House prices on course to hit new peak
Dutch Housing Market Quarterly
- More transactions than ever in a first quarter: 55,911 homes changed owners
- Prices rose by 6.8% in the first quarter compared to the first quarter last year
- House prices appear to be on course to hit a new peak: we expect rises of at least 6.5% for 2017 and 5% for 2018
- Sales of between 225,000 and 235,000 homes expected in 2017
- High levels of confidence in the housing market leading to faster sales
- Owner-occupied homes are still affordable but increasingly less accessible for starters
- Mortgage rates are expected to remain low until the end of the year. In 2018 we expect to see modest rises
- A new government must make a concerted effort with municipalities, housing associations, commercial parties and private developers to solve the growing housing shortage
The growth in the Dutch housing market continued unabated in the first quarter of 2017. The number of sales in the first three months of the year reached 55,911 homes, well over 30% more than in the first quarter of 2016. At the same time, as the rise in the number of transactions, the number of owner-occupied homes for sale also rose, which meant that shortages remained at roughly the same level. This contributed to the rise in house prices in the first quarter of 2017 compared to the last quarter of 2016 – by 2.0%. The year-on-year rise was no less than 6.8%.
In view of the strong growth in the first quarter and taking account of economic growth, rising incomes and persistently low interest rates being forecast, we are adjusting our expectations for this year: for 2017 as a whole we are assuming 225,000 to 235,000 sales and price rises of around 6.5%.
On the mortgage market, new approvals rose further during the first quarter of 2017. Higher repayments and tighter restrictions in the loan-to-value regime contributed to the total gross mortgage debt levelling off in the fourth quarter of 2016. Rising competition between mortgage providers is expected to help keep mortgage rates low in 2017 too.
In view of the ever growing housing shortage and sharp price increases this has caused, it is for the new government to work together with municipalities, housing associations, commercial parties and private developers to come up with new solutions for the housing market. A crucial task here will be to build affordable homes in both the owner-occupied and rental sector.
Chapter 1: Trends in the housing market
1.1 Nationwide trends in the market for existing homes
Further falls in mortgage rates and high levels of consumer confidence have contributed to the sharp rise in the number of sales during the first three months of 2017. With 55,911 sales, this is the strongest first quarter of the year ever measured. The fact that the number of transactions has grown so fast is surprising. Uncertainties abroad – such as Brexit and the election of Donald Trump as the new president of the USA – as well as the parliamentary elections in the Netherlands have apparently not translated into greater caution on the owner-occupied housing market. Normally speaking, the number of sales in the first quarter tends to be much lower than in the fourth quarter because it is beneficial for tax purposes to use one’s assets to buy a house before the end of the year, and any new regulations coming into force on 1 January often lead to an end-of-year spike (see De Vries et al., 2016). Since no tax changes were introduced with effect from 1 January 2017, this seasonal effect can hardly be detected this year: compared to the last quarter of 2016 only 5,000 fewer houses were sold in the first three months (-8%; see Figure 1.1).
While the number of transactions among households in all age groups has risen over the last two years, we are nevertheless seeing a clear shift in their relative shares. During the first quarter of 2015, younger households (<35 years) still accounted for more than half of all house purchases. Figure 1.2 shows that since then their share has fallen to around 45%, while in fact the proportion of young adults in the general population has still been rising slightly. The share of 35 to 55-year olds, on the other hand, has risen substantially. The preconditions on the housing market would appear to favour those moving house more than first-time buyers. They are accumulating more equity in their houses and consequently are in a better financial position to achieve their goal of moving house.
In addition, the trends we see in the different housing types diverge significantly. While the number of sales of apartments, mid-terrace houses and end-of terrace houses have risen by roughly 55% over the past two years, this figure was 78% for semi-detached homes, and even as high as 93% for detached houses (see Figure 1.3). A catch-up effect is evidently in full swing for large houses, which in fact fared particularly badly during the crisis in less urban regions.
