CAIRN-INT.INFO : International Edition

1Over the course of the twentieth century, taxes have become increasingly technical, while also occupying more and more space in public debate. For actors in the political arena, this two-fold development entails the constant over-dramatisation of taxation concerns which, from the nineteenth century to the inter-war period, took the shape of caricatures and slogans. With the stroke of a pen or a pithy motto, attempts were made to convince the average lay person. Statistical issues were already part of the discussion, but they were secondary concerns in the debates over how each social category’s fiscal contributions should be distributed. In more recent times, however, the fiscal debate has become undeniably more complex, with numbers and the law playing a growing part in it. Nevertheless, metaphors and stories, even fictional ones, have remained an essential component of controversies regarding taxes.

2Many works devoted to the analysis of public issues have highlighted the importance of categorisation and framing in the dynamics of public policy. [1] Supporters of the cognitivist approach have argued that political leaders develop interpretive frameworks that are shared by all, but they overlook the different struggles and practices of self-interested actors. [2] In the fiscal realm, the issue of determining which social sectors are affected by injustice and who benefits from a certain measure is crucial for understanding the debates which arise. On the other side of the Atlantic, this perspective has resulted in an enigma: how can we comprehend the fact that political leaders have managed to enact fiscal measures for the benefit of the rich [3] with the full support of the majority of the population – as seen, for example, when George W. Bush passed tax cuts for the wealthiest Americans in 2001. [4] Among the explanations put forth, some have emphasised different social groups’ unequal access to the hidden stakes inherent to some political messages. In the mind of taxpaying voters, support for equality can often be combined with the belief that taxes for the wealthiest should be decreased. [5] In an attempt to explain Americans’ support for the elimination of the estate tax, others have insisted on the political skill of the Republicans in interweaving this issue with a broader campaign in defence of moral values, such as the fight against legal abortion or gay marriage. [6] Following in the footsteps of the linguist George Lakoff, who has argued that metaphors should be treated as full-fledged cognitive tools, Michael Graetz and Ian Shapiro have put forth a different interpretation of this “political mystery”: according to them, the success of the Republican crusade against fiscal reform can be explained by their ability to broadcast a few edifying examples, while those defending taxation often rely on statistics, repeating ad nauseum that inheritance taxes only affect about 2% of all Americans. [7] In other words, the moral rhetoric of those against the tax would necessarily win out over the cold objectivity of numbers.

3This investigation of the role played by rhetoric in the creation of political controversies is still under-developed in France, with the exception of analyses focusing on “windows of opportunity” that tend to limit themselves to the decisions made by high-ranking government officials in accordance with the electoral calendar. [8] And yet, the inception, the success and subsequent disappearance of the “tax shield” provides a good example for us to fully investigate the way in which stories, symbols and statistics are used to justify or criticise a public policy. Voted for by Dominique de Villepin’s government in autumn 2005, the fiscal shield established a tax ceiling, above which the state had to reimburse overly taxed individuals. The maximum tax bracket was initially fixed at 60% of income in 2005, and then lowered to 50% after Nicolas Sarkozy’s victory in 2007. This measure was very rapidly presented as a “political symbol”, heralding in a new fiscal era that was more protective of taxpayers and designed to compensate work and help reduce the global weight of compulsory levies. According to a number of different polls, initially public opinion was heavily in favour of this measure (over 70% in 2005), and then equally heavily against it from 2009-2010. [9] The adoption and then elimination (voted on in 2011, enacted in 2013) of this measure by the same majority six years apart is a unique phenomenon that deserves our attention.

4The tax shield allows us to examine the reasoning behind upward-redistributive policies, which benefit a few thousand of the wealthiest taxpayers. According to the typology established by the American political scientist James Q. Wilson, client politics are clearly different from majoritarian politics, not only from the point of view of the distribution of their costs and benefits (distributed costs and concentrated benefits in the first case, costs and benefits largely shared in the second), but especially with regard to their visibility and exposition. [10] In theory, client politics are more likely to succeed because they are highly opaque. [11] On the contrary, majoritarian politics are completely transparent, thanks to public opinion, polls and symbols – with the result that its promoters can reap significant political gains. The tax shield was novel because it combined properties stemming from both of these political approaches. Even though its effects were concentrated on a minority of taxpayers, political leaders promoting the shield tried to present it as a popular, majoritarian measure, good for rich and poor alike. The clash between the left and the right was specifically centred on how to characterise this public policy, as this would inevitably require stories and symbols to convince public opinion that it was either an example of client or majoritarian politics. In fact, as Wilson argued, it is “the perception of costs and benefits that affects politics”, rather than objective distribution. [12] The interest of studying the tax shield thus lies in this fundamental tension, as well as in the role of stories and statistics in changing perceptions.

5Attempts to credibly introduce and then roll out the tax shield revealed that symbolic storytelling does not always occur in opposition to coherent statistical analysis. The numbers themselves are involved in the creation of narrative. The construction of public taxation issues requires the ability to use both these registers in turn in order to gain support and be credible over the long term. Here we shall thus first study the creation of narratives and a body of metaphors that were likely to sway the convictions of the greatest number of people. Initially, this process consisted of turning the atypical situation of a handful of farmers into a symbol of national value, while remaining discreet about the social characteristics of those who would actually benefit from the reform. Very quickly, however, supporters of the shield were also forced to turn to statistics in order to justify the measure’s majoritarian facet. When the profiles of those benefiting from the measure came to light, this laid fertile ground for stories and symbols to be flipped. From an instrument meant to protect taxpayers from excessive taxes with regard to their income, the tax shield thus became the embodiment of fiscal injustice and client politics.

Methods and materials used in this study

In order to account for the multiple dimensions involved in construing the tax shield as a public problem, we used several types of different materials. We conducted a systematic search for repeated occurrences of the expression in parliamentary debates, Senate reports and Assembly transcripts. In order to cover the widest possible spectrum of the media, we used the Factiva software package for the written press. A search for “tax shield” occurring between 1 September 2005 and 31 December 2011 produced 5,391 results – in daily and weekly newspaper articles (with the notable exception of Le Monde) and press releases from agencies – which reached a peak in 2010 (this year alone producing 2,041 documents, or almost 38% of our whole corpus). We paid special attention to audiovisual media, which we analysed by consulting the Inathèque archives and noting all the instances where the tax shield and tax on wealth were mentioned between 2005 and 2011. A third type of source was used: interviews with two high-ranking officials from the General Directorate of Public Finances and with a member of parliament who was particularly implicated in the controversy over the fiscal shield. The fragmented nature of these interviews can largely be explained by the reticence of the main players to evoke a topic that was still politically fraught at the time of our study, in 2011 and 2012. Likewise, polling institutions refused to share with us the names of those commissioning the polls and the exact results of opinion surveys concerning the shield, widely used by political actors.

Protection against taxes: a “popular” cause?

6The transformation of a bias or a demand into a political problem means at the very start adopting a language that is designed to attract as much support as possible. Just as the term “death tax” was a semantic innovation designed to delegitimise taxes on inherited wealth, [13] the “tax shield” was an expression which positively symbolised the principle of protection against taxes, while conserving a certain degree of ambiguity on the status of the victims thus spared.

The right words to say it: rewording fiscal injustice

7In a country like France, where inequalities in inherited wealth were very significant during the nineteenth and the beginning of the twentieth century, [14] the project for a capital tax has been defended by the left and contested by the right since the 1920s. [15] Postponed for a long time, this measure only became a concrete vision when François Mitterrand came to power in 1981 and created the Impôts sur les grandes fortunes (IGF - Tax on Large Estates). Rightwing members of parliament vigorously opposed the measure straightaway, and, being unable to eliminate it, tabled an amendment via Maurice Schumann (RPR senator from the Nord region) establishing a “ceiling” that would limit taxation to 80% of tax income declared. [16] The majority in parliament rejected the amendment but ever since that time, the ceiling proposed combined two different (and not necessarily correlated) economic ideas: wealth and income.

