Introduction
This book examines the comparative development of fiscal capacity in a number of Asian and African colonial states.Footnote 1 We situate this study in the context of the changing world order in the long century between 1850 and 1960. The end date signals the termination of colonial rule in most of sub-Saharan Africa, with the exceptions of Portuguese Africa (1975) and the semi-autonomous status granted to the Union of South Africa (1910).Footnote 2 The starting date of 1850 is more arbitrary, as colonial control began earlier in some parts of both Asia and Africa. But, broadly speaking, it was only after 1850 – and indeed in many of the colonies which are examined in this book, several decades later – that the colonial powers began to develop the fiscal systems of the territories which came under their control.
We define ‘fiscal capacity’ as the ability of the state, or its representative agents, to collect revenues in order to provide public goods and services. In this definition, fiscal capacity refers not just to the power of the state to assess and collect taxes, but also to its ability to extract non-tax revenues from state monopolies, from enterprises such as railways or from foreign aid. Moreover, fiscal capacity also refers to the ability of the state to raise loans to supplement both tax and non-tax revenues, especially for capital works. All the chapters in this book examine the long-term development of colonial fiscal policies, using this definition of fiscal capacity. They also explore how different colonial administrations tackled the problem of revenue centralization, and what factors influenced both the revenue mix and changing expenditure policies. The chapters also address wider questions on the use of forced labour, debt creation, the impact of increasing global and local trade, and the development of financial systems.
We will argue in more detail that existing concepts and theories of fiscal development have had little to say about the nature and determinants of fiscal development in a colonial setting. Until now, the historical literature has paid much more attention to the rise of the fiscal state in Europe and its offshoots in the Americas, than to the development of fiscal systems in the former colonies and protectorates in Africa and Asia. Fiscal histories that employ a global comparative outlook have either concentrated on the comparative features of fiscal state formation in Eurasia (Yun-Casalilla & O’Brien Reference Yun-Casalilla and O’Brien2012, He Reference He2013), or have focused on the North-South divergence in the Americas (Engerman & Sokoloff Reference Engerman, Sokoloff and Haber1997, North, Summerhill & Weingast Reference North, Summerhill, Weingast, Bueno de Mesquita and Root2000, Sokoloff & Zolt Reference Sokoloff and Zolt2006, Grafe & Irigoin Reference Grafe and Irigoin2012). While there has been a recent upsurge of studies in the fiscal history of African and Asian colonies (Booth Reference Booth2007a, Frankema Reference Frankema2011, Gardner Reference Gardner2012, Wahid Reference Wahid2013, Frankema & van Waijenburg Reference Frankema and van Waijenburg2014, Huillery Reference Huillery2014, Alexopoulou & Juif Reference Alexopoulou and Juif2017), the findings of these studies have yet to be synthesized. This volume brings the experiences of colonies in Asia and Africa into a global comparative framework in order to clear the ground for new theories on the relationship between colonial state-building and the development of fiscal capacity.
The Colonial State in a Changing World Order, 1850–1960
The fiscal systems analysed in this book all developed during the wave of global imperialism that emerged between the independence wars in the Americas (c. 1776–1830) and the end of colonial rule in Africa. The spread of the Industrial Revolution from Britain to the wider Northern Atlantic basin opened up new possibilities to bring overseas areas under direct control. The heightened pace of technological change, and accumulation and application of scientific knowledge brought solutions to a range of problems, including the treatment and prevention of tropical disease (e.g. quinine to treat malaria, smallpox vaccination), communication over large distances (e.g. telegraph) and the preservation of perishable commodities (e.g. refrigerators). Moreover, the major revolutions in transportation, railways and steam ships, in particular, greatly enhanced global trade and opportunities of colonization (Kuznets Reference Kuznets and Kuznets1974, 168, O’Rourke & Williamson Reference O’Rourke and Williamson1999, Williamson Reference Williamson2011, Frankema, Williamson & Woltjer Reference Frankema, Williamson and Woltjer2018). The colonial states, no matter how weak and understaffed many of them were, occupied a critical place in this compulsory process of globalization (Hopkins Reference Hopkins2002, 6–7; see also Burbank & Cooper Reference Burbank and Cooper2010, chapter 10). Colonial states were responsible for maintaining law and order, and for creating a political space for infrastructure development, trade and capital investment, in order to create profitable export sectors. All these aims required money. The fiscal system, it can be argued, formed the backbone of the colonial state.
The expansion of formal and informal empire was led by a handful of Western European powers, but was not an exclusively European undertaking. The United States annexed the Philippines, and replaced its Spanish government in 1900. Threatened by Western incursions in the mid-nineteenth century, Japan decided that an empire was necessary if it was to be taken seriously by the other global powers. In the late nineteenth century, Meiji Japan became an imperial power by adapting foreign production technologies and by engaging in rapid military capacity building.
Although the physical distance from metropolitan centres provided the colonial state with some degree of policy autonomy, improved transportation and communication technologies also meant that metropolitan supervision was more effective than it had been in the early modern era. Both internal and external factors caused changes in colonial policy, including fiscal reforms. In Indonesia, pressures from within the colony and in the Netherlands brought an end to the cultivation system and a transition to the so-called liberal era in the 1860s and 1870s (Fasseur Reference Fasseur1991). In India, the rebellion against British East India Company rule in 1857–8 induced a reorganization of the colonial army as well as the government administration and the financial administration (Prakash Reference Prakash2004). The Meiji reforms in Japan involved a reorganization of the fiscal apparatus, which in turn influenced later Japanese imperialism. The discovery of diamonds in 1867 had an enormous impact on subsequent economic and political development of South Africa, including revenue and expenditure policies (Feinstein Reference Feinstein2005, 108–9, Gwaindepi & Siebrits, Chapter 9, this volume).Footnote 3
While major investments in fiscal development in sub-Saharan Africa were made only after the scramble of the 1880s and 1890s, many African societies experienced major changes in the relations with their trading partners, both African and European, during the middle decades of the nineteenth century. Even though the shift from slave exports to commodity exports occurred at different points in time, this ‘commercial transition’ coincided with improving terms of trade for African commodity exports and rapidly expanding export volumes. The expansion of ‘legitimate commerce’ across the nineteenth century, in turn, created new opportunities for financing the colonial state (Law Reference Law1995, Frankema & van Waijenburg Reference Frankema and van Waijenburg2014, Frankema et al. Reference Frankema, Williamson and Woltjer2018).
By the end of the nineteenth century, when the scramble for Africa had largely been completed, European empires reached the territorial size that they would retain until the 1940s. In this period, the growth of colonial populations, output and trade accelerated, despite the problems caused by the First World War and the Great Depression. While the transport and communication revolution transformed the logistics of empire-building, innovations in military technologies, including improved naval vessels and machine guns, enhanced the superiority of the metropolitan ‘cores’ over the colonial ‘peripheries’. Investments in railways, roads and harbours opened up the more promising hinterlands to increasing numbers of European merchants, engineers and occasional settlers. Their activities concentrated mainly on export production, but food production for domestic consumption also became a policy concern as populations expanded (especially in parts of Asia) or to supply major mining areas (especially in parts of Central and Southern Africa). The gradual and radical shifts in the global division of labour associated with ‘modern imperialism’ also provoked new flows of migrants, capital and commodities within and across both continents. It is in this context of a changing world order, characterized by sharply increasing flows of cross-continental trade and investment capital, that colonial administrations attempted to strengthen and consolidate the fiscal capacity of the colonial state.
