Digital identity in the interest of all:

(1)

Digital identity in the interest of all:

Future or Fantasy?

A research into the differences between the point of view of the aid

provider and people affected surrounding digital identities in

humanitarian CTP

Hiske Doornekamp – S3504328

September 2019

▪ First supervisor: Prof. Dr. A.J. Zwitter, Rijksuniversiteit Groningen

Second supervisor: Dr. C. Churruca Muguruza, Universidad de Deusto

This thesis is submitted for obtaining the Master’s Degree in International Humanitarian Action. By submitting the thesis, the author certifies that the text is from his/her hand, does not include the

work of someone else unless clearly indicated, and that the thesis has been produced in accordance with proper academic practices

(2)

2

Abstract

The humanitarian sector is adopting new technologies at a high pace. Also the use of digital identity systems is explored. One of the possible applications is in humanitarian cash transfer programming. The two most important stakeholders, the people affected and the aid provider, have different interests in the implementation of digital identities in this context. This research aims to gain greater insight into both points of view. To do so, a desk research is combined with interviews and results of co-design sessions.

The results of this research show that the people affected have an interest in a reliable and timely transfer; a flexible, low-cost and easy process; a familiar cash distribution mechanism; data protection; communication and involvement; trust and human involvement; safety; and empowerment and inclusion. The potential for increased empowerment and inclusion is the strongest argument in favour of digital identity systems for the people affected. The aid provider sees chances in institutional efficiency and effectiveness; transparency, accountability & corruption reduction; data-driven decision making; interoperability; and better outcomes people affected. Challenges as viewed by the aid provider are dysfunctional technology, illiteracy among people affected, insufficient institutional capacity; and data protection.

In conclusion, it seems that for people affected there is more at stake than there is to win with the implementation of digital identity systems. For aid providers there are substantial arguments in favour of implementation. Yet, stating that aid providers are in favour of digital identity systems and people affected are against would not capture the necessary nuances, considerations and motivations on both sides. Both stakeholders have an interest in an efficient mechanism that helps realize quick and reliable assistance. For both stakeholders applies that the programme design, the approach of technology, and the approach of identity are crucial aspects that can determine the success of working with digital identity systems.

The researcher gives six recommendations to aid providers to overcome the discrepancy between the two views. These can be categorized into a central position for the people affected; communication and involvement; informed, valid consent; training; approach of technology; and approach of identity.

(3)

3

Abbreviations

AI Artificial Intelligence

CTP Cash Transfer Programming

FSP Financial Service Providers

GDPR General Data Protection Regulation ICRC International Committee of the Red Cross IDP Internally Displaced Person

IFRC International Federation of Red Cross and Red Crescent

Societies

KRCS Kenyan Red Cross

KYC Know Your Customer

LMMS Last Mile Mobile Solutions M&E Monitoring and Evaluation

NGO Non-Governmental Organization

ODI Overseas Development Institute

PRIMES Population Registration and Identity Management Ecosystem PII Personal Identifiable Information

SSI Self-Sovereign Identity

UN United Nations

UNDP United Nations Development Programme UNHCR United Nations High Commissioner of Refugees

(4)

4

1. Introduction

Over the last years, funding for humanitarian action has increased substantially. At the same time, the humanitarian financing gap has never been as yawning as nowadays (Development Initiatives, 2018; High-Level Panel on Humanitarian Financing, 2016). In 2017 for example, funding for UN-coordinated appeals - that are central in humanitarian assistance - was higher than ever before with 14.9 billion USD. Nevertheless, with 10.3 billion USD, the unmet requirements were higher than ever before (Development Initiatives, 2018).

This growing gap is a result of the increase of humanitarian crises in both quantity and duration in the 21st century. In the period 2005 – 2017, the number of crises with an internationally-led response rose from 16 to 30. The length of these crises went up from four to seven years on average (UN OCHA, 2018). In 2017, as a result of natural hazards, conflict and forced displacement, the number of people in need of humanitarian assistance in order to survive increased to over 201 million people, located in 134 countries (Development Initiatives, 2018).

Humanitarian assistance can be described as ‘intended to save lives, alleviate suffering and maintain human dignity during and after man-made crises and disasters caused by natural hazards, as well as to prevent and strengthen preparedness for when such situations occur’ (Development Initiatives, n.d.). The act of showing compassion and helping other people is universal and exists as long as humankind. The foundation of the Red Cross in 1859 marks the beginning of a period where humanitarian action became more organized (Davey, Borton & Foley, 2013). Nowadays there are countless local, national and international actors active in the field of humanitarian action. There are coordinating bodies such as the United Nations (UN) Office for the Coordination of Humanitarian Affairs and structures for cooperation such as the cluster approach, all these actors have their own intentions, specialism and approach. This means humanitarian assistance comes in different shapes. One way to make a categorization is based on the type of transfer modality. The main types are in-kind, vouchers and cash. In-kind assistance has always been dominant, with an estimate of 94 percent of the total humanitarian assistance provided in-kind in 2016 (World Bank Group, 2016). In-kind assistance comes in the form of goods and/or services, such as food, seeds or medicines (CaLP, n.d.a; Harvey & Bailey, 2015a).

This research focuses on assistance in the form of vouchers and cash, referred to as Cash Transfer Programming, Cash Based Interventions or Cash Based Assistance (CaLP, n.d.a). It is debated whether assistance in the form of vouchers1_{is part of cash assistance. The UN}

1_{Vouchers are paper or electronic entitlements to goods and/or services (CaLP, n.d.a; Harvey & Baily,} 2015a).

(7)

7 High Commissioner for Refugees (UNHCR) for example includes vouchers in their definition of cash transfers in their ‘Operational Guidelines for Cash Based Interventions in Displacement Settings’ of 2015, but on the other hand the Department For International Development argues that CTP cannot include vouchers (World Bank Group, 2016). In this research, following the guidelines of the Cash Learning Partnership, vouchers are included in cash assistance. The term Cash Transfer Programming (CTP) and the definition ‘The provision of money or vouchers to individuals or households, either as emergency relief to address basic needs or as recovery support to protect/re-establish livelihood/economic productive activities’ are used (HCT Cash Working Group Philippines, 2016).

