The Long-Term Effects of Student Absence: Evidence from Sweden

Swedish cohorts born in the 1930s, hand-collected historical elementary school records linked to administrative registers over the full life cycle

Swedish children in elementary school in the 1930s-1940s, followed into old age

The only study in this cluster that follows absence to lifetime earnings. Historical school ledgers recording day-level absence are linked to Swedish administrative income, education and mortality registers. The published abstract states the results but does not itself specify the identification strategy; I was unable to open the full text (Economic Journal and the IFS/DICE working paper versions were both paywalled or 403), so the precise source of exogenous variation - and in particular whether sibling/family fixed effects are used - is NOT verified here.

Elementary school absence significantly reduces contemporaneous academic performance, final educational attainment, and labour income across the whole life cycle. The pattern the authors emphasise is dynamic - small immediate learning losses that compound into larger human capital deficits over time - which is consistent with the small per-day contemporaneous effects found in US within-student designs but larger cumulative consequences.

Historical cohort with register linkage, so the outcome measurement is excellent, but the identification of exogenous absence variation is not verifiable from the abstract alone. Absence in 1930s Sweden was still driven by illness, farm labour and family circumstance, all of which have heritable components and independently predict adult income. Until the identification is confirmed this should be read as the best available long-run evidence rather than as a clean causal estimate, and it is not load-bearing for the topic verdict.