BR Research recently sat down in Islamabad with a team of researchers from the insight2impact (i2i) facility who were on a visit to Pakistan earlier this month. A data resource centre hosted by South Africa-based think-tank Cenfri and market facilitator FinMark Trust, the i2i facility is sponsored by Bill & Melinda Gates Foundation and the MasterCard Foundation and structured around five focus areas related to financial inclusion: measurement tools, datasets, open data portals, client research, and applications lab. BR Research had a detailed discussion with the financial inclusion experts - Grant Robertson, Hennie Bester, and Richard Chamboko - on the need for a meaningful conceptual framework and a measurement system for financial inclusion. Some excerpts from that conversation are provided below:

<B>BR Research: Financial inclusion has multiple layers. But developing countries such as Pakistan seem to be confusing new accounts with financial inclusion. What are your thoughts on that?</B>

<B>Hennie Bester:</B> Currently, governments are measuring financial inclusion through the metric of access to bank accounts, even though less than one percent of the financial activities of low-income individuals are actually conducted through them. Without adults actually using these accounts to live their lives, such government policy has limited impact on development. Part of our task is to change that mindset. People manage what they measure. If you are measuring the number of bank accounts, then it is going to give you very few, if any, development outcomes. We have seen this not only in India, but also in Bangladesh and in Indonesia. One of our main insights is that developing countries have got to move from account uptake (ownership) to account usage (value).

Pakistan is amongst the countries with the lowest financial inclusion. Depending on which survey you take, only about 10 to 13 percent of adults in Pakistan are formally included. By 2020, we hear that the Pakistani government has a target to increase that penetration to 50 percent; however, we would urge caution with that approach. In India, the Prime Minister's campaign for inclusion also targeted bank accounts. There, branch managers were incentivised based on the number of active bank accounts they opened. This gave rise to the "one rupee scheme" where branch managers took their own money and put one rupee into each account. And then suddenly, ten million dormant accounts became active, but were contributing very little, if anything, to development. I tell this story to raise caution that setting off the wrong races in Pakistan could result in the same outcomes.

<B>BRR: So what are some of the metrics to assess usage or value?</B>

<B>HB:</B> There are, basically, three categories of usage. First, there are accounts that are not used at all - not even as a store of value. If an account hasn't been used for 12 months, it really has no utility for an individual; second, if the account has only been used once in the past three months, it is probably for some form of remittance; and third, if it is used once a month, it is probably to receive a salary or wage or welfare fund. Besides these three, there are frequent users, who are likely the same individuals who were using their bank accounts on a regular basis five years ago.

There are some usage metrics, which are normally gathered by payment regulators, but the problem is that these are typically aggregated by channel, not by client.

On the other side, there is a completely different kind of research that is being done, called the Financial Diaries. It was originally started by the FinMark Trust in South Africa, around 2004-05. It was later expanded to Kenya, and then to Pakistan. Financial Diaries normally pick two hundred households and track all of their financial transactions for twelve months. That gives you the most granular and utterly fascinating data on how low-income households really use financial services to manage their lives. The challenge is to take this kind of quantifiable and useful insight to financial service providers and governments.

<B>BRR: What are some of the insights from these Financial Diaries?</B>

<B>HB:</B> The diaries tell us that low-income individuals are extremely ingenious, and they maintain multiple relationships - particularly for credit. They may maintain relationships with up to five or ten credit providers, all informal. If they fall into bad favour with one, they use the other one. If they have some connection to the formal sector, it is usually for long-term savings, as they have trust in the registered service provider. These individuals often get disappointed in group savings.

<B>BRR: So, based on such insights, what is i2i's definition of financial inclusion?</B>

<B>HB:</B> We need to understand that simply throwing formal financial services down the low-income sector's throat doesn't work. Most don't need access to formal, multi-functionality accounts and we, at i2i, like to refer more to usage of retail financial services, which is a much wider concept.

Our research shows that most low-income individuals in the world prefer localised financial services, which is basically the point where I am served and the decision-maker for my service is within walking distance of where I am located. These services are mostly informal in developing countries, and, as low-income individuals live lives of great variability, they really value being able to have interaction with someone who can show flexibility and whose service is cheaper. But localised financial services are rarely advocated by governments, with the exception of Thailand, and often create a threshold of value for which remote and, typically more formal providers, compete with. Hence informal financial service persists!

