New Global Findex: What You Need to Know

25 April 2018

The third global Findex was released with much fanfare at the World Bank’s Spring Meetings last week. And rightly so. This eagerly awaited data set is the financial inclusion community’s best demand-side measure of financial inclusion globally. With three years of time series data now available, Findex tells a compelling story of the progress we have made on access to financial services over the past six years, and it also illustrates the challenges we face in bridging the access gap for marginalized groups and increasing usage. What it doesn’t tell us — and what our community needs to examine as more people gain access to financial services — is whether and how financial services benefit the poor, and how to measure and mitigate potential downside risks.

Gains in access, but gender gap persists

Findex shows remarkable progress on access to finance for the poor across developing countries, with 515 million more adults globally reporting account ownership in 2017 than in 2014. India was a star performer in 2017. Roughly 80 percent of the country’s population now has a financial institution account, jumping from 53 percent in 2014. India also reduced its gender gap from 20 percentage points to 6 points. Sub-Saharan Africa also continues to make remarkable progress, with 10 countries in the region reaching at least 50 percent access and four reaching beyond 65 percent. Even very challenging markets like the DRC, Liberia, Malawi and Mali made progress.

As Asli Demirgüç-Kunt and Leora Klapper pointed out in a World Bank blog post last week, despite advances in women’s access to financial services in markets like India and Indonesia, the global gender gap remains unchanged at 7 percentage points. In some countries, low access rates for women bring down the national average significantly. For example, Pakistan would achieve far more than 21 percent access nationally if more than 7 percent of its women were banked. Bangladesh made impressive progress on inclusion generally, but mostly among men: the gender gap grew from 9 percentage points to 29 points between 2014 and 2017.

Progress has also been made in rural areas. The percentage of rural households with an account increased from 44 percent in 2011 to 66 percent in 2017. Big shifts in China (58 to 78 percent), India (33 to 79 percent) and Sub-Saharan Africa (19 to 39 percent) are largely responsible. This is important, but there is clearly still much more to do, especially in Sub-Saharan Africa, where roughly 75 percent of the population lives in rural areas. As in many things, Kenya leads the way in the region. Rural Kenyan households with accounts grew from 38 to 81 percent between 2011 and 2017, largely thanks to the ubiquity of mobile money.

Importantly, Findex enables us to link progress to policy choices. CGAP has published a report on the basic enablers of digital finance, namely, agents, nonbank issuers of e-money, tiered and risk-based KYC and appropriate consumer protection measures. By and large, Findex shows that countries with enabling regulations around agents, e-money and tiered KYC experienced substantial improvements in access. In addition, the rapid progress made in markets that leverage government-to-person payments, like India, Mongolia and Iran, demonstrates the role a broader range of policy choices can play in shifting access. This new evidence should encourage governments to adopt the kinds of policies that improve access.

Usage lags behind access

Important data points are also emerging on usage. Interestingly, they correlate with global supply-side data from GSMA.

Usage continues to lag, especially in markets that have taken large leaps forward in access. While India was a strong performer on access, it still has a usage challenge, with 48 percent of registered accounts remaining dormant. Growth in 90-day activity rates on mobile money accounts in Africa has also been slower than most would like. GSMA estimates these have grown from 26 percent in December 2012 to only 36 percent in 2017. But these figures understate an overall positive trend in usage rates. The denominator of registered accounts in both cases grew rapidly — by 27 percentage points in India since 2014 and by 25 percent in 2017 alone in the case of mobile money in Africa, according to GSMA's 2017 State of the Industry Report on Mobile Money. Given that new accounts are less likely to be used, this suggests that underlying usage rates may be better than we think. At the same time, it points to increasing usage as a central challenge for operators going forward.

We also know from GSMA that a larger number of use cases on an existing platform enhances usage. GSMA data show that the number of use cases besides person-to-person, airtime top-up and cash-in/cash-out is gradually increasing and includes bulk disbursements, merchant payments and bill pay. Findex confirms this, but also shows how much more room we still have to go to fully digitize payments. The number of people in developing countries who made or received a digital payment in the past year grew from 32 percent in 2014 to 44 percent in 2017. So we know that people are increasingly using digital accounts to make payments.

