I can't seem to find the report that they're referencing and the conclusion seems to be directly at odds with the Bureau of Labor Statistics' monthly assessment

I can't seem to find the report that they're referencing and the conclusion seems to be directly at odds with the Bureau of Labor Statistics' monthly assessment of multiple job holders[1]. Does anyone have a link to the publication? [1]https://fredblog.stlouisfed.org/2019/08/moonlighting-in-the-spotlight/ https://fredblog.stlouisfed.org/2019/08/moonlighting-in-the-... EDIT: Here's the report https://www2.census.gov/ces/wp/2020/CES-WP-20-26.pdf https://www2.census.gov/ces/wp/2020/CES-WP-20-26.pdf. One explanation for the discrepancy seems that the number of individuals holding multiple jobs when measured on a monthly basis has been decreasing over time while the number working at more than one job at any time during a quarter has increased over time. People seem to be more likely to occasionally work a second job but less likely to consistently work a second job. The paper's authors disagree with this take but I'm not sure I find their explanation totally convincing.