Fiscal Stimulus, Aussie style

Posted by Jason Harris[1]

The Australian economy escaped from the global financial crisis relatively unscathed. Growth continued almost unabated, the unemployment rate topped out at only 5.8 per cent, and there was no significant loss of productive capacity. In fact, even while the crisis continues, Australia’s major economic challenges centre around managing an economy at full-employment with emerging inflationary pressures and dealing with the structural changes associated with a resurgent terms of trade, issues that many other economies would look upon with envy.

One of the key factors behind this is the aggressiveness of the fiscal stimulus adopted in response to the crisis. This post will focus on the make-up and design of the fiscal stimulus, which was very much in alignment with the IMF’s adage of timely, targeted, and temporary. Going into the crisis, Australia could have been considered a prime candidate for collateral damage.  With a large current account deficit, and heavy reliance on global capital markets for wholesale bank financing, Australia was exposed to the freezing of financial markets. Further, as the crisis developed, there was a sharp slowdown amongst Australia’s major trading partners, and a large decline in the hitherto booming terms of trade.  But Australia avoided the sharp contractions experienced by most other advanced economies.  With the exception of one quarter of slightly negative growth, GDP continued to grow throughout 2008 and 2009.

The overriding consideration in the fiscal policy reaction was to get economic activity up quickly, with money going to the sectors with the highest multipliers, without increasing expenditure permanently. Different types of stimulus have different impacts on the economy. IMF analysis suggests that direct public investment has the largest impact on aggregate demand, with a fiscal multiplier of around 1.4.  Transfers to low income (liquidity constrained) households have a smaller effect, with a fiscal multiplier of around 0.5. Of course, the second dimension that needs to be accounted for is the speed at which the stimulus can be injected into the economy. Transfers can generally happen much faster.

In Australia cash handouts worth approximately 2 per cent of GDP were sent out quickly to low and middle income households, who have relatively high marginal propensities to consume. This happened first in December 2008, through the existing transfer systems, then again in March 2009 through the existing tax system.

These payments had a noticeable impact on consumption, and GDP began to grow again in the March quarter, after a small contraction in the December quarter. The printing of this positive outcome also had a major impact on confidence, with measures of both consumer and business confidence soaring immediately after the National Accounts release. To the general public, it appeared that the combined policy effort was succeeding.

The second phase of the stimulus focused on generating activity in areas that were expected to suffer from a slowdown in activity due to the problems in the financial sector, primarily in the construction sector. This was done largely through “shovel ready” capital expenditures, worth around 2½ per cent of GDP, which could be implemented relatively quickly. This included a large program providing every school in Australia with funds to construct new facilities; a first home owner’s boost, which provided significant extra cash to first home buyers to help stimulate residential construction; direct investments in the construction of social housing; the introduction of a housing insulation program, and an investment tax credit scheme for small and medium sized businesses.

Importantly, these measures were designed to come into operation quickly. The school program, which formed the largest share of this phase, required schools to complete their projects by March 2011. In order to achieve this ambitious time frame, the program provided building design templates, which were to be followed unless non-conformance could be justified, and prioritization of specific projects, such as libraries, halls, classrooms, and refurbishments. Procurement practices for these projects were to follow existing State procurement processes. These measures had noticeable impacts on the construction activity figures in the June and September quarters, following the phase out of the cash payments. Construction activity picked up, investment activity was boosted and first home buyers entered the market in droves.

The third phase focused on longer-term, large public infrastructure investments, such as rail, roads, and ports, which would have the largest fiscal multipliers, but also the longest lead times. It was anticipated that the first two phases of the stimulus would bear the initial load in terms of providing stimulus, while the investments were scoped out and designed. This phase is still ramping up and is worth around 2 per cent of GDP.

Size and shape of the stimulus

The size and timing of the stimulus was very aggressive. The IMF placed Australia’s stimulus—over three years worth around 6½ per cent of GDP—as one of the largest discretionary packages in the G20. A large factor behind the size and speed of this response was the sound fiscal position in which Australia entered the crisis.  A terms of trade boom, combined with a robust and credible medium-term fiscal strategy (see an earlier post on Australia’s fiscal strategy) had delivered consistent budget surpluses and a negative net debt position. This provided the government with the ability to respond to the crisis swiftly, without the concerns about risking fiscal sustainability that other countries may have had to consider when calibrating their stimulus packages.

