220.127.116.11 Resistance and lack of institutional and legal framework for implementing The fifth problem was resistance from the old guard and those who sympathised with them. The sixth was the lack of an institutional and legal framework within which plans could be implemented. The Tender Board was often mentioned as the biggest stumbling block hampering delivery. Geographically, the settlements were still divided and without the existence of seamless local government structures, implementation was to remain the biggest challenge, particularly for the priority areas that were identified by the RDP and the first 100 days.
18.104.22.168 Lack of statistical leadership and vision The sixth and biggest problem perhaps that faced the statistics system in South Africa before democracy, and in the first six years of democratic existence, was the lack of vision for a national statistics system. The focus in the six years of democracy focused on operational effectiveness to the exclusion of an overall strategic framework that determines “why official statistics.”
22.214.171.124 Lack of co-ordination of national priorities The seventh problem that made it difficult to have an effective statistics system was the absence of the digestive capacity to interpret government policies and priorities and appropriate them in a systematic way, where sequencing of processes determine how priorities interlock in a self supporting way. There is a sense that with the demise of central planning in the Eastern Block, South Africa tended to skirt around the issue of co-ordination and probably got to the point of almost loathing it. Statistics in general, responds to policy. The value chain of information starts with defining a purpose in life for a country, and then locating and describing ways of implementation and finally instituting measures of verification. It is in the terrain of verification that official statistics become important. Statistics are a rudder that guides policy formulation, monitors implementation and measures as well as evaluates performance. So they serve a client who is the policy maker and those who implement it. One key pitfall from the experiences of the RDP Office was to subscribe to being lean as a virtue without assessing the capacity required for organising issues of state and governance. The lean and mean philosophy while virtuous, requires to be contexualised. The current efforts by the cluster system to strengthen the centre of government have just begun to address issues of co-ordination and effectiveness at the level of strategy and possibly redress the limitations of the RDP office.
The arena of statistics as it existed and continues to exist, consists of a system that produces what the office best thought was good for government and the stakeholders. However, the office has recognised that it continued to inform at a tangent, lacked a system of indicators and missed the audience. South Africa has not been alone in this dilemma of wandering in statistical wilderness as the series of case studies illustrates. It was battling with problems of a unique nature and these were very country specific.
6.10.7 The watershed meeting on statistics system (the North West axis) In October 1994, the provincial government of the North West hosted a watershed meeting which emanated as a sequel of meetings that the North West hosted bilaterally with all nine provinces and jointly with the then FOSAD equivalent, on the future of national statistics in South Africa. The meeting was attended by all RDP offices and the provincial and local government representatives as well as national ministries.
Although the CSS was invited, they declined to participate as they believed that their transformation process was on course and the planning workshop was inadvisable. A key resolution of the meeting, communicated to both the CSS and the RDP office was that the organisation needed serious surgery including advertising the top position of head of the organisation. Census ‘96 was discussed extensively and implementation plans were also looked into. The conclusion was that the census was in jeopardy and a recommendation for an assessment mission for readiness was made. The mission was conducted and resolved that the census be postponed by six months and be headed by a different individual to the then incumbent. The meeting also discussed the geography of the country and the need to re-organise the geo-political and economic space of South Africa. This could be viewed as the pre-cursor to the Demarcation Board. On national accounts, the meeting resolved that the CSS had to be strengthened in order to take charge of both sides of the national account. This is in order to eliminate the moral hazard of a policy department, in this case the South African Reserve Bank, (SARB), handling national accounts statistics, i.e., rendering the bank the role of referee and player. On the CPI, the meeting resolved that there is need to collect information for the compilation of CPI for the rural areas so as to have a targeted handle on how the rural communities get affected by price changes. It was further indicated that the CPI in the way it is collated cannot be representative and required revamping. It covered only 14 urban areas. More fundamentally the meeting resolved on the creation and funding of a national statistics system and strengthening training of cadets who would participate in the system. Lastly the workshop appealed for a more representative Statistics Council. This workshop was amongst the first few to attempt a strategic focus on the provision of statistics in South Africa.
