FI Bank Barometer H2 2025: What the Lending Slowdown Means for EU Credit Risk Reporting

ByOfficer

Last updated: May 2026

On 25 May 2026, Finansinspektionen published its Bank Barometer for the second half of 2025. The headline: Swedish lending to the public grew just 0.9 per cent year-on-year in Q4 2025, down from 1.7 per cent in Q2 2025. That is a sharp deceleration in the space of two quarters. Net interest income fell. Major banks lost market share. The slowdown was driven primarily by reduced lending to non-financial corporations, not households.

If you file COREP credit risk templates for an EU bank with Nordic exposures, or you run credit risk reporting for any institution watching the European lending cycle, this data matters. A lending slowdown that starts in corporates eventually shows up in your exposure volumes, your risk weight calculations, and your IFRS 9 stage migration numbers. The question is not whether it hits reporting. The question is which templates first and how fast.

This article breaks down the FI Bank Barometer H2 2025 findings, explains where a lending deceleration flows into EU credit risk reporting, and identifies the specific COREP, FINREP, and Pillar 3 templates that reporting teams should watch.

Related reading: our COREP reporting guide

What the FI Bank Barometer H2 2025 Actually Shows

The FI Bank Barometer is a semi-annual descriptive report on the Swedish banking system. It covers lending volumes, profitability, funding, and market structure. It does not assess financial stability. That distinction matters because the report presents raw data without stability judgements, which makes it useful for reporting teams who need facts rather than commentary.

The key findings for H2 2025:

  • Lending to the public grew 0.9 per cent year-on-year in Q4 2025, down from 1.7 per cent in Q2 2025.
  • The deceleration came primarily from reduced lending to non-financial firms.
  • Swedish banks’ average profitability and net profit decreased, driven by lower net interest income and net financial income, plus higher costs.
  • Major banks continued to lose market share. Seven banks still account for over 80 per cent of lending to the public, but mortgage banks, savings banks, and consumer credit firms gained share.
  • Consumer credit lending, which represents about 4 per cent of total Swedish lending, grew fastest. Consumer credit firms drove most of that growth.

One detail teams commonly miss: the shift is structural, not just cyclical. Major Swedish banks are losing corporate lending volume to smaller institutions, while consumer credit firms are expanding at the fastest rate. That split affects how exposures distribute across exposure classes in COREP, and it changes the risk profile of the corporate loan book even before any borrower actually defaults.

Why Nordic Lending Data Matters for EU Credit Risk Reporting

Sweden is not a small market. Swedish banks such as SEB, Handelsbanken, and Swedbank, together with Finland-domiciled Nordea, operate across the Nordics and hold significant cross-border exposures within the EU. Several are designated as other systemically important institutions (O-SIIs) or fall under ECB direct supervision through their euro area subsidiaries.

For reporting teams outside Sweden, the relevance works through two channels.

First, direct exposure. If your institution lends to Swedish corporates, holds Swedish covered bonds, or has interbank exposures to Swedish counterparties, a lending slowdown in Sweden changes the credit quality assumptions feeding your own COREP submissions. That exposure shows up in C 09.01 and C 09.02 (geographical breakdown of exposures by residence of the obligor) if you breach the 10 per cent non-domestic exposure threshold for geographical reporting in Article 5 of the ITS on supervisory reporting (Regulation (EU) 2021/451).

Second, signal value. A deceleration in lending to non-financial corporates in a mature, well-capitalised banking system is a leading indicator. When I see corporate lending growth halve in two quarters in Sweden, my first question is whether the same pattern is forming in the broader EU corporate book. The ECB’s bank lending survey has already flagged a net tightening of credit standards on loans to firms. The FI data is consistent with that supply-side tightening: banks are lending less to corporates.

Reporting teams that treat national supervisory data as purely local are not wrong. But they miss the early signal. By the time the ECB aggregates capture the trend, your institution’s own exposure data already reflects it.

