The Politics of Revenue Bargaining in Africa: Triggers, Processes, and Outcomes

Kjær, Anne Mette (Ed.); Sandvad Ulriksen, Marianne (Ed.); Bak, Ane Karoline (Ed.) Book The Politics of Revenue Bargaining in Africa: Triggers, Processes, and Outcomes Provided in Cooperation with: Oxford University Press (OUP) Suggested Citation: Kjær, Anne Mette (Ed.); Sandvad Ulriksen, Marianne (Ed.); Bak, Ane Karoline (Ed.) (2024) : The Politics of Revenue Bargaining in Africa: Triggers, Processes, and Outcomes, ISBN 978-0-19-196482-4, Oxford University Press, Oxford, https://doi.org/10.1093/oso/9780192868787.001.0001 This Version is available at: https://hdl.handle.net/10419/302594 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/4.0/ The Politics of Revenue Bargaining in Africa The Politics of Revenue Bargaining in Africa Triggers, Processes, and Outcomes Edited by Anne Mette Kjær Marianne S. Ulriksen and Ane Karoline Bak Great Clarendon Street, Oxford, OX2 6DP, United Kingdom Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries © Oxford University Press 2024 The moral rights of the authors have been asserted Some rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, for commercial purposes, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization. This is an open access publication, available online and distributed under the terms of a Creative Commons Attribution – Non Commercial – No Derivatives 4.0 International licence (CC BY-NC-ND 4.0), a copy of which is available at http://creativecommons.org/licenses/by-nc-nd/4.0/. Enquiries concerning reproduction outside the scope of this licence should be sent to the Rights Department, Oxford University Press, at the address above Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America British Library Cataloguing in Publication Data Data available Library of Congress Control Number: 2023940510 ISBN 9780192868787 DOI: 10.1093/oso/9780192868787.001.0001 Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work. Acknowledgements The early ideas for this book emerged a decade ago. In 2013, Jalia Kangave, Mesharch Katusiimeh, Marianne Ulriksen, and Anne Mette Kjær were working on a contribution to an UNRISD research programme on domestic revenue mobilization for financing social policies. The programme was run byKatjaHujoatUNRISD.Whiledoingthiswork,werealizedtherewasalack ofsystematicandcomparativeresearchintocasesofmicro-level revenuebargaining between specific groups of revenue providers and the state. In fact, we knew little of the political dynamics of revenue bargaining in Africa. Jalia, Marianne, and Anne Mette pursued the idea over a couple of years and discussed it on several occasions in Aarhus and Johannesburg. Would aid dependence be further reduced and would this lead to increased efforts at raising domestic revenues and more domestic state–society reciprocity? A first unsuccessful proposal led to our narrowing in on revenue bargaining as the centre of attention in the analytical framework we then developed. This became the central focus in the proposal for the research programme ‘Political Settlements and Revenue Bargains in Africa’, which was granted funding in 2016 by Danida’s Consultative Research Committee (16-03-AU). Consequently, the acknowledgements for this book would be incomplete without mentioning those that made the broader research project happen through commenting and inspiring discussions: Ole Therkildsen and the members of the comparative politics section at the Department of Political Science, Aarhus University. Also, Jalia Kangave, who was initially part of the proposal butwhothengotapositionwiththeIDSinSussex andhadtoleaveus,offered valuable insights. Members of the research section of comparative politics at the Department of Political Science also offered generous comments on the proposal. The research programme ‘Political Settlements and Revenue Bargains in Africa’ (PSRB) www.ps.au.dk/psrb developed into a truly collaborative effort and research on revenue bargaining in particular in Tanzania and Uganda but also Senegal, Togo, and Zambia. The project has also seen cross-country collaborations with important comparative contributions. Professor Bakibinga and Dr Jamal Msami as heads of south teams of the PSRB programme havebeenthebestpartners:thankyou.Alltheseprojects havebeendiscussed continuously at five incredibly engaging and inspiring project workshops in vi Acknowledgements Aarhus in 2016, Entebbe in 2017, Bagamoyo in 2017, Arusha in 2019, and then finally again in Aarhus in August 2022. Thank you to all the programme researchers for always being keen to provide comments and constructive feedback on early as well as later paper drafts. In these project workshops, we were also lucky enough to get inputs from the outside. Thank you to OddHelge Fjeldstad, who joined us in Entebbe and Aarhus; to Matilde Thorsen who was with us in Bagamoyo; and Per Tidemand, who hosted us in Arusha and joined us in Aarhus. The importance of these PSRB workshops for this book cannot be emphasized enough. Quite concretely, the theoretical framework, which is carried through in this book, is largely built on a collaborative effort and sometimes strenuous conceptual and theoretical discussions at these workshops. The book itself came into being when four reviewers provided excellent feedback on the initial book proposal sent to Oxford University Press. The commentswereexceptionally thoughtful andmadeusthinkand workharder on the common framework as well as introductory and concluding chapters. Thank you also to the editors Adam Swallow and Vicki Sunter for their patience and belief in our work. Anna Persson, Lise Rakner, and Anne Mette Kjær organized a panel about revenue bargaining at the ECPR joint sessions in Cyprus in 2018, and we thankparticipantsatthepanel,andAnnaPerssonandChristianvonHaldenwang in particular. Several of the chapters here became a part of the book project in the wake of this workshop. We presented different chapters at various events: the Effective States and Inclusive Development conference in Manchester in September 2019; the IPSA in Brisbane in 2018; the NoPSA in 2021; the annual meetings of the Danish Political Science Associations; and a workshop at the Centre for Emerging African Economies at Roskilde University in December 2019. We thank Sam Hickey, Lars Buur, Rasmus Hundsbæk Petersen, and Lindsay Whitfield for comments on these occasions. The Effective States and Inclusive Development programme at Manchester has been a special source of inspiration, and we thank in particular Sam Hickey and Tom Goodfellow for their willingness to discuss their own book projects and pass on experiences. Furthermore, several annual meetings of the African Studies Association have provided the setting for presenting and discussing many of the book’s empiricalchapters,includinginBostonin2019whereNicvan deWalle acted as discussant on our book panel and gave insightful comments for which we are grateful, and at the Taxation in Africa panel in 2020 where Catherine Boone was another brilliant discussant. Acknowledgements vii For comments on many parts of the book project along the way, we would like to thank the Comparative Politics section at the Department of Political Science, Aarhus University, and the Danish Centre for Welfare Studies at the Department of Political Science, University of Southern Denmark. Mia Woer provided excellent research assistance for the introductory chapter, for which we are thankful. We thank ‘the Bergen tax people’ Odd-Helge and Lise Rakner for continuous discussions about the book project, and taxation and development issues ingeneral.Theseconversationshavebeenimmenselyimportantforthebook, and we are thankful to Odd and Lise for your continuous support. Mick Moore, always a hard but fair critic, has commented on various papers of ours which have developed into parts of the book, and we owe a lot to Mick’s way of thinking about tax and the governance dividend. We owe a special and heartfelt thanks to Ole Therkildsen for his always very clever and thoughtful inputs. To the ‘Tax Ladies’—Rachel Beach, Matilde Jeppesen, and Ane Edslev Jacobsen:Thankyouforalloureveningsofnerdytaxdiscussions,whichhave fed into this book and enriched it. Finally, Annette Bruun Andersen and Natasha Elizabeth Perera have been efficient and patient with us in the language-revision phase and we are eternally grateful to you for bearing with us. 2Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak then politically overruled. In this case, as in many others, the bargaining process ends in (the continuation of) tax exemptions. We argue that this outcome represents a micro-level fiscal contract defined as explicit or implicit agreements between ruling elites (or representatives thereof) and individuals or groups of revenue providers related to revenue provision. While not an agreement concerning an exchange of taxes for goods or services, it is still an agreement in which the bargaining parties recognize their mutuality and interdependence. Tax exemptions may seem a poor choice from a revenue mobilization perspective, but they accommodate the demands of the farmers andthepoliticalconcernsofUgandanpoliticians.Thistensionbetween fiscal andpoliticalinterestsisattheheartofthisbook.Throughoutthechapters,we present micro-level instances of revenue bargaining across five African countries to explore how and under what conditions revenue bargaining emerges, evolves, and leads to fiscal contracts. Our theoretical framework (Figure 1.1) informs the case studies presented in the empirical chapters, Chapters 3–12. The bottom part of the framework illustrates how the revenue-bargaining processes unfold at the micro level. As we theorize in Chapter 2, whether and when revenue bargaining leads to micro-level fiscal contracts is conditioned by what triggered the bargaining, how and where the bargaining processes unfold, and in particular the relative bargaining positions of revenue providers vis-à-vis ruling elites. We focus on the micro level to gain a nuanced and in-depth understanding of the politics of revenue bargaining. However, micro-level instances of revenuebargainingareformed bymacro-level contexts and havemacro-level implications (top part of Figure 1.1). Thus, we situate the microcosmoses of revenue bargaining within countries’ political contexts, that is, the political settlements. The political settlement approach helps to disaggregate state and societal actors and to pay attention to revenue providers’ resources and bargaining power within contexts dominated by informal clientelistic relations in resource-constrained economies. The political dynamics at play across the cases show, among other things, that ruling elites (i.e. individuals who have either formal or informal positions of governing power) often accept tax exemptions in order to maintain their positions of power. We assess whether the micro-level fiscal contracts entail some level of responsiveness and repeated interactions between the ruling elite or representatives hereof and revenue providers; what we term state–society reciprocity. Many of the case studies show that governments prove willing to negotiate with revenue providers—from big business to informal sector workers—and make concessions, although these are often in the form of reduced tax payments. Political settlement Holding power of state and societal actors Trigger Actions, events or changes in circumstances that prompt negotiations Relative bargaining positions Revenue providers' organizational and fiscal importance vis-a-vis ruling elites Revenue bargaining Direct or indirect negotiations between ruling elites and revenue providers related to revenue provision Micro-level fiscal contract Agreements between individual or groups of revenue providers and ruling elites related to revenue provision State-society reciprocity Responsive and repeated interaction between state and societal actors around policy Macro-leve l Micro-level Figure 1.1 The bookʼs theoretical framework. 4Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak While the concessions given by governments do not amount to increased political accountability or improvements in public service delivery, our empirical case studies show that negotiations can prompt responsiveness and repeatedengagements.In somecases, these engagementsbecameinstitutionalizedwithintheformalstate structures, therebyhavingthepotentialto move beyond informal processes of policymaking. Consequently, in this book, we contribute to knowledge about when and how expanding revenue mobilization may prompt institutionalization of state–society interactions around taxation. In the following, we begin by describing the increased focus on domestic revenue mobilization, its effect on the politicization of taxation, and whether current research has found signs of the emergence of fiscal social contracts. Second, we look back at the early contributions on taxation and on state– society relations in Africa and argue that political changes since the 1990s have fertilized the grounds for a link between taxation and state–society reciprocity. Third, we outline our theoretical contributions to the fiscal contract andthepoliticalsettlementtheories(toppartofFigure1.1).Finally,wesetthe stage for studying revenue bargaining by presenting our comparative design, which includes case studies in Tanzania, Uganda, Mozambique, Senegal, and Togo. Chapter 2 will then detail the theoretical framework (bottom part of Figure 1.1), explain the methodological approach, and briefly introduce the empirical chapters in the volume. 1.2 Expanding taxation, politicizing taxation Since the turn of the new millennium, the focus on the importance of domestic revenue mobilization in low or lower-middle income countries has increased immensely among academics, governments, and development practitioners alike. This is most importantly explained by increasing demands for the financing of development and public goods provision such as improved public infrastructure and better public services in health, education, and social protection (IMF, 2018;UN, 2021). The UN’s ambitious Sustainable Development Goals (SDGs), agreed upon by all member states, are a manifestation of these demands and serve to reinforce the need for development financing, beyond what can be achieved by overseas development assistance (ODA). While total net ODA has effectively increased in absolute terms in recent years, even reaching its highest-ever point (USD 162 billion) in 2020, it still only constitutes a very small proportion of the estimated amount needed to Politicization of taxation and state–society reciprocity in Africa 5 finance the SDGs, around USD 2.5 trillion (Doumbia and Lauridsen, 2019; Winckler Andersen and Therkildsen, 2019). The UN has acknowledged this critical gap in the financing of the SDGs, which led to the formulation of the Addis Ababa Action Agenda in 2015, and with it the intention to integrate financing into the work of reaching the SDGs. A designated UN Tax Committee was established to support, in particular, developing countries in strengthening tax systems and domestic revenue mobilization (see e.g. UN, 2021). Increasing debt burdens have further heightened the pressure on governments to expand domestic revenue mobilization to balance debt stocks (Bak, Jeppesen, and Kjær, 2021;Smith, 2021;OECD, 2021), a demand which has become even more pronounced since the COVID-19 pandemic. Attention to domestic revenue mobilization has increased and, since the turn of the millennium, so too has tax revenue across Africa. According to the IMF (2018), total revenue, excluding grants, increased from an average of 14% of GDP in the 1990s to about 18% in 2016, while total tax revenues increased from 11 to 15% in the same period (Moore, Prichard, and Fjeldstad, 2018, 32). The same trend is found in several newly developed datasets (UNU-WIDER, 2021;Cogneau et al., 2021;Jerven, 2022) which have sought to improve both data quality and measurement validity, as well as to bolster avenues for comparison of revenue trends across time and space. Albers and Suesse (2022) carefully calculate real tax revenue per capita from 1900 up until the present. They find that for the average African country, taxation has increasedsincethe1960s,buttookalargejumpattheturnofthemillennium. Cogneau et al. (2021) find that in former French colonies, tax per GDP has increased by 5.9 percentage points of GDP between independence and today. They find that this increase is largely explained by natural resource revenue, but other researchers argue that the increase follows a rise in the capacity to tax society more broadly (Jerven et al., 2022;Moore, 2021). Many so-called modern taxes such as personal and company income taxation, and property taxes had already been introduced in colonial times (Genschel and Seelkopf, 2022). Beyond a sustained period of economic growth in the 2000s, recent expansions in domestic revenue mobilization can be explained at least partly by growing capacity in tax administrations, new technological opportunities, the adoption of tax instruments that target and accommodate challenges posed by the globalized economy, and the economic structures and realities on the ground (Moore, 2021). The expectation among academics, donors, and development practitioners alike is that taxation could be the source of a ‘healthy’ conflict between citizens, bureaucrats, and politicians, and between state and society more 6Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak broadly (Moore, 2004;Bräutigam, 2008;Prichard, 2015;Lucas, 2017). As individuals feel a higher fiscal burden, they could be expected to make demands for a return in the form of improved public goods and services and more political accountability from state actors. In theory, states will then seek to accommodate these demands because it is less administratively and politically costly to tax people if they are themselves more willing to pay. Over time, such interactions around taxation have been theorized to create an exchange-based macro-level fiscal social contract, whereby the state delivers public goods and services in return for tax compliance and revenue (Levi, 1988;Bräutigam, Moore, and Fjeldstad, 2008;Moore, Prichard, and Fjeldstad, 2018;Bak, 2019). Looking across the continent, taxation has been source of some conflict, as individuals and taxpayers have reacted to new or increased taxes. To highlight a few examples from recent years: several African countries have sought to tax social media, which met with strong popular reactions in, for example, Uganda (Reuters, 2018) and Benin (Okunoye, 2019); or to tax mobile money transactions, which took people to the streets in Ghana and Tanzania in 2021 (The East African, 2021;BBC, 2022). In Kenya, the removal of VAT exemptions for fuel led to strikes and widespread protests on social media and in the streets in September 2018 (Miriri, 2018); and in Ethiopia, small business owners shut their businesses to protest against a tax hike (Africanews, 2017). Such protests over specific taxes spur revenue bargaining between revenue providers and state actors, but they relatively rarely revolve around a new demandforpublicservicedeliveryandaccountabilityinexchangeforthetax. Rather, the demand is that the proposed change in taxation is not adopted. Theprotestsaresometimessuccessfulinpromptinggovernmentstorespond. In Benin, for example, the social media tax was never adopted; in Kenya the suggested VAT rate on fuel was halved from 16 to 8%; and in Ethiopia, the tax hike was abolished. Tax grievances sometimes coincide with larger grievances and dissatisfactions with government. One oft-cited case from the 1990s is the protests against increases in VAT in Ghana, which led to large-scale anti-government protests (Prichard, 2015, ch. 3). In the 2017 case of Ethiopia, the so-called tax hike protests evolved into anti-government protests in Oromia state, largely explained by the Oromo people’s already strained relations with the national government (Dahir, 2017). However, such cases are rare and the circumstances often quite specific, as in Oromia. Summing up, there are signs of politicization of taxation in the wake of increasing domestic revenue mobilization on the African continent. But it is not clear what this politicization implies for the expectationof the emergence Politicization of taxation and state–society reciprocity in Africa 7 of fiscal social contracts between state and society, especially ones centred on the exchange of taxes for political accountability and public service delivery. 1.3 Signs of fiscal social contracts? During the past two decades, the scholarly literature on taxation and fiscal contracts has grown immensely, seeking to answer whether taxation leads to fiscal social contracts and whether it has positive so-called governance dividends (Moore, 2004;2015) such as political accountability and representation. It is fairly established that, on the macro level and across continents, higher taxation levels are positively correlated with certain political outcomes. Taxation has been linked statistically to levels of democracy (Ross, 2004;Baskaran, 2014), the quality of governance (Baskaran and Bigsten, 2013;Broms, 2015), and vertical accountability (Dom, 2018). Survey-based research examining the behavioural effects of taxation upon the individuals paying taxes have also found positive effects on various measures of political engagement or accountability demands (Paler, 2013;Broms, 2015;Sjursen, 2018). However, as Dom (2018, 31) notes, the observations of an average correlation on the macro level tells us little about whether and how taxation relates to such political outcomes in individual countries. Nor do individuallevel survey experiments translate into implications about the potential for societal and political changes. While it seems that taxation may have some effects on political outcomes, there are few signs of macro-level exchange-based fiscal social contracts whereby improved public service delivery is provided in return for tax payments (Moore, Prichard, and Fjeldstad, 2018,Guimaraes, Duca, and Ndlovu, 2018;Fjeldstad and Therkildsen, 2020). As Fjeldstad and Therkildsen (2020, 38) state, ‘we do not know of any major political deals (fiscal contracts) in recent years involving an increase in broad-based taxes in exchange for the provision of public services (education, health, roads, etc.)’. There is also a large strand of qualitative case-based research that aims to further explore the effects of taxation on political outcomes and state–society relations (Rakner and Gloppen, 2003;Eubank, 2012;Jibao and Prichard, 2015;Prichard, 2015;Rakner, 2017;Gatt and Owen, 2018;Goodfellow and Owen, 2018;Bak,2019;Schneider,2020).Mostofthis researchdemonstrates that taxation can prompt state–society interactions. However, they remain isolated case studies with few or no cross-country comparisons, and they do not systematically disaggregate the causal mechanisms between taxation 8Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak and its political outcomes. One notable exception is Prichard’s (2015) study of tax bargaining across the diverse political contexts of Ghana, Kenya, and Ethiopia. He finds that taxation can lead to changes in public service provision, tax policy, and administration and expansions in accountability, and he goes some way towards theorizing when governments’ tax initiatives lead to tax bargaining, focusing on a limited set of institutional conditions. The scholarly interest has been mainly focused on broad-based exchanges of taxation in return for something. As Prichard himself later observed (2019, 9), much of the original fiscal contract literature on the macro level has studied large, relatively dramatic shifts in tax collection and accordingly fiscal contracts. Perhaps we have been looking in the wrong places? There is growing evidence that changes in taxation do lead to revenue bargaining around tax policies, but as shown above, rarely at the macro level. Bargaining isconfinedtospecificgroupsoftaxpayersortospecificinstancesoftaxreform or reactions to changes in tax administration. As Moore, Prichard, and Fjeldstad(2018,182)have rightlyargued, we need toexplorenotsomuchwhether macro-level fiscal contracts are emerging, but rather ‘what kinds of tax bargains are likely in different contexts’. Incremental yet significant changes and political effects are also likely to unfold at the micro level, where individuals or groups negotiate with state actors over taxes. A critical and more recent strand of the literature has stressed the importance of contextualizing taxation to understand its effects below the macro level (Fjeldstad, 2001;Kjær, 2009;Bodea and Lebas, 2016;Broms, 2017; Prichard and van den Boogaard, 2017;Meagher, 2018;Bak, 2019;van den Boogaard, Prichard, Beach, and Mohiuddin, 2021). Specifically, this research shows that whether or not societal actors engage with state actors around tax-related issues is conditioned by their capacity for collective action, taxcollection methods, the presence of community-based service, and social circumstances such as gender, ethnic identity, and patronage ties. Local politics and societal structures condition the effects of taxation but, importantly, are also shaped, disrupted, or reinforced by taxation. Thus, when observing the effects of increases in taxation on state–society relations and political outcomes, we need to pay attention to the effects on social and political power structures. Therefore, this book embeds the analysis of revenue bargaining within a theoreticalframeworkcognisantofmacro-andmicro-level powerstructures. This implies that as we study revenue bargaining at the micro level, we pay particular attention to how the power positions of revenue providers vis-àvis ruling elites shape the processes and potential outcomes. We heed Mick Moore’s advice, and ‘go beyond the initial reactions to taxation of each type Politicization of taxation and state–society reciprocity in Africa 9 of actor and take into account the ways in which they then interact, whether conflictual, cooperatively, or in more complex ways’ (Moore, 2007, 16). This enables us to observe how negotiations related to revenue provision result from and feed into existing power structures, and thereby we can determine whether and when such moments could indeed be termed reciprocal interaction around taxation. We gain a deeper understanding of the potential effects of taxation on state–society reciprocity and, in turn, a new perspective on the implications of the recent increased politicization of taxation in Africa. In the following, we further develop our understanding of state–society reciprocity and how it relates to taxation. We present a broader and older literature on state–society relations and disengagement in Africa and relate this to recent political progress on the continent. We argue that, in a context of regular elections, increased political competition, and strengthened civil society, the stage may be set for state–society reciprocity to develop through revenue bargaining. 1.4 Taxation and state–society reciprocity in Africa The idea that expansions in taxation in less democratic developing countries could cause accountability and state–society engagement originates from European state-building experiences. The comparative historical accounts by Levi (1988) unfold the idea that taxation involves some kind of bargained exchanges between states and their citizens. Drawing specifically on the emergence of institutionalized representation and accountability in Britain and juxtaposing rentier states with fiscal states, Moore (2004) developed the proposition of a governance dividend of taxation in contemporary developing countries. Building on these seminal works, the edited volume by Bräutigam, Fjeldstad, and Moore (2008) cemented the potential links between taxation and the development of political institutions and statebuilding in developing countries through both theoretical and empirical chapters. Since then, the fiscal contract literature has flourished. An older literature on state–society relations, state-building, and development in Africa had in fact touched on similar issues, although with a less systematic focus on taxation (Ekeh, 1975;Hyden, 1983;Rothchild and Chazan, 1988;Migdal, 1988;Boone, 1992;Bratton and Hyden, 1992;Guyer, 1992;Harbeson,Rothchild, and Chazan, 1994). Inthisliterature,theconcept of reciprocity featured centrally in the analysis of the African state. The state was often described as being detached from society, as citizens would withdrawfromoftenrepressivestateactions.InVictorAzarya’s(1988)words,they 10 Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak would ‘disengage’ from the state, seeking survival in the reciprocal relations existing in alternative realms. Such alternative realms have been defined in slightly different ways and given a variety of labels, including the ‘economy of affection’ (Hyden, 1983) or in Peter Ekeh’s (1975) term ‘the primordial realm’. Ekeh famously argued that real reciprocity existed in the primordial realm rather than in the formal ‘civic realm’. In the former, citizens felt that their contributions, given in the form of tribute, time, labour, and loyalty, were truly reciprocated. They were willing to invest in the primordial realm because they recognized that here they had duties as well as rights. Within this literature, Hyden (1980;1983) formulated a negative link between taxation and reciprocity. Based on the concept of ‘the peasant mode of production’, he argued that the peasant economy with its limited technological advances and widespread subsistence production did not give grounds for the state to grow structural or functional roots in society. Rulers could not tax surplus production, which would emerge only out of productivity increases, and instead, they had to demand tributes from the existing stock. Consequently, taxation was seen as an imposition and the state an intruder, coercive in nature. Rather than opting fully into the formal system, peasants preferred to have an exit option and would therefore always keep one leg in the informal ‘economy of affection’. Any taxation-induced reciprocity between state and society became difficult, precisely because genuine reciprocity effectively only existed within the economy of affection. Hyden’s contribution was criticized for exaggerating the extent to which peasants were able to exit the system (Bates, 1981), and it was questioned whether customary relations could indeed be reciprocal given the nature of highly hierarchical socio-economic structures. For example, René Lemarchand argued that reciprocation may arise as much from the fear of retaliation as from a ‘structurally induced form of altruism’ (1989, 38–40). In a later contribution on concepts in the study of politics in Africa, Hyden (1992, 9f) develops his understanding of reciprocity, on which we rely here. He argues that state–society relations do not rest merely on immediate quid pro quo exchanges, and that we should bring into the analysis the concept of reciprocity. While reciprocity has much in common with exchange, it differs first and foremost in that it constitutes a continuous relationship, which is based as much on expectations of behaviour as it is on actual behaviour. Reciprocity requires broader agreement and a consensual underpinning. Examining how revenue bargaining and the politics thereof influence state– society reciprocity thus allows us to identify different and more nuanced effects of taxation and revenue bargaining than simple exchanges of taxation for public services and accountability. Politicization of taxation and state–society reciprocity in Africa 11 As is evident from the thriving fiscal contract literature, Hyden’s focus on therelationship betweentaxationandreciprocityremainsrelevant;however, there are reasons to subscribe to a more positive perspective than that which prevailed in the 1980s. 1.4.1 Fertilized ground for state–society reciprocity? Since the publication of these important contributions on state–society relations in Africa in the 1980s and 1990s, the political and economic contexts have changed significantly, with implications for the dynamics of state– society reciprocity. Subsequent generations of political science research have highlighted, among other things, the significance of increased political competition and a strengthened civil society. This changed context may have fertilized the ground for the emergence of revenue bargaining as ‘healthy’ interactions between states and their citizens, which in turn build state– societal reciprocity. Many African countries experienced democratic openings during the 1990s and with them a political competition that caused optimism for the development of state–society reciprocity (Bratton and van de Walle, 1997). Wasthispositiveexpectationrealized?Fromoneperspective,itcanbeargued that progress towards democratic consolidation has been disappointing, as most African regimes did not fully democratize. Indeed, recent scholarly contributions on African politics highlight the stability of hybrid regimes where formal democratic institutions, such as elections, parliaments, and courts,coexistwith authoritarianfeaturessuchastheconcentrationofpower in the president who often prolongs term limits or uses the parliament as a rubber stamp in de facto one-party systems, and repressive practices which infringe on political rights and civil freedoms (Cheeseman and Klaas, 2018;Bleck and van de Walle, 2019. On the other hand, after several hundred competitive elections over the last thirty years, the holding of elections has become the ‘default option of politics’ (Bleck and van de Walle, 2019, 6), and elections have caused changes to African politics as citizens have become voters with expectations and hopes of better lives (Bratton and Logan, 2006). Bleck and van de Walle (2019, 20–22) argue that elections are political moments of ‘heightened citizenship’ and can have both positive as well as negative implications. They might promote exclusionary and demagogic politics as much as they can also promote responsiveness on the part of the government. Generally, the turn to regular competitive elections has implied ‘an unprecedented explosion in political participation’ (ibid, 22), 1980 1981 1982 1986 1989 1998 1999 2000 2018 2019 0,12 0,1 0,08 0,06 0,04 0,02 0 % of GDP 2020 2017 2016 2015 2014 2013 2012 2011 2010 2009 2008 2007 2006 2005 2004 2003 2002 2001 1997 1996 1995 1994 1993 1992 1991 1990 1988 1987 1985 1984 1983 Mozambique Senegal SSA average Tanzania Togo Uganda Figure 1.3 Grants, percentage of GDP, 1980–2020. Source: ʻUNU-WIDER Government Revenue Datasetʼ. Version 2021. https://doi.org/10.35188/UNU-WIDER/GRD-2021. Politicization of taxation and state–society reciprocity in Africa 19 Mozambique’s tax revenue in 2013, 2017, and 2019 are largely explained by large one-off winnings from capital gains taxation of asset sales in the extractive industry (IMF, 2016, 20; 2020, 14). Most recently, the fall in revenue between 2019 and 2020 in the countries where data are available indicates a clear impact of the COVID-19 pandemic on the economies (Fjeldstad and Therkildsen, 2020). The other essential change is that the countries have become less aid dependent. This is shown in Figure 1.3. Throughout the period displayed in the graph, the level of grants, measured as share per GDP, is highly volatile, but since the change of the millennium the tendency, with some variation, is relatively clear: the countries are generally less aid dependent and are thus in need of expanding domestic revenue. Following the fiscal contract theory, these two significant changes (rising domestic revenues and declining aid dependence) would lead to the expectation of a fiscal impetus to revenue bargaining. Upon examination, the economic structures of the five case countries appear typical of lowor lower-middle-income countries (see Table 1.1). The five countries’ GDPs per capita range from USD 587 (Mozambique) to USD 1,585(Senegal),whichareallbelowtheaverageinsub-SaharanAfricaofUSD 1,656. They also score below average on the Human Development Index, and the proportion of the population living below the USD 1.90 poverty line ranges from 38.5% to 63.7%, with only Senegal below the sub-Saharan Africa average. None of the countries benefit from large, stable flows of extractive natural resource rents. The economies in all five countries are still dominated by agriculture. While the sector only accounts for about one quarter or less of overall economic output, agriculture provides livelihoods for the large majority of the countries’ populations. Agriculture thus constitutes between 30 (Senegal) and 72% (Uganda) of total employment. Agricultural productivity, as shown by cereal yields in kilogrammes per hectare, is low, between 835 kg and 2 tonnes, on average almost half of the average yields for South Asia (Abraham and Pingali, 2020, 186). Subsistence agricultural production is widespread and informal commercial sectors are large. From a political settlement perspective, this suggests that the level of capitalist development is limited, and therefore that a capitalist class of any significant size or influence will be limited. Consequently, the political settlements in our case countries are mainly clientelist rather than capitalist, a feature which characterizes all poor countries (Khan, 2010). This also means that few organized societal groups are able to create wealth outside access to 20 Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak rent opportunities nurtured through their relationships with political elites (Khan, 2010, 54–55; Hickey and Hossain, 2019, 26). Ultimately, the lack of economic transformation, together with the consequent absence of a capitalist class, implies that substantial structural barriers to expanding the tax base remain in all our five case countries. This has implications for the politics of revenue bargaining and taxation. Under the conditions of a small tax base, the national budget is similarly constrained. This makes it difficult to secure political stability through the reallocation of domestic revenue to powerful groups (Khan, 2010). Instead, distribution of resources and access to rents must happen off budget and through clientelist or other informal networks. In clientelist settlements like our case countries, building a stable ruling coalition requires that elites balance and accommodate powerful factions, and this is likely to manifest in whether and how revenue bargaining emerges. Furthermore, the holding power of the individual and groups of revenue providers vis-à-vis the ruling elite will be significant for the kind of outcomes that result from such bargaining. Table 1.2 presents the scores of our case countries on a number of political features that have been found to condition where and how we expect revenue bargaining to occur (Moore, 2004;Prichard, 2015;van den Boogaard et al., 2021). All five countries have institutionalized regular elections, which provide one possible political channel through which societal actors can engage state actors over tax policies. The quality of the elections and the openness of the political space do vary between the countries. Senegal stands out among the cases as the most democratic with the highest level of political rights, civil liberties, government effectiveness, and political stability, but a look beyond theseindicatorsleaveslittledoubtthatSenegalstillfallswithinthecategoryof a hybrid regime with strong executive power, politicization of the judiciary, and instances of violent repression, together with strong clientelist networks permeating democratic institutions (Dumont and Kanté, 2019;Kohnert and Marfaing, 2019;Kelly, 2020). At the other end of the scale, Uganda should undoubtedly be considered an electoral autocracy, but there is still meaningful political competition, and political rights and civil liberties are not considerably worse affected than in Tanzania. 