staff Nowcast is currently calling for 1.55% growth:
- Like GDPNow, there has been considerable volatility in this model’s estimate through the third quarter. At the end of June it was looking for just 1.2% growth. This bounced to a high of 2.2% at the end of July.
- Also like the Atlanta model, US consumer data has been the primary contributor of upside revisions, including real personal consumption expenditures, retail sales and the last ADP employment report.
- The negatives this quarter have been tied to the industrial economy, including soft capacity utilization in the recent weak print for manufacturing PMIs.
What we make of this data:
#1: The models agree that Q3 US economic growth is set to run well below the 4-quarter average of 2.8%.
- On the plus side, 1.6% – 1.9% growth is still nowhere near the start of a much feared/anticipated recession.
- On the downside, Q3 will be the slowest growth since Q4 2018 when US/global equity markets were in free-fall.
#2: The US consumer is holding up the domestic economy. If personal consumption were weaker, both models would be tipping over into negative readings for Q3. This is why we have been so focused on American perceptions of recent stock market volatility. Equity prices are the transmission mechanism between macro uncertainty and consumers “talking themselves into a recession”.
#3: The choppiness of the Atlanta/NY Fed model output mirrors recent global capital market volatility, two sides of the same existential question: when will the US economy slide into recession? Both algos have generated estimates at/below 1.5% in the last 2-6 weeks, only saved from further declines by stronger consumer data. If those had come in weaker, we’d be looking at sub 1.0% growth or worse in Q3. No wonder Treasury yields tumbled in August…
For some longer-term perspective let’s finish up with the most closely watched regional Fed model, the New York Fed Recession Probabilities percentage, which stands at 38% just now. While it only uses 2 inputs (3-month and 10-year Treasury yields) rather than dozens (like the nowcasts), it remains at levels consistent with an all-but-certain US recession in the next 12 months.
The internals of the Atlanta/NY Fed models show how the US economy can escape that fate: the American consumer has to keep spending, and business investment/industrial output must reaccelerate, even if modestly. US-China trade negotiations are the fulcrum issue. Productive progress will keep equity markets stable and support consumer confidence. And a successful conclusion will give businesses the confidence to invest. Now, it just needs to happen…
Atlanta Fed GDPNow model: https://www.frbatlanta.org/cqer/research/gdpnow.aspx?panel=3
New York Fed Staff Nowcast: https://www.newyorkfed.org/research/policy/nowcast