After the trough
Dutch Housing Market Quarterly
The Dutch economy emerged from the recession during the course of 2013. For 2014 we envisage modest growth of real Gross Domestic Product (BBP). On the domestic front, there are increasingly positive signs. However, unemployment continues to rise and we expect households to remain cautious both this year and next about stepping up their spending. That said, we do expect to see a small rise in private consumption next year, thanks to an increase in real disposable household income. And the expected gradual acceleration of economic growth (from 1% this year to 1½% in 2015) will lead to a slight drop in unemployment next year.
Housing market: more sales, prices inching up
With hindsight we can say that the Dutch housing market reached a trough in mid-2013, both in terms of prices and sales numbers. Already from the second quarter of 2013 we see a rise in the price index of existing homes (PKB) and in the number of sales contracts.
In the first quarter of 2014 prices rose by 0.5% on the fourth quarter of 2013. When compared to the first quarter of a year ago, we actually see a drop of 1.5%. However, during the past nine months prices have edged up - by 1.3%. The number of property transfers has also risen considerably. Several factors can explain why house-buyers feel now is a good time to buy. The main reason is the marked improvement in affordability. Confidence on the housing market has also improved further. In our view, these positive factors have greater impact on the market than the negative factors such as negative equity and further rising unemployment. We therefore expect to see house sales rise further to 130 thousand – 140 thousand this year (from 110,103 in 2013), and an average price trend of between 0 and +1%.
Both the Dutch economy and housing market will fare better in 2015 compared to 2014. At the same time, affordability will remain favourable, because house prices will not accelerate quickly, mortgage interest rates will remain at the current level and real household disposable income will rise. For next year we expect to see a further modest increase in sales and a nominal price rise of 1 to 3%.
In due course, the construction sector will also reap the benefit of these favourable housing market developments. We are already seeing the emergence of greater interest in newly built housing. However, the landlord's levy, which lessors have to pay over their social housing portfolio will hamper the recovery of the construction market for rental housing. We expect new construction output to stabilise in 2014, followed by an increase of 8% in 2015.
Payment difficulties getting worse
Total mortgage debt declined in 2013 and will rise little if at all in 2014, despite an increase in newly issued mortgages. Mortgage rates with a fixed-interest period of up to five years have fallen to a historically low level and are expected to remain low this year. The favourable outlook for the economy and the housing market does not mean that problems such as payment arrears and negative equity will be resolved quickly. Some one million households are in negative equity and this number will drop only gradually in the coming years as house prices rise again. The number of those in arrears on their mortgage payments has risen sharply and will continue to grow in the years ahead.
Marked regional divergence in price development
The decline in house prices of the past five years has not been equally pronounced in all regions of the Netherlands. In those regions where the price rise before the crisis was most prominent, the drop was relatively largest. Conversely, in regions that saw only a limited price rise before the crisis, the price drop during the downturn was less extreme. The region of De Haaglanden (around The Hague) and a number of regions in the northern wing of the Randstad urban conglomeration are an exception to this pattern.
Chapter 1: Unexpectedly high growth data
Last year, the Dutch economy emerged from the recession earlier than we had expected around the mid-year mark, and with surprisingly high growth figures. In the fourth quarter, Gross Domestic Product (GDP) increased by as much as 0.9%. This was mainly due to strong growth in corporate investment as a result of a one-off rise in car sales. Because it was only a one-off rise, growth for the first quarter of 2014 is likely to show a relatively sharp drop. However, for the rest of this year, economic activity will pick up, leading us to expect a 1% rise in GDP for the year as a whole, compared to 2013 (table 1). The momentum will continue into next year, with growth of up to 1½% expected for 2015, mainly due to improved dynamic on the domestic front.
Exports benefit from return of European growth
Last year, export growth was very weak at only 1.4% year-on-year. We expect exports to pick up during the course of this year, as economic growth returns to the eurozone and increases further in our other important trading partners such as the UK and the US. During the course of 2014, exports are likely to receive a further boost from a depreciation of the euro vis à vis the dollar. Accordingly we expect the volume of Dutch exports to rise by 3¼%, accelerating to 5¼% in 2015, thanks to a stronger eurozone economy but a weaker euro.
More green lights on the domestic front
There are increasing signs of growth on the domestic market too. The fourth quarter saw exceptionally high growth in corporate investment - driven mainly, as we said, by high car sales. Companies were acting in anticipation of tighter tax regulations from 1 January 2014 relating to taxation on the private use of company cars. However, this was not the only growth factor. In the second half of 2013, investment in housing increased, having contracted on a quarterly basis for over two years.
In the final months of 2013 and the first months of this year, consumer confidence improved significantly. In April, the confidence indicator registered -5, just above the long term average of -7.4. The increased confidence is largely due to improved sentiment about the economic situation in general. In April, the sub-indicator 'economic situation for the next 12 months' registered the highest score in seven years. Although the sub-indicator 'purchasing willingness' also showed a slight improvement in recent months, consumers remain cautious when it comes to making large purchases. The improved confidence is therefore not expected to translate into more spending on a large-scale.
There are a number of reasons for the reduced pessimism among consumers. First, fears for a re-emergence of the European debt crisis have abated. Second, purchasing power has risen this year for the first time in four years. Inflation is expected to run at only 1% this year - a sharp drop on last year, when prices rose by 2.5% on average. Consequently, this year will see an end to the decline in real income. Purchasing power is further boosted by a number of tax cuts implemented by the government. These include a lowering of the marginal rate of income tax and further levy cuts for low-income households. Another factor possibly contributing to the improved sentiment is the restoration of calm on a number of important policy issues, such as the housing market and social security.
