Statistiques des recettes publiques en Afrique 2019

Chapter 2. Non-tax revenue trends, 2000-2017

Introduction

A complete picture of public finances requires statistics that go beyond taxation, especially in the case of many African countries that obtain substantial revenues in the form of grants or royalties from oil and minerals. Revenue Statistics in Africa collects statistics on both tax and non-tax revenues, non-tax revenues being government revenues that do not meet the OECD definition of taxation. Although data on non-tax revenues might not come from the same sources or have the same reliability as tax statistics (see Box Box 2.1) they need to be included in any accounting of a country’s total financial resources. This chapter therefore provides cross-country comparisons of non-tax revenue data for the countries included in this publication.1

The main categories of non-tax revenues2 reported as part of this project are (see also Annex Annex B):

• grants from foreign governments or international organisations (budget aid, food aid, capital transfers, current transfers, project grants, programme grants, international debt relief, etc.)

• rents and royalties (such as oil or mining royalties)

• other property income (interest, dividends and other returns on government investment);

• sales of goods and services (which include some administrative fees)

• fines and penalties (including fines and penalties due to tax violations)

• miscellaneous and unidentified revenues (non-tax revenues that cannot be classified according to the other categories).

Box 2.1. Non-tax revenues and data reliability

Data on non-tax revenues are often less detailed than tax revenue data. Some countries might have aggregate figures for total non-tax revenues in some years but do not break this information down beyond grants and other non-tax revenues categories. This may be due to the fact that data on non-tax revenues often comes from several different sources. Revenue Statistics in Africa indicates cases where total non-tax revenues are reported but may be incomplete.

Non-tax revenue data are not necessarily subject to the same level of scrutiny as tax revenues. Many types of non-tax revenue are not part of the main budget process, nor are they defined by legislation so they are not subject to the same oversight mechanisms as tax revenues. Non-tax revenues that are not accompanied by legal compulsions do not always require the same documentation.

Non-tax revenues are often under the responsibility of different authorities than tax revenues, so there is not necessarily a separation made between revenues and expenses. Some revenues such as administrative fees might be used as cost-recovery mechanisms and subtracted from cost figures rather than reported as revenues. This could be the case for local governments or other public institutions for which data only exists on net transfers of funds to the central government.

Grants, legal settlements and mining and oil contracts also often involve large payments by external entities such as multinational corporations and foreign governments that do not necessarily have incentives to maintain transparency. There could also be incentives to collude with local authorities to conceal costs. In addition, some countries may negotiate large payments from resource extraction payments as a lump sum that bundles together many categories of both tax and non-tax revenue, making detailed breakdowns less feasible.

Non-tax revenues as percentage of GDP

In 2017, non-tax revenues as a percentage of GDP varied significantly among the 26 countries included in this publication. They ranged from 18.7% of GDP for Botswana to 0.5% of GDP for South Africa. In every country except Botswana, the Republic of the Congo and Equatorial Guinea, non-tax revenues were lower than tax revenues (Figure 2.1). On average, the amount of non-tax revenue collected in each country was 43% of the amount collected in tax revenues.

Figure 2.1. Total tax and non-tax revenue by country, 2017

Note: Non-tax revenue includes sub-national non-tax government revenues for Eswatini, Kenya, Mauritius and Morocco, the only countries for which such revenues are reported. In 2017, these represented, respectively, 4%, 29%, 6% and 45% of all non-tax revenue collected for each country. South Africa reported some local and regional taxes, but no local and regional non-tax revenues.

1. Non-tax revenues for Botswana and Eswatini include funds from excises and customs duties collected by South Africa and then transferred to Botswana and Eswatini through the Southern African Customs Union (SACU) revenue-sharing agreement.

2. Figures for grant revenues and oil revenues for the Republic of the Congo are those reported by the IMF in its April 2019 World Economic Outlook. Figures for other kinds of non-tax revenue were not obtained for the Republic of the Congo.

The tax-to-GDP ratios and non-tax-to-GDP ratios need to be interpreted with caution for some countries due to incomplete data. See the country tables in Chapters Chapter 5 and Chapter 6 for further information.

Source: Table 4.1 in Chapter Chapter 4 and Table 6.1 in Chapter Chapter 6.

Statlink

Since 2008, many African countries have been increasing tax revenues but still seen their deficits increase (see Box Box 2.2). This is partly because government expenses have in general been increasing as a percentage of GDP but also because total government revenues do not follow the same patterns as tax revenues. Adding tax and non-tax revenues (Figure 2.2), we see that the countries raising the most tax revenues are not the same as those raising the most total revenue. Since 2008, some countries have seen drops in total revenue collected despite increasing their tax revenues, because their non-tax revenues had declined over the same period.