The rise in the number of house sales was evident in all the provinces, but the differences between the provinces are considerable (see Figure 1.4). For example, the number of sales in Overijssel and Gelderland rose by almost 45%, and in Noord-Brabant some 40% more homes were sold in the past quarter than in the first quarter of 2016. Compared to those provinces, the number of transactions in Utrecht and Noord-Holland grew very little: by 17% in Utrecht, and not quite 13% in Noord-Holland. The fact that supply is drying up in these two provinces is preventing further rises in transaction numbers, particularly in their biggest cities: Utrecht and Amsterdam, and so is intensifying the housing shortage there. As a result, the year-on-year rise in sales in the city of Utrecht ended at only just above 3%, while the number of sales in the capital in fact fell by more than 4% (see Figure 1.5). There are a number of potential explanations for the fall in the sales in Amsterdam. The combination of rapidly worsening affordability for first-time buyers and those wishing to move house, and the wearing off of the catch-up effect since the crisis seem to be the obvious ones. But a more wait-and-see attitude among potential sellers in order to profit even more from the rapidly rising prices cannot be entirely ruled out either.
Shortage of houses for sale
The sharply rising number of sales during recent quarters is now producing rising shortages on the housing market. In the last edition of the Dutch Housing Market Quarterly (Vrieselaar et al., 2017) we concluded that the housing market is on course for historic housing shortages. However, based on the shortage indicator of the Dutch Association of Real Estate Brokers NVM, which measures the relationship between the total number of houses for sale and the number of sales in a particular quarter) we do see breaks in the overall downward trend. Just as in the fourth quarter of 2016, buyers had a choice of six homes in the first three months of 2017 (see Figure 1.6).
Consumer confidence and speed of sales
At the same time, consumer confidence and confidence in the housing market are high. Following a relatively sharp fall at the start of the year (-10 points), the Eigen Huis Market Indicator published by the Homeowners’ Association (Vereniging Eigen Huis - VEH) stabilised again in March 2017 and is currently at 110 points (see Figure 1.7). The proportion of optimists therefore exceeds the proportion of pessimists on the housing market, which usually manifests itself in a greater willingness to buy, strong growth in the number of transactions and also, if possible, a greater willingness to pay higher prices.
This high consumer confidence is also reflected in faster sales. During the first quarter of this year houses were on the market for an average of 77 days, a fall of 30 days since the first quarter of 2016. During the last three quarters, speed of sales has stabilised below the long-term average (see Figure 1.8).
Alongside the robust growth in the number of sales, average house prices have also risen strongly. The house price index (Prijsindex Bestaande Koopwoningen – PBK) rose in the first quarter by no less than 6.8% compared to the previous year. These are year-on-year rises that were not even seen in the five years before the crisis (2003-2008) (see Figure 1.9). Compared to the last quarter of 2016 the rise is 2%.
These price rises can partly be explained by the catch-up effect. The fall from the peak in the third quarter of 2008 to the deepest point in the second quarter of 2013 (-20.6%) was followed by a sharp price rise (+16.4%). Even so, prices are still on average 8.6% lower than before the crisis.
Broken down into house types, the house price index also shows that it is precisely the smaller homes (mid-terrace houses and above all apartments) that have become more expensive over the last year. Figures from Statistics Netherlands (CBS) show that detached and semi-detached houses have risen in value year-on-year by 4.9% and 6.3% respectively, whereas apartments were on average more than 10% more expensive in the first three months of 2017 than a year earlier.
What is also striking is that prices in the rental market are also picking up. Figures from the Pararius letting website (2017) show that prices in the private rented sector from the second quarter of 2013 – after which the recovery took hold in the owner-occupied housing market – until the first quarter of 2017 rose by almost 19%. At a national level, the owner-occupied housing sector and the (private) rented sector are both moving in the same direction, where price rises in the private rented sector are slightly higher. In the rented sector too, this is mainly due to increased demand and lagging supply.
House price trends go hand in hand with a divergence between the urban housing markets on the one hand and less urban and rural housing markets on the other hand. Houses sold in urban regions with shortages change hands at significantly higher prices. In Utrecht, these jumped by 8.6% and in Noord-Holland by as much as 10.5%. Similarly, in Zuid-Holland house prices have been rising faster than average (see Figure 1.10). The four large cities appear to have a major hand in this: in Amsterdam, Rotterdam, The Hague and Utrecht house prices have risen much faster than in the Netherlands as a whole (see Figure 1.11). These differences can be largely explained by rising demand among private households for owner-occupied homes, while the number of homes available to buy lags behind. At the same time, we are seeing increasing activity in these markets among buy-to-let investors (de Vries & van Leeuwen, 2017), which is intensifying competition for scarce properties for sale and driving up prices even more.