Humanitarian action in the form of cash assistance has a long history. Examples of crises where both national (governments and civil society) and international actors (UN, Red Cross and Non-Governmental Organizations [NGO’s]) gave assistance in the form of cash, are the Franco-Prussian war of 1870-1871, the Indian famine in the 19th century and Botswana in the 1980s (Harvey & Bailey, 2015a). In the 1990s and 2000s the number of cash programmes slowly started to increase, with the 2004 Indian Ocean Tsunami as turning point. To assist people affected by the Tsunami, several organisations kicked-off their first cash programmes to see whether this was a proper replacement of food assistance. CTP became one of the most researched subjects in humanitarian action in the 21st_{century (Harvey & Bailey, 2015a;}

Harvey & Bailey, 2015b; World Bank Group, 2016).

Evidence shows that CTP in crisis and disaster can complement or substitute in-kind aid effectively in order to meet the needs of the people affected, and herewith reduce immediate hardship. Moreover, looking at the long term, CTP has the potential to be driver of both economic and social development (Bailey & Harvey, 2015b; Hanlon, Barrientos, & Hulme, 2010). Firstly, the research shows that people affected use the transfer efficiently for essential goods and services such as food, shelter, agricultural tools, health care and debt repayment. Rarely the transfer is used for goods such as alcohol, tobacco or arms (Bailey & Harvey, 2015b; Hanlon et al, 2010). Secondly, cash transfers can be one of the most cost-effective and cost-efficient delivery mechanisms because the costs for among others the procurement, transport and distribution of supplies are significantly lower than with in-kind assistance (CaLP, n.d.a). Other important arguments that support the narrative in favour of CTP are human dignity and empowerment. People affected have the opportunity to prioritize their own needs and decide themselves what is best for them. It gives them more control of how they respond to disaster or conflict (CaLP, n.d.a). Finally, evidence shows that the money received through cash transfers is usually spent locally and through multiplier effects stimulates local economies (Bailey & Harvey, 2015b; Hanlon et al., 2010).

(8)

8 Through the 2016 ‘Grand Bargain’, a document aiming to improve efficiency and effectiveness of humanitarian aid, 53 significant donors and implementing agencies committed to implement CTP on a larger scale. It is estimated that in 2016, 10.3 per cent of the total money for humanitarian aid, was spent on CTP. Anno 2018, the debate is not anymore about whether CTP can improve humanitarian aid, but about how it should be adopted to have an optimal outcome (Smith et al., 2018).

Though the humanitarian sector acknowledges the effectiveness and efficiency of CTP, as with any form of assistance, there are also challenges and difficulties. Annex one gives an overview of this as indicated in different (case) studies conducted by researchers and aid providers.

There are different ways how cash transfers can reach the people affected. Firstly, there are unconditional and conditional transfers. In a conditional transfer, there are certain requirements that the person affected needs to meet in order to receive the money. This could be for example the attendance of school or training, building a shelter or the attendance of regular check-ups at a healthcare unit. Cash for Work is a form of conditional cash. Cash is provided on the condition of undertaking designated work. Unconditional cash is cash that is given without any requirements regarding expected behaviour or actions (CaLP, n.d.a; Harvey & Bailey, 2015a). Secondly, there are restricted and unrestricted transfers. This refers to the utilization of the money. With a restricted transfer, the people affected are required to use the received cash for specific goods or services. With an unrestricted transfer, the people affected can use the received cash entirely as they wish (CaLP, n.d.a). There are also different delivery mechanisms. There is direct cash, which is cash distributed by hand (‘cash in envelopes’). This can be done by the aid provider, local financial service providers (FSPs), or other actors such as a post office. Next to direct cash, there is so called ‘e-cash’ (CaLP, n.d.a; Harvey & Bailey, 2015a).

There is a growing interest in e-cash, also referred to as electronic or digital cash.2_{E-cash can}

be stored, spent and received through for example a mobile phone, prepaid ATM or debit card. Depending on the system in place, the money can be used directly to buy goods or access services (such as with M-Pesa, a mobile phone-based money transfer system which allows you to pay with your phone) or the money can be withdrawn from ATM machines, banks,

2_{There is a slight difference between the two. In electronic transfers money is transferred from the} bank account of the aid provider to the bank account or mobile phone of the person concerned. In digital transfers money is transferred as well from the aid provider to the bank account or mobile phone of the person concerned, but the addition is that digital platforms are used to do this (Trulioo, 2018).

(9)

9 payment providers (such as local shops) and mobile phone companies (CaLP, n.d.a; Smith, 2012; Harvey & Bailey, 2015a).

In among others the context of digital cash programmes, digital identities are a topic of interest in the humanitarian sector (USAID, 2017). As discussed later in this research, multiple major players in the field of humanitarian action are developing or already working with digital identities for different purposes. The use of digital identities in humanitarian programmes has potential benefits, but there are also questions about the risks for the people affected who are already in a vulnerable position. Rather than inclusion, does the use of a digital identity lead to the exclusion of people who are (digital) illiterate? To what extent can personal information in a digital identity system place a person at risk when it is shared, leaked or hacked? How is the right to privacy adhered to and to what extent have people affected the real possibility to raise their voice when this right is not safeguarded?

Touching briefly upon the topic of digital identities shows that depending on the point of reference, the view on benefits and risks can be quite different. This research focuses on the dilemma of the differences between the point of view of the people affected and the aid provider in the context of digital identities in humanitarian CTP. This results into the following research question:

‘What are the differences between the point of view of the people affected and the point of view of the aid provider surrounding digital identities in humanitarian cash programmes?’ To answer this question, the research can be broken down into four sub questions:

- Sub question 1: What are positive aspects of the incorporation of digital identities in CTP for the people affected?

- Sub question 2: What are negative aspects of the incorporation of digital identities in CTP for the people affected?

- Sub question 3: What are positive aspects of the incorporation of digital identities in CTP for the aid provider?

- Sub question 4: ‘What are negative aspects of the incorporation of digital identities in CTP for the aid provider?’

Structure

This paper starts with a conceptualisation that explains some technical concepts that are crucial to understand before reading this research. Next is the literature review, which gives insight into the existing academical literature about technology in humanitarian action, technology in cash programming and blockchain in humanitarian action. The review is concluded with hypotheses linked to the problem statement and sub questions of the research.

(10)

10 The methodology follows. The methods of research are explained, the ethical considerations are discussed, and a section is dedicated to the limitations of the methods. After the methodology, the results follow. This chapter is divided into the point of view of the people affected and the point of view of the aid provider. Both sections are a combination of desk research and empirical data. The discussion goes through the results and confirms or rejects the hypotheses. Finally, the conclusion summarizes the answer of the main research question, gives recommendations to bridge the gap between the different point of views, discusses the limitations of the research, and gives recommendations for future research.

