## 1nldea2024006

## Source details

**Canonical URL:** [1nldea2024006](https://www.imf.org/-/media/files/publications/cr/2024/english/1nldea2024006.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2024/english/1nldea2024006.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2024/english/1nldea2024006.pdf.json)

---

### Insurance market structure and vulnerabilities
- Sector size and concentration:
  - Total balance sheet assets: EUR 445bn as of mid-2023 (43 percent of GDP).
  - Life insurers licensed: 22 (end-2022); Non-life insurers licensed: 107 (end-2022).
  - Life sector: three largest life insurers account for 67 percent market share; largest ten cover almost the entire market.
  - Non-life sector: top three firms account for one third; top ten hold 70 percent cumulatively.
- Market dynamics:
  - Sales of new life products (especially individual life) have decreased since early 2000s.
  - In 2022, 31 percent of gross life premiums were unit- and index-linked products.
  - Non-life market dominated by compulsory health insurance: medical expense insurance gross written premiums - direct business: EUR 52,222m (2022).
  - Gross written premiums (2022): total life: 14,612 (gross), net written: 12,289; total non-life direct business: 65,952 (gross), net written: 65,375.
- Investment and mortgage exposure:
  - Loans and mortgages ~23 percent of life insurers’ total assets (constant share).
  - Insurers account for around 9 percent of mortgage loans to households.
  - Four biggest insurers have mortgage subsidiaries for direct consumer mortgage lending.
  - Direct real estate (including mortgage exposures) rose from about 13 percent of total assets in 2016 to 18.5 percent in 2022; indirect real estate from 1.2 percent in 2016 to 3.5 percent in 2022.
  - Domestic investments: 42 percent (life), 50 percent (non-life); 54 percent of equity holdings are domestic.
- Solvency and valuation mechanics:
  - SCR coverage ratios (end-2022, median): life 196 percent; non-life 183 percent.
  - Solvency ratios well above regulatory threshold but below EU average and affected by Long-Term Guarantee Measures.
  - Volatility Adjustment (VA) effect: for VA users SCR ratio would have been ~70 percentage points lower without VA, yet still above 100 percent; no other EU country recorded such a large average benefit from VA.
  - Drivers of VA effect: long durations of technical provisions (life weighted average maturities: life 15.6 years in 2022) and large mortgage exposures not reflected in EIOPA’s representative portfolio.
- Profitability and performance:
  - Investment yields of life insurers fell to 2.7 percent in 2022 after 2018–2021 range 3.6–5.0 percent.
  - Median return on equity (2022): life 1 percent; non-life -3 percent.
  - Combined ratio for non-life hovered between 98 and 99 percent and remained below 100 percent since 2018.

### Pension system structure, transition risks, and key metrics
- Three-pillar structure:
  - Pillar I (AOW): residents accrue 2 percent of flat-rate pension annually for 50 years; retirement age reached 67 years in 2024 and will increase in line with life expectancy from 2026.
  - Pillar II: mandatory occupational schemes (6.0m active members, 9.4m deferred members, end-2022); coverage 87 percent of employees; total pension fund assets EUR 1.59trn (end-2022).
  - Pillar III: voluntary pension savings with tax-deductible contributions up to limits.
- Market size and concentration:
  - Netherlands ranks 4th largest pension fund market in absolute amounts and 3rd as share of GDP: total pension fund assets EUR 1.59trn (end-2022), 154 percent of GDP.
  - Number of pension funds: 174 (end-2022): 47 industry-wide, 109 company funds, 18 others; consolidation from 926 in 2002 to 174 in 2022.
  - Two largest funds hold 47 percent combined market share.
- Transition to Defined-Contribution (WTP):
  - Wet Toekomst Pensioenen (WTP) in force 1 July 2023; transition to DC contracts by 1 January 2028.
  - From 1 January 2028 pension funds can only administer DC contracts; contributions age-independent; accrued DB entitlements converted into DC capital under legal framework (conversion methods: standard method or value-based ALM method).
  - New DC contract types: Solidarity DC (collective accounts without individual explicit stakes; compulsory collective reserve) and Flexible DC (personal investment accounts with explicit stakes; optional collective reserve).
  - Transition framework covers governance, financials, implementation, and bridging plan; bridging plan allows use of “transition FTK” under certain conditions.
- Pension fund funding and risks:
  - Median funding ratio end-2022: 125 percent; only 0.5 percent of pension liabilities managed by funds with funding ratio < 100 percent.
  - Funding requirements: minimum funding requirement ≈ 105 percent of liabilities (policy funding ratio uses 12-month moving average); risk-based capital requirement calibrated to a 2.5 percent annual underfunding probability (on average ≈ 25 percent of liabilities).
  - Derivatives exposures: interest rate swaps notional EUR 442bn (end-2022); market value moved from EUR 112bn (end-2020) to EUR -87bn (end-2022). DNB estimates aggregate interest rate sensitivity EUR 600m per basis point.
  - Liquidity reliance: margin calls and repo market drying-up present potential liquidity risks; currently few funds active in repo markets but transactions can be substantial for large funds.

### Institutional and supervisory framework; practices and gaps
- Supervisory architecture:
  - Twin-peaks: DNB (prudential) and AFM (conduct); both accountable to MoF and MoSA; annual budget process with multi-year strategies; recommendation to define intermediate budgetary safeguards to strengthen supervisory independence.
  - Staffing and remuneration: DNB/AFM derive salary levels from mid-point for the economy; recommendation to benchmark periodically against financial-industry peers.
- Supervisory approach and tools:
  - DNB’s ATM (risk-based) methodology: institutions classified into impact classes I1, I2, I3; end-2022 counts: 90 class 1 insurers, 31 class 2, five class 3.
  - Data-driven supervision: quarterly QRTs and dashboards, Dataloop project for data validations; pilots with NLP and machine learning for pattern recognition and outlier detection.
  - AFM: trend monitoring, firm-specific supervision for five largest insurers (≈80 percent retail market), Data Dashboard Non-Life and plans for Life Dashboard.
- Governance, fit & proper, and internal controls:
  - Fit and proper assessments applied to board members and key function holders; AFM advises and may participate; collective competence evaluated.
  - Governance and internal controls supervision includes ORSA, internal audit, actuarial function and three-lines-of-defense arrangements.
- On-site inspections and enforcement:
  - No minimum frequency of on-sites; smaller insurers face fewer on-sites after ATM introduction; inspectors typically notify at least eight weeks ahead.
  - Enforcement powers under WFT include issuing instructions (articles 1:75, 1:77, 1:79), imposing fines (article 1:80), modifying/withdrawing licenses (article 1:104), increasing technical provisions (article 3:67), temporary capital add-on (article 3:132), recovery plans (article 3:135), short-term finance schemes (article 3:136), activity restrictions (article 3:137).
  - Recommendation to expand on-site inspections for lowest impact class insurers as a backstop.

### Macroprudential, Solvency II, and supervisory convergence
- Solvency II status and measures:
  - Solvency II transposed and applicable since 2016 for most insurers; Basic Regime applies to many smaller insurers; smallest undertakings exempted.
  - Internal models: nine undertakings (three groups) authorized to use partial internal models; DNB first NCA to use new QRTs for internal models and has an Internal Model panel.
  - IFRS 17 implemented 1 January 2023; Solvency II valuation approach remains aligned.
  - Trialogue agreement will modify VA methodology; amended Solvency II Directive to be transposed presumably by second quarter of 2026.
- Macroprudential practice and tools:
  - DNB conducts macro risk analyses, participates in EIOPA stress tests and IAIS monitoring; no insurers currently subject to Solvency II-based macroprudential requirements, but the Solvency II review introduces liquidity tools including dividend bans.
  - DNB conducts country-specific stress testing (project EIPHORIA) and is developing a top-down stress test tool using SII QRT data with modules to include mortgages and loans.
  - Recommendation to allow DNB to cap dividend payments of insurance undertakings on macroprudential grounds (proposed clause in WFT).
- Cross-border and group supervision:
  - Group supervision regimes in WFT; DNB powers as group and host supervisor; colleges of supervisors meet at least annually.
  - Recommendation to require IAIG participation in IAIS Monitoring Phase for the Insurance Capital Standard and to explore stronger cross-border supervisory cooperation with EIOPA.

### Key policy recommendations (selected, with addressees, timing and priority)
- Risks and Vulnerabilities:
  - Intensify analysis of insurers’ mortgage loans and develop suitable stress testing approaches (Addressee: DNB; Timing: ST; Priority: H). (¶14)
  - Closely monitor pension funds’ repo transactions, amend supervisory reporting where necessary, and perform liquidity stress tests incorporating a drying-up of repo markets (Addressee: DNB; Timing: I; Priority: H). (¶33)
- Institutional and Regulatory Structure:
  - Define intermediate steps in the budgetary process to safeguard DNB and AFM independence (Addressees: MoF, DNB, AFM; Timing: ST; Priority: M). (¶41)
  - Periodic benchmarking of supervisors’ salary levels against financial-industry peers (Addressees: MoF, DNB, AFM; Timing: ST; Priority: M). (¶43)
  - Maximize collaboration and systematic data sharing between DNB and AFM (Addressees: DNB, AFM; Timing: ST; Priority: M). (¶51)
- Insurance and Pension Supervision:
  - Review thresholds for impact classes in ATM and consider expanding number of classes (Addressee: DNB; Timing: ST; Priority: M). (¶79)
  - Develop largely automated supervisory tools using advanced technologies (Addressee: DNB; Timing: MT; Priority: M). (¶86)
  - Expand on-site inspections for insurers in lowest impact class (Addressee: DNB; Timing: ST; Priority: H). (¶91)
  - Clarify legal requirement for independent supervisory board members (Addressees: MoF, MoSA; Timing: MT; Priority: H). (¶97)
  - Enable AFM to collect data points relevant to conduct supervision via legislative clause (Addressees: MoF, MoSA; Timing: ST; Priority: M). (¶108)
  - Expedite Life Insurance Dashboard and ensure timely data for Non-Life (Addressee: AFM; Timing: ST; Priority: M). (¶109)
  - Monitor and manage pension transition risks for authorities (Addressees: MoSA, DNB, AFM; Timing: Continuous; Priority: H). (¶124)
  - Monitor effectiveness of internal control functions during transition, emphasize actuarial and internal audit roles (Addressee: DNB; Timing: Immediate; Priority: M). (¶125)
  - Require IAIG to participate in IAIS Monitoring Phase for ICS (Addressee: DNB; Timing: Immediate; Priority: M). (¶130)
  - Explore strengthening supervision of cross-border business with EIOPA (Addressee: DNB; Timing: ST; Priority: M). (¶131)
  - Enact clause allowing DNB to cap dividends of insurers for macroprudential reasons (Addressee: MoF; Timing: ST; Priority: M). (¶137)
- Timing key: C = Continuous; I = Immediate (within one year); ST = Short Term (within 1–3 years); MT = Medium Term (within 3–5 years).
- Priority key: H = High; M = Medium; L = Low.

### Selected financial soundness indicators (highlights, percent)
- Insurance sector (medians):
  - SCR coverage ratio – life: 204 (2022); SCR coverage ratio – non-life: 188 (2022).
  - MCR coverage ratio – life: 448 (2022); non-life: 400 (2022).
  - Investment yield – life: 1.2 (2022).
  - Return on equity – life median: 0.7 (2022); non-life median: -2.9 (2022).
  - Loans, mortgages / total assets excl. unit-linked (life): 24.1 (2022).
- Pension funds:
  - Policy funding rate, median – DB: 123.6 (2022).
  - Investment return, median – DB: -23.0 (2022); DC: -16.7 (2022).
  - Interest rate swaps notional value (end-2022): EUR 442bn; market value (end-2022): EUR -87bn.
  - Liquid assets / total assets – DB+DC: 88.1 (2022).

*Source: IMF staff summary of the FSAP technical note on insurance and pension supervision in the Netherlands (based on material through June 2023).*

### EXECUTIVE SUMMARY __________________________________________________________________________ 5

### EXECUTIVE SUMMARY

### Insurance market structure and vulnerabilities
- The life sector has been steadily shrinking over the last two decades; sales of new life products, especially individual life business, have decreased since the early 2000s, putting pressure on life insurers’ business models.
- The non-life market is relatively saturated and dominated by compulsory health insurance (which covers medical expenses and replaced public health insurance in the 1990s).
- Dutch insurers have become more domestically oriented; large life insurers that received government funding during the global financial crisis were required to restructure parts of their business.
- Investment exposures to illiquid assets and real estate have increased, especially at larger life insurers.
- Insurers are significant providers of mortgage loans:
  - Insurers account for around 9 percent of mortgage loans to households.
  - All four biggest insurers have their own mortgage subsidiaries for direct consumer mortgage lending; several other insurers lend indirectly through mortgage funds.
  - Concentrated mortgage loan exposures contribute to a significant home country bias.
- Solvency metrics:
  - Solvency ratios of Dutch insurers are well above the regulatory threshold but below the EU average and distorted by the mechanics of the ‘Long-Term Guarantee Measures’ in Solvency II.
  - The Volatility Adjustment (VA) has a significant impact on Dutch life insurers, resulting in a higher Solvency Capital Requirement (SCR) ratio for firms using it.
  - The VA effect is driven by balance-sheet features: durations of technical provisions that are considerably longer for Dutch life insurers than for many European peers, and large investment exposure towards mortgage loans not properly reflected in the ‘representative portfolio’ used by EIOPA to calculate the VA.

