Journal Issue
Share
Article

Denmark: Selected Issues

Author(s):
International Monetary Fund. European Dept.
Published Date:
December 2014
Share
  • ShareShare
Show Summary Details

Household Debt in Denmark1

Danish household debt is the highest among the OECD countries, standing at about 300 percent of disposable income. This paper examines Danish household balance sheets and assesses factors that have contributed to the high level of household debt and their policy implications. Various factors seems to account for the size of household debt, including large pension assets, a highly developed mortgage market, the availability of flexible mortgage products such as deferred amortization loans, indirect subsidies through tax preferences for home ownership, and a regulated rental market that limits mobility. While direct risks of high household debt to financial stability have been contained so far, macroeconomic risks with possible indirect impact on financial stability still remain in the event of large adverse shocks. High debt may also be weighing on consumption, contributing to Denmark’s still-significant output gap. Policy measures could be considered to facilitate further deleveraging in ways that minimize adverse effects on growth and to prevent rapid debt accumulation in the future, including reducing tax preferences for home ownership and using the fiscal savings to fund high-multiplier stimulative measures, removing distortions in the rental market, introducing tighter limits on interest only loans, and applying macroprudential tools.

A. Danish Household Balance Sheets

1. Danish household debt stands currently at about 300 percent of disposable income, the highest among OECD countries (Figure 1.A). Household debt has been always relatively high in Denmark, well above 100 percent of disposable income even in the 1970s (Figure 1.B). The recent upward trend in household debt started in the late 1990s, reaching 300 percent of disposable income by 2008. This coincided with the period of rapid house price appreciation that peaked in 2007. The housing boom eventually ended with large house price corrections, but household debt remains elevated with only moderate deleveraging having taken place so far.

Figure 1.Danish Household Balance Sheets

2. Household assets are also high in Denmark (Figure 1.C). Danish household financial assets are among the highest in OECD countries, and together with non-financial assets (mostly housing wealth), Danish household assets stood at over 800 percent of disposable income in 2013.

3. Danish household assets have grown substantially over time (Figure 1.D). While liabilities reached high levels, assets have risen even more rapidly for the past few decades. What is most noticeable about the asset side development of the Danish household balance sheets is the steady increase in pension assets. Pension assets (before taxes) were well below 100 percent of disposable income in the 1970s but continued to grow for the past decades, having reached above 300 percent recently. Assets in the form of securities and shares have also risen significantly in recent years while cash and deposits, the most liquid form of assets, have been relatively stable.

4. While total household assets are high in Denmark, most of the assets are illiquid such as individual pension accounts and housing wealth (Figure 1.E). The size of illiquid assets is larger in Denmark than in most of its peers. On the other hand, the size of liquid assets is comparable with peers, only slightly higher than the peer average. Net worth is more than 500 percent of disposable income in Denmark, which is high but not significantly higher than other OECD countries.

5. Financial net worth gives a somewhat different picture about the financial position of Danish households (Figure 1.F). Once housing wealth is taken out, net worth drops significantly, even though it is still positive. Non-pension financial net worth, which takes out pension assets from financial net worth calculation, therefore taking account of only liquid assets, is negative for Denmark, which is also the case for the Netherlands and Norway. Negative non-pension financial net worth illustrates the importance of housing wealth and pension assets for household financial positions in these countries.

6. Danish household micro data show that high income families tend to have higher household debt while most households have substantial amount of assets. Median debt higher than 100 percent of disposable income is concentrated in families from income deciles 7-10, but 75th percentile numbers show that more indebted families tend to have much higher debt ratio than median in all income deciles. On the other hand, median household assets are substantially higher than median debt levels in all income deciles.2

Distribution of Household Debt by Income, 2011

(Percent of disposable income)

Sources: Danmarks Nationalbank and IMF staff calculations.

Distribution of Household Assets by Income, 2011

(Percent of disposable income)

Sources: Danmarks Nationalbank and IMF staff calculations.

