2 The importance of taxes for development

by Odd-Helge Fjeldstad, Chr. Michelsen Institute, June 2009

2.1 Weak state finances

Average tax revenue in low-income countries was approximately 13 percent of GDP in 2000 (Baunsgaard & Keen 2005), i.e., less then half of the OECD average of 36 percent (OECD 2007). In the past few years, overall government income from domestic sources in sub-Saharan Africa has risen from an average of less than 15 percent in 1980 to a little more than 18 percent in 2005 (Gupta & Tareq 2008). The greater part of this increase comes from higher income from natural resources, and not through the tax system. Domestic revenue from sources not related to natural resources has risen by less than 1 percent of GDP in the last 25 years. In developing countries with abundant natural resources, too, state income from sources other than natural resources has remained more or less unchanged (Keen & Mansour 2008).

In many low-income countries that are net importers of oil, domestic revenue generation has not kept up with the increase in public spending. Consequently, a growing share of their operating budgets are financed through foreign aid. In Ghana, for instance, the share of the operating budget (including debt relief) financed through aid increased from 16 percent of GDP in the period 1997-1999 to 36 percent in 2004-2006 (Gupta & Tareq 2008). The corresponding figures for Tanzania show an increase from 22 percent to 40 percent, and in Uganda from 60 percent to 70 percent. Estimates from the OECD show that the dependence on aid will probably increase in the future, particularly in Africa and some countries in Asia. 1

These numbers show that there are considerable weaknesses in the fiscal base of many low-income countries. According to the International Monetary Fund, tax revenue equivalent to 15 percent of GDP is a “reasonable” minimum level for low-income countries to secure the financing of basic government tasks such as law and order, health, and education (IMF 2005). Many countries do not reach this level. A study by Fox & Gurley (2005) found that as many as 44 of the 168 countries included in the study had tax revenues lower than 15 percent of GDP in the 1990s. Eighteen of these countries are in sub-Saharan Africa. Income from natural resources and aid substitute non-existent tax income, and might ensure that important development goals are reached. However, financing state spending through such sources contributes little to developing the institutional capacity of the state (Moore 2004; Ross 2001). Revenue from aid and natural resources is also generally more unpredictable than tax revenue (Bulír & Hamann 2007).

2.2 Explanations for the low tax base in poor countries

Several factors contribute to explaining the low tax base in developing countries.

1. Corruption and tax evasion are widespread. Studies from various developing countries show that it is not unusual that half or more of due taxes never reach the state coffers because of corruption and tax evasion (Christian Aid 2008; Fjeldstad & Tungodden 2003; Mookherjee 1997). Multinational corporations play a central part for instance through manipulating transfer prices to avoid taxation (see chapter 3 in this report). Tax evasion by domestic taxpayers is also widespread in poor countries (Chand & Moene 1999; Fjeldstad 2006). The willingness of the taxpayers to pay taxes can be low for historical, political, or cultural reasons (Lieberman 2003), but their reluctance can also be attributed to a lack of trust in authorities that consistently misuse public funds. (Rothstein 2000). Tax evasion in poor countries is probably one of the factors that contributes most to corruption in the public sector.

2. The political and economic elite in many developing countries is often not part of the tax base because of tax exemptions and/or evasion (Stotsky & WoldeMariam 1997). In addition, the tax base often does not include people in liberal professions like lawyers, doctors and private consultants (Bird & Wallace 2004).

3. The problems with mobilizing domestic resources are made worse because the liberalization of foreign trade the last 20 years has led to a fall in customs revenue for developing countries. This has been a particularly serious problem for low-income countries. Research from the IMF shows that while rich countries have succeeded in compensating for the fall in customs revenue through other sources of income, mainly value added tax, the poorest countries have only succeeded in replacing about 30 percent of lost customs revenue through other tax bases (Baunsgaard & Keen 2005).

4. It is difficult to collect taxes in poor, agricultural economies. The tax bases are often small, and the cost of collecting large; personal income is seasonal and unstable (Fjeldstad & Semboja 2001; Bernstein & Lü 2008). Where personal income tax generates around 7 percent of GDP in developed countries – and is paid by about 45 percent of the population, the corresponding number for developing countries is only 2 percent of GDP – paid by less then 5 percent of the population (Bird & Zolt 2005).

5. The informal sector, or the unregistered part of the economy, is large, particularly in towns, and this makes tax collecting difficult (see Table 2.1).

   Table 2.1 Informal sector (unregistered) as percentage of GDP (2000)

   Region	Value creation in % of official GDP
   AFRICA	41
   CENTRAL and South AMERICA	41
   Asia	29
   Post-Communist Transition Economies	35
   European OECD-Countries	18
   North AMERICA and OECD countries bordering on the Pacific	13.5

   Source FIAS (2009) based on The World Bank (2007)

6. Poor countries often lack the resources and capacity for building effective tax collection systems. Given the scarce administrative resources, it is rational for tax collectors to concentrate their efforts on the relatively small number of available taxpayers who have the ability to pay, but lack the political contacts that can “protect” them against taxation (Dasgupta & Mookherjee 1998; Svensson 2003).

