Conceptualizing the fiscal state: implications for sub-Saharan Africa

Abstract

This paper contributes to the debate on domestic revenue mobilization and state-building in the Global South by exploring the concept of fiscal states and the common assumption that such states are present in sub-Saharan Africa. We systematically review the diverse understandings of the fiscal state across relevant literatures to revisit its conceptualization. On that basis, we define the fiscal state as a state whose public revenue base is dominated by tax revenue and loans, and where the relationship between taxation and external and domestic borrowing is balanced and thereby sustainable and characterized by interdependence. We distinguish the fiscal state conceptually from the tax, debt, and rentier states and present a typology of these ideal state types, discussing illustrative empirical examples of different states in sub-Saharan Africa. These illustrate that not all sub-Saharan states can be categorized as fiscal states. This is important because when African states are regarded as fiscal states, assumptions are made about their economic structures; yet, to the extent that these are absent, fiscal policy reforms are unlikely to carry long-term positive effects.

1. Introduction

The term ‘fiscal state’ used to be a prerogative of industrialized nation states with a specific narrative aimed at explaining why Europe spearheaded state formation. Within the past two decades, this has changed. Mentions of the fiscal state in scholarly work have increased across disciplines, and the term is now applied in works on state-building globally, including in the Global South. This turn in the literature has occurred in parallel with the emergence of an increased focus in the international development community on how low-income countries can tax more and better. Programmes to support taxation are typically formulated based on an implicit assumption that the prerequisites for building or advancing fiscal states exist in low-income countries. This paper sets out to clarify the defining features of a fiscal state as a first step towards interrogating these assumptions.

The body of literature on fiscal states within the field of economic history is substantial (e.g. Bonney, 1999; Dincecco, 2011; Johnson and Koyama, 2014; O'Brien, 2011; Yun-Casalilla, 2012), but despite frequent application of the term and continuous discussion of ‘the fiscal state in Africa’ (e.g. Albers et al., 2020; Moore, 2021; Waris, 2018), there is little focus on (re)assessing and (re)defining the concept as it has entered new realms. Few scholars have directly examined whether states in low-income countries have developed from tax states (provided that they even qualified as such) to fiscal states. It has not been explicitly discussed what such a development would look like in contemporary lower-income countries or what the defining attributes of a fiscal state are.

This is problematic in two ways. For researchers studying the emergence and consequences of fiscal states, proper conceptualization is key. While concepts that are properly delimited theoretically and empirically have intrinsic value, there is of course also instrumental value. To explain the emergence of a phenomenon, the phenomenon itself, and the subsequent effects hereof, we need to be able to disentangle and conceptually distinguish causes, concept, and effects from each other. While some lessons may be learned from the early European emergence of fiscal states and state-building efforts, the fiscality of many contemporary states has developed and changed under different conditions. Consequently, we need a fresh look at the concept of fiscal states to avoid conflating cause, concept, and effect. Otherwise, such conflation would limit our ability to empirically understand and examine these factors.

Second, the lack of explicit conceptualization of the contemporary fiscal state is problematic in a development policy context. Fiscal policy recommendations are often formulated in a way that seems to take for granted that African states are fiscal states and assumptions are made about their economic structures. Yet, if such assumptions are mistakenly made, effective tax policy implementation and fruitful yields are very unlikely, regardless of any ‘political will’ to tax more. Recognizing the absence or presence of fiscal states is a first step towards explaining why significant efforts to reform the tax system, including the implementation of VAT and administrative reforms, have not substantially increased tax revenue in many African countries (Fjeldstad, 2014; Fossat and Bua, 2013; Jeppesen, 2021; Moore, 2021; Moore and Fjeldstad, 2008).

This article conceptualizes the fiscal state in an empirically relevant and analytically applicable fashion. We pursue this purpose by reviewing the fiscal state literature to explore how the term ‘fiscal state’ has been understood and applied, which in turn leads to a definition of the fiscal state as a state whose public revenue base is dominated by tax revenue and loans, where the relationship between taxation and external and domestic borrowing is balanced, characterized by interdependence and therefore sustainable. The state is here understood in the Weberian sense, referring to contemporary states with centralized authority and sovereign rule (Frankema and van Waijenburg, 2021). We elaborate on and discuss the key defining attributes of fiscal states before we use a typology to illustrate the differences and dynamics between the fiscal state and three other state ideal types: the tax state, the rentier state, and the debt state. We substantiate the relevance of this endeavour by including country illustrations.

2. Reviewing the meaning of the fiscal state concept

Taxation is intimately linked with state-building. Fiscal sociology has produced many historical analytical narratives that describe the concurrent political and economic processes that led to the emergence of the fiscal state (Bonney, 1999; He, 2013; O'Brien, 2011; O'Brien and Hunt, 2007; Yun-Casalilla, 2012). Schumpeter's (1918 [1954]) seminal work marked the advent of the fiscal sociology literature. Schumpeter associates the ‘tax state’ with the transition from a domain state, where the main source of income was the king's own domain, to a situation where inhabitants in the kingdom provided most funds. Hence, a tax state relies primarily on the mobilization of revenue through taxes. Later, as rulers expanded domestic revenue mobilization, they were able to use future tax payments as security in large-scale borrowing. Combined, governments' capacity to borrow, their power to tax, and the interaction between taxation and loan constraints became central to shaping the powers of the modern state (Bonney, 1999). Taxation and borrowing created a nexus that sowed the seeds of today's complex state economies that many would call fiscal states (Moore, 2004).