The average affordability of owner-occupied homes in the Netherlands is very good at present. The affordability index produced by Calcasa shows that the net monthly housing costs also continued to fall for buyers in the first quarter of 2017. New buyers are spending only 14.1% of their income on net housing costs, a fall of 0.3% compared to the last quarter of 2016 (see Figure 1.12). This is to do with the fact that the fall in mortgage rates and rising disposable income in the last three years have had a greater effect on affordability than the sharp house price rises.
Just one indicator for average affordability in the Netherlands is problematic, though. Firstly, the regional differences in house prices are greater than the regional differences in incomes. Affordability of owner-occupied homes is therefore first and foremost a regional matter. Secondly, the Calcasa index looks only at households that recently bought a house, and therefore enjoy favourable mortgage conditions. Figures from the Woononderzoek Nederland (WoON ) show that the actual net housing costs for all homeowners in the Netherlands in 2015 amounted to around 28% of disposable household income. In addition, the affordability index only includes households that succeeded in buying a house. Households that have not been able to get on the property ladder because of high prices are not considered. Accordingly, we might not bedealing with an affordability problem at the national level,, but may well see an accessibility problem in the market for owner-occupied homes: potential first-time buyers are less likely to succeed in buying their first home. The final point is that an isolated survey of the owner-occupied housing market is a shortcoming in itself.
Homeowners in the Netherlands have much lower mortgage-to-gross-income ratios than renters in both the public and private rental sectors. This can partly be explained by the income differences between both groups. On the other hand, even as house prices have been rising in recent years, buyers have been able to benefit from falling mortgage rates, while rents have at best remained the same, but mostly have been rising. It is therefore not surprising that renters are now spending almost 40% of their disposable income on housing costs (Statistics Netherlands, 2016). Figures from the PBL Netherlands Environmental Assessment Agency (Planbureau voor de Leefomgeving) (PBL, 2016) therefore show that the proportion of renters with a payment risk has increased enormously in the period from 2002 to 2015 – from 5% to 18%, whereas this has remained very low in the owner-occupied housing sector (see Figure 1.13). A side-effect of this trend is that many renting households may be less successful in the long term in buying their own home. In view of the tighter lending conditions, potential first-time buyers are being forced to build up much more capital before they can buy a house. This is much more difficult if they are having to pay a higher proportion of their disposable income in rent.
1.2 New homes and investments
The picture is mixed as regards new homes. On the one hand the number of new owner-occupied homes sold has risen by 10% to more than 33,000. The total number of new homes coming on the market, including rented housing, even rose by 12.5% to almost 54,000. Although this rise not as sharp as the rise in sales in the market for existing homes, it is good news for the housing market shortage.
On the other hand, the number of building permits issued in the last year has unexpectedly fallen again. Throughout 2016 as a whole, some 51,000 consents were granted, which is 5% less than in 2015. In view of the time between obtaining permits and putting a new house on the market (an average of sixteen months), it would therefore appear that these new homes will not reduce the pressure on the housing market this year. We expect that the number of new homes will reach around 60,000 only after 2018 at the very earliest, assuming that property developers and housing associations will be able to make more workers and resources available to invest and build new homes.
The fall in the number of permits is seen mainly in the rental sector: while the number of permits for owner-occupied homes in 2016 contracted by 2.8% compared to 2015, the number of permits for new rental housing fell by as much as 10% (see Figure 1.14). This can partly be explained by the weak investment position of housing associations, resulting amongst other things from the landlord levy and the much more stringent conditions concerning commercial letting, known as non-DAEB activities (Services of General Economic Interest) activities, under the 2015 Housing Act. Where during the crisis they still accounted for 50 to 60% of all permits for new rented housing, their share in 2016 fell below 30% (-12,956 permits between 2009 and 2016; see Figure 1.15). Commercial property developers, on the other hand, succeeded in increasing their market share from 27% to 67% (+3,488 permits for rented housing).
Owner-occupied housing development also comes under the non-DAEB activities of housing associations. As already mentioned, this is still possible only under very restricted conditions, which has meant that in 2016 as a whole only 861 building permits were granted to housing associations, almost 7,000 less than in 2009. This fall was partly compensated by self providers. Last year this group in fact obtained more consents than before the crisis. They now account for some 27% (9,712 consents) of all consents for new owner-occupied housing (Statistics Netherlands, 2017).
