Blurring Lines of Responsibility: How Institutional Context Affects Citizen Biases Regarding Policy Problems

Abstract

In conditions of blurred lines of institutional responsibility, individuals make political choices by primarily relying on group-serving biases rather than holding leaders accountable. Whereas literature suggests that leadership cues and partisan biases shape who citizens hold responsible for economic conditions, we still don’t know how murkier lines of responsibility could affect people’s perception of whom to hold responsible for objective conditions. We test the hypothesis that group-serving biases more strongly affect responsibility-assignment in multilevel governing contexts compared to responsibility-assignment in unitary governing contexts. We use original data from an experimental survey that we conducted in 2015 in Bosnia and Herzegovina (BiH), and in 2018 in Albania and Kosovo.

Keywords: lines of responsibility; European Union treatment; equal pay; EU enlargement; Albania, Bosnia and Herzegovina, and Kosovo

Introduction

Extant literature suggests that blurred lines of institutional responsibility might make more difficult for voters to hold governments accountable on objective conditions and policy decisions (de Vries et al., 2011; Hellwig and Samuels, 2008). Location of institutional agents in charge could be tougher in multilevel systems with dispersed authority compared to unitary systems with clearer responsibility assignments. Without clear lines of responsibility, citizens are more susceptible to cues, mental short-cuts and prior biases when deciding who to hold responsible. Existing research suggests that individuals make political choices by primarily relying on group-serving biases rather than finding such a responsibility with decision-makers (Brader et al., 2013; Druckman et al., 2013; Tilley and Hobolt, 2011).

Whether contexts with murkier lines of responsibility lead to more biased decision-making remains an untested assumption (Tilley and Hobolt, 2011, pp. 328). This speculation has major implications for multilevel governance, because the expansion of dispersed governance and international institutions around the world may lead to more biased political behavior regarding responsibility assignment. Indirect observations hint that responsibility assignment is more pronounced in political entities with multilevel governance than in unitary political entities (Page, 2018).

Hypothesis: Group-serving biases more strongly affect responsibility-assignment in multilevel governing contexts when compared to responsibility-assignment in unitary governing contexts.

Rationale for Case Selection

Existing studies on responsibility-assignment have focused on one-state cases with similar institutional contexts that blur lines of responsibility to varying degrees. An important reason why existing research is limited in cross-national comparisons is the challenge of comparing biases in cross-national contexts with varying institutional arrangements. We focus on three EU membership-aspiring countries: Albania, BiH, and Kosovo, with Albania being a EU candidate country since 2014, and BiH and Kosovo being in Association and Stabilization Agreements with the EU as well as potential candidate countries. These cases provide an opportunity to compare opinions in EU-membership aspiring countries with regard to one external institution (the EU)―with individuals divided on support for EU membership―and the governments and mainstream political process are supportive of EU membership. BiH is divided into two autonomous, governing regions/entities relevant for our argument: the Federation of Bosnia and Herzegovina (FBiH) and the Republika Srpska (RS), with low and high levels of institutional clarity, respectively. The FBiH is ruled by ten cantons with their own parliaments as well as an overall FBiH parliament, while the RS is ruled by one unified parliament. Both Albania and Kosovo are unitary states with only two decision levels, the central government and municipalities. With lower levels of institutional clarity, citizens rely more on biases when assigning responsibility for policy outcomes. Biases elicited by experimental frames should be stronger in the FBiH (with murkier institutional clarity) compared to Albania, Kosovo and RS.

EU enlargement yields contextual similarities of political camps (pro- and anti-EU membership), with governments and population majorities supporting EU membership (Hobolt and Spoon, 2012; Hobolt and Tilley, 2014). Hence, the process of joining the EU readily allows for cross-national comparisons. BiH and Kosovo carry the postwar international intervention with federated and unitary political solutions (EUFOR and the UNMIK in Bosnia and Kosovo, respectively), allowing for a comparison of similar contexts.