7. In sum, Danish household balance sheets are characterized as large liabilities and assets, and assets consist mostly of illiquid assets in the form of housing and pensions. In light of these findings, the next section analyzes factors that account for high levels of Danish household debt.

B. What Determines the Level of Household Debt?

8. Social and institutional setups are likely to matter in accounting for household debt. In simple economic models, household consumption is determined based on expected lifetime income, interest rates, and preferences for current and future consumption. The current level of spending, together with the current level of household income, determines the level of saving or borrowing. The life-cycle and permanent income hypotheses suggest that households will tend to take on debt when they are young, gradually shift to positive net worth as their incomes rise over time, and eventually run down their net worth in retirement. While these are well accepted theories, in practice, various institutional factors also affect households’ decision to take on debt. This subsection focuses on institutional aspects – including the pension system, mortgage products, tax preferences for home ownership, and the rental market – to shed light on factors behind Denmark’s high level of household debt.

Pension assets

9. Social institutions affect households’ consumption-smoothing decisions. Where the welfare system is well developed, households take account of transfers and welfare and pension benefits provided by the government as part of their lifetime income. Cross-country differences in household debt are thus likely to reflect, at least partly, differences in welfare system. In particular:

  • Pension contributions tend to reduce the amount of current disposable income, thereby increasing the need for households to borrow more when they are young. Large pension benefits also reduce the need to save for retirement.

  • As noted earlier, the size of pension assets as a share of disposable income is very large in Denmark (see Box 1 for a description of the Danish pension system). Occupational pension schemes, which started in the late 1980s and early 1990s, are likely to have contributed to the persistent increase in pension assets for the past decades. As the system is not fully phased in, the accumulation of pension assets relative to dispoible income is likely to continue increasing over the next two decades or so.

Household Pension Assets and Debt, 2012

(Percent of household gross disposable income)

Sources: OECD and Fund staff calculations.

Box 1.Danish Pension System

The Danish pension system consists of multiple pillars:

  • The first pillar is state retirements and statutory schemes, which are mandatory. The first pillar has two components: (i) Basic pension; and (ii) labor market supplementary pension (ATP). ATP is a fully funded defined-contribution scheme.

  • The second pillar is quasi-mandatory occupational pension schemes which are agreement-based and cover about 90 percent of labor force. These schemes are also fully funded defined-contribution schemes. Occupational pension schemes were established as part of collective bargaining at the end of the 1980s, and they are typically sector-specific. Contributions to occupational pensions range between 9 and 17 percent of gross wages

  • The third pillar is a voluntary supplementary pension schemes, typically managed by banks and insurance companies.

Danish Pension System

Source: Based on Guardiancich (2010).

The Danish pension system resembles the Dutch pension system which also consists of three pillars: The first pillar is basic state old age pension under a statutory insurance scheme. The second pillar is quasi-mandatory occupational schemes which cover about 90 percent of employees (OECD, 2013). The third pillar is voluntary private pension schemes. The key difference between Danish and Dutch pension systems is that while Danish pension schemes are mostly defined-contribution schemes with individual accounts, the Dutch system is mostly based on defined-benefit schemes.

The increasing size in pension assets is likely to have supported, at least partly, the recent rise of Danish household debt. On the other hand, Danish household debt was relatively high even back in the 1970s when the size of pension assets was smaller. This suggests that some other factors were also at play.

Housing market

10. The way the housing market is organized is likely to affect households’ behavior as well. A large portion of household debt consists of mortgages. Thus institutional factors that affect the effective cost of credit and mortgage affordability are likely to influence households’ behavior. These factors include mortgage product innovation and tax preferences for mortgage debt and home ownership.

11. Mortgage market. The Danish mortgage market is highly developed. The Danish market for covered bonds backed by mortgage loans is one of the largest in the world. The stock of covered bonds is equivalent to 143 percent of GDP, more than four times larger than in other countries (IMF, 2014). Mortgage credit is readily available in Denmark, and the average price of mortgage loans is among the cheapest in Europe when adjusted for fees and the cost of pre-payment options (Erlandsen et al., 2006). The well-developed mortgage market has supported Danish households to finance their home purchases and build housing wealth throughout the past few decades.

