COVID-19: Housing market impacts and housing policy responses - an international review

This report is the third in a series of reports written and produced for the ACOSS/UNSW Sydney Poverty and Inequality Partnership by Hal Pawson, Chris Martin and Fatemeh Aminpour at the City Futures Research Centre at UNSW Sydney along with Kenneth Gibb from the University of Glasgow and Chris Foye from the University of Reading. This report has also received support from Mission Australia, National Shelter (on behalf of NSW Shelter and Shelter WA)and Queensland Shelter.

The purpose of this series of reports is to look at the impacts of the COVID-19 pandemic on housing and homelessness policy. This third report looks at the effects of these policy changes in Australia and other high-income countries including Canada, Germany, Ireland, New Zealand, Spain and the US. Studying this range of countries gives the opportunity to compare similar jurisdictions with a variety of housing regimes and national governance systems.

The first report of this series, COVID-19: Rental housing and homelessness impacts – an initial analysis studied the impacts of the pandemic on housing and homelessness policies in a range of countries during 2020. The second report in the series, COVID-19: Rental housing and homelessness impacts in Australia took a deep dive into the pandemic impacts on housing and homelessness in Australia 2020 and 2021.

This third report in the series is the 16th report from the Poverty and Inequality Partnership. This partnership between ACOSS and UNSW Sydney explores the ways in which inequality and poverty relate to measures of disadvantage such as health, housing and homelessness through the inclusion of researchers from multiple disciplines.

We extend our sincere gratitude to the ACOSS members and philanthropists who continue to support this vital research partnership, including Anglicare Australia; Australian Red Cross; the Australian Communities Foundation Impact Fund (and three subfunds – Hart Line, Raettvisa and the David Morawetz Social Justice Fund); the BB and A Miller Foundation; the Brotherhood of St Laurence; cohealth, a Victorian community health service; Good Shepherd Australia New Zealand; Mission Australia; the St Vincent de Paul Society; the Salvation Army; and The Smith Family.

We thank the partnership and supporters of this project for their assistance throughout, as well as the ACOSS Board, UNSW Deputy Vice-Chancellor Equity, Diversity and Inclusion Professor Eileen Baldry, and Ian Jacobs during his tenure as Vice-Chancellor of UNSW Sydney.

Cassandra Goldie & Carla Treloar

Key points

  • Defying initial expectations, many countries have seen surging housing markets during the first two years of the COVID-19 pandemic. By late 2021, for example, both prices and rents were escalating at historically high rates in Australia, the UK and other Anglophone nations.
  • During the period from early 2020 to late 2021, nominal house prices rose in all eight case study countries included in this research – Australia, Canada, Germany, Ireland, New Zealand, Spain, the UK and the US. In sharp contrast to the 2008 Global Financial Crisis, none of the countries has seen any significant periods of nominal price decline.
  • It is the Anglosphere countries where the pandemic house price boom has been most evident. By Q3 2021, annual nominal house price inflation had reached 22% in Australia and New Zealand, 18% Canada and USA, 12% in the UK and 11% in Ireland.
  • General avoidance of pandemic-triggered housing market collapse must be credited, in part, to the remarkable government measures to maintain incomes and cushion economies seen during the first two years of COVID-19. Across the countries covered in this research, common approaches included shoring up benefits, introducing new temporary assistance and forms of furlough or wage subsidies, as well as finance sector support from central banks.
  • As also exemplified across the case study countries, extraordinary measures directly targeted at safeguarding housing systems and protecting at-risk populations also helped to confound initial fears of recession, crashing property values and surging homelessness. Such activities – as implemented in most of these nations – included mortgage payment deferrals, rental eviction moratoriums and emergency accommodation provision for homeless people.
  • Booming housing markets during COVID-19 are probably mainly due to the rock-bottom interest rates and quantitative easing measures that have also formed part of official economic stimulus in many countries. Pent-up household savings will also have contributed. In some of the case study countries (Australia and the UK) an additional factor was direct government-funded housing market stimulus – with hindsight, a misdirected form of official pandemic response. Many households, working through the pandemic and with a strong financial position and housing wealth to cash-in, were able to exercise expansionary housing choices that better met their adjusted housing preferences. Many others, however, have been locked out by resulting house price inflation.
  • Less widely reported than booming prices, housing rents also took off during 2021 across most of the countries in our research. Whereas several countries had restricted rent increases in the early ‘income shock’ phase of the pandemic, almost all had lifted these restrictions when rents began rising, on shifts in demand. By the end of 2021, with the possible exception of Canada, annual increases were topping 8% in all of the Anglosphere nations – a rate of increase generally far exceeding past decade norms. Rent inflation in Australia, the UK and the US was, by this time, running at rates unseen since the 2008 Global Financial Crisis.
  • German and Spanish housing market performance during COVID-19 has contrasted with that seen in the Anglophone world. In the house sales market, Germany’s relatively robust pre-2020 price growth continued on a similar trajectory, whereas Spanish price growth remained subdued. Germany’s rental market likewise appears to have been relatively unaffected by the pandemic, with rent inflation generally continuing to moderate during 2020 and 2021. In Spain, meanwhile, rents continued to decline in nominal terms.
  • Germany’s experience here probably reflects the country’s unusually stable and resilient economic and housing systems, a tradition of conservative mortgage lending, and a stronger social safety net. For Spain a key factor affecting the nation’s economy and housing market during COVID-19 will have been the heavy damage sustained by the dominant tourism industry.
  • Overall, in most of the eight countries covered in this research, rising house prices and rents during 2020 and 2021 will have resulted in generally declining housing affordability. In five of the six countries for which national statistics are available, nominal rent increases (usually in the range 10-15%) exceeded wage increases in the two year period to late 2021.
  • In several countries covered in the research, 2020 and especially 2021 saw both property prices and rents increasing faster for detached houses compared to apartments, and for suburban or rural locations compared to urban locations. These trends probably in part reflect the rapid rise of remote working that has weakened spatial ties to city centre office locations, enabling workers to contemplate out of town moves. As a result, pandemic-triggered damage to housing affordability is likely to be all the greater in the non-metropolitan settings attractive from this perspective.

Research scope

Two years after the outbreak of COVID-19, this report analyses pandemic impacts on housing systems across a range of high income countries during this period, and documents a range of policy responses relating to housing and homelessness. Our review arises from parallel studies initiated in mid-2020 by the UK Collaborative Centre for Housing Evidence (CaCHE), focused on the UK, and by the University of New South Wales, covering Australia.

Mainly undertaken in Q3/Q4 2021, the current study also encompassed six other developed countries: Canada, Germany, Ireland, New Zealand, Spain, and the US. Country selection was influenced by the need to include jurisdictions comparable with the UK and Australia, while also encompassing diversity in relation to housing regime type and national governance systems.

Emergency income protection, labour markets and economic resilience

Across all of our case study countries, the sudden realisation of the need for stringent mobility and social mixing restrictions in early 2020 quickly prompted far-reaching emergency measures enacted at huge public and private cost to protect household incomes and economic relationships. In many countries these included newly created wage replacement schemes, as well as enhanced payment rates for existing social security benefits – the latter implemented as an economic stimulus as much as in recognition of minimum incomes needed to sustain basic living standards.

Different approaches were in evidence across the eight countries depending on a range of factors such as pre-established welfare state approaches (contrast Germany with the Anglophone economies), and in some cases a pre-disposition for prematurely dismantling enhanced welfare payments (Australia and the UK).

Measures implemented in different countries also varied in the extent to which they targeted key sectors disproportionately affected by lockdown, as well as in the treatment of the self-employed, small businesses and non-citizens. More broadly, however, many national responses were a conscious emulation of similar income protection policies and goals following macroeconomic and political international co-operation. The extent and duration of emergency responses also of course varied according to the severity and temporal incidence of the public health crisis itself, with nations like the USA and Spain hit hardest, with the UK close behind.

Some initial interventions (e.g. temporarily increased rates for key social security payments in Australia and the UK for both new and existing claimants) had notably egalitarian impacts. However, the pandemic also accentuated many pre-existing sources of inequality and poverty, something reinforced by the k-shaped unequal economic recovery experienced in some countries in 2021 (i.e. different economic sectors recovering at different rates, times, or magnitudes). This helps to explain the UK controversy on the ending of the enhanced level of Universal Credit but also the interest in countries like Canada to debate making permanent some of reforms to insurance and benefits during lockdown.

Rental housing regulation

All of the case study countries implemented eviction moratoriums early in the pandemic, reflecting the global impulse to ‘stay home’ and the risk of widespread arrears and eviction – to which existing regulatory regimes in most countries contributed by providing only low or moderate tenure security and assurance of affordability. The eviction moratoriums varied significantly in coverage, duration, the legal mechanisms by which they were effected, and by the complexity of the response. The UK and the US, in particular, produced complicated, contested responses. The available evidence shows the impact of the moratoriums on rates of termination proceedings and evictions also varied, but was everywhere substantial.

It is striking how little permanent reform has come about as a result of the moratoriums, most of which ended by late 2021. It is a similar story with countries’ emergency rent regulations. While prohibitions on rent restrictions were a common early response, such restrictions have been ended, even as rental markets have become more pressured. Spain’s rent moratorium, in which large landlords are required to either defer rents or reduce them by 50%, has contrasted with the typical approach of other countries, which was to keep rents being paid. In Australia and the US, rent relief schemes directed cash payments and rebates to landlords, but many of the Australian schemes were undersubscribed, while implementation of the US scheme has been highly problematic.

In a few jurisdictions, significant permanent reforms strengthening tenant security have been enacted (or pledged) during the pandemic, although in most cases these involved processes already in train rather than being in any real sense a ‘COVID-19 response’.


In all of the countries covered in the research, early 2020 saw substantial emergency action to protect existing homeless populations from elevated health risks posed by the pandemic. This mainly involved publicly (or philanthropically) funded placements in hotels for rough sleepers, residents of homeless shelters considered to pose health risks, and sofa surfers no longer welcome to stay with friends or relatives. In some cases (e.g. England, New Zealand) these actions involved co-ordinated, nationally-driven and funded programs. In others (e.g. Canada, Spain, USA) they were more ad hoc locally-initiated measures driven by municipalities or NGOs.

Generally governments were quicker to initiate emergency accommodation (EA) programs in unitary and semi-unitary states (e.g. England, Ireland, New Zealand) than in federations (e.g. Australia, Canada, USA). In some federal states (Australia, Germany), the crisis exacerbated tensions in federal-state relations regarding division of responsibility for homelessness across levels of government.

Epidemiological evidence demonstrates that, imperfect though they may have been, emergency accommodation provision for homeless people measurably reduced the health toll of COVID-19 among those directly affected.

In Australia and the UK, the pandemic served as a stimulus for stepped-up assistance to the most vulnerable homeless populations – help that, for some, extended to being aided to secure longer term housing. This is likely to have meant that a substantial number of chronic rough sleepers and others will have gained a settled home who – in the absence of the pandemic – would not have done so. At the same time, as exemplified in Australia, the extremely limited capacity of the social housing system often saw help of this kind tightly rationed, and thereby extended only to a minority of potential recipients.

On the positive side, several state or national governments within the study remit also pledged significant new social housing investment programs during the pandemic. While relatively modest in scale and/or limited in duration, at least some of these represent commitments that would not have been made in the absence of the public health crisis

COVID-19 and housing market impacts – house sales markets

Partly referencing back to the immediate housing market impacts of the Global Financial Crisis, most economic commentary early in the pandemic anticipated a significant hit to house prices, with one widely-cited Australian bank scenario envisaging a three-year price decline exceeding 30%. Large reductions were also predicted through official sources in Canada and the UK. In practice, of course, the opposite has occurred – at least in most of the countries in this research (see Figure A).

Figure A: Nominal house price change, Q1 2019-Q3/4 2021

Sources: See Figure 5.5

In addition to record low interest rates and quantitative easing, key drivers of rising prices appear to have included strong income support measures; pent-up savings, especially among high income earners; and an increase in the proportion of incomes that households have been willing to spend on housing.

By comparison with our other case study countries, it is the Anglophone nations where a COVID-19 house price boom has been particularly marked. It is probably the combination of strong income support measures and highly liberalised mortgage lending regimes which primarily explains this. However, other policy factors will have contributed. Responding to the onset of the pandemic, central banks reduced interest rates and initiated or expanded quantitative easing, while in countries such as Australia and the UK national governments compounded the resulting housing market stimulus through homebuyer support measures – e.g. the Australian Government’s HomeBuilder house purchase and renovation grant package, and the UK Government’s cut to stamp duty on acquisition of dwellings valued at under £500,000.

House sales markets during the pandemic also tended to see a shift in consumer demand towards larger dwellings (typically houses) and suburban or rural locations, with inner city apartments comparatively shunned. This has been linked to the ‘race for space’, a hypothesized change in housing consumption preferences arguably influenced by the expanded scope (necessity) for working from home. If such changes were to prove enduring they could have the scope to reverse the post-2000s tendency for residential property markets in ‘superstar cities’ (e.g. London, Sydney, Toronto) to ‘outperform’ national norms. Many urban economists, however, doubt that agglomeration economies (the economic benefit of co-located firms and deep labour markets in large cities) will have diminished potency in a post-COVID world, thereby weakening metropolitan primacy.

COVID-19 and housing market impacts – rental housing markets

At the national level, most of the Anglophone countries covered in this study (especially Australia, Ireland, New Zealand, the UK and the US) saw a brief initial reduction in general rent inflation (rent prices rising more slowly) at the start of the pandemic, followed by rapidly accelerating advertised rents during 2021. By late 2021, market rents for available properties in all of these countries were escalating at more than 8%.

Figure B: Annual inflation in market rents, 2018-2021: Anglophone country comparison

As in relation to house prices, rent trends during COVID-19 in Germany and Spain have contrasted from those in the Anglophone countries. In Germany, apparently extending a pre-existing trend, rent inflation appears to have continued to subside during 2020 and 2021, albeit with nominal rents only beginning to actually decline in late 2021. In Spain rent inflation generally fell back sharply during 2020, with nominal decline setting in by early 2021.

In most of the Anglophone countries market rents in some capital (or other large) cities were comparatively hard hit early in the pandemic. This was notably true of London, Melbourne, New York, Sydney and Toronto. The same was true of Madrid and Barcelona. While some of these witnessed substantially revived demand and renewed rent inflation during 2021, there were signs of continuing parallels with the house sales market whereby rent inflation in non-metropolitan markets was often continuing to exceed that of cities. One of the starkest such examples involved Australia’s capital cities (where, by Q4 2021, advertised rents had just regained their pre-pandemic level) and non-metropolitan Australia (where rents had risen by 18% over this same period).

Higher inflation of house rents in comparison with apartments (paralleling house/apartment sale price trends) seems to have likewise reflected ‘the race for space’ – housing consumption preferences more influenced by dwelling size than prior to the pandemic.

In most of the Anglophone countries, notwithstanding market turbulence since March 2020, the pandemic has seen rents rising ahead of earnings, implying worsening of rental affordability that will have in most cases compounded previous trends. In the UK, for example, rents rose by 11% in the two years to Q4 2021, while earnings rose by only 7% over the last two years for which figures are available. Similarly, in New Zealand the comparable figures are 10% versus 8% and in the USA 14% versus 9%. In Australia, where city markets remained generally subdued in late 2021, this effect will have been mainly a non-metropolitan phenomenon – regional rents were up 18% in the two years to Q4 2021.
Relatively rapid rates of rental inflation may mainly reflect unusually low rates of tenancy turnover – a factor that could prove transitory. At the same time, however, at least for some countries (e.g. Australia, New Zealand) such a development is likely to be offset and perhaps even outweighed by the prospective re-start of international migration, a traditional driver of rental housing demand.


The sudden emergence of COVID-19 unleashed a tide of uncertainty and fear. But the expected housing market disorder, mass insecurity and homelessness has been – as yet – largely avoided. Much credit must be attributed to emergency income support provided, in many countries on a previously unimaginable scale. More direct housing and homelessness interventions have also importantly helped in dodging these bullets.

Yet hopes that COVID-19 might serve as a ‘focusing event’ prompting an overdue policy re-set to address embedded housing inequalities have proved largely unfulfilled. In fact, with house prices and rents having been largely inflamed rather than subdued through the first two years of the pandemic, most of our case study countries enter 2022 with housing affordability pressures even more acute than at the outset.

1.1 Research origins, purpose and remit

As COVID-19 exploded across the world in early 2020, it was immediately apparent that there were major implications for housing and housing systems. In the words of the UN Special Rapporteur on Housing (Farha 2020), as voiced on 18 March 2020, housing is the ‘first line of defence against the COVID-19 outbreak’. To shore up this line of defence, many countries saw innovations in housing and homelessness policy, and in income support, formulated and implemented at astonishing speed and scale. At the same time, the economic disruption quickly triggered by the pandemic raised the likelihood of huge housing market disruption with potentially devastating consequences.

Reviewing the situation two years later, this report analyses pandemic impacts on housing systems across a range of high income countries, and documents a range of policy responses directly or indirectly relating to housing and homelessness. Our review arises from parallel studies initiated in mid-2020 by the UK Collaborative Centre for Housing Evidence (CaCHE), focused on the UK, and by the University of New South Wales, covering Australia. CaCHE/UNSW collaboration has already featured in published reports on the crisis in its early stages (Fitzpatrick et al. 2021; Pawson et al. 2021).

Now, to inform a wider international comparative review of COVID-19 and housing, we have researched pandemic impacts and policy responses across six other developed countries in addition to the UK and Australia. This work was mainly undertaken during Q3/Q4 2021, at a time when the pandemic had been ongoing for 18-24 months. In these countries the progress of vaccination programs had, by this time, largely enabled emergence from large scale lockdown restrictions. While the public health emergency was far from over, economic and social interaction had been substantially restored and housing systems were functioning more normally. While Australia and New Zealand were continuing to experience major restrictions during this time, in other countries many crisis housing measures had lapsed and housing market transactions were proceeding largely unimpeded.

Housing is of course a multi-faceted topic and housing policy likewise encompasses many diverse dimensions. In this review we focus on certain distinct parts of this territory:

  • Housing markets
  • Rental housing regulation
  • Homelessness

Also, given their fundamental importance in underpinning housing system consequences of the pandemic, we also review social security and other income protection policy responses enacted across the eight countries. Pandemic impacts across a wider range of housing domains have been researched by CaCHE colleagues in relation to the UK context (see These have encompassed housing institutions and their resilience, domestic abuse, tenant activism, and placemaking (planning).

The research is informed by case studies of the eight countries involving detailed literature reviews, and by interviews with academic colleagues in the six non-UK/Australia jurisdictions, as well as advice from a number of fellow experts across the case study nations. Our methodology is described in more detail in Section 1.3.

1.2 Initial expectations of housing and economic damage due to the pandemic

When the crisis hit in March 2020 and the necessity of economic and public health lockdowns of various kinds was crystallised, immediate, unprecedented and temporary policy interventions were established, albeit with an uncertain duration. Lockdown implied closing large parts of domestic economies, restricting international travel (business, tourism and migration) and quarantining of incoming travellers in some cases. It meant many millions working from the home where most of their time was now spent, often also home-schooling. Especially in 2020, many employees were laid off or entered into short-time working arrangements. At the same time, workers experiencing earned income reductions were widely supported by a range of income protection measures involving benefits, cash transfers, furloughing and other supports for businesses and the self-employed, all with different levels of generosity, comprehensiveness and continuity.

There was a clear recognition of the importance of housing during this period as evidenced by provision of extraordinary supports to keep people in their homes during the crisis and mitigate wellbeing concerns about pandemic-triggered debt, including housing payment arrears. Such assistance included mortgage holidays and deferred payments for owner-occupiers and landlords, suspension of arrears-based evictions, as well as the affordability-enhancing role of the above income protection measures. At the same time, and for more clearly public health-oriented reasons, many countries launched successful programs to move street homeless or people at risk of homelessness into now vacant hotel and related accommodation, and later to prioritise these groups for vaccination.

Not everywhere, but in certain countries, housing market transactions temporarily ceased, short-term lets closed down as a subsector, and housing construction stopped. It was in this context and in part drawing on the experience of the aftermath of the GFC that many commentators and official economic forecasters predicted collapsing housing markets in the wake of the pandemic and expected medium term economic recession, with large percentage reductions in housing prices predicted in Australia, Canada and the UK, among others. This encouraged both direct and indirect measures to support housing markets.

This report examines how, influenced by government interventions to redress perceived threats to stability, housing markets in fact performed during the first two years of the pandemic. It also reviews, in particular, the official measures adopted across our eight case study countries more directly aimed at protecting the welfare and housing security of vulnerable populations.

1.3 Methodology

Case study country and country expert selection

In our choice of countries it was seen as important to select jurisdictions:

  • Comparable with the UK and Australia
  • Encompassing diversity on housing regime type and national governance systems
  • Home to known housing studies colleagues with suitable expertise.

On this basis it was decided that, in addition to the UK and Australia, the case study countries should be Canada, Germany, Ireland, New Zealand, Spain and the United States.

In each case study country we identified two academic expert informants; one to cover each of the following:

  • Social policy/housing/homelessness policy expertise
  • Housing markets/economic policy expertise

Data collection step 1 – targeted literature review

Having kindly agreed to contribute, each country expert was asked for advice on existing published overviews/news sites/web pages containing material about:

  • Key policy changes legislated/implemented in response to the pandemic
  • Housing market developments/housing institution resilience under COVID-19.

Through a targeted literature review, guided as such, the relevant research team member was able to develop an overview note on policy changes/market impacts seen under pandemic conditions.

Data collection step 2 – country expert interviews

Online interviews with country experts were convened to confirm/enhance research team understanding of key policy changes (derived from literature review) and market impacts. These discussions were intended to investigate, in particular, why and how identified policy and market developments had come about, and with what impact. A topic guide was developed for this purpose.

Drawing on recordings of expert interviews, as well as our review of published literature, country working papers were drafted for each of the six non-UK/Australia nations, with one set of six focusing on housing markets and the other set of six on rental housing and homelessness. Additional added value was derived from consultation of interviewees on draft working papers. For some countries other academic colleagues were also invited to input through reviewing working paper drafts. The revised working papers formed the basis for this report.

Data collection step 3 – secondary data identification and analysis

Advised by case study country informants, the research team accessed statistical datasets on housing market transactions and related demand and supply dynamics for analysis to inform Chapters 5 and 6.

1.4 Case study countries: similarities, differences, regimes

The choices made about country selection were to a degree pragmatic. We selected federal systems, unitary systems and exemplars from different welfare regime types in Europe. We were able to exploit good early outputs, and data, access and networks in some countries as well the practical importance of the publication language involved. Inevitably, we have relied to an extent on published journalism, grey literature and early rapid publications – so language was a non-trivial issue and may impact on our evidence in particular places – Germany and Spain. Nevertheless, with the assistance of Google Translate, many domestic language publications from these latter two countries were included in our survey.

It is acknowledged that case study selection also reflected resources and time constraints but we would argue we have selected useful comparators for our overarching focus on learning for Australia and the UK. The federal systems of the USA and Canada are strong candidates for Australia, as is New Zealand, as the country’s near neighbour. Similarly, Ireland, Germany and Spain are well-fitted comparator countries for the UK (and its own constituent nations). At the same time, these countries also provide a range of COVID-19 responses which reflect the varying seriousness of the public health emergency itself, as well their government’s reading of the situation, their own policy stance, policy legacy and path dependency and the available instruments and the accidents of geography. There was also clearly a lot of international rapid sharing of policy innovation (worthy of study in its own right).

While not perfect, Esping-Andersen’s welfare regime typology (and its subsequent amendments) has been widely used across social policy and within housing research to cluster countries. Here, we can argue that we have exemplars of different approaches, even if some are missing (e.g. Nordic/social democratic) but we do have neo-liberal, corporate and southern European models represented, as well as clearly demarked forms of governance (federal, devolved and unitary systems). The welfare regime approach is a useful social policy simplification but it does not always work well when dealing with the wobbly pillar of the housing sector. Recently, in relation to private rental regulation across Europe, Kettunen and Ruonavaara (2021) pointed out that the Nordic countries actually have completely different attitudes to regulation and the primacy of the market.

Given these constraints and recognising that the objective was not to be exhaustive or comprehensive, but rather to generate sufficient breadth of experience and also provide a minimum of comparative analysis, we believe that this selection provided a useful basis for meeting our fundamental study objectives.

1.5 Public health and economic experiences across the comparator countries

During 2020 and 2021 the eight case study countries experienced very different exposures to COVID-19 infections and deaths. All saw outbreaks of severity sufficient to trigger major economic restrictions (or ‘lockdowns’) at national and/or large city scale. Since it is constraints of this kind that are liable to feed through into the functioning of housing systems, it is to be expected that housing market impacts will have been felt in all of the countries concerned. Nevertheless, as shown in Figure 1.1, as measured by associated death rates, the pandemic has been much more serious in the USA, the UK and Spain than in the other northern hemisphere countries.

Figure 1.1

When it comes to the severity of economic damage resulting from the pandemic, the GDP series graphed in Figure 1.2 suggests that, once again, the UK and Spain were far more seriously affected than the other countries in the set. It should be noted that Ireland is excluded because Irish figures on this indicator are distorted by the disproportionate share of economic activity related to high value transactions of foreign-owned corporations that have little material impact on the Irish economy. However, on a Modified Domestic Demand basis, Ireland’s GDP in fact contracted by 17% in the period Q4 2019-Q2 2020 – similar to the Euro area average (15%) (Government of Ireland 2021).

Figure 1.2

1.6 Report structure

Following this introduction, the main body of the report is structured thematically, with each of the main research topics being covered in turn. Because of its crucial underpinning role, we first look at emergency income protection measures that tended to involve various forms of social security enhancement and wage subsidy. This analysis forms Chapter 2. Then, in Chapters 3 and 4 we concentrate on rental housing and homelessness, primarily focusing on policy responses to the pandemic which typically included some level of restriction on rental evictions, as well as enhanced emergency accommodation provision for rough sleepers and others. Chapters 5 and 6 analyse housing market developments seen during 2020 and 2021, as these played out in the house sales and rental property markets. Finally, in Chapter 7, we reflect on the key messages that emerge from the research.


Fitzpatrick, S., Mackie, P., Pawson, H., Watts, B. and Wood, J. (2021) The COVID-19 crisis response to homelessness in Great Britain

Government of Ireland (2021) Budget 2021 Economic and Fiscal Outlook

Kettunen, H. & Ruonavaara, H. (2021) Rent regulation in 21st century Europe – Comparative perspectives; Housing Studies, 36 (9) pp1446-1468

Pawson H, Martin C, Sisson A, Thompson S, Fitzpatrick S & Marsh A 2021. COVID-19: rental housing and homelessness impacts – an initial analysis. Report no. 7 for the ACOSS–UNSW Poverty and Inequality Partnership

Key points

  • The eight nations undertook a remarkable effort to maintain incomes and protect the economy to take people through lockdown and the public health emergency but also to support post-lockdown recovery. Governments pursued similar policies to shore up benefits, introduce new temporary assistance and, in particular, to adopt variants of furloughing or wage subsidies. Allied to this were indirect measures pursued through monetary and fiscal policy to keep finance and business working when demand was shocked.
  • The polices in different ways and to varying extents complemented housing protection measures such as emergency homelessness shelter provision, suspending evictions, deferring mortgage payments and providing direct help to low income households through higher housing support.
  • These were unprecedented and far-reaching policies. They did however operate in a highly uncertain environment and there was a tendency to see cycles of support and intervention rise and fall alongside the successive waves or spikes in virus caseloads, as well as growing stakeholder pressure to return to a ‘normal’ economy. These responses were further complicated by early excessively pessimistic forecasts about economic rebound (as well as the housing market).
  • Different responses could be discerned across the eight countries depending on a range of factors such as pre-pandemic welfare state approaches (contrast Germany with the Anglophone economies), and in some cases a desire to quickly unwind enhanced welfare payments (Australia and the UK).

