Public opinion on the Eurozone fiscal union: evidence from survey experiments in Italy

ABSTRACT

We investigate public attitudes toward the fiscal union: a policy advocated in official European Union documents and designed to address asynchronous economic fluctuations in the eurozone. We employ survey questions and conjoint analyses embedded in population-based panel surveys in Italy, and draw expectations from theories of tax-and-transfer schemes, public insurance, ideology, diffuse support, identity and trust. High-income right-wing individuals with weak European identity and negative assessment of EU membership are more likely to oppose the measure. However, high-income respondents display greater willingness to pay, especially in order to keep the euro, whereas lower-income participants are readier to ditch the currency if the monetary union does not deliver good economic performance. The political feasibility of this policy seems therefore to rest on the willingness to contribute by the core constituency supporting the euro. We also investigate the preferences for the institutional design of the policy.

KEYWORDS:

• Conjoint analysis
• economic policy
• eurozone fiscal union
• Italy
• political economy
• public opinion

Acknowledgments

The experiments have been administered by SWG on behalf of the Italian National Election Studies Association. We thank Samuel Brazys, Damian Bol, Manuele Citi, James Cross, Renato D’Amico, Johan A. Elkink, Vincenzo Galasso, Nikitas Konstandinidis, Bjørn Høyland, Valentino Larcinese, Vincenzo Memoli, Massimo Morelli, Sven-Oliver Proksch, Aidan Regan, Francesco Zucchini, participants at seminars at the Bocconi University, University College Dublin, McGill University, University of Catania and University of Oslo and at the 2015 DSE, ECPR, EPSA, ICPP conference panels.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Fabio Franchino is professor of political science at the Università degli Studi di Milano.

Paolo Segatti is professor of political sociology at the Università degli Studi di Milano.

ORCID

Fabio Franchino http://orcid.org/0000-0001-8774-6494

Paolo Segatti http://orcid.org/0000-0003-1505-2519

Notes

1 The model ignores benefit targeting but, for a given tax rate, individuals may support spending in areas where they benefit the most. For instance, the unemployed may prefer spending on unemployment benefits over alternative destinations.

2 But they are not more supportive of eurobonds. Stock owners instead display similar opinions (see also Bechtel et al. 2014).

3 Moene and Wallerstein (2001, 2003) show how higher earnings decrease demand for purely redistributive policies, increase demand for income-replacement policies and are unrelated to demand for universalistic policies (e.g., health). Their model accounts for both redistribution and insurance by adding an unemployment benefit to Meltzer–Richard formulation of individual utility.

4 The direction may be opposite in case of trustworthiness of own national institutions since higher trust correlates with support for slower integration (Sánchez-Cuenca 2000) and less policy integration (McLaren 2007).

5 ‘Grecia: Italia Terzo Creditore Con 40 Miliardi Di Prestiti’. La Repubblica. 25 January 2015. ‘Debito, Ecco Quanti Soldi Deve La Grecia All’Italia’. Il Giornale. 26 January 2015. ‘Quanto Costa La Grecia Al Contribuente Italiano’. Il Sole 24 Ore. 28 January 2015.

6 ‘Trust is the main obstacle to progress on Greece’. EUObserver, 12 July 2015.

7 The coefficients of high-income respondents are −0.18 (se = 0.055, p < .01) in model 1a and −0.19 (se = 0.056, p < .01) in model 1b if ‘inactive’ participants is the reference category.

8 They are 8.3 percentage points (se = 3.20) and 9.3 percentage points (se = 4.10) more likely to choose an economic programme proposing 3 per cent unemployment rate than, respectively, inactive and unemployed participants. They are 10.3 percentage points (se = 4.54) more likely to select an unemployment rate objective of 1 per cent than participants in higher-income occupations.

9 Low-income respondents are 5.9 percentage points (se = 2.76) and 7.5 percentage points (se = 3.83) more likely to prefer less oversight than inactive and high-income participants.

10 The conditional AMCEs for the ‘ditch euro’ attribute in the unemployed and low income subsamples differ significantly from the AMCEs in the inactive subsample (p > 0.002 and p > 0.009, respectively), and in the unemployed subsample from the high income subsample (p > 0.043).

11 The conditional AMCEs for this attribute in this subsample differ significantly from the AMCEs in the inactive (p > 0.038), unemployed (p > 0.012) and low-income (p > 0.001) subsamples. Attitudes toward additionality of taxation do not differ.

12 F-test p > 0.257, see Figure S9 in the SM. Ideology affects mostly attitudes about national taxation and spending.

13 They also penalize ditching the euro less severely than left-wing participants – a finding that resonates with the peculiarities of Italian politics (Conti and Memoli 2014; Conti and Verzichelli 2012).

14 Those who assess positively EU membership are 54 percentage points (se = 2.46) more likely to reject a programme ditching the euro than respondents evaluating membership negatively.

15 Respondents criticizing membership are, respectively, 5.8 and 6.8 percentage points (se = 2.28 and 2.50) more likely to oppose additional spending than positive and neutral subjects.

16 Positive evaluators are, respectively, 6 and 11.4 percentage points (se = 2.17 and 2.42) more likely to reject national governments as spending institutions than neutral and negative respondents. They are also 11.5 and 10.16 (se = 2.63 and 2.95) percentage points more likely to reject the national over the European Court of Auditors than neutral and negative respondents (10.6 and 9.6 percentage points (se = 2.57 and 2.94) when comparing the national court with the European Commission).

17 Participants of the economic policy conjoint experiment were not asked questions about trustworthiness.

Additional information

Funding

We kindly acknowledge the financial support from the Italian Ministry of Education, University and Research [Ministero dell'Istruzione, dell'Università e della Ricerca] (Project 2010943X4L_003, 2013–16) and from the Cariplo Foundation [Fondazione Cariplo] (Project CP3, Finanziamenti 2013).