ATC080528: Report on Budget Vote 11: Statistics South Africa

Finance Standing Committee

Report of the Portfolio Committee on Finance on Budget Vote 11 (Statistics South Africa), dated 28 may 2008

The Portfolio Committee on Finance having been briefed and deliberated on the Budget Vote 11 and the Work Programme 2008/09 – 2010/11 of Statistics South Africa (StatsSA) reports as follows:

1. INTRODUCTION
On 23 May 2008 the Deputy Minister of Finance, Statistician-General, and senior officials from StatsSA briefed the Committee on the WorkProgramme 2008/09 – 2010/11. StatsSA outlined  its key priorities for 2008/09 in line with its strategy.  

According to the 2008/09 – 2010/11 Work Programme, StatsSA’s legislative mandate relates to providing the state with information about the economic, demographic, social and environmental situation in the country. In the introduction to the presentation, the Statistician-General elaborated on the strategic importance of statistics. According to the Statistician-General, statistics provided by StatsSA help policy-makers in evidence-based decision making, managing results, deepening transparency, and for broadening and deepening democracy. The strategic importance attached to statistics is also emphasised in the 2008/09 – 2010/11 Work Programme. Specifically, the Work Programme points out that the availability of comprehensive, relevant, timely and accurate statistical information is crucial for effective policy making, development planning and decision-making, as well as for monitoring and evaluating.

This report consists of different sections. Section 2 gives an overview of the key priorities of Stats SA for the 2008/09 financial year. Section 3 provides details of Stats SA’s strategic risks and challenges, and an overview of what is envisaged beyond 2008. Section 4 provides a summary of questions by members of the Committee and responses.

2. KEY PRIORITIES  
The Statistician-General gave an overview of the key priorities of StatsSA with a discussion on the characteristics of a statistical organisation, the role of statistics towards evidence based decision-making and the critical societal issues and how statistics relate to them. According to the Statistician-General, the characteristics of a statistical organisation include the following: 

Intellectual;
Technological;
Political;
Logistics; and
Administrative.

These characteristics form the basis for the effective functioning of StatsSA and the Statistician-General made it clear that each characteristic carries an equal weighting. As mentioned above, statistics are needed for evidence based decision-making. In other words, the quality of statistics informs the quality of decisions that are being made. Society needs basic macro economic and social information to inform decision making in both the public and private sectors. Key macro economic and social information could include details of economic growth, price stability, employment and job creation, life circumstances, service delivery, poverty and demographic profiles, and population dynamics. Providing reliable information is perhaps more important today than in the past, specifically in light of changes in interest rates, increasing food prices, labour market dynamics and the fight against poverty.

The following are the key priorities of Stats SA for 2008/09:

Quarterly Labour Force Survey;
Living Conditions Survey;
Reweighted CPI;
Reviewing of Producer Price Index;
Geo-referenced dwelling frame;
National Strategy for Development of Statistics; and
International activities and the International Statistics Institute (ISI).

2.1 Quarterly Labour Force Survey
The aim of the Quarterly Labour Force Survey is to comprehensively measure the dynamics of the labour market. The Labour Force Survey is the primary instrument to measure employment. Re-engineering the Labour Force Survey started in 2006. The Statistician-General pointed out that data collection on the Quarterly Labour Force Survey were initiated in January 2008. The improvements with the new Quarterly Labour Force Survey include reviewed concepts and definitions of key indicators, frequency, timeliness, redesigned questionnaires and processes. According to the Statistician-General, the first results of the Quarterly Labour Force Survey will be published in August 2008, and from then on results will be published on a quarterly basis. The Quarterly Labour Force Survey will provide details of seasonal variation, which will help to inform policy makers.

2.2 Consumer Price Index (CPI)
The aim of the CPI is to measure price changes as experienced by consumers over time. The Statistician-General pointed out the 2005 Income and Expenditure Survey provided key inputs in the new weightings for the CPI basket. The 2005 Income and Expenditure Survey used a diary method whereby respondents were not seen for one day only, but rather over a period of 12 months. This provided useful information on consumer spending patterns. StatsSA is currently running a parallel sample on new products. The new weights for the reweighting of the CPI will be published by June 2008, while the reweighted CPI will be released in February 2009.

2.3 Producer Price Index (PPI)
The PPI shows monthly movement in prices of producers for a part of the economy (excluding services). The PPI has last been reviewed andreweighted in 1999. According to the Statistician-General, the effect of not reviewing the PPI series over the last 10 years is that an outmoded product classification and basket of goods have emerged which is unreliable. The Statistician-General also pointed out that during the reweighting, unexpected problems occurred. The Statistician-General concluded by stating that the PPI is on a stable footing this year.

