The central contribution of this book lies in presenting detailed case studies of the politics of taxation, which provide new insights into how and when connections between taxation, responsiveness and accountability have emerged in practice. However, moving from broad propositions about taxation and governance to detailed empirical research at the country level confronts two closely related challenges. First, existing models have focused on broad causal arguments, while saying much less about the specific ways in which bargaining over taxation may play out in practice in contemporary developing countries. Second, there are good reasons to believe that these processes may be frequently difficult to observe, owing to the fact that bargaining over taxation is likely to be frequently “indirect, taking the form of strategic interaction and mutual behaviour adjustment, rather than public haggling and agreements” (Moore Reference Moore, Bräutigam, Fjeldstad and Moore2008: 38).

This chapter is aimed at translating the existing body of theoretical and historical literature into a more specific set of propositions for empirical investigation through extended case study research. This involves defining and operationalizing key concepts and outcomes of interest in order to construct a model linking taxation, responsiveness and accountability through three clearly defined causal processes. This, in turn, provides the foundation for developing a more dynamic model. This dynamic model seeks to capture the importance of strategic bargaining between taxpayers and governments and to highlight the role of contextual factors in shaping bargaining processes and outcomes. The goal is not to capture every possible process through which taxation may shape levels of responsiveness and accountability. Instead, the focus is on translating a comparatively broad, difficult-to-observe, research question into a set of clearly defined and observable empirical possibilities to guide the case study analysis.

With these goals in mind, this chapter is organized in five parts. It begins by revisiting the key lessons that can be drawn from historical experience, as they provide the foundation for the model to follow. The second section develops the basic model, capturing the causal mechanisms and processes that are expected to link taxation to broader changes in responsiveness and accountability. The third section presents a subset of specific operational outcomes of interest, in order to establish a clear standard against which to assess the case study evidence. The fourth section turns to the role of context and highlights five sets of factors that are likely to shape the character and extent of tax bargaining. Finally, the fifth section incorporates these contextual factors into a dynamic model of tax bargaining, which captures the likely diversity and contingency of connections between taxation, responsiveness and accountability.

2.1 The lessons of history: diverse pathways linking taxation and political change

Seminal models of tax bargaining developed by Bates and Lien (Reference Bates and Lien1985) and Levi (Reference Levi1988) are the most common starting point for discussion of the connections between taxation, responsiveness and accountability. In their model, Bates and Lien (Reference Bates and Lien1985) propose that when taxpayers are mobile, and thus able to evade taxation, governments will have incentives to make concessions to those taxpayers to secure their tax compliance. Levi (Reference Levi1988) takes this basic logic a step further, arguing that the ability of a government to collect tax revenue from citizens depends on three factors: coercion, ideology and, crucially, quasi-voluntary compliance. Quasi-voluntary compliance captures the idea that citizens will be more likely to comply with taxation when the government is perceived to be responsive and accountable, and when other citizens comply as well. Within this framework, governments have incentives to become more responsive and accountable to taxpayers so as to increase quasi-voluntary compliance. This may occur through broad-based concessions to the majority of taxpayers, or may be dominated by elite interests, as governments focus on delivering more modest “side-payments” to the most important larger taxpayers.

While these models derive much of their power from their ability to simplify a more complex reality, Moore (Reference Moore, Bräutigam, Fjeldstad and Moore2008) has argued for the need to dig more deeply into the likely complexity and diversity of processes of tax bargaining in contemporary developing countries. He follows earlier models in proposing that tax bargaining refers to “the exchange of tax revenues (for the state) for institutionalized influence over public policy (for citizens),” but then goes on to propose that “bargaining ranges from (a) direct and explicit haggling and agreement to (b) indirect strategic interaction and mutual behaviour adjustment without direct negotiation.” He thus highlights what is relatively implicit in earlier models: while “tax bargaining” is a usefully broad concept, specific instances of tax bargaining are likely to be extremely varied and relatively more complex than the term “bargaining” may evoke. This suggests the value of further exploring the multiple and relatively distinct processes that may underpin the broader concept of tax bargaining in contemporary developing countries, although without entirely abandoning the obvious benefits of parsimony and simplification.

