Forecast Sensitivities

Construction forecasts are always subject to substantial uncertainty, being sensitive to a wide range of underlying assumptions. For example, construction forecasts are sensitive to the underlying assumptions made about employment, which has a wide variety of flow-on effects in the economy.

Underlying these construction forecasts are our whole-of-economy forecasts. These forecasts were prepared prior to the release of ABS national accounts information for the September quarter 2009. Over the past year, the Australian economy has surprised economic forecasters by managing to avoid recession. One of the reasons for this has been the Australian labour market, which has been very resilient in the face of tough business conditions. The pattern of unemployment throughout this economic cycle has been different to what has been observed in previous downturns. For example, in the 1990 recession, the unemployment rate rose far more quickly before peaking at over 10%. Following the GFC, the unemployment rate is yet to reach 6 %, although it is expected to continue to rise in the short term.

Thus, an alternative scenario has been prepared to estimate the effect of unemployment peaking at around 6%, rather than at around 7% as currently forecast by our macroeconomic model. This analysis has been conducted by assuming higher employment levels, mirroring the observed pattern published in Labour Force data to date by the ABS. The impact that improved labour market performance has on construction activity for 2009/10 is shown below in table 1.

Table 1
Change in Construction Activity, 2009/10 (Real terms)

	Increase in work done
Residential building	+0.1%
Non-residential building	+0.0%
Engineering construction	+0.0%

Note: increases shown as deviations from the baseline scenario.

The impacts in 2009-10 are relatively minor, since the simulation involves very little change to current employment and unemployment levels for 2009/10. Looking further out, the simulation shows a greater impact over construction activity, with the main impact being shown in the residential building. This is because residential building, being driven by owner occupiers, is the type of construction which will be most sensitive to changes in the labour force. The 2010/11 impacts on construction activity are shown in Table 2.

Table 2
Change in Construction Activity, 2010-11 (Real terms)

	Increase in work done
Residential building	+0.7%
Non-residential building	+0.2%
Engineering construction	+0.2%

Note: alternative scenarios are shown as deviations from the baseline scenario.

Overall, these effects are relatively minor – the construction forecasts will not vary widely if there are differences between the modelled macroeconomic scenario and the actual labour force results as published by the ABS.