Improving transparency, increasing competition, and addressing gender disparities are key to reducing fraud and strengthening trust in mobile money markets.
Editor’s note: This episode of VoxDevTalks is also available on Spotify, Apple Podcasts, and YouTube.
In this two-part VoxDevTalks special, Tim Phillips speaks with Francis Annan about his extensive research into mobile money markets in Ghana. The focus of this episode—the first of two—is on understanding and reducing fraud and misconduct in rural financial systems. Annan shares findings from three pivotal studies that investigate the mechanisms of overcharging, market dynamics, and gender discrimination in the digital finance sector.
“Mobile money… is really about improving the financial lives of people, especially the poor.”
In areas lacking traditional banking infrastructure, mobile money—facilitated by local vendors or ‘human ATMs’—has become a cornerstone of economic participation and resilience.
Why mobile money matters in Ghana
Annan opens with an exploration of why mobile money is so essential to economic development in Ghana, particularly in rural settings. Mobile money offers a lifeline to financial services: savings, low-cost remittances, and shock mitigation.
“It has the ability to lift people out of poverty… to remit money in a low-cost way… to build the financial and economic resilience of people in this market.”
These services are delivered by ‘human ATMs’—local agents who act on behalf of telecom providers to allow deposits, withdrawals, and other transactions. Their role is central to delivering financial services in villages where formal banking institutions are absent.
Understanding and addressing overcharging
One of Annan’s key findings relates to the rampant overcharging by these agents. Although fees are set by mobile providers, agents frequently exceed these official tariffs—especially in villages, where regulatory oversight is minimal and price transparency is poor.
“It’s remarkably widespread. One in four, one in five transactions are overcharged.”
Consumers often lack the knowledge or alternatives to respond to overcharging. Many are unaware they are being overcharged, or they have no other vendor to turn to. This means agents often weigh the low risk of getting caught against the potential gains, making cheating more likely.
A two-sided solution: Information interventions
To address this, Annan conducted an experimental intervention that improved pricing knowledge for both vendors and consumers. This ‘two-sided information programme’ was simple yet powerful: consumers were told the official prices, and vendors were informed that consumers now knew those prices.
Annan found that agents in villages where the intervention took place were around 72% less likely to engage in cheating. The programme also helped align consumer perceptions with reality—many had assumed overcharging occurred 60% of the time, though the actual rate was closer to 20–25%.
“You do not only reduce the level of misconduct… but you also actually make consumers better calibrated.”
These low-cost interventions created ripple effects across the market. Even untreated agents reduced overcharging—by about half the rate of treated ones—due to peer pressure and fear of reputational loss.
Can more vendors reduce misconduct?
The second study tackled the problem from a structural angle: limited competition. In many villages, a lack of vendors gave existing agents market power, which encouraged fraud. Annan collaborated with a major provider to introduce new agents randomly in selected villages.
This created a natural experiment to test whether increased competition reduced misconduct. The results were striking:
“Incumbents… were less likely to commit misconduct… there was some pressure on incumbents to promote consumer protection practices in the marketplace.”
Additionally, Annan discovered that new agents who operated non-financial businesses—like selling rice—were more honest. This bundling of services created higher reputational stakes.
“If you get hurt in the market, you don’t only suffer your services for mobile money, but you also suffer your services for rice, if rice is what you sell.”
Interestingly, these bundled stores also saw increased non-financial revenues after becoming agents, revealing positive spillover effects in local economies.
Gender and misconduct
The third and final study investigated the role of gender in mobile money fraud. Using a unique audit study design, Annan recruited trained ‘secret shoppers’ of different genders to visit vendors and record their experiences.
“Female auditors were massively discriminated against. They were much more likely to be cheated… the cheating is more so when female customers visit a female vendor.”
This result surprised even Annan. Contrary to conventional wisdom, female vendors were more likely to overcharge, and female customers suffered disproportionately. In contrast, male-to-male interactions showed evidence of ‘within-gender favouritism’.
He identifies three explanations:
- Misbeliefs about vendor behaviour: People wrongly assumed female vendors were more honest, which allowed them to exploit the perception.
- Gender empowerment gaps: In villages with low female empowerment, women faced greater risk of exploitation.
- Economic pressure on women: Female vendors in lower-income households may rely more heavily on misconduct for supplemental income.
These structural inequalities have serious implications:
“The notion of social distance—either through beliefs structure, empowerment structure, or even income—could act as an important source for misconduct in the market in this discriminatory way that we think can undermine the functioning of financial markets.”
Lessons for financial inclusion policy
Across these studies, a consistent theme emerges: effective financial inclusion must tackle both human and institutional factors. Information, competition, and social norms each play a role in shaping outcomes—digital infrastructure alone is not enough.
“You’re going to need a trustworthy and more accountable financial marketplace, which unfortunately has not been the case.”
Annan’s interventions provide a roadmap for improving digital finance ecosystems, particularly in vulnerable communities. By enhancing transparency, introducing market competition, and understanding social dynamics, mobile money systems can become more inclusive and resilient.
“Mobile money, the market structure is super interesting. It’s a platform to do really good economics, studying questions within the marketplace when it comes to financial services.”
References
Annan, F (2022), “Gender and financial misconduct: A field experiment on mobile money,” Unpublished manuscript.
Annan, F (2025), “Misconduct and reputation under imperfect information,” Journal of Political Economy, 133(5): 1460–1496.
Annan, F (2025), “Randomized entry: The equilibrium effects of entry in digital financial markets,” Unpublished manuscript.