Statistics SA: Strategic Plan and Budget 2008-09

Meeting report

Statistics SA (Office of the Statistician General): Work Programme 2008-11
The Chairperson welcomed the Deputy Minster of Finance, Mr Jabulani Moleketi to the meeting.

Hon Deputy Minister Moleketi reported that the Minister of Finance had an oversight and responsibility for Statistics SA (Stats SA) but did not control statistical outcomes. The Statistician General (SG) was responsible for reporting on all statistics. Over the years there had been significant improvements in building capacity, ensuring that South Africa was the provider of choice on statistics, across a range of areas. The mandate stretched beyond South Africa as it was active also in the international field and would deal with challenges of providing relevant statistics to all stakeholders in the economy.

Mr Pali Lehohla, Statistician General (SG), Stats SA, said that the theme "South Africa that I know" meant that there needed to be appreciation of South Africa in all facets. There were three forms of knowledge: knowledge as information, knowledge as understanding, and knowledge as power. The strategic importance of statistics was to allow decisions to be made on the basis of evidence, and said that statistics would be managed for development results, to deepen transparency, to broaden and deepen democracy and for managing the Paris Declaration on aid effectiveness. Statistics were also important in the anticipation of important changes, including changes in society, climate change and others. In addition Stats SA attended to providing State statistics on prices, Gross Domestic Product (GDP) and census information.

A statistical organisation was characterised by a highly intellectual, technological environment, with an understanding of the political environment within which it operated. The logistics to collect information became huge, and the administration had to be effective and efficient. All had to operate with equal intensity.
Mr Lehohla noted that the role of statistics was to inform decision-making in economic growth, price stability, employment and job creation, life circumstances and the demographic profile. The policy environment was fluid. Policy debates were ongoing about changes in the interest rates, increasing food prices, use of bio fuels, labour market dynamics and the fighting of poverty. All of this must be responded to by “knowing South Africa”.

In 2008/09 the key priorities included the launching of a quarterly labour force survey, with a focus on agriculture in the Eastern Cape, a survey on living conditions to measure poverty, a re-weighted Cost Price Index, review of the producer price index (PPI), a geo-referenced dwelling frame, a National Strategy for development of Statistics (which should include signing up of an international agreement) and international activities through the International Statistics Institute (ISI) to be set up in Durban in 2009.

Mr Lehohla went into more detail on each of these priorities. In respect of capturing the pulse of the labour market, Mr Lehohla noted that data had been collected since January 2008, and the first results would be shown in August, and every quarter thereafter. The seasonal variation would form the basis of a policy and response.

Measurement of consumer price changes was using the diary method over a period of twelve months. A re-weighted Consumer Price Index (CPI) would be published by February 2009.

The Producer Price Index (PPI) showed monthly movement in prices of producers. This had last been re-weighted in 1999. Although there had been stabilisation there were still some challenges to be addressed. Poverty would be measured in multiple dimensions, through engaging with stakeholders and finalising the survey instruments and methodology, publicity and training.

Mr Lehohla pointed out that there were unique geographical identities, and spatially defined planning would result in better investment in service delivery. A complete database of all dwellings had to be completed by 2009/10. The demarcation for the census 2011 would have to be completed during 2010/11. There were protracted social and political processes involved. 43% of the dwelling frame should be completed by March 2009. The National Statistics System (NSS) had to have credibility; and, after considerable work done by Stats SA between 2003 and today, it was now regarded as credible. The stakeholders would be involved, and would be trained on the quality assessment framework, and support would be provided in terms of education, provincial profiling and raising awareness.

South Africa was to host the International Statistics Institute (ISI) in Durban from 2009. This would prove the capability of Stats SA and add legacy. At the moment ISI was not representative, with a mostly male and white contingent. In future it should change and bring in knowledge from Africa and other parts of the world apart from Europe and America. South Africa could add value by notifying of the unique challenges and solutions. An African Young Statisticians Conference was to be held on 1 July. There was a need for greater harmonisation in the Southern African Development Community (SADC). African statistical development was somewhat uneven. The continent was unaware of its population size, but appreciation of the importance of statistics was growing. There would be a round of population and housing censuses in 2010.