1.6.2 The political settlements of our case countries The political indicators only tell part of the story. To understand the political space for revenue bargaining and specifically the contextual conditions of bargaining power of different revenue providers vis-à-vis the ruling elites, we Table 1.1 Economic indicators Senegal Mozambique Togo Tanzania Uganda SSA (sub-Saharan Africa) averageg GDP per capitaa1,381.6 598.8 630.8 1,071.4 894.5 1,656.1 HDIb0.512 0.456 0.515 0.529 0.544 0.547 Poverty head count under USD 1.90 (share of population)c38.5 % (2011) 63.7 % (2014) 51.1 % (2015) 49.4 % (2017) 41.3 % (2016) 42.3 % (2014) Employment in agriculture (share of total employment)d30.1% 70.2% 32.4% 65.1% 72.1% 52.9% Cereal yield (kg per hectare)e1301.9 835.1 1145.6 1568.3 2049.5 1445.2 Natural resource income (share of GDP)f0.016 % 0.04 %∗0.014 %∗∗ 0.015 %∗0.007 % 0.029 % aYear: 2019 (constant 2015 USD), from World Bank national accounts data, and OECD National Accounts data files. bYear: 2019, index from 0–1, from Human Development Report Office, 2020. cRespective years in parentheses, from World Bank, Development Research Group. Data are based on primary household survey data obtained from government statistical agencies and World Bank country departments. dYear: 2019, from International Labour Organization, ILOSTAT database. Data retrieved on 29 January 29 2021. eYear: 2018, from Food and Agriculture Organization, electronic files and website. fYear: 2015–2020 (average of period), from UNU-WIDER Government Revenue Dataset. Version 2021. https://doi.org/10.35188/UNU-WIDER/GRD-2021. Indicator: Non-tax Revenue. Average of non-tax revenue between 2015 and 2020 is made by own calculation. ∗Missing data from 2020. ∗∗Missing data from 2019–2020. gWDI’s definition of sub-Saharan Africa (excluding high income). 22 Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak Table 1.2 Political indicators Senegal Mozambique Togo Tanzania Uganda Electoral democracya0.75 0.37 0.42 0.41 0.31 Political rightsb29 14 15 12 11 Civil libertiesb39 29 27 22 23 Government effectivenessc0.01 −0.72 −0.69 −0.77 −0.58 Political stabilityc−0.02 −1.16 −0.92 −0.41 −0.69 Nature of Political SettlementdStrong dominant Weak dominant Strong dominant Vulnerable authoritarian Competitive clientelist Source: aYear: 2021, index 0–1, from Nazifa Alizada, Rowan Cole, Lisa Gastaldi, Sandra Grahn, Sebastian Hellmeier, Palina Kolvani, Jean Lachapelle, Anna Lührmann, Seraphine F. Maerz, Shreeya Pillai, and Staffan I. Lindberg. 2021. Autocratization Turns Viral. Democracy Report 2021. University of Gothenburg: V-Dem Institute. bYear: 2022, index 0–100, from Freedom House Score. Countries and Territories. 2022. https:// freedomhouse.org/countries/freedom-world/scores. cYear: 2020, index from −2.5 to +2.5, from World Government Indicators http://info.worldbank.org/ governance/wgi/Home/Reports. dYear 2018, Categorization from the ESID data, except for Togo (lacking in ESID), which is based on secondary sources. need to take as our point of departure the case countries’ political settlements (see also bottom of Table 1.2). In the following, we briefly present the political settlement typology and use it both to introduce the typical features of our case countries’ political settlements, and to outline where they differ. Each chapter considers (implicitly or explicitly) how individual or specific groups of revenue providers involved in instances of revenue bargaining are positioned in relation to the ruling coalition and the ruling elites. We follow the original political settlement typology as introduced by Khan (2010, 55). This typology has been discussed and further elaborated by others (e.g. Whitfield et al., 2015;Hickey and Kelsall, 2020;Schulz and Kelsall, 2021), work which we draw on in what follows when categorizing our case countries.² Khan’s typological categorization follows two steps. The first is to distinguish between capitalist and clientelist settlements. As discussed, all our case countries fall into the latter of these two categories: Clientelist ² Particularly, we draw on the Political Settlements (PolSett) dataset (Schulz and Kelsall, 2021), developed by the Effective States and Inclusive Development (ESID) programme based in Manchester, which categorizes countries in terms of their political settlement based on expert surveys and their openly available detailed descriptions. The original Khan typology has subsequently been subject to much discussion, but for our purposes here (to characterize different settlement and therefore to emphasize the diversity of the countries in which case studies are carried out), it suffices. Politicization of taxation and state–society reciprocity in Africa 23 settlements are characterized by the fact that ‘significant holding power is based on sources outside the incomes generated by formal institutions’ (Khan, 2010, 53). While formal institutions such as property rights or regular elections will be in place, they are rarely enforced, because they are at odds with the interests of powerful groups. When this is the case, there is a mismatch between formal institutions and how politics work informally. The second step in Khan’s typology is to categorize different types of clientelist settlement. To this end, Khan (2010, 64–65) distinguishes between the horizontal and the vertical distribution of power. The horizontal distribution of power refers to ‘the power of excluded factions, relative to the ruling coalition’. Some ruling elites can create a broad-based coalition which includes many powerful factions, while others cannot. In the latter case, if excluded factions are strong, the ruling coalition is more vulnerable. The vertical distribution of power distinguishes between higherand lower-level factions within the ruling coalition. Lower-level factions are the ones who mobilize support in local constituencies, and their source of holding power depends largely on how much support they can mobilize for the ruling coalition. The stronger the lower-level factions, the more the ruling elite will have to attend to their interests in order to keep them within the ruling coalition. Betweentheverticalandhorizontalpowerdimensions,therearefourtypes of clientelist settlement: (1) Strong lower-level as well as strong excluded factions imply competitive clientelism, where competition between factions is strong. (2) Strong lower-level combined with weak excluded factions imply weak-dominant party coalitions, in which the dominant party is not threatened by excluded factions but where the strong lower-level factions have the power, for example, to block policy implementation or to achieve good access to rents. (3) Where both lower-level and excluded factions are weak, the coalition is strong-dominant, meets little resistance, and the ruling elite is fairly free to implement the policies it pleases. (4) Finally, when lower-level factions are weak and excluded factions are strong, the political settlement is characterized by a vulnerable authoritarian coalition, which enjoys some enforcement capacity but suffers from instability because it has to resort to repression of excluded factions. Within this typology, Uganda can be categorized as competitive clientelist (Schulz and Kelsall, 2021). Primarily as a result of the colonial borders, the country has since independence been characterized by many different regional, religious, ethnic, and socio-economic cleavages. In the south there were relatively strong centralized kingdoms with a history of conflicts between them, and north of the river Nile there were smaller chiefdoms belonging to different ethnic groups (e.g. the Alur in West Nile closer to 24 Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak Sudan and the Karamojong closer to Kenya) (Karugire, 1996;Reid, 2017, 118). These cleavages made for a fragmented political settlement, that is, a balance of power that had many competing groups and factions. In such a fragmented political settlement, it is always difficult to construct a stable ruling coalition with factions that provide reliable and lasting support for the ruling elite (Kjær, 2015;Whitfield et al., 2015). Several early postindependence leaders in Uganda struggled to create durable coalitions, but since the National Resistance Army (NRA) and its leader, Yoweri Museveni, won a guerrilla war in 1986, the coalition has been relatively durable even if its composition has changed character over the years, with fragmentation and increasing competition between factions. A key driver of the increased fragmentation was the introduction of multi-party elections in 2006 and the simultaneous lifting of presidential term limits. Elections and the introduction ofNRM party primaries have served to increase the power of lower-level factions (Kjær and Katuusiihmeh, 2021). At the same time, power is more dispersed among the elites, and although there is a powerful circle around the president, the military seems more fragmented (Reuss and Titeca, 2017). During the latest elections, opposition candidates have been able to mobilize substantial support (Khisa, Vokes, and Wilkins, 2021). So, even if it is strongly authoritarian and with the exercise of power personalized around the president, the fragmentation and competition between factions gives Uganda the categorization of competitive clientelist. For revenue bargaining, Uganda’s competitive clientelism means that elites will be incentivized to give in to demands from powerful factions to a higher extent than in a strong-dominant party settlement. Chapters 7 (by Kjær and Arinanye) and 8 (by Ngabirano) focus on revenue bargaining with, respectively, Uganda’s agricultural sector and multinational companies in the petroleum sector. In Chapter 9, Khisa, Msami, and Therkildsen examine campaign financing and revenue bargaining in both Uganda and Tanzania. Tanzania is categorized by the PolSett dataset as vulnerable authoritarian with rather strong excluded factions and weak lower-level factions (see Table 1.2). This labelling shows a significant change from the Julius Nyerere period when power was more concentrated with much fewer powerful excluded factions. For a long period after Nyerere’s retirement, the coalition became more fragmented, especially during the Jakaya Kikwete years (Therkildsen and Bourgoin, 2012;Kelsall, 2018). However, this changed when the CCM party had to elect a new chairman in 2015. Factional competition within the party led to an internal party split, whereby the central figure Edward Lowassa broke away to the opposition party Chadema taking with him both financial and political power (Kelsall, 2018, 18). The Politicization of taxation and state–society reciprocity in Africa 25 new party chairman John Magufuli understood the need to control excluded factions and centralize power in the party. In vulnerable authoritarian settlements, the expectation is that the government should be able to enforce some policies and, for example, could expand their tax effort without having to compromise for less in specific instances of revenue bargaining. However, several chapters in this book provide grounds for questioning whether this is the case for the politics of taxation. It should be noted that fieldwork for the chapter contributions was done under the now late Magufuli’s presidency. After two years in power, it is evident that the ruling coalition has changed with the new president Samia Suluhu Hassan, allowing for more political opposition and rejuggling the coalition so as to exclude some of Magufuli’s supporters. It is still early to assess the implications of this change. In Chapter 3, Edslev Jacobsen focuses on revenue bargaining in the urban informal sector in Dar es Salaam, while Chapters 5 (by Ulriksen, Katera, and Msami) and 6 (by Fjeldstad and Rakner) examine the negotiation processes around various tax reforms in Tanzania. Lastly, in Chapter 11, Ulriksen, Myamba, and George study bargaining between the Tanzanian government and international aid agencies around the financing of a cash transfer programme. Senegal is categorized as strong-dominant with weak excluded factions, weak lower-level factions, and concentrated power in the leadership. There has been little change in the political ruling elite over the years despite Senegal’s two political turnovers: in both cases, the new incumbents had served with the former regimes. Nevertheless, around elections marking the end of a president’s two terms (in 2000 and in 2012), momentary fragmentation of the ruling coalition occurred, and political competition opened in the fight around succession. In 2000, political competition was in part facilitated by the fact that structural adjustment programmes and reduced domestic revenue constrained Abdou Diouf’s abilities to maintain the ruling coalition through buying the support of, amongst others, important religious groups (Diop and Diouf, 1990;Koter, 2021). In 2012, Abdoulaye Wade’s attempt at securing an unconstitutional third term was countered by a strong popular movement and a united opposition. Following both elections, the new incumbents succeeded in reconsolidating the ruling coalitions and ensuring stability in the political settlement by clientelist means (Mbow, 2008; Kelly, 2012;Dumont and Kanté, 2019). These bouts of more competitive clientelism and subsequent reconsolidation are important for understanding the revenue bargaining with revenue providers in the informal economy and their brokers as well as the evolution of informal sector taxation in Senegal, as discussed by Bak in Chapter 12. 26 Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak Mozambique’s political settlement is characterized as weak-dominant with strong lower-level factions but weak excluded factions. The current Mozambican coalition has its roots in the peace deals that ended the longrunning civil war between the FRELIMO and the RENAMO resistance. The FRELIMO strategy was a mix of co-opting and dividing RENAMO factions, and this strategy has generally been deemed successful (Vines, 2018;Buur and Salimo, 2018). Under the Guebuza presidency (2005–2012), the coalition became strong-dominant, as Guebuza strengthened control over the party,relyingonapowerfulnarrativeof‘nationalunity’(Macuane, Buur,and Monjane, 2017). However, even if the regime is authoritarian and repressive, control of lower-level factions has weakened somewhat during the recent Nyusi leadership. One driver was the local elections in 2018, which served to strengthen the importance of these factions. In Chapter 4, von Schiller examines the implications of the relations between local and central governments for revenue bargaining in Mozambique. Togo is not included in the ESID’s PolSett dataset, but from our own analysis (Beach, 2018;Chapter 10), we categorize the political settlement in Togo as strong-dominant with weak excluded factions and rather weak lower-level factions. President Faure Gnassingbé has chosen a strategy of co-optation of the leaders of other political parties, and like his father before him, has often resorted to brutal oppression to ensure that electoral results lean in his favour. His father, Eyadéma Gnassingbé, came to power in a coup in 1967 and ruled as a dictator for 38 years. Eyadéma did face a national conference in 1991, which introduced space for political competition and opposition voices; however, he removed term limits shortly thereafter. Faure further manipulated the electoral system to favour his well-organized party, UNIR (l’UnionpourlaRépublique).Theparty,dominatedbyhispeopletheKabiyé, holds the majority of seats in parliament despite representing a lower proportion of the population. There are four main ethnic groups in Togo: the Ewe (and Minas), the Kabiye, the Moba, and the Kotokoli and Tem. The Ewe people live along the more economically advantaged coast in the south and dominate in both the civil service and commercial sectors, while the president’s clan dominates the military and law enforcement in Togo (Beach, 2018). In recent years, there has been increasing pressure for more democratic modes of governing, including the possibility of decentralizing and devolving some power. However, there has been no effective political openingsofar.InChapter10,Beachdescribestwoinstancesofrevenuebargaining in rural and urban Togo. The five case countries of this book have all seen an expansion in domestic revenue mobilization during the last two decades that, like elsewhere, Politicization of taxation and state–society reciprocity in Africa 27 has prompted many instances of revenue bargaining. Some of these are the focus of the case studies in this book. Our case countries are typical of lower-income countries in Africa, with economies largely not having undergone economic structural transformation, with a limited tax base, and with the ruling elites’ continuous need to maintain their ruling coalitions through distribution of resources and access to rents. The case studies set in these five countries provide new knowledge on when and how revenue bargaining emerges, evolves, and leads to fiscal contracts in clientelist political settlements across different degrees of ruling coalition fragmentation. Together, our case countries provide a foundation for comparative and generalizable observations of the politics of revenue bargaining in Africa. 1.7 Conclusion Domestic revenue mobilization has been increasing in Africa, and taxation has become more politicized since the turn of the millennium. However, there seem to be few signs of exchange-based fiscal social contracts emerging at the macro level, where citizens pay taxes in return for public service delivery and accountability. Nevertheless, recent developments in terms of increased political competition and civil society participation offer grounds for revisiting older insights about the potential role of taxation in state– society relations. Perhaps we have been looking in the wrong places for the political effects of taxation. Revenue bargaining, that is, negotiations related to revenue provision, arguably happens on the micro level between, on the one side, individual or groups of revenue providers and, on the other side, state actors or ruling elites. However, systematicand comparative research on micro-level revenue bargaining in Africa remains limited. This book attends to this gap and seeks to deepen our knowledge and understanding of the dynamics and politics of revenue bargaining in Africa. In doing so, this book contributes to two key literatures in the studies of African politics: the fiscal contract literature and the political settlement literature. In combining these approaches, we add a systematic view of politics to the fiscal contract theory while integrating a nuanced revenue dimension to the political settlement theory. Furthermore, a major contribution of this book is to operationalize these two primarily macro-level theories to facilitate micro-level analyses of revenue bargaining. The theoretical framework is elaborated in more detail in Chapter 2. The book then explores instances 34 Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak OECD. 2021. Africa: The Rising Debt Burden Highlights the Need for Further Progress in Domestic Revenue Mobilisation. Paris: OECD. URL: https://www.oecd.org/ tax/tax-policy/africa-the-rising-debt-burden-highlights-the-need-for-furtherprogress-in-domestic-revenue-mobilisation.htm (accessed 15 August 2022). Okunoye, B. 2019. ‘In Africa, a New Tactic to Suppress Online Speech: Taxing Social Media’. Retrieved on 17 November from https://www.cfr.org/blog/africanew-tactic-suppress-online-speech-taxing-social-media. Paler, L. 2013. ‘KeepingthePublicPurse: AnExperimentinWindfalls, Taxes, andthe Incentives to Restrain Government’. American Political Science Review 107(04): pp. 706–725. https://doi.org/10.1017/S0003055413000415. Prichard, W. 2015. Taxation, Responsiveness and Accountability in Sub-Saharan Africa. Cambridge: Cambridge University Press. https://doi.org/10.1017/ CBO9781316275511. Prichard, W. 2019. ‘Tax Politics and the Social Contract in Africa’. Oxford Research Encyclopedia of Politics.https://doi.org/10.1093/acrefore/9780190228637.013. 853. Prichard, W.,and V. vanden Boogaard. 2017. Norms, Powerandthe Socially EmbeddedRealitiesofMarket TaxationinNorthernGhana. AfricanStudiesReview60(1): pp. 171–194. Rakner, L. 2017. ‘Tax Bargains in Unlikely Places: The Politics of Zambian Mining Taxes’. The Extractive Industries and Society 4(3): pp. 525–538. Rakner, L., and S. Gloppen. 2003. ‘Tax Reform and Democratic Accountability in Sub-Saharan Africa’. In Beyond Structural Adjustment. The Institutional Context of African Development, edited by N. Van De Walle, N. Ball, and V. Ramachandran, 77–100. New York: PalgraveMacmillan. https://doi.org/10.1057/9781403981288_ 3. Reid, R. J. 2017. A History of Modern Uganda. Cambridge: Cambridge University Press. Reuss, A., and K. Titeca. 2017. ‘When Revolutionaries Grow Old: The Museveni Babies and the Slow Death of the Liberation’. Third World Quarterly 38(10): pp. 2347–2366. Reuters. 2018. ‘Uganda Police Use Teargas to Disperse Protest against Social Media Taxes’. 11 July. https://www.reuters.com/article/us-uganda-socialmediaprotests/uganda-police-use-teargas-to-disperse-protest-against-social-mediataxes-idUSKBN1K121S. Ross, M. L. 2004. ‘Does Taxation Lead to Representation?’ British Journal of Political Science 34(2): pp. 229–249. https://doi.org/10.1017/S0007123404000031. Rothchild, D., and N. Chazan, eds. 1988. The Precarious Balance: State and Society in Africa. Boulder, CO: Westview. Politicization of taxation and state–society reciprocity in Africa 35 Schneider, A. 2020. ‘Democratic Deepening and State Capacity: Taxation in Brazil and India’. In The Politics of Domestic Resource Mobilization for Social Development, edited by K. Hujo. Cham: Palgrave Macmillan: pp. 173–206 Schulz, N., and K. Kelsall. 2021. ‘The Political Settlements Dataset: An Introduction with Illustrative Applications’. ESID Working Paper No. 165. Dataset available at: The political settlements dataset: An introduction with illustrative applications – ESID (effective-states.org) (accessed 28 August 2023). Sharp, S., S. Sweet, and A. Rocha Menocal. 2019. Civil Society Engagement in Tax Reform. London: ODI Report. Sjursen, I. H. 2018. Essays on Behavioural Development Economics. PhD thesis, Department of Economics, Norwegian School of Economics. Smidt, H. 2018. ‘Shrinking Civic Space in Africa: When Governments Crack Down on Civil Society’. GIGA Focus Africa. No 4. Smith, G. 2021. Where Credit Is Due: How Africa’s Debt Can Be a Benefit, Not a Burden. London: Hurst and Company. Stasavage, D. 2005. ‘Democracy andEducationSpendinginAfrica’. AmericanJournal of Political Science 49(2): pp. 343–358. Therkildsen, O., and F. Bourgouin. 2012. ‘Continuity and Change in Tanzania’s Ruling Coalition: Legacies, Crises and Weak Productive Capacity’. DIIS Working Paper,2012:6. Copenhagen: Danish Institute of International Studies. Tilly, C. 1992. Coercion, Capital, and European States. AD 990-1992. Cambridge: Blackwell Publishers. Timmons, J. F. 2005. The Fiscal Contract: States, Taxes, and Public Services. World Politics, 57(4), 530–567. http://www.jstor.org/stable/40060117 Tripp, A. M. 2017. ‘In Pursuit of Autonomy: Civil Society and the State in Africa’. In Africa in World Politics, edited by J. W. Harbeson and D. Rothchild. New York: Routledge.: pp. 89–110. UN. 2021. Committee of Experts on International Cooperation in Tax Matters. October. UNU-WIDER. 2021. UNU-WIDERGovernmentRevenueDataset. Helsinki: UNUWIDER Available at: https://www.wider.unu.edu/project/government-revenuedataset. van den Boogaard, V., W. Prichard, R. Beach, and F. Mohiuddin. 2021. ‘Enabling Tax Bargaining: Supporting More Meaningful Tax Transparency and Taxpayer Engagement in Ghana and Sierra Leone’. Development Policy Review.https://doi. org/10.1111/dpr.12563. Vines, A. 2018. ‘Elite Bargains and Political Deals Project: Mozambique Case Study. Stabilization Unit. UK. 36 Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak Whitfield, L., O. Therkildsen, L. Buur, and A. M. Kjær. 2015. The Politics of African Industrial Policy: A Comparative Perspective. Cambridge: Cambridge University Press. Winckler Andersen, O., and O. Therkildsen. 2019. Can the SDGs in Low-income Countries Be Financed? And Should We Care? DIIS Woking Papers,2019, no 2. Copenhagen: Danish Institute for International Studies. 2 Unpacking revenue bargaining Triggers, processes, and outcomes Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær 2.1 Introduction In the introductory chapter, we offered an overview of the fiscal contract literature and argued that there is a need for more knowledge about the extent to which increased revenue mobilization leads to revenue bargaining and potentials for state–society reciprocity. In particular, we need to study systematically the micro-level politics of revenue bargaining, and to this end, weneedtotheorizeandstudythefiscalandpoliticalimportancethatrevenue provides. This chapter therefore develops a theoretical framework (see also Figure 1.1 in Chapter 1) to help us answer our research question, namely how and under what conditions revenue bargaining emerges, evolves, and leads to fiscal contracts. We begin by defining the main theoretical concept of revenue bargaining and distinguish it from that of fiscal contracts. Second, integrating fiscal contract theory and the political settlement approach, we disaggregate the concept of bargaining power, add the dimension of revenue providers’ fiscal importance, and theorize how it influences their relative bargaining position vis-à-vis ruling elites. Thereafter, we discuss how revenue providers, given their relative bargaining position, can influence policy and when revenue providers and ruling elites are able to reach agreements in the form of fiscal contracts. Finally, we discuss potential triggers of revenue bargaining, that is, theeventsoractorsthatkick-startrevenuebargaining.Thetheoreticalframework that we develop here is applied in this volume’s empirical Chapters 3–12 that analyse micro-level instances of revenue bargaining in our five case countries: Uganda, Tanzania, Togo, Mozambique, and Senegal (discussed in Chapter 1). We end this chapter with a brief discussion of the methodology applied across the contributions in this volume, followed by an introduction to each of the empirical chapters. Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær, Unpacking revenue bargaining. In: The Politics of Revenue Bargaining in Africa. Edited by: Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192868787.003.0002 38 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær 2.2 The concepts of revenue bargaining and fiscal contract In this book, we define revenue bargaining as implicit or explicit negotiations related to revenue provision between individual or groups of revenue providers and ruling elites (or representatives hereof). This definition differs from that of former literature in two ways. First, it delimits bargaining to the process of negotiation and distinguishes it from the potential outcome. In the fiscal contract literature, the process of revenue bargaining has often been implied and hence conflated with the outcome in the form of a contract (Bak 2019, 53). Earlier definitions have thus described revenue bargaining as ‘the exchange of revenues (for the state) for institutionalized influence over public policy (for citizens)’ (Moore, 2008, 36, emphasis added). We stress that revenue bargaining happens, in principle, prior to the exchange or, as we focus on, an agreement related to revenue provision. With our definition, we avoid the semantic and possibly analytical conflation of revenue bargaining (processes) with bargains (the potential outcomes, e.g. in the form of fiscal contracts). Second, we use the term revenue rather than tax bargaining (as employed, e.g., by Prichard, 2015 and van den Boogaard et al., 2021) because it, per definition, includes other sorts of income sources that are not strictly taxes, such as various sorts of levies, fees, royalties, or development aid. As Wilson Prichard (2019, 7) has argued, ‘a narrow focus in legally defined taxes will exclude large parts of how governance and services are actually financed in Africa’. As a research object, revenue bargaining can be difficult to observe. As Moore (2008, 37–38) suggests, bargaining includes: [a] wide range of types of (political) exchange, ranging from explicit haggling (ʻIf you do this, I will do thatʼ) to indirect, strategic, anticipatory interaction (ʻLet us announce more public spending on health now, in the hope that Parliament will be ready to accept an increase in VAT rates next yearʼ). Revenue bargaining can be direct and observable, such as when Ghanaians in 1999 protested against the 2.5% increase in the VAT rate and the government reacted by earmarking the won revenue the Ghana Education Fund (Prichard, 2015). In line with Moore’s description, we emphasize that revenue bargaining will also—and perhaps more often—be indirect, with no observable negotiation involving two parties with distinct interests. This was the case with the abolition of informal sector personal taxes in Uganda and Tanzania around election time (Kjær and Therkildsen, 2013). Here, the Unpacking revenue bargaining: Triggers, processes, and outcomes 39 governments abolished taxes because they were unpopular, and given that opposition candidates campaigned against them, the rulers would not risk losing rural votes by insisting the taxes remain. The ruling elites thus anticipated a loss of votes to the opposition, and the policy of abolishing the tax was based more on a perception than on a direct bargain with taxpayers. Hence, revenue bargaining could happen anywhere and anytime; as protests in the streets over new taxes, as op-eds in newspapers pushing for newtaxes,atcampaignmeetings,duringclosedconversationsinMPs’offices, as a part of legislative reform process, and at the market floor between vendors and tax collectors. The instances of revenue bargaining analysed in this book span the spectrum and allow us to make comparative observations about when and where revenue bargaining is likely to take place and how these conditions shape the outcomes. Across the chapters of this book, revenue bargaining includes a range of stakeholders. A revenue provider is any actor who contributes funds to the public budget including individual taxpayers, sector groups, local elites, multinational companies, and even international donors. On the state side, we see the involvement of heads of state, members of the government, and politicians elected for national or local offices, as well as bureaucrats. Some of these may be members of the ruling elite, that is, individuals who hold formal or informal positions of political power, and others who may not. For example, a cabinet minister can hold a formal position but might not be a member of the ruling elite given lack of political influence. On the other hand, a person may be a close adviser tothe president but without any formal position of power. Our definition of revenue bargaining is broad in the sense that negotiation simply must ‘relate to revenue provision’. Thereby, we recognize that revenue bargaining may not only concern positive exchanges of taxation for a return but may also result in revenue foregone (Moore, 2015). This includes situations that can be difficult to categorize as cases of revenue bargaining because there might not be a tax in place or a tax could have been abolished, such as in Kampala, Uganda where boda bodas, the motorbike taxis, have been exempted because they constitute an important support group for the ruling elite (Goodfellow, 2015). Judith Tendler (2002) called this type of arrangement ‘the devil’s deal’—you vote for me and I will not tax you—while Holland (2016) terms intentional non-enforcement of laws for the sake of maximizing votes ‘forbearance’. Other powerful groups may be able to avoid paying tax by influencing members of the government to get exemptions. Such exemptions have perceivable consequences for policy and fiscal outcomes. For example, the granting of tax exemptions reduces the overall tax 40 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær take and hence impedes the provision of public services. The influence of powerful groups may also result in the ruling elite prioritizing the provision of club goods, for example power supply for a specific industry rather than, say,countrywideruralelectrificationprogrammes.Thus,specificinstancesof revenue bargaining can result in micro-level contracts that are counterproductive to the macro-level achievement of accountability and broad-based development. To deepen the understanding of when revenue bargaining leads to microlevel fiscal contracts, the two must be conceptually and empirically distinguishable. Where revenue bargaining denotes the processes of implicit or explicit negotiations, fiscal contracts are the potential outcomes of these processes; they occur in cases where the involved parties are able to reach an agreement or compromise. In the literature on taxation and governance, the fiscal contract has most often been associated with a loose conception of a ‘healthy’, broad-based exchangerelationshipbetweenstateandsocietyinwhichtaxesareexchanged forpublicgoodsandpoliticalaccountability(Moore, 2004;2008;Martin and Prasad, 2014). Since the analytical interest of this book is revenue bargaining at the micro level, including its potential outcomes, we need a fiscal contract understandingonthesameanalytical level. Levi(1988)andTimmons(2005) represent a strand of the fiscal contract literature that describes micro-level fiscal relations between state and revenue providers. They both argue that governments strike different bargains with different groups of actors. Specifically, Levi (1988, 12) suggests that in a country, several (fiscal) contracts can exist in parallel, comprising the different deals a ruler enters into with different groups of societal actors in order to stay in power. Such microlevel fiscal contracts denote direct exchanges between a (group of) revenue provider(s) and the ruling elite, and it allows us to capture the immediate outcomes of micro-level instances of revenue bargaining. Specifically, we define micro-level fiscal contracts as explicit or implicit agreements between ruling elites (or representatives hereof) and either individual or a group of revenue providers related to revenue provision. Four points of elaborations are in order. First,themicro-levelfiscalcontractdoesnothavetobeabalancedcompromise in absolute terms that materializes in an equal exchange (Hyden, 1992). The agreement can manifest itself as much in mutual recognition and commitment as in material terms (Hyden, 1992; see also Levi and Sacks, 2009). Second, the contribution by revenue providers can take many forms. The contributioncanbe formal, public(mostoftentax) payments, supporting the ruling elite via the national budget, or informal, private payments to ruling Unpacking revenue bargaining: Triggers, processes, and outcomes 41 elites, for example in the form of campaign financing rather than through state coffers. Third, the ruling elite side of the contract could include positive actions, such as goods or service provision, or negative actions including non-decisions, forbearance, and ‘devil’s deal’ (Tendler, 2002;Holland, 2016; Piracha and Moore, 2016). In other words, positive or negative does not refer to a value judgement but rather to positively delivering a public service or negatively refraining fromtaxing and/or delivering. Here, our understanding of the fiscal contract differs from both Levi (1988) and Timmons (2005), who onlyconsiderexchangesofrevenueforbenefitsorpoliticalinfluence.Fourth, the contract does not have to be in line with the initial proposal. Theoretically, the implication could be larger as well as smaller revenue mobilization by the government. Given the economic and political contexts (as theorized in detail below), agreements are, however, more likely to involve a compromiseofalowertaxburdenfortherevenueproviderandthereforelessrevenue mobilization by the government. Whether formal or informal contributions, positive or negative actions, or an agreement of lower tax payments, the agreementsconstitutecontractualrelationsbetweenrulingelitesandrevenue providers with implications for the shape of the tax system, fiscal outcomes, and for state–society reciprocity. While a case study approach does enable one to get closer to the often difficult-to-observe attributes of an explicit or implicit agreement, mutual recognition, and commitment, it is not an easy task. We suggest several observable implications of micro-level fiscal contracts. One is an exchange. Evidence of an exchange related to revenue provision, whether positive or negative, can be taken at least as an indication of a micro-level fiscal contract. Another approach is to identify when revenue bargaining ends and inquire why it ended and what happened after it did so. Further, a fiscal contract can be implied from contract parties’ expressions of expectations of and commitment to each other. Whereas fiscal contracts are always preceded by revenue bargaining, revenue bargaining does not always end in agreement and a fiscal contract. They might end in non-exchanges. To illustrate different non-contractual outcomes, we can draw on recent efforts to levy a tax on social media across the African continent. In Benin, the government’s proposal to tax certain types of social media prompted widespread mobilization of social media users, both online under the hashtag #TaxePasMesMo (‘Don’t Tax My MegaBytes’) and taking to the streets (Okunoye, 2019). In response, the President withdrew the tax. In Uganda, the social media tax was also met with protests; however, here the government ignored the opposition and went ahead with the taxes without committing to a return. Both cases constitute instances of revenue 42 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær bargaining that did not lead to fiscal contracts, though in two different ways. In the Beninese case, the tax was withdrawn and thus no exchange established. The Ugandan case is, at face value, a case of coercive taxation without an exchange. However, in reaction, many Ugandans decided to quit the targeted internet services (or use VPNs). To the extent that this is seen as tax resistance, one could argue that the revenue bargaining is ongoing. Ultimately, whether a micro-level fiscal contract is present or not is an empirical question, and this book’s empirical chapters provide illustrations of the many kinds of fiscal contracts to which revenue bargaining might lead. Analysing under which conditions micro-level fiscal contracts emerge is an instrumental means to determining the implications of the politicization of taxation. 