Jon Pareliussen
OECD
OECD Economic Surveys: Sweden 2025
Abstract
Even though the number of housing units per person and the supply response to housing prices in Sweden compare well to its OECD peers, supply shortages persist. Allocative efficiency is low, as tax subsidies to homeowners inflate land prices and rent controls lead to poor matches between housing needs and available units. Productivity in construction is above the OECD average but has flatlined in recent decades. Speeding up zoning and permitting could increase the supply of land while limiting local technical requirements could facilitate standardisation and automation. Improved skills and digitalisation in construction, planning and process management could improve efficiency in both zoning and construction. The cyclicality of construction employment leads to skill losses during downturns which could be addressed by counter-cyclical investments in skills.
Housing plays a pivotal role in people’s lives and takes up a large share of household income while directly impacting personal health and well-being. Construction is a sizeable and volatile share of macroeconomic supply and housing wealth and expenditures shape overall private consumption. The housing market has significant repercussions for the effective functioning of the labour market, for financial and macroeconomic stability and for the environment. The share of income households spend on housing has been on the rise for the past few decades across the OECD and in Sweden. This raises questions about affordability. For example, renters in the lowest income quintile spent 32.5% of their income on housing in 2019 on average in the OECD, and 41.5% in Sweden. Rising housing prices have also contributed to falling fertility rates across the OECD (OECD, 2021[1]; OECD, 2024[2]; OECD, 2025[3]).
The main objectives of housing policy, as described in the OECD’s Agenda for Housing Policy Reform (2024[4]) should be to make housing markets more inclusive, more efficient and more sustainable. The main focus of this chapter is on policy reforms to boost the supply of good-quality affordable housing which is well-matched to demand. Financial stability repercussions of housing are discussed in Chapter 1. Aspects of climate adaptation of housing including zoning, planning and insurance markets are discussed in Chapter 2. Greenhouse gas emissions from housing and commercial buildings are very low in Sweden, as energy use for buildings is mostly based on renewable sources. Energy efficiency requirements for newbuilds are high, but there may nonetheless be room for additional energy efficiency measures as illustrated by a strong demand response to electricity prices during the energy crisis (OECD, 2023[5]).
Sweden’s housing sector suffers from well-known structural imbalances (Table 3.1), notably caused by rent controls and tax subsidies to homeowners driving up housing prices and tying up supply in inefficient uses while faster zoning and permitting in some municipalities would unlock land faster. Cyclicality of housing prices and construction is a challenge to macroeconomic stability and at the same time a challenge to housing construction given long lead times and the loss of skills occurring in downturns.
Various measures have been taken to improve the functioning of the housing market over the past decade or two, but these reforms have shied away from addressing well-known root causes of supply-demand imbalances. As previous OECD Economic Surveys have pointed out, housing policy measures can reinforce each other if implemented within a coherent package of reforms, while measures taken in isolation may be less efficient or even counterproductive (OECD, 2019[6]). For example, well-intended policies like the proposed loosening of macroprudential regulations (Chapter 1) or offering tax relief for homeowners while supply remains rigid will drive up land prices and make housing less affordable in the long term while it will make rental housing even less affordable relative to owner-occupied housing. In contrast, a comprehensive reform programme encompassing housing taxes, rent controls, skills and planning would be able to align more affordable housing supply with demand both in terms of housing types and locations much better than what is the case today. Such a reform package would likely also reduce the need for macroprudential regulations and thereby facilitate their loosening.
The existence of housing shortages in Sweden is well-known and discussed in previous OECD Economic Surveys (OECD, 2019[6]). There is broad agreement that shortages stem from population growth not having been met with sufficient increases in the supply of housing, especially in the three largest cities (Stockholm, Gothenburg and Malmö), but also industrial hotspots in the North (OECD, 2023[5]). The Swedish National Board of Housing, Building and Planning makes annual prognoses of housing construction needs, calculated as a function of the existing housing supply, initial shortages, changes in the number of households and a housing reserve, which is a share of vacancies as people move from one unit to the next. According to the agency, 38 000 to 59 000 new housing units have been added each year between 2015 and 2022, reducing the accumulated supply deficit to 150 000 housing units. Together with a projected increase in the number of households and decommissioning of old housing, 52 000 new units should be added each year over the next decade to fully meet demand. However, in a context of high interest rates and falling house prices only approximately 31 000 were built in 2023 and 2024, and a slightly higher number expected in 2025 (Boverket, 2025[7]).
