Indirect Taxation and Gender Equity: Evidence from South Africa

Transcription

1 Indirect Taxation and Gender Equity: Evidence from South Africa South African Country Paper January 2009 Daniela Casale School of Development Studies, University of KwaZulu-Natal Note:Part of the work emanating from the project has been accepted for publication in Feminist Economics 2012 [International Association for Feminist Economics] 1

2 CONTENTS 1. INTRODUCTION 2. BACKGROUND AND SOUTH AFRICAN CONTEXT 2.1 Indirect tax incidence studies 2.2 The tax structure in South Africa 3. METHODOLOGY AND DATA 3.1 Indirect tax incidence methodology 3.2 Description of the data sources 3.3 Description of indirect taxes in South Africa 3.4 Identifying the gender incidence of indirect taxes 4. RESULTS 4.1 Total indirect tax incidence 4.2 Incidence by urban/rural area and by race 4.3 Incidence by quintile and by presence of children 4.4 Incidence by consumption category 5. SIMULATIONS AND POLICY SUGGESTIONS 6. CONCLUDING REMARKS 7. REFERENCES 8. APPENDIX TABLES 2

3 1. INTRODUCTION This study is part of a multi-country project on Making Tax Reforms Work for Women: Mobilizing Taxes for Gender Equality and Women s Empowerment. The main objective of the project is to investigate both explicit and implicit forms of gender bias (see Stotsky 1997a, 1997b) in tax systems in countries of varying levels of development (South Africa, India, Argentina, Mexico, United Kingdom, Morocco, Ghana and Uganda). Explicit bias arises due to specific provisions in the tax law that treat women and men differently. These are typically found in direct taxes. Implicit bias occurs when provisions of the tax law have a differential impact on women and men due to gendered social or economic behaviour, even though the tax law contains no explicit bias. This form of bias is typically found in indirect taxes (i.e. taxes levied on goods and services). The project focuses on the gendered impact of two main types of taxes: personal income taxes (which are direct taxes) and indirect taxes (in particular VAT, excises and the fuel levy). The findings from the personal income tax analysis for South Africa are available in the country paper by Budlender and Valodia (2007). This paper covers the indirect tax incidence analysis for South Africa. The overall findings from the South African case study suggest that the tax reforms of the past two decades have gone a long way in eliminating both explicit and implicit forms of bias in the South African tax system. This is particularly the case for the personal income tax system, while the indirect tax incidence analysis suggests that there may still be some small room to further eliminate the burden of indirect taxes on poor female households. This paper is organised as follows. In the next section, some background to indirect tax incidence studies is provided, as well as an outline of the tax structure in South Africa. Section 3 describes the data and the methodology used, including an explanation of how we identify the gendered incidence of indirect taxes. The results of the incidence analysis are presented in Section 4, while Section 5 provides some discussion on further 3

4 possible policy changes to the indirect tax system. Concluding remarks are made in Section BACKGROUND 2.1 Indirect tax incidence studies There is a large literature, including some studies in developing countries, that examines the incidence of indirect taxes, i.e. the question of who ultimately bears the burden of taxes on goods and services (Bird and Miller 1989; Younger 1993; Younger et al 1999; Ahmad and Stern 1991; Gibson 1998; Rajemison et al 2003; Sahn and Younger 1998; 2003). There have been a few studies in South Africa that have tried to model the impact of value-added taxes in particular (Fourie and Owen 1993; Alderman and del Ninno 1999; Go et al 2005), and one more extensive study by Woolard et al (2005), which explored the incidence of indirect taxes using data from 2000 as part of a more general study of tax incidence for the South African National Treasury. The focus of most of these studies, however, and certainly in the South African case, has been on the incidence of indirect taxes by income group, i.e. on the regressivity or progressivity of the indirect tax system. The goal of this paper and of the broader project has been to extend this work by exploring the gender impact of indirect taxes, and particularly the impact on poorer women, and women living with children. The study of the (gender) implications of indirect taxes is important for a number of reasons. Indirect taxes make up a large portion of the government s tax revenue, especially in developing countries. In South Africa these taxes make up 40 per cent of total government tax revenue, although this is low by developing country standards where the share is generally between 50 and 60 per cent (Barnett and Grown 2004). In addition, there is a global pattern of indirect taxes increasing as a share of total government revenue, as it has become more difficult to tax companies and individuals 4