12. The Mortgage product innovation. Mortgage lending has seen significant product innovation since the late 1990s. The Danish mortgage market was dominated by fixed rate mortgages until variable rate loans were introduced in 1996. The share of variable rate loans made by mortgage credit institutes (MCIs) to households grew from zero to about 35 percent by the end of 2003 and stands at close to 70 percent as of March 2014. “Interest only” (IO) loans—these are called deferred amortization loans in Denmark as they normally come with a deferred amortization period of up to 10 years—were introduced in 2003, joining the group of European countries which have IO loans.3 The use of deferred amortization loans grew rapidly since then, and these loans account for a little over half of outstanding loans by MCIs in recent years. These new mortgage products, particularly the use of deferred amortization loans are likely to have increased the affordability of mortgages among households with limited capital and therefore have contributed to the rapid increase in household debt in Denmark since 2003.

MCIs Outstanding Loans to Owner Occupied Homes and Holiday Cottages

(Share by loan types)

Sources: Danmarks Nationalbank and Fund staff calculations.

Availability of interest-only mortgage loans
19952005
AustraliaNAA
DenmarkNAA
FinlandNAA*
FranceAA
GermanyAA
GreeceNAA*
IrelandAA
KoreaNAA
NetherlandsAA
PortugalNAA
Spain?A
SwitzerlandAA
UKAA
A= available NA=not availableSource: Scanlon et al. (2008)

13. Tax preferences for owner-occupied housing and mortgage debt contribute to high household debt.

  • Denmark, like many other advanced economies, allows mortgage interest deductibility from taxable income. The degree of mortgage interest deductibility has been lowered in stages over time. The 1994–98 tax reform reduced deductible interest payments to about 46 percent, and deductibility was lowered further to 33 percent in 2002 (Erlandsen et al., 2006). Starting from 2012, mortgage interest deductibility for interest payments exceeding DKK 50,000 is planned to be reduced to 25 percent by 2019.4 In 2014, interest deductibility still remains at 33 percent (Table 1). Despite the gradual reduction in mortgage interest deductibility, the degree of tax relief on debt financing cost, measured by the difference between the market interest rate and the after-tax debt financing cost of housing, is high in Denmark among OECD countries, the third most generous after the Netherlands and Czech Republic (Andrews et al., 2011).5

  • Tax rates for property tax in Denmark are higher than those in most peer countries listed in Table 1. However, a tax freeze was introduced in 2002 and there are also special reductions, so the actual tax burden is lower than the rates suggest. The taxable value for property value tax is effectively frozen at the assessed value in 2001, implying that a nominal ceiling is imposed on the property value tax so that a rise in the value of property will not increase tax payments for the home owner.6 In addition, a limit of 5 percent for the annual increase in payments of land taxes was introduced in 2002, further reinforcing the pro-cyclical pattern of house prices and implied tax rates (Callesen, 2013). These limits on property tax payments not only serve as an implicit subsidy for housing but also could amplify economic cycles through the pro-cyclicality.

Table 1.Taxes on Housing
FinlandDenmarkNorwaySwedenIrelandSpainNetherlands
Mortgage interest tax deductibility75% of interest paid are deductible from taxable capital32.7 % of mortgage interest can be deducted in other taxable income.Up to 30% of mortgage interest paid for first-time buyers, and 15% for others up to a certain number of years and up to an absolute ceiling. Only mortgages taken out before end 2012 qualify, and all relief terminates at end 2017.Mortgage interest payments are not PIT deductible for properties purchased after January 1st 2013Fully deductible with no cap (limited to fully extended sicne Jan 2013; earlier non-amortizaing loans continue to be eligible to the full deductibility.)
The value of deduction is reduced by one percentage point per year between 2012-2019 to 25%Fully deductible30% of interest exenditure is deductible from other taxable
(33% during 2002-2012)
Capital gains tax30% applicable to income up to EUR40,000 and 32% on income exceeding that amountExempt if the owner livedExempt if lived for one year0.333% unless owner occupied, with special properties purchased in 2012-1419%< EUR6,000 21%>EUR6,000Exempted if dwelling is main fiscal residence
Inheritance tax (inheritance to family members)15 % of the estate exceeding DKK 264,100 (2012). No tax on spouse.Abolished in 2014Abolished in 200533% >EUR225,000 (to children). No tax for family members living in0.8%-36.5% (2013)€100,000 exampted if donated to reduce mortgage debt (temporary)
Property tax
Municipal taxReal estate taxMunicipal real estate tax (land tax)Municipal taxMunicipal property fee replaced real property tax on private residence in 2008Local property taxTemporary surcharges of up to 10% for 2014 and 2015
0.6-1.35% (0.32-0.75% for primary residence)1.6-3.4%0.2 -0.7 %SEK7,074 fee or 0.75% of the assessed value of the property if that amount is lower0.18-0.25%0.1-0.3% of property
Municipal tax valuationLevied on the tax value of the real property.Levied on the land value. A limit of 5 percent for the annual increase in payments of land taxes introduced in 2002.The taxable value is 20-50 % of the market value.Self-assessed marketLevied on the cadastral value, which is adjusted every eight years with respect to the property’s market
State taxProperty value tax 1% < DKK3,040,000 (taxable value) 3% > DKK3,040,000 (taxable value)
State tax valuationThe taxable value is the lowest of (i) the assessed value as of Jan 1 of the current tax year; (ii) 105% of the assessed value as of Jan 1, 2001; or (iii) the assessed value as of Jan 1, 2002.
Wealth taxNet wealth tax abolished in 2006.No wealth tax, but property value tax,Municipal net wealth tax (individuals) 0%< NOK 1,000,000 0.7%>NOK1,000,000Net wealth tax abolished in 2007No wealth tax (abolished in 1978)Net wealth tax was reinstated for years 2011-2014. 0.2-2.5%No wealth tax
State net wealth tax

(individuals) 0%< NOK 1,000,000

0.4%>NOK1,000,000
Wealth tax valuationThe taxable value is 30% of the market value for the primary residence and 60% for secondary residences.
Tax on imputed rentsTaxable; rates increased from 0.6 to 0.7 percent for houses < €1,040k, from 1.55 to 1.8 percent, for houses above €1,040k
Sources: Deloitte international tax highlights, www.kpmg.com, National tax authorities websites, Sveriges Riksbank (2014), Citizens Information (Ireland).

Property and Land Taxes and House Prices

(Percent; index 2005=100)

Sources: Danmarks Nationalbank and Fund staff calculations.

Rental market

14. The structure of the rental market also affects the demand for housing and mortgage debt. A well-functioning rental market would allow people to make non-distorted choices about housing and asset structure. Distortions in the rental market may make rental properties inaccessible to those household who would prefer to rent (e.g., new families) and thus force households to pursue homeownership earlier than they would otherwise or when it is otherwise not the best option, creating excess demand for housing and mortgage debt, possibly increasing price volatility for owner occupied houses.

15. The rental market exists in Denmark, but it has many distortions.7 The share of rental dwellings is relatively high in Denmark compared with other OECD countries, but Danish rental housing (both private and social) is subject to rent regulation and also receives more direct/indirect public subsidies than owner-occupied housing. Housing allowances are the major direct support, covering both social housing and private rental housing. About 20 percent of the population received cash allowances for rental costs.8 These allowances on average cover 40 percent of the actual rents paid in Denmark.

Home Ownership, 2008

(Percent of total dwelling stock)

Sources: CECODHAS Housing Europe Review, 2012 and Fund staff calculations. For Ireland, the data are as of 2004.

16. Rent regulation for social housing based on cost-based principles implies that rents may deviate from the market-clearing level in both directions. On average, the rent level in social housing is estimated to be below the market-clearing level, being reflected in waiting lists for attractive apartments (Erlandsen et al., 2006). However, given mortgage products innovations and tax subsidies for owner-occupied housing, the after-tax monthly financing costs of owner-occupied housing could be lower than rents in recently constructed social housing in some areas.

17. In the private rental market, tenant rights are highly protected. Open-ended leases are the standard contract, and the landlord cannot terminate the contracts except in certain cases. Tenants are also given the right to sub-let. Buildings built before 1991 are subject to rent control, and more than 80 percent of the private rental housing market stock in Denmark is under rent regulation (Erlandsen et al., 2006). Cuerpo et al. (2014) reports that the degree of rent control in Denmark is the third highest among their sample of 20 European countries. These distortions are likely to have limited mobility and may constrain the private rental market, which is much larger in neighboring Germany.

18. In sum, the interaction of various institutional factors seems to explain high household debt in Denmark. While Denmark’s mortgage market is well functioning and highly developed, there are large direct and indirect subsidies for all types of housing, and the highly regulated rental market is likely skewing housing needs and use in Denmark. The coexistence of a well-developed mortgage market and regulated rental market, together with large tax preferences for owner-occupied housing and mortgage debt seems to have supported the underlying tendency of high household debt in Denmark. The 2000-07 housing boom, combined with mortgage product innovations and the rise in financial assets (especially pension assets), are likely to have contributed to the further surge in household debt since the late 1990s.

C. Macroeconomic Impact of High Household Debt

19. High household debt could pose direct risks to financial stability if the number of mortgage loan defaults rises sharply in the face of adverse shocks. In practice, however, despite the high level of household debt, the household arrears ratio remains relatively low in Denmark, after rising slightly during the global financial crisis. The low level of arrears reflects the strong legal and regulatory framework in Denmark: The LTV ceiling of 80 percent on new mortgage loans limits lender losses in the event of a default. In addition, mortgage loans are full recourse backed by quick repossession and forced sale procedures. A mortgage loan is considered in default after 3 ½ months of non-payment, and forced sale procedures are initiated unless alternative workout procedures are agreed with the borrower. It typically takes less than nine months from the payment becoming overdue to the property being sold. This also contributes to limiting mortgage banks’ potential losses. If the sales of the property do not sufficiently cover the mortgage bank’s claim, the uncovered claim remains as an unsecured claim against the borrower.

Arrears Ratio, Mortgage Banks

(Percent of total number of outstanding loans to households)

Sources: Danmarks Nationalbank and Fund staff calculations.

20. Previous studies have found that direct risks from high household debt to financial stability are limited in Denmark. A series of papers by the Danmarks Nationalbank using Danish household micro data have demonstrated that gross debt is highly concentrated among high-income families as well as households with larger financial assets. The Nationalbank’s stress tests have also shown that most households are resilient to interest rate shocks and the number of families in arrears remains relatively low even in stress scenarios (Danmarks Nationalbank, 2012;Danmarks Nationalbank, 2013). Overall, these studies have concluded that elevated household debt does not directly threaten financial stability.

21. However, high levels of household debt may pose risks to macroeconomic stability by dampening consumption. Case et al. (2005) and Case et al. (2013), examining the link between increases in housing and financial wealth and household spending, find a large effect of housing wealth on household consumption using a panel of country level and U.S. state-level data. Dynan (2012) uses U.S. household level data to examine the effect of leverage and debt overhang on household consumption. She finds that highly leveraged homeowners had larger declines in spending between 2007 and 2009 than less leveraged ones even when the former had smaller changes in net worth. Her results support the view that excessive leverage has contributed to the weakness in consumption. More aggregate cross-country studies have also shown that high household debt is linked to lower consumption. IMF (2012) finds that housing busts preceded by larger run-ups in gross household debt are associated with significantly larger contractions in economic activity, with larger declines in household consumption and real GDP and more increases in the unemployment rate, and the reduction in economic activity persists for at least five years.

Private Consumption and Household Debt

(Percent growth; percent of disposable income)

Sources: OECD and Fund staff calculations.

22. Recent empirical results about Denmark are also in line with the earlier findings. A study by Danmarks Nationalbank, based on Danish household micro data, has also shown that households with high loan to value (LTV) ratios cut consumption more than households with low LTV ratio during the global financial crisis (Andersen et al. (2014)). The larger drop in spending among the high-LTV households reflects the fact that these families consumed a larger fraction of their income than their less-leveraged peers prior to the crisis. These results suggest that consumption was lowered more than would have been the case if Danish households did not have so much high debt levels.

23. These findings suggest a need for measures to facilitate further deleveraging and prevent excessive debt accumulation in the future. Denmark’s real private consumption has been flat in recent years and the deleveraging process has been slow. As a result, the recovery after the global financial crisis has been anemic. While high household debt may have had limited direct impact on financial stability, risk associated with household debt still remains in the Danish economy and could appear somewhere else (e.g., problems with commercial loans due to reduced consumption, especially in retail, commercial real estate and construction sectors) if further large shocks hit the economy. These macroeconomic impacts could feed back into financial stability (most likely in the commercial banks rather than the mortgage banks) if loan problems in the affected sectors are large enough.9 High levels of household debt thus increases the vulnerability of the economy, and therefore policy measures are needed to ease debt levels gradually and facilitate the rebuilding of buffers so that these risks are mitigated. The next section discusses possible policy options.

Real Private Consumption and House Prices

(Billions of 2005 DKK, left; Index: 2010=100, right)

Sources: IMF World Economic Outlook, OECD and Fund staff calculations.

Household Deleveraging

(Percent of disposable income)

Sources: Danmarks Nationalbank and Fund staff calculations.

D. Policy Options

24. Policy changes could be considered in several areas. The discussion in Section B suggests that the functioning of the housing market could be improved by removing various direct and indirect subsidies to owner-occupied housing and by making the private rental market more flexible, while ensuring that risks are better managed in the mortgage market. Possible policy options include the following:

  • Reducing tax preferences for owner-occupied housing and mortgage debt: As discussed earlier, tax relief on the debt financing cost of owner-occupied housing is relatively large in Denmark. Mortgage interest deductibility is expected to be reduced gradually by 2019. Given the current low interest rate environment which limits the effective benefit of interest deductibility and improving housing market conditions, now seems to be the right time to implement the reduction. Some countries are implementing these measures recently. For example, the Netherlands, where mortgage interest was fully deductible, introduced changes in January 2013 (Box 2): Only interest on fully amortizing mortgages is now tax deductible for new loans.10 Mortgage interest deductibility will be also reduced to 38 percent in steps of ½ percent per year. Moreover, the valuation of properties for tax purposes should also made more in line with the market value so as to remove the pro-cyclicality of effective tax rates and house prices due to the tax freeze. Fiscal savings from such tax changes could be used to fund stimulative measures with a high multiplier, thereby providing a “balanced-budget stimulus.”

  • Reforming the rental market to enhance flexibility: Distortions in the rental market should be minimized to allow the rental market to function as a complementary alternative to home ownership. Cueropo et al. (2014) argues that rent controls have a significant destabilizing impact on the aggregate housing market. They also advise against using rent controls for redistribution purposes, noting that more targeted policies would be welfare-enhancing without affecting rental market efficiency. Spain and Portugal have implemented reforms recently to increase the rental market efficiency (The size of rental market is small in these countries). In Spain, rent increases for new leases are now set freely by the contractual parties without the need to explicitly index the increase to the consumer price index. The tenant-landlord relationship is also now more balanced—previously, tenants were heavily protected and landlords’ rights were limited—through measures such as reducing the minimum contract duration from five to three years.

  • Discouraging the use of deferred amortization loans: Adjustable and variable rate loans with and without interest only periods are the dominant form of mortgages in Denmark. These loans played an important counter-cyclical role during and after the financial crisis as borrowers were able to reduce payment obligations by shifting as rates on existing adjustable-rate loans fell and as households shifted from high fixed rates to the lower rates on adjustable rate loans. However, these loans have potential adverse effects on loan default risk and impairments, as they increase the sensitivity of the economy to interest rate hikes. Deferred amortization loans also delay the process of deleveraging. Measures aimed at reducing the use of deferred amortization loans could be considered. For example, eligibility for such loans could be limited to loans with LTVs well below 80 percent or borrowers could be required to amortize the portion of the loan above a lower LTV ceiling, or risk weights and/or generic provisioning for IO loans could be escalated. 11 The recent proposals for Supervisory Diamond for mortgage credit institutions which include limiting short-term bonds and loans with interest only periods go in the right direction. Gradually removing the tax deductibility of interest payments on deferred amortization loans could also be considered to encourage borrowers to use such loans more sparingly. As mentioned earlier, the Netherlands has taken this measure recently by ending tax deductibility for non-amortizing mortgages (Box 2).

  • Using macrorudential tools: To prevent rapid accumulation of household debt in the future and mitigate vulnerabilities, new policy instruments could be considred. The use of limits on LTV ratios and debt service-to-income and/or debt-to-income ratios could be explored.12 In the Netherlands, the LTV limit of 100 percent was introduced in January 2013. It will be reduced further by 1 percent per year until it reaches 100 percent. A lower LTV limit of 50 percetn applies to IO loans. In additon, higher risk weights, increased sectoral capital requirments, or higher provisioning can be also applied to sectors with higher LTVs or to loans exceeding certain thresholds for LTVs.

  • Easing liquidity constraints: The Netherlands has recently taken a measure to temporarily raise the exemption threshold on the gif tax to €100,000 if the recipients use the gift to pay down mortgage debt ahead of schedule (see Box 2). This is unlikely to have a large effect, but this may be a relatively costless way to facilitate deleveraging.

Average Mortgage Rates in Denmark

(Percent)

Sources: Association of Danish Mortgage Banks and Fund staff calculations.

Box 2.Measures to Reduce Household Debt in the Netherlands

The Netherland’s household debt stands at about 300 percent of disposable income in 2013, only slightly higher than in Denmark. House prices fell by 27 percent in real terms in Netherlands, resulting in a loss of housing wealth of about 60 percent of GDP. The loss is concentrated mainly among younger cohorts where an estimated 60 percent have mortgages that are underwater.

To address household debt overhang, the Dutch authorities introduced a series of policy measures in January 2013, targeting the owner occupied housing sector. These measures include:

  • Tax deductibility was also partially removed. Only interest on fully amortizing mortgages is now tax deductible. Unlike in Denmark where deferred amortization loans typically come with a limited interest-only period, IO loans without any amortization are common in the Netherlands.

  • LTV was capped to 106 percent, which will be reduced further by 1 percentage point per year to reach 100 percent in 2018. Lower LTV limit of 50 percent applies to IO loans.

  • Prepayment on mortgage with 10 years or less of remaining maturity is made possible without penalty.

  • A temporary tax exemption (until end 2014) for monetary gifts of up to €100,000 used to reduce mortgages debt was introduced. More than 50,000 households have used the scheme so far. This measure is an attempt to encourage more transfers between the elderly and younger households, thereby easing liquidity constraints facing the latter and helping to boost their consumption and support aggregate demand.

Netherlands shares many similar features to Denmark: A highly developed mortgage market, large pension assets, tax preferences for owner-occupied housing, and distortions in the rental market. Despite these similarities, however, issues facing the Netherlands are somewhat different from those in Denmark, in part due to the differences in the institutional setups.1

For example, the Dutch pension schemes are mostly defined-benefit schemes. Occupational pension accrual rates are equal for all participants and younger employees contribute in excess of the present value of their pension benefits. This implies that savings are transferred from the relatively poorer young to the wealthier old. In this context, Mirkaic (forthcoming) shows that reducing pension contributions to an actuarially fair level and keeping the accrual rate unchanged could free up income to increase consumption and lower debt of younger households by 2-4 percent annually. On the other hand, this issue does not arise in the case of Denmark because the Danish pension schemes are mostly define-contribution schemes with individual accounts.

1/ See Hassine (forthcoming) for institutional details on the Dutch housing market and its mortgage financing system.
References

    Andersen, Asger Lau; Duus, Charlotte; Jensen, Thais Lærkholm,2014, “Household debt and consumption during the financial crisis: evidence from Danish micro data,Working Paper no. 89, Danmarks Nationalbank. http://www.nationalbanken.dk/en/publications/all/Pages/Household-debt-and-consumption-during-the-financial-crisis-Evidence-from-Danish-micro-data-.aspx

    Andrews, D., A.Caldera Sanchez and A.Johansson,Housing Markets and Structural Policies in OECD Countries,OECD Economics Department Working Papers, No. 836. OECD publishing. http://dx.doi.org/10.1787/5kgk8t2k9vf3-en

    Andritzky, Jochen,Resolving Residential Mortgage Distress: Time to Modify?IMF Working Paper, forthcoming.

    Callesen, Per,Property prices, debt and financial stability” in M.Balling and J.Berg edits. Property Prices and Real Estate Financing in a Turbulent World, SUERF Study, 2013. http://www.suerf.org/index.php?option=com_k2&view=item&id=518&Itemid=147

    Cuerpo, Carlos, SonaKalantaryan, and PeterPontuch,Rental Market Regulations in the European Union,Economics Papers 515, April2014, European Commission.

    Danmarks Nationalbank, Monetary Review, Part 2, “The wealth and debt of Danish families2nd quarter, 2012.

    Danmarks Nationalbank, Monetary Review, Part 2, Danish families’ financial robustness, variable rates and deferred amortisation,4th quarter, 2012.

    Danmarks Nationalbank, Monetary Review, Part 2, Danish families in mortgage arrears,3rd quarter 2013.

    Erlandsen, E., J.Lundsgaard and F.Hüfner(2006), “The Danish Housing Market: Less Subsidy and more Flexibility”, OECD Economics Department Working Papers, No. 513, OECD Publishing. http://dx.doi.org/10.1787/046875878368

    Guardiancich, Igor,Denmark - Current pension system: First assessment of reform outcomes and output,European Social Observatory, May2010.

    Hassine, Michelle,Housing Market and Finance in the Netherlands,2014 The Netherlands Selected Issues Paper, forthcoming.

    International Monetary Fund, World Economic Outlook, April2012.

    MicoMrkaic,House Prices, Consumption, and Household Debt Overhang in the Netherlands,2014 The Netherlands Selected Issues Paper, forthcoming.

    OECD, Pensions at a glance 2013, 2013.

    Reiakvan, Lisa Kristine, and HaakonSolheim,2013, “Comparison of household debt relative to income across four Nordic countries,Norges Bank Staff Memo, No.5.

Prepared by Kazuko Shirono (EUR).

However, households at the 25 percentile in debt are not necessarily in the 25 percentile in assets. If the highly-indebted households within each decile tend to be the low-asset households with each decile, then there could still be more distressed households than suggested by these figures.

The text table shows the availability of IO loans as of 2005, which could differ from the current practice. For example, IO loans are no longer offered in Ireland.

This applies to both new and existing mortgages.

This assessment refers to the condition in 2009.

With the tax freeze and other special reductions for houses and flats bought before July 1998 and for pensioners, the average effective rate paid in 2006 was 0.55 percent of the assessment value even though the statutory rate of real estate tax is 1 percent (and 3 percent above a certain threshold) (Erlandsen et al., 2006).

More recent data for Denmark (2013) also shows a similar composition of home ownership. “Others” for Denmark in the text chart is mostly cooperative housing.

Danish Ministry of Economic Affairs and the Interior. Andrews et al., 2011 also report that the share of population receiving cash allowances for rental costs is higher in Denmark than most other OECD countries.

See Andritzky (forthcoming) for discussion on negative feedback loops of household debt.

However, mortgages outstanding as of January 1, 2013 are unaffected.

See the 2014 Financial System Stability Assessment report for a set of recommendations on the mortgage market and covered bonds including ones on deferred amortization loans,

Although Denmark has an LTV limit on mortgage loans that are funded by covered bonds, borrowers can exceed the limit by taking out other loans.

Other Resources Citing This Publication