Dutch housing market rumbles on: even higher price hike expected in 2018
Dutch Housing Market Quarterly
- House prices in the Netherlands rose by an average of 7.6% in 2017
- During the last quarter the price index was even 8.2% higher than a year earlier
- Highest increases are again the major cities and three Randstad provinces, but house prices are also rising fast in Flevoland
- We expect house prices to rise by 8 % in 2018 and around 7% in 2019
- This means that in just two years the average house will cost about €300,000
- This makes houses less affordable for first-time buyers and could lead to rising debts
- The strong fourth quarter boosted the number of sales to almost 242,000 homes – 27,000 more than in 2016
- Flevoland and Drenthe particularly stand out, with the number of sales increasing by more than a fifth
- There is a clear resurgence of market activities in areas outside the Randstad and bigger cities
- This, combined with better affordability and lower prices in these areas, leads us to anticipate a slight increase in the number of sales to approximately 250,000 homes in 2018
Regions outside the ‘Randstad’ are making a strong comeback
The Dutch economy grew faster in 2017 than in the nine previous years, not least because the housing market has also had a fair wind during the last year: thanks to falling unemployment and the almost unassailably high level of confidence among the population in the housing market, no fewer than 242,000 existing homes were sold, more than ever in the Netherlands. The flip side, however, is that houses are becoming more and more expensive, especially in the Randstad and Flevoland. As we predicted in mid-2017, the price index produced by Statistics Netherlands (CBS) and the Land Registry for existing owner-occupied homes rose by no less than 7.6%. In this Dutch Housing Market Quarterly we will examine in greater depth the regional differences in prices and sales and explain why we expect more houses to be sold and price rises to keep accelerating in 2018.
Never before had we seen so many houses sold in just three months as in the third quarter of 2017, we wrote in November. This record didn't hold for long, because in the last quarter of the year even more houses were sold: more than 66,000, as the figures from the Land Registry show. This brought the total number of transactions in 2017 to almost 242,000, which means that roughly 1 in 18 owner-occupied homes in the Netherlands changed hands last year – a turnover rate of about 5.5 percent.
That is not only more than in 2016, but is also a sharp rise compared to 2006, when the all-time high in sales preceding the crisis was set at about 210,000 units. In part this extraordinarily high number can be explained by demographic reasons: since 2006, not only are there 600,000 more households, but the housing stock has also increased by approximately 600,000 homes. Assuming that around 60% of these homes (roughly 360,000 units) have ended up in the owner-occupied sector, and coupled with a turnover rate of more than 5%, this would easily account for almost 20,000 extra sales due to the demographic shift alone. The fact even more homes were sold in 2017, however, would then seem to be the result above all of low interest rates, a catch-up effect in both the housing market and the Dutch economy as a whole, and because the alternative to buying, i.e. renting, has become much more expensive.
A few months ago we expected that the shrinking supply of affordable homes would become a limiting factor in the number of sales. Yet, the shortages on the housing market are not reflected in the number of transactions. Seemingly, they do however translate into higher prices for these homes (see ‘Prices’ further on in this Housing Market Quarterly).
Moreover, buyers appear, and are perhaps forced, to look outside the Randstad. Figure 1, for example, shows that there is little slack left in the provinces of Noord-Holland and Utrecht and the rate of sales is falling, but in Flevoland over 24% more homes were sold in the past year. In fact, there were double-digit increases in every province outside the three Randstad provinces last year. The fourth quarter was no exception to this trend, although Drenthe (+22.4%) and Friesland (+20.3%) led the field in these three months. Particularly striking is that the sharp fall in the number of sales in Noord-Holland in the third quarter (-8.6%) proved not to be a harbinger of things to come in the final quarter of the year: here too, slightly more homes were sold than at the end of 2016.
