Evaluating the Net Effect of the ISDE Subsidy Scheme in the Netherlands – Comparison of Evaluation Methods to Estimate Additionality

The national government encourages Dutch households and businesses to use less natural gas and more sustainable heat, which reduces CO2 emissions. The Sustainable Energy Investment Subsidy (Dutch: Investeringssubsidie Duurzame Energie or ISDE), which has been in place since 2016, provides a subsidy for the purchase of solar boilers, heat pumps, biomass boilers and pellet stoves. The scheme is meant for both private individuals and business users. An important element in the evaluation of the ISDE was the ‘additionality’ of the scheme. This was evaluated by the Dutch organization for applied natural scientific research (TNO) in 2018 and by SEO Economic Research (SEO) in 2019. In this paper the evaluation methods are described, compared and critically reviewed in order to identify uncertainties and limitations of the methods and possible improvements. As is evidenced by the fact that different evaluation methods come to very different results for the additionality (percentages) of the ISDE devices, it becomes apparent that the ‘true’ additionality of the ISDE scheme is hard to determine. It depends on many factors elaborated on in this paper. An important limitation in the evaluation is the absence of a control group, which is why was opted for a survey-based method in which questions were asked about what would have happened without ISDE. The reasons for the (mostly) low additionality are (probably to a large extent) related to the evaluation methods used. Several biases in surveys could result in an overestimation of free-riders, in turn indicating a low additional effect. Quantifying free-riders through surveys remains a challenge. It deserves attention to investigate the possibilities to optimize survey questions as much as possible towards this goal. A possible manner to achieve this is by including additional questions about other non-financial motives for installing the appliance and what other policies played a role. Adding a question about the age of the replaced heating installation could help to identify end-of-life replacements, which is nonadditional. This paper shows that explaining the used definition (scope) of additionality in detail is key to let readers interpret results correctly. It also shows that evaluating whether non-financial motives or another policy instrument played a deciding role in the installed appliances is paramount in order to estimate additionality, which is a complex task for which there is no universal solution.