Bias Perception in Public Opinion Among Political Elites. Comparing Mechanisms Across Five Countries.

In 2018, Broockman and Skovron published an article in the American Political Science Review that demonstrated how US politicians systematically overestimate the conservativeness of public opinion (Broockman David E. & Skovron Christopher (2018) Bias Perception of Public Opinion among Political Elites. American Political Science Review. 112 (3): 542-563). Using a list of statements on cultural liberal issues such as gay rights, abortion, gun control and immigration, they show that US politicians perceive US citizens to be significantly more to the right of the political spectrum that they really are. This conservative bias is true for all statement, and for all politicians, be they senior or junior, Republicans or Democrats (although the bias is stronger among Republicans). In our contribution, we build on these insights with the use of a unique data set from six political systems where we conducted surveys with elected legislators (N=898 with an average response rate of 47%, range: 15%-77%) and citizens (N=11,672). Our aim is twofold. First, we want to establish whether conservative bias in elites' perceptions is unique to the US context or can be found across countries and issues. To do so, we surveyed elected members of parliament (MPs) from Belgium, Switzerland, Canada, Germany and Netherlands to estimate public support for different policies from socio-economic issues such as covering a wealth tax, immigration or issues covering cultural cleavages such as euthanasia. By pairing these estimations with information on actual public opinion, we can investigate the bias in elites' perceptions. Our analyses show that the conservative bias extends beyond the US. Across countries and issues, on average 65% (min.: 0%, max.: 96%) of the politicians overestimate public support in the conservative direction. Building on these insights, our second aim and main contribution to the literature is to establish the causes of this substantial conservative bias. Focusing on our extensive data set of public opinion, we test whether we find evidence in line with the tentative explanation brought forward by Broockman and Skovoron. Namely, that the citizenry holding more conservative opinions is politically more active, and is thus more visible to politicians. We thus test alternative explanations stemming from the more general societal developments which we find also across the countries included in our study, for example the support of parties on the right of the political spectrum. We discuss how our findings translate to studies on policy congruence and what this means for the representational role of politicians.