Unequal pay remains a concern since the initial EU integration efforts (Mac Rae, 2010; van der Vleuten, 2013). The EU conditions accession with equal pay for equal work policies (Avdeyeva, 2010; Koldinska, 2007). Women in Albania, BiH, and Kosovo face higher levels of pay inequality compared to most EU member countries: the average pay gap is 16 per cent in the EU, and in Albania and BiH is 18 percent and 20 percent, respectively (Miluka, 2013; Somun-Krupalija, 2011). In Kosovo, we calculated that the gender pay gap is 11 percent (Trusti, 2018). Such a low pay gap in Kosovo reflects high female employment in the public sector.

If policy outcomes are addressed as bad in terms of the EU membership standards, we expect those who support their state’s independence to blame circumstance and not their government, hence holding them less responsibility compared to those who support EU membership. In turn, EU partisans would assign more responsibility to their government for bad EU policies compared to those who favor independence. We expect these effects to be stronger in multilevel contexts (FBiH) in comparison to unitary contexts (Albania, RS, and Kosovo).

Research Design

We used data from an original survey experiment conducted in BiH in July 2015, Kosovo in October of 2017, and Albania in December 2017-January 2018 (Appendix A). We conducted a two-stage experiment. Firstly, the questionnaire asked respondents for opinion about equal pay, which is addressed (or not) as an EU standard before the respondents report their opinion on this issue. In line with EU membership requirements, we address rights as EU membership conditions. Secondly, the questionnaire randomly assigned the same respondents in or out of another treatment. Therefore, our sampling design produces four treatment groups. For the second treatment, the questionnaire informed respondents about experts’ claim that, compared to other EU member countries, Albania/Kosovo/BiH is worse regarding gender inequality in pay. This message was not provided in the control group. Afterward, the questionnaire asked about how much responsibility they assign to their government for pay inequality. At the beginning of the interview, the questionnaire asked all of the groups about their support for either EU membership or national independence (support for a hypothetical referendum on EU membership), thus dissecting the ‘EU supporters’ and ‘independence supporters’.

Results

Table 1 shows OLS regression models predicting the equal pay responsibility-assignment scores (on a 0–10 scale: No responsibility – Full responsibility for the government), with two independent dummy variables for the treatments (information about experts’ claims about bad policy outcomes regarding pay inequality, compared to the reference category of no policy outcome information, along with the EU frame treatment), a dummy variable for one’s EU membership preference and the interaction between the three. Respondents leaned towards assigning responsibility to the government: means of 6.4 in the FBiH, 6.7 in the RS, 6.9 in Kosovo, and 7.5 in Albania. Voting in favor of a hypothetical referendum on country’s EU membership was popular among the survey respondents: 89 percent in the FBiH, 57 percent in the RS, 91 percent in Kosovo, and 93 percent in Albania. There were no differences across the countries between supporters of membership and independence regarding level of responsibility assignment (at traditional levels of statistical significance).

Moreover, the treatment effects are larger (and with higher levels of statistical significance) in the FBiH compared to the RS, Albania, and Kosovo. These results suggest support for our hypothesis: the effects are stronger in the FBiH where biases should affect more opinion formation about institutional responsibility due to the entity’s complex multilevel governance. Appendix B shows graphs of the treatment effects. Independence supporters provide survey responses that exonerate their government from gender equal pay policy.

Conclusions

Arguably, biased responsibility-assignment would be more prominent in political systems where lines of responsibility are murkier compared to other systems (Tilley and Hobolt, 2011), but until now we lacked direct evidence about the effect of institutional context on these biases. Under the conditions of blurred lines of institutional responsibility, individuals can adjust their perceptions based on their EU biases related to whom to hold responsible about gender unequal pay. Compared to those who support EU membership, state independence supporters in the FBiH are less likely to hold their government responsible, and instead blame circumstances when they receive information concerning bad conditions. We found empirical support for the expectation that citizens adjust who they think is responsible for objective conditions when institutional context makes it difficult to discern the responsibility locus. The EU frame affected independence supporters who assigned less responsibility to their government for pay inequality.