The North West administrative and political leadership axis jointly with the RDP office, provided the space for challenging quite successfully what was becoming a myopic self-serving statistical hegemony based in Pretoria. This led to visible changes at the Central Statistical Services from July 1995 and subsequent improvements in timeliness and quality, the latter albeit gradual and problematic.
This workshop offers South Africa a golden opportunity to act strategically in terms of managing statistics. If there is anything to learn from Australia, Canada and more recently Mozambique and Uganda, it is ways of providing a strategic framework for the management of statistics and methods of operationalising these activities within this framework. From the ABS we can learn that: a “strategic management process delivers ever changing strategies which move the ABS in the directions set by the objectives in the plan. Some strategies in the plan will not be fully realised; on the other hand, new strategies will emerge from across the organisation and be implemented. The success of this process depends on how well these directions are understood, accepted and pursued by all staff.”17 6.10.8 The 1996 Census Hopes were pinned on the 1996 Census results, however, these brewed their own controversy, particularly the preliminary results which suggested that we were not as many as we usually thought. The release of a final result that was two million people more did not help to build the image of the organisation.
The census has largely helped to ameliorate the paucity of data for planning. However, the worst government wide mistake committed was the selling of the data. While the price per dataset was not prohibitive, the notion of selling turned departments and users off as they saw this practice as unorthodox and ran counter co-operative governance. Besides, it was argued that the census had already been paid for by the fiscus. The facts facing Statistics South Africa were however that, they had to balance their budget and the sale of data to recover R 7million was real. Government was ruthlessly chasing the reduction of government deficit and any over-expenditure was dealt with ruthless discipline. This era brought about discipline and reduced the deficit to levels that are unprecedented across government. The abrupt discontinuation of the October Household Survey programme is bound to create discontinuities in the ability of government to measure progress. Ad hoc implementation of surveys mitigates against any notion of statistical dependability in the future. The leadership of the country has begun to apply pressure on the statistical system and require it to deliver.
6.10.9 Political leadership and statistical challenges In the January 2002 Cabinet Lekgotla, after an extended discussion on communication, the Lekgotla focused on what it is that is to be communicated, i.e., the content of communication. It was concluded that such content will be of a nature of facts and figures, i.e., of a statistical nature. The President in the same Lekgotla alluded to the desire to paint a picture of seven years of democracy. Such a picture depends on the indicators to be provided by the statistical system.
In the last two State of the Nation Addresses that the President made, I have noted that in 2001, only two sets of statistics out of the five that were used came from Statistics South Africa. In the State of the Nation Address of 2002, none of the figures coming from Statistics South Africa were used. The question that this state of affairs begs is how relevant is the organisation, how accessible are the statistics it produces and what is the level of public trust the statistics are accorded.
In the same Cabinet Lekgotla, as already indicated elsewhere in the paper, the President gave a clear instruction to have the national statistics system to work and deliver. Furthermore, in an encounter of the Directors General with the President, of the four agenda items tabled, one was on statistics. There is no doubt that at the highest level of political authority, statistics are not only viewed as important, but are urgently required for managing public affairs. If you can’t measure it you cannot manage it.
The Minister of Finance has posed questions on the veracity of statistics and what he sees is that they fail to add up. He has called upon the organisation to embark on an agenda for improvement.
The conclusion one draws from this series of interventions is that there is political will and leadership to get the statistics of this country on a sound footing. What is needed is for the bureaucracy to locate its position within the already established value chain and align its strategy. This workshop guides such positioning.
6.10.10 Agenda for statistical improvement By 1999, Parliament passed the Statistics Act, Act 6 of 1999 which empowered the Statistician General to co-ordinate official statistics. The Act also raised the level of authority of the organisation. In November 2000, a Statistician General was appointed. The new leadership has focused on improving leadership and changing management practices. The drive has been towards a flatter structure, promoting a learning organisation and allowing lateral information flows. Managing and encouraging exchange of experiences and personnel on the continent has witnessed a commendable measure of success.
