Check against delivery
The economic forecasts for the 2009-10 Budget were prepared in the face of unprecedented turbulence in the global economy. With increasing signs that the Australian economy was heading into recession, it became clear that the traditional approach to preparing the economic forecasts and projections for the Budget was in need of revision.
These opening remarks seek to explain the economic forecasts and projections in the Budget, as well as providing the rationale for the changes in methodology that have been introduced this year.
Since the Updated Economic and Fiscal Outlook (UEFO) was published in early February, we have seen a further deterioration in the outlook for the global economy. A significant development has been sizeable falls in world trade. The IMF is now forecasting that global trade volumes in 2009 will be 11 per cent lower than in 2008, a contraction that would be unprecedented since the Second World War.
In the period since UEFO, the release of the National Accounts showed a fall in Australian real GDP in the December quarter of ½ a per cent. This and a range of other economic indicators since February point to a more rapid deterioration in domestic economic conditions than previously expected.
As a result, the Budget is forecasting a recession in the Australian economy. Numerical macroeconomic forecasts have been provided in the Budget since the late 1960s, and this year’s Budget is the first to forecast a recession over that time.
By recession, we mean a sustained period of either weak growth, or falling real GDP, accompanied by a significant rise in the unemployment rate.
The forecasts for economic growth were revised down at Budget to zero in 2008-09 (previously 1 per cent) and a contraction of ½ a per cent in 2009-10 (previously growth of ¾ per cent).
Unemployment is now forecast to peak at 8½ per cent in 2010-11 with inflation moderating to 1½ per cent by June 2011.
All these forecasts have associated margins of error. Margins of error have been a feature of economic forecasts for as long as they have been made, and always will be.
Having said that, the economic forecasts in the Budget represent Treasury’s best professional judgement based on the available evidence and our professional understanding of how the economy tends to evolve.
In the Budget, we were presented with the challenge of forecasting not only the extent of the domestic recession, but also the nature of the recovery. This has required some changes in methodology, and it is to these that I now turn.
The first change was to include an additional forecast year in this Budget, for 2010-11. Under the old methodology, we would have projected economic growth in that year at its trend rate of 3 per cent. In our view, that would have been overly optimistic given the severity of the global crisis we are facing.
Our forecast for 2010-11 is for growth of 2¼ per cent, still below trend and implying a delayed recovery compared with past cycles. Were these forecasts to be realised, the Australian economy would experience a slower recovery than those from either the early 1980s or early 1990s recessions (Chart 1). Indeed, it would represent the longest period of sub-trend growth since the Second World War.
Chart 1: Real GDP growth in recessions and recoveries
Note: Year 0 is 1982-83, 1990-91 and 2008-09.
Source: ABS Catalogue Number 5206.0 and Treasury
At the end of such a weak period of growth, there will be substantial unused capacity in the economy. When the economy eventually recovers, the economy can be expected to grow faster than average as unused resources are brought back into production. This has been the experience in the past although, as explained earlier, we are assuming that the process takes longer than in the previous recoveries.
This brings me to the second change in our methodology.
For the remaining two projection years in the forward estimates – 2011-12 and 2012-13 – rather than assuming trend growth, as in earlier Budgets, we have assumed a period of above trend growth of 4½ per cent. We built these estimates up as follows.
Trend growth in the economy is estimated at 3 per cent. This comprises 1½ per cent growth in the labour force, based on population trends, and 1½ per cent growth in labour productivity – the average over the past 30 years. This produces our standard projection assumption of 3 per cent GDP growth.
But when the economy begins to recover, employment grows faster than trend, as unemployed workers enter or re-enter employment. The experience of the previous two recessions was that the unemployment rate dropped by two percentage points in the first two years of recovery. We have assumed a similar pattern of recovery in the Budget projections. This brings employment growth to 2½ per cent.
Past cycles also reveal a recovery in labour productivity, as unused physical capital in the economy is brought back into production. And the lead up to the current crisis saw a substantial build up in the economy’s productive capacity, particularly in the mining sector. As a result, we have assumed that productivity grows at 2 per cent over these two years, around ½ a per cent above its long run average, again similar to the early years of recovery from the 1980s and 1990s recessions.
Productivity of 2 per cent plus employment growth of 2½ per cent implies growth in real GDP of 4½ per cent.
I should also note that in the first three years of recovery from the recession – that is the years from 2010-11 to 2012-13 – average growth is 3¾ per cent. This is below the average growth in the first three years of the 1980s recovery, which was 4.8 per cent, and the 1990s recovery, which was 4 per cent.
Consistent with recent practice, the Budget also contains medium-term projections. These provide the basis for estimating the longer term path of the fiscal aggregates, and span the gap between the forward estimates and the 40 year projections contained in the intergenerational reports.
The medium-term projections run to 2019-20. In the first four years, 2013-14 to 2016-17, real GDP continues to grow above trend as unemployment falls back to the NAIRU – assumed to be 5 per cent, as it was in both the first and second intergenerational reports. Also supporting growth is a recovery in the participation rate, reflecting encouraged workers returning to the workforce as employment prospects improve. These factors support growth of a little under 4 per cent in the years 2013-14 to 2016-17.
Real growth slows in the period 2017-18 to 2019-20 as population ageing causes a decline in the participation rate, and thus a reduction in trend economic growth. The effect of an ageing population is factored into the projections throughout the entire projection period. But it is not until 2017-18, when unused economic capacity is assumed to have been exhausted, that these effects come to the fore.
Undoubtedly, there will be some variation in growth rates, with some years coming in above our projections and some below. As has always been the case, the projections reflect our view of how fast the economy is likely to grow on average.
We would be the first to agree that no one can predict with any precision GDP growth up to 10 years into the future. But we can make sensible assumptions about the likely average path of the economy through time based on our understanding of long run economic trends and the behaviour of the economy in past cycles.
Average real GDP growth over the 7
years of recovery, 2010-11 to 2016-17, is 3.9 per cent, slower than the average of 4.2 per cent over the comparable period for the 1980s recovery, and 4.3 per cent for the 1990s recovery. Average growth over the 12 years of forecasts and projections in the Budget is a little under 3 per cent – that is, a little under its long run average.
It is also noteworthy that our approach is similar to that used in other jurisdictions. For example, the US Congressional Budget Office uses potential output to set the level of real GDP in its medium-term (10-year) projections. In doing so, the CBO assumes that any gap between actual GDP and potential GDP that remains at the end of the short-term (two-year) forecast will close during the following eight years.
Thus, the methodology adopted by the US Congressional Budget Office for its medium-term projections is almost identical to the methodology adopted for the medium-term projections in the Budget.
I am sure that Committee members will have many questions for us and we are happy to take them now.