VAT rating on books: National Treasury briefing; Statistics SA Strategic Plan & budget 2011: Minister in Presidency and Stats SA briefing

Meeting report

VAT: requests for zero-rating on books: National Treasury briefing
Mr Cecil Morden, Chief Director: Tax Policy, National Treasury, told the Committee that the government had released a White Paper in 1988 indicating its intention to the replace the then one-stage General Sales Tax (GST) with a Value-Added Tax (VAT) system. VAT was introduced on 30 September 1991. A Value-Added Tax Committee (VATCOM) recommended a broad tax base with minimal exclusions. The VAT rate had been at 14% since 7 April 1993.

When the VAT Act, No.89 of 1991 (the Act) was promulgated on 12 June 1991, no provisions were made to exclude, or zero-rate, any goods or services. However, on 17 July 1991 the zero-rating of brown bread and maize meal were announced. On 29 September 1991, following extended public protest, eight additional basic food items were added to the zero-rated list. At the beginning of 1993, further public protest resulted in the zero-rating of an additional nine basic food items. Pressures for a higher VAT rate on luxury goods were allayed, to an extent, by the extension of ad valorem excise duties on luxury goods, including such items as air conditioners, cell phones, video cameras and dishwashers. Vehicles were later added to the list by way of a progressive ad valorem duty formula, which meant that the cheapest vehicles were excluded from this impost.

National Treasury (NT) noted that VAT was the second most important source of revenue for the South African fiscus, after Personal Income Tax (PIT). The South African VAT system appeared to be the second-best performing VAT systems in the world, after New Zealand.

Mr Morden then explained VAT concessions. Because VAT was a consumption destination-based tax, exports were zero-rated and imports were subject to VAT. Most countries excluded non-fee based financial services from the VAT base, due to practical difficulties. Many analysts had shown that in absolute monetary terms, the middle and higher income earners benefited more from the zero-ratings than the poor. However, public opinion was very strongly in favour of maintaining the zero-rating of basic foodstuffs. An effective VAT system, not unduly compromised by well-intended but doubtful concessions, was essential in generating the necessary tax revenues to fund government’s expenditure programmes, including initiatives to help the poor.

Mr Morden repeated that goods and services that were zero-rated included exports, and nineteen basic food items, but also included paraffin, goods subject to the fuel levy, international transport services, farming inputs, sales of going concerns and certain grants by government. Goods and services that were exempt from VAT included non-fee related financial services, educational services, residential rental accommodation, and public road and rail transport. Some of the basic foodstuffs that were zero-rated included brown bread, maize meal, samp, mealie rice, dried mealies and beans, lentils, pilchards/sardines in tins, milk powder, dairy powder blend, rice, vegetables and fruit, vegetable oil, milk and cultured milk, brown wheaten meal, eggs and edible legumes. During the last few years, and especially since the zero-rating of illuminating paraffin in 2001, the Minister of Finance had received several requests from the public to abolish VAT on goods and services such as medicines and medical services, books, electricity and water, red meat, security related expenses, canned vegetables, household candles and toll fees.

It had been argued that exemption was also an elegant way of relieving certain businesses from the administrative burden of complying with the VAT system. In this regard, the threshold below which certain businesses did not have to register for VAT was also an indirect form of exemption. However, whilst VAT exemption might result in administrative relief for businesses that made only exempt supplies, it actually resulted in major administrative complexities for businesses that made both exempt supplies and taxable supplies. There was still a debate on whether zero-rating did in fact help the needy. Mr Morden noted that the zero-rating of illuminating paraffin was a prime example of a zero rating that was well intended, but actually misguided. Most of the benefits of the VAT zero-rating of illuminating paraffin were captured by retailers, in the form of higher profit margins. He noted that provision of relief in the form of reduced rates, and zero-rating for “merit” goods, might even worsen the “regressivity” of a VAT system because these goods were mostly bought by individuals in the middle and higher income groups.

Turning to the requests for zero-rating on books, he noted that the estimated household expenditure for books and stationery in 2009 was R11 310 000, and the estimated VAT amount calculated on the books and stationery amounted to only R1 389 000. Books and stationary included books, newspapers and periodicals, miscellaneous printed matter, and stationery and drawing materials. National Treasury did not support the request to zero-rate books. If there was a case to support students, it had to be done more directly and based on a definite budget. Instead, NT also suggested that resources had to be made available for schools and public libraries to ensure better access to reading material for all citizens.

