Butler Baldwin and Johnson 2006 — The Effects of Occupational Injuries After Returns to Work Work Absences and Losses of On-the-Job Productivity

Summary

Butler, Baldwin, and Johnson (2006) extend the 1995 findings on post-injury employment patterns by quantifying productivity losses for the first three years after injury for Ontario workers with permanent partial disabilities (PPDs). They decompose losses into an absentee effect (weeks not worked) and an earnings effect (reduced wages/hours while present), and estimate determinants of each separately for workers in stable versus unstable post-injury employment. Even workers who return to stable employment incur significant ongoing losses — primarily via the earnings effect — which are invisible to standard return-to-work metrics.

Key Claims

• Productivity losses are large and persistent for both stable and unstable workers: stable group loses $43\%$ in year 1, recovering to $6\%$ in year 2 then rising to $16\%$ in year 3; unstable group loses $58\%$, $29\%$, and $52\%$ across the three years.
• For stable workers, year-1 losses are dominated by the absentee effect (lengthy first spell of absence); years 2–3 losses shift to the earnings effect (lower wages/hours while present).
• For unstable workers, the absentee effect dominates across all three years, reflecting repeated spells of injury-related absence.
• Human capital effects differ by stability group: high school education and specific tenure reduce losses for stable workers; vocational training reduces losses $\approx 15\%$ per year for unstable workers (portable skills substitute for firm-specific capital when workers cycle across employers).
• Accommodation effects also differ: flexible schedules significantly reduce losses for stable workers (years 1–2); light work reduces unstable worker losses in year 3; but flexible schedules can increase unstable workers' absentee losses in year 2 — interpreted as "trial work" failures where accommodated workers re-absent after failing the accommodation test.
• The earnings effect — reduced on-the-job productivity during periods of actual employment — is a previously unmeasured but substantial source of permanent disability costs even among workers who "successfully" return to work.

Concepts Introduced or Extended

• Disability Management and Return to Work

Entities Mentioned

• Richard J. Butler
• Marjorie L. Baldwin
• William G. Johnson

Quotes

"It is commonly believed that a return to work marks a recovery from the disabling effects of an occupational injury and an end to the economic losses associated with the injury, but two recent studies provide evidence that contradicts the common wisdom."

"The long-term effects of permanent impairments are both greater and more persistent over time than was previously thought to be the case."

My Take

The productivity-loss decomposition is the paper's primary methodological contribution — it moves the field beyond counting weeks absent to measuring the earnings gap even during weeks present. The "earnings effect" finding for stable workers is underappreciated: these workers are present and employed, yet consistently earn less than their preinjury benchmark, and this gap actually grows from year 2 to year 3. The trial-work interpretation of flexible schedules increasing unstable workers' absentee losses is speculative but plausible. The same Ontario data limitations apply as in the 1995 paper (pre–Americans with Disabilities Act (ADA), pre-2000, Canada-specific workers' compensation (WC) administration). The three-year restriction that reduces the sample from $\approx 4{,}400$ to $800$ cases is a validity concern, though sensitivity checks with larger samples support the qualitative conclusions.