Association for Public Policy Analysis & Management

A number of studies have used workers' compensation claims linked to administrative earnings data to estimate the causal effect of workplace injury on labor market outcomes for workers in the United States. These studies typically find that injuries resulting in temporary disability can result in meaningful long-term earnings losses and reduced economic well-being, even when the injured worker is not determined to be permanently disabled. However, the impacts of workplace injury have been studied less frequently in the continental European setting, where a range of institutional differences from the United States (including the generosity of disability benefits, employment protections, and the scope of collective bargaining agreements) might alter the relationship between injury, disability, and long-term earnings loss.

This study provides a new international perspective on the impacts of workplace injury by estimating post-injury earnings losses for a large, nationally representative sample of blue-collar workers in Italy. We use a new Italian dataset that combines a national random sample of matched employer-employee data with records from the Italian workers’ compensation agency.

We compare the employment and earnings of injured workers up to five years after the date of injury with outcomes for workers who did not experience any occupational injuries. We also compare workers who suffered a permanent disability with those whose disability was only temporary. We apply propensity score matching to the rich set of variables available in our data to construct a control group for the injured workers that is observably equivalent in terms of job characteristics, employer characteristics, and pre-injury career trajectories. We conduct analyses stratified on gender.

Consistent with studies from the US setting, we find that a severe occupational injury decreases the probability of employment. Injured workers who return to work also experience a decline in their average weekly wages at both three and five years after the injury. Compared to injured men, injured women have a smaller reduction in the weekly wage, but a larger reduction in employment. We also find, as in the US-based literature, that disability ratings are strongly predictive of earnings and employment losses. Finally, we find that the employment of permanently disabled workers does not respond to work disincentives triggered by the payment of disability benefits.

When we examine outcomes for workers with only temporary disabilities, however, we do not find statistically significant evidence of long-term earnings losses relative to matched controls. This is in marked contrast to US-based studies that have found long-term consequences of temporary disability. We view these findings as evidence for the important role that labor market institutions can play in mediating the pathway from injury to long-term disability and earnings losses. At the same time, the results suggest that very severe occupational injuries have detrimental consequences for individuals’ long-term economic well-being regardless of the different degrees of generosity of workers’ compensation policies and rules.