2.3 The politics of bargaining Having distinguished between the concepts of revenue bargaining and fiscal contracts, we also need to be able to see how power relations affect the bargaining processes. Political settlement theory offers a tool to grasp the relative bargaining power of revenue actors. A political settlement refers to the distribution of power among groups and institutions in a society based on which the ruling elites create coalitions of different key political factions (Khan,2010;Whitfieldetal.,2015).Rulingcoalitionsaremaintained,andthe ruling elites’ hold on power secured, by giving powerful groups or factions in society access to rents such as government contracts, land rights, or tax exemptions (North et al., 2013;Kjær, 2015;Bak and Therkildsen, 2022). The holding power of these groups or individuals refers to their ability to engage inandsurviveconflictsand,byimplication,toimposetheirinterestsonother groups or the state. In the following, we first discuss the power of the ruling elite and revenue providers, respectively. Subsequently, as actual bargaining power must be understood as a relational concept whereby the two parties engage in negotiations, we further elaborate on the relative bargaining positions of the ruling elite and revenue providers. The model of bargaining positions emphasizes that negotiating parties go into the bargaining process from a certain position within the political settlement. From the outset, the groups have certain resources and holding power that may put them on either a better, a weaker, or an equal footing with respect to those on the other side of the bargaining table. Initial bargaining positions, even if they change along the way, may thus matter for the fiscal contract outcomes. Unpacking revenue bargaining: Triggers, processes, and outcomes 43 2.3.1 The bargaining power of ruling elites Literature onAfrican politics tendstoviewruling elitesasratherautonomous policymakers because, as they often deal with weak political opposition parties and civil society organizations, they are rarely constrained in making policy decisions (van de Walle, 2001). However, rulers do not act in isolation; they are responsive to the groups with strong holding power, whose support they need in order to remain in power (Khan, 2010;North et al., 2013). The power of the ruling elite depends on the political, economic, and administrative resources in the ruling elite’s possession (Levi, 1988, 19), which again are derived from the political settlement. Ruling elites are politically resourceful if factions opposing them are not strong—and if lower-level factions supporting the ruling elite are weak, merely complying and supporting rather than acting as veto-players from within (see Khan, 2010;Whitfield et al., 2015). On the other hand, ruling elites are less powerful when such factions are strong and therefore need to be accommodated. Hence, decisions to increase revenue can be politically costly if taxes target important constituents. A political settlement perspective shows that given the need to maintain legitimacy and support, rulers may not be able to maximize revenue, as is often assumed. This follows Levi’s (1988, 10) argument that rulers’ ability to maximize revenue is constrained by their need for supportive constituents. Further, as Fjeldstad and Therkildsen (2020, 39) point out, while macro-level agreements around some level of redistribution from rich to poor through taxation is rare in West and East Africa, some kind of redistribution does occur as rulers distribute resources in exchange for political loyalty to maintain stable ruling coalitions. Withregardto economic resources, the availability of non-taxrevenuesuch as aid or natural resource revenues can change the autonomy of the ruling elite, simply because they depend less on domestic taxpayers and can afford to disregard their interests. Although, as regards aid, ruling elites instead become dependent on aid agencies (Ulriksen 2013). Aid dependency weakens the bargaining power of the ruling elites vis-à-vis donors whereas the availability of incomes from extractives would strengthen bargaining power towardstaxpayersanddonorsalike(ibid.).Relatedtoeconomicresourcesare also the costs of bargaining, including the costs of acquiring information, of the actual bargaining, and of subsequently implementing the policy resulting from the bargain (Levi, 1988, 27). Such costs will also influence ruling elites’ bargaining power. 50 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær The focus on expansion of domestic revenue mobilization as a trigger of state–society interactions originates in the European state-building experiences. Non-democratic rulers introduced new taxes in order to mobilize resources to win wars, and in return, they needed to provide openings for political representation (Schumpeter, 1991;Tilly, 1992). Negotiations over revenue in African countries operate in a post-colonial political context characterized by the absence of inter-state war, hybrid institutions, and economies in which the majority remains employed in low-technology subsistence agriculture (Jackson and Rosberg, 1982;van de Walle, 2001;Moore, 2004;Bräutigam, 2008;North et al., 2013). Thus, revenue pressure and the drive to collect revenue in Africa come from many factors other than interstate war. One such factor is rulers’ need to maintain political stability in a situation of what Doner et al. (2005) call ‘systematic vulnerability’. Systemic vulnerability is present under a combination of resource scarcity, perceived threats to the ruling elite, and the need for resources to maintain political stability.While systemicvulnerabilityismoreexpedientinasituationcharacterized by external security threats, vulnerabilities can also arise from a fiscal crisis, a decline in foreign aid, or a shortage of opportunities to take up loans. Many African elites have experienced rising political costs of staying in power. Most African countries have institutionalized, regular, and relatively free and fair elections (Bleck and van de Walle, 2018). Elections offer an impetus to collect revenue to win votes through public service provision (Stasavage, 2005). Running election campaigns is also costly, so incumbent ruling elitesmaywanttoraiserevenue fortheir campaignpurposes, although the converse scenario is also plausible, namely that elections lead to declining revenue as exemptions may be granted in return for support (Bak and Therkildsen, 2022). Elections may also offer the opportunity for a tax to be politicized, as happened with the informal sector taxes in Uganda and Tanzania (Kjær and Therkildsen, 2013). Finally, and less directly connected to elections, threatsfrom,andneedstooffersidepaymentsto,importantgroups (withinoroutsidetherulingcoalition)couldinstigate a sense ofvulnerability and hence motivate the ruling elite to search for revenue (Donor et al., 2005; Whitfield et al., 2015). Summing up, ruling elites in Africa do experience revenue pressure, which is an important driver in governments’ move to increase resource mobilization. However, given that most African states are economically vulnerable, their fiscal need is constant, though it may be latent. Therefore, it alone cannot explain the emergence of revenue bargaining. To broaden our knowledge hereof, we need to look for the immediate factors that trigger revenue bargaining. Unpacking revenue bargaining: Triggers, processes, and outcomes 51 Such factors could include actions by state actors including government officials, local politicians, and tax administration employees, such as an announcement or an adoption of a new tax. But not all attempts to increase taxation leads to revenue bargaining. Therefore, some empirical chapters explore the conditions under which this is more likely to happen. Triggers could also be events such as elections or a change in circumstances that renderanalreadyimplementedtaxthesuddenfocal pointofrevenuebargaining. The tax administration may begin to enforce a tax policy hitherto left unimplemented. Citizens may receive new information about the taxes they pay or learn about a corruption scandal, prompting a resistance towards paying taxes. Similarly, the entry of civil society organizations promoting human rights and tax awareness could trigger a bargaining process. These are just a few of the broad variety of triggers and conditions under which revenue bargaining emerges that are explored in this book’s empirical chapters. 2.5 The bookʼs methodological approach and introduction to chapters The theoretical framework presented and discussed in this chapter has as its point of departure the political settlement and fiscal contract literatures, and is based on a synthesis of the two. However, the purpose has not been to build a comprehensive theory of the politics of revenue bargaining to be tested in the book’s case countries. Instead, we take an iterative approach to ourinquiries.Wehaveconceptualizedkeynotionsandtheorizedtotheextent possible a priori. The theory, the concepts, and the theoretical framework were then continuously revisited and discussed in light of emerging findings at workshops of the Political Settlement and Revenue Bargains in Africa Project,andatconferences,includingadesignatedEuropeanConsortiumfor Political Research joint session in Cyprus, where contributors were present. Within this collaboratively developed framework, the book’s chapters take an explorative approach to uncover new empirical knowledge of the politics and dynamics of revenue bargaining, which in turn allows us to deepen our understandinghereof and to build a micro-level theoretical foundation of the fiscal contract theory. To develop a comprehensive understanding of the politics of revenue bargaining at the micro level, we argue that it is imperative to study a large variation of cases. The book includes both cases where revenue bargaining does unfold and cases where revenue bargaining could have been expected 52 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær but did not occur. Different chapters focus on different aspects of revenue bargaining in their case studies; these include whether and when revenue providers mobilize, whether they are able to get state actors to negotiate, how the bargaining evolves, or what the outcome is. One example is studying how differencesin bargaining power between different groups of revenue providersinfluencetheoutcomeoftheirrevenuebargainingwithstateactors. This is done in Chapter 7 where the bargaining power of agricultural sector actors in Uganda differs depending on their position in relation to the ruling coalition and whether they are able to join forces with other organizations. Another example focuses on instances of collective action prompted by taxrelated grievances and whether these actually lead to revenue bargaining, as inChapter3whereinformalwomentradersin Dar es Salaam,Tanzania, even when organized felt unable to influence whether they were taxed or if they received a return. A third example is whether and how the presence of third parties influences the processes and outcomes of different cases of revenue bargaining which is the focus of Chapter 10 in which local chiefs in Togo play a large role in the bargaining processes between revenue officials and taxpayers. Most of the cases have been selected based on an understanding of the contextual conditions of the revenue-bargaining instances, which in turn helps us to understand and explain variations in the processes and outcomes. All the book’s contributions build on extensive field research, including interviews with concerned revenue providers, relevant state actors, and experts. Some draw as well on participant observation data (Chapters 3 and 10), original survey data (Chapters 4 and 9), or data from primary and secondary sources such as news articles, reports, and legal texts (Chapters 5,6, 7,8,11, and 12). Based on extensive empirical knowledge of the cases combined with original collected data, the authors present the cases of revenue bargaining in qualitative narratives of different forms. This is an instrumental means of uncovering the dynamics of revenue bargaining, how positions and relations of power may change, and whether and under what conditions instances of revenue bargaining lead to some sort of agreement in the form of a (micro-level) fiscal contract. Employing the explorative approach has meant that some case studies are so-called negative cases of fiscal contracts where revenue bargaining occurs but does not lead to an agreement between revenue providers and the ruling elite. To mention one, revenue bargaining between the Tanzanian government and international aid agencies does not lead to an agreement around payment of the social protection programme (Chapter 11). Such a negative case is helpful, methodologically, to rule out spurious causes, and empirically Unpacking revenue bargaining: Triggers, processes, and outcomes 53 it appears to serve as initial engagements, which in some cases shape or directly lead to future revenue bargaining and build reciprocity. We now briefly introduce the empirical Chapters 3–12 of this book. In Chapter3, AneEdslevJacobsendrawsonextensivefieldresearchamong informal women traders in Dar es Salaam, Tanzania. Edslev Jacobsen finds that even if these women are taxed and do experience tax-related grievances, they very often do not act collectively because of their already marginalized position. Only with the aid of an NGO are they able to organize and initiate a dialogue with local government actors. In Chapter 4, Armin von Schiller examines what may lead to revenue bargaining between Mozambican municipalities and their citizens, based on original interview data. He finds that revenue pressure plays a large role and that opposition municipalities face stronger revenue pressure than others because of the unpredictability of central transfers. This incentivizes them to bargain with citizens over revenue in consensual rather than conflictual ways. In Chapter 5, Marianne S. Ulriksen, Jamal Msami, and Lucas Katera examine four different instances of revenue bargaining in Tanzania, based on in-depth interviews with key actors. In their analysis, they focus on the different arenas of bargaining and the strategies employed by the negotiating parties. They find that revenue providers can succeed in getting the government to concede to lower tax payments, particularly when bargaining is moved to arenas that enable the building of structured, repeated, and trusted relationships. In Chapter 6, Lise Rakner and Odd-Helge Fjeldstad examine the 2014 VAT Act reform in Tanzania and the role of lobbying in the bargaining process, based on document analysis and original interviews with involved parties. They find, among other things, that by creating alliances, business associations and other lobby groups worked purposefully and managed toshape the VAT act in accordance with their own interests. In Chapter 7, Anne Mette Kjær and Clayton Arinanye present findings on seven instances of bargaining concerning Uganda’s agricultural sector. Based on extensive field observations, interviews, and policy documents, they show that agricultural sector actors succeed more when they protest against taxes than when they try to lobby for improved service delivery. In the former case, it is easier for them to ally with other actors and easier for the ruling elite to give concessions. In Chapter 8, Dan Ngabirano examines revenue bargaining in the judicial arena in his study of how the Ugandan government took a multinational oil 54 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær company tocourt.Thegovernmentwonand thecorporationwasmadeliable topayalargesumofcapitalgainstax.Basedonananalysisoflegaldocuments combinedwithinterviews,Ngabirano’schapter showsthatrevenueproviders who we know to be economically important for the ruling elites have a weak bargaining position because they are not politically important. In Chapter 9, Moses Khisa, Jamal Msami, and Ole Therkildsen examine the connection between campaign financing and tax exemptions in Uganda and Tanzania through original survey data and interviews. They argue that campaign finance gives donors of such finance more bargaining power than tax paying does. In Chapter10,RachelBeachexaminestwoinstancesofrevenuebargaining in Togo based on extensive fieldwork and participatory observation within the Togolese revenue authority. She finds that a third party’s intervention can bolster as well as invert the bargaining power and position of the revenue authorities in important ways. In Chapter 11, Marianne S. Ulriksen, Flora Myamba, and Constantine George explore the revenue-bargaining process between the international aid agencies—the current revenue providers of Tanzania’s cash transfer programme—and the Tanzanian government. They pay particular attention tothepreferencesofnegotiatingpartiesandhowtheparties seektomaximize their own interests in the negotiations, and they find that even if the international aid agencies provide substantial revenue, the agencies are unable to fully impose their preferred interests on the Tanzanian government. In Chapter 12, the book’s final empirical chapter, Ane Karoline Bak offers a historical analysis of revenue bargaining between the Senegalese ruling elites and the informal commercial sector. Bak finds that the evolution in taxation of the sector can be explained in part by the relative bargaining positions that shift over time depending on the composition of the ruling coalition and the changing bargaining power of the revenue providers and, in particular, of their broker UNACOIS. The chapters all draw on the theoretical framework developed and outlined in this chapter to examine instances of revenue bargaining. Though the book’s chapters focus on different parts of the revenue-bargaining processes, theydooverlapinmultipleways,allowingcomparativeand cross-case, crosssector, and cross-country observations to be made. When relevant and possible, cross-chapter comparisons are referenced in the respective chapters. Not least given the shared theoretical framework, we are able to draw out with confidence some broadly generalizable patterns in the conclusion (Chapter 13). Beyond that, the concluding chapter discusses comprehensively the Unpacking revenue bargaining: Triggers, processes, and outcomes 55 collective answers that the chapters provide to our research question of how and under what conditions revenue bargaining emerges, evolves, and leads to fiscal contracts. 2.6 Conclusion Over the past two decades, domestic revenue mobilization has been expanding and taxation has become politicized across Africa. Concurrently, a literature studying the political outcomes of taxation has blossomed, but found few signs of the expected exchange-based fiscal social contracts between state and society. Despite the proliferation of fiscal contract research, the dynamics and politics of revenue bargaining remain undertheorized. As discussed in Chapter 1, we therefore need to develop a theoretical framework that helps us move from the macro to the micro level and allows us to embed the analysis of revenue bargaining within the economic and political power structures. It hasbeen thepurpose ofthischapter topresentatheoreticalframeworkthat allows us to study the politics of revenue bargaining through case studies of micro-level instances hereof. This theoretical framework is based on the existing fiscal contract theory and the political settlement approach and provides two major contributions. First, the framework contributes to the fiscal contract literature by defining and unpacking the concept of revenue bargaining and, importantly, distinguishing it from its potential outcome: micro-level fiscal contracts. This allows us to separate processes from outcomes, and analyse the conditions that shape each of them as well as their relationship. Second, the framework facilitates bringing politics into the analysis of revenue bargaining. We theorize that both revenue providers’ organizational importance and their fiscal importance for ruling elites determine the strength of their bargaining positions. We include a heuristic device to identify bargaining parties’ initial bargaining position as a point of departure for exploring how and under what conditions revenue providers are able to promote their demands in negotiations related to revenue provision. Equipped with this theoretical framework, the empirical chapters that follow take an explorative approach to understanding processes of revenue bargaining. Besides examining the role of power, the empirical chapters also go beyond the common studies of revenue bargaining by paying attention to the actual immediate triggers besides the fairly static revenue pressure, as well as leaving the imaginary bargaining table behind and going to where instances of revenue bargaining actually take place. 56 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær References Bak, A. K. 2019. When the Fiscal Social Contract Is Not about Tax: Understanding the Limited Role of Taxation in Social Accountability in Senegal. PhD dissertation, Department of Political Science, Aarhus University. Aarhus: Politica. Bak, A. K., and O. Therkildsen. 2022. ‘Democratisation in Tanzania: No Elections without Tax Exemptions’. Journal of Eastern African Studies 16(1): pp. 47–67. Bates, R. 1981. Markets and States in Tropical Africa: The Political Basis of Agricultural Policies. Berkeley: University of California Press. Behuria, P., L. Buur, and H. Gray. 2017. ‘Studying Political Settlements in Africa’. African Affairs 116(464): pp. 508–525. Bleck, J., and N. van de Walle. 2018. Electoral Politics in Africa: Continuity in Change. New York: Cambridge University Press. Bräutigam, D. 2008. ‘Introduction: Taxation and State-Building in Developing Countries’. In Taxation and State-Building in Developing Countries, edited by D. Bräutigam, O.-H. Fjeldstad, and M. Moore, pp. 1–33. Cambridge: Cambridge University Press. Doner, R., B. Ritchie, and D. Slater. 2005. ‘Systematic Vulnerability and the Origins of Developmental States: Northeast and Southeast Asia in Comparative Perspective’. International Organization 59(2): pp. 327–361 Evans, P. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton, NJ: Princeton University Press. Fjeldstad, O.-H., and O. Therkildsen. 2008. ‘Mass Taxation and State–Society Relations in East Africa’. In Taxation and State-Building in Developing Countries, edited by D. Bräutigam, O.-H. Fjeldstad, and M. Moore, pp. 114–135. Cambridge: Cambridge University Press. Fjeldstad, O.-H., and O. Therkildsen. 2020. Implications of the Covid-19 Pandemic for Revenue Generation in Poor African Countries. DIIS Working Paper 2020: 13. Copenhagen: DIIS, Danish Institute for International Studies. Goodfellow, T. 2015. ‘Taming the “Rogue” Sector: Studying State Effectiveness in Africa through Informal Transport Politics’. Comparative Politics 47(2): pp. 127–147. Grindle, M. S. 2004. ‘Good Enough Governance: Poverty Reduction and Reform in Developing Countries’. Governance: An International Journal of Policy, Administration and Institutions 17(4): pp. 525–548. Holland, A. 2016. ‘Forbearance’. American Political Science Review 110(2): pp. 232–246. Hyden, G. 1992. ‘Governance and the Study of Politics’. In Governance and Politics in Africa, edited by M. Bratton and G. Hyden, pp. 3–26. Boulder, CO: Rienner. Unpacking revenue bargaining: Triggers, processes, and outcomes 57 Jackson, R. H., and C. G. Rosberg. 1982. Personal Rule in Black Africa: Prince, Autocrat, Prophet, Tyrant. Berkeley, CA: University of California Press. Khan, M. H. 2010. Political Settlements and the Governance of Growth-Enhancing Institutions. London: SOAS Working Paper. Khan, M. H. 2012. ‘Governance and Growth: History, Ideology, and Methods of Proof’. In Good Growth and Governance in Africa, edited by Akbar Noman, Kwesi Botchwey, Howard Stein, and Joseph E. Stiglitz, pp. 51–79. Oxford: Oxford University Press. Kjær, A. M. 2009. ‘Sources of Local Government Extractive Capacity: The Role of Trust and Pre-Colonial Legacy in the Case of Uganda’. Public Administration and Development 29(3): pp. 228–238. https://doi.org/10.1002/pad.532. Kjær, A. M. 2015. ‘Political Settlements and Productive Sector Policies: Understanding Sector Differences in Uganda’. World Development 68: pp. 230–241. https:// doi.org/10.1016/j.worlddev.2014.12.004. Kjær, A. M., and O. Therkildsen. 2013. ‘Elections and Landmark Policies in Uganda and Tanzania’. Democratization 20(4): pp. 592–614. Leonard, D. K. 2010. ‘Pockets of Effective Agencies in Weak Governance States: Where Are They Likely and Why Does It Matter?’ Public Administration and Development 30(2): pp. 91–101. Levi, M. 1988. Of Rule and Revenue. Berkeley, CA: University of California Press. Levi, M., and A. Sacks. 2009. ‘Legitimating Beliefs: Sources and Indicators’. RegulationandGovernance3(4):pp. 311–333. https://doi.org/10.1111/j.1748-5991.2009. 01066.x. Martin, I. W., and M. Prasad. 2014. ‘Taxes and Fiscal Sociology’. Annual Review of Sociology 40(1): pp. 331–345. https://doi.org/10.1146/annurev-soc-071913043229. Moore, M. 2004. ‘Revenues, State Formation, and the Quality of Governance in Developing Countries’. International Political Science Review 25(3): pp. 297–319. https://doi.org/10.1177/0192512104043018. Moore, M. 2008. ‘Between Coercion and Contract: Competing Narratives on Taxation and Governance’. In Taxation and State-Building in Developing Countries: Capacity and Consent, edited by D. Bräutigam, O.-H. Fjeldstad, and M. Moore, pp. 34–63. Cambridge: Cambridge University Press. Moore, M. 2015. ‘Tax and the Governance Dividend’. In Perspectives on Politics, Production and Public Administration in Africa, edited by A. M. Kjær, L. Engberg-Pedersen, and L. Buur, pp. 159–172. Copenhagen: Danish Institute for International Studies. Msami, J., L. Katera, and M. Ulriksen. 2022. Bargaining Strategies in Recent Tax Reforms in Tanzania. Working paper 16-03-AU. Aarhus: Department of Political Science, Aarhus University. 58 Marianne S. Ulriksen, Ane Karoline Bak, and Anne Mette Kjær North, D. C, J. J. Wallis, S. B. Webb, and B. R. Weingast. 2013. In the Shadow of Violence: Politics, Economics, and the Problems of Development. Cambridge: Cambridge University Press. Okunoye, B. 2019. ‘In Africa, a New Tactic to Suppress Online Speech: Taxing Social Media’. Net Politics blog. 10 January. https://www.cfr.org/blog/africa-new-tacticsuppress-online-speech-taxing-social-media. Piracha, M., and M. Moore. 2016. ‘Revenue-Maximising or Revenue-Sacrificing Government? Property Tax in Pakistan’. The Journal of Development Studies 57(2): pp. 209–225. Prichard, W. 2015. Taxation, Responsiveness and Accountability in Sub-Saharan Africa. Cambridge: Cambridge University Press. Prichard, W. 2019. ‘Tax Politics and the Social Contract in Africa’. Oxford Research Encyclopedia of Politics.https://doi.org/10.1093/acrefore/9780190228637.013. 853. Prichard, W., and V. van den Boogaard. 2017. ‘Norms, Power and the Socially Embedded Realities of Market Taxation in Northern Ghana’. African Studies Review 60(1): pp. 171–194. https://doi.org/10.1017/asr.2017.3. Roll, M., ed. 2014. The Politics of Public Sector Performance: Pockets of Effectiveness in Developing Countries. London: Routledge. Schumpeter,J. A. 1991. ‘The Crisisofthe Tax State’. In JosephA. Schumpeter: The Economics and Sociology of Capitalism, edited by R. Swedberg, pp. 99–140. Princeton, NJ: Princeton University Press. https://doi.org/10.2307/j.ctv173f01t.5. Stasavage, D. 2005. ‘Democracy andEducationSpendinginAfrica’. AmericanJournal of Political Science 49(2): pp. 343–358. Tangri, R., and A. M. Mwenda. 2013. The Politics of Elite Corruption in Africa. New York: Routledge. Tangri, R., and A. M. Mwenda. 2019. ‘Change and Continuity in the Politics of Government–BusinessRelationsinMuseveni’s Uganda’. Journal ofEastern African Studies 13(4): pp. 678–697. https://doi.org/10.1080/17531055.2019.1655879. Tendler, J. 2002. ‘Small Firms, the Informal Sector, and the Devil’s Deal’. IDS Bulletin 33(3): pp. 1–15. https://doi.org/10.1111/j.1759-5436.2002.tb00035.x. Tilly, C. 1992. Coercion, Capital and European States,AD 990–1992. Oxford: Blackwell. Timmons, J. F. 2005. ‘The Fiscal Contract: States, Taxes and Public Services’. World Politics 57(4): pp. 530–567. Ulriksen, M. S. 2013. ‘The Politics of Social Protection Expenditure and Financing in Southern Africa’. Development Southern Africa 30(1): pp. 39–53. van de Walle, N. 2001. African Economies and the Politics of Permanent Crisis, 1979– 1999. Cambridge: Cambridge University Press. Unpacking revenue bargaining: Triggers, processes, and outcomes 59 van den Boogaard, V., W. Prichard, R. Beach, and F. Mohiuddin. 2021. ‘Enabling Tax Bargaining: Supporting More Meaningful Tax Transparency and Taxpayer Engagement in Ghana and Sierra Leone’. Development Policy Review 40(1): p. e12543. https://doi.org/10.1111/dpr.12563. Whitfield, L., O. Therkildsen, L. Buur, and A. M. Kjær. 2015. The Politics of African Industrial Policy. Cambridge: Cambridge University Press. 66 Ane Edslev Jacobsen 3.3 Methodology This comparative case study has been conducted in Tanzania’s largest city, Dar es Salaam, a rapidly growing city with a sizeable informal economy and a large share of informal women workers. By limiting the focus to Dar es Salaam, multiple macro-level historical, social, political, cultural, and economic factors are held constant. At the same time, the study is carried out in seven markets across three municipalities in Dar es Salaam to exploit the variations in taxation and capacity for collective action. Across the municipalities, taxation of market traders varies because the local government in each municipality has the mandate to formulate regulations on market taxation within the municipality. Capacity for collective action varies across the different markets because of the markets’ diverse social composition and organizational landscape. The researched market sites are Mtambani, Magomeni, and Tandale (Kinondoni Municipality); Ilala, Kisutu, and Mchikichini (Ilala Municipality); and Ferry (Kigamboni Municipality). The study is based on interviews andparticipantobservationswith22womenmarkettradersin Dar esSalaam in March–April 2018. Four focus group discussions (FGDs) of four people each and seven in-depth interviews (IDs) were conducted. The interviewees differ in age, educational level, family status, commodity traded, and work history. The participant observations were conducted at a VICOBA (village community bank) group meeting in Ilala and an inauguration event for a VICOBA group in Mchikichini (O1–O2). Cases in which the women reported to collectively having taken united action with the aim of pressuring authorities to accede to their demands are denoted cases of collective action. Cases in which the women have joined or founded an organized group without the intent to influence authorities are denoted cases of collective organizing. Cases in which neither of the above was identified are denoted negative cases. 3.4 Findings 3.4.1 Patterns of collective action and collective organizing Table 3.1 provides an overview of the identified cross-market patterns in collective action and collective organizing. Of the 22 interviewed women, 12 women had never engaged in collective action or collective organizing (marked by ‘–’ in Table 3.1). Of these 12 women, 11 worked in markets in Kinondoni Municipality, and one worked in Ilala market. The remaining We pay, we act? 67 10 interviewed women were all engaged in collective organizing. Spread across five markets, these women reported engaging in different marketbased organized groups. A woman in Tandale participated in a market-based upatu group (informal self-lending group) (ID1). In Mchikichini market, two women were members of VICOBA groups (self-loaning groups), and one of these was also a member of cooperative society for kiosk traders in the market (ID2–ID3, O1). Two women in Mchikichini market were members of the National Women Traders Association (NWTA), which, at the time of research, was in the process of registering as a formal organization (NWTA1).¹ In Ilala market, a woman was a member of both a VICOBA group and the NWTA (ID7, NWTA2, O2). In Kisutu market, four women were members of a VICOBA group and a market-based women’s association accessible for all women in the market (FGD4). Lastly, in Ferry market, two women were VICOBA group members and also participated in a women’s platform, a women-empowerment group in the market and surrounding neighbourhood initiated by the municipal development department (ID4–ID5). Table 3.1 Patterns of collective organizing and collective action Municipality Market Collective organizing Collective action Data source Kinondoni Mtambani – – FGD1 Magomeni – – FGD2 Tandale – – FGD3 Upatu – ID1 Ilala Mchikichini VICOBA – ID3 VICOBA Inauguration event ID2, O2, NWTA∗– NWTA1 Kiosk cooperative society ID3 Ilala – – ID6 VICOBA Electricity fee protest O1, ID7 NWTA – NWTA2 Kisutu VICOBA – FGD4 Women’s association – Kigamboni Ferry VICOBA – ID4–ID5 Women’s platform – Note: ∗NWTA is an abbreviation for the National Women Traders Association. FGD stands for focus group discussion, ID is in-depth interview, and O is observation. ¹ As of 2022, the National Women Traders Association is formally registered as an association. 68 Ane Edslev Jacobsen Onlytwocasesofcollectiveaction(relatedtotaxation)wereidentified.The first case of collective action, the electricity fee protest, refers to an event in Ilala market in the beginning of 2018 when the electricity supply was unstable, with the traders sometimes experiencing hour-long power cuts. At this time, the market leadership, headed by a municipal representative, decided to raise the weekly electricity fee in Ilala market by TZS 500 (Tanzanian shillings)forallstalltraders.Atfirst, mostofthetradersaccepted this because they expected it to be accompanied by electricity improvements. However, as power cuts kept occurring, the traders became discontented with the lack of improvement and demanded the fee be lowered to the previous level. Therefore, several of the VICOBA groups in Ilala market jointly protested to the market leadership. After a few weeks, the market leadership agreed to lower the electricity fee. A woman that had headed the protest reported that, in her opinion, it was the united front by the VICOBA groups that had forced the market leadership to accede to the traders’ demands (ID7, O1). Interestingly, this case illustrates how collective organizing (VICOBA groups) can metamorphose intoavehicleforcollective actiondirectedtowardsmunicipal authorities. The second collective action case, the inauguration event, was a one-time event held by a VICOBA group in Mchikichini market (Ilala Municipality). The women wanted to showcase the potential of VICOBA groups for women’s empowerment and to draw the local authorities’ attention to the challenging economic conditions and gender-related issues experienced by women traders in Mchikichini. For this purpose, the women had managed to invite the District Commissioner of Ilala Municipality and other prominentlocalgovernmentactors.Thewomenheldspeechesandsangabouttheir concerns to persuade the local government to prioritize women’s economic empowerment and to provide better security for women in the market. By the end of event, the women had asked the local government to provide financial support to the VICOBA group to alleviate their economic hardship (O1). Later, a VICOBA member stated that the main sources of their economic hardship were a lack of access to loans and capital, high prices charged by middlemen (madalali), low incomes, and high levies charged by the municipality (ID2). 