7. In many developing countries a few hundred, or maybe a few thousand, taxpayers contribute the greater part of tax revenue (see Table 2.2). In Bangladesh, for instance, less than 1 percent of the population are registered as taxpayers, and 4 percent of these (i.e., less than 0.4 percent of the population) pay 40 percent of the tax take, whereas 50 percent of taxpayers (less than 0.5 percent of the population) pay less than 1 percent of the tax take (Sarker & Kitamura 2006). In Tanzania, with a population of more than 40 million people, 286 companies contribute about 70 percent of domestic tax revenue (Fjeldstad & Moore 2008). According to Baer (2002), 0.4 percent of taxpayers in Kenya and Colombia pay respectively 61 percent and 57 percent of the total domestic tax revenue.

   Table 2.2 Concentration of tax collecting in selected countries (2000)

   	Number of large Tax-Payers	% of Total Number of TaxPayers	% of Tax revenue
   Argentina	3 665	0,1	49,1
   Benin	812	1,0	90,0
   Bulgaria	842	0,1	51,4
   Hungary	369	0,1	42,1
   Kenya	600	0,4	61,0
   Peru	2 430	0,9	64,9
   The Philippines	833	0,2	36,0

   Source Bodin (2003), referred to in FIAS (2009)

8. In the last few years, many developing countries have to a greater extent then before been able to finance their spending from other sources than taxation, for instance through commercial loans, income from oil and mineral resources, and foreign aid. A number of studies show that extensive aid can have negative effects on the recipient countries’ incentives to generate revenue through domestic resources (Bräutigam & Knack 2004; Remmer 2004). Research conducted by the World Bank finds that aid to African countries reduces tax income by an average of 10 percent (Devarajan, Rajcoomer & Swaroop 1999). 2

9. Capital flight and tax havens contribute to entrenching existing tax structures in developing countries. In many countries, particularly in sub-Saharan Africa and Latin America, capital flight is accompanied by increased foreign borrowing. This borrowing is not used to finance investment or consumption, but to finance the capital flight itself (Rodríguez 1987; Boyce & Ndikumana 2005). The ensuing debt burden will probably hurt the poor most, since public spending on the social sectors and on investment in infrastructure must be cut to service the debt. This also affects the development of institutions in the public sector through a falling level of real earnings and increased corruption.

10. Capital flight is a global phenomenon, but there are great regional differences between developing countries (see Table 2.3).

    Table 2.3 Capital flight as share of private assets in Latin America and East Asia

    	(percent)
    	1980-9 (a)	1990-8 (a)	1980-9 (b)	1990-8 (b)
    Sub-Saharan Africa	27.6%	30.1	27.4	30.3
    Latin America & Caribbean	8.5%	9.0	7.5	7.9
    East Asia and the Pacific region	4.5	5.0	2.0	2.7

    Note: (a) all observations; (b) only full data points

    Source Collier et al. (2004, Table 1A, p.22)

11. Collier et al (2004) estimate that capital flight as a share of private assets was about two times higher in Latin America than in East Asia in the 1980s and 1990s. For sub-Saharan Africa, i.e., in the region with the greatest scarcity of capital, capital flight as a share of private assets was on average six times higher than in East Asia in the 1980s, and more than ten times higher in the 1990s. 3 It is probable that capital flight has not only caused, but is also caused by, lower economic growth, macroeconomic instability, and political instability in Latin America and sub-Saharan Africa. However, no matter which mechanisms have been active, capital flight from these regions has probably contributed strongly to the erosion of the tax base, and thereby also to the reduction in resources available for financing the development of public institutions, social services, and infrastructure investment. Capital flight may also have reduced the interest of the political elite in local economic growth and development.

2.3 Tax and governance

The political importance of taxation goes far beyond providing income to finance the public sector, investments, and the basic needs of the population. Historically, state building has been closely connected to the development of the tax system (Tilly 1992; Webber & Wildavsky 1986). 4 However, the tax system has not only contributed to establishing states, but also to promoting the state’s legitimacy and strengthening democracy, as well as to creating economic well-being for the general population. 5 The concept ‘fiscal social contract’ is central to explaining the development of representative states and democracies in western countries. The experiences of Western Europe and North America show that taxation has contributed to making the authorities more representative and accountable by furthering a dialog between the state and civil society on taxation. Mobilizing interest groups (business organizations, trade unions, and consumer organizations) to support, oppose, and propose tax reforms, has been central in this connection. The development of an effective tax administration also stimulated the development of institutions in other parts of the public sector. Systems for recruitment, competence building, and management in the tax administration have been models for the development of other parts of the public administration (Bräutigam et al 2008).