Since its early employment in historical research to explain ‘why Europe’ in the state-building literature, the fiscal state concept has travelled across time and space to other contexts and development levels. On this journey, the concept has grown unclear and diluted. Today, the term often encompasses countries with various political regimes, social institutions, and modes of production. This risky journey towards conceptual stretching is to be expected when a concept is applied across disciplines from economic history to political economy, and it is not necessarily problematic (Sartori, 1970). Nevertheless, the lack of a common understanding, or clear definitions to facilitate comparative thoughts, inhibits interdisciplinary exchange as well as accumulation of knowledge about the dynamics of contemporary fiscal states and, not least, theorization about their causes and effects (Gerring, 1999; Hodgson, 2019). If we have not clearly defined the fiscal state, how can we recognize it when we see it or be certain that we are able to separate its potential preconditions from its defining attributes as well as these from its ascribed effects?

To explore how understandings of the fiscal state differ across fields and empirical contexts, we carried out a systematic literature review. We used the main academic databases, Web of Science and Scopus, employing a search strategy in which we combined keyword and reference searches. Specifically, we searched for sources that mention ‘fiscal state*’ and cite Schumpeter (1918), Moore (2004), Herbst (2000), Besley and Persson (2009), Bonney (1999), North and Weingast (1989), or Dincecco (2011). These constitute key references in the historic fiscal state literature and contemporary, specifically African, state-building literature. To stay clear of broader non-fiscal state-building perspectives, we excluded references to Levi (1988), who focuses primarily on political economy and actors, and Tilly (1992), who has a broader perspective on war and state-making. We expect the contractarian perspectives of fiscal states to be captured by Moore (2004). Second, we conducted a literature search combining the above references with the keywords ‘tax state*’, ‘rentier state*’, and ‘fiscal capacity’ to create a basis for delimiting the fiscal state from other types of states.

To illustrate, the search in Scopus on ‘fiscal state*’ produced 1,258 search results¹, which were reduced to 190 sources after the reference searches. Most of these do not engage the fiscal state concept analytically, though. Rather, they refer more or less in passing to ‘the rise of the fiscal state’ literature, including references such as Bonney (1999), O'Brien (2011), Yun-Casalilla (2012), He (2013), and Torres Sánchez (2007). For the review of fiscal state understandings, we relied only on sources that apply the concept in a manner where defining attributes can be discerned. Our literature review builds on these references combined with the corresponding group from Web of Science as well as key references added based on reviewer comments and inputs from a UNU-WIDER workshop (June 2021).

The fiscal state concept is not the only concept applied in research on fiscality in historical and contemporary state-building. For one, ‘fiscal constitution’ was defined by Brennan and Buchanan (1980) as ‘a prevailing type of fiscal system in a specific country at a given moment in history’ (in Bonney, 1999: 5) and later by Levi (1988: 49) as ‘a polity's rules and procedures for extracting and collecting revenue’. Fiscal constitution is arguably comparable to ‘fiscal regime’ (Monson and Scheidel, 2015: 7) as well as ‘fiscal systems’, most prominently used by Bonney (1999) but also by Schumpeter (1918 [1954]). All these terms allow us to study premodern fiscality – that is before actual states could be observed – as well as conduct analyses that cut across time and space because the state institutions are not included in concept definitions. For example, Frankema and Booth (2020) study fiscal systems and use this concept to investigate variations in fiscal capacity across colonial states. Andersson and Lazuka (2019) and Bonney and Ormrod (1999) use the term ‘fiscal system’ in a similar manner. As our main interest is in understanding the core defining attributes of contemporary fiscal states, we abstain from discussing further these empirically more inclusive analytical concepts or the dominant fiscal systems of the premodern era.

Few references define the fiscal state explicitly. Among those that do, the defining attributes either overlap or exhibit fundamental differences. In the following, we review the diverse understandings and applications of the fiscal state concept. This allows us to settle on a simple and concise definition of the fiscal state, which is both relevant and useful for understanding dynamics within and differences between states and whether SSA countries can be deemed fiscal states.

3. Diverse understandings of the fiscal state

At their core, most definitions of the fiscal state apply Schumpeter's simple understanding of the tax state as reliance mainly on income from taxes and, then, add the element of borrowing. Beyond this common ground, there seems to be a variety of understandings of the concept that raises important questions related both to the intension of the concept (i.e. what are the defining attributes) and its extension (i.e. to which empirical cases does the concept apply?) (Gerring, 1999; Sartori, 1970). Figure 1 presents these key questions that capture some of the differences and, one could say, inconsistencies in fiscal state understandings.

Figure 1. Review of conceptual extension and intension of the fiscal state.

Source: Authors' construction based on Ogden and Richards' triangle (1923).

3.1 Are all states fiscal states?

There is implicit disagreement in the literature concerning whether all contemporary states can be categorized as fiscal states. One group of scholars either explicitly or implicitly assume that all modern states are fiscal states because they have the ability to raise revenue. Differences among them become a matter of how states use their fiscal tools to meet public policy goals such as adjusting market failure or designing welfare policies (Buchanan and Musgrave, 2001; Jensen, 2011; Musgrave, 1996). Such differences between states are taken as expressions of, for example, variations (Musgrave, 1996) or advancements (Jensen, 2011) of fiscal states. Seen through this lens, all states in SSA would be considered fiscal states.