For a few key use cases, Findex shows that there has been a gradual shift away from cash and toward more digital payments. For example, 53 percent of developing country respondents paid a utility bill in 2017, 27 percent sent or received domestic remittances, 19 percent received a government payment and 15 percent received an agricultural payment. For all of these use cases, the proportion of people using accounts for payments rose significantly, with the proportion using cash decreasing across the board. Continuing to encourage a shift away from cash across a range of use cases will help increase overall usage rates for digital finance.

But there is also considerable room for growth. In each of these use cases, the actual proportion of people using accounts is still relatively low, but increasing. Twenty-eight percent of those who paid a utility bill in developing countries did so through an account (up from 15 percent in 2014). Of those who sent or received remittances, 46 percent did so through an account (from 30 percent). Of those who received government payments, 65 percent did so through an account (from 59 percent). And of those who received agricultural payments, 18 percent did so through an account (from 8 percent). Governments are clearly doing relatively well at getting people to use accounts to receive payments, but there is still room to improve. And with the other use cases, there is still significant room for improvement.

What does this mean for poor people?

As a community, we still need to have a better answer for the ways in which access to financial services actually benefits poor people, whether through the direct use of financial services or because finance enables poor people to gain access to other basic services, such as electricity, clean water, health and education.  We have some useful data points from the research community, especially for payments and savings, but we are far from having a complete picture, especially for credit, which is of course the financial service that drives profitability for most providers. While well-being may require separate impact analysis to establish causality, there are things we can measure that are related to well-being — particularly, do these products do harm?

On a recent trip to Kenya, there was widespread concern about an emerging digital credit bubble. CGAP research bears this out. In Tanzania, we conducted a nationally representative phone survey on digital credit and discovered that half of digital borrowers had made late loan payments, and one-third had at some point defaulted on a digital loan. There is a need to help governments track and remediate risks around financial services that are increasingly available to the poor, to improve providers’ ability to supply financial services responsibly, and to help consumers understand the financial products they are using and the consequences of defaulting.

I have been working in the financial inclusion industry for 18 years, and I have never seen the kinds of gains in access we have seen over the past decade. It is an exciting time to be involved in financial inclusion. But we still have much to do to reach marginalized groups, improve usage rates across the board and ensure financial services are provided in a healthy, sustainable manner. We at CGAP look forward to working with our partners to continue the journey.

This post has been updated to reflect the fact that CGAP's Tanzania survey revealed one-third of digital borrowers had at some point defaulted on a digital loan.


Submitted by anuj jain on
There seems to be a conceptual flaw in conflating 'financial inclusion' with 'having an account with formal financial institution, bank or in some cases mobile money'. This is perhaps an institutional bias of the researchers and report in framing the report and analysis; because time and again, and for many years, we know and know well that many people continue to use self-managed, often informal financial services' and prefer to do so because of a number of reasons. We also know that even when we have accounts in formal institutions, many of us, and majority of people living in poverty continue to prefer to use financial services outside registered and regulated formal institutions. Time and again, it has been made clear by people that it is not that institutions don't trust them, it is actually the other way round. People don't trust institutions with their money, and sometimes don't find is either convenient or necessary to have (or use, when they have) an account in the banks. While banks must continue to strive to become instruments of national policies and global mandate to be inclusive in their practices, measure of financial inclusion and this data must be understood with a pinch of salt that those who are NOT in the formal banking system, have their own ways to access financial services. It will be interesting to confirm, for example, if 67 million members of SHGs in India linked to banks are counted in the numbers or not. If 14 million+ members in VSLAs are considered included or not? On the other side, we also know that only a small fraction of those considered included (i.e. having a savings account) have access to loans and other financial products, and not really participating fully in the formal financial system just because they have a savings account. The analysis and report is of lot of use and must be commended for all the efforts and work. Yet, the conclusions need very careful and deliberate analysis. Numbers often hide more than they reveal, as wise ones say.