8Eliminated by the right in 1986, the IGF was reinstated by the left in 1988 and renamed Impôt de solidarité sur la fortune (ISF - Solidarity Tax on Wealth), with the goal of funding the brand-new Revenu minimum d’insertion (RMI - Minimum Welfare Income). Michel Rocard’s government then took on the ceiling notion, which prohibited the total of the ISF and income tax from exceeding a certain threshold (initially set at 70% of taxable income, then raised to 85% in 1991). The goal was to get people to accept the reinstating of a wealth tax by promising that it would not be confiscatory (the mechanism allowed taxpayers to deduct any overpayment from their ISF totals). After 1995’s tumultuous strikes, in order to make other deductions more palatable, Juppé’s government introduced a “ceiling to the ceiling”, which was meant to limit the effects of the ceiling on those with large personal fortunes. Having won 2002’s elections, the parliamentary right renewed its attacks against the ISF, without however demanding its elimination, a move which many had believed to be decisive in Jacques Chirac’s electoral defeat in 1988. In the report that he presented to the Senate in 2004, [17] Philippe Marini (UMP senator from the Oise region) indicted the ISF as being both costly (given its negative effects on competitiveness and the nation’s attractiveness to investors) and unjust (due to a discrepancy between the nominal growth of assets and an out-of-date barometer) and the cause of many relocations. Statistics were at the heart of his argument. Using data from the General Directorate of Taxation’s monitoring agency, which tracks every individual liable for the ISF tax who moves abroad, Senator Marini put the number of exiled persons at between 350 and 370 per year, without in reality knowing if these moves were for tax reasons or not. To make his argument more convincing, he brandished the figure of 2,525 taxpayers forced to relocate over the course of the previous six years, disregarding the number of individuals who had moved back during this period. The result was allegedly disastrous, entailing loss for the national economic fabric and important shortfalls for public finances. [18]

9It was following this report that the term “tax shield” made its appearance in French political debate. The expression was directly borrowed from Italy, where Silvio Berlusconi created a “scudo fiscale” in 2001. [19] Like any lexical transfer, the transition from Italian to French involved a shift in meaning and purpose. The initial goal of the Italian shield was to allow expatriated capital to return to Italy and be “protected” from taxation, after payment of a modest fee of 2.5%. Under the guise of widening its taxpayer base and filling the public coffers, the “scudo” protected expatriated capital from the normal application of the law and instituted a temporary and exceptional right to “amnesty”, according to its architects. As it was subsequently used by the French right-wing, the term “shield” concealed a certain degree of ambiguity regarding the social status of its beneficiaries, who were sometimes those subject to the ISF and other times all taxpayers likely to be excessively taxed. The taxonomic change from “ceiling” to “shield” is not insignificant. While the ceiling targeted solely those individuals with significant assets but small revenues, the shield gained a wider meaning and implicitly linked taxes with an attack or oppression against which taxpayers needed to be protected.

10This notion was in line with the demands of American social movements that had made the tax ceiling a popular measure of social justice. Proposition 13, adopted in 1978 by two-thirds of California’s electorate, limited property taxes to 1% of the building’s value (compared with 3% previously) and capped the increase in property taxes at 2% per year, regardless of inflation. [20] Several other US states then adopted similar measures, inspired by Arthur Laffer’s famous saying according to which, above a certain threshold, “too many taxes kill taxes”. Those defending the tax shield in France nevertheless preferred to refer to its German counterpart, deemed more contemporary and more legitimate, even if in reality they were conflating two different experiments. Germany had in fact eliminated its wealth tax, but without putting any “shield” in place. But the German example still provided legal legitimacy to the ISF’s critics, who referred to the Karlsruhe court’s ruling in 1995 that judged the wealth tax to be unconstitutional when it lead to taxation that was greater than half of a taxpayer’s income. In his attacks on “capping the ceiling”, Philippe Marini conjured up the threat that France could one day be sanctioned by the European Union’s Court of Human Rights for infringing property rights. In this sense, the shield was presented as a way to restore the constitutionality of the ISF. However, the Constitutional Court had several times already confirmed this tax’s conformity with article 13 of the Declaration of the Rights of Man and of the Citizen, underscoring that income could not be used as an appropriate basis to evaluate individuals’ contributive capacities when taxing their personal fortunes. [21]

11The transition from a “ceiling” to a “shield” thus widened the scope of the measure, without it being open to criticism as an example of client politics in favour of the wealthy. Evoking a war-like and protective metaphor, the shield suggested that it could be applied to all taxpayers, thus making taxation fairer and more respectful of the law. Making such a measure more popular would nevertheless mean “framing” the debate, associating it with values, images and symbols that were likely to sway the greatest number of people.

A media parable: mobilising property owners on the Île de Ré

12The shield was to be used to “protect”. In order to legitimise it, the categories of people likely to benefit from it had to be identified. The idea that rich property owners fleeing France were the victims of over-taxation was a notion developed at length by Philippe Marini in his 2004 report. However, the figures accompanying this report were uncertain and disputed. [22] The argument according to which the ISF had become an archaic tax, eliminated by the majority of other European countries, was likewise not enough to legitimise the shield. However, the unique configuration of the Île de Ré provided a way to illustrate the injustices of the ISF, by exhibiting victims that were much more likeable and familiar than capital exporters.

13The parable that sought to present the farmers of the Île de Ré as wretched victims of the tax code had its roots in a particular context. When the bridge connecting the island to the continent was completed in 1988, the demand for luxury tourist infrastructure greatly increased, while the local population’s desire to class the area as a heritage site (identifying sensitive sites, establishing no-construction zones) significantly limited the amount of land available. [23] The combination of these two factors caused the price of land to explode; some properties increased eighty-fold in just twenty years. As long as they worked their land, farmers on the island could ignore this market change, as work-related property was excluded from the ISF’s purview. The spike in prices nevertheless put many at the helm of personal assets totalling several million euros, liable to be taxed like any other assets of the same value. But this situation was not enough to explain the unique political impact of the Île de Ré. The media coverage and political attention this local issue received was handled by “special issue stakeholders” [24] who knew exactly how to transform an incident of local indignation into an issue of national injustice.

14Following several inspections carried out by the tax authorities, landowners came together under the umbrella of the Association de défense des habitants de l’Île de Ré (ADHIR – Association for the Defence of the Residents of the Île de Ré) which, 160 members strong, soon shifted its exclusive focus to condemning the injustices wrought by the ISF, its concerns voiced by its president Valérie Constancin. Its spectacular achievement was to make a few local farmers, who owned properties worth twenty times more than the island’s other farms, look like paupers and representative victims of the ISF on a national scale.

15Born in Antony in 1964, Valérie Constancin completed the first part of her studies in Versailles, before going on to study law at Assas. After moving to the Île de Ré, her spouse’s hometown, she started out in the hotel business, taking over a three-star hotel that allowed her to become one of the island’s wealthiest residents (with a personal fortune of 15 million euros). [25] After running – and losing – as a right-wing candidate in Sainte-Marie’s municipal elections in 2001, Constancin nevertheless did not give up politics entirely: she embarked on a crusade to bring to light the situation of the island’s farmers and fight to eliminate the ISF.

