Housing amid Covid-19: Policy responses and challenges

 Introduction

The COVID-19 pandemic has hit the housing sector particularly hard, but governments have swiftly responded with an array of measures to alleviate the negative consequences of the crisis for tenants, borrowers, builders and lenders. However, some of these measures, unless they are temporary, can stand in the way of a robust recovery and/or impair the responsiveness of the housing market to the evolving needs of society (Figure 1). For example, rental market restrictions help tenants in the short term but typically also weaken supply responses, as they make housing investment less responsive to changes in demand and can pose obstacles to residential mobility. Against this background, this note provides new evidence of the impact of the COVID-19 crisis on construction activity and prospects, reviews the policy responses by governments, and discusses trade-offs between preserving short-term affordability of housing for tenants and mortgage-holders, facilitating mobility and ensuring sufficient, environmentally sustainable supply.

 

Figure 1. Housing policy responses to COVID-19 could lead to long-term distortions

Source: OECD.

  The COVID-19 crisis has severely hit the housing sector

The spread of the COVID-19 pandemic undermined the real estate sector throughout the world. Containment measures involved total or partial shutdowns of construction sites in many countries, and the associated income and revenue losses for households and enterprises adversely affected the outlook for the different segments of the property market, depending on the timing and stringency of confinement and the severity of the public health crisis, which differed across countries. Construction output fell by 40% in April 2020 in the United Kingdom.1 Similarly, the number of housing starts, or new residential construction projects, collapsed by more than 30% in April in the United States and only marginally recovered in May (+4%). On the other hand, also in the United States, the issuance of housing permits, which typically leads residential building activity, provides some glimpse of hope: after a strong decline in April (-21%), permits rebounded quite strongly in May (+14%). The recovery continued in June (+2%) bringing the number of year-to-date permits close to their 2019 levels.

There is a dearth of high-frequency cross-country indicators that are specific to the real estate sector. For a handful of countries, a construction sector version of the Purchasing Managers’ Index (PMI) assesses business conditions and sentiment in the construction sector. The indicator illustrates the dramatic deterioration in business confidence at the onset of the pandemic (Figure 2). Activity began to contract in March 2020 and reached a trough in April, when strict lockdowns applied in the countries for which the construction PMI is available. Although the lifting of the most stringent confinement measures brought some ease, activity in the construction sector continued to contract in May, except for Italy. By June, all countries have regained expansionary territory, except for Germany. Similar indicators point to comparable conditions in the United States: in April, prospects for single-family sales in the US fell to a value of 30, the lowest level since 2012 (NAHB/Wells Fargo Housing Market Index).2 The index recovered slightly in May (37) and even more in June (58) but remains far below the pre-crisis level (72).

 

Figure 2. Construction sector activity plummeted in early 2020

Note: A reading above 0 signals expansion and below 0 contraction. The indicator is equal to the original PMI published by Markit minus 50.

Source: IHS Markit.

Unfortunately, comparable survey data is not available for other OECD countries. To overcome the data gap and gauge construction sector business conditions across a larger set of countries, internet-search data offer additional insights (Box 1, Figure 3). Internet searches related to broad activity in the real estate sector, including rentals, inspection and appraisals, and property listings and agencies, correlate strongly with the construction PMI indicator and confirm the sharp slump in the real estate sector as countries put confinement measures in place. The severity of the slump varied across countries depending on the strictness of containment and confinement measures.

 

Figure 3. The COVID-19 crisis has seriously hurt construction sector conditions across the world

Note: The colours indicate the deterioration of business conditions in the construction sector as measured by the change in the seasonally adjusted and demeaned Google Trends category “Real Estate” during the COVID-19 crisis. The Google Trends-based indicator ranges from ‑50 to +50 so that its change in principle takes value from -100 to +100 though in practice all changes were negative for the period shown on the chart.

Source: OECD calculations based on Google Trends analysis.

