Norway: Selected Issues

Norway: Selected Issues

Abstract

Norway: Selected Issues

The Housing Boom and Macroprudential Policy1

The high and rising house prices and household debt in Norway pose important financial stability risks. To address these systemic risks, the authorities have implemented a number of macroprudential policy measures. This paper empirically assesses the effectiveness of the housing-related measures in the Norwegian context, controlling for other variables that affect house prices and mortgage credit growth. Additionally, a DSGE model is used to examine the potential impact of tightening certain policies.

A. The Norwegian Housing Market and Household Debt

1. Norwegian house prices have risen substantially over the past 15 years. Since 2000, nominal house prices have risen more than 140 percent (more than 80 percent after adjusting for CPI inflation), with average annual house price growth of 9.3 percent from 2000–07 and 4.6 percent since 2008. The increase was even larger in oil-dependent regions and cities such as Stavanger, where prices more than doubled between 2005 and their peak in 2013. The rise in house prices has also been geographically widespread, with prices rising by 70 percent or more in different regions since 2005. Recently, house price inflation slowed during 2015, but accelerated again in some regions in early 2016. Developments have diverged across regions. In the Oslo area and in central Norway prices have continued rising robustly. In Stavanger, where the sizable drop in oil prices since 2013 has had a significant impact, including on unemployment, house prices have begun to decline from very high levels. Nationwide, house price overvaluation is estimated at 40 percent at end-2015 (based on the average of three different standard valuation measures).2

A03ufig1

Regional House Prices

(SA Index: 2005=100)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Statistics Norway and Fund staff calculations.
A03ufig2

Estimated House Price Valuation Gaps in Norway

(Percent)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: OECD and Fund staff calculations.Note: The model is described in the 2013 Nordic Regional Report and 2013 Norway Article IV Selected Issues.

2. Households’ indebtedness has risen along with house prices. Households’ debt level has risen from around 145 percent of disposable income in 2002 to over 220 percent in 2015, higher than in most comparator countries. This has been primarily driven by debt rising much faster than incomes, especially before the 2008–09 global financial crisis (GFC). Households’ debt growth averaged 12 percent per year before the GFC, and, while it has slowed, it has averaged 7 percent annually since then. Over the same period, annual disposable income growth averaged slightly more than 5 percent.

A03ufig3

Household Debt

(Percent of disposable income, 2014)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: OECD and Fund staff calculations.

3. Low unemployment, robust income growth, and declining mortgage rates have contributed to housing demand. Unemployment in Norway has remained relatively low, even following the GFC, which increased housing demand pressures. In particular, until 2014, high oil prices helped keep unemployment down, especially in oil dependent areas such as Stavanger. Housing demand has also been fueled by declining mortgage rates, with real mortgage rates down substantially since 2002 and falling close to zero by end-2015.

A03ufig4

Earnings Growth and Unemployment Rate

(Percent and percent of labor force)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Statistics Norway and Fund staff calculations.
A03ufig5

Interest Rates

(Percent)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Statistics Norway, Norges Bank, and Fund staff.

4. Structural factors have also contributed to high and rising house prices and household indebtedness (Figure 1). Population growth and increasing urbanization—with average population growth over 1.2 percent since 2008 and an average annual urbanization rate of 1.4 percent over 2010–15 according to the CIA World Fact Book—are increasing the demand for housing in the main urban areas. At the same time, restrictions on development and minimum unit size have constrained the responsiveness of the housing supply, driving up prices. Also, there are a number of tax incentives for home ownership and mortgage financing. Compared with other assets, owner-occupied housing enjoys a large discount in tax base calculation for wealth taxation (25 percent of market value for primary dwellings and 80 percent for secondary dwellings). Interest on mortgages is tax deductible, which effectively reduces the debt service costs, thereby incentivizing households to borrow more (IMF, 2013a). Finally, lower risk-weights for housing loans compared to corporate loans have caused banks to shift their loan portfolios towards mortgage lending.

Figure 1.
Figure 1.

Structural Factors Contributing to the Housing Boom

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

5. Overvalued house prices and elevated household debt levels can create systemic macro-financial risks. If house prices substantially exceed fundamentals, this increases the risk of a house price correction. The direct effect on default rates would probably be limited due to households’ financial buffers, the social safety net, and because mortgages are full recourse loans. However, a decline in house prices would weigh on households’ consumption (e.g., due to wealth effects), which would negatively impact output and non-financial corporates (NFCs), as well as increasing unemployment. A downturn would likely increase NFCs defaults on loans, especially loans to real estate developers, which would impair banks’ balance sheets.