Consumers remain cautious
Despite the positive signals, this year we expect to see a further drop in private consumption. Last year, real disposable household income declined for the third successive year as incomes lagged behind inflation, employment contracted and taxes and social security premiums went up (figure 1). Macro-economic gross individual savings rose last year, which meant consumption volume declined at a faster pace than incomes. It is not surprising that savings are on the increase, since many home-owners have a net mortgage debt that is higher than the value of their home. This year, some of these households will continue to deleverage or else build up savings. The decline in employment means that greater purchasing power scarcely translates into higher total real disposable household income. Consequently, private consumption will decline by a relatively large 1% in 2014. Although household disposable income is expected to increase considerably next year, consumption will follow with a lag. Accordingly, private consumption will likely increase in 2015 by only ¼%.
More long-term unemployed
We expect unemployment to rise this year to an average of 7¼% of the workforce (15-74 years), from 6.7% in 2013. According to a recent study by Erken et al. (2014), unemployment is now above the level that could be expected on the basis of model forecasts. The explanation for this is not so much the reduced employment that has inevitably followed the drop in GDP, but an unexpectedly sharp rise in the labour supply. Despite the decline in available jobs, more people are looking for work. Although employment will inch up next year, the labour supply will expand apace. Consequently, unemployment will decline only modestly during the year and will again average 7¼% in 2015. In recent years, the drop in employment has led to a sharp rise in the number of long-term unemployed (longer than one year) (figure 1). Because the long-term jobless tend to find it hard to get work even when the economy picks up, there is a real risk that unemployment will decline only modestly once employment increases.
Chapter 2: Developments on the existing homes market
The second-hand housing market in the Netherlands bottomed out in terms of both prices and sales numbers in 2013. In July, the existing homes price index (PBK) of the Land Registry /CBS stood at 84.0 (vis á vis 2010=100). In March 2014 the index registered 1.3% higher, at 85.1. During the course of 2013, there was also a turnaround in the number of existing homes sold, reaching the highest level since 2010 in the second half of the year. In view of both recent and expected economic developments we foresee no further drop in either house prices or transaction numbers. In 2014, the market got off to a strong start. With a total of 28,963 sales of existing homes, this was the best quarter since 2008. Prices also rose on a quarterly basis – by 0.5%. The outlook for the remainder of this year and for 2015 is favourable.
The first quarter of 2014 had the strongest first-quarter sales performance since 2008. Based on the improved affordability and the more positive sentiment, we expect a further rise in the number of houses sold this year. However, reduced borrowing capacity and the threat of negative equity for a relatively large group of home-owners will dampen the pace of this increase. In 2015 real household disposable income will finally start to go up. And the stabilisation of house prices is expected to contribute to confidence in the property market. Consequently, we expect to see more houses changing hands in 2015.
Further rise in sales numbers
In the first quarter of 2014, 28,963 homes changed hands - the most in a first quarter since the start of the financial crisis in September 2008 (Figure 4). That said, the increase (seasonally adjusted) was less pronounced than in the second half of 2013, when growth averaged 16% on a quarterly basis. In the first quarter, growth was only 3%. In our view, this is a temporary deceleration. It should be borne in mind that a relatively high increase is manifested quicker from a low base.
The Netherlands Association of Real Estate Brokers (NVM) has registered a strong start to 2014. Sales contracts with NVM estate agents account for some 75% of the current market (showing up on average two months earlier than with the Land Registry). The NVM has registered 24,360 agreed sales – the best quarterly performance since 2008. Seasonally corrected, this is 6.6% more houses sold than in the fourth quarter of 2013, which already registered the highest number of sales since the crisis erupted. This increase would indicate higher growth in the number of sales registered with the Land Registry in the second quarter of 2014 than in the first quarter.
Fewer homes on the market
Thanks to the increased volume of sales, there was a decline in unsold housing stock in the first quarter of 2014. In our previous Dutch Housing Market Quarterly we explained that the supply of houses for sale could either rise or fall this year. With the market picking up, more home-owners could be encouraged to put their house up for sale. Meanwhile, in the past quarter, the rise in sales has so far led to a slight further decline in supply - by 4% on a quarterly basis (Figure 5). Thus the bargaining position is shifting gradually in favour of vendors, but for the time being, buyers are in a relatively strong position. This is also clear from the NVM’s tightness indicator, which divides total houses for sale by the number of transactions. This currently stands at 19, compared to 18 in the previous quarter. That said, this is a lower level than the 29 registered at the start of 2013 (Figure 6). An increase in sales will push the indicator down over time, but it is still a buyers’ market. Prior to the crisis, the indicator registered 6 to 7, on average (NVM, 2014).
Reasons for increased sales
The rise in the number of transactions since the second half of 2013 can be explained by three factors. First, affordability has improved despite the tightening or mortgage lending standards. This is because real house prices have fallen by over 30% since the peak of 2008, while real household disposable income has dropped by only 4.5% during the same period. Research by Calcasa (2013) likewise shows that affordability  improved throughout 2013. In the fourth quarter of the year, home-owners spent on average 17.9% of their net monthly income on housing costs. By comparison, this was as much as 30.6% in 2008. Home ownership has not been this affordable, particularly for first-time buyers, since the 1990s (Calcasa, 2013). As Francke and Van der Minne (2013) have shown, some first–time buyers are now buying houses that previously would have been beyond their reach.
A second reason for the rising sales is that pent-up demand has built since the crisis. Many owner-occupiers and potential first-time buyers were keen to move house, but delayed their purchase as prices were falling and they were unsure of their financial situation. Before the crash, over 30% of house-seekers succeeded in achieving their goal. During the slump only 18% managed to do so (WoON, 2012). This procrastination cumulated in an increased willingness among existing home-owners to take the plunge in the first half of 2012 (Figure 7). In view of the decline in sales in 2012 and the first half of 2013, we assume that latent demand has since intensified.