16For the Association, being notified of the first tax assessments was the opportunity to get national attention. In an article from 19 March 2005 titled “Sur l’Île de Ré, l’ISF frappe les non imposables” (“On the Île de Ré, the ISF hits non taxable individuals hard”), Le Parisien condemned the grotesque situation of those households which were subject to the ISF despite barely earning the minimum wage. The information was even more striking because it seemed to go against all common sense. The article was supported by allegedly emblematic stories, which were then repeated across various media outlets and formats. For example, there was the story of Noëllie Plaideau, a former child-minder making less than 1,000 euros a month and who, upon the death of her husband, allegedly owed ten years of ISF tax, or 50,000 euros in all. She was first mentioned in Le Parisien’s article and then reappeared on TF1’s television broadcast later the same day. In addition, Plaideau’s story was mentioned in an article in Libération[26] and in a piece calling for the reform of the ISF in L’Expansion. [27] The story most frequently cited by journalists, however, was that of René and Jacqueline Passé, a retired couple who were supposedly forced to sell one of their two fields to pay the ISF. This couple first appeared during TF1’s news bulletin, lamenting “the loss of an inheritance which belonged to us, which is gone and which we shall never see again”. [28] A few weeks later, the Passés were once again featured on another TF1 program, then cited again in an article in Le Point, [29] and finally present as well on the show C dans l’air, where they lamented the fact that they had never been able to buy anything nice for themselves. [30] Out of the 500 inhabitants of the Île de Ré who were allegedly affected by the ISF, only a few retirees appeared in the media, used to illustrate Valérie Constancin’s argument, which always ended with the same coda: [31] the aberration of these situations revealed the twisted effects of a tax that despoiled honest farmers. The cause of the poor landowners of the Île de Ré reached an even greater audience when it appeared during a primetime entertainment news program hosted by the popular Christine Bravo.


“What really scares me is that this is likely to happen to all of us, it can happen to me even though I’m not from the Île de Ré. All of a sudden, my neighbourhood takes off and I have to pay the ISF because next door, there are billionaires who make the prices go up in my neighbourhood.” [32]

18Valérie Constancin’s strong aptitude for convincing others lay in her ability to generate empathy, not only in the television hosts in front of her, but also in viewers at home. From her point of view, the world was divided into two kinds of people: on the one hand, “landowners on the Île de Ré who are simply acting as good citizens by preserving traditions and wanting to hand down a legacy to their children” and who “end up paying the ISF”; and on the other hand, “billionaires who own Van Goghs and Renoirs and don’t pay the ISF”. Presenting herself as the champion of a “traditionally French population, of the land which forms the backbone of the country”, Constancin revived the long tradition of resistance to taxes which made small rural landowners into the prototypical victims of the fiscal Hydra. She likewise exploited the feeling of injustice generated by the different treatments reserved for different types of assets: the much greater difficulty of taxing movable capital than capital in the form of real estate, the ISF’s exclusion of works of art from its calculations since 1981, etc.

19Much like any literary narrative, this folktale relied on the familiar to create the illusion of reality. Its symbolic power resided in the possibilities it offered to individuals to act in the political arena by transforming anecdotal experiences into established representations. [33] But storytelling talent was not enough to turn a narrative of injustice into a draft law. In the case of the tax shield, the work of special interest advocates entailed working directly with those who enjoyed the political monopoly, so that their status as victims would be recognised – thus turning the story of a few farmers into a symbol of fiscal injustice.

From symbol to instrument: a difficult implementation

20The transformation of a social issue into a political problem first presupposes establishing a way of categorising the groups of people subject to a political measure. [34] With regard to the fiscal shield, the symbol of the Île de Ré farmers played a structuring role, not least due to its ambiguity: here were taxpayers who presented themselves as having modest means, but in fact were at the helm of multi-million euro estates. This very uncertainty regarding the social status of those benefiting from the measure was at the heart of attempts to legitimise the shield and influenced the whole process of its implementation.

Putting the issue on the political agenda

21The media’s parable of the destitute and despoiled farmer from the Île de Ré allowed for public recognition of the need to protect households with low incomes from the ISF. But in order for this symbol to become expressed in law, it was necessary for political actors to own it and find a reason therein to justify their actions. [35]

22In the spring of 2005, Valérie Constancin contacted several members of parliament and convinced Jacques Masdeu-Arus, a UMP member from the Yvelines, to file a proposal for a law to exclude one’s primary place of residence from ISF calculations, since “important disparities regarding the price of real estate throughout the French territory make this tax particularly inequitable”. [36] The ADHIR continued its efforts and joined with the Union nationale de la propriété immobilière (National Union of Property Owners, 200,000 members) and the Contribuables associés pour demander à tous les députés et sénateurs de mettre fin au caractère injuste et aux effets néfastes de l’ISF (Taxpayers’ Association to ask all parliamentarians and senators to put an end to the unfair nature and harmful effects of the ISF). On 23 May 2005, Valérie Constancin was received by Thierry Breton, then Minister of the Economy, who rapidly adopted a position.


“Over the years, this tax has become no longer a wealth tax […] but quite simply one more tax on the savings and homes of our fellow citizens.”

24The campaign to try and stigmatise the ISF by popularising the plight of the Île de Ré farmers quickly bore fruit, creating a sort of fiscal illusion one could term “obfuscation”: a process by which political players deliberately mask the cost of a measure from the public by instead focusing on visible gains. [37] Here the dissimulation entailed maintaining the ambiguous nature of the ISF’s victims’ economic situations, so that the largest number of people could identify with their circumstances. The combination of local mobilisations, thrust onto the national stage, with the arguments of those who for several years had already been looking for ways to limit the ISF – without, nevertheless, seeking to eliminate it entirely – meant that the tax shield finally made the political agenda.

25In the fall of 2005, the parliamentary right-wing jumped on the Île de Ré example in order to advance its tax shield project, as can be seen in the two UMP interventions below, pleading for serious reforms of the ISF.


“What people call the ‘Île de Ré syndrome’ in reality affects thousands of landowners across France. These individuals with modest incomes will always be forced to pay the ISF, not because their income is being taxed, but simply because they possess land that has increased in value.” [38]
“Everyone can remember the absurd case of the Île de Ré, where real estate and land prices have skyrocketed. Small landowners, whether retired or not, suddenly found themselves being asked to pay exorbitant sums, which they could only do by selling their property. It is inherently unjust, even inhumane, to strip these farmers of their lands, where generations have followed each other, working the land without asking anything of anyone. » [39]

27The story of the Île de Ré also worked as an exemplary metaphor, in the sense that many members of parliament presented it as representative of a great many other situations. The image of the island’s small landowners allowed the tax shield to be painted as an act of justice, designed to benefit the greatest number of people; that is to say, all those who, despite not being rich, had seen the value of their property dramatically increase thanks to runaway real estate inflation. It was thus not a matter of legislating for a small minority of wealthy taxpayers, but for those “like everyone else” who by chance would become subject to the ISF even though their incomes had remained quite low. The reform proposal did not so much concern a circumscribed social group as much as a situation in which taxpayers could end up without having anticipated it (symbolically, this was referred to as the Île de Ré “syndrome”). In order to make it seem that most people would approve such a measure, those in power relied on a variety of polls: during the debates on the floor of the National Assembly, the Budget Minister Jean-François Copé announced that more than 71% of French citizens would be favourable to the creation of a fiscal shield as proposed by the government. [40] If the project was supported by public opinion, then clearly that signified that it was a fair measure, and vice versa: this circular and self-fulfilling reasoning was persistent and unassailable. But the support of those polled was in fact concerning a general principle (putting a cap on taxation), due to the absence of reliable data describing the social profile of those who potentially stood to benefit from the reform. Those fighting against the tax shield seized on this argument, much like the Centrist Charles de Courson.


“Is the creation of a `tax shield’ a good idea? No. If 71% of French people don’t agree with me, it is because they don’t know what this measure entails.” [41]

29Opponents also added that, in tax matters, what polls measured above all were the inconsistencies voiced by many citizens when they expressed their preferences:


“Ask French citizens if they want lower taxes. 100% will be in favour. Ask them if they want certain public services to be cut and 90% of them will tell you no. Polls on fiscal policy would make you cut taxes but not spending, and thus make the public deficit worse.”

31In order to lend credence to the idea that it was a measure directed primarily at modest incomes, Copé shared the projections that the Ministry of Economy, Finances and Industry had given him. According to these numbers – which, until 2007, were the only legitimate figures in the public debate – 93,000 individuals were likely to benefit from a 60% tax shield, 81,000 of whom were among the “lowest-income”; [42] 165,000 persons subject to the ISF (out of a total of 330,000) could benefit from the measure. Conclusively, Copé announced that for 87% of the people concerned, the shield would affect taxpayers in the first income decile: working-class or lower middle-class households.

