In 2016, we established the Digital Financial Services Research Group at University of
Washington to investigate the technological challenges in improving access to financial
services for the world’s poor and to promote a global research community at the intersection
of Computing for Development and Financial Services. The research approach was to focus on
specific barriers identified by the financial services community and to develop and pilot
solutions. In this talk, I will present one specific project: An assessment of SMS fraud in
Pakistan. There is the concern that widespread scams initiated by SMS have a negative impact
on the low income population that are the target of financial inclusion initiatives. To investigate
this we collected fraudulent SMS messages in Pakistan through various means and conducted
a range of consumer and stakeholder interviews. Based on our analysis, SMS fraud is
dominated by lottery type schemes, which have the greatest impact on vulnerable, low-income
rural populations. To detect SMS fraud, we offer a simple heuristic for fraud detection that has
a high accuracy rate and is adaptable to evolving fraud schemes. We develop a
recommendation of a fraud mitigation strategy based on targeting fraudster call back
numbers.