Social Policy Model by National Institute for Economic Policy

SOCIAL DEVELOPMENT PORTFOLIO COMMITTEE
10 October 2001
SOCIAL POLICY MODEL BY NATIONAL INSTITUTE FOR ECONOMIC POLICY

Chairperson, Acting: Ms Ramotsamai (ANC)

Documents handed out:
National Institute for Economic Policy presentation (awaited)

SUMMARY
The meeting consisted of a briefing by the National Institute for Economic Policy on a tool that has been developed for assessing the implications and impacts of social policies. It allows parliamentarians and others to assess how many people are being targeted by proposed or existing programs and policies, what the demographic characteristics of the target groups are, what the budget implications of the policies are/will be and provides some indication of the likelihood that a particular policy will achieve its goals. Essentially a modelling tool, it also allows decision-makers to simulate various policy and implementation scenarios to make decisions about which policy will have the best cost-benefit ratio, for example. It can also assist in identifying when a policy as it is legislated is falling short of its objectives in practice, because the actual number of people being reached (obtained from administrative data) can be compared with the number of people the tool predicts should be reached. It works from existing databases of populations statistics, obtained by NIEP from Statistics SA, and is particularly effective in providing an idea of the effect Social Development policies will/should have on poverty, for example.

MINUTES
Social policy model - developed by National Institute for Economic Policy (NIEP)
Dr Adelzadeh started his briefing by saying that the NIEP had been concerned with developing a way to assess the impact of social policy on the people of South Africa, by for example looking at the transfer of tax funds to social policies aimed at poverty alleviation. He said that there is a massive gulf between the rich and the poor in this country, and this can only be diminished through social and economic policies. The NIEP had therefore developed a social policy model which was able to simulate the impacts of major government cash and non-cash transfers, as well as income tax, on individuals and families. The aim of this was to identify how much these affected poverty, who the beneficiaries were, and what impact these had on the budget. The model also allowed for the estimation of the likelihood that any policy would achieve its stated targets, but beyond this, and of more relevance to the committee, perhaps, it offered the opportunity to pre-test a policies goals to ensure that they will achieve the desired effect.

To this end, the model provides three types of analysis:
- Distributional: who is/will be better/worse off due to this policy?
- Budgetary: what are the policies financial and/or feasibility implications for the budget?
- Dynamic effects: what are/will be the dynamic macro-economic impacts of policies, for example what are the implications for growth, inflation etc.

The model works by using a database of individuals and households drawn from the census and government surveys conducted by Statistics SA. This database provides information on the characteristics of the population, and consists of information on 61000 families, with 3800 different columns of information i.e. 3 800 different variables or classification criteria. This data can then be manipulated to provide information on how many people meet the criteria for a specific government policy, for example. Various simulations are then possible, changing the parameters of the policy, to see the effect of changing the criteria or the measures that the policy spells out.

In this way the model simulates current policies, but also allows decision-makers to simulate the effects of proposed changes to eligibility and entitlement. The implications of each simulation on the budget, the number of people living in poverty etc. can then be compared. Dr Adelzadeh said that there were currently twelve modules of the model, but that more were being developed. He also said that it was possible to model different tax and VAT permutations, to find out where money could be obtained to fund different policies. The model also allowed an analysis of the effects of each permutation on the different sections of society.

A question was raised about whether it was possible for someone to receive more than one grant, particularly the old age pension and the child support grant in the context of more grandparents caring for AIDS orphans, for example.

Dr Adelzadeh replied that at the moment, the law stated that no-one could receive more than one grant, but that the rules might need to be relaxed, given the growing number of such cases.

Ms Gandhi (ANC) pointed out that she was aware that in her constituency, the department was open to providing grandparents with the child support grant.

Ms Ramotsamai asked whether the NIEP assessed conditions on the ground, before assessing the feasibility of any policies.

Dr Adelzadeh said that this could be factored into the model, with the parameters being set both as they appear in the law and as it is being applied. This has a benefit in that it allows decision-makers to assess discrepancies and identify weaknesses.

Dr Adelzadeh concluded his presentation by providing several examples of simulations which had been conducted (see document). For example, he provided one simulation which estimated the number of people living in poverty in the country at 24 million, or 56-57% of the population. He then went on to give indications of how this figure was divided by race, gender, family type and province. For example, approximately 22 million of those below the poverty line are black, 4.85 million live in single female-headed households, and the majority live in the Eastern Cape and KwaZulu Natal. He said the particular advantage was that this quantified what may be implicitly know, and help to identify goals for policies. He added that all these figures (presented on individual graphs) could be combined into one set of data. He said that any combination was possible and that his examples were limited by the two-dimensional nature of graphs for presentation purposes.