Mr Lehohla said that it was necessary for Stats SA to sustain the capacity that had built, to avoid losing it. It must continue to train people in all parts of the world, and to attract and retain technical skills. At the moment there was a disproportionate number of foreign staff. Leadership and management were vital and Stats SA had embarked on a NEPAD Leadership Programme. The ISIbalo Programme aimed to develop women in statistics, and young statisticians, and included programmes on maths, statistics and the girl child.

Stats SA’s strategic risks were outlined as data management and information delivery, the accuracy of the business register, the need to build leadership and management capabilities, and strengthening analytical and methodological capabilities. There was also a need for corporate data processing, integrated fieldwork and relocation to the intellectual hub of the city.

Mr Lehohla touched on the issue of climate change, noting that there had to be measurement of carbon credits. Stats SA was on track for the first time in giving comprehensive employment data on a quarterly data, changing consumption patters and a poverty profile.

Finally, Mr Lehohla tabled a comparative prioritisation of challenges from 2003 to 2007,which gave context to the PPI. In 2003 everything was regarded as urgent, and there was little room for innovation. Only in 2007 was the Office able to attend to the PPI, and attention was now being focused upon it.

Discussion
Mr K Moloto (ANC) asked about the National Business Register, noting that Stats SA and South African Revenue Services (SARS) must interact on this serious matter that warranted urgent attention He asked what steps had been taken to ensure that the relevant legislation was being amended, and whether it was being treated as urgent.

The Deputy Minister agreed that the National Business Register was important. It was an inter-disciplinary and inter-departmental matter, that did not just depend on Stats SA. It had to speak to the provisions of the new Companies Act, which would put an obligation on different actors to provide the necessary information. It was work in progress. It would improve the capacity of SARS to generate revenue and make projections more accurately.

Dr Jairo Arrow, Deputy Director General (DDG): Methodology and Standards, Stats SA said that there was a steering committee of Department of Trade and Industry (dti), SARS, National Treasury (NT) and Stats SA. The interviews had been conducted for the appointment of a project manager, and this would be followed by tasks being put in place. The question of the legislation would be raised with the relevant principal Department. There had been progress.

Mr Moloto asked about quarterly GDP releases, and why the transport survey had been prioritised first. He asked if sectors not mentioned were well covered and if the information was adequate.

Dr Rashaad Cassim, Deputy Director General: Economic Statistics, Stats SA, noted that one of the challenges of the quarterly GDP was to dissect the economy better, and in the services centre. The relevance of the GDP was its ability to provide quarterly, and then to provide annual data. Quarterly data would allow for short term decisions. Currently the formal sector was covered, and the informal was based on estimates. The estimate of the services sector was easier,  but it was more difficult when it came to the breakdown of services. For example, a manufacturer may carry out transport, advertising and the like in the course of the whole business, but all of this would be categorised under manufacturing as the major activity. There was therefore difficulty in analysing exactly what fell within the sector. The sectors had been chosen on the basis of what was most important in the economy, and on the basis of ease of measurement. Transport was important, and it was easier to measure than advertising, for instance.

Dr Cassim added that South Africa was the only country where there was a division of labour on estimating the GDP; Stats SA would look at it from the production side, and SARS from the expenditure side. The reason for the dual investigation was the importance of the result. Most statistics agencies would have two units to consider the national accounts, and the two units would balance out their assessments. In South Africa Reserve Bank would estimate one and National Treasury the other. Stats SA needed its own capacity to estimate expenditure (currently being done by Reserve Bank) and by improving its capacity in this area it would strengthen its ability to engage with the Reserve Bank. Stats SA was subject to greater transparency than a bank.
 
Mr Moloto noted that a feasibility study had been embarked upon to improve capacity, and asked what the significance of this exercise had been, and whether shortcomings had been picked up in the previous methods of measurements.

Ms Refiloe Masiteng, Deputy Director General: Population and Social, Stats SA said that all measurements were purpose-driven and would be based on internationally known and acclaimed standards. For instance, measurements on labour would follow International Labour organisations standards.

