Will refugees prove the case for a human-centered data infrastructure?
We are witnessing a historic global refugee crisis. There are today 70.8 million people in the world that were displaced due to conflict or disasters, 23 million of them are registered as refugees. The Syrian refugee population alone accounts for 5 million people.
A common challenge that refugees face is the lack of verifiable documents about their identity, finances, and education. This can make it very difficult to get official papers quickly and access important services. Purchasing a SIM Card, accessing employment and financial services are all contingent on the presentation of an “official” proof of ID.
There is a need for a solution that improves the verifiability and availability of documents especially in transit and immediately after arrival. But we need to think beyond that. Once refugees access services, these services, in turn, generate data (or contextual identities). We need an infrastructure that allows to store this data, make it portable, and secure. There is a huge compounding effect of having such an infrastructure. It can make it much easier to build a reputation, access education and healthcare or pay, save and invest at a reasonable cost. Not only for the refugees, but also for the host communities.
Most of us do not appreciate the importance of this. We don’t realize just how much we benefit from the availability of — and trust in — data about our personal and professional life. Third-parties can verify information about our education, business registration, and assets such as real estate. It forms the basis of our investments, savings, and businesses and thus our economic life.
For refugees, the portability of data is key. Many never return home and keep on the move. Once displaced for five years, a person spends an average of more than 20 years displaced. Often initially separated from their families, they spend years living and working in a country before being reunited with the rest of their family. We must avoid that they have to start over because their digital documents and data do not move with them.
Another essential requirement is data privacy and security. Data, especially in the case of refugees can be abused to track movements or discriminate. Also, the use of biometrics in many registration steps massively increases the stakes of a data breach.
All these requirements can only be met through innovation and change at a systems level. And only if this results in solutions can be leveraged in different geographies and are sustainable in the long term. That is they can exist without depending on funding by a handful of stakeholders.
Decentralized identity has the potential to fulfill many of the requirements. It puts the individual at the core of their personal data. This means data moves with the person and is under their control. It has the potential to empower individuals with data and simultaneously align everyone’s interest. Humanitarian organisations can split the costs of registration and collecting data. Governments, and the private sector can verify data to register refugees for their services. These services can in turn share data on the platform and even monetize it. And refugees have a powerful tool to gather trusted data for the long term.
This doesn’t come without risk. In fact, there are many things to be figured out. First, control comes with responsibility. The individual has to decide when to share data and with whom. Yet, people are rarely in a position to evaluate the consequences of handing over personal data. Also, the power dynamics are such that refugees often do not have a meaningful choice. Second, the implementation of full security and control also requires individuals to handle encryption. That means access to smart devices and storage of encryption keys, which can be lost and are difficult to recover. And lastly, we should ask ourselves why it is necessary to gather and require certain types of data in the first place.
Yet, the potential is huge, and many problems can be addressed through good design choices. Technology will only be one ingredient, however. The trust and data need to come from elsewhere. The entities that issue the foundational credentials for refugees will have a key role to play in driving that paradigm shift. These are in particular the UN agencies and the governments of the host countries. If everyone works together, refugees and the organizations that interact with them could be the first to prove the concept of human-centered data structures at scale.
The refugee identification journey
As mentioned, many refugees lack trusted documents when they arrive in a new country. There are mainly two types of reasons for that. The first type is non-availability. Often, because refugees have to leave in a rush, documents are left behind. There are many accounts of people risking their lives to obtain certification of education before fleeing their homes. Also, documents can be lost, damaged, or taken from them when they are in transit. The longer the journey lasts, the more likely this is to happen.
The second type of reason is not the lack of documents, but the lack of acceptance by third parties. Even if refugees have documents with them, it’s usually impossible to verify the authenticity of that information. This is a very important point. Documents, especially paper documents can be faked. An important feature of what people call digital Identity system is being able to verify data about a person.
The impact of this is especially important at the beginning of a refugee’s “identification journey”, because of the hierarchy of credentials and ID requirements for access to certain services.
Let’s take a look at the typical identification journey of a refugee in the chart below.
Every country has its own system. For instance, in some countries, the government is in charge of doing the first registration of refugees instead of UNHCR. Also, the specific ID requirements for things like work permits and access to SIM registration and financial services differ from country to country.
In addition, every refugee’s situation is different at the outset. Some might have managed to carry documents that allow to register them with some basic ID information and clarify family bonds. Yet, even if these “breeder documents” are not available, UNHCR will still register everyone based on self-claims. This is in the interest of providing them basic assistance and protection quickly. Still, lacking these documents can significantly slow things down at later stages in the process.
A very critical step in this chart is the red arrow. Our chart is not really true to scale. It can take months or even years to obtain the first official credential from the host government. Sometimes, this is due to a lack of political will to officialize refugees’ statuses and enable economic inclusion. But there is another very practical reason. If breeder documents are lacking, or are not easily verifiable, individuals become “difficult cases” and end up in a separate box. Admin officers in the host country rarely have any incentive to proactively handle a difficult case.
