Viewpoints: 2026

If you rely on the GDP estimates from the Federal Reserve Bank of Atlanta's team, the US economy is performing the improbable by growing well beyond its potential, despite two key sectors experiencing no growth and minimal new job creation. The Atlanta Fed's GDPnow estimate suggests that Q4 GDP growth is at an annualized rate of 5.1%. However, that estimate of output growth is at odds with the weak growth picture depicted by data from manufacturing, housing, and employment. That data indicates that GDP growth is occurring at, at best, half the rate.

The US economy is composed of three primary sectors: goods, structures, and services. The Federal Reserve's industrial production metric shows that manufacturing output in Q4 through November is lower than in Q3, suggesting minimal, if any, growth in goods output for Q4.

The structures component includes three sectors: housing, nonresidential, and state and local. Housing starts in October are significantly below the Q3 average, indicating a decrease in residential construction activity. Nonresidential construction has been weak throughout 2025, with data indicating a decline in Q4. Meanwhile, state and local construction spending shows minimal growth. Thus, the structures component is likely to weigh on Q4 GDP.

This leaves the service sector, the largest of the three. The service sector doesn't seem to be expanding rapidly enough to support the GDPNow estimate either. For instance, private sector service job growth amounted to just 112,000 in the three months ending in December.

Moreover, for the GDP output to increase by more than 5%, the income side must show a similar rise. However, the data on jobs, wages, and hours indicate a very modest increase in labor income in Q4. To align with the GDP output growth estimate, there would need to be an extraordinary increase in operating profits to offset the weak labor income. Companies would have to register a record increase in margins to do that, which is unlikely given the price data.

The government shutdown, which lasted over 40 days, affected and delayed several economic reports to such an extent that it's currently very challenging to obtain a reliable assessment of the Q4 economic performance. Nevertheless, if the data on jobs, manufacturing, and housing accurately represent the situation, then GDPnow is more than 50% off.

November was another soft month for state labor markets. No state had a statistically significant change in payroll employment, though the raw increases in California (32,500). New York (17,100), as well as the decline in Illinois (9,700) look to be of some note. However, no state appears to have had a change as large as .2 percent. In the October numbers, the federal cuts were reflected in large, most likely statistically significant, drops in DC, Maryland, and Virginia.

State unemployment rates in November were generally, though not universally, higher in November than in September (there are no unemployment figures for October, due to the federal government shutdown). Delaware and West Virginia’s unemployment rates were .4 percentage points higher in November than in September, and a number of other states saw increases of .3 percentage points. Hawaii’s rate dropped .3 percentage points. The highest unemployment rates were in DC (6.5%), California (5.5%), New Jersey (5.4%), Nevada (5.2%), New Jersey (5.2%), Oregon (5.2%), and Michigan (5.0%). Alabama, Hawaii, North Dakota, South Dakota, and Vermont had unemployment rates under 3.0%, while South Dakota’s 2.1% was the lowest in the nation.

Puerto Rico’s unemployment rate was unchanged at 5.7% (remarkably, there is an October unemployment rate estimate for Puerto Rico) and the island’s job count rose by 1,700.