Mr N Singh (IFP) referred to the printed budget statement, Vote 11, on page 187. He noted that this said that by 2009 the organisation aimed to have reliable methods, skilled staff and skills programmes. He asked whether these did not exist at the moment.

The Deputy Minister stated that this was a continuous process, to allow Stats SA to reach its desired level by 2009. Different partnerships with tertiary institutions, who must be able to deliver skilled people, and in house training, would provide relevantly trained people in the sector. There was also a need for a special mindset and motivation by staff, who would be required to package the numbers in a user-friendly way. He had enormous admiration for their work. The public should be informed that Stats SA was not trying to “police” people when it asked questions, but was trying to change their lives, and the researchers needed special skills that were learned through practice. The data produced involved a huge range of work and skills.

Mr Lehohla noted that there had been significant progress already.

Mr Singh referred to page 200, in relation to personnel. He asked for more clarity, because there was a considerable increase in projected numbers, both of staff and interns. He asked if there was a programme in place to attract these numbers of people.

Mr Lehohla said that there was an aggressive training policy. There was a need for recruitment. The Statistical Institute aimed to train according to its focus on how technology, statistics and geography would go together, and people would be trained specifically to integrate these areas.

Ms Nombuyiselo Mokoena, Deputy Director General: Corporate Services, Stats SA, added that Stats SA would take in interns, and currently was keeping more at the end of their internships. Stats SA was going to all institutions to recruit. It had also begun to speak to professional organisations to help with specific skills. It would try to establish more permanent posts, as these were easier to fill. Bursaries were offered to matriculants and staff members to do degrees, and there were about 30 students at one of the institutes. Last year bursaries had also been offered for students to attend local institutions and these graduates would be used for the pool of recruitment. Other programmes were also in place. There was also a retention strategy, including a programme of Department of Public Service and Administration, who were using Stats SA as a pilot for a skills audit. Once Stats SA moved to their new building, she felt that would also encourage staff to stay because it would create a more conducive working environment.

Mr Singh asked, in relation to climate change, what linkages had been established with the line function Departments of Agriculture (DOA) and Environmental Affairs and Tourism (DEAT).

Mr Lehohla said that Stats SA did interact with DEAT and DOA to the extent that it dealt with natural resources. The issue of climate change only emerged in the last UN Statistics Commission in Norway, to which Stats SA was invited but could not attend. Statistical measurements were likely to take some years but the Nordic countries had already started to examine how to assemble the tools.

Mr S Marais (DA) said that it had been indicated, during the Annual Report presentation, that skills shortages had led to the qualified audit report. He asked what the staff complement was, and whether there were sufficient skilled staff to do the job.

Mr Lehohla said that Stats SA had, since 2005, been moving to improve operational efficiency and effectiveness. This was dependent on the change of the structure to respond to the key objectives, and expansion of management and other layers. Even before the final structure was put in place, there had been recruitment and training, including elevation of the provincial structures. There was a programme and project offices to deal with integration of systems. There was better operational efficiency in planning and running projects. The Institute of Southern African Studies training centre was training on field work and people would be sent there, until the Institute was set up locally.

Ms Mokoena added that Stats SA had moved from disclaimers to a clean audit in the last year. Most of those at high executive level had been in their posts for a while.

Mr Marais said that Stats SA had admitted that there were problems with the PPI statistics, and had sounded caution on how to use these statistics. He asked how the SG would effectively deal with the negative perceptions of consumers and investors, and correct the statistics. He also asked Mr Marais to what extent the new and old analyses of CPI were comparable, and how were they to be interpreted. The Reserve Bank and industry were using these figures and they impacted on the lives of consumers. He asked how the end result was reached and interpreted correctly.

Dr Rashaad Cassim said that one critical challenge of Stats SA was how well it educated and communicated with users about problems and issues. The media was generally constructive, although sometimes issues were blown up out of proportion. It was important to put the CPI and PPI in context. There would be many critical articles on this over the next few months.

