The UK's Non-Bank Lending and Private Credit: A Comprehensive Analysis

The UK's Private Credit Landscape

The United Kingdom is the deepest consumer credit market in this series, measured per head, the most liquid, and the one where the risk most often sits somewhere the reader is not looking.

The headline numbers are reassuring. Households owed £254.8 billion of consumer credit in May 2026, excluding student loans. Arrears on that credit are around 1%, low by historical standards, according to the Bank of England. The sovereign is rated AA by S&P. Contracts are enforceable, insolvency is predictable, and the securitisation market is the deepest in Europe.

And yet the UK's non-bank lenders have a habit of failing with performing loan books. Wonga entered administration in August 2018 with a gross loan book of £77 million and, within months, more than 40,000 redress claims. QuickQuid followed in October 2019; its wronged borrowers recovered 53.5 pence in the pound. Amigo stopped lending in 2020 and paid out £194 million in cash redress and refunds while winding down. Provident closed its doorstep lending business in 2021.

None of these lenders was killed by borrowers who could not pay. All of them were killed by borrowers who had paid, and then complained.

That is the mechanism this report is built around. In the UK, a consumer loan does not stop being a liability when it is repaid. The Consumer Credit Act's unfair relationship test, the binding jurisdiction of the Financial Ombudsman Service and a professional claims industry together allow past conduct to be repriced years after the fact, and in bulk. The largest example is the £9.1 billion motor finance redress scheme, which reaches back to agreements signed in April 2007 and whose legal fate will be argued before the Upper Tribunal between 16 and 26 February 2027.

For a private credit investor, this changes what diligence is for. Credit risk in UK consumer lending is well understood, well modelled and, for now, benign. Conduct risk is retrospective, correlated across lenders and invisible in arrears data. The investable question is not only whether these borrowers will repay. It is what this loan book will later be judged to have owed them.

Role of Non-Bank Financial Institutions

The UK has no single legal term for non-bank lenders. The industry says specialist lenders, the regulator says consumer credit firms, and the statistics mostly break lending down by product rather than by lender type. Whatever the label, the footprint is large.

• Consumer and motor finance. Members of the Finance & Leasing Association (FLA) wrote a record £162.8 billion of new finance in 2025, including £122.5 billion of consumer credit, almost a third of all new consumer credit in the UK. Non-bank lenders wrote £34.4 billion of the £84.3 billion FLA members lent in the first half of 2026, about 41%.
• SME lending. 68% of 2025 SME lending came from challenger and specialist banks or non-bank lenders. Challenger and specialist banks alone held 60% of gross SME bank lending, up from 39% in 2012.
• Specialist mortgages and bridging. Non-bank lenders such as Together, with an £8.4 billion loan book, run entirely on wholesale funding.

The definitional trap is that much of what the UK calls specialist or challenger lending sits inside banks. Close Brothers, Vanquis and Secure Trust are banks, funded by deposits and supervised by the Prudential Regulation Authority as well as the Financial Conduct Authority (FCA). A deposit-funded challenger bank is not a counterparty for a lender-finance facility. The investable universe is the wholesale-funded part of the market: digital consumer lenders, independent motor finance houses, buy now, pay later (BNPL) providers, specialist mortgage and bridging lenders, and SME fintechs.

Focus of This Report

This report looks at UK non-bank lenders as counterparties for private credit: consumer, motor, BNPL and SME lenders that fund themselves through warehouse facilities, forward-flow agreements, securitisation and bonds. It does not cover sponsor-backed corporate direct lending, which is a different market with different risks.

Key Takeaways

• Big, growing and late-cycle. Consumer credit grew 9.8% in the year to September 2026, roughly two and a half times nominal GDP growth, while three of nine MPC members are voting for a rate rise.
• Default is not the main loss channel. The UK's non-bank failures since 2018 came from redress on historic lending, not from borrowers who stopped paying.
• The largest test is still pending. The £9.1 billion motor finance scheme is partly suspended, and payments are unlikely before 2027.
• The tail is being bounded, not removed. The High Court has held the ombudsman to its six-year time limit, and a bill before Parliament would add a 10-year absolute limit and remove automatic unenforceability sanctions. The unfair relationship test stays.
• New perimeter. BNPL became regulated credit on 15 July 2026.
• The opportunity is below the securitisation threshold. Large lenders fund in public markets at tight spreads. Sub-scale lenders, and new originations ring-fenced from legacy books, are where an institutional facility adds value.