5 due to the increased mobility of labour and capital (Khattry and Rao 2002; Barnett and Grown 2004; Aizenman and Jinjarak 2006). The indirect tax base is also very wide. While personal income taxes (PIT) and other direct taxes affect only a small percentage of the population, indirect taxes - because they are levied on consumption - will affect most people. This point is particularly pertinent to a gendered analysis of taxes in South Africa. Unemployment rates among women are currently around 50 per cent, women are far more likely to be engaged in informal work than men, and even where women are working in the formal sector of the economy, their earnings are less likely to be above the tax threshold than men s (Casale and Posel 2005). Using 2005 labour force data, Budlender and Valodia (2007) estimate that 73 per cent of employed women compared to 65 per cent of employed men fell outside of the tax net, and of the PIT paid, women s contributions accounted for only 30 per cent. While there is no explicit bias in the indirect tax system (tax authorities don t have different VAT rates by group, whereas they often do have different PIT rates by group), there will be implicit bias as people have different spending patterns and so will bear the burden of the tax in different proportions. This is particularly relevant in the gender context, as much research now shows that men and women have different spending priorities when they control resources. 2.2 The tax structure in South Africa The structure of taxes in South Africa over the last two decades is shown in Table 1. In the most recent period, direct taxes form around 60 per cent of total revenue, with personal income taxes and corporate taxes being the two largest contributors (30 per cent and 27 per cent of revenue respectively). Indirect taxes make up just under 40 per cent of total revenue, with the main component, VAT, contributing 26 per cent of total revenue. VAT, excises and fuel taxes the indirect taxes that we investigate in this project - jointly make up 33 per cent of total revenue. Unlike other developing countries, South Africa does not currently rely heavily on trade taxes for government revenues. 5

6 Some important changes are evident over the decade, especially the shift away from indirect to direct taxes. This has been driven predominantly by increasing tax revenues from corporations, especially over the last decade as profit rates rose. Although as a proportion of total revenue, individual taxes have not increased, tax revenues from PIT have risen substantially. These changes have been partly due to a marked increase in the efficiency of the tax authority, the South African Revenue Services (SARS), which has been able to extend the tax base and improve the collection rate substantially. Since the political transition, the ANC government has been able to increase the tax/gdp ratio from 23 per cent in 1993 to 28 per cent in 2007/8 (above the targeted rate of 25 per cent) (National Treasury, 2008), thereby creating the fiscal space for increased expenditure. It is worth noting that actual tax collections by SARS have in recent years exceeded budget projections. In 2007/2008 a budget surplus of 1 per cent of GDP was recorded and a surplus of 0.8 per cent was projected for 2008/09 (National Treasury, 2008). 1 Table 1: Tax Structure, South Africa, / / /08 Tax/Source of revenue Revenue % of total Revenue % of total Revenue % of total raised Tax Raised tax raised tax in R'm Revenue in R'm Revenue in R'm revenue Individuals % % % Companies % % % Other 657 1% % % Total direct taxes % % % VAT/GST % % % Excise duties % % % Fuel levy % % % Customs duties % % % Other % 4 044,1 2% % Total indirect taxes % % % Total tax revenue % % % Source: Budlender and Valodia (2007) from National Revenue Accounts, National Treasury, South Africa 1 More detail (also from a gender perspective) on the tax structure and tax reform over the past few decades can be found in the country paper by Budlender and Valodia (2007) and in the synthesis chapter by Budlender, Casale and Valodia (2009). 6

7 3. METHODOLOGY AND DATA 3.1 Indirect tax incidence methodology The methodology most commonly used for calculating the incidence of indirect taxes involves estimating the amount of tax paid by households indirectly through information on their spending behaviour. Most countries, including South Africa, conduct household surveys in which they collect detailed data on households expenditure patterns. The post-tax expenditure values available in these surveys are used with corresponding tax rate and price information for the year in question to calculate the amount of tax paid by each household on each consumption item. Assuming that the tax is shifted forward entirely onto the consumer, the amount of tax paid per item can be calculated as follows where the tax is ad valorem: taxpaidv ij = rate j * (expend ij /(1 + rate j )) where rate j is the tax rate on item j and expend ij is the reported expenditure for household i on item j. For a unit tax, the amount of tax paid by the household per item is calculated as: taxpaids = (expend / price ) * duty ij ij j j where duty j is the per unit duty on item j and price j is the retail price of that item. Tax incidence is then calculated as the percentage of total household consumption expenditure spent on the tax for that item, or in total. The convention in the international literature on tax incidence is to use consumption expenditure rather than income as the base as it is a better measure of wellbeing if households engage in consumption smoothing. A more practical reason for using consumption expenditure rather than income here was that not all countries in the project could access (reliable) income data. 7

8 3.2 Description of the data sources The expenditure data that we use to calculate the tax incidence for South Africa are drawn from the Income and Expenditure Survey (IES) of 2000, a household survey conducted by the national statistical agency, Statistics South Africa (SSA). The IES, which is predominantly used to update the CPI weights, is conducted every five years among a nationally representative sample of about households. It contains very detailed information on the spending patterns of households, with data collected on around 500 expenditure items through face-to-face interviews. 2 There has been some concern expressed about the quality of the data from the 2000 survey (see Simkins 2004), but predominantly on the income information collected. We therefore use only the expenditure data from the IES. We use a cleaned version of the dataset (prepared by Global Insight) which has had many of the anomalies corrected or removed, and we also use revised and updated sampling weights based on the 2001 Census provided by Statistics South Africa (which deal with some of the sampling issues that were of concern). 3 The tax rate and price information that we use to calculate the tax incidence per item was gathered from various government sources: National Treasury Budget Review 2000; South African Revenue Services VAT Guide for Vendors; and the Statistics South Africa retail price survey for The SSA report on the IES 2000 includes the following definition of the respondent/s: The person (or persons) responding in this interview should be a member/members of the household who is/are likely to do the purchases for the household or know the answers to our questions. (Statistics SA 2002: 91). 3 The more recent 2005 IES was released in early 2008, but we have chosen not to use the updated survey information here as some of the expenditure data are not considered reliable (SSA 2008). In particular, the share of spending on food was found to be much lower than in 2000 across all quintiles in the distribution (and compared to other countries of similar levels of development), which would effect our incidence results substantially. 4 The author would like to thank Morné Oosthuizen and Ingrid Woolard who shared their price and excise duty data. 8

