Cash Access Deserts: Mapping Financial Infrastructure Gaps
I stood outside a shuttered bank branch in Rotherham last winter. The ATM had been ripped out six months prior. A pensioner named Doris told me she now takes two buses — forty-seven minutes each way — just to withdraw her own money. Two buses. For her own cash. This is not a footnote in a banking report. This is a Tuesday.
Cash is not dead. It is dying in some places and thriving in others, and the difference is not about progress — it is about power. Banks decide where the machines go. They decide based on spreadsheets, not on Doris. The result is cash access deserts: neighborhoods, towns, whole postcodes where the nearest ATM is a hike, a bus ride, a prayer. The elderly walk further. Small businesses turn away sales because they cannot make change. The unbanked pay predators in corner shops to hold their money. This is not market efficiency. This is market abandonment.
Why deserts form
Banks close branches the way landlords evict tenants: with a letter, a date, and zero conversation. Marginally profitable becomes unprofitable becomes closed. ATM operators follow the same logic. Low transaction volume? Rip it out. Move it to the shopping center where the footfall is.
What is left behind is predictable. Residential zones. Rural roads. Estates where the average account balance does not excite a quarterly earnings call. The machines cluster where the money already is, and vanish where it is not. This is not a glitch. This is the design.
Here is what kills me: the banks think they know the map. They know their own branches. Their own ATMs. Their own transaction data. But they do not know what is across the street. They do not know the competitor machine that broke last month. They do not know the bus route that was cancelled, turning a ten-minute walk into an hour. They optimize on partial data, and partial data produces broken neighborhoods.
Regulators try. Mandates. Incentives. Access-to-cash schemes. But enforcement needs measurement, and measurement needs a map that does not exist. You cannot regulate what you cannot see.
Mapping access with field data
This is where we come in. I have sent contributors into deserts that do not appear on any official map. We walk the streets. We photograph the empty wall where the ATM used to be. We time the walk to the nearest working machine. We note the broken pavement, the unlit underpass, the bus that comes every ninety minutes if it comes at all.
The metric that matters is not radius. It is reality. A machine two kilometers away across a motorway is not accessible. A machine inside a shopping center that closes at six is not accessible to a night-shift worker. Geometry lies. Feet do not.
Our people record the transport links, the barriers, the opening hours, the state of the pavement. We build a map of what it actually takes to get cash, not what it looks like on a spreadsheet. The difference is staggering. I have seen official maps show "adequate access" in zones where residents travel over an hour. I have seen "low priority" labels on estates where every shop is cash-only and every ATM charges £1.99.
Identifying intervention targets
Once you have the real map, the solutions stop being theoretical. Areas with no access within fifteen minutes are critical. Fix them now. Areas with a single machine are fragile — one breakdown away from crisis. Areas with multiple machines but predatory fees have nominal access, not real access. The distinction matters because the intervention changes.
Sometimes the fix is a new ATM. Sometimes it is a shared banking hub in a post office. Sometimes — and this is where the bankers look at me funny — the fix is not a bank at all. It is cashback at the corner shop. It is a mobile van on market day. It is accepting that the old model is dead and the new one has not been born yet.
Field data tells you which fix fits which hole. Without it, you are throwing darts at a board you cannot see.
The regulatory opportunity
For regulators, a field-validated map changes everything. Instead of reacting to closure announcements with angry letters, you see the desert forming before the sand settles. You intervene early. You measure whether the intervention worked. A new ATM should cut access time. If it does not, you picked the wrong corner. The data tells you. The data does not lie.
I have watched too many policy documents gather dust while real people walk too far for too little. The tools exist. The network exists. The only question is whether anyone with authority cares enough to look at the ground instead of the slide deck.
Request a pilot →