Macro-level supply shortages in Sweden seem to be slightly less severe than in comparable OECD countries. The number of dwellings per inhabitant is above the OECD average and at similar levels as in Norway, Denmark, Japan and Belgium (Figure 3.1, Panel A), but it should be noted that Sweden’s share of singe-adult households is the highest in the European Union. The housing supply response to strong population growth in the past two decades was initially hampered by the Global Financial Crisis but strengthened considerably since (Panel B). Previous OECD research also indicates that in Sweden housing supply is relatively responsive to shortages as manifested in prices compared to OECD peers (OECD, 2021[1]). The median housing cost as a share of disposable income was 17.9% in Sweden in 2022, below other Nordics (21.4%) and the 18.5% OECD average (OECD, 2025[3]).
Even as housing supply shortages may be less severe than in its Nordic neighbours, Sweden seems to have a higher incidence of unmet housing needs. The number of people living rough or in temporary accommodation is 5.5 out of 10 000 in Sweden, which is in the lower half of OECD countries for which comparable data exist, but considerably higher than in for example Norway (2.6) and Finland (1.6) (OECD, 2025[3]). Even so, homelessness remains relatively marginal with outcomes not only reflecting housing shortages, but also institutional settings within social services, health- and criminal care. Overcrowding is another measure of how the housing market meets the housing needs of people in different situations. The share of people living in overcrowded housing is 16.4% in Sweden, almost double that of Finland (8.8%) and nearly triple that of Norway (6.0%). Moreover, it has been on the rise for the past decade (Eurostat, 2025[8]) even as the ratio of housing per population has increased (Figure 3.1, Panel C) and shortages in terms of housing units per household have decreased.
Source: OECD, Affordable Housing database; Statistics Sweden; and OECD Analytical Database.
Unmet housing needs are related to Sweden’s dual housing market. The rental segment and the owner-occupied segment of the Swedish housing market are in practice segregated by rent controls into two separate markets. Prices are distorted in both the rental and owner-occupied markets, but the distortions are very different in nature and pull in different directions. This creates tensions when the two markets meet in the competition for constructible land and has implications not only for the matching of housing supply to demand and people’s needs, but also for the overall quality and quantity of housing supply.
In the owner-occupied housing market, prices are set by supply and demand but are inflated by the combination of low and regressive recurrent property taxation and generous mortgage interest deductibility. Housing is an asset. In order to avoid distortions compared to other assets it should therefore ideally be taxed equivalently to other assets. It is also a consumption item. To avoid distortions to rental housing and other consumption goods, the consumption value of living in owner-occupied housing should be taxed equivalently to rental housing (so-called “imputed rent” taxation). The practical implementation of such imputed rent taxation would normally happen through recurrent taxation of property at its updated market value with a rate calibrated to equalise taxation of owner-occupied housing to that of rentals. Few OECD countries live up to this ideal, but Sweden is among the countries outright subsidising home ownership through the tax system. Sweden’s property tax revenues are among the lowest in the OECD, and half of the cost of mortgage-financed owner-occupied housing is subsidised through the tax system (Figure 3.2). This is largely due to the centrally set recurrent tax on residential property being charged at a flat and low rate across all housing units and capped at a low level, combined with generous mortgage interest deductibility (OECD, 2023[5]). While the policy has been justified by its perceived positive impact on homeownership, evidence from Sweden and other countries suggests that mortgage interest deductibility primarily drives up property prices and exacerbates wealth inequality, without significantly improving homeownership rates in the long term (Bourassa et al., 2012[9]; Hilber et al., 2014[10]). In Sweden, this tax treatment is regressive as it subsidises those households with the financial capacity to obtain a mortgage, with the subsidy increasing by property value. It likely leads to over-consumption of housing, drives up the price of land and results in significant tax revenue losses.