The combination of the three phases, and the relative size and timing of each phase resulted in a stimulus package that provided maximum impact early, before gradually easing off, to make room for a transition back to private sector growth (Chart 2 below). 

This profile ensured that the stimulus contributed significantly to growth early on, when it was most needed.[2] Using the Global Integrated Monetary and Fiscal Model, the IMF estimated the base case was for the stimulus to contribute 2 percentage points to growth in 2009, before leveling off in 2010, with no net contribution, and begin detracting from growth in 2011, as appropriate in an economy that is growing at above-trend rates (Chart 3 below).

Importantly, the overwhelming majority of the stimulus is temporary, with little if any longstanding impact on the primary balance.  Even though Australia is in a relatively sound fiscal position, this was still a key consideration in the design of the stimulus. Returning the budget to surplus quickly would put Australia in a stronger position to deal with future shocks, and maintaining low Commonwealth debt levels will help contain debt servicing costs.

Lessons learned

So far, most of the measures have worked as anticipated. The cash handouts were an effective way of stimulating consumption activity. The various tax credits and subsidies provided to households and businesses were effective in generating activity in areas of the private sector which may have otherwise sat back to wait and see how the crisis would play out. The direct government spending was also effective, in adding to aggregate demand directly, and in targeting the industries where the downturn was expected to hit the hardest. The outcome of the public infrastructure component is yet to be felt in full, as it is really only in the early stages of ramping up.  While there were some delays and costing issues with the direct spending—particularly with education investments—these were relatively minor.

By outlining a clear and credible deficit exit strategy at the same time as fiscal stimulus was announced, backed up by a sound medium-term budget framework that demonstrated exactly how the consolidation was to be achieved, the path back to surplus and continued fiscal sustainability was understood and accepted by outside observers.  The budget is now projected to return to surplus in 2012-13, and net debt is expected to peak at 6.4 per cent of GDP.

In terms of public financial management, there are a couple of lessons to be taken away from this episode:

  1. Using existing systems is the key. The cash handouts were particularly successful, and a large part of this was because they were sent out through existing tax and transfer systems, which ensured that the cash got out quickly and efficiently.
  2. Consider the design specifics of the policies early. When pushing large amounts of public monies out the door quickly, particularly on new programs, there is likely to be some waste and inefficiencies.  Steps should be taken to mitigate this at the design stage, although this needs to be balanced against the overall objective of the stimulus, which is to boost aggregate demand quickly. The Australian experience provides a couple of examples of issues with program design, largely through the capital expenditure on school infrastructure and household insulation, both of which had problems relating to value for money, and in the latter case weaknesses in industry regulations.  Both of these programs required modification later on, and in the case of housing insulation, the program was eventually cancelled due to governance, quality, and safety issues. [3] One of the recommendations coming out of these programs is that targeted compliance and audit programs should accompany new programs, particularly those that do not have existing policy infrastructure in place.
  3. Including some scalability into the stimulus package at the design stage might make life easier later. The fallout from the crisis in Australia was much milder than initially predicted. Including scalability into package at the initial design stages—both up and down depending on how circumstances play out—would enable some adjustment as conditions change. This could include enhancements to automatic stabilizers in the stimulus package, to increase the magnitude of their operation. As it is, the stimulus measures were relatively rigid in terms of size, and it would be difficult to reverse them today if desired, largely due to contractual commitments. 
Download Charts

[1] Jason Harris is a Technical Assistance Advisor in the Fiscal Affairs Department at the IMF. He previously worked in the Australian Treasury preparing the Commonwealth Budget, and as an economic and fiscal adviser to the Australian Prime Minister. He also spent two years on secondment to the Papua New Guinea Treasury, helping to prepare the PNG Budget.

[2] The contribution to growth can be thought of as the change in the size of stimulus, adjusted for lags—a growing stimulus will have a positive contribution to growth, while a shrinking stimulus will detract from growth.

[3] Reports by the Australian National Audit Office, and specific reviews into the school investment and housing insulation programs by independent taskforces provide more specific advice on ways to design such programs.

Note: The posts on the IMF PFM Blog should not be reported as representing the views of the IMF. The views expressed are those of the authors and do not necessarily represent those of the IMF or IMF policy.