When nudge comes to shove: making e-health opt-out was always a risky venture

By Stephen Easton

July 19, 2018

The sudden rush to opt out of e-health records this week amid a groundswell of delayed-action privacy concerns demonstrates that governments have to be very careful about how and when they use the most powerful nudge of all.

Default enrolment – opt-out rather than opt-in – is the “the Usain Bolt of nudges” in the words of law professor Cass Sunstein, who helped coin the term nudge to describe government interventions that preserve individual choice, in contrast to mandates and bans.

“They work really fast and they have a massive impact,” explained Sunstein at the recent Behavioural Exchange conference in Sydney (pictured above). This means they also must be used with caution.

He emphasised that automatic enrolment “has far more impact than other kinds of nudges” such as making choices simpler and easier for people, giving them carefully crafted pieces of information at the right moments, referring to social norms, offering non-monetary rewards or simply asking people to stop, think and actively make a choice or a pre-committment.

The golden rules for nudges that avoid a big public backlash warn they must be very transparent, they must avoid ulterior motives, and they must not conflict with people’s values or their own best interests.

The e-health backlash shows that a lot of people now disagree with the government’s view that the records are in their interests. Worries about privacy, information security and the medical utility of the revamped records have grown considerably since four years ago, when an external review panel told then-Health Minister Peter Dutton there was strong public support for the e-health system, and it was clear that it would change to an opt-out model.

Sunstein had come bearing opinion polls, showing strong public support for nudges in a range of countries where they are being pursued by a small but dedicated cadre of public servants.

These statistics showed “people do not object to default rules” per se, he said, unless they are “inconsistent with people’s values or interests” or seem to have ulterior motives like benefiting one group over others in society or helping keep the government of the day in power.

The leading light of the nudge field accepts a key criticism of default options, however: they can “work even in circumstances when they are not in people’s interests, because they can exploit inertia”.

Defaults are powerful game-changers in situations where few people ever get around to opting in, as they did with Labor’s Personally Controlled E-health Record, before it morphed into the My Health Record under the Coalition.  Normally once the change is made, very few ever get around to opting out either.

“That’s a fair point, and it suggests a reason to be extremely careful with the design and transparency of default rules, so that it doesn’t end up that people are stuck, just because they’re devoting their scarce attention to other issues,” said the law professor.

Time is not on anyone’s side

Time is “the most precious commodity that members of our species have” in Sunstein’s view, and he feels this is an extremely important point which classical liberal thinkers like John Stuart Mill and Friedrich Hayek did not pay enough attention to when arguing for a society powered by liberty and freedom of choice.

If he could go back in time and re-write the book Nudge, co-written by his colleague Richard Thaler, Sunstein said they should have emphasised more that the main goal is to help people to make an informed choice in situations where this is difficult due to complexity or the time and effort it would take to work through the options.

He calls this the “navigability” problem. Sometimes it takes too much time and energy or specialist knowledge to exercise informed choice, and being given choice is a burden, like if a taxi driver in a giant metropolis asked a tourist which way they’d like to go to the airport.

Nudgers, according to their guru, need to pay more attention to this and the fact that “people who face economic or other forms of deprivation” usually find navigating life’s many choices harder than others. “And a spotlight placed on the problem of navigability, I think, provides an organising framework that can both help identify good policies and also help us understand the ethical challenges.”

Other nudges usually only impart a relatively small boost to uptake of an optional government offering like the My Health Record. Changing a default option is way out ahead, in a class of its own, hence the Usain Bolt analogy. Recent changes to superannuation rules put a stop to funds defaulting young people into life insurance policies that eat up all their fees, for example.

Defaults consistently flip the figures from most people staying out of whatever system, to the vast majority staying in, or vice-versa.

But this only works when policymakers are sure that nearly everyone would, on reflection, agree the default is beneficial and not harmful. Ideally, they would have opted in if they had found the time, or if someone explained it all to them, and made it easier to make an informed choice.

Being automatically enrolled to vote is a good example where the nudge enhances the ability of citizens to exercise an important right. One good example Sunstein mentioned is a free school meals program in the United States for children from lower socioeconomic backgrounds; a lot of eligible parents weren’t taking it up until they were defaulted into it. Having money put into a retirement savings plan by default has also been generally successful and relatively uncontroversial in many cases overseas, and isn’t a coercive rule like Australia’s superannuation guarantee.