9 3.3 Description of indirect taxes in South Africa As mentioned earlier, indirect taxes contribute just under 40 per cent of tax revenue in South Africa. The main component is value-added tax (VAT), which accounts for 25.7 per cent of total tax revenue, with much smaller shares derived from excise duties (3.4 per cent), the fuel levy (4.2 percent) and customs duties (4.7 per cent) (see Figure 1 below). For this study, we analyse the incidence of VAT, excise duties and the fuel levy only. A brief description of these taxes is provided below (Table 2 contains details). Figure 1. Composition of tax revenue in South Africa, 2007/08 Customs duties 4.7% Fuel levy 4.2% Other indirect 0.3% Excise duties 3.4% Individuals 29.5% VAT 25.7% Other direct 3.8% Companies 28.3% Source: Own calculations from Budget Review 2007/8, National Treasury In South Africa, VAT is a multi-stage single-rate tax levied on the consumption of most goods and services (whether they are produced locally or imported). The VAT rate has remained at 14 per cent on the value of most goods and services since 1993, although there are a number of zero-ratings and exemptions. The following goods and services are zero-rated: 19 basic food items (among them brown bread, eggs, vegetable oil, grains, rice, milk, fresh fruit and vegetables, dried legumes, canned fish), illuminating paraffin, goods which are subject to the fuel levy (petrol and diesel), international transport services, farming inputs, sales of going concerns and certain government grants. The 9

10 zero-rating of basic food items and paraffin/kerosene (used predominantly by the poor as a fuel for cooking, lighting and heating) was implemented specifically to alleviate the burden of VAT on poorer households. 5 The goods and services which are VAT exempt are residential rental and accommodation; educational services (including crèches); public road and rail transport; non-fee related financial services; and medical aid and medicine/medical services provided by public health institutions. Unlike with goods that are zero-rated, suppliers of VAT-exempt goods are not able to claim back the input VAT. This implies that, to the extent that the inputs attract VAT themselves, some of the VAT may be passed on to the final consumer. An effective rate would be between zero and 14 per cent. However, for this project, we rate these goods at zero per cent, given that the largest input cost in these sectors is likely to be labour. 6 Specific unit excise duties are levied on sorghum meal, tobacco products, and nonalcoholic and alcoholic beverages. Details of these duties are provided in Table 2, although it is important to note here that the taxes on tobacco and alcoholic beverages are particularly high. The fuel levy is also a unit tax, levied at cents per litre of petrol and 89.4 cents per litre of diesel. For this study, we calculate the incidence of the fuel levy on petrol and diesel for household use and for private transport only. We do not estimate the impact of a transfer of the fuel levy onto the consumer where fuel is an input in other production processes. However, we do make a rule-of-thumb adjustment for the public transport sector, where it is assumed that the total amount of the fuel levy is 5 Paraffin was only zero-rated in April of Although our data are from October 2000, we have calculated tax incidence as if the zero-rating had applied in 2000 i.e. using the spending behaviour information of households on paraffin from We do this to get a more realistic picture of the current incidence on the poor especially. However, this assumption ignores any knock-on effects that an effective reduction in the price of paraffin would have on other spending patterns. 6 Another way of approaching this would be to estimate the likely effective VAT rate by using a detailed input-output table for South Africa. This is not only beyond the scope of the project, but would also lead to a loss of detail (and precision) as the IES has more detailed expenditure categories than the input-output table for SA and the categories do not correspond exactly with each other. 10

