A newly developed Labor Market Stress Indicator (LMSI) offers policymakers and economists a more nuanced understanding of economic conditions by analyzing unemployment trends across all 50 states and the District of Columbia. Unlike traditional national metrics, this tool tracks how many states are experiencing accelerating unemployment—defined as a rise of at least 0.5 percentage points above the lowest rate recorded in the previous 12 months. The approach mirrors the Sahm rule but applies it at the state level, providing earlier and more geographically detailed signals about potential recessions. n nAccording to research from the Federal Reserve Bank of San Francisco, when 30 or more states simultaneously show rising unemployment, the national economy has historically been in recession. Additionally, in past downturns, about 75% of the U.S. labor force resided in states with deteriorating job markets. The LMSI leverages state-level unemployment insurance claims data, allowing for more timely assessments than official employment reports, which only date back to 1976. By using reconstructed historical data going back to 1948, the indicator offers a comprehensive 77-year perspective. n nIn July 2024, the national Sahm rule triggered a recession alert due to a sharp rise in unemployment. However, the LMSI revealed a different picture: while nearly 70% of the labor force lived in states with rising joblessness—close to the 75% threshold—the number of affected states barely crossed the 30-state mark and quickly declined the following month. This short-lived and geographically uneven pattern contrasted sharply with previous recessions, where both measures remained elevated for extended periods. n nAs of June 2025, only the District of Columbia met the LMSI’s criteria for labor market stress, affecting just 0.2% of the national workforce. A statistical model based on the LMSI estimates the current probability of recession at 5%, well below the 40% threshold observed during past downturns, indicating the economy remains in an expansionary phase. n nThe indicator also highlights regional disparities in economic resilience. States such as Illinois, Ohio, Nevada, and Oregon have consistently shown labor market deterioration during every recession over the past 75 years. In contrast, resource-dependent states like North Dakota, Wyoming, Alaska, and Nebraska have experienced such stress less than half as often, suggesting their economies are less synchronized with the national business cycle. n nThe Federal Reserve’s 12th District, encompassing Alaska, Arizona, California, Hawaii, Idaho, Nevada, Oregon, Utah, and Washington, accounts for one-fifth of the U.S. labor force. Analysis shows that during prior recessions, roughly three-quarters of workers in this region lived in states with accelerating unemployment. In mid-2024, however, the share fell significantly short of that benchmark, reinforcing the conclusion that recent labor market fluctuations were temporary rather than systemic. n nAlthough unemployment rates in California, Idaho, and Oregon rose slightly after mid-2022, conditions across the 12th District stabilized in early 2025. The LMSI’s transparency and simplicity—counting states with worsening job markets—make it a practical addition to existing economic monitoring tools. Its geographic focus proves especially valuable when national indicators send mixed signals. n nLooking ahead, the San Francisco Fed plans to launch a dedicated data page with weekly LMSI updates, enhancing its utility for real-time economic assessment. n n— news from Federal Reserve Bank of San Francisco n
— News Original —nTracking Labor Market Stressn nThe labor market is an important barometer of overall economic health. Rapid increases in the unemployment rate are usually associated with the start of recessions. Therefore, a timely and accurate assessment of labor market conditions is essential for policymakers and market participants. n nAs the Federal Reserve began tightening interest rates in early 2022 to reduce inflation towards its 2% target, concerns grew about slowing the economy to the point where it would fall into recession. Two popular macro alarms sounded last summer. First, in July 2024, the Sahm (2019) rule indicated that the economy might be in a recession. The Sahm rule suggests that, when the national unemployment rate rises 0.5% or more relative to the lowest 3-month average over the preceding 12 months, it usually signals the economy is in a recession. Second, the difference in yields between long- and short-term Treasury securities remained negative from late 2022 until late 2024, a phenomenon known as a yield curve inversion. In the past, such inversions have reliably foreshadowed recessions a year in advance (see, for example, Bauer and Mertens 2018). n nDespite those warning signs, consumer spending, payroll growth, and business investment continued to expand, leaving forecasters divided on whether a downturn was imminent. Given these conflicting signals, we