2.1 Chapter introduction

The economic responses by national governments following lockdown in the wake of the initial pandemic outbreak were unprecedented. These include macro stabilisation, monetary and fiscal policies and the efforts of central banks to support business and finance during the emergency (i.e. not just labour market, income protection and help with housing costs). This had, if not always admitted, clear Keynesian countercyclical spending benefits for business but particularly through furloughing and enhanced benefits, in terms of direct consumer spending (most dramatically exemplified through giving the population one-off cash transfers in the USA). Not only were they more effective than many would have dared to hope in shielding jobs and livelihoods, but they also made a huge contribution to protecting housing security.

In part, national economic policy responses hastily rolled out in many countries from March 2020 were inspired by international guidance (WHO, EU, IMF, etc.). But they also involved emulating policies from elsewhere, even if they were essentially untried domestically (e.g. the UK’s adoption of furloughing), and working with the grain of existing domestic income protection arrangements, as influenced by the extent of income protection deemed appropriate according to national traditions. The responses were also contextualised by the shape of national public health responses (e.g. the early international quarantining of Australia and New Zealand was quite different in nature and impact from the approaches pursued elsewhere). While some countries veered towards more comprehensive responses, others favoured a sectoral focus and took different approaches to the self-employed, support for business vis a vis the employees of those businesses, while others distinguished between intervention through unemployment assistance or by wage subsidy/furloughing.

Across our eight case study countries, governments pursued a range of macroeconomic supports and fiscal and monetary mechanisms while experiencing different GDP shocks (influenced by policy stances premised on initial economic forecasts of differing accuracy around consequent levels of GDP shrinkage, intervention take up, borrowing and indebtedness). In different ways, all eight governments tried to anticipate future needs for intervention by winding back and phasing out income protection interventions, though they all later were obliged to reintroduce or provide second generation supports when further waves of the pandemics and lockdowns continued into late 2021 and beyond. All learned ‘on the job’ as the pandemic progressed but did so unevenly and in ways filtered by their policy settings and underlying willingness to intervene in markets.

A number of key themes are explored in this chapter. First, what were the policies? Second, why did the income protection or economic resilience measures take the form they did in specific nations? Third, how effective were these interventions and why? Fourth, what do these interventions mean for national housing systems? Fifth, what are the longer term implications of this remarkable episode? These themes are developed through the chapter and the focus of the final section.

The structure of the rest of the chapter is in three parts. Section 2.2 sets the income protection measures in a broader national context, links them to macroeconomic questions and recognises the importance of risk and uncertainty in pursuing the policy interventions. Section 2.3 describes the main policies introduced in each of the eight nations studied. Section 2.4 is a broader consideration of the implications of these policies, returning to the five themes introduced above. This includes thinking about the repercussions for national housing systems and also considers the broader comparative lessons of what is observed specifically for Australia and the UK.

2.2 Contexts for intervention

To understand better the shared and the national-specific characteristics of income protection interventions across the eight cases examined here, a number of factors should be considered:
The overarching or background government and cultural attitude to large scale, if temporary, interventions in key markets, especially for social security and the labour market – a touchstone of fundamental debates about the ‘proper’ place of market and state. This can be represented to a degree by welfare regimes (Esping-Andersen 1992) such that different clusters of countries have high-level similar or aligned approaches to the welfare state (especially income protection), the role of the market and the functions of government to steer these arrangements and norms (noting that these regime positions are not immutable, especially over time)1There is a lively critical literature about welfare regimes that notes many of the shortcomings and ad hoc extensions made to make countries fit across different dimensions, something particularly the case with housing policy. We view the use of such clustering devices in a pragmatic and heuristic sense and share some of the misgivings discussed in the literature (e.g. Aspalter 2019). Moreover, our focus here is largely about the similarity of albeit temporary large scale measures across such different types of economy and welfare state..

As we argued in the previous chapter, we have selected eight nations from the global North on pragmatic grounds of familiarity, access, data, levels of economic development and specific forms of comparability (e.g. governance; language, etc.). Adopting the welfare regimes clusters widely used in comparative social policy studies allows us to group together subsets of countries (e.g. USA, Canada and Australia) and also to highlight comparisons across different types of welfare regime (e.g. Germany and the more neo-liberal examples). We also note the governance distinction here between federal regimes (USA, Canada, Australia, Germany), devolved systems (UK), regional systems (Spain) and more unitary systems (Ireland and New Zealand) – important distinctions that further complicate the use of regime blocs.

National responses were in part a reaction to clear international calls for co-ordinated responses from a macroeconomic, and also income protection perspective, in part because of the common requirement to deal with the immediate public health crisis and the need for lockdown and the massive economic shock that that generates.
Large scale rapid income protection measures are also a significant response to reassure the domestic audience, maintain order and to sustain as much normal living as is possible in such uncertain circumstances.

The similarity of immediate responses, filtered by existing national systems, reflects the importance of international organisations sharing good practice in a time of public health emergency and a willingness of national governments to emulate the more attractive or apparently suitable approaches to crisis income protection intervention. It also indicates a degree of state confidence that it could swiftly and efficiently deliver such radical and economy-wide policy innovation during a public health crisis.

On the one hand, there is a relatively co-ordinated response but also a shared nervousness based on the previous large scale shared crisis of the GFC in the late 2000s and beyond. Governments recognised that they were going to rack up considerable borrowing and debt to pay for these income protection measures (and related business support) and several of our cases (USA, Ireland, Spain and the UK) had been significantly politically and financially impacted by the fallout from the financial crisis. It turned out that while these nations did not stint from the large cost of the measures, they were, as with other countries, almost immediately looking at how to exit themselves from these obligations. Arguably, several did this too quickly (e.g. Australia) and were later obliged to restart interventions. It is also the case that governments were now less willing to accept the political damage of austere responses in a public health crisis and anticipated subsequent economic recession or sustained downturn. They also showed much more willingness to embrace more unconventional monetary policy responses (quantitative easing) as part of the countercyclical macro response.

What were the key components of income protections? These are examined in more detail in the next section but important common strands can be seen in Box 2.1

BOX 2.1

  • Wage subsidy to businesses to reduce the cost of labour
  • Furloughing – i.e. the state paying a proportion of salaries to keep people at home (with an element or contribution from employers) not working but with a reasonable income without making them redundant while business demand is impacted by lockdown
  • Financial support for businesses through grants, tax reliefs and cheap rapid provision of low cost loans (as well as central bank maintenance of liquidity and QE operations)
  • Enhanced unemployment assistance and other pre-existing benefits (including the temporary removal of conditionality rules such as job search evidence in the UK and Australia), new benefits associated with the impact of COVID-19 lockdowns, etc. as well as more universal direct cash payments (e.g. USA)
  • Financial supports to the self-employed and small businesses (the diversity of this sector and its own rule complexity made it more difficult to support as comprehensively or consistently, creating the risk of gaps (e.g. as was repeatedly argued in the UK).
  • As the enormity of enforced large-scale economic shutdown quickly loomed in to view in the northern hemisphere spring of 2020, analysts were in near unison in forecasting significant reductions to GDP, house prices and a long economic recession. These forecasts were later replaced by more moderate adverse effects and indeed smaller aggregate effects of subsequent pandemic waves (projected housing market reverses were perhaps among the most egregious forecasting errors especially in countries like Canada (by Canada Mortgage and Housing Corporation (CHMC)) and the UK (OBR), but to an extent this was true in almost all cases examined). Nonetheless, the economic measures to protect incomes, lockdown and home working also produced important distributional consequences that aggregate measures could not fully mitigate. Specific sectors (and exposed places) of the economy such as tourism, hospitality and other personal service sectors were worst hit, whereas online retail and distribution benefited relatively (Duca et al. 2021).

Despite the unprecedented interventions, in countries like the UK, Ireland and the USA, dependence on food banks, charity and reliance on income maintenance soared, reinforcing pre-existing inequalities (e.g. by gender, race, by digital divides, and also the consequences of prior decisions to exclude non-permanent residents in both Australia and the UK, though the UK did include those with no recourse to public funds in their emergency homelessness measures). While an absence of the aggregate measures would have allowed unimaginable economic dislocation (exacerbating the public health crisis), it is important to remember these important unresolved (and sometimes aggravated) questions of distribution.

Stepping back, a unifying challenge faced by governments (and arguably one that is continuing) concerns the uncertainty of the near future public health environment and the consequent measures required to manage future waves and variants of the virus. Optimism bias and also societal exhaustion with COVID-19 constraints on behaviour and freedoms run up against exponential growth of new variants and threats to the capacity of health systems. Even in well vaccinated nations, it is fundamentally hard to plan for the short to medium term. Consequently, governments have been forced back into cycles of lockdown measures (more restricted – in some cases, advisory rather than mandatory) and mitigating income protection followed by release from restrictions and the rundown of these supports.

Latterly, governments across the eight nations studied here have been developing more divergent ways forward – e.g. the UK’s accommodation of higher infection caseloads (and vaccination) in order to maximise the ‘economic freedoms’ from lockdown measures (and implicitly accept higher minimum levels of infection and illness). But underlying these strategies is a fundamental uncertainty about the future trajectory of the virus, about vaccine and treatment efficacy in the face of new variants and future economic scenarios. This is the radical uncertainty discussed by Kay and King (2019) and a context that makes it difficult (and unwise) to plan and commit long-term economic and financial strategies, further compounding these uncertainties.

The next section looks at the income protection responses of the eight national case studies.

2.3 Approaches to income protection and economic resilience

2.3.1 Australia

The Australian Federal Government instituted key income protection measures in March of 2020 which were then phased out between September of that year and March 2021. Further different measures were introduced in the summer of 2021 in response to ongoing local lockdowns at city and State/Territory level. The initial key measures were JobKeeper salary replacement or wage subsidies (originally $1500 for full time and $750 for part-time workers – both fortnightly) and a Coronavirus Supplement, which was added to JobSeeker payments for the unemployed and some other benefits and was originally worth $550 a fortnight (doubling the standard JobSeeker rate). Importantly, the temporarily higher rate of unemployment payment applied to pre-existing benefit recipients as well as people losing work due to the pandemic (and ineligible for JobKeeper).

Remarkably, thanks to these measures, the average income of the poorest tenth of Australian households (decile 1) actually increased in the initial months of the pandemic (Biddle et al. 2020).

Similarly, it is estimated that, when housing costs are taken into account, 2.6 million people were experiencing poverty in June 2020; whereas, in the absence of the special income protection measures this figure would have been 5.8 million (Phillips et al. 2020).

In the second phase of prolonged lockdowns in 2021, Pandemic Disaster payments were introduced by Australian governments, their remit widened after public outcry at the initial exclusion of locked down part-time workers receiving social security payments. Rates varied between $200 and $750 depending on hours worked (they were phased out once states reached 70-80% vaccination levels). These lower payments, compared to JobKeeper, also went to more than a million fewer low-income people receiving other social security benefits (compared to the first phase). The only provision still in place at late 2021 is Pandemic Leave Disaster Payments, which helped casual workers or other workers without access to sick leave when they are forced to quarantine or self-isolate (Pawson et al. 2021; Pennington and Stanford 2021).

According to the IMF (2021), the JobKeeper program paid out $89 billion (4.5% of 2020 GDP). Unlike the other income protection measures, the Pandemic Leave Disaster Payments policy was funded by State and Territory Governments and not the Federal Government. However, the wider program of fiscal interventions aimed at supporting households, businesses, public health and other interventions to support the financial system and encourage economic activity were worth $312 billion or just under 16% of 2020 GDP. At the same time as the lending environment was supported, quantitative easing was undertaken by the Reserve Bank of Australia with $100 billion of secondary market government bond purchases.

2.3.2 Canada

Canada spent about $290 billion (13.2 percent of GDP) in direct aid to households and firms, including wage subsidies, payments to workers without sick leave and access to employment insurance, an increase in existing GST tax credits and child care benefits, and a new distinctions-based Indigenous Community Support Fund. The Government also funded around $85 billion (3.9 percent of GDP) in liquidity support through tax deferrals (IMF 2021).

During the pandemic-induced economic downturn, Canada brought in connected policy interventions that financially supported keeping people in work (the Canada Emergency Wage Subsidy (CEWS)). They also offered enhance levels of social security to those who did lose their employment (the Canada Emergency Response Benefit (CERB), subsequently changed to the Canada Recovery Benefit). Canada also provided financial assistance for COVID-related sickness benefit obliging people remain in their homes (the Canada Recovery Sickness Benefit). In addition, Canada provided a series of specific benefits for caregivers, students, as well as non-standard workers and the self-employed who could demonstrate that their income was reduced. The main policy, CERB, was phased out along with the wage subsidy, CEWS, by October 2021 (Pennington and Stanford 2021).

At the time of writing, several albeit less generous schemes remain in place that support workers unable to work because of local public health lockdowns, as well as measures to help caregivers and workers who have to self-isolate. There are also temporary sector-specific financial supports for tourism and hospitality business sectors and resources for those businesses hardest hit by covid-induced economic slowdown. The Canadian government is now reviewing whether or not to retain reforms that improved the working of the employment insurance system during the crisis so that benefits operated under more streamlined and quicker payment systems (Pennington and Stanford 2021).
Canada’s economy shrunk by 5.3% in 2020 but was already experiencing sharp recovery growth towards the end of 2020 (World Bank) and this seems to have continued in the first part of 2021 according to forecasters. Relevant monetary and macro interventions included: (i) reducing the overnight policy (interest) rate in March 2020 to 0.25 percent; (ii) quantitative easing measures – an extension of the bond buyback program across all maturities, purchases of corporate and government bonds; (iii) supporting additional market liquidity in different ways); (iv) the federal government announced $95 billion in credit facilities (including $13.8 billion in forgivable loans) to lend to firms under stress; (v) supporting the Canada Mortgage Bond (CMB) market by purchasing CMBs in the secondary market; (vi) under the Insured Mortgage Purchase Program, the government planned to purchase up to $150 billion of insured mortgage pools through the Canada Mortgage and Housing Corporation (CMHC) but by October 2020 decided this was no longer necessary.

2.3.3 Germany

As in the rest of Western Europe, the effect of the pandemic on economic and public health outcomes in Germany was negative, significant and sustained. However, even before the pandemic, Germany already had something resembling the Furlough Scheme. Under the “Kurzarbeit”, companies could apply for state income support when 30% of their employees were affected by a drop in demand. In response to the pandemic, the replacement rates were made more generous, and the threshold was reduced from 30% to 10% of employees (OECD 2020a; DB 2021). Under the pandemic Kurzabeit delivered 60% of employee wages for hours not worked (and this went up to 2/3 for parents; after 7 months these figures rise to 80% and 87%, respectively2(in German): Kurzarbeitergeld: aktuelle Informationen ( – with the employer paying the employees for hours actually worked. In total, workers typically received 70% of their pre-pandemic pay (Pennington and Stanford, 2021). Germany supported 6 million workers through this scheme, peaking in the spring of 2020 before falling back to just 2% of eligible workers by August 2021. Germany provided additional income supports to the hospitality, arts and music sectors, as well as to the self-employed.

Together with enhanced child benefit and increased income support for the self-employed, these measures helped sustain average wages (DB 2021) and consequently the pandemic only had a modest impact on the unemployment rate, which increased from 5% in February 2020 to 6.4% in the summer 2020, with the self-employed and marginally employed worst affected (DB 2021; Clark et al. 2021). The impact of the pandemic on the economy more generally has been considerable although once again less severe than in some comparator countries such as UK. Between Q4 2019 and Q3 2020, the volume of GDP dropped by -4% (compared to -8.6% in UK) (ONS 2021). According to the IMF (2021), the Federal Government initiated three supplementary emergency budgets in 2020 and 2021 worth more than 240 billion Euros or 9.8% of GDP and this included public health measures and business support such as grants to small businesses (the latter worth 50 billion Euros).

2.3.4 Ireland

Of our four European case studies, Ireland reported the lowest number of COVID-related deaths per capita: 106 per million compared to 111 per million in Germany (BBC 2021). Looking at GDP figures, one might conclude that Ireland has outperformed economically too, being the only EU country to report economic growth in 2020 (France24 2021). However, these figures are inflated by the high performance of export-markets (Bloomberg 2021). The domestic economy has not fared well. By Q1 2021, consumer spending had fallen by around 12 per cent year on year (Department of Finance 2021), while unemployment had risen to 7.4 per cent, broadly in line with the EU average (19-countries) (OECD 2021).

Nonetheless, the income protection measures introduced by the Irish Government appear to have largely insulated household incomes from the effects of the pandemic. The most significant were the Temporary COVID-19 Wage Subsidy Scheme (TWSS) – which temporarily subsidised up to 70% of incomes, capped at a maximum of 410 Euros per week – and the Pandemic Unemployment Payment (PUP) (Beirne et al. 2021). A recent study suggests that, without these measures, household incomes would fallen by almost 20 per cent in Q2 2020, whereas in reality, they fell by between 0.1-4.2 per cent (Irish Times 2021a).

According to the IMF (2021), Ireland introduced a fiscal program of €24.5 billion (about 14 percent of GNI), covering 2020 and 2021. This program included €20.5 billion in direct support, involving: (i) €11.4 billion labour market support (including job-related social security support), (ii) €2 billion health sector capacity enhancement, (iii) €1.5 billion business support, and (iv) €0.5 billion capital works. The Government’s Draft Budget for 2021 also contains an additional stimulus of 1.7 per cent of GDP focused on extending income support measures, providing support to the hospitality sector, and increasing green, health and housing spending.

2.3.5 New Zealand

In response to the pandemic and the need for the initial lockdown the New Zealand government put together a response and recovery fund worth more than NZ$61 billion (19.3% of GDP through to 2024-25 – IMF 2021). This included larger components such as: Health and related spending, (e.g. cost of managed isolation and vaccines) NZ$5.6 b or 1.6% GDP; permanent increases in social spending to protect vulnerable people (NZ$2.4b or 0.7% GDP); and a wage subsidy to help employers affected by the pandemic (NZ$13.9b or 4.3% GDP). This program was reintroduced during subsequent lockdowns – for example in September 2021 (Pawson 2021).

Other interventions included business tax changes (permanent) and temporary tax loss carry back (NZ$8.6b or 2.7% GDP), as well as Infrastructure investment (NZ$3.8b or 1.2% of GDP, Education spend (NZ$3.4b or 1,1% GDP), as well as Housing and Urban Development (NZ$4.7b or 1.5% of GDP3Transitional housing expenditure increased by 150% from $150.9M in 2018/19 to $253.3M in 2020/21 and emergency housing grants up 272% from $88.1M year ending June 2019 to $320.5M by year ending June 2021.). And, while it cost relatively little (NZ$0.6b), a generous non-taxable unemployment benefit for those who lost work in the lockdown, which was only paid to the newly unemployed and was albeit temporary but (somewhat controversially) more generous than existing social security (Pawson 2021).

The New Zealand economy shrunk by 11% in Q2 of 2020 but recovered so that the overall reduction for 2020 was only 2.9% and this was followed by stronger growth in the first half of 2021 (IMF 2021). At the same time the IMF reported that the New Zealand Reserve Bank cut interest rates, provided liquidity to the financial system and entered into quantitative easing purchasing of bonds. These interventions spilled over into the housing market with temporary relaxation run by the reserve bank and retail banking of mortgage loan to value ratios (May 20202 to March 2021), as well as 6 month deferment of mortgage principal and interest payments (ending March 2021). Alongside this, there were also temporary rent freezes (until September 2020) and during lockdown restrictions prevented tenancy evictions (during March-June 2020).

2.3.6 Spain

Spain has had one of the worst pandemics in Europe, both in terms of economic and public health outcomes. By August 2021, Spanish deaths equated to a death rate of 1800 per million, only just below that of the UK (1970 per million) (BBC 2021). Spain’s economy also experienced one of the sharpest contractions in Europe (IMF 2020), with youth unemployment soaring to 42 per cent by September 2020, the highest in Europe (EuroNews 2020). Spain’s exposure to international tourism and disproportionate use of temporary contracts, and the prevalence of SMEs made it especially economically vulnerable to the pandemic (OECD 2021a; FT 2020), and continue to do so. (FT 2021c).

Spain entered the pandemic with comparatively high levels of public debt (96% GDP) and with considerably less fiscal leeway in its response. In terms of social security, the major interventions were: a moratorium on taxes for self-employed; widening of unemployment benefit access to include temporary and short-term workers; and the introduction of a furlough scheme up to a maximum of 950 Euros per month (OECD 2020). According to Pennington and Stanford (2021), Spain’s Expedientes de Regulación Temporal de Empleo (ERTE, the temporary labour force adjustment plan) subsidised 70% of workers’ net salary for the first 6 months of the program, before dropping to 50%. The program was cut at the end of September 2021 and by then less than 500,000 workers were still covered by the scheme.

With a million more people unemployed than pre-COVID and with already long-term entrenched unemployment and poverty, Spain introduced a new monthly minimum income support payment targeted to lower-income groups. The typical income support benefit was worth 600-700 Euros a month with larger amounts of up to 1015 Euros a month for larger families and single parents. Around 850,000 low-income households received such monthly supplements (TAI 2021). These measures appear to have fallen short of protecting those at the bottom of the income spectrum though: material deprivation increased from 4.7 per cent in 2019 to 7 per cent in 2020 (El Pais 2021).

According to the IMF (2021), the Spanish Government injected 85 billion Euros into pandemic-related programs (7.4 per cent of GDP). This included 24.7 billion Euros for ERTE, 6.5 billion Euros for the self-employed, 1.6 billion Euros for existing income support and 3 billion Euros annually for the new minimum income support system.

2.3.7 United Kingdom

The initial lockdown commenced in March 2020 and was followed by income protection measures in the form of furloughing, enhanced universal credit payments for working age claimants, support for the self-employed, for businesses (e.g. extensive rates relief on local taxes, grants and loans cost about £66 billion). The Job Retention Scheme (furlough) paid up to 80% if wages for those on reduced hours. Initially for full time employees only, it became more flexible in the second lockdown. Many employers contributed the other 20% of costs in order to allow employees to be fully furloughed. The scheme benefited 12 million at its peak at a cost of around £66 billion.

On the social security side, Universal Credit, the main working age benefit, was increased by £20 per week or just over £1,000 a year from April 2020 until the increase was abandoned in October 2021 after costing around £10 billion. The UK Government also (temporarily) removed conditionality rules for key working age benefits and also temporarily restored the 30th percentile local housing allowance for low income private renters.

While furloughing and related measures were seen as temporary, the consequences of further national lockdown post December-2020 led to the extension of these supports which were phased out (or ended, in the case of the enhanced UC payment) by the autumn of 2021. The scheme for self-employed workers carried on through to August 2021. In the first period up to October 2020, 2.3 million or 69% of those eligible took part, and even by January 2021 it was estimated to have cost £20 billion.

In 2020, the UK economy contracted by 9.8 per cent of GDP (IMF 2021). The UK Government massively increased public spending in the light of the pandemic both for the personal sector and for businesses. The Bank of England also intervened heavily through three waves in 2020 of quantitative easing – purchasing of government and other corporate bonds worth in total £450 billion.

2.3.8 United States

The US Federal Governments (Trump and Biden) introduced a succession of mitigation and emergency responses (IMF 2021). The key income support and household measures were:
Coronavirus Preparedness and Response Supplemental Appropriations Act (2020): This included transfers to state level for additional Medicaid spending, as well two weeks paid sick leave, up to six months emergency leave at 2/3 pay for those ‘infected’, food assistance, and transfers to states to fund enhanced unemployment insurance.

Coronavirus, Aid, Relief and Economic Security (CARES) Act (2020): This included $293 billion for one-off tax rebates, $268 billion to expand unemployment benefits, $25 billion for a food safety net for the most vulnerable, $510 billion to prevent corporate bankruptcy, $349 billion in forgivable loans and grants to help small businesses, $150 billion in funding transfers to states and local government, as well as associated support to help the mortgage market and housing sector (discussed below).

Paycheck Protection and Health Care Enhancement Act (2020): This involved $483 billion to support local businesses retain jobs and keep businesses afloat using a range of further interventions similar to those in the CARES Act.

Various Trump Administration Executive Orders, (2020): included measures to circumvent the otherwise ending of programs e.g a further $44 billion in extra unemployment benefits, providing continuing student loan relief, deferring the collection of employee social security payroll taxes, and identifying options to prevent eviction or repossession in the housing market.
Consolidated Appropriations Act (2021) (Trump Administration): This included further enhanced payments of $300 weekly unemployment benefits and direct stimulus payments of $600 per individual – the complete package was worth $868 billion (nearly 5% of GDP).

Biden Administration Economic Rescue Plan (2021) (prior to infrastructure legislation): This delivered another round of assistance worth $1.8 trillion (more than 9% of GDP). The package provided time bound help to the unemployed, other social security recipients and $1600 direct payments to eligible individuals, as well as direct aid to local government and states.
In Q2 of 2020, the US economy contracted by a full 31% of GDP but rebounded strongly thereafter in terms of economic growth and job creation (IMF 2021). Alongside these huge programs, there were also various relevant monetary and fiscal measures introduced after the pandemic struck in early 2020. The Federal Reserve cut interest rates to ultra-low levels, between 0-0.25%. The Federal Reserve also worked with the Treasury under the auspices of the CARES Act to support the flow of credit, to support a variety of loans to the corporate, small business and household sectors, as well as using Fannie Mae and Freddie Mac to assist borrowers by providing forbearance, suspending evictions and modifying loan terms.

We can summarise these eight country stories in the following table. The table, perhaps a little crudely, asks which types of measures predominated in each nation but also whether pre-existing policies were important, if there is any evidence that specific policies were prematurely ended (i.e. where the factors originally prompting intervention remained in evidence), or, indeed that policies introduced as a result of the pandemic are likely to continue in the longer term.

Table 2.1: Income protection variation

2.4 Discussion and implications for housing

2.4.1 Income protection, economic resilience and learning lessons

It is striking that when looking at the national responses we see a combination of modifying existing approaches. This is clearly in the German case but also when we observe the range of income supports already in place, as well as adding new temporary measures to support furloughing, other forms of wage subsidy and help for sickness benefit, combatting isolation and providing more targeted forms of new kinds of benefit help to those on low incomes or newly out of work. In Australia there was a clear sense that support that was more likely to keep people in work during the crisis would significantly enhance the subsequent economic rebound with lower rehiring requirements – lessons learned from previous recessions and uneven recoveries.

There has been much more variety regarding responses to key sectors disproportionately affected by lockdown, the treatment of the self-employed and small businesses. All countries have sought to wind back these measures and most if not all have found they have needed to reintroduce, often scaled back versions in later waves of the virus. To an extent these responses can be attributed to the policy stance of the nations, their set of measures in place prior to 2020. It is also a function of the immediate crisis response in the first months of 2020 which in part reflected the different pace at which the pandemic struck. But it was also unquestionably an emulation of similar income protection policies and goals following macroeconomic and political international co-operation. Moreover, the emergency also facilitated Treasury officials taking the lead in Australia to implement more generous interventions like the JobKeeper package and the doubling of unemployment benefit. Membership of the EU mattered and so did geography (e.g. for New Zealand and Australia’s capacity to enforce travel quarantine and for transmission across porous borders e.g. Canada and the USA). However, geography while significant, was not as important as political will to enact a costly quarantine.

The countries were differentially affected in economic terms by lockdown, with nations like the USA and Spain hit hardest and Australia and New Zealand probably least affected. The initial income protection measures were remarkably effective. In Australia, social housing providers initially anxious about accelerating rent arrears found that, in fact, temporarily higher social security rates saw historic arrears being paid off.

