1The French proverb “something promised, something owed” clearly expresses moral expectations about promises shared by every human society.  [1] As a language act relating to a future action, the promise constitutes the very principle of morality, to the extent that there would be “no ethics without the promise.”  [2] The promise is, in particular, fundamental to democracy, as shown by the extracts from electoral manifestos cited here as an epigraph. The functioning of representative democracy is effectively built on the connection between electoral pledges and governmental policies. The act of defining alternative policies to be submitted to a vote, followed by the enactment of these policies by elected representatives in the form of actual public policies, are central to the legitimacy of representative democratic systems. In a famous article, Jane Mansbridge observes, for example, that the traditional model, which she terms “promissory representation,” “focused on the idea that during campaigns representatives made promises to constituents, which they then kept or failed to keep.”  [3] Representatives are considered accountable, and must bear the consequences for their promises in front of those who elected them.

2This principle is closely connected to that of the “mandate” citizens give their representatives to fulfill a defined mission.  [4] Whereas liberal democracies have proscribed the use of the imperative mandate—which requires that the elected person cannot deviate from the action for which they were appointed—the representative mandate constitutes a privileged normative foundation for legitimizing a political decision.  [5] Electoral pledges are presented to citizens as a means of real control over the political decisions to come, and so elicit high hopes, and correspondingly extreme disappointment when they are not kept: in many democracies, opinion polls reveal a climate of general skepticism towards commitments made in electoral campaigns—and election manifestos in particular are seen as no more than a marketing tool, or even a “sham.”  [6]

3Even if it is often only implicit, skepticism towards electoral pledges is also shared by sociologists of public action, since this field of research tends either to elude the dimension of politics outright, or to conclude that it has only marginal importance.  [7] It has been suggested, however, that reflection on the interaction between politics and policies may be able to breathe new life into the study of public policies.  [8] The dimension of political competition also appears in the well-known concept of the “political stream,” in which the meeting of the stream of problems and the stream of policies leads to the opening of a window of opportunity.  [9] But political streams are conceptualized only vaguely in the writings of John Kingdon himself, which explains why it has been examined above all from the perspective of the personal logic of public decision-making, through the concept of “leadership”  [10] or the sociology of ruling elites.  [11] Only a small amount of public policy research has explicitly addressed the specific role of political parties or elections in the manufacture of public policies.  [12] For most authors specializing in this field, the mechanisms of competition between political parties are largely outweighed by other parameters such as path dependence,  [13] external shocks,  [14] governments’ limited resources,  [15] international constraints,  [16] epistemic communities,  [17] and parallel channels of interest representation.  [18]

4The constraints limiting the maneuvering room available for enacting political alternatives is also visible in the discourse of political actors skilled in strategies of “blame avoidance.”  [19] The shift towards a political world in which it has become “difficult” to keep promises may be one of the main causes for the crisis in trust in contemporary democracies, which results in rising abstention rates and increased distrust of political elites on the part of citizens.  [20] Several authors believe that an excessive disconnect between electoral pledges and public policies could undermine the legitimacy of representative democracy.  [21]

5It is therefore crucial for anyone with an interest in the legitimacy of contemporary representative systems to study the fate of electoral pledges. These promises are important objects both normatively (they specify the perimeter of the representative mandate) and empirically (they help citizens evaluate the quality of a candidate or a government). To analyze these issues empirically, we must study public policies through a specific prism: promises made during electoral campaigns. This also requires us to relate the concepts and tools of the analysis of public action to the literature on the question “do parties matter?” more typically associated with political sociology. This is the aim of this special issue.

6Motivated by the tensions between normative conceptions of political responsibility and their practical dynamics, the present issue aims to demonstrate the factors and mechanisms that determine whether electoral pledges are respected. To this end, we set out a conditional model for promise-keeping, summarized by the following general hypothesis: an electoral pledge is more likely to be fulfilled if the executive branch has both the incentive and the capacity to do so. In order to evaluate the relevance and validity of this conditional hypothesis, this special issue proceeds from the results of the ANR Partipol research project.  [22] The contributions analyze five cases of electoral pledges in France, chosen (as we explain below) because of the divergence between their trajectory and the expectations derived from the existing literature: the “full stop” to privatization in the manifesto of the Parti socialiste (PS) (Socialist Party) for the legislative elections of 1997 (article by Rafaël Cos); the 25% increase in the minimum vieillesse (old-age minimum income) (Isabelle Guinaudeau and Olivier Costa) and legislation granting more autonomy to universities (Isabelle Guinaudeau and Sabine Saurugger), both policies announced by Nicolas Sarkozy in the presidential campaign of 2007; the closure of the Fessenheim nuclear power station (Eva Deront, Aurélien Evrard, and Simon Persico) and the opening of marriage and adoption to same-sex couples (Stéphanie Abrial and Simon Persico), both part of François Hollande’s 60 engagements pour la France (60 Commitments for France).

