An evaluation of export promotion programmes with repeated multiple treatments.

Export promotion programmes usually consist of the provision of multiple services and aids, including consultancy, trade missions and international fairs and business-to-business meetings, of which firms can take advantage either simultaneously or at different moments in time. This constitutes an unusually complex setting for programme evaluation. Relying on assumptions of sequential ignorability extended to the multiple-treatment framework, and exploiting the programme participation data of an Italian region (Tuscany), we estimate the treatment effect of different services and aids on multiple aspects of the firms export performance using a marginal structural model that adjusts for dynamic confounding by means of inverse-probability-of-treatment weights.