Monday 4 June 2018

iRegulate – Apple’s solution to screen addiction and personalised paternalism


This week at their annual developer’s conference Apple released details of new software features aimed at curbing screen addiction. Such features might be considered novel by some and a cynical attempt to avoid legal action by others (Bradshaw, 2018 in the Financial Times), but to some people – by which I mean me – this marks an important step in the development of personalised paternalism.

Personalised paternalism is an interesting idea that exists, at least in theory, as a response to the existing problems standard paternalism in regulatory frameworks. Though, standard paternalism is a bit of a misnomer, so let’s clarify some terminology first.

Without full details, it’s hard to assess how closely Apple’s new software will be borrowing from nudge theory – a branch of behavioural economics (which itself is a branch of economics with a good helping of psychology mixed in) that focuses on adjusting (or, as some might argue manipulating. See, for example, Rebonato, 2014; Arad and Rubinstein, 2015) how a proposition is presented to a decision maker in order to influence their choosing of a specific option (see Thaler and Sunstein’s Nudge (2008) for more). However, some reported elements, such as time-usage reports and mandatory downtime options, clearly have a grounding in the world of nudge.

The trouble with nudges, at least philosophically, is primarily the question of who should nudge? We all like to think we make wise decisions most of the time, but what if you were told you were often wrong in your decision making? And further, what if you were told there was someone who, at no expense to yourself, was willing to nudge you in the right direction? When phrased like this, it’s hard to really be aggrieved at your benevolent advisor. But think about it this way: what if Apple didn’t let you set the downtime options, but instead decided that they understood your usage better than you and chose for you how long you’re access to a specific app was restricted?

This is part of the problem with paternalism (and it is definitely paternalism. Nudges, typically, do not objectively limit freedom of choice – there is some debate regarding nominal freedom of choice, but that’s beyond the scope of this piece). Whilst both you and Apple may agree you need and want to cut down on your screen time, and whilst Apple may be better informed of how to meet this end more effectively, there’s something deeply sour about this approach.

In nudge theory, the scenario is not so explicit. Say you want to pick a pension scheme, but there’s a lot of information out there and you don’t really want to trawl through it all to pick the absolute best scheme for you (assuming you’re even able to do that). Nudge theory may advocate a carefully selected default option pension scheme. The default option effect states that when faced with a decision, people are more likely to pick the default option than an alternative option. If the default pension plan is the optimal pension plan for the average saver, then, so the theory goes, welfare may be maximised by nudging people towards the default plan, rather than allowing them to sift through all the information and potentially pick the objectively wrong policy.

And for most people this system works. Whilst some argue against this idea, conjecturing it’s not possible to know what other people would choose for themselves (again, see Rebonato (2014), or, more broadly, On Liberty by Mill (1859)), this misses a point: in some situations, most people will not know what is the best option for them, and they’ll be happy enough to be nudged in the ‘good enough,’ direction (see Sunstein and Thaler (2003) who cite Beattie et al. (1994) on this point). The stronger criticism comes in the form of a question: can we do better?

This is where personalised paternalism comes in. The problem with either the hard paternalism example of Apple setting the downtime options or the soft paternalism of experts nudging people towards the best – on average – pension scheme is that most people will still be left somewhat unsatisfied by these approaches. Whilst the default scheme might be good enough, it won’t necessarily be optimal for me as an individual. Similarly, whilst on average it might not be wise for a person to spend umpteen hours a day on their phone, you might be able to cope with the effects of screen time much better than the average person. By adhering to averages, everyone will lose out – though some more than others.

To get the best of both worlds, we would need to be able to feed all our preferences, ideas and uncertainties into the system, and then let someone else’s expert judgement determine the best course for us specifically (assuming, of course, you want someone else making the decision). This is personalised paternalism, and as we begin to collect more and more data about ourselves, the potential for the theory to become realised skyrockets. Apple’s screen addiction software may mark an important step in this direction. Here’s why.

Say Apple reports back to you how much time you send playing Candy Crush every week, and you decide you want to cut down. You could specify an amount of time each week you are allowed to play Candy Crush, and let the software work in the background to keep track of everything. Perhaps in the future this software might access data in your calendar to know when you should be working, or your location data to figure out you like to play whilst you’re commuting to work, and it will build your allotted time around these preferable and not so preferable time frames? This might work fantastically, potentially increasing productivity and making downtime feel more rewarding.

Of course, it’s easy to spin all of this as a positive and ignore the negatives. If the goal of this software is to cut down on screen time, is it really the best strategy to let technology coordinate more around your life? Additionally, as I have argued before – following arguments put forward by Bar-Gill (2012) – even in disclosure interventions (when a person is given more information to help them reach a better decision themselves, rather than asked to trust the judgement of someone else who has that information, i.e. a nudge) which the above example is, there are still paternalistic aspects, and aspects that infringe on freedom of choice. For example, if I was told how much time I don’t use my phone every day, my feelings about how I should moderate my use will probably be different compared to the effect of being told how much I use my phone. The decision of how to frame information disclosure is a paternalistic act in itself.

Finally, we should always ask an important question: why do we need it? Personalised paternalism supposes to make our lives better by tailoring regulatory strategies and nudges to our specific requirements; but is it right that we should feel compelled to regulate ourselves? Is it the consumer’s fault that they spend too much time on their iPhone, or Apple’s for making such an appealing device, or King’s for using bright colours and sounds in Candy Crush? There is no absolute answer to these questions, but I believe they’re important ideas to think about nevertheless.

I generally believe Apple’s announcement is a positive one, and when we accept some form of paternalism will always exist, the ability to have some control over it if we want is an important feature to have. But paternalism, nudges and the agency question that surround both remain important questions to consider, and it would be dangerous to see one solution as a panacea.

References

Arad, A, Rubinstein, A (2015) ‘The people’s perspective on libertarian-paternalistic policies,’ working paper no. 5/2015, research no. 00140100. [Date accessed: 04/06/2018] [Online]: https://www.researchgate.net/profile/Ariel_Rubinstein2/publication/321161872_The_People%27s_Perspective_on_Libertarian-Paternalistic_Policies/links/5a129b3e458515cc5aa9e5a4/The-Peoples-Perspective-on-Libertarian-Paternalistic-Policies.pdf

Bar-Gill, O (2012) ‘Seduction by Contract’ Oxford University Press: Oxford

Bradshaw, T (2018) ‘Apple addresses screen addiction with new suite of tools’ Financial Times. [Date Accessed: 04/06/2018] [Online]: https://www.ft.com/content/e4048d90-6824-11e8-b6eb-4acfcfb08c11

Beattie, J, Baron, J, Hershey, J, Spranca, M (1994) ‘Psychological determinants of decision attitude’ Journal of Behavioural Decision Making, 7(2), pp. 129-144

Mill, J S (1859) ‘On Liberty’ in ‘On Liberty, Utilitarianism and Other Essays’ (2015), Oxford University Press: Oxford

Rebonato, R (2014) ‘A Critical Assessment of Libertarian Paternalism’ Journal of Consumer Policy, 37(3), pp. 357-396

Thaler, R, Sunstein, C (2003) ‘Libertarian Paternalism’ The American Economic Review, 93(2), pp. 175 -179

Thaler, R, Sunstein, C (2008) ‘Nudge’ Penguin Books and Yale University Press: London


Cryptocurrencies and Corpocracies

Cryptocurrencies are not libertarian. To be sure, aspects of cryptocurrencies, and the blockchain technology on which they are built, reso...