While there was a considerable positive impact associated with the initial interventions in general, we have noted that the pandemic period has also crystalised and worsened many pre-existing sources of inequality and poverty, something reinforced by the k-shaped unequal economic recovery. This helps to explain the controversy in the UK regarding the ending of the enhanced level of Universal Credit but also the interest in countries like Canada to debate making permanent some reforms to insurance and benefits during lockdown. It should not surprise any reader that the consequences of broadly similar temporary measures had different impacts in quite different economies, systems of governance and social welfare systems.

In parallel to the negative shocks and responses to the lockdowns and health crises, we should not be surprised by the strong economic recovery that followed given the depth of macro intervention, countercyclical stimulus, the use of furloughing that kept worders in jobs, and the inherent automatic impulse of economies to positively respond to a sharp contraction (a form of mean reversion). Aggregate economic resilience does not however imply more equal outcomes and should not obscure the fact that several nations recognised the plight of specific groups and supported them through income protection measures but did not make these permanent even though the base payment rates are demonstrably inadequate to prevent poverty.

Moreover, in developing its fiscal tightening measures to help pay for public borrowing, the UK government having debated a range of tax increases or spending cuts opted to raise national insurance for all low to middle income (i.e. up to the top of basic rate income tax levels) households – an echo of the ‘all in it together’ austerity mantra of the Coalition Government in 2010-15.

What are the longer term implications might follow on from this remarkable episode? It is clear that many countries have participated in national experiments to rapidly shift their income protection systems, grafting on and delivering temporary enhancements to what was already there, augmented by completely new supports, bringing together social security, tax and grant making government functions across federal, regional and unitary systems of government delivery. At the same time many citizens of countries like Australia, the UK, the USA, Ireland and Spain were, for the first time, exposed to the levels of support offered by ‘normal times’ domestic income support – and were shocked to realise how little was on offer4In an interview with senior civil servants in the UK for a sister study, this was a key learning point for the government in the first 2020 lockdown – see Gibb, K et al (forthcoming) COVID-19, income maintenance, help with housing costs and market intervention, CaCHE: Glasgow..

In the case of the UK there is a striking contrast between the decade-long rollout of the new universal credit system for working age means tested benefits and the matter of weeks pandemic income protection took to be set up and delivered effectively. That these policies were delivered so well in the general absence of recent detailed preplanning and scenario futures preparation by governments for such pandemic disasters, is all the more remarkable. Unquestionably, there is a huge amount of learning and data from these experiments across the world which should inform future thinking about national social security systems.

The rapid response through income protection while lockdowns were underway was also one characterised by profound uncertainty about the immediate future and swings of policy optimism and negativity regarding the duration of lockdowns and consequent income protection measures. This was worsened by initially overly pessimistic economic and related forecasts around the world (especially for shorter term projections of how national economies would perform). Again, one would hope that there are lessons and evidence to draw for government finance departments and central banks from this emergency period.

Pervasive uncertainty is a part of the fabric of public policy now and economies and governments must surely seek to invest more in preparedness and scenario planning for future crises be they another pandemic, financial crisis or another form of disaster. Learning lessons from what went well and badly, given the fundamental policy settings for key areas like income protection and fundamental public services, is a third example of drawing on the evidence of the pandemic (and also connecting and comparing lessons from the GFC – Earley, 2021).

One area that has been widely problematic internationally concerns the controversy over the economy-public health trade-off (and whether it exists) and this has been the main area where populist political debates, ideology and the tensions between subject experts and politics have become most difficult and intractable. Governments have been accused, on the one hand, of going into lockdown too slowly, leaving too quickly and having to repeat this cycle several times. At the same time, governments have a strong need to ‘re-open’ the economy, schools and facilitate normal economic activity. They also have to deal with the uncertainties inherent in vaccine roll-out and efficacy as well as the succession of different new variants. Would a more consistent approach to income protection over the entirety of the pandemic been more effective simply in terms of supporting the economic recovery and citizen wellbeing (i.e. not just in redistributive terms)? The evidence from those nations which did seek to retain and evolve such policies, as well as those, like Australia who arguably withdrew too early only to have to bring back other measures later on, will all be instructive.

2.4.2 Housing

Complementing the income protection measures are the key housing interventions we discuss in this report. This can be viewed in two directions. On the one hand, protecting disposable incomes is critical for maintaining housing affordability when otherwise high housing costs and growing arrears would cause critical problems for many households under lockdown. At the same time, mortgage holidays or deferred payments, increased housing-related benefits as well as greater security measures in the private rented sector, all contribute to protecting more marginal households struggling under the pandemic. Removing or reducing housing affordability pressures is evidently an important element of making lockdown and more time working at home, etc. more acceptable and sustainable.

As we discuss later, the housing market has been at the forefront of post-COVID economic recovery with strong price and rent increases, including surging real terms increases in parts of the USA, in Canada, Australasia and across Europe. Private rents are also rising strongly in many cities examined such as Dublin, many parts of Australia, the USA and Canada. Markets have shifted away from city centres towards larger houses in more suburban and rural locations (a phenomenon found across the developed world – Duca, et al, 2021) and segmenting the housing market and widening wealth inequalities. In hindsight, the UK decision to reduce transaction taxes temporarily to support the housing market now looks both unnecessary and wasteful and may have increased the volatility of the market. Similarly, the Australian Government’s HomeBuilder house purchase or renovation grant program (see Section 5.4.6).

The economic recovery from COVID can be further constructively supported by housing activity. Two areas in particular are the opportunities for planners and developers to facilitate rental investment (from market to affordable and social), mixed tenure and in-fill sites in city centres that were hitherto largely non-residential. This could also support regional economic productivity and provide less expensive housing for workers (Maclennan, et al. 2021). Second, the growing importance of housing retrofit to meet climate change targets affords opportunities to build back the economy with labour intensive fabric first construction work and manufacturing/installation and maintenance opportunities in residential renewable energy systems. In British city regions and internationally through developing green mortgage finance, these models support net zero aspirations, the economic recovery from COVID and economic restructuring.


Aspalter, C (2019) ‘Welfare regime analysis: 30 years in the making’, International Social Work, 62 (1) 76-88.

Beirne, K., Doorley, K., , Regan, M., Barra, R. & Tuda, D. (2021) The Potential Costs and Distributional Effect of COVID-19 Related Unemployment in Ireland %281%29.pdf

BBC (2021) “COVID map: Coronavirus cases, deaths, vaccinations by country”

Biddle, N., Edwards, B., Gray, M., & Sollis, K. (2020) COVID-19 and mortgage and rental payments: May 2020 ANU Centre for Social Research and Methods

Bloomberg (2021) Ireland 2020 GDP Grows 3.4% Despite COVID; Economy Driven by Exports. Accessed online

Clark, A., Ambrosio, C., & Lepinteur, A. (2021). The Fall in Income Inequality during COVID-19 in Five European Countries.

Department of Finance (2021) Minister Donohoe notes strong growth in GDP and encouraging indicators for the domestic economy

Deutsche Bank (2021) Outlook for the Germany Residential Property Market.

Duca, J, Hoesli, M and Montezuma, J (2021) ‘The Resilience and Realignment of House Prices in the Era of COVID-19’, Journal of Real Estate Research, online first, DOI 10.1108/JERER-11-2020-0055

Earley, A (2021) Housing Systems, Their Institutions and Their Resilience. CacHE: Glasgow.

El Pais (2021) COVID-19: Coronavirus pandemic pushed poverty in Spain to highest levels since the Great Recession | Economy and Business. Accessed online

Esping-Andersen, G (1992) The Three Worlds of Welfare Capitalism. Princeton University Press.

Euronews (2021) “COVID-19: Spain sees highest youth unemployment rate in EU as pandemic hits hard”

International Monetary Fund (2021) Policy Responses to COVID-19.

Irish Times (2021). Household income would have fallen by 20% without COVID supports, study shows.

Maclennan, D, Long, J, Pawson, H, Randolph, W, Aminpour, F and Leishman, C (2021) Housing: Taming the Elephant in the Economy – A Report to the Housing and Productivity Commission.

OECD (2021) OECD Housing. Policy Toolkit – Synthesis Report.

ONS (2021) International comparisons of GDP during the coronavirus (COVID-19) pandemic.

Pawson, H (2021) New Zealand Working Paper COVID-19 Internal Project Paper.

Pawson H, Martin C, Sisson A, Thompson S, Fitzpatrick S & Marsh A 2021. COVID-19: rental housing and homelessness impacts – an initial analysis. Report no. 7 for the ACOSS–UNSW Poverty and Inequality Partnership

Pennington, A and Stanford, J (2021) International COVID-19 Income Supports: an Update. Briefing Paper, November 2021. The Australian Institute Centre for Future Work.

Phillips, B., Gray, M., & Biddle, N. (2020) COVID-19 JobKeeper and JobSeeker impacts on poverty and housing stress under current and alternative economic and policy scenarios; ANU Centre for Social Research Methods


Key points

  • Eviction moratoriums were implemented in all eight case study countries – a remarkable common response reflecting the early pandemic impulse to ‘stay home’ and, for most countries, the only low or moderate level of security afforded tenants by existing regulatory regimes. Many also intervened in relation to rents to a degree not seen in decades.
  • The ways in which countries responded varied considerably. Germany and New Zealand’s ‘short-run’ moratoriums were over in three months. In Canada, Ireland and Australia, moratoriums and transitional provisions of varying strength and coverage extended somewhat longer into the ‘middle run.’ Spain, the UK and the US had ‘long-run’ responses, repeatedly extending measures or implementing new ones.
  • Where available, national data shows the moratoriums had varying degrees of impacts on termination proceedings and evictions – but always at least a substantial reduction.
  • There is, as yet, little sign of enduring or permanent regulatory reforms arising from the experience of the emergency. Some countries have improved tenure security, but mostly by implementing reforms proposed before the pandemic. There has been less reform directed at rents; although several governments deployed rent regulation against the early ‘income shock’ of the pandemic, few have sought to use them against the later ‘rental cost shock’ caused by shifting demand.

3.1 Chapter introduction

In addition to emergency measures to support household incomes, many governments in our case study countries implemented crisis responses specifically aimed to protect rental housing residents. The most significant of these were eviction moratoriums, and measures to regulate rents and/or provide relief from rental liabilities5Other emergency measures were enacted in some jurisdictions, such as temporary restrictions on access to premises by landlords and agents, and alleviation of some obligations relating to repairs and maintenance; these are not discussed further here.. These moves reflected the remarkable common impulse of people the world over to ‘stay home’, suppressing social and economic intercourse in order to suppress transmission of the virus (Tooze 2021). More specifically, they acknowledged the vulnerability of many private sector renters to the income shock resulting from the suppression of economic activity, and the prospect of sudden, widespread rent arrears and evictions.

In many – but not all – of our case study countries, social as well as private rental housing will have been covered by pandemic-triggered temporary restrictions on landlord freedoms. However, once again reflecting the understanding that exposure to economic hardship will have been concentrated among the low income workers of the private rental market, this chapter generally focuses on that sector rather than public or not-for-profit housing.

In most of our case study countries, the risks posed to private tenant security and affordability by COVID-19 were underpinned by the shortcomings of existing regulatory regimes. Table 3.1 gives a high-level overview of the relative levels of regulatory assurance of these aspects of renting for each of the case study countries6While the prime focus here is on the private rental market, regulatory frameworks in some countries and in certain aspects encompass both private and social sectors, necessarily passing over some significant differences in the legal mechanisms used between countries, and some significant differences between jurisdictions within countries.

For indicators of tenure security, the table indicates whether landlords may take termination proceedings without disclosing reasons or where these may be invoked only on certain legally prescribed grounds, and the degree to which tenants’ circumstances are considered by the relevant tribunal in termination proceedings. This may be limited by laws that make termination in some rent arrears and no-grounds proceedings mandatory (as in most jurisdictions in Australia, the UK and the US), that restrict the tribunal’s discretion to decline termination (as in New Zealand), or that put the onus on tenants to dispute termination proceedings (as in some Canadian provinces, Ireland and Spain).

Table 3.1. Pre-pandemic tenure security and rental affordability, eight case study countries

Like the impulse to ‘stay home’, governments’ immediate moves to prevent evictions were a remarkable common feature of early responses to the pandemic: by the end of March 2020, eviction moratoriums had been announced in all our case study countries. The restrictions themselves, however, varied substantially – including within countries, where they differed over the phases of the pandemic, and differed by sub-national jurisdictions – in terms of coverage (e.g. status of social housing), strength of protection (e.g. types of tenancy termination outlawed) and duration. Another dimension of difference was in the legal mechanisms used for temporary alteration of existing regulatory rules – whether legislated, or effected by executive action. There was also variation in the extent of emergency deviation from regulatory norms regarding rent payment obligations – through restrictions on rent increases, rent variations .

The UK and the US stand out for the complexity, if not incoherence of their responses, which were marked by multitudinous legal instruments, last-minute extensions, and uncertainty – but which evidently prevented many evictions that would otherwise have occurred.

Figure 3.1. Timeline of eviction moratoriums, eight case study countries, 2020-21

As Figure 3.1 indicates, the case study countries can be roughly grouped by the timeframes of their eviction moratoriums: the short-run (Germany and New Zealand), the middle run (Australia, Canada and Ireland) and the relatively long-run (Spain, the UK and the US). The chapter examines each group in turn.

3.2 Short-run regulatory responses: Germany and New Zealand

Of our eight case study countries, Germany’s regulatory response was probably the simplest. In March 2020, the federal government (which has primary responsibility for rental housing regulation) passed emergency legislation stopping, for three months, termination proceedings for rent arrears arising from the economic effects of the pandemic. This short moratorium was followed by a longer period (scheduled to run to 30 June 2022) in which landlords are precluded from taking eviction proceedings for arrears accrued during the moratorium (Ungerer 2020; Ortmanns and Beisken 2021). Considering the relative brevity of the moratorium period, it must be supposed that any arrears accrued during that period would be relatively small. No special measures regulating rents were introduced. In the view of our country expert, Germany’s emergency response was designed around a regulatory regime that was already relatively protective.

New Zealand’s moratorium was similarly short: a legislated national moratorium that for three months stopped the commencement and enforcement of eviction proceedings on most grounds. Prohibited grounds included the landlord or an incoming purchaser using the premises for their own housing, reflecting the heavy restrictions on household movement imposed by New Zealand’s public health orders. Termination proceedings for anti-social behaviour and other urgent grounds remained allowable, as well as for rent arrears – but only where tenant arrears equated to more than 60 days. Rent increases were prohibited, and the government encouraged rent negotiations between landlords and tenants in hardship, but did not set out a special regime or guidance for them. It also temporarily doubled (to NZ$4,000) the maximum amount lent to tenants under its Rent Arrears Assistance scheme. The moratorium expired 25 June 2020.

Subsequently, in August 2020, the government enacted a wider program of tenancy law reform proposed prior to the pandemic, effective February 2021. This included the abolition of no-grounds terminations, significantly improving tenure security. In October 2021, the New Zealand Parliament passed further amendments, empowering the housing minister to make ‘COVID-19 tenancy orders’ for area-specific eviction moratoriums where public health orders also restrict movement. To date these powers have not been used.

As shown in Figure 3.2, landlords’ applications for rent arrears and termination orders dropped by 55% and 67% respectively during New Zealand’s moratorium compared to the previous quarter (the previous year’s data is not available). Subsequently, application rates rose only moderately in late 2020, before easing once again in 2021.

Figure 3.2

3.3 Middle run regulatory amendments: Canada, Australia and Ireland

The emergency response in Canada would be characterised as short-run were it not for the somewhat longer response in the confederation’s largest province, Ontario. Canada’s eviction prevention measures were determined and implemented at the province/territory level, with all but Nunavut and the Northwest Territories announcing moratoriums in March 2020. The moratoriums were relatively broad, with provincial tribunals suspending eviction proceedings on all but urgent grounds. However, the measures were mostly short-lived: half had ended by mid-year and all had ended by September 2020. In British Columbia, tenants in arrears who made regular repayments continued to be protected from eviction through to June 2021. Three provincial governments (Alberta, British Columbia and Manitoba) also prohibited rent increases; all had lifted this restriction by the end of September 2020.

In January 2021, and again in April 2021, Ontario halted eviction proceedings on the reintroduction of stay at home orders, with this new moratorium ending June 2021. Ontario set its rent increase guideline for 2021 at 0%.

Published statistics on Canadian eviction proceedings are not available. Canada Mortgage and Housing Corporation survey data indicate six per cent of rental properties in Canadian metropolitan areas were occupied by tenants in arrears at October 2020; no data are available for the previous year but a slight majority of property owners (58%) reported arrears were more common than in the previous year; 33% said it was the same (CMHC 2021).

Ireland’s COVID-19 eviction moratorium, enacted 27 March 2020, was both broad and strong, prohibiting landlords from giving termination notices on any grounds and stopping all evictions for the duration of the (initial) emergency period (see Figure 3.3). Rent increases were also prohibited. These prohibitions expired 1 August 2020. However, provision was then made for tenants in COVID-related hardship to claim, by declaration to their landlord and the Residential Tenancies Board, protection from eviction and rent increases until July 2021 and, if termination proceedings were taken then at that stage, a longer 90-day notice period. The declaration required the tenant to make contact with advice services to assist making repayment plans. The Irish Government did not implement new rent relief schemes, but did change eligibility rules to make the Rent Supplement payment available to persons receiving the Pandemic Unemployment Payment.

Figure 3.3

In October 2020, the Irish Government passed further legislation providing for eviction moratoriums to be triggered whenever public health orders restricted travel to a 5km radius. These provisions were triggered and eviction moratoriums were in place October-December 2020, and again December 2020 – April 2021. Under these moratoriums, termination proceedings were allowed on grounds of anti-social behaviour and rent arrears, although tenants in COVID-related hardship were again entitled to protections by declaration, which were extended to 12 January 2022. For termination on most other grounds the moratoriums effectively stopped the clock, with notice periods starting again after the end of the emergency period plus a 10-day statutory grace period.

In Australia, a six-month eviction moratorium was announced 27 March 2020 by the National Cabinet (comprising the federal government, and the governments of the six states and two territories). Each of the states and territories then legislated measures of their own design, differing in scope, strength, and other details. Tasmania’s was the most comprehensive, with numerous grounds for termination, including rent arrears, suspended for all tenancies. Other jurisdictions distinguished a core group of tenants in COVID-related hardship for special protection from eviction, either by prohibiting termination proceedings for rent arrears and without grounds (Queensland, Victoria and Western Australia), or the lesser protection of longer notice periods and additional scrutiny of proceedings in conciliation and tribunal hearings (New South Wales). Some jurisdictions (New South Wales, Victoria and Western Australia) also provided a lesser degree of additional protection to tenants more widely, through longer notice periods and greater scrutiny of termination proceedings (Pawson et al. 2021a and b; Martin 2021).

Five Australian jurisdictions also prohibited rent increases during the emergency period (Victoria, Western Australia, South Australia, Tasmania and the ACT, with the ACT’s prohibition applying to COVID-impacted tenants only). At the same time governments generally encouraged landlords and tenants in hardship to negotiate variations to rent obligations, and four (New South Wales, Queensland, Victoria and Western Australia) implemented formal processes to conciliate variations.

Most jurisdictions extended their moratoriums past the original six-month timeframe, but by March 2021 all but South Australia and the Northern Territory had ended (SA and the NT have continued their modest provisions to date). Restrictions on rent increases were lifted by then too (South Australia narrowed its prohibition to COVID-hardship cases in September 2020). Transitional provisions were implemented in New South Wales, Victoria and the ACT, protecting tenants with COVID-related arrears from the emergency period from proceedings provided tenants resumed payments (although in Victoria the protection merely prevented proceedings for payment, and still allowed eviction).

During Australia’s mid-2021 Delta variant outbreak, New South Wales and the ACT reintroduced eviction moratoriums (in July and August 2021 respectively), focused narrowly on COVID-impacted tenants; both were lifted in November 2021. Notwithstanding the continuing (modest) measures in South Australia and the Northern Territory, Australia’s use of restrictions against eviction lightened dramatically in 2021, even as government continued, through public health orders, to restrict movement and intercourse to suppress transmission.

Assessing the impact of Australia’s moratoriums is difficult because of the paucity of official data on tenancy terminations and evictions. As we have reported elsewhere (Pawson et al 2021b), South Australian tribunal data shows landlords’ applications for termination were down 33% in Q4 2020 compared with the same quarter in the previous year. Nevertheless, over 3,000 tenancies were subject to termination proceedings following commencement of the ‘moratorium’.

The incompleteness of Australian eviction moratoriums is also evident in another data fragment, this time relating to New South Wales: in a 10-week period from 26 June 2020, when Sydney entered its long Delta lockdown, the state’s tribunal heard 4,581 applications to terminate tenancies, ordered termination in 1,412 in cases, and 141 evictions were conducted (NSW Legislative Council 2021b). In interviews with representatives of tenants and landlords (Pawson et al. 2021a and b) we found qualified support for the moratoriums from both landlord and tenant advocates. These were considered a readily understood intervention that had effectively protected tenants at a crucial time in the emergency. The main qualifications on support related to moratorium incompleteness, and/or the weakness of the response on rent liabilities, which undermined moratorium effectiveness.

Governments gave little guidance to variation negotiations, and only a small minority of tenancies – between 8-16%, according to different research sources – got a rent variation, with more tenants reporting they were refused or discouraged from asking (Martin, et al. 2021). Most state governments and the ACT also implemented rent relief schemes, delivered through cash payments or land tax rebates, where variations were negotiated for tenants in hardship. Except in Victoria, however, these schemes were significantly undersubscribed (Pawson et al. 2021b). Tenant representatives reported tenants in hardship who could not negotiate rent variations anticipating eventual eviction and moving out themselves.

3.4 Longer-lasting (but still temporary) responses: Spain, the UK and the US

Spain was one of the worst affected countries in the early months of the pandemic, with many households, businesses and agencies shutting down in the first quarter of 2020. Spain’s formal eviction moratorium was introduced by Royal Decree-Law 11/2020 of March 31, originally for two months, but extended repeatedly through 2020 and 2021. At this writing, the moratorium is due to expire 28 February 2022.

The Spanish moratorium has several aspects, and more than our other case study countries, Spain combined eviction protection with regulatory adjustments on rents. Under the terms of the moratorium, ‘vulnerable’ tenants (defined by different household income thresholds, experiencing a loss of income of 40%) have been allowed to apply to courts for the suspension of eviction proceedings. The application triggers a referral to the local social services agency, which prepares a report for the court; this, along with evidence from the tenant and the landlord of their respective hardships is considered by the court in determining the suspension.

For vulnerable tenants of ‘large landlords’ – defined as owners of more than ten urban properties, or properties with a constructed area of more than 1,500 square metres – the eviction moratorium is also a ‘rent debt moratorium’. These tenants may request a 50% reduction in rent; if the landlord does not agree, a moratorium on payment applies as long as the tenant remains vulnerable, up to the end of the emergency period or nine months, with the postponed rent to be repaid over not less than three years.

More widely, Spain’s emergency measures have also provided for the fixed terms of tenancy agreements to be extended, at the request of the tenant, for up to six months in each instance. This prevents landlords both from giving no-grounds termination notices, and from increasing rents between fixed terms.

From October 2020, the Spanish government has paid compensation to landlords of non-paying vulnerable tenants protected by the moratorium, based on losses relative to average rents. It also limited the application of the eviction moratorium to lawful tenants, after squatters were able to claim protection under it first version. According to one of our country experts, both the eviction and ‘rent debt’ moratoriums were politically highly controversial, and perceived by owners as an appropriation of property. The Spanish Government, which includes a strongly anti-austerity socialist party as junior coalition member, may have been less concerned about antagonism with property owners than governments in other case study countries.

Official statistics – see Figure 3.4 – show that Spanish evictions in the second quarter of 2020 dropped dramatically – especially evictions for rent arrears, which were down 90% year on year. Subsequently, however, evictions rose, to 75% of the previous year’s level in Q3, and exceeding the previous year’s level in Q1 2021.

Figure 3.4

UK eviction moratoriums implemented in 2020 and 2021 employed a complicated mix of legal mechanisms. On 26 March 2020, the UK Government announced the suspension of all eviction proceedings in England and Wales for 90 days, subsequently extended to 24 August and (three days before expiration) again to 20 September 2020. the UK Government also extended by regulation the notice period required for most termination actions in England to three months, later extending the period for rent arrears terminations and for no-grounds terminations to six months, subsequently dialled back to four months. Extended notice periods were also implemented by the governments in Wales and Northern Ireland.

When proceedings resumed in September 2020, Ministers directed that ‘egregious’ cases be prioritised, including anti-social behaviour cases and rent arrears exceeding 12 months, with the understanding that the majority of less significant cases would take longer than usual to proceed. The government also sought to apply additional scrutiny and justification to the eviction process in England through the use of pre-action protocols. A device imported from social housing practice (Pawson et al. 2010), pre-action protocols as applied here have aimed to elicit information about tenants’ COVID-related hardships and require landlords to countenance alternatives to eviction. However, because termination is mandatory in some proceedings, including rent arrears and no-grounds proceedings, it was doubtful that English tribunals could give the requirements of the protocols effect (Watts et al. 2021).

From November 2020, the government implemented more substantial protections, this time by prohibiting bailiffs in England from enforcing eviction orders in most circumstances (exceptions included rent arrears exceeding nine months); originally due to expire January 2021, the prohibition was extended three times, eventually ending May 2021.

The Scottish Government implemented its own eviction moratorium, using similar measures, if rather more straightforwardly applied. Eviction proceedings were suspended at the outset of the pandemic, with effect to 9 July 2020; when they resumed, termination notice periods were increased and tribunals afforded more discretion in ordinarily mandatory termination proceedings, which also gave new pre-action protocols some teeth. From December 2020 to January 2021, enforcement of most eviction orders was suspended, and subsequently provision was made for prohibitions on enforcement in areas subject to movement restrictions. Those arrangements ceased 30 September 2021.

UK governments did not regulate rents during the emergency, and a public campaign for the suspension or cancellation of rent arrears was rejected by major parties; the UK Parliamentary Opposition’s stated reasons included that such action would infringe the right to property ownership under the European Convention on Human Rights. Instead, various measures of financial assistance were implemented: in March 2020, Local Housing Allowance was increased to cover 30% of market rents (restoring an earlier link, severed in 2016), and in late 2021 English local authorities were allocated additional funds to support low-income tenants still in rent arrears. The Scottish and Welsh Governments implemented loan schemes for the payment of arrears: the Scottish scheme paid up to nine months’ rent arrears, for repayment over a period of up to five years.

Government data on termination proceedings (‘possession claims’) in England show a dramatic reduction in the pandemic period (see Figure 3.5). Possession claims by all landlords in Q2 2020 were down 88% on the same quarter the previous year, and claims for 2020-21 were down 79% on the previous year; however, the largest reductions were by social housing providers, not their private landlord counterparts. Claims by private landlords or otherwise using the ’accelerated procedure’ (largely no-grounds claims by private landlords) were down 80% in Q2 2020 year on year, and by the fourth quarter the rate was down 40% year on year.

Figure 3.5

In the US, multiple agencies and levels of government have effected eviction moratoriums over the course of the pandemic. The first were de facto moratoriums resulting from court closures across the country in Q2 2020, or where courts decided to defer evictions as non-essential proceedings, in the early days of the pandemic. In March and April 2020, 43 states (and the District of Columbia), as well as numerous municipal jurisdictions, also enacted formal moratoriums, which operated variously by prohibiting landlords from initiating eviction proceedings, or by suspending proceedings or the enforcement of eviction orders (Hepburn et al. 2021). From a comprehensive scoring of state and municipal moratoriums, Benfer characterises most such measures as ‘heterogeneous, patchwork in nature, and largely short-lived, falling far short of model protections’ (Benfer 2021). Most of these local restrictions had, in any case, lapsed by July 2020.