Other scholars, such as Yun-Casalilla (2012: 1), reserve the term ‘fiscal state’ for ‘nations and democratic states as they emerged over the nineteenth century’. Moreover, and relevant for exactly contemporary states, Yun-Casalilla distinguishes the fiscal state from the rentier state by whether they penetrate society to earn income (Ibid: 5). Over time, many scholars have analysed the rentier state as a distinctive state type defined by its dependence on rents from extractive natural resources or donors (e.g. Beblawi, 1987; Chaudry, 1989; Mahdavy, 1970; Moore, 2004; Ross, 2016; Sachs and Warner, 2000). Interestingly, the literature review showed little overlap between references to fiscal states and rentier states, which implies that the two literatures are siloed. The conceptual relation between fiscal and rentier states has thus received limited attention. However, it is clear that rentier states and fiscal states represent contrasting examples of state development functioning. Rentier states are often characterized by their ‘pathologies’ (Moore, 2004), meaning that dependence on particularly oil rents tends to have a series of negative effects on the economy and society. In contrast, fiscal states often involve positive connotations because the diverse revenue base is perceived to have several beneficial effects on the economy and society. Hence, the existence of the concept of rentier state and its widespread application on empirical cases such as Nigeria (Lewis, 2007) or Angola (Hammond, 2011) demonstrates the empirical relevance of this state type. Consequently, the contrasting foundation and empirical relevance of the rentier state ideal type indicates that not all states are fiscal states.

The question concerning the extension of the fiscal state is, of course, closely linked to its conceptual intension. The following three sections are structured around questions that arose from the literature review and that represent dimensions on which the defining attributes of the fiscal state concept vary, sometimes explicitly but often implicitly.

3.2 Is capacity a defining attribute of fiscal states?

The first question concerns whether capacity (administrative, state, or fiscal) is explicitly included as a defining attribute of fiscal states. Many see state capacity as a core attribute of the fiscal state: Boucoyannis (2015: 304) argues that state capacity allowed for the imposition of taxation, and Dincecco and Katz (2014) describe state capacity as states' ability to extract taxation. Cardoso and Lains (2010: 5) associate the rise of the fiscal state with ‘an increase in the ability of centralized states to manage the administrative apparatus for raising taxes, as well as with the ability of the sophisticated financial institutions to manage public debt’. Yet, many other sources do not mention state capacity explicitly as they discuss fiscal states (Musgrave, 1996; Omeje, 2016; Sandbakken, 2006; Zhang, 2017).

Tightly linked to state capacity is fiscal capacity, which potentially adds another layer to the complexity. Fiscal capacity and fiscal states are sometimes used in conjunction and interchangeably, but only few papers mention both fiscal state and fiscal capacity. For example, the recent paper by Albers et al. (2020) is titled ‘The fiscal state in Africa’, but it only mentions and defines fiscal capacity. Therefore, it is difficult to determine if fiscal capacity is seen as a defining attribute of fiscal states or rather as a precondition or effect of its presence. Interestingly, the literature on fiscal capacity yields three times as many hits in Web of Science and one and a half times as many hits in Scopus. One likely reason is that while fiscal state and fiscal capacity might share core denotations, the latter seems to be more easily approached and measured if taken by the number of quantitative studies on fiscal capacity (e.g. Cárdenas, 2010; D'Arcy and Nistotskaya, 2018; Dincecco and Prado, 2012; Jensen, 2011).

Fiscal states and fiscal capacity arguably carry different meanings and attributes because they represent two distinct analytical levels: one characterizes a type of state; the other characterizes attributes of a state. Meanwhile, fiscal capacity warrants its own discussion, and two curiosities are of relevance here. First, fiscal capacity is almost exclusively used in reference to states' ability to tax (e.g. Cárdenas, 2010; D'Arcy and Nistotskaya, 2018; Ricciuti et al., 2018; one important exception being Frankema and Booth, 2020). Connotations of the term fiscal capacity simply imply having capacity to ensure government revenue. Therefore, it would make more sense to talk of types of fiscal capacity such as tax capacity, which would refer to the ability to tax, borrowing capacity, as the ability to borrow, and rents capacity, as the ability to extract rents including from extractive natural resources. Such disaggregation is relevant because studies that use fiscal capacity but refer to, and measure it by, states' ability to collect taxes could simply opt for tax capacity (e.g. Baskaran and Bigsten, 2013; Dincecco et al., 2019; Dincecco and Prado, 2012; Jensen, 2011).² Likewise, Ricciuti et al. (2018) argue that tax ability is just one dimension of fiscal capacity (the other dimension in their argument is impartiality). Second, researchers of fiscal capacity often focus on direct taxes. Even though direct taxes have traditionally been perceived as the most developed form of state income (e.g. Kaldor, 1963), it is debatable whether direct taxes are a central enough tax source in today's developing states for them to be a relevant indicator of fiscal capacity (Bird and Zolt, 2005). This concern is also relevant for the conceptualization of the fiscal state; a narrow focus on this ‘modern tax’ (Genschel and Seelkopf, 2021) can be problematic. The tax landscape in Africa obviously looks different than when the European fiscal state emerged. Direct taxation cannot be assumed to be broad-based in an African context and, hence, this tax type might entail very different qualities and implications. To the extent that fiscal capacity denotes the ability to tax, one risks neglecting the important conceptual distinction between the tax state and the fiscal state and the progression in sources of revenue that differentiate the two.

