Measuring ‘Financial Inclusion’ with a Multidimensional Index: A Conversation with Noelia Cámara, BBVA

Financial inclusion stories and research are published daily, lauding various efforts to bring lower-income people into the formal banking fold. All progress deserves celebration, but also closer examination. When a new initiative takes effect, or a new service deployed, how does that advance us in achieving financial inclusion? A backdrop of sound measurement is critical. A BBVA research team, Noelia Cámara and David Tuesta, recently set out to construct an index that measures the extent of financial inclusion at the country or region level. The index is discussed and applied to 82 countries in the team’s new paper, Measuring Financial Inclusion: A Multidimensional Index. We were especially intrigued to learn that this research incorporates both supply and demand-side data. I recently sat down with Cámara to talk about the project, from challenges in measuring financial inclusion to the implications of the newly-available index.

1. What are the challenges in measuring financial inclusion?

Many issues arise when it comes to measuring financial inclusion. First, there is no single definition for financial inclusion universally accepted in the literature. Most definitions include three dimensions: use, quality, and access. However, when it comes to defining these dimensions, no consensus is found. For instance, the use of financial services is part of the financial inclusion concept, but it is not clear what “use of financial services” really means. Thus, several questions come to the fore: Do we consider having a bank account in the formal financial system to be a necessary condition for financial inclusion? Is having a pre-paid card or microinsurance enough to classify an individual as included? Is using electronic payment intermediation (e.g. paying bills with a mobile phone) a sufficient condition?

Second, and this may be the most important challenge to be addressed in the near future, there is a lack of harmonized data. For example, measuring access can be approached by assessing the number of bank branches and ATMs, but there isn’t always information available about the number of banking correspondents’ outlets. There is also access to financial services through mobile phones and electronic devices like laptops and smartphones. However, having a mobile phone or a computer is not a sufficient condition for access to the financial system; in order to generate broader access and foster financial inclusion, many other things such as regulation, new financial digital channels, or infrastructure are necessary. Finding workable data and measuring these elements of access is difficult. These challenges very much apply to measuring the quality of financial services in a given market, as well.

2. What dimensions do you use to measure financial inclusion? Which dimensions are more important than others?

We postulate that the degree of financial inclusion is determined by the three dimensions of usage, barriers, and access. Barriers are used as a proxy for the quality dimension. We define an inclusive financial system as one that maximizes usage and access, while minimizing involuntary exclusion. Involuntary financial exclusion is measured by a set of barriers perceived by those individuals who do not participate in the formal financial system.

The weights of the variables in our index are calculated endogenously, by applying a technique called Principal Component Analysis (PCA) in two stages. As shown by our estimates, access is the most important dimension for measuring the level of financial inclusion (0.42). Access represents a necessary but not sufficient condition for using formal financial services. The weight for the usage dimension is 0.29, and for barriers it is 0.28.

3. Which counties performed well and which performed poorly?

As expected, developed countries have the most inclusive financial systems, and the African countries are more financially excluded. There are some exceptions among developing countries, such as Mongolia and Thailand, which perform relatively well.

4. Your study is unique in that it combines both supply and demand-side data sets. What are the limitations of relying on supply-side data?

Supply-side data is only accurate in measuring some aspects of financial inclusion, such as access. These data are not conclusive in terms of measuring use of financial services. For instance, with number of bank accounts, we have no information about how these figures are distributed among the population. Supply-side data may overestimate the inclusiveness of financial systems, as one person can have more than one account or loan, which is a very common practice in developed countries. By contrast, having individual micro-data for the demand-side is desirable because it enables us to identify who is actually using the financial services.

5. You write that “We find that the degree of financial inclusion is highly correlated with some macroeconomic variables such as GDP per capita, education, efficiency of a financial system, and financial stability.” What do you think the policy implications are since financial inclusion is so intertwined with broader economic and social issues?

Talking about policy implications from a macroeconomic perspective needs further analysis to check for causality. What we illustrate in our study is that some important macroeconomic variables such as GDP per capita, education, efficiency of a financial system, and financial stability perform similarly with financial inclusion, which then shows that a causal relationship may exist.

6. At CFI, we are very much focused on the quality of services provided to the base of the pyramid, and client protection in particular. Where do you see the role of more qualitative issues like client protection in measuring financial inclusion?

The quality dimension is, so far, the most difficult of financial inclusion to measure. As I mentioned earlier, in our index we try to approach quality of financial services by the perceptions of excluded people. Particularly, the variable trying to proxy the lack of trust in the financial system should capture part of this client protection information. We are aware that this is only a partial approach, particularly since only excluded people provide answers to this question. Our current research focuses on improving the measurement of the quality dimension.

7. As you may know, the CFI has defined financial inclusion as a state in which “everyone who can use them has access to a full suite of quality financial services, provided at affordable prices, in a convenient manner, with respect and dignity. Financial services are delivered by a range of providers, in a stable, competitive market to financially capable clients.” How do you feel this definition corresponds to your index and research and in what ways do they diverge/differ?

The CFI’s definition for financial inclusion includes the three dimensions that we use to define financial inclusion, with quality-related issues included only partially from the perception of excluded people. There are also some aspects such as “respect and dignity” that we are unable to measure with the available information.

The only divergence between the CFI’s definition and our index is the inclusion of the financial capability concept. This is a broad concept that is difficult to define. It includes financial literacy, learning by using financial services, having the money to use the services, and so forth. Financial capability is not only difficult to measure for all the countries in our sample, but we also consider financial capability as a variable related to financial inclusion, rather than one to be included in the financial inclusion measurement. We believe that financial capability could be either a cause or a consequence of financial inclusion.

8. What about the role of insurance, pensions, and other financial services not accounted for in the index?

There are different degrees of financial inclusion. Our index is a basic measure that can be interpreted as a lower bound. We consider a financially included individual as having one of the following financial products: a bank account, savings account, or loan. Although other financial products such as pensions and insurance are important, we do not consider an individual to be financially included if holding only insurance with none of the previous specific financial products.

9. Were there any surprises in your research?

Surprises are common when creating an index. One may have a general idea about the ranking of a group of countries, however, there are always surprises. In our case, Mongolia and Thailand rank much better than expected. These two low-income Asian countries outperform other low-to-middle income countries, their East Asian neighbors, and even some high-income countries. For Mongolia, the degree of financial inclusion in the country is even higher than developed countries such as Sweden, Ireland, and Austria. It may be due in large part to universal cash hand-outs from the government. Thailand, which is ranked higher than Greece, has a high position in the ranking mainly due to the large number of bank accounts and the insurance schemes offered by the government.

10. What gaps in research still remain and which do you think are the most important to address?

For measurement issues, the quality of the data is the most important challenge that we need to address. Working on improving quality and generating accurate data are key for improving the measurement of financial inclusion. Big data generated by financial institutions (provided in an aggregated manner) should substantially improve the measurement of the three inclusion dimensions. Examples of such provider-generated data might include information regarding the range of financial services used by income group and access to financial services by channel. On the other hand, information about the unbanked is only possible through surveys.

Theoretical research and industry alignment is important in terms of operationalizing and acceptance of a definition of financial inclusion. Identifying how mechanisms behind financial inclusion work may help to reach a single, commonly accepted definition.

Have You Read?

Measuring Financial Inclusion: The ‘Basic Set’ Solution

Lean Research – Is It Time to Reimagine the Study of Financial Inclusion?

Reality Check: Three Ways We Hype Financial Inclusion Breakthroughs

Join the Conversation

Stay informed. Subscribe to our newsletter.