32The statistical argument thus became essential to conferring rational legitimacy to the story of the Île de Ré. The lack of any data other than that provided by the Ministry of Finance gave the existing government a significant advantage: depending on its agenda, it could favour one projected scenario or another and choose an auspicious time to make it public (or not), without the opposition being able to use the same tools.

Statistical uncertainty and political acceptability

33After being adopted in the fall of 2005, the tax shield returned to the forefront of public debate during the 2007 election campaign. Under the guise of protecting inherited wealth (presented as the fruit of decades of labour and not the mere handing down of sizeable annuities), Nicolas Sarkozy proposed eliminating inheritance taxes for 95% of taxpayers, lowering the tax shield’s threshold from 60% to 50% of incomes and grouping it together with the Contribution sociale généralisée (CSG – Generalised Social Contribution) and the Contribution pour le remboursement de la dette sociale (CRDS – Contribution for the Reimbursement of Social Debt). Although it was presented as a mere extension of existing policies, this measure actually modified the provisions substantially. Setting the tax shield at 50%, a symbolic gesture, also recalled the popular slogan used by taxpayer associations, according to which citizens should not work more than one out of every two days for the state. Sarkozy used this motto for his own agenda when he said “I want the state to be forced to let everyone keep at least half of what they made”. [43] Once again taking advantage of the approval bestowed upon him by various polls, [44] he tried to impose the idea that this was a measure that the majority of French people approved of, despite the fact that the tax shield was far from being a central point of his campaign, and that the rate initially set at 60% had not even been applied yet: 2005’s income was declared in 2006, and refund applications could be filed between 1 January and 31 December 2007. Those polled were thus expressing an opinion about a campaign promise, without any information about the real costs to public finances or the distribution of its effects on taxpayers. Here we can identify the source of what the political scientist Larry Bartels has represented as a gap between voters’ values and the policies they end up supporting. [45] Without access to real data regarding the tax shield’s effect, it is possible that voters in 2007 sincerely believed that this measure would help low-income taxpayers who had been unjustly burdened.

34As early as January 2007, the powers that be noted the very low number of applications filed to obtain a tax shield and proposed the following explanation: taxpayers were being cautious on the advice of their tax advisors, who feared that every application would give rise to a thorough check of the taxpayer’s entire file. To allay these fears, the budget minister asked the tax services to refrain from in-depth verifications when a tax shield application was filed. In May 2007, a few days after the second round of the presidential election, the first numbers were made public. The gap between the measure’s political weight and the small number of taxpayers actually affected either engendered surprise or merely acted as confirmation of the unequal nature of the measure, depending on one’s political camp. At that point in the year, only 1,780 taxpayers had applied for a refund on the taxes they had paid in 2006, or a mere 2% of the 93,000 individuals originally expected. This gap – which seemed enormous to some – could in reality be explained by the difficulties involved in measuring the number of low-income households liable to benefit from the reform.


“The problem is that we don’t a priori know all the incomes and we have to envision behaviours that, by definition, we don’t know either. Identifying the number of people who could benefit would have meant knowing the number of taxpayers who were eligible for the refund and the number of those who would claim it. For very low incomes, the curve is so flat that things can change dramatically, depending on what parameters you introduce. The uncertainty comes from a margin squeeze: you have to calculate at the same time a part of an income that you don’t know and a minimum threshold below which you have to assume that the taxpayer won’t claim his or her shield. So if you change one parameter or another, the numbers change by the tens of thousands. Simply put: it’s like trying to shoot an arrow at a moving target while blindfolded… So I think that we should cut the people making estimates some slack.” [46]

36Unlike the tax authorities of other countries, France’s Ministry of Finance does not use complex sociological or behavioural economic studies to predict the possible attitudes of its taxpayers, which often leads to sizeable discrepancies between the predicted and actual figures for any given measure. [47] In the fall of 2007, the number of individuals who had actually asked for the tax shield to be applied was still much lower than predicted (there were 2,398 applications by 31 August 2007 and 3,116 by 30 September). Past statistical estimates thus put the government in an embarrassing position: contrary to the numbers announced, the shield seemed only to be helping a tiny number of taxpayers. In May 2007, the media revealed that a single taxpayer, the heiress of the Galeries Lafayette, had already received a seven million euro refund. Faced with numbers that clearly contradicted the stated goal of the reform, the budget minister explained this failure “by the fear of tax monitoring and a lack of sufficient information”. [48] In addition to multiplying reassurances that taxpayers had nothing to fear from his administration, the minister also posted an online calculator for people to check and see if they were eligible for the tax shield and what they could expect to gain from it financially. There were only a few short months left to reduce, albeit not eliminate, the discrepancy between the some 2,000 applications received during the first half of 2007 and the 93,000 potential beneficiaries cited by the Finance Ministry. This situation ultimately raised questions about the legitimacy of a measure in which much had been invested politically.

37Taking stock of the gravity of the situation and the opprobrium that the projections provided three years earlier would attract if they turned out to be completely wrong, high-ranking officials in the tax authorities proposed contacting the “small taxpayers” who were likely to benefit from the tax shield but were simply not aware of it. [49]


“For a long time, we had constantly followed up on the employment allowance measure. So for the tax shield we applied the same principle. We developed a program which churned out a list of possible beneficiaries and we asked our staff to contact these people.”

39While tax officials were busy, as every fall, trying to regularise local taxes, they then received instructions to seek out individuals who were likely to benefit from the tax shield. In total, 80,000 letters were sent out to low-income households, urging them to look into the tax shield. The results did not take long to appear. On Reunion Island, more than 5,000 fiscal households – the vast majority both welfare recipients and home-owners – claimed that the tax shield applied to them. [50] Until that time, those subject to property taxes would play up their very low incomes and always obtained generous tax relief from the tax authorities. From 2007, this situation went from being an exemption offered in cases of poverty to a right that, thanks to the tax shield, taxpayers could use to counter tax demands.

40The follow-up letters sent by the government had a two-fold effect, both qualitative and quantitative. During the last two months of 2007, the total number of refund applications went from 3,000 to 20,177 (13,700 of which were ultimately accepted). The majority of the new applicants to benefit from the exemption were low-income households, which allowed the image of a fair and balanced tax measure to be conserved. The total number was of course still far from the 93,000 originally predicted, but nevertheless lowing the shield to 50% of income was put forth as a solution to this problem – in particular because the Finance Ministry provided a new set of predictions in the summer of 2007, announcing that the measure available starting in 2008 would benefit about 230,000 individuals. The relative failure of the tax shield in its original incarnation at 60% thus appeared to legitimate the idea behind its extension, voted upon in August of 2007.

The vagaries of administrative implementation

41Although the issue of the social status of those poised to benefit from the tax shield was crucial to its legitimisation, it also played a part in the debate over its administrative implementation. The controversy that erupted after 2007’s presidential election between political leaders and the administration regarding the shape that the refund procedure should take was an example of this. The government and certain members of parliament attributed the low number of refund applications to the process used. [51] The shield could be demanded ex post, after taxes were paid and the application was reviewed by the tax authorities. If the application was accepted, the taxpayer in question received the refund in the form of a check in the mail. Although defended by the ministry, this mechanism rapidly drew criticism from political leaders for making things more complicated for taxpayers and preventing them from being able to calculate the deductions for which they were eligible themselves while filling out their tax returns. Elected officials suspected that the administration, prone to excessive regulation, was trying to stall refund measures and thus called for a shift to a disclosure procedure. The President of the Republic shared this opinion [52] and asked for the Ministry of Finance to put an end to administrative sluggishness. [53] A reverse charge system was finally put in place in the fall of 2008, which flipped the burden of proof: now the administration was required to evaluate the tax shield that taxpayers self-calculated.