When a decision was made to move from one type of series to another, there would always be transitional “hiccups”. He was cautious about using the term “error”. A statistician regarded these as harmless, but the political interpretation of this word could be serious. He would prefer to use the term "creating noise". The moving and the transitional phase would thus create noise, and there would be unanticipated complexities. A series that had not been changed for a while became  less relevant to the economy. Stats SA had changed not only to try to get more current schemes, but also to move to international classifications.

The PPI was more complex than the CPI, because with the CPI, Stats SA had enumerators to go out and collect prices, and retained control over that process. Stats SA however was reliant on firms when it came to the PPI. There collection of data took about 2 weeks, and then about three to four days were needed to process that data. Some firms might give incorrect data and this would have to be checked.

There were two matters that had not been anticipated. The transition was more complex than Stats SA had thought it would be. In addition there were new respondents whose response rate had been challenging. There was also the mindset about how to interpret user statistics. Analysts would create an outcry when the PPI was to be reviewed, not least because revision was not common (as it was in other countries) and so any change would be regarded with suspicion. Stats SA needed to educate users on that point.

On the question of the “caution” about the statistics, Stats SA said that it had advised caution, because it had recognised that there were some shortcomings. It had had to take a decision whether to delay putting the PPI out, or whether to produce it, warning that there might be some need for later revisions. A delay would have created a domino effect in to the following months. The SG had been quite open on what the situation was, and was not caught off guard, but was trying to make the difficulties clear to others. Nonetheless, bond analysts whose bonuses relied on predictions were annoyed at the changes.

Dr Cassim added that analysts would treat the PPI as a core indicator, although in fact it was only an indicator for agriculture, manufacturing and mining, thus covering only about 30% of the economy. In the CPI services were an important driver, and because it was so wide, Stats SA devoted about five times more resources to it than to the PPI. Stats SA was presently running a parallel sample of CPI and PPI for the whole year. The market must be educated to see that the month to month must be tracked. Better PPI measuring would not necessarily change the qualitative mass, but these could have effects on the CPI.

Dr D George (DA) said that page 28 set out the statistical process. He noted that there was a lag between parts of this process, which might be problematic since the economy used information received from Stats SA to identify measures and implement them. He asked if there was any indication of the lag for each key indicators.

Dr Cassim said that CPI and PPI covered the whole reference month, and came out 28 days after the reference month. However, retail would only come out 47 days after the reference month. Analysts believed that this was too late. However, this must be compared to Canada, where retail sales information would come out six weeks after the reference month. The USA analysts would prefer to get preliminary data, with a revision, but Canadian analysts wanted to work only on final figures. The relevance of economic statistics would also depend on economic efficiency. The biggest threats were the Business Register and getting firms to respond. Some were actually refusing, despite the publicity campaigns and there might have to come a time when legal force might be necessary, although Stats SA did not want to place undue burdens.

Ms Masiteng clarified the position in relation to the specialised surveys. The statistics for each of the surveys would be released at different time frames; some 4 weeks after each quarter, some only after 9 months if there was a great deal of data. The Living Conditions Survey would be continuous, but this should be released after April 2010 (although the survey itself would end in October 2009) as the data processing would go alongside the collection. Migration data was released on a monthly basis, but it took some time as it was secondary analysis of the Home Affairs data. Mortality data was released yearly. The time lag would be determined by the amount of work

Dr George said that the Minister of Finance had said that the poverty index was political and technical. He asked what the purpose of the poverty index would be and how soon it would be out.

The Deputy Minister noted that the Millennium Development Goals included halving of poverty. It was therefore relevant to know what was being halved. Currently government institutions and NGOs all had different understandings of the measurement of poverty, and having one definition would involve a political process. It was important to empower organised labour to understand the implications of the living wage.

Mr Lehohla added that most of the work had been concluded on this index. Poverty was multi dimensional, ranging from absolute to relative poverty, and perceptions were also important. Stats SA had attempted to confine itself to absolute poverty, because it was possible to measure this, and because the relative measures would be more applicable to those societies that had already dealt with absolute poverty, and South Africa had not yet reached that stage. The future surveys would look at relative poverty as well.