Figure 2.2. Total tax and non-tax revenue by country, 2008 and 2017

Note: See footnotes to Figure 2.1 and country tables in Chapters Chapter 5 and Chapter 6 for further information.

1. Non-tax revenues include SACU revenues.

2. Non-tax revenues only include grants and oil revenues.

Source: Table 4.1 in Chapter Chapter 4 and Table 6.1 in Chapter Chapter 6.

Statlink

Box 2.2. Taxation and debt

The 26 African countries in this report have seen a substantial rise in public debts, after a period of decline in the aftermath of the Heavily Indebted Poor Countries (HIPC) Programme. From 2012 to 2017, the average general government debt rose from 38.8% of GDP to 55.7% (Figure 2.3). It is of even greater concern that debt in foreign currency rose from 22.8% of GDP to 32.5%.1

The recent rise in debt during a period of increasing tax revenues can be attributed to rising expenditures and declining non-tax revenues. Although expenditures for the 26 African governments were low by world standards, they rose from 24% of GDP in 2007 to 25.6% in 2017 while 15 out of 22 countries saw non-tax revenues as a percentage of GDP decline over the same period.

African countries are often seen as risky investments. As of 31 July 2019, with the exception of Botswana, Mauritius, and South Africa, none of the 26 countries had a sovereign debt rated as investment grade by at least one major ratings agency.2 According to the IMF’s analyses through its Debt Sustainability Framework,3 the Republic of the Congo is in debt distress, and four countries (Cabo Verde, Cameroon, Ghana and Mauritania) are at high risk of debt distress.

Rising debt levels in countries with more difficult access to financing carry the risk that they will avail themselves of sources of financing around which there is less transparency, or for which long-term risks could be hidden. These risks are even more complicated in the cases of Burkina Faso, Côte d’Ivoire, Mali, Niger, Senegal and Togo (which use the West African CFA franc), and the Republic of the Congo, Cameroon and Equatorial Guinea (which use the Central African CFA franc) that do not have the option of borrowing in their own currency.

Figure 2.3. General government gross debt, 2005-17

Source: Authors’ calculations based on figures from April 2019 World Economic Outlook, IMF.

Statlink

1. Not counting Burkina Faso, the Democratic Republic of the Congo, Egypt, Madagascar and Nigeria for which data is not available.

2. Sovereign debt ratings as reported on the website tradingeconomics.com/country-list/rating.

3. IMF Debt Sustainability frameworks as reported in August 2019 at the IMF website: www.imf.org/external/Pubs/ft/dsa/DSAlist.pdf

Differences in the sources of non-tax revenues have profound implications when it comes to the risks associated with these sources of revenue in terms of their variability and their vulnerability to economic shocks and political changes. Figure 2.4 shows the contribution of each sub-category to total non-tax revenues for each country. Four distinct groups are apparent in these data:

• Ten countries received a majority of their non-tax revenue in the form of grants in 2017 (in descending order Niger, Rwanda, Togo, Burkina Faso, Senegal, Mali, the Democratic Republic of the Congo, Madagascar, Côte d’Ivoire and Uganda).

• Seven countries received more revenues from rents and royalties than other sources of non-tax revenues (in descending order The Republic of the Congo, Equatorial Guinea, Mauritania, Cameroon, Tunisia, Nigeria and South Africa). For all these countries except South Africa and Tunisia, oil royalties provided the majority of non-tax revenues.

• Botswana and Eswatini were unusual both in the amount of non-tax revenues they collected (18.7% of GDP and 14.3% of GDP, respectively) and because they received a majority of their non-tax revenue (56% and 86%, respectively) from the Southern African Customs Union (SACU) revenue-sharing agreement. This agreement is a revenue stream funded entirely through customs duties and excise taxes but is classified as a non-tax revenue under the category “Miscellaneous and unidentified revenues”. Botswana also received around 44% of its non-tax revenue from mineral royalties and dividends in 2017, which were mostly due to the diamond mining industry.

• For the remaining seven countries, neither grants nor rents and royalties nor SACU revenues were the most important source of non-tax revenues. For Ghana, Kenya, Morocco and the Seychelles, revenues from sales of goods and services, or from administrative fees were the largest contributor. Egypt, on the other hand, collected most of its non-tax revenues from interest and dividends. While Cabo Verde received grants that amounted to 3.6% of GDP in 2017 (making it the third-highest recipient of grants among the countries in this publication), non-tax revenues from other sources were larger in total.

Figure 2.4. Structure of non-tax revenues by country, 2017

Note: Data include sub-national government non-tax revenues for Eswatini, Kenya, Mauritius and Morocco.

