## 1luxea2022002

## Source details

**Canonical URL:** [1luxea2022002](https://www.imf.org/-/media/files/publications/cr/2022/english/1luxea2022002.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/cr/2022/english/1luxea2022002.pdf.md)
- [Structured JSON version](/-/media/files/publications/cr/2022/english/1luxea2022002.pdf.json)

---

### Housing Prices Dynamics
- On average, housing prices grew by about 5 percent per annum in real terms in the last 30 years, three times faster than the average euro area and OECD country.
- Population grew by more than 60 percent, mostly driven by foreign labor force.
- Housing prices accelerated markedly: after hovering around 5½ percent over 2010–18, housing prices have been growing by double digits since 2019.
- Rental price dynamics:
  - Growth of rents has not been commensurate with housing prices overall.
  - Rents in Luxembourg have been growing faster compared to major cities in Europe; Luxembourg City has become one of the most expensive in Europe.
  - BCL analysis: one additional year of tenancy decreases the monthly rent by around 20 euros per 100 square meters.

### Drivers of the Boom in Housing Prices
- Demand-side drivers:
  - Shift in labor force toward high skilled workers and well-paid jobs, increasing demand for higher quality homes.
  - Sociodemographic changes: rising female employment/participation rates and higher share of single-person/single-parent households.
  - Demand surge concentrated in higher income quartiles: share of households intending to buy/build homes rose significantly for the 3rd and 4th income quartiles during the pandemic; bottom quartile demand trended down.
  - Evidence of increased buy-to-let investments (increase in rental listings).
  - Fiscal policy contributions:
    - Public sector jobs account for more than 20 percent of total employment and more than 40 percent of resident employment.
    - Contribution of public wages to growth of compensation of employees exceeded 50 percent during 2018–20.
    - 2017 tax reform abolished taxable rental value, increased mortgage interest deduction for main residence, and increased threshold for deducting home purchase savings contributions.
    - 2015 change to VAT super reduced rate on newly constructed home: ceiling on deductibility replaced ceiling linked to value.
    - Abolition in 2017 of the temporary budgetary compensation tax of 0.5 percent and introduction of several tax credits reduced the tax wedge.
  - Low interest rates:
    - Historically low mortgage rates increased borrowing capacity and expected returns.
    - Acceleration in housing prices accompanied by increase of credit growth.
    - Loan-to-value and share of fixed rate mortgages increased gradually, reducing interest rate risk.
    - Macroprudential authority (National Systemic Risk Committee, CdRS) recommended activation of borrower- and capital-based measures.
- Supply-side constraints:
  - New constructions and construction permits increased until 2017 but stock of housing per capita continued to fall as population growth outpaced dwellings growth.
  - In 2018–19, new dwellings fell even as construction permits kept rising.
  - Supply of new dwellings concentrated in Luxembourg canton and its neighborhood.
  - Stock of social and affordable housing in public hands remains below 2 percent.
  - Construction sector capacity constraints:
    - Decoupling between production (activity) and orders (demand); widening gap associated with rising job vacancy rates.
    - Pandemic exacerbation: 60 percent of firms consider labor shortages a major obstacle to growth; construction costs increased by about 9 percent in 2021H2.
    - Issues: aging labor force, reduced interest in construction jobs (90 percent of employment in sector are cross-border workers and foreigners), dwindling wage differential with neighboring countries, lack of affordable housing, competition from public sector.
  - Land hoarding and concentration:
    - Temporary reduction of taxes on capital gains on immovable assets between July 1, 2016 and December 31, 2018 incentivized land banking.
    - Land prices increased on average +12½ percent over 2018–20.
    - Cost of land estimated to account for 20-42 percent of housing construction sales value (maximum in Luxembourg-Ville).
  - Productivity hurdles in construction:
    - Limited productivity gains; long procedures to obtain permits and lack of availability of landfills increase builders’ costs.

### Potential Impact of Housing Prices
- Affordability and inequality:
  - Share of overburdened households doubled in the last decade to over 10 percent.
  - Housing cost burden (Eurostat definition) has been steadily increasing as prices and rents grew faster than average income.
  - New renters (leases < 5 years) tend to spend 40 percent of disposable income on housing.
  - New homeowners (ownership < 2 years) tend to spend 35 percent of disposable income on housing.
  - Overburden rates higher for foreigners, singles, and single parent households.
  - Despite housing assistance and tax incentives, homeownership rates have declined.
- Competitiveness and labor supply:
  - High housing prices have pushed many workers to look for housing across borders.
  - Cross-border balancing role may weaken due to higher private transportation costs, longer commuting times, and rising neighboring region housing prices.
  - Rising vacancies observed in sectors with low or negative wage differential with neighboring countries (construction and industry).
  - Continued fast housing price growth could reduce ability to attract and retain skilled workers and exert upward pressure on wages, hampering competitiveness.
- Financial stability risks:
  - Households’ debt-to-income doubled in the last 2 decades.
  - Exposures across domestically-oriented banks to real estate sector (developers and construction firms) increased.
  - Staff estimates housing price overvaluation to about 15 percent in 2021Q2 (note large uncertainty; ECB assessment points to an overvaluation of 51 percent).
  - Short-term likelihood of a sharp correction is rather low; medium-term risks remain, mitigated by effective oversight, resilient and highly capitalized banking sector, and high household wealth.

