[FEATURE] Employment Estimates

[Eric-Liu-SANDAG]

Is your feature request related to a problem? Please describe

We may need employment Estimates at some point for data requests for the potential Forecast/SR16 development work.

Describe the solution you'd like

Create employment Estimates at the MGRA level. This could be done using a mix of data sources which include but are not limited to:

• LEHD LODES data
• BLS QCEW point data
• CA EDD Confidential file data

Describe alternatives you've considered

N/A

Additional context

At some point, link the CES working paper here:

[GregorSchroeder]

CES working paper released and added to Census LEHD repository.

https://github.com/SANDAG/Census-LEHD/issues/11
https://www.census.gov/library/working-papers/2025/adrm/CES-WP-25-52.html

[GregorSchroeder]

A potential methodology to produce annual employment estimates would be as follows,

1. Start with the BLS QCEW for the given estimate year
   a. See SANDAG/BLS: Bureau of Labor Statistics ETL Project
2. Use this to scale the most recently available LEHD LODES dataset
   a. LODES is on a two-year lag release schedule (2022 released end of 2024, 2023 released end of 2025)
   b. Scaling done at state level? Within NAICS? Using BLS QCEW Q2?
   i. See Census-LEHD/lodes/documentation at main · SANDAG/Census-LEHD for working papers and documentation to assist in answering these questions
3. Distribute the scaled block-level data down to the LCKey-level using the CA EDD confidential dataset.
   a. See SANDAG/EMPCORE: SANDAG's EMPCORE employment database
4. Aggregate this up to the MGRA-level to produce annual employment estimates.

[GregorSchroeder]

We would also want to include self-employed persons not covered in unemployment insurance. My though would be to use the ACS 5-year B24080 "Self-employed in own not incorporated business workers" that is available at the blockgroup level. This brings up two issues though.

1. How to disaggregate from blockgroup to MGRA and/or LCKey
2. How to or even if to scale since the ACS 5-year operates on a one year lag (2024 released in late 2025).