The Big Picture

Table 1 — Core Indicators

Sources: ONS; Bank of England; Trading Economics; World Bank; rating agencies.

The economy is growing slowly, and the national accounts have just been restated. Figures consistent with Blue Book 2026 put 2025 growth at 1.2%, down from 1.3%, which was itself a downward revision from 1.4%. Growth in 2024 was revised up to 1.1%. Unlike in some markets in this series, the revisions do not change the story: real GDP in the second quarter of 2026 was 2.0% above its level at the end of 2024.

Inflation is the problem. CPI rose to 3.5% in September 2026, from 3.1% in August, its highest rate since October 2025, with motor fuels the largest driver. Core inflation rose to 2.8% and services to 3.7%. The Bank of England expects headline CPI to reach around 3.75% in the fourth quarter and slightly above 4% in the first quarter of 2027, as a protracted conflict in the Middle East feeds through energy prices. Regular pay growth of 3.7% leaves real wages roughly flat.

Table 2 — Monetary and Funding Conditions

Effective rates are Bank of England data for September 2026, except overdrafts (August). Spread data from KPMG's Q4 2025 market update.

The direction of the next rate move is genuinely open. Support for a rise has built meeting by meeting: one vote in April, two in June, and three in both July and September, while the majority pointed to slack in the economy. Elsewhere the cycle has already turned: the Federal Reserve, the ECB and the Bank of Japan all raised rates within nine days of each other in September, and sterling swap markets now price rises rather than cuts. The September inflation print hardened expectations of a rise at the 5 November meeting. For a lender funding at floating rates against fixed-rate consumer assets, a rise compresses margins. For an investor in a floating-rate facility, it raises the coupon. Either way, the pricing gap remains wide: new personal loans at close to 10%, credit cards above 21%, against a policy rate of 3.75%.

Table 3 — Credit Depth

Sources: Bank of England Money and Credit; The Money Charity (May 2026 stock); ONS; FCA; FLA.

Consumer credit is growing at roughly two and a half times the pace of nominal GDP. August's net flow, revised up to £2.7 billion, was a record monthly figure, and September's £1.9 billion was close to the six-month average. Card borrowing, the most expensive and most discretionary product, is growing fastest at 13.1%.

Investor implication: fast book growth flatters arrears ratios, because the denominator grows before new loans have had time to season. A lender reporting stable arrears in a year of 20% book growth has told you less than it appears to. Ask for vintage curves, not portfolio ratios.

Structure and Size of the Non-Bank Sector

The Definitional Trap

There is no official figure for the UK's non-bank share of consumer credit. The Bank of England's monthly release breaks consumer credit down by product, not by whether the lender takes deposits. The broadest official measure covers all lending to households and companies: banks' share of that stock fell from 85% in 2008 to around 80% in 2024. The two usable proxies come from trade and policy bodies, on different bases.

Table 4 — FLA Member New Lending and the Non-Bank Share

Source: Finance & Leasing Association. Covers asset, consumer and motor finance. FLA members wrote £162.8bn in 2025 in total; the FLA's full-year 2025 release prints the non-bank figure as "£2 billion", an evident typo, so 2025 is not shown.

Two cautions. First, FLA data measures new lending, not outstanding stock, and covers only FLA members, who write about a third of new consumer credit. Second, the FLA's non-bank category includes the captive finance arms of vehicle manufacturers. These are subsidiaries of global industrial groups, not independent lenders, and three of them are among the challengers to the FCA's motor finance scheme.

Where Non-Banks Actually Lend

Table 5 — Segment Map

Sources: Bank of England; FLA; FCA; British Business Bank; company disclosures. Mixed vintages, as shown.

