Structural cross-cutting issues

Taxing the informal economy

The Outlook country surveys gather evidence that the “informal economy” – workers and companies operating outside the reach of the law or public administration – is a major obstacle to broadening the tax base and collecting direct taxes. Informality is indeed widespread in developing countries, but highest in sub-Saharan Africa (Table 2). This poses a wide range of economic challenges: not only are taxes not collected, but informal firms are also often less productive and there are no labour and social protection schemes for workers. In short, high informality leads to lower economic growth and greater social exclusion. Informality often arises where the costs of legal employment outweigh the benefits for producers, employers or employees. If entry costs into a regulated economy are unaffordable, people and businesses are forced to remain outside the system (Jütting and de Laiglesia, 2009).

 

Table 2: Share of informal employment in total non-agricultural employment in Africa
%, selected countries 1975-79 1980-84 1985-89 1990-94 1995-99 2000-07
North Africa         47.5 47.3
Algeria 21.8   25.6   42.7 41.3
Morocco   56.9     44.8 67.1
Tunisia 38.4 35 39.3   47.1 35
Egypt 58.7   37.3   55.2 45.9
Sub-Saharan Africa       76    
Benin       92.9    
Burkina Faso     70 77    
Chad       74.2 95.2  
Guinea   64.4   71.9 86.7  
Kenya     61.4 70.1 71.6  
Mali 63.1   78.6 90.4 94.1 81.8
Mauritania   69.4 80      
Mozambique       73.5    
Niger 62.9          
Senegal   76        
South Africa           50.6
Zaire (Dem. Rep. of Congo)   59.6        
Zambia       58.3    

Source: Jütting and de Laiglesia (2009).

Fiscal policy in developing countries must consider capacity, incentives and segmentation. In countries where the informal sector comprises more than half of the economic activity, the question arises as to how governments can pursue fiscal policy in terms of both taxation and expenditure. On the one hand, more firms in the formal sector means increased tax collection and social security contributions for the state. On the other hand, more people covered by social security means increased liabilities for governments as employees become eligible for health insurance, pensions and other benefits where offered.  In addition, the increase in tax revenue from formalising informal firms may be smaller than expected. Indeed informal firms that enter the system are often too small and too poor to make sizeable contributions. However, value-added and sales taxes could still produce a notable increase in tax collection as these also indirectly tax informal activities (Latin American Economic Outlook, 2009).

Ghana has tried a new approach to tax collection. The Internal Revenue Service negotiated an arrangement with the Ghana Private Road Transport Union (GPRTU) to use the union as a tax collection agent under the “Identifiable Groupings Taxation” (IGT) scheme. Simple and easy to administer, IGT calls for small and affordable taxes to be collected daily or weekly from both formal and informal union members. GPRTU retains 2.5% of revenues as an incentive to maximise collection. Although relatively successful, this attempt to make inroads into the informal sector has come at a high cost and created opportunities for corruption. So, while more tax is raised, the amount is below potential (Joshi and Ayee, 2002).

The quality of tax policies and tax administration also plays an important role. Complex tax codes and high compliance burdens imposed by an inefficient tax administration are powerful incentives for small enterprises to remain informal. For example, country surveys reveal that, in Uganda and Zambia, bureaucracy and corruption are identified as barriers against entering the formal sector. While in Togo, informal firms state that complex registration procedures impede their entering the formal sector. Figure 17 plots the number of hours per year it costs businesses to pay taxes. On average, it takes fewer hours to pay taxes in Africa than in Latin America, but more than in the Pacific, Asia and OECD countries. However, on the left hand of the scale are countries where the compliance burden is exceptionally high. See OECD-CTPA (2008) for an extensive discussion of the tax compliance burden.

Tax administration capacity

Administrative capacity constraints have been highlighted throughout the Outlook country surveys as a major obstacle to improving tax policy in Africa. The administrative constraints are such that they limit policy options.

