# Chapter 6 Research Design

Between these two

At this intersection of (pct) taxation and (deliberative) democracy, two related research questions arise:

1. $itm:resolve-better-tax$ Can democratic rule resolve — and if so, how — any remaining disagreement about a priori desirable and doable taxation?

2. $itm:prove-deliberative-democracy$ Can deliberative democracy address — and if so, how — the vastly complex, macro-level abstractions facing our polities?

This dissertation investigates the intersecting set of these two research questions.

In one sense, deliberative democracy serves as the method to resolve a priori disagreement about taxation, and — equivalently — to test whether pluralist democracy may be partly to blame for the absence of better taxation. Deliberative democracy is a good method for research question $itm:resolve-better-tax$ on taxation, both because I axiomatize it to maximize democratic legitimacy, and I hypothesize it to reveal some of the dysfunctions of pluralist aggregation plaguing tax choice (see below).

Conversely, taxation is a good case to test deliberative democracy in research question $itm:prove-deliberative-democracy$, because it affects everyone, but is shrewd in the kind of macro-level abstractions which deliberative fora have so far avoided130 and which may strain the cognitive ability of deliberators.

1. Good deliberation is: Metaconsensus (or well-structured ordinal preferences)
2. Good deliberation is: Intersubjective Consensus (IC) or Intersubjective Rationality People who share the same beliefs and hold the same values express the same preferences and vice versa.

1. As a result of deliberation, metaconsensus in beliefs, values and preferences will be higher. Factors will be fewer, and loadings higher.
2. As a result of deliberation, intersubjective consensus will be higher. People who share the same beliefs about the economy, and hold the same allocative values will express the same tax preferences and vice versa.

This is open also to factors of fundamental disagreements about the role of rationality in democratic decisions (as compared to, say, power, practice).

–>

## 6.1 Open Second-Order Questions

These hypotheticals flow, for the most part, from a priori knowledge. They are grounded in reason, but not — as hypotheticals can never be — in experience. I have, as much as possible, supported a posteriori assumptions with robust empirical findings — but I cannot provide experiential evidence for the complete hypotheticals.

In welfare and tax, comprehensive a posteriori knowledge about hypotheticals is hard to come by.

Inframarginal reforms as those I suggest here strain the limits of economic models, all of which must be grounded in only marginally known behavioral data. For example, we would need to know the price elasticity in the demand for luxury goods, or saving and agree on (as I have shown) irreduceably normative judgments of diminishing returns or an optimal savings rate. Natural experiments are, of course, unavailable and the external validity of any laboratory setting will be strictly limited. There is, in other words, not much else beside a priori knowledge that we might learn about welfare and tax. My reasoning here is hopefully as widely agreeable as any comprehensive account of welfare and tax, but it will surely not go uncontested. Absent empirical clarification, we must then resolve such disagreement in a democratic process. Taxation, in other words, probably cannot be reduced to a first-order question, but it, too reverts to a second-order concern on how to democratically resolve any disagreement we might have about it.

In democracy, we have some, if limited a posteriori knowledge about the deliberative hypothetical I here suggest. Deliberative democracy has been successfully implemented mostly in local settings on policies of limited scope (Fung 2003; Hendrickson and Tucker 2005). In their proposal for “Empowered Deliberative Democracy” Fung and Wright (2001, 17) explicitly suggest what underlies much of the current designs:

1. A focus on specific, tangible problems

2. Involvement of ordinary people affected by these problems and officials close to them

3. The deliberative development of solutions to these problems.

Few attempts at deliberating regional issues of greater complexity have been made, including a ‘Citizens Assembly’ on electoral reform in British Columbia, Canada (confer Ratner 2008).

On the one hand, deliberative theorists as Fung and Wright (2001, 17) have criticized this limited scope of deliberative attempts, focused on the immediate, tangible and “confined to the realm of neighborhood and locale”, as much out of necessity, as out of conviction (Boggs 1997, 759). The associated anti-authoritarian, anti-bureaucratic, anti-functional-differentiation impetus may, as (de Sousa Santos 1998) hopes for participatory budgeting in Porto Allegre, Brazil contribute to a “counterhegemonic globalization” by concentrating on special issues like land rights, urban infrastructure and drinking water in urban settings, but may by the same token remain “Bean n’ Rice works” (de Sousa Santos 1998, 479). Faced, as our polities inevitably are, with macro-level abstractions, such cherishing of “human-scale democracy” above all else may, suspects Boggs (1997), move democracy “in a defensive and insular direction, laying bare a process of conservative retreat beneath a facile rhetoric of grassroots activism” (Boggs 1997, 759).

On the other hand, political psychologists as (Rosenberg 2002) have problematized the cognitive demands deliberation poses. Recent research in political psychology suggests, that — contrary to the bounded rationality assumption — imperfect human reasoning may not only stem from remediable cognitive scarcity, but may be developmentally determined. Rosenberg (2002) has suggested a threefold developmental sequence and typology of sequential, linear and systematic reasoning. His empirical accounts suggest that if any, only systematic thinkers will be able to meet the cognitive demands for reasoned arguments, and egalitarian free speech of deliberative democracy. Moreover, this cognitive competence was found not to be domain specific, and while people may regress to lower levels of competence under high loads or appropriate cues, they are unlikely to easily, if ever, achieve higher than developed levels. “Structurally (more and) less developed reasoning adults” make “not only the adequacy of citizens’ reasoning, but also their equality” a problem for deliberative democracy (Rosenberg 2007, 12, emphasis added).

Few empirical work has been done to address both these ontological and empirical reservations towards deliberative democracy. We know very little about if and how deliberative fora do in the face of macro-level abstractions and unequally limited cognitive ability of deliberators.

## 6.2 Hypotheses

Both research questions $itm:resolve-better-tax$ and $itm:prove-deliberative-democracy$ can be rolled up in one set of hypotheses:

## 6.3 Methods

I test these hypotheses by subjecting people to a , pioneered by James (Fishkin 2009). This ‘Gold Standard’ of deliberative fora (Mansbridge 2010, 55), combines skillfully moderated small and plenary group discussions of randomly sampled citizens, rigorously vetted, balanced briefings and experts with pre-treatment, post-treatment and control group opinion surveys into a quasi-experimental design (as in , p. ).

This design suits me well, because it combines the electoral part of democracy with the talking part (Chambers and Kymlicka 2002, 308) and thereby offers the kind of quantitative data needed to test these hypotheses. Both as a research method and as a second-order hypothetical, it also enjoys high external validity: it delivers a collective choice and has been shown to work in actual policy making.

## 6.4 Expected Results

### 6.4.1 Hypotheses $itm:knowledge-gain$.x are informed by prior research on the “Heuristics and Biases in Thinking about Tax” (McCaffery and Baron 2003), supplemented by systematic misunderstandings specific to broad choices between income, consumption and asset taxation.

My hypothesized misunderstandings flow from anecdotal experiences I have had with laypersons and political leaders. They also shine through in some of the welfare state research, and contrast with my synthesis of an (, p. ) and (, p. ).

### 6.4.3 Hypotheses

are related to public choice and opinion research in political science, showing that popular beliefs often violate vnm-consistence and can fall prey to aggregation dysfunctions. Opinions about taxation, in particular, may reveal anomalies at the individual and aggregate level, because tax is highly complex and demands highly structured choices. Past research has shown that deliberation can alleviate preference structuration anomalies (Farrar et al. 2003). A dp on tax choice might replicate those findings and extend them to another policy area.

### 6.4.4 Hypotheses $itm:interaction-effects$.x inform the political psychology, both of tax and of deliberation.

If the above effects do not, in fact, interact with equity beliefs (hypothesis $itm:interact-equity$, tax choice must be affected by something other than allocative preferences. In (p. ), equity preferences are moves along a curve, and shifts or moves of curves reflect preferences orthogonal to equity. If, as hypothesized, the above effects do not interact with equity preferences, deliberation may reveal (or bring about) popular understandings in line with the suggested institutional of (p. ): If given some thought, people agree that they can shift the trade-off curve between, say, equity and efficiency outwards, and they prefer those institutions that do. Hypothesis $itm:interact-ses$ is broadly related to the post-democracy thesis (Crouch 2004), by which late oecd liberal capitalism and democracy not only disenfranchise the social contract, but do so at a socio-economic gradient. The poor (and middle classes) may be too systematically confused to effectively act on their supposed self-interest, reinforcing a mutual crises of political and economic equality. Lastly, hypothesis is informed by the political psychology of deliberation, and the empirical research on unequal cognitive ability (Rosenberg 2002). If, as hypothesized, people with greater cognitive ability will display greater changes along the aforementioned hypotheses, deliberations on tax, too, may have to consider such unequal abilities in their theory and practice.

More broadly, a dp on tax choice can inform several fields, including research on public finance and the welfare state, public choice and political economy.