If we correct the mortgage amounts and house values contained in the data base of WoON2012 for recent developments, we can estimate that in the first quarter of 2012, 590,000 out of the 850,000 (68%) house-seekers among existing home-owners are not in negative equity. This group can therefore move along the property ladder without incurring residual debt. Releasing this pent-up demand depends on the extent to which confidence returns to the market. According to the home-owners' market indicator of the Home-Owners' Association, this confidence has shown an uninterrupted rise for 14 months since January 2013, and in April this year stood at 88%. This is the highest reading since 2008. In the past, confidence measured with a lag of 8 months had a predictive value for the six-month average of the number of transactions. Although this correlation has weakened somewhat in recent years, we nonetheless think that the marked improvement in sentiment will support a rise in sales. However, the rise will not be as strong as the indicator might suggest (Figure 8), on account of the problems with negative equity (see next section). Home-owners in negative equity are unlikely to be able to move house and will first have to build up savings or pay off additional debt.
A third factor underpinning increased sales is that the rental market remains an under-developed alternative. As a result of the house price slump of recent years, the net monthly costs per square metre for an owner-occupier are lower than for a comparable home in the social rental sector (Volkskrant, 2013). Moreover, private tenants are faced with high, means-tested rent increases. This acts as a stimulus to the demand for houses in the owner-occupied sector. Moreover, according to MVGM (2014)  building sites for the construction of private rental properties are generally too expensive for developers. If local authorities were to reduce land prices for this purpose, it would mean they would have to incur even more losses on their land portfolios. It is therefore unlikely that the private rental sector will constitute a suitable alternative to the owner-occupied sector any time soon.
Impediments to sales growth
Despite the above reasons for increased sales, we see two factors that will impede a rapid acceleration of transaction numbers. First, of the 850,000 potential house-seekers, over 260,000 are saddled with negative equity. Although in theory this residual debt can be included in the mortgage for another house, the monthly costs will be higher for these households. This can stand in the way of movement along the property ladder. Since we do not expect house prices to rise rapidly in the short term, this problem may act as a brake on the increase in house sales for the coming years. Accordingly, in some cases houses are on the market for a long time: 55% of the supply has been for sale for over a year (Figure 9). Thus we see an increasing duality emerge: houses new to the market are selling quicker, but homes that have been for sale for over a year are increasingly harder to shift (NVM, 2014). These houses are possibly over-priced, as many households are reluctant to turn potential negative equity into reality (CPB, 2013).
The second impediment is the expected rise in unemployment. Although a relatively large percentage of unemployed people live in the rental sector (63%, WoON2012), rising unemployment will dampen the demand for property. We expect unemployment to peak in late 2014 (see Chapter 1).
In 2015 we anticipate a slight decline in the number of households in negative equity, on account of nominal price rises and (additional) repayments (See box 1, Chapter 3). Employment will also increase slightly during the course of next year. Accordingly, we expect these impediments to sales growth to lose some of their effect in 2015, which will benefit the recovery of sales.
Conclusions and expectations for sales
We expect that the positive factors will have a greater impact on the recovery of transactions than the impediments. The positive effect is somewhat greater than we originally thought, as is evidenced by the strong sales performance since late 2013 and the pent-up demand that appears to be released in 2014. For the rest of this year we envisage a further rise in the number of houses changing hands, reaching a total of 130 to 140 thousand. In any case, 2014 will show the strongest performance in house sales since 2008.
That said, this expected rise will not compensate for even half of the over 100 thousand fewer sales compared to 2008. However, in (Piljic and Stegeman, 2013) there are various reasons to assume that we will not return to the pre-crisis level of sales. Nonetheless, the envisaged total for this year remains relatively low.
In 2015 we expect to see a further rise in the number of houses sold. The slight increase in household income after years of contraction will contribute to this. Moreover, in view of the modest price rise in recent quarters, we expect to see a further strengthening of confidence on the housing market. Finally, the recovery in prices will mitigate negative equity somewhat in 2015, thus gradually easing movement along the property ladder.
Prices of existing homes (PBK-index) rose by 0.5% in the first quarter of 2014 compared to the previous quarter. During the past nine months the index has inched up by a total of 1.3%. The trough was likely reached in June 2013, and 2014 will the year of a gradual recovery. For 2015, we believe economic recovery and modest income growth will bring a further rise in house prices.
Prices have bottomed out
Although house prices rose by 0.5% in the first quarter of 2014, a comparison with the first quarter a year ago shows a drop of 1.5% (Figure 10). Nonetheless, over the past 9 months prices have risen by 1.3% on balance. Accordingly, we do not find a year-on-year comparison to be a good measure, as this doesn't reflect current developments. If prices remain flat during the coming months, then from May this year, the index will in fact show a year-on-year rise. In view of recent developments and based on our expectations, we conclude that prices already bottomed out in June 2013.
The growth in sales has supported price stabilisation in the recent period. Lower interest rates have also played a part (see Chapter 3). That said, negative equity problems will prevent house prices from rising rapidly. At the same time, credit-restricting measures such as the tightened Nibud mortgage lending standards and the return of the annuity mortgage as well as the lower maximum mortgage will put a cap on price growth.
Differences per segment
Since the start of the economic crisis in 2008, prices of detached houses have declined considerably more (-24.3%) than prices of houses in the less expensive segments, such as apartments (-18.4%) or mid-terrace houses (-16.7%). This makes sense from an economic point of view: expensive houses require greater income elasticity on the demand side, because they have the characteristics of luxury goods. It is these goods that people economise on during a crisis. Moreover, the more expensive price brackets probably benefited more from credit growth before the crisis and from mortgage deductibility. This is reflected in the fact that prices of upmarket homes rose faster during the boom than in the other segments (Figure 12). Now, with prices edging up again, the luxury bracket may be expected to accelerate. However, in our view, rapid price growth in this segment will be countered by the effects of reforms such as abolition of tax-favourable mortgage types such as interest-only mortgages. Moreover, the gradual reduction of mortgage interest deductibility for tax purposes  will affect the highest incomes first, and hence the more expensive end of the market. Yet we see that the price drop in all segments has come to an end in the past quarter. The price of apartments, for instance, rose by 0.5%; semi-detached houses went up by 1%; and detached houses by 0.4%. The percentage of upmarket houses in total sales does not seem to have lost much ground since the crisis (Figure 13). Up to 2008, detached houses accounted for an average of 11.7% of total transactions, compared to 11% currently.