describe in this Economic Letter our new labor market measure that could help alert policymakers to rapidly deteriorating economic conditions on a timely basis. n nGeography matters n nNational averages of economic indicators hide important regional diversity. For example, while the unemployment rate has risen 2 percentage points in Colorado since mid-2022, unemployment has barely budged in Hawaii. Experience shows that recessions become self-reinforcing when job losses spread broadly across states and sectors (Hamilton and Owyang 2012). If economic weakness remains isolated, healthy regions can offset those that are struggling through tourist inflows, interstate supply chains, and federal transfers, thus preventing a nationwide contraction. n nOur new Labor Market Stress Indicator (LMSI) takes a different approach by examining unemployment patterns across all 50 states and the District of Columbia. Rather than relying solely on national aggregates, this indicator counts the number of states experiencing accelerating unemployment—defined as a state’s unemployment rate increasing at least 0.5 percentage point above its previous 12-month low. It can be interpreted as a state-level version of the Sahm rule. n nThe LMSI uses state-level unemployment insurance claims data, which provide a more timely reading into labor market conditions. Since official state unemployment rates are only available from 1976 onward, we rely on the recent work by Fieldhouse et al. (2024), who digitized historical state-level unemployment claims. Based on their statistical model, we extend their state unemployment rate estimates from 1948 to today, thus creating a comprehensive historical data set spanning 77 years. n nThis approach offers two main advantages over national aggregate measures. First, it captures the geographic breadth of economic stress, distinguishing between localized downturns and widespread recessions. Second, it accounts for the fact that different states may experience varying labor market conditions due to regional economic structures and policies. n nA simple state count, but a powerful signal n nFigure 1 shows the LMSI from January 1948 to June 2025, with shaded areas representing official recession periods, as dated by the National Bureau of Economic Research (NBER). Though not specifically for this purpose, the LMSI correlates very well with recessions throughout modern U.S. economic history. The national economy has invariably been in recession each time 30 or more states simultaneously experienced accelerating unemployment. n nThe indicator’s performance becomes even more compelling when we consider the proportion of the national labor force affected when the LMSI signals that a state is in distress. Adjusting for the relative population size of each state in this way provides a more intuitive read on the scope of deteriorating labor market conditions. n nFigure 2 shows this adjusted measure, based on state-level labor force data that are only available since 1976. In previous recessions, roughly 75% of the U.S. labor force has lived in states experiencing accelerating unemployment. By examining these two signals simultaneously, we are better able to distinguish genuine recession signals from temporary or localized labor market disruptions. n nFalse alarm n nThe July 2024 period provides a compelling case study for the LMSI’s value. Why did the national Sahm rule flash red even though there was no apparent recession? It is helpful to consider not only the number of states with stressed labor market conditions but also the share of the labor force that live in such states. For example, a handful of large states—California, Texas, and New York—account for more than one-third of the U.S. labor force. An unusually large jump in any one of them can significantly impact the national unemployment rate, even if labor market conditions in most other states remain stable. n nIn July 2024, nearly 70% of the labor force resided in a state with accelerating unemployment, slightly below the 75% mark typically associated with recession calls, as shown in Figure 2. However, the share of states meeting our LMSI-based rule barely cleared the 30-state threshold and quickly fell within one month, as shown in Figure 1. By contrast, in every post-1980 recession, the LMSI and the share of the labor force in distress quickly climbed and stayed elevated for months. Hence, the more limited geographic pattern of distress and its very short duration were notably different from previous recessions. n nAccording to June 2025 data, only the District of Columbia exhibits accelerating unemployment, corresponding to about 0.2% of the U.S. labor force—a far cry from the 75% mark. n nWe can also use the LMSI to develop a statistical model that estimates the likelihood of recession based on the number of states experiencing accelerating unemployment. This model indicates that the LMSI assigned a probability of at least 40% or higher to all previous recession periods, whereas it