Note: In Panel B, METR stands for “marginal effective tax rate” for owner-occupied, debt-financed housing investments.
Source: OECD Revenue Statistics database; OECD (2021), Brick by Brick.
A first step of a property tax reform should be to phase out the upper limit of the tax. A second step would be to regularly update tax values of properties, as is for example done annually in Norway. This would enable as a third step, to tax property proportionally to its value and proportionally to rent. To ensure these reforms do not disproportionately affect asset-rich but income-poor households, the government could introduce targeted measures such as tax deferrals for low-income households. There would also be potential benefits in delegating property taxation rights to municipalities, as discussed below. The tax base should consist of regularly updated property values, but municipalities could have some flexibility in setting rates locally and in taxing undeveloped land at a higher rate. Alternatively, phasing out mortgage interest deductibility for owner-occupied housing would achieve some of the same aims as imputed rent taxation, but would tax those with low outstanding mortgages compared to the house value too little.
Changes in housing taxation should be phased in over time to minimise adverse consequences on demand and construction in the transition period. Finland and the Netherlands are examples of countries that have phased down tax incentives in recent times. In both instances reforms were phased in over a long period of time. Finland phased out interest deductions on older loans more gradually than on new loans, allowing homeowners to adapt to the reduced benefit while discouraging excessive borrowing. The Netherlands reduced the transaction tax from 6% to 2% in 2011, first temporarily as a way of stimulating the market, but then from 2012 permanently as a way of facilitating residential mobility, restricted interest relief to mortgages that were fully amortisable in 2013, launched a gradual reduction in the maximum mortgage relief rate by 0.5 percentage point a year until 2040 from a starting point of 52% in 2014. In 2017, the phase-out was accelerated, taking it to 37% in 2024, aligning it with the basic income tax rate. The introduction of a new income tax bracket in 2025 increased the maximum rate again to 37.5%. However, reduced mortgage interest deductions have partially been offset by a reduction in the imputed rent tax rate (OECD, 2025[11]; OECD, 2021[1]).
Rent controls are the strictest in the OECD (Figure 3.3). Rents are in most cases set in negotiations between a landlord and a tenant union, but can also be determined in direct negotiations between the landlord and the tenant. If the parties cannot reach an agreement, the dispute can be referred to a Regional Rent Tribunal. There, a reasonable rent for the apartment will be determined, based on an assessment of the apartment’s so-called “utility value”, which includes amongst other factors the size, standard and technical qualities of the unit (bruksvärdessystemet). The rent-setting regulations apply to all rental units, publicly and privately owned.
Index from 0 (least restrictive) to 1 (most restrictive)
Note: The rent control and tenure security indices range from 0 (no restrictions) to 1 (all types of restrictions) according to answers to the 2019 OECD Questionnaire on Affordable and Social Housing.
Source: OECD (2021), Brick by Brick.
Misallocations related to the system are large. For example, misallocations within the existing housing stock in central Stockholm alone have been estimated to entail an annual welfare loss of 0.5% of GDP (OECD, 2017[12]), and the Swedish Housing Agency has estimated that Sweden would have had 40 000 more housing units without rent controls (Sjöberg and Söderberg, 2013[13]). Finland, which deregulated rents in the midst of the 1990s financial crisis saw a 50% increase in private rental supply over a five-year period following the reform (de Boer and Bitetti, 2014[14]).