2.4 Living Conditions Survey
According to the 2008/09 – 2010/11 Work Programme, the aim of the Living Conditions Survey is to measure the extent and distribution of household poverty and monitoring the progress in poverty reduction. The Statistician-General made it clear that measuring poverty should take on a multidimensional approach. StatsSA has engaged with the relevant stakeholders and is in the process of finalising the survey instrument and methodology. As the 2008/09 – 2010/11 Work Programme points out that the survey will be conducted in collaboration with other public policy initiatives that include stakeholders such as the Department of Social Development, National Treasury, the Presidency, and institutions undertaking other government commissioned studies.  The survey will take place over a 12-month period and data collection will commence in November 2008. The 2008/09 – 2010/11 Work Programme states the results will be available in 2010. 

2.5 Geo-referenced dwelling frame
According to the Statistician-General, the aim of the Geo-referenced dwelling frame is to compile a complete, up-to-date database of all dwellings and other structures in the country. Understanding geographical statistics is important as it helps us in understanding the dynamics of geographical settings in terms of investments, service delivery, etc. The Statistician-General pointed out the need to compile an up-to-date list of all dwellings and structures in South Africa. According to the 2008/09 – 2010/11 Work Programme, the dwelling frame will be completed during 2009/10. According to the Statistician-General, 43 per cent of the dwelling frame will be completed by March 2009.

2.6 National Statistics System (NSS)
According to the 2008/09 – 2010/11 Work Programme, the adoption of common concepts, definitions, classifications and standards is the cornerstone of statistical coordination in the NSS. The 2008/09 – 2010/11 Work Programme points out that one of the strategic objectives of the NSS is developing a National Strategy for Development of Statistics (NSDS). StatsSA has prioritized the development of the NSDS to provide a framework for building and sustaining capacity to produce national statistics fit for use according to defined quality standards. According to the Statistician-General, the NSDS is to be compiled in consultation with other stakeholders. Stakeholders will be trained in the South African Statistical Quality Assessment Framework. The priorities in terms of statistical support differ as follows across national, provincial and municipal levels:


National level: Education, Health and Home Affairs;
Provincial level: Provincial profiles; and
Municipal level: Raising awareness of statistics.

It is envisaged that a draft NSDS will be available by March 2009.

2.7 International activities and the International Statistics Institute (ISI)
StatsSA will host the 57th session of the ISI during August 2009. In the 122-year history of the ISI, it will be the first time that a sub-Saharan country hosts this session. The ISIbalo Capacity Building Programme has been developed as a principal activity of the 57th Session of the ISI. TheISIbalo Capacity Building Programme has specific objectives, which are as follows:

Facilitating the broader participation of Africans at the ISI 2009;
The development of segmented cadres of statisticians and demographers in Africa; 
Developing statistical education in Africa;
Promoting research in science with a fragmented history of African achievement; and promoting statistical utilization in Africa for evidence based policy making and creating a diverse range of sharing best practice.

3 STRATEGIC RISKS AND CHALLENGES         
The Statistician-General provided the following list of strategic challenges and risks facing StatsSA:

Data Management and Information Delivery project;
Business register;
Building leadership and management capabilities;
Strengthening analytical and methodological capabilities;
Corporate data processing;
Integrated fieldwork; and
Relocating Stats SA.

In addition to these strategic challenges and risks, Stats SA will be focusing on the following key issues over the period:

Benchmarking Gross Domestic Product;
Living Conditions Survey;
Statistical Training Institute;
ISI in 2009; 
Census 2011;
New buildings in Tshwane; and
Climate change.


In concluding the presentation, the Statistician-General highlighted the following key points:
For the first time, comprehensive employment data will be available on a quarterly basis;
Information will be readily available on the changing consumption patterns in the economy; and
For the first time, a comprehensive and integrated poverty profile is on its way through the Living Conditions Survey.

4. Committee deliberations
Having deliberated and sought clarity related to the 2008/09 – 2010/11 Work Programme, the Portfolio Committee on Finance raised concerns on:

National Business Register;
Release of quarterly GDP figures;
Capacity to measure GDP from the expenditure side;
Additional personnel, staff competency and skills;
Relationship with other Departments;
Producer Price Index (PPI) and Consumer Price Index (CPI) statistics;
Lags of statistical process;
Poverty Index;
Measurement of non observed economy;
Adequate accommodation for 2010;
Choosing of measurements;
Non profit organisations;
Demographic shifts;
Public image of StatsSA;
Self employed survey;
Zero rating of food stuffs;

Responses.
In response to the question on the business register, StatsSA indicated that it is an inter-departmental and inter-disciplinary matter that does not hang on the capacity of StatsSA but relates to some of the key departments. It also relates to the provisions of the Companies Act. Various stakeholders are obligated to provide the necessary information to complete the business register. StatsSA is in the process of addressing this issue as it will improve the capacity of SARS to generate the different revenue resources and provide accurate projections. 