Following this appointment, several sessions have been held with the Statistics Council on the agenda for improving statistics and clarifying roles and responsibilities of Council in relation to Statistics South Africa in general and the National Statistics System in particular.
By January 2001 a document on the creation of a national statistics system was tabled before FOSAD for consideration and FOSAD adopted it. By September 2001 working teams for the NSS were established and by February 2002 a fully functional NSS unit was established in Statistics South Africa. Within a few months of it being operational, cross-departmental work teams have been established and a strategy document on the NSS has been prepared and adopted by the FOSAD meeting of November 2001. In conjunction with the Presidency, Statistics South Africa will lead the implementation of the NSS and this indicator framework workshop provides an opportunity for users, producers and suppliers to exchange views on priorities. A series of bilateral meetings with departments have been rolled out since February 2002 and will continue. Included in these meetings, we focused training of staff in government departments on the use of data holdings that Stats SA possesses.
A quality and methodology team has been set up at Statistics South Africa to monitor, advise and ensure that quality is not only implemented but is seen to be done. A systematic process of browsing through each series and identifying areas for improvement has begun. For instance preparations are afoot to look at Census ‘96 and confront it with that of 2001 even before Census 2001 yields results.
A rigorous statistical training programme has begun. Up to 32 members of staff are studying statistical techniques with 22 of them being fulltime. The organisation has set aside about 3% of its total budget for statistical training and commits to at least each staff member receiving a minimum of six days training a year. Furthermore the organisation has set aside R1 million for appointment of internships for periods not exceeding three months. On the people with disabilities, we have moved deliberately to increase our intake, particularly in data processing and jointly with the department of labour have conducted a very successful training that has brought people with disabilities on board with competencies that meet requirements.
We have started a profound programme for promoting statistical literacy as a means of achieving economic literacy. To this end, in conjunction with the Department of Education and SAMDI we are developing materials from the census, especially the census at schools for training pupils in the handling of data.
The organisation leads the pack in adopting state of the art technology, especially in data processing. The challenge is that of integrating disparate systems into one seamless data holding with interfaces to the user community.
On the agenda at the turn of the century are challenges of measurement on causes of death. Extent and nature of crime, measurement of incidence and prevalence of HIV/AIDS, patterns and levels of enrolment, especially within the context of declining number of pupils sitting matric exams. Employment statistics are being revamped through the use of better sampling scheme.
Notable improvements and collaboration is being made on the GIS as a corporate asset for South Africa incorporated.
6.10.11 Conclusion In South Africa, the first five years of statistical freedom can be noted for what could best be termed as the “data spaghetti era” as well as profound lack of common vision on the nature and direction of the statistics system. There was relentless focus on operational effectiveness with less attention on strategy and direction in which the organisation and the system were to move. A less favourable resource base for building statistical infrastructure hampered rapid development of the organisation. Some series like the October Household Survey, had to be abandoned because of funding problems and have been replaced by yet an important series, the labour force survey which does not however, cover the extent and modules included in the OHS. The political leadership is fully aware of the need for improving the statistics system and indeed has elevated the status of the organisation to the level of a fully fledged department. However, as can be seen in the case of successful statistics systems, these systems take a while to develop, they consume resources as collections to inform increase in number, and they succeed where training is life long. Without strategic management a statistics organisation cannot survive. For South Africa, it appears that like in Mozambique and Uganda, the political leadership has resolved to actively support the coming in to being of the institution.
7.0 The to do’s From the country studies we can learn that there are several “to do’s” for the improvement of the statistics system in South Africa. Below is the list that I trust could guide resolutions that can be implemented:
Auditing the system and its current products;
Producing a strategic plan with key deliverables, identified beneficiaries, costs of not having a national statistics system and associated resource inputs;
Adopting a planning and implementation framework;
Ensuring that statistics are of good quality;
Implementing peer review for production;
Recommitting the work teams to the improvement agenda;
Delivering an indicator template and indicators;
Delivering a picture for the seven years of democracy;
Building the infrastructure for production of statistics, such as field operations;
Training staff and allowing for fulltime study from time to time;