### Authorities’ Response and Policy Proposals (Housing Pact 2.0 and follow-ups)
- Strategy pillars:
  - Boost housing supply: mobilize vacant dwellings and unused land; build more, denser, faster, and at lower cost; increase share of affordable homes.
  - Contain demand pressure and reduce geographic concentration.
  - Increase residential mobility.
  - Reduce under-occupation.
  - Improve effectiveness and coordination of implementation of reforms.
- Measures to increase supply of affordable housing:
  - Collective housing subsidies for construction of affordable housing in Housing Pact 2.0: €350 million since 2019.
  - Land purchases for building reserves purposes.
  - Draft law on ‘Baulandvertrag’ (introduced November 2020):
    - Any plot newly designated as “constructible land” must be developed for housing purposes within a maximum of four years from designation.
    - If owner fails to comply, plot to be used for construction of dwellings of public interest, such as social housing.
- Macroprudential recommendations by CdRS to activate borrower- and capital-based measures to address risks from easy financing conditions and credit growth.

### Affordable housing law (drafts and features tabled December 2021; draft tax law to be tabled by end-2022)
- Affordable and low-cost housing continuum (draft affordable housing law):
  - Affordable rental housing:
    - 50 percent of households in    in  municipalities are eligible for the allocation of this type of housing.
    - Rent calculated according to the concept of the affordable lease, considering the income level of the tenant.
  - Social rental housing:
    - Dwellings often owned privately that fall under the social rental housing management regime and have not benefited from housing subsidies.
    - Only management of the dwellings can be eligible for a financial contribution if the organization has an agreement with the Housing Ministry.
  - Affordable housing sale:
    - Implemented only by public promoters; regime comparable to "subsidized sale" with long lease and right of repurchase.
    - Planned eligibility limits so that 60 percent of communities will have access.
  - Low-cost housing sale:
    - Implemented by public promoters; regime comparable to 'unsubsidized sale' regime.
    - Ceiling designed to make 70 percent of communities eligible for acquisition.
  - Public promoters:
    - Two public promoters ‘Fonds de Logement’ and ‘Société Nationale des Habitations à Bon Marché’ have ramped up activity significantly and plan to maintain the pace.
- Property taxation and taxes on unused or empty dwellings/plots (draft law):
  - Reassessment of the value of each plot of land based on a new formula with reviewed and updated data (current valuation method based on figures from 1941).
  - An additional tax on empty and non-developed plots located in a designated construction zone.
  - Use of recalculated property values as a base for a revised property tax.
  - Housing tax flanking the property tax: houses or flats not used for the owner’s own purposes nor rented will be subject to an additional tax.
- Individual housing allowance law (demand-side bill tabled):
  - Revision of eligibility criteria to allow more people to benefit.
  - Increase of the amounts and ceilings to reduce the effort rate of beneficiaries.
  - Administrative simplification and digitalization of the system.

### IMF Staff Policy Recommendations and Priorities
- Overall emphasis:
  - Efforts to boost housing supply and taxation of unused land and empty dwellings are welcome but will take time and may face implementation challenges.
  - Multi-pronged approach needed: contain demand pressure; mobilize and use existing resources more efficiently; address capacity constraints (labor and land hoarding) and productivity hurdles.
- Demand-side recommendations:
  - Consider tightening macroprudential policy.
  - Prudent recruitment and wage policy in the public sector.
  - Reduce expected returns via property taxation reform.
  - Better target housing assistance.
  - Specific macroprudential suggestions:
    - Introduce debt-to-income or debt-service to income limits and concentration limits on real estate loans for banks.
    - Complement LTV limits and capital-based measures (existing measures include a positive rate of 0.5 percent for the Countercyclical Capital Buffer and a risk-weight floor of 15 percent on mortgage loans granted by internal rating-based banks to the purchase of residential property in Luxembourg).
    - Accompany tightening with compensatory fiscal support targeted to low-income households.
  - Better targeting of help-to-buy policies and tax expenditures:
    - Consider phasing out deductibility of interest payments for the main residence.
    - Make the Bellegen Akt (credit on transfer tax) more progressive and available only to first-time homebuyers.
    - Reduce expected returns to steer demand away from real estate investments (e.g., increasing transfer tax for secondary homes, limit exemptions on capital gains for secondary homes, revisit accelerated depreciation schemes).
  - Fiscal policy considerations:
    - Mindful calibration of public employment, public wages, taxes, and social benefits due to potential unintended consequences for housing prices.
  - Reduce geographic concentration:
    - Spatial planning to promote activities outside current clusters and close to borders.
    - Promote remote work and further develop transport infrastructure.
- Supply-side recommendations:
  - Reform property taxation to increase residential mobility and reduce under occupation:
    - Adopt a simplified framework for property taxation and regularly update house valuations.
    - Lower transaction costs for buyers replacing their main residence.
    - Limit eligibility for exemptions from capital gains tax on inherited properties to 1–2 years after the inheritance.
    - Link tax incentives on rental properties to affordability of rent to increase affordable rental supply.
    - Consider incentives to enhance residential mobility of older people (expand housing options).
  - Mobilize existing resources and increase density:
    - Taxation of unused land to reduce land hoarding and faster mobilization of public land.
    - Construct higher, ecologically sustainable buildings with greater density; scale up “density bonuses” near public transportation.
    - Adapt housing supply to evolving needs and lifestyles (e.g., micro-units, co-living).
    - Municipalities to group Baulucken in private hands to construct larger buildings.
  - Address supply bottlenecks and increase productivity in construction sector:
    - Streamline procedures for permits (establish a one-stop shop for construction permits and a platform to track land under development).
    - Improve availability of landfills and promote circular construction.
    - Promote offsite/modular construction and investment in digitalization (less than 1 percent of new buildings are prefabricated).
    - Increase investment in R&D in partnership with private sector to advance innovative construction techniques and materials.
    - Address labor and skills shortages:
      - Short term: incentives for existing workers (especially aging ones) to stay longer, train unemployed with necessary pre-existing skills, promote reallocation across sectors, limit competition from the public sector.
      - Medium term: attract and retain new workers, provide new training opportunities (including in future skills) to enhance career prospects and wages in the sector.