So, "who dunnit"? Because this dissertation is, at bottom, positivist, I cannot show what — let alone who — prevented a better tax, or a better democracy. Non-events such as as land or consumption taxation and deliberative democracy are always causally underdetermined, just as history is always overdetermined. For a detective, that is a bit of a let-down, but there is still work to be done. I might not find the perpetrator, but at least, I can absolve wrongly accused democracy, and, during the hearing, spread word of the crime.

#### 6.4.4.1 Specifying the PCT.

As is argued in the above, the PCT still requires a lot of grunt work detail. In particular, monetary dynamics of the PCT must be addressed, its formulae and schedule must be specified, an elegant design and administrative process must be drafted and a its implementation must be planned. Specification of the PCT will depend on cardinal quantification of its effects and require extensive economic modeling as well as econometric data.

#### 6.4.4.2 Informing the International Political Economy of Progressive Taxation.

If the PCT is indeed superior, cardinally and inductively, this verifiably desirable and doable hypothetical will put new questions to the literature on welfare retrenchment. Given that a better configuration is possible, political science must answer whether denationalization and tax competition are to blame for the present misery.

Two formal approaches apply here.

##### 6.4.4.2.1 Multi-Level Game Theory.

One is the multi-level game theory of international tax harmonization (Scharpf 1997a). Here, considerable effort should be devoted to understanding the strategic imperatives of governance both at the national and international level to adopt or avoid the PCT. An econometrically informed modeling of the distributive effects within and between countries will be required to gauge this multi-level game131.

##### 6.4.4.2.2 Veto Playing.

Closely related, international tax harmonization should be modeled as a veto-player problem (Tsebelis 2002). Given the anarchic nature of the international realm (taking on a realist view), veto playing appears to be a reasonable approximation of international tax harmonization. In particular, a modeled comparison between tax harmonization at the international and European level with their different veto rights may be instructive to estimate the overall significance of the model.

#### 6.4.4.3 Modeling the Introduction of the PCT.

Lastly, the chain reaction dynamic hypothesized in the above will need deductive, as well as inductive verification and econometric quantification: just how much costs will the PCT early adopter incur to the non-adopters? And how quickly will these costs materialize? How large would a critical mass of countries have to be to introduce and proliferate the PCT? Which countries would be suitable to ignite the chain reaction? –>

Such a difference $$\Delta$$ between “raw” and “enlightened” thinking, beliefs (Caplan 2007) and preferences (Fishkin 2009) is worth knowing for several reasons:

1. There may be agreeable, doable but hypothetical taxes, which may (Pareto)-improve over current taxes (Harberger 1974) or may be fairer (Rawls 1971) and more sustainable (Solow 1956), including a (Mill 1848; McCaffery 2002; Frank 2005b; Seidman 1997), a (???; Buiter 1988) and a (Friedman 1962). By comparison, existing taxes appear unnecessarily wasteful, inequitable and unsustainable.132

These suboptimal tax regimes may present democratic sovereigns with needlessly harsh tradeoffs between efficiency, equity, sustainability and other competing ends. The means of an intact mixed economy (Musgrave 1959; Stiglitz 2011) may consequently erode or even force retrenched welfare states into “permanent austerity” (Pierson 2001; Streeck and Mertens 2010) as it becomes harder for them to reconcile planned allocation and market exchange through taxation.

Public confusion about taxation may be a contributing factor, or even a sufficient — but not a necessary — condition for the non-existence of these supposedly superior taxes.

2. Progressivity in tax, especially, may hinge upon an enlightened, “new understanding of tax” (McCaffery 2005), as the historic mainstay of redistribution — the pit — withers away in the face of its inherent contradictions (McCaffery and Hines 2010). People may understand tax systematically different from how they would under (more) ideal speech, which may cause them to choose less effectively progressive taxes than they otherwise might, or otherwise inadvertently act even against their material self-interest. Additionally, the difference between raw and enlightened thinking about tax may, in itself, depend on the socio-economic status of citizens. Taken together, such socially structured and structuring differences in understanding tax may contribute to further, or perpetuate existing social inequality, even under nominal political equality.

3. The aggregative and representative institutions of liberal-pluralist democracy may be partly outmatched by the vexing complexity (Merton 1968) and tightly concentrated interest (Olson 1971) of late market economies, such as in taxation, violating both their input and output legitimacy (Scharpf 1997b) as well as basic democratic prescriptions for “enlightened understanding” (Dahl 1989). Empirical political sociology bears out such popular dissatisfaction, not with democracy it self, but with incumbents, enamored by special interest, and inattentive to the public good (???; Norris 2011; Putnam, Pharr, and Dalton 2000). If such a “post-democratic” constellation (Crouch 2004) were to plague the polity in its thinking about tax, it may well affect other policy areas too.

## 6.5 State of the field

This positive aspect of this thesis sits at the intersection of two fields of empirical research:

1. Practical experiments with different deliberative fora on varying policy areas,
2. Experimental and survey research about popular understandings of tax.

### 6.5.1 Empirical Deliberation

#### 6.5.1.1 Deliberative Experiments

Deliberative fora have been tried on a number of policy areas and places, including renewable energy in Texas (USA) (Lehr et al. 2003), participatory budgeting in Porto Allegre (Coelho, Pozzono, and Montoya 2005) (Brasil), nanotechnology (Powell and Lee Kleinman 2008), electoral reform in British Columbia (Canada), policing and education in Chicago (Illinois) (Warren and Pearse 2008), and city planning in Philadelphia (Pennsylvannia) (Sokoloff, Steinberg, and Pyser 2005). They also vary in their institutional design (reviewed in Fung 2003). These include several small-n, multi-day designs, such as Citizen Juries (Smith and Wales 2000) or Planning Cells (Dienel 1999), where a few randomly selected citizens hear expert witness and deliberate amongst themselves for a couple of days, all facilitated by trained, non-expert moderators and prepared by a balanced briefing book to draft a Citizen’s Report or similar consensus recommendation. The otherwise similar Consensus Conferences (Einsiedel and Eastlick 2000) are often (much) longer, feature extended learning periods and are typically held on highly technical issues.133

Deliberation has also been tried in large-n, shorter designs with a more quantitative methodology in dps (Fishkin and Farrar 2005), where hundreds of randomly selected citizens receive issue books, deliberate in small, moderated groups and plenary sessions, hear expert witnesses and — instead of reaching a consensus — cast a secret vote or fill out a questionnaire, after the typically daylong deliberation concludes. Often, these survey results are compared to answers provided by participants before the event, and/or to a control group of non-participants, allowing quasi-experimental within-subject or between-subject statistical inferences.134

There is quite limited experience with long, large-n designs, including — to my knowledge — only the (Warren and Pearse 2008) that spanned several months, and included several hundred participants who were offered stipends and accommodation, participated in extended learning sessions, attended public hearings and ultimately agreed to recommend a new electoral system (a stv) for the province (British Columbia Citizens’ Assembly on Electoral Reform 2004).

Across these different designs and purposes, deliberative fora have been shown to improve knowledge of the subject matter in question, to change — but not bifurcate — beliefs (Fishkin 2009), to yield more orderly and orthogonal preferences structures (Farrar et al. 2003) and to boost a sense of political efficacy or mutual trust even among disempowered citizens (Karpowitz, Raphael, and Hammond 2009).

These encouraging findings on the capacity of deliberation to strengthen democratic rule notwithstanding, much of deliberative practice may be criticized for often focusing on the immediate, tangible and remaining “confined to the realm of neighborhood and locale” (Fung and Wright 2001, 17). This includes not only the overtly “human-scale” democratic efforts (Boggs 1997, 759) of participatory budgeting (de Sousa Santos 1998), city planning (Sokoloff, Steinberg, and Pyser 2005) or public stewardship of forests (Cheng and Fiero 2005), but also the much-lauded dps, which have, even when they were about clearly large-scale issues such as the eu (Fishkin 2009) or renewable energy use (Lehr et al. 2003) mostly shied away from the very structured choices and remote abstractions which policy must engage. These dps have, for example, demonstrated that more people prefer eu enlargement and more people would be willing to pay a premium for renewables (Fishkin 2009, 157), but did not ask — and probably not deliberate on, either — whether the eu was an oca, and if not, what fiscal complements might have to be implemented (for example, Mundell 1961), or by which process competing renewable energies should be chosen, especially if they depend on economies of scale and network effects (for example, Krugman 1990). Clearly, these are only two haphazardly selected examples of abstractions (oca) and choices (infant industry regulation) relevant to these topics, but equally clearly, someone, somewhere in the democratic process has to make these calls.