Mr Morden then outlined his second presentation, which dealt with The VAT Treatment of Merit Goods and Services. Mr Morden stated that the Terms of Reference indicated that the NT required a review of current and proposed VAT exceptions and guidance on how to treat future requests for preferential treatment of so-called “merit goods and services”. The key objectives of the study were to develop a sound tax policy position on the VAT treatment of these merit goods and services, to review whether existing merit goods or services currently treated as exceptions were in line with tax policy principles, to develop a sound tax policy position on the VAT treatment of books, medicines, medical services, and municipal services, and to ensure protection of the tax base while ensuring that the VAT system was equitably acceptable and not too complex from an administrative point of view. The study team was required to consider international trends and best practice, the views of various stakeholders, the economic impact of existing and proposed exemptions, the anticipated fiscal impact of existing and proposed exceptions, and the administrative implications of changes to the current status.

The first question was what should be defined as merit goods and services. Arguments used to justify merit status included whether the good or service was in the public’s interest, whether the product was essential, whether poorer households should not have to pay VAT on these products, whether the product was of a cultural nature and whether the product was difficult to tax. International trends showed that VAT relief of merit goods and services should be viewed against the background of generally higher VAT rates, and VAT concessions typically only reduced the “regressivity” of the VAT system when the spending patterns of lower income groups were viewed in isolation. The trends indicated that VAT concessions that could not be targeted specifically to poorer households actually made VAT more regressive, and that these concessions undermined the attainment of other tax objectives and could distort consumer and producer choices.

The NT had not received any proposals for changes to be made to existing zero-ratings and exemptions. However, there was a clear preference for zero-rating over exemptions. There had been no requests for introduction of additional rates. An analysis of household spending showed that existing zero-ratings tended to make up a higher proportion of the consumption spending of lower income households. However, when this expenditure was considered in the context of total household consumption spending overall, they tended to account for a much smaller portion of total spending. As a result, the VAT “savings” enjoyed by lower income groups were much lower than those enjoyed by higher income groups.

The NT estimated the proportion of benefits that suppliers and consumers would receive from the proposed zero-rating of prescription medicines, non-prescription medicines and books. For prescription medicines, the supplier would receive a 63% share of the benefit from the VAT zero-rating. Suppliers would receive a 59% share of benefit from the VAT zero-rating of non-prescription medicines. For books, the supplier would receive an 87% share of the benefits from the VAT zero-rating, while consumers would only receive a 13% share of the benefits.

The NT concluded that “very low” and “low” expenditure groups derived limited benefit from preferential VAT treatment. Targeted assistance in the form of grants and subsidies were required for these groups. Mr Morden reiterated that existing zero-ratings and exemptions served to make the VAT system more regressive. He also repeated that suppliers of books, and of both prescription and non-prescription medicines, would benefit more from zero-rating than would the consumers.

The NT, therefore, rejected the request for the zero-rating of books because it would worsen the regressivity of the VAT system, and consumers were likely to receive a relatively small portion of any benefit arising from zero-rating. In addition, the cost of books had little influence on literacy levels in South Africa, but additional funding for books in the education system would have a bigger impact on literacy levels.

The NT also rejected the request for the zero-rating of medicines because it would worsen the regressivity of the VAT system. The positive price elasticity of demand suggested that the elimination of VAT from medicines would not have the desired effect of increasing consumption of medicines, and very low and low expenditure groups were in any case reliant on the public health sector for access to medicines so lower medicine costs would not automatically improve access to health care of these groups.

Discussion
The Chairperson noted that the zero-rating of books was a matter that the Committee had wanted to discuss for a long time.

Dr D George (DA) noted that there were many countries in the world that provided concessions for books in particular. This was obviously done for a reason. He asked if the NT had looked at these reasons in other countries. He wondered what impact it would have on the fiscal framework. The Minister of Finance had been asked this question but the Committee had not received a straightforward answer. He thought the impact would be very small, as it did not have much of an impact on spending on education. The NT mentioned that existing zero-ratings might not be beneficial for poor people in particular, and that existing concessions could even be removed. This seemed counter-indicative. He asked if NT was really saying that there was no benefit to poorer people from zero-ratings.

