Willing to Pay Project


Introduction

All modern welfare states face a set of very difficult challenges as they adapt to the demographic, economic and fiscal pressures of the early 21st century. These include: a) fiscal pressures of an aging 'core' population; b) political challenges of maintaining public support for adequate social welfare and education in the context of growing ethnic diversity; c) growing public frustration with and even distrust of bureaucratic state institutions and political authority; d) intense pressures to reduce (or at least not increase) taxes for politically powerful constituencies; and e) the continuing pressures to move from manufacturing-based economies towards service-based economies. These competing pressures deeply constrain the political choices available to policy makers in all advanced democratic nations. It is simply not true, however, that these forces push all democratic states in the same direction. Quite the contrary: the empirical evidence suggests that modern democracies are maintaining quite different policy trajectories - even in the face of broadly similar political, economic and fiscal pressures.1   This fact fascinates me and motivates my research agenda.

I believe that to understand the actual policy choices made in different countries, we must examine the interaction between political institutions, public policies, and citizen's preferences. In my past work I have largely focused on two sides of these complex relationships - public policies and political institutions. I have been cognizant of the fact that both what the state actually does, and how it does it, must affect citizens' attitudes and perceptions of what their state ought to do. The truth, however, is that political scientists have never really been able to test this assumption. Understanding how institutions shape and frame people's preferences and consequently their choices is pivotal to a full comprehension of societal regulatory mechanisms. Moreover, developing a better understanding of how and to what extent specific institutions shape and modify people's decisions may allow us to reform and adapt institutional system in a more effective and measured way.

My research will now focus more directly on precisely the ways in which the political and institutional context shapes or affects citizens' preferences. I believe that only when we better understand both what citizens in different polities actually believe about their state, and why, can we build realistic models to understand how their policy systems can be reformed or adapted in the context of the enormous pressures they face today. This research will thus combine the strengths of classical historical institutionalist analysis with recent developments in cognitive and evolutionary science and decision theory.

Methods: History, Institutions, and Experiments

In order to test the relationships between institutions, choices and preferences, I propose to conduct a series of experiments in different countries in which we examine the different trade-offs individuals in different societies make under different conditions. Specifically, we will focus on two sets of redistributive policy issues: Taxation and public pensions. Space does not permit a full elaboration of the precise questions to be asked in each experiment, but the chief purpose will be to build a series of scenarios that will allow us to test how different institutional contexts frame or shape citizens' decisions and thereby better understand how they perceive and process different policy choices and trade-offs.

To be sure, developing nuanced tests of these basic propositions in different national contexts will take time and a rather subtle understanding of the national context itself. In short, to effectively test these ideas we will need to understand the workings of the national political institutions as well as the structure of different tax systems in each of these different countries.

We will conduct several different types of experiments as we progress through the five years of this research project, and adjust the experimental designs as we learn from the interactions between the country specialists and the experimental results. For example, we might conduct experiments similar to Eek and Rothstein's in which they examined the relationships between bribery and social trust in Swedish and Romanian students by exposing the subjects to different scenarios and monitoring their responses as the scenarios were manipulated (Rothstein and Eek, 2009). We will also set up several of the more common 'market scenarios', which are well known to many political scientists. Finally, as the study develops we intend to build a set of so-called 'learning experiments' in which subjects are allowed to run the experiment several times and adapt their strategies (and preferences?) according to the behaviors they witness on the part of other players in the experiment. Each of these experimental methods has advantages and disadvantages. Some of these are preferable if one wants to minimize the impact of the participants' everyday life experience (Hoffman and Hearst 1990). Other experimental tools help to establish a baseline from which one can understand questions like attitudes towards other individuals, trust, perceptions of others, etc. Finally, others can provide insights into the perceptions of subjects within quasi-social situations and/or where communication is allowed and subjects can know how others behave as the experiment is iterated over time.

