One of the best near-term recession indicators is the job market. The release of October’s jobs report showed unemployment rose to 3.9%. Small increases in the unemployment rate have historically been sufficient to trigger a recession. This is called the Sahm Rule. Now, it’s not calling for a recession yet, but it may be getting close if the unemployment rate does not improve from here.
Why It Has Worked Historically
The Sahm Rule is designed to rapidly determine if the U.S. economy is in recession, in part, so that policymakers can respond. It’s designed by economist Claudia Sahm.
Using jobs data is helpful because it’s released weeks ahead of most other economic metrics. This approach has merit because unemployment has a big impact on economic growth. Broadly two thirds of the economy is consumer spending, so when jobs are cut, it’s a fair bet that consumer spending falls, and hence economic activity typically weakens too because consumer spending is its largest component.
Constructing The Metric
The metric examines the current 3-month average of the unemployment rate compared to the low unemployment rate of the previous 12-months. The 12-month low of unemployment is currently 3.4% as occurred twice in January and April of 2023. Then the recent readings of unemployment for the past 3 months (August, September and October) are 3.8%, 3.8% and 3.9% respectively. Therefore should unemployment remains at or above the current 3.9% level for November’s and December’s jobs report or simply spike materially higher in November, then the indicator will call a recession. A recession could be called by this indicator with the next jobs report on December 8.
The Track Record
The indicator has called recessions since the 1960s consistently, but has also seen a few false positives. The Sahm rule always calls a recession, but, less likely, the warning is triggered without a subsequent recession.
It’s important to remember the current level is just below the recession threshold. We’ve seen a similar level to where we are currently four times since the 1960s without a following recession. Still even on current data a 2024 might be argued to be more likely than not. The main concern is that as a recently-developed indicator, it is necessarily fitted to historical data. As such, how it continues to hold up for an extended period “out of sample” remains to be seen.
Still, the logic behind the indicator is sound. Importantly the metric does typically need to hit 0.5%, it’s reached slightly lower levels of around 0.4% several times without triggering a recession. Once the 0.5% level is hit, unemployment does tend to climb during the duration of any recession.
Other Recession Indicators
Other metrics suggest a recession may be coming. The yield curve is a relatively trusted recession predictor on a medium-term view and has been signalling a recession for many months. The markets too, as measured by the S&P 500 index, have generally declined since July, suggesting some economic weakness could be coming. Also, the Conference Board’s Leading Economic Indicators have been declining since April 2022, though those indicators do incorporate variations of the metrics already discussed, so isn’t an entirely unique indicator compared to others such as the stock market and yield curve. Event forecasting site Kalshi currently weighs the chance of a 2024 recession at approximately 50%. If we do see a 2024 recession, it would not be a big surprise.
What Next
The yield curve has been calling for a U.S. recession for some time. If the Sahm indicator joins it in forecasting a recession, the odds of a 2024 recession will increase on most estimates. Events such as a possible government shutdown could raise the odds of a U.S. recession further. However, it’s important to note that the Sahm Indicator is not calling a recession just yet. Upcoming jobs reports will be key.