“At the start, what political leaders wanted us to do was ask us to send a check straight off, starting the first year, to all the people who demanded the tax shield, just because they declared ‘I have a right to 5,000 euros, give them to me’… We replied that we didn’t have the full picture and that we needed taxpayers to give us the information we didn’t have so we could calculate their shield: we’ll refund once we’re sure of the amounts and we do it in hopes of providing the best service possible to users. But for those in politics, asking taxpayers to declare all these income elements that they don’t necessarily know is like an inquisition… The reverse charge system was monstrously complicated because we had no way of knowing if the amount asked for as a shield was the right amount. We couldn’t send out checks into the wild on the basis of totals that we didn’t know.” [54]

43The goal of the tax shield was not just to refund over-taxed individuals. As a tool of public policy, it also helped to redefine the relationship between taxpayers and the administration with regard to principles, bureaucratic processes, and the ministry’s system of communication. [55] It was part of a long evolution, begun at the end of the 1980s, which increasingly sought to characterise relationships between taxpayers and the tax authorities as a type of partnership, taking the shape of a contract. [56] It was not just another exemption or deduction that “ate away” at the French tax system (the infamous “loopholes” which remained secret for a long time [57]), but well and truly a basic rule, contained in the first article of the general tax code. [58] The symbolic, even philosophical, scope of this stipulation was once again picked up by the Minister of Finance and the Economy during a senatorial debate on the government’s “tax package” in July 2007.


“The 50% threshold is obviously full of symbolism. It represents a true partnership between the taxpayer and the state, as is already the case in a number of other European countries. 50/50: this is the equation behind our fiscal contract.” [59]

45The political investment made in this symbol was such that its implementation could not happen silently. The fiscal administration, used to targeting taxpayers who owed money, now had to seek out those to which it owed money as a result of overpayment. It was no longer a question of getting taxpayers to understand the importance of their tax duty, but helping them to advocate for their rights against the administration itself. This switch, an obvious departure from the entirety of the social history of taxes, inspired some reticence on the part of tax officials. [60]

A symbol reversed

46The financial and economic crisis of 2008 quickly and harshly put public finances to the test. The tax cuts decided upon in 2007 – of which the tax shield was part – thus increasingly appeared out of sync with the government’s promises of balancing public accounts. Moreover, the first statistics published regarding the nature of those who benefited from the tax shield stripped the latter of most of its legitimacy. In the space of a few months, the shield became the symbol of client politics which exacerbated inequalities rather than mitigating them.

Concentrated but highly visible effects

47The shield’s application relied upon a paradoxical situation. Having made it a strong symbol of its policy, the government wanted to make its results public. But this desire contradicted a structural trait of France’s fiscal administration, used to secrecy and discretion. The civil servants working at the ministry very quickly understood the political sensitivity of the measure and the care taken by the government to regularly communicate the evolution of the shield’s application.


“Regarding the shield, we immediately felt that it was very politically sensitive and we worked with monthly statistics. Every month we would produce a report; then mid-year a cabinet member would pull out a number to say, look at how the shield is working, we’re at this point now, look at how it’s a measure that’s good for all our taxpayers… that’s when we lost all control, it was out of our hands at that point. But it’s true that we communicated a lot of statistics to the cabinet because we knew it was a sensitive topic.” [61]

49Beyond the aforementioned controversies and conflicts of jurisdiction, here we can also see the implicit ties that bound high-ranking government officials and political leaders. [62] They both shared the hypothesis that better statistical visibility with regard to the tax shield would help its wider acceptance. And yet no automatic causal link exists between the visibility of levies and free consent to taxation: in the United States, where income taxes played a very important role after 1945, compulsory levies were much more hotly contested than in France, where consumption taxes and social security contributions (less visible but also less progressive) have always been predominant. [63] In fiscal terms, opting for transparency is a double-edged sword: although it certainly confers a greater degree of legitimacy to the fiscal system, it also more easily gives rise to contestation. This same reasoning can be applied to the tax shield. Erected as a political symbol, touted by emblematic figures and transmitted by inspirational narratives, the reform was measured against several indicators that were supposed to illustrate its legitimacy: the number of people benefiting, their social profiles, the expected return of fiscal exiles, an increase in France’s attractiveness to investors, etc. But interpreting these facts actually produced a counter-narrative and alternative symbolic figures. In the summer of 2010, the media revealed that Liliane Bettencourt had been able to benefit from the tax shield, all the while concealing several Swiss bank accounts and the ownership of an island in the Seychelles, the Île d’Arros, worth approximately 40 million euros. Over the course of a few years – albeit ones marked by 2008’s financial crisis – the symbol of the tax shield was flipped: officially designed to help low-income taxpayers, it became the embodiment of inequitable policies, more likely to benefit the owner of the Île d’Arros than the inhabitants of the Île de Ré.

50None of the numbers made public by the ministry truly succeeded in confirming the political validity of the tax shield. The statistics coming out of Bercy – no longer predictions but reflections of the actual situation – showed a strong concentration of gains among the wealthiest taxpayers. The refund mechanism revealed a process of inverse redistribution which was expensive both politically and symbolically: in 2009, it thus came to light that, out of the 19,000 or so individuals benefiting from the tax shield, the richest 1,170 households had, on average, received a check for approximately 360,000 euros, consequently hoarding 62% of all the refunds issued as part of the tax shield.

Table 1

Statistics on the application of the tax shield and budgetary developments (2007-2011)

Table 1
2007 2008 2009 2010 2011 Number of individuals benefiting from the tax shield 13,700 15,446 18,764 16,223 13,034 Global cost of the tax shield (in millions of €) 229 563 679 636 735 Average refund (in € per household) 16,700 36,500 36,000 39,000 56,400 Deficit as defined in the Maastricht Treaty (as % of GDP) -2.7 -3.3 -7.5 -7.1 -5.2 Debt as defined in the Maastricht Treaty (as % of GDP) 64.2 68.2 79.2 82.3 86

Statistics on the application of the tax shield and budgetary developments (2007-2011)

Sources: For the numbers concerning the application of the tax shield, see the memo issued by the General Directorate of Public Finances on 5 April 2012, published in Le Parisien on 26 June 2012. For the numbers regarding the deficit and debt as defined in the Maastricht Treaty, see the INSEE, Finances publiques, consulted online on 30 October 2012: <>.

51The other indicator that was supposed to attest to the shield’s economic effectiveness was not ultimately much more successful: according to the ministry’s estimates, the number of tax exiles leaving each year remained the same (around 800), while the number of tax exiles returning increased marginally at best. Also unfavourable to the tax shield’s image was the fact that media attention was less focused on actual numbers than on a handful of celebrities, whose returns were publicised and then contradicted. After having left for Switzerland in 2006 for openly financial reasons, in May 2007 the singer Johnny Halliday announced that he would consider returning to France after Sarkozy’s victory and the adoption of the new fiscal shield. The following autumn, however, the singer was still tax domiciled in Switzerland. Likewise, Denis Payre, a French entrepreneur exiled in Belgium, appeared in the media multiple times between 2005 and 2007 as the symbolic figure of the kind of talent that would return to France if tax legislation were modified, in particular the ISF. When the 60% tax shield was adopted, Payre announced his return to France in the fall of 2006, but then changed his mind. On 9 May 2007, a few days after Sarkozy’s victory, Payre made a second announcement regarding his return to France, now expressing his full satisfaction with the 50% tax shield. Nevertheless, he only returned home in 2009, ultimately citing professional and family reasons as much as fiscal motivations. Neither statistics nor media celebrities seemed to confirm the shield’s attractiveness. After some time, the measure even began to encourage rather than hinder tax optimisation strategies. As early as 2006, bankers and asset managers remarked upon the boon the tax shield represented for large property owners, especially if they could capitalise on the “Île de Ré phenomenon”. [64] In March 2009, the Conseil des prélèvements obligatoires (Council on Compulsory Levies) observed the shield’s perverse impact, which had led some taxpayers to adopt new optimisation strategies in order to decrease their income and benefit from the shield even more. [65] And on the other hand, the small landowners of the Île de Ré who had conferred rural and popular legitimacy upon the shield were scarcely visible any longer. Those whom the shield was supposed to protect – hardworking farmers, economic or artistic talents – were nowhere to be found. Who then was benefiting from the shield?

