28 May 2025
Despite recent ECB rate cuts, the average interest rate on mortgages is expected to increase further. This is because of lagged effects from the latest hiking cycle. The ECB blog shows that the resulting drag on consumption could last at least until 2030.
Mortgage contracts often have interest rates fixed for long periods, or even for the entire duration of the contract. This implies that the average rate paid by households evolves only gradually with changing policy rates. In periods of rapid monetary policy changes, this can lead to paradoxical situations.
For example, interest rates on outstanding mortgages are currently expected to increase further despite recent policy rate cuts – even though the ECB aims to stimulate the economy, this lag can drag on consumption.
The transmission of monetary policy through mortgage interest rate payments is critically dependent on the interest rate environment at the onset of the cycle, as well as on the distribution of mortgage types.
What matters in particular is the share of adjustable-rate mortgages (ARMs) and fixed-rate mortgages (FRMs) of different durations. These types of mortgages are spread unevenly in the euro area, both geographically and across households. This, in combination with the fact that current market expectations are suggesting that the euro area is not returning to the low inflation and low interest rate environment that prevailed until 2021, implies that interest rates on the stock of mortgages will continue to rise. It will have a range of effects on private consumption across countries and households.
Using household-level data from the ECB’s Consumer Expectations Survey (CES), we show that mortgage payments are expected to rise in coming years, despite recently falling monetary policy rates. This could further dampen consumption through the so-called cash flow channel via mortgages. Moreover, we show that the uneven distribution of mortgage types will shape the path of interest payments across income groups at least until 2030.[1]
How are different mortgage types distributed?
About a quarter of mortgages in the euro area are pure ARMs. For the households holding those mortgages the effects of monetary tightening – and subsequent easing – were felt quickly.
The remaining lion’s share are FRMs. By design, many have been shielded from changes in the policy rate thus far. But this will only last until their rate is reset. And here again, mortgage design plays a decisive role – due to the duration of the interest rate fixation period of the loans. Around 10% of all mortgages are FRMs which will reprice to higher rates within the next three years. A further 20% will do so by 2030.
What is more, Chart 1, panel a, shows that there is significant cross-country heterogeneity, with ARMs more prevalent in Spain and Italy and FRMs more common in France and Germany. This means that the uneven impact of policy changes is not only influenced at the household level – geographical location plays a notable role as well.
Perhaps even more intriguingly, the differences between mortgage types within individual countries also have an impact. If a lower-income household[2] has a mortgage, it is more likely to be an ARM (see Chart 1, panel b). For households in the lowest 20% of the income distribution, 32% of mortgages are ARMs, compared with 17% for the highest quintile. This makes them more exposed to interest rate risk in the coming years than higher-income ones.
This structural difference across income groups reflects both economic constraints and different levels of financial literacy.
First, in the short term, it is generally more expensive to choose an FRM when prevailing short-term market rates are low. This is either because markets anticipate a rate increase which is already priced into an FRM, or because lenders demand a risk premium to bear the interest rate risk. So poorer, or more liquidity constrained, households tend to opt for the more affordable ARM contracts.
Second, higher-income households tend to have higher levels of financial literacy. This may also explain why they are more likely to choose mortgages that are fixed for longer periods. With awareness of the sensitivity of ARMs to changing financial conditions, they are willing to face higher up-front costs, and so hedge against interest rate risk.
Chart 1
Mortgagors by remaining interest rate fixation period on their mortgage
By country | By income quintile within country |
|---|---|
percentage of mortgagors | percentage of mortgagors |
 |  |
Source: CES and ECB calculations.
Notes: The statistics are computed from the CES housing module administered in February 2025. The shares are computed over respondents who have a mortgage outstanding and represent percentages of respondents rescaled by the annual survey population weights. Panel a) shows differences in mortgage fixation patterns across countries. Panel b) shows differences in mortgage fixation patterns across income groups within country. Panel b) depicts the structure of fixation of mortgage loans of each income quintile. The dashed blue bars refer to fixed-rate loans that are set to be repriced within the next three years. The blue bars correspond to the share of FRMs with longer fixation horizons. Panel a) depicts the average share of loans repricing, i.e. exposed to interest rate risk, over the period 2025-30 by income quintile. The latest observation is for February 2025.
Pipeline pressures differ across income groups
Let’s have a look how these factors affect average mortgage rates.
Over the 2022-23 tightening cycle, policy rate hikes have pushed up the interest rate paid by households on their outstanding mortgages. Microsimulations based on data from the CES, conditional on lending rates on new loans remaining constant, suggest that the average interest rate paid by households will continue to increase (see the yellow line in Chart 2, panel a). This prognosis holds despite the fact that recent ECB rate cuts have already translated into a fall in lending rates on new mortgages (see the blue line in Chart 2, panel a).
This happens because many FRMs that were issued in the period of low interest rates are still set to reprice at higher rates in the coming years. Additionally, borrowers who gradually repay their lower-rate loans are replaced by new mortgagors taking up new loans at higher rates.