Nationally, the federal congress included eviction moratorium in the Coronavirus Aid, Relief, and Economic Security (CARES) Act. Commencing 27 March 2020, the CARES Act eviction moratorium applied in relation to ‘federally related properties’, including properties supported by federally-backed finance or occupied by tenants with Housing Choice vouchers. It was estimated that these rules achieved coverage of between 28% and 46% of all US renter households (Congressional Research Service, 2020). Where it applied, the moratorium prohibited the commencement and enforcement of eviction proceedings for unpaid rent until 23 August 2020.

Two weeks later, from 4 September 2020, a second national moratorium was imposed, this time by the federal government’s Centres for Disease Control (CDC). This covered all tenants meeting certain income and hardship criteria, including that they had applied for government assistance and would be at risk of homelessness or overcrowding if evicted. Tenants seeking the moratorium protection were required to declare their eligibility to their landlord. Like its CARES Act predecessor, the CDC moratorium prohibited landlords from commencing or enforcing eviction proceedings for non-payment of rent. In Benfer’s (2021) assessment, the requirement for pro-active tenant engagement through submitting to their landlord a written declaration was a distinct weakness in the moratorium. More generally, ‘the patchwork of local, state, and federal moratoria [implemented in 2020 and 2021 proved to be] porous, needlessly complicated, poorly understood, and rarely enforced’ (ibid). Partly for these reasons, local judges differed widely in their understanding of the moratorium and their willingness to enforce it (academic expert interview).

Originally scheduled to expire on 31 December 2020, the CDC moratorium was extended by the US Congress to 31 January 2021, and prolonged three times subsequently by the CDC. After the CDC second extension, the US Supreme Court narrowly rejected a challenge to the moratorium brought by real estate organisations, flagging that any further extensions not expressly legislated would be regarded as exceeding CDC powers. On 23 August 2021, after the third unlegislated extension, the US Supreme Court duly struck down the moratorium. At that time, some state- and municipal-level moratoriums continued; at this writing (early 2022) a handful of municipal moratoriums (mostly in California) remain in place.

In addition to the eviction moratoriums, rental assistance was provided by different levels of government over 2020-21, with similarly chequered implementation. Throughout 2020, rental assistance schemes were implemented by numerous municipal governments, many using federal funds provided under the CARES Act but operating on different terms.

As the original end date of the CDC moratorium neared, the federal government announced an Emergency Rental Assistance (ERA) package, commencing January 2021 with a further supplement in March, totalling $US45 billion – understood to approximate the amount of arrears resulting from the emergency in 2020 (Parrott and Zandi 2021). The ERA was targeted to ‘households with incomes at or below 80% of area median income (AMI) that have experienced financial hardship due to COVID-19 and are at risk of homelessness or housing instability’ (Reina et al. 2021). In practice, however, ERA payments have been usually channelled direct to landlords, with tenants receiving help directly only where landlords declined to participate (e.g. because of reluctance to accept accompanying conditions such as desisting from eviction filings). Payments have been mainly calibrated in relation to arrears (typically extending back over six months), with ongoing rent payment assistance usually limited to three months into the future.

The Congressional decision to delegate ERA implementation to the states followed the precedent of some other large Federal funding programs (e.g. Medicaid). In this instance, however, the model has proved administratively problematic. Even by August 2021 only $US8 billion of the $US45 billion allocated to the program had been spent ‘meaning renters who could be protected from eviction with the funding might be forced out of their homes because of administrative delays’ (Holpuch 2021). Our country expert laid blame for the slow roll-out on the decision not to use existing federally-funded housing authorities to distribute relief, but to set up the schemes from scratch. By late 2021, the flow of relief funds was increasing.

Despite the complicated and contested ways in which they were implemented, it is evident that US emergency responses, especially the eviction moratoriums, had a very significant impact on tenure security. From analysis of evictions data sourced by Eviction Lab, Hepburn et al. (2021) calculate that eviction application rates in the period March-December 2020 were 65% below their historical average, with the greatest reduction in the period of the CARES Act moratorium and early sub-national moratoriums. The spike in eviction applications in the two-week gap between the two national moratoriums (see Figure 3.6) also indicates the strong direct effect of these restrictions. Over the period of the CDC moratorium, Rangel et al. (2021) calculate that eviction applications were down 53% on historical average levels. After its sudden cancellation, eviction application rates lifted, although only to 63% of their historic average (Hass et al. 2021). Over the combined moratorium periods an estimated 2.45 million evictions had been avoided by early 2021 (Rangel et al. 2021).

Figure 3.6. Weekly eviction fillings in 2020 and 2021 as a % of pre-pandemic norm

The extensive US data have also been used to estimate the public health impacts of the moratoriums. In the face of a catastrophic public health situation, the evidence shows eviction restrictions measurably contributed to effective disease control. Hill et al. (2021) found that ‘evictions lead to significant increases in infections’, and that policies to stem evictions are a warranted and important component of COVID-19 control’ (ibid p1). Meanwhile, Jowers et al. (2021) estimated that the eviction moratoriums in 2020 reduced coronavirus infections by 3.8% and deaths by 11% by comparison with the counter-factual scenario. Moreover, Leifheit et al (2021) found that where local eviction moratoria were lifted COVID-19 incidence and deaths increased relative to localities where restrictions remained in place.

3.5 Policy development implications

We have characterised some countries’ regulatory responses as ‘long-run’, but that’s only relative to the shorter-lived responses of other countries, not to the continuing pandemic. Now, in early 2022, almost all the responses are expired, with some having ended prior to further major waves of infections and lockdowns – phenomena that prompted no restoration of emergency restrictions.

The impact of the emergency measures on rental housing policy development for the longer term also appears limited. It is true that some case study countries – most notably Ireland, New Zealand and the Australian state of Victoria – have introduced or pledged permanent improvements to tenure security since March 2020. However, most such moves were already in train prior to the pandemic. Also, particular features of some countries eviction moratoriums – preliminary conciliation and consideration of alternatives to termination, affording tribunals more discretion to decline to terminate – have not featured strongly in new and proposed measures.

Ireland has pledged significant reforms to rental housing regulation following the pandemic experience. In December 2021, the Irish Parliament legislated to abolish no-grounds terminations after the first fixed term of a tenancy, ending their previous cyclical availability, with effect from June 2022. Also in Northern Ireland, where the temporary extension of notice periods to 12 weeks has been repeatedly extended (to May 2022), a bill is currently before the Assembly to make the longer periods permanent. Otherwise, pre-pandemic proposals have been legislated in New Zealand (abolishing no-grounds terminations), and the Australian states of Victoria (abolishing no-grounds terminations after the first fixed term) and Queensland (a modest reform: allowing no-grounds only at the end of fixed terms). Similarly, in February 2022 the UK Government re-confirmed a previously stated intention to abolish no-grounds terminations in England and Wales, and in the US proposals to expand ‘just cause’ evictions are before the New York State Assembly.

On the other hand, there are also examples of reduced tenure security after the pandemic. The Queensland regulatory amendments provide new grounds for termination during fixed term tenancies, including preparation of the premises for sale – a ground first introduced in its eviction moratorium. In Ontario, the provincial government implemented a pre-pandemic proposal to reduce landlords’ termination notice obligations and facilitate evictions shortly after its first eviction moratorium – a move that attracted criticism and in turn may have contributed to government acceding to calls for the province’s second moratorium in early 2021.

It is a similar story regarding rent regulation. Although several governments deployed rent regulation against the early ‘income shock’ of the pandemic, few sought to use them against the later ‘rental cost shock’ caused by shifting demand. Ireland, again, is an exception: its December 2021 legislation reduces to 2% the caps on rent increases in rent pressure zones. Another exception is Spain, which maintained restrictions on rent increases and the partial rent moratorium over the course of 2020-21, and in late 2021 indicated it was considering rent caps on large landlords, but this has yet to be legislated. The current New York proposals would also expand the coverage of its rent stabilisation regime. But in New Zealand, February 2022 saw the Prime Minister ruling out rent controls (after the housing minister suggested ‘nothing is off the table’ in responding to growing affordability problems). Most governments were evidently more comfortable with providing financial assistance to help pay rents than restrict or vary rental liabilities, but the pandemic experience of establishing novel relief schemes has also not as yet led to permanent changes in rental assistance.


Benfer, E. (2021) How Policymakers (and Courts) Sabotaged Eviction Moratoria; The Appeal, 2 April

Congressional Research Service. (2020) CARES Act Eviction Moratorium.

EP Data (2021) Evictions, statistics, data and graphs: Spain.

Hepburn, P., Louis, R., Fish, J., Lemmerman, E., Alexander, A.K., Thomas, T.A., Koehler, R., Benfer, E., and Desmond, M. (2021) US Eviction Filing Patterns in 2020; Eviction Lab, 27 April

Hill, A.L. (2021) The effect of eviction moratoria on the transmission of SARS-CoV-2; Nature Communications 12: 2274

Jowers, K., Timmins, C., Bhavsar, N., Hu, Q. and Marshall, J. (2021) Housing Precarity & the COVID-19 Pandemic: Impacts of Utility Disconnection and Eviction Moratoria on Infections and Deaths Across US Counties; National Bureau of Economic Research Working Paper

Kenna, P., Nasarre-Aznar, S., Sparkes, P. and Schmid, C. (eds) (2018) Loss of Homes and Evictions across Europe. Cheltenham: Edward Elgar.

Leifheit, K.M., Linton, S.L., Raifman, J., Schwartz, G.L., Benfer, E.A., Zimmerman, F.J. and Pollack, C.E. (2021) Expiring Eviction Moratoriums and COVID-19 Incidence and Mortality; American Journal of Epidemiology – Accepted paper

Martin, C. (2021) Australian Residential Tenancies Law in the COVID-19 Pandemic: Considerations of Housing and Property Rights.; UNSW Law Journal, vol. 44 no. 1: 197-226.

Martin, C., Hulse, K. & Pawson, H. (2018) The changing institutions of private rental housing: an international review; Final Report no. 292; Melbourne: AHURI

NSW Legislative Council (2021)‘Answer to Question on Notice – Attorney General – Evictions’ 20 August 2021, Parliament of NSW, Sydney.

Pawson, H., Martin, C., Sisson, A., Thompson, S., Fitzpatrick, S. and Marsh, A. (2021a) COVID-19: Rental housing and homelessness impacts – an initial analysis; ACOSS/UNSW Poverty and Inequality Partnership Report No. 7, Sydney

Parrott, J. and Zandi M. (2021) Averting an eviction crisis

Pawson, H., Martin, C., Thompson, S., Aminpour, F. (2021) COVID-19: Rental housing and homelessness policy impacts in Australia; ACOSS/UNSW Poverty and Inequality Partnership Report No. 12, Sydney

Pawson, H., Sosenko, F., Cowan, D., Croft, J., Cole, M. and Hunter, C. (2010) Rent Arrears Management Practices in the Housing Association Sector; London: Tenant Services Authority

Rangel, J., Haas, J., Lemmerman, E., Fish, J. and Hepburn, P. (2021) Preliminary Analysis: 11 months of the CDC Moratorium; Eviction Lab, 21 August

Residential Tenancies Board (Ireland) (2019-2021) Dispute Resolution Statistics.

Tenancy Services (New Zealand) (2020-2021) Dispute Resolution Statistics.

Tooze, A. (2021) Shutdown: How COVID Shook the World’s Economy. New York City: Viking.

Ungerer, J. (2020) Coronavirus Moratoria in Private Law: Comparing the Position of Tenants in the UK and Germany.

Watts, B., Bramley, G., Pawson, H., Young, G., Fitzpatrick, S. & McMordie, L. (2022) Homelessness Monitor England 2022; London: Crisis

Key points

  • There is (at least as yet) little evidence from any of the case study countries of a pandemic-triggered surge in newly arising homelessness; in North America, however, the crisis may have compounded unsheltered homeless numbers.
  • COVID-19 enhanced the public visibility of rough sleeping, or ‘street homelessness’ as a social problem and, in countries like Australia and the UK, more graphically exposed the disproportionate representation of non-citizens within homeless populations.
  • In all of the countries covered in the research, early 2020 saw substantial emergency action to protect existing homeless populations from elevated health risks posed by the pandemic. In the main, this involved publicly (or philanthropically) funded placements in hotels for rough sleepers, couch surfers and residents of homeless shelters considered to pose health risks.
    Governments were notably quicker to initiate emergency accommodation programs in unitary and semi-unitary states (e.g. England, Ireland, New Zealand) than in federations (e.g. Australia, Canada, USA).
  • In some federal states (Australia, Germany), the crisis exacerbated tensions in federal-state relations regarding division of responsibility for homelessness across levels of government.
  • Epidemiological evidence demonstrates that, by reducing transmission risk, emergency accommodation provision for homeless people measurably reduced the health toll of COVID-19 among those directly affected, as well as benefiting the wider community.
  • At least in Australia and England, the pandemic served as a stimulus for stepped-up assistance to the most vulnerable homeless populations – help that, for some, extended to being aided to secure longer term housing. This is likely to have meant that a substantial number of disadvantaged rough sleepers and others will have gained a settled home who – in the absence of the pandemic – would not have done so.

4.1 Chapter introduction

In many countries, the rapid onset of the COVID-19 crisis in early 2020 prompted alarm about its homelessness implications. These centred, in particular, on two worries. Firstly, that surging unemployment due to economic disruption, through its impact on household incomes, would feed through into reduced ability to pay the rent or mortgage, and ultimately – for some – loss of housing. Secondly, there was the realisation that the pandemic could pose special risks to many ‘already homeless’ people, in terms of their exposure to infection. For many of those in shelter or hostel accommodation, social distancing or self-isolation would be impossible. Similarly, for rough sleepers or ‘unsheltered homeless people’, protecting against COVID-19 demanded hygiene standards that could not be met.

Beyond this, of course, both sheltered and unsheltered homeless populations involve people liable to be in poorer health and at greater risk of mortality than the non-homeless population (Nilsson et al. 2017).

Chapters 2 and 3 have already discussed the ways that pandemic-triggered income support measures and rental housing regulation adjustments helped to mitigate risks that the COVID-19 economic shock would trigger a new wave of homelessness. This chapter therefore centres on the ways that governments sought to protect homeless people from COVID-19 infection – largely through provision of emergency accommodation – and on measures to provide pathways out of EA and into longer term housing. To some degree, active efforts of these kinds were seen in all eight countries covered in this report, albeit highly varied in terms of how they were initiated, undertaken and funded. Our assessment of these programs forms the heart of this chapter.

It should be emphasized at the outset that, at least in several of the countries covered in this report, the pandemic followed on from a period of intensifying housing need and rising homelessness during the late 2010s. In Australia, for example, over the decade from 2006, overall homelessness rose by 30% to 116,000 people on census night 2016 (Australian Bureau of Statistics (ABS) 2018); while rough sleeping in England7Note that in this chapter the focus is on ‘England’ rather than the UK more broadly. This reflects the fact that housing and homelessness policies differ significantly across the four UK jurisdictions and there is insufficient space to cover this diversity here., as officially estimated, was running at over 4,000 in late 2019 compared with less than 2,000 in 2010 (Fitzpatrick et al. 2021a). Meanwhile, New Zealand’s social housing waiting list tripled to 23,000 in three years to 2021 (Corlett 2021), and in the USA unsheltered homelessness rose 31% to 226,000 in the five years to 2020, substantially outweighing a small reduction in sheltered homelessness over the period (US Department of Housing and Urban Development 2021).

The remainder of this chapter is structured as follows. First, in Section 4.2, it looks at the changing incidence of homelessness during the pandemic. Next, in Section 4.3 it describes the EA programs implemented in the case study countries, and the efforts to provide move-on accommodation mounted in some of these. Some reflections on these experiences are presented in Section 4.4, ahead of a brief conclusion in Section 4.5.

4.2 Pandemic impacts on the incidence of homelessness

Although statistics on the incidence of homelessness during the pandemic are only sparsely available across the countries covered by this research, some significant observations can be drawn from the evidence collected for this report.

Statistics on flows of people becoming homeless during 2020 and 2021 are available for Australia and England. Both jurisdictions recorded a significantly decreased incidence of ‘newly homeless’ people during the first 15 months of the pandemic. Relevant agencies in the two countries saw the rate of applications for homelessness assistance fall markedly at the time of initial national lockdowns in Q2 2020, with the numbers remaining relatively suppressed for the remainder of 2020 and the first half of 20-21 – see Figure 4.1

Figure 4.1

As a percentage of the number of homelessness assistance applications during the 15 months preceding the COVID-19 crisis, the number of applications during the first 15 months of the pandemic (April 2020-June 2021) was 88% in Australia and 92% in England. While other factors will have also played a part, the main influence here is probably the COVID-19 rental evictions moratoriums discussed in Chapter 3. More detailed data for England reveals that very large reductions in homelessness arising from the loss of rental housing in 2020/21 were only partially offset by somewhat increased numbers being made homeless due to domestic violence, relationship breakdown and family/friend exclusion (people being asked to leave an existing household by friends or family members) (Watts et al. 2022).

While no comparable data on homelessness flows are available for other countries covered by the current research, reports from North America suggest that – measured in terms of point in time numbers – visible homelessness rose in both the USA and Canada during 2020 and 2021. It has been perceived that unsheltered homelessness rose during the pandemic, at least in certain states. Consistent with this belief, it has been asserted that homeless encampments expanded in 2020:

Before the pandemic, tents could reliably be found in city centres up and down the west coast. Now encampments have cropped up in leafy neighbourhoods all around places such as Los Angeles, Portland and Seattle (The Economist 2021).

Official US homelessness statistics, however, suggest a slightly different story. Albeit that these numbers need to be hedged with many qualifications, the latest national figures reported to the US Congress indicate that sheltered homelessness indeed fell back in the year to January 2021, the point in time national total declining by 8% on the equivalent figure for January 2020 to 326,000 (US Housing and Urban Development Department 2022). At the same time, albeit limited to areas where counts had been conducted, unsheltered homelessness appeared to have remained virtually static at some 233,000 (ibid).

Citing University of Washington housing academic Dr Gregg Colburn, the Economist reported that ‘the pandemic divided the country’s homeless into two groups: those who were able to take advantage of emergency programs, and those who fell through the cracks when shelters shut down’ (ibid). This is a reference to one of the key homelessness policy and practice responses to the pandemic, as further discussed in Section 4.3.

At least during summer, in 2020 and 2021 numerous Canadian cities reportedly saw a rise in visible homelessness in the form of new or expanded temporary encampments:
Encampments are presenting a dilemma for municipal leaders across Canada. Spurred in large part by the pandemic, during which some homeless people have avoided shelters over fears of catching COVID-19 in such congregate settings, clusters of tents have become common sights in multiple cities (Moore and Gray 2021).

Albeit with some nuance, this perspective was echoed in interviews with Canadian housing experts undertaken for this research:

There’s clearly been an increase in visible homelessness – that is without a doubt, for sure (housing industry representative body interview)

I’m not sure homelessness actually rose [during the pandemic]; what happened was that it became more visible … A lot of people [previously housed in shelters] got worried about [the health risks of] congregate living and went outside (Senior City Official interview]

In Toronto, despite 900 people having been rehoused from encampments during 2020, this was reportedly insufficient to offset the associated increase in unsheltered homelessness (Noble and Coplan 2020).

Albeit less well-documented, news reports and expert interviewee testimony suggests that some similar developments were seen in 2020 and 2021 in Spain. While efforts were made to provide emergency shelter for rough sleepers and those displaced from hostels closed for public health reasons (see below), the frequently unsuitable nature of premises used for this purpose, as well as inadequate support arrangements, reportedly led to many residents preferring unsheltered homelessness (Moreno and Roldán 2020).

Over and above fears about infection risks in homelessness shelters, another factor believed to have swelled Canada’s unsheltered homeless population early in the pandemic was the large-scale accelerated release of prisoners near the end of sentences. Early in the crisis Canada saw a 16% drop in its national prison inmate population as authorities sought to reduce prison population densities for public health reasons (Bradley 2020)8Perhaps similarly motivated, a 5% reduction in Australia’s prison population was recorded in 2020 – a notable interruption to 10 consecutive years of steady growth (ABS 2021).. As it would appear, unusually large numbers of ex-offenders were released just at the time that the homelessness shelter capacity was being sharply reduced to lessen potential virus transmission (see Section 4.3).

4.3 COVID-19 homelessness policy and practice responses – emergency accommodation

4.3.1 Overview across the case study countries: leadership and funding

In all of the countries covered in this research, early 2020 saw substantial emergency action to protect existing homeless populations from elevated health risks posed by the pandemic. People targeted for such help generally included residents of crowded hostels or shelters where people shared rooms and/or other facilities, as well as rough sleepers – or ‘unsheltered homeless people’. By and large, such assistance involved publicly (or philanthropically) funded placements in hotels suddenly hugely under-utilised at the time, due to the collapse of international and (to begin with) domestic tourism and business travel. An overview of case study country approaches is shown in Figure 4.2.

Figure 4.2. COVID-19 homelessness EA programs in case study countries

Largely reflecting different governance frameworks, and divisions of governance responsibilities, there was substantial diversity across our case study countries in terms of EA program leadership and funding (see Table 4.1). Thus, national programs were implemented only in the unitary (or semi-unitary) jurisdictions – England, Ireland and New Zealand – albeit that in England program delivery was the responsibility of local governments. In Australia, EA provision for homeless people in 2020 and 2021 could be similarly described as ‘top-down programmatic’ rather than ‘bottom-up organic’, although in this case the initiating agencies were state governments, rather than the national (federal) government.

Table 4.1

In the federal states of Canada, Germany, Spain and the USA, 2020 EA provision was initiated and undertaken in more of a bottom-up way – by municipalities in Canada, Germany and Spain, and by local CoCs in the USA (these latter entities being collaborations usually involving NGOs and local governments).

Associated expenditure in Canada and the USA was, however, supported by national governments. In the latter, Congressional action through the CARES Act 2020 provided some $12 billion for Housing and Urban Development (HUD) programs, including $9 billion that could be utilised for emergency action on homelessness. Part of this funding was specifically intended to ‘help prevent [a COVID-19] outbreak among … people experiencing homelessness and very low-income households … at risk of homelessness’ (National Low Income Housing Coalition, 2020).

In Canada, while much of the cost of EA provision was probably met by municipal and provincial governments, the Canadian federal government also made significant contributions. These included an initial payment of $157 million in April 2020 for activities such as temporary accommodation procurement, personal protective equipment for staff, and associated additional salary costs (CHRA, 2020). In September a further $237 million was committed ‘to help extend … the emergency measures that have been successful in reducing the risk of potential outbreaks among people experiencing homelessness, as well as provide them the flexibility to deliver affordable housing solutions’ (CMHC, 2020).

Australia, Germany and Spain stand out as case study countries where national governments declined to provide significant financial support to underpin EA activity, nor played any strategic advice or co-ordination roles.

4.3.2 EA programs in England, Ireland, New Zealand and Australia

While it is hard to compare the scale or remit of EA activity across countries, it is fairly clear that this was more extensive in the unitary governance nations – England, Ireland and New Zealand – and in the four active Australian states of New South Wales, Victoria, Queensland and South Australia. By and large, the authorisation of new EA bookings was implemented in parallel with national, state- or city-level lockdowns (or ‘stay at home orders’) as these were imposed during 2020 and 2021. In Australia, when lockdowns ended, new bookings generally ceased and existing bookings were maintained only for some former rough sleepers or others displaced from homelessness shelters that needed to remain closed (or occupied only at relatively low density).

England’s ‘Everyone In’ program was probably the largest and most comprehensive of all those reviewed in this report. This was initiated at the very start of the pandemic, on 26 March 2020, when the UK Government instructed local councils to move everyone sleeping rough, or staying in communal shelters, into safe, ideally self-contained, emergency accommodation over the following two days. By May 2020, some 14,610 people were residing in EA hotels and other forms of temporary housing commissioned under Everyone In, and by January 2021, some 37,000 people had been assisted (Shelter 2021).

Crucially, the initial Ministerial instruction made clear the program’s inclusive remit. In this, it effectively overrode two ‘business as usual’ restrictions on local authority obligations towards homeless people. Firstly, their limited rehousing duties towards single people; and, secondly, their constrained ability to provide material help to migrants subject to the No Recourse to Public Funds (NRPF) rule9In Scotland official authorisation to override NRPF exclusion of non-citizens was more explicit. This took the form of guidance from the Convention of Scottish Local Authorities (COSLA), backed by emergency legislation ‘seeking to ensure that those with NRPF had access to support and healthcare during the pandemic’ (Watts et al. 2021)..

Similarly in Australia, state government EA programs initiated in March/April 2020 notably adopted an inclusive approach by temporarily dropping the standard requirement for beneficiaries of such assistance to be Australian citizens or permanent residents. These state government programs were also implemented at considerable scale. In the six months to 30 September 2020 it is estimated that some 40,000 people were placed in EA by these jurisdictions (Pawson et al. 2021a), although this will have included some placements of only very brief duration and not directly related to the pandemic (e.g. short-stay bookings for women fleeing domestic violence outside of lockdown periods). In terms of activity scale it is perhaps more significant that, by 2020 year end, at least 12,000 former rough sleepers had been provided with EA by the four relevant states (Pawson et al. 2021b).

Largely spearheaded by the national Health Department, Irish Government EA activity was initially targeted largely on hostels and other congregate housing facilities considered in need of reduced building occupancy (Lima 2021). Three Dublin establishments closed altogether as advised by the Health Services Executive (Kelleher and Norris 2020). Albeit involving a relatively small number of people, efforts were also made to induct rough sleepers into the program. At least 1,120 new emergency accommodation bedspaces were sourced for homeless people early in the pandemic – including apartments as well as hotels (ibid).

In an observation that somewhat paralleled experience in Australia and England, Ireland country expert Dr Joe Finnerty reported that securing co-operation from ‘hard to help people’ was made easier not only due to the fear of COVID, but also because – by comparison with the low standard temporary shelter usually available to homeless people – ‘the accommodation offer was enhanced’. ‘It wasn’t ‘come on in into an overcrowded hostel’, it was ‘you can have your own room, in a guest house or B&B’, so relatively more attractive’ (Finnerty et al. 2021a).

In New Zealand, meanwhile, it is reported that initial EA funding from the Ministry of Housing and Urban Development went to Housing First Auckland, a non-government agency, to house over 500 people off the streets within a fortnight at the start of the pandemic (Moore 2021). Nationally, an emergency allocation of NZ$108 million underpinned a block booking of 1,600 hotel and motel rooms through to April 2021, as well as associated support services ‘to enable people to stay housed until [the NZ Government] can secure more long-term housing supply’ (NZ Government 2020). In all, it is reported that 3,826 rough sleepers and ‘people living in communal facilities, or crowded houses where it was difficult to limit contact with others’ were placed in motels in the year to March 2021. However, 1,000 people remained in emergency accommodation on 31 March 2021 awaiting the outcome of government efforts to secure longer term housing (Cooke 2021).

As implied in the above discussion, in Australia, England, Ireland and New Zealand, the implementation of EA programs incorporated an assumption of responsibility to assist at least some service users into longer term housing, as well as temporary shelter. This is further discussed in Section 4.3.5.

4.3.3 EA activity in North America in 2020

EA provision for homeless people during the pandemic was less comprehensive or consistent in North America than in the countries discussed above. In the US, in particular, the essentially ‘bottom-up’ initiation of such provision meant it was highly variable across the country. At least in some areas, however, the initial pandemic phase saw CoCs drawing on philanthropic, local state government funding as well as federal resources to expand homelessness accommodation facilities. In New York, for example, the City took on hotels emptied of guests due to the crisis, turning these into temporary homelessness shelters.