### Pension system structure and transition risks
- The Dutch pension system comprises three pillars:
  - (i) a public, pay-as-you-go scheme guaranteeing a minimum pension;
  - (ii) mandatory occupational schemes (fully funded and for most employees participation is automatic via employment contracts); and
  - (iii) voluntary pension savings.
- Size and concentration:
  - Total pension fund assets amount to around EUR 1.5trn, making the Netherlands the fourth largest pension fund market globally in absolute amounts, and third if measured against GDP.
  - Consolidation is ongoing and the sector is highly concentrated: the two largest funds hold a combined market share of 47 percent.
- Scheme design and pressures:
  - Most pension schemes are defined-benefit (DB) and have come under pressure since 2008 when low interest rates resulted in declining funding ratios and loss of confidence.
  - Some funds reduced pension benefits; indexation of pensions was generally not achieved for many years.
  - Funding ratios improved only with rising interest rates since 2022, and many pension funds resumed indexation of benefits.
  - Demographic and labor-market challenges: population living longer, declining membership from younger cohorts, and more self-employed/freelancers who accrue little or no pension in the second or third pillars and will therefore rely heavily on the first pillar.
- Legal and structural reform:
  - On 1 July 2023, the Future of Pensions Act came into force, introducing a transition from the current defined-benefit to a defined-contribution system by 2028.
  - The new system aims to reduce potential tensions between generations, better align risk-taking and risk-bearing across age cohorts, and better suit a modern labor market with more flexible work arrangements.
  - Strengths retained include collective investment policy and a strong governance framework for pension funds.
- Specific pension-related vulnerabilities noted:
  - Pension funds’ dependence on repo markets requires close monitoring.
  - The pension system transition creates potential risks for authorities related to resources and legal risks; internal control functions (including actuarial function and internal audit) are critical during the transition.

### Institutional and supervisory framework
- Institutional design:
  - The system for independent state agencies (including De Nederlandsche Bank (DNB) and Autoriteit Financiële Markten (AFM)) balances powers and accountability.
  - The yearly budget process is well elaborated, incorporating supervisory objectives and key performance indicators; authorities are encouraged to further detail the budgetary structure to provide additional safeguards to supervisory independence.
  - DNB and AFM compete on the labor market for similar experts as financial institutions; the competitiveness of supervisors’ jobs should be measured against similar jobs in the financial industry.
  - Supervisory cooperation between DNB and AFM is well-established.
- Supervisory approach and effectiveness:
  - Supervision of insurers and pension funds is effective in the Netherlands.
  - DNB’s supervisory approach is risk-based, intrusive, and transparent, leveraging well-developed supervisory tools and an ongoing supervisory dialogue.
  - Governance, behavior, and culture frameworks have been applied in practice, delivering findings and enforceable recommendations; DNB has used supervisory boards (including independent members) to channel messages to management.
  - AFM has upgraded its supervisory approach by leveraging data and aiming for a more forward-looking perspective.
- Areas for supervisory refinement:
  - DNB’s new risk-based supervision methodology (ATM) should be regularly reviewed with further backstops, including a few regular on-site inspections even for the smallest insurers.
  - The institutional structure could benefit from periodic benchmarking of supervisors’ salary levels against relevant peers in the financial industry.
  - Closer collaboration and systematic data sharing between DNB and AFM is recommended.
  - Increasingly critical to collect necessary supervisory data and apply advanced technologies to analyze data, including connecting data from different sources.
  - Monitor insurer mortgage lending and pension funds’ repo dependence closely.

### Key policy recommendations (Table 1: Main Recommendations)
- Risks and Vulnerabilities
  1. Intensify the analysis of insurers’ mortgage loans and develop suitable stress testing approaches (Addressee: DNB; Timing*: ST; Priority**: H). (¶14)
  2. Closely monitor pension funds’ repo transactions, amend supervisory reporting where necessary, and perform liquidity stress tests which incorporate a drying-up of repo markets (Addressee: DNB; Timing*: I; Priority**: H). (¶33)

- Institutional and Regulatory Structure
  3. Define intermediate steps in the budgetary process to provide additional safeguards to the independence of DNB’s supervisory function and of the AFM (Addressees: MoF, DNB, AFM; Timing*: ST; Priority**: M). (¶41)
  4. Implement the practice of periodic benchmarking of supervisors’ salary levels against relevant peers in the financial industry (Addressees: MoF, DNB, AFM; Timing*: ST; Priority**: M). (¶43)
  5. Explore further options to benefit from the second and third lines of defense in strengthening supervisory processes (Addressee: DNB; Timing*: MT; Priority**: M). (¶48)
  6. Ensure maximizing opportunities for collaboration between DNB and AFM, especially to jointly engage with financial institutions and to share data, through systematic measures (Addressees: DNB, AFM; Timing*: ST; Priority**: M). (¶51)

- Insurance and Pension Fund Supervision
  7. Review the thresholds for impact classes in ATM, and incorporate additional elements of systemic importance. Consider using a slightly larger (and even) number of impact classes (Addressee: DNB; Timing*: ST; Priority**: M). (¶79)
  8. Explore opportunities to develop largely automated supervisory tools based on advanced technologies – able to learn and adapt by using algorithms and statistical models – allowing processing of large data sets (Addressee: DNB; Timing*: MT; Priority**: M). (¶86)
  9. Expand the number of on-site inspections for insurers in the lowest impact class, as a backstop to the risk-based approach (Addressee: DNB; Timing*: ST; Priority**: H). (¶91)
  10. Further clarify the requirement of independent supervisory board members in law (Addressees: MoF, MoSA; Timing*: MT; Priority**: H). (¶97)
  11. Enact a clause which would allow the AFM to collect data points relevant to conduct supervision (Addressees: MoF, MoSA; Timing*: ST; Priority**: M). (¶108)
  12. Expedite work on the Life Insurance Dashboard, and (also for Non-Life) ensure that timely data is available (Addressee: AFM; Timing*: ST; Priority**: M). (¶109)
  13. Closely monitor and proactively manage potential risks of the pension system transition for the authorities related to resources and legal risks (Addressees: MoSA, DNB, AFM; Timing*: C; Priority**: H). (¶124)
  14. Closely monitor the effectiveness of internal control functions in the transition process, and emphasize the important role of all pension funds’ internal control functions in the transition process, including the actuarial function and internal audit (Addressee: DNB; Timing*: I; Priority**: M). (¶125)
  15. Require the IAIG to participate in the IAIS’s Monitoring Phase for the Insurance Capital Standard (Addressee: DNB; Timing*: I; Priority**: M). (¶130)
  16. Explore, in collaboration with EIOPA, ways to strengthen supervision of cross-border business, including a review of how cooperation through platforms could be strengthened, and promote further supervisory convergence (Addressee: DNB; Timing*: ST; Priority**: M). (¶131)
  17. Enact a clause which would allow DNB to cap dividend payments of insurance undertakings, based on macroprudential reasons (Addressee: MoF; Timing*: ST; Priority**: M). (¶137)

- Timing and priority keys
  - * C = Continuous; I = Immediate (within one year); ST = Short Term (within 1-3 years); MT = Medium Term (within 3-5 years).
  - ** H = High; M = Medium; L = Low.

### Scope and methodology of the FSAP note
- Purpose and coverage:
  - This technical note analyzes regulation and supervision of insurance undertakings and occupational pension funds in the Netherlands as part of the 2024 Financial Sector Assessment Program (FSAP).
  - The analysis is based on the regulatory framework and supervisory practices in place as of June 2023.
  - The FSAP reviewed recent developments and structures of the Dutch insurance and pension fund sectors; focused review of legislative framework and supervisory practices rather than a full observance assessment of global standards.
- Benchmarks and standards used:
  - Insurance oversight benchmark: Insurance Core Principles (ICPs) of the IAIS (November 2019), with focus on ICPs of macrofinancial relevance and where material regulatory changes occurred since the last FSAP (including ICP 17 on solvency, ICPs 7, 8, 16 on governance and risk management, ICPs 1, 2, 9, 24 on supervisory approach, ICP 19 on conduct, ICPs 23 and 25 on group supervision and cross-border cooperation).
  - Pension oversight benchmark: Principles of the International Organisation of Pension Supervisors (IOPS, November 2010), EU legislation including IORP II and Solvency II where relevant.
- Evidence base:
  - The note is based on review of regulations, market analyses, and meetings with Dutch authorities (DNB, AFM), Ministry of Finance (MoF), Ministry of Social Affairs and Employment (MoSA), insurers, pension funds, industry associations, and private sector bodies.
  - The Dutch authorities provided a full self-assessment for the eleven ICPs analyzed, supported by examples of supervisory practices and assessments.

*Source: IMF staff summary of the FSAP technical note on insurance and pension supervision in the Netherlands (based on material through June 2023).*

### 4.      The previous FSAP in 2017 made several recommendations to strengthen the

### 1nldea2024006 - 4.      The previous FSAP in 2017 made several recommendations to strengthen the

### Previous FSAP recommendations and technical notes
- The previous FSAP in 2017 recommended:
  - Closely monitor the Solvency II implementation, specifically on the methodology for the ultimate-forward rate, the loss absorbance capacity of deferred taxes, and the volatility adjustment.
  - Exercise pillar 2 powers, such as dividend payout restrictions and capital add-ons, if necessary.
  - Establish mechanisms to credibly enforce national market conduct regulation with regard to incoming cross-border business.
  - Strengthen certain aspects of the pension law, such as the requirements for advice provided by a pension fund, or the rules on the amount of costs or the suitability of pension products.
  - Consider direct oversight of governance and staff qualifications at group level for service providers to which activities of pension funds have been outsourced.
- Further technical notes of this FSAP:
  - The Technical Note on Systemic Risk Analysis summarizes stress test results and elaborates more on risks and vulnerabilities for both sectors.
  - The Technical Note on Climate Risk Oversight reviews the regulatory and supervisory framework inter alia against the IAIS Application Paper on the Supervision of Climate-related Risks in the Insurance Sector.

### Insurance market structure and consolidation
- Sector consolidation:
  - Life insurers licensed: 22 as of end-2022 (including third-country branches and funeral insurers), down from 36 five years earlier and from 138 in 2002.
  - Non-life insurers licensed: 107 at end-2022 (including third-country branches), down from 251 in 2002.
  - Seven reinsurers were established in the Netherlands in 2022.
- Balance sheet assets and relative size:
  - Total amount of balance sheet assets: EUR 445bn as of mid-2023, corresponding to only 43 percent of GDP.
  - Insurance sector assets relative to GDP: 43 percent in 2023 (2022: 46 percent).
- Market concentration:
  - Life sector: the three largest life insurers account for a market share of 67 percent; the largest ten cover almost the entire market.
  - Non-life sector: top three firms account for a third of the market; top ten hold a cumulated market share of 70 percent.

### Life insurance market dynamics and product trends
- Sales and product mix:
  - Sales of new life products, especially individual life business, have decreased since the early 2000s.
  - In 2022, 31 percent of gross premiums were collected for unit- and index-linked products, just a little more than for ‘other life’ products which include term life insurance and immediate annuities.
- Drivers of decline:
  - Legislative changes (e.g., Dutch Bank Savings Act of 2008) and restricted favorable tax treatment of savings-linked and unit-linked insurance.
  - Rising cost of life insurance due to falling interest rates and increased life expectancy.
  - A major loss of confidence after the “usury policies scandal” (life products sold in the 1990s and early 2000s charged more costs than communicated and promised unrealistically high returns).
- Pension market interaction:
  - Collective (pension) business developed positively; declining interest rates and pension scheme reform resulted in significant transfers, especially from small single-employer pension funds towards insurance providers.
  - With the new pension system, life insurers might gain further market shares in the pension market.

### Non-life market, health insurance dominance, and profitability
- Non-life market composition and trends:
  - Non-life market dominated by compulsory health insurance.
  - Health insurance premiums for hospital care and other cure-based health claims reached EUR 52bn in 2022 (79 percent of the total non-life market).
  - Motor and fire account for 8 and 6 percent of total non-life premiums, respectively.
  - In 2020/21, Covid-19 pandemic disrupted structural growth patterns; hospitals and health professionals cancelled regular medical treatments and revenues were guaranteed through continuity schemes.
  - Profitability improvements: motor and fire profitability gradually improved; property and casualty insurers hit in 2021 by floods in Limburg and Noord-Brabant; signs of a rising trend in cyber claims.
- Profitability metrics:
  - Investment yields of life insurers fell to 2.7 percent in 2022 after fluctuating between 3.6 and 5.0 percent since 2018.
  - Median life insurer return on equity in 2022: 1 percent.
  - Median non-life insurer return on equity in 2022: -3 percent.
  - The combined ratio for non-life insurers has been rather stable and below 100 percent since 2018, although higher than the EU/EEA average.

### Market size, penetration and density
- Global and EU ranks and figures:
  - The Netherlands records the 15th highest insurance penetration globally, mostly driven by large premium income in health insurance.
  - Gross written premiums in 2022: EUR 81bn.
  - The Netherlands hosts the fifth largest insurance sector in the European Union (EU) in 2022.
  - Life insurance penetration (premiums to GDP) in 2022: 1.2 percent (advanced markets average: 3.7 percent; EU: 3.4 percent).
  - Non-life penetration in 2022: 7.3 percent (advanced markets: 5.8 percent; EU: 3.0 percent).
  - Insurance density (premiums per capita) in 2022: USD 657 (life) and USD 4,074 (non-life).
- 2022 premium breakdown (selected figures, EUR million):
  - Total life premiums (gross written): 14,612; Net written premiums: 12,289.
  - Index-linked and unit-linked insurance (life) gross written premiums: 4,572; net written: 4,458.
  - Other life insurance gross written premiums: 4,466; net written: 2,867.
  - Insurance with profit participation gross: 1,993; net: 1,814.
  - Life reinsurance gross: 948; net: 926.
  - Total non-life gross written premiums - direct business: 65,952; gross written premiums - proportional reinsurance accepted: 571; gross written premiums - non-proportional reinsurance accepted: 279; net written premiums: 65,375.
  - Medical expense insurance gross written premiums - direct business: 52,222; net written premiums: 52,167.