In view of the price trend of the past three quarters, with an average rise of 0.2% per quarter, the trough has finally been reached, after five years of decline. The expected growth in sales and the relatively low mortgage interest rates may boost prices further. If prices remain stable to modestly rising throughout the year, the index for 2014 will increase by an average of 0 to 1% compared to last year. In any case we expect that for the first time since 2008, house prices will not drop, and may even register a net rise, year-on-year. For those with potential negative equity, that will be a welcome development.
For 2015, thanks to relatively good affordability, low mortgage rates and growth in real disposable household income, as well as likely increased sales, we expect that the price index will rise during the course of the year. Thus for next year, we envisage a price rise of 1 to 3%.
Pieter van Dalen
Chapter 3: Mortgage trends
Total outstanding mortgage debt declined in 2013 and will rise only little, if at all in 2014, despite an increase in new mortgages issued. Since mid-2011, mortgage interest rates have fallen sharply. The rates for fixed-interest periods of up to five years are historically low, and we do not expect to see any sharp rise this year. The positive developments described in the previous chapters and the favourable economic outlook do not mean that problems such as payment arrears and negative equity will be resolved quickly. Approximately one million houses are in negative equity, and this number will fall only gradually in the coming years as house prices rise again. The number of households in arrears on their mortgage payments has risen sharply and will continue to grow during the coming years.
3.1 Decline in total mortgage debt
Total outstanding mortgage debt declined in 2013 for the first time in decades. In late 2012 total mortgage debt of Dutch households amounted to € 653 billion; a year later this had fallen to € 637 bn (Figure 13), a drop of over 15 bn (2.4%). The decline was caused by a reduction in the number of new mortgages issued and increased repayments by mortgage holders. The latter were prompted by low savings deposit rates and new (fiscal) incentives, such as means-testing for elderly care and the temporary extension of the limit for tax-free gifts (see our previous Dutch Housing Market Quarterly for a comprehensive report). As yet, no data on total mortgage debt are available for the first quarter of 2014; however, data are available on total mortgages issued by banks. On the basis of these figures, it appears that mortgage volume stabilised in the first quarter.
With house sales picking up, more mortgages are being issued. In the first quarter of 2014, the value of new mortgages amounted to € 6.8 billion. This is higher than the amount issued in the first quarters of both 2013 (€ 5.9 bn) and 2012 (€ 6.7 bn; Figure 15). The amount of refinancing mortgages issued is very low at only € 1 bn, which is related to the ongoing low mortgage interest rates.
As outlined in Chapter 2, we expect a further rise in house sales in 2014, with house prices stabilising. The number of new mortgages issued will increase apace. At the same time, we expect the trend of additional mortgage repayments to continue in 2014. On balance therefore, total outstanding mortgage debt will rise little if at all this year.
3.2 Increasing number of homes in negative equity
As a result of the slump in house prices, an increasing number of homes are saddled with so-called underwater mortgages. This means the mortgage debt is greater than the value of the home. The precise number of such properties depends on the definition of mortgage debt and home value. According to Statistics Netherlands (CBS), there were over 1.4 homes in negative equity on 1 January 2013, which is a third of all privately owned residential properties. (Figure 16).
According to Statistics Netherlands, the negative equity - the difference between the mortgage value and the official WOZ value of the property - averaged € 61,000 on 1 January 2013 (CBS, 2014). These calculations do not take account of equity held by home-owners in a savings or investment mortgage.
The Netherlands Central Bank (DNB) however, does factor in accumulated savings, and concludes that in late September 2013 nearly 30% of mortgaged homes were carrying underwater mortgages (DNB, 2014). Of the 4.3 million privately owned homes in the Netherlands, a total of 3.6 million are financed with a mortgage. This means that approximately 1 million residential properties are in negative equity, which is a quarter of all privately owned homes. The DNB data match our own estimates (see Box 1: homes in negative equity).
In fact, there were homes in negative equity before the crisis too. Often, particularly among first-time buyers, the mortgage was higher than the value of the property at the time of purchase. Buyers were allowed to include costs, such as fees and taxes as well as renovation costs in the mortgage. Thus, according to the definition of Statistics Netherlands, on 1 January 2006 - before the crisis - some 14% of privately owned homes were in negative equity. When prices are rising, this negative equity automatically disappears. Prior to the crisis, house prices were rising by over 5% annually (annual average 2001-2006). Accordingly, within two years, a property with a loan-to-value (LTV) of 110% was already 'above water'. By contrast, with declining prices, a mortgage can slip further under water if little or no repayment is being made on the loan.
Even when prices are rising, it can take years before current home-owners in negative equity will be 'above water' again. If house prices rise at a pace of 2% annually from 2014, it will take ten years before they are back at the level of 2008. Home-owners who make repayments on their mortgage will be out of negative equity quicker than those who currently have an interest-only mortgage.
Mortgage lending standards have since been tightened in order to reduce the risk of negative equity. The maximum LTV of new mortgages is being reduced by 1 percentage point annually, to a maximum of 100% of the property value by 2018. Accordingly, mortgages will no longer be underwater at the start. Moreover, since 2013, first-time buyers will only be eligible for tax deductibility on their mortgages if they take out a linear or annuity repayment mortgage. In the case of a linear mortgage, 3.33% of the loan is repaid annually. With annuity mortgages, the amount repaid in the initial years is between 1.5% and 2%, depending on the interest rate; in later years, the amount repaid rises, and the amount over which interest is paid monthly declines. A property will therefore only incur negative equity if the price decline is greater than the percentage repaid.