3.4.2 Understanding the patterns Most of the collective organizing and the two collective action cases are in the markets in Ilala Municipality and in Ferry market in Kigamboni Municipality, whereas the negative cases are mainly located in the markets of We pay, we act? 69 Kinondoni Municipality. In the following, I use the women’s experiences of taxationandlevelofcapacity forcollectiveactiontoexplorewhethertaxation motivated collective action and whether differences in capacity can explain the collective engagement patterns identified above. 3.4.2.1 Motivation: Does taxation trigger collective action and collective organizing, and if so, how? As argued earlier, taxation can motivate revenue providers’ collective engagement if they experience that paying taxes contributes to their economic hardship. Table 3.2 shows the market traders’ self-reported formal payments thattheymaketoconductbusiness.Sometraderspay onaweeklybasis,while others pay every day. The most noticeable difference is between the vegetable traders in markets in Kinondoni Municipality, who reported paying TZS 1,400–2,900 weekly, and the vegetable traders in Ferry market in Kigamboni Municipality, whose weekly payments amount to TZS 7,000 (FGD1–FGD3, ID4–ID5). In 2018, TZS 1,400 corresponded to about USD 0.6, and now TZS 7,000 is about USD 3. While this may seem a small amount—and a small difference between the highest and lowest payment—it is worth noting that most of the women were small-scale traders reporting to have very low incomes, which meant that even small levy changes were felt. However, since most of the women Table 3.2 Market tradersʼ self-reported weekly formal payments (Tanzanian shillings (TZS)) Municipality Market Self-reported formal payments (weekly) Patterns of collective organizing (1) and collective action (2) Kinondoni Mtambani 1,400 – Magomeni 2,900–5,000 – Tandale 1,750–3,000 1: Upatu group Ilala Mchikichini 4,200–4,600 1: VICOBA groups, NWTA, Kiosk coop. society 2: Inauguration event Ilala 4,900 1: VICOBA, NWTA 2: Electricity fee protest Kisutu 4,500 1: VICOBA, women’s association Kigamboni Ferry 7,000 1: VICOBA, women’s platform Note: Data from interviews FGD1–FGD4 and ID1–ID7. 70 Ane Edslev Jacobsen reported not to hold accounts of purchases, sales, or levies paid, it has not been possible to establish exactly how much these weekly payments ate away from the traders’ income. This also meant that it was difficult to establish whether the higher formal taxation in Ferry market put a greater economic pressure on the Ferry traders compared to the traders in the other markets. Women in all markets expressed being under severe economic pressure, and across the cases of collective organizing and action, economic hardship seemed to motivate collective engagement with the prospect of easing the economic pressure (FGD4, ID2–5, O1–O2), among other things. Illustratively, a VICOBA member in Ferry said the following about why she joined the group: . . . because you can save up little by little. If you have 2 million TZS [Tanzanian shillings] in shares, you can borrow up to 6 million, and then you are in a position to bring about development. I have been able to take my children to school, I have built a house, and I have bought a land while in VICOBA. (ID4) In Kisutu market, informal taxation appeared to be a challenge, and some women reported this to be their main reason for joining a VICOBA group. Here,inadditiontopayingstallrenttoIlalamunicipalauthorities,thewomen were unlawfully charged between TZS 30,000and 50,000 a month in stall rent by other traders who had previously occupied the stalls. The women said they had complained to the market leadership, but only individually as it had not occurred to them to use the VICOBA group or the women’s association to voice their complaints (FGD4). As such, this is an example of taxation issues triggering collective organizing but not collective action. Generally, the women said that this economic hardship could, in part, be ascribed to taxation because they had to pay the formal levies even on days where they did not make any sales. However, taxation seemed to be one of many factors rather than the paramount triggering factor for their collective engagement. Another way taxation could motivate revenue providers’ collective engagementisiftheyexperiencealackofreciprocityorunsatisfactoryreturnfortheir payments. To examine whether this motivation is present, it is first important to understand what the women would expect in return for their tax payment. The women themselves explained that they pay the aforementioned levies to be allowed to conduct business in the market and to cover a range of market services.Thewomenreportedthatthelevieswerepaidtothelocalauthorities or the market leadership and that the levies were collected by municipal officials, market leadership, or appointed traders. Results differed as to whether We pay, we act? 71 Table 3.3 Market tradersʼ accounts of what their weekly payments are supposed to fund Municipality Market Garbage collection Cleaning Security Electricity Stall rent Market infrastructure and repair work Kinondoni Mtambani X X Magomeni X X X X Tandale X X X X Ilala Mchikichini X X X Ilala X X Kisutu X X Kigamboni Ferry X X X the traders were aware that the rates and purpose of the formal levies figure in municipal bylaws. Table 3.3 shows the traders’ accounts of what the weekly payments were supposed to fund. According to the traders, the formal levies are supposed to fund different types of market services as well as stall rent (FDG1–FDG4, ID1–ID5). Interestingly, results differed quite a bit regarding whether the women were satisfied with the return they got from their payments and whether they experienced a lack of reciprocity on the part of the local government (and sometimes the market leadership). In Mtambani and Ferry markets, the interviewedwomenwerequitesatisfiedwiththeirreturns(ID4–ID5,FDG1). In the other markets, the women expressed significant discontent with the service provision. In Tandale market, the traders said that the payments they made for ‘market infrastructure’ were supposed to fund a functioning drainage system. However, during the interviews, the women would point to the open sewers with stagnant water and waste surrounding us and say that the local authorities were clearly not upholding their end of the agreement(FDG3). Asimilarproblemwasarticulatedbythewomenin Magomeni market (FDG2). Whether this experience of a lack of reciprocity motivated the women’s collective engagement varied across the markets. In Magomeni and Tandale markets, the dissatisfaction with the service provision did not trigger any kind of collective engagement; here, only one woman reported to be in an upatu group to deal with her economic challenges. Conversely, in other markets, the lack of reciprocity triggered discontent and collective engagement. The electricity fee protest in Ilala market is one such case. Here, the traders 72 Ane Edslev Jacobsen werediscontentedthattheincreaseinelectricityfeeswasnotaccompaniedby electricity improvements, and this made them protest to the market leadershipheadedbyamunicipalrepresentative(ID7,O2).Theinaugurationparty in Mchikichini market is another case. Here, the women’s complaints to the local government were based on dissatisfaction with the poor state of security for women traders in the market now that they actually paid for security (O2). In the cases of collective organizing, the issues of dire economic conditions and lack of service delivery (in particular, security) contributed to the women’s decisions to form and join groups. In these cases, the women said they felt like they had to rely on themselves—and their own organized groups—to handle these challenges (along with the social and gender-related challenges). Arguably, this implicitly speaks to a lack of reciprocity. Most of these traders connected the issues of poor service delivery, substandard market infrastructure, and the experience of having to rely on themselves for tackling these challenges to the issue of a lack of reciprocity on the part of the local government. However, as evidenced, it was only in a few cases that these organized groups metamorphosed into vehicles of collective action to complain to the local authorities about this experienced lack of reciprocity. Summing up, taxation can motivate collective engagement among revenue providers through the two economic grievances. However, even though almost all the traders said they experienced these grievances, there was quite a variation in the collective engagement: over half of the traders did not participate in any form of collective engagement, and I only identified two cases of actual collective action around taxation. Thus, tax-related grievances do notseemsufficienttotrigger collectiveorganizingand,particularly,collective action. 3.4.2.2 Capacity: Preconditions for the link between taxation and collective action? As argued earlier, women traders’ collective level of organizational and relational capacity is likely to condition whether they manage to overcome power-related barriers to collective action. In the following, I show how variations in organizational and relational capabilities help explain the negative and positive cases of collective engagement across markets. 3.4.2.2.1 Organizational capacity First, in the markets where no collective engagement was identified, the organizational capabilities of the women were indeed very limited. For example, in Magomeni market, the women apparently viewed collective organizing as We pay, we act? 73 a potential way to address their economic hardship and voice their discontent with their conditions, but they also expressed a lack ofknowledgeof how to do it in practice. One woman simply put it: ‘We do not how to deal with that [collective engagement]’ (FGD2). Another woman added: ‘The one way, we know, is the way you see us: coming from home to the market here and home again . . . we end up discussing our challenges as normal conversation, but nothing will change [because] us women in this market, we do not have that understanding’ (FGD2). As such, the women experienced having a fundamental lack of know-how in organizational development. Moreover, the women expressed having a limited understanding of their rights and a lack of access to information and channels of advocacy. In Magomeni, a woman said: ‘Things like groups are not here, and there is no one who can speak for us. There should be a person who could enlighten us, telling us that we as women need to fight for our rights by choosing one person who will speak on our behalf’ (FGD2). This quote also offers another interesting insight: the woman talks of an ‘us’ and ‘we as women’ illustrating that womanhood, even in spaces without collective engagement, may invoke a sense of community, thus making womanhood a potentially fruitful identity around which to mobilize. I return to the potential of the collective identity of womanhood later. The women participating in collective engagement had a greater organizational capacity; as such, organizational capacity and engagement seem to reinforce each other. However, from the women’s accounts it was evident that theirorganizationalcapabilitiesto someextenthadprecededandenabledthe emergence of their collective engagement. In Ferry market, for example, one womanhadpreviouslyparticipatedinvocationalandorganizationaltraining facilitated by the NGO CARE International. When she started her business in the market, she used her acquired organizational capabilities to mobilize other women traders in Ferry market into forming a VICOBA group to alleviate economic hardship: ‘I learnt [with CARE] till I understood about leadership, and now I have the ability to teach other people, and teach them so that they can teach other people’ (ID4). Another trader in Ferry indicated that the woman trained by CARE was a decisive human organizational resource: ‘Yes, that woman [ID5] was one of the people who influenced me to join [VICOBA]. She gave me information, and we decided that we should work to be like women we always [see] in the news; we should fight to have our own group’ (ID4). In Ilala, Mchikichini, and Kisutu markets, it was the training by a local civil society organization (CSO) called Equality for Growth (EfG) that sparked their organizationalcapacity. Thewomenin these marketsexplained 74 Ane Edslev Jacobsen that EfG—a CSO established in 2008 with the aim to empower informal women workers in Tanzania—started hosting seminars for the women traders focusing on legal and human rights, advocacy, and building capacity for collective engagement. Later, EfG assisted the women in establishing VICOBA groups, and in 2018, the women traders—primarily centred in the markets of Ilala Municipality—took steps to establish the National Women Traders Association (NWTA) (ID2–ID3, FGD4, ID7, O1–O2, NWTA). The NWTA was registered as an NGO in 2021; has engaged in its first advocacy efforts on behalf of women market traders; and is currently formulating a constitution, building up the internal organization, and accessing funding (Interviews with EfG and the chairperson of NWTA, March–April 2022). In relation to overcoming barriers to collective engagement, it was evident that most of these women’s views on the costs of collective engagement changed as their organizational capacity grew. Previously, they had viewed collective engagement as yet another time-consuming and costly activity, but with the organizational training, most of them said they had learned how to organize in ways that were cost and time efficient (e.g. holding meetings in the market on a regular basis at a time in between peak business hours). As attested by the cases of collective action, some of the women eventually started using their groups as channels of influence to municipal and market authorities. This attests to the fact that organizational capacity—at least over time—can help revenue providers overcome political barriers to collective action and thus become a political resource that the revenue providers can use to their advantage (O1–O2, ID4). It also substantiates that for disadvantaged revenue provider groups there is a certain sequencing in the collective engagement: collective organizing can be a necessary step to overcome, over time, the political barriers of collective action. 3.4.2.2.2 Relational capacity Starting with Tandale and Magomeni markets where collective engagement was almost non-existent, the women’s accounts attested to quite a limited relational capacity. Several women described that there was a lack of social trust in the markets. In Magomeni market, one woman said: ‘That relationship of trust, to be honest, is not there. Here, we just know each other, but whenever you encounter a challenge, you will have to deal with it yourself’ (FGD2). Interestingly, a woman in Ilala market who did not participate in any form of collective engagement voiced a similar experience: ‘Most people herehave groups in thestreets wherethey live, becausehere,theydonottrust each other very much’ (ID6). Other accounts spoke to a lack of reciprocity. In Tandale market, a woman who had previously been in an upatu (selflending) group said that she had stopped participating because ‘people will We pay, we act? 75 cheatyou...theybecomebrutal,becausetheyonlyreceive,butwhenit’stheir time to pay, they won’t pay’ (FDG3). In these contexts marked by a lack of trust and reciprocity, I did not observe a strong sense of community or sense of ‘we-ness’. Rather, the social relations seemed to be dominated by an ‘every person for herself’ notion. Illustratively, when I asked if the traders generally helped each other, one woman responded: ‘No! It’s like this: If I’m experiencing challenges myself, how can I support someone else who has challenges?’ (FGD3). This lack of relational capacity was decisive for the women’s lack of collective engagement. In Magomeni market, a woman said it well: ‘In this market, there is a lot of selfishness, so it is a bit hard to organize’ (FGD2). Conversely, women participating in collective engagement in Mchikichini, Ilala, Kisutu, and Ferry markets displayed a much greater social capital and often a strong collective identity. Interestingly, the women’s relational capacity was not just a product of their collective engagement. Rather, it was evident that their collective engagement was most often preceded and enabled by the women’s close, trust-based, and reciprocal social relationships. A woman in Ferry market recounted that it was her friendship with another woman trader in the market that made her want to form a VICOBA group, explaining that they were ‘like a family’ and helped each other with capital and family-related activities (such as funerals and weddings) (ID5). A woman in Mchikichini market similarly said: ‘I was reluctant to join, because I did not get enough information about it. But after I came here, I met other women and they advised me to join VICOBA, so the fear went away, and I became confident and joined’ (ID2). Friendships between women traders thus arguably help the women overcome psychological barriers to collective engagement. For the positive cases of collective action, high relational capacity clearly preceded the actual action, as both acts happened under the auspices of well-functioning VICOBA groups. 3.4.2.2.3 The collective identity of womanhood: A strong source of community Across the markets, it becomes clear that a strong source of community, and the most prevailing and unifying collective identity, was related to the notion of womanhood—that is, the perceived commonalities related to being women. The interviews offered several insights into why this was the case. In general,itseemsthatthisnotionofwomanhoodisshapedby patriarchalgender norms and power hierarchies. First, the women’s perceived commonality stemmed from their experience that they, as women, had family responsibilities. A woman in Ferry market explained it like this: ‘Look, women, like me, we have huge responsibilities . . . . We women should come forward because we have families to take care of’ (ID5). Second, the notion of womanhood stemmed from generally being subjected to (and having to 82 Ane Edslev Jacobsen Olson, M. 1965. The Logic of Collective Action: Public Goods and the Theory of Groups. Cambridge, MA: Harvard University Press. Polletta, F.,and J.M.Jasper. 2001. ‘Collective Identityand Social Movements’. Annual Review of Sociology 27(1): pp. 283–305. https://doi.org/10.1146/annurev.soc.27.1. 283. Prichard, W. 2015. Taxation, Responsiveness and Accountability in Sub-Saharan Africa: The Dynamics of Tax Bargaining. Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9781316275511. Prichard, W., and V. van den Boogaard. 2017. ‘Norms, Power, and the Socially Embedded Realities of Market Taxation in Northern Ghana’. African Studies Review 60(1): pp. 171–194. Prud’homme, R. 1992. ‘Informal Local Taxation in Developing Countries’. Environment and Planning C: Government and Policy 10(1): pp. 1–17. https://doi.org/10. 1068/c100001. Putnam, R. D., R. Leonardi, and R. Y. Nanetti. 1994. Making Democracy Work: Civic Traditions in Modern Italy. Princeton, NJ: Princeton University Press. https:// press.princeton.edu/books/paperback/9780691037387/making-democracywork. Roever, S. 2014. Informal Economy Monitoring Study Sector Report: Street Vendors. Cambridge, MA: WIEGO. Sewell, W. H. 1992. ‘A Theory of Structure: Duality, Agency, and Transformation’. American Journal of Sociology 98(1): pp. 1–29. Sharp, S., S. Sweet, and A. R. Menocal. 2019. Civil Society Engagement in Tax Reform. London: Overseas Development Institute. https://cdn.odi.org/media/ documents/12926.pdf. Snow, D. A., and S. A. Soule. 2010. A Primer on Social Movements. New York: W.W. Norton. Timmons, J. F. 2005. ‘The Fiscal Contract: States, Taxes, and Public Services’. World Politics 57(4): pp. 530–567. van den Boogaard, V., W. Prichard, R. Beach, and F. Mohiuddin. 2022. ‘Enabling Tax Bargaining: Supporting More Meaningful Tax Transparency and Taxpayer Engagement in Ghana and Sierra Leone’. Development Policy Review 40(1): pp. 1–22. https://doi.org/10.1111/dpr.12563. van den Boogaard, V., W. Prichard, and S. Jibao. 2019. ‘Informal Taxation in SierraLeone: Magnitudes, PerceptionsandImplications’. African Affairs118(471): pp. 259–284. Vanek, J., M. A. Chen, and G. Raveendran. 2015. A Guide to Obtaining Data on Types of Informal Workers in Official Statistics. Cambridge, MA: WIEGO. WIEGO. 2014. The Urban Informal Workforce: Street Vendors. Cambridge, MA: WIEGO. 4 Triggers and strategies of revenue bargaining Evidence from Mozambican municipalities Armin von Schiller There is only one hope. If more people believe in what we are doing, theyʼll start paying tax. That will be our budget. Then we can start to rebuild. Araujo, Mayor of Quelimane, in a BBC interview in 20131 4.1 Introduction Revenuebargainingimpliesanexplicitorimplicitnegotiationbetweenagovernment and revenue providers (see Chapters 1 and 2 in this volume). When considering revenue bargaining and fiscal contracts, there is a natural tendency to focus on national governments as they control most of the public revenue. However, revenue bargaining also happens at the international as well as subnational levels, and, in order to better understand revenue bargaining, it seems that the subnational level is a particularly promising avenue of investigation. Most citizens feel that national and international bargaining processesarereservedformorepowerfulactors.²Theirownpowerandinfluence will tend to be perceived as higher at the subnational level. In addition, incentives to engage in revenue bargaining might also be stronger at this level because, beyond expected influence, the links between revenue and public services are stronger and more visible here. Based on this idea, this chapter explores drivers of revenue bargaining based on a comparative analysis of Mozambican municipalities. The focus is ¹Johnson (2013). ² See discussion on dimensions of bargaining power in Chapter 2. Armin von Schiller, Triggers and strategies of revenue bargaining. In: The Politics of Revenue Bargaining in Africa. Edited by: Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192868787.003.0004 84 Armin von Schiller on the role of factors connected to the concept of political settlements, which has received little attention so far in understanding the tax performance of subnational units (Kjær, 2009;Jibao and Prichard, 2015). Mozambique is a particularly appealing case for this explorative analysis as it is a highly polarized political environment combined with some level of political and fiscal decentralization. As a consequence, a disputed decentralization process over recent decades has led to the creation of 53 municipalities, with some remarkable autonomy in decision-making around taxation coupled with locally elected municipal governments. In particular, the chapter aims to improve the understanding of two aspects: what triggers municipalities to reach out to citizens and initiate revenue-bargaining processes, and what strategies do they employ, including explanationsofthedifferencesinintensityanduse. Inaddition,itassesses the degree to which these activities appear to be successful and actually influence the content of the municipal fiscal contract as the expected final outcome of revenue bargaining (see Chapter 2). To answer these questions, data was collected from 101 semi-structured interviews conducted in Mozambique with representatives of municipal and national-level government. At the municipal level, interviewees included membersofthemunicipalgovernment,themunicipalassembly,localadministration, and civil society organizations (CSOs). At the national level, academic experts, representatives of different ministries involved in the decentralization process, and international donors were interviewed. The sample of municipalities included 11 out of the 53 municipalities in Mozambique, including diverse municipalities in terms of size, province, and intramunicipal level of political competition as well as political alignment with the national government. The results indicate that two factors trigger the predisposition of municipalities to reach out to citizens and initiate revenue-bargaining processes: perceived revenue pressure and political competition (see Chapter 2). Prichard (2015) extensively discusses the relevance of the level of revenue pressure facing governments on their predisposition to bargain. In essence, a government facing stronger revenue pressure is more likely to be open to compromise with taxpayers in order to access revenue. In this analysis, revenue pressure increases for the municipalities as a result of the uncertainty regarding the timeliness and size of intergovernmental transfers. This effect is particularly strong in municipalities governed by parties not in power at the national level. Hence, the results suggest that the potential politicization of the transfer system is a main determinant of revenue pressure.³ Political ³ Evidence on politicization of the transfer system can, for instance, be found in Banful (2011) and Brollo and Nannicini (2012). Triggers and strategies of revenue bargaining 85 competition triggers revenue bargaining by increasing the pressure on local governments to show themselves as performing well. Although this factor is identified as relevant in all municipalities, it appears to be particularly strong in municipalities governed by opposition parties. In terms of strategies, the interview material indicates that among representatives of local governments and administrations, the narrative around a more cooperative approach with citizens is present everywhere but is somehow stronger in municipalities characterized by political competition. This positive perspective towards revenue bargaining does not, however, seem to result in concrete measures. Conversations with representatives of CSOs do not, on the whole, confirm the alleged efforts. When comparing perceptions in municipalities with a lower and higher degree of political competition, the differences are also minimal. Similarly, while self-reported initiatives around revenue bargaining by the municipalities differ between lowand high-competition environments, these seem not to be particularly successful and do not result in substantial variation in the revenue bargaining taking place at the local level. Also, in many cases, it seems that initiatives were driven externally and, in particular, by donors, which may explain the lack of identifiable success. Overall, this chapter contributes to the growing effort to understand the politics of revenue bargaining by focusing on identifying factors shaping the predisposition of local governments to bargain. The analysis also indicates limitations on measures in this area and the necessity of thinking carefully aboutpreconditionsifrevenuebargainingistowork.Followingfromthis,the results also carry relevant policy implications. Enhancing revenue mobilizationandpromotingrevenuebargainingis highontheagendaofinternational donors. This is also based on the assumption—only partly backed up in the literature—that revenue bargaining can have a substantial governance ‘dividend’ by affecting politicians’ responsiveness and citizens’ engagement in public affairs, ultimately leading to more accountable government institutions (see Chapter 1;Fjeldstad, 2014;Moore et al., 2018, chapter 8). The results presented in this chapter call for caution and for thinking carefully about preconditions for revenue bargaining to emerge and for it to represent a meaningful process, able to bring the aforementioned desirable outcomes. In particular, the analysis shows how—where CSOs are too weak to coordinate and represent revenue providers—the outlook for the success of any initiatives around revenue bargaining is slim, even if we assume the political authorities instigated them with good intentions. This is connected to the accepted premise that lack of capacity for collective action undermines the potential for bargaining (Prichard, 2015). 86 Armin von Schiller The chapter proceeds as follows. Section 4.2 describes the political context and the decentralized system in Mozambique in order to understand the context in which municipalities operate. Section 4.3 provides details on the methodological approach, before the results are presented in Section 4.4. Section 4.5 concludes. 4.2 Fiscal and political decentralization system in Mozambique Mozambique has been in an ongoing process of decentralization since the peace accords of 1992. In fact, political and fiscal decentralization represented a crucial part of the peace negotiations (Ames et al., 2010, 103; Maschietto, 2016). However, the decentralization process, especially in its fiscal and political dimension, turned out to be less ambitious than initially envisioned. It took until 1997 to establish autonomous municipal governments in Mozambique (Weimer, 2012, 4). The main reason was political in nature. The Resistência Nacional Moçambicana (RENAMO, Mozambican National Resistance) received unexpected electoral support in the northern and central rural areas at the first multi-party elections in 1994. The potential loss of power in the event of decentralization led the Frente de Libertação de Moçambique (FRELIMO, Mozambican Liberation Front) to contest the planned ambitious decentralization model, which was then reformulated in a new compromise with RENAMO (Forquilha, 2016, 130; Maschietto, 2016, 109). As a result, a dual-structured system of local governance was introduced by law number 9/96, and in Mozambique, two types of decentralization coexist: devolution and deconcentration (Buur, 2009, 99; Maschietto, 2016, 105). Municipalities (autarquias locais) represent the devolution aspect of the system. They have autonomous political power and control a remarkable number of revenue instruments. At the other end of the spectrum, provinces, districts, andlocalitiesrepresentthedeconcentratedpartofthesystem. Atthe time this research was undertaken, deconcentrated units lacked autonomous political power, were directly accountable to the central state, and controlled only a limited number of revenue instruments.⁴ ⁴ This was changed by a constitutionalreform in 2018 (after the time of this research), by which assemblies at all levels are now directly elected and subsequently elect the head of the subnational executive out of their members. How or whether this new change in political decentralization will be matched in fiscal terms is unclear and is the subject of debate at the time of writing. Triggers and strategies of revenue bargaining 87 Table 4.1 Year of establishment and types of municipalities Year of creation Municipality Type of municipality 1997 Maputo A Matola, Beira, Nampula B Xai-Xai, Inhambane, Chimoio, Tete, Quelimane, Lichinga, Nacala, Ilha de Moçambique, Pemba C Chokwe, Chibuto, Maxixe, Manica, Dondo, Mocuba, Gurue, Angoche, Montepuez, Cuamba D Manhiça, Mandlakazi, Vilankulo, Catandica, Marromeu, Moatize, Milange, Monapo, Metangula, Mocimboa da Praia Vilas 2008 Namaacha, Macia, Massinga, Gorongoza, Gondola, Ulongue, Ribaue, Alto Molocue, Moeda, Marupa Vilas 2013 Boane, Praia do Bilene, Quissico, Sussundenga, Nhamatanda, Nhamayabue, Maganja da Costa, Malema, Chiure, Mandimba Vilas Source: Law no. 10/97 (República de Moçambique, 1997); Law no. 11/2013 (República de Moçambique, 2013); MAEFP, (2016) Giventhefocusofthischapteronthe effectoffactorsconnected topolitical settlement on shaping revenue bargaining at the local level, the empirical analysis is centred on Mozambican municipalities, as, in comparison to deconcentrated units, they control substantial tax handles and have some degree of freedom in setting up their strategies and priorities. As shown in Table 4.1, 53 municipalities have been established since then in three waves (1997, 2008, and 2013).⁵ The Ministério da Administração Estatal e Função Pública (MAEFP, Ministry of State Administration and Public Function) divides all municipalities into the following five categories based on a series of vaguely defined socioeconomic and political indicators: type A, the capital city Maputo; type B, the largest provincial capitals besides Maputo; type C, the capitals of all other provinces as well as two other cities; type D, cities that are particularly relevant to the local development of broader regions; and vilas, towns that do not qualify for category D (MAEFP, 2016, 28). According to the MAEFP, around one third of the Mozambican population lives in municipalities. Moreover, ⁵ These 53 municipalities include all of the country’s 23 cities and 30 out of 68 towns (MAEFP, 2016). In theory, the establishment of new municipalities should be based on technical arguments following the official policy, pursued since 1997, of gradualismo (Buur, 2009). However, the decision about where to create new municipalities has been identified to be largely driven by political reasons and FRELIMO’s incentive to favour its own strongholds and avoid potential loss of political power (Forquilha, 2016). 88 Armin von Schiller 60% of Mozambique’s GDP and 85% of its fiscal revenues are generated in municipalities (MAEFP, 2016, 26), Law number 1/2008 (Art. 17; República de Moçambique, 2008b) and decree number 63/2008 (República de Moçambique, 2008a) define the fiscal competencies of the municipalities at the time this study was undertaken. The most prominent taxes are the municipal poll tax (Imposto Pessoal Autárquico—IPA), the municipal property tax (Imposto Predial Autárquico—IPRA), the municipal property transaction tax (Imposto Autárquico de SISA—IASISA), and the municipal vehicle tax (Imposto Autárquico sobre Veículos—IAV). It is important to highlight that discretionary revenue policy is limited. A significantly larger degree of discretion can be exercised in the realm of non-tax revenue—meaning fees, service charges, fines, and short-term loans—than when it comes to taxes, for which, in most cases, general principles are determined according to national law (World Bank, 2014, 109). As is the case in many African countries (see Moore et al., 2018, 159), although municipalities have several instruments with which to generate their own revenue, they are highly dependent on intergovernmental transfers.On average,transfersrepresentwellabove65%ofrevenue.Inreality,this numberisprobablymuchhigherasavailableinformationtendstocomefrom bigger cities with more capable administration and higher revenue potential. Inthissense,theweightoftransfersisespeciallyhighwhenitcomestomunicipalities of type D, with a level around 90% (Ilal and Weimer, 2017;Weimer and Carrilho, 2017, 88). Data quality and availability are poor in Mozambique, but aggregating the evidencefromindividualcasestudiessuggeststhatmunicipalrevenuepotential is highly underexploited (Chimunuane et al., 2010b;2010a;Boex, 2011; Nguenha et al., 2012;2017;Weimer, 2012;Bunk et al., 2017;Weimer and Carrilho, 2017). Politically, Mozambique is a highly polarized country. Its political settlementcan be consideredtohavegraduallyshiftedoverrecentdecadestowards more exclusion and centralization of power and resources. At the same time, competition between factions within the ruling coalition has increased, as well as among the opposition, which is far from united.⁶ At the national level, FRELIMO has been in power since 1994. At the municipal level, however, opposition parties have been able to win some municipal elections, held in ⁶ See extensive analysis of the historical development of the political settlement in Mozambique in Macuane et al. (2018). Triggers and strategies of revenue bargaining 89 1998, 2003, 2008, 2013, and 2018 (MAEFP, 2016, 7).⁷ Since 2009, the Movimento Democrático de Moçambique (MDM), which evolved from within RENAMO, has been challenging the country’s bipartisan political system.⁸ In the municipal elections of 2013, which RENAMO boycotted, MDM won themajorityofvotesinfourmunicipalitieslocatedinthenorthandthecentre of Mozambique (Beira, Quelimane, Nampula, and Gurué), whereas FRELIMO won the remaining 49 municipalities. In 2018, RENAMO participated again in the local elections. MDM suffered a remarkable loss of votes and only kept control of its stronghold Beira. RENAMO won in eight municipalities, including in big influential cities such as Nampula, Quelimane, and Nacala.⁹ FRELIMO still dominates in the vast majority of municipalities. Its power base was confirmed again in the 2019 presidential election in which it reached a support of around 73% of the electorate, although irregularities were denounced.¹⁰ Overall, this section has offered some background to the polarization of the political system in Mozambique as well as the fiscal architecture within which the municipalities operate. It also highlights that local revenue potential remains highly underused. Together, these features indicate the relevance of exploring what leads municipalities to pursue local revenue mobilization in general and revenue bargaining with their citizens in particular. 4.3 Methodological approach The nature of this study is explorative. The main goal is to understand better what factors trigger local governments to engage in revenue bargaining with citizens and how they do it. To answer these questions, 101 semi-structured in-depth interviews were conducted with a range of stakeholders. The list of interviewees includes municipal government officials, municipal administrators, staff from national ministries and agencies, civil society actors, international donors, and academic experts. The interviews were conducted inMaputo(withnationalexperts)andin11municipalitiesbetweenFebruary and May 2017.¹¹ ⁷ At the national level, the first democratic presidential and parliamentary elections took place in 1994 and are held regularly every five years (Bertelsmann Stiftung, 2016, 7). In all presidential and parliamentary elections, FRELIMO gained the majority of the votes. ⁸ For further details concerning the emergence of the MDM party, see Maihack and Plagemann (2010). ⁹Comissão Nacional de Eleições (2018). ¹⁰ For instance, by the European Union Election Observation Mission (2019). ¹¹ Interviews were conducted by all members of the project ‘The Political Economy of Local Taxation in Mozambique’, listed in the acknowledgments. 