Prospects are now improving, as many countries are gradually lifting confinement measures. Recent Google Trend data suggest that the conditions have improved markedly across most countries, without, however, reaching February levels (Figure 4). It is noteworthy at this stage to remember that, following the original PMI’s definition, a positive reading of the index only suggests that activity is expanding, albeit from a very low level, not that output has come back to pre-crisis levels. It will certainly take some time in many countries before residential construction reaches pre-COVID-19 volumes. Besides, there remains considerable uncertainty about the extent to which the economic slump is going to weigh on future demand and the prospects of the sector at large.

 

Figure 4. A rebound followed the construction crash in most countries

Note: Positive numbers mean expansion and negative ones contraction. The chart shows countries for which the Google Trends data are available and where the population is greater than 10 million in 2018.

Source: OECD calculations based on Google Trends analysis.

The United States, where high-frequency data are available, provide one example of a rapid recovery in housing transactions and construction. The sales of new residential homes dropped considerably in March and April: as a result, the inventory of finished yet unsold homes expanded from 5.5 months of supply in February to 6.7 months in April 2020. Sales recovered strongly in May, however, bringing the pace of construction and the inventory back in line with the demand for new homes: months of supply dropped back to 5.6 in May 2020. This stock-flow assessment of the US housing markets supports the view that direct long-lasting effects on home prices may be limited. To what extend second-round demand effects via unemployment and household income alter this view remains uncertain at this point.

 Mitigate short-term social and economic impacts: Rescue and support measures

The partial shutdown of economic activity implemented to fight the pandemic has exposed vulnerable households to sudden income losses and exacerbated the housing affordability challenges that a large number of OECD countries were already experiencing before COVID-19. Many households that already struggled to make ends meet before the crisis now face the risk of material deprivation, including the inability to honour rent and mortgage payments (OECD, 2020[1]). Indeed, before the crisis, low-income households were already devoting a large share of their earnings to servicing their debt and rent obligations, leaving them with little or no scope to absorb a fall in their income after taxes and transfers (Figure 5). Poor housing conditions, especially overcrowding, also appear to have been exacerbating the health impact of COVID-19 (Brandily et al., 2020[2]).

 

Figure 5. Low-income households spend much of their income on housing

Rent and mortgage burden as a share of household disposable income in the bottom quintile of the income distribution, 2018 or latest available year

Note: Rent burden includes private market and subsidised rent. Mortgage burden includes principal repayment and interest payments.

Source: OECD Affordable Housing Database (Adema and Plouin, 2020[3]).

With the onset of the COVID-19 crisis, governments responded with a host of specific measures to protect mortgage-holders and tenants in addition to the support from social safety nets. A number of countries also intervened to help the post-crisis recovery of the construction sector (Figure 6). In most countries, emergency support involved a suspension of eviction procedures, temporary forbearance of rent and mortgage payments, and in some cases moratoria on utility payments. Most governments, at both national and local levels, also took specific steps to shelter the homeless during the lockdown.

 

Figure 6. Most support measures targeted tenants and homeowners

Note: The sample includes all 37 OECD and 9 non-OECD countries. See detailed country-coverage in the Annex.

Source: OECD (2020[4]), DIW, Arena Housing Project.

Countries implemented several specific policies to support tenants and homeowners in addition to letting social safety nets play their role. For instance, Spain introduced a moratorium on rent payments for economically distressed tenants who rent from large-scale landlords. Vulnerable tenants in Spain were also offered financial support to pay rent, in the form of six-month microcredits at zero interest backed by the government and repayable over ten years. Austria, Germany, Mexico and Portugal allowed severely affected tenants to postpone rent payments for the months of strict lockdown. Moreover, Ireland, and Luxemburg deployed financial support to tenants unable to honour rent payments as a result of the crisis. In Portugal, households who have suffered from a loss in income of more than 20% and do not benefit from the right to defer their rent payments may be granted an interest free-loan to honour their rent obligations. In Greece, the government authorised a temporary reduction (of up to 60%) of rent payments for tenants that lost their job during the crisis. Similar measures were taken by local governments and in some cities, such as in Lisbon and Sintra (Portugal), social housing rents have been suspended for several months (OECD, 2020[5]).