We compared four political entities: the FBiH, RS, Albania, and Kosovo. We observed stronger effects in the FBiH, a context with more obscured authority lines. In this case, biases are more important for mitigating inconsistencies between policy preferences and objective conditions. Suggesting that expanding multilevel governance may result in more biased opinion formation, we need more research that would expand both country and policy outcome ranges.

References

1. Avdeyeva O (2010) States’ Compliance with International Requirements, Gender Equality in EU Enlargement Countries. Political Research Quarterly 63: 203–217.
2. Brader T, Tucker J and Duell D (2013) Which Parties Can Lead Opinion? Experimental Evidence on Partisan Cue Taking in Multiparty Democracies. Comparative Political Studies 46: 1485–1517.
3. de Vreese C, Banducci S, Semetko H and Boomgaarden H (2006) The News Coverage of the 2004 European Parliament Election Campaign in 25 Countries. European Union Politics 7: 477–504.
4. de Vries C, Edwards E and Tillman E (2011) Clarity of Responsibility Beyond the Pocketbook: How Political Institutions Condition EU Issue Voting. Comparative Political Studies 44: 339–363.
5. Druckman J, Peterson E and Slothuus R (2013) How Elite Partisan Polarization Affects Public Opinion Formation. American Political Science Review 107: 57–79.
6. Follesdal A and Hix S (2006) Why There is a Democratic Deficit in the EU: A Response to Majone and Moravcsik. Journal of Common Markets Studies 44: 533–562.
7. Hellwig T and Samuels D (2008) Electoral Accountability and the Variety of Democratic Regimes. Journal of Political Science 38: 65–90.
8. Hix S and Marsh M (2007) Punishment or Protest? Understanding European Parliament Elections. Journal of Politics 69: 495–510.
9. Hix S and Marsh M (2011) Second-order Effects plus Pan-European Political Swings of European Parliament Elections across Time. Electoral Studies 30: 4–15.
10. Hobolt S and Spoon J (2012) Motivating the European Voter: Parties, Issues and Campaigns in European Parliament Elections. European Journal of Political Research 51: 701–727.
11. Hobolt S Tilley J and Wittrock J (2013) Listening to the Government: How Information Shapes Responsibility Attributions. Political Behavior 35: 153–174.
12. Hobolt S and Tilley J (2014) Blaming Europe? Responsibility without Accountability in the European Union. New York, NY: Oxford University Press.
13. Koldinska K (2007) Gender Equality: Before and After the Enlargement of the EU: The Case of the Czech Republic. European Law Journal 13: 238–252.
14. Miluka J (2013) Gender Wage Gap in Albania. http://eca.unwomen.org/en/digital-library/publications/2013/11/gender-wage-gap-in-albania
15. Page, D (2018) How the Criteria for Joining the European Union Affect Public Opinion: The Case of Equal Pay between Women and Men in Bosnia and Herzegovina. Journal of Common Markets Studies 56(2): 230-246.
16. Schmitt H (2005) The European Parliament Elections of June 2004: Still Second-order? West European Politics 28(3): 650–679.
17. Somun-Krupalija L (2011) Gender and Employment in Bosnia and Herzegovina – A Country Study. Bureau for Gender Equality International Labour Office – Geneva. http://www.ilo.org/wcmsp5/groups/public/dgreports/gender/documents/publication/wcms_170832.pdf
18. Tilley J and Hobolt S (2011) Is the Government to Blame? An Experimental Test of How Partisanship Shapes Perceptions of Performance and Responsibility. Journal of Politics 73(2): 1–15.
19. Trusti (2018) Raporti vjetor 2018 [Annual Report 2018]. http://www.trusti.org/wp-content/uploads/2018/12/RV2018_shq.pdf
20. van der Vleuten A (2013) The Price of Gender Equality: Member States and Governance in the European Union. Burlington, VT: Ashgate Publishing.
21. Weber T (2011) Exit, Voice, and Cyclicality: A Micrologic of Midterm Effects in European Parliament Elections. American Journal of Political Science 55(4): 906–921.