### Implementation, Monitoring, and Data Needs
- Municipal implementation challenges:
  - Principal-agent problems: local voter welfare often misaligned with national welfare; social housing beneficiaries often nonvoters.
  - Municipalities have low financial incentives to increase housing supply due to the bigger role of municipal business tax compared to property taxes and high costs of additional infrastructure.
  - Housing Pact 2.0 conditions subsidies to construction of social and affordable housing—a step in the right direction.
  - Increase of property taxation could enhance implementation if accompanied by mechanisms to enable public capture of increased land value due to new infrastructure and amenities.
  - Establish coordination mechanisms, define clear goals and responsibilities, and strengthen accountability.
- Data needs:
  - Register of vacant properties is a start.
  - Improve timeliness and availability of data on stock and flows of dwellings (rental properties, finished dwellings, main residence/secondary, etc.).
  - Track land under development and commercialization of dwellings.
  - Collect data about residential mobility (including cross-borders) and commercial real estate (prices, vacancy rates).
  - More granular data on construction sector (including real estate developers) to analyze construction costs and mark-ups and market structure.
  - Data on beneficiaries of tax incentives and housing assistance (by income level) to assess fiscal incidence.

### Income Inequality and Labor Market — Trends and Model Setup
- Trends in inequality and income distribution:
  - Luxembourg’s Gini index (after taxes and social transfers) has been growing faster and has surpassed that of the euro area countries since 2018.
  - Measurement caveat: potential measurement issues due to Luxembourg’s small population and high reliance on foreign labor.
- Key labor market developments (historic patterns to 2019):
  - Participation rate: increased by just under 10 percentage points from the early-2000s to 2019.
  - Unemployment rate: increased from a low of just over 2 percent in the early-2000s to just under 6 percent in the pre-pandemic period.
  - Foreign commuters: the ratio of foreign commuters to national employment has increased by around 30 percentage points since the early-2000s.
  - Age and skill composition shifts: share of older workers (age 46 to 65) increased by around 5 percentage points since the mid-2000s; high-skill share of the population increased by around 20 percentage points since the mid-2000s.
- Structural model features and calibration:
  - Small open economy with heterogeneous workers who can work domestically or commute to a foreign labor market; foreign workers can commute into Luxembourg.
  - Firms employ low-, medium-, or high-skill technologies; workers differ by ability and demographics.
  - Selected calibration moments include exact reported figures (e.g., Unemployment Rate: 2.1% early period; 5.5%–5.6% late period; Foreign Commuters / Active: 46% early period; 70%–72% late period; Participation Rate: 64%–74% across periods; Population Skill Share arrays preserved in source).

### Model Results, Scenarios, and Policy Simulations
- Drivers of past increase in inequality:
  - (i) Technology costs (decreases in cost of medium-skill x_M and high-skill x_H technologies) — primary driver and over-explains the increase in inequality.
  - (ii) Dispersion of worker ability μ — explains a negligible share.
  - (iii) Average ability of age groups ν_a — implies a decline in income inequality but explains distributional winners and losers.
- Forward-looking digitalization scenario:
  - Assumption: cost of medium-skill technology decreases by a further 84 percent; cost of high-skill technology decreases by a further 58 percent. Returns on young and older workers remain the same.
  - Predicted impacts: high-skill share increases; unemployment declines overall; effects on inequality ambiguous, with simulation implying a slightly lower Gini coefficient.
- Policy simulations (three schemes):
  - a. Re-skilling of older workers:
    - Implementation: increase the idiosyncratic component u_i such that u_i′ = u_i 1.3.
    - Results: reallocates older workers from low to medium-skill; lowers inequality modestly; increases unemployment rate of medium-skill workers.
  - b. Support for medium-skill occupations:
    - Implementation: direct subsidy increasing after-tax revenue of medium-skill match y′(s,m) = 1.1 * y(s,m).
    - Results: large increase in share of medium-skill workers; decreases inequality notably; decreases unemployment for medium-skill workers and older age groups; downside: may discourage movement to high-skill occupations.
  - c. Improved worker-firm matching:
    - Implementation: increase matching efficiency M′ = 1.05 * M.
    - Results: no effect on skill distribution; limited effect on income inequality; uniformly positive impact on unemployment across age and skill groups.
- Tradeoffs and recommendations:
  - Combine broad-based policies (improved matching) with targeted policies (medium-skill support or re-skilling) to address inequality while managing distortion risks.
  - Consider public sector employment effects carefully due to potential competitiveness impacts.
  - Design re-skilling and ALMPs mindful of program effectiveness, calibration, and potential distortions; prioritize consolidation and support for effective programs.

### Key Findings and Conclusions
- Housing price dynamics and drivers:
  - Luxembourg’s housing prices have grown rapidly, with affordability concerns and medium-term risks to attractiveness for workers and financial stability.
  - Surge driven by constrained supply (labor shortages and land hoarding) and high demand—mostly the top 50 percent and buy-to-rent transactions.
  - Contributing policies: low property/land taxation, increased untargeted support to homeownership, large recruitments in the public sector, and lower personal income taxation.
  - Low-for-long interest rate environment reinforced demand for housing and land.
- Policy priorities:
  - Multi-pronged approach required: contain demand pressure; use existing resources more efficiently; build more and faster with lower cost; mobilize vacant properties; increase residential mobility and reduce under occupation; increase affordable housing supply by public and private sectors.
  - Implementation is key: housing policy is cross-cutting across social, environmental, fiscal, financial and development domains and involves multiple government entities; requires coordination mechanisms, clear goals and responsibilities, and strengthened accountability.