Ms Masiteng said that there were many measurements for poverty, based on the minimum food need for daily requirements, while the threshold above and below would be measured on the basis of additional non-food components. It was also used for comparisons on how SA was performing on the Millennium Development Goals, and on eradication of poverty. The poverty index should be able to determine the head count. The primary responsibility was to distinguish between the poor and non poor. More work would have to be done to understand relative poverty, and a third leg – the subjective analysis of poverty – would be available in 2010.

 Dr George noted that there was much demographic shift, with municipalities struggling to absorb the new residents. The population census was ongoing, but he asked if there was a plan to identify more closely the number of residents in informal settlements in South Africa, as the anecdotal evidence showed that this was growing, with an impact on service delivery.

Mr Lehohla replied that Gauteng and Western Cape had previously been labour-receiving provinces, but had now changed also to become children-receiving provinces. He tabled a graph on population group, by age groups, comparing the age distribution generally in South Africa, and then for the Eastern Cape, Free State and Gauteng over the years. Children were now moving to the provinces with their parents, and there was an increase of 500 000 households in Gauteng from 2005 to 2007. This was quite dramatic. Gauteng had around 40% of most business, housing and so on. Urban Space policy would have to deal with whether this was desirable or if it posed problems. To an extent this was a spontaneous process. However, the question could be asked how much more effective it could be if controlled by lagging. It was necessary to interact all the statistics.  The statistics should be used to investigate the policy question “What if we did the following…”
The interaction of people and structures across time and space had different results. In the apartheid years, and even up to 2001, it was clear that most black people were far from the business centres of the cities. Current statistics showed a different shift. Space economics must ask what the desirable spread should - for instance looking at a platinum belt as the cross centre of the future.

Ms Masiteng added that in the past the informal settlements were easy to access and understand, but they were mobile so there had to be a plan to monitor the movement and changes. A strategy was still needed for that.

Mr B Mnguni (ANC) said he had had difficulty in linking the work programme document to what had been said, and programme 4 had been renamed.

Ms Celia de Klerk, Head of Strategy, Stats SA, said that measurable objectives (called objectives and measures) were in fact included in the glossy Work Programme. The work programme was not structured to programmes because it was now being compiled in line with the strategy of the organisation. For every measure and estimate there would be a one-on-line alignment. She illustrated this, noting that there would be a matching to outputs. Many of the measurable objectives crossed several lines. The annual reporting was aligned to the work programme.

Mr Mnguni said that his main concern was around the census - as there was no detail given of what was being done to build up to it in each year, other than that the budget had increased. He also asked, in view of the influx, especially from Zimbabwe, how Stats SA would differentiate between citizens and visitors.

Ms Masiteng pointed out that census plans were dealt with in the glossy Work Programme; the plans were in draft form but all activities would be ready by June. The census was the total count, and would include both citizens and non citizens. However, questions would be included to find out where the person was born, where he was five years ago, and when he had moved. The question of legal or illegal residence would be difficult.

Mr Mnguni asked also how to differentiate between the original SA economy and the skills that were being brought in.

Mr Mnguni also referred to the "underground" economy - sex workers and drug dealers, and asked if it was possible to measure the impact of such.

Mr Lehohla said that this was difficult to measure; in some countries it would be an inverse of what the policy were able to capture.

Dr Cassim added that the economic statistics would usually rely on the business register, taking as its starting point the VAT threshhold. The labour force survey also played an important role in tracking employment. Most of the information on the informal sector came from the household surveys. Last year the Bureau for Economic Research suggested that Stats SA had underestimated the informal sectors by about R9 billion. However, although there was a large informal sector in terms of numbers of people and businesses, its contribution in output was quite small. The costs of monitoring it could outweigh the benefit.

Mr Mnguni asked if there was any indication that the infrastructure would be able to accommodate the expected influx in the lead up to 2010

Mr Lehohla said that yesterday Stats SA released a survey showing growth in the revenue being generated in the hotel and accommodation sector, of about 38%. Most was due to the volumes of people using the accommodation. He was not sure whether the number of rooms had increased. He would give a fuller answer after some further investigation.

