Norway: Selected Issues

Selected Issues

Abstract

Selected Issues

House Prices and Labor Mobility in Norway: A Regional Perspective1

A. Introduction

1. House prices in Norway have been growing rapidly in recent years. As of May 2018, nationwide house prices were 55 percent higher than in 2010. The national house price to income ratio remains historically and internationally high. Although house prices fell in 20172—particularly in Oslo, which saw nominal house price declines of 10.5 percent—the correction was short lived. House prices rose again by 7.5 percent during January to May of 2018 on a seasonally-adjusted basis.

uA03fig01

Norway: House Price Indicators

(1995=100)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Sources: OECD, Haver Analytics.
uA03fig02

Real House Prices By Region 1/

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Source: Haver Analytics.Notes: 1/ per square meter in thousand NOK deflated by CPI;2/ Rogaland and Hordaland

2. There has been a significant regional divergence of house price trends since 2013. Real house prices in Oslo now stand 60 above their 2010 level—compared to 35 percent for the whole of Norway. House prices in Oslo have been increasing particularly quickly compared to other regions since 2013. This represents a contrast to the last period of rapid house price appreciation—before the global financial crisis—when house prices grew evenly across Norway.

3. Large differences in house prices across regions can have macroeconomic implications. There is growing evidence that large house price differentials can limit regional labor mobility, thus slowing income and productivity convergence (Ganong and Shoag, 2015; Hsieh and Moretti, 2017). House price differentials—to the extent they translate into higher household debt and debt servicing costs—can make some local economies more sensitive to abrupt house price corrections than the others, thus providing arguments in favor of region-specific rather than nation-wide policies to mitigate financial vulnerabilities.

4. In this analysis we estimate the extent to which the recent regional house price divergence in Norway can be explained by fundamental factors. Section B looks at the recent trends in regional house prices, demand and supply factors more in detail. Section C describes our econometric approach to estimating regional equilibrium house prices, and provides main findings on the extent of house price over- or under-valuation across Norwegian regions. Section D studies the impact of house price differentials on labor mobility in Norway. Section E concludes.

B. Regional House Price Developments in Recent Years

5. In recent years, house prices have been increasing rapidly in Oslo, and growing at a moderate pace in other regions. House prices in Oslo have been growing at a fast rate since 2013, with a real appreciation of over 20 percent in 2016 alone. While real prices in the capital declined by 11 percent between March and December 2017, they picked up strongly again in the first half of 2018, and as of May 2018 they were again above their average 2016 level. In comparison, in the oil regions house prices are still below the levels observed before the 2014 oil price bust, and in the rest of the country the average real annual house price growth between 2013 and 2017 has been only 2.5 percent.

uA03fig03

Norway: Real House Price Growth by Region

(Y/y change, in percent)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Sources: Haver Analytics.
uA03fig04

Real House Prices in Oslo

(January 2012=100)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Sources: Haver Analytics, Real Estate Norway.

6. The Oslo correction happened not long after new mortgage regulations entered into force in January 2017. The new measures included: (i) a debt-to-income (DTI) limit of five; (ii) tightened conditions for applying an amortization requirement; and (iii) a lower limit for the maximum percentage of new mortgage lending in Oslo to deviate from one or more of regulatory requirements. There is evidence that the regulations, especially the DTI limit, have been more binding in Oslo than in the rest of the country. More generally, staffs analysis suggests that macroprudential tools targeted at the housing market in Norway have contributed to improving the composition of household credit, and have had a dampening impact on growth in household credit and house prices.3

uA03fig05

Loan-to-Value Ratio and Debt-to-Income on Repayment Mortgages in Norway

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Source: Finanstilsynet 2017 residential mortgage survey

7. Several factors have likely contributed to strong house price growth in recent years, but not all can potentially account for the regional divergence in price developments:

  • Population growth outpacing residential construction. Over the last decade Oslo has experienced much stronger population growth than the rest of the country. At the same time, the response of housing supply has been very sluggish, and only very recently has there been a considerable increase in the number of new house starts in Oslo. In comparison, in the rest of the country housing supply growth has been more aligned with population changes. Nevertheless, housing supply in the capital region has not been able to keep up with population growth since the mid-2000s. This raises the question why slow supply would lead to price increases only in the last three to four years.