90 Armin von Schiller Table 4.2 Municipalities selected as cases for the comparative analysis Municipality Province Type of municipality Political leadership at the local level (as of 2017) Matola Maputo B Aligned Inhambane Inhambane C Aligned Maxixe Inhambane D Aligned Beira Sofala B Non-aligned Dondo Sofala D Aligned Quelimane Zambézia C Non-aligned Nacala Nampula C Aligned Monapo Nampula Vila Aligned Pemba Cabo Delgado C Aligned Xai-Xai Gaza C Aligned Vilankulo Inhambane Vila Aligned Note: Aligned equals ruled by FRELIMO; non-aligned equals ruled by MDM. The municipalities were selected based on the diverse cases selection strategy (George and Bennett, 2005, 83). Hence, in order to assure maximum variation in the variables of interest, the sample includes municipalities from different provinces, characterized by different levels of administrative capacity, political competition, and political alignment (See Table 4.2).¹² Administrative capacity is proxied by the type of municipality, with cities of type B being considered to have the highest capacity, while type D municipalities and so-called vilas have the lowest (see the previous section for information on legal basis for the different types of cities). Municipalities are consideredalignedifthemunicipalityisgoverned by thesamepartyinpower at the national level (FRELIMO).¹³ Political competition is measured by the margin of victory in the municipal elections to identify strongholds versus more competitive political environments. In 2019, local elections took place in Mozambique. As this was after the data collection, these election results could not be considered; however, the two municipalities labelled as ‘non-aligned’ in this study remained as such after the 2019 elections.¹⁴ Also, the level of political competition and the distribution of political power remained similar in all other ¹² It is also important to underline that practical issues, including accessibility, security concerns, and data availability before the interview phase, also affected the case selection. We particularly rely on data provided by GIZ (2017). ¹³ At the time of data collection, four out of a total of 53 municipalities were governed by opposition parties. Two of them (Beira and Quelimane) are represented in the sample. ¹⁴ In Quelimane, RENAMO won with 59% of the vote, while in Beira, MDM won again with 48% of the vote (Comissão Nacional de Eleições, 2018). Triggers and strategies of revenue bargaining 91 municipalities included in the sample. Particularly noteworthy was that in Nacala, RENAMO regained power by a small margin and that in the disputed municipalities of Monapo and Matola, FRELIMO won but with a very narrow margin. This further reinforces the idea that these were particularly competitive environments at the time of this data collection. In all municipalities, interviews were held with the same set of actors.¹⁵ In terms of politicians, the mayor and (sometimes or) the councillor for finance were contacted. The municipal assembly, including members from different political parties, was also approached. In the administration, the head of finances or revenue and a member of the broader revenue team were interviewed.Inmostcases,supportfromthedonororganizationwashelpfulwhen makingcontact with theinterviewees.Alessstandardized approachwas used to contact CSOs. In most cases, the initial contact was through donors or research partners in Mozambique. The semi-structured interviews were run based on guidelines developed for specific types of actors.¹⁶ Topics addressed included perceptions about the funding situation of the municipality, challenges in exploiting revenue potential and the degree to which revenue is an issue of public debate, and information on activities to reach out to citizens and explain to them how revenue systems work and how money is spent,aswellasthedegreetowhichpoliticalconsiderationsaffecttheinteraction between local governments and citizens regarding revenue matters. All interviewees were asked for their consent and assured confidentiality. Normally, two researchers from the team participated in the interview. In the vast majority of the cases, a translator was involved. 4.4 Identified triggers and strategies of revenue bargaining 4.4.1 Triggers to pursue local tax collection: Perceived revenue pressure and political competition The analysis of the interview material strongly indicates that there are two major triggers that explain the predisposition of municipal governments to engage in revenue bargaining: the unreliability of the fiscal transfer system and political competition. ¹⁵ You can find an overview of the interview partners by municipality and type of actor in the Appendix (Tables 4.A1 and 4.A2). ¹⁶ Guidelines can be provided upon request. 98 Armin von Schiller better preconditions for bargaining in the future. In terms of ownership and sustainability, it is questionable whether municipal governments were, in most cases, the main initiators of information and outreach activities. Rather than the municipalities themselves, it seems that third parties— donors and national-level institutions—play a crucial role in facilitating and initiating measures.²⁶ Examples of this include the successful coordination of civil society through local committees such as SAMcoms in Pemba (I101, CSO representative in Pemba), the introduction of a social audit in Inhambane (I43) with the help of MASC (Fundação Mecanismo de Apoio à Sociedade Civil—the Foundation Support Mechanism for Civil Society), and an information campaign about the municipal poll tax (IPA) launched in Maxixe that was conceptualized by national ministries (I53, member of Maxixe municipal assembly). In the case of participatory budgeting in Nampula and Quelimane, this reliance on donors resulted in the cancellation of the whole programme once the municipality was supposed to use its own funds instead of donor money to implement outreach measures (I8, donor employee). The effectiveness of the measures is also doubtful. Although interviewees in all municipalities stated that outreach activities should continue, and considered them valuable, most interviewees indicate that the activities did not have any clear measurable impact on the tax culture and level of engagement of citizens. Only very few were confident in attributing a positive effect to these initiatives and could name concrete examples of successes (e.g. I47, I55, and I67; public administrators and members of the municipal assemblies in Maxixe, Beira, and Dondo). The limited success of the measures highlights a precondition of revenue bargaining that is commonly ignored. The municipal government might be open to bargaining, but if revenue providers lack a minimum organizational power, they will not only have a weak bargaining position (see Chapter 2), but they will also be unable to engage in revenue-bargaining processes in a meaningful manner. As such, there is only very little anecdotal evidence of a systematic organization of interests among citizens to discuss revenue matters at the municipal level. CSOs are perceived to exert only very limited influence on municipal policymaking, particularly in the revenue area, even when supported in the municipalities when explicitly asked whether they consider that municipal identities are evolving over the years, as well the knowledge of citizens about the division of responsibilities between levels of government. ²⁶ In particular, this point was made strongly in interviews with donor employees (I2 and I8) and CSOs (I43, I58, and I79) as well as members of municipal assemblies (I60, I75, and I82) in aligned and nonaligned municipalities. Triggers and strategies of revenue bargaining 99 by donors. Reasons cited are mostly connected to lack of organizational strength and thematic expertise. In fact, supporting doubts about honesty in the alleged openness to engage in revenue bargaining by some municipalities, some interviewees described cases where municipalities were unwilling to cooperate with them as soon as CSOs started to engage more intensively in public finance topics.²⁷ All in all, municipalities implement many outreach activities that could serve as an invitation to start the process of revenue bargaining. However, the effectivenessintermsofprovidingplatformsforinclusiverevenuebargaining seems to be questionable. Beyond that, even where municipalities might be honestly willing to engage in such processes, it seems that the preconditions are not in place for meaningful revenue bargaining to emerge since CSOs as third parties coordinating and aggregating citizen’s demands, wishes, and arguments lack the organizational capacity to fulfil these roles. 4.5 Conclusion This chapter has explored factors that trigger revenue bargaining between municipal governments and citizens. Furthermore, it has analysed the strategies they use to approach citizens. In this regard, the greatest contribution of this chapter is in going beyond revenue collection and performance towards understanding what leads municipalities to engage in revenue bargaining. This is an essential step to developing a more nuanced understanding of factors that explain differences in predisposition for revenue bargaining across municipalities as well as preconditions required for meaningful revenue bargaining to take place. The results indicate that political competition and perceived unreliability of the transfer system strongly incentivize municipalities to engage in revenue bargaining. The interview material also shows that the narrative about the relevance of revenue bargaining is present in almost all municipalities. Concrete initiatives are more often named in municipalities characterized by high political competition, although they exist in all. However, these efforts are perceived by civil society to be significantly lower, regardless of whether the municipality is aligned with the central government or not. ²⁷ Donor employees raised this issue repeatedly (in particular, I2 and I8). Also, CSOs themselves claimed this in both aligned and non-aligned municipalities (e.g. I43, I58, and I79). 100 Armin von Schiller These results underline that the openness of government will be insufficient to trigger meaningful revenue bargaining. Openness is a necessary, but certainlynotasufficientcondition.Bargainingimpliestheengagementoftwo sides. Hence, either the individual revenue providers or at least the CSOs— as the third party coordinating the voices of the revenue providers—must be sufficiently organized to purposively interact with the local governments. Currently, in Mozambique, the civil society at the municipal level is too weak to fulfil this role. As a result, even if we assume that municipal governments arehonestlyopentostartingrevenue-bargainingprocesses,whichisdoubtful as donors seem the strongest drivers of these initiatives, the scope for revenue bargaining to emerge and have major impacts on the micro-level fiscal contract is small. More tentatively, a question that derives from the results and would deserve further scrutiny is the potential paradox that the efforts of national governments to manipulate transfer systems to disadvantage municipalities governed by opposition parties might, under certain circumstances, have the opposite effect than the one expected. In this line, the additional fiscal pressure might trigger a strengthening of the micro-level fiscal contracts between these local governments and their citizens and, thereby, put these very same municipalgovernmentsinabetterpoliticalpositioninthemidand longterm as well as, more generally, increase the political salience of these municipalities. Wang (2017) describes how China depoliticized its public finance to disincentivize the activity of local politically active social groups. The analysis in this chapter would suggest that in a somewhat reverse logic, putting revenue pressure on the municipalities controlled by the political opposition might encourage the interaction between society and local government, improving their relationship and potentially leading national government to face problems of legitimacy. From a revenue-bargaining perspective, it is important to note that the potentially associated political polarization might subvert any possibility for broad-based fiscal contracts between central government and the broader society at the national level, creating the paradox that a stronger fiscal contract at the local level weakens the national one. Theresultspresented in thischapteralsohaverelevantpolicyimplications. In particular, they send a clear warning. Scholars and international donors have proposed revenue collection to have a governance dividend for state building and government–citizen relationships (Chapter 1; see also Moore, 2015;Prichard, 2015). The analysis shows that even if governments are willing to actively approach and engage with citizens, the desired outcomes will only take place if the counterpart, the citizens, are able to meaningfully engage in the bargaining process. This implies that donors need to approach Triggers and strategies of revenue bargaining 101 thisissuefromamoreholisticperspective. Revenuecollectionbased on more inclusive revenue bargaining demands more than increasing the technical and administrative capacity of governments and more than pushing government to formally offer the channels. Strengthening civil society in this field, institutionally and in terms of technical capacity around fiscal issues, is at least equally relevant. 4.6 Appendix Table 4.A1 Overview of interview partners by municipality Beira 9 Dondo 9 Inhambane 7 Matola 7 Maxixe 8 Monapo 8 Nacala 6 Pemba 7 Quelimane 9 Vilankulo 4 Xai-Xai 10 Note: 17 interviews were held with individuals working at the national level. Table 4.A2 Overview of interview partners by type of actor Local administration 26 Municipal assembly 27 Municipal government 14 CSO (incl. community leaders) 16 Donors 12 National experts (incl. academics and official of ministries) 6 Acknowledgements This chapter presents part of the results of the project ‘The Political Economy of Local Taxation in Mozambique’, implemented within the context of the 52nd Postgraduate Training Programme of the German Institute of Development and 102 Armin von Schiller Sustainability (IDOS)—formerly Deutsches Institut für Entwicklungspolitik/ German Development Institute (DIE). The research was funded by the Ministry for Economic Cooperation and Development (BMZ). Members of the research team included, beyond the author, Saida Bunk, Salvador Forquilha, Dominique Klawonn, Jonathan Krull, Alina Sennewald, Conrad Steinhilber, and Juliane von Boeselager. Their fundamental contribution to this analysis is gratefully acknowledged. The author and project members also wish to thank the partners: Instituto de Estudos Sociais e Económicos (IESE), Project Programa de Boa Governação Financeira (German DevelopmentCooperationin Mozambique), andAssociaçãoNacionaldos Municípios de Moçambique (ANAMM). Without their support, this research would not have been possible. Special thanks go to Katharina Hübner, Amandio Jaquete, Eduardo Nguenha, Bernhard Weimer, Michael Sambo, Bernardino António, Stefan Leiderer, Christian von Haldenwang, and Julia Leininger for their helpful ideas, comments, and support. The project team is also grateful for the indispensable support of a large team of translators, as well as all the administrative support provided by the different partner institutions in Germany and Mozambique. Finally, I want to thank Anne Mette Kjaer, Marianne Ulriksen, Ane Karoline Bak, Odd Helge Fjeldstad, and all other participants of the ECPR Joint Sessions Workshop ‘The Politics of Revenue Bargaining’ for their feedback on previous versions of this chapter. References Ames, B., E. Connerley, D. M. do Rosario, E. Nguenha, and L. Francisco. 2010. Comparative Assessment of Decentralization in Africa: Mozambique In-Country Assessment Report. Vermont: ARD, Inc. http://pdf.usaid.gov/pdf_docs/PNADX220.pdf. Banful, A. B. 2011 ‘Do Formula-Based Intergovernmental Transfer Mechanisms EliminatePoliticallyMotivated Targeting?Evidence fromGhana’. Journalof Development Economics 96(2): pp. 380–390. Bertelsmann Stiftung. 2016. BTI 2016—Mozambique Country Report. Gütersloh: Bertelsmann Stiftung. https://bti-project.org/fileadmin/api/content/en/ downloads/reports/country_report_2016_MOZ.pdf. Boex, J. 2011. An Analysis of Municipal Revenue Potential in Mozambique: Summary Report. Washington, DC: World Bank. Brollo, F., and T. Nannicini. 2012. ‘Tying Your Enemy’s Hands in Close Races: The Politics of Federal Transfers in Brazil’. American Political Science Review 106(4): pp. 742–761. Triggers and strategies of revenue bargaining 103 Bunk, S., S. Forquilha, D. Klawonn, J. Krull, A. Sennewald, C. Steinhilber, and A. von Schiller. 2017. Socio-Political and Administrative Determinants of Municipal Revenue Performance: Insights from Mozambique. Briefing Paper No. 14/2017. Bonn: Deutsches Institut für Entwicklungspolitik. Buur,L. 2009. ‘ThePoliticsofGradualismo:PopularParticipationand Decentralised Governance in Mozambique’. In Rethinking Popular Representation, edited by O. Törnquist, N. Webster, and K. Stokke, pp. 99–118. New York: Palgrave Studies in Governance, Security, and Development. Cheeseman, N., and D. de Gramont. 2017. ‘Managing a Mega-City: Learning the Lessons from Lagos’. Oxford Review of Economic Policy 33(3): pp. 457–477. Chimunuane, O., M. Hassam, S. Hindkjaer, and B. Weimer. 2010a. Estudo sobre Potencial Tributário no Município da Vila de Vilankulo. Ministério para a CoordenaçãodaAcçãoAmbiental(MICOA)incooperationwiththeWorldBank. Chimunuane, O., M. Hassam, and B. Weimer. 2010b. Estudo sobre Potencial Tributário no Municipio da Vila de Marromeu. Ministério para a Coordenação da Acção Ambiental (MICOA) in cooperation with the World Bank. Comissão Nacional de Eleições. 2018. ‘Edital’. http://www.open.ac.uk/technology/ mozambique/sites/www.open.ac.uk.technology.mozambique/files/files/ Resultados%20e%20mandatos%202018.pdf. European Union Election Observation Mission. 2019. Republic of Mozambique. Final Report: General and Provincial Assembly Elections. Election Observation Mission. https://www.eeas.europa.eu/sites/default/files/eueom_moz2019_final_ report_en.pdf. Faguet, J.-P. 2014. ‘Decentralization and Governance’. World Development 53(C): pp. 2–13. Fjeldstad, O. H. 2014. ‘Tax and Development: Donor Support to Strengthen Tax Systems in Developing Countries’. Public Administration and Development 34(3): pp. 182–193. Forquilha, S. 2016. ‘Municipalização em Moçambique: Lógicas e Dinˆ amicas Políticas’. In Desafios para Moçambique 2016, edited by L. D. Brito, C. N. Castel-Branco, S. Chichava, S. Forquilha, and A. Francisco, pp. 129–150. Maputo: Instituto de Estudos Sociais e Económicos IESE. George, A. L., and A. Bennett, 2005. Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press. GIZ. 2017. Base de dados sobre as finanças municipais em Moçambique: 2009–2015 [Data file and code book]. Data collection commissioned by GIZ in cooperation with MEF and implemented by MAP Consultoria. Maputo: Deutsche Gesellschaft für Internationale Zusammenarbeit GIZ–Programa Boa Governação Financeira. 104 Armin von Schiller Ilal, A., and B. Weimer. 2017. O Sistema de Transferências Fiscais Intergovernamentais (STFI) em Moçambique: Uma Análise Fiscal e Institucional. Relatório Final. Maputo: Ministério de Economia e Finaças/GIZ. Jibao, S. S. and W. Prichard. 2015. ‘The Political Economy of Property Tax in Africa: Explaining Reform Outcomes in Sierra Leone’. African Affairs 114(456): pp. 404–431. Johnson, L. 2013. ‘Quelimane: One Square Mile of Mozambique’. 2 April. https:// www.bbc.com/news/world-radio-and-tv-21831810. Kjær, A. M. 2009. ‘Sources of Local Government Extractive Capacity: The Role of Trust and Pre-Colonial Legacy in the Case of Uganda’. Public Administration and Development: The International Journal of Management Research and Practice 29(3): pp. 228–238. Kjær, A. M., and O. Therkildsen. 2013. ‘Elections and Landmark Policies in Tanzania and Uganda’. Democratization 20(4): pp. 592–614. Lambright, G. M. S. 2014. ‘Opposition Politics and Urban Service Delivery in Kampala, Uganda’. Development Policy Review 32(1): pp. 39–60. Macuane, J. J., L. Buur, and C. M. Monjane. 2018. ‘Power, Conflict and Natural Resources: The Mozambican Crisis Revisited’. African Affairs 117(468): pp. 415–438. MAEFP (Ministério da Administração Estatal e Função Pública). 2016. Breve Avaliação da Situação Actual dos Municípios em Moçambique. Maputo: Ministério da Administração Estatal e Função Pública. Maihack, H., and J. Plagemann. 2010. ‘Eine Dritte Politische Kraft in Mosambik?’ GIGA Focus Afrika 8. Hamburg: GIGA. https://pure.giga-hamburg.de/ws/files/ 24543752/gf_afrika_1008.pdf. Maschietto, R. H. 2016. ‘Decentralisation and Local Governance in Mozambique: The Challenges of Promoting Bottom-Up Dynamics from the Top Down’. Conflict, Security & Development 16(2): pp. 103–123. Moore, M. 2015. ‘Tax and the Governance Dividend’. In Perspectives on Politics, Production and Public Administration in Africa, edited by A. M. Kjær, L. Engberg-Pedersen, and L. Buur, pp. 159–174. Copenhagen: Danish Institute of International Studies. Moore, M., W. Prichard, and O. Fjeldstad, eds. 2018. ‘Small Taxes and Large Burdens: Informal and Subnational Revenues’. In Taxing Africa: Coercion, Reform and Development, pp. 147–178. London: Zed Books. Nguenha, E., O. Muianga, and A. António, A. 2017. Empenho e Potencial Tributários dos Municípios em Moçambique: Estudo de casos de Nampula e Quelimane. Maputo: Associação Nacional dos Municípios de Moçambique (ANAMM). Nguenha, E., U. Raich, and B. Weimer. 2012. ‘Subnational Finances: Performance and Sustainability of Mozambican Municipalities’. In Moçambique: Descentralizar Triggers and strategies of revenue bargaining 105 o Centralismo: Economia Politica, Recursos, Resultados, edited by B. Weimer. pp. 197–215. Maputo: Instituto de Estudos Socias e Económicos (IESE). Prichard, W. 2015. Taxation, Responsiveness and Accountability in Sub-Saharan Africa. Cambridge: Cambridge University Press. República de Moçambique. 1997. Lei no. 10/1997 of 31 May 1997. Maputo: Boletim da República. República de Moçambique. 2008a. Decreto 63/2008. Maputo: Boletim da República. República de Moçambique. 2008b. Lei no. 1/2008 of 16 January 2008. Maputo: Boletim da República. República de Moçambique. 2013. Lei no. 11/2013 of 3 June 2013. Maputo: Boletim da República. Wang, Y. 2017. ‘Why Tax Policy Is Not Politics in China: Public Finance and China’s Changing State–Society Relations’. Politics and Policy 45(3): pp. 194–223. Weimer, B. 2012. Municipal Tax Base in Mozambique: High Potential—Low Degree ofUtilisation. DiscussionPaperNo.1/2012. Bonn: DeutschesInstitutfürEntwicklungspolitik. Weimer, B., and J. Carrilho. 2017. Political Economy of Decentralisation in Mozambique: Dynamics, Outcomes, Challenges. Maputo: Institute for Social and Economic Studies (IESE). World Bank. 2014. Mozambique Public Expenditure Review: Addressing the Challenges of Today, Seizing the Opportunities of Tomorrow.http://documents.world bank.org/curated/en/677921468275102771/Mozambique-Public-expenditure-re view-addressing-the-challenges-of-today-seizing-the-opportunities-of-tomorrow. 5 Tax reforms in Tanzania Where and how are compromises negotiated? Marianne S. Ulriksen, Lucas Katera, and Jamal Msami 5.1 Introduction The then Tanzanian Finance Minister, Dr Phillip Mpango, caused uproar in Parliamentwhenheproposed,duringthepresentationofthe2016/2017budget, ‘theamendmentofthe IncomeTaxAct,Cap332,withaview toremoving income tax exemptions on MPs’ gratuity in order to promote equity and fairness in taxation’.¹ The Members of Parliament from both sides of the floor ‘were not ready to accept the idea’.² And they were not the only ones to get upset that June of 2016. The government proposed a range of tax reforms targeting a wide variety of taxpayers, thereby setting in motion protests and attempts to revert the reform proposals. Hence, the spate of tax reforms was a unique opportunity to study the subsequent revenue bargaining between governmentanddifferentrevenueproviders,allowingustoexplorequestions such as how and where are revenue providers engaging with the government around tax reforms, does the government at times compromise on its initial proposal, and what are the micro-level fiscal contract outcomes? We selected four cases where we, based on the theoretical framework in Chapter 2, assumed that revenue providers and government would engage in negotiations over tax reform (see Msami et al., 2022). The cases are: the MPs who saw taxes introduced on their retirement benefits; the security forces whose duty-free shops were abolished; the tourism sector, which ceased to benefit from VAT exemptions on services offered; and the transport sector, where VAT was imposed on auxiliary services. Because the theoretical framework in Chapter 2 does not theorize where and how the bargaining processes ¹ See Tanzania’s The Citizen newspaper: https://www.thecitizen.co.tz/tanzania/news/national/mpsgang-up-against-taxation-of-their-send-off-package-2558092 (accessed 16 June 2022). ² Ibid. Marianne S. Ulriksen, Lucas Katera, and Jamal Msami, Tax reforms in Tanzania. In: The Politics of Revenue Bargaining in Africa. Edited by: Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192868787.003.0005 Tax reforms in Tanzania 107 actually unfold, we explore how targeted revenue providers and government³ engage in negotiations; that is, what strategies are used to convince/win over the other party. Furthermore, we contribute to the taxation literature by studying the role of institutional settings—what we term arenas—in facilitating bargaining. We identify three arenas where revenue providers and government actors engage in bargaining: public, bureaucratic, and political. Findingsfromourfieldwork (see Msamiet al.,2022forthe fullin-depthanalysis) suggest that the public arena comprises few institutionalized spaces for dialogue, which makes engagements between revenue providers and government conflictual and compromise unlikely. Instead, in the bureaucratic and political arenas, where more formal and informal institutions allow for structured, repeated, and trusted relationships to develop, the revenue providers are more likely to push the government to compromise on the initial tax proposal.Hence,fiscaloutcomesofrevenuebargainingmaybelessthanoriginally anticipated in policy proposals, although tax reforms also trigger better organization of revenue providers and the building of relationships between state and societal actors. 5.2 Arenas and strategies of revenue bargaining The (re-)introduction of tax by government can cause initial protests and reactions from revenue providers (see Chapter 1). However, as Moore (2007, 16) argues, ‘any realistic understanding [of the tax bargaining] processes requires that we go beyond the initial reaction to taxation of each type of actor and take into account the ways in which they then interact, whether conflictingly, cooperatively, or in more complex ways’. The nature of these interactions is informed by institutions, which the taxation literature notes as crucial in forming revenue bargaining. Levi (1988) argues that institutions can facilitate bargaining by reducing transaction costs and enabling less confrontational forms of bargaining. Prichard (2015), in his seminal contribution on contextual factors shaping tax bargaining, highlights how institutions shape the feasibility of tax bargaining by allowing taxpayers to engage with government in constructive ways. Essentially, institutions—such as effective parliamentary processes, forums of engagement, and informal relations based on trust and bargaining—are spaces for dialogue where compromises ³ Although we consistently refer to the ‘government’ in this chapter, we understand the government within the political settlement approach to consist of the inner circles around the president, that is, the ruling elite. 114 Marianne S. Ulriksen, Lucas Katera, and Jamal Msami political circles. Additionally, as the force is—by its very nature—a hierarchical and secretive organization, it is expected to be well organized and have a powerful leadership (Therkildsen and Bourgouin, 2012). Thus, from a collective action perspective, the security force can speak from one common preference position. In sum, the four cases shared the experience of being targeted by the government in the 2016 tax reforms. The tourism and transport sectors faced the introduction of VAT payments, MPs’ gratuity payment was taxed, and the security forces’ perk of duty-free shops was abolished. The stark surprise of being targeted for tax payments triggered immediate reactions from the revenue providers (except the security forces) and set in motion a bargaining process. 5.3.1 The tourism sector Before the 2016 tax reform proposal, the main arenas in which discussions took place between the government and the tourism sector were the political and bureaucratic arenas during the national budget processes and through the parliamentary standing committee responsible for tourism. The tourism sector, represented by the Tanzania Private Sector Foundation (TPSF) and TCT (Tourism Confederation of Tanzania), was able to lobby for a continuous VAT exemption to the sector. It was therefore a surprise when the Minister of Finance announced the introduction of VAT during the budget session of 2016/17 in June 2016. After the budget session where the VAT was introduced, the main arena of engagement was in the parliamentary standing committee on tourism. During the meetingsbetween theTCT and theTPSF ononehand andthestandingcommittee of the Parliament on the other, the tourism sector stakeholders sought to revoke the decision. Inthelobbyingprocess, the members oftheTPSF met formally with the standing committee and informally with some members pushing their case. Despite efforts by the TCT and other tourism stakeholders, under the umbrella of the TPSF, the government went ahead with its decision. After the introduction of VAT and the endorsement by Parliament, the main forum of exchange moved to the public arena, and the bargaining became antagonistic. The tourism sector used the media to make their case and drum up popular support (The Citizen, 2016a). It claimed that it was not unwilling to pay taxes but argued that the introduction of VAT would have an adverse effect on its competitiveness. The government used the media to Tax reforms in Tanzania 115 tarnishthetourism sector’s reputationby saying thattheydidnotwant topay taxes (Daily News, 2016). The debate in the media became heated, and parts of the tourism sector acknowledged that constantly attacking the Ministry of Tourism in the newspapers was not constructive. The President closed the debate by saying that the VAT would stay, even if it affected the number of tourists visiting Tanzania negatively (The Citizen, 2016b). Later in 2016, TCT continued to push the government to reconsider reversing the decision. This time, the TCT wrote to the Minister of Tourism to ask for a meeting to discuss the case and reach a mutual understanding. While the government accepted this meeting, it never took place. However, the tourism sector continued to reach out and sought to establish dialogue with the budget committees and other stakeholders like Tanzania Revenue Authorities (TRA) by travelling from Dar es Salaam to the capital, Dodoma. In the first half of 2017, the TCT was invited to make a presentation to the Parliament on the private-sector perspective of the tourism industry. During the seminar, the budget parliamentary committees learnt more about the tourism industry and how a retainment of the VAT would affect the sector. Persons we interviewed in the sector (see Msami et al., 2022) explained how the presentation of material and evidence-based advocacy opened dialogue attheformalmeetingaswellasmoreinformalconversations withparliamentarians wanting to know more. These kinds of dialogues in which different stakeholders in the sector engaged with the government were instrumental in the considerations to remove some of the taxes in the sector from the financial year 2017/18. Table 5.2 provides a schematic summary of the case of revenue bargaining between the tourist sector and the government in Tanzania. Initially, the relationships between the government and the organizations representing the tourism sector were amicable and took place in formal institutions such as the budget and parliamentary committees. However, the government’s announcement to introduce VAT surprised the sector, and bargaining moved to the public arena and turned more conflictual. Despite pressure through media stories, the government did not change its mind. Over the following months, the tourist sector re-engaged with government officials in the political arena, which allowed them to present their case and plea for a reconsideration of how the VAT was implemented. The government did not waive the VAT but offered a compromise by specifying that tax authorities would not impose VAT on licences and fees, which they had done since the VAT was introduced. Hence, the bargaining ended with a micro-level fiscal contract, although the revenue outcome was lower than initially anticipated, as the government waived some VAT payments by the sector. Table 5.2 Summary of the tourism sector case Sequence Arenas Strategies Policy outcome Before announcement June 2016 Political arena: voice in budget and parliamentary committees Dialogue and negotiation ⇒Agreement to postpone introduction of VAT Between announcement and approval by Parliament Political arena Lobby parliamentarians ⇒Government unwilling to change policy After approval by Parliament; second half of 2016 Public arena: use of media Non-compromising strategies: tactics of blame and criticism in the media by both parties However, also some compromising strategies: e.g. writing to minister asking for dialogue (although meetings did not materialize) ⇒Government unwilling to change policy Late 2016 and first half of 2017 Political arena re-established somewhat: Meetings with ministries / budget committee ⇒(re-)engagement in forums of exchange Compromise and influence: Strategies by tourist sector: – engage with different stakeholders in Dodoma (build relationships) – share information on importance of and impact on the sector – question legality of VAT implementation Strategies by ruling elite: – listen and offer dialogue (budget committee) – bring in other government stakeholders ⇒Clarification: no VAT on licences and fees Tax reforms in Tanzania 117 5.3.2 The transport sector The transport sector covers the provision of passenger or freight transport, whether scheduled or not, by rail, pipeline, road, water, or air and associated activities such as cargo handling, storage, etc. Given the various subsectors, organizations, and companies that comprise it, the sector as a whole is not always well organized.⁸ As in the tourism sector, the government introduced tax reforms that affected the transport sector.⁹ In addition to the sector’s non-participation in the decision, the timing and the expected hasty implementation of the reforms caught members by surprise. The rapid pace with which the government moved from the proposal phase to incorporating reforms in the 2016/17 finance bill denied the sector time and space to mobilize a coalition against the tax reforms. In addition to reducing the volume of imports and transited cargo, the reforms precipitated a price war among transporters as the transited cargo sector struggled to remain operational. Increased licence fees forced the closure of many small and medium-scale transporters. Thus, while the industry continued to portray a collective and cohesive external front, there were internal disagreements. Upon learning of the reforms on 1 July 2016, one of the sector’s representatives, the Tanzania Association of Transporters (TAT), convened a series of meetings with members, clients, and affiliates—including the freight forwarders’ association (TAFFA) and the truck owners’ association (TATOA)— to establish the scale of likely impact, a common position, and a strategy for engaging the state to address their concerns. Of primary concern was the maximum time an importer had to clear their goods without incurring additional port and handling charges (wharfage and storage) that would incrementally have an impact on the amount of VAT incurred. At the meetings, members agreed to advance two major positions: request a moratorium on new charges and petition for an increase in maximum allowable time to clear goods from the revised 14 days to at least 90 days. The establishment of a sectoral common position on the reforms provided the TAT with a mandate to find a compromise with the authorities. However, a supporting coalition had to be established discreetly rather than through open channels, as senior ministry officials were wary of being seen as overtly siding with the private sector in its agitation against the government. TAT was forced to rely on informal approaches to present its initial case, which involved direct but discreet personal appeals by one of its members to the ⁸ For an interesting account of the limitation of collective action in the transport sector, see Rizzo (2017). ⁹ For details on the tax reforms see Msami et al. (2022). 