Rental market regulations were also adjusted, at least temporarily in some countries. For example, Ireland, the Netherlands and Spain put in place rent freezes for rental contracts that were renewed during the confinement period. Rental contracts were made more favourable for tenants, by extending duration or allowing premature termination, to facilitate compliance with the restrictions imposed by the lockdown in Argentina, Austria, Belgium, Germany, the Netherlands, Portugal and Spain.

Steps were taken to protect mortgage-holders. Countries like Italy and the United States – in which the proportion of income devoted to the repayment of mortgage debt was among the highest in the OECD countries before the crisis (Figure 5) – implemented measures to protect mortgage holders against the risk of losing their homes, such as the suspension of foreclosure procedures during the period of confinement. This measure was also taken in the Netherlands and the United Kingdom.

The coverage and generosity of housing subsidies were increased in some countries. Ireland, Luxemburg and Russia lowered eligibility criteria for housing subsidies, extending access to financial support to households during the crisis, while the deferral of payments of real estate taxes was authorised in Latvia.

Other measures aimed to support housing finance directly. In some countries, liquidity was provided directly to banks and mortgage lenders (Canada) and, in some cases, by temporarily relaxing macro-prudential regulations imposed on banks (Israel, Norway and Turkey).

In some countries, steps were taken to raise the supply of social housing and support the post-crisis recovery of the construction sector. Where housing was in short supply before the crisis, additional funding and/or easing of lending conditions were put in place to provide liquidity to developers (Argentina, Austria, India, Ireland, Israel, the Netherlands, Russia and the United States). Also, in the Netherlands, the issuance of building permits was facilitated, and additional funding was provided to housing associations to increase the supply of social housing. Councils in big cities in Portugal granted exemptions to income and real estate capital gains taxes to incentivise the lettings on the affordable rental market for homeowners operating in the short-term holiday rental market. The European Union is putting together an initiative to support the renovation of buildings to improve housing quality, especially in the area of energy efficiency (European Commission, 2020[6]).

 Crisis-response measures are needed but involve policy trade-offs

While meeting an important objective of supporting tenants and borrowers during the crisis, several measures pose difficult policy trade-offs over the medium term (Figure 7). For example, if they are maintained for an extended period, measures that aim to preserve near-term affordability may create disincentives for the maintenance and expansion of the housing stock, as well as thwarting residential and labour mobility in the longer term. They may also undermine economic and financial resilience. This section sheds light on these complex interactions and discusses options for addressing them.

 

Figure 7. Housing policy responses to Covid19 involve trade-offs and complementarities

Note: Green means that the policy in the row supports progress towards the objective in the column; orange means that the policy in the row hampers progress towards the objective in the column; white means that there is no known effect.

Source: (Causa, Woloszko and Leite, 2019[7]; Cournède, Sakha and Ziemann, 2019[8]; Cavalleri, Cournède and Özsöğüt, 2019[9]; Causa and Pichelmann, 2020[10]; Bétin and Ziemann, 2019[11]) and authors’ elaborations.

 Laying the ground for a well-functioning, sustainable housing market

Measures that preserve housing affordability for mortgage-holders and tenants in the short term can have adverse longer-term side-effects. They can undermine resilience or compromise the long-term functioning of the housing market.

Tax support

Increasing tax relief for mortgage borrowers directly helps them meet housing finance obligations during the economic crisis. By supporting borrowers, this form of intervention also indirectly benefits lenders through lower loan delinquency, contributing to short-term economic resilience. However, if maintained for too long, tax advantages for mortgage-holders feed into house prices, making them more sensitive to variations in household income. The implications are twofold: first, prices become more unstable, compromising long-term economic resilience (Cavalleri, Cournède and Ziemann, 2019[12]); second, higher prices hurt affordability (Figure 8). These consequences underline the importance of providing a timetable for their withdrawal.