1. For Rwanda non tax revenues aside from grants were not disaggregated in the data and are therefore classified as miscellaneous and unidentified revenues.

2. Complete non-tax revenue figures were not available for the Republic of the Congo. Figures reported here represent only grant revenue and oil revenue obtained from the IMF’s April 2019 World Economic Outlook classified as rents and royalties.

3. For Botswana and Eswatini, “Other revenues” include funds from excises and customs duties collected by South Africa and then transferred to Botswana and Eswatini through the Southern African Customs Union (SACU) revenue-sharing agreement.

4. Interest, dividends, etc.

5. Sales, fines, SACU revenues, etc.

See the footnotes to Figure 2.1 and country tables in Chapters Chapter 5 and Chapter 6 for further information.

Source: Table 6.2 in Chapter Chapter 6.

Statlink

The decreases in non-tax revenues observed between 2008 and 2017 for most (16) of the countries in this publication were driven by declines in all four main subcategories. As shown in Figure 2.5, drops in non-tax revenues have been observed within all four groups of countries mentioned above. The biggest driver of change for the 21 countries for which data was available was the average drop in rents and royalties of 2.1% of GDP, mostly due to lower oil prices since 2014. However, this drop in non-tax revenues between 2008 and 2017 followed a period in the early 2000s when commodity revenues and grants both increased. Of the 17 countries for which data on non-tax revenues since 2000 was available, 14 reported increases in non-tax revenues between 2000 and 2008.

Figure 2.5. Change in the structure of non-tax revenues for select countries, 2000-17

Note: Non-tax revenues were not available before 2002 for Egypt, before 2004 for Botswana, before 2009 for Mauritania or before 2005 for Equatorial Guinea.

See footnotes to Figures 2.1 and 2.4 and country tables in Chapters Chapter 5 and Chapter 6 for further information.

Source: Table 6.1 in Chapter Chapter 6.

Statlink

Although there are long-term trends in non-tax revenues, these revenues tend to vary more from year to year than do tax revenues. Over the past decade, 14 of the 26 countries in this publication showed at least one year-on-year increase in non-tax revenue equivalent to 2% of GDP while 15 countries registered at least one year-on-year decrease of a similar magnitude. Such fluctuations are often driven by resource rents, due to swings in commodity prices or production, or by grant revenues, reflecting the shorter duration and lower predictability of grant flows in comparison to tax revenues. Other sources of non-tax revenue, such as SACU revenues, have also proven highly variable.

• In the Republic of the Congo, combined revenues from grants and oil jumped from 36.5% of GDP in 2007 to 47.2% of GDP in 2008 before declining by 25.7 p.p. the following year. Between 2014 and 2015, these revenues dropped from 35.7% of GDP to 14.7% of GDP. In both cases, these were due to major changes in oil prices.

• Between 2011 and 2012, Mali experienced a sudden drop in grant revenues, going from 4.5% of GDP down to 0.1% during a year of multiple political crises.

• Eswatini’s non-tax revenues fell from 18.2% of GDP to 8.1% of GDP between 2009 and 2010 when its SACU revenues fell by 80% following the collapse in customs and excise taxes in the two years following the global financial crisis.

Grants

The amount of grants reported for 2017 by countries included in this publication varied from less than 0.5% of GDP (Botswana, Equatorial Guinea, Morocco, Nigeria and South Africa) to over 7% of GDP in Niger (Figure 2.6).

Figure 2.6. Grants by country, 2017

Source: Tables 6.2 and 6.3 in Chapter Chapter 6.

Statlink

The 11 countries receiving grant revenues amounting to at least 1% of GDP in 2017 include all the countries classified as low income3 by the World Bank, along with three lower-middle-income countries (Cabo Verde, Côte d’Ivoire and Senegal)4 . For all of these countries except Cabo Verde, grant revenues were a majority of non-tax revenues in 2017.

There is a partial relationship between grant revenues received by countries in this publication and their income per capita (Figure 2.7). However, grant revenues vary among countries with low GDP per capita. For example, Mali, Rwanda and Uganda have similar levels of GDP per capita (in purchasing power parity terms (PPP)) but in 2017 they reported grants as a percentage of GDP ranging from 1.0% (Uganda) and 1.8% (Mali) to 4.7% (Rwanda). Cabo Verde is another outlier, receiving grants equivalent to 3.6% of GDP; no other country with GDP per capita over 3 500 USD collected more than 1.2% of GDP.

Figure 2.7. Grants as a percentage of GDP versus GDP per capita by country, 2017

Source: Table 6.2 in Chapter Chapter 6, and GDP figures from the April 2019 World Economic Outlook, IMF.