Fiscal Capacity Building in a Colonial Context
In most of the colonies of Africa and Asia there was little scope for an ‘organic’ evolution of government structures and fiscal systems, as had occurred in the metropolitan countries over a span of centuries. Until the early decades of the nineteenth century, European colonial control over Asia and Africa had been largely confined to coastal and island settlements, which often functioned as little more than trading posts. Officials charged with running these settlements had neither the resources nor the need to develop new fiscal instruments.
With the expansion of imperialism in the course of the nineteenth century, the costs of maintaining colonies rose. The reluctance of home governments to place much of the financial burden of imperial expansion on the backs of metropolitan taxpayers pressed colonial governments to expand their revenue base as quickly as possible and become fiscally self-supporting. In fact, in several cases they were asked to make contributions to the metropolitan economy, and to this end new fiscal duties were often imposed upon the local population. Colonial administrators could rely to varying extents on revenue institutions that were already in place, but these were often imperfectly understood, and local elites had good reasons for hiding the extent to which they were taxing their people. Therefore, it was not unusual for local people to be confronted with additional revenue demands from colonial officials while at the same time continuing to pay pre-colonial imposts to local rulers. This was especially the case when colonial governments adopted ‘indirect rule’ and left local power structures largely in place.
In the European context, scholars have analysed the rise of the ‘fiscal-military’ state, where the ability of countries to wage war was predicated on growing flows of both tax and non-tax revenues. These imposts were often accepted by the population as temporary demands to meet legitimate defence needs, although in many cases the new taxes became permanent (Tilly Reference Tilly1990, Bonney Reference Bonney1999, Hoffman Reference Hoffman2015). In most Asian and African colonies, external threats were largely eliminated by the late nineteenth century, and the revenues raised by the colonial state were not primarily intended to fight off external powers. The First World War (1914–18) had some fiscal impact on many colonies, especially through the contraction in world trade which affected revenues. Those colonies which provided significant numbers of troops to the metropolitan country, such as India, shouldered most of the costs of their upkeep (Roy, Chapter 3, this volume). Yet, on the whole, colonial revenues were first and foremost needed to secure internal order. Colonial control often required considerable investments in local armies and police forces, as well as roads, railways and ports. Moreover, revenues were needed to pay the salaries of government officials who administered the government departments. Without the trust and commitment involved in the defence of a ‘nation’, raising revenues on behalf of a foreign government, which often had little legitimacy in the eyes of local people, required a more intensive combination of coercion and co-option. In many colonies across Asia and Africa, officials found this combination difficult, if not impossible, to implement.
One of the ideas advanced by the bellicist theory of fiscal development in Europe, is that constant interstate warfare resulted in an imagined ‘social contract’ between citizens and the state. Gradually, this would become an increasingly important marker of ‘national’ identity, solidarity and unity in the face of the violence perpetrated by outsiders. Indeed, nationalism was one of the major social-political forces in Europe both before and during the period of our study (c. 1850–1960). But on whose behalf did colonial governments collect revenue and provide public services? Many subject populations were ethnically heterogeneous, if not fragmented, and did not share a common identity. Moreover, co-optation of specific ethnic groups, social classes or elite factions hampered the idea of a state working in the public interest. The legitimacy problem of the colonial state was aggravated by the importation of labour from India and China into parts of East Africa and Southeast Asia, as well as from Korea into the Japanese colony of Manchuria. This gave rise to economies where small- and medium-scale businesses were owned by people who were neither locals nor from the metropolitan country. To the extent that national identities were sharpened during colonial occupation, it was generally in response to perceived oppression from the colonial power, and to the migrant workers from other parts of the world, who were either regarded as tools of colonial exploitation (e.g. coolies) or as colonial middlemen (e.g. tax farmers, entrepreneurs). Rather than underpinning colonial state formation and fiscal capacity expansion, nationalist ideologies promoted by independence movements thus became a major destabilizing factor of existing political-fiscal arrangements.
Fiscal Extraction and the Costs and Benefits of Empire
The national historiographies which have developed on the colonial experience of particular metropoles, not just in European countries but also in the United States and Japan, reveal great variation in their assessment of the costs and benefits of empire. In Dutch, Belgian and Portuguese historiography the consensus view is that the possessions in Asia and Africa were, on the whole, beneficial to the development of the metropolitan economy.Footnote 4 Assessing the contribution of empire to Portuguese economic growth in the early-modern era, Costa, Palma and Reis (Reference Costa, Palma and Reis2015, 1) concluded that “eliminating the economic links to empire would have reduced Portugal’s per capita income by at least a fifth”. For the modern era, Lains (Reference Lains1998) has argued that the net benefits of Portuguese Africa to the Portuguese economy between 1885 and 1975 were also substantial.
The net contribution of the Netherlands Indies to the Dutch treasury was also considerable. At the peak of the cultuurstelsel (cultivation system) in the mid-nineteenth century, it has been estimated at 52 per cent of the total metropolitan budget (Fasseur Reference Fasseur1991, van Zanden & van Riel Reference Van Zanden and van Riel2000, 223). The Dutch experiment (c. 1830–60) inspired the Belgian king Leopold II to set up a similar system in the Congo Free State (Frankema & Buelens Reference Frankema, Buelens, Frankema and Buelens2013). Although the rates of extraction were less clear after 1870 (van der Eng Reference Van der Eng1998, Gordon Reference Gordon2010), private sector remittances remained substantial, especially from the plantation and mining sectors in both Indonesia and the Congo (Booth Reference Booth1998, 210–14; Buelens & Frankema Reference Buelens and Frankema2016). Scholars have pointed to the exceptional profitability of Belgian investments in Congolese mining and plantation sectors during the 1920s to 1950s (Buelens & Marysse Reference Buelens and Marysse2009, Vanthemsche Reference Vanthemsche2010).
The debates in Britain and France have been less conclusive and often more politicized (e.g. Gann & Duignan Reference Gann and Duignan1967, Ferguson Reference Ferguson2002, Lefeuvre Reference Lefeuvre2006). Davis and Huttenback (Reference Davis and Huttenback1988) argued that empire-building not only brought about a redistribution of resources between the United Kingdom and its colonial dependencies, but also transferred resources within British society from the general British taxpayer to a select class of merchants, entrepreneurs and investors who greatly benefitted from the ‘free’ naval and military protection of their overseas activities (O’Brien & Prados de la Escosura Reference O’Brien and Prados de la Escosura1998, Offer Reference Offer1993, Edelstein Reference Edelstein, Floud and McCloskey1994, 213–14). More recently, scholars have explored the extent to which British colonies benefitted from access to metropolitan capital markets and favourable conditions for government loans (Ferguson & Schularick Reference Ferguson and Schularick2006, Gardner Reference Gardner2017).