11 passed on to the consumer and that fuel constitutes 30 per cent of input costs in this sector. Table 2. Indirect tax rates and specific duties Tax Item Ad valorem rate/specific duty VAT VAT-rated Most goods and services (incl. imports) 14% Zero-rated goods Exempt goods -19 basic food items (brown bread, dried mielies and mealie rice, brown bread flour, samp, eggs, fruit, vegetables, dried beans, lentils, maize meal, rice, pilchards in tins or cans, vegetable cooking oil, milk, cultured milk, milk powder and dairy powder blend, edible legumes and pulses of leguminous plants i.e. peas, beans and peanuts) -Paraffin -Exports -Petrol and diesel -Farming inputs -Sales of going concerns -Certain grants by government -Residential rental and accommodation -Educational services (including creches) -Public road and rail transport -Non-fee related financial services -Medical aid and medicine/medical services provided by public health institutions 0% Assumed to be 0% Excise duties Preparations of sorghum for making beverages 33 cents/kg Mineral water and non-alcoholic beverages 8 cents/litre Beer 2239 cents/litre of absolute alcohol Sorghum beer 745 cents/100 litres Unfortified wine 6790 cents/100 litres Fortified wine cents/100 litres Sparkling wine cents/100 litres Spirits /100 litres of absolute alcohol Cigars cents/kg Cigarettes cents/10 cigarettes Cigarette tobacco 6412 cents/kg Pipe tobacco 3893 cents/kg Fuel levy a Petrol cents/litre Diesel 89.4 cents/litre Source: Budget Review 2000, Department of Finance, South Africa Notes: a The levy consists of a fuel levy component and a Road Accident Fund component. 11

12 3.4 Identifying the gender incidence of indirect taxes The biggest methodological challenge that we faced in the project was how to estimate the gender incidence of indirect taxes. This is because sex is an individual attribute, but expenditure is collected at the household level in most surveys (and often occurs at the household level, especially where spending is on indivisible/public goods). It is common practice in the literature which estimates the incidence of indirect taxes on individuals by race or by income, for example, to simply assume equal sharing in the household. So if the total amount of indirect tax paid is R1000 a month in a family of four, individual incidence would be equal to R250 per person. Assuming equal sharing in the household and calculating an individual incidence on that basis did not seem satisfactory for a study on the gender impact of taxes, given that intra-household allocation of resources is not always equal. The project considered adopting different sharing rules for different classes of goods, but this proved to be highly contentious and in the end largely arbitrary, as for most countries little case study (or other) research exists on the intra-household allocation of resources that could inform our choice of sharing rules. Instead, it was decided among the project participants to use an alternative approach to measuring the gender impact of taxes that would also be more feasible for a cross-country comparative study. This involved classifying households as being either more female or more male and then analysing the tax incidence on the individuals within these households. We use three definitions to classify households as being male-type or female-type households. The first simply takes into account the presence of male and female adults in the household; the second and third try to take into account gendered spending power in the household by adding the dimension of control over resources, measured through employment status and household headship. Details are provided below and in Table 3, which shows the distribution of individuals across the various household types. 12

13 Table 3. Distribution of individuals across household categories and by quintile (%) Quintile All Q1 Q2 Q3 Q4 Q5 Presence of adults Adult male majority Adult female majority Equal number adults Employment status Male breadwinner Female breadwinner Dual earner No employed Headship Male-headed Female-headed Average p.c. monthly expenditure per quintile R66.26 R R R R Source: Own calculations from IES 2000 Notes: Data are weighted. The first definition, which uses the presence of male and female adults (aged 18 years and older) to classify households by gender, results in three categories of household: adult female majority households (where adult females outnumber adult males), adult male majority households (where adult males outnumber adult females) and equal number adult households. In South Africa, 42 per cent of individuals live in adult female majority households, 22 per cent live in adult male majority households with the remaining 36 per cent living in households where there are an equal number of adult males and females. The employment status definition 7 classifies households into four categories: female breadwinner households with at least one employed adult female and no employed adult males; male breadwinner households which contain at least one employed adult male and no employed adult females; dual earner households with at least one employed adult male and one employed adult female, and households with no employed. In South 7 In the IES 2000, employment status is based on the following question and prompt, During the past seven days, did do any work for pay, profit or family gain? Formal/informal work, working on a farm, casual/seasonal work, etc. 13

14 Africa, this latter group consists mostly of households where either pensions and grants (predominantly through the government s social welfare programme) or remittances from migrant workers form the main source of income (87 per cent of these households) (own calculations from the South African Labour Force Survey of September 2001). There are more individuals living in male breadwinner households (26.4 percent) compared to female breadwinner households (21.6 per cent), with another 24.2 per cent living in dual earner households. It is not surprising in a country with an unemployment rate of around 40 per cent for the last decade (using an expanded definition that includes the non-searching unemployed) that the largest proportion of individuals in South Africa (27.8 per cent) live in households where there are no employed members. The headship classification categorises households as either male-headed or femaleheaded. The following excerpt from the SSA report on the IES 2000 provides the definition of headship used in the survey: At Statistics SA we have a clear definition of a household head. Respondents may have a different idea of what household head means, and you must explain to them what Stats SA wants. The head is the person in whose name the dwelling is registered. It may be the person who owns the dwelling, or is responsible for the rent, or gets the dwelling through their work, or through their relationship to the owner. If two or more persons have equal claim to be head of the household, or if people state that they are joint heads or that the household has no head, then choose the eldest as the head. A head of a household must be a member of the household. If the head of the household is an absentee head i.e. he/she does not reside at the dwelling unit for at least 4 nights a week, then choose the spouse of the head. If the spouse is not a household member, choose the oldest resident person as the head. (Statistics SA 2002: 90). It is interesting to note that a large number (41 per cent) of individuals in South Africa live in female-headed households. This is due to, among other reasons, a high incidence of labour migration, particularly among men (resulting in the female in the household being reported as the de facto resident head), and relatively low levels of marriage and partnership in South Africa. 14