7This introductory article first addresses the limitations of the existing literature on the influence of elections on public action, including the work of specialists focused on promises, and then presents the conditional model for whether promises are respected, which structures the whole project. We then explain how we selected the case studies to evaluate the explanatory power and limitations of this model, before presenting the contributions to this special issue and noting interpretations that recurred across all of the cases studied.

Different ways of understanding how elections shape public policies

8Much work in political sociology and comparative politics has attempted to determine the influence of competition between political parties on public policies, which turns around a central question: do parties matter? To answer this, three successive waves of research have addressed different aspects of the link between inter-party competition and public policies: the first aimed to measure the impact of a change of government on policies enacted; the second aimed to evaluate the correspondence between the issues presented in manifestos and the public policy agenda; and the third examined the fate of electoral pledges.

9The first wave focused on the effect that a change in the party in power had on public policy. Primarily, this literature explored in greater detail the idea that such changes would lead to shifts in the policies enacted. With regard to public finances—the policies that have been most studied  [23]—left-wing governments do not spend significantly more than conservative ones, contrary to the expectations created by a conception of a left-right division centered on the state’s role in the economy. Other economic indicators, such as the balance of trade, unemployment, or inflation,  [24] also have only a limited correlation with the party in power. Similarly, more sectoral public policies such as foreign affairs  [25] or the environment  [26] remain relatively unaffected by the particular party in government. The major indicators studied—the public deficit, unemployment, the proportion of the budget devoted to defense or education, etc.—are ultimately fairly stable, and their development does not seem to be clearly influenced by changes of party control.  [27]

10Starting from this observation, a second wave tried to go beyond the overly simplistic conceptualization and operationalization of “parties,” which involved an implicitly static view of representation, with parties representing fixed interests typically involving an opposition between “left” and “right.”  [28] Rather than analyze the effects of “parties,” authors focused instead on “manifestos,” which can vary over time for a given party. This resulted in the production of large databases that encode the thematic content of manifestos, such as the Comparative Manifesto Project (CMP) and the Comparative Agendas Project (CAP). The hypothesis shared by these studies is that, in competition between parties and between public policies, a high degree of differentiation exists between parties that are keen to reinforce and capitalize on their credibility in a given sector—that is, their “issue ownership.”  [29] Social democratic parties, for example, could be expected to make a priority of defending the welfare state and extending rights—and act accordingly once in power—while conservative parties may favor policies related to sovereignty or ones that would be advantageous to businesses. Here again, the results are unclear, or at least variable, between those studies that identify a correlation between issue emphasis in manifestos and budgetary (or legislative) priorities,  [30] those that contest this result on grounds of methodology  [31] or substance,  [32] and those that conclude that there is a conditional effect dependent on the institutional capacity of the government,  [33] the type of public policy, and the timing within the electoral cycle.  [34]

11Although these approaches represent advances in incorporating the major transformations of parties and manifestos over the last few decades, major limitations remain: the convergence of electoral and governmental priorities suggests that elections shape public policies but is still far from proving this, since no precise information is provided on what was promised and what was actually carried out. The limitations of an approach centered exclusively on the visibility of issues are all the more significant as the priorities asserted during a campaign by the different parties tend empirically to overlap  [35] and therefore do not necessarily offer a relevant index of differentiation between parties.

12A more recent wave of studies has therefore focused on the fate of the commitments presented to the electorate. This approach makes it possible to consider the development of electoral offers in more precise terms: it involves studying the proposals for change that are explicitly formulated for the attention of voters, and their realization (or abandonment) in public policy. It also enables us to reposition the study of the connection between party competition and public policy in its national and historical context: a promise for a given change in public policies will have a different meaning depending on the period or country in which it is formulated. Above all, a promise-based approach makes it possible to evaluate more directly the congruence between promises formulated during the electoral campaign and what is carried out in government.  [36] This led to the emergence, in the 2000s and 2010s, of an international research group focused on electoral promises, present in twelve OECD countries: the Comparative Political Pledges Group (CPPG). The group’s aim is no longer to work on manifestos as a whole, but to concentrate on electoral pledges: “a commitment to carry out some action or produce some outcome, where an objective estimation can be made as to whether or not the action was indeed taken or the outcome produced.”  [37] Their results offer a more nuanced vision of the relationship between political competition and public policies: although they observe a general increase in the number of promises made,  [38] they do not confirm the very negative perceptions of electoral promises being quickly forgotten once an election has passed. On the contrary, they tend to validate more clearly than previous approaches the link between elections and public policies. The level of manifesto commitments that are carried out is far from negligible, with an average of 50% and levels of up to 85% in the United Kingdom.  [39]