In terms of the Gross Domestic Product (GDP) and its relevance to the economy, StatsSA reported that GDP is an important factor to provide data on a quarterly and annual basis. Quarterly data is important because it allows policy and decision–makers to make short-term decisions. Currently the entire economy is covered in GDP. The challenge lies with the generic services sector and the overall estimate of the services sector. Transport, communication and manufacturing were chosen, because of their importance to the economy and the fact that these sectors are easier to measure.

StatsSA highlighted that South Africa is the only country where a division of labour exists when estimating the GDP. The statistical agency calculates GDP from the production side while the SARB temporarily calculate GDP from the expenditure side. StatsSA highlighted that it requires their own capacity in estimating the expenditure side. Because of capacity constraints, national accounts on the production side alone pose challenges. StatsSA believes by improving their capacity to measure the expenditure side will strengthen their ability to engage SARB on the expenditure side. Whether this will be a legitimate issue will be an important debate. 

Regarding recruitment, StatsSA has put in place capacity building programmes and internship programmes to attract more people. StatsSA is also communicating to professional organisations to assist in acquiring certain skills.  Bursaries have been offered to matriculates to obtain certificates, diplomas and degrees.  A retention strategy, that will be finalised this year, has been put in place to retain employees.

StatsSA has undertaken partnerships with tertiary institutions over and above in-house collaboration and in-house training that will ensure thatSouth Africa will provide trained people within the sector. Tertiary institutions must also play their role to deliver appropriate skilled persons. Young people should be motivated to start studying in the discipline of statistics. Stats SA has taken the initiative by sending young people to Tanzaniato further their studies. Training must focus on how technology, statistics and geography are related to one another. 

 
StatsSA interact with other departments to the extend of the National Account of Natural Resources. The issue of climate change as a measurement has only emerged from the UN Statistics Commission. Although Statistical measures on these will take one to three years, some countries have already conduct studies on how to measure climate changes.

Regarding the PPI, Stats SA highlighted that the PPI is an indicator for 30 per cent of the value added of the economy. One critical challenge thatStatsSA faces is how well it educates and communicates users about a number of problems and issues that it has. In moving from the old to the new transition some unanticipated complexities arises. When a series exists that has not been change for some time, it becomes less relevant to the economy because of the change of dynamics in the economy. New and international classifications have been introduced and products are classified differently. The PPI is far more complex than the CPI. For the CPI, StatsSA has enumerators that on a monthly basis collect prices from stores. StatsSA has control over this process whereby enumerators have to give the price in at a certain day. StatsSA is dependant on firms where it comes to the PPI. This is a series where StatsSA collects information for two weeks and 2/3 days after collection process the information. Information in terms of data that some firms provide may be incorrect. When StatsSA released the PPI, it had indicated there were certain problems which would be resolved and that revisions could be expected.

In the case of the Consumer Price Index (CPI), StatsSA is totally cautious because the CPI is based on wages, which is a key input into monetary policy. The resources that are put into the CPI are about five times more of what is put into the PPI. A lot of energy and effort has been put on the CPI.  

Time lags differ from size and burden per survey. For example when the statistics on the Community Survey were collected in February 2007, it was only released in October 2007. The survey could not include the level of municipalities, which was release by March 2008. In terms of the re-engineering of the QLSF, release will only commence four weeks after collection of each quarter. The DHS was release nine months after the data was collected.  The Living Conditions Survey will be a continuous survey, but the collection of data will be concluded at the end of October 2009. As this is a continuous survey, the data processing will go alongside the data collection and will be released after April 2010. Migration data will be released on a monthly basis, but it would be the month of the previous year. Mortality data will be released every year. The mortality data for 2007 will be released in 2008. The time lag will be determined by the amount of work of data collection. 

Regarding the poverty index, StatsSA reported that progress has been made and that the technical work is almost concluded. Policy discussions will have to begin, however, poverty is multi-dimensional in terms of absolute poverty, relative poverty and perceptions around poverty that are critical and important. StatsSA is confine with absolute poverty, that is measurable and also the relevant measure. The poverty survey will take into account relative poverty. The poverty index is based on the minimum food need for daily requirement. Within this there would be a threshold above and below that would be determined as the poverty line.