Mr Marais noted that statistics obviously had to be interpreted. He referred to an extract from an article read out by Mr Lehohla, and asked if the article was implying that overall the lives of people in the Eastern Cape had improved, or if there were improved statistics in the Eastern Cape because so many of the poor people had moved to other provinces.

Mr Lehohla said that the article had dealt with agriculture, and had indicated that the number of employees in agriculture had increased, and that Eastern Cape accounted for upwards of 60% of this. Stats SA did not yet know what was happening in that sector. If there was the possibility of increasing employment, then policy may begin to look at the policy successes that generated employment in agriculture, and link those issues with the social pensions and similar matters. This was an interesting observation of Mr Marais. There had been improvements across the country, but Eastern Cape was also showing improvements, and the article suggested that increasing numbers in Eastern Cape had improved their skills. Analysts would now have to look at whether the migration had had an effect on this.

Mr Marais asked how the decisions were taken on what Stats SA wanted to do, and whether there was a subjective, or objectively bench mark approach

Mr Lehohla said that methods were “a consequence of the possible, which would be a combination of light and instruments to be used”. Statistical methods were often limited to the areas where there was knowledge, and the unknown remained unknown. Statistics would measure what was capable of being observed. Stats SA adopted a positivist philosophy - accepting that there were always things that would remain unknown. The limitation of practice as statisticians was that there would be legalised errors (such as finding a child of 14 at work, although he should not be working until the age of 15) and intentional omissions, which would be omitted for pragmatic reasons, such as lack of resources, or non mature methods.

Dr Cassim said that Stats SA had to be assured that the methodologies it was using were assured. There was an internationally driven programme on how the national accounts would be used to estimate GDP. This programme set out how to treat figures, and what assumptions to make. There were international and Stats SA manuals on how and why decisions and calculations were made. On the CPI, decisions were taken elsewhere about quality adjustments; but Stats SA did not do so. For each set, there were manuals that explained how things were done. Stats SA, when it took the decisions about the re-weighting of the CPI and PPI, had undergone peer reviews by international experts. In many countries, the importance of the PPI was that overall weighting was not calculated from sales, but from value-add. PPI must be consistent with national accounts so that it reflected properly with GDP. User groups, who were not in positions of conflict, would also assist the process.

Mr Marais said that Stats SA had mentioned that they were able to anticipate some matters, and had stated that they were not caught off guard with the PPI. He wondered if the xenophobia had been anticipated, and if so, why the anticipated events had not been communicated to politicians, so that they could have responded differently.

Mr Lehohla clarified that he had not said that Stats SA had anticipated xenophobia. He said that the practice of statistics, to be useful, must anticipate change – whether this would occur through disease, climate change, xenophobia and the like. This was not to say that statisticians could tell when such changes  would occur.

Dr Cassim said that perhaps Stats SA did not anticipate the full extent of the problem with PPI, but it did make the problems known through publication when it was aware of them. This had been a learning experience, as the process perhaps should have been done more gradually. The lack of anticipation was the reason why Stats SA was not more explicit to Members.

Mr Moloto asked for clarity on what the measurable objectives were and whether they had been achieved. He referred to the Work programme, page 53, which set out the assessment of health facilities. He asked what the significance of this was.

Ms Masiteng responded that the Health Matrix Network was a global network, and would include matters such as infant mortality rate, epidemics and so forth. This depended also on birth registration statistics, and mortality, which was not captured by Department of Health but by Department of Home Affairs. It concerned how administrative surveys could be used in a complementary manner between livelihood, population and health.

Mr Moloto noted the reference to the non-observed economy and to non-profit institutions and he asked why this exercise was being undertaken and how it would benefit the national account.

Mr Lehohla said that Non Profit Organisations (NPOs) were becoming more important players in the economies of developing countries. They could contribute to the GDP through volunteer work, and in fragile and settled countries the impact might be significant, although it was not necessarily so in South Africa. The non profit sector was huge in Sudan and Rwanda, for example. Churches were not in the arena of GDP measurement because in theory they did not generate income, although their value add was huge. This was an item on the agenda for the International Statistics.