(11)

11

2. Conceptualisation

This chapter explains the key concepts of this research:

- Identity;

- Identity systems; - Digital identity;

- Foundational and functional identity;

- Centralized, federated and self-sovereign identities; - Blockchain.

Identity

Identity is a complex concept consisting of someone’s individual, relational and social identity. An individual identity can be described as ‘… a vehicle through which one can build his or her global life story, incorporating goals, beliefs and values (Nagy & Koles, 2014)’. A relational and social identity refer to the social interaction of a person and the (perceived) membership of social groups, such as ethnicity, nationality, religion, gender and family (Nagy & Koles, 2014). The concept of ‘identity’ has no unambiguous definition, but is rather a social construct, subject to its historical, social and cultural context. The answer to the question ‘Who am I’ largely depends on external factors: other people are the mirror in which a person sees oneself (Tatum, 2000).

An identity in the context of an identity management system – as in this research - is ‘a set of permanent or long-lived temporal attributes associated with an entity’ (Camp, 2014). It is important to acknowledge the social and emotional value of an identity to understand the sensitivity of identity issues.

Identity systems

An identity system consists of different elements. A passport can be used as an example to explain these elements. The first component is the identifier. An identifier distinguishes a person, place or thing. In the example of the passport, the passport number is the identifier. For a bank account this is the IBAN, and for a car this is the license plate. Secondly, there are attributes. An attribute is a characteristic of the entity. The passport lists for example the nationality and height of an individual as attributes. Then there is a personal identifier. This is a (set of) identifier(s) linked to an individual, based on attributes that are difficult or impossible to amend. In the case of the passport, the personal identifier is for example the date of birth and name of an individual. These are components one could lie about, but they cannot be changed. Finally, an identity system includes a biometric method of identification, such as a photo in the case of a passport (Camp, 2004).

(12)

12 The function of an identity system is morefold. Firstly, it can be used for identification. Identification is ‘the association of a personal identifier with an individual presenting attributes’ (Camp, 2004). This could for example mean the acknowledgement of the claimed name of a physical person. Secondly, an identity system can authenticate. There is a distinction between identity authentication and attribute authentication. Identity authentication is ‘proving an association between an entity and an identifier’ (Camp, 2004). A passport that shows the link between a physical person and its name is an example of identity authentication. Attribute authentication is ‘proving an association between an entity and an attribute’ (Camp, 2004). Here you can think about the linkage between a painting and a certificate of authenticity. The final function of an identity system is authorization: the decision to allow a specific action (Camp, 2004).

Digital identity

In the current era, due to globalization and technological innovation, people can interact digitally with other people, companies and institutions while being located on the other side of the country or even the world. Identity is no longer just present in the physical world. Like a non-digital identity, a digital identity can be explained in different ways, is multi-dimensional and is subject to its context. Some argue that the term ‘identity’ as a subset of all personal data is too narrow and not accurate any longer, because in the current connected world of big data, a person’s whole digital footprint is connected to one’s digital identity (Voshmgir, 2018). In this research, the definition ‘a set of electronically captured and stored attributes and credentials that can uniquely identify a person’ is used (IBRD & World Bank, 2018)’. Next to digital identity, there is Personal Identifiable Information (PII) and personal data. PII is any data that can be used to clearly identify an individual. This could be a national insurance number, mailing address or phone number, but also an IP address, login details and social media posts (Jennings-Temple, 2018). Article 4 of the General Data Protection Regulation (GDPR) of the European Union defines personal data as ‘any information relating to an identified or identifiable natural person’. This means all PII is personal data, but not all personal data is PII (Seaman, 2016). Following the definitions used in this research, the terms PII and digital identity are close to interchangeable: both are defined as data that leads to the identification of an individual. The difference is that a digital identity is about a set of information and PII can be both a single piece and set of information (Seaman, 2016).

Under the GDPR there are so called special categories of personal data, which are sensitive personal data. This includes genetic, biometric and health data, as well as personal data revealing racial and ethnic origin, political opinions, sexual orientation, religious, philosophical or ideological convictions or trade union membership (EU Regulation 2016/679: article 9).

(13)

13 To better understand the components of a digital identity, the Digital Identity Lifecycle of the World Bank (see figure 1) can be followed.

The lifecycle consists of three stages: registration, issuance of credentials, and authentication for service delivery or transactions. The registration process starts with enrollment of a person that claims to have a certain identity. Enrollment is done through the collection of key identity attributes. This is usually a combination of biographical data, such as a person his or her name, date of birth and address, and biometrical data, such as fingerprints. The second phase of registration is validation, which means that the existence and uniqueness of the identity are examined. This happens through checking the presented attributes against existing data (IBRD & World Bank, 2018). After the registration, the issuance of credentials takes place. Credentials are claims by third parties about an identity. Examples of credentials are a driver’s license, which claims that a person is capable of driving a car, and a university degree that claims that a person has specific knowledge and skills (Longley & Sporny, 2018). For an identity to be considered digital, the credentials issued must be electronic. This means that the data is stored and communicated electronically. Electronical credentials can be categorized into something a person knows, something a person possesses, and something a person is. Something a Figure 1: Digital Identity Lifecycle and Key Roles (World Bank Group and Center for Global Development, 2017)

(14)

14 person knows could be for example a code or password. Something a person possesses could be an identity card or mobile phone. Something a person is entails biometric data such as a fingerprint (World Bank Group and Center for Global Development, 2017). Like with non-digital identities, the process of authentication is the acknowledgement of the credentials which makes the identity legitimate. Ultimately, this means the person is being granted access to a service. The ways to be authenticated and obtain access can usually be classified in the same three categories as the electronical credentials: something a person knows, possesses or is (World Bank Group and Center for Global Development, 2017).

Foundational and functional identity

Identification systems can be designed with either a functional or foundational function. A functional identity scheme is put in place for a specific application, such as elections, tax or health insurance (World Bank, 2016). A foundational scheme is intended primarily to provide identity as a public good, not attached to a specific service (USAID, 2017). In principle, only governments and the UNHCR can issue a foundational identity. This means that other actors who are concerned with identity are talking about functional identity. This is how it works in theory. In reality, the line between functional and foundational identity is blurred. Think for example about functional identities such as social security numbers, driving licenses or health insurance cards that become, in practice, a foundational identity. In for example Bangladesh, Haiti and Mexico, voter identities became national identities (GSMA, 2016; World Bank, 2016).