Sweden has no official social housing sector based on a long-standing view that the rent-controlled sector should provide affordable quality housing for everyone, but municipalities operate emergency housing. However, rent controls have not made housing affordable for low-income households, with 44% of people in rental housing in the bottom quintile of the income distribution spending more than 40% of their disposable income on rent. This is a considerably higher share than in the OECD on average (36%), in Denmark (34%) or in Norway (37%), similar to Iceland (43%) and lower than in Finland (51%). In contrast, only 8% of low-income owner-occupiers with a mortgage are in the same situation (Figure 3.4). Large apartments in the most attractive housing sub-markets receive the greatest indirect subsidies. For example, the households occupying rental apartments larger than 180 square metres in Stockholm had an average income equivalent to the top 100th income percentile (Fritzon, 2020[15]).
Share of population in the bottom quintile of the income distribution spending more than 40% of disposable income on mortgage and rent, by tenure, 2022 or latest year available
Note: Housing costs cover only those relating to mortgage costs (principal repayment and interest payments) and rental costs (for both private market and subsidised rental housing). See the OECD Affordable Housing database for data on total housing cost burden (which includes mortgage principal and interest repayment, rents, structural insurance, mandatory services and charges, regular maintenance and repair, taxes and utilities (including electricity, water, gas and heating)). Housing costs are considered as a share of household disposable income, which includes social transfers (such as housing allowances) and excludes taxes. In Chile, Colombia, Mexico, and the United States, gross income instead of disposable income is used due to data limitations. In the United Kingdom, net income does not adjust for local council taxes and housing benefits. In Türkiye, net income does not adjust for personal income taxes. Results only shown if category composed of at least 100 observations. Data for Australia, Japan, New Zealand, Switzerland, United Kingdom and the United States refer to 2021, for Hungary, Norway and Türkiye to 2020, for Canada to 2019, for Iceland to 2018, for Israel to 2017.
Source: OECD Affordable Housing database.
The growing disconnect between the value of living in a rental housing unit and the price paid has made rent controls largely untenable as the rental and the owner-occupied market meet in their competition for constructible land. Over time, rental units have been converted to owner-occupied housing cooperatives and construction of new rental units has been slow. Private companies largely focus on owner-occupied housing while rentals are mostly left to municipal housing companies, whose supply is also held back by limited profitability under rent controls (Engerstam, Warsame and Wilhelmsson, 2022[16]; Hermansson and Song, 2024[17]). The imbalances have led to undersupply of rental housing, rationing, a black market for rentals and years-long queues in the main agglomerations. Average queuing time in the Stockholm Region before getting an apartment was for example 8.8 years in 2024, and more than 800 000 people were queuing to get an apartment (35% of the total population in the region) (Bostadsförmedlingen, 2024[18]).
In response to this, new rental housing has since 2006 been allowed to opt for charging so-called “presumption rents”, where the rent level is agreed upon between the landlord and a local tenant association outside of the standard utility value system. The intention is that the rents should cover the cost of construction, including land purchase, and a “reasonable” profit. Presumption rents have brought the new-built and refurbished part of the rental sector closer to a market price. They have contributed to rental supply in the past few years, in the context of low interest rates, generally favourable economic conditions and a now lapsed subsidy programme.
Even though presumption rents are a clear improvement compared to the default rent controls, there is no evidence of any value added compared to market rents justifying such regulation in the first place. Procedures to increase rents are still regulated and rents are thereby subject to litigation. This happened for example in 2022, when decisions to increase rents in line with cost inflation on management and maintenance were struck down by the Svea Court of Appeal (Hovrätten). Presumption rents are valid for 15 years before transitioning to the utility value system. In municipalities where most existing dwellings are older and have significantly lower (regulated) rents, the book value of new buildings subject to presumption rents is often written down immediately after completion to reflect the prevailing rent level. This discourages both municipal and private housing companies from investing. This is particularly problematic in municipalities experiencing major industrial investments or large business establishments where rapid housing expansion is needed to accommodate a growing workforce. This is for example a considerable challenge in Northern Sweden’s, where considerable investments in green industries are on-going and planned, as discussed in the previous OECD Economic Survey of Sweden (2023[5]).
