ODISSEI Lunch Lecture: The equitability of policy interventions: the Dutch decentralization of the social domain

On Thursday 13 October between 12-13 hrs, Mark Verhagen (University of Oxford) will give an online Lunch Lecture on his use of Dutch administrative data at Statistics Netherlands (CBS) and machine learning to study the effects of decentralization of government tasks: ‘Using Causal Trees and Registry Data to Study the Equitability of Policy Interventions: The Dutch Decentralization of the Social Domain‘.

We warmly invite you to attend this lecture, learn more about the use of register data and machine learning to study effects of policy interventions, and participate in the Q&A and discussion after the lecture.

Registration for this event has closed. If you want to attend, you can request the link by sending an email to communications@odissei-data.nl.

About this Lunch Lecture

Decentralization of governmental tasks is frequently applied in social care systems, but key questions regarding its merits and risks remain. Chief among them is the question of how the transfer of responsibility from central to local entities affects equity. Existing research often makes use of aggregate measures to study the effects of the decentralisation but those measures do not provide sufficient insights into how an intervention might have differently effected certain groups in the population. As a consequence, those studies do not provide policymakers with the necessary information to fully judge whether the effects of a policy change have been equitable across demographic groups. 

Machine learning methods allow for identification of demographic groups that are reacting differently to a policy intervention and can provide a more complete picture of effects of decentralisation in a population. To illustrate the potential of those methods, Mark Verhagen studies the decentralization of the social domain in The Netherlands, using rich administrative data of Statistics Netherlands (CBS). During his talk, Verhagen will describe the application of causal trees and causal forests methods to address the question of equity of policy interventions. In his study he found considerable heterogeneity in the policy’s effect on social assistance use, both across geographic regions and demographic groups.

About Mark Verhagen

Mark Verhagen is a PhD candidate at Nuffield College and the Leverhulme Centre for Demographic Science, University of Oxford. He has a background in Sociology (Oxford) and Econometrics (University of Amsterdam). In his research he applies methods from the domains of machine learning and pattern recognition to answer a wide range of social science questions.


The ODISSEI Lunch Lectures highlight methodological issues and innovations in Social Science.

Relevant links

Photo by AltumCode on Unsplash