13This literature encourages the thought that political parties and elections should be better integrated into public policy analysis. But it also shows that, approached from the point of view of promises, the influence of electoral competition is by no means systematic. First, government activity is not limited to the field delimited by electoral manifestos: a government must act in many domains that it will not have anticipated during its electoral campaign. Above all, these studies show that certain promises—a large minority—are not kept, while others are. However, the explanations for these differences provided by the authors of the CPPG have so far been limited. As we shall discuss, these authors have especially shown the influence of institutional factors, demonstrating that the division of power (coalitions, minority governments, etc.) tends to reduce the number of promises kept. Their control variables also suggest that a higher proportion of “non-budgetary” promises are kept.  [40]

14It seems we can and must extend the focus considerably: although elections and manifestos shape public policies only in certain cases, we must characterize these cases, beyond simple institutional and budgetary configurations. Rather than simply evaluating the correlation between promises and policies, we should identify the conditions in which this is established: why are certain promises respected more than others? How do “constraints” emerge and how much freedom do elected representatives have? Can we identify factors that shape how the heads of the executive branch use this room for maneuver to carry out their mandate? This special issue explores these questions by mobilizing a renewed conceptual and methodological framework that will allow us to extend the conclusions emerging from the macro-statistical approach that has dominated until now in the analysis of electoral pledges.

Studying the conditional influence of electoral pledges on public policies

15Several elements can limit or, conversely, increase the influence of electoral commitments on the policies carried out. It is indispensable to identify these conditions if we are to understand the variations observed in the literature. Although several authors have called for the development and testing of a conditional model of promise-keeping,  [41] attempts to do this have so far focused on a limited number of factors, mainly institutional ones. Party manifestos are situated at the juncture of elections and public action, and it therefore seems relevant to build once again on electoral studies, which enables us in particular to consider the incentives for promising and realizing certain measures, and on the sociology of public action, which shows that policymakers have limited resources and considerable constraints that weigh upon the construction of policies.

16However, until now, the dialogue between pledge research and these subdisciplines has been limited. This can surely be explained by the fact that this literature has only recently developed—perhaps because of preconceived ideas that electoral manifestos are irrelevant, or because electoral pledges as an object of study existed at the juncture of several sub-disciplines. Analyses of manifestos have long followed established paths, tending to capitalize on existing databases (notably those of the CMP and the CAP), whose creation required considerable resources. The turn towards encoding and analyzing electoral pledges, which involved significant risk and effort, only arose in the context of an increase in the visibility of manifestos and a growing awareness, in the field of electoral sociology, of the limitations of approaches centered on the salience of issues, given the observation that electoral priorities are largely convergent.  [42] Besides the recent emergence of electoral pledge analysis, its limited connections with such structurally important fields of political science as the sociology of public action and electoral studies can undoubtedly be explained by the difficulty—if not impossibility—of integrating sufficiently synthetic and relevant information in large longitudinal and comparative databases, whose construction remains the main aim of the CPPG.

17This dialogue seems to us particularly promising for enlarging the conceptualization of the “constraints” elected representatives are subject to, but also for taking account of the negotiations that take place within the available room for maneuver. Conceptually, our innovation is to introduce into the study of electoral pledges concepts from public policy analyses and the literature on party competition. We have in mind especially the theories of issue-based competition,  [43] the literature on electoral cycles,  [44] Kingdon’s “multiple streams framework” (MSF),  [45] the analysis of the “publics” (clienteles or target populations) of distributive policies,  [46] mechanisms of delegation to unelected agencies,  [47] strategies for blame avoidance,  [48] and the political  [49] and operational  [50] dimensions of the capacity to change a public policy.

18The connections between these perspectives leads us to a central, structuring hypothesis: the outcome of an electoral pledge depends on the capacity of leaders to stand by it, but also on their incentives to use their resources to do so.  [51] It is not uncommon for a single factor to affect both the capacity and incentives to keep one’s word—this could be the case, for example, where the executive branch has limited popularity, which could simultaneously provide an incentive to hold more closely to its manifesto, while also limiting its political capacity to enact reforms. It nonetheless seems heuristic to distinguish between what representatives can do and what they wish to do, since both conditions are logically necessary for a commitment to be kept.