Centralized, federated and self-sovereign identities

Identity systems come in different forms. A way to distinguish the different forms, is into centralized, federated and self-sovereign identity systems. Most identity lifecycles are managed centrally. This means that one single organization or institution has the power to control the cycle. It does not mean that the identity only has a purpose for that one actor. In for example national identity card systems, the government is the single actor that controls the system, but the identity card is widely accepted as mean of identification (Dunphy & Petitcolas, 2018). The federated identity system is another form of identity management. In such systems, information established for access to a particular service can be used for access to other services as well. A well-known example is Facebook: with a Facebook profile, a person can access other online platforms such as Instagram or Pinterest (Dunphy & Petitcolas, 2018). The final category is self-sovereign identity (SSI) systems. Rather than a third party that controls the identity and its data, the person itself, the so-called end-user, manages. Identity attributes can be added only by the end-user or when that person asks others to do so. The same applies to sharing data: the end-user decides which attributes he or she wants to share and which ones not (Othman & Callahan, 2017). This form of identity management needs to be facilitated through an eco-system of trusted and equal entities that enables the recording and exchange

(15)

15 of the attributes. Blockchain is a technology that can do this (Dunphy & Petitcolas, 2018; Othman & Callahan, 2017).

Blockchain

Blockchain can be defined as ‘a decentralized database, which stores a registry of assets3_and

transactions across a peer-to-peer network’ (Zwitter & Boisse-Despiaux, 2018). The technology can be adopted for different purposes, but was initially created in 2008 to solve the problem of the double spending of the digital currency Bitcoin. With blockchain the same Bitcoin could only be spent once. This can be explained as follows. Unlike in centralized systems, in blockchain there is not a one actor that registers the transactions, but all actors in the peer-to-peer network register the transactions that are made. On a certain point this information needs to be ‘sealed’ with a unique key that everyone in the network accepts. This process is called mining. To seal the information all actors in the network receive a mathematical problem which they need to solve. The number that comes out of this calculation and where the majority in the network agrees upon will be the seal of the information. This number is called ‘Proof Of Work’, because the number is the proof that calculations have been made. After being sealed, the block with data is stored on all devices in the network. The devices, usually computers, are called nodes. Every node in the network needs to record the exact same thing. Skipping, adding or adjusting information is not possible. Together these nodes form the infrastructure of the blockchain. Following this process, double spending is not possible and the record of the transactions is in theory immutable, secure and transparent (Coindance, n.d., CoinCentral, n.d., Mamoria, 2017, Zwitter & Boisse-Despiaux, 2018).

(16)

16

3. Literature Review

This chapter goes into academic literature about digital solutions in humanitarian action. The first part is about technology in humanitarian action in general. Next, technology in cash programming is discussed, and the final part is about blockchain in the humanitarian sector.

3.1. Technological innovation in humanitarian action

After three industrial revolutions in the second half of the 18th_{century, the beginning of the 21}st

century marks the start of the fourth industrial revolution (Schwab, 2016). The first revolution took place around the construction of railroads and the steam engine in the period 1760 – 1840. The ability of mass production enabled by the invention of electricity and the assembly line were central in the second revolution in the late 19th_{century and into the early 20}th_century.

The 1960s mark the third revolution, the so-called digital revolution, which started with the development of semiconductors but also entails the internet coming into existence in the 1990s (Schwab, 2016). The fourth revolution, that is currently taking place, builds on the digital revolution and is marked by among others mobile internet, sensors that are small, powerful and cheap, artificial intelligence (AI) and machine learning. Digital technologies are being integrated into daily life and transform global industries (Schwab, 2016), including the humanitarian sector (Sandvik & Raymond, 2017).

The positioning of the humanitarian sector towards digital technology has been receptive. The event commonly seen as the turning point, is the 2010 Haiti earthquake. From then, aid providers have adopted technology on a large scale (Sandvik & Raymond, 2017; Sandvik, Jumbert, Karlsrud, & Kaufmann, 2014). This development of digital innovation in the aid sector can be approached from different angles. From a growing body of literature seems that technology can be seen as a tool to enhance efficiency, effectiveness and accountability (Dette, 2016; Sandvik et al., 2014). Dette indicates four technological tools that have been widely used throughout the humanitarian sector:

- Phone-based feedback mechanisms and two-way communication; - Digital data collection with smartphones and tablet computers;

- Remote sensing with satellites or unmanned aerial vehicles and location tracking (GPS);

- Broadcasting and community production of radio shows (Dette, 2016).

In the enthusiasm about the possibilities of technologies, the risks and unintended consequences, that can be detrimental or even lethal, are sometimes overseen. Therefore, some academics argue that the humanitarian sector should instead approach technology from

(17)

17 a framework with a greater focus on the potential harm (Dette, 2016; Sandvik & Raymond, 2017; Sandvik et al., 2014). The potential harm can be captured in eight challenges:

1. Mishaps and mistakes

This refers to the problems that arise following programme design or implementation that are not devised well. Examples are a project with mobile vouchers in areas with low connectivity and a project distributing tablets among field staff in areas without (sufficient) plugs to charge the devices (Dette, 2016).

2. Negative impact of new devices

Technology driven interventions may fail to have a positive impact. Despite the popularity of crowd-seeded initiatives such Voix des Kivus, Pham and Vinck, that information provided by communities through SMS is often not used in the actual response. Moreover, it is not only the absence of a positive impact, there is the possibility of direct and indirect negative effects of technology. An example is of the use of incomplete or low-quality data that effects the credibility and accuracy of the policy and decisions based on this data. A fundamental problem as discussed by the scholars Latonero and Gold, is that we cannot know all the positive and negative impacts of new technologies. This results into difficulties in making informed decisions (Sandvik & Raymond, 2017). Other negative effects could be for example in the form of governments restricting people’s ability to communicate, or even increased violence against the people that the technology aimed to protect. An example of this is the Eyes on Darfur project of Amnesty International in 2007/2008 which monitored violence in villages at risk of attack. The violence of the government against the people in these villages during and in the subsequent years of the programme increased. The collected data in combination with qualitative data showed that this increase was a retaliation for the programme of Amnesty (Sandvik & Raymond, 2017).

Also not all staff of aid providers welcomes new technology. Firstly, training on how to uit requires time. Secondly, it can limit the working methods. Think about digital survey forms and reporting platforms with word limits that restrict freedom in answering. Also, technology can create a new power balance within organizations. Think about data in central databases that are difficult to access for field staff with limited connectivity (Dette, 2016).

Lastly, technology can come with new gender bias. Think about the tendency of radio to cater more to male than female listeners, and more men than women who keep the phone within a household (Dette, 2016).