*Source: 1luxea2022002 - 1. Summary of Objectives and Policy Proposals; 15. In December 2021, the government followed up with a draft ‘affordable housing law’; IMF Staff calculations and model analysis as presented in the Luxembourg chapter.*

### 1. Summary of Objectives and Policy Proposals ___________________________________________ 12

### 1. Summary of Objectives and Policy Proposals

### A. Housing Prices Dynamics
- On average, housing prices grew by about 5 percent per annum in real terms in the last 30 years, three times faster than the average euro area and OECD country.
- Population grew by more than 60 percent, mostly driven by foreign labor force.
- Housing prices accelerated markedly: after hovering around 5½ percent over 2010–18, housing prices have been growing by double digits since 2019.
- Rental price dynamics:
  - Growth of rents has not been commensurate with housing prices overall.
  - Rents in Luxembourg have been growing faster compared to major cities in Europe; Luxembourg City has become one of the most expensive in Europe.
  - BCL analysis: one additional year of tenancy decreases the monthly rent by around 20 euros per 100 square meters (possible explanations: landlords not adapting existing contracts, transaction costs, information asymmetry).

### B. Drivers of the Boom in Housing Prices
- Demand-side drivers:
  - Shift in labor force toward high skilled workers and well-paid jobs, increasing demand for higher quality homes.
  - Sociodemographic changes: rising female employment/participation rates and higher share of single-person/single-parent households.
  - Demand surge concentrated in higher income quartiles: share of households intending to buy/build homes rose significantly for the 3rd and 4th income quartiles during the pandemic; bottom quartile demand trended down.
  - Evidence of increased buy-to-let investments (increase in rental listings).
  - Fiscal policy contributions:
    - Public sector jobs account for more than 20 percent of total employment and more than 40 percent of resident employment.
    - Contribution of public wages to growth of compensation of employees exceeded 50 percent during 2018–20.
    - 2017 tax reform abolished taxable rental value, increased mortgage interest deduction for main residence, and increased threshold for deducting home purchase savings contributions.
    - 2015 change to VAT super reduced rate on newly constructed home: ceiling on deductibility replaced ceiling linked to value.
    - Abolition in 2017 of the temporary budgetary compensation tax of 0.5 percent and introduction of several tax credits reduced the tax wedge.
  - Low interest rates (“low for long”) effects:
    - Historically low mortgage rates increased borrowing capacity and expected returns.
    - Acceleration in housing prices accompanied by increase of credit growth.
    - Loan-to-value and share of fixed rate mortgages increased gradually, reducing interest rate risk.
    - Easy financial conditions boosted developers’ demand for land, raising land prices.
    - Macroprudential authority (National Systemic Risk Committee, CdRS) recommended activation of borrower- and capital-based measures.
- Supply-side constraints:
  - New constructions and construction permits increased until 2017 but stock of housing per capita continued to fall as population growth outpaced dwellings growth.
  - In 2018–19, new dwellings fell even as construction permits kept rising.
  - Supply of new dwellings concentrated in Luxembourg canton and its neighborhood.
  - Stock of social and affordable housing in public hands remains below 2 percent.
  - Construction sector capacity constraints:
    - Decoupling between production (activity) and orders (demand); widening gap associated with rising job vacancy rates.
    - Pandemic exacerbation: 60 percent of firms consider labor shortages a major obstacle to growth; construction costs increased by about 9 percent in 2021H2.
    - Issues: aging labor force, reduced interest in construction jobs (90 percent of employment in sector are cross-border workers and foreigners), dwindling wage differential with neighboring countries, lack of affordable housing, competition from public sector.
  - Land hoarding and concentration:
    - Constructible land abundant but concentrated among a few private owners and real estate developers.
    - Temporary reduction of taxes on capital gains on immovable assets between July 1, 2016 and December 31, 2018 incentivized land banking.
    - Land prices increased on average +12½ percent over 2018–20.
    - Cost of land estimated to account for 20-42 percent of housing construction sales value (maximum in Luxembourg-Ville).
    - Oligopolistic land ownership and hoarding can slow release of homes and shift supply toward high-end dwellings.
  - Productivity hurdles in construction:
    - Limited productivity gains; high share of labor cost and high land prices may have reduced investment in technology and skills.
    - Long procedures to obtain permits (e.g., environmental requirements) and lack of availability of landfills could increase builders’ costs.

### C. Potential Impact of Housing Prices
- Affordability and inequality:
  - Share of overburdened households doubled in the last decade to over 10 percent.
  - Housing cost burden (Eurostat definition) has been steadily increasing as prices and rents grew faster than average income.
  - Income inequality widened due to polarization of labor market; bottom quintiles disproportionately affected.
  - New renters (leases < 5 years) tend to spend 40 percent of disposable income on housing.
  - New homeowners (ownership < 2 years) tend to spend 35 percent of disposable income on housing.
  - Overburden rates higher for foreigners, singles, and single parent households.
  - Despite housing assistance and tax incentives, homeownership rates have declined.
- Competitiveness and labor supply:
  - High housing prices have pushed many workers to look for housing across borders.
  - Cross-border balancing role may weaken due to higher private transportation costs, longer commuting times, and rising neighboring region housing prices.
  - Rising vacancies observed in sectors with low or negative wage differential with neighboring countries (construction and industry).
  - Continued fast housing price growth could reduce ability to attract and retain skilled workers and exert upward pressure on wages, hampering competitiveness.
- Financial stability risks:
  - Households’ debt-to-income doubled in the last 2 decades.
  - Exposures across domestically-oriented banks to real estate sector (developers and construction firms) increased.
  - Staff estimates housing price overvaluation to about 15 percent in 2021Q2 (note large uncertainty; ECB assessment points to an overvaluation of 51 percent).
  - Short-term likelihood of a sharp correction is rather low; medium-term risks remain, mitigated by effective oversight, resilient and highly capitalized banking sector, and high household wealth.