Some of these more technocratic considerations may be rightly relegated to experts, but for others — including the abstractions of tax — this may not be possible or desirable as discussed in the above. In terms of Landwehr and Holzinger (2010), dps may not be sufficiently “coordinative” to be fully deliberative, and not only because they do not enforce a collectively binding decision (2010, 377). Absent abstractions, “coordinativeness” suffers because the policy options (for example, “faster eu expansion”) in which preferences are expressed (2010, 375) are ill-defined (for example, what kind of european integration?). If such expressed preferences are allowed to detach from ontologically given policy options (for example, currency union with our without fiscal complements), people can effectively “exit” from substantial agreement (Landwehr and Holzinger 2010, 377): “deliberators” can just talk, which may still be desirable, but stretches the concept too thin (Thompson 2008, 502).

To show its stripes — or limits (!) — deliberative democracy must be tried on such highly structured choices and abstract issues, too. So far, with the notable exception of some Danish (small-n) Consensus Conferences and the (large-n) ca (on its complexity, Blais, Carty, and Fournier 2008), such applications are lacking in empirical experiments with deliberation. If deliberative practice continues to blank out governing abstractions and structured choices — the very stuff of the modern world and its political rule — it risks moving “in a defensive and insular direction, laying bare a process of conservative retreat beneath a facile rhetoric of grassroots activism” (Boggs 1997, 759).135

One strand of positive research treats deliberation as the independent variable, investigating its effects (including Bächtiger (2005) on legislatures, Hibbings and Theiss-Morse (2002) on power differentials or Jackman and Sniderman (2006) on better-grounded judgements). In a similar vein, (Steiner et al. 2005) studied in how far deliberatively reached decisions “correspond to criteria of social justice” (Steiner 2012, 13), by which they seem to mean a utilitarian maxim (Bentham 1789) or the (???) difference principle (Steiner 2012, 95).136

In other work, deliberation is the dependent variable, investigating its causes (including Steiner et al. (2005) on institutional aspects or Steiner (2012, 13) on issue polarization). Similarly, (Landwehr and Holzinger 2010) compare how the discursiveness and coordinativeness of parliamentary debates and citizen conferences on regulating embryonic stem cell research impacts resultant preference change, their proxy for deliberation. Employing a more sophisticated and theory-grounded operationalization than most, Landwehr and Holzinger (2010, 377) analyze speech acts to code for arguing (for example, “to claim”) and bargaining (for example, “to offer”).

Lastly, researchers simply seek to operationalize deliberation. For instance, (Stromer-Galley 2007) suggests a coding scheme, including “reasoned opinion expression” when people provide “a definition, a reason for holding the opinion, an example, a story, a statistic, or fact, a hypothetical example, a solution to the problem, further explanation for why the problem is a problem, an analogy, $...$ or any further attempt to say what they mean or why they have taken the position that they have”.

These academic reflections all try to abstract away the substantive issues at stake in the respective deliberative encounters. Social scientists, here, as often, are loath to get caught in first-order conflicts over, say, school policy: “$...$ measuring the factuality or quality of reasoning was deemed unnecessary” (Stromer-Galley 2007, 10). This impulse is understandable, even commendable: Social scientists are, by definition, in the business of explaining the second-order social conflict over normative and positive first-order questions (compare Gutmann and Thompson 2004, 12) and must generalize from the particular to build or test social theory. Yet, since deliberation is a discourse ethic (Gutmann and Thompson 2002, 153), it is difficult to analyze without the actual, domain-specific discourse. The operationalizations employed vary widely, ranging from the crudely statistical (Steiner 2012)137 to the quick-and-dirty semantical (Steiner 2012)138 to promisingly speech-act linguistic (Landwehr and Holzinger 2010).

The dqi — even in Steenbergen et al. (2003)’s more exacting formulation — and other such operationalizations “can be applied easily and reliably to a wide range of deliberative contexts” (Steenbergen et al. 2003, 22), but that very decontextualization makes it hard to adjudicate whether really, universal validity was discussed, let alone reciprocally acclaimed. I share the uneasyness of “critical theorists and postmodernists” (Steiner 2012, 11) with this measure of quality of deliberation — though I do not think you have to be either to object, nor for the same reasons.139

Rather than how many justifications were given in total, or how often people referred to counterarguments (Steenbergen et al. 2003, 29), I want to know whether these reasons corresponded in substantive terms to those they disagree with. I provide a hypothetical exchange from deliberating taxation to illustrate my point:

Foundationalist140 I believe that everyone should be entitled to incomes generated from uncoerced exchange with others.

Social Democrat141 Yeah, obviously we are not going to take away all of a person’s income, they should be entitled to it, but maybe we’ll take some of it because we need that for education. That’s not going to hurt them.

Georgist142 I agree there may be something inviolable about owning the value added in free exchange with others. But, Foundationalist, do you think that should also extend to owning unimproved land and natural resources? It seems to me that no one has really created these values, and whoever reaps rents from them does so only because of the coercive powers that be.

Formally speaking and under dqi coding, both Social Democrat and Georgist refer to Foundationalist’s argument, yet their speech acts are of strikingly different quality. Social Democrat, rather sloppily, reiterates the argument and proceeds to balance it with another good (education) and suggests a compromise. We can see how the dialogue would proceed; Foundationalist — presumably deontologically bound to inviolable property — would refuse to trade off property against opportunity. Talking past one another, Social Democrat and Foundationalist may not have exchanged, let alone agreed on universal validity claims. By contrast, Georgist — maybe provisionally — embraces the foundationalist’s argument and points to an unexpected implication in foundationalist terms. It is much harder to see how their dialogue would proceed; presumably, Foundationalist would not — ex ante — have been open to large-scale land reform or taxation (as Georgism implies), and may not much warm to the idea. Still, she may find it necessary to either clarify her position further, or to concede Georgist’s point, because his may just be the “forcelessly forceful” better argument. Foundationalist and Georgist are grappling with the formulations and implications of what may become a universally acceptable moral claim, and in so doing, have crept a little closer to communicative rationality.

Crucially, this kind of Habermasian adjudication cannot be done in the abstract; researchers have to look for ideal speech in substantive terms as in this case, a foundationalist theory of property.

By abstracting away the actual arguments from a normative theory vested in their force, these researchers, as Steiner (2012) concedes, “take the position that large parts of the Habermasian version should be relaxed” (Steiner 2012, 150). For Steiner (2012), “the crucial point should be that participants in a political debate acknowledge that others also have reasonable arguments $...$ If this is accomplished, the other side is humanized” (2012, 150). This may well be true and important research, especially for divided societies on which Steiner (2012) here reports, but it stretches the concept of communicative action too thin.

To be sure, there may be such a loosely-defined thing as “deliberation” — distinct from “debate”and “discussion” (Landwehr and Holzinger 2010, 384) — and we should study its antecedents and consequences, as much of the above-cited research does. In this vein, deliberation raises positive questions: What “middle-range theories” associate antecedents and consequences (Mutz 2008)? To the extent that this deliberation is a normative project at all, it espouses a consequentialist outlook: deliberation is good, because it leads to “more public-spirited attitudes; more informed citizens; greater understanding of sources of, or rationales behind, public disagreements; a stronger sense of political efficacy; willingness to compromise; greater interest in political participation; and, for some theorists, a binding consensus decision” (Mutz 2008, 524).

Habermas prescription for communicative action, again, is different. Communicative action is not a positive statement about causes and effects open to falsification, because it poses a regulative principle (Kant 1781), which “is unachievable in its full state, but remains an ideal to which, all else equal, a practice should be judged as approaching more or less closely” (Mansbridge et al. 2010, 80). Communicative action is also not a consequentialist ethic, but deontological, because “it has the force to legitimate” (Habermas 2008, 147) — or even teleological, because “reaching understanding is the inherent telos of human speech” (Habermas 1984, 287). As such a normative political theory, communicative action cannot be operationalized into a set of treatments. Just because you subtract power, inequality, disrespect (Steenbergen et al. 2003), even “irrationality”, or any other combination of procedural criteria from communication does not necessarily make it ideal speech — even though the inverse may well be true, as these and other criteria may be necessary conditions for ideal speech. Legitimating collective choice through ideal speech is the irreducible, final reason of Habermas (1984)’ political theory whose achievement can only be found in substantive arguments.

Still, to thereby “simply define” deliberation as “intrinsically good” does not, as Mutz (2008, 524) suspects, render “empirical research irrelevant”. As Steiner (2012, 13) reminds us, the “level of deliberation” must be “measured and not merely assumed”. By definition, the normative impetus of communicative action cannot be falsified, but we may find that at any given time, any given people under any given circumstances, fail to meet Habermas’ exacting standards (Thompson 2008, 499). Such failures can serve “as indicators of contingent constraints that deserve serious inquiry and $...$ as detectors for the discovery of specific causes for existing lacks of legitimacy” (Habermas (2006, 420, emphasis in original), compare also Thompson (2008, 502)).