The map shows where the investable non-bank activity is concentrated: digital unsecured lending, independent motor finance, BNPL, specialist property and SME credit. These are also, not coincidentally, the segments where the redress question has been most active.

BNPL is the fastest structural change. Deferred payment credit grew from £60 million in 2017 to more than £13 billion in 2024, and a fifth of UK adults used it in the year to May 2024. It became regulated credit on 15 July 2026.

Pricing

No General Rate Cap, and One Very Specific One

The UK has no usury ceiling for mainstream consumer credit. The only hard price cap applies to high-cost short-term credit (HCSTC), defined as credit with an APR of 100% or more and a term of up to 12 months. Since 2 January 2015 it has had three parts:

• interest and fees of no more than 0.8% a day of the amount borrowed;
• default fees capped at £15;
• total charges capped at 100% of the amount borrowed.

Before the cap, rates ran as high as 4% a day, with a typical customer taking six loans a year. On 6 October 2026 the FCA confirmed it would keep the cap at its current level and would not proceed to the second stage of its review, choosing instead to monitor the market through new product sales data.

Outside HCSTC, prices are set by the market. Personal loans averaged 9.85% in September 2026 and interest-charging cards 21.71%. Near-prime and subprime products price much higher: Amigo charged up to 49.9%.

The Price That Gets Set Afterwards

The UK's binding price control is not a ceiling at origination. It is a judgement made after the fact.

The motor finance scheme shows exactly how that judgement is calculated. The FCA's economic analysis found that discretionary commission arrangements raised the APRs borrowers paid. It turned that finding into an "APR adjustment" of 17% for agreements from April 2014 and 21% for earlier ones. Most eligible consumers receive the average of that estimated loss and the commission paid, plus interest at Bank Rate plus 1% with a 3% floor. Around 90,000 cases closest to the Supreme Court's Johnson judgment receive all commission plus interest. In about one case in three, awards are capped.

That is a regulator repricing a 17-year back book, agreement by agreement, using a formula.

Table 6 — How UK Price Controls Work

Investor implication: a UK consumer lender's model should carry expected conduct cost per account as a line item next to expected credit loss. A lender that cannot show you that line has not priced its own largest risk.

{{cta-component}}

Regulatory Framework

1. Statute, Supervisor and Ombudsman

Consumer credit sits under two statutes: the Financial Services and Markets Act 2000 (FSMA), which governs authorisation and supervision, and the Consumer Credit Act 1974 (CCA), which still sets many of the rights attached to individual agreements. The FCA took over consumer credit regulation from the Office of Fair Trading on 1 April 2014. Deposit-taking lenders are also supervised by the Prudential Regulation Authority. Disputes go to the Financial Ombudsman Service, whose decisions bind firms and which is free for consumers.

2. Licensing and the New BNPL Perimeter

Lending to consumers requires FCA authorisation. From 15 July 2026 this extends to third-party lenders offering deferred payment credit, the regulatory term for interest-free BNPL. Lenders must either be authorised or hold a temporary permission, and those on temporary permissions have six months to apply for full authorisation. Credit provided directly by a merchant stays outside the regime, and agreements made before 15 July 2026 do not acquire the new protections. Complaints about agreements made from that date can go to the ombudsman, which expects about 2,000 in the second half of 2026/27. Experian's bureau data, which covers providers that report to it, shows 98.5% of 2025 BNPL balances repaid on time.

3. Conduct Rules

Three layers apply. The FCA's consumer credit sourcebook sets affordability and creditworthiness requirements. The Consumer Duty requires firms to deliver good outcomes and fair value. And the HCSTC price cap applies to the highest-cost segment.

4. Unfair Relationships and the Supreme Court

Section 140A of the CCA lets a court reopen a credit agreement if the relationship between lender and borrower is unfair, judged on everything that happened, including after the agreement was signed. In October 2024 the Court of Appeal found that car dealers acting as credit brokers could not lawfully receive lender commission without the customer's informed consent, which froze the auto ABS market. In August 2025 the Supreme Court sided with lenders on two of the three appeals but upheld the finding of an unfair relationship in the Johnson case, where an undisclosed commission was very high. Auto securitisation restarted soon afterwards.