### D. Authorities’ Response and Policy Proposals
- Government strategy built on Housing Pact 2.0 and a broader multi-pronged approach to reduce housing market imbalances:
  - Strategy pillars:
    - Boost housing supply: mobilize vacant dwellings and unused land; build more, denser, faster, and at lower cost; increase share of affordable homes.
    - Contain demand pressure and reduce geographic concentration.
    - Increase residential mobility.
    - Reduce under-occupation.
    - Improve effectiveness and coordination of implementation of reforms.
- Measures to increase supply of affordable housing:
  - Collective housing subsidies for construction of affordable housing in Housing Pact 2.0: €350 million since 2019.
  - Land purchases for building reserves purposes.
  - Draft law on ‘Baulandvertrag’ (construction plot contract) introduced November 2020:
    - Any plot newly designated as “constructible land” must be developed for housing purposes within a maximum of four years from designation.
    - If owner fails to comply, plot to be used for construction of dwellings of public interest, such as social housing.
- Other policy-relevant observations:
  - Need to address capacity constraints in construction sector, land hoarding, and permit delays to speed up supply response.
  - Macroprudential recommendations by CdRS to activate borrower- and capital-based measures to address risks from easy financing conditions and credit growth.

*Source: 1luxea2022002 - 1. Summary of Objectives and Policy Proposals*

### 15.      In December 2021, the government followed up with a draft  ‘affordable housing law’,

### 15.      In December 2021, the government followed up with a draft  ‘affordable housing law’,

### Affordable and low-cost housing continuum (draft affordable housing law)
- Affordable rental housing
  - 50 percent of households in    in  municipalities are eligible for the allocation of this type of housing.
  - Rent calculated according to the concept of the affordable lease, considering the income level of the tenant.
- Social rental housing
  - Dwellings, often owned privately, that fall under the social rental housing management regime and have not benefited from housing subsidies.
  - Only the management of the dwellings can be eligible for a financial contribution, provided the organization in charge of the management has previously concluded an agreement with the Housing Ministry.
  - These dwellings are not subject to the affordable rent rules.
- Affordable housing sale
  - Implemented only by public promoters.
  - Regime comparable to the current regime of "subsidized sale" with long lease and right of repurchase.
  - Planned eligibility limits so that 60 percent of communities will have access to it.
- Low-cost housing sale
  - Implemented by public promoters.
  - Regime comparable to the current 'unsubsidized sale' regime.
  - Ceiling designed to make 70 percent of communities eligible for acquisition of this type of housing.
- Public promoters
  - Two public promoters ‘Fonds de Logement’ and ‘Société Nationale des Habitations à Bon Marché’ have ramped up activity significantly in the past years and plan to maintain the pace.

### Property taxation and taxes on unused or empty dwellings/plots (draft law to be tabled by end-2022)
- Two components of the draft law:
  - Reassessment of the value of each plot of land based on a new formula with reviewed and updated data (current valuation method based on figures from 1941).
  - An additional tax on empty and non-developed plots located in a designated construction zone.
- Use of recalculated property values as a base for a revised property tax.
- Housing tax flanking the property tax:
  - Houses or flats not used for the owner’s own purposes nor rented will be subject to an additional tax.
- Detailed modalities currently being discussed.

### Individual housing allowance law (demand-side bill tabled)
- Key elements:
  - A revision of the eligibility criteria to allow more people to benefit from the allowances.
  - An increase of the amounts and ceilings to reduce the effort rate of the people who benefit from the allowances.
  - Administrative simplification and digitalization of the system.

### Policy proposals and priorities (IMF staff recommendations)
- Overall emphasis
  - Efforts to boost housing supply and taxation of unused land and empty dwellings are welcome but will take time and may face implementation challenges.
  - Multi-pronged approach needed: contain demand pressure; mobilize and use existing resources more efficiently; address capacity constraints (labor and land hoarding) and productivity hurdles.
- Demand-side recommendations
  - Consider tightening macroprudential policy.
  - Prudent recruitment and wage policy in the public sector.
  - Reduce expected returns via property taxation reform.
  - Better target housing assistance.
  - Specific macroprudential suggestions:
    - Introduce debt-to-income or debt-service to income limits and concentration limits on real estate loans for banks.
    - Complement LTV limits and capital-based measures (existing measures include a positive rate of 0.5 percent for the Countercyclical Capital Buffer and a risk-weight floor of 15 percent on mortgage loans granted by internal rating-based banks to the purchase of residential property in Luxembourg).
    - Accompany tightening with compensatory fiscal support targeted to low-income households.
  - Better targeting of help-to-buy policies and tax expenditures:
    - Consider phasing out deductibility of interest payments for the main residence.
    - Make the Bellegen Akt (credit on transfer tax) more progressive and available only to first-time homebuyers.
    - Reduce expected returns to steer demand away from real estate investments (e.g., increasing transfer tax for secondary homes, limit exemptions on capital gains for secondary homes, revisit accelerated depreciation schemes).
  - Fiscal policy considerations:
    - Mindful calibration of public employment, public wages, taxes, and social benefits due to potential unintended consequences for housing prices.
  - Reduce geographic concentration:
    - Spatial planning to promote activities outside current clusters and close to borders.
    - Promote remote work and further develop transport infrastructure.