52The Bettencourt scandal sealed the shield’s narrative fate in the summer of 2010, demonstrating once again the power of great scandals to expose social truths and push moral boundaries. [66] In addition to its political and legal facets, this scandal helped to give a face to the several hundred taxpayers who most benefited from the shield. In particular, public perception of the shield’s effects shifted from a debate on averages to a debate on extremes: people no longer talked about a vague group of 1,000 taxpayers who got a theoretical 300,000 euros each, but instead singled out one individual who had received 30 million euros, or about 5% of the measure’s total cost. The shield was no longer represented in the same way, statistically speaking: instead of abstract numbers supposed to represent tens of thousands of individual cases, now the shield represented such highly concentrated financial benefits that its political acceptance grew trickier and trickier. What is more, the image projected by the Bettencourt scandal was the exact opposite of the one originally targeted by the shield: a rich heiress, who had combined the benefits of the tax shield with optimisation strategies and tax evasion, had now replaced the figure of the hardworking small landowner. The shield had ultimately been stripped of the stories that had legitimised it, which in turn led to a drop in its popularity.

A resilient symbol

53With hindsight, it would be possible for us to deem the government’s persistence in defending the tax shield until the fall of 2010 as incomprehensible or gauche, in particular amidst the financial crisis, the deterioration of government accounts (see Table 1 above), and the echoes of the Bettencourt scandal. In fact, the case of the tax shield poses the question of the reversibility – or irreversibility – of political symbols. How could the government justify eliminating a measure that had been erected as the symbol of a new era in tax legislation? How could the same majority that had adopted the shield a few years prior now negotiate politically for its abrogation?

54The warning signs of public disaffection regarding the tax shield had not been lacking, however: an inversion in polling trends, defeats in mid-term elections, pressure from a number of local elected officials and parliamentarians in favour of eliminating what had quickly been identified as an “electoral albatross”. While figures from the ruling party expressed their doubts (Alain Juppé and Jean Arthuis from 2009, and Jean-François Copé in March 2010 after the regional elections), Éric Woerth, the Minister of Labour, Solidarity and Civil Service, declared in May 2010: “We will not touch the tax shield”. Although this sentiment was reiterated several times by the president, the shield’s legitimacy was finally, albeit cautiously, called into question in the fall of 2010. Laying the groundwork for a policy reversal, the prime minister also asserted that the shield’s elimination was not a political “taboo” (October 2010), after which the new Finance Minister, François Baroin, declared in an interview that the shield had become “a symbol of injustice”. The rapid deterioration of public finances no doubt played an important role in the government’s ultimate decision. Between 2007 and 2010, public debt had increased twenty points with regard to GDP, while the shield’s cost had almost trebled (from 229 to 636 million euros). During that same period of time and especially after 2009, the benefits of the shield were reaped by a smaller and smaller number of wealthy taxpayers. The symbolic value of the shield continued to fade as its financial cost rose (despite its rather moderate price tag compared with other cornerstones of 2007’s tax reform package, such as the exemption of overtime from income tax).

55Abolishing the tax shield took several months: the first doubts were expressed in the fall of 2010; followed by the preparation of a large-scale reform of taxes on wealth in the spring of 2011; then the prime minister’s official announcement in March 2011 of the suppression of the tax shield during a conference organised in Bercy on the theme “Taxes and wealth”. This exit strategy sought to minimise the measure’s importance by incorporating it into a larger reform perspective: eliminating the tax shield should not be seen as a form of failure or resignation, but rather the counterpart to ISF relief, where thresholds were raised and rates lowered. Putting an end to the tax shield was presented as a non-event, overshadowed by a wider reform that had not been able to be carried out in 2005 or 2007. In an attempt to avoid altering the symbol of the ISF, the parliamentary right thus created another symbol, whose ultimate political discredit allowed for a reform of wealth taxes in 2011, rather than the pure and simple elimination of the ISF.


57* *

58Ultimately, how should we interpret the adoption and eventual rejection of the tax shield from the point of view of analysing how public issues are constructed? The realisation that a tax measure targeting a tiny minority of individuals, among a population’s richest, could be widely supported by a country as a whole, has been the subject of two different kinds of interpretations in American research. Some scholars have suggested that democratic debate has been commandeered by lobbies, think tanks and the media, all of whom are suspected of manipulating public opinion and emphasising myths rather than an objective analysis of tax transfers. [67] The adoption of measures heavily favouring the rich would, in this perspective, be symptomatic of the dysfunction of our democratic institutions, controlled by elites mostly concerned with protecting their own interests. [68] According to this hypothesis, citizens should be granted greater access to information, in order to have them perceive the non-egalitarian nature of measures such as getting rid of the estate tax or creating a tax shield. Other researchers have refuted the idea that support for inequitable policies stems solely from individuals’ inability to be truly and fully informed: a measure which benefits a minority of already well-off individuals could in fact be heartily supported by people who are voting less for their own personal interests, and more in accordance with their values and their vision of the future. The wave of popular disaffection regarding the redistribution of wealth could thus be explained less in terms of the elites’ manipulation, than in terms of a profound shift in the population’s dominant values. It would thus not be sufficient to present the statistical proof of the extreme concentration of such and such a fiscal reform to hinder its popularity; one would have to utilise stories, symbols and values to reveal the immorality of a specific measure.

59The tax shield is situated at the intersection between these two competing interpretations; it thus compels us to incorporate a dynamic facet into our analysis of how public issues are constructed. Initially, the reform was widely accepted, according to opinion polls (and despite the absence of hard data concerning its application). Public support relied on a relatable story (the farmers of the Île de Ré), claims of expertise (examples provided from Italy and Germany), and statistical data (the ministry’s predictions). These three elements which contextualised the tax shield all had one thing in common: they concealed the socio-economic status of its future beneficiaries. Political leaders were able to maintain this veil of secrecy by presenting the tax administration’s approximate predictions as hard facts. Consequently, taxpayers could simultaneously support the tax shield and the idea that the wealthy should pay more taxes. [69] If symbols can only be refuted with difficulty, in the sense that their credibility relies on largely incontestable narratives, the same cannot be said of statistics, depending on whether we are speaking about predictions or proven figures. The government’s initial choice to construe the tax shield as a political symbol by using statistical arguments ultimately highlighted the discrepancies between the ex ante projections provided by the Ministry of Finance and the ex post figures available after its application. The visibility of transfers benefiting the wealthiest was accentuated by the refund mechanism, which in turn lent credibility to the theme of reverse distribution.

60This new statistical formulation, far removed from the initial projections, was not however enough to single-handedly invalidate the legitimacy of this public policy. In order for the media’s narrative to effectively reverse the symbolic order that it had helped to construct, a third step was needed: elaborating an alternative narrative, with its own symbols and leading figures, to give a face to this new statistical argument. This reversal occurred simultaneously with the Bettencourt scandal, substituting a new story for an old one and identifying the people who were truly benefiting from the tax shield, whose socioeconomic characteristics were at antipodes with those of the poor victims the measure was originally designed to protect. The framing of the debate was thus shifted: the presumed intention of the tax shield’s creators that it should be used as a measure of fiscal justice seemed to fade away, and instead the public began to see “clientelist” tendencies. The tax shield now looked more and more like a “loophole” advantageous to a few individual interests. Both the strength and the fragility of political symbols can be observed here; they are simultaneously resources for constructing public issues but can also become sources of stigma when reality proves to deviate too much from the stories imagined. It is this dynamic and ever-evolving aspect, founded on the confrontation between imagined narratives and the statistics that are supposed to grant them objective status, which allows us to understand the underlying paradox inherent to the tax shield. More generally, this case study shows just how essential the manipulation of stories and symbols has become in defining public policies, including those which would initially appear to be the most dependent on statistical reasoning. Far from being neutral and dispassionate instruments of public debate, numbers are themselves also a part of the narratives proffered by political players. [70]


