Perth Voice Interactive

AT the end of CHOOK reporter David Bell’s psychology honours’ year, he was back in the midst of reporting on councils grappling with ways to incentivise tree retention or discourage the use of natural gas, and got to thinking about a few of his favourite (often weird, sometimes terrible) policies proposed by behavioural sciences to fix the world. 

Proposal 1

Remove all driver-side airbags and replace them with a large metal spike protruding from the steering wheel, pointed at the driver’s chest.

Economist Gordon Tullock proposed this as a thought experiment, suggesting that people would drive more cautiously if they knew any collision could lead to their impalement. It’s plausible that this would save the lives of pedestrians and others outside the car who avoid being hit as everyone drivers more slowly and cautiously, but probably has a lot of other non-obvious downsides (apart from just the increased commute times and extremely risky emergency car rides to the hospital, no one wants to die because a dope on E-plates slammed into the back of their car).

Proposal 2

Make people sign their signatures at the top of a document, rather than the bottom, to encourage honesty.

The theory goes that by having to sign a document before filling it out, people will be more honest about what they write afterwards, having been reminded that they’ve agreed to state the truth. The researchers claimed they’d trialled this in a real-life experiment using car insurance forms, where drivers are asked to state how many miles they drove that year (with higher mileage increasing their insurance payments). Supposedly, customers who had to sign at the top before declaring their mileage reported they’d driven an average of 2,400 miles more than customers who signed at the bottom, and the researchers explained the higher rates as reflecting more honest reporting.

But it probably doesn’t work. 

Follow-up experiments, including by some government departments, didn’t show the same kind of results. The original data is widely suspected to have been fabricated by someone involved in the initial research, and the authors have now retracted the study. 

Proposal 3: 

Randomly distribute a small number of exploding cigarettes into each batch (the proportion of exploding to regular cigarettes should reflect the long-term risk dying from smoking). 

Okay, not really. But less-lethal versions of making consequences more immediate have been utilised and do have some merit. 

Distant outcomes are hard for us to fathom. The behaviourist BF Skinner saw that both humans and animals learned more quickly when their reward was given immediately. On the flipside, a distant (and uncertain) punishment isn’t much of a deterrent: “The more remote the predicted consequences, the less likely we are to follow advice,” he wrote in 1987; “why not arrange immediate consequences that will have the effect that remote consequences would have if they were acting now?”

While Skinner was a humanitarian and didn’t go so far as to suggest exploding cigarettes, the idea of making consequences more immediate underpins many functional policies. Not many people think of the long-term impacts of throwing a plastic bottle on the ground. But they’re more likely to think of the more immediate consequences of being fined for littering, or being shamed by others, and some might even be motivated by the 10-cent return. And if we can’t go ahead with the exploding cigarettes idea, making them more expensive makes for a reasonable immediate disincentive.

Proposal 4

Pay people a bounty to hunt problem pests. 

This is a great policy to increase the number of whatever you’re trying to get rid of. According to the original anecdote set during the reign of the British Raj in India, when the government tried to reduce the number of venomous cobras around Delhi by offering a bounty for every dead snake, the region’s cobra population instead boomed. According to the tale related by economist Horst Siebert, people began to breed cobras to claim the reward. The incentive scheme was scrapped, and some breeders allegedly released their cobras into the wild, and the plan to cut down cobra numbers instead saw the population boom. Whether the original anecdote was accurate, the power of these ‘perverse incentives’ would be demonstrated in many examples since, from government attempts to reduce feral pig numbers via a bounty in the United States, to recent reports of people 3D-printing guns in order to get payments from illegal firearm buyback schemes. 

If a local tree-protection policy results in pre-emptive loggings by landowners who don’t want to be saddled with a protected tree – as Vincent council was warned about last month (“Tree policy withdrawn after carnage warning,” Voice, November 23, 2024) – then we can add our local example to the Wikipedia page for perverse incentives.