Marseille (Reference Marseille1984) and Lefeuvre (Reference Lefeuvre2006) have argued that French colonies received substantial net inflows of funds to the detriment of French taxpayers, especially in the decades after 1930. Marseille (Reference Marseille1984) has also pointed out that in the interwar years in particular, uncompetitive traditional industries were sustained by captive colonial markets after the loss of European markets, with adverse effects on the competitiveness of French industries. Following earlier incomplete attempts by Bobrie (Reference Bobrie1976), Coquery-Vidrovitch (Reference Coquery-Vidrovitch1982) and Marseille (Reference Marseille1984) to calculate the net transfers between France and its colonies, Huillery (Reference Huillery2014, 1) estimated that the net payments to French West Africa (the AOF) were negligible, on average 0.29 per cent of total metropolitan public expenses, while the burden on African subjects was high. This view finds support in the work by van Waijenburg (Reference Van Waijenburg2018), who has shown that the implicit tax burden of forced labour in French West Africa was extremely severe, particularly up to the 1920s.Footnote 5
The Japanese case is also important, not least because many students of comparative colonialism, following the influential work of Myers and Peattie (Reference Myers and Peattie1984), have tended to accept that Japanese policies in Taiwan, Korea and Manchuria were more ‘developmental’ compared with the extractive or exploitative policies pursued by the European powers. In fact the evidence is mixed. In Taiwan, the Japanese government transferred funds to the colonial budget in the early years of colonial occupation, but, after 1909, surpluses from the colony were used to fund transfers on both government and private account to metropolitan Japan (Booth & Deng Reference Booth and Deng2017, 94). But both Korea and Manchuria maintained large balance of payments deficits, funded by inflows from Japan. An American study of the Japanese economy came to the conclusion that, at the end of the 1920s, from a fiscal point of view the colonies were a liability rather than an asset (Moulton Reference Moulton1931, 180). Large capital inflows to both Manchuria and Korea persisted until the 1940s.
Fiscal Development and Economic Growth
The debates on the costs and benefits of empire for European metropoles are closely connected to the debates on the (long-term) consequences of colonial extraction for colonial societies. An influential strand of literature stresses the intimate link between colonial repression, fiscal extraction and institutions designed to exploit colonial populations (Jamal Reference Jamal1978, Young Reference Young1994, Acemoglu, Johnson & Robinson Reference Acemoglu, Johnson and Robinson2001). Other studies, especially on sub-Saharan Africa, have argued that colonial state budgets were small and, if anything, led to understaffed bureaucracies and underinvestment in public services rather than high tax burdens (Kirk-Greene Reference Kirk-Greene1980, Frankema Reference Frankema2011, Gardner Reference Gardner2012). In the context of Asian colonies, the impact of the cultivation system on living standards in Java and the role of the colonial state in the Indian famines of the late nineteenth century have been extensively debated (see for Java: Elson Reference Elson1984, chapter 11; see for India: Roy Reference Roy2012).
The scanty evidence we have suggests that colonial economies did grow over extended periods of time, mostly in the order of an annual 0.5 to 1.0 per cent per capita (see for Africa 1870–1950: Prados de la Escosura Reference Prados de la Escosura2012; see for Asia 1870–1950: Maddison Reference Maddison2010). This growth was largely driven by the expansion of export sectors and related infrastructural investment. Rapid growth was realized in times of mineral or cash-crop export booms, but these usually tapered off after one or two decades at best. Structural change and economic diversification, and in particular the development of competitive manufacturing industries, remained weak. The biggest exception among the cases studied in this book is South Africa, where greater political autonomy after 1910 facilitated the adoption of import substitution policies in the industrial sector from 1924. Along with systematic labour repression, these policies resulted in a manufacturing share of 20 per cent in GDP by 1960, in addition to 13 per cent for mining (Feinstein Reference Feinstein2005, 144, Austin, Frankema & Jerven Reference Austin, Frankema, Jerven, O’Rourke and Williamson2017, 356). Yet, pervasive racial discrimination meant that little of the growth translated into improvements in living standards for the indigenous African majority (Feinstein Reference Feinstein2005, 67–71).
To what extent has fiscal development stimulated or limited colonial economic growth? In the absence of a historical counterfactual it is hard to formulate even a tentative answer to this question. The few countries that, at least in name, remained independent (e.g. Liberia, Ethiopia, Thailand) did not fare any better than their colonial neighbours, but it remains questionable to what extent this tells us anything about the relationship between fiscal development and economic growth. Some have argued that the access of colonial states to metropolitan capital markets has enhanced their possibilities to borrow and invest cheaply, in contrast to poor countries that remained independent (Ferguson & Schularick Reference Ferguson and Schularick2006; see for a critique: Accominotti et al. Reference Accominotti, Flandreau, Rezzik and Zumer2010). However, colonial powers in general tended to be rather conservative in the supply of colonial loans and could have done much more if they had wanted to. Only after 1945, were substantial grants-in-aid provided for investment in colonial development programmes. Moreover, in many African colonies, marketing boards were syphoning off export sector surpluses that remained unutilized for capital investments or expansion of local banking.
Investments in health and education received low priority, although there was considerable variation across colonies and metropolitan powers. For instance, the United States channelled much more resources into the education system of the Philippines than the Dutch did in Indonesia under the so-called ‘ethical policy’ (Frankema 2013). Whereas school enrolment rates in British Africa rose faster than in French Africa before 1940, the British relied much more on private funds supplied by missionary organizations to finance mass education, while the French emphasized the role of public educational investment by the colonial government (Frankema Reference Frankema2012, Cogneau & Moradi Reference Cogneau and Moradi2014).
Any evaluation of the relationship between fiscal development and colonial economic growth will have to consider the transition that colonial states made from a ‘night-watchman state’ with minimal resources towards a more ‘developmental state’ (Booth Reference Booth2007b, 67–8), with small but growing investments in health, education and infrastructure. This transition was supported by increasing opportunities for borrowing in metropolitan capital markets, increasing grants in aid, and some degree of tax diversification towards ‘modern’ personal and corporate income taxes and general sales taxes. All these developments reflected the ambition to ‘modernize’ the colonial state, but the effect of these reforms has remained limited in the majority of cases.
This brings us to one of the key insights of this book: colonial economies and societies were difficult to engineer and the solution to the revenue problem thus varied enormously both across and within empires. Since local economic, political and social conditions were crucial in shaping the opportunities and constraints to fiscal capacity building, the economic legacy of the colonial ‘fiscal state’ may be best considered as unique to each colony and often even each region. Specific blends of political, legal and commercial institutions evolved which, in most cases, required significant modification to support a process of ‘modern’ economic growth.
Fiscal Capacity and the Colonial State: a Theoretical Vacuum
We now return to the issue of fiscal capacity, and its relation to state capacity. The existing literature accepts that state capacity includes fiscal capacity, but they are not synonymous concepts. State capacity also involves the ability to protect subjects of the state against the threat of internal or external violence, expropriation and other offences; to effectively implement government policies; and to run a state bureaucracy. State capacity also includes the concept of ‘legal capacity’, which Besley and Persson (Reference Besley and Persson2009) define as the capacity of the state to support markets with appropriate institutions. In this book, we use the term ‘colonial state’ to refer to a governance system imposed on indigenous people in a distinct territory by a foreign power for a prolonged period of time, whereby the central government ultimately relies on the military, administrative and technological backup of the metropole to secure internal order.