15 Also evident from Table 3 is that the female-type households and those with no employed tend to be among the less well-off, concentrated in the lower quintiles of the expenditure distribution. In contrast, the male-type households, the dual earner households, and the equal number adult households are more heavily concentrated at the upper end of the expenditure distribution. This means that any tax policy that has positive gender equity implications, will also result in strong income equity outcomes. The last row of Table 3 shows the average monthly per capita expenditure of households in quintiles 1 to 5. These figures highlight the very unequal distribution of expenditure in South Africa, with the relative increase in average expenditure from quintile 4 to quintile 5 being much larger than the increases across the lower quintiles (a five-fold increase compared twofold increases across the lower quintiles). For South Africa, there is a large overlap across the three gendered household classifications, evident from the cross-tabulations in Table 4. For example, just over 80 per cent of female-headed households fall into two employment status categories: 40.7 per cent are in the female breadwinner category and 40.6 per cent are in the no employed category. The majority of female-headed households, 71 per cent, are in the category of female adult majority households. About 73 percent of female adult majority households fall into the categories of female breadwinner (36.6 per cent) and no employed households (35.9 per cent). The tax incidence results, which are presented in the next section, therefore tell a similar story regardless of the gendered household definition that is used. Before moving on to the empirical results, it is important to reiterate the key limitation of our analysis: that we are unable to estimate an individual incidence for men and women because we do not have individual level information on expenditure or consumption. So while we refer to implicit bias in favour of or against male-type households, it is important to recognise that women living in those households will also bear part of the tax burden. Table 5 below shows the distribution of males and females across household types. To take one example which illustrates this point: although the majority of women 15

16 live in either female breadwinner or no employed households, 21 per cent and 23 per cent live in male breadwinner and dual earner households respectively. Table 4. Cross-tabulations of individuals by household classification a) Employment status by headship Male-headed Female-headed Male breadwinner Female breadwinner Dual earner No employed b) Presence of adults by headship Male-headed Female-headed Adult male majority Adult female majority Equal number adult c) Employment status by presence of adults Adult male majority Adult female majority Male breadwinner Female breadwinner Dual earner No employed Source: Own calculations from IES 2000 Notes: Data are weighted. Equal number adults Table 5. Distribution of males and females across household classifications MALES FEMALES Number % Number % Presence of adults Adult male majority Adult female majority Equal number adults Employment status Male breadwinner Female breadwinner Dual earner No employed Headship Male-headed Female-headed Source: Own calculations from IES 2000 Notes: Data are weighted. 16

17 4. RESULTS 4.1 Total indirect tax incidence The results of the incidence analysis are presented in Tables 6 to 10 and in the Appendix Tables A1 to A7. Table 6 reports the overall tax incidence for the different household types using the three gendered definitions of households. Due to the strong correlations across household categories reported above, the story that emerges from these results is consistent regardless of which household definition is used. Total indirect tax incidence is lower in female-type households than in male-type households, by around a full percentage point on a base of approximately 8 per cent. This result holds for the different types of taxes as well, i.e. VAT, excise duties and the fuel levy. The pattern of incidence among households with no employed members is similar to the pattern among femaletype households, while the dual earner and equal number adult households resemble the male-type households in their tax incidence. Table 6. Overall incidence by household types (tax as a percentage of expenditure) Total Tax VAT Excise Tax Number of Fuel Tax Households Headship Male headed *9.06 *7.17 *0.96 * Female headed Employment Categories Male breadwinner *9.36 *7.36 *1.12 * Female breadwinner Dual earner *9.15 *7.13 *0.89 * None employed *7.84 *6.99 *0.49 * Household Sex Composition Adult male majority *9.23 *7.29 *1.1 * Adult female majority Equal # adult *8.84 *7.12 *0.85 * Source: Own calculations from IES 2000 Notes: Data are weighted. * Reports statistical significance in equality of means t-tests with unequal variance at 5% level. Reference category in italics. For example, tax incidence in female-headed households is tested against tax incidence in male-headed households. 17