### Foreign participation, ownership shifts, and supervisory responses
- Ownership and reinsurance patterns:
  - Increasing share of foreign insurers in the Dutch non-life market.
  - Five years ago, strong interest from private equity firms in taking over insurers at prices above business priced on the swap curve; investors aimed at bringing forward profits and receiving dividends.
  - Occasionally accompanied by asset-intensive reinsurance to group-internal reinsurers in jurisdictions with less strict supervisory regimes compared to Solvency II.
  - DNB has pointed out the risks involved in asset-intensive reinsurance.
- Supervisory measures:
  - It is no longer possible for non-EU/EEA insurance undertakings to carry out insurance activities in the Netherlands without a local establishment of an insurance undertaking.

### Insurtech, technology, and supervisory monitoring
- Insurtech developments:
  - Insurtechs active with services from insurance pricing to robo advice and prevention, digitalization services, fraud detection and claims handling.
  - Example innovations: motor insurance policies rewarding safe driving via a device in the policyholder’s car; robo-advice concepts for disability insurance.
  - AFM has formulated points for attention for automated advisory services and outlined how technology could affect the insurance sector in the next decade.
- Supervisory monitoring:
  - DNB monitoring path: explorative research in 2017/18 → thematic identification of risk management and business model issues → investigation into artificial intelligence and its application by insurers.
  - Observations: advanced data analysis could allow insurers to differentiate premiums and customer acquisition to the point where uninsurability may become a problem; insurtech innovations like cloud solutions used for operational excellence.
  - DNB has investigated the development of artificial intelligence and its applications by insurers.

### Investment allocation, mortgage exposure, and illiquid assets
- Asset allocation (selected figures and trends):
  - Life sector asset shares as of end-2022: sovereign bond holdings 15 percent, equity holdings 9 percent of total assets.
  - Non-life sector asset shares as of end-2022: government bonds 14 percent, equity holdings 13 percent of total assets.
  - Since 2016 steady increase in real estate exposures:
    - Direct real estate (including mortgage exposures): from about 13 percent of total assets in 2016 to 18.5 percent in 2022.
    - Indirect real estate: from 1.2 percent in 2016 to 3.5 percent in 2022.
  - Loans and mortgages account for a rather constant share of 23 percent (life insurers).
- Mortgage lending role:
  - Insurers account for around 9 percent of mortgage loans to households.
  - The four biggest insurers each have mortgage subsidiaries for direct consumer mortgage lending; several other insurers lend indirectly through mortgage funds.
  - Insurers typically take vertical slices of mortgage funds on their balance sheet.
  - In 2023, DNB analyzed exposures, risk management and lending conditions and collects data from the larger insurers on loan-to-value ratios.
- Geographic home bias and diversification:
  - Domestic investments account for 42 percent of life insurers’ investments and 50 percent of non-life insurers’ investments.
  - Other relevant investment markets: Germany 9 percent, France 8 percent, United States 6 percent.
  - 54 percent of equity holdings and participations are domestic.
  - Bond exposures more diversified: 22 percent of government bonds and 16 percent of corporate bonds are domestic.
  - Credit quality of the bond portfolio is very high.
- Illiquid assets:
  - Increase in allocation to illiquid assets, specifically at larger life insurers, funded by reducing allocation to listed bonds and equity.
  - Increases mainly in direct lending, private equity and debt (funds), infrastructure, and collateralized loans.
  - Access to some illiquid asset categories remains uneven between the largest insurers; partnerships with asset managers with illiquid asset expertise are used.
  - DNB observes institutional investors adding illiquidity by turning to private assets in sustainable and impact investments.

### Derivatives usage
- Main derivative instruments: interest rate swaps and FX forwards; to a lesser extent equity put options.
- Central clearing:
  - New derivative positions are almost exclusively centrally cleared.
  - The largest six insurers account for 97 percent of the outstanding notional and fall under the central clearing obligation.
  - About a third of open-interest rate derivative positions have a central counterparty as counterpart.

### Policy recommendation (from the FSAP)
- Recommendation 1:
  - Intensify the analysis of insurers’ mortgage loans and develop suitable stress testing approaches. This should include an obligation for insurers providing mortgages to periodically update the information about the interest-only mortgage client’s financial situation.

*Source: IMF staff calculations and analysis as presented in the provided FSAP extract.*

### 1.2 percent in 2022, also below the 5-year average.

### 1nldea2024006 - 1.2 percent in 2022, also below the 5-year average.

### Insurance sector performance
- Underwriting performance of the non-life sector is profitable with combined ratios below 100 percent; hovering between 98 and 99 percent, it is above the EU/EEA average.
- Return on equity (median):
  - Life insurers: 1 percent in 2022, after fluctuating between 2 and 9 percent in previous years.
  - Non-life insurers: -3 percent (median) in 2022, after an average of 7 percent in prior years.
- Solvency Capital Requirement (SCR) coverage ratios (end-2022 context):
  - Life insurers: 196 percent.
  - Non-life insurers: 183 percent.
  - These SCR coverage ratios are significantly lower than the EU/EEA average.

### Impact of Long-Term Guarantee (LTG) measures and Volatility Adjustment (VA)
- The Volatility Adjustment (VA) has a significant impact on Dutch life insurers, resulting in a higher SCR ratio for firms using it.
- For VA users, the SCR ratio would have been around 70 percentage points lower without this measure, but still above the regulatory threshold of 100 percent.
- No other EU country recorded—on average—such a large benefit from using the VA.
- Structural drivers of the VA impact on Dutch life insurers:
  - Durations of technical provisions are considerably longer for Dutch life insurers than for many European peers.
  - Large investment exposure towards mortgage loans is not properly reflected in the ‘representative portfolio’ used by EIOPA to calculate the VA.
- Regulatory note: The trilogue agreement on the Solvency II review introduces a modified methodology for the calculation of the VA. The amended Solvency II Directive will need to be transposed into national law, presumably by the second quarter of 2026.

### COVID-19 effects on insurers (Box 1)
- 2020: initial uncertainties had a major, though temporary, impact on the value of insurers’ investment assets and posed a large financial risk.
- Insurance claims increased in some lines (particularly travel insurance and legal assistance); new business in short-term travel insurance declined.
- Claims in motor insurance and P&I declined due to lower mobility.
- DNB actions during the pandemic:
  - Urged insurers not to pay dividends or buy back shares, following ESRB and EIOPA recommendations.
  - Allowed health insurers to provide liquidity support to health care providers.

### Pension system structure (three pillars) and key metrics
- Pillar I: AOW (public, pay-as-you-go)
  - Residents accrue 2 percent of the flat-rate pension annually for 50 years.
  - Retirement age reached 67 years in 2024; from 2026 onwards it will be increased in line with life expectancy.
  - Around 3.6 million pensioners received total benefits of EUR 43bn in 2021.
- Pillar II: Mandatory occupational schemes (main form of pension savings)
  - 6.0m active members, 9.4m deferred members (end-2022); 87 percent of employees covered.
  - Pension contributions deductible up to EUR 128,810 (2023).
  - Total pension fund assets: EUR 1.59trn (end-2022).
  - Netherlands ranks 4th largest pension fund market in absolute amounts; as a share of GDP ranks third at 154 percent.
- Pillar III: Voluntary pension savings
  - Contributions deductible within tax framework; various providers (banks, life insurers, asset managers).

### Demographics and implications for pensions
- Population:
  - Grew from around 13 million in early 1970s to 17.6 million in 2022.
  - Expected to increase to more than 20 million by the early 2060s.
- Labor participation:
  - Reached 57.7 percent in 2022, including the self-employed.
- Median age and dependency ratios:
  - Median age 42.7 years in the Netherlands vs 44.4 years in Europe.
  - Old-age dependency ratio: 30.1 in 2020 (Netherlands) vs 32.0 in the EU; projected 52.6 (Netherlands) vs 55.7 (EU) for 2080.
- Population pyramid peaks at around 52 years, indicating a significant increase in retirees in the late 2030s.
- About 8 percent of pension fund members (active and inactive) live abroad; contributions for members living abroad are about 1 percent of total contributions.

### Pillar II details and fund structure
- Coverage:
  - Participation is automatic for most employees; coverage of 89 percent of employees in 2022.
  - Self-employed coverage around 6 percent.
- Scheme features:
  - Most schemes are defined-benefit (DB), typically career-average salary basis up to salary cap EUR 128,810 (2023) with maximum annual accrual of 1.875 percent.
  - Conditional indexation based on wage growth or prices; many funds have not afforded indexation since the global financial crisis and only re-started indexation in 2022.
- Types of pension funds:
  - 174 pension funds (end-2022): 47 industry-wide pension funds, 109 company pension funds, and 18 others.
  - Consolidation: number of pension funds declined from 926 in 2002 to 174 in 2022.
  - Two largest funds hold a combined market share of 47 percent.
- Innovations:
  - Premium Pension Institutions (PPIs): seven exist, provide DC lump sum options.
  - General Pension Funds (APFs): five APFs with total assets under management of around EUR 25bn.

### Pension fund assets, allocation, and cashflows
- Total pension fund assets: EUR 1.59trn (end-2022).
- Sector flows (2021 example):
  - Inflows (contributions and net transfers): EUR 48bn.
  - Outflows (benefits and expenses): EUR 38bn.
- Asset allocation (end-2022):
  - Bonds: 41 percent of total assets (stable since 2020).
  - Shares and other equity (including alternative assets): 34 percent (down from 36 percent six years earlier).
- Fixed-income quality:
  - One third of fixed-income assets rated in the highest rating category.
  - Another 48 percent in investment grade.
- Compared to the EEA average:
  - Equity holdings are significantly lower.
  - Derivatives exposures are considerably higher.

### Derivatives exposures and liquidity considerations
- Notional values and market values (end-2022):
  - Interest rate swaps: notional value EUR 442bn (up from 171bn two years earlier); market value declined from EUR 112bn (end-2020) to EUR -87bn (end-2022).
  - DNB estimates aggregate interest rate sensitivity of the funds’ derivatives portfolio at EUR 600m per basis point.
  - Forward-rate agreements for currency risk: nominal value EUR -392bn and market value at end-2022 of EUR 14bn.
- Liquidity risk:
  - Pension funds use derivatives primarily to hedge interest rate and currency risks—margin calls could pose a liquidity risk.
  - Under normal circumstances, pension funds have sources of liquidity, in particular cash, to meet margin calls.
  - In extreme market situations, pension funds might be dependent on secondary liquidity through the repo market; currently only few Dutch pension funds are active in the repo market (with limited amounts), but individual transactions can be substantial given the large size of some funds.

*Source: IMF staff calculations based on DNB.*

### 33.      Recommendation 2: Closely monitor pension funds’ repo transactions, amend supervisory

### 1nldea2024006 - 33.      Recommendation 2: Closely monitor pension funds’ repo transactions, amend supervisory

### Recommendation summary
- Recommendation 2: Closely monitor pension funds’ repo transactions, amend supervisory reporting where necessary, and perform liquidity stress tests which incorporate a drying-up of repo markets.
- Additional recommendations appearing in the section:
  - Recommendation 3: Define intermediate steps in the budgetary process to provide additional safeguards to the independence of DNB’s supervisory function and the AFM.
  - Recommendation 4: Implement the practice of periodic benchmarking of supervisors’ salary levels against relevant peers in the financial industry.
  - Recommendation 5: Explore further options to benefit from the second and third lines of defense in strengthening the processes of supervision.
  - Recommendation 6: Ensure maximizing opportunities for collaboration between DNB and the AFM, especially to jointly engage with financial institutions and to share data, through systematic measures.

### Pension funds’ investment performance (2021–2022)
- DB schemes:
  - Median fund investment performance in 2022: -23 percent.
- DC schemes:
  - Median fund investment performance in 2021: 9.7 percent.
  - Median fund investment performance in 2022: -16.7 percent.
  - Overall 2022 median DC performance also reported as -17 percent in the text.
- Distributional note: "The distribution across pension funds has not widened substantially in 2022, so basically all funds were affected by the market turbulences in a similar way."

### Funding ratios and sensitivities
- Sensitivities: Funding ratios are sensitive to equity and interest rate developments because many pension funds hold material equity investments and typically hedge their interest rate risk only partially.
- Historical pressures: Since 2008, low interest rates have put pressure on the funding ratio, pushing many funds below legally required levels.
- Recent recovery:
  - At end-2022, the median pension fund recorded a funding ratio of 125 percent.
  - Only 0.5 percent of all pension liabilities were managed by funds with a funding ratio below 100 percent.
- Visuals referenced (not reproduced here): Figures illustrating pension funds’ investment performance distributions (DB and DC, 2020–2022) and funding ratios distribution (2018–2022) and liabilities per funding level bucket (in percent).

### Third pillar (voluntary individual pension savings)
- Scope: Provision of pensions through annuity products and retirement savings plans issued by insurers, banks and asset managers; supplements AOW and occupational pensions; used by those not in permanent employment or self-employed.
- Product types: Unit-linked products, with-profit policies, products with minimum return guarantees.
- Payout forms: Annuity or lump sum that must be used to purchase an annuity.
- Tax treatment: Comparable to treatment in the second pillar.