Although countries in East Asia followed a development path different from that of western countries, the tax system was an important component in the development strategy of this region, too. In South Korea and Taiwan, taxation contributed to supporting economic policies that furthered development and the building up of public institutions in general (Shafer 1997). The tax system in Taiwan required the authorities to develop extensive databases for a wide range of enterprises and households. To a large extent, this contributed to curbing the development of the informal sector that is characteristic in many other developing countries. In the 1950s, South Korean authorities focussed strongly on developing the tax system, particularly for personal and corporate taxes (34 percent of tax revenue derived from direct taxes). This laid the foundations for a broad-based tax system under the regime of president Park in the 1960s. Later, this was the basis for the development of state information systems and databases that made it possible for the authorities to target state credits, subsidies, and other interventions towards individual companies in the process of industrialisation. In Latin America and Africa, Costa Rica and Mauritius can point to similar experiences: The tax system was a key factor for the development of an accountable and functioning state. When the state depends on tax income from wide sections of citizens and businesses, the authorities have incentives to expand their presence also in rural and peripheral areas. This presupposes, however, that the state develops an institutional apparatus for registering its citizens and businesses, and an effective tax administration.

In this perspective, the challenge for poor countries is not necessarily to collect more tax, but to tax a larger share of their population and businesses. For several reasons, including economic structure and history, this is not easy. The informal sector is often substantial, and difficult to tax. Another important challenge is to avoid taxes that taxpayers generally regard as unfair, and that require a large measure of coercion to collect (for instance poll tax). Such taxes have been common in many poor countries, and it is characteristic for them that only a small proportion of dues are paid, and the cost of collection is high. They have also blocked the development of a fiscal social contract. However, the past few years have seen substantial resources invested in changing the attitudes and behaviour of tax administrations towards taxpayers (Fjeldstad & Moore 2009). Experience from a number of countries has shown that taxpayer behaviour can be changed by reforming the tax system. In some countries, this has given the public a more positive attitude to the tax system, and has led to the mobilization of interest groups that demand better public services. For example, the authorities in Ghana, Tanzania and Uganda have all increased their fiscal space through higher domestic revenue mobilization in the period 2000-2006 (Gupta & Tareq 2008: 45).

However, in many poor countries, the authorities have no incentives to enter into a dialogue and negotiate with organized groups in society. This is one of the main reasons for bad governance. A complex set of historical factors, including state formation through colonization, has frequently led to a concentration of economic and political power with a small elite. This elite generally does not pay taxes, and is relatively unaffected by organised interest groups in society. The state is powerful vis-à-vis its citizens – and is not answerable to them, but it is weak in terms of capacity for implementing policies. The elite lacks both the will and the ability to build a civil society. The need of the elite to negotiate with organised domestic interest groups is lessened further by the global context, where the political elite has access to enormous resources from sources other than the taxation of citizens, particularly income from natural resources (Leite & Weidmann 1999; Christian Aid 2008) and organized crime (Bayart et al 1999; Chabal & Daloz 1999). Developing countries whose income derives mainly from sources other than taxation of their citizens, for instance from natural resources like oil and minerals, are generally characterized by bad governance and poor public institutions (Ross 2001). Among the few exceptions are Botswana and Malaysia. Bad governance is often correlated with the state being independent of revenue from taxation of citizens and businesses. Access to substantial foreign aid can also contribute to detaching the state from its citizens, and reducing the need for tax reforms (Bräutigam & Knack 2004).

2.4 Conclusion

The development of effective tax systems is a great challenge for many countries. The total tax revenue is generally low, and tax evasion is widespread. The tax bases are often very narrow, where a small number of firms contribute the greater part of the tax revenue. The informal sector is large and growing. Taxation of international business transactions has become ever more complicated to handle for local tax administrations with limited resources for employing the necessary qualified personnel. Furthermore, capital flight and tax havens contribute to undermining the tax base. It is estimated that capital outflows from Africa represent 7.6 percent of total GDP on the African continent. This implies that African countries as a group are net creditors of donor countries.

Discussions of the importance of taxes for development have, until recently, been nearly completely absent. 6 This is now changing. Taxation is about to become a central – if not the central – topic in the development debate – also in developing countries. In August 2008, for instance, tax administrators, finance ministers and politicians from 39 African countries met in Pretoria, together with representatives of the OECD, the World Bank, IMF, and several bilateral aid organizations. 7

The “Pretoria-communiqué” concludes that more effective tax systems are central for a sustainable development because they can:

• mobilise the domestic tax base as a key mechanism for developing countries to escape aid or single resource dependency;

• reinforce government legitimacy through promoting accountability of the government to tax-paying citizens, effective state administration and good public financial management; and

• achieve a fairer sharing of the costs and benefits of globalisation.

State authority, effectiveness, accountability and responsiveness are closely related to the ways in which governments are financed. It matters that governments tax their citizens rather than live from oil revenues and foreign aid, and it matters how they tax them. Taxation stimulates demands for representation, and an effective revenue administration is the central pillar of state capacity.