The diverse historical experiences captured in the work of Charles Tilly (Reference Tilly1992) are a valuable starting point for understanding this likely complexity. In simplified form, Tilly argues that in early modern Europe the rising costs of warfare forced states to extract an increasingly heavy tax burden from taxpayers, with significant consequences for the character of states and of political power. Most famously, he argues that in England the need for taxation led to conflicts between taxpayers and the monarchy, eventually resulting in tax bargains that granted taxpayers greater representation and legal protection in exchange for higher taxes. However, while the comparatively straightforward example of tax bargaining in England has become the starting point for much contemporary discussion,Footnote ¹ Tilly's model is significantly more diverse and expansive, and this diversity holds important additional lessons.

While European states faced a common need to expand revenue collection, Tilly argues that differences in the characteristics of individual states led to different revenue-raising strategies and divergent outcomes. He particularly highlights the importance of differences in the extent to which states were characterized by two elements: capital and coercion. Capital-intensive states were home to powerful commercial classes that controlled access to significant and mobile economic resources, while coercion-intensive states were largely agrarian and lacked a large commercial class of significant influence or wealth. This difference was at the root of profoundly different responses to the need for revenue.

In capital-intensive states, the primary challenge was to gain access to the mobile economic resources controlled by capitalists. This was achieved by ceding institutionalized political power to capitalists in exchange for their willingness to contribute tax revenue. Such bargaining was necessary because capitalists were otherwise able to resist taxation by force or, more simply, through tax evasion facilitated by the mobility and secrecy of their assets. The tax bargains that resulted fragmented political power and limited the impetus for the construction of powerful central states. This was exemplified by the city-states of Italy, where councils of merchants and capitalists governed and the central state remained weak.

By contrast, states where capital was weak tended to turn towards coercion to extract a small surplus from the rural poor. These states generally relied on bargains between central states and the landowning elite, according to which the central state granted land and power to elites in exchange for the ability to extract tax revenue from a peasantry that was both immobile and weak. This system of revenue mobilization was coercive and unequal but nonetheless gave rise to powerful central bureaucracies capable of extracting significant revenue. This model extended throughout much of Eastern Europe and Scandinavia, and was exemplified by experiences in Russia, where an increasingly powerful central state gradually extended its coercive power over the landed nobility and the subjugated peasantry.

The most successful and inclusive tax bargains, as in the English case, occurred where the relative power of capitalists and of coercive states was most balanced. The strength of capitalists provided strong incentives for the state to reach a revenue-enhancing compromise, while states were strong enough to establish themselves as the central political actors. However, even such cases of relatively successful and inclusive tax bargains come with important caveats. Most obviously, the extension of responsiveness and accountability was not initially universal, but was restricted to a limited group of elites and large taxpayers. Equally important, there were important differences in the particular pathways leading from taxation to the expansions of responsiveness. This can be seen in a comparison between the English and French cases addressed by Tilly, while the experience of the American Revolution, which gave rise to the oft-quoted revolutionary slogan “no taxation without representation,” offers a third meaningfully distinct pathway.

The experience of seventeenth-century England epitomizes the ideal of relatively explicit tax bargaining, as it involved a comparatively explicit exchange of tax compliance for the expansion of accountability. By the beginning of the seventeenth century, Britain had established a relatively independent court system to defend the rule of law and a parliament with legal control over taxation. Yet, despite the legal status of Parliament and the courts, the ambitions of the monarchy and the escalating costs of war led successive kings to defy both Parliament and the courts by resorting to unilateral revenue-raising strategies, which included new customs charges, forced loans, the sale of monopolies, property seizures and the sale of peerages (North and Weingast Reference North and Weingast1989; Tilly Reference Tilly1992). Over time, these abuses by the Crown contributed to growing anger amongst taxpayers and Parliamentarians and ultimately, to civil war, which began in 1642 and eventually resulted in the overthrow of the monarchy. The decision to reinstall the monarchy in 1660 was taken only on the condition that it “confirmed the power of Parliament within the British state, especially when it came to revenues and expenditures” (Tilly Reference Tilly1992: 157).

