Evaluating the impacts of government business assistance programmes: approaches to testing additionality

Evaluation of government policies and programmes should try to test for the additionality of impacts over what would otherwise have occurred. Such testing can enhance the evaluator's objective analysis of outcomes. We report on our experiences in developing and testing a largely quantitative approach to these issues, using econometric techniques to evaluate a business assistance programme. We use a new wide-coverage business database linking firm administrative and survey information. This enables us to compare the changes in performance of assisted firms against matched similar unassisted firms. The econometric estimates find a statistically significant impact of the business assistance programme on the participating firms. In addition, theses econometric results are consistent with qualitative studies that examined contextual issues in more depth. We discuss the advantages and risks for this approach as compared with alternatives, as well as the challenges for developing and resourcing the required capabilities.