COREP Credit Risk Templates: Where the Slowdown Shows Up

A lending deceleration does not appear in one template. It ripples across several, and the timing depends on whether your institution uses the standardised approach or IRB.

C 07.00: Standardised Approach (CR SA)

Template C 07.00 captures original exposures, credit risk mitigation, and risk-weighted exposure amounts under the standardised approach. A corporate lending slowdown shows up here as declining original exposure amounts in the “Corporates” row, but only if the slowdown is large enough to offset new originations and drawdowns.

The common mistake: teams focus on the risk-weighted exposure amounts (RWEA) column without checking whether the shift is coming from exposure volume changes or risk weight reassignment. A reduction in performing corporate exposures with a 100 per cent risk weight, replaced by higher-risk consumer credit exposures at 75 per cent (retail) or 150 per cent (higher risk), changes your RWEA composition even if the total exposure stays flat. The Swedish data shows exactly this pattern: corporate lending shrinking while consumer credit grows.

C 08.01 to C 08.07: IRB Approach

For IRB banks, the picture is more granular. Templates C 08.01 through C 08.05 break exposures by IRB exposure class, PD grade, and LGD. C 08.07 reports the split between exposures under SA and IRB for banks using both approaches.

Where the lending slowdown matters for IRB: if corporate lending volumes drop, the PD distribution in your corporate portfolio can shift. Institutions that reduce new origination tend to keep their existing book, which ages. An aging corporate loan book, with fewer new high-quality originations diluting the pool, drifts toward higher average PD over time. That drift shows up in C 08.01 as a migration across PD bands.

The trap nobody talks about in training: a lending slowdown can improve your near-term RWEA numbers because you have fewer exposures, while quietly worsening your average PD. The next quarter, the RWEA benefit from lower volumes reverses as the PD migration catches up. I have seen teams celebrate a Q4 RWEA decrease without noticing the PD drift in the same submission.

C 09.01 and C 09.02: Geographical Breakdown

If your institution has exposures to Swedish obligors above that geographical-reporting threshold, C 09.01 (SA) and C 09.02 (IRB) require a geographical breakdown. A lending slowdown concentrated in Sweden means the Swedish country row in these templates shows declining exposure volumes, but potentially rising RWEA density if the remaining exposures are higher-risk.

This is where Nordic cross-border banking complicates things. A Swedish parent’s lending slowdown does not always mean the Finnish or Danish subsidiary sees the same pattern. Report the geography of the obligor, not the geography of the parent. Teams that consolidate Nordic exposures under “Sweden” because the parent is Swedish get the geographical breakdown wrong.

IFRS 9 Staging and FINREP: The Slower Burn

The FI Bank Barometer does not report IFRS 9 data directly. But a lending slowdown, combined with declining bank profitability and reduced net interest income, creates conditions where IFRS 9 expected credit loss (ECL) models need careful attention.

Stage Migration Under a Lending Slowdown

IFRS 9 requires institutions to assess whether credit risk has increased significantly since initial recognition. A lending deceleration to non-financial corporates does not automatically trigger Stage 2 migration. But it often correlates with deteriorating macroeconomic forecasts, which feed into forward-looking ECL models.

The real problem starts when the slowdown is sector-specific. Swedish corporates in real estate, construction, and SME services are more exposed to the lending cycle than exporters or utilities. If your IFRS 9 model uses sector-level PD adjustments, a broad-based lending slowdown can mask the concentration risk within sectors where the deterioration is sharper.

Teams commonly get this wrong by treating the lending slowdown as uniform across all corporate sub-segments. It is not. The FI data shows consumer credit growing while corporate lending shrinks. That divergence should show up in your staging analysis as differentiated treatment, not a single macro overlay applied to all exposures.

FINREP F 18.00: Performing and Non-Performing Exposures

Template F 18.00 captures the breakdown of financial assets by performing, Stage 2, and non-performing/Stage 3 status. A lending slowdown feeds into this template on a lag. Reduced new origination means the denominator (total gross carrying amount) grows more slowly, so even a modest increase in Stage 2 or Stage 3 exposures pushes the ratios up faster.