(18)

18 3. Loss of trust and reputation

People affected are not always receptive of new technologies. An example is that local staff that use phones are sometimes being perceived by communities as less friendly. Moreover, in the context of conflict, armed groups could be more suspicious or even ban aid providers that use technologies such as phones or aerial imagery. Potentially, also the quality of survey responses and information obtained through digital devices decreases (Dette, 2016).

4. Digital vulnerabilities and digital harm

There is an absence of collective awareness and preparedness in the humanitarian sector about the risks related to the use of technology. Digital attacks can cause substantial harm. One because aid providers are more vulnerable because of the increased dependency on connectivity, two because there is more information that can be used against people or passed on to third parties. Common ethical, technical, and rights-based standards to address this do not exist. This is a result of a lack of willingness (following the fear to lose face when releasing incidents), an underestimation of the issue, and a lack of experience and knowledge about how to evaluate the potential harm (Dette, 2016; Sandvik & Raymond, 2017).

5. Privacy risks and irresponsible digital data

Digital attackers can re-identify people by combining the growing numbers of datasets with personal data of communities or individual (Dette, 2016).

6. Increasing inequality and power imbalances

Technology is not neutral, nor does it eliminate institutional or political power. Moreover, technology in itself should be seen as an institutional power. It can ‘deepening the processes of creating inequalities’ as actors with power introduce new technologies that enhance their own power position and undermine the engagement of the actors without power. New technologies are not passively adopted by societies. Rather, there is a mutually constitutive relationship between technology and society (Dette, 2016; Sandvik & Raymond, 2017; Sandvik et al, 2014).

7. Dependence on non-humanitarian actors and sectors

With the implementation of technology, tech-companies and other private actors get involved in the humanitarian sector. Unlike humanitarian actors, they are not committed to the

(19)

19 humanitarian principles4_{, but the dependency on them increases. This can compromise the}

values and objectives of aid providers (Dette, 2016).

8. Double-standards and hypocrisy compromise humanitarian principles

Recently, European policymakers have been put to a halt in their plans to use advanced biometrics in citizen registration by civil unrest on the topic. At the same time, biometrics are encouraged for refugee registration. Examples like these raise questions about double-standards and stress the need for safeguarding ethical double-standards for technology in the humanitarian sector (Dette, 2016).

Hypocrisy is also seen in the phenomenon of disaster experimentation. In his paper ‘Ebola: A Big Data Disaster’, McDonald writes the following about this:

‘The chaos of humanitarian disaster often creates an implied social license for experimentation with new approaches, under the assumption of better outcomes. Vested interests dominate the public discussion of humanitarian data modelling, downplaying the dangers of what is essentially a public experiment to combine mobile network data and social engineering algorithms. In the case of using mobile network data to track or respond to Ebola, the approaches are so new–and generally so illegal– that most advocacy focuses on securing basic access to data. Advocates for the release of CDRs often paint an optimistic picture of its potential benefits, without applying the same rigor to the risks or likelihood of harm. This trades on the social license created by disaster to experiment with the lives of those affected, under the implicit assumption that it can’t make the situation worse’ (Sandvik & Raymond, 2017). Following this statement, Sandvik and Raymond conclude that the current trend of disaster experimentation can be seen as ‘fundamentally treating highly vulnerable populations affected by extreme crisis events as experimental subjects of largely untested, non-consented, and remotely applied technological intervention’ (Sandvik & Raymond, 2017).

3.2. Technology in humanitarian Cash Transfer Programming

The use of technology in CTP has great potential, but there are also risks. This section goes into both sides of the narrative.

The first argument in favour of the use of technology in the context of humanitarian CTP is efficiency and cost-effectiveness. With so-called ‘pull’ methods, the people affected are ‘pulled’

4_{The humanitarian principles of Humanity, Neutrality, Impartiality and Independence are derived from} the seven Fundamental Principles of the Red Cross. The four principles have been captured in multiple crucial documents, such as the Geneva Conventions that form the basis of International Humanitarian Law. Currently approximately 500 humanitarian organizations have committed themselves to the humanitarian principles through signing the Code of Conduct for theInternational Red Cross and Red Crescent Movement, and non-governmental organizations in disaster relief

(20)

20 to a pay-point such as a government office, bank or post office. With a ‘push’ mechanism, the transfer is ‘pushed’ electronically to the people affected, for example to a smart card or mobile phone. Push mechanisms can be advantageous because there is no more need for the presence of staff at a pay-point, nor for transport, security, and insurance for physical movement of the cash. In addition, with push methods, the risk of leakages through staff fraud is seen as significantly reduced (Devereux & Vincent, 2010).

Secondly, technology has the potential to increase flexibility for the people affected. Delivery through ATMs or mobile phones eliminates partly or fully the need to be at a certain day and time at a certain location. Consequently, the time and costs of transport and queuing can be (partly) eliminated as well (Devereux & Vincent, 2010).

The potential for (financial) inclusion is the third argument for the use of technology. Push mechanisms as described above could introduce people to financial infrastructure (such as (mobile) banking services, ATMs, etc.) that formerly were not accessible to them. Research shows that this access has a positive effect on the economic equality and growth of a country. It can also increase security. For example, after the implementation of M-Pesa in Kenya, carrying substantial amounts of cash – formerly necessary in some professions such as in the taxi-branch – could be avoided. The positive effect of financial inclusion could be beyond the target group, especially with programmes in rural communities in isolated areas. The increased demand for financial infrastructure following cash programmes from certain areas could be an incentive for FSPs to expand their activities to these places (Devereux & Vincent, 2010).

Not only could the increased access to (financial) infrastructure be a gain for the users, for governments this is an opportunity as well. It could be a way to streamline the delivery of a range of commodities and services to citizens. Smart cards can be used to receive cash transfers, but also for other services such as recording a person’s medical history or a voting record (Devereux & Vincent, 2010).

There is an ongoing discussion about the digital divide, and whether the Global South is left behind in the current developments fuelled by technology. Some research points out that this is not the case, and that new technologies reach people that are actually living in places with the least developed infrastructure. These people skip the step of using conventional landlines and go from non-technology means of communication, straight to satellite-based telecommunication systems. An example that illustrate this is the rapid adoption of M-PESA in Kenya. Within three months after the launch of the service, 111,000 people registered and transferred a total value of 6 million USD. This included many people who never used a mobile phone before at all (Devereux & Vincent, 2010). A pilot in Lesotho with people in vulnerable positions (older people, people with disabilities, women, poor farmers, etc.) shows that even

(21)

21 illiterate people without any former experience with technology are able to learn how to use phones. As reported by community leaders, the newly learned skills empowered and increased confidence among the users (Devereux & Vincent, 2010).

