Stealth stagnation: why the U.S. economy is not as robust as you think

By Jean Boivin and Alex Brazier

The prevailing narrative about the U.S. economy is that it’s ‘resilient’: despite rapid rate hikes, economic growth has held up and may even be accelerating. Yet looking back, we see a different picture: stagnation that’s gone largely unnoticed thanks to measurement issues and a misleading cyclical view of the labor market. Let’s unpack that.

On a broad measure of activity, the U.S. economy has basically flatlined since the end of 2021, growing less than 1% in total over the period – the weakest 18 months ever seen outside a recession. See the Growth never so weak outside recession chart.

What is this broad measure? It’s an average of two ways of measuring economic activity.¹ One (GDP – Gross Domestic Product) adds up total spending by households, businesses, government and on exports; another (GDI – Gross Domestic income) adds up incomes and profits of households and firms.

In reality, the two should match. But no measure is perfect and gaps sometimes open up between them. Right now, the measured level of GDP is very high relative to GDI. GDP says the economy has grown by 1.9% in the past 18 months. GDI says it has contracted by 0.5%. Statisticians can’t find the income to match their estimates of spending. See the Flatlining activity chart.

Both measures can be revised over time. GDI might be revised up if more income is found. We expect a small upward revision to GDI due to changes in the way the Fed’s interest income on its past asset purchases will be recorded. But we also think it striking that consumer spending is measured to be so much higher than implied by reliable sales tax revenues, so it’s quite possible that estimates of GDP get revised down. The truth is probably somewhere between GDP and GDI. Research from the U.S. Bureau of Economic Analysis suggests the average of GDP and GDI is the most reliable guide to what’s really going on – a view shared by the NBER committee that judges when the U.S. has gone into recession.

The gap between the measures tends to be wider in volatile times. It was last this big just before the recession in the Global Financial Crisis of 2008. See the Wide gap between measures chart. Fed research suggests GDI is better at flagging when the economy is about to take a turn (for the better or the worse) – which could speak in favor of putting even more weight on it.

Partly because the GDP data come first and tend to set the tone. But more importantly: the creation of nearly 7 million jobs – not normally a sign of a weak economy – is misleading people, we think. In a typical business cycle, stagnation would be associated with job destruction.

But this is not a business cycle. Around 15 million jobs were cut almost overnight when the pandemic hit and the economy shut down. Companies have been hiring back as activity restarted, and 18 months ago they were still 7 million short of finishing that process.

Against that baseline, jobs have actually been destroyed. And that’s even clearer when we account for the fall in the number of hours each employee is now working – that’s equivalent to losing more than half a million jobs since 2021 and takes overall job gains down closer to 6 million in an economy that would have had to add 7 million jobs just to get back to square one.

Seen in this context, recent job numbers are far from a sign of economic strength – like the GDP/GDI average, they reveal stealthy stagnation.