Housing shortages continue to increase
Given the forecast for the growth in the number of households in the Netherlands until 2020 by an average of 62,500 households a year (Statistics Netherlands, 2016), the pressure on the housing market will continue to intensify. Since around 10,000 homes are demolished each year, a production in excess of 70,000 homes is needed simply to prevent the housing shortage from increasing any further. We have discussed at length the reasons for the low production in De Vries et al. (2016), which include capacity problems in the construction sector, municipalities deferring their land sales and investment restrictions on housing associations. This situation has still not improved, and so it is clear that the new government, together with municipalities, housing associations, the commercial construction sector and private developers must rise to the challenge to build more homes. In view of recent developments in the housing association sector, the biggest challenges would appear to lie here (see also Hendriks, 2017). However, we believe that a wider discussion also needs to entail a broad discussion on how the allocation of new areas for development can be better coordinated with the transformation potential of existing buildings and the built environment (see also PBL, 2016).
1.3 Adjusted forecasts for 2017 and 2018
In the previous Dutch Housing Market Quarterly (Vrieselaar, 2017) we predicted that the house price index would rise by 5% in 2017 and that between 220,000 and 230,000 homes would be sold.
The trends we have described in the market for existing owner-occupied homes and new homes have led us, however, to adjust our prices, and to a lesser extent the transactions, in this Dutch Housing Market Quarterly. We expect that persistent housing shortages, rises in incomes (see Badir et al., 2017) and sustained low interest rates (see Chapter 2) will push up prices further, leading to an average price increase of at least 6.5% in 2017. We also expect that the housing market will only begin to cool slightly in 2018. Although the shortages in the housing market will increase, higher interest rates and high price rises will ultimately reduce affordability. We therefore expect prices to keep rising, although levelling off slightly, to at least 5% in 2018 (see Figure 1.16). As far as transactions are concerned, we expect to see between 225,000 and 235,000 sales in the market for existing owner-occupied homes in 2017 (see Figure 1.17), depending on trends in the number of houses for sale.
We believe that it is also up to the new government to keep a close eye on these trends in the housing market. The fact that the housing market is currently breaking record after record is increasingly difficult to explain merely in terms of a catch-up effect and should not be unequivocally applauded. If the housing market continues to grow at this rate and the supply of both new and existing homes lags behind rising demand, housing costs will swallow up an ever larger proportion of disposable household income, certainly in the rental sector but also in the owner-occupied sector. Since the accessibility of the owner-occupied housing market will decline further under these conditions and will be reserved for households with a relatively high income and/or from a wealthier background, we will see a negative effect on aggregate wealth accumulation and wealth distribution in the Netherlands. This effect will be intensified by the ever widening price differences between the hot markets, such as Amsterdam, Utrecht and Rotterdam, and the weaker housing market regions outside the Randstad. Moreover, the macro-economic risks that arise with surmounting private debt and increasing non-payment risks should not be underestimated.
 Initial figures for new mortgage approvals for the month of April 2017 may possibly indicate a cooling in transaction activities on the Dutch housing market (HDN, 2017).
Chapter 2: Trends in the mortgage market
2.1 Number of mortgage approvals rising, mortgage debt stable
During the first quarter of 2017 some EUR 17.7 billion, seasonally adjusted, was lent in new home mortgages (see Figure 2.1). This was a rise of 39.9% compared to the first quarter of 2016. Approvals for refinancing or extra borrowing amounted to EUR 5.2 billion, seasonally adjusted, in the first quarter of 2017 (see Figure 2.1), a rise of 48.9% compared to the first quarter of 2016. For the remainder of 2017 we expect the level of new approvals to rise further, in line with the rise in the number of house sale transactions (see Chapter 1).
The total gross mortgage debt fell slightly in the fourth quarter of 2016 by EUR 823 million, to EUR 664 billion (see Figure 2.2). Since mid-2015 the volume of approvals has again been greater in each quarter than the volume of repayments and gross mortgage debt has been rising. The fourth quarter of 2016 was the first quarter to break this trend. Figures for total mortgage debt in the first quarter of 2017 were not available at the time of writing. Nonetheless, we do expect a rise in total mortgage debt in 2017 that shadows the rise in the number of mortgage approvals.
Set against the expected rise in the number of new mortgage approvals are the extra repayments on mortgages. As long as interest rates on savings remain low, many households will choose to repay their mortgages or build up their capital in a bank savings mortgage rather than putting their savings into a regular savings account. We expect these extra repayments to continue for the time being. The reintroduction of the increased gift tax exemption may also lead to extra repayments in 2017.