If we accept this more demanding definition of deliberation as the venue in which our approximating of ideal speech is ascertained, as I here wish to do, our research designs must be focused on argumentative substance. Others have done this, and have produced encouraging insights. (Ratner 2008) interviewed ca members after the fact, and found that most members considered Habermas’ validity claims honored — especially compared to the proper tragedy that was the following referendum campaign in which many ca alumni participated. While the analysis is based on interviews, not process data, it maintains the Habermasian standard and does not spare the reader qualitative data, including snippets recounting substantive disputes over electoral design. (Gutmann and Thompson 2004), too, reporting on deliberations over medical rationing, stick to their (quasi-Habermasian) standard, argumentative reciprocity, and offer examples of how it is met or violated in substantive arguments over health care.

Some may object that this approach collapses the procedural theory of democracy into a substantive theory of justice, and that it consequently requires researchers to take sides. This neither need, nor ought to be so. Habermas (1984)Communicative Action, much like Rawls (1971)’ contemporary Theory of Justice is neither purely procedural, nor substantive, but elegantly posits the discursive conditions under which substantive claims ought to be made. The contrast between procedural and substantive theory, here, as elsewhere, is misleading:

This kind of public philosophy would avoid the dichotomy that has come to dominate contemporary discussions of political theory, which poses a choice between basing politics on a comprehensive conception of the good, on the one hand, or limiting politics to a conception of procedural justice, on the other. We can and should avoid choosing either of these approaches exclusively. — Gutmann and Thompson (2004, 90)

Consequently, researchers need not take sides, nor follow (dubiously legitimate) expert rule (Blok 2007; Haas 1992). Emphatically, “alternative conceptions of the common good” (Cohen 1989, 23) are possible and desirable. Researchers must, however, familiarize themselves deeply with the issue to be deliberated, penetrate the discursive fluff and distortions to relentlessly unearth causal and moral disagreements which reasonable people may have over the issue. Researchers can then analyze their rich, qualitative, ideally process data in terms of these expected disagreements, to assess in how far the observed deliberation lives up to the standards of communicative action. Ideally, deliberators will struggle to formulate universal validity claims to express and resolve these very disagreements. In short, researchers need not have a position on the issue, but they must have a stake in the abstractions, disagreements and uncertainties plaguing the issue.

Communicative action on taxation, in particular, must be analyzed in substantive terms for these reasons. Tax is fraught with deep-seated causal and moral disagreement. To test whether people in any given deliberation on tax really have exchanged universal validity claims to — potentially — resolve these disagreements, their actual communication must be compared to some ideal standard of expected arguments, tentatively identified by researchers.

Pluralist and Deliberative Democracy
Pluralist Democracy Deliberative Democracy
Knowledge Miracle of Aggregation Wisdom of Crowds Schools as the Operative Metaphor
Legitimacy Special Interests (Alternative!) Common Goods
Equal Participation Representation Veil of Ignorance
Speech Power Communicative Action
Preferences Pre-Social Intersubjective
Hypothetical Examples of More or Less Valid Claims Concerning the Optimality of Taxation
Comprehensibility Truth Truthfulness Rightness
desirable-tax, tax-optimality “A deadweight loss occurs if otherwise pareto-improving exchanges do not occur”.“Taxes can sometimes prevent people from trading things on the market, which they otherwise would have exchanged to at least mutual benefit.” “Taxes are anti-growth”. “Under some circumstances, taxes can depress market activity without raising revenue”. “Lower taxes benefit everyone, because growth trickles down.” “If and to the extent that people react to incentives, and that we can really compare their utility, perfect markets make some people better off without hurting others, compared to how things were before”. No one would get any work done, if all their rewards get taxed away!"”It seems that markets do a good job in solving some problems, and we can, to that extent and in those domains accept that people will also be influenced by threats and rewards
doable-tax, tax-incidence “The tax burden is distributed according to the relative price elasticities of sellers and buyers in any given taxed transaction.” “The ultimate burden of a tax may differ from who nominally pays it.,Taxes are almost always levied on some trade, and their ultimate burden is distributed among the traders according to how sensitive they are to price changes.,The less you can react to a price change, the more of a tax you bear.” “With a Financial Transaction Tax (FTT), we can take back the money from those who caused the financial crisis.” “To the extent that a FTT discourages short-term trades, it does not raise any revenue.,To the extent that a FTT raises revenue, behavior does not change.,The burden of a FTT will be complexly determined by relative elasticities.” “Employers already pay half of social insurance”. “Social insurance is a payroll tax and falls entirely on labor, no matter the nominal distribution.,The burdens are shared between employers and employees according to their relative price elasticities.” “Big corporations should pay their fair share.” “Corporations are not moral subjects; they should not be liable for redistributive or general revenue taxes.,However, the rich people who hold a lot of stock in corporations should pay taxes on such income, or wealth.”
Hypothetical Examples of More or Less Valid Claims Concerning the Incidence of Taxation
Less Valid More Valid
Comprehensibility The tax burden is distributed according to the relative price elasticities of sellers and buyers in any given taxed transaction The ultimate burden of a tax may differ from who nominally pays it.,Taxes are almost always levied on some trade, and their ultimate burden is distributed among the traders according to how sensitive they are to price changes.,The less you can react to a price change, the more of a tax you bear.
Truth With a Financial Transaction Tax (FTT), we can take back the money from those who caused the financial crisis. To the extent that a FTT discourages short-term trades, it does not raise any revenue.,To the extent that a FTT raises revenue, behavior does not change.,The burden of a FTT will be complexly determined by relative elasticities.
Truthfulness Employers already pay half of social insurance Social insurance is a payroll tax and falls entirely on labor, no matter the nominal distribution.,The burdens are shared between employers and employees according to their relative price elasticities.
Rightness Big corporations should pay their fair share. Corporations are not moral subjects; they should not be liable for redistributive or general revenue taxes. However, the rich people who hold a lot of stock in corporations should pay taxes on such income, or wealth.

### 6.5.2 Misunderstanding Taxation

Survey and experimental researchers have investigated how people misunderstand taxation and its abstractions. (McCaffery and Baron 2003), tracking the heuristics and biases research program (Kahneman, Slovic, and Tversky 1982) suggest that voters fall short of (???) in their choice of tax: they incorrectly believe that taxation has a trivial, “flypaper” incidence (McCaffery and Baron 2004a), they favor (indirect) taxes not labeled or not visible, they fail to aggregate different taxes — especially the proportional or regressive payroll and vat — confuse progression in absolute and percentage terms and inconsistently prefer bonuses over penalties. Worried, (McCaffery and Baron 2004b) suspect that politicians may exploit these and other misunderstandings and that current tax systems may be suboptimal as a result of voter irrationality. They also suggest that people may overestimate effective progression, and that tax regimes may, as a result, be less redistributive than voters intend and believe them to be (McCaffery and Baron 2004b).

In a similar vein, (Sausgruber and Tyran 2011) show that people mistake the nominal incidence of a tax with its effective incidence on the buyers and sellers in a taxed market transaction.143

Under the resulting “tax-shifting bias” (Sausgruber and Tyran 2011), or, equivalently “flypaper theory of taxation” (McCaffery and Baron 2003), participant buyers often choose suboptimally high taxes on sellers, which they end up paying (partly) themselves. Sausgruber and Tyran (2011) furthermore find that participants are less likely to tax-shift if they are given accurate information on tax burdens exogenously and if they have experienced its income-reducing effects in the laboratory marketplace. Troublingly for deliberative democrats, they also find that (albeit minimally defined) initial deliberation does not reduce tax-shifting, but that it “makes initially held opinions more extreme rather than correct” (2011, 164), corroborating concerns for (deliberative) group polarization (Sunstein 1991).

These are rigorously researched and disconcerting findings that have not received nearly the attention they deserve from welfare state scholars and other political economists. These and other misunderstandings of tax may constitute partial, possibly sufficient — but not necessary — causes for changing tax regimes and constricted, or retrenched welfare states.

These are also — much like the sponsoring heuristics and biases program — no definitive, comprehensive list of misunderstandings, but an evolving, loosely connected set of relatively low-level cognitive effects. More research is needed:

• how do people (mis)understand some of the broader and more complex issues bearing on taxation, such as economic growth, savings rates or government and market failures?
• how — if at all — do the suggested low-level cognitive heuristics and biases filter up to the very structured choices of base, schedule and timing of taxation asked of the oecd polities?

As these pressing questions are addressed to develop a coherent account of social change from misunderstanding taxation, research designs as those by McCaffery and Baron (2004b) or Sausgruber and Tyran (2011) may well face two kinds of limits:

1. The complexity of higher-order considerations may overwhelm both designs.

Laboratory models in behavioral economics as those by Sausgruber and Tyran (2011) have the distinct advantage of directly testing the misunderstood causal effect in question (in this case, tax-lse), and rendering participants with an immediate experience of the effect (here, lower incomes). As the causal effects in question become more complex — for example, liquidity or employment decisions under different taxes — laboratory models face ever harsher tradeoffs between internal and external validity (for a review and dissenting opinion, see Jiménez-Buedo and Miller 2010). Trivially, economy-wide choices — for example, between a pit and pct — are very hard to model in the laboratory; if they were any easier, economic policy would be a non-issue.

