/ 23 February 2001

Spend better, not more

Alta Folscher

Spending on health, education, social welfare and housing will not keep up with population growth over the next three years, according to government projections.

After several years of real growth in social spending, this change reflects the government’s view that existing funds should be disbursed more effectively.

Spending on these areas is critical to poverty alleviation. The high levels of poverty in South Africa the development report of the United Nations Development Programme measures absolute poverty at 45% impacts negatively on future growth and development prospects.

The budget can be used in several ways to address poverty. However, the main instrument to directly alleviate the effects of poverty remains spending on social services, including health, housing and welfare. Social services also remain critical in addressing the root causes of poverty, through education, social development and primary health care spending.

Changes in total spending on social services in South Africa do not always reflect their poverty bias. Even though the sector has seen on average 3,9% real increase in spending from 1995/96 to 2000/01, spending over the next three years will not keep up with population growth of 2,5% a year.

Furthermore, the social service share of the budget is set to decline over the next three years. Why is this? There are three aspects that should be taken into account.

Firstly, spending on social services is driven by personnel expenditure. Changes in wage policy in 1996 drove spending on personnel up across all sectors in the period 1996/97 to 1998/99, but disproportionately so in the social services, where overall spending peaked at 60% of non- interest expenditure (19,1% real increase over the first two years of the period). Similarly, in 1998/99 and 1999/00 the general curbs on spending on personnel caused a decline in social spending. It has since stabilised.

Secondly, spending in the social sector has been the victim of shifts in priorities. Budget 2000 saw a real increase in defence spending of 32% after real declines since 1994. This increase was partly paid for by the stabilisation of social services spending.

Thirdly, the government has argued that no further reprioritisation towards social services is necessary. Any further improvements in the number and quality of services in the sector should come from spending the existing money better. On the face of it, it is hard to argue with this. In 1995 South Africa was already spending more on social services than other middle-income countries. However, our spending outcomes were much worse.

Shifts between departments in social spending since 1995/96 show a marginal reprioritisation towards health and welfare spending. Since spending in health and welfare is more demand-driven than in education and the forward projections of spending are in part based on historical spending patterns, there is the danger that HIV/ Aids spending pressure will increasingly put education at a disadvantage. Given the shortage of skills in South Africa the long-term impact of this trend should not be disregarded.

Health spending over the next three years in terms of Budget 2001 is R9,6-billion less than the Budget 2000 projections, while welfare has gained a total of R22,2-billion and education R10,4-billion. The decreased allocation for health is of concern in the light of the pressure on health budgets from HIV/Aids. While the increased welfare spending is to be welcomed for the same reason, the decreases in projected health spending should have been prevented by allocating more of the extra money available in the national budget over the medium term towards social services.

Social spending since the mid-1990s has also been characterised by a shift within departments between programmes aimed at the historically more advantaged communities (for example, tertiary health and education spending) towards programmes aimed at historically disadvantaged communities (for example, primary health care). In welfare policy changes in the type and magnitude of social entitlements were aimed at improving equity. However, take-up of the entitlements is still skewed towards historically advantaged communities and areas.

Budget 2001 admits that relative spending on social services, especially in the provinces, is set to decline over the next three years. This, it says, would be off-set by efficiency gains in the sector, which would be even more critical as the real increases in social spending lag behind population growth. So far, however, there has been little evidence of such improvements despite similarly constrained budgets.

Ironically the potential cause for continued poor service delivery over the next three years again lies in personnel expenditure. The quality and quantity of services delivered is a function of the number of people employed, their skill levels and motivation. Curbs on personnel expenditure since 1997/98 have meant that capacity has been eroded: it was often those with better skills and who could find jobs elsewhere that exited the public service. Departments also tended not to fill vacancies. To argue that a continued decline in the personnel expenditure share of social spending would free up resources for efficiency gains represents a misunderstanding of the people-intensive nature of social services, especially in the light of already depleted skill levels.

The fact remains that when budgets are squeezed, public servants have two choices: they either deliver the same services for more, or they deliver fewer or poorer services. In the absence of quantifiable service delivery indicators, it is difficult to make a judgement on which of the alternatives will prevail. Anecdotal evidence suggests it may be the second. Reflection on the connection between pay and service delivery supports this.

Alta Folscher works for Idasa’s Budget Information Service