Throughout the study, historical institutionalist country specialists will work intensively with the experiments in each round, both so that we can refine the experiments in ways that can make them more realistic within different national contexts, but equally importantly so that we can build experiments that will test the specific hypotheses generated by the country specialists. It is my experience that different institutional and policy structures have different implications - or at least salience - in different national contexts. We want to know how these trade-offs are perceived and, perhaps, if they can be manipulated.2

The initial phase of this project we will bring together scientists who have had experience working with and building experimental models and social scientists who are expert in political institutions and tax policy regimes in our four nations (see list of key collaborators below). The general objective will be to build on each others' expertise. The strength of the comparative institutional scholar is that he or she has a much deeper understanding of not only the formal institutions in a given country, but also the informal norms and expectations that are likely present in that country. The strength of the experimental scientist is that he or she is better positioned to design specific experiments that will enable us to test for differences in norms, expectations, perceptions of fairness, attitudes toward redistribution, and willingness to pay. The foundational idea of this project is that by combining these strengths we will be able to build better models to test the arguments and assumptions made by historical institutionist country specialists, and thereby build better and more verifiable theories for explaining cross-national variation.

I will work with researchers and scholars from all four countries as well as the tax specialists and experimental scientists. We will 'kick off' the project with a large multi-day conference/workshop. The objective in this conference will be to learn from one another, share ideas, and lay out a more detailed plan for our research project as it progresses. I am well aware that given the complexity of multiple schedules of different scholars around the world that it would be complicated, to say the least, to arrange our schedule over a multiyear period.

In the first year of the project we will assemble our researchers and begin some initial pilot experiments in two of our four countries (Italy and the United States). There are a variety of types of experiments that have been developed by experimental scientists. In the initial phase of the project we expect to organise pilot experiments regarding public goods and trust. We want to establish what I will call a "baseline" for the different countries in an attempt to understand some basic variations in social norms with respect to willingness to pay taxes. I do not anticipate clear and easily interpretable findings at this stage of our project. It is my hope however that we will be able to learn from our initial experiments and through interactions between the experiment designers and the country experts build more realistic and understandable experimental models which can be then be retested in each of our countries. This will be an iterated process in which we build models, test them in the field, attempt to re-evaluate our findings so that we can then retest our various hypotheses.

We will reassemble my country specialist collaborators as well as the external scientists and tax experts in the first, third and final years of the project. These larger conferences will have slightly different functions as the research has developed and we have become more expert. Throughout the project, however, the principal investigator and the postdoctoral fellows will travel to each of the countries in which particular experiments are being held at the time of the experiments. I believe it will be very important to have a consistent treatment across the countries and across time.

I expect that this study will generate interesting and useful findings with respect to the intersection of institutional structures, tax regimes, and citizens' willingness to pay taxes and/or finance intergenerational redistribution that will be of use to both national scholars and comparative political economists. These findings will be published in both national and international journals throughout the course of the study. We will also write at least one edited volume in which we compare the national findings. I expect at least one major university press monograph to be written on the history, institutions and tax policy choices the countries in the study. Finally, I will write (or co-author) a monograph that attempts to build a bridge between our development of cognitive science and our models of institutional and welfare state change.

Conclusion

It is impossible to live in Italy for four years, as I have now done, and not understand that citizens' perceptions of their society and their government is profoundly negative (especially when compared to a country like Sweden). But it was not simply Mr. Berlusconi who caused these perceptions. Instead, it strikes me that Berlusconi was more an example, or symptom, of the basic perceptions that Italians have about their public institutions and leaders. But, while I think this is a reasonable assumption, no one has actually tested the underlying logic of this explanation. More importantly, no one has attempted to test and compare the underlying understandings of political choices in countries as diverse as Italy, Sweden Britain, and the United States. Once again, we think we know what is going on in citizens' minds when they consider policy choices like the willingness to pay taxes, or the choice about which kind of pension reforms they are willing to accept, but we do not know much about their real perceptions. My interest in the interactive relationships between policy choices, institutions and ideas lies at the core of the research I hope to engage over the next several years. Simply put, I am interested in exploring and explaining the multiple paths and different choices made in different democratic welfare states.


View the full ERC Grant Statement for this project.



1 For example, despite intense 'tax competition' in Europe, tax burdens have not fallen in recent years: Indeed, the average tax burden in the OECD grew from 33.5% in 1990 to 38.5% in 2007. Even in the heavily taxed EU-15 countries tax burdens have remained quite stable, growing from 38.2% in 1990 to 38.7% in 2007. Similarly, there is great pressure on health spending in the OECD, and while we have seen the introduction of some privatization in several countries there is no evidence to suggest that the various systems are 'converging' on a common policy.

2 An important distinction is "between-subject" and "within-subject"-designs. Between subject design applies one treatment to all subjects and compares them. Within-subject design applies many different treatments to each subject. We will conduct both between and within subject design in order to attempt to single out both individual differences and manipulate the data by holding more variables constant.