### Institutional and regulatory structure — supervisory independence, powers, and resources
- Supervisors: DNB (prudential supervisor) and AFM (conduct supervisor).
- Accountability: Primarily to the Minister of Finance (MoF) and the Minister of Social Affairs and Employment (MoSA); operations can be audited by the Court of Audit.
- Budget framework:
  - Four-year budget limits informed by multi-year strategies and subject to the Cost Framework.
  - Yearly budget process includes submission to panels of supervised institutions, approval by Supervisory Boards, submission to MoF and MoSA before December 1, and publication post-approval.
  - The ministries assess proposals on (i) formal compliance with law, (ii) procedural agreements, (iii) definition of key performance indicators, and (iv) factual or procedural errors.
  - Observed concern: Cooperative and informal practice does not provide sufficient safeguards in case of disagreement and may not shield against political interventions.
- Staffing and remuneration:
  - DNB and AFM derive salary levels from the mid-point for the economy; recommended practice is periodic benchmarking against relevant financial-industry peers and consideration of gender and broader diversity aspects.

### Organization of supervision and internal governance
- DNB organization: Clear internal governance with a three lines of defense model; uses ‘Planning & Control Cycle’ to measure and report performance.
- Second and third lines of defense:
  - Risk Management & Strategy department (2nd line) provides independent views; recommendation to extend review time and implement regular follow-ups.
  - Internal Audit (3rd line) conducts annual thorough audits; recommendation to introduce concise and focused audits (‘quick scans’) and some recurrent audits with predefined frequency.

### Supervisory cooperation and enforcement
- Domestic cooperation: Strong cooperation between DNB and AFM under WFT and a 2016 cooperation agreement; Article 1:46 WFT requires cooperation to align rules and guidance where both supervisors’ mandates intersect.
- Suggested enhancements:
  - More joint meetings/examinations with industry to improve the perception of supervisors speaking with one voice.
  - Creation of a standardized, flexible IT solution for secure data and information sharing to facilitate both regular and ad hoc exchanges.
- International cooperation: Extensive set of MoUs; post-Brexit arrangements effective January 1, 2021 include a multilateral MoU and a bilateral MoU between EIOPA and UK authorities; a Coordination Agreement between DNB and the UK Prudential Regulation Authority.
- Enforcement powers under WFT (selected):
  - Issue instructions to adhere to a particular line of conduct (article 1:75, WFT).
  - Impose instructions preventing an insurer from conducting business from a branch-office or providing services in another EU Member State (article 1:77, WFT).
  - Issue instructions under penalty (article 1:79, WFT).
  - Impose an administrative fine (article 1:80, WFT).
  - Modify, withdraw or limit licenses (article 1:104, WFT).
  - Impose an instruction to increase the technical provisions (article 3:67, WFT).
  - Impose a temporary capital add-on (article 3:132, WFT).
  - Request a recovery plan (article 3:135, WFT) for breach of SCR.
  - Request a short-term finance scheme (article 3:136, WFT) for breach of MCR.
  - Restrict insurer activities (article 3:137, WFT).
- Enforcement process: Decisions to start enforcement are taken by division directors or the Executive Director of DNB responsible for Supervision; final draft penalties submitted for approval to the director of DNB responsible for Supervision; objection and appeal mechanisms exist (appeal to District Court in Rotterdam), with writs generally not having suspensive force except for provisional relief.

*Source: IMF staff calculations based on DNB data.*

### 57.      The Netherlands has transposed the EU’s Solvency II Directive (2009/138/EC) including

### 57–72. Insurance and Pension Regulation in the Netherlands

### Insurance regulation — Solvency II and Basic Regime
- The Netherlands has transposed the EU’s Solvency II Directive (2009/138/EC) including secondary legislation into national law; it has been applicable to the large majority of Dutch insurers since 2016.
- Solvency II exclusions:
  - Does not apply to funeral expenses and benefit-in-kind insurers.
  - Does not apply to the majority of small insurers; these fall under the Dutch “Basic Regime”, a risk-based supervisory framework comparable to Solvency II in its set-up but tailored to this target group.
  - The Basic Regime was evaluated in a report submitted to the Dutch Parliament in 2020.
- The smallest insurance undertakings are exempted from prudential supervision but remain bound by legislative obligations; exempt insurers are required to inform prospective policyholders of their exempt status.
- Table 5 key statistics (as presented):
  - Number of insurers: 113 (Solvency II insurers), 23 (basic insurers), n.a. (Exempted insurers)
  - Gross written premiums (GWP): EUR 80.1bn (Solvency II insurers), EUR 55m (basic insurers), n.a. (Exempted insurers)
  - Thresholds (in EURm):
    - Solvency II: GWP > 5; Technical provisions (TP) > 25
    - Basic insurers: GWP > 2.2 and < 5; TP > 10 and < 25
    - Exempted insurers: GWP < 2.2; TP < 10
  - Max coverage: EUR 12,500
- Internal models:
  - Nine insurance undertakings (belonging to three groups) authorized to use a partial internal model for capital requirement calculations.
  - Main risks covered in these internal models include interest risk, spread risk, equity risk, non-life premium and reserve risk, life trend risk, and life level risk.
  - Ongoing monitoring process includes annual reporting on overall appropriateness and issues identified during the year.
  - DNB implemented new quantitative reporting templates on internal models ahead of the EU-level implementation in 2024; these templates are fully operational and support a DNB dashboard for detailed analysis of internal model results.
  - DNB conducted specific on-site inspections and sector-wide studies (e.g., expert judgments, aggregation models) that resulted in findings requiring follow-up, including model changes.

### Accounting and valuation standards for insurers
- DNB does not approve professional accounting standards.
  - Dutch accounting standards prepared and approved by the Raad van de Jaarverslaggeving.
  - International accounting standards prepared and approved by the International Accounting Standards Board.
  - Application of IFRS within the EU covered by EU Regulation (EC) No 1606/2002.
- Solvency II was created in line with developments in the IAIS, IASB, and IAA, leading to an economic risk-based approach that incentivizes proper risk measurement and management.
- IFRS 17:
  - Implemented as of January 1, 2023.
  - Provides a valuation methodology similar to Solvency II principles.
- Valuation frameworks (Table 6 highlights):
  - Assets:
    - IFRS: IFRS 9 for financial instruments; both fair value and amortized cost accepted depending on business model and SPPI test.
    - RJ 290: comparable for Dutch GAAP; both fair value and amortized cost accepted depending on classification.
    - Solvency II: Assets valued at the amount for which they could be exchanged between knowledgeable willing parties in an arm’s length transaction.
  - Liabilities:
    - IFRS: Per 1 Jan 2023 IFRS 17 for insurance liabilities (fulfillment principle; building block approach: best estimate, risk margin, contractual service margin). For financial liabilities, IFRS 9 (predominantly amortized cost).
    - RJ 605: allows valuation of insurance liabilities according to Solvency II principles (subject to Dutch legal requirements).
    - Solvency II: Liabilities valued at the amount for which they could be transferred, or settled, between knowledgeable willing parties in an arm’s length transaction.
- Since introduction of Solvency II, no changes to general purpose accounting or to Solvency II valuation have been introduced.

### Governance, internal controls, and risk management (Insurers)
- Governance, internal controls, and risk management requirements are set out in the Solvency II Directive and delegated acts and transposed into Dutch law.
- Pillar 2 provisions mainly relate to the system of governance, including the Own Risk and Solvency Assessment (ORSA).
  - Requirements align with WFT provisions on sound and ethical operational management.
  - Coverage includes organizational structure, operational management, expertise and trustworthiness of directors and key officers, risk management, internal control, key functions, and outsourcing.
  - ORSA is part of the risk management system and focuses on insurers’ self-assessment of risks and associated capital needs.

### Group supervision (Insurers)
- Solvency II emphasizes group supervision and group supervisor powers.
- WFT provides three regimes for group supervision: (i) limited risk insurers, (ii) insurance groups, and (iii) financial conglomerates.
- DNB powers:
  - As group supervisor: can issue an instruction to a mixed financial holding or insurance holding to adhere to a particular line of conduct when the holding or an insurer in the group violates a provision of the WFT.
  - As host supervisor: entitled to impose measures on request of the group supervisor on holdings with registered offices in the Netherlands that are part of an insurance group when insurers in the group breach provisions in other member states. Measures include instructions to adhere to a particular line of conduct (with or without penalty) or to impose an administrative fine.

### Pension fund regulation — legal framework and recent reforms
- Primary legislation:
  - Pension Act (PW) for company pension funds and industry-wide pension funds.
  - Mandatory Occupational Pension Scheme Act (Wet verplichte beroepspensioenregeling) for occupational pension funds.
  - Secondary legislation based on these laws imposes prudential requirements and sets material requirements on control structures and relationships among employees, employers, and pension funds.
- IORP II:
  - Revised IORP II Directive came into effect in January 2017, introducing stricter governance requirements (risk, audit and actuarial key functions) and the Own Risk Assessment.
  - European Commission is scheduling a review of IORP II; likely themes: (i) governance and prudential standards, (ii) cross-border activities and transfers, (iii) information to members and beneficiaries and other business conduct requirements, (iv) shift from defined-benefit to defined-contribution, (v) sustainability, and (vi) diversity and inclusion.
- Context and need for reform:
  - Over the last decade, confidence in the pension system—particularly the second pillar—has declined.
  - Low interest rates constrained indexing and led some funds to reduce pension benefits.
  - Demographic pressures: longer life expectancy and declining numbers of younger contributors.
  - Increasing number of self-employed and freelancers with limited accrual in second or third pillar.

### Transition to a Defined-Contribution Pension Regime (WTP) — key elements
- Wet Toekomst Pensioenen (Future of Pensions Act, WTP) came into force on 1 July 2023.
- Main implications for pension funds:
  - Pension funds can only administer DC contracts from 1 January 2028 onwards.
  - In these DC pension plans the contribution will be age-independent.
  - Default: current DB entitlements of pension funds will be converted into DC capital in the new DC contracts.
- Transition requires decision-making and implementation efforts involving social partners, pension funds, pension administrators, and supervisors between 1 July 2023 and 1 January 2028. The WTP includes a transition framework covering governance, financials, operational implementation, and the bridging plan.

- Main elements of new DC contracts:
  - Pension funds must establish age dependent risk profiles for participants based on specific uniform measures.
  - Two types of DC plans: solidarity scheme and flexibility scheme.
    - Solidarity DC: involves personal investment accounts but no explicit stakes in specific investment funds at individual level; returns on collective assets are split across individual pension accounts according to specific legal rules.
    - Flexible DC: close to DC plans already known; personal investment accounts with explicit stakes in specific investment funds.
  - Both types offer annuities that allow continued exposure to investment risk during disbursement.
  - New contracts involve a separate collective reserve (compulsory in solidarity DC; optional in flexible DC) intended for collective risk-sharing; legal conditions apply to maximum level of this reserve.

- Transition framework components:
  - Governance:
    - Social partners must formulate a transition plan to support decision-making.
    - Pension funds must formulate an implementation plan; roles of parties are explicit in law.
    - Pension funds accountable for accepting social partners’ decisions and must consider fairness and manageability of risks.
    - Objection to conversion of accrued pension benefits is not possible at the individual level.
  - Financial:
    - Conversion splits collective assets into individual pension accounts.
    - For conversion of past accrual, two methods: the standard method or the value-based asset-liability management method. Both set limits to value allocation but allow some decision-making freedom.
    - Transparency requirements on intergenerational effects: parties must consider (i) market consistent valuation old vs. new contract, (ii) benefit projection old vs. new contract in positive, neutral and adverse scenarios.
    - The shift to age-independent contributions implies a plan change compared to the current “average premium system” and involves intergenerational effects that must be explicitly considered, including potential compensation arrangements.
  - Implementation:
    - Implementation plans must explain control of financial and non-financial risks (governance, process, operational, IT risks).
    - Pension funds must explain management of pension administration risks.
    - Legislation requires an external auditor to review data quality and requires consideration of remedial actions before implementing the transition.
  - Bridging plan:
    - Under certain conditions, pension funds can apply an amended prudential framework (“transition FTK” rather than “nFTK”) that relaxes conditional indexation and benefit-cutting regulations until conversion.
    - If chosen, pension funds must draw up a bridging plan.

### Investments (Pension funds)
- Investment regulations follow the prudent-person principle.
- Article 13a of the Decree FTK requires a pension fund to set a long-term strategic investment policy covering objectives, policy principles, risk appetite, composition of proposed investment portfolio, room for deviation (limits), and policies for management of relevant risks.
- Pension funds must translate the strategic investment policy into an investment plan with specific target levels and ranges for each investment category.
- Specific limitations:
  - Pension funds prohibited from investing more than 10 percent of assets in the parent company (article 135 Pension Act).
  - Investments in derivatives permitted only if they contribute to risk reduction (hedging) or facilitate effective portfolio management.
- Reporting and supervision:
  - Pension funds must report full investment portfolio including target levels and upper/lower bounds; enables DNB periodic assessment of deviations between actual and target allocations.
  - DNB examines limit-setting processes in on-site investigations.
- Custody and depositary:
  - Pension law does not make the use of a depositary (as defined in IORP Directive Chapter 3) mandatory; internal controls and processes are deemed stringent enough.
  - Appointment of an external custodian is not subject to regulatory approval.
  - No explicit requirement for custodian independence in Dutch pension law.
  - DNB expects IORPs to design operations to guarantee “controlled and sound business operations” (article 142 Pension Act) and interprets assets not kept safe and segregated from the asset manager in a custody account as not controlled or sound.