118 Marianne S. Ulriksen, Lucas Katera, and Jamal Msami minister and his permanent secretary. It was also indicated in interviewees that these meetings involved exchange of money for support. Thus, the TAT was forced to co-opt the ministry, which then guided it in how toengage with the powerful budget and transport parliamentary standing committees. The TAT also claimed that it had to furnish payments to prominent members of these committees (as well as a few prominent opposition MPs) in exchange for their support. Collectively, the purchase of support allowed the TAT to establish a powerful coalition of actors capable of heaping and sustaining pressure on the treasury. One of the coalition’s first acts was large-scale awareness-building of the adverse effects of the reforms on the transport sector. This was done in collaboration with the TAFFA and the TPSF. Awareness-building occurred through media briefings, interviews on national print and digital media, as well as audio and television—with multiple appearances by the TAT’s top officials. Through its members and networks, the TAT succeeded in influencing the national media to publicize its cause adequately. A key element in its media strategy was that ordinary Tanzanians would lose their jobs due to loss of business and decline in import volumes. The TAT’s message was simple: the reforms were hurting national livelihood and pride. This did strike a chord with the authorities as they moved to reassure the public and neighbouring landlocked countries of the viability of the port of Dar es Salaam. In March 2017, the TAT, along with other members of the TPSF, was invited by the Treasurytopresenttheir caseataseriesofpre-budgetary meetingsinvolvingparliamentary committeesandthegovernmentin Dodoma. At the meetings, only the TAT and the tourism lobby were given substantialtime to present their cases, mainly because they—through their media engagements in the public arena—were at the forefront in complaining about the introduction of VAT, which would reduce their competitiveness and affect the national economy due to loss of jobs and revenues. It appears that the returntothepoliticalandbureaucraticarenasinwhichtheTATcouldpresent details on the implications of the VAT to the sector caused the government to waive VAT charges on ancillary services, and the 2017/18 Finance Act passed in July 2017 confirmed the revisions of the reforms sought by the TAT. Summing up (see Table 5.3), the introduction of VAT and other tax charges cameasasurprisetothe transport sector.Initially, the sectorwaspoorlyorganized and lacked a unified voice in its interactions with the government, but the tax reform prompted the sector to organize and find a common position through the TAT. The sector was able to gain increased access to the political arena by buying support in the bureaucratic arena and via a successful Tax reforms in Tanzania 119 Table 5.3 Summary of the transport sector case Sequence/ timeline Arenas Strategies Policy outcome At announcement (July 2016) Introduction of tax a surprise ⇒no access to political arena Right after announcement (July–Sept 2016) Gaining access to bureaucratic arena Sector establishes internal collaboration and buys access to side-lined ministry Sept 2016–Mar 2017 Public arena Awareness-building in the media Nov 2016–June 2017 Increased access to political arena (invited by treasury) Direct negotiation in discreet meetings ⇒waiving VAT charges on ancillary services public campaign in the news media. Thus, a micro-level fiscal contract was reached, although the government would receive less revenue than initially anticipated. 5.3.3 The members of parliament The ruling party, CCM (Chama Cha Mapinduzi—‘Party of the Revolution’), has since the inception of multi-party elections in 1995 accounted for 71– 93%(currentlyabout93%) ofMPs. Theopposition,comprisingfourpolitical parties, has largely presented itself as a unified block to counter its minority position. In June 2016, as part of the government’s push to reform the tax system, the government proposed to waive tax exemptions on MPs’ gratuities, meaning that MPs would pay 30% in tax of their gratuity payment at the end oftheirfive-year term.Thiscasestudydocumentstherevenuebargainingfollowing the tax proposal, which led to a compromise: waivers of exemptions on gratuity payments would be a 5% levy (rather than 30% tax), and it would take effect in 2020 after the end of the current parliamentary term. Stakeholders involved in gratuity tax reforms were adamant that the bargaining outcome represented a poor compromise, with major concessions given by the government. Apparently, the motive behind attempts to tax MPs stemmed from the ruling party’s loss of parliamentary seats, increasing costs of elections, and the need to weaken the financial base of opposition MPs. However, due to multi-party representation, the government had to disguise these attempts at weakening parliamentary democracy by imposing the tax 120 Marianne S. Ulriksen, Lucas Katera, and Jamal Msami on the ruling party’s own MPs. This caused cross-party mobilization against the tax, with MPs on both sides of the political divide adopting creative strategies to agitate against reforms. The MPs realized that they could not easily find strong popular support due to past public opposition to hikes in MPs’ salaries and gratuity payments, first in 2009 and then in 2014. Consequently, MPs sought to influence public opinion more discreetly in two principal ways: CCM MPs emphasized that their gratuities and allowances provided out-of-pocket assistance to their constituents, whereas opposition MPs publicly endorsed the reforms and challenged the government to extend them to the exempted gratuities of other political elites, notably the President and the Prime Minister.¹⁰ While the former strategy sought to highlight the productive uses of gratuities and the burden facing individual MPs, the latter challenged the reach and legitimacy of the reforms. These measures backfired, however, as the president revealed his earnings and indicated his willingness to be taxed. Riding on a wave of favourable public opinion, the president appealed for public support for reforms by highlighting the discrepancies in earnings between MPs and ordinaryTanzanians,urgingMPsto liveby their politicalpromisesofserving ordinary Tanzanians by acquiescing to the proposed reforms. Not be deterred, MPs enlisted the support of the powerful religious and civil society community, keenly aware of the mobilization powers of mainly religious bodies, owing to their contribution to social services as well as the strong religious values of the Tanzanian society. The head of the powerful parliamentary budget committee offered concessions to religious groups by suggesting a revision of proposed tax rates to the groups in return for their support in fighting the gratuity tax reforms. Facing the prospect of an undesired alliance between MPs and civil society, the government swiftly moved to placate religious groups by proposing a maintenance of exemptions for the religious and civil society community. This left MPs in a difficult position, unable to mobilize support from any legitimate external constituency. Facing a lack of options, MPs threatened to wield a collective veto by rejecting the 2016/17 finance bill. The president Magufuli retaliated by threatening the use of constitutional powers to bypass Parliamentand/ordissolveitandcallageneralelection. WithinCCM, senior figures were wary of the financial and political consequences of fresh elections and appealed to the council of elders, consisting of past party chairmen and secretary generals, who helped convey a message to the president for a ¹⁰ https://en.igihe.com/news/tanzania-leaders-gratuity-for-income-tax;https://www.ippmedia.com/ sw/makala/wabunge-ccm-walivyogeuka-mbogo-kodi-ya-kiinua-mgongo. Tax reforms in Tanzania 121 negotiated compromise. This was reached with a downward revision of the tax rate from 30 to 5%, rescindment of intended tax reforms on allowances, and postponement of application of taxes until 2020. In sum (see also Table 5.4), MPs were not able to speak with one voice, even though the majority was against the tax reform. Engaging with the political leadership in the public arena, both CCM and opposition MPs were confrontational but used different arguments to appeal to the public. However, the strategy of using media stories backfired when the President proved unwilling to make any concessions. Moreover, an attempted alliance with civil society organizations proved unsuccessful. In the end, it was in the political arena that the revenue providers were able to reach a compromise with the government. The MPs collectively threatened to veto the budget, thereby—as with the transport sector—revealing the importance of mobilizing collective action. Nevertheless, it appears that informal institutions and relations within the CCM party’s inner circles were decisive in enabling MPs to push the president to soften his initial stand. Once again, as with Table 5.4 Summary of the MP case Sequence Arena Strategies Policy outcome Initial proposal (2016) Public arena Use of media – CCM MPs emphasize their out-of-pocket assistance to constituencies – Opposition MPs challenge for reform to be extended to others in the ruling elite President retaliates in media to get public support ⇒No change Latter half of 2016 Public/political arena MPs seek alliance with civil society, i.e. religious groups (possibly due to strategic political positions/knowledge) Government makes separate agreements with religious groups ⇒No change (but for civil society maintenance of exemptions) First half of Political arena Overt: threat to reject finance bill 2017 Covert: Key CCM members seek to convince President to give concessions ⇒Government gives concessions, although policy remains (5% instead of 30%, and by 2020) 122 Marianne S. Ulriksen, Lucas Katera, and Jamal Msami the tourism and transport sectors, a micro-level fiscal contract was reached, although the government compromised its initial stand. 5.3.4 The security forces The army, police, and prison forces comprise a large share of civil servants in Tanzania, and the forces are strong politically because their devotion to national service and national security is considered very important (URT, 2016). The security forces had been benefitting from duty-free shops operating at the Tanzania Peoples’ Defence Force, Police and Prisons’ barracks and selling a range of products, both durables and non-durables, at prices that excluded both VAT and excise duty. Legally, only members of the Tanzania Peoples’ Defence Force, Police Force, and the Prison Service had access to theshops,butmisusewaswidespread,causinglossofgovernmentrevenue.In addition, it is claimed that senior officials sold goods from the duty-free facilities in other shops at cheap prices in unfair competition with other traders paying VAT. This behaviour created difficult business environments for those complying with the tax laws. There were persistent public complaints about the misuse of duty-free shops, and the business community criticized the unfair competition.¹¹ Concurrently, in its preparation of the 2016/17 budget, which was characterized by VAT reforms in many sectors, the government saw an opportunity to reform the tax exemptions of duty-free shops and announced removal of the exemptions for the financial year 2016/17. Compensation was introduced in the form of an additional salary payment, a flat rate of TZS 100,000/month to all soldiers regardless of rank. This amount was expected to cover the costs of taxes that army members now had to pay in the shops. According to the Minister of Finance and Planning during the budget speech of 2016/17, the government considered this the best targeting mechanism while also avoiding misuse. It is difficult to fully detect what happened, but it is probable that this case involved a pre-emptive concession by the government to counter possible opposition to the tax. Arguably, there was also informal bargaining between government and top-level officials in the security forces, who may have been more interested in a solution benefiting the forces broadly. At least, it is interesting to note that lower-level members of the security forces were positive about the new compensation scheme, whereas mid-range senior officials ¹¹ https://www.thecitizen.co.tz/tanzania/news/national/no-more-tax-exemptions-for-military-shops- -2557650 Tax reforms in Tanzania 123 Table 5.5 Summary of the security forces case Sequence Arena Strategies Policy outcome Before proposal Public arena Public outcry and complaints by business sectors provide legitimacy of reform of duty-free shops Political arena Although hard to observe directly, government gives pre-emptive concession, either in agreement with top army officials or on its own ⇒pre-emptive concession Announcement of proposal Public arena Tax reform of disputed army barracks announced ⇒tax reform, but compensatory scheme introduced at the same time were more critical (Msami et al., 2022). Perhaps this is not surprising, as a flat payment would be proportionally more lucrative at lower salary levels. To sum up (see also Table 5.5), the general move by the government to reform the tax system as well as public concerns about the misuse of dutyfree shops urged the government to remove tax exemptions. However, the security forces are clearly of such political importance that the government pre-empted any upsets by introducing a compensatory scheme despite the financial implications for tax mobilization. While senior officials decried the changes as they had been benefitting disproportionately from the exemptions, their influence did not counter possible agreements made between top-level officials and the government or the general support for the new scheme among low-level members in the security forces. In addition, the public arena was not a place for engagement between revenue provider and government as they had other arenas for dialogue. Instead, public debates gave the government’s tax reforms some legitimacy as it made the government appear proactive in dealing with a widely recognized problem. Again, the final compromise was most likely settled in the political arena, although we cannot prove this due to the secretive relations between the army and the government: either the government came up with a compensation scheme itself, or it was agreed in consultation with top-level officials. 5.4 Case comparison and conclusion In this chapter, we contribute to the understanding of how and when revenue bargaining leads to direct government concessions, and we have an 130 Odd-Helge Fjeldstad and Lise Rakner the Big Four global accounting and professional services firms,⁴ as well as a former deputy commissioner general of the Tanzania Revenue Authority. Individualbusinesspeoplealsolobbiedfortheirspecificinterests.Inaddition, public-sector agencies, the Ministry of Agriculture, and the ministry responsible for tourism mobilized against the bill. The lobbyists argued that the abolition of exemptions would make the country unattractive for investors and leave Tanzanian companies uncompetitive in domestic and regional markets. Working together in a concerted fashion, business associations and other lobby groups succeeded. Even though the governing party, the CCM, holds a majority vote in Parliament (Bunge) and the bill had been drafted by a technical team appointed by the government, the VAT Act that the President signed in January 2015 reintroduced many exemptions that had been abolishedinthedraftbillandrestoredtotheMinisterofFinancesomediscretionary power to grant further exemptions. A member of the VAT technical team in the Ministry of Finance described the new act as ‘old wine in a new bottle’.⁵ AnalysingtheprocessleadinguptothenewVATAct2014,ourstudymakes three contributions. First, it shows that tax lobbying in Tanzania has become better organized and coordinated through the larger business associations, sometimes under the umbrella of the Tanzania Private Sector Foundation. Their capacity for collective action has been strengthened and contributed to shaping the ability of business groups to make demands of the government (Prichard, 2015). In the context of the VAT Act 2014, well-organized associations were able to mobilize more powerfully for their demands, partly by engaging professional tax consultants and lobbyists to promote their position to parliamentarians and senior government officials. The enhanced role of the international accounting and consultancy firms in revenue bargaining in Tanzania is a reflection of the importance of lobbying and the substantial resources spent on influencing policymakers and legislators. Professional tax advisors with in-depth knowledge of the tax legislation target their lobbying towards influential stakeholders, including parliamentarians and public agencies who would be affected by the proposed legislation. ⁴ The Big Four are the world’s four largest international professional services networks, offering audit, tax, consulting, advisory, actuarial, corporate finance, and legal services. Ordered by size they are PricewaterhouseCoopers (PwC), Deloitte, Ernst & Young (EY), and KPMG. In 2012, they had a combined turnover of USD112 billion, with 2,800 offices and over 700,000 employees worldwide. All four companies have businesses in over 150 countries (based on information published in the official global websites of the firms PwC, Deloitte, EY, and KPMG). ⁵ Interview, Dar es Salaam, 11 March 2015. Lobbying in tax policymaking: The case of VAT reform in Tanzania 131 Second, the study adds important nuances to our understanding of legislative performance in singleand dominant-party regimes. Descriptions of African legislatures in the popular press and academic works often depict these institutions as uniformly weak and inconsequential for political and policy outcomes. However, recent scholarly work challenges this view (Opalo, 2019;Collord, 2019). This research underlines how in dominant partyregimes,parliamentaryinstitutionsplayasignificantrole,asparliament has become an important bargaining arena between powerful political elites and the executive; see Chapter 2 of this volume). Tanzania is considered one ofthemostinstitutionalized dominantone-party regimes in Africa(Nyirabu, 2002;Morse, 2014;Collord, 2019). As it is an institutionalized, dominantparty state, it could be assumed that once government is united, there is very little room for lobbying Parliament. Arguably, as a unitary actor, we would expect reforms approved by the executive to be approved by Parliament with relative ease, in contrast to less institutionalized political systems characterized by political alternations and deinstitutionalized party systems (Rakner and Svåsand. 2013;Opalo, 2019). However, in connection with the VAT reform,we findthatthegovernmentwas not abletoinstruct individualmembers of parliament who proved receptive when confronted by an increasingly professional group of lobbyists. In Tanzania, during the past decade, political elites have sought to entrench their bargaining power through legislative reforms, creating the institutional resources needed to better extract benefits from the executive. The bargaining process behind the 2014 VAT Act is illustrative of the complexities of implementing tax reforms in contexts of multiple interests (see also Chapter 5 on revenue-bargaining processes in Tanzania that include a multiplicity of actors and strategies). Third, our study challenges the argument that bribery is the preferred way to influence policy decisions in developing countries whereas lobbyism is more common in rich countries (Harstad and Svensson, 2011). Our study shows that this is an oversimplification that does not reflect how policy influencing actually takes place in poor, weakly regulated countries. Lobbyism might be more prevalent than previously assumed in the literature (Fjeldstad and Johnsøn, 2017). When interest groups grow in influence, companies and business associations may develop a preference for lobbyism over corruption. Our analysis of the implementation of the VAT reform in Tanzania shows that there are numerous entry points for lobbying. Having escalated in the period after 2000, elite bargaining around policy reform in Tanzania now involves private-sector financiers who contest for parliamentary seats, fund MPs’ campaigns, and lobby in an ad hoc, issue-based manner (Collord, 2019, 219). 132 Odd-Helge Fjeldstad and Lise Rakner The analysis draws on findings from research carried out in Tanzania during the period 2014–2016. We analyse the development of the VAT Act, from the process started in December 2012, to the presentation of the draft bill in Parliament in May 2014, through to its final vote in Parliament in November 2014, and its final signing by the Executive in December 2014. Our main sources of data are: (i) careful reading of the VAT bill (draft May 2014), the amended bill (November 2014), and the VAT Act (December 2014); (ii) interviews with representatives of the business community, individual business people, partners of the Big Four international accountancy and advisory companies, domestic tax practitioners, IMF representatives, staff of the Ministry of Finance and the Tanzania Revenue Authority (TRA), and sector interests; and (iii) reports and newspaper articles collected for the period December 2012–June 2016. Our findings resonate with other case studies in this book, and, as such, contribute to substantiating and elaborating the theoretical framework. The remainder of the chapter is divided into five sections. In the next section, the VAT reform is presented. The third section explores the conditions that triggered the VAT-reform initiative and provided entry points for revenue bargaining. The lobbying process and its organization are examined in the fourth section. The fifth section discusses the outcome of the revenue bargaining as reflected in the new VAT Act. A concluding section completes the chapter’s analysis. 6.2 The case: Background for the VAT reform VAT was implemented in Tanzania in July 1998, advised by and with technical assistance from the IMF.⁶ The VAT was expected to broaden the tax base, leading to substantial increase in tax revenue without distorting investment decisions. With few exemptions and zero-ratings, the original VAT Act of 1997 was perceived by the IMF and international tax advisors as a ‘best practice model’ for VAT design in a developing-country context. Subsequent developments, however, deviated from the that model.⁷ When the new act was introduced in 1998, it was estimated that VAT would generate revenues equivalent to 6% of GDP. However, the VAT regime was gradually undermined by exemptions, and revenues never exceeded 4.5% of GDP. ⁶ As early as 1991, introduction of VAT was proposed by a Tax Commission appointed by the government (United Republic of Tanzania, 1991, chapter 12). However, before 1997 little progress was made in this area. ⁷ Interview with senior officer, Research and Policy Department, TRA HQ (member of the VAT Technical Team), Dar es Salaam, 11 March 2015. Lobbying in tax policymaking: The case of VAT reform in Tanzania 133 Over time, amendments of the Act incorporated an expanding number of exemptions and zero-ratings of goods and services and of persons entitled to receive exempt supplies. Some of these exonerations were promoted by the Tanzania Investment Centre (TIC) in order to attract investments. Generous tax incentives granted to multinational companies, especially in agriculture and extractive sectors such as mining, led domestic enterprises to lobby for tax exemptions to adjust for the perceived unfairness of the tax regime.⁸ The VAT Act provided discretion and hence a space for domestic companies to request for exemptions. The Minister of Finance had power to grant exemptions and could channel requests for exemptions through the Task Force on Tax Reform (composed of representatives from the public and private sectors), a process that made it relatively easy to add exemptions.⁹ Other ministries and government agencies were also involved in tax policymaking. Fragmented policymaking and bureaucratic competition made it easy for interest groups and individuals to lobby for their interests.¹⁰ Accordingto Ali Mufuruki,aleadingbusinessowner in thecountry,a‘tax-incentives industry developed to advise clients on how to legally access incentives’.¹¹ The Big Four became a big part of this industry. The numerous exemptions and zero-rated goods and services complicated the underlying VAT structure, caused complexity in the tax system, and added to widespread leakages through the many opportunities for abuse and avoidance. This also had adverse effects on revenue generation. By 2012, Tanzania ranked among the countries with lowest VAT productivity among the Southern African Development Community (SADC) and East African Community (EAC) countries. An IMF aide-mémoire from 2012 describes theTanzanianVAT regimeasfollows: ‘The ever-expandinglistofpreferences has deteriorated VAT tax base. Tax exemptions of inputs add to tax cascading and numerous tax relieves seriously undermine VAT revenue productivity’ (Kreloveet al., 2012, 46).By2012,seniormanagersintheMinistry ofFinance and the (Tanzania Revenue Authority) TRA recognized that the VAT regime needed reform.¹² The Minister consulted the IMF, an IMF mission was put ⁸ By the end of 2014, 80 per cent of the strategic investors granted tax exemptions by the Tanzania Investment Centre were domestic companies. Interview with Director of Investment Facilitation, TIC, Dar es Salaam, 17 November 2014. ⁹ Interview with two senior officers, Research and Policy Department, TRA-HQ, Dar es Salaam, 11 March 2015. ¹⁰ Interview with IMF representative, Dar es Salaam, 25 August 2014. ¹¹ Interview, Dar es Salaam, 10 November 2014. The late Ali Mufuruki was a co-founder and chairman of the CEO Roundtable of Tanzania, a policy-dialogue forum that brings together more than a hundred CEOs of leading companies in Tanzania. ¹² Interviews with members of the VAT Technical Team from the Ministry of Finance and the TRA, Dar es Salaam, 11 March 2015. 134 Odd-Helge Fjeldstad and Lise Rakner in motion, and external VAT consultants were engaged. A technical reform team composed of staff from the Ministry of Finance, the Tanzania Revenue Authority, and the Office of the Attorney General was established. According to the IMF, the original VAT regime in Tanzania from 1997 was almost ‘textbook’, but some structural features affected its intentions negatively over time and contributed to the low tax-to-GDP ratio. Two problems were noted: (i) poorly structured refunds (paid late, leaving businesses cash strapped, legitimizing the ‘cry for exemptions’),¹³ and (ii) the stated aim of providing incentives for foreign capital (FDI). Amendments of the VAT Act of 1997 introduced manyexemptionsandzero-ratingsforintermediate inputs tovarious industries and sectors, originally intended in part to work around tax refund problems, in part to enhance investment and protect the local market, and in part to reduce the tax burden on consumers and some sectors, especially agriculture (Krelove et al., 2012). 6.3 The VAT-reform initiative and conditions that triggered revenue bargaining While revenue concerns played a major role in the initiation of the VATreform process, the reform was also triggered by pressure from various stakeholders demanding more tax fairness. The excessive VAT exemptions were routinely lamented by civil society groups, Members of Parliament, as wellastheControllerandAuditorGeneral(CAG), who inhisvariousreports pointed to the increasing and escalating exemptions (IMF, 2008;Uwazi, 2010;TRA, 2011;Ndunguru, 2012;CAG, 2013;IFC, 2013).¹⁴ The extensive discretionary power of the Minister of Finance to grant exemptions was also noted as a major problem by all parties. Based on recommendations from the IMF, pressure from the international donor community as well as increasing pressure from the main opposition party Chadema, the Parliament’s Public Accounts Committee (PAC) in 2012 directed the Controller and Auditor General to conduct a special audit of the exemptions to ascertain whether the waivers were beneficial to the country. In 2013, the government decided to develop a new VAT Act and established a technical reform team to lead this work. The team was composed ¹³ In Tanzania, like in many other many African countries, it may take several months, and sometimes up to a year, to process refund claims (Carter, 2013). ¹⁴ In an interview in The East African (8 February 2014), Mr Rakesh Rajani, directorof the civil society organization Twaweza East Africa, questioned the government’s commitment to review the exemptions, saying the job should take less than six months to accomplish, but that the government had been dragging its feet. Lobbying in tax policymaking: The case of VAT reform in Tanzania 135 of staff from the Ministry of Finance, the Tanzania Revenue Authority, and the Office of the Attorney general, with support from an external VAT expert recruited by the IMF. During 2013 and 2014, a new VAT bill was drafted. According to the IMF resident representative in Tanzania, the original draft was very ambitious, almost aiming to be a ‘perfect VAT Act with very limited number of exemptions’.¹⁵ The new bill included a few exemptions and zero-ratings (food and basic necessities), but all special reliefs for named bodies were abolished (URT, 2014a). The draft bill removed the discretionary power of the Minister of Finance to grant and modify tax exemptions (ibid., 88). The bill stipulated that exemptions be approved by Parliament and that new exemptions should be created or modified by Parliament only. The draft VAT Bill was sent to Parliament from the Ministry of Finance in May 2014. The work leading up to the draft bill was run by the technical reform team supported by the IMF. According to a member of the technical team, the multi-stakeholder Task Force on Tax Reform ‘was not consulted since it was seen as being part of the problem’.¹⁶ The first version of the bill was drafted by a technical expert engaged by the IMF. Thereafter, the Tanzanian VAT Technical Team made some ‘improvements of the draft act during first half of 2014 to make it better fit the Tanzanian context’.¹⁷ The IMF’s resident representative confirmed that the Tanzanian team did not take on board all recommendations from the IMF, claiming they would not work in the Tanzania context.¹⁸ In this process, the technical team also had consultations with some Cabinet members. According to members of the technical team, an early draft of the VAT bill was sent to the Confederation of Tanzania Industries (CTI), the Oil and Gas Association of Tanzania (OGAT), the Tanzania Petroleum Development Corporation (TPDC), the Tanzania Private Sector Foundation (TPSF), and the Big Four. However, according to the TPSF, no consultations took place prior to the bill being presented to Parliament in May 2014.¹⁹ Conceding this point, the IMF representative referred to the drafting of a bill as a rather closed process, where technical assistance and outside forces are somewhat limited in their inputs: ‘By Tanzanian standards there were some consultations, but consultations could have been made better, more widely, and ¹⁵ Interview, Dar es Salaam, 27 October 2015. ¹⁶ Interview, Dar es Salaam, 11 March 2015. ¹⁷ Interview with members of the technical team, Dar es Salaam, 11 March 2015. ¹⁸ Interview, Dar es Salaam, 27 October 2015. ¹⁹ Interviews with Ali Mufuriki (CEO Roundtable Tanzania), 10. November 2014; Lathifa Sykes (CEO Hotel Association of Tanzania), 19 March 2015; and Edward Furaha (Research and Policy Manager, TPSF), 27 October 2015. 136 Odd-Helge Fjeldstad and Lise Rakner involved more interest groups for example from the private sector’.²⁰ He further reflected that: ‘IMF had expected that the Cabinet would approve the May 2014 Bill and agree on the Bill before it was sent to the Bunge. This did not happen. The line ministries—especially tourism and agriculture—were notawarethattheexemptionsfortheirsectorshadbeenremovedandreacted when they saw the bill’.²¹ It is not surprising that different stakeholders hold different views when describing a consultation process that happened some time back, a process with which at least one party (the private sector) was unhappy. However, based on evidence acquired through interviews with key stakeholders, it is evident that the process of drafting the 2014 VAT bill was neither participatorynorconsultative.Keystakeholdersintheprivatesectorwerenotproperly consulted, and some Cabinet members and ministries, especially Tourism and Agriculture, had not been provided with adequate information about the short-term consequences of the proposed changes of the VAT Act. The fact thatthe Parliament’sBudgetCommitteesentthedraftbill backtotheCabinet for a redrafting adds credence to the argument that stakeholder consultations were limited in the first stage of drafting the VAT bill. The above analysis illustrates that revenue bargaining can be triggered by reforms drafted by technocratic government staff (or external actors) tasked with how to increase revenue. The case further illustrates that such ‘technocratic reforms’ have distinct limits, as stakeholders who have not been consulted in the initial stages of a reform may seek to influence the design of the reform before it is finally tabled as an act. 6.4 Political settlement and arenas of revenue bargaining The political settlement in Tanzania is rooted in a strong institutionalized party dominating the parliament upheld and legitimized through its ties to the country’s business elites (Ishiyama and Quinn, 2006;Babeiya, 2011; Gray, 2018).²² The ruling party the Chama Cha Mapinduzi (CCM)²³ has governed Tanzania since independence with remarkable stability (Phillips, 2010;Coulson, 2013;Lofchie, 2014). Whereas legal reforms have officially ²⁰ Interview, Dar es Salaam, 27 October 2015. ²¹ Interview with IMF representative, Dar es Salaam, 27 October 2015. ²² The CCM’s majority position does not include Zanzibar, where the opposition is strong. The main opposition party in mainland Tanzania, Chadema, has succeeded in winning a number of subnational elections since 2010 in several of the major cities, including Dar es Salaam and Arusha. ²³ Chama Cha Mapinduzi means ‘Party of the Revolution’ in English. Lobbying in tax policymaking: The case of VAT reform in Tanzania 137 detached the party from the state, most state institutions are still informally linked behaviourally to the ruling party. The lack of separation between the state and the party (Makulilo, 2008) makes the CCM a state-party or a party-state.Theparty remainsthe‘partyofchoice’foranyoneseekingapolitical career (Therkildsen and Bourgouin, 2012). As noted by one Tanzanian describing the CCM’s likelihood of winning the 2005 elections, ‘CCM has its roots everywhere . . . . It has taken up all the earth. There is no room for other partiestogrow’(Phillips,2010,109).Signallingtherulingparty’scontrolover commercial business, Babeiya (2011, 95) argued that: ‘The use of threats also continues to guarantee CCM much support from the business community as businesspersons fear repression if they do not support the incumbent party’. With economic liberalization, the CCM’s ability to retain centralized control over wealth accumulation began to unravel due to a combination of economic decline, growing informalization, corruption, and, ultimately, economicliberalization(Kiondo,1994;Mmuya,1998;Babeiya,2011).Asargued by Collord (2019), Tanzania’s changing economy saw CCM leaders adopt a new pattern of politicized accumulation, one characterized by ad hoc connections between an expanded private-sector elite and various factional networkswithintheCCMitself.BusinessownersofAsianoriginwontwoout of four by-elections between 1992 and 1995, an unprecedented occurrence in Tanzania (ibid., 118). Despite the CCM’s dominant position and executive dominance, Babeiya (2011, 97) argues that individual MPs’ influence relies heavily on private resources in their election campaigns, and their individual resources are shaping voting behaviour: ‘The power of the purse is currently one of the tickets to pass the nomination and election test in both party and intra-party competitions’. Corresponding to the earlier discussion of government capture, arguably this culture of private wealth shaping electoral outcomes, businesspeople are venturing into political careers, suggesting a strong link between business and politics (see Chapter 9 of this volume). The political settlement outlined above could be expected to have yielded a different outcome of the VAT revenue-bargaining process. In a state where oneparty dominates,holdinga majority oftheparliamentary seats,wewould not expect a law endorsed by the government to be altered by lobbying or treatment in Parliament.However, because individual MPs rely heavily on resources and business support to finance their electoral campaigns, the VAT bill was susceptible to changes during the vote in Parliament. During the bargaining process, the draft bill was substantially reshaped. For instance, many itemsthatarereferredtoas‘VATable’ intheMay2014Bill areexemptedinthe VAT Actpassed inParliamentinNovember2014.Whilecentralactorsingovernment, particularly in the Ministry of Finance and the Tanzania Revenue 138 Odd-Helge Fjeldstad and Lise Rakner Authority, and in the parliamentary opposition, NGOs, the media, and some donorshadarguedforasimplifiedVAT Actwithadrasticallyreduced exempt list, Tanzania ended up almost in ‘status quo’. Interviewswithbothmembersofthetechnicaldraftingteamandrepresentatives from the private sector established that the draft bill of May 2014 was met with protests from the private sector.