  • [1]
    For a review of the existing literature, see Daniel Cefaï, “Les cadres de l’action collective. Définition et problème”, in Daniel Cefaï, Dany Trom (eds), Les formes de l’action collective. Mobilisations dans des arènes publiques (Paris: Éditions de l’EHESS, 2001), 51-98.
  • [2]
    For a more in-depth critique of this approach, see Fabien Desage, Jérôme Godard, “Désenchantement idéologique et réenchantement mythique des politiques locales”, Revue française de science politique, 55(4), 2005, 633-61.
  • [3]
    John Roemer, “Why the poor do not expropriate the rich: an old argument in new garb”, Journal of Public Economics, 70, 1998, 399-424; Isaac Martin, “Redistributing toward the rich: strategic policy crafting in the campaign to repeal the Sixteenth Amendment, 1938-1958”, American Journal of Sociology, 116(1), 2009, 1-52; Paul Pierson, Dismantling the Welfare State. Reagan, Thatcher, and the Politics of Retrenchment (Cambridge: Cambridge University Press, 1994).Online
  • [4]
    Jakob Hacker, Paul Pierson, Winner-Take-All Politics. How Washington Made the Rich Richer and Turned Its Back on the Middle Class (New York: Simon & Schuster, 2010).
  • [5]
    Larry M. Bartels, “Homer gets a tax cut: inequality and public policy in the American mind”, Perspectives on Politics, 3, 2005, 15-31.Online
  • [6]
    Thomas Frank, What’s the Matter With Kansas? How Conservatives Won the Heart of America (New York: Metropolitan Books, 2004).
  • [7]
    Michael J. Graetz, Ian Shapiro, Death by a Thousand Cuts. The Fight over Taxing Inherited Wealth (Princeton: Princeton University Press, 2005).
  • [8]
    Patrick Hassenteufel, Andy Smith, “Essoufflement ou second souffle? L’analyse des politiques publiques ‘à la française’”, Revue française de science politique, 52(1), 2002, 53-73 (57).
  • [9]
    Several polls published in 2010 reported a growing rejection of the measure among public opinion. 67% of those polled favoured the elimination or suspension of the measure in March 2010 (CSA poll conducted on 31 March 2010, cf. press release from the Agence France Presse on 1 April); this number rose to 71% (in favour of elimination plain and simple) in October of the same year (BVA poll conducted on 5 and 6 October 2010, cf. press release from the Agence France Presse on 8 October 2010).Online
  • [10]
    James Q. Wilson, American Government. Institutions and Policies (Lexington: D. C. Heath & Company, 1989 [1st edn 1980]), chapters 15 and 22.
  • [11]
    More recently, see the analyses of Pepper D. Culpepper in Quiet Politics and Business Power. Corporate Control in Europe and Japan (Cambridge: Cambridge University Press, 2011).
  • [12]
    J. Q. Wilson, American Government, 427. Further on (page 441) he adds: “A political conflict is in large measure a struggle to […] alter perceptions and beliefs”.
  • [13]
    M. J. Graetz, I. Shapiro, Death by a Thousand Cuts; George Lakoff, Don’t Think of an Elephant! Know Your Values and Frame the Debate (White River Junction: Chelsea Green Publishing, 2004), xiii.
  • [14]
    Thomas Piketty, Gilles Postel-Vinay, Jean-Laurent Rosenthal, “Wealth concentration in a developing economy: Paris and France, 1807-1994”, American Economic Review, 96(1), 2006, 236-56. In 1913, the richest 1% of French people owned more than 55% of the total inherited wealth. This number has greatly decreased over the course of the twentieth century and was only 21% in 1994.
  • [15]
    Nicolas Delalande, Les batailles de l’impôt. Consentement et résistances de 1789 à nos jours (Paris: Seuil, 2011).Online
  • [16]
    See the Senate’s session on 24 November 1981 (Journal officiel de la République française, 3069).
  • [17]
    Philippe Marini, “L’impôt de solidarité sur la fortune: éléments d’analyse économique pour une réforme de la fiscalité patrimoniale”, Information report no 351 drafted for the Senate Finance Committee, annexed to the minutes of the proceedings on 16 June 2004.
  • [18]
    See Philippe Auberger’s comments condemning a “confiscatory tax system, resulting in relocations of capital, investments and jobs”, National Assembly, Parliamentary Debates, 1st session of 20 October 2004.
  • [19]
    Ferdinando Targetti, “Due anni di politica economica”, Il Mulino. Rivista bimestrale di politica et di cultura, 2, March-April 2003, 283-94.
  • [20]
    Isaac Martin, The Permanent Tax Revolt. How the Property Tax Transformed American Politics (Stanford: Stanford University Press, 2008).
  • [21]
    This position was reaffirmed in decision number 2010-99 QPC, from 11 February 2011 (consulted online on the website of the Constitutional Court, 22 November 2012).
  • [22]
    For example, Jean-Pierre Brard, a member of parliament with Communist affiliations, stated that according to the Tax Board, the number of those subject to the ISF and choosing to relocate for financial reasons remained insignificant (National Assembly, Parliamentary Debates, 19 October 2004, 3rd session).
  • [23]
    Céline Barthon, L’île de Ré (Plomelin: Éditions Palantines, 2005).
  • [24]
    Annie Collovald, Brigitte Gaïti, “Des causes qui ‘parlent’”, Politix, 4(16), 1991, 7-22.
  • [25]
    Cf. François Malye, “Enquête sur les millionnaires de l’île de Ré”, Le Point, 8 September 2005.
  • [26]
    “Ré, île de millionnaires sans le sou”, Libération, 5 April 2005.
  • [27]
    “Impôt non exonéré d’effets pervers”, L’Expansion, 28 September 2005.
  • [28]
    TF1 news show hosted by Jean-Pierre Pernault, 3 May 2005, 1:17pm (INA Archives).
  • [29]
    Le Point, 8 September 2005.
  • [30]
    C dans l’air, 22 September 2005, 6:33pm (INA Archives).
  • [31]
    According to narrative theory, the coda is the retrospective evaluation – or moral – of the story which allows the reader to move from the narrative “there and then” to the “here and now” of the present moment. See Jérôme Bruner, Pourquoi nous racontons-nous des histoires? Le récit au fondement de la culture et de l’identité individuelle (Paris: Retz, 2002), 33.
  • [32]
    On a tout essayé, France 2, 24 March 2005, 7:11pm (INA Archives).
  • [33]
    Pierre Bourdieu, Langage et pouvoir symbolique (Paris: Seuil, 2001), 210.
  • [34]
    Joseph Gusfield, La culture des problèmes publics. L’alcool au volant: la production d’un ordre symbolique (Paris: Economica, 2009 [1st American edition: 1981]).
  • [35]
    David A. Rochefort, Richard W. Cobb, The Politics of Problem Definition. Shaping the Policy Agenda (Lawrence: University Press of Kansas, 1994).
  • [36]
    National Assembly, proposal for law number 2268 concerning the suppression of primary residences from the ISF’s purview, lodged 14 April 2005.
  • [37]
    Tino Sanandaji, Björn Wallace, “Fiscal illusion and fiscal obfuscation: an empirical study of tax perception in Sweden”, IFN Working Paper, 837, 2010.
  • [38]
    Speech by Dominique Tian, UMP MP from Bouches-du-Rhône, National Assembly, Parliamentary Debates, 19 October 2005, 2nd session.
  • [39]
    Speech by François Guillaume, UMP MP from Meurthe-et-Moselle, National Assembly, Parliamentary Debates, 19 October 2005, 2nd session.
  • [40]
    He cited a study done by the Institut CSA without giving an exact reference, cf. National Assembly, Parliamentary Debates, 16 November 2005, 2nd session.
  • [41]
    National Assembly, Parliamentary Debates, 16 November 2005, 2nd session.
  • [42]
    National Assembly, Parliamentary Debates, 16 November 2005, 2nd session.
  • [43]
    Speech given by Nicolas Sarkozy at the Porte de Versailles, 14 January 2007.
  • [44]
    According to the poll conducted by TNS Sofres-Fondation Jean Jaurès-Le Nouvel Obs from 29 March 2007, 60% of those polled said they were in favour of the tax shield being lowered to 50% (53% of left-wing voters and 76% of right-wing voters). On the use of polls, see Érik Neveu, “Les politiques de communication sous la présidence de Nicolas Sarkozy”, in Jacques de Maillard, Yves Surel (eds), Les politiques publiques sous Sarkozy (Paris: Presses de Sciences Po, 2012), 47-69.
  • [45]
    Larry Bartels, Unequal Democracy. The Political Economy of the New Gilded Age (Princeton: Princeton University Press, 2008).
  • [46]
    Interview with a top-ranking civil servant with the General Directorate of Public Finances (tax division).
  • [47]
    By comparison, the study of taxpayer behaviour has been a fairly developed field in the United States for several decades now. One of the subject’s leading specialists is the economist Joel Slemrod, who edited Why People Pay Taxes. Tax Compliance and Enforcement (Ann Arbor, University of Michigan Press, 1992) and Behavioral Public Finance (New York, Russell Sage Foundation, 2006). The Office of Tax Policy Research at the Michigan Ross School of Business (Slemrod is its director) officially aims to produce and disseminate expertise for political leaders, administrations and companies.
  • [48]
    Éric Woerth’s interview in Le Monde, 25 July 2007.
  • [49]
    Here, the issue with the tax shield can be linked to the practice of “non-recourse” observed in relation to social security. In particular, see Philippe Warin, “Non-demand for social rights: a new challenge for social action in France”, Journal of Poverty and Social Justice, 20(1), 2012, 41-53.
  • [50]
    Interview with a high-ranking official in the General Directorate of Public Finances (tax division).
  • [51]
    This argument came out of the interviews conducted by two journalists with several different members of parliament who were specialists on economic and financial issues: Thomas Bronnec, Laurent Farques, Bercy, au cœur du pouvoir. Enquête sur le ministère des Finances (Paris: Denoël, 2011).
  • [52]
    “Bouclier fiscal: les bénéficiaires n’auront bientôt plus à réclamer leur dû au fisc”, Le Figaro, 29 August 2008.
  • [53]
    “Sarkozy sermonne Lagarde”, L’Express, 11 September 2008.
  • [54]
    Interview with a high-ranking official in the General Directorate of Public Finances (tax division).
  • [55]
    Philippe Bezes, Alexandre Siné (eds), Gouverner (par) les finances publiques (Paris: Presses de Sciences Po, 2011).
  • [56]
    Nicolas Delalande, Alexis Spire, Histoire sociale de l’impôt (Paris: La Découverte, 2010).
  • [57]
    Katia Weidenfeld, À l’ombre des niches fiscales (Paris: Economica, 2011).
  • [58]
    Article 1 stipulates that “direct taxes paid by a taxpayer cannot be more than 50% of his income”.
  • [59]
    Speech given by Christine Lagarde, Minister of the Economy, Finances and Employment in the Senate, 25 July 2007. Online
  • [60]
    Alexis Spire, Faibles et puissants face à l’impôt (Paris: Liber/Raisons d’agir, 2012). Online
  • [61]
    Interview with a high-ranking official in the General Directorate of Public Finances (tax division).
  • [62]
    Cf. Pierre Bourdieu, “La force du droit”, Actes de la recherche en sciences sociales, 64, 1986, 3-19.
  • [63]
    Monica Prasad, Yingying Deng, “Taxation and the worlds of welfare”, Socioeconomic Review, 7(3), 2009, 431-57.
  • [64]
    Olivier Morlighem, “L’agriculteur de l’île de Ré et le bouclier fiscal”, La Tribune, 31 March 2006.
  • [65]
    “The tax shield can lead, in certain cases, to a situation where those liable, perhaps after having reorganised their assets and income, are de facto exempted from not only the ISF, but also other direct taxes like property and local taxes. If new levies on wealth were to be introduced, the question of the tax shield’s level should be posed” (“Le patrimoine des ménages”, Report issued by the Conseil des prélèvements obligatoires, March 2009, 273).
  • [66]
    Luc Boltanski, Élisabeth Claverie, Nicolas Offenstadt, Stéphane Van Damme (eds), Scandales, affaires et grandes causes. De Socrate à Pinochet (Paris: Stock, 2007).
  • [67]
    Martin Gilens, Affluence & Influence. Economic Inequality and Political Power in America (Princeton: Princeton University Press, 2012).
  • [68]
    Jakob Hacker, Paul Pierson, “Abandoning the middle: the Bush tax cuts and the limits of democratic control”, Perspectives on Politics, 3(1), 2005, 33-53.
  • [69]
    The survey titled “Perception des inégalités et sentiment de justice” (Perceptions of inequality and feelings of fairness) conducted in fall 2009 showed that 65% of those polled felt that those with high incomes should be taxed at a higher rate. See Michel Forsé, Olivier Galland (eds), Les Français face aux inégalités et à la justice sociale (Paris: Armand Colin, 2011).
  • [70]
    We would like to thank Philippe Bezes, Lucas Guffanti and the participants of the general seminar held by Sciences Po’s Centre d’études européennes on 12 June 2012, as well as the journal’s anonymous readers for their comments on previous versions of this article.