  • Oil shock. The rapid decline in oil prices in 2014 has had an impact on the whole Norwegian economy, but the regions with a larger dependence on oil, such as Rogaland and Hordaland, have been hit much more—translating into slowdowns or declines in house prices in these areas, relative to the rest of the country.

  • Low interest rates. A gradual reduction of the Norges Bank’s policy rate since late 2014, and low global interest rates have led to a considerable decline in mortgage rates in Norway in recent years. However, unless borrowing costs have not declined evenly across regions, lower mortgage rates cannot explain the faster house price growth in Oslo than in other regions.

  • Preferential property taxes. Property tax rates differ across municipalities in Norway, and several municipalities do not impose property taxes at all. In Oslo, the property tax was introduced only in 2016, and it is levied on a relatively small share of the properties (the most expensive ones). Despite local differentiation, property taxes are overall relatively low (the maximum rate is 0.7 percent of a property’s value) and the national tax system is very generous to mortgage-takers.

uA03fig06

Household Real Disposable Income by Region

(2005=100, per person)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Sources: Statistics Norway.
uA03fig07

Population and Residential Building Stocks in Regions

(2000=100)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Sources: Statistics Norway, IMF staff calculations.

8. Differences in supply and demand factors across regions may not be large enough to fully explain recent house price dynamics. In particular, a rapid pace of house price growth in some areas raises questions whether fundamental factors alone can explain the growing differences across regions. To quantify the impact of fundamental factors on house prices and the extent of potential regional overvaluations, in Section C we construct an econometric model and use it to estimate equilibrium house prices at the regional level.

C. Estimation of Regional House Price Overvaluation

9. To estimate potential overvaluation of regional house prices in Norway, we use a two-stage approach. The existing literature has primarily focused on detecting national house price overvaluations, as sufficiently long time series of data are necessary to model the long term relationship between house prices and fundamental factors4 (see Box 1 for an overview of tests of asset price overvaluations). In comparison, house prices and key explanatory variables are often not available at the regional level, or regional data have only few observations. This is also the case in Norway, where time series of many regional variables start from early or mid-2000s, and are available only on annual basis. To overcome the short sample issue, we thus apply a two-stage approach, where we first estimate national equilibrium house prices in Norway. The national equilibrium prices then inform the regional (second stage) regression, which is specified in terms of deviations of variables from their national-level averages.

Tests for House Price Overvaluations: Literature Overview

Methods used to identify deviations of house prices from their equilibrium levels (or “bubbles”) can be divided into three groups:

  • Cointegrating equations and error correction models. House prices (or house price to income/rent ratios) and the fundamental factors, such as housing stock and borrowing costs, are assumed to co-move closely over the long term. Bubbles are then identified as short-term deviations from the estimated long-term relationship (Meen 2001; Ambrose et al. 2013; Geng 2018).

  • Econometric tests of time series of asset prices. The time series of house prices are tested for the presence of bounded variance, and for stationarity. The overvaluations are identified as periods during which house prices present explosive or non-stationary behavior. Similarly, the existence of the cointegrating (error-correcting) relationship between house prices and fundamental factors can be tested within separate periods. See Gurkaynak (2005) for an overview of the time series tests.

Focusing on Norway, Anundsen (2016) considers a mix of cointegrating-equations methods, and econometric time series tests to analyze the behavior of house prices. He finds no evidence of house price overvaluation in Norway at the national level as of 2016, consistent with our analysis (see below).

10. In the first step, we estimate the national equilibrium house prices for Norway using approach in Geng (2018). In her model, real house prices are a function of a range of fundamental and policy-related factors (Box 2). The regression is estimated using data for 20 advanced economies between 1990:Q3–2016:Q4—which should be sufficiently long to properly identify a long term cointegrating relationship. We estimate national equilibrium house prices in Norway5 as the fitted values from two specifications in Geng (2018): one that includes only the fundamental factors commonly used in the literature, i.e. household real income and wealth, building stock, and interest rate (model 1); and one that has a larger number of significant explanatory variables, and that includes policy measures (model 2). Residuals from the two models are then identified as deviations of actual house prices from the equilibrium values, i.e. over- or under-valuations.

