AFM | DNB Monitor of Loan Standards and Financial Stability 2026

AFM | DNB Monitor of Loan Standards and Financial Stability 2026

Summary

This report is the first monitor by the AFM and DNB on the relationship between mortgage loan standards and financial stability. The monitor was prepared at the request of the Minister of Finance and the Minister of Housing and Spatial Planning (VRO). The goal of the monitor is to provide annual insight into the risks arising from mortgage lending, so that financial stability is more prominently taken into account when setting loan standards. The monitor uses a fixed set of indicators that map relevant developments among households, financial institutions, and the housing market.

Since the introduction of statutory loan standards in 2013, vulnerabilities have gradually decreased. The average Loan-to-Value (LTV) and Loan-to-Income (LTI) ratios have declined, making households better equipped to withstand price drops or income shocks. However, for new loans, a slight increase in LTV and LTI ratios has been visible since 2022. Among other factors, this means that financial stability risks remain present. Mortgage debt is high from an international perspective, new mortgages still involve high LTVs, and an increasing number of households – particularly first-time buyers – are using almost their entire borrowing capacity. As a result, the interplay between debt accumulation and high house prices remains a significant systemic risk.

Prudent loan standards therefore remain essential for financial stability. Easing loan standards would, given the strained housing market, lead to more risky lending behavior and further upward price pressure. Stricter loan standards can reduce financial stability risks, but in the current tight market, they will have a short-term impact on accessibility. A broader policy mix – including reducing fiscal incentives and increasing the housing supply – is necessary to reduce structural vulnerabilities.

1 Introduction

Mortgage loan standards are important for financial stability because they ensure responsible credit provision to households. Since 2013, the loan standards have been legally anchored and are set annually by the cabinet in the Temporary Regulation on Mortgage Credit. Last year, the effectiveness and efficiency of the loan standards were evaluated for the first time. Following this evaluation, the Minister of Finance, jointly on behalf of the Minister of Housing and Spatial Planning (VRO), requested the AFM and DNB to monitor financial stability in relation to mortgage loan standards annually. Additionally, the CPB (Netherlands Bureau for Economic Policy Analysis) was requested to monitor the relationship between loan standards and the accessibility of the owner-occupied housing market. In this way, financial stability and the accessibility of the owner-occupied housing market play a more prominent role in setting loan standards. This report was prepared by the AFM and DNB and contains the first report on loan standards in relation to financial stability.

The housing market constitutes an important source of financial stability risks. Dutch households have relatively high mortgage debts, and these mortgages form a significant part of the loans of banks and other credit providers. Furthermore, there is a clear relationship between mortgage lending and the development of house prices.1 Rapid increases in house prices can – in combination with generous credit provision and increasing debt – increase the vulnerability of the financial system to downward shocks.

1 For example, a recent DNB study finds that easing loan standards in the Netherlands translates into rising house prices. 2 See, for example, BIS (2023). 3 This standard sets a maximum on the portion of income that may be spent on mortgage costs. The maximum mortgage costs can subsequently be converted into a maximum loan amount relative to income. 4 This aligns with the proposals made in the evaluation by SEO.

Loan standards make an important contribution to limiting these financial stability risks by capping the mortgage debt households can incur.2 The Netherlands has two statutory loan standards: one based on income (Loan-to-Income or LTI standard)3 and one based on the value of the property (Loan-to-Value or LTV limit).

With this annual monitor, we aim to provide insight into the development of risks to financial stability arising from mortgage credit provision. We do this based on a fixed set of quantitative indicators, with which we can (in the long term) make changes over time clearly visible. We look at indicators that make the size and buildup of vulnerabilities among households and financial institutions visible, and at indicators that show the buildup and materialization of risks. We measure the buildup of vulnerabilities, among other things, via the share of households with a high LTV and/or LTI, the share of households facing a rate review in the short term, and the share of interest-only mortgages. For the buildup and materialization of risks, we look, for example, at the development of house prices and at credit risks at financial institutions.4

The report is structured as follows. The next chapter describes the main developments based on the indicators. The report concludes with conclusions and recommendations. The Annex contains some additional indicators.

1 Introduction

2 Development of Indicators and Financial Stability Risks

This chapter describes the recent development of the indicators related to mortgage loan standards, and the main implications for financial stability. First, we outline the developments in the housing market and mortgage debt. Subsequently, we delve deeper into specific risk characteristics of the mortgage debt, such as the Loan-to-Value (LTV), Loan-to-Income (LTI), and rate reviews. These indicators together provide a good picture of the vulnerabilities among households and financial institutions.

Housing Market and Mortgage Debt

The Dutch housing market is strained and shows signs of overvaluation. Although the rise in house prices has recently flattened, house prices in the Netherlands have risen significantly harder (21 percent) than incomes in the same period (14 percent) since mid-2023. The price-income ratio has therefore increased by almost 9 percent during that period (Figure 1). The ratio is still lower than at the beginning of 2022, but with the recent increase, the Netherlands deviates from the eurozone as a whole, where the price-income ratio remained fairly constant over the past two years. The share of housing transactions where the asking price was exceeded was almost 75 percent in 2025 (Figure 2).