Box 1: Homes in negative equity
Estimates are provided by various organisations of the number of households in negative equity, i.e. whose mortgage is higher than the estimated value of their home. Based on our own analysis of the micro-database WoON2012 of CBS/BZK, we conclude that in the first quarter of 2014, approximately one million home-owners (28%) would face a residual debt to the tune of about € 50,000 if the property were sold. In arriving at this estimate, we took account of additional repayments made in 2012 and 2013, savings built up in savings and investment mortgages and the development of the house price index (CBS). The majority (65%) of households in negative equity, would have a residual debt of at least € 20,000 upon sale of the property.
Young home owners and houses in the lowest price brackets are over-represented among the mortgage-holders in negative equity. For example, 33% of negative equity is found among home-owners under the age of 35, while this group accounts for only 20% of total mortgage debt (Figure 16).
If we assume no extra repayments, but an annual house price rise of 2% through 2025, then the number of households in negative equity drops from 28% of current mortgage-holders in the first quarter of 2014 to 4% by 2025. Figure 17 shows several possible scenarios based on annually rising prices. In each scenario, the percentage of households in negative equity will remain relatively high in the coming years.
Negative equity in itself does not lead to higher monthly costs or to payment difficulties. However, if a mortgage is in arrears, then negative equity is an additional concern. Where there is surplus equity in a home, the owners have a capital buffer. If they get into payment difficulties, the house can be sold and they can begin again with a clean slate. However, those in negative equity are left with residual debt if the house is sold, which has to be either re-financed or paid back from other sources - if any - such as savings.
3.3 Increase in mortgage arrears
In recent years there has been a large rise in the number of mortgages that have fallen into arrears. The Credit Registration Bureau (BKR) registers arrears on mortgage payments longer than four consecutive months (120 days). According to these records, on 1 April of this year, 100,581 individuals were in arrears of over 120 days on their mortgage. As there are often two names on a mortgage deed, the actual number of households is lower than this. We are talking about no more than 2% of the 3.6 million households with a mortgage. Data on securitised mortgage portfolios also show that the number of mortgage arrears in the Netherlands is very low, compared to other countries.
It should be pointed out that more recent repayment problems may not be visible in the BKR data.
Box 2: Background to BKR-registration
The BKR reports on loans issued by credit lending institutions, mail-order credit and re-structuring credit. For the time being, mortgages are only registered with the BKR if they fall into arrears (negative registration). In the case of other loans, these are registered with the BKR by the issuer at the time they are extended (positive registration). In January 2014, nearly 740,000 consumers were in payment arrears on loans - that is 8.6% of the 8.6 million consumers with a BKR-registered loan. In this case too, we do not have a complete picture of private payment arrears. This is because the BKR has no data on rental arrears, tax debts or unpaid energy bills.
It should be noted that in the case of non-mortgage credit, there can be more than two consumers with unpaid debts within a household (for mortgages, it is usually one or two individuals per household). It is therefore difficult to estimate the number of households involved.
These data chiefly give a picture of long-term mortgage delinquency. Recent arrears, or frequent payment problems lasting less than 120 days are not reported to the BKR. However, if a payment settlement has been negotiated with the mortgage lender within four months, this is registered with the BKR.
Conversely, BKR lists a number of acute problems that have already been resolved. Home-owners who are in the process of paying off arrears from the past (for example during a temporary moratorium) remain in the BKR books until the full amount of the arrears has been paid off. It is therefore difficult to determine, based on these figures, the precise number of home-owners who are behind in their mortgage payments. That said, the data give an idea of the trend: the number of BKR registrations has more than doubled since 2010, and is still rising. Compared to the pre-crisis period, the number has as much as trebled.
Payment difficulties on other debts have also increased. Besides arrears on mortgages, the BKR also registers arrears on other loans, including mail-order credit and re-structuring credit. In January 2014, nearly 740,000 consumers were in payment arrears on loans - an increase of some 25% since early 2011. It is not unthinkable that this much larger group also contains home-owners who are (still) paying back their mortgage, although failing to repay other non-mortgage debt.
If attempts are unsuccessful to negotiate a payment settlement or to sell the house in the usual way via an estate agent, then as a last resort, the bank can repossess the home (forced sale). However, these so-called fire-sales generally fetch lower prices than a regular sale. Both banks and home-owners therefore try to avoid this eventuality. Data show that despite rising payment arrears, the number of forced sales since 2012 has fallen from over 2,500 to approximately 2,000 annually (Figure 19). This drop should not be seen as a lessening of payment problems, but as a sign that banks supervise home-owners who have fallen into arrears more closely.
Mortgage delinquencies are generally the result of a large drop in income, for example following job loss or divorce. When people lose their job the effects are felt with a lag, because they are initially entitled to unemployment benefits. Payment problems tend to arise at the end of this period if no new job or other source of income has been found. Thus the number of arrears may rise further in the coming years, even if unemployment should decline.
If the payment problems are prolonged, the house may have to be sold. This process also takes some time. This means that altogether, there is quite a long lapse between job loss, the accumulation of payment arrears and the ultimate sale of a house. To illustrate, we compare the development of unemployment with losses incurred on mortgages as reported by the Dutch Mortgage Guarantee Fund (NHG) (see Figure 20). In the 1980s, the peak mortgage losses occurred two years after the peak in unemployment.
The picture emerging is that the economic recovery is underway, but that debt payment problems will continue to increase for a number of years. The recovery will take place in stages. Since mid-2013, both economic growth and the housing market have been picking up, while on the other hand job vacancies are still declining. From 2015 employment will also pick up. Payment problems and defaults on mortgages - and on other debts - will likely continue to increase in the coming years, in line with the delayed effects outlined above.
3.4 Mortgage interest rates decline further
Mortgage interest rates have declined since mid-2011, particularly the rates for fixed-interest periods of less than ten years (Figure 21). In March 2014 the average rate for newly issued mortgages was 3.3% for a fixed interest period of one to five years, and 3.9% for a fixed-interest period of five to ten years (DNB). A year earlier, the rates were 3.8% and 4.6% respectively. The rates for fixed interest mortgages of up to five years are now even lower than the previous record low of 2005.