3.3 What are the revenue sources of fiscal states?

A second key question that arises concerns revenue sources. There seems to be consensus that fiscal states are financed through taxes and borrowing, but the relative weight given to these revenue sources differs in the literature.

Most scholars mainly associate the fiscal state with the ability to tax (Albers et al., 2020; Cardoso and Lains, 2010; Dincecco and Prado, 2012; He, 2013; Yun-Casalilla et al., 2012), but few define the characteristics of taxation further. Thus, He (2013) emphasizes the centralized nature of tax collection, while O'Brien (2005: 30–31) implies that moving from direct taxes (such as land taxes) to easier-to-collect indirect taxes made Britain the most effective fiscal state in Europe at the time.

There is little discussion of how dependent a state needs to be on taxation vis-à-vis other revenue sources for it to qualify as a fiscal state. To the extent that this remains un(der)specified, there is a risk of conceptual overlap between the tax state and the fiscal state. Meanwhile, most scholars do mention that fiscal states – in addition to being able to tax – have the ability to issue paper money or borrow (e.g. Bonney, 1999; Cardoso and Lains, 2010; He 2013; Moore, 2004; Piano and Salter, 2021). While issuing paper money was a central part of early state-building, many contemporary states are not monetarily sovereign; their currencies are not in demand, they are indebted in foreign currencies, they are part of monetary unions, and they face global pressure not to conduct monetary policy (Gevorkyan and Kvangraven, 2016; Smith, 2021; Sylla, 2020). Hence, monetary policy is not included in the definition here.

As for borrowing, the rise of the fiscal state is attributed to when it became possible to secure (mainly commercial) loans (Bonney, 1995). This allowed the state to achieve ‘an economy of scale in mobilizing financial resources’ in the long term (He, 2013: 5) and gave it ‘superior capacity … to raise the financial resources to cope with emergencies’ (Moore, 2004: 301). Emergencies, most notably wars (e.g. Dincecco and Prado, 2012), meant that tax revenues were insufficient and prompted rulers to borrow. Borrowing was the outcome of taxation in that states' fiscal systems provided the basis for credible commitment to lenders who gave significant loans on the basis of future tax revenues. As Bonney argues, fiscal states' powers were to be made up by ‘the capacity of governments to borrow, and the interaction between the constraints on borrowing and the power to tax’ (1995: 505).

Curiously, while borrowing is undisputedly a defining attribute of fiscal states, as highlighted above, empirical research on contemporary fiscal states mainly tends to focus on tax collection (Albers et al., 2020; Beramendi et al., 2019; Besley, 2020; Moore, 2021). The reason might be that in today's complex economies, borrowing can be obtained through many other foundations than expectations about future tax revenue, for example, through expectations to states' growth potential or through natural resource collateralization (AFDB, 2021; Mihalyi et al., 2020; Smith, 2021). It might be that the ability to borrow is taken for granted in today's global economies. Hence, the literature instead (over)emphasizes the importance of taxation in the state economy. Importantly, this seems to reflect the fact that debt and taxes have often been presented as substitutes, i.e. if a state lacks tax revenue, for example in times of crisis, it can instead borrow money to cover expenses. Yet, this understanding de-emphasizes the important interplay between the two revenue sources. We argue that a key feature of the fiscal state is exactly that taxes and borrowing are complementary rather than substitutes. For example, the soundness of the tax base will influence the ability to gain favourable loans. This complementary interdependence between taxes and debt is part of what makes the fiscal state more sustainable than other state types. We will return to this point later.

3.4 Do fiscal states need to spend in certain ways?

A final significant question concerns whether fiscal states are defined solely by their sources of revenue or the character of the states' expenditure is also a defining attribute. The scholarly contributions analysed in the above do not address this question explicitly. Some contributions include the composition of public expenditure as part of the definition of a fiscal state whereas many other contributions do not. What is clear, therefore, is that there is significant unclarity in the literature.

Scholars who include the expenditure side in their understanding of the fiscal state consider whether domestic revenues are used to finance a government's policy objectives or publicly negotiated common goods, including redistributive purposes (Bird and Zolt, 2015; Buchanan and Musgrave, 2001; Cardoso and Lains, 2010; Martin et al., 2009). They, for example, connect the rise of fiscal states to the emergence of democratic rule, modern state bureaucracy, and the first signs of public services beyond protection of citizens (Buchanan and Musgrave, 2001; Mehrotra, 2013; Yun-Casalilla, 2012).

Musgrave (1996) distinguishes between four types of fiscal states according to the structure of states' expenditure. His point of departure is public finance and the argument that fiscal states need to be categorized according to whose needs the state accommodates and the extent to which the state redistributes and meets market flaws. This is very different from other branches of the literature reviewed that mainly focus on the state's revenue side. For instance, Albers et al. (2020) are clearly interested in discerning whether fiscal states are emerging in sub-Saharan Africa based on long term analysis of whether domestic revenues are increasing.