5. The Redress Machine

Table 7 — Financial Ombudsman Service Complaints

Source: Financial Ombudsman Service annual reports and complaints data. Financial years run April to March. The 2024/25 total of 305,726 covers complaints about products; the annual report's 305,918 adds 192 about claims management companies.

The ombudsman's caseload more than doubled for irresponsible lending in 2024/25, then fell sharply. The cause of the fall is instructive. From 1 April 2025 the service began charging claims management companies and other professional representatives £250 per case after their first ten, with £175 credited back if the consumer wins. In the following quarter, new irresponsible lending complaints fell to 4,500, against 24,900 a year earlier. Over the full year, professional representatives' share of the caseload fell from half to 18%. Motor commission complaints, paused while the FCA built its scheme, fell to 400 in October–December 2025 against 14,400 a year earlier. The ombudsman expects about 188,000 new cases in 2026/27.

From that low base, the flow has started to edge up again. The ombudsman received 57,000 complaints in July–September 2026, about 3,400 more than in the previous quarter, and credit card complaints rose to 6,400 from 4,700 a year earlier. It links part of the rise to cost-of-living pressure and to consumers using AI tools to bring complaints themselves. Professional representatives brought only 1,800 of the 53,600 cases in April–June. The claims channel is shifting from paid intermediaries to consumers, and the case fee does not reach them.

The courts are drawing lines too. On 24 June 2026, in Barclays and others v Financial Ombudsman Service, the High Court quashed four ombudsman decisions that had read earlier Supreme Court judgments as creating a continuing duty to correct unfairness, which would have let unaffordable-lending complaints reach back beyond the six-year time limit. The FCA intervened on the lenders' side, the first time it has done so against the ombudsman.

The lesson for an investor is that complaint volume is not a fixed property of a loan book. It is a function of the economics of claims management and the reach of the time limits, both of which policy and the courts can change in either direction.

6. The Motor Finance Redress Scheme

Table 8 — The Scheme at a Glance

Sources: FCA PS26/3 and statements; trade press.

The 2007 start date is not arbitrary: that is when consumer credit complaints came within the ombudsman's reach. Most of the industry, including Lloyds, Barclays, Santander and Close Brothers, chose not to challenge. Provisions are large and concentrated in banks: Lloyds, which owns Black Horse, the UK's largest motor finance lender, held £1.95 billion, unchanged at its third-quarter results on 29 October 2026; Barclays £430 million, also unchanged in the third quarter; Close Brothers about £320 million; Bank of Ireland €429 million at December 2025. Vanquis, which did not use discretionary commission arrangements, holds £3 million.

If the challenges succeed, liability does not disappear. The FCA estimates that resolving the same complaints individually, through firms, the ombudsman and the courts, would cost more than £6 billion more than the scheme, and it has told lenders to plan for the possibility of no scheme and no pause in complaint handling.

7. What Changes Next

The Financial Services and Markets Bill, announced in the King's Speech on 13 May 2026, passed the House of Lords on 15 September and is now before the House of Commons. It carries two reforms that matter here. Neither is law yet. A third change sits outside the bill.

1. Ombudsman reform. A 10-year absolute time limit for bringing a complaint, counted from the firm's act or omission, with exceptions set by the FCA. The "fair and reasonable" test would be met where a firm complied with the relevant FCA rules. Two changes are already in force from 1 October 2026: wider grounds to dismiss complaints, and an amended rule (DISP 3.6.4R) confirming that decisions apply the standards in force at the time of the act complained about, not today's.
2. CCA reform. Most of the Act's prescriptive information requirements move into FCA rules, and the automatic sanctions attached to them, unenforceability and loss of interest and default charges, are repealed. The unfair relationship provisions in sections 140A–C, along with sections 56 and 75, stay in legislation for now pending further policy work.
3. Claims management. The FCA opened a market study into claims management services on 19 May 2026, with a final report due by 19 May 2027. It has said that, "motor finance aside, there are no other mass redress events on our radar".