### Accounting principles, valuation, and funding rules (Pension funds)
- Valuation and accounting:
  - Assets and liabilities of pension funds are valued at fair value for general accounting purposes.
  - Dutch accounting guideline RJ 610 specifies valuation of pension liabilities based on the yield curve of risk-free nominal rates published by DNB.
  - No separate valuation standards for prudential supervisory purposes; funding ratios are calculated on fair-value for minimum funding requirements.
- Funding requirements (FTK and Pension Act):
  - Two capital requirements:
    - The minimum funding requirement, derived from the EU’s IORP Directive, is equal to around 105 percent of the value of the liabilities. A 12-month moving average of the funding ratio is used (‘policy funding ratio’).
    - The risk-based capital requirement is calibrated so that the probability of underfunding within one year is no more than 2.5 percent. On average, the capital requirement is approximately 25 percent of liabilities, but can differ significantly between pension funds.
- Underfunding and recovery:
  - If underfunded, a pension fund must submit a recovery plan to DNB.
  - The FTK provides a flexible recovery period without a fixed end date; the recovery plan must be revised annually using a new recovery period (of the same length). The recovery period can be no more than 10 years, and the plan must show full coverage by the end of that period.
  - Under the FTK, reducing accrued pension rights and pension benefits is a measure of last resort; pension funds must demonstrate other measures have been applied and that reductions would otherwise be necessary only if unavoidable and not disproportionately harmful to members or employers.
  - Benefit reductions can be spread (“smoothed”) over the full recovery period to prevent sudden significant reductions.
  - Backstop requirement: if the funding ratio stays below the minimum funding requirement for five consecutive years, the pension fund must reduce benefits immediately to recover to the minimum funding requirement.

*Source: DNB; excerpts from the Netherlands Financial Sector Assessment (sections 57–72).*

### 73.      The principle of sound and ethical business operations plays an important role in

### 1nldea2024006 - 73.      The principle of sound and ethical business operations plays an important role in

### Sound and ethical business operations; governance, risk management and controls
- Legal and regulatory basis:
  - Article 143 Pension Act: principle of sound and ethical business operations in relation to governance, risk management and risk controls.
  - Section 21b FTK: pension funds must not exclusively or mechanically rely on ratings issued by a rating agency for creditworthiness assessment.
  - Section 22 FTK: periodic feasibility test assessing consistency between the financial business structure, the estimated pension results and the relevant risks.
- Required frameworks and procedures for pension funds:
  - Document and implement policies with respect to the control of risks.
  - Have a policy aimed at sustainable control of financial and non-financial risks (the latter includes compliance risks).
  - Implement procedures and measures for documentation of incidents and take measures when an incident occurs, aimed at controlling materialized risks and preventing repetition.
  - Provide for a systematic analysis of integrity risks and document and implement an integrity policy.
  - Clear and comprehensive requirements exist with respect to the prevention of conflicts of interest.

### Board responsibilities and documented objectives
- Article 102a of the PW: the pension fund’s board must document the objectives and principles, including the risk appetite, after consultation with the other bodies of the pension fund.
- Uses and role of documented objectives and principles:
  - Serve for internal decision-making, accountability, advising and supervision within the pension fund.
  - Guide all bodies within the pension fund and staff; important for sound and ethical business operations including total risk management and risk control.
- Additional governance rules: Article 31, part 1 and Article 143 of the Pension Act and the Code Pension Funds provide further rules for the corporate governance framework of pension funds.

### DNB supervisory approach to internal control systems
- Supervisory framework basis:
  - DNB’s supervisory approach of internal control systems is based on the IORP directive (EU 2016/2341).
- Requirements for pension funds under supervision:
  - Administrative and accounting procedures.
  - An internal control framework.
  - Appropriate reporting arrangements at all levels of the institution.
  - An effective risk management function capable of identifying, assessing, monitoring, mitigating and reporting on key risks in a timely way and promoting a sound risk culture.

### Transparency and disclosure requirements for pension schemes
- Pension Act requirements: pension scheme administrators must provide information to active members, deferred members, pensioners, former spouses, and other dependents.
- Specific disclosure obligations and timings:
  - On joining the pension scheme: the employer or pension administrator must provide details of the pension arrangement (so called ‘Pensioen 1-2-3’).
  - Once a year: members receive a Pension Benefit Statement indicating the level of the pension entitlement for old age pension and—if included in the scheme—the entitlement to survivor’s pension; includes indexation and expected levels under a normal, pessimistic, and optimistic scenario.
  - On leaving employment: confirmation of end of membership plus information on accrued capital / benefits and future indexations.
  - At pensionable age: confirmation including relevant information on benefit level and - if included - entitlement to survivor’s pension.
  - Each pensioner annually: a variable pension overview of the benefit paid over the past year and payment for the coming year.
  - Online pension status overview: pension administrators must provide information on members’ pension entitlements to the online pension status overview.
  - Specific-instance notifications: requirements to inform members and beneficiaries in cases such as indexation, employer failing to pay premiums, or changes in the regular rate of accrual.

### Risk-based supervision: DNB (ATM) and classification
- DNB supervisory method (ATM):
  - Risk-based; supervisory priorities derived from institution risks, macroprudential developments and sector-impacting risks.
  - Proportionality applied mainly according to size, to a lesser extent complexity and substitutability; simpler practice for smaller institutions.
  - Method is, where possible, data-driven and automated, using automatic scores and enforcing justification for deviations.
- Annual classification and programs:
  - Institutions classified into impact classes I1, I2 and I3; impact classes derive from the Risk Tolerance in Supervision Statement.
  - ‘Basic program’: starting level of supervisory activities and identifies an institution’s risk profile.
  - ‘Risk-based program’: includes supervisory activities and mitigating activities to lower excessive risk ratings to conform to supervisory risk tolerance.
- End-2022 impact class counts:
  - 90 class 1 insurers
  - 31 in class 2
  - five in class 3
- Recommendation 7:
  - Review the thresholds for impact classes in ATM, and incorporate additional elements of systemic importance, e.g., complexity, interconnectedness, and substitutability.
  - Consider using a slightly larger (and even) number of impact classes, e.g., four, for more granularity.

### Data-driven supervision: DNB and AFM practices and developments
- DNB data-driven tools and indicators:
  - Developed (early-warning) risk indicators based on QRTs, ad-hoc data-requests and annual qualitative information.
  - Quarterly quantitative reports used to calculate risk indicators: solvency ratio, capital score, liquidity score, economical calculated capital ratio, combined ratio, interest rate risk ratios and others.
  - Dashboards per (re)insurer and market/submarket (life, health, non-life).
  - Data specialists and supervisors assess reports; management team discusses market-level overview.
- Non-financial risk data tools and plans:
  - Introduced a survey tool improving data quality and user friendliness for undertakings.
  - Plan to build in 2023 a ’low code’ solution for business experts to build and manage risk models for non-financial risks, improving efficiency and time-to-market.
- Use of advanced techniques and pilots:
  - Experimenting with Natural Language Processing (NLP); use cases include:
    - Examining board minutes for influence of key function holders during on-site inspections.
    - Searching accountants’ reports for material findings.
    - A dashboard enabling supervisors to search for words across qualitative regulatory reporting.
  - Pattern recognition and outlier analyses based on machine learning to improve reporting quality and develop data quality validation rules.
  - Pilot: enabling insurers to use DNB’s validation rules in their own systems before submission.
- DNB reporting cadence:
  - Quarterly reports discussed in management team; developed reports for senior management.
- Recommendation 8:
  - Explore opportunities to develop largely automated supervisory tools based on advanced technologies – able to learn and adapt by using algorithms and statistical models – allowing processing of large data sets.
- AFM data-driven initiatives:
  - AFM’s supervision centered on effects on consumers and markets; medium- and long-term strategy informs yearly planning.
  - Trendmonitor and sector analyses used to create a Heatmap combining top-down and bottom-up information to set annual supervisory priorities.
  - For five largest insurance companies (representing almost 80 percent of the retail market), firm-specific supervision (account management) with dedicated supervisors.
  - AFM uses a ‘Company Canvas’ (being digitized into an ‘Institutional View’) and prepares an Annual Review for each of the five largest insurers.
  - AFM data products: ‘Data Dashboard Non-Life’ exists; plan/initiate ‘Data Dashboard Life’. 2023 analysis undertaken to determine necessary data for monitoring life market risks.
  - AFM uses survey data and web scraping for monitoring distribution and remuneration policy.

### On-site inspections: frequency, methodology, follow-up
- Inspection scheduling and frequency:
  - No minimum frequency of on-site inspections at insurers; no targets in terms of a specific percentage inspected annually.
  - DNB’s methodology: on-site inspections scheduled if an account supervisor deems it necessary based on risk assessment.
  - On-site inspections are not a standard component for small insurers; introduction of ATM in 2021 resulted in a declining number of on-site inspections for small and medium-sized insurers.
  - For pension funds, number of on-site inspections at smaller entities has declined, though overall engagement is intense during the transition phase.
- Notification practices:
  - While ad-hoc unscheduled inspections are legally possible, institutions are almost always informed at least eight weeks before an on-site inspection.
  - Every insurer receives a ‘supervisory calendar’ in December containing all supervisory activities scheduled for the coming year.
  - AFM can carry out unscheduled inspections, but only when needed, mostly for illegal market practices; recently there has been no need for unscheduled inspections in insurance or pension fund sector.
- Communication and follow-up of findings:
  - Findings communicated via final on-site report and a closing meeting; account supervisor present.
  - Follow-up performed by account supervisor, usually via a letter addressing required follow-up.
  - If DNB on-site report results in findings, a risk-mitigation program is started by the account supervisor.
  - Risk-mitigation program may require follow-up on each separate inspection finding, but lower-risk findings can be deprioritized (still expected to be followed up, but without specific monitoring).
  - AFM requests feedback from supervised institutions on its findings; for formal enforcement aimed at recovery, AFM always monitors the effect; for informal enforcement, follow-up choice depends on violation, chance of recidivism, and importance.
- Recommendation 9:
  - Expand the number of on-site inspections for insurers in the lowest impact class, as a backstop to the risk-based approach.

### Governance, internal controls and risk management supervision
- Supervision approaches:
  - Day-to-day supervision: governance, risk management and risk control of individual insurers and pension funds supervised daily, especially for larger pension funds; supervisors have regular meetings with the board, accountability body, senior management, and outsourced service providers.
  - Thematic sectoral reviews: typically include desk research and interviews and may result in overviews of good practices (example: liquidity risk management).
  - On-site investigations aimed at the governance process.

*Source: 1nldea2024006 - 73.      The principle of sound and ethical business operations plays an important role in*

### 93.      The assessment by DNB is based on the risk-based ATM approach. Insurers and pension

### 93.      The assessment by DNB is based on the risk-based ATM approach. Insurers and pension

### Supervisory approach and annual review
- DNB’s assessment is based on the risk-based ATM approach.
- Insurers and pension funds in Impact Class 3 are always assessed through an annual review consisting of desk research and interviews to verify risks.
- If risks are outside DNB’s risk appetite as supervisor, a risk mitigation process is started.
- In practice, at least two to four meetings per year are held by the supervisor with each of the large institutions as part of the annual review; key function holders are included in these meetings.
- Additional meetings can take place in the case of an ongoing risk mitigation program.

### Fit and proper requirements for board members and key function holders
- Board members and key function holders of insurers and pension funds are required to be ‘fit and proper’ individuals with the necessary knowledge, experience, and integrity.
- DNB ensures persons who effectively run the institution are fit and proper by assessing new holders; key function holders are always assessed when assuming a new position.
- When assessing fitness and propriety, DNB considers—inter alia—professional qualifications, experience in the insurance or pension fund industry, time commitment, track record of compliance with laws and regulations, and personal integrity.
- Assessment methods include reviewing the application and a potential interview.
- For entities supervised by both DNB and the AFM (such as insurers):
  - AFM advises DNB on these assessments and can participate in the assessment interview.
  - Both supervisors must agree to the appointment of such candidates.
  - DNB informs AFM when reassessments are considered and AFM may decide to participate.
- The result of an individual’s propriety assessment is valid until new facts or circumstances arise, regardless of the individual’s (future) role; DNB assesses if new circumstances should result in a reassessment.
- Severe impacts on propriety may lead to the release of the individual’s responsibility within the insurer or pension fund.
- The result of the fitness assessment is applicable to a specific role; individuals are to be assessed whenever their role changes or they are proposed for a different role in scope of DNB’s fit and proper assessment.
- New facts or circumstances are to be notified to DNB by both the individual and the insurer or pension fund.

### Collective competence, governance reviews, and supervisory dialogue
- The collective competence of the board provides input for the individual assessment to assure sufficient collective suitability is maintained.
- DNB considers potential conflicts of interest and potential impact if an individual were to act inappropriately.
- DNB reviews governance structure to ensure sufficient checks and balances to prevent conflicts of interest and ensure proper oversight.
- In 2022, DNB reviewed the board effectiveness of a board that changed from a two-tier to a one-tier board and conducted a deep dive to assess a board after changes that impacted board effectiveness.
- DNB has established a dialogue with insurers’ and pension funds’ supervisory boards to convey important messages and receive feedback on strategic developments; independent members were especially invaluable in crisis situations.
- Current practice of appointing independent supervisory board members is well established; further clarification in law would ensure consistency across institutions and in time.
- Recommendation 10: Further clarify the requirement of independent supervisory board members in law.

### Remuneration, behaviour, culture, and internal controls
- Insurers and pension funds are required to have a written remuneration policy reflecting the size of the organization and the nature, size and complexity of its operations; remuneration should add to good governance and should not encourage taking more risk than acceptable.
- Remuneration principles must also apply to outsourcing partners.
- Behaviour and culture are explicit elements of DNB’s supervisory methodology.
- DNB conducts deep dives into the system of governance and the effectiveness of internal control functions; the annual assessment of non-financial risk includes behaviour and culture.
- DNB issued a good practice on insurance risk management in 2018.
- For several insurers, DNB performed on-site inspections on the design and operation of the risk management function.
- Internal control functions were also assessed at the group level for Internationally Active Insurance Groups (IAIGs).