Dr George followed up on the points he had made about transition. He wondered if it should not be called a transition and that people should be told so, so that they could anticipate any problems.

Ms S Makoto (ANC) said that in the Strategic Plan there was mention of trust and confidence building in Stats SA. She asked how far Stats SA had gone in improving the relationship with the media, and ensuring that the analysts were effectively understanding the information being given out.
 
Mr Risenga Maluleka, Office of SG, said that the recognition of the scientific nature of statistics, the trust placed in official statistics and the internal practices and history of the organisation were all important. The understanding of scientific methods was also vital. The Statistician General would regard the term “sampling error” in a very different light from the media, who might sensationalise the “error” aspects. Stats SA had started to work with specific journalists now that there was a communication unit. The media’s bias when there was a dispute would also influence the way in which matters were reported. Comments from public commentators were sometimes taken as a basis for the evaluation of official statistics. Corporate governance issues were also important. In the past this had been a serious challenge. There was a need to be transparent at all times, but communications and skills were also important.

Mr Trevor Oosterwyk, Head of Communications, Stats SA, added that service providers had been secured to train staff on how to deal with the media. There had been better service delivery to and relationships with the media. This helped to improve the construction of the message getting across. On January 8 there had been a sensationalist and patently wrong article, and this was taken to the press ombudsman. A hearing was scheduled on 3 June. Stories such as this reflected negatively on the reputation of Stats SA, and should not be left unchallenged.
 
Ms Mokoto asked what the results of the survey were on numbers of employees and self employed in South Africa, and what government should be doing to assist the second economy, as the impact, even of the informal sector, was important.

Ms Masiteng said that there were estimates around the informal sector, which would include people simply making a subsistence-type of living for themselves. 700 000 of the population in 2005 had said that they were running non-registered businesses. The impact was for the policy analysts to assess.

Mr Lehohla said that Stats SA had done two reports on self employed. The size of government intervention had certainly, according to the small, medium and micro enterprises (SMMEs), been less than MAFISA lendings. 75% of businesses started with money borrowed from friends. Government was only really providing between half a percent and five percent of funding to start small businesses. The policy question was what must be done, but that was not for him to say.

Mr Mnguni said that the issue of the poor and increasing food prices was important, and he asked if there were any recommendations on zero rating of food stuffs and assisting the poor.

Mr Lehohla said that Stats SA had observed that the increase on the fuels of petrol, paraffin and diesel, per litre, was relatively higher in the case of paraffin, and government had adopted a policy in 2000 to zero-rate paraffin. However, the real benefit was not passed to the beneficiaries. That raised the question what policy must do. The community survey would show the percentage of the population that used paraffin for lighting. There must be more interaction. Stats SA would supply the evidence, but policy options must then be drawn from this.

Dr Cassim said that the CPI should be made more useful to the policy debate. CPI should perhaps be judged not against expenditure, but rather on income for those in the lowest 25%. He hoped to expand on how the CPI could become more user friendly. Tension would come in during wage negotiations, when unions would realise what the CPI was for workers in a particular category.

Mr Singh asked if there was  is there information available on the most commonly-consumed products by the poor, and whether they were VAT rated.

Mr Lehohla said that the information from the poverty expenditure survey would generate information on products consumed. Stats SA could provide this information. That would be in the document on poverty measurements, including the percentage proportions of each product.

Ms Mokoto asked about cost efficiency; and asked to what extent Stats SA was ensuring that its services achieved value for money and no wastage of resources, and whether it was able to execute the activities under the budget. Some Departments had done virements, and she asked whether it was adhering to the requirements.

Ms Mokoena said that Page 101 of the glossy Work Programme outlined the initiatives to minimise costs These included better management, lessening of the budget for some items and ensuring elimination of wastage. The Bid and Budget Committees met regularly to ensure adherence to policy issues. The Budget Committee was sitting more regularly to monitor expenditure throughout the year, not just at the time of adjustment estimates. Some training programmes were now being done in house. There was more work to be done. There had been virements at the end of the year.

The meeting was adjourned.