The Millennium Development Goals are one of the objectives that have been embraced by South Africa and amongst them is the halving of poverty. It is one of the objectives of the South African government to half poverty by 2014. Government institutions, NGOs and various stakeholders are engaging one another to have consensus around poverty that will have an impact on what the poverty line should be. StatsSA will provide the technical input, but the political process is entirely off the hands of StatsSA.

Regarding the non-observed economy, StatsSA indicated that although this is a very complex area to measure, estimates from some countries are based on an inverse of what the police are able to capture and could be calculated 20 times the value in the non existing economy. The statistical system that economic statistics relies on is the business register, labour force survey and to some extend, the income and expenditure survey. Tracking illegal activity for publishing has to be estimates and cannot be official due to the nature of the activity.

With respect to accommodation, StatsSA reported that it had released the previous day figures of growth in the accommodation sector, not in the number of houses, but in the revenue that it has generated. March-to-March growth indicates about 38 per cent. There are also increases in the volumes of people that are using accommodation.

Regarding how systems are chosen, StatsSA indicated that it is moving into operation efficiency and operation effectiveness. During the past five years, StatsSA had position itself through addressing the kind of problems around the data sets it had. Operation efficiency and effectiveness will depend how well the systems are harmonised and how well the structure responds to it. In order to achieve this, StatsSA had realised the need to expand the management and other layers. This could be achieved through the approved structure. At a provincial level, the structure had been elevated to that of a chief director. There is also a programme manager that deals with the difficult issues of integration. Officials from StatsSA, on a rotational basis, will attend intense training on fieldwork in Istak during this year. Simultaneously, a training programme has been developed to train field operators. There is also clear review with the assistance of international experts when decisions are taken. International peer review and user groups were second factors, which determine how to measure. 

Regarding non-profit organisations, StatsSA responded that the effectiveness of these institutions depend on how it contribute to GDP. In South Africa this is not significant although it might be significant as volunteer work. More affluent societies are now doing volunteer and humanitarian work. It is becoming an increasingly important factor in the developing society, particularly in counties that are very fragile. The non-profit sector becomes important as South Africa develops.

Regarding demographic shifts, StatsSA highlighted that Gauteng and Western Cape were previously not only labour receiving provinces but are now also children receiving provinces. This is because children migrate with their parents to Gauteng. The number of household in Gautengbetween 2001 and 2007 moved from 2.5 to 3 million. These are the dramatic shifts that the economy is facing. The issues that policy has to grapple with are whether this  is in terms of urban space function, or other elements that will lead to diseconomies. 

In Eastern Cape, the number of employees in agriculture has increase significantly. Eastern Cape accounts for 60 per cent in the growth of employment in agriculture. This could also be seen in another cycle of the labour force survey.  A possible question for the improvement of the figures is whether people did not move from the Eastern Cape to Gauteng and the Western Cape?

StatsSA reported that its public image is largely affected by the contested views when the statistics are produced. The corporate and public image will also be affected by the scientific nature and trust in official statistics, internal practices within StatsSA and the history of the organisation that operated without a communication unit. StatsSA engage, train, and work with journalists on certain events during lock-up sessions. Service providers have been appointed to train senior staff members on how to deal with the media. This has resulted in an improved relationship with journalists. 

Regarding the self-employed survey, Stats SA indicated that it had concluded two surveys; one in 2002 and the other in the 1990’s. The results of the size of government intervention were, as reported by the self-employed are far less (loans) of what they received from informal companies to start their businesses. Seventy five per cent of small businesses that were created received money that they borrow from friends or raised by their family. Both surveys indicated that less that half a per cent traces what government schemes might be.

With respect to zero rating, StatsSA reported that it had observed overtime that the increase in cents per litre in petrol, paraffin and diesel as of 1997, 1998 and 1999 were higher for paraffin. Government adopted a policy in 2000 to zero rate paraffin. The benefit, however, was not passed to the beneficiaries. When prices increase per cents per litre it increases more for paraffin. The community survey indicates the percentage of population that uses paraffin for cooking and lighting. StatsSA supply the statistics and evidence, but requires policy to deal with policy options that are at the disposal of government. 

5. Recommendations
Based on the deliberations with Statistics South Africa on the Work Programme 2008/09 – 2010/11, the Committee recommends that:

StatsSA provides a detailed report to the Committee on progress made towards rectifying its public image;
StatsSA provides a detailed report to the Committee on progress made towards achieving targets as outlined under each key priority for 2008/09; and
StatsSA to timeously inform the Committee of any statistical information that could have a negative impact on the economy.