### Supervision of outsourced providers and governance inspections
- Governance, risk management and risk controls of major pension administrators and asset managers may be subject to DNB’s supervision pursuant to the outsourcing rules contained in the PW or pursuant to the WFT if third parties are themselves supervised entities.
- In the latter case, the focus is on the solidity of the relevant third party (governance, risk management and risk control assessed from the third-party’s point of view).
- In preparing for the transition to a new pension system, in 2022 on-site inspections were conducted to look into the governance of pension funds.

### Conduct supervision — Insurance sector (AFM role and risks)
- The AFM does not mediate individual customer disputes; the independent Financial Services Complaints Institute (Klachteninstituut Financiële Dienstverlening, KIFID) handles mediation and decisions.
- AFM analyses complaints from consumers to determine market problems; annual complaints data submitted by financial institutions are a source for AFM’s risk-based work.
- Key market supervision risks in insurance are a mix of legacy issues and evolving risks (e.g., technological developments).
- Legacy issues include unit-linked business (the “usury policies scandal”), profit-sharing policies, closed-books, and life insurance, plus enforcement of the ban on commissions and professional competence and remuneration.
- AFM has conducted research into profit-sharing insurance policies and continuously monitors settlement of unit-linked insurance policies.
- Developments like climate change and cyber crime affect insurability and can lead to uninsurability and unfair competition from abroad.
  - In 2021, AFM conducted a study into the impact of climate change: damage and losses caused by climate change are increasingly uninsurable.
  - In 2023, AFM published main conclusions of an exploratory study into the growing market for cyber insurance: ‘Cyber insurance: comparability requires attention of insurers’.
- In 2020/21, AFM investigated compliance with product development rules regarding the Insurance Distribution Directive and the ban on commission by authorized agents; shortcomings led to informal and formal enforcement actions.
- AFM also investigated compliance with professional competence standards and the obligation for intermediaries to publish their remuneration policy.

### Conduct supervision — Pension funds (complaints, information, and recommendations)
- The new Pension Act introduces requirements regarding complaints handling by pension funds and insurers:
  - Required: internal complaints and disputes procedure.
  - Defines a complaint as any expression of dissatisfaction.
  - Requires a complaints administration and a general description of the complaints procedure to be available to (former) participants, scheme members, former partners of scheme members, and pension beneficiaries.
- In the preparation phase, AFM will discuss focal points of its supervision with the pension sector and communicate good practices to meet new legal requirements.
- AFM research in 2022 produced key recommendations:
  - (i) evaluate your own complaints procedure and review in particular the definition of ‘complaint’;
  - (ii) ensure the complaints procedure is easy to find on the pension fund’s website via a maximum of one or two clicks and explain in simple language how the complaints procedure is structured;
  - (iii) measure satisfaction after handling a complaint.
- AFM’s Trend Monitor identifies important trends and related risks in the financial sector to inform strategy and annual planning; example: survey into use of pensions products with variable benefits and consumer interest alignment.
- Recommendation 11: Enact a clause which would allow the AFM to collect data points relevant to conduct supervision.
- Recommendation 12: Expedite work on the Life Insurance Dashboard, and (also for Non-Life) ensure that timely data is available.

### Supervision of the Pension Transition — DNB role, tasks, and preparations
- Since 2019, DNB has acted externally as economic advisor to the government and as the pension fund supervisory authority; since mid-2021 DNB started internal projects for the transition:
  - Development of assessment frameworks for risk-based supervision
  - Development of guidance for pension funds
  - Setting up a supervisory program
  - Setting up the internal organization
  - Monitoring and nudging the pension sector to prepare for the transition
- DNB supervises pension funds’ compliance with the law regarding the transition, in particular:
  - (i) the quality and governance of the decision-making,
  - (ii) the adequate management of financial risk,
  - (iii) the adequate management of operational and IT risk.
- With respect to conversion of past pension accrual into the new pension system, pension funds must report their intent for conversion to DNB, after which DNB must take a formal decision whether or not to impose a prohibition on the conversion.
- To achieve a fair and adequately managed transition, DNB aims to ensure sufficient supervisory capacity, avoid procedural bottlenecks, and provide transparency; remedial actions should take place as early as possible in the transition period.
- Separate assessment-frameworks have been developed for five main building blocks of the transition:
  - (i) bridging plan,
  - (ii) defining the participants’ risk profiles,
  - (iii) the new DC contract,
  - (iv) the conversion decision,
  - (v) non-financial risk management.
- The assessment frameworks describe how DNB will assess compliance with law and regulations and will be updated continuously.
- DNB developed guidance (Q&A’s, factsheets, good practices) for the pension sector on these five themes and discussed this guidance through roundtables, informal consultation with stakeholders (e.g., the MoSA, the Dutch association of pension funds, the Dutch association of insurers), and public consultation.

### DNB supervisory program — four phases and organizational readiness
- Supervisory Program in Four Phases:
  - Phase 1: The preparatory phase, in advance of a fund’s reporting their intent to convert.
  - Phase 2: The phase starting with a fund’s reporting of the intent to convert and ending with DNB’s decision whether or not to impose a prohibition.
  - Phase 3: The phase between DNB’s decision and the actual implementation of the conversion.
  - Phase 4: The phase after the conversion (the execution of the new DC contracts).
- Phase 1 activities: monitoring the pension sector’s progress, nudging the sector to prepare, sector communication (round tables and seminars), risk identification and mitigation at pension fund / pension administration organization level, including on-site inspections.
- Phase 2 activities: set up efficient assessment processes to handle many conversion intents in a relatively short time, including IT processes for processing reporting and supporting information; internal processes for assessment, record-keeping, decision-making and communication; identifying knowledge/skill profiles and required staff capacity; exploring IT automation opportunities.
- DNB organizational measures:
  - Additional supervisory capacity secured in consultation with the government and based on analysis of required number of supervisory staff.
  - Two new departments established:
    - (i) a department central in resource planning;
    - (ii) a department to steer the process of the assessments of pension funds’ conversion decisions.
  - Internal education: a training program is ongoing; further training for specific roles will likely be necessary.
  - Close cooperation with the AFM to ensure supervisory roles align.

### AFM supervisory program and priorities for the pension transition
- AFM identified main conduct risks under the new pension framework and during the transition phase:
  - Pension scheme members have an insufficient understanding of what happens to their pension (entitlements) during the transition phase;
  - Pension scheme members have an insufficient understanding of the status of their pension at the retirement date;
  - The pension scheme does not meet the needs and risk preference of the scheme members;
  - Scheme members possibly do not make the appropriate choices concerning their pensions;
  - Unrealistic expectations are raised in communications;
  - Scheme members cannot understand the developments related to their pensions;
  - The quality of pension advice is inadequate.
- Preparatory Phase AFM activities: advisory role in regulatory development, advocacy for transparency and understandability, guidance material for communication plans and effective communications, sharing expectations and guidance via roundtables and the AFM’s Transition Bulletin, attention to timely and clear recalculation information, discussion with the Dutch Pensions Register on supervision impact, exploratory research (e.g., survey into correctness of pension information, information provision on indexation, complaints handling, transparency and accountability on operating costs).
- Transition Phase AFM thematic reviews planned on:
  - communication during the transition phase including consequences of transformation of pension entitlements;
  - implementation of the risk preference inquiries;
  - timely, correct, clear, and balanced information provision of the transition Pensions Benefit Statements;
  - guidance for scheme members’ decision-making regarding their pension;
  - the pension fund’s statement on the establishment of its pension schemes;
  - complaints handling;
  - the quality of pension advice.

*1nldea2024006 - 93.      The assessment by DNB is based on the risk-based ATM approach. Insurers and pension*

### 122.      In the Post-Transition Phase, the AFM plans to supervise the timely, correct,

### In the Post-Transition Phase, the AFM plans to supervise the timely, correct, clear, and balanced information provision of the Pensions Benefit Statements for DC schemes and the recalculation letters.

### Post-Transition supervision of pension information and member guidance
- AFM plans to supervise:
  - timely, correct, clear, and balanced information provision of the Pensions Benefit Statements for DC schemes and the recalculation letters.
  - the online pension status overview to ensure that information will be timely and completely available for everyone.
  - risk preference inquiries.
  - the provision of guidance to scheme members in decision making regarding their pension (options).
  - the pension fund’s statement on the establishment of its pension schemes (clarifying the objectives, structure, risk appetite and options in respect of the scheme).

### Transition management and supervisory cooperation
- Activities conducted in cooperation with DNB, with biweekly meetings of both authorities in the Transition Management Board.
- Activities are communicated regularly, e.g., through the monthly transition bulletin.
- DNB and the AFM have launched a Platform with experts from pension funds and insurers (including representatives of the relevant associations) to regularly discuss practical questions throughout the transition period.
- DNB and the AFM jointly survey the pension sector to understand progress in preparing for the new framework; this survey is sent out twice a year as of 2022, and relevant findings are communicated with the sector.

### Recommendations related to the pension transition
- Recommendation 13: Closely monitor and proactively manage potential risks of the pension system transition for the authorities related to resources and legal risks. In particular, any communication from the authorities during the transition phase needs to be carefully phrased in order to minimize the risk of litigation.
- Recommendation 14: Closely monitor the effectiveness of internal control functions in the transition process, and emphasize the important role of all pension funds’ internal control functions in the transition process, including the actuarial function and internal audit. Effective systems of internal controls will be essential to ensure a high quality and good governance of the conversion decision-making and adequate management of financial and non-financial risks.

### Consolidated supervision and supervisory cooperation (insurance groups)
- For insurance groups under DNB supervision, coordination arrangements follow the Solvency II Directive; college of supervisor meetings take place at least annually with open exchange of information culminating in a joint risk assessment.
- Until 2023, two Dutch insurance groups were designated as IAIGs; one moved legal domicile to Bermuda, changing group supervisor.
- One IAIG under DNB supervision established a Crisis Management Group (CMG) several years ago.
- AFM is currently not participating in supervisory colleges of Dutch insurance groups but is an active member in relevant working groups: IAIS Market Conduct Working Group and EIOPA Committee on Consumer Protection and Financial Innovation.
- DNB coordinates closely with AFM on conduct of business matters.
- 2021 measure: Netherlands introduced a ban on the provision of cross-border services by third-country insurers; 38 insurers were affected by this measure.
- Recommendation 15: Require the IAIG to participate in the IAIS’s Monitoring Phase for the Insurance Capital Standard.
- Recommendation 16: Explore, in collaboration with EIOPA, ways to strengthen supervision of cross-border business, including a review of how cooperation through platforms could be strengthened, and promote further supervisory convergence.

### Macroprudential supervision practices and authorities' tools
- Solvency II reporting data used for macroprudential supervision include balance sheet, solvency position, composition of capital requirements; DNB also uses ad hoc data requests for particular risks (example: (i) return guarantees and (ii) mortgage exposures).
- DNB participates in bottom-up EIOPA insurance stress tests and contributes to the IAIS annual global monitoring exercise.
- DNB conducts macro risk analyses via the Financial Stability Department and supervisory divisions in a ‘micro-macro dialogue’; identifies most important macro risks twice a year in an internal risk report.
- Financial Stability Report published twice a year; example: autumn 2022 edition discussed the impact of inflation on insurers’ technical provisions.
- DNB publishes part of the Solvency II reporting data at aggregate and individual entity level on its website (including balance sheet data, investments, profit and loss accounts, solvency data, benefits, and premiums).
- Currently, insurers are not subject to macroprudential requirements based on Solvency II; the trilogue agreement on Solvency II introduces macroprudential tools related to liquidity risk management, including a ban on dividend payments. The amended Solvency II Directive will need to be transposed into Dutch law, presumably by the second quarter of 2026.
- 2019: Act on the recovery and resolution of insurers came into force, requiring systemically relevant insurers to draft a resolution plan.
- DNB performs a systemic relevance assessment of individual insurers every three years based on IAIS guidance, considering size, interconnectedness, substitutability and non-traditional activities; systemic relevance informs supervisory classification and intensity, and public interest test for resolution.
- Historical note: 2015 Aegon Group designated Global Systemically Important Insurer by the FSB; subsequent enhanced supervision and recovery/resolution planning. 2019 FSB suspended designation of insurers as systemically important; final discontinuation in 2022. FSB now bases considerations on IAIS Holistic Framework.
- Recommendation 17: Enact a clause in the WFT which would allow DNB to cap dividend payments of insurance undertakings, based on macroprudential reasons.