The within-subject designs of cognitive psychology as those by McCaffery and Baron (2004b) also offer great internal validity when the heuristics and biases are low-level. Participants rate their agreement with, or the truth, of several statements they are presented with. Experimenters vary the context or wording of substantively identical statements to identify contradictions or mental shortcuts within the answers of individual participants. McCaffery and Baron (2004b) provide supposedly “de-biasing” information as within-subject treatments. The problem with higher-level concerns — say, whether to tax income or consumption — is that equivalent de-biasing treatments would eventually need to be an introductory economics course, covering at least the circular flow in the economy, some macroeconomics of aggregate demand and the Haig-Simons identity of income. As the required syllabus grows, it not only becomes impractical for participants with limited time and patience, but it also ignites a combinatorial explosion of necessary treatments to figure out just which (of supposedly many) insights from the de-biasing did the trick.

2. More problematically yet for the deliberative-minded researcher, both designs cannot problematize, let alone resolve, any remaining substantive disagreements over taxation and its economic abstractions. Both behavioral economics and cognitive psychology stipulate that there is a (vnm-)rationality, and merely chronicle whatever human deviations from it they find. Again as with the heuristics and biases research program in general, researchers and deciders must be aware of these deviations from rationality in tax, but these deviations do not add up to a theory, much less an account of social change in taxation or democracy. We know that von Neumann and Morgenstern (1944) say one (internally consistent) thing, and other homo sapiens mostly something else, but we learn little about the conditions of this divergence, let alone ways to reconcile it.

Things get even thornier for a heuristics and biases research program of tax, when higher-level concerns are introduced and — as they likely will — turn out to be controversial. McCaffery and Baron (2004b) or Sausgruber and Tyran (2011) posit expert rationality against their participants, and, whenever they differ, find their participants at fault. This works well enough for disentangling percentage and absolute tax burdens, and tolerably for demonstrating a tax-shifting bias144, but leaves the researcher empty-handed when faced with genuine disagreement between experts, or challenges as the legitimacy of expertise. To be sure, there may well be the vnm-rational, human-nature-compatible optimal taxation, that renders all other proposals regrettable misunderstandings — but this hypothesized agreement and its communicative conditions must be tested, not axiomatized. The heuristics and biases research program can inspire a sociological account of misunderstood taxation, but it cannot bring it to its conclusion. When faced with inevitably controversial higher-level issues such as the choice between a pit and a pct, both cognitive psychology and behavioral economics will smack of expertocracy and remain fruitless because they allow no recourse to a second-order theory (Gutmann and Thompson 2004, 125) — such as Habermasian deliberation — to resolve disagreement.

## 6.6 Research Questions

The deliberative experiments on taxation proposed here builds on these two strands of research as much as it probes its limits:

1. Taxation as a Case. A deliberative forum on taxation, including its structured choices and abstractions, adds another — yet rare — experiment on a technocratic and complex policy issue to the empirical record. From this perspective, deliberation is the research topic, and taxation is a case to test the limits and capacities of this prescription for political rule.145
2. Deliberation as Method. Lastly, deliberating taxation will provide data to better understand possible misunderstandings of tax and, thereby, clarify remaining controversies over it. From this perspective, optimal taxation and popular deviations therefrom are the research topic, and deliberation provides the method to investigate, or even resolve these differences in a normatively attractive manner (Rawls 1971; Habermas 1984).

These two research topics all fold into the aforementioned research question: how will different people understand taxation differently, if they participate in a deliberative process closer to the ideals of communicative action than the status quo of representative democracy?

## 6.7 Hypotheses

Both research questions, about 1) the remaining disagreements about tax and 2) the ability of deliberative democracy to cover abstract and structured issues can be rolled up into one set of hypotheses.

Fishkin (2009) suggests that deliberation increases knowledge, and changes preferences as a function of such knowledge gain. Similarly, I hypothesize that people will gain in knowledge about taxation, including resolving some systematic misunderstandings of concepts and causal relationships, some variants of which are already documented (McCaffery and Baron 2003; Caplan 2007). I furthermore hypothesize that as a function of these knowledge gains, people will change their preferences on the desired base and schedule of taxation. Additionally, deliberation will increase self-assessed autonomy and competence of participants, and they will find the presented alternatives of base and schedule more meaningful. Learning interventions will further drive knowledge gain and improve self-assessments, but — if it is sufficiently balanced — will not interact with attitude change.

1. $itm:think-different$ If ordinary citizens are given the possibility to deliberate welfare and taxation, they will think differently about it. Specifically,

1. $itm:knowledge-gain$ Knowledge Gain. People will gain in knowledge. Specifically relevant for tax choice,

1. $itm:bastard-keynesianism$ Bastard Keynesianism. People will understand that an economy can have an arbitrary (sub-@Solow1956) savings rate, and that — equivalently — if lowered slowly, aggregate consumption will not depress aggregate demand.

2. $itm:real-myopia$ Real Myopia. People will understand that the future prosperity of an economy is, in part, determined by present net investment, and that — equivalently — real, not nominal (for example, gdp) indicators track the savings rate of an economy.

3. $itm:bastard-hayekianism$ Bastard Hayekianism. People will understand that an economy can adopt an arbitrary government quota (if, possibly, at a cost), and that — equivalently — not all (but some) taxed resources are lost. Taxation is not exclusively or unconditionally a negative-sum proposition.

4. $itm:flyper-theory$ Flypaper Theory. People will understand that ontologically and empirically, only and always natural persons bear the burdens of taxation, and that — equivalently — the flypaper theory of tax incidence is false.

This hypothesis is similar to, but not identical to McCaffery and Baron (2003)’s findings of tax aversion, by which people mistakenly prefer taxes not labeled as such and/or indirect taxes and a disaggregation effect, by which people fail to integrate taxes on same bases.

5. $itm:ordoliberal-mess$ Ordoliberal Mess. People will understand that Pigouvian and general-revenue taxes follow opposing logics, and that — equivalently — a Pigouvian tax should not raise revenue.

This hypothesis is similar to, but not identical to McCaffery and Baron (2003)’s findings of a disaggregation effect, by which people fail to integrate different taxes even on same bases.

6. $itm:false-omnipotence$ False Omnipotence. People will understand that taxation will often (if not always) cause unintended consequences in markets, or that — equivalently — taxation can cause a negative-sum dwls. Taxation is not exclusively or unconditionally a zero-sum proposition.

7. $itm:false-omniscience$ False Omniscience. Relatedly, people will understand that taxable income and wealth (but not consumption) cannot be measured independent of market-pricing, or that — equivalently — any taxation based on fiat evaluations may have unintended consequences in markets.

2. $itm:universal-validity$ Universal Validity People will find the above abstractions meaningful and accept a universal validity of underlying causal and moral arguments.

3. $itm:attitude-change$ Attitude Change. Partly as a function of the above effects people will prefer a PCT, conditionally supplemented by a WT, LVT and NIT over the present tax regimes of PIT, VAT and Payroll.

2. $itm:meta-assessment$ Meta Assessment

1. $itm:autonomy$ Autonomy. People will consider themselves more effectively autonomous in discussing and choosing basic tax regimes.

2. $itm:competence$ Competence. People will consider themselves more competent in discussing and choosing basic tax regimes.

3. $itm:meaningful-choice$ Meaningful Choice. People will find the choices of tax base and schedule to be more meaningful to express their political autonomy.

3. $itm:learning-interventions$ Learning Interventions. A learning interaction targeted at one of the above-listed knowledge gains will further that knowledge gain.

4. $itm:interaction-effects$ Interaction Effects. The above effects will:

1. $itm:interact-equity$ Not interact with people’s equity beliefs, or — equivalently — people will not change their thinking about tax as a function of their allocative preferences.

2. $itm:interact-ses$ Interact negatively with people’s socio-economic status, or — equivalently — people with greater socio-economic status will change their thinking about tax the least.

3. $itm:interventions$ Intervention-interaction

1. Of the above effects, meta assessments will interact positively with learning interventions.

2. Of the above effects, attitude changes will not interact with learning interventions over and above the effect mediated by knowledge gain.