When rents in the existing stock are regulated, the moving chain which would normally be an important mechanism to match supply and demand does not work. Higher-income renters in attractive older rental apartments will stick to their underpriced housing and shy away from more expensive new-builds (Karpestam, 2022[19]; OECD, 2019[6]). Regardless of whether the rents in these new-builds are set in the market or by presumption rents they will tend to be unaffordable to low-income renters. Some low-income renters will manage to get a hold of an apartment in the low-quality end of the existing stock, for which vacancies open from time to time. Others will rent too small apartments for their needs. These continued imbalances are reflected in rental vacancy statistics. There are no immediately available vacancies in the existing stock (under the utility value system) in any of the three main cities of Stockholm, Gothenburg and Malmö, while the very few that exist in neighbouring municipalities are unattractive due to their small size. In the rest of the country, vacancies are mostly found in unattractive areas. There are more vacancies in the new-built (presumption rent) rental housing stock. These apartments are for the most part not rented because of their high rent levels, even though there does not seem to be a systemic issue of these apartments staying empty in the long term (Sveriges Allmännytta, 2025[20]; Hermansson and Song, 2024[17]).
These imbalances are the key reason why overcrowding is more common in Sweden than in comparable countries. It is also why there is a very strong socioeconomic gradient to overcrowding in Sweden, with overcrowding being very common among low-income households relative to high-income households and among the foreign-born (OECD, 2023[5]; OECD, 2019[6]). Overcrowding is directly related to tenure (Figure 3.5). The situation discourages mobility (Karpestam, 2022[19]) and reinforces spatial segregation, the neighbourhood clustering of people of similar socio-economic backgrounds (Hermansson and Song, 2024[17]). People entering the rental market, like immigrants or young people, tend to live in unattractive suburbs with high levels of concentrated welfare dependency and unemployment (Fritzon, 2020[15]).
2023
Note: The overcrowding rate is defined as the percentage of the population living in an overcrowded housing unit. A person is considered as living in an overcrowded housing unit if the household does not have at its disposal a minimum number of rooms equal to: one room for the household; one room per couple in the household; one room for each single person aged 18 or more; one room per pair of single people of the same gender between 12 and 17 years of age; one room for each single person between 12 and 17 years of age and not included in the previous category; one room per pair of children under 12 years of age. 2019 for Iceland.
Source: Eurostat.
Rent controls were implemented during the second world war. The rules on rent setting have been significantly changed since then. The utility value system was introduced in the late 1960s and the Rent Negotiation Act (hyresförhandlingslagen) in the late 1970s. Vested interests in the sector have since built up, notably among long-time tenants who are well-organised in tenant unions. A reform programme taking this complex political economy into account could be implemented by first allowing rents in new-builds to be set freely. Rent controls in the existing stock should then be phased out over a long enough period, for example 10 years, with a clearly defined and communicated phase-out schedule, to give existing tenants time to adjust and thereby reduce resistance to the reform. Finland successfully abolished rent controls without any phase-down period from 1991. Czechia abolished rent controls only in new contracts in 1993 creating a dual rental market favouring insiders (de Boer and Bitetti, 2014[14]), not dissimilar to the current situation in Sweden with presumption rents and default rent controls.
A social housing sector should be developed to provide affordable housing on a means-tested basis. This could build on existing structures and institutions, and could also be used to smooth the transition to market rents by supporting people in the legacy system on a needs-based case-by-case basis. Some principles should be observed to secure a well-functioning social housing sector. Means-testing should happen at regular intervals. If it takes place only at the time of entry, it tends to lower mobility by locking people into their current housing unit. Raising rents to market levels for tenants whose circumstances have improved will free resources and housing units to more economically vulnerable households. A social housing sector should also be designed to promote mixed neighbourhoods and avoid stigma connected to certain areas or buildings. France for example requires that a certain percentage of apartments in new builds are reserved for social housing (OECD, 2021[1]). Another approach could be to provide social housing through the existing municipal housing companies. Unless they are compensated for providing units with rents below market, this would put them at a competitive disadvantage to private companies. Compensating public and privately owned housing companies for providing social housing on a level playing field basis might be a better solution both from an efficiency point of view and in order to achieve more mixed neighbourhoods. Sweden has considerable experience in leveraging the private sector to provide publicly financed services in for example education, labour market policies and social care. Such a model could also apply to social housing. This could potentially also benefit developers, who could collateralise social housing contracts to part-finance construction projects.