Mr Morden answered the question on the fiscal impact of the zero-rating system by noting that the supplier of books would receive 87% share of the benefits from zero-rating. This varied depending on the type of product and the market structure. While it was desired that most of the benefit could be passed to the consumer, the reality was that this would not happen.

Mr Morden then dealt with the reasons why other countries had made special concessions for books. He said that there were benefits to the poor when it came to the zero-rating of books, but these were marginal compared to the benefits the rich would receive. It was a question of who benefited, and in what proportion. Suppliers of books would receive just over six times the benefit that the end-consumer would receive from the zero-rating. NT therefore considered that it would be preferable to apply the money received from VAT on books to boost literacy programmes and public libraries, as well as social grants. He acknowledged that zero-rating resulted in some benefit to the poor, but stressed that this must be seen against the cost of granting that benefit.

Mr N Koornhof (COPE) asked what the NT defined as “books”, and whether this would include both electronic and printed versions. He also wanted to know if the 14% VAT gained from book sales could be kept aside as a conditional grant for libraries.

Mr Morden replied that one of the difficulties lay in definitions. In New Zealand, during the debates on zero-rating of books, the question arose whether pornographic magazines could also be included, so there were real problems. The “categories” into which books fell were very broad, and trying to put this into a legal definition would be very complex indeed. He was not saying that it could not be done, but that it would create substantial discussion and dilemnas. For this reason, he could not give an absolute answer as to what would be regarded as a “book”.  

Mr S Swart (ACDP) stated that the zero-rating of books was a positive idea. However, the fiscal implications had to be understood. He wanted to know how the NT had calculated that suppliers would receive an 87% share of the benefits from the VAT zero-rating of books. 

Ms Z Dlamini-Dubazana (ANC) noted that it could be true that the cost of books did not affect literacy. However, there were instances where learners had to wait months to get books. NT had said that the government had to budget for learners books. She asked if the NT had a plan or a model in place for how government would do this. She also asked what feedback the NT had given to the public regarding the requests of products that were proposed should be zero-rated.

Mr Morden answered that the issue of literacy had to be tackled in a more aggressive way. However, he did not think the zero-rating of books would help anything at all.

Mr Morden added that there were various people asking for the zero-rating of various items. The NT asked itself why people made certain requests for zero-rating and what they would achieve out of it. Usually, the people that were thought to derive some benefit from the zero-rating did not actually benefit in the end. He reiterated that the best way to have consumers at large obtain a benefit was to have targeted relief, or to allocate more money to libraries.

Mr Morden said that the point the NT was trying to make was that the government had to choose the mechanism by which it wanted to achieve its objectives. A broad research study had shown that zero-rating concessions was probably the weakest route to take. There were many better-targeted interventions that the government could take.

The Chairperson said that he did not want to complicate the matter, but when the issue was first raised, it related to school books. However, in order to address this matter, it was necessary to look at the whole range of interventions that the government had made in moving towards free education. The President's State of the Nation Address (SONA) has announced various interventions that were still going to be made into the education departments. The issue of school books would then be discussed in more detail. The Committee would put the matter of zero-rating books to rest for a while, although there might be a need to raise the issue again.

Statistics SA:  Work Programme and 2011-2014 Strategic Plan and Budget
Hon Trevor Manuel, Minister in the Presidency: National Planning, said that Statistics South Africa (Stats SA) was not a standard government department. Its independence was vested in its own statutes, which limited the power of the Minister over that body. One of the issues that would receive focus in the presentation on Stats SA was the Census 2011. He noted that the two documents that were important for today’s briefings were the Stats SA's Strategic Plan 2011-2014 and its Work Programme for the fiscal year 2011/12. Members would be able to see a strong correlation between the two documents. He did not think that entities needed to have a new strategy drawn up every year. A strategy should rather be drawn over the longer duration, with more focus upon the work programme within the strategy. However, the Public Finance Management Act (PFMA) called for a strategic plan to be tabled every year.