typically shows probabilities below 10% during expansion phases. n nBased on the latest data, the model predicts a recession probability of just 5%, suggesting that we are in an expansion phase and far from recession. Thus, the rapid improvement in labor market conditions suggests that the July 2024 signal reflected temporary factors rather than the onset of a sustained economic downturn. n nCanaries in the coal mine n nThe LMSI also provides valuable insights into regional economic conditions and how they tie in with national recessions. For example, in Figure 3, we rank states by the number of times during the previous 12 recessions they exhibited accelerating unemployment, as measured by our version of the state-level Sahm rule. On one side of the spectrum, Illinois, Missouri, North Carolina, New Hampshire, Nevada, Ohio, Oregon, Pennsylvania, Rhode Island, South Carolina, Tennessee, and Washington have flashed red in every single one of the past 12 recessions. On the other side, states like Nebraska, South Dakota, Kansas, North Dakota, Wyoming, Alaska, Hawaii, Iowa, New Mexico, Oklahoma, and Louisiana have flashed red less than half the time when a recession occurred in the national economy. n nWhat is different about that latter set of states that are not closely connected to national recessions, for example the Dakotas, Wyoming, and Alaska? These are states that rely heavily on industries that extract natural resources, where swings in the unemployment rate are less likely to be tied to the national business cycle. These differences allow us to get a clearer picture of the sources of labor market stress that may have national implications from those that are more regional in nature. n nThe 12th District n nThe Federal Reserve Bank of San Francisco is responsible for the 12th Federal Reserve District, which includes the states of Alaska, Arizona, California, Hawaii, Idaho, Nevada, Oregon, Utah, and Washington. Combined, these states represent one-fifth of the national labor force. We may therefore ask, How does the 12th District compare with the national economy? n nThe blue bars corresponding to 12th District states in Figure 3 offer a clue. Those states are evenly distributed across the spectrum: States like Washington and Oregon are the clearest bellwethers of national distress in labor markets, while states like Alaska and Hawaii are considerably more isolated from domestic economic conditions. n nConsidering that District states appear to evenly represent the range of state experiences, we consider another question: How well would a version of the LMSI perform if we base it only on 12th District economies? n nFigure 4 presents the share of the District’s labor force in states with accelerating unemployment, based on the method we used for the national economy. n nIn all previous NBER recessions, about three-fourths or more of the 12th District labor force resided in states with accelerating unemployment. In July 2024, the fraction of District states in this category fell well short of the 75% threshold. This reinforces the previous observation that recent national labor market signals were uneven and probably reflected temporary factors more than a generalized weakening of the economy. n nIt is important to note that since June 2022, unemployment rates in California, Idaho, and Oregon have each ticked up by close to a percentage point. However, during early 2025, labor market conditions appear stable across the District. n nConclusion n nThe Labor Market Stress Indicator (LMSI) described in this Letter provides a valuable complement to national recession indicators by revealing geographic variation in labor market weakness. The LMSI’s transparent methodology—simply counting states with accelerating unemployment—makes it easy to interpret while providing valuable insights into the geographic distribution of economic stress. This geographic lens becomes especially valuable when national indicators provide mixed or unclear signals about the state of the U.S. economy. n nAs of mid-2025, our indicator shows that the U.S. economy is not currently in recession and that labor market conditions remain relatively stable. The flexibility, clarity, and ease of interpreting the LMSI make it a useful addition to the standard tools used by economists, policymakers, and the public to monitor economic conditions. A new LMSI data page with weekly updates will be available on the SF Fed website in the coming months. n nReferences n nBauer, Michael D., and Thomas M. Mertens. 2018. “Economic Forecasts with the Yield Curve.” FRBSF Economic Letter 2018-07, March 5. n nFieldhouse, Andrew J., David Munro, Christoffer Koch, and Sean Howard. 2024. “The Emergence of a Uniform Business Cycle in the United States: Evidence from New Claims-Based Unemployment Data.” Brookings Papers on Economic Activity, Spring 2024, pp. 265–319. n nHamilton, James D., and Michael T. Owyang. 2012. “The Propagation of Regional Recessions.” Review of Economics and Statistics 94(4), pp. 935–947.