The ink had hardly dried on the 2011 MediumTerm Budget Policy Statement (MTBPS) when Moodys announced that it was revising its outlook for South Africas credit rating from stable to negative. Two months later, about a month before the release of the 2012 national budget, Moodys was followed by Fitch with a similar decision. Both rating agencies had therefore come to the conclusion that the credibility of South Africas medium-term budget was suspect, based on the same information.

In its response to the respective downgrades, the National Treasury attributed the rating agencies decisions to the deteriorating sovereign debt position globally, in particular in Europe, and the resultant spate of sovereign downgrades, rather than to factors specific to South Africa. It also referred to the medium-term budget as a definitive outline of the future fiscal intentions of the South African government, intimating that it should be accepted as cast in stone. Unfortunately this merely takes us into a circular argument.

Credibility is of course not only a matter of the credit rating agencies stance, but more importantly that of the capital market. With regard to governments borrowing costs, bond yields are not displaying the persistent upward trend that would be indicative of a credibility crisis.

Credibility is not something that can be claimed at will; it has to be earned by behaving consistently in the promised manner over an extensive period of time. Put differently, it is ones track record that counts. A quick examination of South Africas recent fiscal history should therefore give a clear indication of whether the claim of credibility is justified or not.

Annexure 1 sets out how government expenditure, revenue and the budget balance progressed with respect to the budget period from 2006 to 2009 (viz. during the Mbeki/Manuel era), from the first estimates in the applicable MTBPS up to the final outcome three years later for the fiscal years in question.

The most prominent feature of how the budget progressed during this period is the consistently large underestimation of government revenue (in excess of 25% from the first estimate for each year). The result was that the concomitant excess in government expenditure could be easily absorbed and the budget balance turned out better than initially anticipated.

Annexure 2 repeats the same exercise for the subsequent three fiscal years up to 2011/12 (viz. during the Zuma/Gordhan era).

The result is the opposite of that for the preceding three fiscal years up to 2008/09 revenue is close to the initial target, expenditure is much higher, and thus also the deficit.

Whereas government revenue exceeded the initially expected amount by 28% on average in the three fiscal years up to 2008/09, government expenditure was on average 22% higher than first estimated in the most recent three fiscal years. If accuracy was the determining factor, the credibility of the pre-2008/09 budget period would therefore have been more problematic than the current budget position (which it was not)!

Government expenditure and revenue: average % deviation from first estimate

But clearly excess expenditure is not regarded with the same benevolence as excess revenue. As long as excess revenue is absorbed into permanently higher spending (which was only partially the case in 2006-2009, as the graph clearly shows) it does not create any legacy problems.

However, excess expenditure is a different matter due to the difficulty of reducing government spending, especially recurrent current expenditure such as wages and salaries. Furthermore, excess revenue largely as a result of improved tax compliance is a technical matter, while excess expenditure is a political issue that needs to be dealt with as part of the political process.

So how credible is the current medium-term budget then?

The 2011 MTBPS promises a nominal rate of increase of 8,4% per annum in government expenditure for the three fiscal years up to 2014/15, resulting in expenditure as a percentage of GDP declining from 32,9% in 2011/12 to 31% in 2014/15. On the other hand, revenue is forecast to increase by 11% per annum during this period, causing a decline in the budget deficit from 5,5% of GDP in 2011/12 to 3,3% in 2014/15 and the stabilising of the gross loan debt to GDP ratio at approximately 42%.

The credibility of the revenue projections depends primarily on how realistic the underlying economic projections are. GDP is forecast to grow at 4% real per annum, and here the first year is already likely to disappoint with growth of around 2,5%. To achieve the projected values for the budget deficit and government debt would therefore require that government expenditure increase at an even slower rate. Is this a realistic expectation when the assumption that the wage and salary bill, which accounts for 40% of expenditure, will decrease in real terms is already suspect?

Government expenditure is furthermore under upwards pressure as a result of the increasing role government plans to play in the economy for example, the funding of the proposed special economic zones has not been spelled out yet.

In its response to the rating agencies, the National Treasury furthermore boldly states that the fiscal guidelines proposed in the 2011 Budget Review form the foundation of the fiscal stance. These guidelines are based on three principles, viz. counter-cyclicality, long-term debt sustainability, and intergenerational equity.

In applying these principles, the National Treasury proposes the adoption of an annual target for the structural budget balance, explicit costing of existing and new programmes, and setting a timeline for bringing the budget balance back on target.

The question is whether Government has bought into the proposed guidelines. A bold statement of support for the National Treasurys proposals and evidence that they are being adhered to are required. Fiscal credibility does not depend on the National Treasury alone, and the doubts that have been expressed in this regard are an indictment of the lack of visible general support for the sanctity of the budget process.