To determine whether to implement a technology-based programme, one should look at the size of the programme: both in terms of the number of people of the target group and the duration of the programme. It is expensive and inefficient to set up a programme based on advanced technology for a target group of only a few hundred or thousand people and/or for a short period of time. Setting-up a digital database is quite an investment which leads to high costs per person / transfer. Looking at a larger scale, marginal costs decrease rapidly when the target group becomes larger (Devereux & Vincent, 2010).

A final benefit is that technology incorporation could entail aspects that make it more difficult to commit fraud than with manual systems. Decreased fraud could for example follow from stringent bank reconciliation procedures and verification methods that use biometrics (Devereux & Vincent, 2010).

On the other side of the coin, there are the risks and challenges that come along with the use of technology in humanitarian CTP. A first question that raises is the question about the roles in managing data. Humanitarian programmes are often a collaboration between different actors such as NGO’s and the governmental institutions, and there are always donors involved. Who owns and controls the data, who can access sensitive PII and who is responsible for ensuring data protection? This is an important question because data such as ethnicity could be used against people by governments or other parties. Think about systematic exclusion or retribution, especially in the context of (ethnic) conflict. In addition, data could be misused by politicians to gain power. In Malawi for example, politicians used databases to target potential recipients in order to be able to make promises about old-age pension, and herewith to increase their popularity in the elections (Devereux & Vincent, 2010).

Another risk comes along with the monopoly of a certain technology provider. Difficulties will arise when the actor with the monopoly, for some reason, can or does not continue with the provision of the technology. Dependency on this single actor will lead to a temporary or even permanent shut down of the cash programme, leaving the people affected without assistance (Devereux & Vincent, 2010).

3.3. Blockchain in the non-profit sector

Blockchain has been described as ‘one of the most fundamental inventions in the history of computer science’ and is ascribed the potential to ‘radically transform the way in which we live and transact with each other’ (Hughes, 2017). There are opportunities in the form of saving costs and enhancing efficiency within international payment systems, insurance policies, and

(22)

22 risk management procedures (Kewell, Adams, & Parry, 2017; Zwitter & Boisse–Despiaux, 2018). However, the technology is not only of interest for the profit sector. Actors in development work and humanitarian action are exploring the options of the adaption of blockchain as well (Hughes, 2017; Kewell et al., 2017; Van den Berg, 2018; Zwitter & Boisse-Despiaux, 2018).

There are different purposes of blockchain that are being explored or already deployed in the sector of development and humanitarian action. Firstly, the technology is used to increase transparency, reliability and ease in the supply chain in the humanitarian sector. Greater transparency can be about where and how materials and products are extracted or made, and about whether there is no conflict, child labour or other ‘unethical’ activity in the process. This can help fight corruption, a significant challenge in the aid sector. Hughes (2017) refers to blockchain as ‘a new cyberweapon to the fight against corruption and other economic crimes and holding offending sates to account’. Greater transparency can also be about the financial supply chain. Blockchain has the potential to enable donors and organizations to trace funding and prevent abuse or misallocation of funds from happening (Van den Berg, 2018; Zwitter & Boisse–Despiaux, 2018)

Another application is in ownership and tenure rights and registration. The rent or ownership of land and housing is often the most important asset of people living in poverty in the Global South. In India for example, landlessness is – even more than caste or illiteracy - one of the main reasons for poverty. Documents that prove ownership are therefore of great importance. Nonetheless, reality learns that written documents are not always existing in the first place, and can be easily stolen, misplaced, destroyed, or manipulated. Using blockchain, registration will be permanent, transparent, corruption-proof, faster and cheaper (Van den Berg, 2018; Zwitter & Boisse-Despiaux, 2018).

A third application of blockchain is in financial inclusion. Globally, around two billion people do not have access to formal financial services. This is a result of both the lack of financial infrastructure in developing countries and that commercial banks see people with very limited means as unprofitable (Van den Berg, 2018). Blockchain has the potential to speed up and decrease costs of money transfers, making transfers of small amounts economically viable (Kewell et al., 2017). Besides this, World Bank reported that in 2016 the value of remittances sent to developing countries was as high as 430 billion dollars: three times the amount that has been sent to these countries as Official Development Aid. The costs of these transactions are high. 7.4% transaction fee is the global average for the transaction of a remittance and the percentage is even higher for regions such as sub-Saharan Africa. Some see blockchain-based services as a solution: with the design of a free app that enables people to act as their

(23)

23 own bank and cryptocurrency to transfer of remittances. There are already several start-ups working to make payments to and from Africa more affordable. Examples are BitPesa (working with blockchain and bitcoin) and SureRemit (working with their own cryptocurrency) (Van den Berg, 2018).

The possibility to ‘colour’ coins is another interesting application of blockchain technology. Colouring means that specific attributes that represent morals, principles and ethics, are added to coins and payments done with the coloured coins should be in line with these values. The idea is already deployed. CarbonCoin is an energy efficient digital currency that plants trees. And there exists a cryptocurrency with Muslim values (Kewell et al., 2017). Looking at the humanitarian sector, this could be translated into earmarked donations or money streams that can only be spent in the pre-assigned projects, and/or for pre-selected goods and services (Zwitter & Boisse-Despiaux, 2018).

A digital identity backed by blockchain is another useful instrument that one is starting to explore. Approximately 1.8 billion people globally do not have formal proof of their identity. As a result, these people often do not only have difficulties accessing basic services such as education and healthcare, but also, they cannot participate in elections and in the formal labour market. One of the problems in the current system is that people are depending on central authorities for identity management. Blockchain has the potential to dismantle this top-down structure and decrease or eliminate reliance on a single actor. Besides changing the nature of existing identity systems, the technology could also help giving vulnerable groups of people (e.g. refugees) access to documented proof of identity (Chen, 2018; Kewell et al., 2017). There are different initiatives of actors working with blockchain for digital identity systems. ID2020 is a public-private partnership that works towards increased access to identity. They developed a blockchain based identity management system working with biometrics, aiming to ensure immutable records of identity for stateless and undocumented people (Chen, 2018). Other examples are a digital identity system for a voucher programme developed by AID:Tech for a refugee camp in Jordan and a project of UN Women using Blockhain to give women access to micro-loans and facilitate safe money transfers among female entrepreneurs (Zwitter & Boisse-Despiaux, 2018).