  • For example, in theory, relying more on income tax and exemptions on basic consumer items would enable more redistribution of resources. But where administrative capacity is weak, personal income tax is less progressive than expected. Firstly, only wages, mostly earned in large private firms and in the public sector, are taxed. Secondly, personal income earned on capital is typically not taxed. Capital, real estate income and other revenues of high earners in the informal sector are thus outside the reach of tax administration.
  • For a variety of reasons, VAT exemptions in Africa are often regarded by experts to be regressive (Box 1). Strategies that copy those used in countries with high administrative capacity can be counter-productive. In Morocco, before a 2005 fiscal reform, generous VAT exemptions undermined the potential of VAT introduced in 1986.

The vast majority of countries in the Outlook survey cite the lack of skilled staff as a major impediment to tax collection. The constraining factors encourage corruption, as highlighted in the country notes of Cameroon, Comoros, Guinea-Bissau and Nigeria.

The surveys have shown that despite great progress in adopting Information and Communication Technology to increase revenue collection, more can still be done. South Africa offers e-filing for payroll taxes, while Botswana, Cape Verde and Cameroon have e-taxation platforms. These initiatives require educational campaigns to motivate individuals and enterprises to use these systems. In Cape Verde, about 15% of companies used the new e-taxation system in 2009. Algeria, Angola, Côte d’Ivoire and other countries are looking for ways to incorporate new technology in their taxation systems.

Box 1: Regressive nature of VAT exemptions for commodities in Africa

Paradoxically, "traditional" VAT exemptions on commodities (Chambas, 2005) compromise poverty-reduction strategies. We immediately think of the reduction in tax revenue and the resulting fall in the financing of public spending as a result of tax exemptions. But we generally forget another direct effect of VAT exemptions.

In an open economy, the price levels of tradable goods are a result of CIF (cost, insurance and freight) prices, including taxes. If there are VAT exemptions on commodities, the prices of commodities, especially food products, are lower than if they were subject to VAT, since the exemption means that VAT is not applied at the border. This decline in domestic prices affects not only exempted products, but also substitute products. A VAT exemption for rice, for instance, could reduce the price paid to local producers, and even to producers of substitute products for rice. Indeed, the effects of a VAT reduction generally fall not on the marketing and processing channels but on the producers, who are paid lower prices.

Consequently, exemptions lead to a fall in the price of commodities -- usually food products -- thus benefiting consumers, especially poor people in urban areas. But the lower prices of products have a negative effect on local agricultural production, reducing producer prices and therefore the revenue of local producers, especially farmers, many of whom are also poor.

Furthermore, the more modern producers of agricultural products are worst affected in terms of competitiveness, since the significant input supplies they require are subject to VAT. As a result of the exemptions, even if local producers choose to be subject to VAT in order to benefit from the rebate mechanism, they must definitively bear the VAT on their intermediate consumption subject to VAT. The disadvantage of residual VAT places producers in a situation in which they have negative protection against imports not subject to any VAT (the VAT exemption applies at the customs barrier).

VAT exemptions are therefore counter-productive to the aim of reducing the poverty of the poorest agricultural producers and they slow down the development and modernisation of the agricultural sector.

Source: Jean-François Brun and Gérard Chambas, CERDI.

Getting donors to help

Whether, overall, aid helps or hinders public resource mobilisation remains unclear. However, it is a well established fact that the share of aid aiming to strengthen it is still very small.

There is increasingly widespread concern that the availability of foreign aid may reduce incentives for governments to raise domestic revenue. This may, in turn, negatively affect the quality of governance by reducing pressure for state capacity development and reducing incentives for government to bargain with citizens over taxes, as discussed earlier. Obviously, tax administrations are genuine in their desire to raise revenue, but the availability of large foreign aid flows may reduce the urgency with which revenue collection is pursued. There is an additional risk that aid-dependent governments, in particular, will shy away from politically demanding income, property and local tax reforms, while these are precisely the areas that are likely to be most important to tax-governance linkages. In practice, however, it is very difficult to test the impact of aid on domestic revenue because of the many factors that shape tax collection. The debate amongst academics remains largely inconclusive (Box 2). Eventually, aid donors are an integral part of the political economy of public resource mobilisation, even more so as they increasingly propose financial and technical assistance to increase revenue collection.