- Supply-side recommendations
  - Reform property taxation to increase residential mobility and reduce under occupation:
    - Adopt a simplified framework for property taxation and regularly update house valuations.
    - Lower transaction costs for buyers replacing their main residence.
    - Limit eligibility for exemptions from capital gains tax on inherited properties to 1–2 years after the inheritance.
    - Link tax incentives on rental properties to affordability of rent to increase affordable rental supply.
    - Consider incentives to enhance residential mobility of older people (expand housing options).
  - Mobilize existing resources and increase density:
    - Taxation of unused land to reduce land hoarding and faster mobilization of public land.
    - Construct higher, ecologically sustainable buildings with greater density; scale up “density bonuses” near public transportation.
    - Adapt housing supply to evolving needs and lifestyles (e.g., micro-units, co-living).
    - Mobilize constructible public land more rapidly and focus resources on large projects with big impact.
    - Municipalities to group Baulucken in private hands to construct larger buildings.
  - Address supply bottlenecks and increase productivity in construction sector:
    - Streamline procedures for permits (establish a one-stop shop for construction permits and a platform to track land under development).
    - Improve availability of landfills and promote circular construction.
    - Promote offsite/modular construction and investment in digitalization (less than 1 percent of new buildings are prefabricated).
    - Increase investment in R&D in partnership with private sector to advance innovative construction techniques and materials.
    - Address labor and skills shortages:
      - Short term: incentives for existing workers (especially aging ones) to stay longer, train unemployed with necessary pre-existing skills, promote reallocation across sectors, limit competition from the public sector.
      - Medium term: attract and retain new workers, provide new training opportunities (including in future skills) to enhance career prospects and wages in the sector.

### Implementation, monitoring, and data
- Municipal implementation challenges
  - Principal-agent problems: local voter welfare often misaligned with national welfare; social housing beneficiaries often nonvoters.
  - Municipalities have low financial incentives to increase housing supply due to the bigger role of municipal business tax compared to property taxes and high costs of additional infrastructure.
  - Housing Pact 2.0 conditions subsidies to construction of social and affordable housing—a step in the right direction.
  - Increase of property taxation could enhance implementation if accompanied by mechanisms to enable public capture of increased land value due to new infrastructure and amenities.
  - Diversify public projects to avoid stigma linked to social housing.
  - Establish coordination mechanisms, define clear goals and responsibilities, and strengthen accountability.
- Data needs
  - Step up efforts on data collection: register of vacant properties is a start.
  - Improve timeliness and availability of data on stock and flows of dwellings (rental properties, finished dwellings, main residence/secondary, etc.).
  - Track land under development and commercialization of dwellings.
  - Collect data about residential mobility (including cross-borders) and commercial real estate (prices, vacancy rates).
  - More granular data on construction sector (including real estate developers) to analyze construction costs and mark-ups and market structure.
  - Data on beneficiaries of tax incentives and housing assistance (by income level) to assess fiscal incidence.

### Key findings and conclusions
- Housing price dynamics and drivers
  - Luxembourg’s housing prices have grown rapidly in recent years, with affordability concerns and medium-term risks to attractiveness for workers and financial stability.
  - Magnitude of price growth is among the highest among OECD countries and major cities.
  - Surge driven by constrained supply (labor shortages and land hoarding) and high demand—mostly the top 50 percent and buy-to-rent transactions.
  - Contributing policies: low property/land taxation, increased untargeted support to homeownership, large recruitments in the public sector, and lower personal income taxation.
  - Low-for-long interest rate environment reinforced demand for housing and land.
- Policy priorities
  - Multi-pronged approach required: contain demand pressure; use existing resources more efficiently; build more and faster with lower cost; mobilize vacant properties; increase residential mobility and reduce under occupation; increase affordable housing supply by public and private sectors.
  - Implementation is key: housing policy is cross-cutting across social, environmental, fiscal, financial and development domains and involves multiple government entities; requires coordination mechanisms, clear goals and responsibilities, and strengthened accountability.

*Italic source: 1luxea2022002 - 15.      In December 2021, the government followed up with a draft  ‘affordable housing law’,*

### 1. Income inequality in Luxembourg has been growing relatively fast and is now higher than

### 1luxea2022002 - 1. Income inequality in Luxembourg has been growing relatively fast and is now higher than

### Trends in inequality and income distribution
- Luxembourg’s Gini index (after taxes and social transfers) has been growing faster over the past several years and has surpassed that of the euro area countries since 2018.
- Measurement caveat: there are potentially important measurement issues with the Gini inequality measure given Luxembourg’s small population and high reliance on foreign labor.
- The model interprets increasing inequality as primarily driven by increasing returns to higher skilled workers due to falling technology costs.

### Key labor market developments (historic patterns to 2019)
- Participation rate: increased by just under 10 percentage points from the early-2000s to 2019 (the pre-pandemic period).
- Unemployment rate: increased from a low of just over 2 percent in the early-2000s to just under 6 percent in the pre-pandemic period.
- Long-duration unemployment: the share of job seekers experiencing long duration unemployment has increased gradually.
- Labor market tightness: increase in unemployment has coincided with an increase in labor market tightness, suggesting unemployed workers face more difficulties finding jobs.
- Foreign commuters: the ratio of foreign commuters to national employment has increased by around 30 percentage points since the early-2000s; domestic cross-border commuters have remained relatively flat, accounting for around 5 percent of the active population.
- Age composition:
  - Share of older workers (age 46 to 65) in the population increased by around 5 percentage points since the mid-2000s, almost entirely matching a decrease in prime age workers (age 31 to 45).
  - Share of older job seekers has increased by 10 percentage points (over double the increase in the population), while share of younger job seekers (age 15 to 30) has declined by a similar rate.
- Skill composition:
  - High-skill share of the population increased by around 20 percentage points since the mid-2000s, driven by equal declines in the low- and medium-skill shares.
  - Share of high-skill among job seekers increased by around 10 percentage points but did not keep pace with population shifts.
  - Share of medium-skilled within job seekers has increased despite a decline in the medium-skilled share of workers — suggesting a hollowing out of opportunities for medium-skilled workers.