18 While there are statistically significant gender differences for all three types of taxes, the largest gender differentials are reported for the excise duties and the fuel levy. Most of the implicit bias against male-type households is being driven by the larger expenditure in these households on alcohol and tobacco and on fuel for private transport. The gender difference in the incidence of the fuel levy would have been even more pronounced if we had not adjusted for the passing on of the fuel levy to consumers in the public transport sector. This is because female-type households are relatively more intensive users of public transport, while male-type households are relatively more intensive users of private transport. 4.2 Incidence by urban/rural area and by race In this and the following sub-sections only the results for the employment status categories are presented. The full set of results for the other two definitions is available in the Appendix in Tables A1 A7. Table 7 reports on the incidence of indirect taxes within urban and rural areas. In both urban and rural areas, total incidence is higher in male-type and dual earner households compared to female-type and no employed households. In rural areas, this is the case for all three types of taxes, i.e. VAT, excises and the fuel levy; whereas in urban areas this result is being driven by the higher incidence of excise taxes and the fuel levy on male-type and dual earner households. In urban areas, no employed households bear the highest incidence of the VAT compared to the other household types, while dual earner households bear the lowest incidence of the VAT. This explains why the incidence of VAT in female-headed households is higher than in male-headed households in urban areas, because a large proportion of female-headed households contain no employed members (see Table A1). For male-breadwinner, female-breadwinner and no employed households, indirect tax incidence is higher in urban areas than in rural areas. In dual earner households, tax 18

19 incidence is higher in rural areas, driven mostly by the higher incidence of VAT and excise duties on this type of household in rural areas. Table 7. Incidence by employment status in urban/rural areas (tax as a percentage of expenditure) URBAN RURAL Employment categories Total Tax VAT Excise Tax Fuel Tax # of HHs Total Tax VAT Excise Tax Fuel Tax # of HHs Male breadwinner *9.43 *7.34 *1.1 * *9.22 *7.39 *1.18 * Female breadwinner Dual earner *9.1 *7.03 *0.81 * *9.27 *7.33 *1.05 * None employed 8.45 *7.38 *0.59 * *7.48 *6.76 *0.42 * Source: Own calculations from IES 2000 Notes: Data are weighted. * Reports statistical significance in equality of means t-tests with unequal variance at 5% level. Reference category in italics. For example, tax incidence in female-breadwinner households is tested against tax incidence in male-breadwinner households within urban and within rural areas. Table 8 presents the incidence results by race group. In South Africa, the majority of the population, 78 per cent, is African, with a further nine and ten percent of the population Coloured (or of mixed race) and White respectively. Just under three per cent of the population is of Asian or Indian descent. The results suggest that for all four race groups, male-breadwinner households bear a larger tax burden than female-breadwinner households, overall, and for the specific types of taxes. However, some of the differences are not statistically significant, especially for the Indian race group, because of a very small number of observations in these categories. 8 In general, individuals living in Indian and Coloured households tend to have a higher indirect tax incidence than in African and White households. Coloured male-breadwinner and dual earner households bear the highest incidence of total indirect taxes in South Africa, driven largely by the VAT and excise tax incidence. Indian male-breadwinner and White dual earner households bear the highest incidence of the fuel levy. The findings are 8 There are also some exceptions to this general finding when analysing the results for the other household definitions. Table A2 in the Appendix shows that Coloured and Indian female-headed households bear a higher VAT incidence than Coloured and Indian male-headed households. And Indian and White femaledominated households bear a higher VAT burden than Indian and White male-dominated households, but these latter differences are not statistically significant. 19

20 likely to be driven more by the income class that these households fall into than their race. African households are concentrated at the lower end of the income distribution, while Indian and especially White households are concentrated at the upper end of the income distribution. Coloured households tend to fall in the middle of the distribution and, as we will see in the next section, tax incidence falls most heavily on the middle quintiles in South Africa, except for the fuel levy, which is highly progressive. Table 8. Incidence by employment status and race (tax as a percentage of expenditure) AFRICAN COLOURED Employment Total Excise Total Excise Categories Tax VAT Tax Fuel Tax # of HHs Tax VAT Tax Fuel Tax # of HHs Male breadwinner *9.32 *7.44 *1.13 * *10.36 *7.98 *1.63 * Female breadwinner Dual earner *9.11 *7.25 *0.87 * * *1.32 * None employed *7.7 *6.93 *0.46 * *9.44 *8.15 *0.87 * INDIAN WHITE Employment Categories Total Tax VAT Excise Tax Fuel Tax # of HHs Total Tax VAT Excise Tax Fuel Tax # of HHs Male breadwinner 9.84 * * * Female breadwinner Dual earner *0.66 * None employed 9.03 * * *8.84 * Source: Own calculations from IES 2000 Notes: Data are weighted. * Reports statistical significance in equality of means t-tests with unequal variance at 5% level. Reference category in italics. For example, tax incidence in African female-breadwinner households is tested against tax incidence in African male-breadwinner households. 4.3 Incidence by quintile and by presence of children The results in Table 9 indicate that female-breadwinner households and those with no employed bear a lower tax incidence than male-breadwinner and dual earner households, regardless of which expenditure quintile the households are in. This is the case for total indirect taxes and for the different types of taxes. For all four of the employment status household categories, total indirect tax incidence tends to fall most heavily on the middle quintiles, particularly quintiles three and four, 20

