Constructing a Coincident Indicator of the US Labor Market

The state of the labor market has been the subject of heated debates.

A divergence has emerged between popular employment indicators, calling into question which indicator we should trust.

Non-farm payrolls continue to post monthly gains, even after downward revision of the benchmark, while the unemployment rate has risen by nearly 100 bps over the past 15 months.

How do we harmonize these seemingly contradictory messages to accurately assess the current state of the labor market?

Building on the work of business cycle legend Geoffrey Moore, we can create a “Coincident” labor market composite index that helps reduce some of the confusion.

As the name suggests, a “Coincident” indicator assesses current labor market conditions. A “Leading” indicator sees future turning points or changes in the Coinciding Indicator.

Non-agricultural payrolls, aggregate weekly hours, household employment level, U3 The Unemployment Rate and the Insured Unemployment Rate have all been reliable indicators of the labor market for decades.

As the chart below shows, nonfarm payrolls and aggregate weekly hours continue to rise, while household employment peaked in November 2022. The unemployment rate is rising rapidly, and the insured unemployment rate is off the cycle low, but has held flat for the past few months.

Rather than choosing which indicator is the best and ignoring the rest, it is much more useful to aggregate all these confidence measures into a single composite index. Any individual indicator is subject to failure at a tipping point due to measurement errors or other idiosyncratic factors, but aggregating different metrics from different surveys and reports helps identify some of the conflicting signals.

To create this coincident employment index, we first standardize each indicator for volatility and then aggregate an equal weighted mix of these standardized values.

The chart below shows the quarterly average level of the EPB Coinciding Employment Index. The most important point to note is that the Coincident Employment Index has declined so far in Q3, now below the average in Q2 and Q1.

The chart below shows the periods when there was a quarterly decline in the EPB coincident employment index of more than 0.1%.

A false signal was recorded in 1995, which was an economic recession that required the Federal Reserve to cut interest rates.

All other instances of quarterly declines in this 5-factor composite index occurred around or during recessionary periods.

The labor market slowed before posting this quarterly decline in Q3, as the chart below shows the three-quarter slope from 2022.

At the beginning of 2022, the EPB Coincident index rose sharply because the labor market was very hot. The growth rate started to decline throughout 2023, and now, the index has a negative slope for the last three quarters.

The chart below shows the periods in which the EPB Coincident Employment Index declined compared to two quarters ago. As the chart shows, there are no spurious signals since 1973, and the signal overlaps closely with recessionary periods.

This signal was triggered:

• A quarter after the 1974 recession began

• A quarter before the 1980 recession began

• 0 quarters before the 1982 recession began

• One quarter before the 1991 recession began

• 2 quarters before the 2001 recession started

• A quarter before the 2008 recession began

On average, the signal was triggered 0.66 quarters before the recession started.

To reiterate, this is a Coincident measure that is not designed to be forward-looking, but rather to provide an accurate assessment of current conditions.

This composite indicator approach is enhanced by coupling signals from Coincident Indicators with signals from Leading Indicators such as the Employment Index, which we regularly highlight in our research publications.

The EPB 5-Factor Coinciding Employment Index is flashing red, signaling a potential contraction in the labor market.

At the moment, readings from five reliable measures of the labor market are mixed, but when we equate all factors and take a collective reading, we see weakness, which likely calls for interest rate cuts.

The Coincident Employment Index is subtracted from the level of household employment and the U3 unemployment rate, which is partially offset by non-farm wages and aggregate weekly hours.

The biggest risk for the Federal Reserve and market participants is to ignore parts of the labor market basket and focus only on variables that confirm previous views.

The EPB Coincident Employment Index is beneficial because it is a broad measure and a perfectly consistent way to track the current state of the labor market on a monthly basis.

Of course, it is important to track an even wider range of data points, including critical leading indicators, but this index provides a massive improvement in interpreting the state of the labor market.

Our objective reading from the Coinciding Employment Index is that the labor market is weakening and possibly contracting, which has a history of coinciding with economic downturns and Federal Reserve interest rate cuts.

We provide clients with an updated reading of this coincident employment index after each Employment Report.