The colonial state thus defined made use of foreign agents to run key positions in the administration, the army and, in most cases, the major export sectors of the local economy. Almost all colonial states analysed in this book were states where local or indigenous inhabitants comprised a large majority of the population; the main exception was British Malaya, where large-scale in-migration from China and India had resulted into a minority position for the indigenous population by the 1930s. Native populations were at least partially integrated into the colonial fiscal system in all colonies. This distinguished them from the much smaller number of settler colonies where indigenous populations remained largely outside the fiscal system, or were even pushed entirely beyond the settler’s land frontier.
It was argued earlier that economic and political theories aiming to explain the dynamic relations between fiscal capacity and state capacity have been mainly inspired by the European experience (Tilly, Reference Tilly1990, Besley & Persson Reference Besley and Persson2009, Reference Besley and Persson2010) and, to a lesser extent, by the cases of China and Japan (He Reference He2013). This leaves an important gap in the literature. The theoretical models of Besley and Persson emphasize the complementary nature of fiscal capacity and legal capacity. Their models reveal, in line with Tilly’s argument, that the positive feedback loops between fiscal capacity and state support for markets are enhanced by war threats, and constrained by the presence of natural resources, which erode incentives to invest in fiscal capacity and market development.
The concept of ‘fiscal modernization’ offers another framework with which to evaluate specific stages in colonial fiscal development. In his comparative study of Britain, Japan and China, He (Reference He2013, 3–13) defined ‘fiscal modernity’ in terms of several criteria. The first relates to the capacity of the state to collect revenues at a local level and to transfer these into a central consolidated fund. According to He, revenue centralization requires a threshold level of commercialization and monetization to levy sufficient indirect taxes on both foreign trade (import and export taxes) and domestic transactions (excises), which can be remitted to the centre using bills of exchange or other methods. The second condition is that the central state uses its revenue generating capacity to establish a long-term floating debt position. In the military-fiscal states of early-modern Europe, governments could borrow to finance exceptional expenses such as warfare. However, after military campaigns had ended, states often redeemed their debts in order to start the next war with a blank sheet. The transition towards structural floating debt is thus a key feature of fiscal modernity (Dincecco Reference Dincecco2009, Reference Dincecco2011).
A third condition relates to the expenditure side of the budget. In order to justify taxes and other imposts, states had to be seen to be delivering goods and services which protected, and indeed enhanced, the welfare of tax-payers. The idea that the state, instead of churches or other charitable institutions, had a responsibility to look after the welfare of the general populace gained ground in parts of Europe and North America during the nineteenth century. This idea was tightly connected with a fourth condition, fiscal modernity, or the transition towards responsible and representative government. Improved accountability and extension of the franchise in Europe and the western offshoots had considerable implications for fiscal policy, as negotiations over the distribution of the tax burden and the provision of public goods became more open. These negotiations were conducted by representatives of most social groups, and went hand in hand with the rise of political parties representing class interests. The push for responsible state government in the British North American colonies developed, in part at least, as a result of the resentment against British refusal to delegate taxation powers to the local population, and was expressed in the famous slogan “no taxation without representation”.
These theoretical and conceptual approaches have clear limitations for understanding the varying patterns of fiscal capacity building under colonial rule. One key reason was that the feedback loops between fiscal capacity and legal capacity, as well as the balance between tax legitimacy and tax enforcement were fundamentally different in a colonial setting. In fact, most colonial fiscal systems found it difficult to progress beyond the first criterion of fiscal modernity as defined by He (Reference He2013). The borrowing capacity of the colonial state, for instance, did not depend primarily on the risk assessment of private investors, but rather on the political decisions taken by metropolitan officials who would weigh the value of colonial development projects – mainly infrastructure – against the risk of default (Sunderland Reference Sunderland2007).
Although by the early twentieth century, the European imperial powers were justifying their control over much of Asia and Africa in terms of a ‘civilizing’ mission, they did not define this in terms of establishing ‘responsible government’, let alone a welfare state. The lack of tax legitimacy was usually compensated by higher degrees of coercion, often exercised through intervention in land and labour markets. Even though commercialization was a priority for all colonial governments, a positive feedback loop between fiscal capacity and legal capacity was not self-evident. Factor markets had to be controlled in order to push the price for labour, capital and land below what a free market would have established. In addition, export sectors had to be taxed in order to syphon off part of the income from international trade, including mineral exploitation. The presence of natural resources did not always crowd out other taxes either, because direct taxes or coercion were often needed to make labour available for the exploitation of mines or plantations. Public expenditures were often not intended to strengthen domestic economic linkages, but rather to open up specific production and consumption opportunities to metropolitan investors, producers and consumers. Fiscal capacity was thus stimulated by, and in turn used to strengthen, monopolies and monopsonies.
A complicating factor was that the great majority of colonial populations were poor, and could not reasonably be expected to pay more than a small fraction of their meagre incomes to the government. Even before the problems brought about by the world depression in the 1930s, colonial governments faced the threat of tax revolts if governments made excessive demands. Colonial governments without effective institutions of representative government could limit the risk of revolt by pursuing various strategies, including outright violence to enforce tax compliance. But this required a credible commitment on the part of the state to use force, and in the longer run could make the problem of legitimacy worse. In most cases, the compromise reached was to limit the range of taxes imposed, and their amount. This minimalist strategy involved the avoidance of direct taxes, if possible, and a strong emphasis on less visible indirect taxes.Footnote 6
Colonial states also critically depended on indigenous rulers to enforce taxes in the expectation, not always realized, that these rulers had greater legitimacy in the eyes of local people. This meant that part of the revenues had to be shared with indigenous intermediaries such as sultans, chiefs, local warlords, or village heads, many of whom had good reason to frustrate attempts at centralizing revenues. But in many indirectly ruled territories, colonial revenue officials had no alternative. Since taxation with representation was by definition impossible, the legitimacy deficit could only be overcome by force, by reliance on indigenous elites acting as intermediates in the process of revenue centralization or by keeping taxes low.
The growing difference between metropolitan and colonial approaches to revenue mobilization had important implications for the composition of revenues. In particular, the colonial state permitted the use of policies to extract resources that might have been used in the metropolitan country in earlier times, but had fallen into disuse, or had even been declared illegal under the modern fiscal state. For example, colonial governments in parts of Asia continued to rely on revenues from the sale of opium, long after the sale and use of the opium derivatives had been either banned or strictly controlled in the metropolitan countries. This ‘narco-colonialism’, to use the phrase of Bayly and Harper (Reference Bayly and Harper2004, 33), sustained the growth not just of British colonialism, but also French and Dutch activities in Asia.
An even more striking example was the large-scale use of forced labour. Although colonial occupation in many parts of Africa was justified by the suppression of the slave trade and slavery, the use of forced labour was often defended, not just in Africa but in Asia as well, as part of the overarching civilizing mission. Forced labour, it was argued, would improve labour discipline among ‘lazy natives’. In fact, it was often considered the only policy tool available to stimulate commercialization and economic development among indigenous populations who were reluctant to abandon their traditional, subsistence-oriented way of life for the uncertainties and indignities of wage labour (e.g. see for French West Africa: van Waijenburg Reference Van Waijenburg2018; see for the Netherlands Indies: Bosma Reference Bosma and Stanziani2013). Another example is large-scale alienation of land, which was then leased or sold to foreign investors, made available for railway and road construction or just appropriated in order to push more indigenous people into the wage labour market.