Arguably, it is difficult, if not impossible, to include both the revenue source dimension and an expenditure side in the definition of a fiscal state while maintaining the ambition to develop an analytically useful concept. Including the expenditure side complicates the concept and empirical identification. And, more importantly, as will be theorized below, rather than constituting a defining attribute of the fiscal state, a contractual relationship between state and individuals is part of the feedback mechanisms that can reinforce and maintain the fiscal state. To illustrate, if a present-day fiscal state invests in increasing productivity on a long-term basis, this can strengthen the state's ability to tax and borrow in the future. On the other hand, unwise investment decisions might negatively affect future revenue sources. Assuming that the revenue sources and their composition then change, the state may enter into negative dynamics that will undermine its fiscality and make it fall out of the category of a fiscal state. Timmons (2005) is but one contribution showing how state expenditure is explained by the composition of revenues pointing to e.g. how a high share of property tax in the USA incentivizes the government to spend a lot on private property protection. Steinmo (1993) shows how variations in welfare states can be explained by types of tax systems. These contributions need to analytically separate public expenditure policies from revenue sources since the latter explains the former. In addition, it is possible for a state to have an advanced system of social services without being a fiscal state, as is the case with some rentier states. Looking at the types of, say, social expenditure alone in these states would not necessarily capture the rentier state dynamics behind them. Likewise, debt states might use debt to finance growth-enhancing investments (i.e. expenditures), which in the future may lead to a change in the revenue source composition and, thus, in financial state type. We return to this point when presenting our typology and debating the sustainability of the different state types in section 4.2. Consequently, we argue that expenditure is not part of the fiscal state definition but instead influence its ability to persist in the future.

If the output side of public finances is included, the conceptual intension increases, extension decreases, and, importantly, conceptual complexity (and confusion) grows. In sum, the blurriness of the fiscal state concept inhibits a proper understanding of the implications of the fiscal state.

3.5 Summing up

Based on the literature review and the four conceptual questions that arose from it, we can make several observations. (1) While some contributions seem to assume that all contemporary states are fiscal states, there are grounds for arguing that this is not necessarily the case. (2) When considering how to define a fiscal state, attributes such as state capacity and fiscal capacity are important features. However, both are elusive concepts themselves and therefore not beneficial in an analytically applicable definition. (3) Though some scholars will analyse fiscality of a state by its ability to tax, other revenue sources are important to include, especially borrowing. (4) Lastly, many define fiscal states by the expenditure side or its link to revenue. Meanwhile, by including this in the definition, we risk confusing defining attributes with effects thereof.

Based on these conclusions, a definition of a fiscal state should enhance parsimony and clearly differentiate the concept's defining attributes from its causes and consequences (Gerring, 1999; Hodgson, 2019). This implies that (1) the concept should not include the size and composition of public expenditure, which can instead be explained by the nature of a fiscal state; (2) it should be clearly distinguishable from other effects such as welfare, governance, and inequality; and (3) it should be distinguishable from its preconditions, such as the level of commercialization of the economy. While conceptual parsimony imposes limits on the depth of the concept, it increases its theoretical and field utility. In addition to parsimony and differentiation from causes and consequences, we seek familiarity and resonance (Gerring, 1999) by maintaining the core defining attributes identified above. We find that only a definition based solely on the characteristics of state revenue can accommodate these criteria. The remainder of the article will present and discuss the chosen definition, its key components, and illustrate the conceptual delimitations with empirical examples.

4. Definition and conceptualization of the fiscal state

Accommodating the criteria above, we define the fiscal state solely on the state's revenue side. Drawing on Moore's (2004) and Bonney's (1999) definitions, we define the fiscal state as a state whose public revenue base is dominated by tax revenue and loans, and where the relationship between taxation and external and domestic borrowing is balanced, characterized by interdependence and thereby sustainable. The definition presumes that the state is a sovereign legal entity in the sense that it can use centrally collected tax revenues to take on long-term borrowing from the market (Frankema and Van Waijenburg, 2021; He, 2013). Centralization of state power has been seen as an important precondition for the rise of the fiscal state (e.g. Besley and Persson, 2009; Cárdenas, 2010; Cardoso and Lains, 2010; O'Brien and Hunt, 2007; Tilly, 1992). Here, sovereignty and centralization are not elaborated but are in a sense assumed in the definition.

Besides its specific focus solely on the revenue side, our definition distinguishes itself not by the fact that it includes taxation and borrowing, which is indeed quite common, but rather by its particular emphasis on the balance between them. To some extent, all contemporary states can tax and borrow money; however, the level of taxation, debt burden, and their internal relations vary greatly. We are not interested in a state's capacity to tax or borrow per se but rather in states' relative dependence and composition of revenue sources as this holds implications for the sustainability of states' public finances. A balance of taxation and borrowing is essential to the fiscal state (Bonney, 1999; Hart et al., 2018). We briefly elaborate on this balance before introducing a typology of fiscal states based on our definition. The typology includes illustrative examples of contemporary African states to demonstrate features of each ideal type.