### Structural model of the Luxembourg labor market — setup and calibration
- Model features:
  - Small open economy with heterogeneous workers who can work domestically or commute to a foreign labor market; foreign workers can commute into Luxembourg.
  - Workers differ by underlying ability determining productivity; firms employ low-, medium-, or high-skill technologies determining returns to ability.
- Calibration strategy:
  - Calibrated to match labor market features for an early period (2000) and late period (2019).
  - Key calibration moments include unemployment rate, average labor market tightness, 80–20 labor income ratio, relative skill shares, participation rates, and commuter ratios.
- Selected calibration moments (Data / Model; Early Period (2000) and Late Period (2019)):
  - Wage Rate: 2/1.00 1.00 1.22 1.21
  - Unemployment Rate: 2.1% 2.1% 5.5% 5.6%
  - Average LM Tightness: 1/ 3/0.30.3 0.5 0.4
  - 80-20 Ratio: 1/ 4/ 4.2 4.3 5.3 5.6
  - Domestic Commuters / Active: 4.7% 4.8% 4.7% 4.7%
  - Foreign Commuters / Active: 46% 46% 70% 72%
  - Participation Rate: 64% 63% 72% 74%
  - Participation Rate Old: 29% 31% 45% 45%
  - Population Skill Share: [38 , 45 , 17] [38 , 44 , 18] [27 , 32 , 41] [27 , 32 , 41]
  - Job Seekers by Age: 1/ [31 , 38 , 31] [37 , 40 , 22] [20 , 37 , 43] [24 , 39 , 37]
- Selected untargeted moments (Data / Model; Early Period (2000) and Late Period (2019)):
  - 95-5 Ratio: 10.6 9.7 17.3 9.3
  - 90-10 Ratio: 6.6 6.6 9.2 7.3
  - 75-25 Ratio: 3.6 4.3 4.5 5.6
  - St. Dev. Wage: -1.8 -1.4
  - Gini Coefficient: 264 132 35
  - Search Duration Share >1 year: 1/34% 2% 44% 21%
  - Unemployment Rate (%) By Age: [1.7 , 1.9 , 2.7] - [2.7 , 4.7 , 8.6]
  - Unemployment Rate (%) By Skill: [2.7 , 2.2 , 0.9] - [9.1 , 6 , 2.9]
- Interpretation of parameter changes between early and late periods:
  - Matching efficiency M declines while firm vacancy posting cost Γ also declines — interpreted as lower ex-ante match rates and lower recruiting costs (internet, online job boards), potentially greater sectoral mismatch.
  - Cost of high-skill technology x_h decreases, increasing returns to high-skill workers (higher wages at the top end).
  - Labor market becomes increasingly favorable to younger workers — interpreted as lower returns to experience or faster skill obsolescence.
  - Participation increases, especially among older workers; growing share of foreign commuters interpreted as increased relative benefits of being active in the domestic labor market.
  - Changes not shared equally: higher-ability workers capture comparatively more benefits.

### Model results and scenarios
- Drivers of past increase in inequality (three channels):
  - (i) Technology costs: decreases in cost of medium-skill x_M and high-skill x_H technologies — the primary driver and over-explains the increase in inequality (implying inequality would be lower if technology costs had not declined).
  - (ii) Dispersion of worker ability μ — explains a negligible share of the increase in inequality.
  - (iii) Average ability of age groups ν_a — implies a decline in income inequality but important for understanding winners and losers (prime age and older workers net losers; younger workers net winners).
- Forward-looking digitalization scenario:
  - Assumption: cost of medium-skill technology decreases by a further 84 percent; cost of high-skill technology decreases by a further 58 percent. Returns on young and older workers remain the same (digitalization increases rate of skill obsolescence).
  - Predicted labor-market impacts:
    - High-skill workers’ share increases; medium-skill share relatively little affected.
    - Unemployment declines overall; prime age workers benefit the most; older workers benefit but not enough to offset their initial unfavorable position.
    - Effects on inequality ambiguous: top-end income increases relative to bottom end, but wider adoption of high-skill technology benefits more workers, leading to a slightly lower Gini coefficient in the simulation.

### Policy simulations (three schemes and implementation in model)
- Overview: three broad policy schemes simulated to combat increasing inequality; each corresponds to distinct active labor market policy categories.
- a. Re-skilling of older workers
  - Implementation: increase the idiosyncratic component 푢푢_푖푖 of worker ability such that 푢푢_푖푖′ =푢푢_푖푖1.3.
  - Interpretation: re-skilling programs (e.g., those by ADEM), on-the-job training, lifelong learning programs, or professional-chamber efforts to develop new skills throughout careers.
- b. Support for medium-skill occupations
  - Implementation: direct subsidy increasing after-tax revenue of a medium-skill match: 푦푦′(푠푠,푚푚) = 1.1∗푦푦(푠푠,푚푚).
  - Interpretation: occupation-specific subsidies, tax benefits, support for firms hiring medium-skill occupations, or increased public hiring of disadvantaged workers.
- c. Improved worker-firm matching
  - Implementation: increase matching efficiency so that 푀푀′ = 1.05∗푀푀.
  - Interpretation: improvements to job boards, job counselling services, apprenticeship programs, language programs, faster regulatory processing for specialty licenses, broader accreditation of skills to help firms identify well-suited employees.