This is not to say that constraints to fiscal expansion paralysed colonial governments. The emancipatory forces that were unleashed by the deployment of African and Asian soldiers in the First and Second World Wars forced many colonial states to revise their expenditure priorities. Even though colonial soldiers were selectively recruited, with a focus on ethnic groups that were both trusted and supposed to have martial qualities, such as the Sikhs or the tirailleurs Sénégalais, colonial administrators felt the need to strengthen the legitimacy of the colonial state. As this need was translated into more explicit development agendas, some restrictions on the use of forced labour were introduced.
These restrictions were supported by the new international organizations that emerged in the early twentieth century, especially the International Labour Organisation, which was founded in 1919. All European colonial powers became members, although some, such as France and Portugal, were more reluctant to abolish forced labour than others, such as Britain (van Waijenburg Reference Van Waijenburg2018). Increasingly, British colonies in both Africa and Southeast Asia dealt with the problem of labour shortages by encouraging in-migration from the Indian sub-continent and China. The Dutch were reluctant to allow in-migration but encouraged the Javanese to move to the less densely settled islands outside Java.Footnote 7 But they continued to impose labour obligations (heerendiensten) on indigenous populations outside Java until 1942.
The greater emphasis on development expenditures, to the extent that they occurred, eventually also placed a time bomb under the colonial state-building project. Educated indigenous elites, some of whom had studied at universities in the metropole, started to play an important role in the rise of various independence movements in Asia and Africa. There was no way that the call for emancipation and representation could be accommodated by administrations whose survival was predicated on restricting access to political as well as economic markets in order to safeguard the interests of European, American and Japanese settlers, planters, merchants and enterprises.
Lessons from a Comparative Approach
Globalization and Colonial Fiscal Development
What then, are the lessons that can be learned from the comparative approach taken in this book? First of all, the comparative evidence indicates that the opportunities to engage in international or imperial trade were the single most important determinant of the cross-colony variation in budget size. Figure 1.1 shows the correlation between per capita export revenues (y-axis) and per capita government revenues (x-axis) in 1911 (a census year in the British empire) on a logarithmic scale. This relationship remained strong throughout the colonial era. In most colonies, a large part of the revenues from international trade came from custom duties, and especially import duties. But there were also a range of other taxes that co-evolved with international trade. These included state revenues from shares in export companies (e.g. mining firms), corporate and individual income taxes, harbour duties, excises and export taxes. Government monopolies on items such as salt, alcohol and opium were also important in some colonies, as these were often viewed as a way of taxing Asian migrant workers, including Indians and Chinese, who were otherwise considered hard to reach. In Africa, the revenues from state marketing boards also became significant, especially from the 1930s onwards.
Figure 1.1 Gross public revenue per capita (x-axis) versus export revenue per capita (y-axis), c. 1911 (current £)
Figure 1.1 also shows that there was no sharp distinction between the per capita export revenues generated in African and Asian colonies. In both regions, the development of exports showed considerable variation. Booth (Chapter 2, this volume) argues that colonial governments in the Philippines, Indonesia and British Malaya boosted tropical export crops such as sugar, coffee, tea, fruits, rubber, palm oil and tobacco. They did this through facilitating the establishment of large estates, usually owned by interests in the metropolitan country, although in the Philippines local investors were also involved. In addition, smallholder cultivators of export crops were important in Indonesia, Thailand, Burma and South Vietnam,
Roy (Chapter 3, this volume) argues that the problem of fiscal inertia in British India (i.e. a persistently small tax base in per capita terms) was partly due to a reluctance to squeeze the agricultural sector too hard, and partly due to a lack of alternatives, including limited per capita involvement in international trade. Most colonies in sub-Saharan Africa and Southeast Asia experienced an impressive terms of trade boom during the nineteenth century (Williamson Reference Williamson2011, Frankema et al. Reference Frankema, Williamson and Woltjer2018), but British India did not experience a similar windfall. On the other hand, the early abolition of taxes and other restrictions on domestic trade led to the establishment of a vast customs union across the Indian sub-continent, in which the main urban economic zones were connected by railroads. In other words, even with a small central budget and a limited basis for indirect taxation, domestic markets became significantly better integrated under colonial rule (Studer 2008, Bogart & Chaudary Reference Bogart and Chaudhary2013).
Frankema and van Waijenburg (Chapter 6, this volume) argue that the coastal colonies in West Africa enjoyed a notable advantage over inland colonies, and that this locational advantage led to different fiscal policies in British and French colonies. The federal structure adopted in French West Africa functioned as a vehicle to integrate areas that were by themselves fiscally unsustainable. Per capita exports in French West Africa remained low from 1913 to 1938 in comparison with Ghana, and most parts of colonial Asia, including the Federated Malay States, which were part of the larger federation of British Malaya (Booth Reference Booth2008, table 3). The French also created a fiscal federation in Indochina, as López Jerez (Chapter 4, this volume) explains, although here too, export revenues per capita were low in comparison with other Asian colonies.
Gardner (Chapter 7, this volume) also indicates that in the settler colonies of East Africa, lower volumes of trade in the early colonial era put a limit to fiscal expansion and supported the introduction of direct taxes, which were both more costly to levy and more directly felt by local tax-payers. Differences in export development also played a role in Portuguese Africa. Alexopoulou (Chapter 8, this volume) shows that Mozambique had to rely more on direct taxes, while the Angolan government relied more on trade taxes, including those derived from lucrative diamond exports. Gwaindepi and Siebrits (Chapter 9, this volume) show how the mineral discoveries in South Africa created ample opportunities for fiscal expansion, which they see reflected in high railway investments and which also encouraged heavy state intervention in the labour market.
The disadvantage of dependence on trade taxes was the vulnerability of colonial budgets to world market volatility. The great depression of the 1930s hit the budgets of the main commodity exporters particularly hard and resulted in austerity measures in colonies as diverse as Ghana, Indonesia, British Malaya and Angola. The terms of trade of most tropical commodities underwent a prolonged decline during the first four decades of the twentieth century. Falling relative prices were compensated by expanding production volumes or by the introduction of new commodities, such as the rubber and palm oil in British Malaya and Indonesia (Booth Reference Booth2007a, 56–60, Buelens & Frankema Reference Buelens and Frankema2016). But even in these colonies, the depression years certainly imposed hardship on many people who lost access to wage employment, and had to accept lower prices for most export commodities.
Local Conditions versus Metropolitan Visions
This volume also offers support for the view that local economic, social and political conditions were more important in the design of colonial fiscal systems than metropolitan blueprints for fiscal governance (Frankema & van Waijenburg Reference Frankema and van Waijenburg2014). Apart from the different opportunities for the creation or expansion of export markets, some of which developed spontaneously, while others were enforced, the nature of pre-colonial tax systems also mattered a lot. In some places, such as Equatorial Africa, there was no system available that could be adopted by the colonial government. Squeezing revenue out of the local population through forced labour seemed to be the only option.
In British India, on the contrary, the Raj was very reluctant to introduce new taxes and modified the existing system of land taxes as a financial basis for the colonial state (Roy, Chapter 3, this volume). This system proved inelastic. In French Indochina, the French decision to impose indirect rule in most regions meant that existing power structures remained in place right down to the village level. Local elites had little incentive to alter existing revenue systems and the French lacked the capacity to force them to change. New revenues imposed by the French thus added to existing tax obligations for most rural people (López Jerez, Chapter 4, this volume).