21 with the poorest quintile paying a smaller share of expenditure on tax than the richest quintile. For VAT and excise duties, the incidence is predominantly on the middle quintiles, while the fuel levy is strongly progressive. 9 Table 9. Incidence by employment status and quintile (tax as a percentage of expenditure) Total Tax VAT Excise Tax Fuel Tax # of HHs Total Tax VAT Excise Tax Fuel Tax # of HHs Quintile Male breadwinner Female breadwinner 1 *8.17 *6.98 * *8.95 *7.4 *1.08 * *9.64 *7.78 * *9.92 *7.53 *1.31 * *9.36 *6.97 *0.99 * Total *9.36 *7.36 *1.12 * Dual earner None employed 1 *7.95 *6.73 * *6.39 *0.39 * *9.24 *7.45 *1.23 * *7.82 *7.12 *0.44 * *9.5 *7.7 *1.04 * * *0.59 * *10.07 *7.78 *1.01 * *8.96 *7.67 *0.72 * * *0.61 * *8.72 *6.73 *0.56 * Total *9.15 *7.13 *0.89 * *7.84 *6.99 *0.49 * Source: Own calculations from IES 2000 Notes: Data are weighted. * Reports statistical significance in equality of means t-tests with unequal variance at 5% level. Reference category in italics. So, for example, tax incidence in female-breadwinner households in quintile one is tested against tax incidence in male-breadwinner households in quintile one When disaggregated by the presence of children (aged 17 years or younger) in the household, some differences in the patterns of tax incidence across the quintiles emerge. These results are presented in Table 10, but the regressivity / progressivity of the tax is more clearly visible in Figures 2 to 5 below. For both male-breadwinner and femalebreadwinner households without children, the incidence of excise duties tends to be more regressive and the VAT incidence more proportional (i.e. the inverted-u shape of the VAT curves is less pronounced) compared to those households with children. 9 These results are largely consistent with those in Woolard et al (2005). Using the data from the IES 2000, they find that the VAT and excise incidence falls largely on the middle deciles when expressed as a percentage of total expenditure. However, when expressed as a percentage of total income, they find the incidence of these taxes to be regressive. They note that this appears to be an artifact that is the result of the large mismatch between income and expenditures in the bottom and top deciles (Woolard et al 2005: 46). Because of concerns with the reliability of the income data in this survey, we do not try to replicate our results using income as a base when calculating tax incidence. 21

22 Figure 2. Total tax incidence by employment status, quintile and presence of children Tax/exp share (%) Quintile Male W Child Male W/O Child Female W Child Female W/O Child Dual W Child Dual W/O Child None W Child None W/O Child Figure 3. VAT incidence by employment status, quintile and presence of children 8.5 Tax/exp share (%) Quintile Male W Child Male W/O Child Female W Child Female W/O Child Dual W Child Dual W/O Child None W Child None W/O Child 22

23 Figure 4. Excise incidence by employment status, quintile and presence of children 3 Tax/exp share (%) Quintile Male W Child Male W/O Child Female W Child Female W/O Child Dual W Child Dual W/O Child None W Child None W/O Child Figure 5. Fuel levy incidence by employment status, quintile and presence of children Tax/exp share (%) Quintile Male W Child Male W/O Child Female W Child Female W/O Child Dual W Child Dual W/O Child None W Child None W/O Child 23

24 Table 10. Incidence by employment status, presence of children, and quintile (tax as a percentage of expenditure) Quintile Total Tax VAT Excise Tax Fuel Tax # of HHs Quintile Total Tax VAT Excise Tax Fuel Tax # of HHs Male Breadwinner WITH children Male Breadwinner WITHOUT children 1 *8.13 *6.97 * *1.85 * *8.88 *7.38 *1.01 * * *1.85 * *9.61 *7.83 * *9.82 *7.51 * * *1.12 * *10.22 *7.51 * *8.83 *6.64 *0.65 * *9.84 *7.26 *1.29 * Total *9.14 *7.34 *0.98 * Total *9.95 *7.39 *1.51 * Female Breadwinner WITH children Female Breadwinner WITHOUT children Total Total Dual Employed HHs WITH children Dual Employed HHs WITHOUT children 1 *7.94 *6.72 * *1.35 * *9.22 *7.45 *1.19 * *9.72 *7.43 * *9.48 *7.71 *0.97 * *9.79 *7.67 *1.79 * *10.02 *7.73 *0.95 * *10.35 *8.09 *1.39 * * *0.56 * * *0.74 * Total *9.11 *7.14 *0.86 * Total *9.39 *7.02 *1.03 * None Employed WITH children None Employed WITHOUT children 1 *6.98 *6.38 *0.37 * * * *7.83 *7.14 *0.43 * *7.76 * * * * * *0.93 * *7.87 *0.42 * *9.01 *7.46 *1.04 * * * * *0.62 * Total *7.7 * * Total *0.85 * Source: Own calculations from IES 2000 Notes: Data are weighted. * Reports statistical significance in equality of means t-tests with unequal variance at 5% level. Reference category in italics. So, for example, tax incidence in female-breadwinner households with children in quintile one is tested against tax incidence in male-breadwinner households with children in quintile one. With few exceptions, the main gender findings hold: regardless of the presence of children in the household or the quintile, female-breadwinner households and those with 24

