A Perfect Track Record, Broken
The Sahm Rule, created by former Federal Reserve economist Claudia Sahm, had a clean historical record. Since 1970, it had identified every US recession with zero false positives. The rule is simple: when the three-month moving average of the unemployment rate rises by 0.5 percentage points or more from its 12-month low, the economy is in or entering a recession.
In July 2024, the three-month average reached 0.53 percentage points above its 12-month low. The rule triggered. But no recession followed.
This wasn't just any indicator failing. The Sahm Rule had become something of a gold standard among recession watchers precisely because it never cried wolf. Unlike the yield curve, which had given false signals before, or the Conference Board's Leading Economic Index, which could be noisy, the Sahm Rule was the reliable one. When it spoke, people listened.
The unemployment rate had been sitting at historic lows around 3.4% through early 2023, then began climbing steadily. By April 2024, it hit 3.9%. By July, it reached 4.3%. The math was straightforward: the three-month average had risen 0.53 percentage points from its 12-month low of 3.5%. The rule triggered exactly as designed.
Why It Failed This Time
Sahm herself explained the failure in a Bloomberg opinion piece on August 7, 2024. The unemployment rate rose not because demand for workers weakened (the traditional recession mechanism) but because the supply of workers increased. Post-pandemic immigration drove a significant increase in the labor force. More people looking for work raised the unemployment rate even though the number of employed people was also growing.
The numbers tell the story clearly. Between January 2023 and July 2024, the civilian labor force grew by approximately 2.8 million people. Meanwhile, employment grew by roughly 2.4 million. The difference between those two numbers explains the rising unemployment rate. In a typical recession, you see employment falling while the labor force stays roughly stable or shrinks as discouraged workers leave. This was the opposite pattern.
Immigration data supports this interpretation. The Congressional Budget Office estimated that net immigration in 2023 was about 3.3 million people, well above the pre-pandemic average of around 1 million annually. Many of these new arrivals entered the workforce, boosting labor supply faster than the economy could immediately absorb them.
The Sahm Rule was designed during a period when labor supply changes were gradual and predictable. The post-pandemic immigration surge created a labor supply shock that the rule was not calibrated to distinguish from a demand-driven recession. The underlying assumption was that unemployment rises primarily signal weakening demand for workers, not expanding supply of workers.
The Labor Market's Mixed Signals
Other labor market indicators were sending conflicting messages throughout this period. Job openings, measured by the JOLTS survey, remained elevated compared to pre-pandemic levels. The quits rate, which typically falls sharply in recessions as workers become reluctant to leave their jobs, stayed relatively stable. Initial unemployment claims, one of the most timely recession indicators, showed no sustained increase.
Wage growth also behaved differently than in a typical recession scenario. While it moderated from the peaks of 2021-2022, it remained positive in real terms for most workers. In recessions, wage growth typically decelerates more sharply as employers gain bargaining power and workers accept smaller increases to keep their jobs.
These contradictory signals should have been a warning that something unusual was happening. The Sahm Rule was flashing red, but other labor market indicators were showing yellow or green. The traditional recession playbook wasn't matching the observed data.
The General Lesson About Indicator Failure
Every indicator, no matter how long its track record, operates on assumptions about the structure of the system it measures. When those structural assumptions change, the indicator can fail. The yield curve's 2022-2023 inversion without recession is another example. The Conference Board LEI's extended decline without recession is a third.
The yield curve inversion, for instance, typically predicts recession because it reflects expectations that the Fed will need to cut rates aggressively to combat economic weakness. But in 2022-2023, the inversion reflected expectations that the Fed would need to cut rates after successfully bringing down inflation, not after triggering a recession. The structural assumption about why yield curves invert had changed.
Similarly, the Conference Board's Leading Economic Index fell for 16 consecutive months starting in early 2022, its longest decline since the financial crisis. Historically, such sustained declines preceded recessions. But the index was heavily weighted toward components like building permits and stock prices that were being affected by sector-specific factors (housing market disruption from rate changes, tech stock volatility) rather than broad economic weakness.
The practical response is not to abandon indicators when they produce a false signal, but to understand the assumptions they depend on and monitor whether those assumptions still hold. If the assumption underlying the Sahm Rule is "unemployment rises primarily because of demand weakness," and you observe that unemployment is rising because of supply increases, you can interpret the trigger differently than you would if demand weakness were the driver.
Structural Breaks and Model Decay
Economists call these failures "structural breaks." The relationships that held for decades can shift when underlying economic dynamics change. The rise of gig work, remote employment, demographic shifts, and policy changes can all alter how economic indicators behave relative to recession timing.
This is why prediction markets often outperform individual indicators. Markets aggregate information from multiple sources and can adapt more quickly to changing relationships. When the Sahm Rule triggered in July 2024, recession probability markets on platforms like Kalshi and Polymarket remained relatively subdued, suggesting that informed traders were skeptical of the signal.
Implications for Composite Approaches
This is precisely why composite approaches that track multiple indicators are stronger than relying on any single one. When the Sahm Rule triggered in 2024 but the yield curve was un-inverting, credit spreads were tight, and other recession indicators were not confirming, the composite view correctly suggested "not a recession" even though one component was flashing red. The convergence requirement filters out isolated false signals.
The New York Fed's Weekly Economic Index (WEI) provides a good example of this approach. It combines ten daily and weekly indicators to produce a real-time reading of economic growth. Throughout the period when the Sahm Rule was triggering, the WEI remained positive, suggesting continued economic expansion.
Credit markets told a similar story. Investment-grade credit spreads stayed near historic lows throughout 2024, indicating that bond investors saw little recession risk. High-yield spreads widened modestly but remained well below levels typically associated with recession. If the economy were truly entering recession, you would expect credit markets to price in higher default risk.
Professional forecasters also remained optimistic. The Survey of Professional Forecasters, conducted by the Philadelphia Fed, showed recession probabilities remaining below 30% throughout 2024, even after the Sahm Rule triggered. This suggests that economists who study these indicators full-time were able to look through the false signal.
Building Better Composite Models
The key is building composite models that weight indicators based on their current reliability rather than their historical performance. An indicator that has worked for 50 years but is based on outdated structural assumptions should receive less weight than one that captures current economic dynamics accurately.
This is where tools like momentum trading engines can be valuable. They can track which economic indicators are currently providing the most predictive power and adjust weightings accordingly. Rather than treating all indicators equally, they can emphasize those that are working in the current environment.
Practical Applications for Traders and Analysts
For traders and analysts, the Sahm Rule's 2024 failure offers several practical lessons. First, always examine the underlying drivers when an indicator triggers. If unemployment is rising due to labor supply increases rather than demand weakness, the recession implications are very different.
Second, use multiple timeframes and data sources. The Sahm Rule looks at unemployment over a 12-month period, but weekly initial claims data updates much faster. Continuing claims, job openings, and quits rates all provide additional context that can help distinguish between different types of labor market changes.
Third, pay attention to what markets are pricing in. When economic indicators give mixed signals, asset prices often provide valuable information about how informed investors are interpreting the data. The fact that stock markets reached new highs in late 2024 while the Sahm Rule was triggered suggested that equity investors weren't buying the recession signal.
Finally, consider using whale finder tools to track how large, sophisticated traders are positioning themselves during periods of indicator uncertainty. These traders often have access to more detailed data and analysis that can help separate signal from noise.
The Sahm Rule will likely remain a useful recession indicator going forward, but its 2024 failure serves as a reminder that no single metric should drive major economic or investment decisions. Understanding why indicators work, not just that they have worked historically, is essential for navigating periods when structural relationships shift.