Political science has been struggling for a number of years with a democratic puzzle: how is it that upwardly-redistributive policies, which grant significant tax cuts to rich people, seem to be supported by a majority of voters? Is this due to misinformation, political obfuscation, or a shift in the moral attitudes towards taxation and redistribution? Through an analysis of the French ‘bouclier fiscal’ (tax shield), which was created in 2005 and abolished in 2011, we stress the major role played by symbols, narratives and numbers in the evolving representations of tax inequality. Initially conceived as a tool that would protect all taxpayers from excessive taxation, this measure became increasingly perceived as embodying client politics, especially after 2009-2010. We argue that symbols, when used in politics, may help frame public issues, but also lead to political weakness when reality appears to be too distanced from constructed narratives.

Nicolas Delalande
Nicolas Delalande is a research fellow at Sciences Po’s Centre d’histoire. His publications include: Les batailles de l’impôt. Consentement et résistances de 1789 à nos jours (Paris: Seuil, 2011); “L’économie politique des réformes fiscales: une analyse historique”, Revue de l’OFCE, 122, 2012, 35-59; and “Quand l’État mendie: la contribution volontaire de 1926”, Genèses. Sciences sociales et histoire, 80, 2010, 27-48. His research focuses on the sociopolitical history of the state, the history of the political economy and the different forms of wealth redistribution (Centre d’histoire de Sciences Po, 56 rue Jacob, 75006 Paris).
Alexis Spire
Alexis Spire is a research director at the CNRS and a member of the Centre d’études et de recherches administratives politiques et sociales (CERAPS, Université de Lille 2). After having worked on immigration policies (Étrangers à la carte (Paris: Grasset, 2005); Accueillir ou reconduire (Paris: Raisons d’agir, 2008)), he now currently works on the power wielded by tax administrations and social relations with regard to taxes. On these topics, he has published the following works: Faibles et puissants face à l’impôt (Paris: Liber/Raisons d’agir, 2012); “La domestication de l’impôt par les catégories dominantes”, Actes de la recherche en sciences sociales, 190, 2011, 58-71; and (with Nicolas Delalande) Histoire sociale de l’impôt (Paris: La Découverte, 2010) (Université de Lille 2, CERAPS, 1 place Déliot, 62000 Lille).
Translated from French by 
Sarah-Louise Raillard
Uploaded on on 03/03/2014
Distribution électronique pour Presses de Sciences Po © Presses de Sciences Po. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait