The Gender Gap in Financial Inclusion: What Can We Learn From Microfinance and Banking Data?

Photo credit: Olli’s Place

The UN Capital Development Fund (UNCDF) SHIFT Programme and Pulse Lab Jakarta teamed up to support four financial service providers and the regulator in Cambodia in analysing anonymised financial records. The support focussed on building capacity by jointly investigating the factors affecting savings and loans mobilisation. Below, we share our findings to date.

Global progress, but a gender gap persists

Globally, there has been great progress in advancing financial inclusion. Yet, there remains a gender gap in account ownership, savings, credit, and payments behaviour. In 2014, 58 per cent of women worldwide held an account, compared to 65 per cent of men.

Interestingly, among the member states of ASEAN, men and women have almost equal access to financial services. For example, national financial inclusion data (FinScope) shows that in Cambodia there is only a small gender gap of +2 per cent; in Myanmar the difference is also small (-4 per cent); and in Lao PDR it is again slightly positive in favour women (+1 per cent).

While men and women may have an equal access to banks and microfinance institutions, an open question remains as to whether they have similar usage patterns of financial services. This question sits in the context of borrower exit rates varying between 28 and 39 per cent per annum (Mix Market 2015), and 45 per cent of the savings accounts being reported dormant (Findex 2014). Alongside stakeholders in Cambodia, UNCDF SHIFT and Pulse Lab Jakarta felt a deeper analysis into the gendered aspects of financial service usage was merited.

Analysing customer journeys

The partners analysed a unique set of customer data obtained from four large Cambodian financial service providers — AMK, AMRET, Sataphana Plc and VisionFund — for the period 2010 to 2015 to capture a medium-term overview of loan and savings mobilisation. Readily available data from the management information systems of the financial service providers was gathered by the UNCDF SHIFT Programme and Pulse Lab Jakarta applied its big data analytical expertise to uncover insights.

The research project has curated around 5.4 million anonymised savings and loan records from 2.6 million customers to examine savings account dormancy and borrower exit, and to improve financial service usage. To put these figures in perspective, Cambodia has an adult population of just under 10 million people.

During the first phase of the research, the different datasets were analysed separately for comparative purposes and the data were disaggregated by gender. Data were analysed using various statistical methods, such as descriptive analysis, distribution analysis and simple survival analysis using hazards ratio.

Initial insights

It has been established above that men and women have equal access to credit and savings services in the region, but our data analysis suggests that the actual usage patterns in terms of the loans and savings amounts mobilised are much lower for most women. In 2015, women on average had around 600 US$ lower savings balance and 1200 US$ lower average loan amount.

These data also enable more powerful insights, not just breaking down data simply by men and women, but doing further sex-disaggregated data analysis by age categories, marital status, different products, and geographies (rural, urban, and between provinces).

Figure 1 Average loan and savings amounts by gender

Initial results show that the majority of customers (69 per cent) did not utilise their savings account, owning total deposits balances that did not exceed 5 US$ (the minimum opening deposit for some financial service providers is around 2 US$). Women more often had low value savings balances below 5 US$ (75 per cent) compared to men (55 percent), indicating the majority of their accounts are dormant. Savings mobilisation remained heavily concentrated in the capital, Phnom Penh, and among a small number of depositors. There is, therefore, a significant untapped potential for savings mobilisation beyond Phnom Penh, and for women in particular, and a significant potential for reducing borrower exit.

Concerning loan mobilisation, the data show men taking on higher segment-loans (SME-loans > 5000 US$ vs. micro < 1500 US$) compared to women. The gap in loan amount is partly explained by differences in types of product uptake: women more often take group loans; men, on the other hand, prefer to take up individual loans, which are generally higher in total amount compared to those of women. Men and women have a somewhat equal borrower retention. While younger customers are often perceived to be less bankable, the analysis equally indicates that they are more likely to experience a lengthier savings relationship with the financial service provider and equally likely to stay in the borrowing relationship.

These results highlight the need to tailor individual loan products better to the specific needs of women, and that there can be more extensive promotion of term deposits over savings accounts. These findings need to be validated by more extensive consultations with stakeholders.

Next steps

Pulse Lab Jakarta and UNCDF SHIFT are currently undertaking the second phase of analysis on the microfinance datasets, which develops a survival model to investigate long term customer exit by incorporating socio-demographic variables (gender, age, marital status). The project could also be extended to other Cambodian financial service providers as well as include data from 2016 and 2017 to ensure results cover more recent market events. The research would also benefit from additional qualitative analysis and focus group discussions.

Furthermore, we hope to look into other research applications such as how economic shocks or natural disasters influence saving and loan activity. Can we identify coping mechanisms, based on these insights, at an individual or community level, derive measures of resilience at a community level, or understand better the communal and societal recovery after a disaster?

These insights are in the context of the wider capacity development of the regulator and financial service providers in Cambodia to conduct data analytics and to shape policy based on the insights. UNCDF SHIFT is piloting a family savings product to address some of the issues of savings account dormancy, and is working with the National Bank of Cambodia to integrate the findings into the national financial inclusion strategy.


Pulse Lab Jakarta and the UNCDF SHIFT are grateful for the generous support from the Government of Australia.