Housing market quarterly report: Higher house prices anticipated despite additional supply

The fact that the demand for owner-occupied houses still far exceeds their supply is evident from recent housing figures. In April, the latest month for which figures are available, home prices were 0.7% higher than a month earlier, and more than 10% higher than a year earlier (see figure 2). This made it the tenth consecutive month with a double YOY growth rate. The same is not true of the largest cities. In Amsterdam, house prices actually fell slightly in the first quarter compared to the previous quarter. This was also the case in the last quarter of 2024. As a result, houses in the capital are now just over 3% more expensive than during the previous peak in 2022, while houses throughout the Netherlands are on average 10% more expensive than in the third quarter of 2022 (see figure 3). House prices in The Hague and Rotterdam are also closer to the previous peak than on average in the Netherlands.

Presumably this is due to the sale of rental housing by private investors. Whereas previously they bought many houses to rent out precisely in the large (student) cities, thus contributing to the rapid house price increases in those cities, they are now offering these houses for sale again. For example, many more houses were for sale in major cities in recent months than a year earlier (see also figure 9). This combination of more supply and less demand seems to have a strong price-pressing effect locally. It is notable, however, that prices in the municipality of Utrecht have continued to rise steeply, despite the fact that here, too, relatively many former rental houses have been put up for sale. Why Utrecht differs from the other three major cities is not clear.

Even twice modal income often not enough to purchase a home

In recent years, the ability to buy a home on two modal annual incomes has disappeared rapidly. This is because home prices have risen faster than the amount people can borrow based on their income. We recently showed that a median household income, the amount in which one half of households earn more and the other half earn less, has been insufficient to buy a median-priced owner-occupied home since 2018. By 2023, with such an income, you could only finance one in three homes that were sold that year.

Even households with an income of twice modal, which puts them among the 35% highest-earning households, have limited choices (based on a 4% mortgage interest rate). Regionally, the situation is most dire in the northern part of the Randstad to Eindhoven and its environs (see figure 4). In municipalities such as Amsterdam, Haarlemmermeer, Amstelveen but also on the Utrechtse Heuvelrug, barely one in three owner-occupied homes sold in 2024 was financially accessible to a household with two modal annual incomes.

But incomes obviously vary considerably between regions. Consequently, some regions have relatively large numbers of households with incomes of twice modal or even more. In 2024, this was true for as many as 43% of households in regions such as Utrecht, the Gooi and Vecht region and the Leiden and Bollenstreek agglomeration. In large parts of Friesland and Groningen, this share remained at 26% to 28%.

A median household income appears to be insufficient in all regions to finance the median-priced owner-occupied home. Households in Greater Amsterdam and the Haarlem Agglomeration appear to be the worst off. In 2024, households with median household incomes for those regions could finance barely 6% of homes sold in that year. Urban regions like Utrecht, The Hague, Arnhem-Nijmegen and Rijnmond fared slightly better, but even there about four out of five homes sold are out of reach of households with median household incomes. Regions where the situation is even slightly better are found mainly outside the Randstad, such as in Friesland, Limburg and Zeeland.

Existing owner-occupied housing is expected to become 8.6% more expensive on average in 2025 than in 2024. This leaves our estimate for 2025 unchanged from three months ago. For 2026, however, the estimated price growth has been revised upward, from 5.7% to 6.6% (see figure 5). We now expect house prices to rise month-on-month next year at about the same rate as this year. That the annual figure for 2026 is lower than the figure for 2025 is due to a relatively strong carry-over effect in the 2025 figure (see also: 'How to read economic figures'). Below we describe the main factors that drive our estimate.

Economic picture fractionally more economical, trade war an uncertain factor

Due to the trade war that the US is waging with the rest of the world, the economic outlook for the Netherlands is somewhat more subdued than we assumed in March. We expect somewhat lower economic growth and slightly higher unemployment in 2025 and 2026 than in our previous estimate. That may reduce the demand for (owner-occupied) housing somewhat. But at the same time, wages in the Netherlands are rising slightly faster than anticipated, which further pushes up demand for owner-occupied homes. So on balance, from a macroeconomic perspective, we see no substantial differences from our previous estimate. Even in a further escalation of the trade war, we expect limited impact on the Dutch housing market, given the sometimes opposite effects. Such as lower economic growth and higher unemployment on the one hand (less demand for owner-occupied houses), but on the other hand in the same scenario lower interest rates (more demand for owner-occupied houses).