4.1 A balance between taxation and borrowing

Most states today can tax and borrow, but we argue that the presence of a balance is what sets fiscal states apart from other states that depend exceedingly on loans, state monopolies, or natural resource extraction. Such a balance renders fiscal states more resistant to economic crises and shocks. Yet, the two revenue sources are not necessarily intimately linked. The loan market is diverse and consists of many different political and commercial interests. China, for example, lends money to lower-income countries in Africa as a part of political deals without necessarily considering their tax base (Bräutigam, 2020; McCormick, 2008).³ This detachment between taxation and borrowing is not a new phenomenon. Frankema and Booth's (2020) observations of colonial states can be paralleled to contemporary states. They emphasize that decisions to take loans were not based primarily on risk assessments by private investors (or indeed only on tax capacity in the colonies) but rather on the decisions of officials of the metropole weighing the cost of development projects in the colony against the risks of default (Ibid.: 12). This was one reason for the lack of fiscal modernization of the colonial state and, conversely, an argument for why a balance and interdependence between taxation and borrowing is important.

To assess the balance between tax and debt, one must include the terms on which a state borrows. Borrowing is a means to stabilize an economy. It can be used to finance short-term measures of crisis management when tax revenue momentarily drops, or to support long-term growth-enhancing investment policies (AFDB, 2021; Dincecco and Prado, 2012; Moore, 2004; Smith, 2021: 99–101). Most recently, however, debt has in many countries led to what IMF describes as fiscal stress rather than economic stability during the COVID-19 pandemic (AFDB, 2021; Smith, 2021: 101–02). Importantly, whether debt has negative effects on fiscal sustainability does not necessarily depend on the size of a country's debt (Chandler, 2020). Instead, the conditions surrounding its debt such as the maturity of the loan, the interest rate, and the currency in which the loan is taken are determinants of a country's ability to service and repay debt obligations (Smith, 2021: 91). Most often, borrowing based on stable tax revenue rather than growth potential or natural resource collateralization can ensure better and more stable loan conditions. This is because borrowing based on natural resource rents or growth potential renders the state more vulnerable. For example, if the prices of natural resources drop, if debt-financed project fails to generate predicted returns, or if the global market starts retracting such as during the financial crisis or the COVID-19 pandemic, this can limit borrowing opportunities or increase interest rates. The more tax revenue the government can collect now and in the future, the better its means for servicing debt, which in turn is likely to reduce the cost of borrowing. A balanced composition between tax and debt thus provides for better loan conditions and sustainability and makes fiscal states more resistant to crises.

We emphasize that fiscal states are characterized by a revenue composition dominated by a balance of taxation and borrowing because this alerts researchers to the importance of sustainability. Admittedly, including sustainability in the fiscal state concept risks adding to its elusiveness. However, we contend that sustainability applied in its most basic meaning is in fact fairly intuitive (Merriam-Webster, 2022). The relationship between taxation and borrowing is sustainable when it is maintained at length without interruption or weakening and when neither of the revenue sources are depleted or permanently damaged through its use. To assess sustainability, analyses need to be contextual, including both national and international factors, and they must also include a time dimension. One example of such an assessment is IMF's debt sustainability framework, which includes various dimensions of domestic capacity such as institutions and the tax base. Importantly, such an assessment acknowledges the deep-rooted variations in debt sustainability between countries at different levels of economic development (IMF, 2013, 2018; Smith, 2021: 96–97). Accordingly, this also implies that economic growth would be a poor indicator when assessing fiscal sustainability because it does say much about the underlying conditions of fiscal sustainability including the breadth and diversity of the tax base (for a discussion of preconditions of the fiscal state, see Bak et al., 2021). Hence, we would risk mischaracterizing states where growth is driven by, for example, large external rents or extensive borrowing, as we elaborate below.

4.2 Typology of ideal state types

According to the presented definition, not all states are fiscal states. This is consistent with the existence of other state types emerging from the literature review, such as the rentier state. In terms of the dominance of and balance between revenue sources, a rentier state exhibits a different structural basis than fiscal states. Hence, we argue that fiscal states and, for example, rentier states are distinct (ideal) state types rather than variations within a group. To illustrate, we have developed a state typology to demarcate the fiscal state from other ideal state types. Figure 2 operates with continuums in a coordinate system and thus emphasizes the dynamic room for state transitions as well as the grey zones between the ideal types.

Figure 2. Distinguishing between ideal state types defined by their revenue side.

Note: The dotted lines are included as figurative indications. The line without an arrow exemplifies when taxes and borrowing become dominant. The line with an arrow exemplifies when a balance is achieved between taxes and borrowing.

Source: Authors' construction.

Other typologies map states according to their revenues. Bonney and Ormrod (1999), for example, have a helpful model of fiscal systems. However, our typology distinguishes itself in two ways. First, where Bonney and Ormrod's model explains historic fiscal systems, the present typology has the potential to understand contemporary states. Second, the Bonney-Ormrod model tends towards presenting a somewhat linear view of state development, not least given the inescapable historical empirical basis of their model. In contrast, our typology presents a dynamic perspective as it allows for states to move between the depicted ideal types.

As defined above, (1) fiscal states' revenue base is dominated by taxes and loans with a balance and interdependence between taxation and domestic and external borrowing. Thus, fiscal states exhibit relatively complex state financing. This differs from (2) tax states that have a relatively simple public economy primarily financed through taxes, and (3) debt states that might have some tax collection but predominantly rely on state revenue from borrowing, which is not necessarily linked to expected future tax revenue. Lastly, (4) rentier states neither tax nor borrow much but attain their main financing from aid or natural resource extraction. A last group of states could be termed non-fiscal in the sense that they have very limited state resources, such as fragile states experiencing long-term conflict. In Figure 2, these states would be placed with the rentier state but with a fundamentally different political economy and much less room for transitioning towards becoming a fiscal state. As these are ideal state types, real-world states typically lie in the grey zones between them. For this reason, we stress the importance of identifying the dominant revenue source(s) when determining empirically whether a state can be deemed a fiscal state. Consequently, if a state's finances are dominated solely by either tax revenue, aid or natural resources, or debt, the state does not classify as a fiscal state.