Indeed, ODA often includes components designed to increase revenue collection, such as direct funding for tax reform; conditionality that requires increased or at least constant, domestic revenue generation; requirements for local matching funds for aid projects; and/or demands for increased social spending which, indirectly, generate pressure for greater revenue mobilization. Figure 18 shows that public sector financial management represented 2% of aid spent on technical cooperation in Africa in 2008. Given that tax administration is a subset of technical cooperation for public sector financial management, donors’ help in building African tax administrations is less than 2%. The Outlook country surveys confirm that there is a lot of room to increase aid in this area. Finally, the issue of whether aid-funded goods and services should be taxed by recipient governments is discussed in the next section.

Box 2: Does aid help? A review of the academic debate

The renewed interest for public resource mobilisation in Africa comes at a time when the effectiveness of foreign aid on the continent is again being questioned. Against the “aid fatigue” argument, proponents of aid have been saying that the returns from aid-funded investment in development can be enormous. They argue that a “big push” in aid funds is required to turn a vicious circle of poverty and under-development into a virtuous circle of poverty reduction and shared economic prosperity.This “big push”, first popularised in the 1950s and 1960s (Easterly, 2005; Guillaumont and Jeanneney, 2006), is now advocated by the UN under the lead of academic Jeffrey Sachs. Aid considered as a “subsidy” provides temporary financial assistance to encourage long term revenue collection, investment in physical and human capital, and the establishment of the institutions of a developmental state (Brautigam and Knack, 2004). Aid-as-subsidy played this role in Botswana, South Korea or Chinese Taipei (Brautigam, 2000, Moss et al., 2006).

Conversely, Ross (2004) makes the case that, like resource rents, foreign aid hurts incentives for good governance in Africa and elsewhere. The so-called “resource curse” argues that unearned income undermines incentives to build local institutions and a social contract with the population. Aid is suspected to have a similar effect of discouraging revenue collection, distorting expenditure decision-making and undermining the incentives to build state capacity. Under this view, aid is not only a crutch delaying institutional development, but potentially undercuts those effects (Moss et al. 2006, ). When a government specifies expenditure needs and donors match these needs with budget support, the public budget constraints are softened and so there is no incentive to raise revenues. Aid would also lead to overall increases in government spending (Remmer, 2004). As politicians have less of a need to prioritise expenditure within a budget constrained by revenue collection, it would weaken government’s capacity to identify budgetary trade-offs. In addition, as Heller and Gupta (2002) argue, the fiscal uncertainty of dependence on external assistance makes long-term planning extremely difficult for countries.

Collier (2006), however, argues that aid has “a less damaging effect on governance than oil if it is provided in “purposive” ways, and accompanied by mechanisms of scrutiny, expertise and management techniques that can add value, and create some pressure for accountability”. Besides, the “resource curse” thesis, applied to either natural resource rents or aid flows, finds little robust evidence in the academic literature.

Amongst the studies that focus more specifically on the impact of foreign aid on tax revenue and tax administration, several conclude that this impact is negative (Remmer, 2004; Gupta , 2005; Devarajan, Rajkumar and Swaroop, 1999; Brautigam and Knack, ibid.; Knack, 2009). Gupta (2007), however, finds a negligible, or even positive, impact of aid on tax revenue. As for studies by Brun (2007) and Cottet and Amprou (2006), they do not clarify whether aid helps or hinders domestic resource mobilisation. Much depends on the type of aid flow and the circumstances in the recipient country. In their 2004 study, Gupta et al., focus on the revenue response to foreign aid, separating total net aid into grants and loans to test whether the impact of grants on domestic revenue is different from that of (concessional) loans. The study suggests that some governments may consider grants to be free substitutes for tax revenue. By contrast, loans must be repaid, which provides incentives for governments to at least maintain tax revenues at current levels if not to increase them (Brautigam, ibid.). Finally, aid is thought to work best in states with high quality public institutions (Burnside, Craig and Dollar, 2000; Brautigam, ibid.).

Theme 2011

Experts from different fields analyse what measures should African governments take in order to engage effectively with emerging economic partners in Africa, such as China, India, Brasil or Turkey.

 

Tax expenditure surveys


Jean-Philippe Stijns
, co-author of the "Public Resource Mobilisation" study, highlights Morocco's practices while observing their taxation policies.