Can Blockchain Help Improve Financial Inclusion? A Comparative Study

Abstract

The financial inclusion of poor populations and/or refugees can be hindered by difficulties in proving or reluctance to disclose their identity. Bank accounts and mobile money services require that identities be provided. Financial digital services based on blockchain technology can provide anonymous authentication to poor/refugee populations and be a first step towards financial inclusion. We scrutinize several examples of such projects by comparing them with blockchain-based digital identity or financial inclusion programs that are not necessarily restricted to poor/migrant populations. We use social network activity as a proxy for the failure or success of such projects. We find that blockchain projects targeted to migrants and poor individuals are more likely to fail than are those targeted to all. We more closely examine one particular case to check the consistency of our proxy. We present plausible explanations for our result: the discrepancy between the needs of populations of low socioeconomic status and the proposed blockchain-backed financial services and the fact that maintaining such services is energy intensive.

Keywords:

• blockchain
• financial inclusion
• refugees
• Finland

JEL Classification Codes:

• F22
• F24
• E42
• O15
• O19
• I38
• G53

Notes

1 There are relations among the notions of money, access to credit, and record-keeping. See Narayana Kocherlakota (1998) and William Luther and Josiah Olson (2015).

2 Although they differ in some aspects (in particular, a refugee may not necessarily be poor), in this article, we treat populations as having a common characteristic: being financially excluded notably because of identity problems.

2 As stated in a previous note (), as currencies are anonymous (as has traditionally been said, goods and services must be delivered to the bearer, regardless of who he or she is, of currencies on demand), it is difficult to signal a “good” track record of credit and transactions payments to a third party when using cash only. In other words, it is difficult to prove “who you are” in financial terms.

4 Mark Latonero et al. (2019) provided an insightful field study with migrants’ testimonials. Some migrants noted that they fear being tracked and sent back to their home countries. Moreover, the field interviews in the aforementioned study indicated that migrants also care about privacy and exhibit a certain mistrust about official organizations, for instance, “when you arrive, all you have is your name, surname… You have to give it everywhere. [You] don’t know what happens. What are they asking for it?” (Latonero et al. 2019, 5).

5 Initially, there are two categories, but as the data description table shows, there are not enough projects in each category to perform robust statistical tests. We then merge these projects and call their target “migrants-poor” and, sometimes, “migrants” for short.

6 Results for Facebook present qualitatively similar results and are available as Supplementary Information upon request from authors.

7 Among those projects targeting migrants and poor individuals, three are tagged as digital identity and five are tagged as financial inclusion, and among those projects targeting “all”,” nine are tagged as digital identity and thirteen are tagged as financial inclusion, which makes the odds of a project being tagged as these two factors quite similar.

8 Interestingly, both targets show a similar history: their average “age” is approximately sixty-seven months, with similar medians, dating the average mushrooming of blockchain projects back to around early 2017.

9 In unreported results, we first confirm that the observations are not normally distributed, and thus, we undertake nonparametric tests.

10 In Table A.1, we see that the Twitter page, started in 2014, shows no news since early May of 2017. Moreover, 1,956 days had elapsed until the day of access in September 2022.

11 Further inquiry should be conducted to understand what happened, most notably by translating Finnish documents. We use only documents originally available in English.

12 However, we observe only one case in which coding could become debatable: the Kora project. At its inception in 2017, this project emphasized helping “poor” individuals. Five years later, in 2022, it touts that it helps “businesses,” which would entail an “all” type of coding. However, we code this project as helping “poor” individuals. This choice turns out to not support our hypothesis, as this project has a very active profile. Nevertheless, this situation does not qualitatively change our result. In fact, coding this project as “all” would even strengthen our conclusion.

Additional information

Notes on contributors

Sébastien Galanti

Sébastien Galanti is at the University of Orléans, France.

Çiğdem Yilmaz Özsoy

Çiğdem Yilmaz Özsoy is at the University of Istanbul Ayvansaray, Turkey.