Despite this apparent agreement, it was not long before conflicts between Parliament and the monarchy re-emerged, thus contributing to a second overthrow of the Crown in 1688, dubbed the Glorious Revolution. That same year, Parliament elected to offer the Crown to the future King William and Queen Mary, but did so only after another institutional revolution, which strengthened the position of Parliament and the courts and shifted the locus of effective power away from the monarchy. This represented a classic tax bargain, as taxpayers agreed to accept increased tax collection in exchange for the institutionalized expansion of governmental accountability. This tax bargain figured centrally in the flourishing of the British state, as it facilitated an increase in the level and stability of tax revenue, a correspondingly increased capacity to borrow abroad, and the construction of a more sophisticated bureaucratic apparatus (Brewer Reference Brewer1989; Tilly Reference Tilly1992).

A century later in France, the need for tax revenue to meet the escalating costs of warfare and debt repayment similarly sparked a period of political conflict that ended with the establishment of a more capable, responsive and accountable government. However, this outcome lacked the relatively explicit tax bargaining of the English case, instead emerging from revolution and an extended period of political upheaval. By the second half of the eighteenth century, the escalating costs of war had led the French state to the brink of bankruptcy. At the time France relied on an inefficient feudal tax system in which the clergy and nobility largely avoided taxation, while the greatest burden fell on agricultural producers and a growing bourgeoisie. Efforts to increase the share of revenue collected by the central state proved both ineffective and unpopular, and led to increasing conflict between the state and taxpayers. By 1789, the fiscal situation had become sufficiently desperate that the king was compelled to summon the Estates General, comprised of the clergy, the nobility and the more broadly representative Third Estate, in order to debate a new fiscal regime. This meeting failed to yield a tax bargain in which increased revenue was exchanged for greater accountability. Instead, the summoning of the Estates General, which was prompted by fiscal crisis and had not occurred since 1614, catalysed the coalescing of a much wider array of grievances amongst the members of the Third Estate, who had soon declared themselves the new National Assembly.

The initial phase of the French Revolution lasted until Napoleon seized power in 1799 and yielded two new constitutions during a complex period of political upheaval and conflict. By the end of the process, the feudal system of rural relations had been largely eliminated and replaced by a powerful and centralized bureaucratic state, with a much-expanded role for the emerging bourgeoisie. Moreover, while the revolutionary government was overthrown after only a decade, it nonetheless introduced the tradition of republican government and contributed to the long-term expansion of accountability. Thus, the Revolution, which was precipitated by a fiscal crisis, gave rise to dramatic political changes. However, the path from tax conflict to political change did not rely on bargaining in the relatively explicit sense of English experience. Instead, it was persistent and unresolved resistance to taxation that precipitated a national crisis and generated pressure for dramatic change through a long, complex and often violent process (Tilly Reference Tilly1992).

A final, and again dramatically different, example relates to the role of taxation in the genesis of the American Revolution. Throughout the latter half of the eighteenth century there had been a debate in the American colony over whether the British Parliament had the right to impose taxes, with the Americans and their allies arguing that under British law only their own representatives could tax them. During the 1760s, this objection led to the repeal of various taxes, including the stamp tax, but the British maintained a tax on tea. This was both a symbolic assertion of power and a means to raise very modest revenue to pay the salaries of various colonial officials, in an effort to reaffirm their loyalty to Britain. In 1773 the issue came to a head when the British government passed the Tea Act, aimed at strengthening the legal standing of the tax on tea. This prompted the so-called Boston Tea Party during which protesters not only refused to allow a shipload of taxed tea to be unloaded, but also destroyed it by throwing it overboard. While this episode had relatively little practical impact on government revenue, it became a symbolic flashpoint for the independence movement, as it highlighted questions about the autonomy and rights of the colony and became a catalyst for popular mobilization. It thus contributed to spurring the onset of the American Revolutionary War that began in 1775 and culminated in the achievement of formal independence in 1783 (Morgan and Morgan Reference Morgan and Morgan1953; Bailyn Reference Bailyn1967; Adams Reference Adams1998).