Home buyers can borrow more and thus bid more

We now assume slightly higher wage growth than three months ago, allowing homebuyers to borrow as well as to bid more for homes. Assuming a 4% mortgage interest rate (by way of example), a household with an income of twice the median will be able to borrow about 12,000 euros more this year than in 2024 due to wage growth and the adjustment of the Nibud lending standards. In 2026, another EUR 18,000 will be added (see figure 4). We assume that this additional borrowing will increase house prices.

Moreover, it is conceivable that the borrowing space will increase a bit further in the coming period, as capital market interest rates are expected to fall slightly through the middle of next year. These interest rates determine how much it costs mortgage lenders to raise the money they can lend back to mortgage customers. The (expected) development of capital market rates are therefore a gauge of the (expected) development of mortgage rates. And by and large, the lower the mortgage rates, the more homebuyers can borrow, and the more expensive houses will become.

In addition, there are signs that, besides a mortgage, first-time buyers are using money from family more often and more frequently to buy a home, either as a gift or as a loan. Family mortgages and family gifts also contribute to higher home prices.

Sale of former rental properties offers some relief; more pessimistic about housing construction

Recently, former (private) rental homes have been sold on a large scale. This has caused an increase in the number of houses for sale and the number of transactions (see also section "Sale of former rental houses makes many more transactions possible, especially in the big cities"). Without this additional supply, house prices would probably have risen more sharply. All the more so because, according to NVM figures, former private homes are on average of lower quality, causing a composition effect for which the existing house price index does not fully correct. So the house price figures now reported may underestimate the actual rise in house prices, especially in places where many former rental properties are sold.

We expect the sell-off wave to soften somewhat toward 2026, as fewer tenants have temporary leases. Furthermore, we are slightly more pessimistic about housing construction than three months ago (see also section 'Housing outlook is lackluster in the near future'). Both are expected to cause the market to tighten in 2026, which is one of the reasons we have revised our house price estimate for 2026 slightly upward.

Fewer price increases anticipated in west of the Netherlands

Prices in the three largest cities have risen less sharply due to the turnaround by buy-to-let investors, and we therefore expect average price increases this year to be lower in the COROP regions in which these cities are located - Greater Amsterdam, Greater Rijnmond and the Agglomeration of The Hague. We foresee the largest average price increases in the provinces of Utrecht, Flevoland and in the COROP regions in Groningen and Limburg. In North and Central Limburg, and East Groningen and Delfzijl and surroundings we have seen house prices rise sharply for some time. They are also the regions where prices are furthest above the previous national price peak of 2022. In Delfzijl and surroundings, houses were more than 22% more expensive in the first quarter of this year than in the third quarter of 2022.

More than 17% more homes were sold in the first four months of this year than a year earlier (see figure 8). This rebound is driven by the sale of former rental properties. Especially in the large (student) cities, where buy-to-let investors were previously very active, the so-called buyout wave led to a sharp increase in the number of houses for sale. In the first five months of the year, the five largest municipalities had more than 30% more houses for sale on average than a year earlier (see figure 9).

This additional supply makes more transactions possible in the current housing market, with a housing shortage and a demand surplus. Amsterdam, The Hague, Rotterdam and Utrecht therefore saw very strong increases in the number of home sales in the first quarter, up to 28% more than a year earlier. The number of sales there in the first quarter was also well above the average number of first-quarter sales in the previous five years. Interestingly, the number of sales in Overig Utrecht and Overig Zuid-Holland was actually lower in the first quarter than in the previous five years (see figure 10).