²⁴ The private sector argued that the drasticremovalofexemptionswouldimply severechallengesforthebusiness environment in Tanzania. For the private sector, the key issue was fiscal predictability. They argued, for instance, that the tourism sector needed more time to adjust. The Parliament Budget Committee demanded more consultations due to what it to be considered weaknesses and shortfalls in the draft bill. Public hearings on the VAT bill starting in August 2014 led to extensive lobbying, especially for exemptions, and amendments of the bill. The debate that took place in the Cabinet as the bill was returned suggests that the Cabinet had not fully grasped the implications of the removal of exemptions for the private sector and some line ministries. The Cabinet was not united. Some ministries, in particular those of Agriculture and of Natural Resources and Tourism, were much in favour of tax exemptions for enterprises in their respective sectors. The Minister responsible for tourism personally lobbied for exemptions in Parliament. Members of the main business associations, the IMF, the TRA, and the Ministry of Finance all confirmed in interviews that lobbying gathered momentum after the drafted bill was presented to Parliament in May 2014. However, initial attempts to ensure that the new VAT Act avoided the problems of exemptions and loopholes associated with the previous act did not succeed. IMF initially suggested that Parliament should be involved in developing the new act, in order to reduce exemptions and sensitize MPs on the costs of exemptions. However, as Parliament only became involved late in the process when the bill was presented for a vote, the opportunity to build a common understanding in Parliament on the costs of exemptions was lost. This again left Parliament open for lobbying during the period after the draft bill was presented to the legislature. As a result, paradoxically, considering that the debate about the new VAT Act started with opposition MPs publicly shaming the VAT exemptions inherent in the amended VAT Act of 1997, Parliament did not emerge as a constituency favouring reducing exemptions. The claim that MPs were being lobbied individually by representatives of the private sector was confirmed by the private sector. As Edward Furaha ²⁴ Interviews with representativesfromTPSF,12November2014 and 27 October2015;managersfrom TRA, 14 November 2014 and 11 March 2015; and officials from the Ministry of Finance, 11 March 2015. Lobbying in tax policymaking: The case of VAT reform in Tanzania 139 of the TPSF said at the time, ‘Individual members who do not know issues are normally made to understand issues in the process of lobbying’.²⁵ A typical strategy would be to start with general issues that touch everyone (such as VAT imposed on funeral services) and use these as evidence of the impacts of thedraftbillofMay2014ifimplemented.ThedirectlobbyingofMPswasalso confirmed by public-sector representatives. According to a senior TRA manager: ‘MPs listen to too many lobby groups . . . . Allocation of responsibility is problematic . . . . MPs are swayed by the Big Four and TPSF. . . .… Clearly, governmentshouldhave donetheirownexplanation,aspoliticiansweremisled by lobbyists.’²⁶ While tax practitioners from KMPG claimed that KMPG had not been involved in the lobbying of MPs, they nevertheless conceded that extensive lobbying of individual MPs had taken place and that ‘the exemptions are all back and the draft VAT bill has gone through a complete overhaul’.²⁷ The VAT law changed considerably between its submission to Parliament for debate in May 2014 and the final vote in November. Lobbying against the VAT bill was coordinated by TPSF towards members of the Budget Committee. They also had meetings with the Ministry of Finance, presenting the private sector’s position. Much of this lobbying was done by tax experts from PwC engaged by the TPSF. According to Lathifa Sikes, representative of the Hotels Association of Tanzania (HAT), PwC and the TPSF were present in Dodoma to lobby MPs.²⁸ She argued, however, that they did not pay bribes to MPs: ‘We went to Dodoma and argued our case’. This observation is important and suggests that it is too simplistic to expect that businesses will prefer to bribe rather than lobby if both strategies can achieve the same goal. While it is difficult to substantiate that lobbying took precedence over bribery, the rise of large,reasonablywell-organized business associations,combined with stricter perceived enforcement of international legislation and conventions against corruption, lead us to tentatively conclude that lobbying was considered a preferred strategy by theTanzanian businesssector. Arguably, ‘revenue bargaining’ took a number of forms. The debate around the VAT Act witnessed little direct bribery. Instead, bargaining was done through several channels,includingindividuallobbyingwherecompaniesacted in their individual capacity, by individual business associations such as the HAT, and by coordinated efforts by the TPSF. ²⁵ Interview, Dar es Salaam, 27 October 2015. ²⁶ Interview with senior manager in TRA, 14 November 2014. ²⁷ Interview, Dar es Salaam, 18 March 2015. ²⁸ Interview, Dar es Salaam, 19 March 2015. 242 Marianne S. Ulriksen, Flora Myamba, and Constantine George 200 180 160 140 120 100 80 60 40 20 0 Irish aid USAID Gates Foundation GOT UN agencies Sida DFID IDA Fiscal years USD (million) 2013/14 2014/15 2015/16 2016/17 2017/18 Figure 11.1 PSSN financing trends by donors. Source: TASAF. expertise, technical assistance, and facilitation. Finally, the funding by the World Bank (IDA³ in the figure) is largely through loans (Ulriksen 2019). As mentioned, although the PSSN and the precedent pilot programmes were donor funded from the beginning, it was decided that the PSSN should be fully government financed by 2020 when a new phase of the programme (PSSN II) would be implemented. In the meantime, it was agreed in 2015 that the government should cover around a third of the budget per year—the financial requirements were estimated at USD 300 million per year, with the obligations divided between the government (USD 100 million), development partners (USD 100 million), and credit from the World Bank (USD 100 million).⁴ In practice, the financial requirements as well as the government’s actual contribution were downscaled. For instance, the estimated amount required for the operation of the PSSN between July 2015 and June 2016 was set to approximately USD 170.7 million. Of this, the government initially committed to pay out USD 44 million (TZS 70 billion) to the PSSN from the 2015/2016 national budget, thereby covering 25.7% of the estimatedprogrammecostsforthatperiod.However, intheend,thegovernment only contributed USD 7 million, which represents only 4.1% of the overall programme costs and only 15.9% of the government’s initial commitment (URT et al., 2016). Thus, in 2016, the government’s contributions were lower than anticipated. Moreover, at the time, there was much uncertainty about the government’s long-term financial commitments as the government did ³ IDA, short for the International Development Association, is part of the World Bank and has the objective of fighting extreme poverty. ⁴ As informed by TASAF in 2016. Who should pay? 243 Table 11.2 PSSN—funding requirements and available funds 2016–2020. 2016/17 2017/18 2018/19 2019/20 FUNDING REQUIREMENTS CCT costs 126,720,000 126,720,000 126,720,000 126,720,000 PW costs 26,285,714 26,285,714 26,285,714 26,285,714 Livelihoods 660,000 713,000 713,000 713,000 Inst. strengthening and op. costs 12,503,000 12,503,000 12,503,000 12,503,000 TOTAL 166,168,714 166,221,714 166,221,714 166,221,714 AVAILABLE FUNDS Balance from previous year 12,156,607 IDA-Additional financing 100,263,388 99,736,612 DFID 28,000,000 26,000,000 18,000,000 17,000,000 Government of Tanzania 7,000,000 Sida 17,200,000 17,500,000 17,500,000 17,500,000 Gates Foundation 893,720 1,966,620 908,860 UN Agencies 655,000 USAID 5,000,000 5,000,000 5,000,000 TOTAL 166,168,714 150,203,232 41,408,860 39,500,000 Financing Gap (16,018,482) (124,812,854) (126,721,714) Source: URT et al. (2016, 10–11)—joint review conducted by the United Republic of Tanzania (URT), World Bank, UNDP, UNICEF, ILO, DFID, and Sida. not provide its implementing body (TASAF) with any clear indication of its financial support of the PSSN programme (see Table 11.2), other than committing to 2016/2017 the same (low) funds as for the previous budget year. During our fieldwork, it was notoriously difficult to get a clear picture of the financing of the PSSN programme. This was in part because there were several new donors contributing in a variety of ways. However, in 2017, the estimates received suggested that the government’s own financial contribution still amounted to about only 4% of the full programme costs.⁵ Thus, around 2017, the Tanzanian government had not lived up to its financial ⁵ According to TASAF officials, the end of 2017 (predicted) financing for the full PSSN programme implementationstandsatUSD749.3million. ThegovernmentofTanzaniawill contributeUSD30million (= 4%). The rest of the funding will come from the following sources: IDA: USD 420 million (original credit USD 220 million, additional financing USD 200 million); DFID: USD 170 million; SIDA: USD 85 million; Organization of Petroleum Exporting Countries (OPEC): USD 16.5 million; USAID: USD 10.0 million; Irish Aid: USD 10 million; Bill and Melinda Gates Foundation: USD 4.01 million; and UN Agencies: USD 3.6 million. 244 Marianne S. Ulriksen, Flora Myamba, and Constantine George commitments, and there were good reasons to doubt that it would do so in the future. As an anonymous government official informed us: The government was expected to gradually take over the funding by 100% in 2020. I do not see that happening. Every year the government has been budgeting for PSSN . . . however, disbursement is another thing. The donors continued to push the government to increase its financial commitments, and the negotiations became particularly heated around 2017 and 2018 when the new phase (PSSN II) was on the table. Negotiations were difficult, in part due to the conflicting preferences of the Tanzanian government and the development partners, as we explain in the following section. 11.4 Conflicting preferences In 2015, the Tanzanian government agreed to gradually take over funding of thePSSN, partly becausekey politiciansandbureaucratsin governmentwere persuaded that the programme could reduce poverty (Ulriksen, 2019). That was also the year that a new president (still for the ruling party, CCM) was elected. Late⁶ President John Magufuli became known for his productivist ethos and centralization of decision-making (Cheeseman et al., 2021), which changed the government’s position on the social protection programme. The new President emphasized hard work and had a strong dislike for ‘handouts’ (Paget, 2021). In addition, with the President’s centralization of power, the policy coalition of domestic technocrats and development partners that had initially driven the introduction of the PSSN (Ulriksen, 2019) became less influential. Instead, the discourse in the government at large swung in favour of those sceptical of cash transfers and away from the proponents of the programme (Jacob and Pedersen, 2018). These developments make our case study even more interesting and relevant. During the period in focus, 2016– 2018, the contrasting position of the government vis-à-vis the donor agencies became even more pronounced, the negotiations consequently more difficult, and the ability to find a compromise even more remarkable. However, as presented in the following, the two parties’ preferences differed at the outset. The Tanzanian government did not regard the PSSN as a priority, whereas the donors were, unsurprisingly, supportive of it. These contrasting preferences played into the negotiations of the new phase of the programme (PSSN II) as explored in section 11.5. ⁶ President Magufuli died in March 2021, after our period of analysis. Who should pay? 245 11.4.1 The preferences of the Tanzanian government Although the Tanzanian government had agreed to the PSSN programme, there was a feeling that the programme fitted poorly with the fifth government’s development priorities. The CCT component in particular was criticized, and the condemnation became more pronounced under President Magufuli, who came to power in November 2015. As one source explains: The CCT does not match with the ideology of the ruling party and the philosophy of economic development. Providing free money to the market economy must create friction, and the majority did not agree it was the right way to go. This [the programme] raises the question of sustainability and dependency. This project design was not from inside but driven by the thinking of the funder [the World Bank]. (Government official) The hesitation towards the CCT came largely from a general belief that cash transfers will not help people out of poverty. There had been complaints that even beneficiaries of the public works component found it hard to improve their situation because the daily wages are very low. Many did not feel they saw tangible results. The feeling is that there is so much money given with little or no impact. This creates resistance. I know of a Regional Commissioner who said very openly that he doesnʼt want TASAF people at his office. But I also think such people are less informed about the programme. (Expert) There was also scepticism towards handing out ‘free money’, and negative stories on the misuse of cash transfers flourished. The media revelations of ‘ghost beneficiaries’—that some PSSN beneficiaries were not poor (enough) to be part of the programme—raised even more public misgivings about the PSSN.⁷ Although the numbers of ‘ghost beneficiaries’ were determined to be small (around 55,000 out of 5.4 million), the public attention magnified this scepticism among politicians. President Magufuli, who came to power on an anti-corruption stance (Paget, 2021), was particularly vocal regarding his reservations about the PSSN programme. Cashtransfersfitted poorlywiththePresident’s slogan‘hapakazi tu’, which roughly translates to ‘here is only work’. The President was concerned with poverty, but in achieving development, he emphasized the importance of ⁷Kennedy (2016);Makene (2016);The Guardian (2017b). 246 Marianne S. Ulriksen, Flora Myamba, and Constantine George productivity and infrastructure where people must lift themselves out of poverty through entrepreneurship and hard work. As one source explains: I donʼt think the government will take over the World Bank support for the PSSN given the hapa kazi tu ideology. To support this [claim], one may recall the recent drought scenario whereby the current government refused to support [give free food to] people who were starving. (Expert) Furthermore,asmayberemembered, when the governmentcommitteditself to the PSSN programme, there were champions in the government that successfully advocated for the programme despite some scepticism about cash transfers and long-term sustainability (Ulriksen, 2019). With the administration under President Magufuli, these champions were either no longer in government or silenced: To a large extent, the Presidentʼs stance is likely to influence other government leaders. When a president like Magufuli says no to putting money into the PSSN programme, who are you to say yes or vice versa? It is obvious his interest is more on construction. He is, therefore, limiting any champions, particularly within the government, who would have raised their voices on behalf of the poor beneficiaries. (Expert) Thus, although there had been reservations towards cash-transfer programmes in Tanzania in the past, the scepticism increased and become more vocal. This may, in part, explain the government’s limited financial commitment. Another issue is the general constraint of financial resources, which caused the government to afford the programme low priority in terms of financial disbursements. Being a lower-middle-income country, Tanzania has limited financial resources and many urgent needs and policy areas that require attention. Consequently, it is not surprising that all stakeholders we interviewed mentioned resource constraints as a general challenge facing all sectors. This is also supported in government documents; for example, in the Five-Year Development Plan 2016/17–2020/21 (URT, 2016, 15) it is stated that inadequatefinancing(underfundinganddelayofdisbursements)isakeychallenge impacting negatively on the implementation of development plans in the country. However, the Tanzanian government also spends less on social protectioncomparedtoothercountriesintheregion(Ajwadetal.,2018).The Who should pay? 247 resource envelope was further tightened by loan repayments and high recurrentcosts.Thus, some interview sources explained that Tanzania wasobliged torepay several World Bank/IMF loans that were taken around 2012/13, and that about one third of government revenue was for repaying loans. Another priority named through interviews was to pay salaries to public servants (at both centralandlocallevels, as wellaswithinthesocial services, e.g.toteachers), which absorbs a large bulk of the budget. Nevertheless, even within the existing funds available, other areas took priority. For instance, one interviewee argued that the government was particularly resource constrained because the country had to fully fund the general elections in 2015 (donors have, in the past, funded large parts of the election costs), which impacted directly on the PSSN programme: There was a time where the budget for PSSN was taken off due to election expenses. Nevertheless, this has not largely impacted the programme because PSSN components were not rolled out at one time but rather gradually. (Government official) The government, even more so with the new President, also prioritized infrastructure and construction, and if World Bank loans were to be taken, it should be for such areas and not for the PSSN where the World Bank had already committed funds. As a source explained: The government is still in a high need of soft loans to be taken for the construction, infrastructure, and other productive and economic activities. If the World Bank is willing to give loans for investment in infrastructure, then we will take [them]. Those are the loans we are looking for. (Government official) Within the social sector, the recently introduced provisions of free education to ensure that children from poor families get an education, and free health provision for children under five and the elderly above 70 years, were priority areas for the President. Some sources argued that this made the PSSN even less relevant than before: Evidence from the TASAF [pilot cash-transfer programme] showed increased schoolenrolment,butthegovernment is nowadvocatingthat there isfee-freeeducation and says this is a more effective way to improve education than the TASAF programme. (Government official) 248 Marianne S. Ulriksen, Flora Myamba, and Constantine George The new government has come with the policy of free education and [is] pushing for free mandatory health insurance especially for the poor and aged, which were the main arguments for the creation of the TASAF. Now, what is the separation between the two [PSSN and education and health services]? (Government official) In fact, the previously mentioned scepticism towards cash transfers clearly played a role, not just in terms of giving the PSSN lower priority than other policy areas but also in arguing that other programmes would be more effective in addressing poverty. A centrally placed government official within the Ministry of Finance’s Poverty Eradication Department went to great lengths to explain to us the advantages of savings and investment schemes that could encourage poor citizens to do ‘productive’ activities: I wish I could promote the savings programmes. I think the CCT is not sustainable; I donʼt think it is the right way to go. If we get more money from donors, we can continue with the CCT but minimize it while putting more focus on productive programmes. (Government official) In sum,theTanzaniangovernmentwasscepticalofthePSSN, particularlythe cash-transfer component, and found other social programmes more fitting to their development plans. This contrasted with the donors’ prioritizing the PSSN and desiring for the government take ownership. 11.4.2 The preferences of the donors The donor community in Tanzania, as in many other low-income African countries, had high regard for social protection programmes such as the PSSN: PSSN is one of the most effective and largest social protection programmes in Africa. It is well managed and is making a huge impact. The programme is also strong in that it is implemented through the government system rather than through a parallel donor system. (Donor representative) Whereas the World Bank initiated the PSSN, other donors started to support the programme due to its poverty-reducing potential. In a written Who should pay? 249 communication, a donor representative provided this explanation, following this question from the authors: Authors: Is your rationale for supporting the PSSN due to its ability to reduce poverty or (/and) as part of [your countryʼs] strategy to expand social protection/social security/social rights? Donor representative: It is mainly the first rationale, that is to reduce poverty and also that we believe it will contribute to better learning outcomes for children, not least girls (as you may know, the programme has conditions related to school attendance), and also that it will contribute to employment and income among the poorest (through the public works component)—these are results that are emphasized in our country strategy. While aid agencies see the value of the PSSN, they wanted the government to take ownership to ensure its long-term sustainability. It was also argued that the PSSN would be affordable: The financial commitment should be for the long term, and we believe that domesticfinancingispossible inthelongrun.They canafforditbut needtomaketheright decision and over time gradually take over from the donors. (Donor representative) However, as explained, the Tanzanian government was hesitant to prioritize funds to the PSSN, and this put the donor agencies in a tricky position. While wanting to ensure that the programme was not donor dependent, and therefore wanting the government to take over the programme, the aid agencies were also aware that it would be difficult to (threaten to) pull out, given the support that poor and vulnerable households receive: We have continued with the support even if we donʼt see much coming from the governmentʼs side, because we understand that pulling out our financing has a direct effect on the poor and vulnerable people who benefit from the programme. (Donor representative) Thus, the government and aid agencies had conflicting preferences with respect to the PSSN, which caused difficult negotiations. As one donor representative put it: ‘The relationship is not always easy. Both sides have their own vision, mission, and priorities for social protection’. Given the government’s limited interest in the PSSN as compared to other issues, one might 250 Marianne S. Ulriksen, Flora Myamba, and Constantine George ask why it then committed to the programme in the first place. One source offered an explanation: It is a tradition for the government to commit when donors ask it to do so. The government says yes, and they just default; this goes across all sectors. The government doesnʼt feel any liability to that. (Expert) Whatever the reasons, the status in the period of our study (2016–2018) was that the Tanzanian government was unsteadily to live up to its financial commitments to the PSSN programme. In the following, we explore how the governmentanddonorcommunity soughttopushthroughtheir preferences. Thisengagementbecameespecially heated during2018wherethedesignand future financing of the PSSN II were negotiated. 11.5 Revenue bargaining: Who should pay? The following analysis explores the bargaining between the Tanzanian government and international aid agencies over the programme design of the PSSN II and the disbursement of funds. We show how the Tanzanian government sought to influence the design of the programme to better fit its priorities, particularly by reducing the importance of the cash-transfer element. While the international donors were able to maintain a strong cashtransfer element in the PSSN II, they were unable to hold the government to its earlier commitments regarding disbursement of funds. 11.5.1 Programme design The first phase of the PSSN programme was set to run through to 2019. Thus, for our period of analysis, bargaining increasingly revolved around the continuationoftheprogramme(i.e.PSSNII).As mentioned,PresidentMagufuli had been highly sceptical about the cash-transfer element of the PSSN, and his strong public reactions affected the relationship between the government and the development partners. For instance, in July 2017, the President harshly criticized cash transfers, and he went so far as to order an immediate stop to the provision of cash transfers to refugees in Tanzania.⁸ Although, he did not mention any ⁸The Guardian (2017a). Who should pay? 251 specific names, his criticism seemed to be pointed at organizations offering humanitarian aid to refugees in Tanzania, including the World Food Programme and the United Nations High Commissioner for Refugees.⁹ In another incident, towards the end of 2017, the government refused at the eleventh hour to sign a memorandum of understanding (MoU) with Irish Aid that would otherwise have secured additional funds for the PSSN. The reason given was that the government preferred finances to be directed to the new PSSN II, with more ‘productive’ components, than to the existing programme dominated by the CCT (interview with government officials). The turn-around caused confusion in the donor community and left several Tanzanian government officials embarrassed as they were involved in the lengthy process of developing the MoU and keenly aware of donor funding cycles that encouraged disbursements to be made within the calendar year. The President subsequently directed the TASAF to review the PSSN and to consider some revisions to the design of the PSSN II (interview with government officials). Consequently, in August 2018, an alternative programme design document for the PSSN II was circulated. In the alternative design, the cash-transfer component was almost entirely removed and the focus was instead predominantly on public works and productive inclusion (i.e. trainingandcoachingtoenhanceparticipants’savingsandproductiveactivities).¹⁰ In governmentdiscourse,thealternativedesignwasnotachangeoftheentire PSSN but merely a change in priorities of the different programme components so that the focus was less on the ‘wasteful’ cash transfers and more on the other ‘productive’ components (interview with government officials). This was, however, in strong contrast to the development partners’ views of the PSSN II in which they regarded cash transfers as essential: In case there will be huge changes within the PSSN, we will need to reassess the programme and our commitment to financing it. Our interest is to keep the percentage ratios [of the programme components] as they are currently. . . . Donors would resist money going away from the cash transfer as it is essential for the other components, at least as a start-up. (Donor representative) At this point, in August 2018, a possible compromise between the government and development partners seemed unlikely. However, in the months that followed, the government slowly came around and seemed to accept ⁹ Ibid. ¹⁰ As we have been informed through conversations with stakeholders. 258 Marianne S. Ulriksen, Flora Myamba, and Constantine George The Guardian. 2017a. ‘JPM tells Burundian refugees to “go home”’. 21 July. The Guardian. 2017b. ‘TASAF in line for overhaul as JPM eyes another “boil”’. 13 January. Ulriksen, M. S. 2019. ‘Pushing for Policy Innovation: The Framing of Social Protection Policies in Tanzania’. In The Politics of Social Protection in Eastern and Southern Africa, edited by S. Hickey, T. Lavers, M. Niño-Zarazúa, and J. Seekings, pp. 122–147. Oxford: Oxford University Press. https://doi.org/10.1093/oso/ 9780198850342.003.0005. URT. 2016. National Five-Year Development Plan 2016/17–2020/21: Nurturing Industrialization for Economic Transformation and Human Development. Dar es Salaam: Ministry of Finance and Planning. URT. 2018. Tanzania Social Action Fund, Productive Social Safety Net, Phase Two (PSSN II): Program Design Document. Prepared by Government Project Preparation Team. Dar es Salaam: United Republic of Tanzania, President’s Office. URT, World Bank, UNDP, UNICEF, ILO, DFID, and Sida. 2016. Final Aide Memoire of the Joint Review and Implementation Support Mission Conducted in October 2016 on the Tanzania Productive Social Safety Net Project (PSSN). Dar es Salaam: Government of Tanzania. von Gliszczynski, M., and L. Leisering. 2016. ‘Constructing New Global Models of Social Security: How International Organizations Defined the Field of Social Cash Transfers in the 2000s’. Journal of Social Policy 45(2): pp. 325–343. Whitfield, L., ed. 2009. The Politics of Aid: African Strategies for Dealing with Donors. Oxford and New York: Oxford University Press. Whitfield, L., and A. Fraser. 2010. ‘Negotiating Aid: The Structural Conditions Shaping the Negotiating Strategies of African Governments’. International Negotiation 15(3): pp. 341–366. https://doi.org/10.1163/157180610X529582. 12 Brokered fiscal contracts Shifting bargaining positions of Senegalʼs informal commercial sector Ane Karoline Bak 12.1 Introduction In this chapter, I trace three decades of revenue bargaining in Senegal’s informal commercial sector. Drawing on the literature on the integral role of brokers in Senegal (e.g. Boone, 1992;Blundo, 2006;Beck, 2008;Gottlieb, 2017,2022), I focus on brokers’ influence on revenue-bargaining processes and specifically their outcomes. I find that what essentially become brokered fiscal contracts¹ can be explained by shifts in the relative bargaining positions of the ruling elites vis-à-vis the informal commercial sector and their brokers, in particular UnionNationaledesCommercants etIndustriels du Sénégal (UNACOIS).² During the 1990s, recurring attempts at taxing the informal sector were resisted by actors in the informal commercial sector, organized by UNACOIS, and aided by a fragmented ruling coalition and weak ruling elites. In the 2000s, support of the informal commercial sector was politically importantforPresidentAbdoulayeWade’s effortto(re)consolidatetheruling coalition. While an informal sector tax was installed, it largely remained a tax on paper and not in practice. However, an organizational fragmentation of the informal commercial sector in the 2000s weakened its bargaining position vis-à-vis the ruling elites in the 2010s, and therefore, in the past decade, informal commercial actors’ fiscal burden has increased with several new tax instruments and tentative expansions in tax collection. ¹ That is, fiscal contracts where an intermediary is sine qua non in the negotiation and upholding of the contract terms. ² UNACOIS changed its name in 2007 to UNACOIS-JAPPO. I refer to the organization as UNACOIS in the beginning of the chapter and to UNACOIS-JAPPO when discussing the later period. Ane Karoline Bak, Brokered fiscal contracts. In: The Politics of Revenue Bargaining in Africa. Edited by: Anne Mette Kjær, Marianne S. Ulriksen, and Ane Karoline Bak, Oxford University Press. © Oxford University Press (2024). DOI: 10.1093/oso/9780192868787.003.0012 260 Ane Karoline Bak I show that to understand the relative bargaining positions of the ruling elite and revenue providers (as theorized in the theoretical framework in Chapter 2), we should study their changing bargaining powers in parallel. In Senegal, fluctuations in the ruling coalition shape the resources and bargaining power of the ruling elites. On the other side, the bargaining power of UNACOIS as a broker is conditioned on its ability to organize and represent both small and large traders in the informal commercial sector, but its political influence is also conditioned by ruling coalition fluctuations creating political openings. The chapter demonstrates the instrumental value of a long-term historical approach and the importance of recognizing the distinct role of brokers when studying revenue bargaining in clientelist political settlements, in Senegal and beyond.³ The analysis is based on data collected from several field research stays between 2017 and 2022, including a survey of informal street vendors, tax-collection data, newspaper articles, interviews with former and current employees in the tax administration and representatives from the formal and informal economy, and reports and analyses of Senegal’s informal commercial sector. 12.2 Senegalʼs informal commercial sector After the agricultural sector, the commercial sector is perhaps Senegal’s most important economic sector. Including both formal and informal businesses, it accounts for 40.3% of the economic turnover and employs 32.8% of all workers (ANSD, 2017, 33, 42). Focusing on the informal (non-agricultural) sector, trade and commerce provide primary employment to more than one million Senegalese across urban and rural areas (ANSD, 2019, 121). The informal commercial sector provides 66% of the informal economy’s turnover (ANSD, 2017, 24, 34), but the tax contributions of the informal commercial economy arelimited. Between 2004 and 2007, the informal commercial sector contributed on average 16.4% of total direct business income taxes (Benjamin and Mbaye, 2012, 60). More recently, a report stated that Senegal’s entire informal economy contributes around 3% of tax revenues (DPEE, 2018). Informal businesses are defined by the absence of formal account-keeping due to lack of either skill or will. While economic actors are not registered under the dominant central government tax schemes such as company ³ Brokers here is arguably comparable to the third parties that Beach (Chapter 10) describes to be involved in revenue bargaining in Togo. Brokered fiscal contracts 261 incometax,value-addedtaxation(VAT),orwithholdingtaxesforemployees, even the smallest vendors pay a range of taxes, fees, and levies (Bak, 2019;ILO, 2020) as elsewhere in Africa (Moore, Prichard, and Fjeldstad, 2018, ch. 7). Senegal’s informal commercial sector is incredibly diverse, comprising street-level vendors and trade business owners (in French, respectively marchands ambulants and commercants), all with some sort of fixed structure (Mbaye et al., 2014). Some have a table on the side of the road, some have a proper shop (in a structured market or in a building), and some large informal traders import and sell to smaller informal retailers. The former mostly function on a day-to-day basis, the latter group is registered and each has a business identification number (Numéro d’Identification Nationale des Entreprises et Associations, henceforth NINEA) which gives access to administrative procedures, banking, and, not least, import and export. There are vast overlaps between Senegal’s formal and informal economies. Most formal businesses engage in some informal activities (Marfaing and Sow, 1999;Greven, 2017;Mbaye, Benjamin, and Gueye, 2017;Ekomié et al., 2020), which is made possible by poor coordination and limited cross-referencing between the NINEA database, Senegal’s tax authorities (Directorate Générale des Impôts et Domaines, henceforth DGID), and Customs (ILO, 2020, 30). For a long time, Customs did not differentiate between formal and informal business (Benjamin and Mbaye, 2012, 59), making it possible to be a registered business without paying taxes, and to import goods, pay customs, and sell the goods in large retail networks while remaining effectively informal, that is, not registering as taxpayers or declaring profits to the DGID. This is a key reason why large, rich traders can reside outside the tax net. It is thus difficult, and not very useful, to apply a narrow definition to the informalsector,nortodrawahardconceptualdistinctionbetweentheformal and the informal economies. Instead, it is imperative to unpack and understand the sector’s dynamics empirically and, especially, the role and influence of brokers in revenue bargaining. 12.2.1 Informal sector brokers and the implications for fiscal contracts in Senegal Senegal’s informal commercial sector includes small and large revenue providers with diverse trades and often diverging interests. Therefore, unionization and membership of business associations are vital sources of 262 Ane Karoline Bak organizational and fiscal bargaining power. When unions and business associations, or effectively their leaders, sit down at the imaginary negotiation table, they bring the organizational bargaining power given by the potential mobilization of the revenue providers they represent, as well as political and social capital that many of their members often do not enjoy. The organizations and their leaders negotiate or broker on behalf of their members regarding issues such as tax. As shown in this chapter, one business association, UNACOIS, has been central in revenue bargaining on behalf of Senegal’s informal commercial sector (see also Thioub, Diop, and Boone, 1998). Conceptualizing UNACOIS’s role as broker helps us understand its varying influence on revenue-bargaining outcomes over the years, which, I argue, is a central explanatory factor of the evolution in informal sector taxation. Brokers are defined as third-party intermediaries who utilize their power, oftengivenbyalargenetworkandgoodconnections,tofacilitateanexchange between two other parties. Some scholars see brokers as central actors in vote buying who provide exclusive access to goods and services in countries wherecitizen–politicianlinkagesaremoreclientelistthan programmatic(see e.g. Kitschelt and Wilkinson, 2007;Stokes et al., 2013;Gottlieb, 2017;Berenschot, 2019). Others see brokers as significant actors positioned at the fault lines between state and citizens, aiding in needs translation and policy implementation in weak states (Mosse and Lewis, 2006;Blundo and Le Meur, 2008;Bierschenk and Olivier de Sardan, 2014). I focus on so-called organizational brokers (Holland and Palmer-Rubin, 2015), which are comparable to interest groups but differ in their control over what is exchanged between members and state actors (Scott, 1972, 95; Blundo and Le Meur, 2008). Brokers balance interests in what is essentially a three-way relationship and negotiation. First, they have to consider state actors’ interests; responding to and accommodating them is imperative for brokers to maintain their privileged access and potential influence. Second, organizational brokers in particular have to care for their members’ social, collective, and organizational interests (Holland and Palmer-Rubin, 2015). Brokers may represent collective interests for altruistic reasons or to keep members satisfied, and if they fail to do this, members could sanction them, for example by stripping them of their power or by leaving the organization. Third, brokers have interests and can exploit their exclusive position to gain rents, patronage, or other goods (Stokes et al., 2013). For business associations, leaders are often successful businesspeople and are likely to bring their personal business interests to the negotiation table, which could conflict with those of their members. Brokered fiscal contracts 263 The act of balancing these interests implies that brokered fiscal contracts, which sometimes result from revenue bargaining, might serve diverse sets of, sometimes conflicting, interests. Brokers may exploit their exclusive position to profit more than their members, who may receive limited information. However, there is a strong element of interdependence between brokers and the revenue providers they represent. While brokers can provide large groups of revenue providers with organizational power and political voice, the strength of brokers’ bargaining position vis-à-vis ruling elites is largely dependent on the extent of their network and the ability to mobilize it. If brokers fail in balancing interests and leave their members dissatisfied, their organizational bargaining power and their bargaining position can be undermined. The followingsectionsdescribe UNACOIS’sbalancingact in revenue bargaining regarding taxation of Senegal’s informal commercial sector, how it changed over time, the consequences for UNACOIS and the broader group of revenue providers, and the outcomes of revenue bargaining. 