That metropolitan visions of fiscal organization, whatever they might have been, were of limited influence, is obvious when we look at revenue outcomes across colonies in Asia and Africa in the early twentieth century (Figure 1.1). Revenue systems adopted within particular empires varied considerably. Alexopoulou (Chapter 8, this volume) shows that even within Mozambique there were three distinct revenue systems geared towards local economic and political structures, with limited opportunity to impose a uniform system. If we examine the much larger French and British empires, the variation is even more striking. Revenues range from land and opium taxes in India and parts of Southeast Asia to ‘plural wives’ taxes in Tanzania, from cattle taxes in French West Africa to rice export taxes in Indochina. Local economies and polities thus shaped both the opportunities and the constraints to fiscal capacity building.
This does not mean, however, that metropolitan identity did not matter at all. The different preferences of the metropolitan powers for direct or indirect rule, for the use of forced labour and the availability of investment capital from the metropolitan economy or elsewhere all had a bearing on different practices of revenue generation. In addition, the fiscal systems of the home governments undoubtedly had some impact on colonial practice. As López Jerez (Chapter 4, this volume) points out, the French only began to implement an income tax in France in the late 1930s; before that they depended very largely on indirect taxes and non-tax revenues. In Indonesia, income taxes assessed on both individual and corporate incomes grew in importance in the last decades of colonial rule, as they did in British India (Kumar Reference Kumar, Kumar and Desai1983, table 12.7). To a considerable extent, this reflects British and Dutch practice at home, although it should be noted that in British Malaya, income taxes were not imposed until after 1945.
Metropolitan visions of colonial development were expressed more clearly in the expenditure side of the budget. For instance, the contrast in education spending between the Philippines and Indonesia reveals a very different view of the long-term aims of colonial rule (Booth, Chapter 2, this volume). For the Americans, education was the principle means of preparing the indigenous population for self-government, and eventual independence. For the Dutch, who considered their colonial role in the Indies as a permanent condition, the drive to increase access to education was much more limited.
As independence movements developed in India, Indonesia and Vietnam in the early twentieth century, British, French and Dutch colonial governments feared that too much access to education in the language of the colonial power would create unrealistic expectations about employment, which would in turn lead to more demands for self-government if not complete independence. Differences in metropolitan visions were also clear in the attitudes towards Christian missionaries and their role in the provision of ‘public’ services such as health and education (Cogneau & Moradi Reference Cogneau and Moradi2014, Frankema Reference Frankema2012).
As Booth and Deng (Chapter 5, this volume) demonstrate, Japanese policies in Korea and Taiwan were predicated on the idea of forcible assimilation, and their spending patterns prioritized this goal. This was reflected especially in programmes of railway construction, and in the large-scale private investments in both colonies, as well as in Manchuria in the 1930s. Access to primary education was also expanded, although there were few opportunities for young people in the colonies to get access to post-primary education. The figures on numbers enrolled in education as a percentage of the total population varied considerably across colonial Asia in the late 1930s. The highest figure was in the Philippines, followed by Taiwan and Thailand where many young people were enrolled in monastic schools. This was also true for Burma, where the figure was about the same as for Korea; Indonesia and Indochina were lower, although still higher than in Ghana or Nigeria (Booth Reference Booth2008, table 12).
Of course, all expenditure decisions were taken within the context of tight budgets, but the relative shares of the budget that went into the security forces, both soldiers and police, varied enormously. Roy (Chapter 3, this volume) shows how the Indian army ate away a big chunk from already limited budgets, not only to maintain order in India, but also to secure it in other parts of the empire. Alexopoulou (Chapter 8, this volume) also indicates that the size of the Portuguese African army was comparatively large, and that security expenses remained very high up to the 1930s. Other colonies could rely much more on external assistance in case of need, from, for example, the British Navy, or imperial recruits, thus supressing the costs of a large permanent local army.
Alexopoulou and Frankema (Reference Alexopoulou and Frankema2018) have argued elsewhere that such resource pooling mechanisms were easier to organize in the British Empire than in the empires of the weaker metropolitan states, such as Portugal or the Netherlands. Indonesia, along with Manchuria and independent Thailand allocated over 30 per cent of budgetary outlays to law and order and defence in the 1930s, compared with only 8 per cent in the Philippines and 7 per cent in Taiwan (Booth and Deng, Table 5.8, this volume). Defence expenditures in both the Federated Malay States and Ghana were even lower, at less than 3 per cent (Booth Reference Booth2008, table 6). In the Philippines, there was a small military presence which was paid for from the American budget, while 36 per cent of Philippine budgetary outlays went to health and education compared with 12 per cent in Indonesia and only 4 per cent in French Indochina.Footnote 8 In Portuguese Africa, where the government spent much larger amounts on the security forces compared with the more advanced parts of British Africa, the budget available for welfare spending remained very small (Alexopoulou & Frankema Reference Alexopoulou and Frankema2018; Gardner, Table 7.3, this volume).
Africa versus Asia: Distinct Patterns?
Were there distinctive ‘Asian’ and ‘African’ patterns of colonial fiscal development? It appears from Figure 1.2 that the average Asian colony had a somewhat larger budget in 1911, but the variation in budgets and in population size was too large to make strong inferences on the basis of these estimates. In terms of spending capacity, the variation within both regions appears at least as large as the cross-continental differences. It seems that individual colonies followed distinct patterns of fiscal development, some of it based on mining (Zambia, South Africa), others on tropical export crops (Uganda, Ghana, Nigeria, French Indochina) or a combination of both (Indonesia, Malaysia, Belgian Congo, Angola). There was a group of landlocked African colonies where revenues remained small because of limited opportunities (e.g. Niger, Chad, Mali) and two large Asian colonies (British India, French Indochina), where central governments found it difficult to centralize fiscal revenues that were well established at local levels.
Figure 1.2 Gross public revenue per capita in 1911 and 1938 (in current £)
Three generalizations may be made, notwithstanding the exceptions that are present in all cases. First, colonial states in Asia tended to have had a longer experience of colonial governance, at least in their core areas. With the exceptions of French Indochina, the Japanese colonies and the Cape Colony (South Africa), Asian colonial governments had more time to integrate, impose, expand or consolidate fiscal policies than colonial governments in sub-Saharan Africa. Second, tax systems in Asia were, on the whole, more deeply rooted in local agricultural systems, and were underpinned by more fine-grained systems of tax assessment (income, wealth, especially land and consumption). Put differently, by the early twentieth century, Asian colonial states had a better capacity to collect information-intensive taxes. A third generalization relates to the greater emphasis colonial governments in Asia placed on agricultural development, and on developing infrastructure, especially irrigation systems, to boost agricultural output.
The development of land taxes illustrates the second of these points. By the early twentieth century, land taxes were important in several parts of colonial Asia, but remained virtually absent in colonial Africa. Historical differences in the spread of sedentary agriculture probably played an important role in explaining this difference. Nomadic pastoralism was more deeply rooted in various tribal societies living in Africa’s extensive savannah belts, large parts of which were too dry for crop cultivation. Farmers were probably more mobile in forested parts of sub-Saharan Africa than in the rice producing regions in Asia, where farmers had adapted to (or were earlier coerced into) more sedentary lifestyles. Taxing shifting cultivators at the central level was very difficult as Frankema and van Waijenburg (Chapter 6, this volume) argue. The absence of land taxes in most parts of Africa and their importance across many parts of colonial Asia, also implied differences in pre-colonial administrative structures.