There were a few strategic areas that had been set out. These included the expansion of statistical production, addressing the information gap in statistics, addressing the quality gap for statistics, and attending to the statistical skills capacity gap. These were the four issues that formed the basis for the strategy. Another major objective for this year was the Census 2011. The Statistics South Africa Act required, in section 7(2), that a census be conducted in 2001 and every five years thereafter. This was an issue that Parliament should discuss, as there was a lot of interest in the conducting of the census for the future.

Mr Pali Lehohla, Statistician-General: Statistics South Africa, stated that official statistics were seen as a public good because they were used for evidence-based decision-making, for informing planning, for monitoring programme implementation, for evaluating programme performance, and for research and historical value. He stated that Stats SA’s goal was to increase the supply of official statistics in order to inform development outcomes.

Stats SA’s strategic objectives included expanding the statistical information base by increasing its depth, enhancing public confidence and trust in statistics, improving productivity and service delivery, leading the development and coordination of statistical production, investing in the learning and growth of the organisation, and promoting international cooperation and participation in statistics. Stats SA wanted to expand the country’s statistical information base by focusing on population dynamics, economic growth and transformation, prices, employment and job creation, life circumstances, service delivery and poverty, safety and security, health and education, sustainable resource management, food security, land reform and rural development.

Stats SA’s priorities for 2011/12 consisted of releasing the Census 2011, achieving statistical support and coordination for education and health as well as crime, and focusing on capacity development and African statistical development. Stats SA hoped to achieve the demarcation of the country into 104 000 enumeration areas by the end of September 2011, in preparation for the Census 2011. Communication and marketing of the Census was an ongoing project. 110 district census offices were established and Stats SA hoped to recruit and train 156 000 staff for the Census by October 2011. The aim was to visit 14 million households and count everyone living in South Africa. A post enumeration survey would be conducted at the end of November 2011 and data processing would commence in January 2012.

Stats SA said there were conflicting sources of information in health statistics and education statistics, and that crime was only reported from administration records. Discussions were under way for conducting a health survey, and there were interdepartmental collaborations to streamline health statistics and education statistics.

The ISIbalo Institute was developed because of the low numbers of skills in statistics and the poor culture of use of evidence in decision-making. The Institute offered foreign study training for statistical skills, internship intake from university and high school, and methodology training with Michigan University.

The Medium Term Expenditure Framework (MTEF) showed that Stats SA’s budget was divided between an administration programme, an economic statistics programme, the population and social statistics programme, the methodology and standards programme, statistical support and informatics, corporate relations and survey operations. The medium term estimate for 2011/12 was R3 240 909 000, rising to R1 698 304 000 for 2012/13, and R1 717 720 000 for 2013/14. (see attached presentation for full details).

Discussion
Mr Swart noted that there was a problem of under-counting the numbers of people in South Africa. He knew there were certain areas on which Stats SA would try to focus. He asked for more information on this.

Mr Swart understood that the issue of immigrants posed a huge challenge to Stats SA. Immigrants, especially illegal immigrants, would try to evade the census. He asked how the Committee could assist Stats SA with this matter.

Mr Swart thought that the Minister had implied that the Committee needed to change the census legislation because South Africa had missed the census for 2006. He assumed that the costs and other implications required some sort of amendment. He asked Stats SA for its thoughts on the matter.

Mr Lehohla replied that Members, as Parliamentarians, had a role in educating immigrants about the census. A Communications Campaign was launched by Stats SA in October 2010, which was premised on an enormous and extensive public awareness and education programme. Stats SA wanted people to know why they should participate in the Census. This objective could not be done by Stats SA alone. It needed partnerships with other departments and entities, as well as with influential and credible people in the various communities.

He said that the challenge would be how they would ensure that immigrants, especially, illegal ones, were made to feel comfortable and assured of the confidentiality of the information that Stats SA wanted to collect.

Mr E Mthethwa (ANC) asked how Stats SA dealt with people migrating from one place in the country to another, to avoid double counting. He also asked what assistance Stats SA would need from the Members of Parliament and their constituencies to ensure that things went smoothly when the census was being conducted.

Minister Manuel answered that migration was a problem that had been around for quite a while.