25 no employed members bear a lower incidence than male-breadwinner and dual-earner households, for total indirect taxes and for the different types of taxes. 10 Within each household category, households with children bear a lower total indirect tax burden than those without children, driven mostly by the differences in the incidence of excise duties and the fuel levy. There are some exceptions to this when analyzed by quintile: for example, female breadwinner and no employed households with children in the middle quintiles have a higher total tax incidence than those female breadwinner and no employed households without children. This is generally being driven by the VAT and fuel levy incidence being higher in those quintiles among the households with children. 4.4 Incidence by consumption category Table A5 reports total indirect tax incidence by consumption category for the employment status categories (note that this table and the table with the results for the headship and household composition categories are in the Appendix due to space constraints here). A comparison of the results for the male- and female-breadwinner categories only is graphed below in Figure 6. We allocate the large number of consumption items in the IES into 25 main categories (loosely based on the United Nations Classification of Individual Consumption According to Purpose or COICOP). We find that, even though female-type households bear a lower overall indirect tax incidence, some interesting gender biases do emerge when the data are disaggregated into consumption categories. The gender differences that emerge are largely consistent with the broader international literature on gendered spending patterns. Female breadwinner households bear a greater 10 Two exceptions are that female-breadwinner households without children in the lowest two quintiles bear a higher incidence of the fuel levy than the other categories of household without children in those quintiles, and no employed households with children in the highest quintile bear a higher burden of the fuel levy and VAT than most other household types in that quintile. 25

26 tax incidence on food (non zero-rated items as well as sugar/confectionary items), utilities, children s clothing, personal care items (both necessity and other more nonessential items), fuel for household use, and education (although education is exempt, textbooks and stationery attract VAT in this category). Male breadwinner households bear a greater tax incidence on meals out, non-alcoholic beverages, alcoholic beverages (particularly beer), tobacco, adult s clothing, private transport, fuel for transport, medical expenditure (as private health care attracts VAT), communication and recreation. Again, dual earner households in the most part resemble the male breadwinner households, and no employed households resemble the female breadwinner households in their spending patterns. Consumption items for which taxes are generally more progressive (for all household types) are housing, meals out, private transport, fuel for transport, communication and recreation; while items for which taxes are more regressive are food, children s clothing, personal care necessities, fuel for household use and education. The tax incidence on non-alcoholic and alcohol beverages and tobacco generally falls most highly on the middle quintiles. (This is also displayed in Figures 7 to 9 below, which plot tax incidence by quintile on food, alcohol and tobacco.) It is interesting to note that many of the items for which the tax is more regressive and that might also be considered good or necessity items are those consumed more intensively by female breadwinner and no employed households. 26

27 Figure 6. Incidence by commodity group: male and female breadwinner Misc Gambling Recreation Communication Education Medical exp Fuel for transport -Public/Collective transport -Private Transport Transportation Domestic and household services Furniture, etc Fuel for HH use -Other -Baby products -Necessities Personal care -Children's clothing -Adult clothing Clothing and footw ear Tobacco -Beer -Wine -Spirits Alcoholic beverages Non-alcoholic beverages Meals out -Sugar/confectionary -Other -Basic Food -Utilities -Housing Housing and utilities Male breadwinner Female breadwinner Tax/Exp share (%) Source: Own calculations from IES 2000 Notes: - denotes subcategory 27

28 Figure 7. Food tax incidence by employment category and quintile Tax/exp share (%) Quintile Male Female Dual None Figure 8. Alcohol tax incidence by employment category and quintile Tax/exp share (%) Quintile Male Female Dual None Figure 9. Tobacco tax incidence by employment category and quintile Tax/exp share (%) Quintile Male Female Dual None Source: Own calculations from IES 28

29 When the tax incidence results by consumption item are further disaggregated by presence of children in the household, the findings from earlier are reinforced (see table A7 in the Appendix which provides the disaggregated results for the headship category only). A comparison between male-type households with children and female-type households with children, finds that male-type households with children bear a higher incidence of taxes particularly on housing, meals out, alcoholic beverages, tobacco, adult s clothing, private transport, fuel for transport, communication and recreation; while female-type households with children bear a higher burden on food, children s clothing, basic personal care items and other non-essential personal care items, fuel for household use and furniture, equipment and household maintenance items. Both male-type and female-type households with children bear a lower incidence overall compared to the households without children, but a higher incidence on certain consumption items such as: housing, food, children s clothing, personal care (esp. necessities and nappies), fuel for household use, furniture etc and education. In contrast, male-type and female-type households without children bear a higher incidence on meals out, non-alcoholic and alcoholic beverages, tobacco, other non-necessity personal care items, adult s clothing, transport, fuel for transport, (private) medical expenditure, communication and recreation. These results suggest that, if we had to divide spending very crudely into good /necessity items and bad /luxury items, the presence of women (with spending power) and children in the household is associated with a greater proportion of spending on the former basket of goods. 29