A randomised route to better government - Op Ed, The Canberra Times


The Canberra Times, 2 April 2019.

In the mid-1990s, researchers embarked on a massive clinical trial. Over 16,000 post-menopausal women volunteered to be randomly assigned to a treatment group, receiving hormone replacement therapy, or to a control group. Many had expected that the study would back the common view – basedon observational studies – that taking Estrogen plus Progestin was good for the health of these women. 

Five years in, the trial’s data and safety monitoring board stopped the study. Not only did the research show no protective impact of hormone replacement therapy, but it those receiving the treatment were significantly more likely to have health problems. When the results were published in the Journal of the American Medical Association in 2002, they had a seismic impact. Millions of women worldwide were taking the supplements. Thousands of GPs had to tell their patients: ‘sorry, we were wrong’. Better evidence bruised a few egos, but it also saved lives. 

A century earlier, such adherence to rigour and data would have been unthinkable. As one commentator notes, before there was evidence-based medicine, we had thousands of years of ‘eminence-based medicine’. 
Today, the challenge is to make public policy more scientific. Not only have researchers shown the power of randomised trials, ubiquitous data has lowered the cost of careful public policy research. In Randomistas: How Radical Researchers Changed Our World, I described how governments from New Zealand to the Netherlands are adopting a ‘what works’ philosophy. Rather than muddle through with glib slogans and prejudiced beliefs, the strategy is to focus on building a better feedback loop. As the British Cabinet Office summed it up: ‘Test. Learn. Adapt.’

That’s why we have announced that a Shorten Labor Government would establish an Evaluator General in Treasury, tasked with carrying out high-quality evaluations across government. From taxation to social policy, good evaluation allows us to scale up the most effective programs and close down those that don’t work. The Evaluator General would receive funding of $5 million a year, allowing it to significantly boost the volume of randomised trials in the federal government.

Labor’s Evaluator General complements our Evidence Institute for Schools, modelled on the successful examples of the US What Works Clearinghouse and the UK Education Endowment Foundation. As Tanya Plibersek puts it, the Evidence Institute will ‘put an end to decades of ideological battles about school education’ and ‘take politics out of the classroom’.

In other areas, experts have called for more rigorous evaluations. When the Productivity Commission convened a roundtable discussion on evaluation of Indigenous programs, one expert pointed to the problem of ‘a litany of poor policies being recycled’. In respectful collaboration, better evaluation has the potential to improve the lives of Indigenous Australians. In its ‘Shifting the Dial’ report, the Productivity Commission made quality evaluation (and more randomised trials) one of its top governance reforms.

The randomised approach is already spreading rapidly in business. Netflix, Coles, United Airlines, Amazon and Google have built randomised trials into their corporate model. Intuit founder Scott Cook aims to create a company that’s ‘buzzing with experiments’. Whatever happens, Cook tells his staff, ‘you’re doing right because you’ve created evidence, which is better than anyone’s intuition’. If you used the internet today, it’s likely you were part of a randomised trial.

Randomised trials flourish where modesty meets numeracy. Unlike in the movies, most positive change happens steadily, not through miracles and magic bullets. From social reforms to economic change, our best systems have evolved gradually. What matters is that we encourage them to get smarter over time. Nicholas Gruen, a longtime champion of the idea of an Evaluator General, argues that rigorous testing of new ideas lets us ‘grow the intelligence for the system to successfully innovate’. 

In the early-twentieth century, Australia was known as ‘the social laboratory of the world’. We were among the first nations to extend the franchise to women, to set a minimum wage, and to provide support to the sick, elderly and unemployed. Since that era, we can point to only a few world-leading policies, such as HECS and the NDIS. 

To claim the mantle of the world’s most progressive reformers, Australian public policy must be driven by a rigorous evidence base. An Evaluator General will play a key role in ensuring this. By institutionalising evidence-based policymaking, we can spend less time in ideological battles, and more time finding practical solutions that work. We can be just as passionate about raising living standards, but a little more scientific and critical about particular programs. Good policymaking isn’t just abstract philosophy – we can learn from medical researchers too.

Andrew Leigh is the Shadow Assistant Treasurer.


Authorised by Noah Carroll ALP Canberra.

Be the first to comment

Please check your e-mail for a link to activate your account.

Stay in touch

Subscribe to our monthly newsletter


Cnr Gungahlin Pl and Efkarpidis Street, Gungahlin ACT 2912 | 02 6247 4396 | [email protected] | Authorised by A. Leigh MP, Australian Labor Party (ACT Branch), Canberra.