Editorial

Undocumented Trust: Could vouching offer a more inclusive approach to identity?

To provide truly inclusive identity services, everyone in society must be able to access them – regardless of if they have trusted ID like a passport or driving licence. So how can we create trust without documents? Accenture’s Ilze Skujina and Rachel Moss suggest vouching could be the answer.

Posted 14 February 2022 by Christine Horton


Having a trusted form of identity is essential for many aspects of modern life. However, not everyone has access to the most commonly accepted forms of official ID, like a passport or driving licence, making it difficult to do even basic tasks like opening a bank account or taking out a phone contract. Solving this issue could help create a more inclusive society and there are a number of potential new solutions that could help.

Many of these exist at the cutting edge of digital technology. However, one concept that could have a major impact actually comes to us straight from the Victorian era, vouching. Making it work will mean resolving some key questions. Who can vouch for whom? How do we prevent exploitation from taking place? And how can we be sure the vouch is trustworthy? But, before turning to these we must explore a theme that came up time and again at the recent Think Digital ID for Government conference, trust.

What is trust and how is it built?

There is a good reason why the idea of trust comes up time and again. It is built into the fabric of cultures and societies of all shapes and sizes, but it is itself a complex concept. Martin Mathews argues that trust contains a fundamental paradox. Whilst it is a mechanism that is used to reduce social uncertainty, it also contains, within itself, uncertainty. Applied to identity it raises clear challenges. Can we ever really be sure that our trust in an identity service is well founded? How do we know what we don’t know? This is where trust in the provider of a service is so important.

Trust in its most basic form can be described as reliance. If I rely upon you to do something, I have to trust that you will behave in a certain way. As the French Anthropologist Marcel Mauss puts it “one trusts completely, or one mistrusts completely”. In this seminal study, Mauss outlines how trust was generated through gift giving between geographically disconnected Polynesian islanders as a way of inducing reciprocity through a process that, itself, becomes increasingly trusted as more people participate in it.

Applying this principle to identity creates some interesting implications. Firstly, someone either trusts a service or they don’t, there is no middle ground. Secondly, a reciprocal relationship between identity providers and users will help build stronger trust between both parties. And thirdly, trust in a system can be developed over time through repeated engagement.

Rather than a one-way relationship between the identity provider and user, we have a give and take. In a successful model, both parties will gain from the relationship and through repeated touchpoints the parties come to trust the system it is built on. Not only would this mean that an identity solution that is more inclusive is more trusted because of wider adoption, it also offers an alternative model to the standard approach for identity verification. One built on a system rather than depending on documentation from trusted authorities.

How can we create trust without documents?

To provide truly inclusive identity services, everyone in society must be able to access them – including those without official documentation. A ‘vouch’ is one approach to providing this sort of verification. In this system a person vouches for someone else by declaring they know them as their claimed identity.

This is not a new phenomenon. Vouching has been used through time to build trustworthy identities. Victorians used it as an introduction to society, a “guarantee that the person being introduced was both respectable and worthy of knowing”. Introductions like this were especially important in small villages as without one, the individual would not be accepted into these close-knit societies.

This method is still in use to a certain extent within a variety of contexts such as passport renewals and mortgage applications. However, in most cases, the person doing the vouching has to be someone in a ‘position of authority’ (e.g. a doctor or local government officer). That is to say, people that are deemed ‘trustworthy’ within this context.

In contrast, Victorians were obliged to rely on the judgement of their friends and family and had to trust that the person being introduced to them was worthy of that introduction. In this model trust is not something that people simply gain by being in positions of authority, though this no doubt helped, it was developed over time and through repeated interactions.

Applied to a digital world this model could have a major impact for people. Take accessing welfare services online as an example. Rather than relying on documentation to verify someone’s identity, a provider could ask the new user to provide a list of referees who are also service users. The existing relationship with these users confers trust and creates an alternative for people who might struggle with documentation.

Interestingly, whilst forms of ID focus on the individual being identified, in this case it is the voucher that needs to be trusted. So, defining this is key. As society changes this definition of trustworthiness might change with it. A future-proof system should, therefore, take into account how these changes might affect the person being identified if their vouchers are suddenly deemed to be untrustworthy (or the other way around).

Further inspiration

The latest Think Digital Identity for Government conference highlighted the importance of looking to social science to think about how humans behave. Along with the development of the UK digital identity & attributes trust framework, we could apply this approach to help us build trust and inclusivity with our solutions. Vouching is relatively new to digital identity, so it is unsurprising that there are still a number of open questions, and it is far from the only solution to create a more inclusive identity system. However, exploring approaches like this could help to broaden access to more people, which should be the goal of any identity system.

This content is provided for general information purposes and is not intended to be used in place of consultation with Accenture’s professional advisors.