Download the FinNeeds Indicators here.
Element |
Indicators |
---|---|
Needs |
For each of transfer of value, liquidity, resilience and meeting goals
Note: liquidity is a single use case (need) |
Device portfolio to meet needs |
For each need, or per specific use case as relevant:
|
Usage |
Usage intensity indicators defined as relevant per use case or need category:
|
Drivers |
|
Outcomes |
|
The needs measurement framework is at the heart of the FinNeeds approach and it focuses on use cases as the starting point.
Each use case can be classified into one of the four categories of financial needs:
Each of these financial needs is made up of a combination of use cases. People exhibit a specific use case which is then classified within a particular need “bucket”.
The diagram below shows examples of % of adults in the sample who expressed specific transfer of value, resilience and goals use cases (Over the past 12 month period) in one of our pilot studies:
Once we know what people’s use cases are and how these classify across the four needs, the next step is to understand how people meet their needs and what role formal financial services play compared to informal or social means of meeting needs. There are various ways people can meet their financial needs or use cases, which are known as the “financial devices” used.
Many financial inclusion surveys already track the devices people have, but the FinNeeds approach tracks what devices people use when responding to each use case and need.
A financial device is any physical, social or electronic mechanism that stores, accumulates, distributes or transfers value, and can be used to meet a financial need. Financial devices is a broader concept than financial services, it is what a person makes use of to meet a financial need. For example, cash at home or savings in gold or other assets would be a “personal device”, while assistance from family and friends would be a “social device”.
Financial devices can be classified in terms of provision or products:
Download the FinNeeds Measurement Framework note here.
The chart below shows the range of financial devices used to address the liquidity need in one of our pilot studies.
We know from conventional financial inclusion measurement frameworks what financial services or devices people have, but often there is not much detail on how people engage with these devices and how usage patterns differ between different use cases and types of devices. The usage measurement framework explores usage-towards-needs according to four indicators, the specifics of which will differ per use case and device:
Usage is most objectively gauged via financial service provider transaction data. For example, tracking account activity, mobile money or card transactions. Consumer surveys or other demand-side data cannot give as granular or objective a picture of usage, but they are currently the only way to gauge usage patterns of informal and social devices.
Download the Drivers Measurement Framework note here.
Indicators: Usage intensity indicators defined as relevant per use case or need category:
Example outputs from our pilot studies
The table below illustrates usage intensity clusters identified for point-of-sale customers in one of our pilot studies:SegmentSEG 1 | SegmentSEG 2 | SegmentSEG 3 | SegmentSEG 4 | SegmentSEG 5 | SegmentSEG 6 | |
---|---|---|---|---|---|---|
Average number of transactions per month since customer first seen | ||||||
Average number of transactions per month since customer first seen | 4.5 | 4 | 5.5 | 23 | 60 | 120 |
Average value per transaction | ||||||
Average value per transaction |
N 7,300 $20.28 |
N 9,200 $25.56 |
N 28,000 $77.78 |
N 9,500 $26.39 |
N 11,400 $31.67 |
N 11,600 $32.22 |
Median spend per month since customer first seen | ||||||
Median spend per month since customer first seen |
N 24,000 $66.67 |
N 25,000 $69.44 |
N 100,000 $277.78 |
N 210,000 $583.33 |
N 690,000 $1,916.67 |
N 1,500,000 $4,166.67 |
Average unique MCC codes used | ||||||
Average unique MCC codes used | 15 | 10 | 20 | 50 | 80 | 100 |
Proportion of total customers in segment | ||||||
Proportion of total customers in segment | 39% | 13% | 10% | 22% | 13% | 4% |
Proportion of total spend contributed by segment | ||||||
Proportion of total spend contributed by segment | 5% | 2% | 7% | 20% | 42% | 24% |
To derive this table, we applied statistical clustering techniques to cluster digital payment users into different “intensity of usage” groups. Segment 1 consists of low users of point-of-sale devices across the frequency and monetary value indicators. This is the largest segment, at 39% of the clients. On average people in Segment 1 have four transactions per month with an average monthly expenditure of NGN 24,000 at point-of-sale machines. In contrast, Segment 6 has high usage, with 120 average monthly transactions and an average spend of NGN1.5 million at point-of-sale machines. Segment 6 consists of only 4% of the sampled clients. The different segments give a usage measure which can be used to better understand how people interact with a device across multiple dimensions.