Rental market restrictions

Rental market restrictions, ranging from eviction moratoria to rent deferrals and freezes, have helped to keep vulnerable households in their homes and provided a degree of income protection for many renters. However, a tightening of rental market regulations entails longer-term adverse side-effects, which need to be considered in the design of recovery policy packages. In particular, a tightening of rent controls makes it more difficult for people who don’t already rent a dwelling to rent one. Such a tightening reduces the rates of return on real estate investment and creates uncertainty for developers and lenders, which discourages investment in real estate, making housing supply considerably less responsive to changes in demand. As a result, over time housing becomes scarcer and more expensive relative to income when regulations remain tight (Figure 8). By potentially resulting in supply shortages, a tightening of rental market regulations may also exacerbate speculative housing bubbles and increases in household debt, undermining economic resilience. Indeed, tight rental market regulation is associated with a higher probability of incidence of financial crises and more severe cyclical downturns in economic activity (Cournède, Sakha and Ziemann, 2019[8]).

Tighter landlord-tenant restrictions tend to reduce residential mobility. Quantitatively, opting for tight versus flexible settings in landlord-tenant regulation can have significant consequences for labour mobility Figure 9). Indeed, excessive protection of tenants often implies that people with uncertain labour market prospects, such as low-wage or non-standard workers, find it difficult to sign a lease, because landlords, who anticipate a difficult eviction in case of non-payment, require evidence about the stability of tenants’ income. Obstacles to residential and consequently labour mobility will be particularly unwelcome in post-COVID19 economies, given the need to adjust and facilitate the reallocation of labour and capital towards sectors and activities with promising economic prospects.

In light of the adverse long-term consequences of overly tight rental market regulations, public authorities would do well to adopt a calendar for the phasing-out of COVID19-related restrictions. Doing so would avoid letting emergency measures become new bottlenecks to the long-term efficiency of housing markets that would ultimately undermine affordability, inclusiveness and sustainability objectives.

 

Figure 8. Housing affordability would suffer from inaction

Additional years of income required to buy a 100m2 home in the absence of structural reforms towards best practices

Note: The figure shows the range of costs (in terms of additional years of income to acquire a home) across countries stretching from 0 for countries that are already operating at best practices to “Maximum” for the country furthest away from best practices.

Source: Cournède, De Pace, Hoeller and Ziemann (2020[13]).

 

Figure 9. Rental market policies have strong implications for residential mobility

Tighter rental market regulations are linked with lower residential mobility, while public spending on housing is associated with more mobility

Note: OECD calculations based on estimates from Table 1 in the source. The dot is the average estimated probability to move evaluated at average policy and household characteristics. The distance between the Min/Max and the average is the change in the estimated probability associated with a policy change.

Source: Causa and Pichelmann (2020[10]).

Public spending on housing

Higher government spending on social housing, in addition to helping short-term affordability, underpins mobility (Figure 10). The two most important categories of social spending on housing are allowances, which help tenants pay their rent, and capital spending to finance investment and expand the housing stock. Over time, there has been a shift from capital spending to allowances in most countries, which benefits residential mobility, as allowances are more portable than entitlements to social dwellings. However, it can involve large deadweight costs, as part of the spending on allowances can end up in higher land prices, if land-use restrictions make the supply of land for building purposes overly rigid.

By contrast, expanding capital spending on social housing, coupled with provisions ensuring that eligibility is portable, can generate benefits for both near-term affordability and long-term supply with limited adverse consequences for mobility. Furthermore, this kind of direct intervention in the market provides an opportunity for governments to promote and accelerate the spread of construction techniques that are aligned with environmental-transition sustainability objectives.

 

Figure 10. Public capital spending on housing has strongly declined unlike allowances

Public capital transfers and public direct investment in housing development, and public spending on housing allowances and rent subsidies, OECD-25 average, % GDP

Note: The OECD-25 average is the unweighted average across the 25 OECD countries with capital transfer and gross capital formation data available for all years between 2001 and 2018. It excludes Australia, Canada, Chile, Iceland, Israel, Japan, Korea, the Netherlands, New Zealand, Turkey and the United States. Spending on housing allowances does not include spending on mortgage relief, capital subsidies towards construction and implicit subsidies towards accommodation costs. See the source for more details.

Source: OECD Affordable Housing Database (Adema and Plouin, 2020[3]).