From promises to policies: The workings of promise-keeping (and -breaking)

46This approach, which is more conducive to a dialogue between political sociology and the analysis of public policies, enables us to reveal the causal sequences, actors, and mechanisms of influence that statistical studies do not identify. Of course, concentrating on a single case necessarily limits any ambition to create a general rule, and the conclusions of each article need to be tested on a broader scale. Nonetheless, our selection of promises with varied characteristics, formulated by governments with particular ideologies and forms of leadership, elected in different political and socioeconomic contexts, puts us in a strong position to suggest that the types of conditionality that are repeated and invariable between cases play a structuring role across the French context.  [93] As it happens, several observations have indeed emerged in parallel.

47Our case studies, and borrowings from public policy analysis, enable us to qualify the role of capacity, at least as a determining factor. They also enable us to think of this capacity as plural: whether it is institutional, political, or operational, the capacity to act involves different dimensions that are not simply dependent on one another. These different forms of capacity are also variable. They develop over time and, since they are neither given nor fixed, they can be gauged more or less successfully by political actors. The latter can overestimate or underestimate their political, operational, or institutional capacity.

48The different forms of capacity are not analyzed as essential conditions, but as resources whose lack can be compensated for. Accordingly, the articles on the LRU Act and the rise in the minimum vieillesse show that elected Presidents can, in the dynamic of power relations with the civil service and the Finance Ministry, convert the resources provided by their “democratic mandate” or high popularity into political or budgetary room for maneuver. As the literature on electoral cycles suggested, the period immediately following an election seems particularly favorable to the fulfillment of promises, but this freedom diminishes over the course of a term of office or if the economy turns for the worse. Technical devices—using decrees, inscribing the measure in a more flexible legal framework, or instigating the legislative process for the measure based on an inaccurate costing—can all be mobilized to secure the adoption or funding of a promise deemed strategically important. The case of the commitment to close the Fessenheim nuclear power station shows that the obstacles the executive branch encounters can even be instrumentalized in themselves in order to “avoid blame,” while justifying the abandonment of one’s promise.

49On the other hand, the cases of unkept promises often resulted from the difficulty of moving from the electoral sphere to the governmental sphere—in other words, obstacles that were unforeseen (or deliberately ignored)  [94] when the manifesto was written. The studies presented in this special issue emphasize this plurality of conditions, and the difficulty, for those involved in writing manifestos, of precisely evaluating the room for maneuver that would be available while taking into account all these factors. The cost of promised policies can therefore turn out to have been poorly evaluated, as was the case for the increase in the minimum vieillesse in 2008. The article on Hollande’s abandoned attempt to close the Fessenheim nuclear power station shows that he clearly overestimated his institutional capacity in this regard.  [95] Hollande also seems to have overestimated his political capacity to pass legislation opening marriage to same-sex couples. This explains his decision to leave time for debate, which provided the opportunity for more mobilizations against the legislation. Conversely, the LRU Bill, which had been pursued over several years in political and administrative networks, was able to be carried out just two months after the legislative elections.

50Where we find greater room for maneuver to carry out the manifesto than we would expect, this raises the question of the executive’s incentives to keep its word, since it is ultimately the executive that plays an essential role in identifying which priorities for action will receive the most generous resources. These incentives can also develop after an election has passed. Viewed as a whole, the contributions to this special issue bring to light a central factor, which is overlooked in the literature on electoral pledges: the characteristics of the actors they affect, either positively or negatively. As Table 1 shows, whether a promise is respected depends to a large extent on the nature of the target social groups, and the type of party actors involved in this commitment.

51The contributions to this special issue show that the incentives to respect a promise depend first, crucially, on the resources of the beneficiary groups. A promise is more likely to be considered a priority when it benefits a part of the electorate that is both strategically important and viewed positively, as was the case for the retirees with modest incomes targeted by the increase in the minimum vieillesse. This case suggests that the fulfillment of a promise that consists in privileging certain target groups is catalyzed by the approach of elections (either national or intermediate) and by dips in the popularity of the executive branch. Conversely, bringing about equal marriage, which targeted a group that is a social minority, perceived in very different ways by the public and very negatively by some—highly mobilized—parts of the population, turned out to be more difficult than expected.

52Strong incentives can also derive from the mobilization of political activists and influential interest groups, such as the Conference of University Presidents, especially when these actors possess significant resources and act within a (neo-)corporatist system that guarantees they will be involved in every stage of the enactment of public policies. Conversely, when the cost of a promised measure is concentrated on such actors, it can create a major obstacle—as seen in the role played by bosses of large public companies who wanted their capital freed up during Jospin’s term, or the obstacles created by EDF in the Fessenheim case.