Once again, conflict over taxation was an essential component of the eventual expansion of responsive and accountable government, but in a manner unlike that which prevailed in either England or France. The process in Britain, while not without conflict, relied on a relatively explicit tax bargain between taxpayers and the government in the face of a fiscal crisis, while the process in France saw resistance to taxation precipitate a fiscal crisis which, in turn, opened the door to dramatic and tumultuous political changes. By contrast, the tax conflict that helped spark the American Revolution involved neither explicit tax bargaining nor a fiscal crisis. Instead, it revealed the ability of debates about taxation, even when relatively insignificant in simple revenue terms, to catalyse political action around broader political grievances. Meanwhile, all of these examples stand in sharp contrast to experiences elsewhere in Europe, where states either profited through continued and often brutal coercion or, in many other cases, were wiped off the map as a result of their failure to raise revenue effectively.

The overarching message that emerges from even this relatively superficial exploration of historical cases is that the relationship between taxation, responsiveness and accountability is unlikely to be simple and linear. Instead, any relationship is likely to be conditional on contextual factors, and to operate, where it does exist, through multiple causal processes, ranging from explicit tax bargaining to much more indirect processes of political contestation and mobilization. This basic insight frames the model to follow. It ultimately posits three distinct causal processes linking taxation, responsiveness and accountability, and these three processes roughly mirror the differing experiences of Britain, France and the United States. Importantly, the model and the case studies to follow do not insist on a linear relationship, but instead emphasize contingency and the contextual factors that may shape outcomes.

2.2 A model of tax bargaining

Against this backdrop, the remainder of this chapter is devoted to developing a detailed model of tax bargaining, moving progressively from the relatively general and abstract to a more focused operationalization of the variables of interest. This section begins by laying out the core elements of the model, which are presented in two parts. The first part summarizes the broad mechanisms that are likely to link taxation to the expansion of responsiveness and accountability. The second part moves from these broad mechanisms to the identification of three specific causal processes through which taxation may contribute to broader political change in practice. The use of the terms “mechanisms” and “processes” here builds loosely on McAdam, Tarrow, and Tilly (Reference McAdam, Tarrow and Tilly2001: 24–27): mechanisms are defined as specific actions or events that may result from a particular cause across a variety of contexts, while processes are defined as particular sequences of events through which these mechanisms combine with other contextual factors to produce particular outcomes.Footnote ²

2.3 Narrowing the focus and defining specific outcomes

The model developed so far provides a broad framework for analysing the causal processes linking taxation, responsiveness and accountability; the core elements of the model are summarized in Figure 2.1. Within this basic model, efforts by the government to raise tax revenue may prompt taxpayers to engage in either tax resistance or collective action to demand reciprocal concessions. These two causal mechanisms may, in turn, prompt three alternative causal processes leading to gains in responsiveness and accountability: direct tax bargaining, tax resistance and changes in government, or strengthened political capabilities of taxpayers. However, while this basic model provides a starting point for analysis, the research challenge is not only conceptual, it is also practical, as these causal processes may frequently be very difficult to identify in practice. With this challenge in mind, the next step in developing the research strategy lies in more precisely defining and operationalizing the specific processes and outcomes of interest.

Figure 2.1 Taxation and the expansion of responsiveness and acco-untability.

The difficulty of observing tax bargaining in practice can be captured through two simple illustrations, which inform the approach adopted here. As a first example, it is likely to be difficult to observe the long-term impact of taxation on taxpayers’ expectations and levels of political engagement. Some forms of political engagement, such as mass protest, are easy to observe. However, if the experience of paying taxes subtly shifts expectations and levels of engagement over several years, or even decades, those changes will be difficult or impossible to observe using a case study approach.Footnote ⁸ A second example deals with the difficulty of identifying the motives of decision makers. In some cases, tax considerations are likely to be central to explaining government actions, and thus are easy to identify. However, it is equally possible that tax considerations may consistently function at the margin of decision-making, thus appearing politically insignificant in isolation despite a potentially significant aggregate impact across multiple decisions over time.

These challenges of observation make it particularly important to define clear and observable indicators against which to measure the presence or absence of the three causal processes proposed by the model. However, in focusing on comparatively observable processes and outcomes, such a strategy ignores less observable, but still potentially important, processes through which taxation may influence the responsiveness and accountability of governments. The implication is that clear and observable examples of the three causal processes are likely to also be indicative of the existence of the more subtle behavioural changes predicted by the model. That is, the actual importance of taxation in shaping political outcomes likely exceeds that which can be observed effectively using the case study methodology that is the focus here.