Though we define the ideal types along a revenue source dimension, we theorize that states can evolve and move between these ideal types as a result of changes in the composition of revenue sources – that is, the balance and dominance of revenue(s), as well as changes in expenditure demands. Changes in state expenditure can influence revenue composition through several feedback mechanisms: for example, if debt service expenditure becomes too high; if public investments in a sector contribute to sector growth and, in turn, more sectoral taxation; or, as indicated in fiscal contract theories (e.g. Levi, 1988), if more spending increases quasi-voluntary compliance and, in this way, eases and increases tax collection. Hence, while the input side of revenue and the output side of spending are theoretically and empirically distinct, the output side manifesting the level of quid pro quo can help improve, sustain, or undermine the input side (Moore, 2004, 2021; O'Brien and Hunt, 2007; Prichard, 2015). While states can potentially move, for example, towards becoming fiscal states, they tend to be caught in the status quo as their main sources of revenue are likely to be self-reinforcing.

The rentier state receives rents primarily from foreign actors; they accrue directly to the state, and ‘only a few are engaged in the generation of this rent (wealth), the majority being only involved in the distribution or utilization of it’ (Beblawi, 1987: 50). In many countries, the reliance on rents has demotivated productive investment in the broader economy and limited economic diversification (e.g. Beblawi, 1987; Karl, 1997; Ross, 2016; Savoia and Sen, 2021). Reliance on rents will tend to be self-reinforcing; as the large amounts of ‘unearned income’ provides little incentive to invest in economic diversification, which implies that the composition of revenue comprising a heavy reliance on rents is maintained, if not exacerbated in the future (Ross, 2016). To illustrate, Nigeria gets close to an ‘archetypal oil nation’ and, thus, rentier state (Watts, 2004: 50). More than 95 per cent of foreign exchange earnings derive from petroleum exports, and oil revenues constitute 65–80 per cent of government revenues (depending on data source and time of assessment) (Fagbemi and Adeoye, 2020; Hailu and Kipgen, 2017). Nigeria's ‘rentier state’ characteristics have been self-reinforcing, and moreover, there is a broad consensus that its oil dependence has led to governance challenges including large-scale corruption, a lack of diversification of production, and other features of the so-called ‘resource curse’ (Auty, 1993; Lewis, 2007). However, this does not mean that Nigeria could not move towards other state ideal types. Its dependence on oil can make its indebtedness larger in times when oil prices are low, which could push it towards becoming a debt state. Alternatively, political decisions to make productive investments beyond the oil sector (i.e. its expenditure decisions) could lead to gradual diversification of the economy. Such interventions would likely serve to diversify the revenue base in the longer run, which would improve the balance between different revenue sources and increase its fiscal sustainability. In other words, there is no deterministic relation between extractive natural resources and their negative effects. The relation is mediated by political and institutional variables (Lewis, 2007; Rosser, 2006).⁴

Debt states tend to be caught in a trap of accumulating debt as one loan leads to another to pay off earlier ones. Meanwhile, it becomes continuously difficult to meet recurring expenditures and the spending demands that caused the loans in the first place. This can potentially lead to debt overhang or the need for debt relief. While a state's revenues are highly unlikely to consist only of loans, loans can be a dominant source of state financing which is obtained, for example, based on debt collateralization, such as commodity-secured loans, or growth potential (AFDB, 2021: 4–5, 64–65; Mihalyi et al., 2020; Smith, 2021: 116). This makes debt states vulnerable to drops in commodity prices, changes in global liquidity, credit ratings, or significant increases in borrowing costs and loan conditionalities (e.g. Gevorkyan and Kvangraven, 2016). To illustrate, while Zambia is highly dependent on its copper industry and collects various types of taxes, it has moved towards becoming a debt state. Zambia's debt load became increasingly unsustainable during the 2010s as a result of ‘reckless’ over-borrowing, a depreciation of the Zambian currency, and borrowing for consumption instead of investments (Smith, 2021: 87–88, 143–145; ZIPAR, 2018). If cobber prices had not decreased in the period or borrowing had been used differently, the Zambian economy might have evolved more positively. Yet, this highlights the vulnerability of relying heavily on fluctuating resource rents and unbalanced debt, thus demonstrating the lack of sustainability in Zambia's financing. Consequently, Zambia became the first African country to default on its Eurobond loans in 2020 (Bräutigam, 2021; Smith, 2021: 116). Today, Zambia's debt makes up more than 100 per cent of its GDP (Benson, 2022; Smith, 2021: 88). This accumulation of debt occurred despite significant debt relief and cancellations throughout the 2000s, although it could be argued that this history of debt cancellation also led to moral hazard risk when obtaining new loans (Bräutigam, 2021). Zambia's movement towards becoming a debt state highlights how debt accumulation tends to be actor-driven. For example, the Zambian parliament agreed to raise the ceiling on external borrowing, and the Zambian government ignored warnings from IMF in the mid-2010s (Smith, 2021: 160; ZIPAR, 2018).