Together with mortgage rates and income growth, rigidity of supply is a major factor behind cumulative housing price growth over the past decade (André and Chalaux, 2023[21]). The cost of constructing a housing unit is the sum of the construction cost and the land price. From 2000 to 2021, the total cost of constructing an owner-occupied apartment increased by 182%. The construction price increased by 150% while the price of land increased by 370%. For rental apartments the construction cost increased by 132%, with the cost of construction increasing by 116% and the price of land by 317% (Sveriges Allmännytta, 2023[22]). Overall, the value of land has increased 27-fold since 1980, and is by far the most important driver of increasing housing prices. Construction cost inflation has not been much higher than consumer price inflation during this period (Figure 3.6). Weak competition has also been raised as a price driver, but this does not seem to be a very important factor. There are many companies in most sub-sectors, large companies have lost market share over time and profit margins are low (Lind, 2020[23]). Land price inflation should be tackled from the demand side by taxing property as discussed above, but a balanced solution would also need to address the supply of land through more efficient decision-making on zoning and permitting and possibly through property taxation.
Note: The value of land and dwellings comes from the national wealth statistics (variables AN2111A and AN111 respectively). For construction cost, the construction cost index excluding wage drift for multi-dwelling buildings has been used.
Source: Statistics Sweden.
Municipalities have the full authority of zoning and permitting within their borders. This means that only the municipality has the right, through detailed zoning and planning, to decide on the use of a given piece of land and what can be built there. These decisions determine housing supply by making land available and by regulating height, density and other requirements. Across the OECD, municipalities can release too little land for various reasons, including for example resistance to densification from incumbent property owners, lack of incentives and weak coordination between municipalities in response to regional challenges. Most Swedish municipalities want to grow, but a number of issues may still prevent them from making timely zoning and planning decisions.
Coordination issues between municipalities, regions and the national level, including a cumbersome environmental permitting procedure are complicating infrastructure projects such as roads, railroads and electricity transmission lines. These are also major challenges to the green transition, as discussed in the previous OECD Economic Survey of Sweden (2023[5]). Possible measures to improve the situation range from improved mechanisms and incentives for coordination to municipal mergers and lifting planning authority to a higher level of government for projects of high societal importance straddling municipal borders.
New developments come with needs for infrastructure, schools, health care and other public services, while tax income will only increase once new workers are there and pay municipal income tax. Most municipalities borrow through the municipal financing vehicle Kommuninvest at favourable terms due to its AAA-rating and low risk premium. However, many municipalities are indebted, and higher interest rates combined with the obligation to balance budgets from year to year may prevent them from making such investments in a timely manner and may therefore hold back development. Reforming property taxation by removing the ceiling as discussed above, while delegating the authority to tax property based on its updated market value and the authority to set rates to municipalities would diversify the local tax base and enable and incentivise development. Local autonomy to define the tax rate could be given along various dimensions, including for example by differentiating between commercial real estate, primary and secondary housing. Further autonomy could be considered for climate adaptation purposes (Chapter 3) and for example to incentivise investments in clean electricity generation (OECD, 2023[5]). A higher tax rate on uninhabited land not used for agriculture or forestry could further incentivise the release of land. A local property tax could risk deepening existing fiscal disparities, as property values vary significantly between municipalities. Adjustments in the income equalization system might be needed to balance these differences while preserving local incentives.