In African states, the role of slavery in fiscal systems was larger, while feudal relations were more developed in parts of Asia. But even within Asia, there were considerable differences in the extent to which governments assessed land taxes, and their importance in local and central revenue systems. Some colonial authorities viewed accurate land cadastres as essential not just for assessing taxes but also for establishing property rights, and estimating output of key staples such as rice.
Given these benefits, some governments were prepared to invest considerable sums in drawing up and maintaining accurate land records, but in other regions, land cadastres either did not exist, or were considered very inaccurate. Even within one colonial jurisdiction, there were often wide variations. The Dutch had very accurate records for Java and Bali, but in most other islands, little attempt was made to develop accurate land records. Within British India, land taxes formed a much higher proportion of agricultural output in Burma than elsewhere in India, or indeed in most other Asian colonies (Booth Reference Booth2014, table 2).
Taxes on domestic consumption at the retail level were another indicator of a more fine-grained system of tax assessment. Even though a general sales tax, let alone a value-added tax, was not implemented anywhere in colonial Asia or Africa, there were a variety of excises and consumption taxes in Asia, and their contribution to the budget was, in many cases, substantial. These consumption taxes were often first introduced as revenue farms, auctioned by the central government, but later integrated into central collection systems (Wahid Reference Wahid2013). Consumption taxes also existed in African colonies, but not on a similar scale. In African colonies a combination of custom duties, poll, head or hut taxes and railway revenues tended to dominate the revenue mix.
A longer legacy of revenue centralization also had implications for state investments in administration, infrastructure and other public services. Again, one has to be careful not to overstate the Asia–Africa differences. The wealthiest colonies in Africa did perform better than the poorest colonies in Asia. But if we look at the four indicators listed in Table 1.1, it is difficult to escape the conclusion that the Asian colonies included in this book did have higher railway densities, higher per capita output of electricity (except for the major mining areas in Africa), more currency in circulation per head of the population, a significant lead in educational attainment and higher literacy rates. At the same time, Table 1.1 illustrates the variation within both regions, and especially the impressively rapid development of infrastructure in the Japanese colonies of Korea and Taiwan (Booth & Deng, Chapter 5, this volume) and the distinct development path of South Africa (Gwaindepi & Siebrits, Chapter 9, this volume).
Table 1.1 Comparative indicators of economic development in Africa and Asia, c. 1950
| Railway density 1950 | Electricity output 1950 | Currency in circulation 1938 | Years of school attainment 1950 | |
|---|---|---|---|---|
| Km/1000 Km2 | kwh per capita | £ per capita | population 15+ | |
| British West Africa | 3.6 | 7.8 | 0.44 | 0.6* |
| French West Africa | 2.1 | 2.0 | 0.27 | 0.6 |
| British East & Central Africa | 1.9 | 7.6 | – | 1.1 |
| Portuguese Africa | 2.5 | 6.0 | 0.25 | 0.5** |
| South Africa | 17 | 876 | 1.61 | 4.2 |
| India | 17 | 14 | 0.39 | 1.0 |
| Indochina | 1.9 | 6.2 | – | 2.0 |
| Indonesia | 11 | 8.4 | 0.26*** | 1.1 |
| Philippines | 3.8 | 25 | 0.98 | 2.8 |
| Federated Malayan states | 25 | 93 | 1.32 | 2.1 |
| Thailand | 6.1 | 3.4 | 0.71 | 3.4 |
| South Korea | 28 | 20 | 0.90 | 4.5 |
| Taiwan | 26 | 172 | 1.09 | 4.3 |
Sources: Railway length, electricity output and currency from Mitchell (Reference Mitchell2007; tables F1, D24, G1). Attainment estimates from Barro and Lee (2013). Population data for African colonies from Frankema and Jerven (Reference Frankema and Jerven2014), for Asian colonies from Maddison (Reference Maddison2010). * Population weighted average of Gambia, Sierra Leone and the Gold Coast, excluding Nigeria. ** Mozambique only, no data for Angola. *** The Indonesian estimate probably excludes paper currency and is therefore too low.
The third generalization relates to investment in agricultural development. The evidence, although not conclusive, does suggest that government expenditure on agricultural development in Asia was not just higher, but also more focused on food crop cultivation, especially in those areas that were densely settled and where colonial officials were worried about food security. The Dutch had invested heavily in developing irrigation systems and transport networks in Java from the late nineteenth century to 1930, and were turning their attention to other islands when the global depression forced them to curtail expenditures. The French invested heavily in irrigation and drainage projects in both the Red River and the Mekong deltas (López Jerez, Chapter 4, this volume). Although there were some attempts to develop irrigation systems in West Africa, they appear to have been poorly planned, and were not successful in raising food output. In view of the very different paths of rural transformation in Asia and Sub-Saharan Africa after 1950, deeper comparisons of the role of the colonial state in agricultural development seem to be warranted.
Fiscal ‘Modernization’ under Colonial Rule: a Contradiction in Terms?
In what sense can we talk about fiscal modernization under colonial rule? All the chapters in this volume indicate that colonial governments managed, sooner or later, to increasingly centralize revenues. In large parts of Africa, where pre-colonial states were either weakly developed or entirely absent, resources were mainly pooled at communal or tribal levels. In some regions, the colonial state had to build fiscal systems from scratch. State-controlled slave production systems were difficult to incorporate into colonial revenue policies, as were farmers practising shifting cultivation, nomadic pastoralists and other households that remained largely outside the monetary economy. In addition, colonial officials often had to tolerate taxes in labour and kind, because they lacked the ability to stop them, or because they did not want to alienate indigenous leaders.
Centralization of revenues did not necessarily imply diversification. High revenues per capita could be extracted by the state with a limited number of tax and non-tax instruments. The case of British Malaya illustrates this point; in both the Straits Settlements and the Federated Malay States, revenues per capita were high by Asian standards, but mainly derived from excises on opium, tobacco, alcohol and petroleum products, together with non-tax sources (Booth, Chapter 2, this volume). Elsewhere colonial officials had to make a greater effort to diversify revenue sources. In Africa, the need for diversification was especially great in colonies that lacked an economically significant foreign trade sector. In Ghana and Nigeria, where exports per capita had increased quite rapidly between 1900 and 1940, direct taxes and customs duties accounted for a higher proportion of total revenues than in any Asian colony by the late 1930s (Booth Reference Booth2008, table 8). But in other parts of West and East Africa, there was a heavy reliance on forced labour to put the wheels of commerce in motion (van Waijenburg Reference Van Waijenburg2018, Frankema & van Waijenburg, Chapter 6, this volume).
Forced or corvée labour declined in importance in most Asian colonies, and in Thailand after 1900, but it persisted in some parts of Indonesia (Booth, Chapter 2, this volume). The fiscal systems of the major mining economies were relatively straightforward, but also proved vulnerable to world market shocks. The treasuries of British Northern Rhodesia and the Belgian Congo, for instance, became increasingly dependent on copper receipts after 1950, but this led to a dramatic collapse in the 1970s when world copper prices plummeted and both countries fell back to the very bottom of the African national income rankings (Abbeloos Reference Abbeloos, Frankema and Buelens2013).