11. The cross-country panel regressions suggest that at the national level house prices were moderately overvalued in 2017. Both models imply a real overvaluation of house prices (of 10–20 percent) in the periods prior to the Nordic banking crisis in the early 1990s and before the global financial crisis. The models also show that real house prices were around 10 percent above their equilibrium levels in the early 2000s. For 2017, model 1 suggests a real overvaluation of national house prices of 19 percent, while model 2 suggests that prices were broadly in line with fundamentals in 2017 on average.

uA03fig08

Model-Implied Real House Price Overvaluation in Norway

(In percent, national level)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Sources: Geng (2018), staff calculations.

Estimating National Equilibrium House Prices in Norway

In Geng (2018), equilibrium real house prices are a function of fundamental and policy factors. The model is estimated on a panel of 20 advanced economies, over the period 1990:Q1–2016:Q4, with country fixed effects.

article image
*** p<0.01, ** p<0.05, * p<0.1

Coefficient estimates from the two model specifications applied to our analysis are presented in Table 1. In model 1 only core fundamental factors commonly used in other studies—real household income and wealth per capita, building stock, and mortgage rate—are included as explanatory variables. In model 2, the per capita income, net financial wealth, and the mortgage rate are also interacted with country-specific elasticity of housing supply with respect to house prices (s in Table 1). This allows for capturing the variation in responses of house prices to changes in the three variables across the countries—proportional to the responsiveness of national housing supply. Additionally, two policy variables: a measure of severity of rent controls and of generosity of the tax system towards mortgage debt are interacted with demand and supply variables to allow for more variation in coefficients across countries.

Table 1.

Norway: Estimation of Regional Equilibrium House Prices in Norway: Results

article image
*** p<0.01 , ** p<0.05, * p<0.1Notes: All explanatory variables are in percentage deviations from the national averages. Specifications (1) and (2) use the percentage deviation of regional prices from the national equilibrium price derived in model 1 and in model 2, respectively. The reported versions of the two specifications differ with respect to the number of explanatory variables. The inclusion of the squared stock of dwellings per capita in the specifications in columns 2 and 5 captures a non-linear effect of new dwellings on house prices; without it the housing supply is not significant. The model-implied regional house price overvaluations presented in the main text are derived using specifications without the level and the square of stock of dwellings per capita, but the results are very similar when using model specifications in columns 2 and 5, i.e. with the stock of dwellings. Results are robust to using levels of explanatory variables instead of percentage deviations, while including both time- and county- fixed effects.

A potential drawback of the panel regressions is that the coefficient on the mortgage rate is relatively small for Norway. In comparison, other Norway-specific studies find an elasticity of house prices with respect to the borrowing rate of around (or above) 10 percent (e.g. Anundsen, 2016). The rationale is that the share of flexible-rate mortgages in Norway (over 90 percent) is very high compared to other countries, making house prices particularly sensitive to changes in the interest rates. Our estimate is much closer to the ones found in other cross-country regressions (see Turk, 2015 for a literature overview).

12. Model 2 is our preferred specification, but to capture model uncertainty, we report results from both specifications. In both models, coefficients are estimated with high precision. Model 1 captures variables for which the relationship with house prices can be derived from a theoretical life-cycle model (Meen, 2001), but model 2 is our preferred specification as the overvaluation time series for model 1 (model residuals) are non-stationary over the sample period. However, to account for model uncertainty, we use results from both specifications to derive regional equilibrium house prices.

13. In the second stage, we specify the regional house price regression in terms of deviations from the national equilibrium prices. To estimate the regional equilibrium house prices, we regress deviations of regional house prices from the national equilibrium price—derived in stage one—on deviations of regional fundamental factors from their national averages. In other words, in the regional regression we estimate the extent to which regional deviations of house prices from the national equilibrium can be explained by the differences in fundamental factors across regions. Formally:

pjtdev=αj+β*Xjtdev+ϵjt,

where pjtdev is a percentage deviation of the house price in region j from the national equilibrium price pt*,andXjtdev is a vector of explanatory variables specified in terms of percentage deviations from the national averages in period t. The fitted values from the regional regression are then added to the national equilibrium prices from stage one. This provide us with the estimates of regional equilibrium house prices over time:

pjt*=pt**(1+pjtdev^)

We use annual data for 19 Norwegian counties between 2005–2016 and estimate a panel regression with county fixed effects. We take regional house price data from Real Estate Norway. As explanatory variables, we use regional registered unemployment, population aged 20–50 years, residential building stock, real income, public housing, and property taxes from Statistics Norway. To correct for the impact of serial correlation and cross-section dependence of error terms, we use Discoll-Kraay standard errors (Table 1).

14. For 2017, the results suggest a real house price overvaluation of 10–20 percent in Oslo, no overvaluation in the oil regions, and a mild overvaluation of 5–10 percent the rest of Norway. House prices in Oslo are estimated to be overvalued by 11 percent in 2017 when applying national equilibrium prices from model 2, while those in oil-dependent regions are estimated to be somewhat undervalued. House prices in other regions seem well explained by fundamentals. Using estimates from model 1 as national equilibrium prices yields much higher overvaluations across all regions in 2017:6 from around 30 percent in Oslo, to 15–20 percent in the rest of the country. Given the higher weight we put on the results from model 2, and considering the size of confidence bands around the estimates,7 we estimate the house price overvaluation in Oslo at around 10–20 percent in 2017. For the rest of the country, the results suggest that house prices are well aligned with fundamentals in the oil regions, and we estimate the house price overvaluation to be around 5–10 percent in the non-oil, non-Oslo Norway.

uA03fig09

Regional House Price Overvaluation Implied by Model 1

(In percent)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

uA03fig10

Regional House Price Overvaluation Implied by Model 2

(In percent)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

15. The two-stage approach is based on some important assumptions, and thus the results should be interpreted with caution. First, an important implicit assumption is that variables used in national-level regression and not available at the regional level (such as mortgage rate or financial wealth) do not vary significantly across regions. Otherwise, the regional regression will suffer from omitted variables bias. Second, in the absence of regional consumer price indices (CPI), we use national price indicators to obtain real values—while it is possible that prices of the same goods differ across regions.

uA03fig11

Model 2: Overvaluation in Oslo with 95 Percent Confidence Band

(In percent)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

D. Regional House Price Differentials and Labor Mobility

16. Evidence from other countries suggests that large house price differentials can have a significant impact on internal migrations. Barriers to labor mobility may reduce its effectiveness as an adjustment mechanism, through which employment, income, and productivity converge across regions. Papers studying labor mobility in the UK and the US find that: (i) housing regulations tend to be more strict in high-income areas (Hilber and Robert-Nicoud, 2013); (ii) housing regulations tend to lead to higher house prices (Hilber and Vermeulen, 2015); and (iii) strict housing restrictions and rising house prices work as a barrier to interregional migration of low-skilled workers and regional income convergence (Ganong and Shoag, 2015; Arregui and Górnicka, 2018), negatively affecting national GDP (Hsieh and Moretti, 2017).

17. Internal mobility in Norway is relatively high by international standards. Annual regional migration flows have oscillated around 2.5 percent of the total population in recent years, which is high compared to other European countries (close to 1 percent of national populations in EU-15 countries on average), and comparable to the levels observed in the US.8 During the recent global financial crisis internal migrations were highly correlated with the differences in unemployment rates across regions, helping to mitigate the impact of the recession.

uA03fig12

Norway: Internal Emigration by Region

(In percent of local population)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Source: Statistics Norway.
uA03fig13

Norway: Internal Emigration and Unemployment Rates in Counties in 2008–2010

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Source: Statistics Norway.

18. However, there is evidence of increasing outflows of prime-age cohorts from Oslo. Outflows of 30–49-year-olds from agglomeration centers are common, as prime age cohorts start families and move further from the city centers in search for bigger and more affordable dwellings. However, the net outflow of prime age cohorts from the capital—mostly to the surrounding regions—has been increasing over the last decade, raising questions about the connection with the increasing house price differentials between Oslo and the rest of the country.

uA03fig14

Net Internal Migration of Prime Age Cohorts by Region

(In percent of population 30–49 years old)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Source: Statistics Norway
uA03fig15

Internal Migration Flows to and out of Oslo

(Number of persons)

Citation: IMF Staff Country Reports 2018, 280; 10.5089/9781484377109.002.A003

Source: Statistics Norway.

19. We estimate the impact of house price differentials on internal migrations between 19 Norwegian counties. We use annual data between 2005–2016 and focus on the migrations of 30–49-year-olds, for whom house prices are an important factor when deciding on the place to live. For younger cohorts, education choices and locations of universities are often the most important reasons for moving. In the absence of data on bilateral migration flows between counties by age, we use the total net internal migration of persons aged 30–49 years (in percent of a county’s population of the same age cohort) as the dependent variable:

netmigrationjt=αj+β1ujt1dev+β2wjt1dev+β3hpjt1dev+β4popjt1dev+ϵjt,

where ujt1dev,wjt1dev,hpjt1dev,popjt1dev stand for the unemployment rate, real labor compensation per person, real house price, and total population in county; in period t, defined in terms of deviations from their cross-county averages. Unemployment and wages capture the economic dimension of internal migrations, population—the structural trend of increasing urbanization (migrations to the cities). All dependent variables are lagged to address potential endogeneity issues. If house prices are a barrier to regional mobility, one would expect the coefficient on the house prices to be negative: a higher real house price relative to the national average should reduce the net migration to county j.

20. Results show a statistically significant, although moderate in magnitude, impact of house prices on internal migrations in Norway. The coefficient on house prices is negative and significant across specifications, and is robust to excluding Oslo and Akershus from the sample (Table 2). It implies that—in the case of Oslo—a 25 percent increase of house prices above national average increases the net outflow of the prime age cohort by almost 10 percent. This magnitude is comparable to estimates obtained by other studies for the case of the U.K.9

Table 2.

Norway: Estimation of Internal Migrations of Prime Age Cohorts in Norway: Results

article image
*** p<0.01, ** p<0.05, * p<0.1Notes: All explanatory variables are in percentage deviations from the national averages. The dependent variable is the total net internal migration of persons aged 30–49 years by county in percent of the local population of the same age cohort. We use prices of detached houses from Statistics Norway to measure house price differentials. Specifications (1) and (2) include time-invariant variables: the Oslo dummy, average employment in non-services sector and distance from Oppland—the most central county (to capture remoteness of a county—which should affect net migration negatively). Specifications (3)–(7) include county fixed effects. In models (2) and (4) real income is included as an alternative measure of compensation. Neither compensation nor income are significant – likely reflecting very small variability of each along time and across regions in Norway. Model (6) is estimated excluding Oslo and the neighboring county—Akershus. Specification (7) includes lagged dependent variable to correct for autocorrelation of residuals. Results are robust to using price of flats, house price to income ratio, and to using net migration of 30–59 years old cohort as the dependent variable. The coefficient on house prices is not significant when using an aggregate house price index (which captures different types of dwellings)—suggesting a segmentation of the housing market within the prime age cohort.

E. Conclusions

21. There is evidence that differences in house prices across regions in Norway exceed the levels implied by the variation in fundamental factors. In general, large deviations of house prices above the equilibrium make them more vulnerable to significant corrections. In this context, we find that in 2017 real house prices in Oslo exceeded the equilibrium levels implied by empirical models by around 10–20 percent. While there was a considerable correction in the Oslo market in the second half of 2017, prices picked up strongly again in the first half of 2018, and as of May 2018 they are again above their average 2016 level. In the rest of the country, house prices seem well aligned with fundamentals in the oil regions, and we estimate the house price overvaluation to be around 5–10 percent in the non-oil, non-Oslo Norway in 2017.

22. We find evidence of a statistically significant, although quantitatively moderate effect of regional house price differentials on internal migrations in Norway. Consistent with evidence from other countries, large house price differences are found to be a factor preventing migrations across Norwegian counties. While internal migration in Norway is high by international standards, continued house price divergence across regions can potentially contribute to weakening income and productivity convergence across regions going forward.

References

  • Ambrose B., P. Eichholtz, and T. Lindenthal, 2013, “House Prices and Fundamentals: 355 Years of Evidence,” Journal of Money, Credit and Banking, Vol. 45, No. 2–3, pp. 477491.

    • Search Google Scholar
    • Export Citation
  • Anundsen A., 2016, “Detecting Imbalances in House Prices: What Goes Up Must Come Down?”, Norges Bank Working Paper, 11/2016.

  • Arregui, N. and L. Górnicka, 2018, “Regional Disparities in Labor Productivity in the U. K.,” IMF United Kingdom Selected Issues Papers, 2018.

    • Search Google Scholar
    • Export Citation
  • Ganong P. and D. Shoag, 2015, “Why Has Regional Income Convergence in the U.S. Declined?Journal of Urban Economics, Vol. 102, pp. 7690.

    • Search Google Scholar
    • Export Citation
  • Geng, N., 2018, “Fundamental Drivers of House Prices in Advanced Economies,” IMF Working Paper 18/164.

  • Gurkaynak, R., 2008, “Econometric Tests of Asset Price Bubbles: Taking Stock,” Journal of Economic Surveys, Vol. 22, No. 1, pp. 166186.

    • Search Google Scholar
    • Export Citation
  • Hilber, C. and W. Vermeulen, 2015, “The Impact of Supply Constraints on House Prices in England,” Vol. 126 (591), pp. 358405.

  • Hilber, C., and F. Robert-Nicoud, 2013, “On the Origins of Land Use Regulations: Theory and Evidence from U.S. Metro Areas,” Journal of Urban Economics, Vol. 75(C), pp. 2943.

    • Search Google Scholar
    • Export Citation
  • Hsieh, Ch. and E. Moretti, 2017, “Housing Constraints and Spatial Misallocation,” NBER Working Paper.

  • Meen, G., 2001, “The Time-Series Behavior of House Prices: A Transatlantic Divide?”, Journal of Housing Economics, Vol.11 (1), pp. 123.

    • Search Google Scholar
    • Export Citation
  • Murphy, A., Muellbauer, J., and G. Cameron, 2006, “Housing Market Dynamics and Regional Migration in Britain,” Economics Series Working Papers 275, University of Oxford.

    • Search Google Scholar
    • Export Citation
  • Turk, R., 2015, “Housing Price and Household Debt Interactions in Sweden,” IMF Working Paper 15/276.

1

Prepared by Lucyna Górnicka and Yuanyan Sophia Zhang. We would like to thank the Real Estate Norway for providing us with data on regional house prices, and Nan Geng for sharing cross-country data and results of her analysis. We thank staff at the Norges Bank, as well as Francesca Caselli, Jacques Miniane, and Mico Mrkaic (all IMF) for useful comments.

2

Note that the annual average of real house prices in 2017 was nonetheless 10 percent higher than the average observed during 2016 for two reasons: (i) house price increases cumulated during 2016, reducing that year’s average figure and (ii) the correction in 2017 mostly occurred in the second half of the year, thereby not pulling down the 2017 annual average by that much.

3

IMF (2018), “Macroprudential Policies and Housing Prices,” EUR Departmental Paper, forthcoming.

4

For another example of a regional analysis of house prices see e.g. Ho, G. (2016), “The Great Divergence: Regional House Prices in Denmark,” Selected Issues Paper.

5

We make an out-of-sample prediction for 2017.

6

This is consistent with an average overvaluation of 19 percent at the national level implied by the model 1.

7

For Oslo, the 95 percent confidence band around the estimate of overvaluation is 6–18 percent for 2017 when using model 2, and 29–39 percent when using model 1.

8

The Economist: America settles down, July 5th, 2012.

9

For example, Murphy et al. (2006) finds that a 25 percent increase in house prices in Greater London would result in an increase of the population outflow by around 20 percent.

Norway: Selected Issues
Author: International Monetary Fund. European Dept.
  • View in gallery

    Norway: House Price Indicators

    (1995=100)

  • View in gallery

    Real House Prices By Region 1/

  • View in gallery

    Norway: Real House Price Growth by Region

    (Y/y change, in percent)

  • View in gallery

    Real House Prices in Oslo

    (January 2012=100)

  • View in gallery

    Loan-to-Value Ratio and Debt-to-Income on Repayment Mortgages in Norway

  • View in gallery

    Household Real Disposable Income by Region

    (2005=100, per person)

  • View in gallery

    Population and Residential Building Stocks in Regions

    (2000=100)

  • View in gallery

    Model-Implied Real House Price Overvaluation in Norway

    (In percent, national level)

  • View in gallery

    Regional House Price Overvaluation Implied by Model 1

    (In percent)

  • View in gallery

    Regional House Price Overvaluation Implied by Model 2

    (In percent)

  • View in gallery

    Model 2: Overvaluation in Oslo with 95 Percent Confidence Band

    (In percent)

  • View in gallery

    Norway: Internal Emigration by Region

    (In percent of local population)

  • View in gallery

    Norway: Internal Emigration and Unemployment Rates in Counties in 2008–2010

  • View in gallery

    Net Internal Migration of Prime Age Cohorts by Region

    (In percent of population 30–49 years old)

  • View in gallery

    Internal Migration Flows to and out of Oslo

    (Number of persons)