Evaluation Scotland Wales
The UK Strategy for Financial Wellbeing is taking forward the work of the Financial Capability Strategy Opens in a new window


Borrower subgroups and the path into distress

Evidence type: Insight i


In recent years there have been increased concerns about individual borrowers falling into financial distress, against a backdrop of rising household leverage during a period of falling interest rates. The Financial Conduct Authority’s (FCA’s) 2017 Financial Lives Survey showed that 15% of UK adults had recently missed or were struggling to make domestic bill and credit repayments. While numerous studies have investigated both how the composition of consumer debt varies across individuals and the experiences of those in financial distress, limitations associated with the survey datasets used have made it hard to fully explore the granularity of these patterns.

The study

This 2020 report from the Financial Conduct Authority uses a rich and detailed administrative dataset which contains the credit files of a large and representative sample of UK borrowers to fully investigate the composition of consumer debt.

The panel dataset that is used is based on 1% of individuals with a credit file at a UK Credit Reference Agency (CRA) between January 2015 - February 2018. CRA data covers the vast majority of credit users, as well as most consumers that have a current account or a contract with a utilities or telecommunications provider. The data included detailed monthly information about peoples’ credit portfolios, including types of debt, outstanding balances and details of any arrears, as well as balances and turnover. After excluding certain individuals from the dataset (for example, those who were debt-free or who were not responsible for the bills), a final sample of 428,097 individuals is included in the analysis.

A statistical technique known as cluster analysis is used to divide credit users into distinct groups based on observed characteristics and debt usage. The following variables are used to define the clusters:

  • Total debt balances;
  • First and second mortgages as a proportion of total debt;
  • Standard-cost credit as a proportion of total debt;
  • High-cost credit as a proportion of total debt;
  • Outstanding household bills as a proportion of total debt;
  • Credit score;
  • Age;
  • Total monthly incomings into current accounts.

The cluster analysis identifies four data-driven subgroups of borrowers:

  1. Mortgage holders (38% of individuals);
  2. Standard-cost borrowers (34%);
  3. High-cost borrowers (18%);
  4. Those with household bills only (4%).

For the purpose of the analysis, financial distress is defined as at least one of the following happening in a given month:

  • Arrears of 90 days (or a default) on any credit card or bill;
  • A County Court Judgement (CCJ) is issued;
  • A credit account is passed to a debt collector;
  • An individual is declared bankrupt.

Key findings

  • Incidences of financial distress vary considerably between the four clusters. Almost one-in-five high-cost borrowers (18%) fell into financial distress over the period that the data was recorded, compared to 8% of standard-cost borrowers, 6% of mortgage holders and just 5% of those with household bills only.
  • Those who go on to experience financial distress share common characteristics regardless of the cluster they are in, including being typically younger and on lower incomes, with lower credit scores and holding more expensive forms of debt.
  • For mortgage holders, those that fall into distress actually have slightly lower debt balances than the non-distressed because they have slightly smaller mortgages on average.
  • Mortgage holders and standard-cost borrowers both have substantial amounts of unused credit that can be used for any form of spending, in contrast to high-cost borrowers and the household bills group.
  • Unlike other clusters, the income of high-cost borrowers is slightly higher for the distressed than for the non-distressed, possibly partly reflecting the differing availability of credit to individuals with different characteristics, as well as differences in demand.
  • The authors state that their work informs discussions about whether individuals likely to fall into distress in the future can be identified in advance, and acts as a catalyst for future discussions and research in this area.

Points to consider

  • Methodological strengths/weaknesses: The methodology used appears sound and the findings are tested for statistical significance, so we can have a fair degree of confidence in the robustness of the findings.
  • While methodological details are scarce, much of the analysis presented is from secondary resources that are robust and reliable (for example the Office for National Statistics).
  • Generalisability/ transferability: This report is of significant interest to creditors, stakeholders and policymakers in the field of consumer borrowing, and of particular relevance to those looking to understand debt patterns among subgroups of the population.
  • Relevance: The findings are situated in a UK context, though some of the learnings may be transferrable to countries with similar financial regulatory environments.

Key info

Year of publication
United Kingdom
Contact information

Adiya Belgibayeva, Karen Croxson, Zanna Iscenko, Jesse Leary and Jonathan Shaw

Financial Conduct Authority