"Material interests, identity, and linked fate in three countries" 2019. British Journal of Political Science. Forthcoming.

In this article, I examine the theoretical connections between identity and linked fate, extending the latter concept across three countries and four types of (potential) identity groups. This belief, that what happens to one's ethnic group, religious group, region, and class shapes one's own life chances, is an understudied middle ground between ideational and material drivers of political attitudes.

Using both experimental and observational analyses, I show that the strength of individuals' belief in linked fate and that belief's consequences vary in systematic and predictable ways. From the very material effect of labor market uncertainty to the highly ideational effect of regional identity, linked fate is a cognitive bridge between two very different kinds of social-psychological experiences that can and should be applied across a wide range of countries and groups.

"Identity, interregional redistribution, and decentralization." (with Francesc Amat)

This paper examines the link between Catalan identity and preferences for interregional redistribution and decentralization in Spain. We exploit two plausibly exogenous causes for identity, the presence of a third person in the household when the interview was conducted to identify and the place of birth of the respondent's mother. Both an instrumental variables approach and a causal mediation analysis (Imai et al 2011) are used to estimate (local) average causal effects. If we assume that there is no direct effect of the third person on redistributive preferences, we can confidently estimate a causal effect in the instrumental variables framework. On the other hand, if we assume that identity is exogenous, we can show that the impact of the third person and the mother's place of birth are both mediated by identity. As a result, these two methods, along with the sensitivity analyses available under each framework, can be usefully combined to examine the robustness of the assumptions necessary for causal inference. From a substantive point of view, we show that group identity has a negative causal effect on preferences for inter-regional redistribution (in a relatively rich region) and a positive effect on preferences increased decentralization.