In light of digital identity systems, it is interesting to have a look at the ‘identity problem’. Whether there is blockchain involved or not, it is of importance to realize that in the current paradigm, it is assumed that the absence of formal identity papers is a problem that needs to be solved. As Ismaev refers to as the ‘moral issue of a persistent identification creep’. The assumption that identification is a cornerstone for systems leads to the design of more systems that require identification from individuals. Herewith, the need for the ‘solution’ is enforced. The

(24)

24 identity problem is created and measured along the parameters of systems and its efficiency rather than the actual needs of the individuals who are interacting with it (Ismaev, 2019).

There is a clear link between identity systems and electoral processes. Incomplete, inaccurate or outdated birth- and death registration does not only mean exclusion of elections for certain people, but on the other hand it also means that there is more space for ghost voters and ‘rogue voters’5_{. All-together, when identity systems are sub-optimal, elections can easily be}

manipulated. Unfair elections can cause tension, violence and even conflict (Hughes, 2017).

Finally, the development of smart contracts backed by blockchain is a topic of interest. Smart contracts are ‘computer protocols that facilitate, verify, and enforce the performance of a contract: self-executing code’ (Kewell et al., 2017). This means that it is a decentralized, automatic contract that - by using a pre-designed set of conditions that need to be met - can make the decision whether to execute or not, without the interference of a central authority. Not only does this eliminate the reliance on a subjective third party, this also increases speed and efficiency (Kewell et al., 2017). A suggested application of smart contracts is combining forecast-based financing with contracts between funding- and implementing partners. Objective indicators to forecast a crisis could be used in smart contracts to activate funding for quick (preventative) action, and the allocation of funds increasingly based on needs rather than political willingness. Another area of application is in supply chain management (Zwitter & Boisse–Despiaux, 2018).

Risks of blockchain

The potential of blockchain for the humanitarian and development is promising, but there are also risks.

Looking at blockchain to increase transparency in supply chains, there are some notes to be made. Transparency about the origins and journey of materials and products is only valuable when the data is truthful. The problem is mostly false data entry, not tampering with data retroactively. Using blockchain does not take away that data input is depending on humans, and therefore is possibly subject to corruption and fraud. In addition, in situations of conflict and corruption, powerful actors will most likely be hesitant to adopt blockchain. And, even when blockchain is used, it cannot be assumed that data input is truthful. A possible scenario is that blockchain will be widely adopted in supply chains, except for by those who have something to hide. This could lead to avoidance of the latter. On the long term, this could mean that the process will be improved or replaced by better practices. On the short term however, this will

5_{Rough voters are non-citizens, living in neighbouring countries, that participate in an election. This} does for example occur when ethnic groups have been split up in different countries after colonization and try to support their leaders in all countries.

(25)

25 hurt the people that are not responsible for the bad practices, but just work for them to make a living. These people are often already in a vulnerable position and will then lose their livelihood as well (Van den Berg, 2018). Next to this, in for example a context of political conflict, full transparency is not always desirable. If the donors of an aid provider are publicly known, parties to the conflict could accuse the aid provider of having a double agenda. An example is the 2003 Iraq war, where Western NGOs and the UN who were financed by the American government faced repercussions because they were perceived as puppets of the United States (Zwitter & Boisse–Despiaux, 2018). Also, as a result of more transparency in the financial supply chain, operational freedom could be restricted because there is no more space for using financial means in negotiating for access to the most vulnerable (Zwitter & Boisse–Despiaux, 2018).

A limitation of blockchain for ownership and tenure registration is that it does not fully address the cause of conflict over control. More than inadequate registration, it is a political problem. Even when ownership is clear, conflicts over land and other property continue because there is a discrepancy between official records and who lives on the land. Whether blockchain is a useful tool depends on the presence of a government that can be trusted and is able to back-up the ownership rights. In the worst-case scenario, government collaboration could work out the other way and enhance favouritism (Van den Berg, 2018).

There are multiple challenges when looking at the use of blockchain to enhance financial inclusion. Smartphones and internet are not accessible for everyone. In August 2014, McKinsey reported that globally, nearly 4.5 billion people - more than half of the global population - did not have regular access to the internet. Moreover, in the least developed countries only one out of ten people was online, 3.5 billion (of the 4.5 billion) people without access were living in just twenty countries, one billion of these people were illiterate, and between 1.1 and 2.8 billion people could not access internet because of a lack of network coverage in the area they were living (Hughes, 2017). Another concern is that cryptocurrencies are not universally accepted as currency and most of the cryptocurrencies are highly subject to fluctuations, which adds a significant risk for the people who receive aid in cryptocurrency (Van den Berg, 2018). Thirdly, there are political considerations when it comes to financial inclusion through blockchain. In general, governments are hesitant about cryptocurrencies. There is not yet a legal framework that regulates the use of it and in combination with the pseudonymous nature, cryptocurrencies can easily be used by criminals outside of the control of governments (Van den Berg, 2018).

A doubtful feature of blockchain in general is the great amount of electricity that the technology in its original form requires. According an estimate of Aste in 2016, every second, a billion

(26)

26 watts of electricity is consumed globally to produce Proof-of-Work, and this is only for the currency Bitcoin. Though initiatives to reduce energy exist, the technology always requires servers and computers for the procession of transactions. The latter means that blockchain cannot work in its optimal form in countries with irregular access to energy and network (Kewell et al., 2017; Van den Berg, 2018; Zwitter & Boisse–Despiaux, 2018).

Next, there are questions about data protection when using blockchain, especially where PII is involved. The question raises whether a decentralized system such as blockchain is suitable to register such sensitive data. There are potential solutions to address privacy concerns, such as zero-knowledge proofs6_{, but how realistic the actual implementation of such solutions is, is}

not yet clear. Looking at blockchain in light of the GDPR raises other questions about data ownership, prevention of the misuse of data and the right to be forgotten. Immutability can prevent actors from tampering with data, but at the same time it impedes the right of the data subject to change or delete his or her data. Another undesirable effect is that mistakes and wrong transactions cannot be deleted from the chain (Zwitter & Boisse–Despiaux, 2018).

A pitfall of the current debate about blockchain is the idea that blockchain per definition means decentralization, transparency, and immutability. One should understand that these are outcomes of intentional choices in the design of blockchain-based system. There are for example permissioned and permissionless blockchains. Permissioned blockchains can be more efficient and faster, but at the cost of reduced security, immutability and censorship-resistance (Kewell et al., 2018; Zwitter & Boisse–Despiaux, 2018).