The rising number of sales is having an upward effect on the total mortgage debt. We expect this effect to be only slightly greater than the expected extra repayments, and so the level of mortgage debt in 2017 will rise slowly, just as in the first three quarters of 2016.
2.2 Mortgage rates still low
Mortgage rates remained low during the first quarter of 2017 (see Figure 2.3). The average rate for fixed-rate periods of longer than ten years is now 2.86%, while the variable rates are still below 2%, at 1.95%. A little more than a year ago, in March 2016, these rates were still within a bandwidth of 2.3-3.1%.
In the short run we expect mortgage rates to remain low (De Groot, 2017). Firstly, In the run-up to the American elections, capital market rates rose slightly but have now fallen back again. In view of the uncertainty surrounding economic growth and the relaxed monetary policy being pursued by the European Central Bank (ECB) it is likely that capital market rates will remain low for the time being. The ECB will continue its extensive bond-buying programme until at least the end of 2017 in order to lower long-term interest rates. We expect a ten-year swap rate of around 0.8% over a period of three months and 0.75% over a period of six months. Our 12-month forecast is 0.6%. Secondly, compared to banks, insurers and pension funds are less dependent on external finance because they issue mortgages as an investment of the pension or insurance premiums paid by their members. These parties are playing an ever greater role in the mortgage market and through increased competition are exercising downward pressure on mortgage rates.
Since mid-2015, most homeowners have been choosing a long fixed-rate period. More than half of the total value of new or renewed mortgages is tied up in a fixed-rate period of six to ten years (see Figure 2.5), with a ten-year period being the most popular (HDN, 2017). Around 24% of the total value of new and renewed mortgages are fixed for longer than ten years. Within this category, most choose a fixed-rate period of twenty years. As soon as interest rates start to rise again, this share may well increase again, as was the case in 2006 and 2007.
When deciding on a long fixed-rate period, it is not only the historically low interest rate itself, but also what is known as the notional interest rate that plays an important role. The maximum mortgage sum will depend, apart from the value of the house, on income and interest rates. If borrowers choose a fixed-rate period shorter than ten years, it is not the actual interest rate but a notional interest rate of 5% that is used in the calculations. As mortgage rates fall, the amount that house buyers can borrow on the basis of the notional interest rate becomes relatively smaller. Those who want to borrow as high an amount as possible are therefore increasingly choosing a fixed-rate period of at least ten years. In fact, among house buyers younger than 35 years, more than 80% choose a fixed-rate mortgage of at least ten years (DNB, 2016).
All in all, the mortgage market is moving in the same direction as the housing market. As the number of sales rises, and because we expect the number of new mortgage approvals to be greater than the extra repayments, existing mortgage debt is expected to rise.
Badir et al. (2017), Economische Kerngegevens 2017 (only in Dutch), Economisch Kwartaalbericht, Rabobank, Utrecht.
CBS (2016), PBL/CBS Regionale prognose 2016-2040 (only in Dutch), Centraal Bureau voor de Statistiek, Den Haag.
CBS (2017), Meer vergunningen nieuwbouwwoningen vierde kwartaal (only in Dutch), Centraal Bureau voor de Statistiek, Den Haag.
DNB (2016), Overzicht Financiële Stabiliteit (only in Dutch), voorjaar 2016, De Nederlandsche Bank, Amsterdam.
De Groot (2017), Rente en Valuta: Nieuw realisme? (only in Dutch) Economisch Kwartaalbericht, Rabobank, Utrecht.
De Vries et al. (2016): Dutch housing market benefits from favourable market conditions. Dutch Housing Market Quarterly, Rabobank, Utrecht.
HDN – Hypotheken data netwerk (2017): Ondanks daling in maand april blijft het aantal hypotheekaanvragen in 2017 stijgen (only in Dutch), HDN, Amsterdam.
Hendriks (2017), Verslikken de woningcorporaties zich in hun eigen ambities?, FTM.
PBL – Planbureau voor de Leefomgeving (2016), Betaalrisico's in de huur- en koopsector, 2002-2015 (only in Dutch), PBL, Den Haag.
Vrieselaar et al. (2017) Scarcity expected to slow down growth of Dutch housing market. Dutch Housing Market Quarterly, Rabobank, Utrecht.
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 May 15th, 2017.
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, Jesse Groenewegen, Nic Vrieselaar and Leontien de Waal.
Barbara Baarsma, Director Knowledge Development Rabobank
© 2017 - Coöperatieve Rabobank U.A., Nederland