B. Macroprudential Policy Developments

6. Norway has implemented additional capital buffer requirements for banks. In particular, the national legislation based on CRDIV and CRR introduced a countercyclical capital buffer (CCB), a systemic risk buffer (SRB) set at 3 percent of risk weighted assets (RWA) for all banks, and an additional capital buffer for domestically systemically important institutions (D-SIIs). Three financial institutions have been designated D-SIIs and will have to hold additional capital worth 2 percent of RWA from July 1, 2016. The CCB was also activated, which required banks to have additional capital worth 1 percent of RWA by mid-2015, and it will be increased to 1.5 percent of RWA by end-June 2016.

7. While increasing capital buffers strengthens banks’ resilience, some of the improvement in regulatory capital ratios was due to changes in risk weighted assets. Following the introduction of Basel II standards in 2007, banks using the internal ratings based (IRB) approach lowered risk weights on mortgages markedly. Banks also shifted the composition of their loan portfolios towards mortgages, which typically have lower risk weights than corporate loans. As a result, regulatory capital ratios (CET1 capital over RWA) improved much more than the simple leverage ratio (CET1 capital over total assets).3

A03ufig6

Tier 1 Capital Ratios

(Percent)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Norges Bank and Fund staff calculations.

8. This eventually prompted the FSA to tighten constraints on how banks using the IRB approach calculate risk weights for mortgage loans. Both the loss given default (LGD) and probability of default (PD) parameters of the IRB models were raised in 2014–15. Effectively, this pushed up average risk weights on residential mortgages from around 10 percent to around 20–25 percent. Additionally, the Norwegian authorities obtained the cooperation of the Danish and Swedish regulators in applying these changes to risk weight calculations for mortgage loans made in Norway by branches of Danish and Swedish banks operating there.

9. The authorities also introduced measures specifically targeted at containing mortgage credit growth (also see Appendix I). They include a maximum loan-to-value (LTV) ratio, an affordability test, and an amortization requirement for loans with an LTV ratio above 70 percent. These measures had been introduced as guidelines in March 2010 and then converted into regulations in mid-2015. The LTV ratio guideline had been set at 90 percent in March 2010, before being lowered to 85 percent in December 2011, and then made a stricter requirement at the 85 percent level in mid-2015. The affordability test requires that borrowers have the capacity to service the debt in the event of a 5 percentage point increase in interest rates, since more than 90 percent of mortgages are variable rate loans. The last requirement is that annual amortization payments worth 2.5 percent of the principal be made on loans where the LTV ratio is above 70 percent.

Heatmap of Macroprudential Policy Measures in Nordic Countries

article image
Source: Fund staff.Note: Green indicates implemented; yellow indicates legislated or planned, but not yet fully implemented; red indicates no measure currently planned.

10. However, there is some leeway in the implementation of these measures. In particular, the LTV ratio requirement can be satisfied with additional collateral or “other guarantees,” which means a loan can de facto be larger than 85 percent of the house it’s being used to purchase. Also, up to 10 percent of the value of new lending each quarter does not have to satisfy these conditions.

11. Norway’s macroprudential policy toolkit is one of the most developed amongst the Nordics, but in some areas it is less ambitious. Currently at 85 percent, the LTV limit is relatively high compared with peers. In the absence of a standard debt-to-income (DTI) or debt service-to-income (DSTI) ratio limit, a financial accelerator mechanism can lead to a positive two-way feedback between credit growth and house price inflation due to the procyclicality of LTV limits, which allow lending to grow more quickly as house price increases accelerate. In addition, the new liquidity coverage ratio (LCR) rule adopted late last year does not include a separate krone LCR requirement. Similar to Norway’s banks, Sweden’s banks rely on substantial foreign currency wholesale funding, so the Swedish authorities have implemented additional LCR requirements for different currencies (i.e., for krone, dollars, and euros separately). However, implementing such a measure in Norway is complicated by the fact that currently there are not enough domestic currency denominated high quality liquid assets for Norwegian banks to satisfy a krone LCR of 100 percent.

A03ufig7

Limits on LTV Ratios

(Percent)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Fund staff calculations.

12. Structural policies can complement macroprudential policies. As noted above tax deductibility of mortgage interest is a key structural factor contributing to high household debt in Norway. Finland provides an example of a country that is gradually eliminating the tax deductibility of mortgage interest, reducing the share of the mortgage interest that can be deducted at the capital income tax rate by 10 percent per year until 2019. This Finnish experience could provide a useful case study for other Nordic countries considering a similar reduction. Additionally, relaxing planning and building restrictions could reduce supply constraints and house price growth.

C. Effects of Macroprudential Tools on Credit and House Price Growth

13. Evidence from other countries and cross-country studies suggest that macroprudential tools have been effective in containing credit and housing booms. Several studies have found that an increase of risk weights on a specific targeted segment of consumer loans was effective in limiting growth of that type of loans in Australia (Bank of England, 2014) and Brazil (IMF, 2013b), as well as in cross-country evidence (Arregui and others, 2013). In addition, a number of studies have found that a tightening of LTV and DTI (or DSTI) ratios is associated with a decline in mortgage lending growth, thereby reducing the risk of the emergence of a housing bubble.4 However, there is less evidence on the simultaneous use of these tools in a specific country setting, and so far no studies have focused on the impact of these macroprudential tools specifically for Norway.

14. To gauge the impact of existing macroprudential measures, we estimate an empirical model with two separate equations for mortgage credit and house prices. A careful assessment of the effectiveness of macroprudential measures requires controlling for the economic environment in which they were taken. While the measures may not have led to an observable significant slowdown in house prices and credit growth, they may have been successful in preventing an even stronger increase. Following Krznar and Morsink (2014), we assess the effectiveness of housing-related macroprudential tools controlling for other factors using two separate equations for mortgage credit and house prices:

Yt=α+βXt+γPt+ɛt

The empirical model relates the year-on-year growth rates of mortgage credit or house price (Yt) to a matrix of control variables (Xt, either current or lagged), and a vector of macroprudential policy tools that have been put in place (in the mortgage credit equations, Pt). The mortgage credit equation includes mortgage rate, unemployment rate, house price growth, a global risk aversion variable proxied by the VIX index, consumer confidence in house purchase, and oil price growth. In addition, several variables on the supply side are included, such as construction output growth and annual change in banks’ funding cost. In the house price equation we include: (i) demand factors, such as mortgage credit growth, population growth, consumer confidence in house purchase, and income growth; (ii) supply factor, such as the growth rates of number of completed houses and transfer of dwellings; and (iii) a measure of market tightness, i.e., turnover time.

A03ufig8

Mortgage Credit and Real House Price Growth

(Year on year percent change)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Statistics Norway and Fund staff calculations.

15. The mortgage credit equation includes measures of two macroprudential policy tools. There are no macroprudential variables in the house price equation since it is assumed that macroprudential measures affect house prices indirectly through the mortgage credit. The two existing policy instruments added to the baseline specification to assess their impact are mortgage risks weights on the supply side and LTV limits on the demand side.5 Information on banks’ planned changes in LTV limits (with lags) applied to customers from the Norges Bank Lending Survey is used as proxy for the LTV policy instrument given that: (i) the date the measure was put in place may deviate from the actual time of implementation given that the LTV limits existed as a guideline until July 2015; (ii) macroprudential measures can affect credit growth with delays.6 All the variables are (or after interpolation) at monthly frequency in a sample from 2007M8 to end-2015 (see Appendix II for definition and sources of variables).7 To correct potential serial correlation and heteroscedasticity in the error terms, the Newey-West estimator is used.

16. The estimation results suggest a significant role of mortgage risk weights and LTV limits, among other factors, in shaping developments in mortgage credit (Table 1). Responses of mortgage credit growth to control variables mostly behave as expected (column (1)). The significance of most coefficients and the fit of the model improve after the inclusion of policy instruments (columns (2) and (3)). In particular, higher mortgage interest rates, banks’ funding costs, and mortgage risk weights are associated with slower mortgage credit growth. A 10 percentage point increase in mortgage risk weights is estimated to significantly reduce mortgage credit growth by 2.4 percentage points. Tightening of LTV limits start to have a statistically significant dampening impact on mortgage growth only after several months. A ten point change in the net balance measure of banks’ tightening of their LTV limits over the next 3 months would reduce mortgage credit growth by 0.6 percentage points. The estimated parameters are used to calculate counterfactuals for mortgage credit growth without changes in mortgage risk weights or LTV limits, which show that credit growth would have been substantially higher recently without the rise in risk weights.

Table 1.

Mortgage Credit Growth and Macroprudential Tools

article image
Source: Fund staff calculations.Note: Dependent variable is mortgage credit growth (year-on-year). Significance at 1, 5, and 10 percent levels indicated by ***, **, and *, respectively. Newey-West consistent variance estimator is used to calculate the coeffcients’ standard errors, which are reported in brackets below the coefficient estimates.
A03ufig9

Mortgage Credit Growth, with and without Policy Changes

(Percent)

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Statistics Norway and Fund staff calculations.

17. Through their impact on mortgage credit growth, macroprudential instruments can also affect how quickly house prices grow (Table 2). Mortgage credit growth has a positive and significant impact on the growth of house prices. Hence, tightening a macroprudential instrument such that it slows mortgage credit growth would also translate into lower house price inflation. For instance, a 10 percentage point increase in mortgage risk weights would reduce house price growth by 1 percentage point. In addition, the negative and significant coefficient on construction output growth suggests that measures to increase the supply of housing, such as streamlining the development process, can help to rein in house price growth.

Table 2.

House Price Growth

article image
Source: Fund staff calculationsNote. Dependent variable is real house price growth (year-on-year). Significance at 1, 5, and 10 percent levels indicated by ***, **, and *, respectively. Newey-West consistent variance estimator is used to calculate the coeffcients’ standard errors, which are reported in brackets below the coefficient estimates.

D. Potential Impact of Tightening Macroprudential Policy

18. Given data limitations, we also examine the potential impact of tightening macroprudential policy using a version of the DSGE model from Chen and Columba (2016). The model is (roughly) calibrated to the Norwegian economy. Values for most standard parameters are taken from the Norges Bank’s NEMO model (Brubakk and others, 2006). Other parameters are calibrated such that the steady state of the model approximates key moments of the data (averaged over 10 years), including the household debt to disposable income ratio, which is the primary variable we focus on in the analysis of the impact of macroprudential policies. Appendix III contains details of key calibrated parameters and steady state ratios. Even though the model has been roughly calibrated to the Norwegian economy, the results should be taken as illustrative.

19. Policy changes in the LTV ratio cap, amortization requirement, and mortgage interest tax deductibility are examined. For the LTV limit, we look at the impact of a 5 percentage point reduction in the maximum LTV ratio (with the reduction occurring over 3 years). For tightening the amortization requirements, we increase the required amortization of the mortgage (over the course of 4 years) such that it reduces the maturity of new mortgages by 5 years. For the tax deductibility of mortgage interest, we model reducing the tax rate at which mortgage interest can be deducted by half (over the course of 10 years), which entails a 14 percentage point reduction in the tax rate.8

20. While tightening all of the macroprudential tools lowers the debt-to-income (DTI) ratio, there are differences in the size and timing of the impact. The upper chart in Figure 2 illustrates the impact on households’ DTI ratio from tightening the different instruments after 5 years and in the new steady state. The tightening of both the LTV cap and amortization requirement causes a reduction in the DTI of nearly 20 percentage points in the new steady state. However, the tightening of the LTV cap achieves almost half of the steady state impact on the DTI ratio in the first 5 years, while the tightening of the amortization requirement achieves less than one-third of the steady state impact in the first 5 years. Reducing the tax deductibility of mortgage interest by half over 10 years ends up having less of an impact on the steady state DTI ratio, at just over 11 percentage points, than tightening the LTV cap or amortization requirement. However, the reduction in tax deductibility achieves almost all of its steady state impact in the first 5 years. This is because households adjust their borrowing behavior to be consistent with the effective debt service (including any savings from interest tax deductibility) they will face with the eventual reduction in tax deductibility even before that reduction is fully phased in.

Figure 2.
Figure 2.

Impact of Tightening Macroprudential Instruments

Citation: IMF Staff Country Reports 2016, 215; 10.5089/9781498345705.002.A003

Sources: Fund staff calculations.

21. One concern about tightening macroprudential policy is that it will have a negative impact on households’ consumption, but the results indicate any impact is small. As a measure of the impact on the economy of tightening macroprudential policies, we look at the effects on households’ (goods) consumption. Similar to the analysis of the impact on the DTI ratio, we measure the impact on households’ consumption after 5 years and in the new steady state. As the lower chart in Figure 2 illustrates, besides the amortization requirement after 5 years, for all of the instruments there is actually a small positive effect on consumption after tightening them.9 Of course, this result is dependent on the various assumptions of the model and should be considered with caution. However, it does indicate that tightening macroprudential policies will not necessarily have significant negative effects on economy in the medium-term, especially if the tightening is gradual.

E. Policy Implications and Conclusions

22. Systemic risks from overvalued house prices and high household debt levels suggest that macroprudential policy measures should be tightened further. Both the empirical results in this paper and international experience suggest that tightening LTV limits and risk weights can slow mortgage credit and house price growth. Steps thus far to do so have been welcome, but there may be scope for gradually tightening these instruments further, including amortization requirements. At minimum, given divergent house price developments across regions, it may be useful to tighten these instruments for mortgages in certain regions (e.g., the Oslo area). Adding other macroprudential instruments would complement the current set of tools. In particular, LTV limits tend to be procyclical and this could be addressed by adding a debt-to-income (DTI) cap or a debt service-to-income (DSTI) limit to supplement the current affordability test.

Addressing structural factors contributing to high household debt and house prices would reinforce the impact of macroprudential policy measures. As illustrated above, gradually reducing the tax deductibility of mortgage interest can lower households’ indebtedness. Moreover, increasing the supply of housing by relaxing restrictions on development can slow the pace of house price growth. Even regardless of the financial cycle and risks, these structural factors should be addressed to help reduce distortions in financial and real investment decisions.

Appendix I. Details of Macroprudential Policies in Nordic Countries

Details of Macroprudential Policies in Nordic Countries

article image
Sources: National authorities, IMF NRR (2013), Sweden Article IV, Denmark FSSA 2014, Jin, Lenain and Brien (2014) and Norges Bank Financial Stability Reports 2014 and 2015.

Loan growth: Growth in lending to the individual customer segments must be below 15% per year. The four customer segments are private homeowners, residential rental properties, farms and other commercial The borrower’s interest rate risk: The proportion of loans where the loan-to-value (LTV) exceeds 75 per cent. the lending limit, and where interest rates are only locked for up to 2 years, must be less than 25 per cent. Applies only loans to individuals and loans for rental housing. There may be waived loans with cover in the form of interest rate swaps and the like.

Repayment Freedom of loans to private: The share of interest-only loans in LTV band above 75 per cent. the lending limit must not exceed 10 per cent. of the total loan volume. Grace loans count regardless of location in priority order.

Loans with short funding: The share of loans refinanced will by. quarter be less than 12.5 per cent. of the total loan portfolio and a year less than 25 per cent. of the loan portfolio. Large exposures: The sum of the 20 largest exposures should be less than the Institute’s actual core capital.

It is further clarified that banks are not covered by the supervisory diamond to mortgage banks. However, there will be launched reports for the banks’ home loans so that FSA can monitor progress and intervene if the risky loans that supervisory diamond shall limit, move a large proportion of the banks.

Appendix II. Data Description for Empirical Analysis

Variable Definitions and Sources

article image

Appendix III. Key DSGE Model Policy Parameters and Ratios

Key DSGE Model Policy Parameters and Ratios

article image
Sources: Chen and Columba (2016); Fund staff calculations.Note: GDP used in the table is mainland GDP. Parameterization of the model aimed for steady state ratios similar to ratios in the data averaged over 10 years (in most cases). Policy parameters were set to resemeble those of actual policies for the LTV and tax rate parameters, while the maximum maturity on new mortgages (which implies a minimum amortization rate) was set to help achieve the targeted steady state household debt-to-disposable income ratio. The full set of parameters used to calibrate the Chen and Columba (2016) DSGE model to Norway are available upon request.

References

  • Ahuja, Ashvin and Malhar Nabar, 2011, “Safeguarding Banks and Containing Property Booms: Cross-Country Evidence on Macroprudential Policies and Lessons from Hong Kong SAR,IMF Working Paper, WP/11/284 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Arregui, N., J. Benes, I. Krznar, S. Mitra, and A. Oliveira Santos, 2013, “Evaluating the Net Benefits of Macroprudential Policy: A Cookbook,IMF Working Paper, WP/13/167 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Cerutti, Eugenio, Stijn Claessens, and Luc Laeven, 2015, “The Use of Effectiveness of Macroprudential Policies: New Evidence,IMF Working Paper, WP/15/61 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Chen, Jiaqian, and Francesco Columba, 2016, “Macroprudential and Monetary Policy Interactions in a DSGE Model for Sweden,IMF Working Paper, 16/74 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Crowe, Christopher, Giovanni Dell’Ariccia, Deniz Igan, and Pau Rabanal, 2013, “How to Deal with Real Estate Booms: Lessons from Country Experiences,Journal of Financial Stability, No. 9, pp. 300319.

    • Search Google Scholar
    • Export Citation
  • Bank of England, 2014, “The Financial Policy Committee’s Power to Supplement Capital Requirements,Policy Statement (January) (London: Bank of England).

    • Search Google Scholar
    • Export Citation
  • Igan, Deniz and Heedon Kang, 2011, “Do Loan-to-Value and Debt-to-Income Limits Work? Evidence from Korea,IMF Working Paper, WP/11/297 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • International Monetary Fund (IMF), 2013a, Norway 2013 Selected Issues, IMF Country Report No. 13/273, 2013.

  • International Monetary Fund (IMF), 2013b, “Key Aspects of Macroprudential Policy—Background Paper,IMF Policy Paper (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Krznar, Ivo and James Morsink, 2014, “With Great Power Comes Great Responsibility: Macroprudential Tools at Work in Canada,IMF Working Paper, WP/14/83 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Kuttner, Kenneth N. and Ilhyock Shim, 2013, “Can Non-interest Rate Policies Stabilise Housing Markets? Evidence from a Panel of 57 Economies,BIS Working Papers, No. 433 (Basel: Bank for International Settlements).

    • Search Google Scholar
    • Export Citation
  • Lim, Cheng Hoon, Francesco Columba, Alejo Costa, Piyabha Kongsamut, Akira Otani, Mustafa Saiyid, Torsten Wezel, Xiaoyong Wu, 2011, “Macroprudential Policy: What Instruments and How Are They Used? Lessons from Country Experiences,IMF Working Paper, WP/11/238 (Washington: International Monetary Fund).

    • Search Google Scholar
    • Export Citation
  • Norges Bank, 2013, “Financial Stability Report,Norges Bank.

  • Reserve Bank of New Zealand, 2014, “C30 New Residential Mortgage Lending: Loan-to-Valuation Ratio (LVR),Statistics (Wellington: Reserve Bank of New Zealand).

    • Search Google Scholar
    • Export Citation
  • Wong, Eric, Tom Fong, Ka-fai Li and Henry Choi, 2011, “Loan-to-Value Ratio as a Macroprudential Tools—Hong-Kong’s Experience and Cross-Country Evidence,HKMA Working Paper, No. 01/2011 (Hong Kong: Hong Kong Monetary Authority).

    • Search Google Scholar
    • Export Citation
1

Prepared by Nathaniel Arnold and Nan Geng. We would like to thank Jiaqian Chen for assistance with the calibration of the DSGE model based on Chen and Columba (2016).

2

Though it is a standard measure of overvaluation, the price-to-rent ratio may be a less useful measure in Norway’s case, where the rental market is small, and overstate the degree of overvaluation.

3

The gap between the regulatory capital ratio and simple leverage ratio would have been wider without the transitional rule. Under the transitional rule in Basel II, an IRB bank’s total risk-weighted assets could not be lower than a given percentage rate of what is would have been under Basel I. The limit was 95 percent in 2007, 90 percent in 2008, and 80 percent since 2009. See 2013 Financial Stability Report by the Norges Bank for more details.

4

These studies include individual country case studies (e.g., Igan and Kang (2011) on Korea, Wong et al. (2011) on Hong Kong, Crowe and others (2013) on U.S., RBNZ (2014) on New Zealand, Krznar and Morsink (2014) on Canada) as well as cross-country studies (e.g. Ahuja and Nabar (2011), Lim and others (2011), Kuttner and Shim (2013), Cerutti and others (2015)).

5

Unlike Krznar and Morsink (2014), we do not use dummies for the months following implementation of macroprudential measures, since most of the recent measures were only made binding in mid-2015.

6

Negative net percentage balances for maximum LTV ratio denote tighter credit standards.

7

The sample is constrained by the availability of data on banks’ funding costs, mortgage risk weights, and the Norges Bank Lending Survey.

8

While the tax deductibility of mortgage interest is technically not a macroprudential tool, it influences how much households borrow, so it is interesting to compare changes in tax deductibility to other macroprudential instruments.

9

Though the shorter run dynamics differ slightly for each instrument, the main factors behind this positive impact on (goods) consumption are that the tighter macroprudential policy limits (i) increase the relative cost of borrowing to consume housing versus the cost of consuming goods, (ii) lower debt levels lead to lower debt service costs in the steady state, allowing for higher consumption. In the new steady state for reduced tax deductibility, borrowers in the model consume less housing, but more goods, while savers also consume slightly more as the revenue to the government from reducing mortgage interest tax deductibility is redistributed as transfers.

Norway: Selected Issues
Author: International Monetary Fund. European Dept.