Kilde's view: the reforms bound the future, not the past. A 10-year longstop means a loan written today can still be complained about in 2036, and the unfair relationship test survives untouched. Repealing automatic unenforceability removes a technical-breach tail that has haunted securitisation lawyers for two decades, which is genuinely good for new originations. Legacy books remain exposed. In practice that argues for financing new, well-documented originations, and for buying back books only where the seller keeps the redress liability.

Credit Quality

Table 9 — Credit Quality Indicators

Measured the conventional way, UK consumer credit quality is good. Arrears are low, households are not stretched in aggregate, and the most vulnerable households hold a small share of the debt. The Bank of England's Financial Policy Committee judged households and businesses broadly resilient in July 2026.

What the headline hides:

1. Growth flatters the ratio. With consumer credit up 9.8% in a year, a large part of every lender's book is too young to have defaulted.
2. Redress is not arrears. None of the losses that ended Wonga, QuickQuid or Amigo would have shown up as arrears on the day the loans were written. Amigo set aside £345 million for complaints on a book that was paying.
3. The squeeze is ahead, not behind. If CPI passes 4% in early 2027 as the Bank expects, while pay growth stays near 3.7%, real incomes fall for the households that use the most expensive credit.
4. The US is a step ahead. The Bank of England notes that the share of US consumer credit more than 90 days in arrears has risen markedly over three years. UK arrears have not followed, but UK funding markets have already reacted to US credit events (see the funding stack below).

The falsifier for the benign credit view is a turn in employment. Unemployment eased to 4.7% in the three months to August, from 4.9%. If it turns up again while real pay turns negative, arrears on subprime cards and loans will move first, and quickly.

Major Players

No UK supervisor publishes a non-bank market-share table of the kind available in Poland or Estonia. The picture has to be built from company disclosures.

Table 10 — Selected Lenders

Sources: company disclosures, Experian data as reported, trade press.

Table 11 — Business Models and Investability

The largest non-banks are not the opportunity for a private lender. Lendable and Together already fund in public markets at spreads a facility cannot match. The investable middle is the lender with a book of tens to a few hundred million pounds: too small or too young for a rated term deal, too large for equity alone, and willing to pay for committed capacity.

The Funding Stack

The UK has the deepest non-bank funding market in Europe. UK placed securitisation issuance reached €38.2 billion in 2025, and UK RMBS, at €25.6 billion, was the largest single collateral category after pan-European CLOs. In the second quarter of 2026 alone, placed UK RMBS reached €12.4 billion. Auto ABS, frozen by the 2024 Court of Appeal ruling, came back after the Supreme Court: €2.8 billion across five deals by mid-May 2026.

Table 12 — The Funding Ladder

Sources: AFME; Morningstar DBRS; company announcements; Bank of England.

Three features of this stack matter for an institutional lender.

The public market is open but not frictionless. UK listed securitisation volume, including deals retained by their originators, fell 19% in 2025 to about £53 billion across 75 deals. UK AAA RMBS spreads widened to a peak of 56 bps after the collapses of First Brands and Tricolor in the United States. A lender that depends on refinancing a warehouse into term ABS is exposed to sentiment it does not control.

Scale is the gate. A 95% advance rate at under 1% over SONIA is available to a lender with an £8 billion book and a long record of public deals. It is not available to a lender with £80 million of two-year-old loans.

Ring-fencing is the product. A facility to a special purpose vehicle that buys only new, criteria-compliant loans separates the financing from the originator's legacy redress liabilities. In the UK, that separation is worth as much to the investor as the collateral itself.

Market Risks

Table 13 — Regulatory Risks

Table 14 — Credit, Funding and Macro Risks

The falsifier for this report's thesis. The argument is that conduct liability is the binding risk in UK non-bank lending and that it is becoming bounded. Two developments would break it: the Upper Tribunal quashing the motor scheme while the 10-year ombudsman limit fails to become law, or complaint volumes returning to 2024/25 levels despite the new case fees. Either would mean the liability is still open-ended, and every UK consumer facility would need to be priced for it.