This is an issue because authorization to work in the domain of expertise or to open a business are often dependent on this first foundational credential. However, it is at this very point, immediately after displacement, where it is the most crucial to have continuity, be able to stay in work and become at least partly self-reliant again. Being paralyzed for months as a consequence of lack of opportunity significantly increases the risk of mental health issues and in some cases even drug abuse and crime, all of which are interrelated. The problem with this situation is its non-ergodic nature. Not many people recover from extended periods of drug abuse and mental health issues. Arguably, there is a lot of attention for success stories of the 0.1% of refugees that “made it against all odds”, but it is safe to assume that for every success case there’s a lot of silent evidence of the potential that was lost in the first weeks and months after displacement.
Can technology help speed up identification and access to services?
There are basically two pathways for improvement of the above process:
First, the ID requirements can be softened. It would be extremely helpful if refugees were able to register a SIM card immediately after having been registered by UNHCR. Also, linking work permits to the alien instead of requiring an additional registration process can be impactful. This change is reflected in the chart below.
Second, we can try to speed up the different steps, in particular the registration for the alien ID.
Both will require advocacy, political will, and regulation changes. However, technology can bring something to the table: Verifiability.
The lack of verifiability is a big pain point at many stages during the identification journey. The Communications Authorities and telco operators in some countries refuse to accept UNHCR or WFP issued credentials for SIM registration. One, not the only, important reason is the lack of verifiability. There’s no way to check digitally for the SIM registration agent if the provided identity information is true. True, in this case, means it is the same data that was registered by the UNHCR. Using decentralized identifiers, these organizations could issue digital verifiable credentials that have verifiability built in. While at it, these organizations could also digitize and certify whatever other documents the refugees carried with them. This could help speed up the alien card process by making it easier for the officer to verify information.
Why decentralization?
Ok, fair, digital verifiable credentials could make things easier. It can speed up the process and increase trust in breeder documents that might then become sufficient for registering a SIM card. Hopefully, this will help some people get into work and become self-reliant more quickly. However, this will not be enough to justify a fundamental change in how we structure data around people.
We have mentioned earlier that every service that a refugee can access creates more collections of data in turn. We can call them contextual identities. With ever more data being aggregated there are two options. Either every service provider (this includes NGOs, the government, and the private sector) store that data in their own databases. Usually, these databases don’t talk to each other, and if they do, we must ask what entitles these entities to share personal data. The other option is to put the individual at the core and in control of their personal identity and data. Which is what we call decentralization.
The second option allows individuals to build rich data personas. There are countless projects by the private sector, NGOs and governments to create employment and livelihood opportunities, especially in the gig economy and e-commerce. These applications are the actual access channels and are therefore extremely important. The data assets that these solutions create need to be made available, not only for individual empowerment but also for data-informed decision making and governance. Otherwise, we create a huge missed opportunity for positive spill-overs between these different initiatives. This doesn’t mean that this data is flowing around in cyberspace for everyone to use. Decentralization gives individuals control and maybe for the first time ever an opportunity to actually monetize their personal data. It also makes data breaches at scale much less likely, simply because there is no central point of attack.
It also fulfills the requirement of portability that we put forward in the introduction. Many refugees will be on the move with intermittent periods of settling down several times. Frequently, this has to do with family reunification. During the Syrian refugee crisis, it was common for families to split up during the journey. Let’s consider the example of a refugee who arrives in Jordan and is unable to join his family (not only immediate family but also cousins or friends). He obtains skills certifications from UNHCR and manages to find work in Jordan. He is a huge asset to his employer and manages to save up a fair amount of money. In addition, he accesses and reimburses several microcredits to invest in skill building and get over some low-income periods. After 3 years there is an opportunity to join his family elsewhere. If he moves, he risks a lot. His credit history will not be recoverable by a financial institution in the new country. His work experience and education are not easily verifiable for any employer and he’s therefore likely to fall through the cracks just because of the comparatively higher cost of verification and the lack of trust (The very same problem he faced when arriving in Jordan).
Are refugees going to trigger the next evolution in technology?
The unprecedented scale of the refugee movement and its impact on the political landscape in rich countries in Europe and North America has created a greater sense of emergency and incentive for action among governments and international organizations. This creates a situation (even if it is for the wrong reasons) were a real paradigm shift seems at least possible. A system where a person’s records are stored in centralized databases is no longer adequate for an increasingly mobile world. It is time to flip the equation and put the individual at the core of their personal identity and data. The technology is there and it has the potential to become an important part of the digital infrastructure of a transitional economic system. Such a system would also benefit populations in low-income countries. By solving the trade-off between data aggregation and security, it can make economies more efficient and contribute to regional integration.
This doesn’t mean undermining the authority and position of trust of nation-states and international institutions. On the contrary, these entities will be crucial to the establishment of this new paradigm as trust providers and issuers of trusted data.
Human-centered data infrastructures seem like the logical next step after human-centered product design and computation (smartphones and laptops). And important figures like Tim-Berners Lee have embarked on that mission. Yet, decentralized identity hasn’t yet exactly taken off anywhere. Refugees and the organizations that interact with them might just be the first ones proving the case at scale.
