Economic theory suggests that an extension of the maximum length of entitlement for unemployment benefits increases the duration of unemployment. Empirical results for the reform of the unemployment compensation system in Germany during the 1980s are less clear. The analysis in this paper is motivated by the controversial empirical findings and by recent developments in econometrics for partial identification. We use extensive administrative data with the drawback that registered unemployment is not directly observed. For this reason we bound the reform effect on unemployment duration over different definitions of unemployment. By exploiting the richness of the data we use a nonparametric approach without imposing critical parametric model assumptions. We identify a systematic increase in unemployment duration in response to the reform in samples that amount to less than 15% of the unemployment spells for the treatment group.