Posted by: Josh Lehner | January 8, 2013

State Employment Revisions, 2012q2

This morning the Bureau of Labor Statistics released the 2012q2 QCEW data for all states. Regular readers will recognize that this is the series we use to benchmark the state employment data. What follows below is a quick update on the state by state employment revisions discussed more thoroughly in previous posts (here for 2012q1 and here for 2011q4.) The QCEW is a near complete count of employment and due to the much larger coverage size (relative to the CES), the data release lags a few extra months. With this latest release we now have a good understanding of when and where the state level data revisions will be for the first half of 2012.

This first graph shows year-over-year employment growth for both the national data and the sum of individual states (plus D.C.) The dark blue line is the currently published national data while the dashed dark blue line is the national data adjusted for the preliminary benchmark announced a few months ago. The red line is the sum of states based on the CES while the light blue lines is the sum of states based on the latest QCEW.

There are 4 main items to highlight. First, the last state benchmarked data point is June 2011 so the currently published CES data (available through November) is now 17 months off benchmark. Second, this can, and does, lead to estimation errors in the monthly CES. Based on the QCEW the state level revisions will be up significantly when officially released in March 2013. However the magnitude of the revisions at the state level does not appear to be worsening in 2012, meaning BLS is making better estimates than they did originally in 2011 after taking over the state level estimates. Third, the previously announced national benchmark for March 2012 (+453,000 private sector) closes the national – sum of states gap. Fourth, however, by June the QCEW sum of states again outstrips the national data, albeit for just the one month of data. If the 2012q3 data continues to show this pattern (QCEW above national CES) then expectations are that the next national benchmark will again be upward.


The fact that the QCEW is so much stronger over the past year indicates that the monthly state level employment data is significantly undercounting employment. In June, the CES sum of state data show private sector year over year job growth of 1.8% while the QCEW show growth of 2.4%. In terms of the number of jobs, the difference is about 700,000. This difference, in magnitude, is about 50% larger than the national benchmark (0.6% compared with 0.4%). With that being said, each individual state will differ. The graphs below show estimated benchmarks for June 2012. The graph on the left is the revision as a percentage of the currently published CES figures, while the graph on the right is the size of the revision in thousands of jobs.


The 2012q3 QCEW data is scheduled to be released March 28, 2013.


  1. Josh, this is interesting and timely analysis of the data.

    Could you also add a graph showing the same data as in the first chart, but for each series, instead of year-over-year percent growth, show it as an index with January 2011 (or June 2011) being equal to 1.000? This would aid in understanding how far off track the original estimates diverge from overall cumulative growth.


  2. […] the current published series is not getting larger. This appears to be the case across many states, as discussed previously. A more complete look at each industry is available here (PDF): […]

  3. […] know, one sleeps on the QCEW data at their own peril when examining employment trends – see my last state by state look. Working with Mark over email, the following takes a look at these expected revisions across all […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s


%d bloggers like this: