Menu

Q E A S

Queensland Economic Advocacy Solutions

header photo

A History of Federal Budget Forecast Accuracy

This week has seen economic forecasts underpinning the Federal Budget questioned over their degree of optimism.

It is crucially important that the Federal Budget is able to accurately forecast economic growth.  Much is riding on their accuracy in particular the revenue raising requirement and the resultant operating surplus or deficit.

We often hear the leaking of information by the Government of the day in the weeks before a Budget along the lines of 'we have experienced serious write downs in revenue'.  Often true but it is ultimately based on the accuracy of the original forecast.

With the significant improvement in the Federal Budget occurring in the outer years (a small surplus of $7.4 billion in 2020-21), and receipts to government to achieve this predominantly determined by health of the economy, a lot is on the line.

In essence forecasting accuracy will determine the veracity of the Federal Government’s claim of a surplus occurring in 2020-21.

Accordingly I have analysed over the last decade the Federal Budget forecast for economic growth against the actual result as measured by the Australian Bureau of Statistics. The Budget is held on the second Tuesday in May each year and issues forecasts for the four following financial years.  The most important forecast is for Gross Domestic Product (GDP) the measure of how much the economy is growing.

The results for GDP are below:

  • Accuracy for the immediate following financial year*: On average the Budget forecast over the ten year period was out by 0.4 percentage points (median 0.4) with a maximum variation of 0.95 percentage points. The average forecasted growth rate over this period was 2.8 per cent.  In seven of the ten years the Budget forecast overstated the economic growth that actually occurred as measured by the ABS. 
  • Accuracy for 2nd Year: On average the Budget forecast over the ten year period was out by 0.55 percentage points (median 0.5) with a maximum variation of 1.2 percentage points. The average forecasted growth rate over this period was 3.2 per cent.   In seven of the ten years the Budget forecast overstated the economic growth that actually occurred as measured by the ABS.
  • Accuracy for 3rd Year: On average the Budget forecast over the ten year period was out by 0.6 percentage points (median 0.5) with a maximum variation of 1.0 percentage points.  The average forecasted growth rate over this period was 3.3 per cent.  In eight of the ten years the Budget forecast overstated the economic growth that actually occurred as measured by the ABS.
  • Accuracy for 4th Year: On average the Budget forecast over the ten year period was out by 0.75 percentage points (median 0.6) with a maximum variation of 1.9 percentage points.  The average forecasted growth rate over this period was 3.3 per cent.  In seven of the ten years the Budget forecast overstated the economic growth that actually occurred as measured by the ABS.
  • Average annual growth for GDP as measured by the ABS over the decade was 2.8 per cent.

The past serves as a guide for the future.  From the above analysis we can conclude that Budget forecasts on balance overstate the economic growth that actually occurs.  The accuracy of the Budget forecast is very good for the immediately following financial year (86%) but falls away as expected in the outer years (77%).

Notwithstanding the significant implications of any variation on the Budget outcome my assessment of the forecasts is that there is room for improvement but their accuracy is good given the difficulty in this ‘art’.

Having said that the probability of the very small $7.4 billion surplus in 2020-21 is indeed questionable based on the last decade and rolling that experience forward.  Any variation or inaccuracy to forecasted economic growth is almost always a negative to the budget outcome.

The problem is this, the Government is 100% certain to spend the money that is budgeted but we never have 100% certainty of the money being earned.  In short we spend money that we have not yet collected.

*I have removed 2009 -10 from the calculation as a result of the impact of the GFC.  The Budget forecast was that the Australian economy would contract by 0.5 per cent where as the actual result was a growth rate of 2.0 per cent! 

I will release the same analysis for the Queensland Budget next week.

Go Back

Comment