Statlink

Grant revenues to African governments have some overlap with Official Development Assistance (ODA) as defined by the OECD Development Assistance Committee (DAC).5 However, grants and ODA are not exactly the same. Grants covered by this publication focus on revenues received by countries from foreign governments and international organisations whereas ODA is a measure of donor countries’ efforts to support development in low- and middle-income countries. ODA not only covers flows captured under “grants” in this publication (budget aid, food aid, capital transfers, current transfers, project grants, programme grants, international debt relief, etc.) but also encompasses concessional loans, in-kind technical assistance, in-donor administrative costs, in-donor refugee costs and scholarships as well as other activities that do not generate a (net) revenue flow to the country. Therefore, the volume of ODA received by a given country will be different from – and usually higher than – the volume of grants reported in Revenue Statistics.

Statistics collected by the OECD DAC on ODA flows to African countries include public flows from all members of the DAC and some countries outside the DAC,6 as well as major international organisations such as the World Bank, the IMF and the UN. Statistics on development co-operation provided by some large non-OECD economies, such as Brazil, China and India, are not included.

Comparing grant revenues to ODA inflows (Figure 2.8) shows that countries receiving the most ODA tend to receive more grant revenues. However, the relationship is not linear; Niger received the highest share of grant revenue as a percentage of GDP in 2017 but Rwanda received the most ODA as a percentage of GDP.

Figure 2.8. Grants versus net ODA received for selected countries, 2017

Note: Figures for ODA include donations from the 30 member countries of the OECD Development Assistance Committee (DAC), the 20 non-DAC countries, as well as multilateral institutions, and private donations. Some major donor countries, such as China are not yet accounted for in these data.

Source: Table 6.2 in Chapter Chapter 6 and OECD Development Assistance Committee (DAC) 2019 for the ODA data.

Statlink

Grant revenues have been volatile between 2008 and 2017, with five countries receiving grants in excess of 6% of GDP in at least one year (Figure 2.9). Increases were not sustained, however: in all five cases, there was at least one single-year drop in grants over this period of at least 2.6% of GDP. In total, 14 countries among the 26 participating in this publication registered a one-year drop in grant revenues equivalent to at least 1% of GDP. The largest shifts in grant revenues were observed in 2009, 2012 and 2014.

Average grant revenues as a percentage of GDP for the countries in this publication have declined to 1.5% of GDP since they peaked at 2.3% of GDP in 2009. Changes in grant revenues often depend on circumstances unique to the country. Egypt, for example, saw a spike in grant revenues in 2014 following the normalisation of international relations after its internal political crisis, while Mali saw its grant revenues disappear in 2012 before resuming the following year.

One important source of variation in grant revenues is debt relief programmes such as the Heavily-Indebted Poor Countries (HIPC) programme. This was launched in 1996 by the IMF and the World Bank as a successor to the Multilateral Debt Relief Initiative (MDRI) with the objective of alleviating the debt burden of lower-income countries facing significant debt distress.

Figure 2.9. Grants for selected countries, 2007-17

Source: Revenue Statistics in Africa comparative data available at dx.doi.org/10.1787/be755711-en.

Statlink

Of the 39 countries eligible for these initiatives, 33 were in sub-Saharan Africa. Fourteen countries in this publication (Burkina Faso, Cameroon, the Republic of the Congo, the Democratic Republic of the Congo, Côte d’Ivoire, Ghana, Madagascar, Mali, Mauritania, Niger, Rwanda, Senegal, Togo and Uganda) received debt relief under HIPC and MDRI (IMF, 2016_[1] ). This led to spikes in grant revenues in some cases: in Niger, grant revenues rose to 46.3% of GDP in 2006 following a debt write-off.

Resource revenues and other property income

One of the most important sources of revenue for governments aside from taxation originates from their status as property owner, especially as land owner. Such revenues are reported as property income in Revenue Statistics in Africa and encompass various distinct forms of income. (Figure 2.10)

Rents and royalties are revenues generated from land the government owns, usually by prospecting and extracting non-renewable resources or from harvesting government-owned farms and forests. Interest and dividends are returns on government-owned investments in corporations. In general, royalties from oil and minerals are kept distinct from other forms of investment income but the issue can become complicated. Many African governments create public corporations that manage oil revenues or maintain revenue-sharing agreements with private corporations that exploit these natural resources. The classification of such funds is not always clear. For example, in Botswana, a significant proportion of property income comes from companies investing in diamond extraction.

Figure 2.10. Rents and royalties by country, 2017

Note: Rwanda is excluded as the data are not available.

1. Cabo Verde property income is not disaggregated.

Source: Authors’ calculations based on Tables 6.5-6.30 and Table 4.17.