Appeals of planning decisions and a lack of resources dedicated to zoning and planning can also lead to delays. Average lead times for zoning and permitting of construction projects increased from 45 months in 2015 to 56 in 2024. Lead times for the full zoning and permitting process vary from around three years in the fastest municipalities to around six in the slowest ones, without apparent systematic differences between large and small municipalities (Figure 3.7). These differences reflect a combination of how much resources municipalities put into zoning and planning in terms of manpower and capital, like investments in digitalisation, and the productivity of these inputs. The latter depends crucially on the skills of the municipal workforce, both individual knowledge and experience and efficiency of process management (Sveriges Allmännytta, 2023[22]).
Average lead times for multi-dwelling buildings for the period 2016-23
Source: Nationell ledtidsindex, 2024.
The planning process is also a determining factor of productivity in construction. Construction sector productivity is relatively high in OECD comparison, but productivity growth has been weak (Figure 3.8). Planning decisions also define the value of land, and municipalities are large landowners, including in several of Sweden’s cities. Land regulated for high-density development of owner-occupied housing will for example be worth more than land regulated for low-density rental housing. This means that even though municipalities always have to sell their own land at market price, they are themselves to an extent defining the price (Sveriges Allmännytta, 2023[22]). When zoning their own land, municipalities have incentives to maximise its value. Incentives for zoning of privately owned land are not necessarily as strong. Also, municipalities tend to interpret the building code differently and add their own requirements in their zoning decisions. This adds cost and complexity to construction projects, and it reduces the scope for standardisation and modular construction (Hermansson and Song, 2024[17]). Process management in planning and zoning could be improved by investing in skills, digitalisation and automation and disseminating best practices.
Gross value added per hour worked, construction
Source: Eurostat.
The construction process is inherently complex, with supply chains involving a large number of companies with few employees, and a high degree of subcontracting from both Swedish and foreign companies. Labour division and specialisation generally increases productivity, but breaking down jobs into many subcontractors within complex supply chains is not always optimal. It places high demands on skills and supply chain management and can also create ambiguity of responsibility. Specialisation can also negatively affect employee skills development (Hermansson and Song, 2024[17]).
Construction is highly cyclical, with volatility of construction employment much higher than in the overall economy. Housing price booms also tend to increase the severity of the subsequent contraction (Albuquerque et al., 2025[24]). Contracting housing prices after rapid price increases during the pandemic have been at the centre of the current downturn, and construction has seen particularly pronounced job losses (Figure 3.9). A share of those who lose their jobs, typically 30-35% (Chapter 4) will find jobs in other sectors, and weak construction demand will reduce the attractiveness of construction-related educational choices, which means that when demand picks up again there are fewer skilled construction workers available to meet it. A scarcity of skilled workers will tend to increase wages and construction costs while at the same time reducing productivity as people with less relevant skills and experience are recruited.
A more pro-active approach to scale up investments in skills during downturns, including management and supply-chain management skills, would provide a pool of skilled workers for the subsequent upturn. This would strengthen the capabilities of the sector to run complex projects and adopt efficiency-improving technologies, thereby boosting productivity. Sweden’s purchaser-provider system for higher vocational education could be particularly well-suited to implement additional study places during downturns. Concretely, the Agency for Higher Vocational Education could, when financing study places from private providers and municipalities, include options to scale up student numbers, designed in such ways that they do not increase the cost of provision per student. Such a no-cost option can be achieved if it can only be exercised by mutual agreement. The same model could potentially be used in adult upper secondary vocational education and training and in Labour Market Training, which are shorter training courses organised for the unemployed by the Public Employment Service. Targeting support through the new Student Finance for Transition and Retraining Programme better towards people at risk of unemployment and towards people who have received redundancy notice could facilitate participation. Since downturns can last for many years and their duration is uncertain, such readiness to scale up should not be confined to construction only, but rather be part of a wider long-term strategy to address sectoral labour shortages, strengthen automatic stabilisers and support workers in dealing with structural change (Chapter 4).
Total employment, seasonally adjusted and smoothed
Source: Statistics Sweden.