Overall, the average rate for outstanding mortgages is expected to increase further as long as it remains below the rate on new loans.
Beyond those general effects, for the reasons described above, the transmission to mortgages has been uneven across income groups.
The increase was steeper among lower-income mortgagors, driven primarily by their higher share of ARMs. The rate paid by households in the bottom 20% of the income distribution reached approximately 3% in 2024, and 2.7% for budget-constrained households – defined as those who spend over 75% of their total income on housing and food expenses. Conversely, for households in the top 20% the average rate was still just above 2% (Chart 2, panel b).
Looking ahead, lower-income mortgagors will continue to experience slightly steeper increases until 2026. After this point, the average rate paid by richer households will start catching up.
Yet by 2030, the rate paid by lower-income borrowers will still have increased more – mostly because higher-income borrowers will still hold a greater share of long-fixation-period mortgages which won’t be repriced by then.
Chart 2
Mortgage interest rates by origination and income quintile
By origination | By income quintile |
|---|---|
annual percentage | annual percentage |
 |  |
Sources: ECB (MIR, BSI, CES) and ECB calculations.
Notes: The chart depicts the interest rate paths for the new business rate (yearly average), which is assumed constant after the last observation, and the outstanding amounts (end of year) of household mortgages. The microsimulations project the rates on outstanding amounts using the household-level distribution of rates paid and remaining interest rate fixation periods. For each projected year we calculate the interest rate paid by each household by adjusting the interest rate of the existing loans with expiring fixation periods in the year they expire, and the ARMs with the country’s latest available information for the new business rate on mortgages, to which we add a risk premium by income quantile calibrated to the macro country risk premia. The latest observations are for February 2025 for the CES, and March 2025 for MIR and BSI.
Higher interest payments dent household consumption
So, how does this translate to the economy at large?
Households can respond to rising interest payments by reducing their consumption or their savings – they have done both in response to higher interest payments. The CES shows that around 46% of mortgagors have reduced their usual spending in the past 12 months, either in response to, or in anticipation of, higher interest rate payments.
Those with ARMs, who were the first to feel the pinch from higher rates, were also the most likely to have increased principal repayments on their outstanding loans (Chart 3, panel a).
That means they were paying back their loans faster in response. It makes sense if the savings they would otherwise have made were earning interest at a rate lower than the one they were paying on their newly repriced mortgages. Instead of leaving their money in a low-interest savings account, they're using it to reduce higher-interest debt, which saves them money.
This pattern is also likely to occur as FRMs are repriced in coming years.
48% of households with a mortgage plan to continue to limit their consumption over the next year, although the extent differs depending on the type of mortgage and the respondent’s interest rate expectations.
Unsurprisingly, households with higher interest rate expectations plan to reduce their consumption more (Chart 3, panel b).
Households with FRMs that are about to expire are especially sensitive to the level of interest rates they anticipate. This can be deduced from the difference in planned consumption reduction between those expecting low versus high rates (yellow bars in Chart 3, panel b).
So, overall, this lagging impact on demand shows that the composition of mortgage contracts is crucial for the transmission of monetary policy, and this can lead to paradoxical developments.
Chart 3
Principal repayments and anticipated consumption reduction for different types of mortgagors
Principal repayments – past 12 months | Planning to reduce usual consumption – 12 months ahead |
|---|---|
percentage of respondents | percentage of respondents |
 | |
Sources: CES and ECB calculations.
Notes: This chart shows the responses to the question “Please think about the level of interest rates over the next 12 months. Which of the following actions are you planning to take over the next 12 months because of this interest rate level?”, as answered by mortgagors with fixed and adjustable interest rates. Fixed interest rates about to expire refer to interest rates on fixed-term loans that will reprice in the next two years to the market interest rate at that point. Panel a) shows the percentages that selected “Increase either principal payments or one-off payment on a mortgage”, while panel b) shows the percentages that selected “Reduce usual spending”. In panel b) we show the responses for different levels of interest rate expectations for the next 12 months. High and low interest rate expectations refer to the top 20% and bottom 20% of interest rate expectations. All responses are weighted using survey population weights.
The views expressed in each blog entry are those of the author(s) and do not necessarily represent the views of the European Central Bank and the Eurosystem.
Check out The ECB Blog and subscribe to receive future posts via email.
For topics relating to banking supervision, why not have a look at The Supervision Blog?
This ECB Blog complements other work trying to understand the drivers behind the ongoing consumption recovery. See, for example, Bobasu, A., Gareis, J. and Stoevsky, G. (2024), “What explains the high household saving rate in the euro area?”, Economic Bulletin, Issue 8, ECB; and Baumann, A., Caprari, L., Kocharov, G. and Kouvavas, O. (2025), “Are real incomes increasing or not? Household perceptions and their role for consumption”, Economic Bulletin, Issue 1, ECB.
Lower incomes are defined on a country-by-country basis, rather than across countries, so the prevalence of more ARMs for lower-income households is not due to country effects.