[Paper Review] Further Education During Unemployment

Pauline Leung, Zhuan Pei|arXiv (Cornell University)|Dec 28, 2023

Labor market dynamics and wage inequality4 citations

TL;DR

This study uses high-quality Ohio administrative data to estimate the labor market effects of unemployed workers enrolling in postsecondary education, finding a six percent average earnings gain three to four years after enrollment, driven primarily by industry-switchers entering healthcare. The analysis bridges dynamic treatment effect methods and reveals significant private and social returns to retraining, with break-even points at 8 and 14 years, respectively.

ABSTRACT

Evidence on the effectiveness of retraining U.S. unemployed workers primarily comes from evaluations of training programs, which represent one narrow avenue for skill acquisition. We use high-quality records from Ohio and a matching method to estimate the effects of retraining, broadly defined as enrollment in postsecondary institutions. Our simple method bridges two strands of the dynamic treatment effect literature that estimate the treatment-now-versus-later and treatment-versus-no-treatment effects. We find that enrollees experience earnings gains of six percent three to four years after enrolling, after depressed earnings during the first two years. The earnings effects are driven by industry-switchers, particularly to healthcare.

Motivation & Objective

• To estimate the labor market effects of retraining among unemployed workers in the U.S., broadly defined as enrollment in postsecondary institutions.
• To address the research gap on the long-term impacts of direct enrollment in community colleges and technical programs, as opposed to government-sponsored training programs.
• To bridge two strands of dynamic treatment effect literature by estimating both treatment-now-versus-later and treatment-versus-no-treatment effects.
• To conduct a cost-benefit analysis from private and social perspectives, assessing break-even timelines and return on investment for unemployed workers.

Proposed method

• The study links administrative data from Ohio’s unemployment insurance system, quarterly wage records, and postsecondary enrollment and credential data from public institutions.
• It employs a matching method to compare labor market outcomes of unemployed workers who enrolled in postsecondary education (enrollees) with observably similar non-enrollees within two years of layoff.
• The method estimates dynamic treatment effects by comparing outcomes across time, accounting for staggered enrollment timing and identifying treatment effects relative to a counterfactual of delayed or no enrollment.
• It uses a longitudinal sample of UI claimants from 2004 to 2011, with follow-up through 2017, to assess long-term earnings and employment outcomes.
• The cost-benefit analysis computes private and social net present values, break-even points, and internal rates of return (IRR) based on earnings gains and investment costs.
• It replicates prior findings on UI benefit extensions increasing enrollment, using temporal variation in benefit duration to validate the causal interpretation of enrollment effects.

Experimental results

Research questions

• RQ1 What are the long-term earnings effects of unemployed workers enrolling in postsecondary education, particularly in community colleges and technical programs?
• RQ2 How do earnings outcomes vary by timing of enrollment, and what is the effect of enrolling now versus later?
• RQ3 To what extent are earnings gains driven by workers who switch industries, especially into healthcare or construction?
• RQ4 What are the private and social returns to retraining, and when do the investments break even in terms of net benefits?
• RQ5 How do extended unemployment insurance benefits influence the likelihood of retraining among unemployed workers?

Key findings

• Enrollees experience a temporary earnings decline during the first two years post-enrollment, followed by a sustained average earnings gain of six percent in the third and fourth years after enrollment.
• The earnings gains are primarily driven by workers who switch industries, particularly those entering healthcare-related fields after completing relevant coursework.
• A longer follow-up of an early subsample indicates that the earnings gains persist and widen over a ten-year period, suggesting durable labor market benefits.
• The private investment in education breaks even after 8 years, while the social investment (including government subsidies) breaks even after 14 years.
• The private internal rate of return (IRR) is 15 percent, and the social IRR is 8 percent, indicating strong economic justification for retraining policies.
• A 10-week extension in unemployment insurance benefits is associated with a ten percent increase in enrollment, equivalent to approximately 1,200 additional enrollees per year in Ohio.