In FINREP F 06.01 (breakdown of loans and advances to non-financial corporations by NACE code), the sector split becomes visible. If corporate lending declines across Sweden and the broader Nordics, the NACE breakdown shows which sectors lost volume and which did not. That information should cross-check against your IFRS 9 staging: sectors losing lending volume should have their macro overlay assumptions reviewed.

CRR3 and the Output Floor: Why the Timing Matters

CRR3 applied from 1 January 2025 (it entered into force on 9 July 2024). The output floor, reported in C 02.00, ensures that IRB banks’ total risk exposure amounts do not fall below a percentage of what the standardised approach would produce. The transitional floor starts at 50 per cent in 2025, rising to 55 per cent in 2026, and reaching 72.5 per cent by 2030.

A lending slowdown interacts with the output floor in a way that is not obvious. IRB banks reducing corporate lending volumes see their IRB RWEA decrease. But the standardised RWEA used as the floor reference may not decrease at the same rate, because SA risk weights are fixed while IRB PD-based weights fluctuate. The result: a bank that shrinks its corporate book might find the output floor bites harder, not softer, because the IRB benefit from low PDs disappears while the SA floor reference stays proportionally larger.

Under Article 465(2) CRR, the transitional cap limits floored total risk exposure amounts to 125 per cent of unfloored amounts until 31 December 2029. A declining corporate portfolio that concentrates in higher-PD obligors can push the unfloored IRB RWEA up, which raises the 125 per cent cap ceiling, but the floor reference also rises if the SA-equivalent calculation assigns those exposures higher risk weights.

Reporting teams filing C 02.00 need to model both directions. The mistake is to assume that less lending means less capital. Under CRR3, less lending of the right kind (low-PD corporates) can mean more capital relative to the remaining portfolio.

What Reporting Teams Should Do Now

The FI Bank Barometer is one data point from one jurisdiction. It does not trigger any reporting obligation by itself. But it provides concrete numbers that should inform how you review your own submissions.

Check Your Nordic Exposure Concentration

Pull your C 09.01/C 09.02 geographical breakdown for Swedish obligors. Compare the Q4 2025 exposure volume to Q2 2025. If the trend mirrors the FI data, your credit risk team should already know. If they do not, the reporting team found it first. That conversation needs to happen.

Review IFRS 9 Macro Overlays for Corporate Lending

If your ECL model uses Nordic GDP growth, unemployment, or lending growth as macro variables, the FI data confirms a material change in one of those inputs. Check whether your model has picked this up. If it uses EU-level aggregates only, the Nordic signal may be diluted.

Stress-Test the Output Floor Sensitivity

For IRB banks, run a scenario where corporate lending volumes decline 10-20 per cent while PD distributions shift upward. Check whether the output floor binds earlier than your baseline projection. Report the result to your capital planning team, not just the regulatory reporting team.

Monitor EBA Reporting Framework Changes

The EBA’s supervisory reporting simplification consultation and the framework 4.3/4.4 roadmap will change COREP credit risk templates over 2026-2027. A lending cycle downturn coinciding with a reporting framework change means double the implementation work. Start mapping which of your existing processes will need to accommodate both the data changes (from the lending slowdown) and the template changes (from the EBA framework update).

Frequently Asked Questions

Does the FI Bank Barometer trigger any EU reporting requirement?

No. The FI Bank Barometer is a descriptive report from the Swedish supervisor. It does not create new reporting obligations. Its value is as a data source that informs your own credit risk analysis and quality checks on COREP/FINREP submissions.

Which COREP templates are most affected by a lending slowdown?

C 07.00 (standardised approach credit risk) and C 08.01 through C 08.05 (IRB approach) capture exposure volumes and risk weights that change when lending decelerates. C 09.01 and C 09.02 (geographical breakdown) show the country-level impact. C 02.00 (own funds requirements) captures the output floor effect.