Statlink

Revenues from natural resources

Rents and royalties among the 26 countries participating in this publication consist mostly of mineral and oil revenues. In 2017, over half the rents and royalties they collected were oil revenues (royalties and revenues from revenue-sharing agreements with oil companies), which accounted for a majority of rents and royalties for Cameroon, the Republic of the Congo, Egypt, Equatorial Guinea, Nigeria and Tunisia. Meanwhile, minerals were the source of most rents and royalties for Botswana, Burkina Faso, the Democratic Republic of the Congo, Ghana, Kenya, Niger, South Africa and Uganda. Fishing accounted for all revenue recorded as rents and royalties in Seychelles and for 80% in Mauritius. About 4% of total rents and royalties for the 26 countries were not resource-related; these included revenues from the construction and operation of airports in Kenya and Togo, for example.

Natural resource revenues tend to be a highly volatile source of financing. Five countries among the 26 generated rents and royalties from natural resources equivalent to more than 6% of GDP over the past ten years and they all registered substantial year-on-year declines in these revenues at some point over this period (Figure 2.11). In the Democratic Republic of the Congo and Côte d’Ivoire, rents and royalties dropped to less than a quarter of their peak. Cameroon’s rents and royalties dropped to 32.8% of their peak, a difference equating to over 4% of GDP.

Rents and royalties for the Republic of the Congo are on a different scale than for other African countries and the impact of the volatility of resource rents is far larger. As a result of the size of these revenues as a percentage of GDP, drops in oil prices in 2008-09 and 2014-15 were associated with drops in oil revenues (and therefore drops in total non-tax revenues) equivalent to over 20% of the country’s GDP.

As well as the impact of price volatility, changes in production can also have a major impact on resource revenues, for example due to new mining operations or the exhaustion of old mines. In the case of Botswana, diamond sales meant for FY 2015/16 were delayed for a year due to buyers and sellers not being able to agree on a price. This led to a 41% difference in mineral royalties between FY 2015/16 and FY 2016/17.

Figure 2.11. Rents and royalties for select countries, 2007-17

Note: Figures for the Republic of the Congo and Equatorial Guinea cannot be presented using the same scale as the one used for the other selected countries. Cameroon is featured in both charts in order to serve as a reference.

Source: Revenue Statistics in Africa comparative data available at http://dx.doi.org/10.1787/be755711-en.

Statlink

The impact of resource rents on government finances among some of the reporting countries reflects a dependence on natural resources that has implications for both the country’s finances and their economic development. In a context where natural resources can have both positive and negative impacts on a country’s economy, there are multiple mechanisms by which governments can capture some of this wealth in order to finance investments as part of a more sustainable economic development strategy (see Box Box 2.3).

Box 2.3. Natural resources and public finances1

There are various mechanisms by which natural resources can generate revenues or savings for government:

Rents and royalties are the most direct means of deriving revenue from natural resource wealth. The government charges fees to companies and individuals in exchange for the right to access government lands. It does this in its capacity as landowner and these fees are generally decided through negotiation. These are recorded under property income.

Payment for services provided by the government is another source of non-tax revenue paid for by businesses in the primary sector. This can include, for example, payments for environmental inspections or for the construction of infrastructure. These are recorded under sales of goods and services.

Public ownership (full or partial) of a corporation that exploits natural resources on the government’s behalf results in government revenues in the form of profits and dividends. These are recorded under property income.

Taxes or levies can be designed to target companies or individuals that access natural resources. It may sometimes be difficult to make a distinction between a tax on companies operating on public lands and a royalty charged to such companies.

Taxes targeting the activity of natural resource exploitation could be introduced, such as an excise tax on the sale of materials extracted from public lands or a tax on mining that targets the activity, rather than the individual or company exploiting natural resources. These, too, will be recorded as taxes.

Companies and individuals exploiting natural resources are generally subject to the same taxes (such as income taxes and value-added taxes) as the rest of the population. Revenues from these general taxes will be included in government financial statements but not necessarily attributed to sectors exploiting natural resources.

Companies and individuals may use some of the wealth they obtain from natural resource extraction to build infrastructure or provide services. Where this satisfies demand for public investment or services, it could result in savings on government expenditure but it would not be recorded as revenue. In some cases, as a condition for access to publicly-owned natural resources, governments may demand that companies and individuals provide certain services, for example paying for at least part of the cost of constructing and maintaining roads to mines.

1. (Mansour and Rota-Graziosi, 2013_[10] ) note that taxes on oil extraction can include corporate income tax, excise taxes on energy products, un-refunded sales taxes such as VATs and non-tax revenue can include royalties, profit sharing, dividends received from state enterprises, and other investment income received from government direct participation in extractive enterprises.