Actions of this kind were needed to enable reduced occupancy of existing shelters rendered unsafe in pandemic conditions, but especially to facilitate quarantine for homeless people testing positive for COVID-19. People remaining in, or being returned to, shelters during periods of high community transmission of the virus were a concern for the Centres for Disease Control and Prevention (CDC) as well as for homelessness advocates, since evidence quickly accumulated that such settings posed very high infection risks (CDC 2020; Colburn et al. 2020).

Nevertheless, those assisted by such efforts accounted for ‘only a small fraction of people experiencing homelessness’ (Moses et al. 2020). The vast majority of homeless people in most cities and localities remained in unsheltered locations or congregate settings (ibid).

In Canada, meanwhile, several Canadian cities and provinces (including Toronto, Montreal, British Columbia and Saskatchewan) initiated EA provision from the start of the pandemic. This included booking hotel rooms, acquiring rental buildings and opening additional shelters (Lee and Hamidian 2020). In Toronto, 40 temporary facilities were opened to move 2,300 people from overcrowded shelters to new shelters or hotel programs to permit social distancing (BGM Strategy Group 2020):

We had about 8,000 people in the [homelessness] shelter system, and over a period of about three months the shelter teams had to almost double the amount of real estate that those people were being housed in. They rented hotel [rooms for this purpose] (Senior City official interview).

Across Canada, it was estimated that 10,000 rough sleepers and other homeless people were initially accommodated under EA programs (Pomeroy 2020).

4.3.4 EA activity in Germany and Spain

The relatively modest scale of EA provision in Germany probably reflects the fact that rough sleeping is comparatively rare. This reflects the relatively strong social safety net provisions that prevail, as well as the strict municipal obligation to shelter roofless people. Nevertheless, concerns about an inability to enable social distancing quickly led to closure of many small shelters early in the pandemic, leading to an overall national reduction in such provision (EFNOWH 2021). In large cities there was some utilisation of hotels for roofless people, also partly to enable de-concentration of homeless hostels and shelters. In certain cases this involved municipalities renting space in youth hostels and hotels. Associated expenditure was mainly borne by the cities themselves rather than by regional governments, or by the federal authorities. Notably, in the City of Hamburg NGOs were funded by a philanthropic benefactor (a tobacco company) to lease hotels for homeless people.

To the extent that Spain’s municipalities played the lead role in emergency action to protect homeless people from the pandemic, there may have been some commonalities between the Spanish and German experiences of COVID-19. A key contrast would be the much higher pre-pandemic levels of rough sleeping in Spain where there is no comparable government obligation to prevent rooflessness. Under its definition, a major NGO estimated that ‘40,000 people live on the streets’ (Caritas 2020).

Perhaps even more so than in Germany, national and regional governments in Spain made little or no contribution, even in terms of advice or guidance, let alone financial support (expert academic interviewee). Most of the accommodation made available was in the form of large non-residential buildings such as sports facilities, much of which was criticised as highly unsuitable from a public health perspective because of the impossibility of social isolation (Calvo et al. 2020). As noted above, it is reported that many abandoned such premises, preferring street homelessness.

Emergency accommodation provided by Spanish municipalities and NGOs in response to the pandemic was generally offered for the duration of confinements or lockdowns. When such restrictions ended, so did the service. There does not appear to have been any ‘pandemic dividend’ to the extent of improving an emergency accommodation resident’s prospects of being helped into longer term housing.

4.3.5 Transitioning EA service users into longer term housing in Australia, England and Ireland

Under all of the national EA programs identified in Section 4.3.2, whether explicitly or implicitly, the relevant governments took some responsibility for transitioning people out of hotels and other EA facilities into longer term housing.

Consistent with the long-established local authority obligation to secure settled accommodation for certain homeless people (by law)10Under legislation in place since 1977, local authorities in England are legally obliged to secure settled accommodation for households assessed as homeless and in priority need (that is, containing a pregnant woman, dependent children or vulnerable adults)., associated action has been implemented most comprehensively in England. This process was formally initiated in a Ministerial letter to local authorities on 28 May 2020 requesting that they ‘set out next-step plans for accommodating and supporting rough sleepers brought off the streets during the pandemic’ (Cromarty 2021 p11). To this end, authorities were asked to ‘[undertake] individual assessments and consider a range of options to ensure people’s housing, health and care needs were met’ (ibid).

By January 2021, according to UK Government statistics (DLUHC 2021), 26,167 people had been transitioned from hotels and other Everyone In premises into longer term housing – i.e. ‘settled accommodation’ (tenancies of at least six months) or supported housing (which might amount to a permanent tenancy, but could involve only a short-term placement). Separately, it has been reported that, by early 2021, the total number of people helped by Everyone In was 37,000 (Shelter 2021). So, taking the total number of people transitioned into longer term housing by January 2021 as stated above, that would imply that 71% of all EI placements had been helped as such.

In Australia, where homeless people enjoy no comparable legal right to housing, policymaker aspirations to transition EA service users into longer term housing were, unsurprisingly, interpreted more narrowly than in England. Nevertheless, unassisted by the federal administration, state governments pledged substantial new funding to transition into longer term housing, some of those placed in EA in 2020. Rigorous assessments were used to ration the limited number of funded ‘accommodation and support packages’ to those with complex needs. For others, the end of their hotel booking often meant a return to low standard rooming houses or shelters.

For those assessed as qualifying for a rehousing package – around 3,500 former rough sleepers with complex needs in NSW and Victoria alone – this process will have resulted in a safe, secure and supported pathway to a secure tenancy in some form of social housing. While no pre-pandemic benchmark figure is available, it seems highly likely that the number of complex needs rough sleepers rehoused out of EA in these states in 2020 and 2021 will have been large by comparison with the equivalent cohort accommodated in social housing over previous years. Additionally, it will have to some extent addressed a growing backlog of chronic rough sleepers built up during the late 2010s.

Likewise, in Ireland, it appears that action to rehouse single homeless people into longer term accommodation also expanded in 2020 and 2021 as a knock-on consequence of EA provision during the pandemic. In 2020 such lettings totalled 1,006 – a 74% increase on 2019 (Lima 2021). Moreover, local authority and housing association lettings to this group, some occurring under Housing First programs, also expanded in 2020 and 2021 (Finnerty et al. 2021a; Finnerty et al. 2021b).

While equivalent data are unavailable for any other Australian state, it is estimated that the NSW EA move-on housing pathway will have helped to transition to longer term housing, around a third of NSW rough sleepers assisted via EA in 2020 (Pawson et al. 2021b). As argued above, the numbers involved were very notable. Set alongside the proportion of all Everyone In service users transitioned into longer term housing in England, however (implicitly up to 71% – see above), this is a relatively low figure.

4.4 Longer term outcomes

Across a number of the countries covered by the research it can be argued that pandemic-triggered homelessness action may have generated beneficial insights, new ways of working or policy developments that will outlast the crisis. In other ways, the experience failed to evoke the lasting changes that many advocates had hoped might result.

At least in some countries, a number of positive outcomes should be noted as having resulted from the way that homelessness was managed during the public health emergency. Firstly, some practitioners and advocates have argued that the COVID-19 experience usefully prompted wider recognition that crowded shelter accommodation is problematic from a public health perspective. As such – at least in England, Scotland, Ireland and in some Australian cities – the crisis may prove to have been a stimulus resulting in a modernisation of emergency housing provision, with outdated low standard shelter-type accommodation no longer considered acceptable (Fitzpatrick et al. 2021b; Pawson et al. 2021b; Watts et al. 2021 and 2022).

Similarly, in the German context it was considered by our expert interviewee that ‘acceptance that there must be better temporary accommodation [for homeless people] might be a little bit a consequence of the pandemic’ (Busch-Geertsema interview). This refers mainly to a possibly widening official recognition that multiple occupancy rooms are unsatisfactory. ‘This is not only a matter of human dignity, but it’s really a hygienic problem … to house people in shared air’ (Busch-Geertsema interview).

Secondly, it is clear that in many countries, the extraordinary action taken to protect homeless people during the crisis stimulated enhanced collaborative working between government and non-government organisations, and across domain boundaries – particularly the housing-health divide. In some countries, including Australia and Ireland, this clearly benefited from stepped-up official attention to street homelessness that had been already established during the late 2010s.

Thirdly, the pandemic’s impact in concentrating attention and funded assistance to single homeless people has been celebrated among homelessness workers in Australia and England (Pawson et al. 2020b; Watts et al. 2022).

Fourthly, longer term investment in social housing largely or partly to the benefit of homeless people was prompted by the pandemic in at least some of our case study countries:

  • During 2020 and 2021 the Australian states of Victoria, Queensland, Western Australia and Tasmania announced self-funded social housing development programs totalling nearly $10 billion, and set to add more than 20,000 units to the construction pipeline over the period to 2024 (Pawson et al. 2021b)
  • The Canadian Federal Government committed $2.5 billion to its Rapid Housing Initiative to quickly develop some 7,500 units of accommodation specifically targeted to high needs homeless people – see Box 4.1
  • In 2020 the UK Government pledged £433 million to a rough sleeper accommodation program in England aimed at accommodating ‘up to 6,000 rough sleepers’, underpinned by funded support services – the program being accelerated to assist in rehousing people out of COVID-19 EA (Cromarty 2021)
  • The Irish Government’s 2021 ‘Housing for All’ plan pledged construction of 9,500 social housing units, annually, during the 2020s (for reference, this averaged 3,240 in the period 2016-19 (Farrell and O’Callaghan 2020)).

The social housing investment programs summarised above may be fairly characterised as mainly of quite short duration and of modest scale. The cited Australian state government programs, for example, will be insufficient to entirely halt the historic decline in social housing representation as proportion of all housing – even across the 3-4 year period they are planned to run (Pawson et al. 2021b, Table 6.2, Figure 6.6). Nevertheless, with the probable exception of New Zealand (where program planning was well-advanced before the pandemic), it is unlikely that any of the governments concerned would have pledged such expenditures in the absence of COVID-19.

Box 4.1: Canadian Federal Government ‘Rapid Housing Initiative’

In October 2020 the Canadian Federal Government announced its Rapid Housing Initiative (RHI), a $1billion program to create ‘up to 3,000 new permanent, affordable housing units across the country’ (CMHC 2021). Targeted to people in severe housing need, the scheme aimed to ensure that resulting housing would be available within 12 months of funding allocation. In the event, the initial allocation reportedly generated 4,700 units (Ibid). Additionally, 4,500 units are expected to be developed under a second funding tranche of $1.5 billion announced in April 2021. Described by a leading housing industry figure (interviewed for this research) as ‘probably [Canada’s] most significant housing program … in the past 20 or 30 years’, the program has largely involved modular construction of studio apartment blocks by municipal housing authorities, utilising Federal Government capital grant funding. With municipalities needing to look to the provincial tier of government for operational funding, the RHI is also a case study of multi-level governance.

On the other side of the ledger, however, the pandemic cast into sharp relief some homelessness policy challenges previously less widely recognised. A point similarly applicable in both England and Australia is the way that the crisis exposed the issue of homelessness involving foreign citizens. In the UK case, the relevant term here is ‘[people with] No Recourse to Public Funds’ – a designation used to capture various migrant groups ineligible for social security benefits or social housing. Associated difficulties came to the fore as a problem faced by local authorities in rehousing EA residents into longer term housing, with the result that, as officially reported, NRPF people accounted for ‘around half’ of the 4,000 people being accommodated under Everyone In by London local authorities as at 30 September 2020 (National Audit Office 2021).

More broadly, some international experts interviewed for this research doubted that the experience of managing homelessness during the pandemic had stimulated any significant systemic change that could help in reducing new homelessness beyond the crisis. Thus, as argued by our German expert adviser, Professor Volker Busch-Geertsema:
The consequences were not dramatic enough to get better public awareness of [the fact] that homelessness is really a housing problem and you cannot solve it without getting better access to housing for homeless people.

At least in the UK, however, experts believe there may be more positive lasting effects. For example, communal homelessness shelters have been effectively barred in Scotland, and there is a prospect that – in part thanks to the policy momentum generated in response to the pandemic – Wales may broaden local authority rehousing responsibilities to encompass single people.

4.5 Chapter conclusion

To a lesser or greater extent, unusual intervention to protect homeless people was seen in many high income countries during COVID-19. Crisis responses played out somewhat contrastingly in different countries, partly reflecting diverse systems of governance. The story that unfolded was to some extent a tale of unitary states versus federal systems. Thus, under the unitary frameworks prevailing in England, Ireland and New Zealand, national EA programs were quickly launched at the start of the crisis. In at least some federal states, meanwhile, the crisis proved to be something of a test for inter-governmental relations. Among those covered in this study, this was most obviously the case for Germany where the national government remained disinclined to accept any direct responsibility for addressing homelessness (Busch-Geertsema interview), and in Australia where the relevant Federal Government minister pushed back strongly against calls for a nationally funded post-COVID social housing investment program – on the grounds that this would ‘usurp’ state government duties (Coorey 2020).

Nevertheless, irrespective of the level(s) of government involved, substantial unbudgeted expenditures were committed to EA programs and move-on housing. In part, such decisions may have been motivated by concern for those directly involved, particularly given the well-known prevalence of ill health (and, therefore, virus vulnerability) among homeless populations. At the same time, it should be emphasized that extraordinary actions and expenditures primarily resulted from a reframing of homelessness as a policy problem in the circumstances of the pandemic – a problem of public health rather than individual welfare (Fitzpatrick et al. 2021b). Thus, while their comments were made with specific reference to Australia, conclusions drawn by Parsell et al. (2020 p9) have wider relevance:

In relation to the homelessness response, it would be a mistake to think that these interventions arise primarily from a concern for the impact of COVID-19 on the health of the homeless … what has driven the recent response has rather been the reframing of homelessness from an individual to a public health crisis, where the vulnerabilities experienced by the homeless are identified as a threat, not only to their own health, but also to that of the public more broadly.

With this in mind, however, it is important to note that in countries where COVID-19 infections were extensive, epidemiological evidence demonstrates that emergency homelessness interventions had a measurably beneficial effect for those directly affected, as well as for the broader community. Thus, to the extent that first wave COVID fatalities were minimised in Ireland’s homeless population, success can be claimed for the health-motivated interventions enacted in this phase of the pandemic. Whereas 23 deaths had been predicted among homeless people in the absence of isolation and shielding, only one such death was actually recorded (O’Carroll et al. 2020). Moreover, in England researchers concluded that, during the COVID-19 first wave, the Everyone In program prevented 266 deaths, as well as 21,092 infections, 1,164 hospital admissions and 338 ICU admissions (Lewer et al. 2020).

Finally, it should be emphasized that in countries where EA placements were accompanied by acceptance of longer term housing responsibility for at least some of those assisted, the result is almost certain to have meant that a substantial number of disadvantaged rough sleepers and others will have gained a settled home who – had there been no pandemic – would not have done so.


ABS (2018) Census of Population and Housing: Estimating Homelessness;

ABS (2021) Prisoners in Australia

BGM Strategy Group (2020) COVID-19 Interim Shelter Recovery Strategy: Advice from the Homelessness Service System; report to City of Toronto

Bradley, J. (2020) Canada’s correctional institutions released thousands of inmates in the early days of COVID-19; National Post, 14 August

Calvo, F., Turró-Garriga, O., Solench-Arco, X., & Lorenzo-Aparicio, A. (2020). What happened with persons experiencing homelessness during the confinement? Study about the perception of professionals about the measures taken during the state of alarm consequence of the COVID-19. RES, Revista de Educación Social, Número 31

Caritas (2020) Caritas warns that COVID-19 has triggered the demand for places for homeless people by 25%

Centres for Disease Control and Prevention (2020) Assessment of SARS-CoV-2 Infection Prevalence in Homeless Shelters — Four US Cities, March 27–April 15, 2020; Morbidity and Mortality Weekly Report

CHRA (2020) COVID-19 response page.

CMHC (2020) ‘Canada to Rapidly Create Affordable Housing and Support the Homeless’, media release, 21 September;

CMHC (2021) Rapid Housing Initiative (RHI)

Colburn, G., Fyall, R., Thompson, S., Dean, T., McHugh, C., Moraras, P., Ewing, V. and Argodale, S. (2020) Impact of Hotels as Non-Congregate Emergency Shelters

Cooke, M. (2021) Ministry has no clue on where 42 percent who left COVID-19 motels went; Radio NZ, 29 July

Coorey, P. (2020) Housing a key plank of budget: Exclusive; Australian Financial Review, 6 August

Corlett, E. (2021). ‘Pressures have built up’: how can New Zealand solve its social housing crisis? The Guardian, 11 July

Cromarty, H. (2021) Coronavirus: Support for rough sleepers (England); House of Commons Library; London: UK Parliament

Department for Levelling Up, Housing and Communities (2021) Live Tables on Homelessness

European Federation of National Organisations Working with the Homeless (EFNOWH) (2021). The impact of COVID-19 on homeless service providers & homeless people: the migrant perspective.

Farrell, C. and O’Callaghan, D. (2020) Spending Review 2020: Analysis of Social Housing Build Program; Dublin: Government of Ireland

Finnerty, J., Cullinane, C. and Buckley, M. (2021a) The Experiences of Simon Communities’ Service Users During the Pandemic; Dublin: Simon Communities of Ireland

Finnerty, J., Cullinane, M. and Buckley, M. (2021b) Sustaining the Accelerant, Dublin: Simon Communities of Ireland

Fitzpatrick, S., Watts, B., Pawson, H., Bramley, G., Wood, J., Stephens, M. & Blenkinsopp, J. (2021a) Homelessness Monitor England 2021; London: Crisis

Fitzpatrick, S., Mackie, P., Pawson, H., Watts, B., Wood, J. (2021b) The COVID-19 crisis response to homelessness in Great Britain; Collaborative Centre for Housing Evidence

Kelleher, J., Norris, M., 2020. Day Services for People Who are Homeless in Dublin: A review commissioned by the Dublin Region Homeless Executive (DRHE). University College Dublin, Ireland

Lee, M., & Hamidian, A. (2020) Comparing provincial economic responses to COVID-19; Policy Note 23 April

Lewer, D., Braithwaite, I., Bullock, M., Eyre, M. T., White, P. J., Aldridge, R. W., . . . Hayward, A. C. (2020) COVID-19 among people experiencing homelessness in England: a modelling study, The Lancet Respiratory Medicine doi:

Lima, V. (2021) The Impact of the Pandemic on Services Oriented Towards Single Homeless Persons. (2021) New Zealand

MHCLG (2021) Coronavirus (COVID-19) emergency accommodation survey data: January 2021

Moreno, E.S., & Roldán, I.F. (2020) Social exclusion and COVID-19: the impact of the pandemic on the health, wellbeing and living conditions of the homeless, Instituto Universitario De Desarrollo Ycooperación Universidad Complutense De Madrid. www.faciam.og

Moses, J. (2020) Responding to COVID-19: Conversations with Homeless System Leaders; National Low Income Housing Coalition

National Audit Office (2021) Investigation into the housing of rough sleepers during the COVID-19 pandemic

National Low Income Housing Coalition (2020) Congressional Leaders Agree to Coronavirus Response Package with Funding for Homelessness and Housing

Nilsson, S., Laursen, T., Hjorthoj, C. and Nordentoft, M. (2017). Homelessness as a predictor of mortality: an 11-year register-based cohort study. Soc Psychiatry Psychiatr Epidemiol. 53(1): 63-75

O’Carroll, A., Duffin, T. and Collins, J. (2020) Harm reduction in the time of COVID-19: Case study of homelessness and drug use in Dublin, Ireland; Int J Drug Policy

Parsell, C., Clarke, A., & Kuskoff, E. (2020) Understanding responses to homelessness during COVID-19: an examination of Australia, Housing Studies, 1-14 doi:10.1080/02673037.2020.1829564

Pawson, H., Martin, C., Sisson, A., Thompson, S., Fitzpatrick, S. and Marsh, A. (2021a) COVID-19: Rental housing and homelessness impacts – an initial analysis; ACOSS/UNSW Poverty and Inequality Partnership Report No. 7, Sydney

Pawson, H., Martin, C., Thompson, S., Aminpour, F. (2021b) ‘COVID-19: Rental housing and homelessness policy impacts’ ACOSS/UNSW Poverty and Inequality Partnership Report No. 12, Sydney

Pomeroy, S. (2020) Recovery for All: Proposals to Strengthen the National Housing Strategy and End Homelessness; Canadian Alliance to End Homelessness

Shelter (2021) Everyone In: Where are they now? Briefing, August https://assets.ctfassets. 

Key points

  • At first, the pandemic led to a freezing of national housing markets, as reflected in a steep drop in housing transactions, housing construction, and mortgage market approvals.
  • Defying pessimistic forecasts, however, nominal house prices increased in all case study countries over the course of the pandemic to late 2021.
  • Within countries, 2020 and especially 2021 saw house prices increasing faster for detached properties compared to flats, and for rural/suburban locations compared to urban locations. By late 2021, however, there were signs of rising demand for urban locations.
  • We identify three different house price trajectories pre- and post-pandemic among our case study countries: Spain, Germany and the Anglosphere.
  • Spanish house price inflation was relatively flat pre-pandemic and remained subdued during 2020 and 2021. Spain’s heavy reliance on the international tourism industry, together with its relatively tight mortgage lending regime, are likely explanations for this performance.
  • Germany’s house price inflation was relatively high pre-pandemic and remained as such during 2020 and 2021. Robust income support measures may help explain the absence of any initial pandemic price dip, while its conservative mortgage lending regime probably underlies its subsequently moderate growth rate, relative some other countries.
  • It is the Anglosphere countries where the pandemic price boom was steepest in 2020 and 2021. In the year to Q3 2021, nominal house prices rose by 22% in New Zealand, 19% in Australia and USA, 17% in Canada, and 12% in Ireland and the UK.
  • The rapid price inflation seen in Anglosphere countries post-pandemic is likely to be explained by a combination of factors, most notably: highly liberalised mortgage lending regimes; quantitative easing and the reduction of interest rates by central banks; strong income support measures and pent-up savings, especially among higher-income households; and the introduction of government measures to support first-time buyers and stimulate the housing market.

5.1 Chapter introduction

When the COVID-19 pandemic first hit in early 2020, the immediate concern among most advanced economies was that the resulting economic shutdown would reduce incomes and employment, compromising credit markets, thus driving down housing demand, and house prices with them. These forecasts drew parallels with the previous economic crisis, which had deflated house prices through the same causal chain.

As it transpired though, by 2021, albeit with the pandemic still far from over, house prices had actually risen across all of our case study countries, and especially in the Anglosphere. In this chapter we analyse this phenomenon. We look to explain the key factors driving resurgent house price inflation, and the significant variations observed between, and within, countries.

Chapter 6 seeks to do the same with private rents, an increasingly important metric, given the large and strongly growing private rental sectors nowadays found in most advanced economies.

The analysis that follows is necessarily provisional. Tracking the dips and dives of housing markets is always complicated by the heterogeneity of the housing stock and the infrequency of housing transactions. In any one day, week or month, relatively few houses are actually sold; and those houses that are transacted are unlikely to be representative of the privately owned stock more generally. This underlying difficulty with measuring housing markets over space and time is compounded in our case by the relatively short, yet extremely febrile, time-period in which we are interested, and the international and real-time scope of our analysis.

Thankfully, there are national and international statistics bodies who convert crude housing transactions data into indices which track like-for-like over time and between countries, but these indices are not infallible, especially at the international scale. Thus, throughout this chapter our motto, paraphrasing Keynes, is that it is better to be vaguely right than to be precisely wrong, or to say nothing at all.

This chapter starts, below, by outlining some of the key pre-existing housing market trends which characterised our eight case studies in the decade or so leading up to the pandemic. The following section then reviews the initial pandemic response, and the subsequent house price stabilisation and boom that took many by surprise. The next section, and the main one, then looks at the drivers of house prices over the last two years or so, which it categorises under four sub-headings: income and wealth of nations, the distributional effects of the pandemic and pent-up savings, mortgage financing, housing as an investor asset class, and shifting housing preferences. The final section concludes by reflecting on, among other things, the growing tension between national governments pursuing home ownership and central banks pursuing financial stability.

5.2 Pre-pandemic housing market trends

Before diving into the two years of the pandemic, we first need to briefly sketch the key global housing market dynamics with which the COVID-19 pandemic interacted and overlaid. All of our case study housing markets were affected by the 2008-09 global financial crisis, though some more than others. It was Spain, Ireland, USA and the UK, where the GFC hit the housing market hardest (see Figure 5.1). Between 2008-2012, housing market confidence in these countries evaporated, and mortgage credit contracted, leading to a significant decline in house prices which has persisted over the subsequent decade. Even in nominal terms, prices in Ireland and Spain had barely regained their 2007 levels by 2020. Declining house prices and credit constraints have also stunted housing supply capacity. In Spain, new housing starts remained at around 10% of their pre-GFC peak at the end of the 2010s (Ouasbaa and Viladecans Marsal 2021), and in the UK new housing construction had only just bounced back fully to pre-GFC levels by this time (ONS 2021).

Figure 5.1

In terms of global housing markets though, the GFC’s most wide-ranging effect has been on mortgage lending. Since the securitisation and globalisation of mortgage markets beginning in the 1980s, international financial markets have become increasingly intertwined with national housing markets, especially those Anglosphere nations with liberalised mortgage lending regimes. Major non-US banks such as Deutsche Bank, Credit Suisse and Barclays were all major dealers in US sub-prime mortgage-backed securities (BBC 2016); so when these assets disintegrated the effect on credit markets was global.

The policy response also took place at a global scale, with central banks and other housing finance regulators across the world introducing more stringent lending requirements (e.g. Basel III reforms). This has particularly affected those countries hardest hit by the GFC, where lending to developers and mortgagees had been very loose. In Spain and Ireland, gross mortgage lending immediately prior to the pandemic was less than 30% of pre-GFC levels, and in the UK it was less than 70% (EMF 2021). However, macro-prudential policies were also introduced by central banks in Australia, New Zealand and Canada (He et al. 2016). The one exception to this trend was the famously prudent mortgage lending regime of Germany (He et al. 2016). At the start of the pandemic, gross mortgage lending there was almost double the levels observed pre-GFC (EMF 2021).

The tighter mortgage lending regime and macro-prudential policy particularly affected prospective first-time buyers who lacked capital. Partly as a consequence of this – but also the more fundamental, long-term decline of the returns on labour relative to capital (Arundel 2017; Christophers 2018; Meen 2018; Piketty 2018) – rates of home ownership plateaued or declined during the 2010s (see Figure 5.2). With the social rental representation continuing to shrink across most of our case studies, a growing proportion of young and low-income people are now having to reside in the private rental sector (nearly 70% of 16-29 year olds in Spain (Ouasbaa and Viladecans Marsal 2021)).

Figure 5.2

In a number of our case study countries, housing market conditions and home ownership rates during the 2010s were also affected by relatively high rates of migration-fuelled population growth. At least in Australia, New Zealand and (albeit to a lesser extent) the UK, such conditions were widely identified as contributing to the pressurised housing markets seen in these countries for much of the decade. In particular, since new migrants are usually renters at least initially, large inward migration flows are liable to stimulate private rental growth – thereby depressing home ownership rates.

A final key dynamic affecting all our case studies, has been the ongoing global decline in interest rates and loosening of monetary policy. Through much of the 2010s, global economic growth flatlined in an era of “secular stagnation”. In an attempt to promote or stabilise investment, central banks maintained interest rates at the very low levels set in response to the GFC, or steadily reduced rates to minimal levels over the decade (see Figure 5.3). Some also injected liquidity through quantitative easing.

Together, these central bank interventions have arguably been the major driver of house price inflation across advanced economies over the last decade. They have lowered the cost of borrowing for those able to access mortgages – especially the already asset-rich (Meen 2018) – while also shifting investment into property in the search for yields, as returns have continued to decline in other asset classes (Byrne 2021). One important impact has been the growing appetite of institutional investors, as well as private individuals, for rental property acquisition (Fuller 2021).