Variable Metaconsensus in Taxation (x3)
Values Beliefs Preferences
Subject What makes a tax desirable? What makes a tax doable? Which tax do we want?
Existing Work axiology, desirable-tax ontology, doable-tax better-tax
Example Statements “People should pay taxes on what they earn on their wealth” “A tax can be easily collected right at the source, in the firm” “We should tax the income of corporations.”
Example Extracted Factors Efficiency vs. Equity Withholding Taxation Corporate Income Taxation
Meta-Consensus High(er) loadings on relatively few(er) factors. People understand that people can disagree on efficiency vs. equity. High(er) loadings on relatively few(er) factors. People understand that one of the choices would be to tax at the source. High(er) loadings on relatively few(er) factors. People understand different taxes, and recognize it in different formulations.
Example Extracted Secondary Factor Indirect Taxation (all columns)
Intersubjective Rationality People who value wealth taxation, and think withholding taxes are easier to collect will prefer a Corporate Income Tax. (all columns)
Delta Before and After Deliberation Tax Misunderstanding: Flypaper Theory (all columns)
Variable Metaconsensus in Taxation (x5)
Values (specific) Values (General) Beliefs (General) Beliefs (Specific) Preferences
Subject What is desirable? What makes a tax desirable? What is doable? What makes a tax doable? Which tax do we want?
Existing Work axiology desirable-tax ontology doable-tax better-tax
Example Statements “Growth is good” “A good tax should encourage people to save for the future.” “Markets are a reasonable way to organize much of human activity.” “It is more feasible to tax things whose value is known.” “People should be taxed on what they spend.”
Example Extracted Factors Consequentialist Ethic Time-Discounted Stable Incomes Pro-Market Liquidity Taxation Progressive Consumption Tax

## 6.8 Public Finance and Welfare State Research

If the above hypotheses (especially $itm:attitude-change$) are confirmed, welfare research has a lot more to explain. If under a legitimate democratic process, people resolve remaining a priori disagreements to select a different tax than we currently observe, welfare state research must provide a second-order theory of the observed (illegitimate) social choice of suboptimal taxation. I cannot provide, let alone test that second-order theory of tax choice in this dissertation, but — if hypotheses are confirmed — can at least suggest that the democratic forum may be important. Pluralism may not be the only reason why we have no better tax, but it might well be one of several reasons.

## 6.9 Public Choice and Political Economy

Public choice has long specified how individual preferences can be aggregated, what it requires and how it can fail (for an overview, Mueller 2003). Conversely, (classical?) political economy has formulated theories of how (material?) interest and power can corrupt and alter preference aggregation (for an overview, Robbins 1976). Both those fields, too, can benefit from deliberating hypothetical taxes.

Public choice, for once, is mostly an a priori and often an insular enterprise, merely chronicling — but not explaining — the inconsistent preferences and structuration anomalies it faces. For example, portfolio theory clearly differs from observed loss (not risk!) averse strategies (Kahneman 2011), but we know relatively little about how culture or institutions mediate this gap. Deliberation is one (possibly attractive!) institution to narrow the gap between a priori public choice and a posteriori policy preferences. The hypotheses tested in this dp on taxation problematize the institutional link between optimal and actual choice.

Public choice is also hard to bring to bear on real-life political choices, because it deals in such high abstractions and the very consistency in preferences it demands are often subject to (second order) controversy. It can be difficult to argue how any particular policy may be explained by, or differ from a theory of public choice, as long as there is disagreement over possible alternatives and ontological concepts. This dissertation synthesizes and develops consistent preferences on taxation, and puts them to a deliberative test. If successful, it may provide an insightful, practically relevant case study of policy and help resolve the controversy over preferences.

Conversely, (classical) political economy can gain from a dp on taxation, because it, too, facing disagreement over alternatives and material ontologies (Keynes (1936) vs Hayek (1931), as in Wapshott (2011)), too often reverts to second-order theory, sometimes getting lost in an unproductive muddle of ideas and interests, rather than one or the other as independent variable.146 This dissertation develops a — hopefully — agreeable material ontology for the mixed economy (), and synthesizes a desirable and doable alternative which is then subjected to a specific, legitimate second-order process: deliberation. By comparing cognition — or ideas (!) — as a function of this second-order process, and by relating it to socio-economic outcomes and conditions of deliberators, this dissertation also tests one specific theory of the political economy. Based on transparent, indeed quite orthodox assumptions, I hope to show that material interest may, through whichever agency or structure, have found a clever way to corrupt democratic rule without resorting to crude power. If confirmed, my hypothesis suggest that — for whatever reason — people systematically misunderstand tax in a way that favors the rich, and more generally, the benefactors of the status quo.

## 6.10 Conclusion

In Schumpeter’s thundering words, taxation and democracy are the two sides of social contract (, p. ). They, above much else, balance contradicting social integration and individuality in modernity and moderate its key antagonistic institutions: states and markets.

Surely, such a broadly defined research agenda risks hubris, but it also promises to formulate and maybe, tentatively answer some important questions for the social sciences. I may well — figuratively — be biting off more than I can chew, and it sometimes shows (for example in ). All I can do is to tentatively mull over a mouthful at a time, and spit out again that which I cannot address. This — forgive the metaphor — rumination takes a lot of pages to write and read, but at least, it is a relatively transparent way of reasoning. Maybe, in the spirit of Umberto Eco’s anti-library (as recounted in Taleb 2007), explicating what I do not know or cannot address in this dissertation will make for better questions.147 Still, to digest anything in full, I must concentrate. Counterintuitively, the often economic or even technical abstractions of taxation and public choice provide a great guide here. This mostly a priori knowledge, too, requires a lot of pages to write and to read, but, akin to enzymes in the digestive tract, it helps me break down complex issues to ever simpler, fewer and clearer questions (as in and ) that can be handily metabolized into a new, and hopefully original argument that I put to the test.

Admittedly, too, underlying this research are normative convictions about political and economic opportunity, and even human destiny. Somehow, such a prescriptive impetus has become unacceptable in much of (German) social science on the welfare state. I must indeed avoid “politics-based evidence making” (The Economist, 2012) or any other conflation of what should be, with what is. However, even from a merely positive perspective, and even to readers who may not agree with these convictions, the hypotheticals they raise pose analytically worthy questions: If there is indeed another, superior fiscal configuration on which we could democratically agree — but do not — this absence needs explaining.

If anything, finding that which needs a social explanation, must be a worthwhile attempt for the social sciences.

It, too, is how social sciences can contribute to human progress, by politicizing the status quo.

“Politicization is the realization that established social norms, social practices, and social relations are contingent rather than sacrosanct, that things could also be different and that citizens, individual and collectively, have political agency by means of which alternatives can be explored and implemented. This recognition that things could also be different has always been the igniting spark of the emancipatory-progressive movements, and politicization has always been their key strategy.”
Blühdorn (2007, 313)

### References

Fung, Archon. 2003. “Survey Article - Recipes for Public Spheres - Eight Institutional Design Choices and Their Consequences.” The Journal of Political Philosophy 11: 338–67.

Hendrickson, David C, and Robert W Tucker. 2005. “Revisions in Need of Revising - What Went Wrong in the Iraq War.”

Fung, Archon, and Erik Olin Wright. 2001. “Deepening Democracy - Innovations in Empowered Participatory Governance.” Politics Society 29: 5–41.

Ratner, Robert S. 2008. “Communicative Rationality in the Citizen’s Assembly and Referendum Processes.” In Designing Deliberative Democracy – the British Columbia Assembly, edited by Mark E Warren and Hillary Pearse, 145–65. Cambridge University Press.

Boggs, Carl. 1997. “The Great Retreat - Decline of the Public Sphere in Late Twentieth-Century America.” Theory and Society 26: 741–80.

de Sousa Santos, Boaventura. 1998. “Participatory Budgeting in Porto Alegre - Toward a Redistributive Democracy.” Politics Society 26: 461–510.

Rosenberg, Shawn W. 2002. The Not so Common Sense : Differences in How People Judge Social and Political Life. New Haven, CT: Yale University Press.

Rosenberg, Shawn W. 2007. “Rethinking Democratic Deliberation - the Limits and Potential of Citizen Participation.” Polity 39: 335–60.

Fishkin, James S. 2009. When the People Speak. Oxford, UK: Oxford University Press.

Mansbridge, Jane. 2010. “Deliberative Polling as the Gold Standard.” The Good Society 19 (1): 55–62.

Chambers, Simone, and Will Kymlicka. 2002. Alternative Conceptions of Civil Society. The Ethikon Series in Comparative Ethics. Princeton, N.J. [u.a.]: Princeton Univ. Press.

McCaffery, Edward J., and Jonathan Baron. 2003. “Heuristics and Biases in Thinking About Tax.” Los Angeles, CA.

Farrar, Cynthia, James S Fishkin, Donald P. Green, Christian List, Robert C Luskin, and Elizabeth Levy Paluck. 2003. “Experimenting with Deliberative Democracy - Effects on Policy Preferences and Social Choice.” In ECPR Conference, Marburg, Germany. Marburg, Germany.

Crouch, Colin. 2004. Post-Democracy. Cambridge, UK: Polity Press.

Scharpf, Fritz W. 1997a. Games Real Actors Play: Actor-Centered Institutionalism in Policy Research. London UK: Westview.

