How banking data can fill the evidence gap on income volatility… and more

If you have been following JRF’s insight infrastructure latest endeavours, you might be aware that we have recently announced a partnership with Smart Data Foundry to explore how banking data can help us untangle stories of economic security and income volatility. However, if you are yet to catch up with this new exciting project, you can find some useful information in my previous blog “Inside The Vault”, as well as in SDF’s “Breaking Down Data Barriers”.

This time around, I’d like to share some examples of what the project will mean in practical terms. In short, we are working on two main assets: 1) a publicly available and free-to-use dashboard, powered by income and expenditure data, as well as some key socio-economic indicators; 2) a Safe Haven that researchers will use to securely access the raw data. But let’s break those two down a bit further.

1) Knowledge is power, and power should be always shared, proportionally and respectfully.

Dashboards might feel a bit overused and overrated nowadays, you might say. Couldn’t agree more, if it wasn’t for the fact that it might be useful to distinguish between those summarising and streamlining access to large datasets, and those set to unlock new data and tell meaningful and actionable stories. The latter is exactly what we have in mind. There’s only so much value in good data if this fails to provide a clear sense of direction, highlighting new and undetected patterns and, more importantly, if such tools are not accessible or useful to intended users.

SDF has extensive expertise in building such tools with users in mind. Some examples include pioneering COVID-19 dashboards, highlighting the effects of the pandemic on people’s finances in real time:

These maps show a significant reduction of income across most of the UK compared with pre-lockdown periods during the early stages of the pandemic in 2020. In many cases, this was in line with the 80% payment rate issued as part of the Furlough Scheme. There are exceptions to these in some areas, most notably Greater London. However, data shows that by August 2020, most areas are in recovery as compared to a pre-lockdown period as restrictions were lifted before the second wave later in that year.

What is evidenced is the distinct drop in all types of expenditure for the younger and poorest groups at the start of the pandemic, with the later recovery seen across other income bands not fully mirrored by this set. This is less apparent in the older age bands, with the expectation that the relatively increased stability of income (based on pension rather than salary or benefits) allows that bracket to better weather financial shocks. Interestingly, in the wealthier brackets, expenditure as a whole increases despite initial dips at the onset of the first lockdown.

Finally, SDF can apply this analysis in specific settings, such as in their award-winning collaboration with East Renfrewshire Council building a cost-of-living dashboard. Linking financial to administrative data at a data zone level, SDF have been able to create a number of indicators of financial distress that can then be used by the council to better target support to those who need it most.

2) I don’t care about fancy tools, give me the data!

Having a data science background, I’m fully sympathetic to those wanting to get access to the underlying data. And I truly believe there is great value in that. We don’t have all the answers and, most certainly, we might not come up with all the right questions.

By allowing researchers and analysts to access the raw data, we want to enable others to discover patterns we may miss. Furthermore, our initial focus will be on income volatility and economic security, we are conscious that there are multiple dimensions revolving around poverty, hardship and exclusion that could and should be brought to light.

By jointly reviewing proposals with SDF, we will make available around 20 licenses for interested parties willing to use this data for research and analysis that can truly contribute to social change.

We are already aware of some good ideas and potential applications. For example, analysis of the effects of changes in the macroeconomy (e.g., inflation, raising interest rates) on the incomes of lower-paid individuals and the dynamics of the economic inequality more generally; or the effects of high-interest rates on household finances and financial vulnerability with rising mortgage rates and rents from the end of 2021 onwards; research into understanding the drivers and risks of financial vulnerability based on banking transactions.

The potential of this data is significant due to its uniqueness, accuracy, national coverage and, most importantly, its near real-time access allowing researchers to start observing the changing human behaviour in response to events as they occur as well as to go back in time to track our lives over the years, ultimately providing key insights into a changing society.

3. So, who is this for?

There’s a final point I believe it’s important to emphasise. This partnership, the tools, analysis outputs, and how we work with researchers will evolve and improve over time. It is true we have specific audiences in mind for the project’s first iteration (i.e., researchers and academia, think tanks, policy, and advocacy organisations), however, as enablers, we want to improve the number of organisations accessing this data for social good, as well as its potential applications.

In addition to users and consumers of data, we are also keen for the financial institutions providing the data in the first place to increase in number. There’s plenty of great work being done under their corporate responsibility banner but this project has the potential to bring their contributions to long-lasting positive social change to a whole new level. More data means more insights could be generated; our understanding of society will improve; and closer to real-time information will help how, individually and collectively, we can make better data-driven decisions; plan service provision; and provide effective advice.

Whether you feel this project speaks to you or are simply curious, more will be revealed at our upcoming webinar on 21st February. You can sign up here: Unlocking the Potential of Banking Data for Good