A Better Way to Analyze Job Market Risk
Dr. Peter Linneman on which sectors are poised for growth and which are likely to contract.
We have regularly discussed a covariance analysis that examines how employment behaves in individual metropolitan areas based on national economic changes. We applied the same analysis to major employment sectors (as defined by the Current Employment Statistics coding system). The analyses are based on monthly employment data from January 1980 through June 2025 (534 time periods for each regression). Specifically, for each employment sector, we estimate a statistical equation, which summarizes how a 100-basis-point percentage change in total national employment affects sectoral employment growth rates. The equation consists of a constant “alpha” for each market and a “beta,” which is a multiplier applied to the national percent change in employment.
Before discussing the alpha and beta coefficients, it’s important to understand the relative strength of the models. The R-squared statistic represents the portion of the dependent variable (sector growth) explained by the independent variable (total employment) in each regression model. In other words, R-squared indicates how well the model “fits” the data and is therefore able to explain historical performance and predict future performance.
Both goods and service-producing sectors have high R-squared statistics, with the highest (most desirable) seen in the private service-producing sector (highlighted in ”go” green). The regression models for trade, transportation and utilities (including wholesale and retail trade) and professional and business services also have high explanatory values. However, mining and logging, utilities and government (highlighted in cautionary yellow) all have low R-squared values, indicating little explanatory power.
The alpha indicates sectoral growth that is independent of national growth. If there is no national job growth, then the alpha is the expected annual percentage change in sectoral MSA employment. Thus, higher alpha sectors have built-in employment growth dynamics. The high alpha sectors, with green highlighting include financial activities, professional and business services, private education and health services, and state government. In contrast, low alpha sectors are highlighted in yellow and are dominated by goods-producing sectors such as mining and logging and manufacturing.
The beta for the U.S. is definitionally equal to 1.0. A sector with a beta of 1.0 registers (on average) an increase of 100 basis points in employment growth (plus its alpha) when national employment rises by 100 basis points. A beta that is less than 1.0 indicates that the sector does not boom (or bust) to as great an extent as the national economy, while a beta of greater than 1.0 indicates that such a sector experiences swings of greater magnitude (around its trend) than percentage changes at the national level.
The high beta sectors (highlighted in orange) are again dominated by goods-producing segments, such as construction and durable goods manufacturing, but also include the leisure and hospitality sector. In contrast, the most stable, low-beta sectors include utilities and government (highlighted in gray). Note that unlike the other metrics, the desirability of beta values depends on the economic outlook, with higher betas sought when the economy is growing.
The breakeven statistic reflects the interaction between the alpha and the beta, indicating the level of U.S. employment growth required to sustain positive job growth within a specific sector. In general, lower breakeven points (highlighted in green), indicate greater resilience within a sector relative to changes at the national level. For example, the private education and health services sector has the lowest breakeven point of -5.8 percent. This means that as long as U.S. employment growth is greater than -5.8 percent, sector growth is expected to be positive. State and local government and the financial activities sectors also have low breakeven points. In contrast, utilities, mining and logging and manufacturing have high breakeven points (highlighted in yellow), indicating that total U.S. employment growth must be relatively high in order for those sectors to see contemporaneous positive growth.
When the national employment outlook is strong, expect sectors with high alphas and high betas to do well. However, when the national outlook is weak, sectors with high alphas and low betas will best hedge against an economic downturn.
You must be logged in to post a comment.