There is a great deal of variation in the size and type of natural resource endowments in African countries. The contribution of this natural wealth to countries’ GDP also varies, as does its impact on government budgets. For example, according to World Bank estimates of the amount of national wealth generated by natural resource extraction, the Republic of the Congo and Equatorial Guinea (both mostly oil) and South Africa (mostly minerals) were the three countries that generated the highest natural resource wealth per person in 2017.7 However, due to South Africa’s larger and more diversified economy, natural resource extraction accounted for 5.1% of GDP (Figure 2.12) while resource rents collected by the government only amounted to 0.2% of GDP.

Figure 2.12. Estimated natural resource extraction, 2017

Source: World Bank (2019), World Development Indicators for estimates of the total value of natural resource production in each country.

Statlink

The impact of countries’ natural resource endowment on public finances also varies greatly. Countries such as the Republic of the Congo, Mauritania and Equatorial Guinea generate significant public revenues from their natural wealth while Togo and Uganda, for example, reported public revenues from natural resource rents equivalent to no more than 0.1% of GDP despite their natural resource wealth production exceeding 13% of GDP.

In three countries (the Democratic Republic of the Congo, Niger and Uganda) wealth generated by forests exceeded 10% of GDP but, as noted above, these did not contribute much to natural resource rents. Although countries such as Cameroon and Niger did collect some royalties on logging, forest revenues were a small part of the non-tax revenues for the countries in question.

A number of the 26 participating countries are heavily impacted by swings in the prices of individual commodities (Figures 2.13). For example, since 2000, the price of gold has risen by as much as 28.1% over the previous year and has fallen by as much as 15.5%. The price of oil was 36.4% higher in 2008 than it was a year earlier, while in 2015, it was 47.3% lower than the previous year. Figure 2.13 shows the effect of such volatility, with a rise in the value of gold exports from Burkina Faso and Mali in 2012 the steady decline in oil exports from the Republic of the Congo and Nigeria between 2013 and 2016.

Figure 2.13. Price of key commodities, 2000-18

Source: World Bank Commodity Price Data (The Pink Sheet).

Statlink

Other non-tax revenues

Certain countries generate substantial non-tax revenues from the normal operations of government. These revenues can be divided into sales of goods and services, fines and penalties, and miscellaneous and unidentified non-tax revenues. The challenges in compiling non-tax revenues outlined in Box Box 2.1 are often encountered in these kinds of revenues, since they are typically not collected by tax administrations and are often not defined by legislation. They might not even be captured by government budgets.

For all these categories of revenues, there can be arguments about the degree to which they are unrequited or compulsory, and therefore whether they are more properly classified as taxes. The question of which administrative fees are taxes and which are not is discussed in Annexes Annex A and Annex B. However, this is not always clear for the question of fines and penalties.

Institutions and governments that have less authority, and thus less tax collection power, may rely more on these kinds of non-tax revenues. This means that the lack of information in this publication on local government finances might under-estimate the non-tax revenues collected by these institutions.

Sales of goods and services and administrative fees

Governments produce goods and services both as a market- and as a non-market participant. As a market participant, governments sometimes provide goods and services that can also be obtained from private companies (such as selling food or providing transportation). At the same time, they also provide services in the course of administering programmes and executing laws that are unique to the function of government. Fees charged for these non-market services are generally classified as administrative fees.

Administrative fees are often difficult to classify since they occupy a grey area between payments for services (which are non-tax revenues) and compulsory unrequited payments collected during government operations (which are taxes).8 Court fees and fees for driver’s licences, passports, patent registrations and marriage certificates tend to be classified as non-tax revenues. When administrative fees are classified as taxes, they are typically assigned to one of the following categories:

• 4400 Taxes on financial and capital transactions (for example, taxes on sales of land);

• 5200 Taxes on use of goods and performing activities (for example hunting licences, vehicle registrations);

• 6000 Other taxes (for example, sales of fiscal stamps, where stamps are used to pay for taxes and administrative fees).

Five countries generated revenues from sales of goods and services and administrative fees equivalent to at least 1% of GDP in 2017 (Figure 2.14): Cabo Verde (3.0% of GDP), Seychelles (2.2% of GDP), Ghana (1.8% of GDP), Morocco (1.5% of GDP) and Egypt (1.1% of GDP).

The composition of these revenues varied by country. In Cabo Verde, 96% of revenues listed under “sales of goods and services” were fees. In Ghana, most revenues listed as sales of goods and services were collected by Municipalities, Departments and Agencies or by Metropolitan, Municipal and District Assemblies. In Morocco, 61% of revenue from government sales of goods and services are collected at the local level.