We should be careful not to push too far the argument that indirect trade taxes crowded out more information-intensive direct taxes. After all, the revenues from trade taxes stimulated investments in bureaucratic capacity, which were in turn a necessary condition for the development of a more sophisticated tax administration. The poorest African colonies may have relied to a larger extent on direct taxes, but the hut, poll or head taxes collected by local chiefs did not reflect greater ‘administrative capacity’. They did not entail a detailed assessment of the incomes accruing from indigenous farms, households or enterprises, let alone their total wealth. Nor did colonial governments necessarily have firm control over the chiefs or village heads who collected direct taxes. Many had good reasons for under-reporting the amounts they were collecting. More in general, tax avoidance remained a structural phenomenon.
He (Reference He2013) has emphasized the importance of commercialization in facilitating the rise of the fiscal state in both Britain and Japan, and it is probable that in many colonial territories the development of a modern financial system was hampered by the fact that most of the population participated in the cash economy only to a limited extent. Outside the main cities, few people had access to banks, and savings were often held in gold, cattle or land. By the early twentieth century, governments were borrowing to fund infrastructure projects but the borrowing was almost always conducted in metropolitan financial markets. Even so, the domestic money supply did grow relative to GDP in some colonies, although greater monetization was restricted to urban dwellers, and those rural producers who marketed all or part of their output (see, for example, on India and Japan: Goldsmith Reference Goldsmith1983; see on Indonesia: Van Laanen Reference Van Laanen, Booth, O’Malley and Weidemann1990).
When we examine the evidence on government borrowing, we again see striking differences between colonies. The Portuguese regime under Salazar after 1933 was extremely reluctant to expand colonial debt (Alexopoulou, Chapter 8, this volume). But this reluctance was not widely shared. British India, Nigeria and South Africa adopted different policies. In India, a large unified currency zone emerged with a greater reliance on internal debt financing, and a rising share of local investors in government bonds (Roy, Chapter 3, this volume). In South Africa, the mineral discoveries set the stage for large investments in railway infrastructure and created ample leverage for debt financing by the colonial government. Moreover, as Gwaindepi and Siebrits demonstrate, the autonomy of South Africa granted in 1910 also freed the way for the creation of a modern fiscal system that was more advanced than in any of the other African or Asian colonies discussed in this volume (Chapter 9, this volume). Indeed, the size of national income, whether driven by large population numbers or abundant mineral resources, expanded the possibilities of colonial debt creation.
In many parts of Asia and Africa by the early twentieth century, we observe major cross-colony differences in expenditures on health and education, but nowhere did they come close to the amounts that were being spent in the metropoles. The Americans did channel a substantial share of the annual budget in the Philippines into education, but elsewhere allocations to health and education were much lower (Booth & Deng Reference Booth and Deng2017, table 4). In the Southeast Asian case, we have the counterfactual of independent Thailand, but while the Thai government did not lag far behind most Asian colonies in terms of development, neither did it forge ahead (Table 1.1). On numerous occasions between 1850 and 1950, the Thai government prioritized defence expenditures over development expenditures on, for example, irrigation or education. The main non-colonized counterfactual in Africa was Ethiopia, which also ran a central budget that was large enough to field a substantial army in its attempt to avoid Italian invasion. At most, we can argue that colonial states in Africa initiated revenue centralization in those regions which did not have a central polity, and that they accelerated welfare spending compared with what might have happened in the absence of colonial control. But this is speculative, as indeed it is in most parts of colonial Asia.
The final criterion of fiscal modernization, the development of accountable government, was by definition impossible under colonial rule. Even though independence movements managed to obtain some political concessions and exercise greater influence in some colonies, fiscal sovereignty was only achieved after decolonization. But nowhere did this transition result in the kinds of constraints on executive power that operated in metropolitan countries such as Britain, France, the United States or the Netherlands. Institutions such as effective audit boards were often slow to develop after independence, and in many former colonies they are still weak, or non-existent. Fiscal sovereignty became more difficult to achieve in the 1980s, as structural adjustment programmes made independent governments subject to conditions imposed by foreign donors and international organizations such as the IMF. In the early twenty-first century, the majority of African states still cope with external financial dependence, while the majority of Asian states enjoy greater sovereignty in fiscal and monetary affairs.Footnote 9
Future Agenda
Now that the first steps towards a comparative understanding of fiscal developments in colonial Asia and Africa have been made, it is worth asking which lines of research are worth pursuing in future work. We see three interrelated priorities.
First, and most important, we need to connect the largely separate literatures on colonial and post-colonial fiscal development. Are there any common patterns in fiscal development after independence, which can be traced back to a shared colonial heritage, or should we rather focus on understanding cross-country variations? One of the reasons we grapple with the question of ‘colonial legacies’ is that we lack a sound analytical framework to distinguish the structural conditions of fiscal capacity building which emerged under colonial rule, from the changes brought about by the many political revolutions in Africa and Asia since independence. In addition, civil or interstate wars, exogenous economic shocks caused by volatile world markets and the unpredictable dynamics of global political relations have all affected fiscal policies. The structural conditions inherited from the colonial era are important for understanding path dependence in both fiscal resources and the institutions underpinning country-specific systems of public finance. The more contingent events are important in analysing deviations from the historical path. Such an analytical framework is crucial in explaining not just the persistence of specific taxes in the post-colonial era, but also changes in the composition of revenue and the allocation of government expenditure. More broadly, we need a framework to understand the political economy of fiscal reforms in the post-colonial era.
Second, and directly connected with the above, there is a need for a deeper analysis of the development of public debt spanning the colonial and post-colonial eras. One of the most influential and widely shared episodes of post-colonial public finance was the debt crisis of the 1980s. To what extent did colonial legacies play a role in this crisis, and to what extent did the responses to this crisis erase these legacies? On the eve of independence, all countries were indebted to varying degrees, but nowhere did government borrowing reach proportions that could overturn the entire fiscal system. This changed in the third quarter of the twentieth century, but the mechanisms driving this change remain poorly understood, at least from a historical perspective. The evolution of government borrowing in the global South is an extremely important aspect of both political and financial globalization, linking fiscal policy-making at the country level to global financial relations, but there is a need for more comparative research on this topic.
Third, in spite of improved quantitative data that can be used to estimate and interpret fiscal developments across countries, there are more subtle changes in the conduct of fiscal policy which are harder to monitor. Changes in information technology have enabled some states to organize the assessment and collection of taxes, as well as the implementation of government expenditures in a much more efficient and transparent way. These changes in turn facilitate more complex methods of socio-economic governance and targeted expenditures than was feasible fifty years ago. In this respect, the changes in countries such as South Korea, Taiwan and South Africa have been nothing short of revolutionary. At the same time, there are many countries, especially in Africa, where the development of reliable flows of information needed for the functioning of the fiscal state has remained disappointing. To date, we understand only imperfectly to what extent varying colonial legacies can explain the development of state capacity in the post-colonial era, although it is clear that the education and training of government officials, both before and after independence, must have played an important role.
In other words, more comparative research is needed to obtain a deeper understanding of the complexities of fiscal state formation. We hope this volume will serve as a valuable contribution to this broader agenda.