53Considerable asymmetries of resources also appear when we examine the promises made by party actors. The incentives to enact proposals from the manifesto are stronger when these proposals have been defended from the beginning by the head of the executive branch, his supporters, and more generally by the majority group of the party. Conversely, incentives are less strong when the promises were made in the context of the electoral campaign as concessions to a minority political grouping or a potential coalition partner. This asymmetry is very clear in the article on the promise to bring an end to the privatization of public companies, formulated in 1997 by the Socialist Party on the insistence of the “Socialist Left,” a party grouping that was marginalized immediately after the election in favor of more “modernizing” elements. Similarly, Hollande made the promise to close Fessenheim nuclear power station only reluctantly, in the context of negotiations for a coalition agreement with EELV, whose representatives had only limited resources after the election to press for a more rapid decision. Conversely, analyzing the increase in the minimum vieillesse, the LRU Act, and same-sex marriage shows how powerful the French president’s tools are—both formal ones (personal powers, mission statements for ministers, mechanisms for controlling their activities, rules for circumventing the parliamentary system) and informal ones (authority over ministers and the civil service, influence on the party, access to the media)—for carrying out the measures that are judged to be a priority.

54More generally, this poses the question of the networks in charge of the drafting of the manifesto and how much this personnel overlaps with those who will belong to the cabinets of the President and ministers after the election. From this point of view, the presidentialization of the parties and political life,  [96] which has given the candidates increasing autonomy relative to their parties, might be expected to increase the rate of promise fulfillment. Indeed, unlike Jospin—who had to negotiate his manifesto with the different groups inside the Socialist Party, and also with the two parties in the pre-electoral coalition—the candidate for a more presidentialized party is relatively free to formulate promises, and to put together a campaign team and later a governmental team, all of which should make it easier to take decisions. The analysis of the LRU Act shows, for example, that the same networks and individuals worked on Nicolas Sarkozy’s presidential manifesto and later, in ministerial cabinets and the Élysée, on the policies that were adopted during his time in office.

55A crucial advantage of the qualitative approach used here is the opportunity it offers to observe how promises are made, constructed, invested, and sometimes abandoned or redefined, how they take on consistency (or fail to do so) over the course of their history, responding to the anticipation of constraints, budgetary compromises, and the permanently shifting room for maneuver for the enactment of the manifesto. Far from being fixed, promises are subject to competing interpretations by citizens, political elites, and administrative actors. This variation in the way these promises are perceived and enacted explains the gulf between the concrete effects of a promise that has been kept, and the effects predicted—as was the case for the increase in the minimum vieillesse, which did not reach all the populations it was supposed to benefit—or the appearance of resistance that was not anticipated—for example, the extension of marriage to same-sex couples provoked a mobilization that the executive branch could undoubtedly have avoided. Finally, the close observation of the sequence of events leading from the formulation of a promise to its realization often requires us to question an excessively linear conception of this process. In fact, rather than a straight trajectory between the field of the electoral campaign and that of public policy, there is a to-and-fro during which certain reflections from technical and expert domains are taken up by political officials.

56Each contribution to this issue therefore demonstrates the richness of promises as a way of viewing the functioning of the French political system. This perspective allows us to test the reality of a foundational principle of representative government: the fulfillment of electoral commitments. Despite the fact that they may fail or give up, our work suggests that political officials usually try to carry out the promises made during the campaign. We need not conclude from this that they possess moral virtues or convictions: they know that they are accountable for their promises in front of the concerned groups, the organizations of civil society, the media, and parties. However much effort officials make, the articles in this issue cast light on the multiple factors affecting promise-keeping: a multidimensional capacity that allows the executive branch to take advantage of the conversion of several types of resource in order to secure the realization of promises, whose relative priority is largely defined in positive or negative terms by the groups targeted.

57These conclusions are also relevant to the public policy literature. Even if the public policies directly associated with an electoral promise are far from the central part of governmental action, this special issue allows us to indicate the conditions in which elections and party actors are able to play a role in the creation of public policies with a particular status. This is particularly the case in prominent domains, giving rise to dynamics of competition between issues, or to substantial mobilization by social groups and interest groups, especially if these are largely perceived positively. From a methodological point of view, finally, our approach shows the cross-fertilizations that can occur between quantitative and qualitative approaches: the selection of cases guided by existing statistical analyses enables new hypotheses that warrant testing at a larger scale.  [97] We hope this special issue casts a useful light on how the machinery of the Fifth Republic guarantees, with varying effectiveness and the occasional malfunction, transmission between electoral promises and public action.  [98]