How quickly does a lending slowdown show up in COREP?

COREP is reported quarterly. A lending slowdown that begins in mid-2025 appears in Q3 and Q4 2025 submissions. The exposure volume effect is immediate. PD migration and staging effects lag by one to two quarters, depending on how quickly your models update.

Should I adjust IFRS 9 overlays based on Swedish data alone?

Not based on the FI Barometer alone. But if your portfolio has material Swedish exposure and your ECL model uses Nordic macro variables, the data should feed into your next model parameter review. The risk is not adjusting too early. The risk is ignoring a confirmed deceleration until it is already reflected in your non-performing exposure ratios.

Does the lending slowdown affect Pillar 3 disclosures?

Yes. Template EU CR4 (standardised approach credit risk exposure and CRM effects) and EU CR6/CR7 (IRB) reflect the same exposure and RWEA data that COREP captures. If your COREP numbers shift, your Pillar 3 disclosures shift with them. The narrative section of your Pillar 3 report should explain material changes in credit risk exposure, including any concentration or geographical shifts linked to the Nordic lending cycle.

Is the Swedish lending slowdown unique or part of an EU-wide trend?

The ECB’s bank lending survey rounds through early 2026 reported a net tightening of credit standards on loans to firms across the euro area, with loan demand mixed by segment. The FI data confirms the pattern in Sweden with specific numbers. Whether the magnitude is comparable across all EU jurisdictions is an open question, but the direction is consistent.

How does this interact with CRR3 implementation?

CRR3’s revised standardised approach and the output floor are in force from 1 January 2025. A lending slowdown during the transition period changes the relative bite of the output floor because IRB and SA RWEA respond differently to volume and PD shifts. Banks should model the interaction, not treat lending cycle effects and CRR3 implementation as separate workstreams.

Key Takeaways

  • The FI Bank Barometer H2 2025 shows Swedish lending to the public grew just 0.9 per cent year-on-year in Q4 2025, halving from Q2 2025, with the slowdown concentrated in non-financial corporates.
  • Swedish major banks lost market share while consumer credit firms grew fastest, changing the exposure class mix that flows into COREP credit risk templates.
  • COREP templates C 07.00 (SA), C 08.01-C 08.07 (IRB), and C 09.01/C 09.02 (geographical breakdown) are the first to reflect a lending deceleration in quarterly submissions.
  • IFRS 9 stage migration effects lag the lending data by one to two quarters. Forward-looking ECL models should incorporate the confirmed Nordic macro deterioration now, not after defaults materialise.
  • Under CRR3, a declining corporate portfolio can increase the output floor bite because IRB risk weights respond to PD shifts while the SA floor reference uses fixed risk weights.
  • FINREP F 18.00 (performing/non-performing breakdown) and F 06.01 (NACE sector split) will show the lending cycle impact with a lag. Sector-level analysis matters more than aggregate numbers.
  • The coincidence of a lending slowdown with EBA reporting framework changes (4.3/4.4) means reporting teams face simultaneous data and template changes through 2026-2027.

Related Articles

  • COREP Reporting Explained – Comprehensive guide to COREP templates, filing requirements, and common errors for EU credit institutions.
  • CRR3 Output Floor Phase-In 2026 – How the output floor works under CRR3, the transitional schedule, and reporting implications for IRB banks.
  • FINREP Reporting Explained – Guide to FINREP templates including asset quality reporting, IFRS 9 staging, and the performing/non-performing breakdown.
  • COREP Reporting Errors – Common mistakes in COREP submissions and how to avoid them, including credit risk template pitfalls.
  • ECB SREP 2026 Priorities – The ECB’s supervisory priorities for 2026, including the reverse stress test on geopolitical risk and liquidity reviews.

Sources and References

Disclaimer: The information on RegReportingDesk.com is for educational and informational purposes only. It does not constitute legal, regulatory, tax, or compliance advice. Always consult your compliance officer, legal counsel, or the relevant supervisory authority for guidance specific to your institution.

Similar Posts