When two branches of the Federal Reserve Bank paint starkly different pictures of the U.S. economy, the result is less clarity and more confusion. That’s the case this week as the Atlanta Fed and St. Louis Fed released third-quarter projections for gross domestic product growth that diverge dramatically.
According to the Federal Reserve Bank of Atlanta’s GDPNow model, the economy is on track to grow at an annualized rate of 3.1% as of September 10. That estimate was slightly higher than the 3% projection issued on September 4, illustrating how the model shifts as new data comes in. But the Federal Reserve Bank of St. Louis sees a very different story. Its Real GDP Nowcast, updated September 5, projects third-quarter growth at just 0.57%.
The wide gap between the two estimates doesn’t reflect contrasting views of the economy so much as different forecasting tools. Both institutions stress that their Nowcasts are separate from official government figures and not intended as formal forecasts. The Bureau of Economic Analysis, part of the Commerce Department, releases the definitive GDP data.
The Atlanta Fed’s GDPNow relies solely on incoming data, using a formula designed to mimic the BEA’s methodology without any subjective adjustments. Each new release—from retail sales to industrial production, jobs data, construction numbers, and international trade—automatically recalibrates the growth estimate. For transparency, the Atlanta Fed even provides public access to its model spreadsheet, spelling out how each datapoint affects the projection.
The St. Louis Fed, by contrast, draws on what it calls “economic content from key monthly economic data releases.” Its methodology, outlined in research on constructing nowcasts, is built around a macroeconomic news index that interprets incoming reports through a different model. As the St. Louis Fed explains, it doesn’t replicate GDP accounting as the Atlanta Fed’s model does; instead, it uses a distinct framework rooted in statistical forecasting.
Both banks produce what are known as Nowcasts, estimates that track the likely pace of current-quarter GDP growth before the BEA issues its first official release. Similar models are also maintained by consulting firms and Wall Street banks. For businesses and analysts navigating uncertain conditions, comparing across models—and weighing them against the BEA’s published data—offers at least some guidance through the uncertainty.
© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.