Since the price rises and extra repayments on mortgages have reduced the number of households in negative equity throughout the Netherlands and unemployment, including (hidden unemployment), is falling to levels reminiscent of the pre-crisis years, the Dutch housing market is surging ahead across a broad front, not only in the Randstad. A major factor here seems to be that not only the Randstad but also larger cities outside of this region are becoming too expensive for average earners, meaning that peripheral and rural regions are becoming relatively more attractive again. Indeed, research by Statistics Netherlands (CBS) has shown that more and more young families have been moving away from the major cities in recent years. Figures 2 and 3 show that one result of this process is a shift in the demographic composition of house buyers in the Netherlands. In 2007, homebuyers were often younger than 35 years, and usually first-time buyers, while in 2017 it has been precisely the older homebuyers, usually those already on the property ladder, who have been most active in the housing market (see Figure 2 and 3). It should be mentioned that the situation before the crisis was very different in terms of the labour market and financial institutions: young adults today are less likely to have a permanent job and can borrow significantly less. One could argue that the distribution of buyers in 2017 is therefore probably more in line with the present intergenerational distribution of assets and income.
Flevoland stands out in particular with regards the prices buyers pay for a house. In 2014 and 2015 the province was bumping along the bottom of the chart with year-on-year price rises of between 0.9 and 1.4%. In the top three, firmly occupied by Noord-Holland, Utrecht and Zuid-Holland, the price of an average home rose by more than 5 per cent annually in the past years. But in 2016, Flevoland moved up to follow hot on the heels of the three Randstad provinces. In the second quarter of 2017 the polder province even overtook Zuid-Holland and recorded the third highest price rises. And in the fourth quarter, prices in Flevoland even rose by 9.8% year-on-year, pushing the province of Utrecht into third place and moving closer to first-placed Noord-Holland (see Figure 4). In the latter province, prices rose for the fourth quarter in succession, by more than 10% annually. An average house there at the end of 2017 cost some 30,000 euros more than a year earlier.
If we compare the number of transactions in 2017 with price increases per province (Figure 5), we see that there is a negative, relatively strong and statistically significant correlation between the two: in provinces where sales are stagnating, such as Utrecht and Noord-Holland, prices are rising faster than average. And conversely, in provinces where the number of sales has risen strongly such as Drenthe, Friesland and Gelderland, house price rises have lagged somewhat behind the other provinces. This indicates that in the more relaxed markets of the peripheral provinces the balance between buyers and sellers is still more even, while there is a typical sellers’ market in the Randstad and all big cities. Flevoland was a significant outlier in the past year, as both prices and sales shot up. This could be down to the fact that supply in that region was still relatively high, and because a city such as Almere has become an important overflow city for buyers from both Noord-Holland and Utrecht. Also, it is not unreasonable to assume that those making an offer for relatively cheap Flevoland have the high prices of these two provinces at the back of their minds, which could push up prices in the ‘polder’ considerably.
Focus: our forecasts for 2018
Prices: houses becoming 8% more expensive
With house prices having become available also for the fourth quarter of 2017, we were able to (re-)estimate our price model for 2018. Based on these, we forecast that prices will rise by 8 per cent in 2018 and 7 per cent in 2019 (see Figure 6). In concrete terms, this means that the price of an average house in the Netherlands could well be around €300,000 in 2019 (Figure 7).
Our model-based estimates turn out even higher, but since our model does not take account of the stricter lending standards (i.e. stricter LTI and LTV norms) which have been in place since 2011, we have ultimately taken the price rises mentioned above as the basis. Eventually, the existing loan-to-income standards, in particular, will prevent prices from rising endlessly: at a certain moment buyers will be confronted with severe borrowing constraints. Yet, for the Netherlands as a whole, this should not be a key problem this year. This is partly due to relatively good affordability in a majority of housing market regions but also because buyers are increasingly using their own means to finance purchases; that is, there are more buy-to-let investors as well as buyers with wealthy parents active on the market.