Figure 1 Price-Income Ratio Eurozone index, 2015=100 Source: OECD.

Figure 2 Overbids Share of transactions above asking price percent Difference between asking and transaction price on overbid (right axis) Source: NVM/Brainbay.

Due to the recent increase in credit growth, the mortgage debt of Dutch households remains high. Mortgage debt grew by more than 5 percent in the second quarter of 2025 compared to a year earlier (Figure 3). This is the largest increase since 2008. The total mortgage debt relative to GDP was just under 80 percent in the past two years (Figure 4). Despite the gradual decline since 2012, household debt remains high from an international perspective. For comparison: in the eurozone, the average debt ratio of households was approximately 50 percent in 2025.5

Figure 3 Mortgage Credit Growth percent per year Source: CBS.

Figure 4 Household Mortgage Debt percent of GDP Source: CBS.

5 This refers to the total debt of households, i.e., both mortgage credit and other forms of credit.

LTV and LTI Ratios

Since the introduction of statutory loan standards in 2013, households have on average become less vulnerable to price and income shocks. The average LTV of all outstanding Dutch mortgages is currently just above 50 percent (Figure 5). Almost 8 percent of mortgages have an LTV higher than 90 percent. This is significantly less than in 2013, when – after a sharp drop in house prices in the preceding years – more than 40 percent had an LTV higher than 90 percent and a third of all mortgages were 'underwater'. Due to the lower average LTV, homeowners and mortgage lenders are less vulnerable in the event of a house price drop. Where there was clearly a decline in the average LTV in the period 2015-2022, this seems to have flattened since 2022. The average LTI also shows an improvement compared to 2013 and is currently 3.2 (Figure 6). The average LTI decreased mainly in the period 2023-2024, partly due to income increases in those years, but has recently risen slightly. Of all mortgages, 9 percent have an LTI higher than 4.5.

Figure 5 Development of LTV Ratio Average LTV Source: DNB.

Figure 6 Development of LTI Ratio Average LTI (right axis) Source: DNB.

For new loans, the average LTV has increased in recent years, meaning that first-time buyers in particular remain vulnerable to a house price drop. More than half of first-time buyers6 take out a loan with an LTV higher than 90 percent (Figure 7). Since 2022, a net increase in the average LTV ratio for first-time buyers has been visible, reaching 89 percent in 2025. However, a similar pattern can be seen among movers, although the average LTV for this group is lower at 74 percent. The years-long gradual decline in the LTV of new mortgages has thus not continued since 2022. As a result, the LTV for new loans remains high, especially compared to other countries, where a maximum standard of 80 or 90 percent often applies.

Figure 7 LTV of New Loans Average LTV (right axis) Source: DNB.

6 We distinguish between first-time buyers and movers based on age. See the definitions in the Annex.

It is also noticeable that households taking out a new mortgage are increasingly seeking the limits of what they can borrow based on their income. In 2025, first-time buyers borrowed an average of 92% of the amount they could maximally borrow based on their income; for movers, this was 83% (Figure 8). This continues the trend of an increasing share of households borrowing almost maximally. This increases the risk for this group that mortgage costs become problematic after an income drop, potentially leading to more defaults and lower consumption.

Figure 8 Borrowing Capacity Used for New Loans Average borrowing capacity used (right axis) Source: DNB.

Other Risk Characteristics

Refinancing risks for households are limited in the short term. When the fixed-rate period expires, there is a risk that the new rate is higher, leading to an increase in mortgage costs. This risk appears limited for households in the short term. By the end of 2025, for approximately 4.5 percent of outstanding loans, the rate needs to be reviewed in the coming year (Figure 9). This share is relatively low because many households fixed their rates for a long time during the period of very low mortgage rates in 2020-2022 (Figure 10). Since the beginning of 2022, mortgage rates have risen significantly. Among other factors, households who need to review their rates in the coming year are expected to face higher rates in three out of four cases (Figure 9). It is likely that future refinancing will also lead to higher rates in most cases. In the period up to and including 2030, the fixed-rate period expires for 30 percent of mortgages. However, most households currently choose a fixed-rate period of 10 years when taking out a new loan.

Figure 9 Refinancing Within One Year percent, share in outstanding debt Source: DNB.

Figure 10 Maturity of New Loans percent, share in new production Source: DNB.

The declining trend in the share of interest-only loans in total mortgage debt continued in recent years. In the third quarter of 2025, almost 40 percent of mortgages at Dutch financial institutions consisted of loans on which no regular repayment is made, and for which no savings or investment product is linked (Figure 11).7 It is also noticeable that interest-only loans generally have a relatively low LTV. The share with an LTV higher than 75 percent is 7 percent for interest-only loans, while it is almost 20 percent for total mortgage debt. This reduces the risk that the value of the collateral is insufficient to repay the loan at the end of the term. This does not alter the fact that interest-only loans can be riskier in other respects than amortizing loans.8 For example, lenders have limited insight into the extent to which customers can continue to bear mortgage costs after retirement.

