Applied Fiscal Research Centre, UCT: Fiscal Risk Management
SUBMISSION TO THE PORTFOLIO COMMITTEE ON FINANCE
18 February 1999

Fiscal risk management
Recent turbulence in global markets and associated effects on interest rates and exchange rates has put in sharp relief the risk associated with macroeconomic policy, as well as the imperative to manage that risk as part of good fiscal governance. In fact this volatility seems to be becoming a structural feature of an increasingly globalised economy.

It is stating the obvious that the realism of the macro framework is dependant on the credibility of the underlying assumptions on which it based. A realistic budget is based on assumptions that have a high probability of occurring and when it is formulated with the intent to actually implement the revenue and spending policies it delineates. In this budget, the macroeconomic projections underpinning the fiscal framework were revised downwards (see \* MERGEFORMAT Table 1 below).

Table 1: Comparison of 1998 budget and 1999 budget macroeconomic projections: fiscal years 1997/98 to 2000/01

Source: Budget Review 1999

However, there is a very real concern that these forecasts may still be over-optimistic. To the extent that projections are not realised, this destabilises the budget and creates a demand for increased public sector borrowing. Last year aggregate fiscal discipline was maintained largely due to larger than anticipated revenue collection which offset sharply rising debt service costs due to higher interest rates. However this buoyancy in tax revenues was not driven by economic growth, but by other potentially short-term gains such as improved revenue collection, changes in the structure and growth of remuneration, etc. It would be unfortunate if this entirely fortuitous outcome could detract from risk management issues – which are likely to become even more pressing in the future.

Table 2: Comparison of budgeted and revised estimates for 1998/99

Some degree of variance around projections is to be expected, given the uncertainty inherent in the policy environment. But the robustness of a set of policy proposals depends on the extent they can be adhered to even in the presence of exogenous shocks in either the external or domestic environment. The ability to maintain fiscal discipline – if it is still deemed to be desirable – would depend very much on the extent that discretionary expenditures impart flexibility in response.

The bulk of non-interest expenditure occurs at provincial level. Yet the Division of Revenue Bill makes provision that, in the even of revenue projections not being realised, the national government will make good the shortfall (Which has a cogent economic rationale, given that provinces de facto have no taxation powers and limited own revenu sources). Provincial expenditure therefore in effect becomes non-discretionary within the fiscal year, placing the burden of adjustment on the national sphere only. Provinces are thus insulated from macroeconomic risk, but the risk is still borne by the national government which will most probably resort to capital markets to raise the necessary funding. This could compromise deficit targets. While provision has been made for a contingency reserve, this cannot be regarded as more than a palliative.

Another source of fiscal risk are contingent liabilities. The Reserve Bank’s forward cover losses of R13.5 billion during 1998 are a striking example of contingent liabilities being triggered by systemic disruption of world markets. The Department of Finance and State Expenditure compile a Statement of Liabilities and Financially related assets which lists off-balance sheet items like contingent liabilities (R143 235 million). But for management of continent liabilities it is imperative to have some estimates of their future costs. Cognizant of the need to limit contingent risk before a crisis is precipitated by external events, the Department has increased guarantee fees to 1.5% of a loan, generating fees of R4.3 million in 1998/99. But a cash accounting system (like that currently in use) would register the fees as income in that period, without recording the concomitant increase in fiscal exposure of the government.

There have been calls to improve the deficit measure on a cash basis by changing pension fund arrangements that convert an explicit liability to an implicit future liability. This however merely displaces the liability temporally, reducing the current (cash based) deficit, while increasing exposure to contingent liability. Having comprehensive estimates of future contingent costs would impose the discipline of a longer time horizon on decision-making.

The national department of finance has shown an enthusiasm for embracing best practice in debt management. The issue of broader fiscal exposure however is an area which still needs attention.