The sudden rush for the exit in just the first few days of the My Health Record opt-out period – and a clear sense of indignation among many critics at the presumption that they would want to have an enduring national e-health record automatically created, with some problematic terms and conditions – suggests a miscalculation.

L-R: Tara Oliver, Robert Slonim, Cass Sunstein and Martin Parkinson. All images: Department of the Prime Minister and Cabinet.

How to maintain public support for nudges

Sunstein addressed a list of common objections to nudging in his speech, and said some of these helped develop his new code of ethics, which he called a bill of rights for nudging.

“The really new data is from Australia and it turns out that among the nudges that are under serious consideration, or actually have been adopted by countries that have been interested in behavioural insights, … the approval rates are really high,” he said.

Approval ratings for nudging that is done by the book are typically around 80% and above, and are not correlated with political party affiliation, but they drop sharply if people think they are being manipulated or deceived. The five principles of ethical nudging are based around these surveys.

The data suggest there should be overwhelming public support for nudges that stick to these principles, especially in comparison to mandates and bans that force people to behave how the government decides, because of the main difference: nudges preserve choice. The e-health opt-out process only offers choice for a limited period.

The predictable (and predicted) backlash also demonstrates another of Sunstein’s tips for would-be nudgers: bias towards the present is one of the strongest of all. “Next year is a foreign country,” he said, in contrast, “and people aren’t sure whether they’re going to visit.”

Sunstein felt he should list some of the strongest and most pervasive biases since academics propose new ones all the time. “It might be bias-overperception bias,” he quipped.

The switch to opt-out has clearly been on the cards for four years, but only now a lot of people are seriously thinking about it. It may be fair to say the average punter could have been better informed by government, but it is strange to see newspaper columnist Janet Albrechtsen admit she only realised last Thursday and still blame everyone but herself for this.

Critics with big media platforms have had four years to oppose making e-health records the default option, as privacy advocates have done all along — people like Justin Warren of Electronic Frontiers Australia who revealed there would only be a three-month window to opt out in February this year.

Present bias combined with another of Sunstein’s top three — the fact that “losses loom larger than gains” — might also help explain the backlash.

Perceived risks like losing some privacy or freedom, or even just losing the ability to delete the record forever after October 15, are probably amplified in the minds of citizens now that the brief opt-out window has opened.

The claimed benefits probably seem less important than avoiding these perceived risks, as they are in the future, on the other hand. And some of those benefits, as with other government data mining projects, go generally to the whole community, making it unlikely they could trump perceived risks to the individual.

If it is OK, why do people think it’s not OK?

Sunstein has never advocated tricking people by exploiting their brain’s autopilot mode; he says genuine nudges always allow people every chance to make an informed, deliberative choice.

“Nudges should be transparent and not hidden. They should be subject to public scrutiny and review.”

One common warning he accepts is that nudges have the potential to offend human dignity. People don’t mind being guided gently but they hate being pushed or dragged.

“Forcing people to choose when they don’t want to — that is offensive to their dignity because it requires them to devote their attention to something to which, on reflection, they don’t want to devote their attention.”

He also agrees that some so-called nudges “have illicit ends” like support for special interest groups or the political party in government, and they could also infringe on individual rights if applied badly.

But if governments violate rights, use policy to partisan ends, or force citizens to do things against their will, this is intrinsically unwelcome regardless of whether it is done through a twisted interpretation of nudge or some other means.

The star attraction at the conference, hosted this year by the Behavioural Economics Team of the Australian Government, made another interesting observation with relevance to the My Health Record backlash as well, in response to a question from an Australian Taxation Office employee.

How could nudges help change “community perceptions” that the tax system is unfair and the biggest companies and highest salary-earners don’t pay as much as they should, asked the ATO staffer.

Sunstein said the first step should be to work out why people don’t think the tax system is fair already, then “try to address it directly and honestly”– because he believes it is best to remove any obstacles to a desired outcome than to try to force it.

“And so I wonder, what’s the obstacle to people thinking it’s OK, if it is OK?”

In this case, the Australian Digital Health Agency might try to “directly and honestly” address the reasons why a significant slice of public opinion is at odds with what the government’s expert advisers honestly believe, rather than bunkering down and returning fire in a propaganda war.

Cass Sunstein’s Bill of Rights for Nudging
Why Martin Parkinson wants a second generation of behavioural economics
Reflections from the founder of Australia’s national nudge unit
New journal delves into behavioural approaches for public administration
The OECD’s six principles for spreading behavioural insights across government

About the author
Inline Feedbacks
View all comments

The essential resource for effective
public sector professionals