Capital market rates are currently low and we expect these to rise only modestly in 2014. In April the 10-year swap rate stood at around 1.75%. We expect this to rise to 2.0% over six months, and to 2.3% over twelve months.
Furthermore, recent issues by banks and insurers of residential mortgage-backed securities (RMBS) and covered bonds show that spreads have narrowed in recent months. These spreads can be seen as an indicator of the risk premium that banks have to pay if they raise money on the capital market to issue mortgages. A narrower spread means that financing conditions have improved for the financial institutions. And because we also expect the rise in the swap rate to be modest, we do not consider a sharp rise in mortgage rates likely. Accordingly, we do not expect financing conditions to form an impediment to recovery on the housing market.
Chapter 4: Recovery in construction not visible until 2015
2013 was a tough year for the entire home-building sector. Only 26 thousand building permits were issued, which is a historic low. However, this year we will see an increase in that number. That said, these homes still have to be built, so the recovery in construction will show up with a lag. After a minor decrease in 2014, we expect to see 8% growth in 2015 in the volume of new housing output.
4.1 Home-building market in a nutshell
Statistics Netherlands (CBS) distinguishes three types of contractors and two property types (owner-occupied and rental) in the residential sector. Roughly speaking, these are 1) private individuals who have a house built for their own use; 2) builders (building contractors and project developers) who mainly build homes for the owner-occupied sector; and 3) housing corporations who mainly commission the construction of homes in the rental sector (table 2).
In the coming years, the construction of new rental housing will remain under pressure because of the landlord’s levy. Housing corporations and other lessors of social rental housing will have to pay a government levy of approximately € 775 per regulated rental unit. The total amount will reach € 1.7 billion in 2017 (Janssen, 2014). According to Central Public Housing Fund (CFV, 2014), the housing corporations will only remain financially sound if they reduce construction output. In the long term, however, rents in the regulated segment will rise further and ultimately it is the tenants who will pay the landlord's levy (Janssen, 2014). The effect of the levy is clearly visible in the low number of applications for the construction of rental properties.
In the owner-occupied sector, applications for building permits have fallen even more sharply since 2007. This sector is demand-driven. During the recent slump, demand plummeted and building permits contracted by 70% compared to the peak in 2007 (table 1). Thus the drop in demand in this segment has been more severe than on the second-hand housing market.
A number of stages can be defined on the market for new homes in the owner-occupied sector. The vast majority of these homes are developed in projects. Once 70-80% of a development has been pre-sold, the builder or developer applies for a building permit. The time between this application and the start of construction is at most three months. It then takes an average of 15 months for the project to be completed. Statistics are available for these three stages. The pre-sale of new housing is registered by the New Housing Monitor (MNW), while the building permits and the new completions are registered by Statistics Netherlands (CBS). The pre-sales data indicate at an early stage whether construction output is going to contract or grow.
4.2 New housing market sensitive to shocks
Economic conditions have considerable influence on pre-sales of new housing. This is because the Dutch housing market is dynamic , which means that the new homes sector, which consists mainly of homes in the highest price bracket, is extra sensitive to a crisis (Van der Heijden, Dol en Oxley, 2011). In the Netherlands it is mainly existing home-owners who buy newly built houses. During the crisis these market parties became unsure whether they would be able to sell their existing home quickly enough and at what price. Moreover, they are often faced with double costs: they have to pay the mortgage on their existing home as well as the interest payable on the construction of their new home. Usually these extra costs are paid via a special account. In addition, more and more home-owners are facing negative equity problems. This too, has affected the demand for new housing.
As a result of these developments, the number of building licenses issued declined rapidly during the crisis, reaching a historic low in late 2013. With only 26,184 permits issued, this was 30% less than in 2012, and down 70% on 2007 (Figure 22). Despite this fall-off, the number of new completions rose in 2013 to 51,000 (Figure 23). Theoretically, this could be the result of accelerated development of shelved projects that were already licensed. However, that seems illogical, because there was no increase in demand in 2012 and early 2013. Instead, we assume the number is the result of changes made in the method of registration by Statistics Netherlands (see also Priemus en Van der Heijden, 2014).
During the recession, developers had to re-designate their construction projects in order to adapt to the changing market. While they had previously concentrated on the most expensive market segment - often in large-scale developments - they now had to shift their focus to the cheaper price bracket. Moreover, they reduced the average project size considerably, thus making it easier to reach the sales threshold of 70-80%. Figure 24 gives a summary of this development, showing the number of newly built houses by price bracket.
4.3 End to declining sales
We appear to have reached the lowest point of housing construction. Pre-sales are rising and are likely to have bottomed out at this point (MNW). In the first half of 2013 the number of houses sold was historically low, but the market was negatively influenced at the time by the fiscal reforms to the mortgage market implemented in January 2013. At the end of the previous year a relatively large number of households had purchased a (newly constructed) house in order to benefit from the existing tax-favourable mortgage type before the reforms took effect. As a reaction to this, home sales fell sharply in the first half of 2013. Sales picked up in the second half of the year, even registering the best six months since 2010 (Figure 25). In the first quarter of 2014, developers sold 4,749 homes off the plans - a rise of 35% compared to the same quarter in 2013 (MNW). In line with the recovery in the existing homes market (Chapter 2), we expect sales of newly built homes to gradually rise further in 2014 and 2015.
Not until 2015 do we expect a recovery in construction output. This is because building only commences after 70-80% of a project has been pre-sold. The average selling time (the number of days between sale of the first unit in the project and the last) of individual houses (versus apartments) has improved over the years (Figure 26). This is partly the result of improved market conditions but also because projects have become smaller. This makes it easier to achieve the minimum sales threshold of 70-80%, thus reducing the average selling time. However, the average selling time for multi-home complexes (e.g. apartments) remains high. This is because construction companies can only sell the last 20-30% when construction is nearly completed. We see it as a positive sign for the construction market that the selling time is getting shorter after years of going in the opposite direction.
Because pre-sales are rising and the average selling time is declining, construction output will start to pick up later this year. However, because it takes an average of fifteen months for a project to be completed, we don't expect to see an increase in the number of actual completions until next year.
Conclusions and expectations
In view of the very low number of applications for building permits in 2013, we expect that construction output will reach a trough in 2014. According to estimates by the Economic Institute for Construction (EIB, 2014), which we endorse, this trough will come to just over 45,000 new completions (both owner-occupier and rental units). At the same time pre-sales of new homes will pick up this year.
For 2015 we expect to see a recovery in home-building. Indeed there may be a catch-up spurt, to reduce the housing shortage, as output has been consistently low throughout the crisis. In recent years, output has fallen short of demand by 20,000 homes annually (see also our regionale woningmarktvisie). In our Rabobank Construction update we provide more detailed information about our expectations for market volume (sales) across the entire construction sector. With regard to residential construction, we are assuming a contraction of 2% in 2014 followed by 8% growth in 2015. For the construction sector as a whole, we expect output to stabilise (0%) in 2014 and to rise by 2.5% next year.
Chapter 5: Five years of slump on the housing market; a regional perspective
Over five years after the start of the crisis, there is a glimmer of light on the housing market. We expect both sales and house prices to pick up throughout the country. It is therefore timely to take a look back at the impact the crisis has had on regional variation in house prices in the Netherlands.
5.1 Price decline and price rise
The house price index that Statistics Netherlands publishes at national level is not available in a regional breakdown. Therefore, an analysis of regional differences in the development of house prices has to be conducted on the basis of the median house price as recorded by the NVM (Netherlands Association of Real Estate Brokers). The national price development that was discussed in Chapter 2 reflects a different set of statistics than the regional analysis in this chapter. The national price index follows the average price development of the total housing stock. The regional median house price is based on random house sales and is therefore sensitive to the composition of the set of houses sold. For instance, if more expensive, detached houses are sold in a given period, the median price will rise, even though the price index itself could drop.
Following a period of surging house prices from early 2005, the median house price in the Netherlands (NVM) reached its highest level ever in the second quarter of 2008: €250,800. Since then, prices have declined by over 18% (€43,800), up to and including the fourth quarter of 2013. However, this decline was by no means uninterrupted (Figure 27). Prices fell sharply from the second quarter of 2008 through the first quarter of 2009, and likewise, though at a slower pace, from the second quarter of 2011 through the third quarter of 2012. During these two periods of decline, prices fell by twelve and ten percent respectively. Expensive, detached houses were particularly hard to shift. In the intervening quarters, however, we saw a period of rising prices - reaching almost six percent. The more expensive bracket may well have picked up somewhat during this period.
Although we now talk of five years of crisis on the housing market, there was clearly no long-term uninterrupted decline in the median price of houses sold. Moreover, it appears prices have bottomed out since the end of 2012. Since then, the NVM has registered a price drop of only 2%, with the median price rising again in the last quarter of 2013.
House prices determined by income and living environment
House prices are largely determined by income levels and to a (much) lesser extent by the quality of the natural environment and tightness on the housing market (Dutch Housing Market Quarterly, February 2013). This was the case prior to the crisis, and continues to be so. The country's highest house prices are found in the ‘rump’ of the Netherlands, where affluence is high, and particularly in the leafy residential areas in the northern wing and the Randstad urban conglomeration (Figure 28). Prices are lowest, on average, in the north of the country, Zeeuws-Vlaanderen, Limburg and in the Rijnmond (Rhine delta area).
Regional income level – regional affluence – is not only instrumental in determining the house prices in a region, but together with income development and the presence of historic features in the landscape and built environment (e.g. wooden ramparts or listed buildings) it also contributed to regional variation in the rising prices before the crisis. Figure 29 shows the effect of a historic feature on price development in the pre-crisis period. In regions containing visible historical landmarks, house prices rose more acutely than in other regions. A number of the most urbanised regions in the northern wing of the Randstad urban conglomeration – the cities of Amsterdam, Utrecht and Haarlem – topped the list for affluence and at the same time contain many historic landmarks. These regions also exhibited the strongest house price growth (Figure 30).
5.2 Price drop also a correction
While affluence and the development of affluence have no direct impact on regional variation of the price decline in recent years, they do play a background role. Our analysis shows that the price fall was steepest in urban regions that had experienced a strong price rise before the crisis, or where the housing stock was expanded considerably after the peak (Figure 31). On the other hand, the price decline was mitigated by a large and strongly increasing demand for housing – rather than the actual urban character of a region itself. The price decline was also tempered by the historic character of a region as well as tightness on its local housing market. Accordingly, the price decline was less marked than average in regions where pressure on the housing stock is greatest.
Altogether, the price decline in the rump of the country was sharper than in the peripheral regions, although both regions are represented in both the regions with the largest price drop – such as South-Eastern-Brabant and South-Western-Friesland – and in regions with the smallest price drop such as Haaglanden (conurbation surrounding The Hague) and Zeeuws-Vlaanderen (Figure 32).
The correlation between the rising prices up to 2008 and subsequent declining prices indicates that at least part of the price drop reflects a correction of the sharp rise in the preceding years. In the Randstad urban conglomeration, this distinction is embodied by the Haaglanden region at one end of the spectrum, and Greater Amsterdam at the other end. Between 2005 and 2008, house prices in Haaglanden rose much less sharply than in Greater Amsterdam; but since 2008, the price decline in and around The Hague has been much less marked than in the Amsterdam region. Composition effects can also play a part. The price rise up to 2008 may be (partly) attributable to a relatively large share of expensive houses in total sales; the subsequent price drop, may have (partly) been caused by a smaller percentage of these houses in total sales. This composition effect could mean that sales of more expensive houses around The Hague have fallen less since 2008 than in Greater Amsterdam. This scenario reflects the diverging picture of employment in both regions. Employment in Amsterdam, which is dominated by commercial services, suffered much more in the crisis than did jobs in The Hague, where the government is a major employer.
Apart from this composition effect, corrections took place between the four episodes that can be distinguished within the entire period of declining prices since 2008. Price development in one episode has the reverse effect in the next episode. The pendulum subsequently swings back in the third episode, only to be reversed again in the fourth. For all episodes, the price development in the previous episode was the strongest (negative) factor behind the price development in the next episode (Figure 33). The initial price fall in the autumn of 2008 was strongest in the regions where the price rise in the preceding years had been greatest. The price rise in 2009/11 was strongest where the drop in the autumn of 2008 had been greatest; and the price drop in 2011/12 was steepest where the rebound in 2009/11 had been strongest. This nationwide pattern appears to indicate that buyers - in recent years too - seize the opportunity to enter the market when house prices are low, but become more cautious when prices start to rise or the decline slows down.
Levelling effect of price decline
Despite the regional differences in price level and development we have identified, the average deviation of the regional house price from the national house price has scarcely changed. While the strong price rise of the 2005-2008 period was accompanied by an increase in the average deviation, in the subsequent years of declining prices, the average deviation of the regional house price vis á vis the national house price lessened again. In 2005, this deviation amounted to an average of 15.5%; at the peak of the housing market it was 17.3%, and during the last quarter of 2013 it was 16.4%. Accordingly, the price rise had a de-levelling effect on the regional variation in house prices, and the same is true for the recovery in house prices in 2011-12. The price drop since 2008 had the opposite (levelling) effect.
The levelling of regional house prices is expressed in the decline in the difference between the median house price in the North of the Netherlands, Zeeuws-Vlaanderen and Limburg compared to the national average (Figure 34). During the boom years, prices in these regions were lagging behind the national average. The growing divergence with the national average in Haaglanden and in a number of regions in the Northern wing of the Randstad urban conglomeration - where prices also rose relatively sharply during the boom - indicates however, that the price decline of the past five years can have not only a levelling effect, but also a de-levelling effect. The main factors at play in these regions were the extent of the demand for housing, tightness on the housing market and the historic character of the living environment.
Although there has been talk of a crisis on the housing market for over five years now, and the median house price registered by the NVM has declined steeply since 2008, there is no question of an uninterrupted decline, and nor has the drop been equally sharp across all regions. At regional level, the price decline since the top of the market has moreover led to a relatively strengthening of the position of a large number of low-price regions that had previously lagged behind much more. In these regions, the relative gap vis á vis the national average decreased. These peripheral regions have emerged from the crisis with the least losses. Besides this, a number of regions in the North-Eastern wing of the Randstad urban belt saw an increase in their deviation from the national average. These sought-after high-quality residential environments perform better both in times of upswing and downturn.
 To be reduced annually by 0.5%-point from 52% to 38% by 2042. It currently stands at 51.5%.
 Although the tightening of mortgage lending standards is not expected to impede sales, it will dampen any rapid acceleration. See Section 2.
BKR (2013). Hypotheekbarometer april 2014.
BKR (2014). Kredietbarometer januari 2014.
Calcasa (2013). The WOX Quarterly Q4 2013.
Centraal Fonds voor de Volkshuisvesting (2014). Financiële Meerjarenverkenning Corporatiesector 2013-2022.
DNB (2014). Overzicht Financiële Stabiliteit, voorjaar 2014.
Dijkhuizen, van M. (2013). Verliesaversie op de woningmarkt,CPB.
Erken, H., Doll, M. en M. Raterink (2014). De oploop van de werkloosheid verklaard, Economisch Statistische Berichten, 99 (4677), 42-45.
Francke, M. en A. van den Minne (2013). Starters slaan hun slag! Real Estate Research Quarterly, 4: 33-40.
Gualthérie van Weezel, T. en S. Kooistra (2013). Woning huren vaak duurder dan kopen.Volkskrant 19-10-2013.
Heijden, van der H., Dol, K. en M. Oxley (2011). Western European housing systems and the impact of the international financial crisis, Journal of Housing and the Built Environment, 26(3): 295-313.
Janssen, B. (2014). De verhuurderheffing: een onderzoek naar de impact op beleggers en verhuurders, Amsterdam School of Real Estate.
Oevering, F. (2014). De Nederlandse woningmarkt in regionaal perspectief. Rabobank, Utrecht.
Piljic, D. en H. Stegeman (2013). Anders wonen, naar een werkende woningmarkt. Rabobank, Utrecht.
Priemus, H. en H. van der Heijden (2014). Raadsels rondom actuele woningproductiecijfers nog niet opgelost.
Schilder, F. en J. Conijn (2013). Verhuizen in de crisis, Amsterdam: ASRE.
Rijksoverheid (2013). Uitkomsten Woononderzoek Nederland 2012.
The Dutch Housing Market Quarterly is a publication of the Economic Research Department (KEO) of Rabobank Nederland. The view presented in this publication has been based on data from sources we consider to be reliable. Among others, these include EcoWin, Land Registry, NVM, DNB, CPB and Statistics Netherlands.
This data has been carefully incorporated into our analyses. Rabobank Nederland 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.
The Economic Research Department is also on the internet: www.rabobank.com/economics
For more information, please call the KEO secretariat on tel. +31 (0)30 – 216 6666 or send an email to ‘firstname.lastname@example.org’.
Special thanks to Leontien de Waal, construction analyst Rabobank, for her contribution to Chapter 4.
Pieter van Dalen, Frits Oevering, Theo Smid, Leontine Treur and Paul de Vries
Tim Legierse, head of Head Domestic Macroeconomic Research, Economic Research Department
Pieter van Dalen, Selma Heijnekamp and Reinier Meijer
© 2014 - Coöperatieve Centrale Raiffeisen-Boerenleenbank B.A., the Netherlands