It's the economy, stupid!

Australia has done it better

The economic management record of Kevin Rudd as a prime minister is constantly under attack in this campaign. To put it bluntly, I fail to see why. As I already discussed in other articles and as I will discuss in more detail below, data show that Australia has done better than most other countries since 2008.

2008-2012/13: very hard not to see the good economic performance of Australia

Data sourced from the World Economic Outlook Database of the International Monetary Fund and the United Nations Development Programme indicate that Australia today has the tenth highest gross domestic product per-capita worldwide, the tenth lowest rate of unemployment among the 35 most advanced economies worldwide, the sixth lowest net debt to GDP ratio among the 26 most advanced economies for which data are available, and the second highest human development index.

If we look at the past few years, then we see that Australia is one of only three countries out of all 35 most advanced economies that maintained a positive rate of GDP growth in both 2008 and 2009. In fact, GDP growth in Australia between 2008 and 2012 averaged around 2.5% a year. This is the sixth highest growth rate in the group of advanced economies and the second highest growth rate outside East Asia. Over the same period 2008-2012, employment in Australia has grown by more than 8%. This is the fifth largest increase among all advanced economies.

Compare this performance to what has happened in the US and Europe and the words of Nobel Prize Joseph Stiglitz will immediately make perfect sense: “Australia, you don’t know how good you’ve got it”.There is no way that you can look at these numbers and reasonably argue that Australia did not do well in the last five years, given the international crisis.

Of course, one can always claim that Kevin Rudd and Julia Gillard have no merits and that this strong performance is entirely attributable to factors other than domesti economic policy. But such a claim is not substantiated by facts and data. In fact, I want to give here an account of a little experiment which in my view indicates that Labour governments had a lot to do with the good economic performance of Australia in 2008-2012/13.

Australia vs. its “synthetic-self”

The problem in assessing the economic management record of a government is the lack of a counter-factual. In other words, we cannot observe how the Australian economy would have evolved since 2008 without Rudd and Gillard as prime ministers.

However, we can construct a “synthetic” Australia and use it as a benchmark to assess “real” Australia. This synthetic Australia should bear as much resemble as possible with real Australia before 2008. Then, any discrepancy in performance between synthetic Australia and real Australia in the period 2008-2012/13 can be largely attributed to the governments of real Australia.

In a nutshell, synthetic Australia is just a weighted average of appropriately selected countries whose economic dynamics are similar to those of real Australia. The weights are chosen to ensure that the difference between synthetic and real Australia in the period 1980-2007 is minimised.

More specifically, the construction of the synthetic benchmark involves three steps. First, given a reference economic indicator, say GDP growth, I collect annual data from the World Economic Outlook of the IMF for each of the 35 most advanced economies in the world. Then, I compute the squared difference between growth in each country and growth in Australia by year. I sum up all the annual squared differences and identify the ten countries with the lowest sum of squared differences. Second, I run a regression of growth in Australia on growth in the ten countries identified in the first step. I use a likelihood ratio test to choose which countries are redundant and which countries instead add explanatory power to the regression. Third, I select the combination of countries that maximises the explanatory power of the regression (as measured by the adjusted R2). The estimated coefficients of this regression are the weights used to construct synthetic Australia.

I have done this exercise for two key economic indicators: the rate of GDP growth and the rate of employment growth. Figure 1 plots GDP growth for real Australia (blue line) and synthetic Australia (red line) between 1980 and 2013. Figure 2 instead reports employment growth for real Australia (blue line) and synthetic Australia (red line) over the same period of time.

The synthetic proxies seem to do provide a good approximation of real Australia between 1980 and 2007. Over the period 1980-2007, average growth in real Australia and synthetic Australia was identical (3.38%) and the correlation between the two variables was 0.84. Over the period 2002-2007, average growth in real Australia was 3.611% and average growth in synthetic Australia was 3.606%. But in the period 2008-2013, real Australia has outperformed synthetic Australia: average growth has been 2.613% in real Australia compared to 2.153% in synthetic Australia. That is, real Australia has grown 0.45 points faster than its benchmark. In particular, if we look at the first two years of the first Rudd government (2008-2009, which were also the years of the deep worldwide recession caused by the Global Financial Crisis), the cumulative difference in growth between real Australia and synthetic Australia was 2.1 percentage points.

Turning to employment growth, the average over the period 1980-2007 was 1.955% in both real and synthetic Australia and the correlation between the two series was 0.83. Between 2002 and 2007, employment grew by 2.55% in real Australia and by 2.47% in synthetic Australia. The difference however becomes much larger for the period 2008-2013: employment growth in real Australia has been 1.693% against 1.053% in synthetic Australia. The cumulative difference between real and synthetic Australia in the rate of job creation in 2008-2009 was almost 2.25 percentage points.

The resemblance between synthetic Australia and real Australia in the period 1980-2007 indicates that synthetic Australia is indeed a good benchmark; that is, synthetic Australia is a valid counter-factual to assess the performance of actual Australia. The data then suggest that in 2008-2013 real Australia has outperformed synthetic Australia. Because of the way in which it is constructed, synthetic Australia does not incorporate the policy of the governments of real Australia in the period 2008-2013. So, to some considerable extent, the superior performance of real Australia must be due to these government policies. The difference between real Australia and its synthetic-self is particularly striking for the period 2008-2009. While the world was being hit by the Global Financial Crisis, Australia managed to stay afloat, avoided a recession, and still created more jobs than what most other similar economies did.

What was the key contribution of the Rudd (and Gillard) government? My opinion is that the great merit of the government was to adopt a strongly counter-cyclical fiscal policy stance. Certainly, to some extent Australia did benefit from a stable and strong demand of resources from Chinese manufacturing, but the fiscal expansion that started in 2008 was the right move at the right time.

In conclusion, I can only go back to where I started: there are simply no empirical foundations to the statement that Rudd (and Gillard) mismanaged the economy. In fact, the data tell exactly the opposite story.