Figure 5.3

Finally, in this section, it is worth noting that – at least in certain countries – the first decades of the 20 first century also saw significantly growing housing market divergence between ‘superstar cities’ and the countries in which they were located. In three of our case study countries, as shown in Figure 5.4, the period 2005-18 saw markedly increased house price differentials between key cities and national norms, likely driven by urban planning constraints (Hilber and Mense 2021) and global investor demand (see also Alter et al. 2018).

Figure 5.4

5.3 Housing markets during the pandemic

5.3.1 Initial impacts

By March 2020, all of our case study countries had introduced social distancing rules and other far-reaching public health restrictions. Citizens in advanced economies had already been voluntarily social distancing en-masse11According to the IMF, two-thirds of social distancing in advanced economies was voluntary i.e. pre-dated introduction of government restrictions (COVID’s Impact in Real Time: Finding Balance Amid the Crisis – IMF Blog_; but the government measures formalised these restrictions and bolstered them with the disciplinary powers of the state. These moves prompted a sudden contraction in demand for restaurants, hairdressers, cinemas and other aspects of the service economy on which advanced economies are so reliant. Together with the sheer uncertainty around the lethality and transmissibility of the virus, this led stock markets to plummet and provoked a “dash for cash” amongst investors (e.g. Hauser 2020).

With a credit crunch and a sharp drop in incomes in prospect, house price forecasts were predictably gloomy. In Australia, for example, the Commonwealth Bank (the nation’s largest mortgage lender) envisaged a 32% price reduction over three years in its worst-case scenario outlook (Janda 2020), Evan Siddell, then CEO of the CHMC predicted an annual house price decline of -18% , a figure which he later recognised as far too pessimistic (Globe and Mail 2021). Even in Germany – one of the most stable housing markets in the world (Muellbauer 2018) – there was significant concern about falling property values, with some commentators speculating that prices could drop by as much as 20% in cities such as Berlin and Munich (Prof. Michael Voigtlander).

Such pessimism was not detached from reality. As a rule of thumb, the first indicator of a house price crash is a decline in the number of transactions, as loss-averse sellers take their homes off the market rather than sell at a lower value. And in Spring 2020, this indicator was flashing red in a number of our case study countries. In Ireland, for example, housing transactions declined two-thirds (year on year) by May 2020 (Allen-Coghlan et al. 2020). The decline was steeper still in Spain where house purchases plummeted by 60% in a quarter (Alves and San Juan 2021).

Of course, part of this suspension in housing market activity was a direct consequence of social distancing requirements. In the UK, for example, the Business Secretary asked mortgage lenders to extend mortgage agreements to “a date at which completion can take place,” and asked homebuyers to “move their dates for completion but also their dates moving into new homes.” (Business Times 2020). Similar measures were taken in our other case studies.

But amidst uncertainty, house prices across many countries also appear to have initially declined during the pandemic. By May 2020, UK house prices had fallen 1.7 percent since the previous month (BBC 2020), while Irish prices had experienced the first negative year on year change since 2013 (Allen-Coghlan et al. 2020). Similarly, in Australia, the official ‘eight capital cities’ index registered a 1.8 per cent price reduction in the June quarter of 2020 (ABS 2021a).

On the supply side of the housing market, meanwhile, the pandemic also had an immediate, tangible effect on housing construction activity. As with transactions, pandemic impacts were both direct and indirect. Under social distancing guidelines, building sites in many countries could only function at a limited capacity, if at all. In much of the OECD, the initial emergency period saw a steep decline in new dwellings completed as construction sites were shut and commencing projects placed on hold (OECD 2020a). Indirectly though, the initial uncertainty around house price trajectories also led to an initial dip in the number of building permits sought. Despite construction having one of the shortest lockdowns, Spain saw the number of building permits fall 20% in 2020, a decline that was almost entirely attributable to the first half the year (Alves and San Juan 2021).

Within a few months, however, these downward trajectories in housing transactions and housing construction had stabilised or even reversed. In Ireland, almost 27,300 units were commenced in the 12 months ending June 2021, the fastest growth in any period since the GFC (Banking and Payments Federation Ireland 2021). In Australia, meanwhile, having fallen from 16,000 to 13,000 from February to June 2020, residential building approvals rose to 19,000 by December 2020 before peaking at nearly 24,000 in April 2021 (ABS 2021b). Underpinning these supply recoveries was the increased confidence that pervaded housing markets more generally: house prices were not plummeting as widely predicted, but were growing, and even booming.

5.3.2 House price stabilisation and boom

By the middle of 2021, house prices had at least stabilised in all our case studies and by September 2021, all were on an upward trajectory. Of the 56 countries sampled by Knight Frank’s Global House Price Index over 90 per cent recorded house price increases year on year in Q2 2020 and Q1 2021 (Knight Frank 2021a). As shown by Figure 5.5, there was considerable variation in house price inflation trajectories in the immediate pre-pandemic period, as well as in the extent to which trajectories were altered by the crisis.

Figure 5.5

Table 5.1. Case study house price trajectories

Table 5.1 above crudely categorises our case studies in terms of their house price inflation in the immediate years pre and post-pandemic. In one corner of the matrix is Spain, where house price inflation was relatively flat going into the pandemic and has remained so post-pandemic. How we judge a trajectory does of course depend upon where we draw the start and end point. But even if we trace house price inflation from 2013, the post-GFC nadir of house prices in Spain, then (nominal) house price inflation in the subsequent eight or so years has still been lower than in Britain and Australia. This has remained the case post-pandemic: a report of 22 countries found that by Q1 2021, Spain was the only country to have recorded a year on year decrease in (nominal) house prices (FT 2021a).

In the opposite corner is Germany where house price inflation was relatively rapid going into the pandemic and has remained so post-pandemic. Historically, house prices inflation has been extremely low and stable in Germany (Muellbauer 2018) but this has notably changed in the last decade or so, during which nominal German house price inflation has outstripped most advanced economies, including the UK and Australia (Economist 2021). Like Spain, however, the trajectory of German house prices has not been affected very much by the pandemic, remaining on much the same upward trajectory as before.

In the final corner are those countries where house price inflation has dramatically accelerated post-pandemic; the most extreme example being New Zealand. House prices there had already been rising relatively rapidly prior to the pandemic, but according to the Knight Frank Global House Price Index (2021b), nominal house prices increased by 22% in the year running from Q3 2020- Q3 2021. The post-pandemic surge in house prices was only slightly smaller in Australia, USA (both +19%) and Canada (+17%). The UK and Ireland both reported a more modest pandemic-boom of +12%, but this was well above the paltry 1% inflation recorded over the same 12 months from (Q3) 2018 to (Q3) 2019 (Knight Frank 2019).

The empty corner, ironically, is the one where we would have expected most case studies to feature. But to our knowledge, no advanced economies saw a steep, persistent decline in house prices post-pandemic.

5.4 House price drivers in 2020 and 2021

5.4.1 Framing the analysis

What then drove these house price trajectories? Whereas market goods have undergone price inflation in the short-run as a result of pandemic-induced supply bottlenecks, the same cannot be said for housing markets. One of the distinguishing features of housing markets, is the relative irresponsiveness (or ‘inelasticity’) of supply in the short-run. Because houses take a long time to construct, the level of new supply today had already been largely pre-determined by the market conditions three or four years ago. A more significant element of our story lies in the turnover of the existing supply, where successive closures and re-openings of the housing market, together with volatile housing preferences and expectations, have driven significant variation in the amount and type of housing stock available at any one point in time (seen Chapter 6 in particular).

Overall though, it is the demand-side of the equation that we need to understand to explain why house prices have their particular trajectory over the last two years. The key demand-side determinants of house prices are now fairly well-established, at least in simple econometric terms (Meen 2018). An individual’s ability to purchase a house is determined first and foremost by their job security, income and wealth. If they require a mortgage, as most households do, then their spending will also be conditional on the mortgage market, most notably the tightness of bank lending and the level of mortgage interest rates. Interest rates are also important in another respect as, along with other factors, they determine the relative attractiveness of housing as an asset class. Finally, at the aggregate scale, housing demand is also affected by the simple number of households, which in the short-term, is most likely to be affected by immigration both within and between countries, as well as household formation choices.

These are all very tangible factors which can be calculated and observed, but demand for housing is also determined by a whole range of intangibles, two of which have appear to have fluctuated most dramatically during the pandemic. The first is house price expectations: to state the obvious, individuals will be willing to spend more on a house presently if they expect it to appreciate rather than depreciate in value over time. Up to a point, therefore, house price prophecies are self-fulfilling. The second intangible is housing tastes and preferences (as reflected in the income/price elasticity of demand): the pandemic appears to have brought about a major change in tastes, with significant ramifications for house prices.

Because the pandemic has simultaneously shocked so many of the variables above, and in ways that are non-linear over time, it will take years to unpick the precise causal effect of the pandemic on housing markets. Even in the US, where econometric studies on post-pandemic housing markets are most developed, there are still opposing views about relatively simple matters such as whether housing demand increased less in urban areas than elsewhere (see Zhao et al., 2021 vs Brueckner et al., 2021). The purpose of this chapter is less to isolate out the effect of each variable independent of the other, and more to understand how these variables interacted with each other in different national contexts, and to approximate their overall effect on housing markets across our eight case studies.

5.4.2 Income and wealth of nations

As economies responded to the pandemic by shutting down or raising their national borders, the effect on global trade was predictably dramatic. Every economy the world over, was affected by the pandemic, but whether it was because of competency or chance, some countries coped better with the pandemic better than others. Nations such as Germany with strong pre-existing social welfare states had the plumbing and mandate already in place to expand income support when the pandemic hit, whereas the USA, in contrast, had to develop an emergency welfare framework almost from scratch. Economies such as Spain which had greater levels of public debt, lower economic growth and a smaller tax-base found it more difficult to finance the expanded public-expenditure that the pandemic demanded. Geography was also an important factor in determining a country’s response to COVID-19. As island economies, New Zealand, Australia (and arguably less so the UK) were better able to insulate themselves from the direct effects of the public health crisis, albeit at a cost. Finally, some countries had the bad-luck of specialising in industries that were particularly hard hit by the pandemic.

5.4.3 Spain and reliance on the tourism industry

It is difficult to think of any major industry that was more affected by the pandemic than the tourism industry. In 2018, international tourism was the largest export category after fuels and chemicals (UNWTO 2019), but as flights were grounded, cruise ships set anchor and borders closed, this market essentially collapsed, with international tourist arrivals plunging 65% during the first semester of 2020 (UNWTO 2020).

This had severe ramifications for those national economies most reliant on tourism. Although only suggestive, correlations indicate that tourism dependence alone can explain over half of the cross-country variation in growth performance among EU countries during the pandemic (Milesi-Ferretti 2021). These effects also appear to have fed through into the housing market: with house price inflation slower, and housing transactions less frequent in those countries and cities most reliant on tourism (Duca et al. 2021).

Among our case studies, it was Spain that was most reliant on the lifeblood of international tourism. In 2018, tourism accounted for 14% of all employment (the closest comparator being New Zealand where it accounted for 8% employment) (OECD 2020b). And it was those regions most dependent on tourism where incomes dropped most significantly, a demand shock that was only partially moderated by the government’s income support measures which did not cover those informal, undeclared jobs common in the tourism industry (Taltavull 2021)

Foreign direct investment in the housing stock also stalled: the number of foreign transactions (which have historically accounted for approximately 8% of total transactions) has dropped more steeply than the number of domestic transactions (Alves and San Juan 2021), and it was those regions which went into the pandemic with high levels of foreign buyers who recorded the steepest drop in overall transactions (Alves and San Juan 2021).

Our analysis is far too provisional to conclude that Spain’s reliance on the tourism industry was the main reason why house prices flatlined – mortgage financing, patchy income support, and flows of capital are all additional factors which we touch on in this chapter. Nonetheless, the fact that Spain’s economy was the hardest hit of our case studies, and that it also experienced the lowest house price inflation, seems far from coincidental.

5.4.4 Distributional effects of the pandemic and pent-up savings

As the example of Spain shows, not all employment sectors were affected equally by the pandemic, and not all households received adequate income support to buffer them from the effects of the pandemic. Examining the economic impacts of the pandemic through a distributional lens helps us better explain changes in housing markets, which are highly segmented by income, wealth and tenure, especially in the short-run.

In terms of its direct economic and public health impact, the pandemic disproportionately affected low-income households. Partly, this was because the pandemic latched onto and exacerbated pre-existing health and structural inequalities; in the UK, the local areas with the highest COVID-19 mortality rates for under-65’s already had lower life expectancy, higher deprivation and child poverty (Health Foundation 2021).

But the pandemic also added on new layers of inequality. Without the space to self-isolate, low-income households were likely to be more vulnerable to the spreading of the virus, as supported by the strong UK-correlation between rates of overcrowding and COVID-19 mortality (Inside Housing 2020). Moreover, it was low-income jobs that relied most on being physically present, and that carried most economic and health risk as a result of COVID-19. Middle and high income earners on the other hand, were much better able to work from home.
That said, the initial data suggests that the first few months of the pandemic narrowed not widened income inequalities (O’Donoghue et al. 2020, Almeida et al. 2021, Clark et al. 2021), as government economic interventions largely buffered the effect of the pandemic on low-income households. This was particularly true in Australia (Davidson 2022). With each wave of the pandemic though, government measures have been progressively less interventionist, and there is now some emerging evidence – from Sweden (Angelov and Waldenström 2021), and Germany (DW,2021) – that the unequal economic effects of the pandemic have fed through into rising income inequality. Moreover, if the economic recovery is k-shaped, as some are suggesting, then without redistributive measures, it will feed through into rising income inequalities (Hutton 2021)

There is more evidence of the pandemic widening wealth inequalities. A household’s wealth is a function of both their existing assets and their savings behaviour, and the pandemic has worked through both mechanisms to exacerbate wealth inequalities. Taking savings behaviour first, the pandemic has dramatically curtailed the ability of households to do a range of activities, including eating out, going to the theatre or on holiday. These are all activities on which high income households spend disproportionately large amounts of their income, and the pandemic has meant these households have saved this portion of their incomes instead. In contrast, the goods and services that low-income households spend most of their money on – utility bills, rents – have remained necessary during the pandemic.

The consequence of the asymmetric consumption patterns above is that there is been a glut of savings, but it has been highly skewed towards high-income households (Hughson 2021). In the UK, the almost one in three of the poorest families (i.e. bottom two income deciles) saw their savings decrease during the pandemic (in richest two income deciles, it was only one in ten) (Leslie and Shah 2021). Similar evidence was reported in Spain where 7% of the population experienced severe material deprivation in 2020, up from 4.7% in 2019 (El Pais 2021).

With greater levels of savings, high income households have been better positioned to trade-up in the housing market. This is reflected in housing transactions data which implies that it is primarily high-value houses which are driving the house price inflation. In the UK, post-pandemic growth in the number of transactions for homes over £500K has dramatically outstripped that for homes less than £500K (Hudson 2021). Similarly, in New Zealand the biggest price increases during the pandemic were recorded on high value properties ( 2021).

In sum, through curtailing the expenditure of high-income households, the pandemic has incentivised and facilitated them to save a greater proportion of their incomes or purchase assets such as housing, thus widening wealth inequality. But this is only one half of the wealth inequalities story: as wealth inequalities have also been driven by another branch of government policy, that of monetary policy.

5.4.5 Mortgage financing

Over the last decade, interest rates across the world have generally continued to fall and this had fed through into lower mortgage holding costs. In response, home buyers have been able to take on greater amounts of mortgage debt, leading to an increase in housing demand and, in the absence of a sufficient supply-side response, higher house prices.

In the initial months of the pandemic, as incomes and house prices looked to be on the decline, the immediate concern was that this could lead to a surge in mortgage arrears and foreclosures, potentially setting off the same vicious circle that occurred during the GFC. Governments and banks therefore responded quickly by introducing “mortgage holidays” which allowed mortgagees to defer their payments. Mortgage holiday facilities were introduced in all eight of our case study countries, but their take up varied significantly from country to country.

In Germany, where mortgage lending is famously prudent, these facilities were scarcely used, and were phased out after only three months (Prof. Voigtlander; EMF 2021). In the UK, mortgage holidays were widely adopted but only briefly: 1/6 borrowers had a deferral in place by June 2020, but this proportion shrunk to 1/84 by the end of the year (UK Finance 2022). It was the US, where take-up of mortgage deferrals was most sustained: over 7 million home-owners took out a mortgage holiday there, and by the end of 2022, 2-million were still more than three months behind on payment by (CFPB 2021a). Unsurprisingly, take up was especially high among lower-income households, and for highly leveraged loans (with high LTV ratios at origination) (CFPB 2021b); a pattern which was also observed in Ireland (e.g. Gaffney et al. 2021).

In the US, there is some evidence that the mortgage forbearance program contributed to a tightening of lending underwriting standards and an increase in the mortgage rate margins, as banks factored in future default risk (Golding et al. 2020). Overall, though the pandemic has not led to the shrinkage in mortgage credit that some feared. Indeed, after initial pandemic lockdowns ended in many countries in mid-2020, new mortgage lending generally continued to expand, although in some countries only modestly. In Spain, where house price inflation has stayed relatively low, the number of new mortgage approvals in first nine months of 2021 was roughly equivalent to the same period in 2019 (INE 2021). In the other countries with higher house price inflation, the expansion has been more dramatic. In the UK, mortgage approvals in May 2021 were over a third higher than their 2014-19 average (Bank of England, 2021a). New mortgage credit issued in New Zealand more than doubled in the second half of 2020 and remained near this elevated level through most of 2021 (see Figure 5.6). In the 12 months ending May 2021, Ireland saw more mortgage approvals than in any 12-month period since the data series began in 2011 (Irish Examiner 2021).

Figure 5.6

Besides the wider rise in housing demand (discussed above), there were two more specific drivers of this mortgage market expansion. The first took place at a global scale, and involved central banks as the chief protagonist. Over the course of 2020, central banks made a total of 207 interest rate cuts (Sahin and Girgin 2021). Through lowering interest rates more generally, they also reduced the cost of mortgage debt for the lending banks, which fed through into lower mortgage interest rates. In Europe (EU27, EFTA and UK), for example, average European interest rates on mortgages dropped to an average 2.11% in 2020, a new all-time low, 26 bps lower than in 2019 (EMF 2021)

In the US, the Fed also pushed down mortgage interest rates through quantitative easing measures. In total, the Fed purchased $580 billion in “agency-MBS” (i.e. those guaranteed by Ginnie Mae, Fannie Mae and Freddie Mac) during the two-month period of March–April 2020, and an average of $114 billion per month since (Frame et al. 2021). In offering to buy back mortgage-backed securities (among other assets), it reassured banks that that there would be willing buyer for any securitised mortgage lending, thus compressing agency-MBS yields and encouraging banks to expand mortgage lending (Frame et al. 2021; Brookings 2021).

Outside the US, the mortgage-backed securities market is considerably smaller but quantitative easing measures were still significant. By buying-up assets such as corporate and government bonds en-masse, central banks sapped the yields on these safe-assets, sending investors further afield to find returns. In the UK, for example, the market for riskier residential mortgage-backed securities (dominated by non-bank lenders) more than doubled from 2020 to 2021, partly as a result of such quantitative easing measures (Bloomberg 2021).

5.4.6 Anglosphere governments: first-time buyer demand-side subsidies

The second way in which states drove up house prices through mortgage finance took place at a national scale, and involved those ‘home-owning nations’ in the Anglosphere whose political economy was predicated on sustained house price inflation. For these countries, governments’ immediate priority was to prevent the house price crash that many were predicting. This objective was most explicitly manifested in UK Government’s decision to cut stamp duty (a transaction tax) for properties up to £500K (an intervention which did little to help first-time buyers who were already exempt from stamp duty tax for properties up to £300K).

Putting a floor under house prices though was not an obvious step towards expanding rates of home ownership. Although mortgage interest rates had dropped to record low levels, banks remained hesitant to lend to first-time buyers. Not only did they require higher loan-to-value ratios, but the security of their income stream was also undermined by the onset of the pandemic. Many eurozone countries saw banks tightening the terms and conditions and credit standards on mortgages and an associated increase in the mortgage rejection rate (ECB 2021). Moreover, central banks have generally kept their macro-prudential policies in place, limiting the extent to which banks could lend to first time buyers even if they wanted.
The difficulties of first-time buyers being able to access finance compounded tensions between elected governments on the one hand, who promised to expand rates of first-time ownership, and central banks on the other, who would prefer to lend to existing home-owners and buy to let landlords due to their lower credit-risk. In some cases, this tension has manifested itself publicly, such as in Ireland where the government openly criticised the central bank (and private banks) for the tightness of their lending criteria for first-time buyers (Irish Times 2020). In other cases, central banks appear to have given ground, such as the UK where the Bank of England recently dropped post-GFC mortgage affordability tests for first-time buyers, although only after the government explicitly told them to consider the accessibility of first-time buyers when formulating macro-prudential policy (Barker 2021; HMT 2021).

For the most part though, governments have attempted to (temporarily) resolve this tension in two ways. The first was to simply give first time buyers more money, helping them outcompete other buyers. In Ireland, for example, the maximum tax rebate for “Help to Buy” scheme (originally introduced in 2017) was increased from €20,000 to €30,000, allowing first home buyers (FHBs) an additional £10K of tax refund (RTE 2021). In a similar vein (although not strictly limited to FHBs) the Australian Government’s HomeBuilder grant scheme (announced June 2020) offered AUD$25,000 payments (later reduced to AUD$14,000) to people buying new, or substantially renovating existing, dwellings, with contracts signed by 31 March 2021. In combination with similar schemes run by some of Australia’s state governments it was possible for first home buyers in certain jurisdictions to secure a ‘no strings attached’ cash subsidy of up to AUD$55,000 (Rowley et al. 2020, Table 1). By mid-2021, as a demand-driven program, it was estimated that HomeBuilder alone would incur $2.5 billion in federal government expenditure (Australian Government 2021).

The second, and more common government response though, was to introduce or expand existing equity support or mortgage guarantee schemes, which transferred the credit risk associated with first-time buyers onto the public balance sheet, thus freeing up banks to lend to them at higher LVRs than would have been otherwise acceptable under central bank macro-prudential policies. Pre-pandemic, the UK government had already established an extensive equity support scheme (Help to Buy) which was due to expire in 2023. However, as the number of high (>95%) LTV mortgage products plummeted from over 400 pre-pandemic, to almost-zero by the end of 2020, the government agreed to underwrite a proportion of the credit risk for high LTV loans (HMT 2021). Although take-up has been very low – only 666 buyers by end of December 2021 – the symbolic significance should not be understated; as commentator Neal Hudson remarked “the Government showed banks it was willing to take a stake in the market and other lenders have followed suit, even outside of the scheme itself” (Telegraph 2021).

In Ireland, the new government’s Housing for All bill similarly proposed introducing an equity support scheme for first time buyers purchasing new homes, which was modelled on the UK’s existing Help to Buy scheme (The Journal 2021). In Canada too, the criteria for government-guaranteed mortgage-loan insurance were (temporarily) relaxed (Government of Canada, 2020). Meanwhile, the Australian Government expanded its mortgage guarantee schemes operating since the start of 2020 under its National Housing Finance and Infrastructure Corporation (NHFIC), more than doubling the initially announced beneficiary quota.

Demand-side subsidies and equity support schemes of the kinds described above appear to have had some effect in accelerating the number of first time buyer mortgage approvals. In Ireland, between April 2020 and April 2021, such approvals increased by 130%, significantly outstripping the increase in mortgage approvals overall (98%). In Australia, grants and loan assistance measures were credited as the major driver of a sharp spike in first time buyer property acquisitions which saw such transactions rising by 65% in the period May 2020-February 2021 – reaching a level higher than at any time since 2009. However, demand-side assistance which enhances FHB purchasing power is unlikely to have much of long-run effect on home ownership rates, as without a supply-side response, such measures tend to be capitalised into house prices. Nevertheless, the intuitiveness of their underlying logic – that the best way to increase home ownership rates is to subsidise home-owners – makes them as popular with electorates as they are unpopular with economists. The way that, with hindsight, schemes like HomeBuilder clearly contributed to 2021 house price inflation at undesirable rates seems to have vindicated this latter view. If direct government stimulus to support the construction industry is a desirable part of an economic stability package, economists would generally favour targeting such aid to non-market social housing12For example, by a margin of four to one, leading Australian economists and other experts participating in a 2020 survey believed that in supporting housing sector stimulus through fiscal measures, such action should be directed at the non-market sector rather than at market housing (Maclennan et al. 2021)..

In stark contrast to the Anglosphere, neither the Spanish nor German government introduced any property tax cuts or first-time buyer equity support measures. Indeed, for Germany the opposite was true. Only a year into the pandemic (March 2021), the German government phased out the “Baukindergeld” – a subsidy for families buying or building their first homes (up to 24K Euros for a four person household). Thus, while the UK was rolling out measures to prop up home ownership and house price inflation, the German government was rolling them back.

5.4.7 Housing as an asset class: landlords and investors

Aided by cheap lending, and income support measures, housing demand amongst owner-occupiers has played a significant role in driving house price inflation during the pandemic. Also important though, has been the role of landlords. Since the global financial crisis, investors – both individual and corporate – have been turning to rental housing in search of higher yields and relatively stable returns. Institutional investors, in particular, have emerged as a major actor in the private rental sectors of most advanced economies, including Germany, USA and Ireland (Christophers 2021; Fuller 2021). In the case of the latter, private rented sector (PRS) investment has been driven almost entirely by institutional landlords (Byrne 2021).

In some countries, the pandemic appears to have incentivised and enabled a new surge of institutional investment in the rented sector. As interest rates dropped once again, and inflation rose to levels not observed for four decades, rental income streams (and capital gains) become even more attractive. This was especially true in Germany where 2020 was the second strongest year of all time in the residential investment market (transactions of > 50 apartments) and private equity funds registered their highest acquisition volume since 2012 (Savills 2021). While this investment interest was mainly focused on the private rented sector, some more conservative investors also bought up subsidised (‘social-rented’) apartments, which featured in almost one in seven transactions in Germany (Savills 2021). In Ireland too, the multi-family/PRS market is now the most active property investment class in Ireland (Irish Times 2021a and 2021b).

In some countries, however, the pandemic also saw something of a popular backlash against the large-scale landlords. Post-GFC, institutional investment in market rental housing was encouraged by many governments as another means of increasing housing supply and “professionalising” the sector. Over the last decade, this benign portrayal was starting to wear thin and, with the pandemic, some countries cut back or eliminated official support measures (at least with regards to the purchasing/management of existing private rental housing). In Spain, for example, the moratorium on rental evictions explicitly distinguished between small-scale landlords and large scale landlords, and the restrictions were much more aggressive towards the latter. In Ireland, the government recently introduced a higher, 10 per cent rate of stamp duty on the bulk-buying of homes (Irish Times 2021b).

When it comes to private individual small-scale investors (‘buy to let landlords’) who continue to dominate most markets across our case study countries, activity patterns during 2020 and 2021 did not necessarily resemble those described above. In the UK, for example, mortgage advances to buy-to-let (BTL) borrowers during these years remained generally in the same range as in preceding years (FCA statistics). In Australia, meanwhile, housing finance issuance to this cohort remained at unusually low levels for most of 2020, before rising sharply from the end of the year (ABS Lending Indicators), by which point it may have become plain that a property market crash had been avoided.

5.4.8 Shifting housing preferences and their price effects

The final driver of rising house prices – and possibly also rents (see Chapter 6) – has been the shift in revealed housing preferences. The widespread introduction, and reintroduction, of social distancing measures (especially including working from home – WFH) has radically altered people’s lives: shifting consumption away from services and towards goods. In doing so it has changed the type of housing and neighbourhood that people prefer, shifting demand from some parts of the housing stock to others, as well as increasing the importance of (and demand for) housing overall.