Miscellaneous or unidentified revenue

Some notable categories of non-tax revenue do not fit within any of the categories listed above. These include:

capital transfers not included elsewhere

voluntary donations to government agencies from individuals or private corporations (not including donations from international organisations, which are classified as grants)

private payments to government made as a result of major court settlements or insurance claims (such as the EUR 100 million payment Arcelor Mittal agreed to pay to the government of Senegal after a court case following the cancelation of an iron mine (RFI, 2014_[2] ))

payments covering different categories in the classification list where a breakdown is not available

payments whose proper classification is unknown due to a lack of data. Rwanda did not disaggregate non-tax revenue outside of grants, so these were included in this category

SACU revenues to Botswana and Eswatini.

Miscellaneous and unidentified revenues can be a significant component of non-tax revenues in certain countries. These include exceptional voluntary contributions to government in Tunisia, capital transfers from special statutory funds in Mauritius and, in Morocco, payments made to government in exchange for the right to compete with state institutions in the provision of services.

Figure 2.14. Sales of goods and services and administrative fees and other revenues related to administration, 2017

Note: Non-tax revenue includes sub-national tax revenues for Morocco and South Africa and sub-national non-tax revenues for Kenya and Morocco, the only countries for which such revenues are reported.

The Republic of the Congo, Nigeria, and Rwanda are excluded as data are not available. Senegal reported no revenue from sales of goods and services.

Source: Revenue Statistics in Africa comparative data available at http://dx.doi.org/10.1787/be755711-en.

Statlink

The volatility of miscellaneous and unidentified revenues may be due to large capital transfers, revenue streams that are short-lived or funds that are reclassified as unidentified due to lack of information. The higher values for this category could reflect uncertainty as to the true amount of revenue within other non-tax revenue classifications.

Southern African Customs Union revenues

The Southern African Customs Union (SACU) has five member countries (Box Box 2.4). Differences in the values of GDP, GDP per capita and internal trade for each country lead to different results from the application of the SACU revenue-sharing formula.

In 2017, Botswana and Eswatini had far lower tax-to-GDP ratios than South Africa (12.2% and 17.0% versus 28.4%, respectively). When non-tax revenues (including SACU revenues) are added, Botswana and Eswatini had revenues as a percentage of GDP that were slightly higher than South Africa (31.0% and 31.3% versus 28.9%, respectively).

Despite this apparent convergence in 2017, SACU payments have historically followed very different trajectories in the three SACU countries in this report. The economic shock of the global financial crisis rendered many economic forecasts of the time inaccurate; because SACU payments were calculated based on these forecasts, major adjustments to SACU payments were required in 2010 and 2011 to compensate for years of overpayments into the SACU pool (Mongardini et al., 2011_[3] ).

This had a disproportionate effect on the smaller SACU economies such as Eswatini, which suffered a larger drop in SACU revenues following the financial crisis than Botswana and South Africa while being the only country of the three that up to that point had been collecting more revenues from SACU payments than was collected by its own tax agencies. A similar situation occurred in 2014, when a slowdown in South Africa’s economy again caused forecasts to be missed; Eswatini and Botswana’s SACU revenues declined as a percentage of GDP in 2015 and 2016, although this trend was reversed in 2017.

Box 2.4. SACU revenue

The Southern African Customs Union (SACU) incorporates Botswana, Eswatini, Lesotho, Namibia and South Africa. It is headquartered in Windhoek, Namibia. Its vision is “an economic community with equitable and sustainable development, dedicated to the welfare of its people for a common future”.

SACU is the oldest customs union in the world, having been founded in 1899 between the British colony of Cape of Good Hope and the Orange Free State Boer Republic. Subsequent agreements in 1910 and 1969 included Botswana, Eswatini and Lesotho. Following Namibia’s independence in 1990 and the end of apartheid in South Africa in 1994, new negotiations led to the current SACU agreement, which was signed in 2002.

The SACU agreement provides for free movement of SACU manufactured products within the union, without application of tariffs or duties. It also provides for common external tariffs and for the payment of customs and excise duties into a common pool to be shared between the SACU countries under the revenue-sharing formula set out in the Annex to the agreement.

The revenue-sharing agreement includes three components:

• A customs component, which divides the gross amount of customs duties according to the value of goods each country imports from other SACU countries in a given year (as a percentage of intra-SACU imports);

• An excise component, which divides the gross amount of excise duties according to each country’s GDP as a percentage of SACU’s combined GDP; and

• A development component, which is funded from 15% of the excise component and is weighted toward less developed SACU countries using a formula based on GDP per capita.