The next step is to compare and contrast the profiles of different usage clusters to inform strategy and policy interventions to increase usage. For example, are high usage customers likely to be more educated and higher-income than low users? Is there a relationship between geographic location (close or far from access points) and usage? Or is there some other element, such as compulsion or having a specific other financial service, that explains intensity of use? And how does intensity of use differ across use cases?
The drivers of usage conceptual framework seeks to explain why people choose the devices that they do to meet their needs. This framework draws on human decision-making models, augmented by financial-inclusion-specific research, to consider how individuals make decisions about the use of financial services. Understanding these drivers can help policymakers to address key policy problems such as high use of informal financial services. Financial service providers can also use this framework to identify which factors to consider when trying to drive use of specific financial products. Four broad classes of drivers are defined:
The drivers that matter most will differ across use cases and will be based on factors such as local context, device type, provider type and consumer characteristics.
Download Drivers Measurement Framework note.
One of our pilot demand-side surveys asked people to rate the reasons for choosing each device in their portfolio, based on a list of reasons such as “it’s convenient”, “I’m comfortable with it”, “others like me use it”, etc. These reasons were designed to be classified into two categories: functional reasons (benefits, convenience, cost considerations) and relational reasons (trust, sense of belonging, comfort). The results suggested that respondents tended to use:
All socioeconomic classes skew towards relational factors, except the very top two high-net worth categories, who emphasise functional benefits.
In a separate statistical modelling exercise of the determinants of usage intensity, conducted on bank transaction data as part of the same study, it was found, not unexpectedly, that income is by far the largest driver of usage, followed by education and relationship status. The demand-side survey responses, however, suggest that more may be at play than just demographics in explaining uptake and usage of different types of financial devices.
The FinNeeds framework seeks to understand the mix of financial devices that people have, as well as the way in which they engage with their financial devices (usage patterns), towards fulfilling their financial needs. Outcomes of use refer to the extent to which financial services enable people to meet their needs.
The outcome indicators aim to establish the percentage of people meeting each need. For example, the percentage who classify into different degrees of resilience, or the percentage of people maintaining liquidity. We then overlay the usage measurement framework to infer insights on the overall “success” of financial services in achieving positive need outcomes. For example, are those who have formal insurance more likely to be resilient than those who are not insured? Or are those with formal credit able to meet the same goal faster than those without? The answers to these questions have implications from a policy perspective. It can help to inform (a) an assessment of financial inclusion and financial sector welfare impact and (b) where the primary challenges are and what can be done to change the situation.
As outcomes are fundamentally different across the four financial needs, separate outcomes measurement frameworks must be created for each of the four needs, or even at a specific use case level. For example, measuring whether one is resilient (is able to recover from unexpected financial shocks in a timely manner) is different in scope from measuring whether a person is able to meet his or her goals (is able to save a large amount of money to pay for an expected expense). So far, i2i has developed initial indicators for liquidity and resilience outcomes. Further refinement is needed.
The chart below gives an indication of the percentage of the sample population in one of our pilot studies classified into one of three liquidity outcome categories: Those who experienced no instances of illiquidity (being unable to meet expenses in a regular income cycle) over the past 12 months were classified as having experienced “no distress”. Those who were unable to maintain liquidity once in the past year were classified as having experienced “some distress” and the rest were classified as being under “severe liquidity distress”. The diagram also shows which types of financial devices those with distress took up to deal with the shortfall experienced. For the no distress category, the bars denote overall device uptake. The blue line indicates the percentage of adults in the sample classifying in each category:
By comparing device uptake across the three segments, it was interesting to see that those with distress are relatively high credit users but also, counterintuitively, have higher incomes than those without distress. This suggests that the middle class is overextended. Also telling is that those with distress are more likely to turn to social devices than those without, a tendency that increases as the level of distress increases. This suggests that, even if people have formal devices, it is not their first port of call when they are struggling to balance the books and may therefore not serve the intended liquidity outcome-of-use purpose.
Learn how to apply the #FinNeeds approach to measuring financial services usage
Sign up here for our monthly Data Briefing on data and innovation in the financial services sector.
And, to receive insight2impact updates, subscribe to the insight2impact quarterly newsletter.
Please write to us and give us your feedback. info@i2ifacility.org