A field experiment on overdraft alerts

• Description of the programme
• The study
• Key findings
• Points to consider

Description of the programme

Although there has been rapid growth in consumer engagement with digital banking, as well as a growing number of money management solutions and applications, a substantial percentage of the UK population incur overdraft and unpaid item charges. It is estimated that 19 million people use their overdraft each year, while firms made £2.3 billion from overdrafts in 2016. While in some cases these charges can reflect a demand for conveniently accessed credit, it is likely that some of the charges could be avoided if consumers had been more aware of their financial position. Research from the FCA in 2018 found that sending consumers a text message alert just before they incur charges for unarranged overdraft usage or unpaid items reduces these charges by 21-25%.

Despite this, few people voluntarily signed up for the alerts, with only 3-8% registering for any type of alert by 2015. However, more recently all major UK banks have enrolled their customers to receive ‘just-in-time’ alerts, if not by their own accord then by Government policy that mandated enrolment (‘auto-enrolment’) in 2018.

This 2018 evaluation from the FCA, collaborating with two UK financial institutions, tests whether consumers would benefit from:

• Just-in-time alerts for arranged overdraft use;
• Early warning alerts for arranged and unarranged overdraft usage;
• Early warning alerts for unpaid items.

The study

The main aim was to assess the effect of messaging in the reduction of total overdraft charges, as well as analyse secondary outcomes such as digital banking usage, balances, transaction patterns and the length of time people were overdrawn.

The field trial was conducted in conjunction with two major UK retail banks and involved data from over one million personal current account (PCA) consumers between November 2017 and February 2018 (the date auto-enrolment became mandatory). There were four separate trials:

1. Trial A provided unarranged overdraft and unpaid item alerts for all consumers, against a control group that received no alerts.
2. Trial B involved a low balance alert for consumers with unarranged overdrafts only.
3. Trial C involved a low balance alert for consumers with no overdraft facility.
4. Trial D involved low balance and usage alerts for consumers with both arranged and unarranged overdrafts.

The control groups for trials B, C and D were also enrolled into the alerts tested in Trial A, but did not receive the extra interventions detailed above.

A further telephone survey was also conducted with a sub-sample of 4,007 clients (2,956 from the treatment group, and 1,051 from the control groups), to gauge the effects of alerts on awareness, measure attitudes towards automatic enrolment and learn more about actions taken after receiving the alert.

Key findings

• Trial A – the average consumer sees a reduction of between 13-18% in unarranged overdraft and unpaid item charges when enrolled into general alerts. This is equivalent to between £0.39 and £0.46 per month.
• Trial B – There was no statistically significant evidence that low balance alerts help consumers with unarranged overdrafts avoid using their overdraft.
• Trial C - There was no statistically significant evidence that low balance alerts help consumers without any overdraft facility reduce the charges levied on them. When consumers are encouraged to self-register for these alerts the registration rate is less than 10%.
• Trial D - The average consumer sees a reduction of 3-8% in arranged overdraft charges when enrolled into alerts that warn of arranged overdraft usage in real time, equating to between £0.28 and £0.45 per month. However, enrolling consumers into low balance alerts did not lead to a further reduction, and there was no effect on charges of notifying consumers who were approaching their arranged overdraft limit.
• The survey suggested that consumers overwhelmingly relied on their own liquid savings, cuts to non-essential spending and informal credit to avoid overdrafts.
• Survey respondents did not find the alerts distracting or annoying, with no evidence of ‘alert fatigue’.
• The research provides support for automatic enrolment of consumers into further alerts regarding their overdrafts. However, the evidence in support of low balance alerts is weak.
• The authors conclude by stating how modern digital approaches to interventions can allow randomisation and implementation to happen relatively easily, allowing the scaling up of similar experiments.

Points to consider

• Methodological strengths/weaknesses: Findings are tested for statistical significance, adding more robustness to the analysis.
• Control groups are used, allowing findings from the intervention group to be compared with groups not receiving the same intervention for benchmarking purposes.
• Generalisability/ transferability: The evaluation is of significant interest to those looking to commission or authorise interventions designed to improve consumer actions regarding their personal debts, and in particular their overdrafts.
• Relevance: The findings are based on consumers from two major UK financial institutions and should not necessarily be seen as representative of the entire market.