Easing land-use restrictions

Easing land-use restrictions is a way of facilitating the recovery of homebuilding and better aligning the supply of housing with evolving demand and the needs of society, while also paving the way for enhancing the efficiency of housing markets in the future. Streamlining the permit process would also help in the short-to-medium term, including by lowering prices (Figure 8). Fostering residential construction could also accelerate the transition to a low-carbon economy provided that the new buildings are required to comply with sufficiently ambitious environmental standards. In addition, reforms that make it easier to redevelop brownfield into residential areas offer a way of expanding housing supply without encroaching on agricultural land.

Facilitating construction and redevelopment would also allow accommodating the possible long-term change in housing demand that the COVID-19 crisis may prompt. There is a possibility that the COVID-19 crisis may lead to lasting mutually linked changes in housing demand and work organisation. Preferences could shift in favour of living in lower-density areas and working remotely. This could slow or even reverse urban-rural divergences. First, such a shift would relieve demand pressures in overly-dense areas. Second, flexible workplace amid more teleworking would free office space for conversion to residential units in city centres, provided land use can accommodate the change. The combination of such demand and supply effects could reduce regional home price differentials and contribute to reducing residential segregation.

Reforms to ease land-use restrictions deliver greater benefits if conducted within an integrated spatial planning framework across government sectors and hierarchies. This approach avoids that different sectors and levels of government enact policies that may conflict with one another. The goal should be to encourage housing construction and improve affordability while enhancing neighbourhood liveability and avoid excessive spatial divergence in the access to public services, transportation systems and social infrastructure. In particular, many national governments would do well to play a more active role in land-use governance through coherent national strategies that clearly specify the boundaries and specific conditions for development that can be utilised by local governments. In addition, strategic spatial plans that span functional territories help to encourage inter-municipal collaboration and internalise the negative externalities that municipal competition may entail.

 Supporting economic and financial resilience

Mortgages and property-related loans account for a large share of total bank credit, which underscores the importance of the real estate market for banks’ business models (Figure 11). The increase in loan delinquency due to the slump in economic activity and the mortgage payment forbearance measures put in place in many countries are putting pressure on lenders. For example, in the United States, where indicators allow for tracking mortgage performance at high frequency, delinquency rates rose in April by a record amount of 3.1%, which is three times as much as the previous record from November 2008 (Black Knight, 2020[14]).

The difficulties that households experience in meeting their mortgage obligations have had spillover effects on housing finance. For example, the impact is visible in the sharp fall of residential Real Estate Investment Trust (REITs) values in the United States and other advanced economies (Figure 12). REITs invest in structured real estate products, such as residential mortgage-backed securities (RMBS), which are losing value due to the deteriorating credit quality of the underlying mortgages. Since REITs are usually leveraged using their RMBS holdings as collateral, their lenders are demanding extra collateral or early repayments to meet their liquidity constraints. To the extent that REITS sell some of their RMBS holdings to meet these margin calls, the decline of RMBS prices is amplified, creating spillover effects among REIT shareholders, their lenders (mostly banks and repo market participants) and other RMBS investors.

 

Figure 11. Real estate loans account for a large share of banks’ assets

Real estate loans as a percentage share of total bank loans, 2018

Source: OECD Resilience Database.

 

Figure 12. Residential real estate investment trusts have lost much of their value since March

Note: The changes are calculated on Refinitiv REIT price indices.

Source: OECD (2020[15]).

There is hence a strong case for supporting both borrowers and lenders during the recovery. Measures to this effect can combine mortgage payment forbearance for borrowers with liquidity facilities for lenders as well as the temporary adaptation of prudential regulations. The objective is that payment forbearance does not automatically generate write-offs or margin calls, which can, in turn, themselves prompt fire sales. However, the crisis-related relaxation of macro-prudential regulations may over time undermine economic resilience. Tighter macro-prudential policy settings, such as requiring lenders to hold more capital against risky mortgages or lower limits on the amount households can borrow against the value of the house they purchase, are associated with higher quality lending, which reduces the risk of future crises (Cournède, Sakha and Ziemann, 2019[8]). Consequently, it is important to phase out these measures and revert to prudent macro-prudential policy settings, once the recovery is well underway.