In this publication, revenue from excises, tariffs and customs duties are included as tax revenues in the SACU country that collected the revenue. They are included under headings 5121 (Excises) and 5123 (Customs and import duties) in the tax revenue tables. Revenues received from the SACU pool are included as miscellaneous revenue in the non-tax revenue tables, as seen in Table 6.13 for Eswatini. In the case of South Africa, where payments exceed the revenue share received from the SACU pool, the payments net of the share received are recorded as a memorandum item in the non-tax revenue table (Table 6.22).

Sources: (SACU, 2017_[4] ), (SACU, 2014_[5] ).

References

AUC/OECD (2018), Africa’s Development Dynamics 2018: Growth, Jobs and Inequalities, OECD Publishing, Paris/African Union Commission, Addis Ababa, https://dx.doi.org/10.1787/9789264302501-en.

IMF (2016), Heavily indebted poor countries (HIPC) initiative and multilateral debt relief initiative (MDRI): Statistical update, International Monetary Fund, http://www.imf.org/en/Publications/Policy-Papers/Issues/2016/12/31/Heavily-Indebted-Poor-Countries-Initiative-and-Multilateral-Debt-Relief-Initiative-PP5024.

Mansour, M. and G. Rota-Graziosi (2013), “Tax Coordination, Tax Competition, and Revenue Mobilization in the West African Economic and Monetary Union”, IMF Working Paper, Vol. 13/163, https://www.imf.org/en/Publications/WP/Issues/2016/12/31/Tax-Coordination-Tax-Competition-and-Revenue-Mobilization-in-the-West-African-Economic-and-40756.

Mongardini, J. et al. (2011), In the wake of the global economic crisis : adjusting to lower revenue of the Southern African Customs, International Monetary Fund, https://www.imf.org/external/pubs/ft/dp/2011/afr1101.pdf.

OECD (2018), Global Outlook on Financing for Development, OECD Publishing, http://www.oecd.org/dac/global-outlook-on-financing-for-sustainable-development-2019-9789264307995-en.htm.

OECD (2017), Revenue Statistics 2017, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264283183-en.

OECD/AfDB/UN ECA (2010), African Economic Outlook 2010, OECD Publishing, Paris, https://dx.doi.org/10.1787/aeo-2010-en.

OECD et al. (2018), Revenue Statistics in Latin America and the Caribbean 2018, OECD Publishing, Paris, https://dx.doi.org/10.1787/rev_lat_car-2018-en-fr.

RFI (2014), Le gros chèque d’Arcelor Mittal au Sénégal, http://www.rfi.fr/economie/20140605-le-gros-cheque-arcelor-mittal-senegal.

SACU (2017), Southern African Customs Union Agreement 2002 (As amended on 12 April 2013), http://www.sacu.int/list.php?type=Agreements.

SACU (2014), Implementing a Common Agenda Towards Regional Integration: 2014 Annual Report, http://www.sacu.int/docs/reports_annual/2014/annual_report.pdf.

Notes

← 1. There is no comparison with the averages of the 24 Latin American and Caribbean (LAC) countries (see (OECD et al., 2018_[8] )) and the 36 OECD countries (see (OECD, 2017_[7] )) (OECD, 2018_[11] )) as these publications do not cover non-tax revenues.

← 2. Please see the Interpretative Guide to non-tax revenue in Annex Annex B of this report for the definitions of these revenue categories.

← 3. World Bank Country and Lending Groups are reported at the following website datahelpdesk.worldbank.org/knowledgebase/articles/906519

← 4. In 2019, Senegal graduated from low-income to lower-middle-income status, the only country in this publication to change income status. Here, we are using the July 2018 classification because it is based on data for 2017.

← 5. The official definition of ODA is: “flows to countries and territories on the Development Assistance Committee List of ODA Recipients and to multilateral institutions which are provided by official agencies, including state and local governments, or by their executive agencies. In addition, each transaction must be administered with the promotion of the economic development and welfare of developing countries as its main objective; and be concessional in character and conveys a grant element of at least 25% (calculated at a rate of discount of 10%)”. Further information is available at: www.oecd.org/dac/financing-sustainable-development/index-terms.htm#ODA

← 6. A full list of countries and organisations reporting statistics to the DAC is available at : www.oecd.org/dac/financing-sustainable-development/development-finance-standards/dacdatasubmitters.htm

← 7. The World Bank refers to “natural resource rents”. This is calculated as the difference between the value of natural resource production in the countries according to commodity prices and the estimated cost of resource extraction. It is not to be confused with the definition of resource rent used by the OECD, for which it is defined as a form of government revenue. Please see the Interpretative Guide for more details.

← 8. See paragraphs 9-13 of the OECD Interpretative Guide in Annex Annex A for an explanation of how administrative fees are classified in this publication.