### Key Financial Soundness Indicators — Insurance sector (selected figures, in percent)
- Capital adequacy:
  - Assets / liabilities: 2018 114.5; 2019 112.8; 2020 112.6; 2021 113.7; 2022 115.0; 2023-Q3 114.8
  - SCR coverage ratio (Solvency II) – life, median: 2018 216; 2019 199; 2020 187; 2021 187; 2022 204; 2023-Q3 193
  - SCR coverage ratio (Solvency II) – non-life, median: 2018 186; 2019 180; 2020 186; 2021 192; 2022 188; 2023-Q3 175
  - MCR coverage ratio (Solvency II) – life, median: 2018 518; 2019 452; 2020 433; 2021 451; 2022 448; 2023-Q3 426
  - MCR coverage ratio (Solvency II) – non-life, median: 2018 422; 2019 402; 2020 414; 2021 422; 2022 400; 2023-Q3 393
  - Unrestricted Tier 1 capital / eligible own funds: 2018 91.0; 2019 90.0; 2020 90.7; 2021 91.5; 2022 90.9; 2023-Q3 91.0
  - Tier 3 capital / eligible own funds: 2018 2.4; 2019 3.3; 2020 2.8; 2021 2.8; 2022 3.3; 2023-Q3 3.1
- Profitability:
  - Growth in gross written premiums - life: 2019 5.2; 2020 -1.5; 2021 -2.0; 2022 -4.3
  - Growth in gross written premiums - non-life: 2019 4.3; 2020 3.3; 2021 2.0; 2022 3.2
  - Combined ratio - non-life, median: 2018 99.2; 2019 99.3; 2020 98.3; 2021 99.2; 2022 98.5
  - Return on equity – life, median: 2018 3.1; 2019 8.5; 2020 5.6; 2021 6.4; 2022 0.7
  - Return on equity - non-life, median: 2018 2.9; 2019 7.7; 2020 7.5; 2021 9.2; 2022 -2.9
- Asset composition and quality:
  - Bonds / total assets excl. unit-linked: 2018 44.9; 2019 44.6; 2020 42.9; 2021 41.9; 2022 32.9; 2023-Q3 31.0
  - Equity, participations / total assets excl. unit-linked: 2018 6.9; 2019 6.5; 2020 7.4; 2021 10.3; 2022 11.6; 2023-Q3 12.1
  - Loans, mortgages / total assets excl. unit-linked: 2018 23.3; 2019 21.7; 2020 22.6; 2021 24.0; 2022 24.1; 2023-Q3 23.5
  - Below-investment grade / total fixed-income assets: 2018 0.8; 2020 1.1; 2022 1.2
  - Weighted average of maturities (in years) – life: 2018 14.8; 2020 13.9; 2022 15.6
  - Investment yield – life: 2018 1.8; 2019 2.6; 2020 2.5; 2021 2.1; 2022 1.2
- Liquidity:
  - Lapse rate (based on contract numbers) – life: 2018 3.7; 2019 3.7; 2020 4.1; 2021 3.8; 2022 3.3
  - Lapse rate (based on volume) - life: 2018 6.7; 2019 6.8; 2020 6.5; 2021 6.0; 2022 5.4
  - Liquid assets / total assets excl. unit-linked: 2018 48.3; 2019 48.5; 2020 46.4; 2021 45.8; 2022 36.3; 2023-Q3 33.2
- Reinsurance:
  - Risk retention ratio (net premium / gross premium) - life: 2018 97.0; 2019 94.6; 2020 88.9; 2021 86.8; 2022 84.1
  - Risk retention ratio (net premium / gross premium) - non-life: 2018 98.3; 2019 98.2; 2020 98.2; 2021 97.8; 2022 97.9

- Note: "Liquid assets include government and corporate bonds, listed equity, cash and cash equivalents."  
- Source line in the appendix: "Source: IMF staff calculations based on DNB and EIOPA data."

### Key Financial Soundness Indicators — Pension Fund sector (selected figures, in percent)
- Funding level:
  - Policy funding rate, median – DB: 2018 112.1; 2019 107.1; 2020 101.5; 2021 113.9; 2022 123.6; 2023-Q3 123.7
  - Share of liabilities of PF below 110 percent funding ratio – DB: 2018 78; 2019 76; 2020 79; 2021 32; 2022 23
  - Share of liabilities of PF below 100 percent funding ratio – DB: 2018 62; 2019 63; 2020 65; 2021 1; 2022 0
  - Share of liabilities of PF below 90 percent funding ratio – DB: 2018 0; 2019 0; 2020 0; 2021 1; 2022 0
- Contributions and benefits:
  - Change in contributions – DB+DC: 2018 0.8; 2019 8.5; 2020 13.5; 2021 -2.1; 2022 15.6
  - Change in paid benefits – DB+DC: 2018 3.4; 2019 4.0; 2020 5.2; 2021 4.5; 2022 5.8
- Asset composition and quality:
  - Equity investments – DB+DC / Total assets: 2018 61.0; 2019 58.8; 2020 56.1; 2021 56.9; 2022 51.2; 2023-Q3 49.6
  - Bond investments – DB+DC / Total assets: 2018 28.0; 2019 28.4; 2020 29.1; 2021 32.7; 2022 31.4; 2023-Q3 31.0
  - Loans – DB+DC / Total assets: 2018 1.1; 2019 1.1; 2020 1.2; 2021 1.4; 2022 2.5; 2023-Q3 1.6
  - Non-investment grade – DB+DC / Total fixed-income assets: 2018 9.1; 2019 9.6; 2020 10.1; 2021 11.8; 2022 11.3; 2023-Q3 11.5
  - Foreign-denominated assets – DB / Total assets: 2018 53.3; 2019 55.8; 2020 56.4; 2021 57.0; 2022 50.5; 2023-Q3 49.8
  - Investment return, median – DB: 2020 9.0; 2021 3.7; 2022 -23.0
  - Investment return, median – DC: 2020 4.7; 2021 9.7; 2022 -16.7
- Liquidity:
  - Liquid assets / total assets – DB+DC: 2018 89.6; 2019 87.9; 2020 86.5; 2021 90.7; 2022 88.1; 2023-Q3 88.7
  - Contributions / benefits – DB+DC: 2018 105.7; 2019 110.3; 2020 119.0; 2021 111.5; 2022 121.8

- Notes from appendix:
  - "Equity investments include holdings in investment funds."
  - "Liquid assets include cash and deposits, listed equity, investment funds, and bonds."
  - Source line in the appendix: "Source: IMF staff calculations based on DNB data."

*Source: IMF staff calculations based on DNB and EIOPA data.*

### Appendix III. Follow-up to 2017 FSAP Recommendations

### Appendix III. Follow-up to 2017 FSAP Recommendations

### Institutional set-up — Cooperation under the twin‑peaks model
- Recommendation: Strengthen cooperation between DNB and AFM, with single approaches for culture, governance and integrity supervision; single approach for data exploitation and data analytics; earlier formal information sharing on problem files.
- Actions taken (Implemented):
  - AFM and DNB hold bilateral meetings at senior management level every six weeks to discuss matters potentially relevant to each other.
  - Team leadership from several departments (Account supervision, Policy departments and Expert centers) meet regularly every six weeks and meet twice per year to discuss the rolling calendar of supervisory activities and zoom in on specific topics (examples: supervision on ESG, and risk-based research).
  - When examining similar topics (examples: compliance function supervision, Interest Only Mortgages, New Pension Regulation), each supervisor operates within its own mandate and supervisory findings are shared by Account Supervision or in supervisory colleges.
  - DNB and AFM exchange methodologies where possible: AFM learning from DNB’s updated risk-based methodology (ATM); DNB inspired by AFM’s method for selecting informal intervention strategies.
  - Joint work on exploring supervision of Diversity, Equity and Inclusion.
  - Investigations found only a few instances where both supervisors requested similar information from supervised entities after offering the sector an extended time window to respond.
  - Data sharing occurs; systems and data analytics remain separate but knowledge is shared and efforts are increasing to integrate data systems (examples: CRM data and licenses). Checks ensure reporting requirements do not overlap and reporting burden is minimized.
  - Information sharing on problem files is standard practice based on the Cooperation Agreement; intensified and formalized in protocol files on solvency problems at insurers. AFM has taken seats at management level in DNB steering groups.

### Institutional set-up — Regulatory powers and policy rules
- Recommendation: Enhance powers of DNB and AFM to introduce technical regulation in accordance with national and EU legislation.
- Actions taken (Largely implemented):
  - Dutch instructions for formal legislation (‘Aanwijzingen voor de regelgeving’) generally limit granting regulatory powers to independent administrative bodies; minister approval may be required.
  - DNB and AFM can adopt policy rules that are self-binding on them when exercising supervisory powers but are not legally binding on supervised institutions; these policy rules are nonetheless highly relevant for supervised entities when complying with regulatory requirements (power based on Article 4:81 AWB).
  - Example: joint ‘Policy rule of DNB and AFM on fitness assessments’.
  - DNB and AFM have delegated regulatory powers based on provisions in the Dutch Financial Supervision Act (WFT) allowing de facto legally binding technical requirements specifying higher legislation.

### Institutional set-up — WNT‑3 remuneration cap and budgets
- Recommendation: Consider exemptions from the WNT‑3 for both supervisory authorities.
- Actions taken (Not implemented):
  - The Law on Standards for Remuneration for Senior Officials in the Public and Semi-Public Sector (WNT-3) would impose the mentioned salary cap on DNB and AFM. The current government decided not to submit the proposal for WNT-3 to parliament, preventing enactment of the law.
  - The MoF has not provided DNB and AFM greater autonomy in setting supervisory budgets; no actions foreseen. In practice, the WNT limits salaries for management levels.
  - Consequence: DNB has difficulty filling vacancies for insurance supervisors; salary differences affect recruitment (example: market add-ons for actuaries that DNB cannot match).

### Institutional set-up — AFM organizational structure and resources
- Recommendation: AFM should apply flexibility in resource allocation and strong guidance under a stable board of directors.
- Actions taken (Implemented):
  - Conduct of business supervision in the retail financial services sector has operated in current product-oriented team structure for seven years with positive experiences.
  - Product-oriented teams increased market knowledge and supervision tenacity; supported by cross-sector expert teams.
  - Ongoing review and optimization of organizational structure; Mr Jos Heuvelman member of AFM executive board since 2018 and reappointed in 2022 for another four-year term.
  - Central management of substantive risks strengthened; priorities are clear and flexibility in allocation is used consciously. Project-based approach to risk enables natural adjustments in capacity allocation and supervisory intensity.
  - Coordination via horizontal partnerships and steering groups (examples: coordination of data driven initiatives; Intermediary steering group for distribution channels; coordination of financial innovation).

- Recommendation: Allocate a substantial budget for big data capabilities at AFM.
- Actions taken (Implemented):
  - AFM allocated a substantial budget for (big) data supervision through a dedicated program office (“Spot-on”).
  - Program entails around 40 permanent staff in a central expertise unit maintaining and expanding the “Managed Data Platform” and building new applications and tools.
  - Multiple data analysts work in specific risk areas.
  - Roughly 10-15 percent of supervision resources are dedicated to data supervision.
  - Several big data initiatives implemented, especially in capital markets supervision and areas with reporting requirements; big data tools also in accounting and pension funds.
  - Efforts constrained by legal restrictions on consumer data and privacy laws; legal means to request consumer data currently limited. A new law proposal would make this possible but still needs to go through legislative processes.

### Institutional set-up — RMS follow-up on recommendations
- Recommendation: Risk, Management and Strategy Department should follow up on implementation of departmental recommendations.
- Actions taken (Largely implemented):
  - DNB’s Risk, Management and Strategy Department (RMS) evaluations conclude with findings and recommendations for investigated departments.
  - RMS reports in its annual report on the status of follow-up of implementation.
  - RMS requests a reflection report from investigated departments about a year after evaluation; the department reflects on changes made driven by the RMS evaluation.
  - RMS provides input to the Internal Audit Department on future audits to ensure follow-up needing extra attention can be considered for audit.
  - RMS is not responsible for implementation; responsibility lies with the concerned departments.

### Insurance oversight — Supervisory review, thematic work and methodologies
- Recommendation: Ensure sectoral thematic work is fully exploited by accompanying growth with account management and line supervision resources.
- Actions taken (Implemented):
  - Selection of supervisory themes is jointly performed by account and expert center teams and decided in the management team.
  - Themes are chosen and planned based on resources from expert center and account teams to execute investigations and follow up findings.
  - Supervisory teams systematically allocate resources to perform planned sectoral thematic work.
  - Sectoral investigations predominantly led by specialists from the Expert Center; on-sites conducted and managed by the on-site department.

- Recommendation: Tune FOCUS! to align with Solvency II as detailed reporting emerges.
- Actions taken (Implemented):
  - ATM methodology has succeeded FOCUS!. ATM uses Solvency II based QRTs (for financial risks) and qualitative questionnaires (for non-financial risks) to calculate data-driven ‘risk scores’ for several risk categories. These risks are used to prioritize supervisory actions.

- Recommendation: DNB to carry out random on-site inspections on data quality and validity; join selected AFM inspections of auditors; confirm predictive power of FOCUS! regularly.
- Actions taken (Largely implemented):
  - DNB’s Expert Centre Operational Risk and IT works on improving data quality; data quality addressed in on-site inspections and thematic supervisory work (example: review of LAC_DT in 2022).
  - DNB holds regular meetings with accountants and actuaries to discuss expectations and general findings.
  - DNB’s Expert Centre Data works continuously on improving data quality. Project “Dataloop” developed within DNB assesses data quality of supervisory reports, allowing validations from Python, SAS and XBRL; signals visible to reporting practitioners and supervisory institutions with chat function for communication; signals can be “approved” or “resubmission”; centralized communication, more efficient assessments and an audit trail with historical reporting assessments.
  - Dataloop currently used internally at insurers and pension funds; external roll-out to the insurance sector planned.
  - FOCUS! has been recently replaced by ATM; the predictive power of ATM has not yet been evaluated.

### Consolidated supervision — Intragroup transactions monitoring
- Recommendation: Given detailed information on intragroup transactions (IGTs), monitoring contagion risk could benefit from an IT tool.
- Actions taken (Partly implemented):
  - DNB has no sectoral dashboard for monitoring reported IGT data due to heterogenous and dynamic nature of reported data and limited risks identified in regular supervision.
  - DNB reviews IGTs on a risk-based and individual level (examples: assessing internal company reports, starting mitigation processes to address specific intra-group risks).
  - DNB performed a thematic review on IGTs in 2018; in recent years specific high levels of IGTs were phased out due to supervisory actions.
  - Regular sectoral in-depth analysis could be useful to identify risks, trends or outliers; DNB will consider incorporating this in a next ATM methodology update.

### Supervisory cooperation — EU coordination and supervisory convergence
- Recommendation: Authorities, working with EIOPA, should support coordination among member state supervisors to minimize supervisory arbitrage.
- Actions taken (Implemented):
  - Contributing to a level-playing field motivates DNB’s involvement in EIOPA; DNB cooperates with other NCAs to enhance supervisory convergence and minimize arbitrage.
  - Example: DNB strengthened supervision on contract boundaries for disability contracts and worked with other NCAs when undertakings under different supervision applied less strict approaches to ensure consistent application.
  - AFM increasing international collaboration and harmonization due to cross-border markets; AFM actively involved in influencing the European supervisory agenda.