2.4 Contextual factors affecting tax bargaining

The preceding discussion provides a picture of the multiple ways in which conflicts over taxation may shape the expansion of responsiveness and accountability, while also translating comparatively broad concepts into a set of more specific and observable outcomes of interest. To complete the model, these alternative causal processes need to be placed in broader context. Whereas the discussion so far has identified multiple possible causal processes, we equally need a framework for thinking about why specific outcomes may emerge in any individual case. That is, we need to understand why in some contexts taxpayers may respond to taxation through collective action, while in others reliance on quieter forms of tax resistance will be more likely. Why in some contexts governments will be strongly inclined to bargain with taxpayers, while in others they are likely to rely heavily on coercion. The goal is not, it should again be emphasized, to present a highly deterministic model: processes of tax bargaining are expected to be embedded within broader processes of political contestation and change, and thus subject to significant uncertainty. The goal is instead to highlight broad factors that are, on average, likely to drive distinctive patterns over time and across diverse contexts.

In seeking to identify relevant contextual factors it is useful to again turn to historical accounts of tax bargaining, in which contextual factors play a prominent role. This is most explicit in Tilly's (Reference Tilly1992) account of taxation and state-building in early modern Europe. In his model, it is the balance between coercion and capital in individual states that is essential to determining whether the need for taxation results in consensual tax bargaining or the expansion of coercion, while he also focuses attention on the importance of factors shaping the capacity for collective action amongst taxpayers. Looking to similar historical cases, Levi (Reference Levi1988) draws particular attention to the importance of political institutions in facilitating tax bargaining and shaping outcomes. Moore (Reference Moore, Bräutigam, Fjeldstad and Moore2008: 51) highlights similar issues, including the importance of government time horizons, access to non-tax revenue, factors conducive to collective action, economic structure and the level of institutionalization of the state. Finally, a long tradition in political science has highlighted the possibility that different tax types may produce divergent incentives and outcomes (Martin and Gabay Reference Martin and Gabay2007).

Given the extent to which the outcomes of tax bargaining are likely to be intertwined with broader political processes, the range of potentially important contextual variables is enormous. As such, a core analytical challenge lies in simplifying this comparatively complex reality within a manageable analytical framework. To do so, the analysis is organized around a set of five structural factors that are expected to shape the extent and character of tax bargaining through relatively distinct pathways. These structural factors are, in turn, expected to be shaped by a wide range of more specific and idiosyncratic contextual variables within individual cases.

Thus, for example, the extent of taxpayer capacity for collective action is a structural factor expected to shape the nature of tax bargaining by influencing the ability of taxpayers to become collectively mobilized and to make effective demands on government. This structural factor is, in turn, expected to be shaped by a wide range of more case-specific contextual variables, including the extent of political openness or repression, the strength of elite leadership and business associations, the unity of taxpayer interests, the strength of pre-existing civil society and the nature of expectations.

Table 2.1 captures the five structural factors that are the focus here, summarizes the pathway through which each is expected to shape the extent and character of tax bargaining and provides an illustrative list of more specific contextual variables that are expected to shape these structural factors in individual cases. Notably, some contextual variables appear in the table more than once, with the strength of business elites, for example, shaping both taxpayer capacity for collective action, and the scope for tax resistance. This overlap points to the potential for individual variables to shape tax bargaining through multiple pathways.

The appeal of this focus on structural factors is twofold. First, relying on categories tied explicitly to the political logic of tax bargaining allows the role of contextual factors to be incorporated explicitly within the broader model. That is, there is a clear and explicit link between each structural factor and the mechanisms and causal processes defined so far. Second, it provides a relatively parsimonious and portable framework that can accommodate the particular features of individual cases. A focus on more specific contextual variables, in the absence of this broader framework, would result in the potentially limitless proliferation of relevant variables in response to the idiosyncratic features of individual cases. By contrast, a focus on these structural factors captures the intuition that a large and diverse set of contextual variables nonetheless are likely to shape tax bargaining in a relatively standard set of. With this overarching framework in mind, the following sections describe each structural factor in greater detail.