Relevant skills in municipalities and construction companies are a prerequisite to making the most of digitalisation and new technologies, including automation, standardisation and prefabrication. Construction is one of the least digitalised economic sectors, while the potential to boost productivity by means of digitalisation is considerable. Construction Information Management (CIM) systems can for example improve the speed and efficiency of information exchanges between different actors involved in the zoning, permitting and construction phases, and reduce the risk of erroneous communication. The Swedish Housing Agency is running a project to increase the use of CIM in the public sector. Extending this project to the private sector to ensure wide adoption of the same standards and guidelines would be helpful (Hermansson and Song, 2024[17]).
References
[24] Albuquerque, B. et al. (2025), “Not all Housing Cycles are Created Equal: Macroeconomic Consequences of Housing Booms”, IMF Working Papers, Vol. No. 2025/050.
[21] André, C. and T. Chalaux (2023), “Housing prices during the COVID-19 pandemic: insights from Sweden”, Kwartalnik Nauk o Przedsiębiorstwie, Vol. 67/1.
[18] Bostadsförmedlingen (2024), Bostadsförmedlingens hyresmarknadsrapport 2024, (accessed on 6 April 2025).
[9] Bourassa, S. et al. (2012), “Mortgage Interest Deductions and Homeownership: An International Survey”, SSRN Electronic Journal.
[7] Boverket (2025), Behov av bostadsbyggande 2024–2033, Swedish Housing Agency.
[14] de Boer, R. and R. Bitetti (2014), “A Revival of the Private Rental Sector of the Housing Market?: Lessons from Germany, Finland, the Czech Republic and the Netherlands”, OECD Economics Department Working Papers, No. 1170, OECD Publishing, Paris.
[16] Engerstam, S., A. Warsame and M. Wilhelmsson (2022), “Long-Term Dynamics of New Residential Supply: A Case Study of the Apartment Segment in Sweden”, Buildings, Vol. 12/7.
[8] Eurostat (2025), EU-SILC survey, (accessed on 25 February 2025).
[15] Fritzon, G. (2020), “Rent Controls: How they damage the housing market, the economy and society”, European Policy Information Center, Stockholm.
[17] Hermansson, C. and H. Song (2024), “Produktiviteten i bygg-och anläggningssektorn - Orsaker, åtgärder och konsekvenser”, Background report to the Productivity Commission.
[10] Hilber, C. et al. (2014), “The Mortgage Interest Deduction and its Impact on Homeownership Decisions”, The Review of Economics and Statistics, Vol. 96/4.
[19] Karpestam, P. (2022), “Area income, construction year and mobility of renters in Sweden: two hypotheses about the impact of rent control”, International Journal of Housing Markets and Analysis, Vol. 16/7.
[23] Lind, H. (2020), “Mikael Anjou: Den ineffektiva byggbranschen – en förändringsagenda”, Ekonomisk Debatt, Vol. 3/2020.
[3] OECD (2025), OECD Affordable Housing Database, (accessed on 25 February 2025).
[11] OECD (2025), OECD Economic Surveys: Netherlands 2025, OECD Publishing, Paris (forthcoming).
[4] OECD (2024), An Agenda for Housing Policy Reform, OECD Publishing, Paris.
[2] OECD (2024), Society at a Glance 2024: OECD Social Indicators, OECD Publishing, Paris.
[5] OECD (2023), OECD Economic Surveys: Sweden 2023, OECD Publishing, Paris.
[1] OECD (2021), Brick by Brick: Building Better Housing Policies, OECD Publishing, Paris.
[6] OECD (2019), OECD Economic Surveys: Sweden 2019, OECD Publishing, Paris.
[12] OECD (2017), OECD Economic Surveys: Sweden 2017, OECD Publishing, Paris.
[13] Sjöberg, P. and B. Söderberg (2013), “Bostadsbristen och hyressättningssystemet : ett kunskapsunderlag”, (accessed on 1 March 2025).
[20] Sveriges Allmännytta (2025), Rapport: Går byggandet i takt med efterfrågan?, Public Housing Sweden.
[22] Sveriges Allmännytta (2023), Markens pris – rapport om allmännyttan och byggandet, Public Housing Sweden.