Understanding Changes to Access to Cash in Canada

The Bank of Canada has issued a Staff Discussion Paper that focuses on the retail side of the cash distribution system examining changes in cash accessibility as the result of the COVID-19 pandemic ¹. It found that 97% of Canadians had access to an Automatic Banknote Machine (ABM) and that this had not changed during the pandemic.

Cash usage has been falling for a while in Canada. In late 2020, 20% of transactions were in cash and 40% of low value transactions, those under $15, used cash.

One of the risks of lower volumes is a decline in the economies of scale leading to higher costs and in turn to a rationalisation of the cash infrastructure. Given that cash is in great demand in a crisis, and that once the network reduces it is hard to regenerate it, understanding changes to access to cash is important.

Methodology: density-based metric

Every bank branch has at least one ABM and the location of branches is known. In this study these are referred to as Financial Institution (FI) ABMs whether they are in the branch or separate from the branch. There are also ABMs operated by Independent Service Operators and these are known as White Label (WL) ABMs. Again, their locations are known.

Since ABMs are how most Canadians get access to cash, tracking access to ABMs is seen by the Bank of Canada as a reliable measure of access to cash.

The last Canadian census was in 2016. The census results are logged in census sub-divisions (CSDs), with the CSDs representing municipalities or comparable areas for statistical reporting and aggregation. There are 5,100 CSDs in Canada. Urban populations are defined as having more than 1,000 people and a population density of 400 per km² or more. Anything different is regarded as rural. A small population has 1,000-29,999 people, medium 30,000-99,999, and large 100,000 or over.

WL ABMs generally have more limited functionality than FI ABMs. They seldom accept cash deposits, have lower withdrawal limits and distribute a more limited range of notes. They also usually charge network access and convenience fees.

Results

Between November 2019 and February 2022, the number of FI ABMs fell 1.06%. Between February 2020 and August 2020 WL ABMs fell 7.79%, but by February 2022 had recovered to be slightly more than before the pandemic.

In February 2022 about 30% of CSDs had at least one FI ABM, 23% had only a WL ABM and nearly 50% had no ABMs. The percentage distribution remained fairly stable over the course of the pandemic.

In terms of the percentage of the population with an ABM in a residential CSD, 92% had at least one FI ABM, 5% had only a WL ABM and 3% no ABM at all. Again, this was stable through the pandemic.

The CSDs without an ABM were almost all rural. 90% of small urban centres had one ABM but half of rural CSDs had no ABM, 25% an FI ABM and 20% only a WL ABM.

When the results were studied by the percentage of the population in each type of CSD, all large and medium CSDs had one ABM as did the vast majority of rural CSDs. 78% of rural CSDs had at least one FI ABM, 14% only a WL ABM and 8% had no ABM at all.

Study limitations

What this approach does not take into account is ‘cash-accessing deserts’, where low income people or minority neighbourhoods do not have access to ABMs.

No account is taken of other means of getting cash, such as cash-back in shops.

This approach pays no regard to the geographical distribution of ABMs and consumers within CSDs. Data shows that ABMs tend to be clustered together and so this approach can over-state access. It also pays no attention to the proximity of people to ABMs in neighbouring CSDs.

The Bank of Canada is now looking at further work to create a distance-based metric. This would look at the travel distance between households, whether in or outside of the CSD. Even this spatial density metric would still not record either the capability of the ABM or their fees. Can the ABM accept cash deposits, what is the access to the ABM inside a branch that is not always open etc?

Equally it is well known that charging fees changes consumer behaviour. There is a tendency to make fewer withdrawals but to take out more money. If there is a shift to WL ABMs, such a change in behaviour would reduce convenience fee income, which could undermine the economics of the WL ABM.

Conclusion

97% of Canadians have access to cash in their communities, but for rural communities this is 92%. The pandemic had no adverse effect on these figures and the ability of Canadians to access cash.

1 - ‘Canadians’ Access to Cash Before and During the COVID-19 Pandemic’. Heng Chen, Marie-Hélène Felt. Staff Discussion Paper 2022-15.