As offices and workplaces have remained closed, and in-person meetings replaced by Zoom calls, the proportion of people working from home has increased dramatically. The economist Nick Bloom noted of the US and UK: pre-pandemic, only about 5 per cent of paid days were spent entirely at home, but during the pandemic, it was roughly 50 per cent (FT 2021b). An official Australian estimate is that ‘up to 40 per cent of workers’ were ‘forced … to experiment with working from home’ during the first year of the pandemic (Productivity Commission 2021 p2).

This has affected the amount that house-buyers (and renters) are willing to spend on different aspects of a house and neighbourhood. More specifically, analysts from the Bank of England looked at the effect of the ‘race for space’ on house prices in the UK, and decomposed it into three different factors (Bank of England 2021b). The first is compositional in nature: the pandemic has changed the type of properties being traded, increasing transactions for detached houses and decreasing transactions for flats. Because, on average, the former are worth more than the latter, this has increased the average value of properties being transacted and explains about 10 percent of house price inflation since January 2020. A similar compositional shift was observed in Spain where an increasing share of transactions in Spain have been for new, larger, single-family houses (Alves and San Juan 2021).

The second dimension of the race for space’ also relates to property type, but this time the argument is that the pandemic not only changed the types of properties being transacted but also changed the amount that people were willing to pay for certain aspects of a property (a change in the valuation of specific important hedonic attributes). This is more difficult to calculate, but the Bank of England analysis suggests the pandemic increased the price that buyers were willing to pay for a house compared to a flat with similar characteristics (e.g. similar area, number of bedrooms) and this partial increase in demand explains about 20 percent of the overall house price growth.

The same process has played out in the Australian housing market where pandemic era price inflation has been much more evident for houses rather than apartments. In the year to July 2021, for example, (typically larger) houses appreciated in value by 18.4 per cent, while the comparable figure for (typically smaller) apartments was only 8.7 per cent (CoreLogic Australia 2021a).

The final dimension of the race for space was spatial, and relates to the type of location people wanted to live in. For those who want to remain close to the city, but who no longer need to commute regularly, it appears to have led to a “doughnut effect” as demand shifted from the centre to the suburbs. Others, including those can work entirely from home, or who can no longer gain employment in the city, have moved even further away from the city.

On this spatial dimension, the Bank of England analysis found that a faster increase in the value of properties outside of London, compared to similar properties in London, explained about 15% of the total uplift in UK house price inflation. Similar house price trends have been observed in other ownership markets. In Australia the year to November 2021 saw property prices in the country’s eight (state/territory) capital cities climb by 20.8 per cent, while the comparable increase elsewhere was 24.3 per cent (CoreLogic Australia 2021b). While property prices fell by 3 per cent in Barcelona in the year to end February 2021, prices rose by 1.7 per cent in Barcelona province (FT 2021c). In Germany, house prices have risen faster in the suburbs of major Southern/Western cities, as city-dwellers joined the race for space (Berlin HYP and CBRE 2021). By late 2021 these trends have to some extent eroded historic margins between prices and rents in capitals and other superstar cities and other regions. As Chapter 6 demonstrates though, there are now emerging signs from rental data that, with the petering out of COVID-19 a real possibility, housing demand is flowing back towards the cities.

Figure 5.7

5.5 Chapter conclusion

This chapter has sought to review the effect of the pandemic on housing markets in advanced economies, with a particular focus on our eight case studies. Our analysis has been situated at two spatial scales, seeking to understand the different trajectories of house prices both between countries and within countries.

Within countries, the story of the pandemic so far is that it has shifted demand away from city centres and flats, towards more suburban and rural areas and houses. There are emergent signs, mainly from rental markets, that this post-pandemic trend is going into reverse, as people relocate back into the city. So will cities regain their pre-pandemic dominance? Ultimately, much will depend on the evolution of the virus and individual preferences. If shutdowns remain a way of life, then this will continue to blunt the agglomeration economies on which cities depend, making them less appealing places to live. Even if the virus becomes endemic though, rendering shutdowns redundant, then there is no guarantee that changes in housing preferences and the normalisation of virtual working practices will fall away entirely, as we may have reached a new work from home equilibrium.

In comparing the housing market outcomes between countries, we drew a three-fold distinction between our case studies, based on their pre- and post-pandemic house price trajectories. In one corner was Spain, where house prices inflation was low going into the pandemic and has remained low since. The reliance of the Spanish economy on tourism, and its poor economic health as a result of the last GFC, meant that it was particularly vulnerable to the shock in demand brought about by the pandemic. This was reflected in GDP and youth unemployment figures, and it was also reflected in the low house prices.

In the opposite corner was Germany, where house price inflation was rapid going into the pandemic and has remained so since. Thus, the negative effects of the pandemic did not really feed through into housing outcomes: house prices continued to rise, and there were very few rent or mortgage payment defaults. One explanation for this lies in the relatively comprehensive German welfare state, which protected the incomes of renters, landlords and home-owners from the shock of the pandemic. The other lies in the significant personal savings that Germans put away – German households save about 10 per cent of their disposable income, twice as much as the average EU or American (FT 2018) – which provides another buffer for income losses (albeit one that is extremely uneven).

But why did German house prices not accelerate post-pandemic at the same rates observed in our other Anglosphere case studies? The answer is no doubt complicated, but it seems to relate to the distinct relationship that these countries have towards home ownership and housing as an asset class. New Zealand, Canada, UK, USA, Australia – all these countries idealise home ownership as a superior tenure, but also have a highly leveraged political economy that is predicated on rising house prices. Recent history and economic theory all indicate that it is nigh-on impossible to sustainably increase home ownership and maintain house price inflation without major growth in incomes. And yet, this is exactly what Anglosphere governments have been seeking, or at least claiming, to do for decades, driving wealth inequalities in the process.

One way of coping with this contradiction, albeit unsustainably, has been to compensate for the lack of income growth by expanding credit markets. Pre-GFC, banks and credit markets were allowed to do this themselves but with catastrophic consequences. Banks have since become more wary in their lending and central banks stronger in their macro-prudential interventions. This has ushered in an uneasy relationship between central banks and national governments, each pursuing different objectives: central bank restricting lending to FHBs in the name of general financial stability; national governments seeking to expand lending to FHBs in the name of home ownership (Barker 2021).

When central banks expanded quantitative easing and cut interest rates in response to the pandemic, their concern was for the stability of the financial system at large, but the bluntness of their instruments together with the mutual dependence of the financial market and the housing market in Anglosphere nations, meant house prices rises were an unavoidable side-effect. Even if they don’t know it, highly leveraged home-owners and landlords now exercise considerable (infra-) structural power, as the price of their house is tied up with the stability of the financial system more broadly.

This expansion of mortgage credit, however, followed the market logic rather than government objectives, flowing disproportionately to existing home-owners with lower credit risk, rather than FHBs. In response, national governments sought to rebalance the mortgage market in favour of FHBs, either by de-risking lending to FHBs, or by expanding the demand-side subsidies on offer to them. But both of these interventions likely had the effect of increasing house prices further.

If annual house price inflation ‘stabilised’ at single digits, around 2-7% say, then it is difficult to imagine much concern on the part of central banks or governments. But house prices cannot simply be turned up and down by the state like a dial, and in several Anglosophere countries, the pace of house price inflation has now accelerated well beyond this threshold, concerning both national governments, because it very obviously pushes home ownership out of reach for FHBs, and central banks because rising house price expectations could bring about a house price bubble. In the US, for example, several members of the US Federal Reserve recently raised concerns about the inflationary effect of interest rates and quantitative easing on house prices (Montoriol-Garriga and Ondina 2021).

There are a range of demand-side policies that could be introduced to dampen house price inflation in the short-run, such as increased housing taxation and tighter loan-to-value lending ratios, but central banks lack the mandate to intervene in such a targeted fashion while national governments lack the political inventive. Only in New Zealand, where house price inflation is most rampant, has the government taken any major steps to dampen house price inflation. In February 2021, the government there formally added a clause to the RBNZ’s mandate, despite the Governor’s objections, instructing it to consider housing prices in making monetary policy decisions, and the RBNZ has since reinstated restrictions on loan-to-value ratios and increased interest rates twice. The government for its part has increased the taxation of property, such as the phased removal of mortgage interest tax relief. But these measures have provoked criticism from the electorate and it remains to be seen whether governments with weaker political mandates have the political capital or will to implement them.


Alter, A., Dokko, J., & Seneviratne, M. (2018). House price synchronicity, banking integration, and global financial conditions. International Monetary Fund

Arundel, R. (2017). Equity inequity: Housing wealth inequality, inter and intra-generational divergences, and the rise of private landlordism. Housing, Theory and Society, 34(2), 176-200

ABS (2021a) Residential Property Price Indexes: Eight Capital Cities; Australian Bureau of Statistics

ABS (2021b) Building Approvals, Australia;

Allen-Coghlan, M., McQuinn, K., & O’Toole, C. (2020). Assessing the impacts of COVID-19 on the Irish property market: An overview of the issues. Quarterly Economic Commentary, 93.

Angelov, N and D Waldenström (2021), “COVID-19 and income inequality: Evidence from monthly population registers”, CEPR Discussion Paper 16333.

Almeida, V, S Barrios, M Christl, S De Poli, A Tumino, and W van der Wielen (2021) The impact of COVID-19 on households’ income in the EU, Journal of Economic Inequality, forthcoming.

Alves and San Juan (2021) The Impact of the COVID-19 Health Crisis on the Housing Market in Spain

Australian Government (2021) HomeBuilder Factsheet 

Bank of England (2020) How has Covid affected household savings? Bank of England

Bank of England (2021a). Mortgage approval data. Accessed online via BuiltPlace;

Bank of England (2021b) Bank Overground blog, 13 December

Barker (2021) The Financial Policy Committee and the Housing Market – Macroprudential Matters

BBC (2016) Deutsche Bank agrees $7.2bn penalty with US regulators – BBC News

BBC (2020) House prices see largest monthly fall for 11 years, says Nationwide – BBC News

Berlin HYP and CBRE (2021) Housing market report: Berlin 2021

Bloomberg (2021) Riskier Mortgage Bonds Thrive Off U.K. Pandemic Property Boom – Bloomberg

Brookings (2021) What did the Fed do in response to the COVID-19 crisis? (

Brueckner, J., Kahn, M. E., & Lin, G. C. (2021). A New Spatial Hedonic Equilibrium in the Emerging Work-from-Home Economy? (No. w28526). National Bureau of Economic Research

Business Times (2020) Frozen UK home market leaves buyers stranded, Real Estate – The Business Times

Byrne, M. (2021) ‘Institutional Investment in the Private Rental Sector in the wake of the COVID-19 pandemic: a review of international “grey literature” and reflections on the Irish context’ School of Social Policy Social Work and Social Justice, UCD

Christophers, B. (2018) Intergenerational inequality? Labour, capital, and housing through the ages. Antipode, 50(1), 101-121.

Christophers, B. (2021). Mind the rent gap: Blackstone, housing investment and the reordering of urban rent surfaces. Urban Studies, 00420980211026466

Clark, A E, C D’Ambrosio, and A Lepinteur (2021) The fall in income inequality during COVID-19 in four European countries, Journal of Economic Inequality, forthcoming.

CoreLogic Australia (2021a) Housing Market Update: August 2021

CoreLogic Australia (2021b) Housing Market Update: December 2021

CFPB (2021a) Housing insecurity and the COVID-19 pandemic (

CFPB (2021b) Characteristics of Mortgage Borrowers During the COVID-19 Pandemic. Accessed here:

Davidson, P., (2022) A tale of two pandemics: COVID, inequality and poverty in 2020 and 2021 ACOSS/UNSW Sydney Poverty and Inequality Partnership, Build Back Fairer Series, Report No. 3, Sydney

Duca, J, Hoesli, M and Montezuma, J (2021) ‘The Resilience and Realignment of House Prices in the Era of COVID-19’, Journal of Real Estate Research, online first, DOI 10.1108/JERER-11-2020-0055

DW (2021) Social inequality in Germany is on the rise, 12 May

El Pais (2021) Covid-19: Coronavirus pandemic pushed poverty in Spain to highest levels since the Great Recession | Economy and Business | EL PAÍS English Edition (
European Mortgage Federation, EMF (2021) EMF-Q1-2021_v2.pdf (

Economist (2021). Global House Prices. Accessed online:

ECB (2021) Bank lending survey

FT (2018) Why are Germans so obsessed with saving money?

FT (2021a) House prices climb to record levels in US and Europe

FT (2021b) Nick Bloom: It is becoming pretty clear now that hybrid working is here to stay

FT (2021c) How a dearth of tourists has transformed Barcelona’s property market

Frame, Greene, Hull and Zorsky (2021) Fed’s Mortgage-Backed Securities Purchases Sought Calm, Accommodation During Pandemic –

Fuller, G. W. (2021). The financialization of rented homes: continuity and change in housing financialization. Review of Evolutionary Political Economy, 2(3), 551-570

Gaffney, E., McCann, F., & Stroebel, J. (2021) The economics of mortgage debt relief during a pandemic (No. 6/FS/21). Central Bank of Ireland.

Guardian (2021) Bank of England plans to remove interest rate rule for mortgages

Globe and Mail (2021) CMHC boss Evan Siddall acknowledges ‘errors’ on last year’s prediction of housing collapse – The Globe and Mail

Government of Canada, 2020 Government of Canada Announces Additional Measures to Support Continued Lending to Canadian Consumers and Businesses –

Golding, E et al (2020) The Mortgage Market as a Stimulus Channel in the COVID-19 Crisis. Mimeo.

Hauser, A. (2020) Seven Moments in Spring: COVID-19, financial markets and the Bank of England’s balance sheet operations; Speech, Bank of England

HMT (2021) Mortgage Guarantee Scheme 210301_Budget_Supplementary_Doc_-_mortgage_guarantee_scheme.pdf (

HMT (2021) Remit and recommendations for the Financial Policy Committee – March 2021

He, D., Nier, E. W., & Kang, H. (2016). Macroprudential measures for addressing housing sector risks. BIS Paper, (86p).

Health Foundation (2021) What geographic inequalities in COVID-19 mortality rates and health can tell us about levelling up

Hilber, C. A., & Mense, A. (2021). Why have house prices risen so much more than rents in superstar cities?

Hudson (2021) Market commentary – September 2021

Hughson (2021) What are the implications of Covid-19 for wealth inequality? – Economics Observatory

Hutton, W. (2021) Don’t let pandemic losers slide further in a K-shaped recovery

Inside Housing (2020) The housing pandemic: four graphs showing the link between COVID-19 deaths and the housing crisis

Irish Examiner (2021) Mortgage approvals surge, adding further pressure to the housing market

Irish Times (2020) Mortgage seekers face tighter criteria over COVID-19 wage subsidy

Irish Times (2021a) Dublin’s private rented sector proves to be investors’ choice

Irish Times (2021b) Why have foreign funds targeted the Irish property market?

INE, 2021. Mortgage Approvals data. Accessed online;

Janda, M. (2020) CBA warns Australia risks 32 per cent house price crash in a ‘prolonged downturn’, flags $1.5 billion coronavirus-hit to bank; 13 May

Knight Frank (2021a) Global House Price Index – Q3 2021

Knight Frank (2021b) Global House Price Index – Q3 2021

Knight Frank (2019) Global House Price Index – Q3 2019 global-house-price-index-q3-2019-6859.pdf (

Leslie and Shah (2021) (Wealth) gap year

Maclennan, D., Long, J., Pawson, H., Randolph, B., Aminpour, F. and Leishman, C. (2021) Australian experts views of housing and the economy: Abstract dreamings or real directions? Sydney: UNSW City Futures Research Centre

Meen, G. (2018) Policy approaches for improving affordability. Glasgow: UK Collaborative Centre for Housing Evidence (CaCHE)

Montoriol-Garriga and Ondina (2021) The impact of the pandemic on international housing markets: Is there a risk of overheating?

Milesi-Ferretti (2021) Economic repercussions of the COVID-induced tourism collapse in the euro area

Muellbauer, J. (2018). Housing, debt and the economy: A tale of two countries. National Institute Economic Review, 245, R20-R33.

Ouasbaa and Viladecans Marsal (2021) Spain’s once-substantial rental market is now one of the smallest in Europe, Brookings Institute

OECD, (2020a) Housing amid COVID-19: Policy responses and challenges

OECD (2020b) OECD Tourism trends and policies 2020

OECD (2021) Affordable housing database

ONS (2021) Housebuilding UK: permanent dwellings started and completed

O’Donoghue, C, D M Sologon, I Kyzyma, and J McHale (2020), “Modelling the distributional impact of the COVID-19 crisis”; Fiscal Studies 41: 321–336.

Piketty, T. (2018). Capital in the twenty-first century. Harvard University Press.

Productivity Commission (2021) Working from Home; Research Paper; Canberra: Commonwealth of Australia

Rowley, S., Crowe, A. Gilbert, C., Kruger, M., Leishman, C. and Zuo, J. (2020) Responding to the pandemic, can building homes rebuild Australia? Final Report No. 341; Melbourne: AHURI

RTE, 2021 What supports are available for first-time buyers? (

Siddell (2021) (2021) New Zealand’s priciest properties quickly becoming more expensive

Sahin, T. and Girgin, Y. (2021) Central Banks Cut Interest Rates 207 Times in 2020, 18 January

Taltavull, P. (2021) Informe sobre el mercado de la vivienda; España 2020; 2º semestre Observatorio de la valoración

The Journal (2021) Explainer: How will the Government’s Affordable Home Shared Equity Scheme work?

Savills (2021) German residential investment market in 2020

Telegraph (2021) Rishi Sunak’s flagship mortgage scheme flops

UNWTO (2019) International Tourism Highlights, 2019 Edition

UNWTO (2020) International tourist numbers down 65% in first half of 2020

UK Finance (2021) Lenders continue commitment to supporting mortgage customers into 2021

Zhao, Y. (2020). US housing market during COVID-19: aggregate and distributional evidence.

Key points

  • Most of the Anglophone countries covered in this study (especially Australia, Ireland, New Zealand, the UK and the US) saw a brief initial stall in national rent levels at the start of the pandemic, followed by rapidly accelerating rent inflation during 2021. By late 2021, in Australia, the UK and the USA, rents were rising at rates unseen since 2008.
  • Rent trends during COVID-19 in Germany and Spain have contrasted from those in the Anglophone countries. In Germany, apparently extending a pre-existing trend, rent inflation appears to have continued to subside during 2020 and 2021, albeit with nominal rents only beginning to actually decline in late 2021. In a somewhat similar pattern, pre-pandemic annual rent inflation of around 5% in Spain was, by early 2021, replaced by rent deflation.
  • In most of the Anglophone countries market rents in some capital (or other large) cities were comparatively hard hit early in the pandemic, but in many cases recovered strongly in 2021 (somewhat less so in Sydney and, especially, Melbourne).
  • Albeit based on more limited available data it would appear that German and Spanish sub-national rent trends have been highly contrasting. In Germany, at least among the major cities, rent inflation appears to have been trending along a largely common downward trajectory during the pandemic. In Spain, by comparison, 2021 regional rent trends have been more variable.
  • Higher inflation of house rents in comparison with apartments (paralleling house/apartment sale price trends) seems to likely to have likewise reflected ‘the race for space’ – housing consumption preferences more influenced by dwelling size than prior to the pandemic.
  • In most of the Anglophone countries, notwithstanding market turbulence since March 2020, the pandemic has seen rents rising ahead of earnings, implying worsening rental affordability that will have in most cases compounded pre-crisis trends.

6.1 Chapter introduction

As discussed in Chapter 5, many countries have seen major impacts on housing asset prices associated with the COVID-19 pandemic. In most of our case study jurisdictions notable developments have included a largely unanticipated price boom that picked up steam from late 2020. However, as this chapter demonstrates, the public health emergency has also triggered extraordinary turbulence in private rental housing markets.

Particularly in countries with minimal social housing provision (e.g. Australia, Canada, New Zealand, US), private rent levels have important social welfare implications in that they concern the housing sector that accommodates most low income households. In Australia, for example, private rental housing provides for double the number of low income households residing in social housing (Pawson et al. 2021a p38). Moreover, in a country like the US the pre-pandemic norm was that ‘while extremely low income renters make up one-in-three renters, they [have faced] increasing competition for rental housing from more affluent households’ (Schwartz 2021 p38).

Private tenants are exposed to housing price fluctuations in ways that owner-occupiers and social renters are not. The market price of rental housing at any one time is directly relevant to those seeking their first tenancy or needing to move between tenancies. Moreover, in most countries covered in this research residential tenancy laws allow landlords to increase rents for existing tenancies in line with the market (see Chapter 3, Table 3.1). As a result, where market conditions inflate rents demanded for new tenancies, this is liable to flow through to the broader tenant population relatively quickly.

Given the above, it is changing rent levels during COVID-19 that form the key focus of this chapter. First, we recount observed trends in rent levels across our eight case study countries as recorded during 2020 and 2021. After an international overview at country level, this analysis progresses to investigate sub-national rent trajectories seen during the pandemic to date. Next, in seeking to explain these patterns we discuss some of the market demand and supply drivers that may have contributed. Then, returning to national level analysis, we seek to assess pandemic impacts to date in relation to rental affordability. The chapter then closes with some brief concluding remarks.

In analysing rent trends both within and across countries it is important to recognise the diverse ways that rent data are analysed and reported. Rent price statistics published by property data firms usually refer to ‘asking rents’ – i.e. the advertised rent for properties available to let during a given period. ‘Agreed rents’ refer to the rents actually contracted at the commencement of tenancies (which will be related, but not necessarily identical, to asking rents). Both of the above measures calibrate ‘entry rents’ for new tenancies commencing in a specified timeslot (month, quarter, year). Other rent indices (e.g. government-commissioned population surveys) analyse rents relating to the much larger body of tenancies in existence at any one time or during any given period.

Before embarking on the substantive discussion it should also be emphasized that residential rental price data is far less easily available than house price data of the kind analysed in Chapter 5. This is true both within most countries and also in terms of international comparative economic statistics (e.g. IMF or OECD). Moreover, such rent statistics as are available tend to be published by property data firms rather than by government agencies. The methodology underlying analyses of this kind is not always fully disclosed; in particular, data coverage or comprehensiveness sometimes remains uncertain. Some measure slightly different dimensions of market performance (e.g. mean versus median values), and vary in sophistication (e.g. in only some cases mix-adjusted or hedonically modelled). For all of these reasons, the following analyses need to be treated with caution.

6.2 Rent trends during COVID-19

6.2.1 Market-wide national trends

Subject to the qualifications summarised above, it would appear that the eight countries fall into three broad categories as regards overall rental market trends in 2020 and 2021: firstly, the Anglophone countries (Australia, Canada, Ireland, New Zealand, UK, US); secondly, Germany; and, thirdly, Spain. Notably, this typology is the same as that identified in relation to house sales market trajectories in Chapter 5.

In the Anglophone countries (perhaps with the partial exception of Canada – see below), after a brief dip at the start of the pandemic, rent inflation began to pick up in late 2020, before accelerating sharply during 2021 (see Figure 6.1). In Ireland, this trend had seen rental prices re-gaining the extraordinarily high growth rates recorded in the mid-2010s (which prompted the introduction of rent controls in 2016). In Australia, the UK and the US, by late 2021 rents were rising at rates unseen since the 2008 Global Financial Crisis.

Figure 6.1

While directly comparable data for Canada are not available, potentially relevant statistical evidence is published on the site, drawing on rental property listings data. These data indicate that, on a ‘price per square foot’ basis, market-wide ‘asking rents’ remained virtually static through 2020 and early 2021. Only in Q4 2021 did these values began to recover. As discussed below, however, when disaggregated by geography and property type parts of the Canadian rental market were subject to much more turbulence in 2020 and 2021 than that national trend would imply.

Germany’s rental market appears to have behaved very differently to those of the Anglophone countries during 2020 and 2021. The city-level data shown in Figure 6.2 indicate that rent inflation generally tended to fall back during late 2020 and into 2021, albeit having notably spiked in four of Germany’s largest cities at the very start of the crisis (Q2 2020). In apparent contrast with Anglophone country trends, rent inflation in four of Germany’s five largest cities had fallen to below zero by the end of 2021. In two of these cities – Munich and Frankfurt – rents had been generally declining throughout the year. Viewed across the period from 2019, one interpretation of the pattern is that rent inflation was already trending down in advance of the pandemic, with that trajectory generally continuing from Q3 2020. In that sense it might be arguable that COVID-19 has had relatively little impact on pre-existing market trends.

Figure 6.2

Overall, as in relation to house sales (see Chapter 5), Germany’s rental housing market once again appears to have performed quite differently from markets in comparator countries during the crisis.

While statistics on Spanish rents are much less readily available, it would appear that – at the national scale – the country has seen rents falling significantly during the pandemic. This has occurred not only as a short-lived development at the start of the crisis, but as an ongoing trend running well into 2021. Thus, according to the real estate company Idealista, average rents per square metre across Spain declined by 3.8% during the first 12 months of the pandemic ( . As explained below, however, the story is much more complex at the sub-national scale.

6.2.2 Geographical variations

At least in some of our case study countries the national trends discussed above conceal substantial sub-national diversity in terms of pandemic impacts across rental markets. Relevant data on these dimensions of market performance are somewhat sparse and/or of only limited comparability across countries. Nevertheless, at least in several of the case study jurisdictions common trends have been apparent. In particular:

  • Significant ‘advertised rent’ reductions (nominal declines, rather than simply reduced rates of inflation) recorded in some capital (or other large) cities during 2020 and in some cases persisting into 2021.
  • 2021 rent inflation in non-metropolitan areas tending to outpace capital (and/or other large) cities.
  • Higher rates of inflation affecting houses as opposed to apartments or units.

At least in Australia, the UK, and the United States, the initial phase of the pandemic saw rents falling sharply in some large cities13Albeit that Australian evidence shows significant variations across the eight state/territory capital cities – pandemic rent trajectories for Sydney and Melbourne were quite different from those for Adelaide and Perth – see Figure 6.5.. Available data for major cities in several case study countries illustrates patterns somewhat similar to national trends (Figure 6.1), but with the important qualification that – as experienced in Sydney, Melbourne, London and New York – annual changes in advertised rents actually fell significantly below zero for substantial periods in 2020 and 2021 (see Figure 6.3).

As shown in Figure 6.3, Greater London’s average advertised rents declined sharply in Q2 2020, with this trajectory maintained all the way through to March 2021. Nominal average rents fell by 5% from March to June 2020, with a second decline in late 2020 following a partial mid-year recovery (Homelet . Only by late 2021 had Greater London rents regained their pre-pandemic levels. By this time, however, they were escalating rapidly.

Figure 6.3

In Canada, meanwhile, according to Toronto’s rental market was much more impacted by the pandemic than most other parts of the country. Mean advertised rents (all property types) fell by 21% in the year to December 2020. While part of this may have been due to compositional change in the stock of advertised properties, Toronto condominium rents dropped by 19% over this same period – albeit that by late 2021 this metric had nearly regained its pre-pandemic value.

In the US, rents fell by 6% in both New York and San Francisco during 2020 (see Figures 6.3 and 6.4). As shown in Figure 6.3, however, these markets had begun to recover by year end 2020. Thanks to continuing increases in most other markets, America’s national mean rent continued to rise throughout 2020. As shown in Figure 6.4, such trends were particularly borne out in some large sun belt cities. By the end of 2021, for example, nominal mean rents in Phoenix had risen by 50% in less than four years.

Figure 6.4

In Australia, an apparent tendency for rents in Australian non-metropolitan areas to outpace those in capital cities during 2020 and 2021 has been much discussed (Pawson et al. 2021a). For example, in the year to September 2021, average asking rents rose by 12.5% for ‘regional Australia’ compared with 7.5% for Sydney, Melbourne and the other six state/territory capital cities (CoreLogic Australia 2021c). Moreover, in interpreting these figures it is also important to factor in the substantial rent reductions seen in some capital city housing markets at an early phase of the pandemic. Thus, calibrated in terms of simple nominal media rents, the contrast between the 2020-2021 rent trajectories of Australian capital cities and non-metropolitan areas is even more stark – see Figure 6.5. During the two years to Q4 2021, regional rents rose by 18% whereas metropolitan rents had, by the end of this period only just recovered to their starting level.

In fact, as shown by more detailed (also previously unpublished) Domain statistics, the phenomenon of regional rents ‘outperforming’ the relevant capital city during this period was especially marked in New South Wales (+17% versus -2%), Tasmania (+26% versus +8%) and Victoria (+15% versus -7%) – see Figure 6.6. Also notable from Figure 6.6. is that Western Australia, where geographical isolation and a closed border largely enabled avoidance of economic restrictions during the first two years of the pandemic, the pattern was completely different.

Figure 6.5

Figure 6.6

Sub-national analysis of rent trends in Spain is also highly instructive as a reminder that national-level statistics may conceal as much as they reveal. Across the country as a whole, rents reportedly continued to fall through 2020 and 2021 (Idealista 2021). At the same time, however, they rose in 2021 in 12 of Spain’s 19 autonomous communities and in 41 of its 50 provinces. In Castile and Leon, in La Rioja and Murcia, for example, rents reportedly rose by more than 6% in 2021 whereas they fell by 11% in Madrid and by 14% in Barcelona. Perhaps significantly, the Balearic Islands also recorded a particularly steep reduction of 14% (ibid).

6.2.3 Property type variations

At least for some of our case study countries there is evidence of substantial recent rent trend divergence according to property type. To some extent reflecting patterns observed in the house sales market (see Chapter 5), there are indications that house rents have likewise tended to rise significantly faster in 2021 than apartment or unit rents. In Sydney and Melbourne, for example, apartment rents fell further and remained depressed for longer in 2020 and 2021 than house rents (Pawson et al. 2021a). And, across Australia, house rents increased by 10% in the year to October 2021 while apartment rents rose by only 6% (CoreLogic Australia 2021d). Similarly, in the US, rental inflation for detached rentals was 12% in September 2021 (year on year), compared with only 8% for attached properties (CoreLogic US 2021).

Figure 6.7

The Irish Residential Tenancies Board (RTB), meanwhile, reported that ‘[w]hile the early quarters of the pandemic were associated with a rapid drop in inflation for both housing types, the [2021] rebound has been more pronounced for houses than apartments’ (RTB 2021 p14). By Q3 2021, house rents were escalating at over 10% p.a. while apartment rents were rising at under 7%. Probably underlying this, RTB figures showed that for both houses and apartments, the level of late 2021 rent inflation was highly correlated with property size – see Figure 6.7.
Similarly, according to Canadian rent trends during the pandemic to date have been highly diverse according to property type. Thus, at the national scale, although condominium rents dropped by 16% during 2020, single family house rents remained virtually unchanged, while (non-condo) rental apartments saw rents continuing to rise during this period.

6.2.4 Rental market trends across and within case study countries: summing up

The preceding sections have highlighted a number of striking commonalities and contrasts across and within countries in terms of rental market trends in 2020 and 2021:
At a country level, there appears to have been substantial similarity across the Anglophone nations – after being moderately dampened for a brief period at the start of the pandemic, rent inflation accelerated to historically high levels during 2021.

Across all of our case study countries it is Spain’s experience that has been the most notable outlier, with rents continuing to fall through 2021.

Across most of our case study jurisdictions certain capital (and other large) cities experienced unusually large and/or sustained rent reductions in 2020 and into 2021.

While data are available only for certain countries, continuing or renewed rent inflation in 2021 has tended to be substantially greater as it has affected houses rather than units.

In the next section we discuss some of the market demand and supply dynamics that may have contributed to these observed trends.

6.3 Rent drivers

6.3.1 Overview

There is significantly less econometric evidence on the determinants of rents than the determinants of house prices. In theory though, since a renter is only consuming housing services for a period of time, and not purchasing the underlying asset, rents ought to be affected by a narrower subset of those factors driving house prices (as discussed in Chapter 5). Incomes, demographics and housing tastes all remain fundamental, but macroeconomic variables like mortgage availability and interest rates are much less significant, as are expectations of future price changes (speculation).

A systematic analysis of factors potentially affecting rental market trends during the pandemic has been undertaken with respect to Australia (Pawson et al. 2021a pp37-59). That evaluation discussed a range of demand-side factors including:

  • International migration – international students14In the UK, where most domestic higher education students attend a university remote from their home town, this cohort also forms a major private rental demand cohort., permanent migrants
  • Internal inter-regional migration
  • Tourism
  • Household income changes
  • The rise of working from home.

Supply-side factors considered included:

  • Private rental stock utilisation (vacancy) rates
  • Tenancy turnover
  • Net stock additions (losses) from the owner occupied sector
  • New supply of housing built for renting (mainly apartments commissioned as off the plan acquisitions by individual landlords)

An equivalent analysis of all these drivers in the case study countries is beyond the remit of this report. Instead, the remainder of this section discusses certain factors considered likely to have particular significance in explaining the market trends summarised in Section 6.2.4.

6.3.2 Variations in public health outcomes

To some extent the national and sub-national variations in rental market trends in 2020 and 2021 highlighted above may be attributable to differential effects of the pandemic in relation to public health and economic performance. For example, consistent with the country’s largely successful exclusion of the virus during much of 2020 and 2021, the rental market in New Zealand – and also in the nation’s largest city, Auckland – stands out as largely unaffected by the crisis (see Figures 6.1 and 6.3). At the opposite end of the spectrum, the relatively high COVID-19 death rate recorded in Spain and the country’s particularly deep recession (see Chapter 1) could have directly or indirectly contributed to that country’s particularly subdued rental market. As further discussed below (see Section 6.3.3), some of this is probably bound up with the unusual structure of the Spanish economy.

At a sub-national level, the varied rental market trajectories of different Australian states and territories in 2020 and 2021 can also be partly correlated with contrasting impacts of COVID-19 and its associated economic fallout (e.g. Melbourne (heavily affected) versus Perth (almost unscathed) – see Figure 6.6).

6.3.3 Shutdown of international travel

While countries varied in the severity of border restrictions imposed during the pandemic, all will have been affected by the vastly reduced volume of international travel during 2020 and 2021. The rental housing market impacts are likely to have transmitted particularly via:

  • Reduced demand for short-term rental housing from overseas tourists – thereby encouraging some property owners to transfer properties (back) into an already well-supplied mainstream market
  • Reduced demand for medium to long-term rental accommodation from incoming migrant workers and international students.

As shown in Table 6.1 the countries covered in the current research are quite diverse in their vulnerability to housing market (and other economic) impacts of reduced international travel during the pandemic.

Table 6.1. National vulnerability metrics related to shutdown of international travel

As far as tourism is concerned, Spain stands out as far more vulnerable than the other countries covered in this study. This will have had a direct impact in terms of reduced demand for rental accommodation, but will also have contributed to the wider economic damage wrought by the pandemic as experienced in Spain in terms of employment and incomes (see Figure 1.2). Within the country, the geography of recent Spanish rental market trends as described in Section 6.2.2 is also likely to have been affected – as noted there, the tourist-dependent economies of Barcelona and the Balearic Islands reportedly saw some of the largest rent declines in the country in 2021.

Similarly, in Australia, the geography of rental market change in 2020 and 2021 is likely to have been affected by the spatial concentration of short-term rental ‘AirBnB’ properties. AirBnB listings for entire homes contracted by 17% in Sydney and 22% in Melbourne during the early months of the pandemic, with substantial impacts on inner city and beachside suburbs within these cities where such properties have been concentrated (Pawson et al. 2021b p53).

With a relatively large population component accounted for by international students (see Table 6.1), and with the country’s international border largely closed for two years from February 2020, Australia’s rental market is also likely to have been affected by this factor to a greater extent than the other countries in this study. Once again, given the uneven spatial distribution of international student settlement, these impacts will have been highly geographically concentrated. With many large universities located in inner city Sydney and Melbourne, demand for rental accommodation in these areas was substantially reduced as a result.

6.3.4 The race for space

As discussed in Chapter 5, the ‘race for space’ describes an apparently marked shift in housing consumption preferences seen in high income countries during 2020 and 2021. As quantified by the Bank of England in relation to the UK house sales market (Bank Underground 2021), this has involved the attribution of greater importance to property size (and possibly also private outdoor space) than was true prior to the pandemic. It has been explained partly as a response to domestic confinement due to public health restrictions and, in partly as a reflection of the substantial increase in working from home seen during the pandemic across most high income countries.

As shown in this chapter (see Sections 6.2.2 and 6.2.3) the past two years have also seen rental market trends possibly explicable in terms of the race for space. Disproportionate rent increases for houses, as opposed to apartments have been documented in several of our case study countries. Similarly, at least in certain jurisdictions – not least in Australia – there is evidence of non-metropolitan markets tending to ‘out perform’ major cities (as shown in Figure 6.6).

6.3.5 Rental housing supply

In seeking to explain rent trends, commentary and analysis tends to focus mainly on change in rental housing demand components (e.g. wage growth, international migration), while supply-side factors receive less attention. As listed in Section 6.3.1, there are a number of supply-side influences that could play a part here. In most countries, statistics that could shed light on such factors are thin on the ground. However, drawing on Australian evidence we can briefly discuss two supply-side dynamics with the potential to result in local or national impacts. It may well be that phenomena similar to those identified in Australia have been recently influencing markets in other countries similarly affected by pandemic conditions.

The size of the private rental stock is subject to change over time. In most of the countries covered in this study the past 10-20 years have seen this portfolio expanding relative to the overall body of occupied dwellings. This may come about through landlord acquisition of formerly owner occupied homes to be rented out. Alternatively, it may result from the completion of dwellings newly built for rental use. In certain areas of inner Sydney and inner Melbourne the onset of COVID-19 happened to coincide with a large influx of recently completed individual-landlord-owned rental apartments – the tail end of an ‘investor-driven’ high density building boom seen during the mid-2010s (Scutt 2016). This will have compounded the impacts of rent-depressing demand-side factors (e.g. reduction of student housing demand) on these localities, thereby helping to explain the extraordinary large rent reductions seen in these inner cities during 2020 (Pawson et al. 2021a).

A potentially more pervasive supply-side influence on rent levels is the tenancy turnover within the established stock of rental dwellings. There is reason to believe that the associated flow of properties becoming available for rent has substantially diminished during the pandemic. In Australia this has been reflected in the unusually small numbers of rental properties coming onto the market in 2020 and 2021. Thus, according to CoreLogic Australia, September 2021 saw new rental listings running at 23% below their 2016-2019 average, while total rental listings were 25% below the equivalent norm (Pawson et al. 2021a, Figures 3.16 and 3.17). Similarly, in Ireland homes advertised as available for rent on 1 August were at a 15-year low, with housing markets outwith Dublin most seriously affected ( 2021).

It may be that similar conditions experienced in other countries help to explain the sharp escalation of advertised rents generally in evidence in most of the Anglophone countries during the latter part of 2021 – see Figure 6.1).

6.4 Overall pandemic impacts on rents and affordability

In most of the case study countries for which such data are available, the pandemic saw only brief 2020 rent reductions at the national scale. Subsequently rents in most Anglophone countries have risen at remarkable rates (see Figure 6.1). As shown in Table 6.2, the net result is that by late 2021, in Ireland, New Zealand, the UK and the US, typical rent levels were more than 10% higher than those at the start of the pandemic.

Table 6.2. Overall national change in (nominal) rents during the pandemic to late 2021 compared with (nominal) change in earnings

In Australia the equivalent national (all property types) increase was 7.1%, but this conceals a huge divergence between capital city and regional rent trends during the period (see Figures 6.5 and 6.6). It is of course acknowledged that, as highlighted in Section 6.2.3, national norms also conceal substantial regional variations in other case study countries. Nevertheless, the general point can still be made that across most of the countries for which data are available (Australia, Ireland, New Zealand, UK, US), market rent norms had, by late 2021, increased during the pandemic by an amount exceeding parallel increases in wage rate norms, in most instances by a substantial amount. Only Canada, for which rent trend data may be less reliable, apparently bucks the trend.

Low income populations residing in rental housing are likely to be substantially reliant on state benefits – often indexed to general inflation – rather than on wages. Other than as affected by temporary supplements during the early part of the pandemic (e.g. as in Australia and the UK), such payments are unlikely to have risen by rates in excess of earnings.

Therefore, while this is acknowledged as a very broad brush statement, it would seem that the rental housing market impacts of the pandemic up until late 2021 have generally resulted in declining rental housing affordability. As observed in Chapter 3, these pressures have evoked no regulatory response of the kind deployed in the early months of the COVID-19 crisis to counter the pandemic-triggered household income shock. Indeed, governments in most countries were lifting restrictions on evictions and rent increases just as the rental price boom took off.

6.5 Conclusions

The COVID-19 pandemic and its associated economic turbulence have proven highly disruptive in rental housing markets, just as in relation to house sales transactions and prices. In most of the countries covered by the current research the extent and diversity of rental market volatility seen during 2020 and 2021 has been substantially greater than at any time since the 2008 Global Financial Crisis.

Across the eight high income countries covered in this research, observed rent fluctuations during the pandemic have resulted from major shifts in both demand and supply market drivers, as triggered by the public health crisis. Contrasting rental housing experiences in different countries partly reflect variations in the pandemic’s economic impacts – as particularly exemplified by Spain, where the damage to national income resulting from diminished tourist travel has been especially serious. Analysis of rental market trajectories at the sub-national-level, however, highlights substantial inter-regional and property type diversity that limits the value of national level analyses.

For most of the Anglophone countries covered in the study, one of the most striking findings is the rapid acceleration of rent inflation during 2021. This is all the more remarkable in countries such as Australia and New Zealand where, despite being previously understood as substantially driven by high rates of immigration, rent inflation has escalated in parallel with international borders essentially closed other than to returning citizens. Part of the explanation may lie in lower rates of rental housing turnover, suppressed by both eviction moratoriums and by the effect of travel restrictions and economic disruption on voluntary tenant moves.

On the demand-side, impacts might have arisen from savings having accrued to mid-high income renters (through cutting back on service consumption), and the possible willingness of benefiting households to increase rent spending as a result. If so, it might be expected that rent increases will have been weighted towards the higher end of the market – perhaps an area for more in-depth research.

With the Omicron wave still engulfing many countries at the time of writing a further set of pandemic impacts on rental housing markets may well be yet to emerge. Irrespective of that, a key question is the durability of the changes identified during COVID-19 to date, especially those resulting from the race for space.


ABS (2022) Average weekly earnings, Australia

Central Statistical Office (Ireland) (2022) Earnings and Labour Costs

CoreLogic Australia (2021a) August 2021 Housing Market Update – online seminar presentation, 23 August

CoreLogic Australia (2021b) Monthly Chart Pack, September 2021

CoreLogic Australia (2021c) Monthly Chart Pack, October 2021

CoreLogic Australia (2021d) Housing Market Update, November 2021

CoreLogic US (2021) US Single-Family Rents Up 10.2% Year Over Year in September – CoreLogic (2021) Q2 Rental Price Report: An analysis of recent trends in the Irish rental market 2021 Q2

FRED Economic Data (2022) Median usual weekly earnings of full-time wage and salary workers by SEX, quarterly, seasonally adjusted

Homelet (2021) HomeLet rental index report, November 2021

Idealista (2021) “How much does it cost to rent in Spain? Prices drop during the pandemic”

Javed, A. & Squires, G (2021) Rent Report September 2021; The Property Foundation

Massey University (2022) New Zealand Residential Rental Survey

ONS (2022) AWE: Whole Economy Level (£): Seasonally Adjusted Total Pay Excluding Arrears

Pawson, H., Martin, C., Thompson, S., Aminpour, F. (2021a) COVID-19: Rental housing and homelessness policy impacts; ACOSS–UNSW Poverty and Inequality Partnership Report No. 12,Sydney

Pawson, H., Martin, C., Sisson, A., Thompson, S., Fitzpatrick, S. & Marsh, A. (2021b) COVID-19: rental housing and homelessness impacts – an initial analysis; Report no. 7 for the ACOSS–UNSW Poverty and Inequality Partnership

Residential Tenancies Board (2021) Rent Index Q3 2021

Scutt, D. (2016) 3 charts that show Australia’s high rise construction boom is scaling to new heights; Business Insider, 21 July

Statistics Canada (2022) Employment, average hourly and weekly earnings (including overtime), and average weekly hours for the industrial aggregate excluding unclassified businesses, monthly, seasonally adjusted

Statistics NZ (2021) Labour market statistics (income): June 2021 quarter

7.1 COVID-19 and housing – expectations defied

The centrality of housing in national economies in all developed countries meant that the potential housing system impacts of COVID-19 abruptly emerged as a high priority concern in early 2020. Not only do housing production sectors in most such nations account for a significant proportion of GDP, but land and residential property values crucially underpin consumer spending and investment. When, around March 2020, most governments suddenly realised the impending need for severely restricted social interaction, this brought with it an immediate appreciation that housing markets could be badly destabilised as a result. As it turned out, of course, many such anxieties proved unrealised.

Albeit at very high cost, most OECD governments intervened with unexpected firmness in 2020 to head off prospective mass unemployment that could have triggered widespread housing payment defaults and the wider knock-on economic damage that would likely have followed. Importantly, mortgage lenders were encouraged to show flexibility to borrowers experiencing income losses, while landlords were temporarily prevented from terminating tenancies. Largely thanks to accompanying interest rate cuts, quantitative easing and effective income support, most countries in our study in fact saw only a brief dip in property prices and rents in mid-2020. By 2021, despite repeated waves of disease that continued to dampen wider economic performance, many were experiencing rapidly accelerating inflation in both house prices and rents.

In general, forecasters missed the distributional profile of (a k-shaped) recovery which allowed wealthier and economically resilient households to move to preferred housing (so-called COVID-buyers) – often involving a shift to a larger home, sometimes in a more suburban or rural location (Duca et al. 2021). Arguably, analysts also underplayed the pre-existing pressured status of several of these housing markets – e.g. in many regions of the United States, in New Zealand, in many western European nations and in Canada, as well as parts of Australia). That a medium term consequence of the pandemic was to significantly worsen housing affordability was unheralded, but looks now like an important legacy of the crisis. In retrospect, policies to stimulate the housing market premised on the expected downturn, such as transaction tax relief in the UK and ramped-up home buyer grants in Australia, look in hindsight to have been wasteful, volatility-inducing and unnecessary.

7.2 Diverse market outcomes

Notwithstanding many commonalities across our case study countries on COVID-19 housing policy responses as well as housing market impacts, the first two years of the pandemic also saw substantial contrasts – not only between, but especially within, countries.

At the national level, the German and Spanish housing systems appear to have experienced crisis effects markedly at variance from those in the Anglophone world. In Germany, pre-pandemic trends in both house prices and rents seem to have been relatively unaffected by the global economic turbulence of 2020 and 2021. This probably speaks of greater socio-economic resilience associated with a stronger welfare state. It may also testify to a more resilient housing system in which housing is somewhat less a speculatively traded assert, and rental regulation similarly provides greater security than that typically found in Anglophone (and many other) countries.

Spain’s housing market remained comparatively subdued in 2020 and 2021, perhaps largely reflecting the country’s heavy dependence on tourism which made it especially vulnerable to wider pandemic-induced economic damage.

Nevertheless, not least in Spain, but also in most of the Anglophone countries the pandemic to date has also witnessed major sub-national variations in housing market impacts. In the Spanish case it seems likely that this mainly reflects contrasting regional economic structures. The hypothesis would be that especially tourism-dominated housing markets have been particularly hard hit by a decline in accommodation demand from both tourists themselves and from the low income workers servicing the tourist economy. Similarly, in some other countries, spatially diverse housing market trends in 2020 and 2021 likely result substantially from the effect of closed borders in obstructing international student travel. This has certainly been a major factor in the ongoing flatness of inner city rental housing markets in Sydney and, especially, Melbourne.

Lockdowns also briefly halted the housing market transaction and construction activity proceeding in the UK and this was also associated with big temporary supply shocks – e.g. the de facto closing of the entire short-term lettings market. Lockdown also materially impacted on the normal course of business – e.g. the delivery of affordable supply programs was hampered by the public sector and public agencies shifting to remote working.

On the other side of the coin, at least in some of our case study countries, the pandemic appears to have indirectly triggered changes in housing consumer behaviour that have altered housing market trajectories in favour of suburban and non-metropolitan areas rather than inner cities, and in favour of larger dwellings (typically houses) rather than smaller properties (typically apartments). This may have come about due to the mass transfer of service industry work-stations from offices to homes and/or as a reaction to prolonged lockdowns as these were experienced by small apartment residents.

The big question that arises from such changes is the extent to which large-scale working from home is now embedded and, if so, whether associated shifts in housing preferences are similarly locked in. If the answer to that question is also yes, it could perhaps signal a reversal in the 2010s trend of ‘superstar city’ housing markets diverging from their national norms (see Figure 5.4). Or, alternatively, it could presage a restructuring in sub-national housing market pecking orders where ‘lifestyle’ localities are promoted closer to the top of such rankings.

7.3 COVID-19, housing and inequality

Much that it unleashed a tide of uncertainty and fear, the sudden onset of the COVID-19 crisis also prompted hopes that it might serve as a ‘focusing event’ with the potential to prompt an enduring policy re-set. For many of the countries we examined, and despite thorough-going income protection, social solidarity and economic security measures, the trajectory of housing outcomes reported in earlier chapters reflects the pre-COVID inequalities and policy settings in those countries.

In some countries, 2020 emergency actions to stave off mass unemployment and/or to utilise social security systems as a vehicle for economic stimulus had a markedly egalitarian impact. Social security benefits adjustments of this kind might have seemed to imply a pandemic-prompted official recognition of pre-existing inequalities or injustices. The UK Government’s 2020 decision to restore the decayed link between housing allowance rates and local rent levels serves as a good example. However, subsequent confirmation that adjustments of this kind were purely temporary made clear that the pandemic had failed to trigger the enduring policy re-set many had hoped for.

In the bigger picture, the crisis had in fact tended to crystalise and compound many pre-existing sources of inequality and poverty, a shift further reinforced by the k-shaped unequal economic recovery, as widely unfolding during 2021.

7.4 Looking to the future

There is scope for the economic recovery from COVID to be constructively supported by housing activity. Firstly, as summarised in Section 4.4, across our case study countries a number of national and state governments pledged new social housing investment programs in the course of the pandemic, with some of these clearly stimulated by the crisis (as opposed to being already in train before it). Secondly, there are opportunities for planners and developers to facilitate rental investment (from market to affordable and social), mixed tenure and in-fill sites in city centres that were hitherto largely non-residential, and where demand for office space may have been permanently eroded. This could also support regional economic productivity and provide less expensive housing for workers (Maclennan et al. 2021).

Thirdly, the growing importance of housing retrofit to meet climate change targets affords opportunities to build back the economy, with labour-intensive fabric-first construction work and manufacturing/installation and maintenance opportunities in residential renewable energy systems. In British city regions and internationally through developing green mortgage finance, these ideas combine net zero aspirations, economic recovery from COVID and restructuring economies in a genuinely forward looking way by investing in renewing and sustaining the housing stock. A good example of this ambition is the Glasgow City Region strategy, which seeks to use the retrofit of more than 400,000 homes across the region as the centrepoint of a regeneration effort that would green the economy, reskill the labour force and combine fabric first refurbishment with decarbonised energy systems.

The recovery from COVID will not in any case have a smooth course unaffected by other emerging challenges. Countries like the USA and the UK and also across the EU face significant re-emerging inflation and rising interest rates for the first time in a generation. While the UK faces the unique supply chain disruption associated with exiting the EU, all of Europe faces a new energy cost crisis requiring large scale public spending interventions (e.g. in Germany, France and the UK). Early 2022 saw energy tariff increases of more than 50% announced in the UK. Even with government support, halving these increases, millions face severe unenviable cost of living trade-offs.

Across the world, COVID-19 dominated economic policymaking and public discourse throughout 2020 and 2021. Albeit in key respects contrary to predictions, the direct and indirect impacts on housing systems have been profound. The pandemic, of course, remains ongoing at the time of writing (early 2022). Moreover, the social and economic impacts of the crisis will take years to play out, with their full implications likely becoming only gradually apparent. For example, not all of the novel housing market trends we have identified as having emerged in 2020 and 2021 are likely to be sustained. Nevertheless, we believe this rapid overview and assessment of COVID-19 effects on housing markets and housing policy to date has yielded worthwhile insights of potential value in informing future official decision-making, and in providing a basis for more narrowly focused in-depth research.


Duca, J., Hoesli, M. and Montezuma, J. (2021) The Resilience and Realignment of House Prices in the Era of Covid-19; Journal of Real Estate Research, online first, DOI: 10.1108/JERER-11-2020-0055

Maclennan, D., Long, J., Pawson, H., Randolph, B., Aminpour, F. and Leishman, C. (2021) Housing: Taming the Elephant in the Economy; Sydney: UNSW


The authors thank the UNSW-ACOSS Poverty and Inequality Partnership for supporting the project, as well as the following research co-funder organisations:

  • Mission Australia
  • National Shelter (on behalf of Shelter NSW and Shelter WA)
  • Queensland Shelter

Thanks are additionally due to Penny Dorsch (ACOSS), co-ordinator and production manager of UNSW-ACOSS publications, and to colleagues who kindly commented on an earlier draft of this report: Professor Carla Treloar (UNSW Social Policy Research Centre), Marion Bennett (Mission Australia), Charmaine Crowe (ACOSS) and Jacqui Phillips (ACOSS). The research team is also most grateful to Dr Nicola Powell ( who kindly provided unpublished rents statistics to inform Chapter 6.

The authors are additionally indebted to an international cast of colleagues who crucially contributed to the research as expert advisers on COVID-19 impacts and policy responses in their countries:

Canada: Steve Pomeroy, Senior Research Fellow, Carleton University
Canada: Jim Dunn Professor, McMaster University and Director, Canadian Housing Evidence Collaborative (CHEC)
Canada: David Crenna, Former senior public servant, Government of Canada
Germany: Michael Voigtlander, Professor of Real Estate, Institute of the German Economy
Germany: Volker Busch-Geertsema, Professor, European Observatory on Homelessness
Ireland: Conor O’Toole, Associate Professor, Economic and Social Research Institute
Ireland: Joe Finnerty, School of Applied Social Studies, University College Cork
New Zealand: Graham Squires, Professor, Economics and Finance, Massey University
New Zealand: Kay Saville-Smith Research Director, Centre for Research, Evaluation and Social Assessment
Spain: Paloma Taltavull de la Paz Professor of Applied Economics, University of Alicante
Spain: Fran Calvo Garcia Universitat de Girona
Spain: Sergio: Nasarre-Aznar, Professor, Faculty of Law, University Rovira i Virgili
United States: Steve Malpezzi, Professor Emeritus, University of Wisconsin
United States: Alex Schwartz, Professor of Public and Urban Policy, The New School, New York

Suggested citation

Pawson, H., Martin, C., Aminpour, F., Gibb, K., Foye, C. (2022) ‘COVID-19: Housing market impacts and housing policy responses – an international review’ ACOSS/UNSW Sydney Poverty and Inequality Partnership Report No. 16, Sydney


ISSN: 1326 7124
ISBN: 978 0 85871 097 9

© Australian Council of Social Service and UNSW, Sydney

This publication is copyright. Apart from fair dealing for the purpose of private study, research, criticism or review, as permitted under the Copyright Act, no part may be reproduced by any process without written permission. Enquiries should be directed to the Publications Officer, Australian Council of Social Service. Copies are available from the address above.