12.3 Three decades of revenue bargaining The following analysis includes case studies of the attempts in the 1990s to generalizeVAT totheinformalsector;theadoptionoftheinformalsectortax, contributionglobaleunique(CGU)in2004anditsreformsin 2006, 2012,and 2019; and the adoption of advance payments on imported goods in 2012 and 2019 targeting large informal importers. These cases represent initiatives that were implemented with the key objective to increase taxation of the informal commercial sector. 12.4 The 1990s: Attempts at generalizing taxation Economicchangesinthe1970s and1980scausedaninformalizationofSenegal’s economy.⁴ First, there was a strong move towards a Senegalization of commerce to compete with long-established, foreign-owned firms, while the formerly fairly well-functioning productive formal sectors suffered under poor and increasingly centralized political management (Thioub, Diop, and Boone, 1998, 70–71). The mid ’80s were characterized by economic recession, which only worsened as the international financial institutions (IFIs) ⁴ I rely here on Thioub et al. (1998), who traced changes in Senegal’s political economy in the wake of economic liberalization, including how it manifested in instances of revenue bargaining (though not conceptualizing as such). 264 Ane Karoline Bak pushed through structural adjustment programs (SAPs). As economic liberalization provided new rent opportunities, and capital in the informal sector grew competitive, a new group of self-made capitalists emerged, and the business class broadened (Thioub, Diop, and Boone, 1998). Combined, the informalizationandliberalizationoftheeconomynarrowedthetax baseconsiderably. The IFIs pressured the Senegalese government led by President Abdou Diouf to expand taxation of the informal economy, which became a key reason for the government’s continuous attempts at generalizing VAT to the informal sectors. The government’s first attempt was in late 1989, when VAT and custom duty rates together increased by 5–6%. In reaction, traders in Dakar closed their shops and stopped all business activity in the capital (Boone, 1994; Thioub, Diop, and Boone, 1998). In the wake of this successful action, UNACOIS formed with the purpose of representing this new, largely informal business class to protect informal economy interests, and UNACOIS came to lead the organized opposition against the government’s following attempts to increase taxation of the informal economy. The second attempt, in 1991, to apply VAT to the informal commercial sector met similar opposition. In 1993, UNACOIS reacted to negotiations regarding VAT and distribution monopolies by threatening, among other things, to stop import and export. In response to the final attempt in 1995–1996, UNACOIS called for the well-known operation villes mortes (operation dead cities). The shut-down of commerce in all major cities failed, but the operation demonstrated the potentialorganizationalpowerofUNACOIS, its capacityofcollectiveaction, and its geographical reach. Ultimately, the traders succeeded in countering the regime’s attempts at taxing the informal economy. 12.4.1 A fragmented ruling coalition and the creation of a strong broker The instances of revenue bargaining in the 1990s did not end in a brokered fiscal contract: they did not establish a positive exchange of informal sector taxation for public services, nor did they seem to end in a devil’s deal where non-taxation was exchanged for political support (Tendler, 2002; see also Chapter 2). If anything, non-taxation was exchanged for maintenance of commerce, economic activity, and social order. As elaborated in what follows, two conditions arguably explain this outcome: a poor and increasingly fragmented ruling coalition, and the making of UNACOIS as broker for an organized informal commercial sector. Brokered fiscal contracts 265 The Parti Socialiste de Sénégal (PS) had been in power since independencein1960basedonarulingcoalitionbuiltaroundcloserelationsbetween the political and economic elites. The ruling coalition was characterized by a domestic business class that was deeply dependent on rent opportunities givenbytherulingelitesandawide-reachingpoliticalmachinesecuringelection victories (Boone, 1992;Thioub, Diop, and Boone, 1998). This domestic businessclassremainslargelytiedtothe structuresofreligiousbrotherhoods, especially the Mourides, whose wealth and capitalist strength were built during colonial times (Cruise O’Brien, 1971;Diop and Diouf, 1990;Boone, 1992;Villalón, 1995;Golub and Hansen-Lewis, 2012;Koter, 2021). The SAPs and economic liberalization coupled with poor political management diminished the economic resources of the ruling elites. Though President Diouf had made strides to centralize and consolidate his power during the 1980s and 1990s (Diop and Diouf, 1990), the structural economic changes limited the resources available to distribute patronage and maintain the religious groups’ function in the political machine (Boone, 1990;Young and Kanté, 1991;Schaffer, 1998;Koter, 2021). Hence, the foundation of the ruling coalition withered, and the coalition became fragmented as elsewhere in Africa during these years (Whitfield et al., 2015). Among revenue providers in the informal commercial sector, the instances of revenue bargaining prompted mobilization and organization. Under the conditions of an increasingly fragmented ruling coalition, potential revenue providers in the informal commercial sector became an important support group. Moreover, UNACOIS was able to utilize its organizational power to position itself as a broker between the two and increase its political influence. This established a relatively strong bargaining position of the revenue providers vis-à-vis the ruling elites (see Table 2.1, Chapter 2). Therefore, when protests and strikes paralyzed Senegal’s economy, the Diouf regime was pressured to accommodate the potential revenue providers and abandon its attempts at expanding its tax effort. 12.5 The 2000s: Installation and reform of the informal sector tax In 2004, a tax reform was passed, which included the installation of the contribution globale unique (CGU), an annual tax on informal economy activities applied to traders with a turnover below 50 million XOF (approx. €76,000), and service providers with a turnover below 25 million XOF 266 Ane Karoline Bak (approx. €38,000).⁵ The CGU substituted and synthesized several taxes, including income tax on industrial and commercial profits, VAT, and the business tax, patente. The CGU was calculated based on estimated turnover with fixed amounts for payments according to 20 turnover brackets. The tax brackets varied between the trade and service sectors. For traders, the effective tax rates varied between 1.5 and 9% and increased progressively with the level of turnover (UNACOIS, 2012).⁶ The motivation for the unified tax scheme was to ease the act of paying taxes, create a first fiscal acquaintance between informal economic actors and tax authorities, and promote tax compliance.⁷ The CGU is an informal economy tax because, although declarative, it does not require provision of financial statements. Business owners need to register, make yearly declarations, and pay accordingly. It is primarily declarative in theory, though. For the most part, the tax is collected by a group of officials who tour the markets once a year (tax inspector 3, Dakar, 26 February 2018; municipal employee, Dakar, 1 March 2018). They assess turnovers based on vendors’ goods and collect the relevant amount of tax. Hence, the CGU is effectively as much a presumptive as a declarative regime. The CGU represents a first brokered fiscal contract between the informal commercial sector and the ruling elites, facilitated by UNACOIS, who, according to the narrative of the CGU, proposed it. This has been stated publicly by the DGID (APS, 2013) and in interviews with a UNACOIS representative (Dakar, 23 February 2018) and a DGID representative (Dakar, 20 February 2018). Although fighting generalization of VAT in the early ’90s, UNACOIS had proposed an instrument that promotes informal sector taxation, at least on paper. However, in the wake of the adoption of the CGU in 2004, UNACOIS criticized the tax brackets for being out of sync with the traders’ means (Sud Quotidien, 2004;Barry, 2004) and claimed that it had not been consulted in the formulation of the final tax brackets (ibid.). Nothing came of these protests in 2004, but the tax brackets became the focal point of subsequent revenue bargaining and several reforms of the CGU. In 2006, following consultation with UNACOIS (UNACOIS, 2012), the fixed amounts were lowered for the 15 highest brackets but increased for the fourthand fifth-lowest.⁸ For the highest bracket, the reduction was notable: from 4,200,000 XOF (approx. €6,407) to 1,500,000 XOF (approx. ⁵ Loi no. 2004-12 du 6 février 2004 modifiant certaines dispositions du Code Général des Impôts. ⁶ See Appendix, Table 12.A1 for overview of the CGU calculus and its reforms. ⁷ Expose des motifs, Loi no. 2004-12 du 6 février 2004. ⁸ See Appendix, Table 12.A1 for overview of CGU reforms. Brokered fiscal contracts 267 €2,290). Furthermore, the tax scale was split in two: one for traders dealing in food products and cement, and one with markedly lower taxation for all other traders. Further reforms of the CGU in 2013 and 2018 will be discussed later. 12.5.1 The balancing act During its first decade, the CGU was but a contract on paper, and very few businesses were enrolled. In 2004, 2,663 firms were effectively subject to the CGU, and in 2007, the number was 4,970 (Benjamin and Mbaye, 2012, 60f) compared to 281,000 informal economic entities in 2003 in Dakar alone (DPS, 2003) and an estimated 774,000 (non-agricultural) entities across Senegal (ANSD, 2013, 16). While the total revenue collected from the CGU doubled between 2004 and 2007, the average tax collection per firm fell, and revenue from the CGU did not exceed 1.5 billion XOF (approx. €2.3 million), which equals less than 1% of total direct taxes collected (Benjamin and Mbaye, 2012, 60). By 2012, revenue collections had fallen to 0.7 billion XOF (approx. €1.1 million) (Sy, 2016). Based on its limited reach into the informal sector, the fiscal outcome of the CGU was negligible in terms of revenue collection and experienced tax burden. In addition, street-level vendors will, when asked, most likely say that they never have heard of the CGU, and that they pay the patente (even though their payment receipts actually say CGU).⁹ The installation and poor implementation of the CGU can be explained, at least in part, by the interests of then President Abdoulaye Wade and UNACOIS’s balancing act. The fragmentation of the ruling coalition in the 1990s increased political competition leading up to Senegal’s presidential election in2000.Therewasrealuncertaintyaboutelectionresults(Beck, 2008,7),and in the country’s first transfer of power, the Parti Démocratique Sénégalais (PDS) and its leader, Abdoulaye Wade, took over from PS. In the wake of his election, President Wade sustained his presidential power by catering to religious leaders and (re)creating avenues for patronage financed in part by large-scale corruption, for example around public tenders (Beck, 2008, 223ff; Mbow, 2008;Kelly, 2012). He reconsolidated the ruling coalition and ensured centralized power and control of central state institutions (Dumont and Kanté, 2019). Moreover, he openly wooed people whose basic livelihood was secured by working in or relying on the informal economy (Dankoco ⁹ In a survey by Gottlieb (2022), only 6% of firms had heard about the CGU. Author’s interviews and focus groups with informal actors corroborated this (e.g. focus group with presidents from five associations representing informal commercial professions, Dakar, 30 March 2022). 274 Ane Karoline Bak For the CGU, the number of taxpayers registered and declarations increased between 2012 and 2022. In 2013, there were 2,035 CGU declarations (Gottlieb, 2022), which almost doubled between 2012 and 2019, and then rose by more than 50% in the two following years. Between 2012 and 2019, the revenue collected from CGU likewise rose from 0.72 (Sy, 2016) to 0.89 billion XOF (approx. €1.5 billion) (Table 12.1). As shown in Table 12.1, collections rose equally within the much shorter period between 2019 and 2021 to 1.15 billion XOF (approx. €1.75 million). The data in Table 12.1 should be complemented with data on revenue collected by teams that visit market areas once a year. In 2021, 29,372 taxpayers were reached, paying in total 1,531,197,361 XOF (approx. €2.3 million),²² which means that many more taxpayers are reached than those registered. In fact, these tours collect more revenue than is collected through declarations, even though an average taxpayer pays 52,131 XOF (approx. €80), which is lower than the average tax payments made by registered taxpayers.²³ Still, the sum of presumptive and declarative taxpayers comprises only a small share of the informal commercial sector. Despite an increase in taxpayers and tax revenue under the CGU, they remain low in absolute terms. The 2.68 trillion XOF (approx. €4.1 million) equals 0.1% of the projected tax collections for 2021 of 2.56 trillion XOF (approx. €3.9 billion), and in real terms, the current revenue does not exceed that of 2008 (Gottlieb, 2022). However, interviews with representatives from the informal and formal economies conducted in March 2022 suggest that both smaller and larger commercial businesses have experienced asignificantincrease in fiscal burden. Thiswas mentioned, also unprompted, by several interviewees (e.g. representative CNES, Dakar, 29 March 2022; representative from UNACOIS-JAPPO, 1 April 2022). For the Acompte BIC, it was only possible to obtain data for the most recent years, even though it has been in place since 2012. As shown in Table 12.2, Table 12.1 CGU, 2019–2021 Registered taxpayers Declarations Revenue collected (XOF) 2019 3,991 889,093,121 2020 5,889 1,024,570,010 2021 18,206 6,487 1,149,220,868 Source: DGID (Dakar, March 2022). ²² Called Paiements par Anticipation, which means advance payments but is in practice a lump-sum payment. Data provided by the DGID (Dakar, March 2022). ²³ This can be ascribed to the legal stipulation of a maximum lump-sum payment of 100,000 FCFA. Brokered fiscal contracts 275 Table 12.2 Acompte sur lʼimportation (XOF) 2018 7.7 billion 2019 8.4 billion 2020 9.0 billion Source: DGID (Dakar, March 2022). the revenue collected by this instrument increased between 2018 and 2020. As the PCF has only been in place since September 2021, it is still too early to obtain data for at the time of writing this chapter. One interviewee in the DGID called PCF ‘a step’ towards regulating the fiscal problem posed by the informal sector (tax inspector 4, Dakar, 22 March 2022); another said that the preliminary data suggest that the PCF is securing new revenue (tax inspector 2, Dakar, 18 March 2022), while a representative of UNACOIS argued that this was not the case (Dakar, 1 April 2022). 12.7 Conclusion As elsewhere, taxation of the informal economy has received much political attention in Senegal. From a revenue-bargaining perspective, informal economy taxation is quite peculiar: It applies to large parts of the population but generates relatively limited tax revenues; taxed informal actors often experience a high fiscal burden compared to their means, and most often, receive a limited return of public or semi-public goods compared to actors in the formal sector (Pimhidzai and Fox, 2012;Rogan, 2022). A critical strand in the tax and governance literature questions the global effort to expand informal economy taxation by highlighting equity issues (Prichard, 2018; Dube and Casale, 2019;Rogan, 2019), questioning the potential economic gains (Moore, Prichard, and Fjeldstad, 2018;Moore, 2022), and emphasizing the complexity of historical experiences, social institutions, and power relations that such taxation plays into (Tendler, 2002;Kjær, 2009;Meagher and Lindell, 2013;Prichard and van den Boogaard, 2017;Meagher, 2018). Thischaptercontributestothisstrandofresearchintwoways.First,itillustrateshowtounpackthedynamicsandpoliticsofinformaleconomytaxation. The informal economy comprises many sectors each with their economic structures, each their organizational and political dynamic, and each their kinds and levels of taxation.²⁴ As theorized in Chapter 2, economic actors ²⁴ See e.g. Chapter 3 in this volume. 276 Ane Karoline Bak from different sectors might be of more or less fiscal and political importance to the ruling coalition. To deepen the analysis, this chapter focuses on one of Senegal’smosteconomicallyimportantsectors:theinformalcommercialsector. Tracing three decades of revenue bargaining with this sector, the chapter demonstrates the explanatory power of this volume’s theoretical framework: from strong and successful resistance against taxation to a brokered fiscal contract with little effective taxation, and, recently, to an expansion in the tax portfolio and an increased fiscal burden felt by the sector. The evolution is linked to the changing bargaining power and, in particular, shifts in the relative bargaining positions of the ruling elites vis-à-vis the revenue providers and, in particular, their broker, UNACOIS. To my knowledge, this chapter is the first long-term study of politics of taxation in Senegal, both in relation to the informal economy and broadly. The second major contribution of this chapter is that it highlights the consequences of the brokered nature of fiscal contracts between informal economic actors and ruling elites. In clientelist settings, as with all five case countries in this volume, it is imperative to consider the influence of brokers in revenue bargaining, especially when researching taxation of the informal economy that includes a large group of small, often politically marginalized, economicactors.Here,brokersmustbalancethediverseandoftenconflicting sets of interests. Furthermore, as an inherent feature of clientelism, competition for access to rents and side deals, also internally in organizations, creates tensions and frequent organizational fragmentation, such as was the case for UNACOIS.²⁵ These conditions became determinant for their bargaining power, which, in turn, had consequences for the brokered fiscal contract between the ruling elites and Senegal’s informal commercial sector. ²⁵ Thanks for Catherine Boone for pointing this out. 12.8 Appendix Table 12.A1 Overview of CGU reforms 2004–2019 2004 2006 2012 2019 Sub-regimes Two sub-regimes: 1. service providers 2. traders Three sub-regimes: 1. service providers 2. traders of food products and cement 3. traders of other products Three sub-regimes: 1. service providers 2. traders of food products and cement 3. traders of other products Two sub-regimes: 1. service providers 2. traders Threshold Service providers: XOF 25,000,000 Traders: XOF 50,000,000 Service providers: XOF 25,000,000 Traders: XOF 50,000,000 All sub-regimes: XOF 50,000,000 Both sub-regimes: XOF 50,000,000 Number of turnover brackets Both sub-regimes: 20 All sub-regimes: 20 Service providers: 5 Both categories of traders: 3 Both sub-regimes: 1 Tax levels Service providers: XOF 10,000–3,000,000 Traders: XOF 5,000–4,200,000 Service providers: XOF 10,000–2,000,000 Traders of food products and cement: XOF 5,000–1,250,000 Traders of other products: XOF 5,000–1,500,000 Service providers: 4–8% Traders of food products and cement: 1–2.8% Traders of other products: 2–3.8% Service providers: 5% Traders: 2% 278 Ane Karoline Bak Acknowledgements I am grateful for comments provided by Anne Mette Kjær, Marianne S. Ulriksen, Catherine Boone, Jessica Gottlieb, Laurence Marfaing, the Taxation in Africa panel atASA 2020,and colleaguesfromAarhusUniversity,theUniversityofSouthernDenmark, and the Political Settlements and Revenue Bargains in Africa project. Amath Mbaye and Mohamed Seydina Mbaye provided invaluable research assistance in Senegal. Regretfully, this project did not, as initially planned, turn into a joint project with the late Professor Fatou Gueye from Université Cheikh Anta Diop de Dakar (UCAD). May she rest in peace. References ANSD. 2013. Enquête Nationale Sur Le Secteur Informel Au Sénégal (Ensis 2011). Dakar. ANSD.2017. AgenceNationaledela Statistiqueet delaDémographieRapportGlobal Du Recensement Général Des Entreprises. Dakar. ANSD. 2019. 181–182 Agence Nationale de la Statistique et de la Démographie Enquête Régionale Intégrée Sur l’Emploi et Le Secteur Informel (ERI-ESI) Sénégal, 2017 RAPPORT FINAL. APANEWS. 2020. ‘Sénégal: La Direction Des Impôts Dément La Hausse de l’impôt Sur Les Sociétés’. APS. 2013. ‘L’acompte BIC sans Effet Sur Les Prix à La Consommation (DGID)’. Agence de Presse Sénégalaise. Bak, Ane Karoline. 2019. When the Fiscal Social Contract Is Not about Tax: Understanding the Limited Role of Taxation in Social Accountability in Senegal. Aarhus: Politica. Balde, Osseynou. 2020. ‘La DGID Vise Les Importateurs Qui Ne Payent Par La Fiscalité’. L’AS in Seneplus.https://www.seneplus.com/economie/la-dgid-viseles-importateurs-qui-ne-paient-pas-la-fiscalite#:~:text=LaDirectionGénéraledes Impôts,de toutes leurs obligations fiscales. Barry, Moustapha. 2004. ‘Lutte Contre La Fraude: L’Unacois Dénonce Les Faveurs Faites à La CSS de Mimran’. Wal Fadjri, 12 May. Beck,Linda J.2008. Brokering Democracy inAfrica: The Riseof ClientilistDemocracy in Senegal. New York: Palgrave Macmillan. Benjamin, Nancy, and Ahmadou Aly Mbaye. 2012. The Informal Sector in Francophone Africa: Firm Size, Productivity and Institutions.http://elibrary.worldbank. org/doi/book/10.1596/978-0-8213-9537-0. Brokered fiscal contracts 279 Berenschot, Ward. 2019. ‘Informal Democratization: Brokers, Access to Public Services and Democratic Accountability in Indonesia and India’. Democratization 26(2): pp. 208–224. Bierschenk, Thomas, and Jean-Pierre Olivier de Sardan (eds.). 2014. States at Work: Dynamics of African Bureaucracies. Leiden: Brill. Blundo, Giorgio. 2006. ‘Dealing with the Local State: The Informal Privatization of Street-level Bureaucracies in Senegal’. Development and Change 37(4): pp. 799– 819. Blundo, Giorgio, and Pierre-Yves Le Meur. 2008. ‘Introduction: An Anthropology of Everyday Governance: Collective Service Delivery and Subject-Making’. In The Governance of Daily Life in Africa: Ethnographic Explorations of Public and Collective Services, edited by Giorgio Blundo and Pierre-Yves Le Meur. Leiden: Brill, pp. 1–9. Boone, Catherine. 1990. ‘The Making of the Rentier Class: Wealth Accumulation and Political Control in Senegal’. Journal of Development Studies 26(3): 425–449. Boone, Catherine. 1992. Merchant Capital and the Roots of State Power in Senegal, 1930–1985. Cambridge: Cambridge University Press. Boone, Catherine. 1994. ‘Trade, Taxes, and Tribute: Market Liberalizations and the New Importers in West Africa’. World Development 22(3): pp. 453–467. Brown, Alison, Michal Lyons, and Ibrahima Dankoco. 2010. ‘Street Traders and the Emerging Spaces for Urban Voice and Citizenship in African Cities’. Urban Studies 47(3): pp. 666–683. Cruise O’Brien, Donald B. 1971. The Mourides of Senegal: The Political and Economic Organisation on an Islam Brotherwood. Oxford: Clarendon Press. Dankoco, Ibrahima, and Alison Brown. 2017. ‘Commerce of the Street in Sénégal: Between Illegality andTolerance’. InRebelStreetsand theInformal Economy:Street Trade and the Law, edited by Alison Brown, pp. 165–182. London: Routledge. Diaw, Abdou. 2019. ‘Strategie de Recettes á Moyen Terme: Focus Sur Le Nouveau Dispositif de Mobilisation Des Ressources’. Le Soleil. Diop, Momar Coumba, and Mamado Diouf. 1990. Le Sénégal Sous Abdou Diouf. Paris: Karthala. DPEE. 2018. Impacts d’une Amélioration de La Productivité Du Secteur Informel Surl’économie Sénégalaise. DPS. 2003. Le Secteur Informel Dans l’Agglomeration de Dakar: Performances, Insertion et Perspectives. Dakar. Dube, Godwin, and Daniela Casale. 2019. ‘Informal Sector Taxes and Equity: Evidence from Presumptive Taxation in Zimbabwe’. Development Policy Review 37(1): pp. 47–66. 280 Ane Karoline Bak Dumont, Gérard-Francois, and Seydou Kanté. 2019. La Géopolitique Du Sénégal De Senghor à l’élection de Macky Sall. Paris: L’Harmattan. Ekomié, Jean-Jacques et al. 2020. ‘The Informal Sector in Francophone Africa: Dominant Characteristics, Scope, and Trends from Our Survey Data’. In Formal and Informal Entreprises in Francophone Africa: Moving Toward a Virabnt Private Sector, edited by Ahmadou Aly Mbaye, Stephen S Golub, and Fatou Gueye, pp. 79–102. Ottawa: International Development Research Centre. Gellar, Sheldon. 2013. ‘The Rise of Citizen Movements and the Consolidation of Democracy under the Abdoulaye Wade Regime (2000–2012)’. In Le Sénégal Sous Abdoulaye Wade: Le Sopi à l’épreuve Du Pouvoir, edited by Momar-coumba Diop. Dakar and Paris: CRES–Karthala, pp. 119–151. Golub, Stephen S., and J. Hansen-Lewis. 2012. ‘Informal Trading Networks in West Africa: The Mourides of Senegal/The Gambia and the Yoruba of Benin/Nigeria’. In The Informal Sector in Francophone Africa: Firm Size, Productivity, and Institutions, edited by Nancy Benjamin and Ahmadou Aly Mbaye, pp. 173–194. Washington DC: World Bank and Agence Française de Développement. Gottlieb, Jessica. 2017. ‘Explaining Variation in Broker Strategies: A Lab-in-the-Field Experiment in Senegal’. Comparative Political Studies 50(11): pp. 1556–1592. Gottlieb, Jessica. 2022. ‘How Economic Informality Constrains Demand for Programmatic Policy’. American Journal of Political Science.https://onlinelibrary. wiley.com/doi/10.1111/ajps.12746. Greven, Thomas. 2017. ‘OrganizingtheInformalEconomy inSenegal’. InWork,Institutions and Sustainable Livelihood, edited by Virginius Xava, Debdulal Saha, and Rajdeep Singha. Singapore: Palgrave Macmillan, 303–322. Holland, Alisha C., and Brian Palmer-Rubin. 2015. ‘Beyond the Machine: Clientelist BrokersandInterestOrganizationsinLatinAmerica’. ComparativePoliticalStudies 48(9): pp. 1186–1223. ILO. 2020. Diagnostic de l’Économie Informelle Au Sénégal. Genève: Bureau international du Travail (BTI)–ILO. https://www.ilo.org/wcmsp5/groups/public/---ed_ protect/---protrav/---travail/documents/publication/wcms_735752.pdf. Kelly, Catherine Lena. 2012. ‘Senegal: What Will Turnover Bring?’ Journal of Democracy 23(3): pp. 121–131. Kelly, Catherine Lena. 2020. Party Proliferation and Political Contestation in Africa: Senegal in Comparative Perspective. Cham: Palgrave. Kitschelt, Herbert, and Steven I. Wilkinson. 2007. Patrons, Clients, and Policies. Edited by Herbert Kitschelt and Steven I. Wilkinson. Cambridge: Cambridge University Press. Kjær, Anne Mette. 2009. ‘Sources of Local Government Extractive Capacity: The Role of Trust and Pre-Colonial Legacy in the Case of Uganda’. Public Administration and Development 29: pp. 228–238. Brokered fiscal contracts 281 Kohnert, Dirk, and Laurence Marfaing. 2019. ‘Senegal: Presidential Elections 2019: The Shining Example of Democratic Transition Immersed in Muddy PowerPolitics’. SSRN Electronic Journal (92739): pp. 1–7. Koter, Dominika. 2021. ‘Religion and Electoral Competition in Senegal’. In The Oxford Handbook of Politics in Muslim Societies, edited by M. Cammett and P. Jones. Oxford Academic, 1–21. Malou, Jean Pierre. 2013. ‘Contentieux DGID-Unacois Sur La Taxe de 3 %— L’acompte n’a Pas d’incidence Sur Les Prix’. Sud Quotidien. Marfaing, Laurence, and Mariam Sow. 1999. Les opérateurs économiques au Sénégal: entre le formel et l’informel. Paris: Karthala. Marfaing, Laurence, and Alena Thiel. 2013. ‘The Impact of Chinese Business on Market Entry in Ghana and Senegal’. Africa 83(4): pp. 646–669. Mbaye, Ahmadou Aly, Nancy C. Benjamin, Stephen S. Golub, and Jean-Jacques T. Ekomie. 2014. The Urban Informal Sector in Francophone Africa: Large versus Small Enterprises in Benin, Burkina Faso and Senegal. Development Policy Research Unit Working Paper 201405. DPRU, University of Cape Town. Mbaye, Ahmadou Aly, Nancy Claire Benjamin, and Fatou Gueye. 2017. ‘The Interplay Between Formal and Informal Firms and Its Implications on Jobs in Francophone Africa: Case Studies of Senegal and Benin’. In The Informal Economy in Global Perspective: Varieties of Governance, edited by Abel Polese, Colin C. Williams, Ioana A. Horodnic, and Predrag Bejakovic. Cham: Palgrave Macmillan, 213–231. Mbaye, Ahmadou Aly, Stephen S. Golub, and Philip English. 2015. ‘Policies, Prices and Poverty: The Sugar, Vegetable Oil and Flour Industries in Senegal’. In Race to the Next Income Frontier: How Senegal and Other Low-Income Countries Can Reach the Finish Line, Policy Research Working Papers, edited by Ali Mansoor, Salifou Issoufou, and Daouda Sembene. Washington, DC: IMF, 50. https:// documents.worldbank.org/en/publication/documents-reports/documentdetail/ 474811468183562115/policies-prices-and-poverty-the-sugar-vegetable-oil-andflour-industries-in-senegal. Mbaye, Ahmadou Aly, Stephen S. Golub, and Fatou Gueye (eds). 2020. Formal and Informal Enterprises in Francophone Africa: Moving Toward a Vibrant Private Sector. Ottawa: International Development Research Centre. Mbow, Penda. 2008. ‘Senegal: The Return of Personalism’. Journal of Democracy 19(1): pp. 156–169. Meagher, Kate. 2018. ‘Taxing Times: Taxation, Divided Societies and the Informal Economy in Northern Nigeria’. Journal of Development Studies 54(1): pp. 1–17. https://doi.org/10.1080/00220388.2016.1262026. 282 Ane Karoline Bak Meagher, Kate,andIldaLindell. 2013. ‘Introduction:Engagingwith AfricanInformal Economies: Social Inclusion or Adverse Incorporation?’ African Studies Review 56(3): pp. 57–76. Moore, Mick. 2022. ‘Tax Obsessions: Taxpayer Registration and the Informal Sector in Sub-Saharan Africa’. Development Policy Review. Moore, Mick, and Odd-Helge Fjeldstad. 2008. ‘Tax Reform and State-Building in a Globalised World’. In Taxation and State-Building in Developing Countries, editedbyDeborahBräutigam,MickMoore, andOdd HelgeFjeldstad. Cambridge: Cambridge University Press. Moore, Mick, Wilson Prichard, and Odd-Helge Fjeldstad. 2018. Taxing Africa: Coercion, Reform and Development. London: Zed Books. Mosse, David, and David Lewis. 2006. ‘Theoretical Approaches to Brokerage and Translation in Development’. In Development Brokers and Translators, edited by David Mosse and David Lewis, pp. 1–26. Bloomfield: Kumarian Press. Ndiaye, Ameth Saloum. 2017. Does the Implementation of Tax-Related Reforms and Institutions-Related Reforms Offer Scope for Sustained Tax Revenue Mobilization in Senegal? Addis Ababa. Eleventh Edition of the African Economic Conference on “Governance for Structural Transformation”, Addis Ababa, Ethiopia, 4–6 December 2017. https://aec.afdb.org/sites/default/files/2019/10/08/does_the_ implementation_of_tax-related_reforms_and_institutions-related_reforms_ offer_scope_for_sustained_tax_revenue_mobilization_in_senegal_.pdf Pimhidzai, Obert, and Louise Fox. 2012. ‘Taking from the Poor or Local Economic Development: The Dilemma of Taxation ofSmall Informal Enterprises in Uganda’. CSAE Working Paper. Washington, DC: World Bank. Prichard, Wilson. 2018. ‘What Might An Agenda for Equitable Taxation Look Like?’ Policy Brief. Brighton: ICTD. Prichard, Wilson, and Vanessa van den Boogaard. 2017. ‘Norms, Power and the Socially Embedded Realities of Market Taxation in Northern Ghana’. African Studies Review 60(1): pp. 171–194. Republic of Senegal. 2019. Projet de Loi de Finances de l’année 2020. Dakar. Riedl, Rachel Beatty, and Ndongo Samba Sylla. 2019. ‘Aspirations and Realities in Africa: Senegal’s Vigorous but Constrained Election’. Journal of Democracy 30(3): pp. 94–108. Rogan, Michael. 2019. Tax Justice and the Informal Economy: A Review of the Debates. WIEGO Working Paper No. 41. Manchester. Rogan, Michael. 2022. ‘Taxation and the Informal Sector in the Global South: Strengthening the Social Contract without Reciprocity?’ In Social Contracts and Informal Workers in the Global South, edited by Laura Alfers, Martha Chen, and Sophie Plagerson. Noarthampton: Edward Elgar, pp. 85–105. Brokered fiscal contracts 283 Schaffer, Frederic Charles. 1998. Democracy in Translation: Understanding Politics in an Unfamiliar Culture. Ithaca: Cornell University Press. Scheld, Suzanne. 2010. ‘The “China Challenge”: The Global Dimensions of Activism and the Informal Economy in Dakar’. In Africa’s Informal Workers: Collective Agency, Alliances and Transnational Organizing in Urban Africa, edited by Ilda Lindell. London: Zed Books, pp. 153–168. Scott, James. 1972. ‘Patron–Client Politics and Political Change in Southeast Asia’. American Political Science Review 66(1): pp. 24. Stokes, Susan C., Thad Dunning, Mercelo Nazareno, and Valeria Brusco. 2013. Brokers, Voters, and Clientilism. New York: Cambridge University Press. Sud Quotidien. 2004. ‘Nouvelle Contribution Globale Unique Ou Impôt Synthétique: L’Unacoisdefs Rejette La Démarche Imposée Par l’État’. Sy, Mamadou. 2016. ‘Fiscalité—Le Renforcement Des Centres de Gestion Agréés Préconisé’. Le Soleil. Tendler, Judith. 2002. ‘Small Firms, the Informal Sector, and the Devil’s Deal’. IDS Bulletin 33(3). Thioub, Ibrahima, Momar-Coumba Diop, and Catherine Boone. 1998. ‘Economic Liberalization in Senegal: Shifting Politics of Indigenous Business’. African Studies Review 41(2): pp. 63–89. UNACOIS. 2012. Livre Blanc de l’UNACOIS: Moderniser Nos PME et Revitaliser Nos Regions. Dakar. Villalón, Leonardo Alfonso. 1995. Islamic Society and State Power in Senegal: Disciples and Citizens in Fatick. Cambridge: Cambridge University Press. Whitfield, Lindsay, Ole Therkildsen, Lars Buur, and Anne Mette Kjær. 2015. The Politics of African Indutrial Policy: A Comparative Perspective. New York: Cambridge University Press. Young, Crawford, and Babacar Kanté. 1991. ‘Governance, Democracy, and the 1988 Senegalese Elections’. In Governance and Politics in Africa, edited by Goran Hyden and Michael Bratton. Boulder, CO: Lynne Rienner Publishers, pp. 57–74. 290 Ane Karoline Bak, Anne Mette Kjær, and Marianne S. Ulriksen 13.1.2 Findings on processes Once revenue bargaining is set in motion, how do cases of revenue bargaining evolve? We make the following four findings that highlight how bargaining positions can change, how bargaining processes evolve, and the outcomes at which they eventually arrive: (1) organizational importance matters more than potential revenue contributions; (2) revenue bargaining can build cross-sector alliances and bargaining power; (3) elections shift the relative bargaining positions of revenue providers and ruling elites; and (4) arenas of bargaining matter for strategies and fiscal contract outcomes. 13.1.2.1 Organizational importance matters more than potential revenue contributions In this book, we examine in tandem the influence of revenue providers’ organizational and fiscal importance for their bargaining positions vis-à-vis the ruling elites. Organizational importance is tightly associated with the concept of holding power in the political settlement theory, and research has demonstrated how ruling elites’ need for political support explains variations in policy outcomes across countries and within countries across sectors. The power thatactorscangainbasedontheirfiscalimportancetotherulingelites has received less attention in this literature. We argue that it is necessary to add a revenue dimension to the concept of holding power and emphasize fiscal importance as an important source of bargaining power for revenue providers. This allows us to examine the importance that ruling elites ascribe to fiscal concerns in comparison with the need for political support. What we find is that a revenue provider’s potential fiscal contribution to the revenue base, even when potentially large, matters less than the revenue provider’s political importance. Across the cases of revenue bargaining, ruling elites prove to be keenly aware of the role of groups of revenue providers as supporting factions in the ruling coalition. For example, the ruling parties in Tanzania and Uganda very clearly prioritized the party’s own political campaigns over government revenue when they allowed big businesses tax exemptions (Chapter 9). In Tanzania, the government also avoided protests from the security forces, an important faction, by pre-emptively introducing a compensatory payout at the time when the VAT exemptions of army shops were removed (Chapter 5). While the removal of the duty-free shops’ exemptions hardly improved revenue mobilization, the government was able to change an unpopular and misused benefit without upsetting the politically important security forces. In Togo, the government compromised on its revenue potential by involving the local Chief in improving tax collection in his area (Chapter 10). That the Chief assisted tax collectors from the far-away [Document text truncated for crawler view.]