Tsebelis, George. 2002. Veto Players - How Political Institutions Work. Princeton, NJ: Princeton University Press.

Caplan, Bryan. 2007. The Myth of the Rational Voter - Why Democracies Choose Bad Policies. Princeton, N.J.: Princeton University Press.

Harberger, Arnold. 1974. Taxation and Welfare. New York, NY: Little, Brown and Company.

Rawls, John. 1971. A Theory of Justice. Boston, MA: Harvard University Press.

Solow, Robert M. 1956. “A Contribution to the Theory of Economic Growth.” Quarterly Journal of Economics 70 (1): 65–94.

Mill, John Stuart. 1848. Principles of Political Economy. London, UK: Unknown Publisher.

McCaffery, Edward J. 2002. Fair Not Flat - How to Make the Tax System Better and Simpler. Chicago, IL: University of Chicago Press.

Frank, Robert H. 2005b. “Progressive Consumption Taxation as a Remedy for the U.S. Savings Shortfall.” The Economists’ Voice 2 (3).

Seidman, Lawrence S. 1997. The USA Tax - A Progressive Consumption Tax. Boston, MA: MIT Press.

Buiter, Willem H. 1988. “Debt Neutrality, Professor Vickrey and Henry George’s.” Cambridge, MA.

Friedman, Milton. 1962. Capitalism and Freedom. Chicago, IL: University of Chicago Press.

Musgrave, Richard A. 1959. The Theory of Public Finance: A Study in Public Economy. Intern. St. New York, NY: McGraw-Hill.

Stiglitz, Joseph E. 2011. The Price of Civilization. New York, NY: Random House.

Pierson, Paul. 2001. “Coping with Permanent Austerity - Welfare State Restructuring in Affluent Democracies.” In The New Politics of the Welfare State, edited by Paul Pierson, 43:369. Oxford, UK: Oxford University Press.

Streeck, Wolfgang, and Daniel Mertens. 2010. “Politik Im Defizit - Austerität Als Fiskalpolitisches Regime.” Cologne, Germany.

McCaffery, Edward J. 2005. “A New Understanding of Tax.” Michigan Law Review 103 (5): 807–938.

McCaffery, Edward J., and James R. Jr. Hines. 2010. “The Last Best Hope for Progressivity in Tax.” Los Angeles, CA.

Merton, Robert K. 1968. “Social Theory and Social Structure.” In Social Theory and Social Structure. New York, NY: Free Press.

Olson, Mancur. 1971. The Logic of Collective Action; Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press.

Scharpf, Fritz W. 1997b. “Economic Integration, Democracy and the Welfare State.” Journal of European Public Policy 4 (1): 18–36. https://doi.org/10.1080/135017697344217.

Dahl, Robert Alan. 1989. Democracy and Its Critics. New Haven, CT: Yale University Press.

Norris, Pippa. 2011. Democratic Deficit – Critical Citizens Revisited. Edited by Pippa Norris. Oxford, UK: Oxford University Press.

Putnam, Robert D, Susan J Pharr, and Russell J Dalton. 2000. “Introduction: What’s Troubling the Trilateral Democracies?” In Disaffected Democracies – What’s Troubling the Trilateral Countries?, edited by Susan J Pharr and Robert D Putnam. Princeton, NJ: Princeton University Press.

Lehr, R L, W Guild, D L Thomas, and B G Swezey. 2003. “Listening to Customers - How Deliberative Polling Helped Build 1,000 MW of New Renewable Energy Projects in Texas.”

Coelho, Vera Schattan P, Barbara Pozzono, and Mariana Cifuentes Montoya. 2005. “Participation and Public Policies in Brazil.” In The Deliberative Democracy Handbook – Strategies for Effective Civic Engagement in the Twenty-First Century, edited by John Gastil and Peter Levine. New York, NY: Wiley.

Powell, M., and D. Lee Kleinman. 2008. “Building Citizen Capacities for Participation in Nanotechnology Decision-Making: The Democratic Virtues of the Consensus Conference Model.” Public Understanding of Science 17 (3): 329–48. https://doi.org/10.1177/0963662506068000.

Warren, Mark E., and Hilary Pearse. 2008. “Introduction - Democratic Renewal and Deliberative Democracy.” In Designing Deliberative Democracy – the British Columbia Assembly, edited by Mark E. Warren and Hilary Pearse. Oxford, UK: Oxford University Press.

Sokoloff, Harris, Harris M. Steinberg, and Steven N. Pyser. 2005. “Deliberative City Planning on the Philadelphia Waterfront.” In The Deliberative Democracy Handbook – Strategies for Effective Civic Engagement in the Twenty-First Century, edited by Peter Levine and John Gastil. New York, NY: Wiley.

Smith, Graham, and Corinne Wales. 2000. “Citizens’ Juries and Deliberative Democracy.” Political Studies 48: 51–65.

Dienel, Peter C. 1999. “Planning Cells and Citizens’ Juries - Foundation of Political Engineering of the Future.” Wuppertal, Germany: Bergische Universität Wuppertal.

Einsiedel, E. F., and D. L. Eastlick. 2000. “Consensus Conferences as Deliberative Democracy: A Communications Perspective.” Science Communication 21 (4): 323–43. https://doi.org/10.1177/1075547000021004001.

Fishkin, James S, and Cynthia Farrar. 2005. “Deliberative Polling - from Experiment to Community Resource.” In The Deliberative Democracy Handbook – Strategies for Effective Civic Engagement in the Twenty-First Century, edited by John Gastil and Peter Levine. New York, NY: Wiley.

British Columbia Citizens’ Assembly on Electoral Reform. 2004. “Making Every Vote Count - the Case for Electoral Reform in British Columbia.”

Karpowitz, C. F., C. Raphael, and a. S. Hammond. 2009. “Deliberative Democracy and Inequality: Two Cheers for Enclave Deliberation Among the Disempowered.” Politics Society 37 (4): 576–615. https://doi.org/10.1177/0032329209349226.

Cheng, Anthony S., and Janet D. Fiero. 2005. “Collaborative Learning and the Public’s Stewardship of Its Forests.” In The Deliberative Democracy Handbook – Strategies for Effective Civic Engagement in the Twenty-First Century, edited by John Gastil and Peter Levine. New York, NY: Wiley.

Mundell, Robert A. 1961. “A Theory of Optimum Currency Areas.” The American Economic Review 51 (4): 657–65.

Krugman, Paul R. 1990. Rethinking International Trade. Cambridge, MA: MIT Press.

Landwehr, Claudia, and Katharina Holzinger. 2010. “Institutional Determinants of Deliberative Interaction.” European Political Science Review 2 (03): 373–400. https://doi.org/10.1017/S1755773910000226.

Thompson, Dennis F. 2008. “Deliberative Democratic Theory and Empirical Political Science.” Annual Review of Political Science 11 (1): 497–520. https://doi.org/10.1146/annurev.polisci.11.081306.070555.

Blais, André, R. Kenneth Carty, and Patrick Fournier. 2008. “Do Citizens’ Assemblies Make Reasoned Choices.” In Designing Deliberative Democracy – the British Columbia Assembly, edited by Mark E. Warren and Hillary Pearse. Oxford, UK: Oxford University Press.

Bächtiger, André. 2005. The Real World of Deliberation – A Comparative Study of Its Favorable Conditions in Legislatures. Bern, Switzerland: Paul Haupt.

Hibbings, John R., and Elizabeth Theiss-Morse. 2002. Stealth Democracy – Americans’ Beliefs About How Government Should Work. Cambridge, UK: Cambridge University Press.

Jackman, Simon, and Paul M. Sniderman. 2006. “The Limits of Deliberative Discussion – A Model of Everyday Political Arguments.” The Journal of Politics 68 (02): 272–83. https://doi.org/10.1111/j.1468-2508.2006.00405.x.

Steiner, Jürg, André Bächtiger, Markus Spörndli, and Marco R Steenbergen. 2005. Deliberative Politics in Action – Analysing Parliamentary Discourse. Cambridge, UK: Cambridge University Press.

Steiner, Jürg. 2012. The Foundations of Deliberative Democracy – Empirical Research and Normative Implications. Cambridge, UK: Cambridge University Press.

Bentham, Jeremy. 1789. Introduction to Principles of Morals and Legislation.

Stromer-Galley, Jennifer. 2007. “Measuring Deliberation’s Content – A Coding Scheme.” Journal of Public Deliberation 3 (1).

Gutmann, Amy, and Dennis F. 2004. Why Deliberative Democracy. Princeton, N.J.: Princeton University Press.

Gutmann, Amy, and Dennis F. 2002. “Deliberative Democracy Beyond Process.” The Journal of Political Philosophy 10: 153–74.

Steenbergen, Marco R, André Bächtiger, Markus Spörndli, and Jürg Steiner. 2003. “Measuring Political Deliberation: A Discourse Quality Index.” Comparative European Politics 1 (1): 21–48. https://doi.org/10.1057/palgrave.cep.6110002.

