Industry Projections Methodology
Updated over a week ago

Lightcast's industry projections are built from Lightcast final industry data. Industry data comes from the Bureau of Labor Statistics' Quarterly Census of Employment and Wages (QCEW) dataset, plus some supplemental datasets that provide information for industries not covered by QCEW.

Projections are calculated for each 6-digit NAICS and county combination individually. For each county-industry combination, we begin with final Lightcast industry data and produce three 10-year projection lines based on 5, 10, and 15 years of historical employment data, respectively. The 5-, 10-, and 15-year historical trends might look like this:

We then project each of those trend lines forward 10 years into the future. The lines are simple projections of historical trends:

The three projections, based on the various historical trend lines, are then combined into one projection. The projection line is damped, meaning that projected growth or decline is slowed a bit each year. At this point our projection looks like this:

So far, our industry projection has been based solely off of Lightcast industry data. Next, we make adjustments to outside sources. First, we adjust to the National Industry-Occupation Employment Matrix (NIOEM), another BLS dataset. NIOEM includes national-level employment projections data. We adjust our projections by using the year-over-year percentage change in employment shown in NIOEM.

We also adjust to long-term industry projections as published by each state. These projections are published approximately every two years. We incorporate the state-level total projection figures into our projections, adjusting their weight depending on the age of the data. The older the data is for any given state, the less weight it is given.The final result of this process is Lightcast industry projections:

Sources: QCEW, NIOEM, each state's industry projections (example: Illinois)

Did this answer your question?