Thanks mate!

I show that while there is only one question on the form that a person might answer differently in-person as opposed to online (and without assistance), it is a very important one, and one which is important in determining the level of support they receive from the state.

While many questions are the same regardless of the platform through which we encounter the question, it's not true for all questions. In this paper, I use the example of the Job Seeker Snapshot to explore how answers can differ depending on who is there to help you provide them.

As I argue in this paper, the shift to technology-generated interactions with the state can generate additional administrative burdens, and it can change the way we respond to some questions.

The answer I give in the paper is that increasingly it is citizens doing the administrative work that was previously done by the state (and also us as customers that are doing the work for private enterprises as well).

I had a paper that came out at the start of the year looking at administrative burden as a constraint on freedom that I think is useful for thinking about what’s going on in this space - it’s the deliberate restriction of citizens being able to exercise their rights.

Yeah, that’s definitely what Don highlights in his recent writing on what’s happening (which I think was the basis for some of Oliver’s segment).

It’s such a fascinating space at the moment, and in the context of our continual shift towards a digital welfare state it’s also getting harder for service users to see a real person to talk through things when they go wrong, which is about to happen a lot in the U.S.!

I haven’t had a chance to watch it yet but I’ve seen Don Moynihan talking about it a bit.

To put it bluntly, the project is an attempt to understand why some paperwork sucks, and how it could be better. I look forward to sharing more work from the project soon and unpacking why some forms are so bloody frustrating...

This work is part of a larger project which looks at the mechanisms that can make administrative documentation hard to interact with. I've been trying to take a very granular approach to different documentation to identify exactly where things go wrong, and how they can go wrong.

It is a particularly striking example of where behavioural insights have been applied in a subtle manner to produce an outcome, with little consideration of the harm that might occur by doing so. It shows us how subtle choices matter.

As I argue in the paper, the techniques used to generate this particular response were harmful to the economic dignity of the people who received this letter, and framed them in a negative light.

In the paper I examine how the initial Robodebt correspondence letter was designed using behavioural insights to produce a particular response - to get people to accept the debt - in an application of clawback logic.

This is a paper about why policy design choices matter, and about how subtle design choices - really specific choices about which words to use - can be deeply harmful.

It’s exciting to have people engaging with this piece of work!