### Recovery and resolution — Winding up regulation, guarantee schemes, resolution framework
- Recommendations:
  - Introduce regulation allowing fast advance payments on all insurance liabilities and require approval of both supervisors for portfolio transfers of insurance and pensions, except when resolution powers apply.
  - Evaluate introduction of an insurance guarantee scheme as an additional resolution tool.
  - Introduce a planned resolution framework similar to the EU Bank Recovery and Resolution Directive with regard to the insurance business model.
  - DNB, with EU authorities, should consider cross-border failure arrangements to ensure coordination and flexibility between resolution tools and authorities.
- Actions taken (Implemented):
  - Since January 2019 Dutch bankruptcy law includes a provision allowing the administrator and/or the resolution authority to continue regular payments and claim payments to policyholders after the failure of an insurance company; if remaining assets are insufficient to pay 100 percent of policyholder claims, a reduction in payments may be applied.
  - For portfolio transfers, DNB is the main responsible authority (see Section 3.5.1a. of the Financial Supervision Act).
  - An evaluation on the possible introduction of an insurance guarantee scheme is ongoing; a study on the topic was completed and sent to Parliament.

*IMF staff summary of Appendix III. Follow-up to 2017 FSAP Recommendations (source content).*

### 2022. A public consultation was launched in Spring 2023

### 2022. A public consultation was launched in Spring 2023

### Public consultation and follow-up
- A public consultation was launched in Spring 2023 to receive input from stakeholders.  
- Responses to the consultation are currently being analyzed.  
- As yet, there is no official government position on the matter and further work is being carried out to arrive at a conclusion.  

### Insurance sector — legislative and supervisory framework
- Since January 2019 the Insurance Recovery and Resolution Act (an amendment to the Financial Supervision Act) is in force.  
  - The law is based on the EU Banking Recovery and Resolution Directive with due regard to the insurance business model.  
  - Differences vs. banking: in insurance there is no risk of an insurance run; the resolution authority therefore has more time for its work, implying less frequent and detailed resolution planning is needed in Dutch legislation compared with the Banking Recovery and Resolution Directive.

### Solvency II implementation — company-specific parameters
- IMF recommendation: Evaluate methodology to allow for company-specific parameters for small size companies and credit insurers under Solvency II.  
- Action taken: Implemented.  
  - In the Netherlands, only one firm has requested to use undertaking-specific parameters.  
  - DNB uses the work from the dedicated EIOPA working group that looked into this Solvency II methodology in 2022.

### Solvency II implementation — internal model approval and model drift
- IMF recommendation: Maintain strong focus on internal model approval of material changes and on identification and monitoring of model drift risk; provide supervisory responses within market reporting timeframes without compromising quality.  
- Action taken: Implemented.  
  - DNB has a strong focus on requests for material (and minor) internal model changes.  
  - DNB installed an Internal Model panel (including heads of relevant department and the director of insurance supervision) that decides on all material changes, on-sites and mitigation regarding internal models.  
  - DNB set up a framework for monitoring ongoing appropriateness of internal models comprising:  
    - DNB principles for supervision in the use of internal models  
    - How DNB monitors internal models as part of the supervisory review process  
    - Process for internal model applications and changes  
    - Roles and responsibilities  
  - Identification and monitoring of potential model drift: for every (partial) internal model a report is prepared and discussed in the internal model panel.  
  - DNB actively uses the IMOGAPI (Internal Model Ongoing Appropriateness Indicators) tool of EIOPA.  
  - On a yearly basis a monitoring report for every (partial) internal model insurer is made describing strengths, weaknesses, attention points, changes, etc.  
  - DNB is the first NCA in the EU to use the new QRTs for internal models and the first to have made reports to analyze these.  
  - Requested decisions of DNB regarding model changes are of high quality and within time frames.  
  - An EIOPA on-site on DNB’s internal model supervision had no major findings and had a generally positive outcome.

### Solvency II implementation — VA and UFR monitoring
- IMF recommendation: Closely monitor elements that limit economic valuation of liabilities; use Pillar 2 powers and take well-defined actions for Volatility Adjustment (VA) and Ultimate Forward Rate (UFR) impacts.  
- Action taken: Implemented.  
  - DNB monitors on a quarterly basis the value of the liabilities and the solvency position for insurers with and without the UFR and VA.  
  - DNB developed a sustainable approach and forward-looking methodology to monitor the risk that future profits and developments will not be sufficient to cater for the UFR drag.  
    - Methodology includes a risk assessment updated every quarter and is an important component of risk assessment and scoring in overall supervision methodology.  
    - An escalation ladder was developed to guide measures at each identified risk level.

### Solvency II implementation — transition monitoring and technical elements
- IMF recommendation: Closely monitor transition into Solvency II, including UFR methodology, tax-loss absorbance capacity of tax credits, and VA methodology in next revision.  
- Action taken: Implemented.  
  - DNB was active in the 2020 Solvency II review to make the UFR methodology as economic as possible (investigations in the last liquid point and improving extrapolation method).  
  - DNB made proposals for the dynamic VA to diminish overshooting effects.  
  - DNB developed a forward-looking methodology to monitor the risk that future profits and developments will not be sufficient to cater for the UFR drag.  
  - DNB is investigating the use of LAC_DT within the sector and has published a question and answer.  
  - DNB closely monitors the VA and the development of the dynamic VA within the Solvency II framework.

### Macroprudential supervision — stress testing and search for yield
- IMF recommendation: DNB should carry out country-specific stress tests in addition to the EIOPA stress testing.  
- Action taken: Implemented.  
  - DNB is developing a top-down stress test tool based on SII QRT data. Several modules have been developed and back tested; next step is to include mortgages and loans.  
  - DNB contributes to the EIOPA Project Group building a top-down stress test methodology using SII QRT data.  
  - DNB’s project EIPHORIA assessed the impact of a low interest requirement on Dutch undertakings; the analysis included a projected low interest rate environment and can be considered a stress test. Some automatic calculations in the ATM model are stress based.

- IMF recommendation: DNB is recommended to closely monitor search for yield activities in the sector.  
- Action taken: Implemented.  
  - DNB conducted research in 2022 to visualize search for yield in Dutch financial institution investment portfolios; research concluded insurers engaged in this behavior more than banks and pension funds.  
  - Main asset allocation changes observed: shift from government bonds to mortgages for life insurers; non-life insurers increased share of corporate bonds; overall increase in allocation to alternative investments (private assets such as infrastructure).  
  - DNB regards Solvency II review regarding the Volatility Adjustment as an important policy initiative and liaises with the Ministry of Finance.  
  - DNB conducts thematic inspections in the insurance sector to establish quality of risk management regarding exotic investments such as private assets.  
  - Actions taken include:  
    - In declaration-of-no-objection (DNO) processes DNB paid special attention to limit excessive search for yield through limitations in the DNO and issuing supervisory expectations;  
    - DNB compared excess return objectives of large insurers to assess outliers;  
    - DNB is assessing adherence to the prudent person principle in a thematic review, including potential need for extra regulation.

### Supervisory reporting — insurance intermediaries and insurers
- IMF recommendation: Revise reporting requirement for insurance intermediaries and increase frequency from annually to quarterly; enhance reporting requirements for insurers with market conduct indicators.  
- Action taken: Not implemented.  
  - Reporting requirements for intermediaries (in total ~6,500) have been fully revised, but annual frequency is maintained. AFM considers yearly reporting appropriate given market size and heterogeneity; current yearly process takes around four months and increasing frequency would impose costs outweighing benefits.  
  - Regarding insurers: insurers yearly report information about consumer complaints and more detailed data on financial results of specific products, which AFM uses for conduct supervision.

### Conduct of business — product oversight and cross-border enforcement
- IMF recommendation: Include product regulation independently of provider location and establish mechanisms to credibly enforce national market conduct regulation for cross-border operations under freedom to provide services.  
- Action taken: Implemented.  
  - Product oversight and governance (POG) norms require financial institutions to develop products matching target market needs and ensure distribution to target market while preventing distribution outside it.  
  - Supervisory mandate regarding POG for cross-border services is divided between home member state (national conduct authority) and host member state depending on sectoral legislation (MiFID II, Solvency II, IDD, CRD, etc.) and whether services are executed via a branch.  
  - AFM aims to prevent mis-selling by supervising organization and application of POG in practice. To enforce POG for FPS operations the AFM cooperates bilaterally with other NCAs and contributes to ESMA and EIOPA (examples: penholder in ESMA Product Governance Guidelines, POG chapter in EIOPA Supervisory Handbook, participation in ESMA common supervisory action on POG, EIOPA POG peer review). AFM specifically cooperates with an NCA regarding POG supervision on a cross-border firm.

### Pension fund oversight — consolidated supervision and licensing
- IMF recommendation: PW should require group supervision.  
- Action taken: Not implemented.  
  - No direct supervision of companies to which pension funds outsource activities. MoF and MoSA investigated need for direct supervision of pension service providers; conclusion was that financial services to pension funds are diverse and already affected by PW, WFT and European regulations, and case for more supervisory powers not made convincingly.

- IMF recommendation: Introduce a license requirement for pension funds.  
- Action taken: Partly implemented.  
  - Licenses are required for APF (General Pension funds) and PPI (contribution-based pension institution) because these entities actively aim to acquire clients. Other pension funds generally execute a scheme for a company or sector and do not compete in a market for pension schemes.

### Pension fund oversight — macroprudential, liquidity, reporting, investments
- IMF recommendation: DNB carry out country-specific stress tests in addition to EIOPA stress testing.  
- Action taken: Largely implemented.  
  - Pension funds are required to report periodically to DNB the impact of prescribed market shocks (interest rate, currency) on their liquidity positions.

- IMF recommendation: Pension funds should be required to have liquidity plans that consider effect on hedging positions under stress.  
- Action taken: Implemented.  
  - Pension funds required to report periodically to DNB the impact of prescribed market shocks (interest rate, currency) on liquidity positions. Requirements for adequate liquidity management follow from prudent person principle and “in control operations” requirement in pension legislation.

- IMF recommendation: Enhance reporting requirements for pension funds with market conduct indicators complementing DNB reporting.  
- Action taken: Implemented.  
  - As of reporting year 2019, AFM conducts a broad annual survey amongst pension providers collecting market conduct indicators voluntarily from second pillar providers. No issues in collecting data. Included datapoints:  
    - Number of Pension Benefit Statement (“Uniform Pension Overviews”) provided to participants and number not provided.  
    - Number of complaints in respect of accrual of pension including number resolved and number resulting in change of pension rights.  
    - Whether a correction policy is available, including whether publicly available.  
    - Number of pensioners faced with reclamation of pension benefits.  
    - Number of complaints in respect of pension benefit payments including number resolved and number resulting in change of pension benefits.  
  - AFM translates survey responses into key risk scores and suggested supervisory interventions. In 2023 AFM improved process to translate market information into a robust risk scoring model. For many indicators, pension providers are compared against market average and peers to identify outliers. Two market conduct risk indicators in risk scoring model:  
    - In-depth analysis of not provided Uniform Pension Overviews per pension fund over years 2019, 2020 and 2021.  
    - In-depth analysis of complaints per 100,000 participants per pension fund over years 2019, 2020 and 2021.  
  - Goal: identify outliers with unusually high missing Uniform Pension Overviews or complaints over multiple years (potential long-term issue) and identify issues in a specific year vs previous years (potential short-term issue). Special attention to funds not reporting or reporting low data quality.

- IMF recommendation: Tighten prudent person regime closer to Solvency II definition.  
- Action taken: Largely implemented.  
  - New pension system provides that pension fund investment policy should be in line with risk attitudes of different age cohorts, tightening prudent person principle. Further tightening deemed unnecessary given new pension system.

### Pension fund oversight — recovery planning, feasibility tests, conduct of business
- IMF recommendation: Reinstate requirement to charge a pension contribution that does not increase coverage deficit under a recovery plan.  
- Action taken: Not implemented.  
  - After Parliament agreed on new Pension Law on 30 May 2023, recommendation not relevant as under new law from January 1, 2028 premium contributions can only be made in DC schemes.

- IMF recommendation: Introduce an additional feasibility test for pension funds with a projection period of about 10 years.  
- Action taken: Not implemented.  
  - Post-Parliament agreement on new Pension Law on 30 May 2023, recommendation not relevant as under the new law only DC schemes are allowed.

- IMF recommendation: Introduce mechanisms to ensure pension participants receive financial advice.  
- Action taken: Implemented.  
  - WFT covers financial advice on products; PW determines agreements made by employers and employees’ representatives. New PW strengthened provisions on information to and guidance of pension plan participants.

- IMF recommendation: Harmonize WFT and PW with respect to quality of advice and suitability of product.  
- Action taken: Partly implemented.  
  - WFT and PW have specific legal scopes; PW strengthened provisions on information and guidance but full harmonization was not aimed for.

### Pension reform — objectives and outcomes
- IMF recommendation: New pension system should aim to reinstate qualities of the system, confidence in a guaranteed defined benefit and add portability.  
- Action taken: Implemented.  
  - Key aims of new pension system: re-inspire confidence while keeping key strengths of current systems. To avoid undue promises under non-sustainable DB-schemes, going forward only DC-schemes are allowed. DC design improves portability. A collective reserve with clear requirements for filling and emptying ensures solidarity and collective risk sharing remain.

*Source: Excerpt from 2022 chapter (public consultation launched in Spring 2023) — IMF country report text.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2024/english/1nldea2024006.pdf_