One of the key factor behind the high price increases is that we expect interest rates to remain at a low level this year with an upwards trend not starting before the fourth quarter. This process will then continue into 2019, albeit at a low level (see De Groot, 2017). These low interest rates work into high price increases in two ways. On the one hand, low interest rates mean that owner-occupied homes in the Netherlands remain relatively affordable. Here it should be mentioned however, that affordability measured as net housing costs as a share of net disposable income has been under pressure because of the strong price rises in the last two years (see Figure 8), particularly in the more expensive regions like Amsterdam and Utrecht. On the other hand, low interest rates, which we see not only for mortgages but also for savings accounts, mean that demand from buy-to-let investors for owner-occupied homes is expected to remain strong.
Other factors – those, which have not been explicitly included in our model – contribute to our expectations of these high price increases in the coming two years. The sharp fall in the number of homes for sale is a major factor: competition between potential buyers is intensifying, forcing them more often to make a quick decision and make an offer high in excess of the asking price. The average selling time has now already fallen to 52 days, which we haven't seen since the housing market boom at the end of the 1990s. Meanwhile, according to the Dutch Association of Real Estate Brokers (NVM), offers above the asking price are made for more than a quarter of the houses for sale.
Furthermore, it would seem that this shortage will not be resolved any time soon: although in 2017 the number of new homes coming on the market did rise by 14% to 64,000, the highest number since 2009 (see Statistics Netherlands, 2018), this rate of new development is not catching up by any means with the shortage that arose during the crisis and is also not in line with the growth in the number of households of about 70,000 per year. Interestingly enough there are marked local differences in the amount of new building. Figure 9 shows that provinces like Flevoland and Utrecht rely more heavily on the new-built stock in satisfying housing demand. Looking ahead: Based on a falling number of building permits granted in 2016, and with an average completion time of 22 months, we do not expect new housing developments to be a major factor in dampening the pressure on the housing market in 2018 either. Yet, a positive sign then is that the number of building permits issued in 2017 has clearly increased (see Figure 10).
Another important factor is that the Dutch population remains optimistic about the housing market: based on the buyers’ confidence indicator of the Dutch Homeowners' Association we indeed see high and stable confidence, with the only concern being the rapid price growth. Nonetheless, prospective buyers are still willing to pay high prices and the propensity to move house has even increased in recent quarters (see OTB, 2017).
Sales: moderate increase to 250,000 houses
Apart from adjusting our price expectations, we see reason to adjust our sales expectations upwards too. In contrast to our previous Housing Market Quarterly, we no longer assume that there will be a fall in the number of transactions, but in fact expect the number of sales to rise slightly to 250,000 sales in 2018, a rise of 3.4% compared to 2017. This changing view is not only due to the ongoing desire for Dutch homeowners to move house, but also because fewer households are in negative equity in a growing number of regions in the Netherlands. As a result, there is a higher likelihood that this desire to move house can actually be turned into an actual move. In fact, since last year the areas outside the major cities have been busy in a catch-up race, which first-time buyers and households moving up the property ladder had already experienced earlier in cities such as Amsterdam and Utrecht, but also cities such as Groningen, Zwolle, Nijmegen, Eindhoven and Maastricht. A further factor is that households with considerable equity in these cities can move to municipalities elsewhere. How significant this effect is requires further research, however.
As already mentioned, with the exception of Amsterdam and Utrecht the falling number of houses on the market seemingly does not directly lead to fewer sales; instead it more or less directly translates into higher prices. Indeed, the number of homes on the market has been falling for several years, while the number of homes sold this year has risen by more than a tenth. Accordingly, the correlation between the annual change in supply and the annual change in transactions is weak and not statistically significant.
