The role of property in representation: The Case of German MPs

In recent years, numerous studies have consistently demonstrated that policy responsiveness of members of parliament (MPs) is biased: Affluent citizens are better represented than the poor and the middle class (e.g., Bartels 2016; Gilens and Page 2014; Elsässer, Hense, and Schäfer 2020). This diagnosis holds across countries and policy domains (Elkjær and Klitgaard 2021; also Bartels 2017). Interestingly, the biases in European welfare states are as pronounced as in the United States. In this light, Elkjær and Klitgaard (2021) conclude that hypothesized mechanisms, such as (the lack of) left-wing parties, economic inequality, (low) participation rates, and (lax) campaign finance rules do not seem to drive biases in policy representation significantly. The actual mechanisms underlying differential responsiveness remain a puzzle of democracy research. This paper contributes to closing this gap and adds to the studies on MP-constituent interactions as a potential driver of biases (e.g., Broockman 2013; Butler and Broockman 2011). Specifically, we examine MPs’ differential accommodation of policy demands depending on their own and their demanders’ property. For our empirical analysis, we rely on data from a survey experiment, which we conducted as part of a large-scale survey with German MPs on the national and state level in 2022. In the experiment, we expose MPs to a hypothetical situation in which citizens directly ask for support concerning a specific policy and manipulate the characteristics of these citizens. Our paper presents how the self-reported likelihood for reactions of MPs is shaped by the material status of demanders relative to inter alia their own material status, party affiliation, left-right self-placement, gender, or level of government. In doing so, we contribute to understanding a mechanism for inducing policy responsiveness biases and ultimately contribute to better understanding the functioning of representative democracy more generally.