The Strange Case of Low Financial Inclusion Using Digital Payment Services in Mexico

Ivonne Acevedo
Miguel Székely
July 28, 2021


Mexico followed, in past years, what appeared to be a textbook formula for expanding access to and use of digital financial services for its citizens. And yet, less than half of its adult population reported having a bank account only two years ago, which is lower than the Latin American average of 55.1 percent, and significantly below the upper-middle-income country average of 73.1. Inclusion for digital financial services is scanter: only 32 percent of adults were users by 2017, and in the same year, 5.5 percent were users of mobile money accounts according to the most recent data available.

Access to digital financial services is important; it provides a lower cost platform for accessing financing for generating economic activity, building up assets, reducing vulnerability to shocks and expanding consumption alternatives, among others. Thus, low levels of inclusion mean smaller chances of creating opportunities and improving well-being.

Analytical framework

In a recent paper, we explore the reasons why digital payment services are used to a much lower extent than would be expected in Mexico, applying a new methodology that uses the analogy of a decision tree: the quest starts by identifying top branches of the “tree” that describe the broad potential causes. The mission of the analysis is to navigate the tree, searching for lower branches that are the ultimate root causes of limited access. These root causes are called binding constraints. Figure 1 represents the decision tree we used for Mexico.

Figure 1. Determinants of inadequate financial inclusion in the use of digital payment services, Mexico

Note: DSP = digital service provider.
Source: Claessens and Rojas-Suarez (2020).

To search for the binding constraints, we followed the following four principles—taken originally from Hausmann, Klinger, and Wagner (2008) —to identify a root cause:

  1. The prices of financial services serve as indicators to determine whether binding constraints are likely on the demand or the supply side of the tree
  2. A sign of a binding constraint is that its relaxation is associated with a significant improvement in digital financial inclusion
  3. A constraint is also likely binding if the agents affected by it are trying to overcome or bypass the constraint
  4. Agents less exposed to the constraint are more likely to thrive, compared with the segment of the population more exposed to it, if it is a binding constraint.

Searching for the binding constraints in digital payment services usage

One sign that the underlying problem is on the supply side of the equation is that fees and commissions for the use of different digital payment services are higher in Mexico than in other countries in Latin America. There is also evidence that rural and low-income populations face higher real costs—including long distances and time—to access financial service providers (Table 1).

Table 1. Total costs of using digital financial services by socioeconomic class as a percentage of international poverty lines, Mexico, 2018

Indicator Poor
US$5.5 a day (2011 PPP)
US$5.5–US$13 a day (2011 PPP)
US$13–US$70 a day (2011 PPP)
Poverty headcount (percentage of the population) 22.70% 46.0% 29.60%
Total costs for using digital financial services
Making withdrawals using ATMs from consumer’s own bank
Cost = fees ($0) + travel cost and social value of time to access ATM
67.8% 28.7% 5.3%
Making withdrawals at ATMs from other banks
Cost = fees + travel cost and social value of time to access ATM
97.7% 41.3% 7.7%
Full payment cycle for using mobile money services
Cost = cash-in fee + transfer fee + travel cost and social value of time to access a correspondent agent + cash-out fee
63.5% 26.9% 5.0%
Note: The international poverty line established by the World Bank for upper-middle-income countries is US$5.50 a day (2011 PPP prices). The vulnerable class has a daily per capita income between US$5.5 and US$13 (2011 PPP prices); the middle class is defined as those with a daily per capita income between US$13 and US$70 (2011 PPP prices). The PPP conversion factor used in the estimation is 8.88861 (local currency unit to international dollars), based on data from the World Development Indicators (World Bank 2021a). The percentages are calculated using the upper cutoff value for each socioeconomic class.
Source: Banxico (2020c), ENIF 2018 (INEGI 2018a), CEPEP (2018), and data from the LAC Equity Lab of the World Bank.

Our exploration of the lower-level branches of the decision tree reveals that an important inhibitor for the proliferation of digital financial services is that Mexico has stricter regulations than other countries for authorizing financial agents (normally called corresponding agents), which might discourage smaller shops from becoming agents, reducing the cash-in and cash-out access points. As a result, (high cost) retail chain stores are the most common type of corresponding agent in Mexico, but not everyone can afford them.

For potential digital payment service providers, regulation can also be an inhibitor since rules are complex and difficult to comply with. For example, the electronic payment institutions (IFPEs) defined in the Law for Financial Technology Institutions, can receive cash deposits only if they ask for additional, burdensome authorization. When verifying whether agents are trying to bypass the regulation, we find that several are forming alliances with financial entities that are already authorized to offer digital financial services, rather than going through the whole process themselves, even if this implies sharing profits.

Another important distinct factor we find is that, in Mexico, there is strong preference for cash, which suggests that a large proportion of potential consumers of digital payment services do not perceive a benefit from using them. This prevents the formation of a critical mass of users that is required to achieve economies of scale, such that digital financial providers do not find it profitable to offer services at prices that are affordable to low-income consumers. The removal of this binding constraint could significantly increase the use of digital payment services in the country. Figure 2 shows that Mexico has one of the lowest rates of cashless payments per capita among countries with available data.

Figure 2. Average number of cashless payments per inhabitant per year, 2019

Source: CPMI-BIS (2020).


In sum, our study finds that, despite efforts in the right direction, additional measures and fine-tuning are necessary to fulfill the objective of achieving adequate usage of digital financial services in Mexico. We pointed out that the two most critical areas to tackle are regulation and the lack of a critical mass of users that could enable a larger supply at affordable prices. Addressing these issues, while consolidating the implementation of the recent laws for digital service providers, needs to be an essential component of the authorities’ road map for the coming years. Although challenging, given Mexico’s level of development and advancements in the past, this goal seems achievable. Easing the binding constraints is necessary for unleashing the potential for digital financial services to play a central role in improving the standard of living of the large sectors of the population that have not yet benefitted from these services.


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.