Moreover, the decline in sales we had previously forecasted for 2018 was partly based on trends in the third quarter of 2017, where a sharp fall in the number of sales in Noord-Holland and Utrecht seemed to foreshadow things to come in 2018. Strikingly enough, this decline did not continue in the fourth quarter, but there was indeed a huge rise in the number of sales in the rest of the country and only a marginal shrinkage and stagnation in the number of sales in the province of Utrecht and North-Holland. These remaining ten provinces together account for roughly three-quarters of the total sales market. A fall in the number of transactions in the Netherlands as a whole compared to 2017 would therefore indicate a significant break in the trend, where growth should also stagnate outside the Randstad. From our viewpoint, the more realistic scenario, however, is that the number of house sales in the country as a whole will more or less level off, with a major decline only realizing once affordability worsens significantly in all regions on the Dutch housing market.
Rise in mortgage debt remains limited
The previous and forecasted price rises are, as we have often argued, not really positive news – at least not for those household with underwater mortgages and lenders who have high share of underwater mortgages on their books. In the longer term, this extreme house price growth is unsustainable with a increased of stronger market downturns once a turning point occurs. Not only would this be bad news for the housing market in itself but would it leave deep marks on the Dutch economy as a whole. This is because the economy and the housing market in the Netherlands are strongly interwoven: rising house prices and sales drive economic growth not only directly through investments in the home and home-related, but also more indirectly through higher consumption elsewhere because homeowners generally feel richer and are more prone to spending money when they see they value of their houses increase (see also DNB, 2018). By contrast, stagnating or falling house prices can reinforce a downturn in the economy as homeowners keep a tight rein on their expenses; a situation that became painfully clear during the last crisis.
Another fundamental problem is that high price rises can lead to higher mortgage debt in the already highly indebted housing market in the Netherlands, despite the tighter lending conditions such as the loan-to-value standard. For the time being and on an aggregate level this has not become apparent yet. Figures 12 and 13 show that up to the second quarter there was a slight rise in net mortgage debt, which was, however much less pronounced than the rise in house prices. In combination with the fact that a growing group of individuals are able to buy a house mostly or even fully using their own assets, this could imply a polarisation of debt. Those who (fully) rely on a mortgage to buy a house have to loan more money because of the rising house prices. While this is perhaps less problematic for the stability of the financial system, on a household level this divide between lowly and highly indebted buyers can make some home-owners more vulnerable. Additionally it can lead to more inequality in wealth because a growing group of people will have no access to the owner-occupied housing market.
 The two-tailed p-value is 0.013.
 This is based on an error-correction model which uses real interest rates, real incomes, and nominal house prices as its input variables.
 It should be noted that new homebuyers are required to make repayments on their mortgages in order to qualify for mortgage relief, which can reduce the mortgage debt of this group at a macro level too. This means that they are slightly less vulnerable to price shocks, and by definition will already be building up their assets (compared to homeowners with an interest-only or partly interest-only mortgage).
The Dutch Housing Market Quarterly is a publication of RaboResearch Rabobank. The view presented in this publication has been based on data from sources we consider to be reliable. Among others. these include Macrobond, Land Registry, NVM, DNB, CPB and Statistics Netherlands.
The date of completion is February 7th, 2018.
This data has been carefully incorporated into our analyses. Rabobank accepts. however. no liability whatsoever should the data or prognoses presented in this publication contain any errors. The information concerned is of a general nature and is subject to change.
No rights may be derived from the information provided. Past results provide no guarantee for the future. Rabobank and all other providers of information contained in this brochure and on the websites to which it makes reference accept no liability whatsoever for the brochure’s content or for information provided on or via the websites.
The use of this publication in whole or in part is permitted only if accompanied by an acknowledgement of the source. The user of the information is responsible for any use of the information. The user is obliged to adhere to changes made by the Rabobank regarding the information’s use. Dutch law applies.
For more information. please call the RaboResearch secretariat on tel. +31 (0)30 – 216 6666 or send an email to ‘email@example.com’.
Christian Lennartz and Nic Vrieselaar and Lisa Hoving
Menno Middeldorp, Head of RaboResearch Netherlands, Rabobank
© 2018 an - Coöperatieve Rabobank U.A., Nederland