7 The ECB also counts life and investment mortgages as interest-only loans. These form 4 percent of the mortgage debt. 8 See also recent publications by AFM and DNB on this subject.

Figure 11 Interest-Only Loans Share of interest-only loans Source: DNB.

Finally, credit risks for financial institutions remain low, although a larger share of mortgages has recently been classified as risky. The share of non-performing loans (stage 3) at banks for Dutch mortgages has been at a low level of approximately 0.8 percent for years (Figure 12).9 However, the share of mortgage loans with increased credit risk (stage 2) has increased in the past two years. This is largely explained by tightening measures that banks have implemented in the risk management of interest-only loans. Despite the recent increase in the share of stage 2 loans, the direct credit risks of Dutch residential mortgages remain limited, partly because about a quarter of outstanding mortgages have NHG (National Mortgage Guarantee) coverage (see Figure A3 in the Annex). A house price drop therefore affects financial institutions mainly indirectly via its macroeconomic consequences.

9 Data from the Credit Registration Bureau (BKR) also shows that the number of households with mortgage payment arrears is low.

Figure 12 Credit Risks of Banks percent, share in mortgage portfolio Source: DNB.

3 Conclusions and Recommendations

Since the introduction of statutory loan standards in 2013, vulnerabilities arising from the large mortgage debt of Dutch households have gradually decreased. As a result, households are on average better able to absorb the consequences of a potential house price drop or negative income shock. The risks to financial stability have thus decreased. However, the mortgage debt of Dutch households remains high compared to other countries, and we see that the persistently strained situation on the Dutch housing market encourages homebuyers to engage in risky lending behavior. LTVs of new loans remain high, and households are increasingly seeking the limits of their borrowing capacity. In particular, first-time buyers are having to take on more debt to be able to buy a home. As a result, the vulnerability of first-time buyers in particular to a house price drop or negative income shock has increased on net since 2022. The combination of high debt levels and risky new loans means that the interplay between the housing market and mortgage debts remains an important systemic risk for the Dutch economy.

Prudent loan standards remain an important condition for limiting financial stability risks. From the perspective of financial stability, easing loan standards is therefore undesirable. This applies all the more given the current large geopolitical and macroeconomic uncertainty. Given the current shortage on the Dutch housing market, it is also likely that easing loan standards will encourage homebuyers to engage in more risky lending behavior to increase their chances of buying a home. This would drive up house prices further and increase risks to financial stability.

10 See the Accessibility Monitor for the Owner-Occupied Housing Market by the CPB.

The Netherlands has a high LTV limit compared to other countries. Against this background, international organizations such as the IMF and the European Systemic Risk Board (ESRB) advise the Dutch government to apply stricter loan standards to limit risks to financial stability. At the same time, tightening loan standards could further worsen the accessibility of the owner-occupied housing market, which is currently under pressure. For example, lowering the LTV limit could make it temporarily more difficult for first-time buyers to buy a house. In the current housing market, tightening loan standards is therefore relatively costly. In a more balanced housing market, with a better-functioning rental market, a lower LTV limit could promote financial stability.

The relationship between financial stability and loan standards depends on various factors. For example, a reliable valuation of homes is essential for the effectiveness of the LTV limit and for a good assessment of the credit risk of mortgage loans. Furthermore, this relationship must be viewed in the context of broader housing market policy. In the Netherlands, home ownership is stimulated by favorable fiscal treatment of the owner-occupied home. In combination with generous loan standards, this leads to high mortgage debts and house prices, which contributes to risks to financial stability. The pressure on the housing market is further increased by the fact that too few new homes are being built. A gradual phase-out of the fiscal stimulus for home ownership and an expansion of the housing supply contribute to households being less inclined to borrow maximally, thereby limiting debt accumulation and risks to financial stability.

3 Conclusions and Recommendations

Annex

The indicators in this Annex are not in the main text of the report, but are part of the set of indicators that we want to monitor annually to provide insight into the development of risks to financial stability arising from mortgage credit provision.

Figure A1 LTI of New Loans Average LTI (right axis) Source: DNB.

Annex

Data and Definitions

For this monitor, we use granular mortgage data from DNB, which contains information on a large number of characteristics of individual loans. This data relates to mortgage loans on the balance sheets of Dutch banks, insurers, pension funds, and investment institutions. In total, the dataset contains data on 87 percent of all mortgage loans on the balance sheets of financial institutions in the Netherlands. This is equivalent to 85 percent of the total mortgage debt of Dutch households.

The reporting of mortgage data by banks has been suspended from mid-2022 to mid-2025. The figures on bank loans in that period are based on information on outstanding mortgages as of mid-2022 and end-2025.

We distinguish between first-time buyers and movers based on age. We refer to a first-time buyer when the oldest person in the household taking out the mortgage is younger than 36 years.