Related to the need for design decisions are the concerns around the absence of a legal framework and a general lack of coordination and direction of the development of blockchain in the non-profit sector. To avoid legal and ethical problems and to reach Blockchain its full potential, one need to ensure compliance with among others international humanitarian law, human rights law and the humanitarian principles7_{(Chen, 2018; Hughes, 2017; Zwitter &}

Boisse–Despiaux, 2018).

6_{With the encryption scheme Zero-Knowledge Proof data can be verified without disclosing the actual} data. In transactions there are ‘verifiers’ and ‘provers’. The prover aims to prove that something is true to the verifier without revealing any additional information about this thing: the focus is only on the final output. The input or calculations remain unknown for the verifier (Ray, n.d.)

7_{The humanitarian principles of Humanity, Neutrality, Impartiality and Independence are derived from} the seven Fundamental Principles of the Red Cross. The four principles have been captured in multiple crucial documents, such as the Geneva Conventions that form the basis of International Humanitarian Law. Currently approximately 500 humanitarian organizations have committed themselves to the humanitarian principles through signing the Code of Conduct for the International Red Cross and Red

Crescent Movement, and non-governmental organizations in disaster relief (Anggraeni 2012; European

(27)

27

Another interesting note is the argument of Atzori, who states that ‘computational technologies are no substitute for human endeavour’, referring mainly to the matters of policy, reason and conscience. This is not an argument against the use of blockchain, but it is an important question to keep asking ourselves: What can and should be the role of blockchain vis-à-vis humans? (Hughes, 2017) Looking at smart contracts for example, the automatization could lead to inflexible processes without the human margin of appreciation that is crucial in the tensed setting of humanitarian action (Zwitter & Boisse–Despiaux, 2018).

In line with the discussion about technology versus human input, is the argument of Ishmaev (2019) about the important distinction between the moral and technical concept of blockchain backed SSI. The moral desirability of the implementation of a SSI system depends upon the recognition of this gap and the capacity to bridge it (Ishmaev, 2019). Looking at the moral or normative concept of SSI, it is interesting to have a look at the ten normative principles of Allen (2017). These principles can be categorized as tangible aspects (‘Systems and algorithms must be transparent’ and ‘Information and services about identity must be transportable’); general, governance aspects (‘The rights of users must be protected’, ‘Users must have access to their own data’ and ‘Disclosure of claims must be minimized’); and ethical aspects (‘Users must agree to the use of their identity’, ‘Identities must be long-lived’ – under the condition that the right to be forgotten is upheld, and ‘Users must control their identities’)(Allen, 2017; Ishmaev, 2017). Ishmaev refers to the first principle of Allen as most significant:

‘Existence – Users must have an independent existence

Any self-sovereign identity is ultimately based on the ineffable ‘I’ that’s at the heart of identity. It can never exist wholly in digital form. This must be the kernel of self that is upheld and supported. A self-sovereign identity simply makes public and accessible some limited aspects of the ‘I’ that already exists.’ (Allen, 2017)

On the other hand, there is the technical term of SSI, which refers to the broad family of technical standards and solutions (Ishmaev, 2017).

3.4. Hypotheses

In this section hypotheses are formed based on the literature. Figure two shows the problem statement, sub questions and hypotheses and how they are linked to each other.

(28)

28 Problem statement – The existence of a

difference between the point of view of the people affected and the aid provider surrounding the use of digital identities in

humanitarian CTP

Sub question 1 – What are positive aspects of the incorporation of digital identities in CTP for the people affected? Sub question 2 – What are negative

aspects of the incorporation of digital identities in CTP for the people affected? Sub question 3 – What are positive aspects of the incorporation of digital identities in CTP for the aid provider? Sub question 4 – What are negative

aspects of the incorporation of digital identities in

CTP for the aid provider?

Hypothesis 2 – Negative aspects can be found in the

consequences of mishaps/mistakes, inadequate data protection; aid could be delayed or even ceased in case of technical dysfunctionalities; existing inequality can increase; people affected can become subjects of

unethical or untested practices. Hypothesis 1 –

Positive aspects are that digital

identity could increase access to other (financial) services, it could provide more flexibility in how to receive the transfer

and it can empower.

Hypothesis 3 – Positive aspects are the arguments

of increased efficiency, cost-effectiveness and accountability. In addition, working with digital identities reduces the vulnerability of the programme to fraud. Hypothesis 4 – Negative aspects can be found in the

consequences of mishaps/mistakes;

discomfort of field staff; reputational risks; difficulties with data protection

and the compromising of

humanitarian principles.

(29)

29

4. Methodology

In this chapter follows first information about the design and methods of this research. Next, the ethical considerations and the limitations of the methods are discussed.

Topic selection

This research was written in collaboration with 510, an initiative of the Netherlands Red Cross. The mission of 510 is ‘to shape the future of humanitarian aid by converting data into understanding, put it in the hands of aid workers, decisionmakers and people affected, so that they can better prepare for and cope with disasters and crises’. One of the projects that 510 works on is 121, and end-to-end CTP system. Together with Tykn.tech, a tech start-up focusing on digital identities and blockchain, they look into the potential of blockchain to improve cash-based assistance by developing a system with a digital wallet that can be used to transfer money to people affected. In the scope of this project, 510 is collaborating with graduate students. The author of this paper approached 510 based on her interest in the research topics on their website. Initially, the research would be about the ethical considerations around the use of digital identities in humanitarian cash programmes. In consultation with Joachim Ramakers, who works for 510 and supervised this thesis, the topic was narrowed down to the current research focus.

Research strategy

Digital identity is a topic that only recently became of relevance in the humanitarian sector. There is only a limited amount of (academic) research available about this topic and the problem is not yet well-defined. Therefore, this research falls into the category of exploratory research (McCombes, 2019).

This research is qualitative. The focus is on details and wording, rather than on the quantity of data and generating general statements - as the case is with quantitative research (Bryman, 2012). Qualitative research is predominantly inductive: it starts with in-detail information, which leads to the generation of a theory. Secondly, qualitative methods have interpretivism as epistemological orientation, which means the emphasis is on the ways in which individual people interpret their social world. Another feature of qualitative research is constructionism, because it approaches social reality as an ever-developing result of people’s creation (Bryman, 2012). Qualitative methods are the proper choice for this research, because the aim is to gain more insight into the thoughts about a specific component (namely, a digital identity) in people’s lives or work. In order to really understand someone’s point of view, there is great value in detailed information such as quotes. Every individual has his or her own opinion. This

Digital identity in the interest of all: