Further recovery in 2015
Dutch Housing Market Quarterly
In the short period between the finalization and publication of this Dutch Housing Market Quarterly, new Nibud standards have been published. We have written an additional Economic Comment about the effects of these new standards: Stricter Nibud standards will limit price recovery on the Dutch housing market in 2015
- The number of sales and house prices will continue to rise in 2015 thanks to the improved economic situation and good affordability of owner-occupied houses
- We do not expect mortgage interest rates to rise in the short term
- Ending or scaling back incentive measures have a negative effect on the recovery
- In 2014 around 145 thousand homes will be sold, and in 2015 between 145 and 165 thousand. Next year house prices will rise by an average of 1½-3½ % compared to 2014
- The recovery will vary considerably from region to region
Economic recovery important for the housing market
For 2014 we anticipate a rise in the Dutch GDP volume by an average of ¾% compared to 2013, largely thanks to exports and investments. In 2015 consumer expenditure among households will also make a positive contribution and we expect a growth in GDP of 1½%. As employment rises, unemployment will fall slightly next year to an average of 6½%. The rise in disposable household income and a recovery on the labour market are important for the housing market.
Further increase in house sales and prices
The recovery in the housing market continued through the third quarter of 2014. The number of sales continued to rise, albeit somewhat slower than in the preceding quarters. On the other hand, the house price index rose more sharply than in the preceding quarters. With an increase of 0.9% compared to the last quarter and 1.8% compared to last year, this was the strongest price rise since the crisis broke out in 2008. The sales contracts registered by the Dutch Association of Real Estate Brokers (NVM) (on average two months prior to the sale of the house) point to a further slight rise in the number of sales and house prices in the fourth quarter of this year. We expect the number of sales in 2014 to reach around 145,000, and the average house price in 2014 will be 1% higher than the average in 2013.
Mortgage interest rates fall to an all-time low
Nominal mortgage debt did not rise in 2014. The increase in new mortgage approvals was more than compensated by the extra repayments. We do expect, though, that thanks to the recovering housing market in 2015 the balance will shift and the nominal mortgage debt will rise slightly. Since mid-2011 mortgage rates have fallen to historically low percentages. We do not expect any significant rises this year or next year. Nonetheless, should this happen it will not result in a strong increase in affordability risks among households.
Further recovery in 2015
In 2015 we expect further growth in the owner-occupied housing market. Positive factors such as further economic recovery, high levels of consumer confidence, low mortgage rates and sales of a large number of new homes are expected to have a greater effect than negative factors such as ending or scaling back incentive measures, the negative equity problem and credit-restricting measures. We anticipate a rise in the number of sales in 2015 to 145,000-165,000. The average house price index is also expected to rise in 2015 by 1½ -3½% compared to 2014.
Chapter 1: Economic recovery taking further shape
Real Gross Domestic Product (GDP) rose strongly in the second quarter of this year by 0.7% compared to the first quarter, following a contraction of GDP volume in the first quarter. The dynamism is largely the result of the normalisation of gas production following the mild first quarter. For 2014 as a whole we anticipate a rise in real GDP by ¾% compared to last year (Table 1), mainly thanks to exports and investments. In 2015 household consumer spending will also make a positive contribution and we expect GDP to grow by 1½%. This year unemployment will remain high at 6¾%, but thanks to rising employment, unemployment will fall slightly next year to an average of 6½%.
Exports will pick up next year
Growth in export volumes last year was very modest at just 2.2% year-on-year. Following a strong first and second quarter, exports were much weaker in the third quarter of this year due to the global slowdown in growth. For 2015 we anticipate a growth in exports of 5¾%. Economic growth is picking up among major trading partners, and exports are also being given an extra boost through the fall in the value of the euro compared to the US Dollar and the British Pound. in our baseline scenario we do not take account of any potential escalation of the Russia/Ukraine conflict. However, this remains a significant downward risk for the European economy, and with it the growth of Dutch exports.
Consumers not (yet) persudaded
Consumer confidence declined in August and September, the likely cause being tensions in Ukraine and the Middle East. The deterioration of sentiment was observed above all in the sub-indicator that measures consumer sentiment for the economic climate in the coming twelve months. Consumer confidence did rise again in October, and despite the international tensions of past months the willingness of consumers to buy remained virtually stable. This part-indicator says the most about the inclination of households to spend. Confidence in the housing market rose further in August and September. The 'Eigen Huis' market indicator of the Homeowners' Association (Vereniging Eigen Huis) homeowners reached 100 in September, its highest level since the series began in 2004.
Real disposable household income will again rise slightly this year as the inflation rate slows and some aspects of the tax and premium burden are eased. Despite the rise in disposable income, this year we still expect a contraction in private consumption volumes because some households will still be focusing on paying off their debts or building up capital. Even so, there are signs pointing to a faster recovery in consumer spending than we currently expect. For example, with the exception of June consumption of durable goods has been rising annually since November last year (Figure 1). We expect the recovery on the housing market will allow for further growth in the consumption of durable goods for the rest of this year and next year. Real disposable income will rise further next year as a result of the recent recovery on the labour market, further real rises in wages and potentially lower pension premiums. We expect household consumption to grow by 1% next year compared to this year.
Employment in the private sector is picking up
We are also seeing clearer signs of recovery on the labour market. Unemployment fell further in September, to 6.5% of the labour force (Eurostat/ILO definition). The labour force actually in work increased by 35,000 persons in the third quarter, whereas in the second quarter there was still a slight fall. The rise in the number of vacancies and temporary hours worked via employment agencies during the past quarters would indicate a further recovery of employment in the private sector. This year total employment is still likely to contract compared to last year as the number of public sector jobs continues to fall as a consequence of cuts. Next year, however, the economic recovery will mean that total employment will grow each year.
The figures on unemployment trends also show that the labour market is slowly beginning to recover. For example, net flow between the group of unemployed workers and the labour force in work was negative during the first quarter (Figure 2). This means that the group of unemployed people who found work was larger than the group of those in work who became unemployed in the same period. This is a clear break with the pattern of previous years, when there was usually a net addition to the group of unemployed from the labour force in work.
Chapter 2: Market for existing owner-occupied houses
During the third quarter of 2014, 39,160 homes were sold. Compared to last year, this represents a seasonally adjusted rise of 5.6%, but a slower rate of growth than during the first quarters of this year. The rate of growth is therefore slowing slightly, but the recovery is still on track. There is a time lag in the response of the Existing Homes Price Index (Prijsindex Bestaande Koopwoningen - PBK) of the Land Registry/CBS to the rise in sales and so it showed an acceleration on a quarterly basis: +0.9% compared to the previous quarter (Figure 3) and +1.8% compared to the third quarter of 2013.
Positive factors for a further rise in the number of house purchase transactions and average house prices in 2015 are considerable confidence in the housing market, falling mortgage rates, sales of newly built houses picking up, the slight increase in real disposable household income and the recovery on the labour market. Against this backdrop there are negative effects: ending or scaling back incentive measures, the problem of negative equity and credit-restricting measures. On balance we expect the positive factors to make a greater contribution, which will enable the housing market to continue its recovery. As a result, we anticipate a further rise in the number of house sales and house prices in 2015.
Further rise in the number of transactions …
During the third quarter of 2014 some 39,160 existing owner-occupied houses changed hands. If we adjust this for seasonal effects, the quarterly increase of 5.6% is significant, with the four-quarter total continuing to rise further to 138,000 sales (Figure 4). Compared to the third quarter of 2013 this represents a sizeable increase of 39%. The rise compared to a year ago was visible in every province. The province of Utrecht experienced the strongest rise compared to the third quarter of 2013 (+52%), but the provinces with the least growth - Drenthe and Limburg - also showed rises of 19% and 24% respectively.
… more sales contracts …
The Dutch Association of Real Estate Brokers (NVM) saw the number of sales contracts mediated by its members remain at the same level in the third quarter (NVM, 2014). The number of NVM sales contracts reflects current market developments for some 75% of the market (on average two months earlier than the Land Registry). With 29,015 sales contracts, this was the best quarter since 2008. Seasonally adjusted it represents a rise of 2% compared to the previous quarter (Figure 5). This is a weaker rise in sales compared to the first two quarters of 2014, when the number of sales rose by more than 6%, and indicates a lower growth in the number of transactions recorded by the Land Registry in the fourth quarter.
… and slightly less for sale …
The rise in sales is also changing the composition of the unsold housing stock. A rise in the number of transactions leads (ceteris paribus) to a fall in the supply of existing homes for sale. At the same time the recovery on the housing market has prompted some homeowners to put their house up for sale. The average supply in the third quarter ultimately fell only slightly quarter-on-quarter, by a little more than a thousand homes (Figure 6). This indicates that the composition is changing: new supply is becoming available thanks to increasing confidence, while on the other hand more homes are being sold.
… are creating a tighter market
The slight decline in supply and a rise in sales is now causing the market to shift slowly but surely in favour of sellers. The NVM's shortage indicator, which divides the total number of houses for sale by the number of transactions, has remained the same at 16. Seasonally adjusted, we observe a 2.3% decline in the indicator from 16.6 to 16.2 (Figure 7). However, it is still a buyers' market. During the years preceding the crisis, the indicator hovered around a value of 6 to 7, which the NVM regards as a normal value. We do note, however, that the extent of the shortage heavily depends on the region and the type of home.
Two contrasting scenarios in selling times and supply
The average selling time fell in the third quarter this year to 122 days, a further improvement on the 133 days in the previous quarter and 160 days in the first quarter of 2014 (Figure 8). However, two contrasting scenarios can be observed as regards selling times. Houses recently put on the market, and generally more competitively priced, sold relatively quickly in this quarter, but the likelihood that houses that have been on the market for a longer period of time will be sold in the next quarter is still relatively small. We see the same trend in another measure; the length of time houses remain on the market. This continued the downward trend from 402 to 395 days. In the previous quarter this was the first fall since the crisis broke out. Figure 9 shows that a relatively large number of new homes have been added to the supply which are sold within one or two quarters, but at the same time the proportion of homes that have been on the market for longer than three years has risen to 18% of the total supply. The asking price for these houses is likely to be too high for a quick sale.
There are various reasons why households are sticking to such high asking prices. On the one hand they do not want to make a loss on the sale due to loss aversion (Van Dijkhuizen, 2013). These include households which are certainly able to finance any residual debt in a new mortgage and/or can use their savings to buy a new house. On the other hand, there are also households that do not have this option. This group therefore has to stick to a relatively high asking price, and therefore find it difficult to sell their house.
Differences between house types
Within this general rise in sales, we see significant differences according to house type. During the crisis, owner-occupied houses for first-time buyers became more affordable with falling house prices and low mortgage rates. Against this, however, some homeowners wishing to move up the housing ladder were confronted with negative equity (a mortgage debt greater than the value of the house) which prevented them from moving house. The result was that the share of first-time buyers in house sales rose, and with it the market share of apartments, the type of home that first-time buyers generally buy. Now that the housing market is picking up again, the first thing we notice is that the market share for mid-terrace houses is rising slightly (Table 2). This is in line with the study by Francke and Van der Minne (2013), who argue that first-time buyers can now buy homes which previously were out of their reach. Secondly, the market share of the generally more expensive house types is slowly but surely recovering compared to the average share following the crisis. Thanks to the recovery, home owners wishing to move are now more likely to be able to sell their homes and take the next step on the property ladder. We expect a gradual increase of market share of the more expensive segments during the coming quarters.
Further recovery in 2015
In 2014 improved affordability, falling mortgage rates, rising confidence, the release of delayed house-seeking activity and relatively high rent rises positively affected the number of sales of existing owner-occupied houses. Looking ahead to 2015, there are various factors which point to a further recovery on the housing market, but at the same time there are factors inhibiting this trend.
Factors pointing to more sales in 2015
First and foremost, the economic context is very important. In Chapter 1 we anticipated that the Dutch economy will grow this year and the next, but this growth is still fragile. For next year we do expect higher employment compared to 2014 and a further rise in real disposable income. This will make an important contribution to a further rise in the number of transactions in 2015.
Secondly, the affordability of owner-occupied homes will remain good in 2015. House prices have only risen slightly, while at the same time mortgage rates have fallen this year to an all-time low (see Chapter 3).
Thirdly, confidence in the housing market is relatively high. This confidence, measured by the Eigen Huis market indicator, has been rising continually for 21 months since January 2013 and reached a value of 100 in September 2014. This is its highest level since the introduction of the indicator in 2004, and points to a further rise in sales (Figure 10). The ongoing rise in confidence is remarkable, since consumer confidence in the general economic situation in fact fell as a result of geopolitical tensions in August and September (see Chapter 1). The problem of negative equity does in fact mean that we do not expect the number of sales to rise as fast as such confidence might suggest.
The fourth important factor affecting the rise in sales of existing properties concerns recent developments in construction. The number of new homes sold on the open market in the first half of 2014 (10,662) shows that the market for new homes is slowly recovering as well (Figure 11). Based on developments in the first half of the year, we anticipate a rise in sales of 7,500 homes to 22,000 this year. We also expect to see a further rise in sales of new homes in 2015.
The market for existing housing is benefiting strongly from this recovery. Each new-build house sold in fact translates into an average of 2.1 extra sales in the market for existing housing (Dam, 2010). This is because the completion of new homes initiates a chain of house moves. The increase of 7,500 new homes we have forecast will then lead in the longer term to around 15,000 extra transactions in the housing stock compared to 2014. Since the extra supply will only enter the market gradually, we expect a positive effect of roughly 10,000 transactions for existing homes in 2015.
Potential demand not driving sales up further in 2015
A considerable proportion of potential demand has been released in 2014. This demand built up during the crisis period (Ministry of the Interior and Kingdom Relations, 2013). Households were delaying buying a house as prices were falling, they were uncertain about their financial situation and adopted a wait-and-see attitude in response to the ongoing discussions on reforms in the housing market. A study by Boumeester (2013) showed that the potential demand for houses (both new-build and existing) did not fall during the crisis (the P bars in Figure 12). As the annual number of moves to owner-occupied homes has fallen substantially during recent years (the G bars in Figure 12) there has been a build-up of households with plans to move house.
We believe that most of the potential demand among first-time buyers has been released in 2014, partly supported by the temporary crisis measures taken by the Government, such as the Starter Loan and a relaxation of the gift tax exemption (see next paragraph). The release of potential demand among first-time buyers will be weaker next year compared to this year, which will have a negative effect on the volume of house sales. We are already seeing this in the market share by age; the number of buyers younger than 30 years is falling slightly.
However, the potential demand among those moving up the property ladder has not yet seen much activity because of the negative equity problem. With an average price rise of 1% this year (compared to last year) and 2% in 2015, we expect that around 19,000 households planning on moving house will move out of negative equity in 2014, and around 30,000 in 2015. This means that next year there could be a relatively higher number of households actively seeking to move compared to this year. In addition, the financing burden for these households in negative equity will be lower next year; from 2015 the mortgage interest payments will be deductible from the residual debt for fifteen years. In 2014 this was still only possible for ten years. This is an extra incentive for those wanting to move up the property ladder to become active in the housing market next year. In addition, last year households were making extra repayments on their mortgages (see Chapter 3). This means that more homeowners with plans to move could become active in 2015. Finally, homeowners are also less pessimistic about their own financial situation (see Chapter 1) and the situation on the housing market itself (Figure 10). This will ensure that next year a larger group of households than in 2014 will be able and have the courage to take the step to move house with a residual debt.
To conclude, on the one hand the positive effect of first-time buyers who, following a period of waiting, will enter the market will be reduced next year. On the other hand we expect that the positive effect of households with ambitions to move will in fact be greater in 2015 than in 2014. On balance we therefore expect for 2015 that there will be no significant growth contribution to the number of house sales from the potential demand.
Factors pointing to fewer sales in 2015
The main factors negatively affecting a further rise in sales in 2015 are the phasing out of temporary incentive measures. Firstly, the temporary relaxation of exemptions from gift tax up to € 100,000 will end on 1 January 2015. However, households will still be able to make gifts in 2015 of at least € 52,256, but only to a child up to 40 years, instead of to everyone. This will be more than enough, however, to repay the median residual debt of € 28,000 (De Vries, 2014). We believe that the phasing out of the measure will reduce the number of sales in 2015 by roughly two thousand transactions (for a full explanation underpinning this expectation, see Van Dalen, 2014, Dutch only). Secondly, many municipalities are likely to stop their incentive schemes for stimulating their local housing market through the Starter Loan, which will have a negative effect on transactions of about 3,500 (see box 1).
Finally, next year there will be a slight fall in the maximum mortgage sum in relation to the value of the house (loan-to-value, LTV), the cost threshold of the National Guarantee Scheme (NHG) and the maximum rate at which households may deduct their mortgage interest payments from their income (for a detailed explanation of these policy measures, read Van Dalen and De Vries, 2014). We do not expect these changes to have any negative effect on transactions in 2015, although they will put a brake on property price rises (see 2.2).
Box 1: The Starter Loan will affect sale volumes in 2015
Since 2002 municipalities have been stimulating the housing market by means of the Starter Loan. This is an initiative of the Dutch Municipal Housing Incentive Fund (Stimuleringsfonds Volkshuisvesting Nederlandse gemeenten - SVn) to bridge the difference between the purchase costs and the maximum amount that the first-time buyer can borrow under the standard terms of the NHG. A limited number of housing associations also offer starter loans when selling their properties. The conditions for the Starter Loan differ between municipalities or associations. The average loan is around € 28,000 and the average purchase price is € 157,400 (EIB, 2014).
Thanks to the Starter Loan, those wanting to get on the property ladder will be more likely to buy a house and/or buy a more expensive house than they would be able to do without this loan. This is because first-time buyers do not have to pay any interest on the loan or repay the loan capital during the first three years. As a result, the Starter Loan does not affect the maximum financing burden laid down by law in the temporary scheme for mortgage credit. It is expected that the income will grow, and that after three years the first-time buyer will be able to pay interest and repay the mortgage. The annual report of the SVn shows that after three years 12% of first-time buyers have repaid the loan, 58% can pay an annuity on market terms, and 31% have insufficient payment capacity to bear the full housing costs. The SVn includes in its calculations what percentage of applicants will be able to bear the housing costs and at what moment. Every three years the mortgage payments are adjusted in line with the income.
Financing of Starter Loan and government crisis measures
SVn provides the Starter Loan from revolving funds or from funds based on the redemption method. With revolving funds, filled by the participating municipalities, the interest and repayments are put back into the fund for providing new starter loans. With the redemption method, a one-off subsidy of 26% of the loan is provided by the municipality. Only 11 of the 300 participating municipalities have chosen to finance the Starter Loan through the redemption method.
In 2013, as a crisis measure to stimulate the housing market, Minister Blok for Housing and the Central Government Sector made € 50 million available for the Starter Loans. The government contribution is being used to finance 50% of the Starter Loans, the other half being financed by the municipality (SVn, 2013). 4,163 loans were provided in 2013 and the SVn expects to provide 8,000 loans this year with government funding. It is expected that the government contribution will have been fully used by the end of 2014.
Negative effect for house sales in 2015
It is unclear what choices the national and local governments will make for 2015. We estimate that Minister Blok will provide a new stimulus contribution. If this becomes reality, on the basis of telephone enquiries the SVn is assuming that 50% of all municipalities will withdraw the Starter Loan scheme, resulting in a reduction of 4,000 loans compared to 2014. But as 28% of first-time buyers would have bought a house anyway, even without a Starter Loan being available (EIB, 2014), the effect on the number of sales is more modest (2,880 sales). Even so, the effect will be greater than merely the decline in purchases by first-time buyers because subsequent transactions will then not take place (property chains). Taking everything into consideration, we expect a negative effect on the number of house sales of around 3,500 transactions.
Conclusions and transaction expectations
This year has seen a strong recovery in the number of transactions in the housing market. In view of the sales already achieved and the latest figures by the NVM, the total number of transactions is expected to reach around 145,000. This is a sharp improvement on the 110,094 sales in 2013 and makes the year 2014 the best year since 2008 as far as the number of house sales is concerned. The causes of this recovery are increased confidence, falling mortgage rates, improved affordability and the economic recovery.
The housing market is expected to continue its recovery in 2015. The most important factors for further recovery in 2015 are rising employment, high levels of confidence and a rise in real disposable household income. We expect that the effect on the number of sales in the supply and the rise in sales of new homes will at least neutralise the negative effect of the withdrawal or scaling back of temporary incentive measures, such as the starter loans and the higher exemption limits on gift tax. One uncertainty here is precisely how the further release of potential demand among those wishing to move up the property ladder and among first-time buyers will develop in 2015.
On balance we expect the number of transactions in 2015 to be between 145,000 and 165,000. With a further rise of 1% per quarter (seasonally adjusted) in 2015 around 157,000 transactions will take place (basic trend, Figure 13). This is our basic scenario. Note, however, that this is significantly less than the average growth of 5% per quarter in 2014. Part of this is catch-up growth and a rise due to the temporary incentive measures. In a more positive scenario in which the economy recovers more strongly than we currently predict, the estimated increase on a quarterly basis is an average of 3%, which would result in around 165,000 sales (rising trend, Figure 13). In a negative scenario we assume that the market will not recover further and the number of transactions will fall back to the average of 2014. In that case, 145,000 houses will change ownership (Figure 13).
Further price recovery in the third quarter …
House prices rose in the third quarter of 2014 compared to the second quarter (+0.9%). The comparison with the same quarter of last year also reveals a rise (+1.8%, Figure 14). The price index is now 3.1% higher than the low point in June 2013. The nominal price level is now the same as it was at the start of 2003. In real terms the price index is back at the level of 1999 (Figure 15); 27% lower than at its peak in 2008.
A major reason for the recovery in prices is that confidence in the owner-occupied housing market has completely returned (and is even at its highest point since 2004), partly as a result of the present policy stability. This has contributed to extra demand for owner-occupied houses, which has strengthened the negotiating position of sellers in 2014. Secondly, owner-occupied houses are readily affordable. Calcasa (2014) states that affordability improved further in the first half of this year. Buyers spent an average of 17.5% of their net monthly income on net housing costs. This is still less than the 18% in the third quarter of last year, when house prices had just bottomed out. Since the 1990s, owner-occupied houses have never been so affordable as now due to the fall in prices from 2008 to 2013 and the historically low mortgage rates, particularly for first-time buyers (Calcasa, 2014).
… is also visible in every segment
The rise in the number of purchases is putting upward pressure on house prices in all segments. However, in line with trends in the number of sales per house type, we are also seeing that house prices in the cheaper segment have risen faster in the past quarter than the more luxury segment (Table 3). We expect that the abolition of tax-advantageous forms of mortgages such as the savings-based and interest-only mortgage will also put a stronger brake on price rises in the more expensive segments.
Regional price differences still considerable
Within the Netherlands there are substantial differences between the regions. For example, transaction prices in areas with relatively more expensive houses and/or a shrinking population have not risen so strongly on average as in the other areas. Price changes in the third quarter of 2014 compared to the same quarter in 2013 vary significantly from +15.5% in the NVM area of Duiven Westervoort to -9% in Elst. The pattern here is that prices in urban areas are rising faster than in the more peripheral areas. In the four major cities of Amsterdam (+4%), Rotterdam (+6.5%), The Hague (+5.5%) and Utrecht (+5.2%) house prices as a whole were rising faster than the national average. In some peripheral regions such as southwest Friesland and northern Limburg house prices were still falling by 5.2% and 3.1% respectively compared to the third quarter of last year. At the same time, house prices have also risen slightly in various peripheral regions (Figure 16). This is in line with our view that in times of recovery it is the urban areas which pick up first, and then the rural areas will pick up gradually in their wake (Oevering, 2014).
House price expectations
House prices bottomed in June 2013 and are now cautiously rising again. The increase in the number of transactions gives sellers a better negotiating position, and falling mortgage rates raise the affordability of owner-occupied houses. In view of the developments so far, the house price index will increase by an average of 1% this year compared to 2013.
We do expect house prices to continue to recover during 2015, with rising employment and disposable household income being important economic factors contributing to this. In addition, the expected increase in sales and low mortgage rates will put upward pressure on prices. At the same time owner-occupied homes will remain readily affordable in 2015.
Even so, we do not expect to see a rapid rise in the price index. The maximum mortgage compared to the value of the house (loan-to-value, LTV) will fall by 1 percentage point, compared to 2014. The maximum rate at which the homeowner may obtain mortgage interest relief will also fall by 0.5 percentage point to 51% in 2015. These policy measures are expected to exert slight downward pressure on prices. In addition, the NHG threshold will be reduced further in 2015 from € 265,000 to € 245,000. This too is expected to have a very limited negative effect (Van Dalen, 2014). The already tightened NIBUD standard, the ongoing negative equity problem and the withdrawal of tax-advantageous forms of mortgages for first-time buyers are not expected to have much of a negative effect on price trends, but will put a brake on the rate of increase. We therefore do not expect the kinds of price rises well in excess of wage rises that we had before the crisis. However, house prices will rise slightly in 2015 in line with the upward pressure on sales. We therefore anticipate a further rise in average house prices in 2015 of between 1½ and 3½% compared to 2014.
Chapter 3: Mortgage trends
Mortgage debt has not risen in 2014, despite an increase in the number of new mortgages issued. Thanks to a rise in house prices, fewer houses are now in negative equity. At the same time, the rise in sales means that homeowners are more likely to have to finance residual debt. Mortgage rates are at a historic low and we expect them to remain low for the time being. A low interest rate is favourable not only for first-time buyers and those wanting to move up the housing ladder, but also for homeowners whose fixed-rate period is soon to expire.
3.1 Rise in newly issued mortgages
With house sales picking up, the number of mortgages issued is also rising again. In the third quarter of 2014 new mortgages to the value of € 9.3 billion were issued. This is more than in the third quarter in the years 2011-2013, but lower than the third quarters of 2009 and 2010 (Figure 17). This is remarkable, in the light of the rise in the number of house sales. As already discussed in the previous edition of the Dutch Housing Market Quarterly, the value of new mortgages lags behind the number of transactions because the average purchase price has fallen, and moreover buyers pay a greater percentage of the purchase costs from their own resources or with family funds.
Despite a total value of new home mortgages of € 24 billion since the start of 2014, total mortgage debt has not increased. At the end of 2013 the total mortgage debt among Dutch households was € 632 billion; at the end of the second quarter of 2014 this was virtually unchanged at € 631 billion (Figure 18). Incidentally, no figures by the DNB on total mortgage debt are available yet for the third quarter of 2014; there is however data on the total value of mortgages issued by the banks. Based on this information, it would appear that the mortgage debt has also stabilised in the third quarter.
The mortgage debt referred to is a gross amount, not taking account of any mortgage-linked savings or insurance products. At the start of 2012 De Nederlandsche Bank (DNB) estimated this accrued capital at 30–45 billion euros, or 5-7% of the outstanding mortgage debt (DNB, 2012). In the meantime this capital will have grown further, meaning that the net mortgage debt has fallen even more than the gross mortgage debt since 2012.
The fact that total mortgage debt has not increased despite a rise in the number of mortgages issued can be explained by the fact that homeowners are continuing to make extra repayments on their mortgages. These extra repayments have been motivated, amongst other things, by the low interest rates on savings and a rise in gifts. Negative equity can also be a reason to make extra repayments, in order to reduce the residual debt when moving house in the future. The role of gifts on the housing market has been discussed in our economic report (in Dutch) Afloop verruiming schenkingsvrijstelling heeft beperkt negatief effect in 2015. Although the level of extra repayments may fall in 2015, we still expect households to continue to make extra repayments on their mortgages during the year. As long as interest rates on savings are low, households will choose to repay their mortgage or build up capital in a savings-based mortgage rather than adding to their regular savings.
As discussed in Chapter 2, for 2015 we expect a further rise in the number of house sales and a slight rise in house prices. As a result, the total value of new mortgages issued will increase further. On balance, the total outstanding mortgage debt is expected to increase slightly in 2015.
3.2 Negative equity declining, residual debt increasing
The rise in house prices is good news for households in negative equity. At the start of this year around 1 million households, or 28% of all homeowners, had a net mortgage debt that was higher than the value of their homes. Median negative equity stands at around € 28,000. The average amount is in fact € 51,000 but that is caused by a few outliers with a very high negative equity of € 100,000 or more. Repayments and capital accrual in mortgage-related assets lead to only a very gradual fall in negative equity. If prices remain unchanged it will take until 2025 before the number of homeowners in negative equity has been halved. If prices continue to rise, this process will be much faster.
With an expected price rise of 1% in 2014 and 2½% in 2015, by the end of next year more than 200,000 homeowners currently in negative equity will find themselves 'above water' again. If house prices continue to rise by 2% a year, the number of homeowners in negative equity will be halved by 2019.
A detailed description of households in negative equity and the effect of price rises can be found in our Special Report (in Dutch) Herstelvermogen van huishoudens die onder water staan.
If a house in negative equity is sold, the 'paper' negative equity changes into an actual residual debt. The rise in the number of house sales means that more households are facing residual debt. We are therefore seeing a greater demand for residual debt finance, as the visitor statistics for the website restschuldinfo.nl show. Interest on residual debt qualifies for income tax relief, as is interest on home acquisition debt. On Budget Day (‘Prinsjesdag’), Parliament announced that it will extend the period over which interest on residual debt is deductible from ten years to fifteen years.
3.3 Mortgage interest rates at a historic low
Mortgage rates have been falling since mid-2011 and reached an all-time low in the third quarter of 2014. In particular, mortgages with fixed-rate periods longer than ten years fell sharply in the previous quarter (Figure 20). In September 2014 the average rate for new mortgages was 3.1% with a fixed-rate period of one to five years, 3.6% for a fixed-rate period of six to ten years, and 4.0% for a fixed-rate period longer than ten years (DNB). A year earlier these were still 3.6%, 4.4% and 4.8% respectively. Mortgage rates for all fixed-rate periods are now lower than the previous historic low of 2005.
Capital market interest rates have dropped sharply in recent months. The 10-year euro swap rate stood at around 1.1% in October 2014, an all-time low. We expect the long swap rates to possibly fall just a little further, with the 10-year swap perhaps falling to 1.0% over a period of 6 months. We do not expect any sharp rise in the long-term interest rates in 2015. Our twelve-month expectation for the 10-year euro swap rate is 1.4%.
Moreover, residential mortgage-backed securities (RMBS) and covered bonds recently issued by banks and insurance companies show that the 'spreads' have narrowed in recent months. These spreads can be seen as an indicator for the risk premium that banks have to pay if they want to raise capital market financing for mortgage lending. A fall means that the financing conditions for financial institutions have improved. The Dutch banks successfully passed the ECB's 'balance sheet test’, comprising an asset quality review and a stress test. A further explanation of this balance sheet test can be found in our recent report (in Dutch) Asset Quality Review: is de ECB geslaagd? The result is positive for the financing conditions of the Dutch banks. A negative surprise could have caused a rise in risk premiums.
We do not expect mortgage interest rates to rise during the coming months. Later in 2015 the mortgage rates may possibly rise a little if capital market interest rates rise again slightly, but we do not believe a sharp rise to be likely. Accordingly, the current favourable financing conditions will continue to contribute to the recovery on the housing market.
Chapter 4: Interest rate trends explained
Today's low interest rates prompt all kinds of questions. Why is the interest rate so low at the moment? Will it start rising again in the future? And can households afford to keep paying their mortgage if interest rates rise? In this chapter we explain why a rise in interest rates will not cause long-term payment problems for homeowners.
4.1 Why are interest rates so low?
The current low interest rates in the Netherlands are the result of both temporary factors and structural long-term developments.
The recent drop in capital market interest rates is due to a combination of low inflation and the easing of monetary policy. Low inflation is related to low economic growth in the eurozone. In order to tackle excessively low inflation the ECB announced in September that it would buy up asset-backed securities and covered bonds in order to inject extra liquidity into the financial system. This is not yet a step towards actual quantitative easing (buying up government bonds), but the markets believe that the chance of this actually happening has increased. In the USA, where the economy is growing more strongly, the programme for buying up bonds is already being phased out, but its impact on capital market interest rates in Europe is probably small because it is precisely such easing measures that are being taken in Europe. For a more detailed discussion of the ECB's policy and the capital markets' reaction, see the chapter (in Dutch) Rente en Valuta from our Economic Quarterly Report of September. For the coming years we expect a modest economic growth in the eurozone, which will be coupled with relatively low inflation and low interest rates. These expectations are described in our Special (in Dutch) Het groeiperspectief van de Eurozone. Should the economy grow more strongly again, this would mean that inflation and the nominal interest rates would rise again.
In the long term, structural factors will play a role such as the worldwide rise in savings and an increased demand for safe investments. The leading economists Paul Krugman and Olivier Blanchard write in the recently published Ebook of VoxEU (2014) that this also relates to demographic factors: the industrialised countries are faced with an ageing population and people are saving more to provide for their old age. As a result, the supply of capital has risen structurally during the past decades. There is also a structurally lower need for credit among businesses, caused by a shift from capital-intensive heavy industry to less capital-intensive ICT. This may explain the fall in real interest rates since the 1980s (Figure 21).
The structural developments described above explain why we do not expect real interest rates - the real interest rate is the nominal interest rate corrected for the level of inflation - to reach the same high levels as in the 1980s. In the rest of this chapter we will discuss to what extent a limited rise in interest rates affects the monthly debt service of homeowners.
4.2 Interest rates and mortgage payments
In the Netherlands the impact of interest fluctuations on households' mortgage payments is limited. In addition, many homeowners have lower mortgage payments than the maximum they could afford.
The average gross interest rate that Dutch homeowners are paying is very stable. Figure 22 shows that the gross interest rate paid in the Netherlands on mortgages (the dark blue line in the figure) has been around 4.5% for many years, whereas the average interest rate for new mortgages in the same period fluctuated between 3.5% and 5.5%. After deducting mortgage interest relief (according to the CBS some 90% of households have mortgage interest relief of 42%), this works out at around 2.6% net. Thanks to the stability of the average gross mortgage rate, this also means that the part of the interest borne by the government through mortgage interest relief is stable. As such, mortgage interest relief does not form a significant budget risk for the government.
The reason why the average mortgage rate on outstanding mortgages is hardly fluctuating at all is because homeowners in the Netherlands opt for a long fixed-rate period, particularly if the interest rate is low or is starting to rise. Figure 23 shows the fixed-rate period for new mortgages (including refinancing) and extensions. During the period from 2005 to 2006, mortgage rates were low and an increasing number of homeowners chose to take out a mortgage, refinance or extend their mortgage for a long fixed-rate period. Even when mortgage rates began to rise in 2007, households chose a long fixed-rate period: around 70% of the amount tied up in new mortgages and extensions was for a fixed-rate period of longer than 5 years. When interest rates began to fall sharply again in 2009, particularly for a short fixed-rate period, only 40% of the amount tied up in new mortgages and extensions was for a fixed-rate period longer than 5 years.
The choice for such a long fixed-rate period has been increasingly forced upon homeowners in recent years by the high notional interest rate (‘toetsrente’). This interest rate determines, alongside the income, the maximum size of the mortgage if a fixed-rate period of less than ten years is chosen when buying a house - see box: How is the maximum mortgage sum calculated? For those who want to borrow as high a sum as possible, the incentive is then to choose a fixed-rate period of at least ten years. These house buyers will then be immune from interest fluctuations for the coming ten years.
If, on the other hand, a house buyer decides on a shorter fixed-rate period, the maximum permitted mortgage payments (and therefore the maximum mortgage amount) will be calculated on the basis of the high notional interest rate. The actual mortgage payments will then be lower than the calculated maximum, creating a buffer to absorb any future rises in mortgage payments after the fixed-rate period has expired.
Box 2: How is the maximum mortgage sum calculated?
Since 2007 the Code of Conduct for Mortgage Loans (Gedragscode Hypothecair Financieren -GHF) has laid down how mortgage providers must calculate the maximum amount of the loan compared to the income (‘loan-to-income’) and the interest. This incorporates a number of extra safety measures. The Netherlands Authority for the Financial Markets (AFM) oversees compliance with the Code of Conduct. The Code of Conduct was tightened in 2011 and since 2013 the procedure described below has been required by law (Wet Tijdelijke regeling hypothecair krediet - Interim scheme for mortgage loans Act).
The maximum percentage of the gross income that households can use to service their mortgages is determined each year by the NIBUD (National Institute for Family Finance Information). The current percentage is from 16.5% for those on low incomes to 37.5% for those on high incomes. The maximum amount that can be borrowed is based on this so-called 'finance charge burden’ (financieringslastpercentage) and the mortgage interest rate. This calculation always assumes a full annuity repayment and the interest rate that belongs to a long fixed-rate period, irrespective of the form of mortgage and the fixed-rate period that the client ultimately chooses.
With a fixed-rate period of less than ten years, the actual interest rate cannot be used, but instead a notional interest rate of at least 5%. Only with a fixed-rate period of ten years or longer is the actual mortgage interest rate used to calculate the maximum amount that can be borrowed. Table 4 shows how the notional interest rate has been adjusted in recent years.
To summarise, we can conclude that the average impact of a rise in interest rates on households' mortgage payments is limited by the mortgage interest rate deduction and because homeowners choose a relatively long fixed-rate period. Most households can also cope with a rise in mortgage payments because their actual payments are lower than the maximum calculated by the mortgage provider. The impact of an interest rate rise for an individual household depends on the chosen mortgage form and the fixed-rate period. In the next paragraph, we will discuss the consequences for various groups of existing and new homeowners.
4.3 Who is affected by a rise in interest rates?
In the current slump we do not expect to see any sharp rises in interest rates. And in a scenario in which interest rates do rise, in most cases this will go hand in hand with stronger economic growth and a rise in inflation and wages, so that a rise in mortgage payments need not be a problem. Only in exceptional situations, such as an inflation shock caused by a spike in oil prices, could a sharp rise in inflation and interest rates occur simultaneously with low economic growth.
If mortgage rates were to rise again in the future, only first-time buyers and those whose fixed-rate period for all or part of their mortgage loan expires at that time would be affected. Those whose fixed-rate period expires during the coming years are mainly existing homeowners with an interest-only and/or a (bank) savings-based mortgage predating 2013. During the period prior to 2013, most homeowners chose an interest-only and/or (bank) savings-based mortgage. As these forms of mortgage make maximum use of the mortgage interest relief available, the actual monthly debt service for such a mortgage is lower than with an annuity mortgage. Among homeowners with a partial interest-only and/or (bank) savings-based mortgage, there is room in the budget for higher housing costs. This may partly explain why few homeowners in the Netherlands find themselves in difficulties when mortgage payments rise following the expiry of a fixed-rate period. Long-term mortgage payment problems are not so much the result of a rise in mortgage payments, but tend to have their origins in a significant drop in income following unemployment and/or divorce.
Since 2013 first-time buyers have been choosing en masse for an annuity mortgage, because they will only qualify for mortgage interest relief if they choose an annuity or linear mortgage. With an annuity mortgage the mortgage interest relief is lower than with a savings-based mortgage predating 2013. This makes it slightly more vulnerable to interest rate changes, as discussed in our Special (in Dutch) De gevolgen van de terugkeer van de annuïteitenhypotheek.
A small proportion of first-time buyers choose an annuity mortgage with a short fixed-rate period. As described in paragraph 3.2, this group too still has scope in the budget: their actual mortgage payments are lower than the maximum they could afford according to the NIBUD. Therefore, they can compensate for a rise in mortgage payments by modifying their consumption pattern.
The majority of first-time buyers choose a fixed-rate period of ten years. If they borrow the maximum amount on an annuity basis, there is no longer any margin for this group between their actual payments and the maximum housing costs which according to the NIBUD they would be able to afford. On the other hand, they will be immune from interest rate rises for a period of ten years. Only in 2023-2024 may they have to face a rise in their nominal mortgage payments caused by interest rate rises. Whether and to what extent the mortgage payments will rise in real terms will depend on inflation. A rise in mortgage payments will hardly be a problem for many first-time buyers. During the first ten to twenty years of their careers, generally enjoy a strong rise in real wages. Real wage rises for those in their forties and fifties, on the other hand, are less pronounced. If they borrow the maximum mortgage amount based on their current income, when the fixed-rate period expires some of them could possibly find it more difficult to cope with a real rise in mortgage payments. If on the other hand they choose a linear mortgage, their payments will be higher in the initial years, but lower after ten years.
Households which choose for a variable interest rate for all or a large part of their mortgage would appear to be the most vulnerable to interest rate fluctuations, but their mortgage payments are relatively low. Moreover, they have the option of converting their mortgage into one with a longer fixed-rate period as soon as they see interest rates beginning to rise. It should be noted that it can only be established retrospectively whether fixed or variable has been the best choice (Legierse, 2009)
All in all, we expect mortgage rates to remain low for the time being. If interest rates do rise, the majority of households in the Netherlands can cope with a rise in their nominal mortgage payments. Major income shocks, such as those resulting from unemployment or divorce, pose a greater risk for payment problems than changes in interest rates.
 All other variables that may influence the supply therefore remain constant.
 This also includes homes that remain unsold. That is why the duration of the supply is significantly higher than the average sale time of homes in that quarter.
 These are houses that are built for sale by property developers, building companies and housing associations. Of this segment, approx. 75%-80% of the new homes are included in the New Housing Monitor (MNW).
 Part of the net monthly income that is used to pay the net housing costs at the moment of purchasing a house.
 The colours green to yellow indicate the extent of price rises compared to the other segments. Yellow represents the least rise compared to the other segments, green the most.
 Note that this is the median house price of the NVM, which does not make any adjustment for composition effects.
 In Zeeuws-Vlaanderen the annual rise in house prices this quarter has been 23.7%, but there was a significant difference in the sale of more expensive houses compared to the previous year.
 Note that these are the prices of the NVM areas. Actual trends may vary from municipality to municipality.
 The maximum LTV is falling each year by 1 percentage point, from 106% in 2012 to 100% in 2018, thus limiting the financing capacity of buyers.
 Note that this means that the maximum purchase value of an owner-occupied house with NHG has fallen from approx. € 250,000 to € 230,000, as the NHG threshold includes purchasing costs payable by the purchaser, usually 6%.
 This chapter focuses on the impact of interest rate changes on the affordability of a household’s monthly debt service. We do not discuss the impact of interest rates changes on house prices. A change in house prices affects the household’s net wealth position, but a change in house prices alone does not affect the affordability of the household’s monthly debt service.
 A household can combine a variety of mortgage forms or fixed-rate periods. The figure shows that 20% of the total number of the new or extended amount incurs a variable interest rate. It is therefore not the case that 20% of households choose a mortgage with a fully variable interest rate.
Calcasa (2014). The WOX Quarterly Q2 2014.
Dalen, P. van (2014a). Vertrouwen Nederlandse woningmarkt naar hoogste punt sinds 2004.
Dalen, P van (2014b). Gevolgen veranderingen NHG in kaart.
Dalen, P. van (2014c). themabericht over schenkingen moet nog komen.
Dam, F van (2010). Nieuwbouw, verhuizingen en segregatie, Planbureau voor de Leefomgeving.
Francke, M. and A. van den Minne (2013). Starters slaan hun slag! Real Estate Research Quarterly, 4: 33-40.
Kosten Koper (2014). Kosten koper berekenen.
Legierse, T (2009). Hypotheekrente: Variabel of vast?
Oevering, F. (2014). De stad als kraamkamer: Steden zijn de motor van de demografische ontwikkeling.
NVM (2014). Woningmarktcijfers 3e kwartaal 2014.
Teulings, C. and R. Baldwin (2014), Secular stagnation: Facts, Causes and Cures, VoxEU.org eBook.
Vries, P. de (2014a). Herstelvermogen van huishoudens die onder water staan.
Vries, P. de (2014b). De gevolgen van de terugkeer van de annuïteitenhypotheek.
Wijffelaars, M. and J. van IJzerloo (2014). Mondiale economie kabbelt voort.
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 Macrobond, 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 ‘economics@rn,rabobank,nl’,
Pieter van Dalen, Theo Smid, Leontine Treur and Paul de Vries
Tim Legierse, head of Head Domestic Research, Economic Research Department
Pieter van Dalen, Reinier Meijer and Selma Heijnekamp
© 2014 - Coöperatieve Centrale Raiffeisen-Boerenleenbank B,A,, the Netherlands