{{cta-component}}

Opportunities for Private Credit Investors

Table 15 — Entry Routes and the Decisive Diligence Question

Route 4 already has precedents. Secure Trust Bank sold its vehicle finance business to funds managed by LCM Partners in 2026, a private credit manager buying a book a bank no longer wanted to hold.

Table 16 — Indicative Return Spectrum (Directional)

Directional only. UK pricing varies with security package, tenor, eligibility criteria, redress protections and issuer scale. Any yield quoted without a matching security and covenant description is not a usable number.

Kilde's view: the UK does not compete on yield. A senior secured facility here pays less than a comparable facility in a frontier market, and the conduct risk is real. What the UK offers instead is the strongest legal and data infrastructure in the series: enforceable contracts, predictable insolvency, deep credit bureau data and a live securitisation exit. Conduct risk in the UK is disclosed, litigated in public and increasingly bounded by statute. That makes it a risk that can be structured around, through eligibility criteria, ring-fenced vehicles and redress covenants, rather than merely priced. For a diversified private credit portfolio, the UK is a lower-volatility anchor, not a yield play.

How the UK Compares

Table 17 — The UK in Context

Sources: Kilde market analyses; Bank of England; FLA; national regulators. Figures are the latest available and not as at the same date. The UK per-capita figure uses the May 2026 stock and a GBP/EUR rate of about 1.20.

Among the markets covered in Kilde's market reports, the UK has about seven times Estonia's consumer credit per head and twice Spain's. It is one of two markets in the table without a general price cap, and the only one where the decisive price control operates after the loan is made. Poland offers volume under a known ceiling; Spain offers a reform-driven repricing; Estonia offers a euro-denominated platform into frontier markets. The UK offers depth and infrastructure, with a conduct liability that has to be engineered out of the structure rather than diligenced out of the borrower.

Conclusion

The UK's non-bank lending market is large, growing fast, well-funded at the top, and governed by a price control that operates in hindsight.

The numbers are clear. £254.8 billion of consumer credit, growing 9.8% a year. Arrears around 1%. A Bank Rate of 3.75% with three committee members voting to raise it, and inflation at 3.5% and heading above 4%. A non-bank share of about 41% of FLA-member new lending, and 68% of SME lending through challengers and non-banks. A securitisation market that placed €38.2 billion in 2025.

The distinctive risk is retrospective. The lenders that failed here failed on redress, not defaults, and the largest redress exercise in the market's history, £9.1 billion on motor finance, has yet to pay out. Its fate will be argued in February 2027. Reform will cap complaints at ten years and remove automatic unenforceability, but it will not close the unfair relationship test or touch existing books.

The distinctive opportunity follows from the risk. Large lenders fund themselves in public markets. The lenders below that threshold need committed capacity, and they need it in structures that separate new, well-documented originations from legacy liability. An institutional facility that provides both, with complaint-rate covenants and redress-aware eligibility criteria, is buying the best legal infrastructure in this series at a price that reflects a risk it has already structured away.

About Kilde

At Kilde, we specialise in secured private credit — and that is all we do.

Kilde is a regulated investment platform that helps accredited investors earn a predictable monthly income through high-yield, senior-secured private credit. We deal exclusively with licensed non-bank financial institutions that have a proven track record and strong financial performance. Each borrower is rigorously vetted using Kilde's proprietary credit scoring model, which analyses over 20 million data points on loans and repayments to ensure that only the most reliable counterparties receive funding.

Our investments typically have a maturity of 3 to 36 months, with collateral coverage exceeding 100%. Monthly cash coupons provide predictable income, while an early redemption option on select deals (typically on a rolling 3-month cycle) offers additional flexibility.

Strict financial covenants protect investor capital throughout the loan tenure, including limits on leverage ratios, interest coverage and repayment rates. Continuous monitoring of financial statements and borrower performance helps mitigate risk, while structured loan agreements under Singapore law provide a strong legal framework for enforcement.

Kilde bridges the traditional trade-off between wealth preservation and growth. Where conventional fixed income forces a choice between low-risk, low-return safety and higher-risk, higher-return vehicles, Kilde's private credit model delivers both: strong yields with disciplined risk management.

Sources

{{sources}}