We do expect the sell-off wave to soften over the next 12 months. The fact that so many landlords were and are able to sell their houses smoothly is probably because a significant proportion of tenants (about half according to Companen's estimates) have temporary leases of one or two years. After the introduction of the ‘Wet doorstroming huurmarkt 2015’ (a law directed to increase residential mobility) temporary leases took off. But as of mid-2024, temporary leases are prohibited in most cases. By 2026, therefore, the majority of tenants will have indefinite leases. This limits the opportunities to sell. According to figures from Kadaster – as mentioned before, this is The Netherlands’ Cadastre, Land Registry and Mapping Agency – private landlords and corporate landlords sold nearly 5,000 rental properties to owner-occupiers on balance in the first quarter of this year (figure 11). This is fewer than in the last quarter of 2024. Sales of second homes – in which there has also been a solid increase – add to this. Again, some of these are ex-rental homes. While a drop in sales in the first quarter is fitting for the season, this drop is somewhat sharper than in, say, the first quarter of 2024. This may be a sign that the sell-off wave is already softening somewhat.

We expect 227,000 existing owner-occupied homes to change hands this year, some 10% more than in 2024, followed by 223,000 sales in 2026 (figure 12). Together, that's slightly more than we were still assuming in the previous estimate, mainly because the sale of former rental properties will allow even more sales this year than we were counting on three months ago.

For 2026, we now expect somewhat fewer transactions. Indeed, next year, a decline in the sell-off wave is likely. Once virtually all temporary leases have ended and the bulk of tenants are back on indefinite contracts, it will be more difficult for landlords to sell the property as long as the tenant does not terminate the lease. Nevertheless, the number of transactions is still expected to remain fairly high in 2026, as we assume that the number of houses for sale will be high enough by then to allow for a relatively large number of sales. It does ensure that, as the market tightens, the number of homes for sale will gradually decrease again.

Housing outlook is lackluster in the near future

Housing construction can provide additional supply of existing owner-occupied housing. But it doesn't look like housing construction is going to take off anytime soon. Therefore, compared to three months ago, our expectations for housing construction are somewhat tempered, especially for the somewhat longer term. There are still many homes in the pipeline, but municipalities have been issuing fewer permits for new homes lately. That is usually a harbinger for new construction, albeit the decline is usually less sharp than the permit dip. For example, figure 13 shows that the decline in the number of building permits issued since 2021 corresponds two years later to the decline in new construction since 2023.

The national government aims to add 100,000 homes each year through housing construction, real estate transformations or by splitting existing homes. But despite this policy goal, annual new construction production has fluctuated around 70,000 homes since 2019. Including splitting and real estate transformation, the total comes out at around 86,000 homes per year. Last year, the figure was below that. Moreover, about 10,000 houses are also demolished each year. The dip in new construction registered since last year is partly the result of the drop in demand in the housing market in 2022/2023 (see also figure 14). In early 2022, mortgage rates rose in a relatively short period of time, causing an increasing group of buyers of – relatively expensive – new-build homes to drop out. The cost of building materials also rose, leaving developers with few options to lower new construction prices to shore up demand.

For a long time, sales of newly built homes for sale were on the rise. But now their sales seem to be stabilizing slightly (figure 14). Earlier, we expressed the expectation that sales of new construction homes would reach the ceiling after 2025. It is still too early to say, but we may already reach that ceiling this year. And that while we are still well below the sales numbers we saw in 2021.

Right now, downside risks dominate. In addition to nitrogen issues and grid congestion issues, the Kaderrichtlijn Water (Water Framework Directive) may further complicate housing construction starting in 2027. Housing construction is also plagued by structural bottlenecks, including a lack of building sites, slow processes and red numbers. The government had plans to address (some of) these bottlenecks. But with the recent fall of the Dutch Cabinet Schoof, it remains to be seen if and when this will happen. However, the plan to freeze rents in part of the social housing sector seems to be off the table. In time this will likely lead to more housing construction than would have been the case if the rent freeze had gone through. After all, less rental income would have significantly depressed the construction of social housing and made the realization of mixed housing projects more difficult.