The least stable of the four types is the tax state. As taxation increases, so does demand for expenditure, and an established infrastructure for taxation provides a basis for beneficial borrowing, which is likely to set off a movement towards becoming a fiscal state. Accordingly, the tax state seems to have status as a historical rather than a contemporary phenomenon (He, 2013). Evolving from a tax state to a debt or a fiscal state is obviously conditioned by the nature of economic transformation as well as the extent of access to the lending market. Access to loans is based on the state's ability to make credible commitments to loan providers due to an expanded and diversified economy. In addition, serious economic crises or conflicts could undermine taxation, either diminishing state finances altogether or necessitating loan-taking, which could push the country towards a status as a non-fiscal or debt state.

The fiscal state ideal type is the most stable of the four types due to its reliance on a balance of and interdependence between taxes and loans. The balance allows it to acquire loans on good terms based on its assurance in tax revenue, and tax revenue remains relatively stable because favourable loans help ensure a stable economy even in times of crisis. The balance and interrelation thus imply that the state's public finances are more sustainable and stable, and for this reason fiscal states are likely to be conducive to long-term economic development. As the revenue side reinforces a positive nexus, only few situations, such as unsustainably high expenditures, can lead them off track. To illustrate, Senegal would be much closer to the fiscal state ideal type than Nigeria or Zambia. Senegal's tax-to-GDP level is about 16–18 per cent; just at or above the Sub-Saharan African average (Bak, 2019; Niang and Mbaye, 2020; UNU-WIDER, 2021). A considerable proportion of Senegal's tax revenue is personal or corporate income tax (30 per cent), while VAT constitutes around 40 per cent. Senegal's revenues from natural resources or aid are not significant. Non-tax revenues thus consisted of 1 per cent of GDP. The domestic revenue take has allowed Senegal to maintain considerable debt levels, which were projected to peak at 68 per cent of GDP in 2022 (IMF, 2021). However, due to the COVID-19 crisis and the effects of the war in Ukraine, debt levels are now expected to reach 75 per cent (IMF, 2022), and Senegal's debt has been assessed by international financial institutions to be sustainable with a moderate risk of over-indebtedness and limited ability to absorb shocks (IMF, 2022). Compared to Nigeria and Zambia, Senegal was moving towards becoming a fiscal state; however, the balance between taxation and borrowing seems to be tipping, as in many other countries, during the COVID-19 crisis. While Senegal's per capita GDP growth has only been slightly higher than that of Zambia or Nigeria, Senegal's tax base is more diverse, and its reliance on taxes relative to rents is higher. Thus, Senegal's case shows that it is not economic growth per se that could make a fiscal state but the potential for fiscal sustainability that build on the level of diversification of the economy (Bak et al., 2021).

In sum, we argue that a fiscal state is defined and identified by its input side. In practice, data on taxation (excluding natural resources revenue) and loans are the main indicators to use, but actually assessing the balance requires a qualitative assessment. Furthermore, it is imperative to take both conditions of revenue acquisition and factors on the expenditure side into consideration as these can improve, sustain, or undermine the revenue input. An extensive assessment is beyond the scope of this paper. Instead, we discuss the implications of our fiscal state definition and typology for understandings of fiscality and (ideal) state types.

5. Conclusion

While the different strands of the fiscal state literature originate from the same Schumpeterian roots, there is a tendency not to interlink historical and contemporary branches. Naturally, there are huge disparities in the international and political conditions of state-building now and then. However, due to the lack of grounding and perhaps aversion to drawing on historical state-building analysis, fiscal state-building today tends to be regarded mainly in terms of its institutional characteristics with less regard to the significance and necessity of economic changes for the emergence of fiscal states.

In this paper, we revisit the fiscal state concept with the objective of arriving at a definition which can be applied to identify the presence of fiscal states in SSA. After reviewing the diverse understandings of the fiscal state across literatures, we arrive at a definition that is both in line with the core defining attributes suggested in the literatures and analytically relevant and applicable to contemporary Africa: a fiscal state is a state whose public revenue base is dominated by tax revenue and loans, and where the relationship between taxation and external and domestic borrowing is balanced and thereby sustainable and characterized by interdependence. Specifically, we emphasize the importance of borrowing, which has received insufficient theoretical and empirical attention in the literature, especially related to contemporary African states. Furthermore, our definition highlights the balance between taxation and borrowing, which we argue to be central to deciding whether a state is really a fiscal state or rather tends towards a rentier, non-fiscal, or debt state.

Applying this definition would arguably imply that most SSA countries would not be categorized as fiscal states. Between the illustrative examples mentioned, Senegal would come closest, but even here, the balance is tipping towards a debt state. This apparent lack of fiscal states in SSA can to a large extent be explained by the absence of economic preconditions. Taxing more requires, first and foremost, a taxable surplus, and this requires economic transformation and diversification; without such a tax base, loans are acquired on other and more vulnerable foundations. This conclusion has important implications not only for how analyses of African public finances are presented and for subsequent policy recommendations, which tend to focus overwhelmingly on increasing domestic taxation. Applying our typology to analyse the fiscality of SSA states contributes to better understandings of the policy options available to governments, for example, when dealing with external shocks such as the COVID-19 pandemic. There are limits to what governments can accomplish out of sheer will if they do not first take account of their fiscal foundation and their actual economic room for manoeuvring.