Working Paper: CEPR ID: DP17288
Authors: Alexander Bick; Adam Blandin
Abstract: Economists have recently begun using independent online surveys to collect national labor market data. Questions remain over the quality of such data. This paper provides an approach to address these concerns. Our case study is the Real-Time Population Survey (RPS), a novel online survey of the US built around the Current Population Survey (CPS). The RPS replicates core components of the CPS, ensuring comparable measures that allow us to weight and rigorously validate our results using a high-quality benchmark. At the same time, special questions in the RPS yield novel information regarding employer reallocation during the COVID-19 pandemic. We document that 26% of pre-pandemic workers were working for a new employer one year into the COVID-19 outbreak in the US, at least double the rate of any previous episode in the past quarter century. Our discussion contains practical suggestions for the design of novel labor market surveys and highlights other promising applications of our methodology.
Keywords: COVID-19; Labor Market; Employer Reallocation; Surveys
JEL Codes: J63; J64
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| COVID-19 pandemic (H12) | employer reallocation (J63) |
| pre-pandemic employment (J68) | separation by March 2021 (J12) |
| separation (Y40) | employment with a new employer (J63) |
| shorter employer tenure (J63) | likelihood of separation (J63) |