Discussion
Mr Olckers (NNP) asked if the model was able to determine the effect of for example the zero-rating on bread, for the poor as opposed to the rich.

Dr Adelzadeh said that this could be done by using another tax figure such as the standard 14% for VAT, and then comparing the distribution effects. He said that he had examples of that as well. He said that it could also be used to assess not only the impact on the number of people living below the poverty line, but could also measure the effect on the poverty gap, for those who remained below the poverty line.

He then went on to explain the services offered by NIEP. He said that the model had been designed to be user-friendly, so anyone could make use of it. It was also possible for users to connect to the Internet and access it, without any NIEP involvement. He also said that the NIEP had funding to train parliamentarians and asked if the committee felt that this was something which would be useful to them, and whether the members were interested in receiving such training.

Mr Olckers (NNP) asked what means test the model used for a Basic Income Grant (an example provided by Dr Adelzadeh). He also asked whether, when the graph was displayed, the criteria were visible.

Dr Adelzadeh replied that, as the income grant did not exist, they had used the means test for the old age pensions, but that any criteria could be set.

Ms Coetzee-Kasper (ANC) asked whether the department used this information. She also asked whether the legislature could use the information to change the budget.

Ms Ramotsamai asked what NIEP's interaction with the Taylor Commission of Enquiry into integrated social development was.

Dr Adelzadeh said that hundreds of scenarios had been run for the Taylor Commission over the last year, and the interaction was quite frequent. With regard to the department, he said that they had presented to the executive committee and the Director General on about four occasions. A training program had also been developed for department staff, and they were very interested in the model.

Ms Chalmers (ANC) asked whether they had looked at whether or not the Child Support Grant had caught up to the old Maintenance Grant in terms of coverage etc.

Dr Adelzadeh said that this had not been done, but that they had looked at the gap between those eligible for the Child Support Grant and those actually receiving it. He said that this was due to the changeover between the two systems, as well as certain other factors.

Ms Ramotsamai asked whether there was a scenario incorporating free healthcare into a poverty reduction program. She also asked if they looked at where government could find the money to pay.

Dr Adelzadeh said that work had been done on free health with the department on occasion, but that this had been affected by the Taylor Commission. He said that as yet, there was no comprehensive module available. He said that in terms of funding, work had been done on tax and VAT, but that changing macro-economic policies could be factored into the model, such as reducing the deficit-reduction target etc. He said that they had presented this to the Department of Finance, who did not have the models, as they were a relatively new innovation. He then provided one example of assessing different financing options, such as a progressive tax. He illustrated the effects of this on various sectors of the population. He did however caution that this assumed the ideal, and did not factor in administrative practicalities of trying to collect all the tax.

Ms Chalmers (ANC) asked if it was possible to cost youth crime prevention strategies, early childhood interventions etc., and whether work was done with the Department of Education in terms of job creation and skills development.

Dr Adelzadeh said that the model allowed for both cash and non-cash transfers to be assessed, so it was possible to assess a 'non-cash transfer' such as a support service which would have some impact on poverty. He said that there were ways of calculating the cash implications of these services, in terms of expenditure and impact.

Training by the Institute
Mr Rangata, responsible for training at the NIEP, continued the presentation. He noted that the core business of the NIEP was research, but that last year a capacity building section had been established. This had two foci: to work with previously disadvantaged universities and teaching institutions; and with decision-makers such as those in government. In terms of work with government, he said that training was due to begin with the Department of Trade and Industry next week, and with the Finance Committee either later in the year or early in 2002. He said that if the committee were interested, it would be possible to design a training program around its members' specific needs. He also said that there was training for both COSATU and NEDLAC. He reiterated that training for parliamentarians was completely funded by the African Capacity-building Foundation.

Discussion
Ms Chalmers asked if the training was on computers and therefore required some competency on computers.

Mr Rangata said that it depended on the needs of the members. It was possible to go into depth on the way the model constructed its scenarios and assessed the impacts, but that this largely required an economic background. But for most policy-makers, this was not required.

Dr Adelzadeh said that most people were not interested in exactly how it worked, but that for some in various departments in particular, training in model building could be provided.

Ms Gandhi (ANC) said that the committee, if not government as a whole, had a problem with financing, in terms of seeing whether departments were using funding properly, as well as identifying gaps etc. she asked if this would assist in identifying problems and gaps.

Dr Adelzadeh said that it could do that.

Ms Ramotsamai said that the briefing had been vitally important, because the committee acted in a mainly oversight capacity, and it seemed clear that the model had some applications in this area. She suggested that the clerk of the committee could put together some ideas, in order to develop a training program and set targets for that. She thanked the presenters, and asked that the committee might be provided with documentation on the model and the presentation.

Mr Rangata offered brochures which he had brought to the committee.

The meeting was adjourned.