A Microsimulation Model of the U.S. Unemployment Insurance System
Table of Contents
Table of Contents
Report Publish Date: December 2021
This report simulates U.S. unemployment insurance before and during the COVID-19 pandemic and estimates how policy reforms would affect who benefits from the system.
Primary Takeaways
The authors found that:
- Two policies that make weekly benefits the same across states slightly reduce the number of weeks paid (98.3% and 97.6% of baseline). They increase the average weekly benefit (102.3%) and raise total unemployment insurance costs (105.5%).
- Setting unemployment insurance at 26 weeks increases the number of weeks paid (114.5% of baseline). The average weekly payment falls (87.3%), but total program costs rise (113.2%).
- Including self-employed workers increases the number of weeks paid (118.5% of baseline) and total program costs (121.9%). The average weekly payment falls (84.4%).
- Including unauthorized immigrants or full-time students increases the number of weeks paid (105.6% and 102.1% of baseline) and total program costs (105.2% and 102%). The average weekly payment falls slightly (94.7% and 97.9%).
Overview and Objectives
The COVID-19 pandemic put millions of people out of work and exposed weaknesses in U.S. unemployment insurance policies and programs. The authors wanted to show how various possible reforms might affect the system.
Hypothesis or Approach
The authors reviewed the literature on unemployment insurance and used Current Population Survey data to develop a microsimulation model that replicates the rules and standards of the system. They modeled unemployment compensation and pandemic-era unemployment benefits. They also simulated policy reforms to see how they would change weeks compensated, total expenditures, and average payments compared to the baseline of pre-pandemic unemployment compensation.
How This Influences Change
Understanding how different reforms affect the system can help policymakers make decisions that support more workers.
Research: Go Deeper
To better understand the kinds of long-term reforms that are needed, we use the Current Population Survey (CPS) to develop a microsimulation model of the major elements of the U.S. unemployment insurance (UI) system prior to and for the duration of the COVID-19 pandemic. We build on methodology developed by Cortes and Forsythe (2021) and Ganong et al. (2020) to develop the microsimulation model. Our microsimulation model consists of replicating the rules/eligibility standards of three components of the UI systems in each U.S. state: 1) non-monetary eligibility, 2) monetary eligibility, and 3) benefit generosity to determine a weekly benefit amount (WBA) for each respondent in our CPS Basic Monthly sample. In the pandemic period, which we define as starting in March 2020, we model full and partial Unemployment Compensation (UC or traditional UI), Pandemic Unemployment Assistance (PUA), Pandemic Emergency Unemployment Compensation (PEUC), and Federal Pandemic Unemployment Compensation (FPUC). We benchmark the results of our model using publicly available aggregate administrative data reported by the DOL and BEA for number of UI weeks compensated, total expenditures, and average payments. We use this model to simulate a series of counterfactual policy reforms to the unemployment insurance system to test how alternative decisions by policymakers could have changed population-level and subgroup outcomes.
University of California Berkeley, December 2021
Figure 1: Change in Unemployment Rate Following Onset of U.S. Recessions
Research Team
This study and report was conducted and created by the following people.
- Hawkins J
- Swanbeck S
Similar Research to Explore
Pandemic-Driven Health Policies To Address Social Needs And Health Equity
Household Experiences in America During the Delta Variant Outbreak