Submitted by Joey Kemler on
Hello, I enjoyed this blog post immensely. It is great to see the increases in access but you do well to acknowledge the fact that we still don't know how/if this is improving the lives of the poor. I do believe these gains to financial access are important in the long run growth of developing economies. You also mention that credit is the service that provides profitability for most providers. I was wondering though if systems like M-Pesa and M-Shwari have tested the idea of using advertisements in their systems? As a way to create a new revenue stream. In the blog post regarding how M-Shwari works ( I believe there would be an interval or gap, where an advertisement could be placed (between the time it takes to find out if you receive a loan or not). This could improve the sustainability of such financial services. It was just a thought I had as I was writing a literature review about digital financial services. If you know of any experiences implementing this type of advertising or the roadblock thats are hindering it, I would greatly appreciate you sharing these with me. Thank you.

Submitted by Larry Reed on
I'm surprised that, with all the money being invested in financial inclusion, we still have not been able to answer the question of whether it does any good for people living in poverty. That should have been one of the first things we analyzed, and yet, here we are, 30 years and billions of dollars after starting and we continue to try to rapidly scale a service without being able to tell whether it is doing any harm.

Submitted by Greta Bull on
Larry, Thank you for taking the time to share your views. As Dan rightly states, there is more and more evidence coming out to support the financial inclusion work. It helps us to identify examples of best practice and build our knowledge. In terms of impact evidence, it is extremely difficult to make sweeping generalizations for several reasons. Firstly, financial inclusion encompasses many products. A positive impact finding in one country or context may not replicate in another, but it does expand our understanding of which pieces are necessary. Secondly, research methodologies and data continuously evolve, which means we may answer the same question very differently today than we had in the past. Thirdly, it is possible that we’ve expected too much from financial inclusion. We know that on its own, financial access is not likely to lead to dramatic reductions in poverty (although recent evidence from Kenya suggests that in some cases it can make a significant dent). But, it can be an important enabler for broader development goals such as education, health, energy, and clean water. Lastly, pushing the financial sector to be more transparent, fair and responsible is a goal we can all work toward, without waiting for a RCT to tell us why it matters. Looking forward, digital financial services could provide a huge boost for financial inclusion, but many thorny questions remain -- particularly, ensuring that digital financial services actually reach the poor rather than perpetuating a digital divide. Also, new players, most notably, super platforms, pose new challenges for policy makers. For these reasons, consumer protection (data protection and privacy, cyber security and competition policy) and value for customers are priorities in CGAP’s next five-year strategy. We will also engage more deeply with the research community to make sure that evidence is actionable, relevant, and answers the sector’s most pressing questions about why this work matters.

Submitted by Dan Radcliffe on
Larry: You’re right that the financial inclusion community has under-invested in rigorous impact research and the rhetoric has often gotten ahead of the impact evidence. However, there is a strong and growing body of evidence that access to finance improves household welfare. Perhaps the best literature review on the subject is this NBER working paper After reviewing the impact literature, the authors had four conclusions: 1) The impact evidence on microcredit is mixed; 2) The impact evidence on savings indicates replicable positive household welfare impacts; 3) The impact evidence on insurance is small but potentially promising; 4) The early impact evidence on payments and digital financial services indicates significant benefit. So the upshot is that savings, payments, digital financial services and (potentially) insurance have positive impacts, though there is much still to learn, particularly around digital credit and insurance. A range of digital credit randomized experiments are underway so we will have more to say on this topic soon!

Submitted by mustafa ramzan on
impact measurement methodologies have a role to play in assessing whether impact was there or not.Economists are dogmatic about RCTs whereas impact investors and microfinance practitioners feel lean data measurement model(something private sector has been using for decades) is the right method.i think financial inclusion with lack /absence of employment and entrepreneurship opportunities both among middle class and base of the paramid will have little to no effect.on human welfare!

Add new comment