Reconsidering the economic impact of the EU manufacturing and export provisions

Abstract

Background

The European Commission is considering two provisions which would permit limited generic drug production in the EU prior to supplemental protection certificate expiration. A recent study by Vicente and Simões addressed the potential economic consequences of these provisions and concluded that the provisions could result in 178, 698 million (M) new units produced, 8,890 new direct jobs, 26,670–35,560 new indirect jobs, and 37 new medium-sized pharmaceutical enterprises over the period 2014–2022. This paper assesses the accuracy of these findings through a critical review of the Vicente and Simões analysis.

Methods

We replicated the Vicente and Simões model and applied a series of adjustments including (1) correction of an arithmetic error, (2) introduction of a counterfactual analysis, (3) substitution of evidence-based parameters and (4) accounting for statistical uncertainty. Like Vicente and Simões, we model effects on four key outcomes: units produced, medium-sized pharmaceutical enterprises launched, and direct and indirect jobs created. Given data limitations, we were unable to replicate the Vicente and Simões export provision model.

Results

Our adjustments yield estimates of the economic effects that are substantially smaller than those reported by Vicente and Simões. Our revisions suggest the provisions would lead to 124,992 M new manufactured units (a 30% reduction relative to Vicente and Simões), 1,898 new direct jobs (78% reduction), 6,642 new indirect jobs (75–81% reduction), and 14 new medium-sized pharmaceutical enterprises (62% reduction). When parameter uncertainty is considered, the estimate of the number of direct and indirect jobs created is not statistically distinguishable from zero.

Conclusion

The Vicente and Simões analysis contains several limitations, most notably a substantial arithmetic error. While difficult to assess, the study implicitly assumes that increased demand leads to job gains in the generic sector, but that reduced demand does not lead job losses in the branded sector. Addressing these limitations substantially reduces the estimated economic effects of the proposed EU provisions.