Mr Howard Gabriels, Chairperson, Stats SA, stated that there were some areas of improvement when it came to doing statistical analysis of the population, with substantial improvements shown over the past few years.  There was a campaign, by the Department of Home Affairs (DHA), to register people who were in hospitals, including births. Many other initiatives were being undertaken to improve the population register. Because Stats SA did not release a census in 2006, it was very well prepared for the current one. The Census 2011 was important because every survey that Stats SA would be doing afterwards would be “grossed” up to the statistics produced in the Census. In order to have a good census, the country needed a good “dwelling frame”, which had to be updated regularly. Stats SA was extremely pleased with the progress it had made over the last five years. Stats SA needed everyone’s help to persuade each household to complete the questionnaires honestly. The better the information that Stats SA could collect from the households, the more useful it would be for building South African society.

Mr Lehohla added that it would be helpful if Members could go to their constituencies and tell the communities about the Census 2011, encouraging them to participate in it.

Ms Dlamini-Dubazana thought it was a serious problem that the Minister's powers were limited, especially when it came to assessing the outputs of the census. This effectively meant that the Minister had the responsibility of finalising the census, but he did not have to worry about the outcomes. This would result in a problem with accountability. She proposed that the Committee should look into this matter.

Ms Dlamini-Dubazana noted that she had recently spoken to the Minister of Finance and asked him if the NT would need more money in view of the imminent release of the Census 2011 campaign. He said the allocated budget must have been enough, since the NT had not complained about it. She expected that the NT would over-spend on its budget. 

Ms N Sibhidla (ANC) said that she had read something in the media at the beginning of the year that challenged the information received from Stats SA. She asked how Stats SA was dealing with this. The presentation showed that Stats SA intended to ensure that its management systems and quality management was in place. She wanted to know how Stats SA would contribute to job creation.

Minister Manuel clarified that a private agency, called Adcorp, published a report saying that Stats SA’s statistics on unemployment were wrong, and claiming that the unemployment figure should have been closer to 8%. Most countries agreed that 5% or 6% unemployment virtually meant that a country was operating at full employment. An 8% unemployment rate meant that South Africa did not have an unemployment problem. If this was indeed true, then he suggested that the President could “de-emphasise” job creation in the work programme of the country. He hoped Adcorp’s report would not be seen to mask one of the biggest problems that the country had ever faced. Most people in the country knew the extent of the unemployment problem. It was fundamentally important that South Africans trusted Stats SA’s output, even when the figures showed things they did not want to see.

Mr Lehohla dealt with the queries about Stats SA’s own contribution to job creation by explaining that Stats SA aimed to recruit and train 156 000 staff members by the end of the year, but noted that these would be for temporary jobs that would be offered. Further to this, Stats SA was looking at taking in matriculants and university students to train for the statistics sector.

The Chairperson thought that the independence of Stats SA had to be discussed. The question was what benefits the independence of Stats SA offered to the country as a whole. He asked Stats SA to elaborate on this. A challenge arose when entities used their independence as a pre-emptive right, and this was a difficult problem that faced many countries.

Minister Manuel replied that an examination of other world events could isolate two experiences. His first example was the 2008 Greek election, when the current government came into power. Although there were, at the time, published statistics, as soon as the new government came into power it realised that the statistics had no veracity at all. The numbers had been tweaked to make the previous government look good. This was a huge issue in governments, and each successive government had to ask how trustworthy were the statistics put out by statistical agencies for the purpose of evidence-based decision-making. Tunisia was another country that experienced these kinds of problems.

Mr Gabriels agreed that it was very important to debate the independence of Stats SA. It was important to understand the role of the Council, which was to advise the Statistician-General and the Minister about the methods, norms and standards that were applied and used to produce statistics. For the reasons set out by Minister Manuel, it was important to protect the national statistics agency from political interference. There was a whole range of protective measures articulated in the Fundamental Principles of Official Statistics, to which almost all countries in the world had been a signatory. 

The Chairperson noted that whenever the Committee met with Stats SA, Members felt that they needed more time to discuss issues. He addressed the matter of aligning the strategic plans and the PFMA. He suggested that the Committee should look into whether there should be annual strategic plan presentations, or if Committees should be looking rather to entities’ progress on an annual basis, using the five-year plans. This was an important issue that merited further discussion. Another issue was the public awareness of Stats SA and the Census 2011. The Committee Members offered their assistance in helping their constituencies to understand what the Census was, and to encourage people to participate in it.

The meeting was adjourned.