New contracts reported by anti-labor consultants in LM-20 Filings

November 13, 2021

Anti-labor consultants hired by employers are required to report details about their hiring to the Department of Labor’s Office of Labor-Management Standards. Law firms that do not directly engage with employees but only provide advice to employers on how to fight unions are currently exempt from reporting.

This graph shows the upper and lower bound for the number of contracts between these consultants and employers.

Ignoring data before 2012, when the published number of clients were not reliable, the big change is a sharp decline in the number of new hirings since 2016.

If the “Persuader Rule” had gone into effect after 2016 instead of being rescinded by the Trump administration, there probably would have been a sharp increase instead. That rule would have required reporting by law firms that provide advice to employers on how to combat unions.

Data Notes #

Consultants are supposed to file notice of their hiring within 30 days in an LM-20 form. If they have already submitted a form for that fiscal year, they are supposed to submit an amended form.

Different consultants seem to interpret “amended form” differently.

Some consultants, resubmit all the information previously submitted plus the information about the new hiring. So if the original LM-20 form listed one client, and the consultant was hired by a new client, the consultant would submit an amended form that listed two clients.

Other consultants only submit new information in an amendment. So, in the same scenario, the second “amended” form would only list one client, the new one.

The Department of Labor publishes information about the number of clients listed in an LM-20 form. We can use that to construct a lower and upper estimate of the number of new contracts signed per year.

First, we can just look at the final amendment in a fiscal year. If we sum up all the number of clients listed in that final amendment of every LM-20, that would be a minimum estimate of the number of new hirings in that year. This method will undercount the hirings done by consultants that file amendments that only contain information about new hirings.

Second, we can sum up all the listed clients for every amendment, we can get an maximum estimate. This estimate will overcount hirings done by consultants that file amendments that keep the same information from previous versions and add new information.

We could get a more accurate count by counting the distinct employers listed as clients, but the data quality is not adequate to do that yet. I will clean up the data to that, though.