2.5 Incorporating context and strategic interaction in the model

As noted previously, a primary advantage of an analytical focus on these structural factors is that it is then possible to incorporate them explicitly into the basic causal model developed earlier in this chapter. Doing so produces a more dynamic model in which specific outcomes emerge from particular strategic choices by key stakeholders, which are in turn shaped by broader contextual factors. This reflects Moore's (Reference Moore2007: 16) observation, noted earlier, that “any realistic understanding of these processes requires that we go beyond the initial reactions to taxation of each type of actor, and take into account the ways in which they then interact, whether conflictually, cooperatively, or in more complex ways.” Accordingly, what follows assumes two groups of actors – taxpayers and governments – and spells out the stylized decisions faced by each group, and the outcomes that result. The broad logic of this more sophisticated model is laid out in Figure 2.2, which highlights the simplified decisions available to each set of actors at each stage of the model, as well as the structural factors that are expected to influence these decisions.

Figure 2.2 Strategic interaction and tax bargaining.

The model begins with a government seeking to expand tax collection to finance its activities. The government is immediately confronted with the first “decision-point” in the model, which is reflected in the top “level” of Figure 2.2 and is summarized in point (1) of the same figure. One option is to pre-emptively bargain with taxpayers, offering commitments and concessions in exchange for popular acceptance of, and compliance with, expanded taxation. Alternatively, the government may seek to coercively impose new taxes, or strengthen enforcement of existing taxes, without making any reciprocal commitments to taxpayers. The choice made by the government is likely to be shaped by a variety of features of the broader context, including the urgency of government revenue needs, expectations about the ability and willingness of taxpayers to engage in tax resistance or collective action, and the ability of existing institutions to lower the costs of bargaining and facilitate pre-emptive agreement.

If the government opts to expand taxation coercively, without making pre-emptive concessions to taxpayers, then taxpayers may respond in one of three broad ways – summarized in point (2) of Figure 2.2. First, they may simply accept taxes without overt complaint, resulting in a case of expanded taxation without tax bargaining. Second, taxpayers may engage in tax resistance through evasion and avoidance, as a way of undermining government revenue-raising efforts and/or as a strategy for prompting government concessions. Third, they may seek to engage in collective action and public mobilization aimed at demanding that the government make reciprocal concessions to taxpayers or else repeal the tax. The course of action adopted by taxpayers is again likely to be shaped by context. Important factors include the scope for resistance through evasion and avoidance, the capacity of affected taxpayers to engage in effective collective action and the incidence and political salience of the particular taxes in question.

Confronted by taxpayer resistance or collective action, the model returns to governments, as they are forced to decide how to respond – with this decision summarized in point (3) of Figure 2.2. The government is presented with two main choices: it may compromise by offering reciprocal concessions to taxpayers or it may refuse to bargain with taxpayers and attempt to expand taxation through coercion. Where governments opt to compromise, as with the Ghanaian example summarized at the outset of this book, the result is an example of direct tax bargaining, with the government offering specific concessions in response to taxpayer demands. The alternative is that the government refuses to compromise, choosing instead to ignore or repress collective action and to tolerate any loss of revenue resulting from tax resistance. The way in which governments manage this decision is likely to reflect broadly the same contextual factors shaping the earlier decision about whether or not to pursue pre-emptive bargaining with taxpayers. The critical difference at this stage is that the capacity of taxpayers to engage in resistance and collective action will have been revealed, with bargaining facilitated by greater certainty about the relative power wielded by competing groups (e.g., Fearon Reference Fearon1995).

The final uncertainty that is addressed by the model relates to the long-term political consequences when governments consistently refuse to compromise in response to taxpayer resistance and collective action. This final stage is summarized in point (4) of Figure 2.2. One possibility is that conflict over taxation may simply fail to spur any meaningful bargaining or political change, as taxpayer resistance and collective action simply fade into history after a failure to achieve short-term gains. However, the model developed here suggests that, in many cases, these processes of unresolved conflict may nonetheless generate indirect pressure for political reform over the longer-term, despite the absence of short-term compromise.