*Source: https://www.imf.org/-/media/files/publications/cr/2022/english/1luxea2022002.pdf*

### 12. The model results suggest that policy responses to increased inequality feature important

### 12. The model results suggest that policy responses to increased inequality feature important tradeoffs that may hinder their effectiveness

### Policy simulations: effects on skill distribution, inequality, and unemployment
- Policies are calibrated to match a qualitatively similar decline in the unemployment rate of older workers as a benchmark.
- Figure 6 assesses impacts on outcomes related to the skill distribution, income inequality, and labor market opportunities.

a. Re-skilling of older workers
- Leads to a reallocation of older workers from low skill to medium-skill occupations.
- Given the relatively low ability of older workers, the policy has little effect on the share of high-skill workers in the economy.
- Lowers inequality by increasing the wages of older, low-skill workers (who have the lowest wages in the economy).
- Effects on inequality are relatively minor because older workers have a relatively low participation rate and therefore do not account for a large share of workers.
- Results in an increase in the unemployment rate of medium-skill workers; this is driven by marginally lower-ability workers being added to medium-skill occupations.

b. Support for medium-skill occupations
- Has a more substantial effect on the skill distribution, causing a large increase in the share of medium-skill workers.
- Decreases inequality, notably the Gini, by allowing some low-skill workers to adopt the medium-skill technology and because a slower adoption of the high-skill technology tends to lower inequality.
- Does not have a major impact on the extreme tails of the income distribution since those workers are employed in either the low- or high-skill occupations.
- Decreases unemployment for medium-skill workers and the older age groups.
- Downside: many workers who would otherwise be employed in high-skill occupations find it beneficial to remain in medium-skill occupations.

c. Improved firm-worker matching
- Has no effect on the skill distribution because it does not affect the value of matched firm-worker pairs.
- Does not have a large effect on income inequality for the same reason.
- Produces a relatively uniformly positive impact on the unemployment rate across age groups and skill levels.
- Older and low-skill workers tend to benefit more in unemployment outcomes, reflecting their initially weaker starting positions.

### Tradeoffs, complementarities, and practical considerations
- Broader policies (e.g., improved matching):
  - Can benefit the largest group of workers.
  - Have little impact on inequality or the skill distribution and may miss the most disadvantaged workers.
- Targeted policies (e.g., medium-skill support):
  - Can directly support disadvantaged groups and lessen inequality.
  - Risk distorting investment or adoption of higher skills (workers may remain in medium-skill occupations instead of moving to high-skill).
- Policies targeting inherent worker characteristics (e.g., re-skilling programs):
  - Avoid the distortion tradeoff.
  - Effectiveness is limited by the size of the targeted worker group (e.g., older workers constitute a smaller share of participants).
- Recommended approach: use a combination of broad-based policies and directed policies to address labor market issues stemming from digitalization and skill-biased changes.

### Additional design factors abstracted from the model (implications for policy design)
- Public sector employment:
  - The public sector is a large employer, especially of domestic workers, but is not directly considered in the model.
  - Public employment could support medium-skill workers or improve opportunities for older, low-skilled workers, but faces similar tradeoffs as medium-skill support schemes (e.g., potentially distorting incentives to invest in skills).
  - Increased public employment could introduce labor-market pressures that weigh on Luxembourg’s international competitiveness.
- Between-sector skills and reallocation:
  - Post-pandemic recovery has produced sectoral imbalances with high vacancies and relatively high unemployment by historical standards.
  - Matching policies would benefit from increased focus on between-sector reallocation.
  - Re-skilling policies should account for skill requirements of growing sectors.
- Job-risk minimization:
  - Policies should be mindful of minimizing job risks from future digitalization and automation.
- Program effectiveness and calibration:
  - Training and re-skilling programs are generally found to be more effective at stimulating employment in the medium term, particularly programs with longer horizons (e.g., lifelong learning).
  - ALMPs (Active Labor Market Programs) tend to have limited effectiveness in some evaluations, and tax benefits or subsidies to private firms may distort hiring incentives (e.g., increased short-duration hires) or crowd out other worker types.
  - Given Luxembourg’s relatively high ALMP spending per unemployed worker and regulatory complexity, consolidation and increased support for effective programs may be more suitable than adding additional programs.

### Model structure highlights (relevant for interpreting results)
- Worker heterogeneity:
  - Workers differ by ability s_i, demographics a_i (young, prime-age, older), employment status, and preferences.
  - Preferences over occupations follow a Frechet distribution with CDF e^{-ε^{-α}}.
- Productivity and technologies:
  - Firms choose among low-, medium-, and high-skill technologies (ℓ, m, h).
  - Technologies parameterized so that σ_h > σ_m = 1 > σ_ℓ = 0 and x_h > x_m ≥ x_ℓ = 0; high-skill technology is more complementary to worker ability and more costly to adopt.
  - Firms pay a constant share 1−β of revenues to workers as wages.
- Labor market matching:
  - Vacancies are created by paying flow cost Γ.
  - Matching rate for a worker of ability s is M u_s η v_s^{1−η}, where M is matching efficiency, u_s unemployed workers of ability s, and v_s vacancies for ability s.
  - Matched pairs are destroyed at rate δ, in which case the firm exits and the worker becomes unemployed.
- Foreign commuters:
  - Foreign economy has L* workers with the same demographic and ability distribution; unemployed workers entering the foreign market have expected utility U*.
- Calibration notes:
  - Demographic returns to labor market captured by ν_a (e.g., returns to experience for older workers or new digital skills for young workers).
  - Idiosyncratic productivity component u_i follows a Pareto distribution with CDF 1−u^{−μ}.

*Italic: Source: IMF Staff calculations and model analysis as presented in the Luxembourg chapter.*

---


_Source: https://www.imf.org/-/media/files/publications/cr/2022/english/1luxea2022002.pdf_