Mutz, Diana Carole. 2008. “Is Deliberative Democracy a Falsifiable Theory?” Annual Review of Political Science 11 (1): 521–38. https://doi.org/10.1146/annurev.polisci.11.081306.070308.

Kant, Immanuel. 1781. Kritik Der Reinen Vernunft. Konigsberg, Germany.

Mansbridge, Jane, James Bohman, Simone Chambers, David Estlund, Andreas Follesdal, Archon Fung, Cristina Lafont, Bernard Manin, and José Luis Martí. 2010. “The Place of Self-Interest and the Role of Power in Deliberative Democracy.” Journal of Political Philosophy 18 (1): 64–100. https://doi.org/10.1111/j.1467-9760.2009.00344.x.

Habermas, Jürgen. 2008. Ach, Europa – Kleine Politische Schriften XI. Frankfurt am Main, Germany: Suhrkamp.

Habermas, Jürgen. 1984. The Theory of Communicative Action. Berkeley, CA: University of California Press.

Habermas, Jürgen. 2006. “Political Communication in Media Society: Does Democracy Still Enjoy an Epistemic Dimension? The Impact of Normative Theory on Empirical Research.” Communication Theory 16 (4): 411–26. https://doi.org/10.1111/j.1468-2885.2006.00280.x.

Blok, Anders. 2007. “Experts on Public Trial – on Democratizing Expertise Through a Danish Consensus Conference.” Public Understanding of Science 16 (163). https://doi.org/10.1177/0963662507062469.

Haas, Peter M. 1992. “Introduction - Epistemic Communities and International Policy Coordination.” International Organization 46 (1): 1–35.

Cohen, Joshua. 1989. “Deliberation and Democratic Legitimacy.” In The Good Polity - Normative Analysis of the State, edited by Alan Hamlin and Philip Pettit. Oxford, UK: Blackwell.

Kahneman, Daniel, Paul Slovic, and Amos Tversky. 1982. Judgement Under Uncertainty - Heuristics and Biases. Edited by Daniel Kahneman, Paul Slovic, and Amos Tversky. Cambridge, UK: Cambridge University Press.

McCaffery, Edward J., and Jonathan Baron. 2003. “Heuristics and Biases in Thinking About Tax.” Los Angeles, CA.

2004a. “Masking Redistribution (or Its Absence).” In Behavioral Public Finance, edited by Edward J. McCaffery and Joel Slemrod. CLEO Research Paper Series. Los Angeles, CA: Russell Sage Foundation.

McCaffery, Edward J., and Jonathan Baron. 2003. “Heuristics and Biases in Thinking About Tax.” Los Angeles, CA.

2004a. “Masking Redistribution (or Its Absence).” In Behavioral Public Finance, edited by Edward J. McCaffery and Joel Slemrod. CLEO Research Paper Series. Los Angeles, CA: Russell Sage Foundation.

2004b. “Thinking About Tax.” Los Angeles, CA.

Sausgruber, Rupert, and Jean-Robert Tyran. 2011. “Are We Taxing Ourselves? - How Deliberation and Experience Shape Voting on Taxes.” Journal of Public Economics 95 (1-2): 164–76. https://doi.org/10.1016/j.jpubeco.2010.10.002.

Sunstein, Cass R. 1991. “Preference and Politics.” Philosophy and Politics 20 (1): 3–34.

Jiménez-Buedo, María, and Luis M. Miller. 2010. “Why a Trade-Off? – the Relationship Between the External and Internal Validity of Experiments.” Theoria 69: 301–21.

von Neumann, John, and Oscar Morgenstern. 1944. Theory of Games and Economic Behavior. Princeton, NJ: Princeton University Press.

Mueller, DC. 2003. Public Choice III.

Robbins, Lionel. 1976. Political Economy, Past and Present. London, UK: Macmillan.

Kahneman, Daniel. 2011. Thinking, Fast and Slow. New York, NY: Farrar, Straus and Giroux.

Keynes, John Maynard. 1936. The General Theory of Employment, Interest and Money. New York, NY: Harcourt Brace.

Hayek, Friedrich A. 1931. Prices and Production. London: Routledge.

Wapshott, Nicholas. 2011. Keynes Hayek - the Clash That Defined Modern Economics. New York, NY: W. W. Norton.

Taleb, Nassim Nicholas. 2007. The Black Swan - the Impact of the Highly Improbable. New York, NY: Random House.

Blühdorn, Ingolfur. 2007. “The Third Transformation of Democracy - on the Efficient Management of Late-Modern Complexity.” In Economic Efficiency, Democratic Empowerment, Contested Modernization in Britain and Germany, edited by Ingolfur Blühdorn and Uwe Jun. Lanham, MD: Rowman Littlefield Publishers.

1. For this insight, I am indebted to Franziska Deutsch who first noted in 2011 that tax was an interesting case because it affected everyone, but most people knew little about it.

2. For example, László Kovács (2004), then Commissioner for Taxation and Customs asked that tax rates must be allowed to differ 6-8% simply to make up for the remote location of some markets.

3. In the oecd-world, those are varying combinations of pre-paid, proportional consumption taxes such as vat and income taxes, often with different schedules for labor, capital and corporate incomes. Wealth taxes, often only on some asset classes, exist in some jurisdictions, but — with the exception of local US property taxes — are not a large revenue source.

4. Consensus Conferences were first developed and hosted in Denmark by the Technology Assessment Board to collect informed popular input for regulating science and technology.

5. Because participants can always drop out at will and will know when they are in the treatment condition, treatment cannot be strictly randomly assigned.

6. A comically naive example of such retreat comes from Piombino, Italy, where schoolchildren were “involved” (sic!) in urban renewal of the town piazza:

“After school trips (…) they had to make drawings of how the piazza should look. These often very colorful and joyful drawings were exhibited and also put on the website of the town.”

Praises Steiner (2012, 29): “In this way, children learned a good lesson about practical politics with a very concrete case of policy-making” (sic!). These and other exercises in “involvement” may be pedagogically or otherwise valuable, but emancipatory deliberation they are not. In fact, the example of children seems awfully revealing as an operative metaphor for this kind of “deliberative” citizen: let them draw cute pictures, or busy themselves with otherwise inconsequential activities, so that they feel heard.

7. Steiner (2012) presents a questionable summary of Rawls (1971): > This principle states that in a political decision the most disadvantaged social groups should profit the most (Steiner 2012, 95).

He finds it expressed when a parliamentarian says:

The government’s priority must be to help those who have the greatest difficulty in paying rather then spreading it evenly across the range.

The difference principle actually states that “social and economic inequalities are to $...$ be to the greatest benefit to the least advantaged members of society” (Rawls 1971, 266), a complex and conditional claim to moral validity that is characteristically degraded in both the operationalization and the coded snippet.

8. Summarizes Steiner (2012): “It is noteworthy that of the 1,027 speech acts, only 5 were interrupted by other participants, which indicates a calm and not boisterous atmosphere during the experiments” (2012, 47).

9. $O$ur dqi defined a sophisticated level of justification as having a certain complexity in the sense that an argument is justified with more than one reason and that these reasons are logically linked with the postulated outcome” (Steiner 2012, 64).

10. (King 2009) criticizes that coded interpretations are presented as objective fact, rather than inter-subjective.

11. Inspired by the Philosophy of Robert (Nozick 1974).

12. Not very inspired, and by no one in particular.

13. Inspired by the Philosophy of Henry (???)

14. By definition, taxes in a market economy — a pleonasm — must always be levied as some potential market exchange occurs. This is straightforward for income and consumption taxes, where the earning and spending transactions are taxed, but also holds for wealth taxes and even other instruments of expropriation (such as inflation). Taxation always requires some real (or imputed) market valuation of the given tax base, which in turn necessitates an actual or hypothetical transaction between market participants. In their laboratory experiment, participant buyers frequently tax sellers more than themselves, forgetting that — under perfect competition, especially factor rent and price flexibility — buyers and sellers will share any given tax on their transaction irrespective of the nominal incidence.[^tax-lse]

15. Tellingly, Sausgruber and Tyran (2011) already take pains to remind readers that tax-lse only holds when factor rents and prices are fully flexible — a condition that is as controversial as is it is difficult to verify, even after 60 years of macroeconomic debate about little else (Wapshott 2011).

16. Taxation is a good case, because — as Franziska Deutsch succinctly noted in 2011 — it affects everyone, but most people know little about it.

17. Ideational perspectives surely are important, and rationalist epistemologies not the only way to knowledge, but, as I explain in more detail in (p. ), first-order alternatives must be clarified first.

18. For this comparison, I am indebted to Maike Schulz, who — citing Eco — reminded me in 2013 that an ever elongating reading list need not be a bad thing.