For financial inclusion, you could start with the formal accounts - how often do people use it and what do they use it for? But we start from the other side: what do people really need financial services for? We have identified four needs through our country research. The biggest one is the "transfer of value" to families back home. Transferring value is a market. How big is that market? How many people are using it? What's the quantum of value that is transferred on average? And what must you do as a formal provider to meet that market?

The second is the people's need for liquidity, to smooth consumption and meet monthly expenses. People do that through multiple means. The third is about resilience, having access to an amount of money in an emergency, particularly in the context of health shocks. And the fourth need is to meet life goals: how do you put together a larger sum of money to do bigger things, ie to invest, to make purchases, etc.

<B>BRR: In the context of Pakistan, what is the biggest need? Is it payments, credit or savings?</B>

<B>HB:</B> It is payments. There is a small need for credit, which is pretty low-down on the list. First come payments; then savings, and then credit. The whole financial system now seems to be re-aligning around digital payments. The conventional financial system is becoming a big dinosaur.

<B>Richard Chamboko:</B> This is a cycle. Payments come first, then other financial needs. Payments build the pipeline.

<B>BRR: If commercial banks are becoming like dinosaurs, how are global payment schemes, like Visa and MasterCard, adapting?</B>

<B>HB:</B> I think a lot of them are exploring. In many countries, they have partnered with a mobile payments provider to provide card functionality. The road to a cashless economy is through cash. It is a basic cost issue - transaction cost is virtually zero for cash-to-cash transactions in the eyes of a customer; but if I pay you in digital value and you are unable to pay it further in digital value, then you have to convert. The moment conversion between digital money and cash starts to approach zero, then you can start getting closer to a serious digital economy.

<B>BRR: How do you push the conversion cost between digital cash and paper cash to zero?</B>

<B>HB:</B> You have to have a ubiquitous network of payment agents, basically increase points where they can convert the two kinds of cash. But to make the infrastructure as cheap as possible, you need to bring other financial services in. Most non-bank-led systems, such as Safaricom in Kenya, are under pressure because they are unable to cross-subsidise their payments division by income from lending or taking savings to do lending. Cross-subsidisation is relatively easy for bank-led regimes.

<B>BRR: What are some of i2i's data tools that can help capture the extent of financial inclusion?</B>

<B>Grant Robertson:</B> Our mandate is around improving the quality and use of survey data and geospatial data in the financial inclusion space. Geospatial data (or GIS) is a very powerful technique to lay out various datasets, but not many countries have official directives to force financial institutions to map out their infrastructure.

We are working in that space to build communications around geospatial data, doing workshops, and running webinars. Starting from basic access points, such data can really inform financial inclusion strategies. So the idea is to map the uptake and then move towards measuring usage through GIS maps.

<B>BRR: While financial inclusion targets the lower-income sector, do you think one of its objectives could be digitisation of the economy?</B>

<B>RC:</B> Financial inclusion's fundamental goal is to improve the welfare of the people. We may play around with the definition, but I am not sure if we can set boundaries.

<B>GR:</B> Digital financial products and services are readily available to high- and middle-income sector, and they don't need a lot of encouragement. It is the lower level that needs services, technology and literacy provided.

<B>HB:</B> For me, financial inclusion is about access of entire population to financial services. In Pakistan, a lot of people that are not poor are not using formal financial services - for varied reasons, such as not wanting to pay taxes. The one thing about digitisation of economy is that transaction data is becoming an alternative to collateral. That's why traditional banking, which is based on secured lending, is going down. The digital platforms are tracking people's transaction flows, which can then be used to assess someone's ability to repay without the need for collateral, and then lend people money in an economy that is largely informal.

But the conflict of digitisation is that a country with a low tax-base will see its government start taxing the transaction flows. To what extent the laws protect the digital transaction flows from the revenue authorities will be critical. Taxation policy will be the biggest stumbling block in the way of digitisation of the economy since real issues regarding data privacy are going to be around government intervention. People prefer cash because it is anonymous.