T02P10 - Bias and Representation in Policy Making

Topic : Comparative Public Policy

Panel Chair : Anne Rasmussen - ar@ifs.ku.dk

Panel Second Chair : Dimiter Toshkov - d.d.toshkov@cdh.leidenuniv.nl

Objectives and Scientific Relevance of the panel

Call for papers

Session 1


Dimiter Toshkov - d.d.toshkov@cdh.leidenuniv.nl - Leiden University - Netherlands

Anne Rasmussen - ar@ifs.ku.dk - University of Copenhagen and Leiden University - Denmark

Civil society as a link between citizens and government. Or not?

Meta Novak - meta.novak@fdv.uni-lj.si - University of Ljubljana, Faculty of social sciences - Slovenia

Damjan Lajh - damjan.lajh@fdv.uni-lj.si - University of Ljubljana, Faculty of Social Sciences - Slovenia

Civil society is often understood as a link between the public and the government. It is another channel for the citizens to influence the public policy making. Since civil society is not elected, the principle of representation often does not apply to the civil society. Nevertheless, it is somehow expected that more vivid the civil society is and more interests it represent, more representative it is. If civil society organisations want to be successful and have an impact on public policy making, these organisations tend to professionalise. The activity of the organisations is thus led by the staff, most of decisions taken in the organisation are formed by organisational staff while members are becoming less and less included and the link between the organisation and its constituency is becoming weak. Organised civil society is thus structurally distant from supporters and members. Inner leadership of the organisation is often elitist which prevents members from taking part in organisational activities of organisation. If the thesis of civil society as a link between the public and the government wants to withstand it is important that interests of citizens (and not of organisational staff) are represented, otherwise we may face inequalities in representation.


The objective of this paper is to observe the role of members in the civil society organisations: to what extent are members included in activities of organisations, how important are they in decision-making within the organisation, as well as in influencing on public policies. We will limit our analysis to the case of Slovenia. Slovenia as one of the newer EU member-states and also one of the smallest with only 2 million inhabitants is often neglected in the systematic studies of impact of civil society on public policy making. Nevertheless, the case of Slovenia may be relevant also for other Central and East European post-socialist countries, Baltic countries, and countries from the Western Balkans. The statistical analysis will be performed on Web survey data gathered among the population of national civil society organisations (N=439). The survey took place between November 2015 and February 2016. The survey is part of international Web survey among national interest groups and civil society organisations. The results and analysis of the survey thus may be tested also for other EU countries.

Lobbying and Legislative Complexity in US States

Vannoni Matia - matia.vannoni@unibocconi.it - Bocconi - Italy

The term kludgeocracy refers to a situation where the status quo in public policy is difficult to change and hence, any change requires inevitably more and often unnecessary complexity. In this scenario complexity is valuable for those seeking rents from the government, such as powerful entrenched interests, because it is harder for the citizen to see who is benefiting from what, lowering the visibility of the issue at stake and because entrenched interests have an informational advantage over citizens. We empirically investigate whether and to what extent lobbying affects legislative complexity, using panel data on the 50 US states since WWII and relying on changes in campaign finance laws across the states in the sample as exogenous treatments. In so doing, we use a novel measure of legislative complexity over time for each state statute, based on recent developments in linguistics and tax legislation simplification. More specifically, we measure the readability of statutes, by looking at features such as word and sentence length, but also at syntactic (e.g. the use of different types of sentences, like subordinates) and lexical features (e.g. the frequency of common words with respect to specific jargon). Moreover, we use natural language processing to derive the set of legal phrases which refer to exemptions and loopholes.

Coalition Government and Policy Responsiveness in Western Europe

Dimiter Toshkov - d.d.toshkov@cdh.leidenuniv.nl - Leiden University - Netherlands

Anne Rasmussen - ar@ifs.ku.dk - University of Copenhagen and Leiden University - Denmark

Whereas recent research has expanded the study of policy responsiveness to a broad range of political systems, existing cross-national studies focus primarily on the impact of political institutions and  only rarely pay attention to the way responsiveness is embedded in the patterns of party government, and the role of coalitions in particular. We examine how coalition conflict and government positions influence policy making and moderate the relationship between public opinion and public policy. Our study is based on a new dataset that tracks policy-making activity with regard to 306 specific policy issues in three countries (Denmark, Germany, and the United Kingdom) over four years and combines information on public preferences with measures of government positions, coalition conflict, and media salience. We find a systematic but relatively weak in substantive terms positive impact of public support on the likelihood and speed of policy change. While coalition conflict has only a weak negative direct effect on policy change and no effect on responsiveness as such, we find that more right-wing governments are both less likely to enact policy changes and less responsive to public opinion.

The impact of China’s advocacy groups on the policy making and its determinants

Emina Popovic - emina.popovic@fu-berlin.de - Freie Universitaet Berlin - Germany


In non-democratic regimes, non-state actors are considered to have less impact on the policy-making than those in democratic political systems where interest groups have a wide variety of options to express their views and positions. There is, however very less empirical research done on advocacy groups’ influence in autocracies, which could confirm or contradict this theoretical expectation. This study fulfills some of the literature gaps by studying Chinese advocacy groups’ impact on policy-making through examining whether there is an inequality in interest representation and if so toward whose preferences. Except investigating whose policy positions are reflected in Chinese policy outcomes, the study informs us on what determines the influence that advocacy groups exert in the decision-making. More than 80 interviews are conducted with specific interest groups, citizen groups, and journalist that were involved in the policy debate, or used some of the public political activities to influence public opinion in relation to the certain policy issue. 


Process tracing method is used to examine advocacy groups’ policy preferences, their advocacy strategies, developments through the policy-making process and final policy outcomes for 17 environmental policy proposals. Policy position where compared to final policy outcomes as a way of assessing the influence of advocacy groups on the certain policy issues. In so doing, statements of the informants that took part in selected policy-events were scrutinized and complemented with publicly available sources and documents provided by advocacy groups and journalists. For data analysis, qualitative case studies are combined with Qualitative Comparative Analysis (QCA) following Ragin’s (2006) guidelines on integrating QCA and process tracing in a single analysis. Preliminary findings indicate that the success of NGO’s direct lobbying remains limited, yet it shows that through their successful use of voice strategies their positions, to the certain extent, get reflected in environmental policy outcomes. However, in the policy proposals where public opinion was mobilized, the analysis shows the bias towards specific interest groups. The findings are pointing out to Chinese government’s responsiveness to the views of the general public when they are expressed through street protest. In the policy proposals, where increasing environmental regulation was opposed by specific interest groups and was not supported through street protests, policy outcomes were closer to business groups’ preferences. The implication of these results might be that Chinese government is balancing between avoiding social unrest and protecting business interests in environmental affairs. Expect to reveal the relationship between interest groups, public opinion and policy outcomes in investigated policy areas, the findings offer some important implications for policy-making in China in general.




Delegated representation in the 21st Century: the experimentation of shared mandates.

Ricardo Cavalheiro - ricardo.cavalheiro@udesc.br - State University of Santa Catarina - Brazil

Leonardo Secchi - leonardo.secchi@udesc.br - State University of Santa Catarina (UDESC/ESAG) - Brazil

The objective of this essay is to analyze an alternative form of delegated representation that promises to soften the representative crisis in legislative bodies: shared mandates.

A shared mandate is a form of legislative representation in which constituents keep control of the mandate through direct participation. In shared mandates, the legislator voluntary sacrifices her voting autonomy to empower constituents, making the mandate more responsive and accountable. Shared mandates works under an agreement between a legislative representative and citizens for cooperatively exercise the legislative power of a mandate. Thus, a shared mandate is a form of delegated representation in which citizens determine the voting preference of their representative and her legislative activity. Shared mandate means that the legislative mandate does not belong to the political party or the incumbent representative but, in fact, to a group of citizens sharing the mandate.

The theoretical inspiration for shared mandates is the delegated form of representation and direct democracy applied to the legislative power. A shared mandate is a delegated model of representation (Pitkin, 1967), a type of promissory representation (Mansbridge, 2003), and an example of the ambassador and the pared-down delegate type of the Rehfeld´s distinction of representativeness (Rehfeld, 2009). Delegated representation is an alternative to the widely used trustee model in which representatives are not expected to correspond strictly to the constituents´ preferences, but they rather have an ethical obligation towards “the general interest” (Burke 1774; Pitkin 1967). As there is an information asymmetry between citizens and politicians, the political agent is tempted to defend other interests, which may be conflictive to those of constituents. In few words, the trustee model is prone to fail in its very basic element: trust. Facing this, shared mandate presents itself as a strategy to bridge this gap through binding consultation of constituencies that inform and determine the representative´s position in legislative matters.

In view of this, the following research question is presented: What formats of shared mandate are used in contemporary democracies as an alternative to confront the representative crisis? To answer this question the study uses a multiple case study discuss different formats of shared mandates in countries with evidence of its utilization: Sweden, Australia, Argentina, Brazil and United States. The variables to analyze the empirical date are the following: initiative, size, eligibility, access, permanence, distribution and extent of power, costs, benefits, and decision-making process. The study breaks down the variables in different analytical categories in order to analyze the empirical cases.

Given that the objective of this study has common objectives to those of the T02P10 panel, such as understanding “Which role do interest groups and political parties provide when it comes to achieving policy representation?”, it seems clear that the discussion about share mandates and delegated representation has the potential to bring light to the panel discussion.

Export PDF