Where citizens are engaged in tax resistance, and the government refuses to make concessions, this may result in reduced tax revenue. This may, in turn, weaken the fiscal position of the government, increasing the likelihood of a change in government and of subsequent reform (tax resistance and changes in government). Whether or not this occurs is likely to depend – to an important degree – on the scope for tax evasion and avoidance, and on the extent to which that additional revenue is pivotal to the political prospects of the government (revenue pressure). Similarly, political engagement and collective action by taxpayers, even where unrequited in the short-term, may contribute to strengthening the political capabilities of taxpayers in the long-term. This outcome is likely to be dependent on the capacity of civil society actors to transform short-term political mobilization into long-term collective action, and on the extent to which tax debates remain politically salient over the longer-term.

Ultimately, this model is attractive because it captures the potential diversity and complexity of the links between taxation, responsiveness and accountability. Whereas most contemporary discussion has focused implicitly or explicitly on direct tax bargaining, the model here captures multiple alternative pathways, emerging from differences in (a) government efforts to impose taxation, (b) taxpayer responses, and (c) the interaction between those two groups. This focus on repeated interaction allows alternative outcomes to emerge as a result of different strategic choices by taxpayers and governments, shaped by broader contextual factors.

A final note about the model is that it focuses on capturing alternative causal processes rather than on predicting specific types of outcomes. This is not an omission, but reflects the fact that there is no compelling reason to expect particular causal processes to lead to distinctive outcomes within the broad universe captured by the concepts of responsiveness and accountability. Instead, the specific concessions made by governments are likely to reflect the relative power wielded by different groups, idiosyncratic features of taxpayer demands and the particularities of government strategies.

2.6 Conclusions

Drawing on a careful reading of historical evidence, and building on earlier models, this chapter has presented the elements of a dynamic model of tax bargaining. The model proposes that the need for taxation can contribute to the expansion of responsiveness and accountability by increasing the bargaining power of taxpayers vis-à-vis the government. The need for taxation may increase the relative bargaining power of taxpayers through the potential for tax resistance and by encouraging political engagement and collective action. These may, in turn, give rise to three causal processes linking taxation, responsiveness and accountability: direct tax bargaining, tax resistance and changes in government, and strengthened political capabilities of taxpayers. Which of these specific outcomes, if any, emerges in practice is expected to be shaped to an important degree by a core set of contextual factors: the level of revenue pressure facing governments, the scope and potential for tax resistance, factors affecting taxpayer capacity for collective action, the existence of institutions that facilitate bargaining, and the incidence and political salience of taxation.

This model serves not only to clarify the causal logic linking taxation, responsiveness and accountability, but is also an integral part of the broader research strategy. These processes are likely to be difficult to observe, as they are expected to be frequently long-term and intertwined with broader political processes. The model developed here thus provides the foundation for trying to make sense of a messier empirical reality. However, there remains a need for care and nuance in evaluating the case study evidence that follows.

On the one hand, even if tax bargaining is significant to long-term political outcomes, there are still likely to be many cases in which episodes of increased taxation yield little or no evidence of direct or indirect tax bargaining. In even the best circumstances taxation remains a fundamentally coercive act. It is often only with a longer-term view that conflicts over taxation may come to be seen as part of larger processes of bargaining and political transformation. The question is thus not whether there will be episodes of taxation without bargaining, but whether those episodes will be so frequent, or so lacking in any evidence of public engagement and mobilization, as to outweigh competing evidence of a positive relationship between taxation, responsiveness and accountability. More simply, the question is whether evidence of tax bargaining is sufficiently strong to meaningfully enhance standard accounts of political contestation and change in developing countries.

On the other hand, precisely because processes of tax bargaining are expected to be difficult to observe, case study evidence is likely to underestimate the full political importance of tax bargaining. That is, if the case studies reveal readily observable tax bargaining, there is every reason to believe that additional, harder to observe processes are also operating below the surface. The case studies to follow are correspondingly an effort to capture comparatively visible examples of tax bargaining, but, in so doing, to also shed light on the existence of the broader causal mechanisms and processes described in this chapter. In turn, this reliance on case study evidence will allow for new insights into the diversity of the causal processes of interest, and into the contextual factors shaping outcomes both within and across countries. With this in mind, the chapters that follow present in-depth empirical evidence from case studies of Ghana, Kenya and Ethiopia.