Technical ReportPDF Available
Housing
and poverty
in Europe:
Examining the
interconnections
in the face of rising
house prices
Rod Hick, Marco Pomati
and Mark Stephens
April 2022
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices
2
Acknowledgements
This report presents findings from a research project funded
by the Economic and Social Research Council [Grant number
ES/S016694/1]. We thank the ESRC for providing this
funding and the reviewers and panel members who consid-
ered the original proposal and provided helpful suggestions
in relation to our programme of research.
We thank our advisory board members - Tristan Carlyon,
Caroline Dewilde, Freek Spinnewijn, Martin Lux, Michelle
Norris and Emma Williams – for their suggestions and
feedback on our work. As part of this project, we held two
webinars on the topics of housing deprivation and housing
affordability, respectively. We thank Collette Bennett,
Caroline Dewilde, Freek Spinnewijn, Vicky McCall and Alice
Pittini for presenting their work and participating alongside
us at these events, and to those who attended for their
participation and feedback too. Our thanks are also due to
Barry Diamond for typesetting this report. We alone are
responsible for the content of this report and any oversights
or errors are ours alone.
Suggested citation: Hick, R., Pomati, M. and Stephens, M.
(2022), ‘Housing and poverty in Europe: Examining the
interconnections in the face of rising house prices’, Cardiff,
Cardiff University.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 3
Executive Summary
This report presents an overview of findings from a
two-year research study examining the relationship
between housing and poverty in a comparative European
context. Debates about housing, poverty, and social
policy more broadly have taken place in ‘disjointed
literatures’ for many years and a starting point for the
project that led to this report was the belief that these
debates have operated at too great a remove for too long.
Ours was a comparative, quantitative study and consisted
of analysis of data from the European Statistics on
Income and Living Conditions (SILC) survey between
2004 and 2018, the final year that we have data for the
United Kingdom.
Our project had four research questions. These were:
- How, if at all, have European housing systems
changed over the last 10 years?
- What is the association between housing deprivation
and poverty, and how does this vary between
countries and over time?
- What is the relationship between housing costs and
poverty, and how does this vary between countries
and over time?
- To what extent does independence for young people
vary between countries and over time, and how does
this relate to their incidence of poverty?
Following the Global Financial Crisis, house prices
evolved in different directions in different parts of Europe
– falling in the Anglophone nations and in Southern and
Central and Eastern Europe, and rising in Scandinavia
and in much of Continental Europe. However, since about
2013, we have observed consistent house price rises in
all parts of Europe.
European housing systems have undergone significant
change in the period between 2004 and 2018.
Homeownership has declined, including – sometimes
especially – for low-income households. Young people
have started to leave the family home later than before
and are less likely to become homeowners when they do
live independently. In a relatively short observation
window of less than 15 years, these are significant
developments.
Housing deprivation varies significantly across the
nations of Europe and is heavily concentrated in some of
the poorer nations of Central and Eastern Europe and in
some nations in Southern Europe. In each welfare
regime, renters experience greater rates of housing
deprivation than owners, but differences between richer
and poorer countries continue to dominate. This picture
is also observed with respect to poor and non-poor
households, too: poor households face an elevated risk of
housing deprivation everywhere, but non-poor
households in Central and Eastern Europe face greater
levels of severe housing deprivation than poor
households in the UK and Ireland.
The extent of housing affordability problems also varies
significantly across the nations of Europe, though
between-country differences are less obviously related to
levels of wealth or explained by welfare regime
classifications. Within countries, housing affordability
problems are often, but not always, more prevalent for
market renters than mortgaged homeowners.
Considering change over time, our research finds that,
controlling for a variety of compositional characteristics,
the position of market-rate renters vis-à-vis mortgaged
homeowners in relation to housing affordability has
deteriorated between 2010 and 2018 across most
European nations.
Our research has three implications for research and two
for policy. In terms of implications for research, the
strength of our findings in relation to levels of wealth and
some housing outcomes – especially, housing deprivation
– leads us to believe that the significance of economic,
as opposed to political, explanations in accounting for
between-country differences needs to be re-evaluated
within comparative study. Second, the relative
consistency with which we observe tenure-related
differences in housing outcomes – across measures and
countries, leads us to conclude that some of the claims
made in relation to its non-comparability between
countries are exaggerated. Third, there is a need to
improve the quality of data in SILC in relation to its
housing variables. It is also vital that the UK, which has
not participated in SILC since 2018, re-joins this survey.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 4
In relation to implications for policy, first, we find that the
severe housing deprivation measure is made up of two
very different sub-components – deprivation in housing
conditions and overcrowding – which are not highly
correlated and respond to different drivers. This leads us
to believe that it would be preferable for Eurostat to
monitor these subcomponents separately. Overcrowding
has been reduced to negligible levels in some nations,
while the deprivation of housing conditions seldom falls
below 10%, suggesting it has a higher “lower bound”. This
in turn suggests that significant reductions in severe
housing deprivation are likely to come via the
overcrowding rate. Since this appears strongly related to
absolute differences in wealth, it appears that economic
growth – assuming this is translated into a growth in
household incomes – would lead ultimately to a
significant reduction in housing quality problems, as
captured by the severe housing deprivation measure.
Our second policy recommendation concerns the growing
significance of housing tenure for housing affordability.
We have found evidence that the relative position of
market-rate tenants has deteriorated relative to
mortgaged homeowners in the period between 2010 and
2018, after controlling for a series of compositional
variables. One reason this matters is that households
renting their housing at market rate are disproportionately
likely to be poor. This speaks to the need to take housing
tenure more seriously as a stratification marker in
European societies and of the need to ensure housing
policy does not significantly favour owner-occupiers at the
expense of renters. This points to a number of possible
policy options, the effectiveness of which might depend
on the particular country or even city context, and might
include policies targeted at the private rental sector, the
reduced rental sector or homeownership. Since supply
shortages will tend to result in higher prices or rents, it
also underscores the importance of a well-functioning
housing market where such shortages are avoided.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 5
Introduction: From ‘Disjointed Literatures’
to the Turn Towards Housing
The significance of housing in terms of the experience of poverty and deprivation is both multi-faceted
and self-evident. Shelter is one of our core human needs, without which life itself is threatened – being
adequately housed is thus intrinsically important and the absence of adequate housing can be
considered a fundamental deprivation. But housing is also instrumentally important: high housing costs
can prevent families from meeting their non-housing needs and can push them into poverty.
Despite these important interconnections, debates about
housing, poverty, and social policy more broadly have taken
place in ‘disjointed literatures’ for many years (Stephens and
van Steen, 2011) – at separate conferences organised by
distinct learned societies and published in the pages of
different academic journals. A starting point for the project
that led to this report was the belief that these debates have
operated at too great a remove for too long.
This separation became more problematic following the
Global Financial Crisis of 2007-8, where developments in the
property market of the United States – in particular, the
securitisation of mortgage finance and extension of “sub-
prime” lending – precipitated an international economic
crisis. The Eurozone crisis that began in 2010, which can be
seen as the second act of the Global Financial Crisis, led to
Greece, Ireland, Portugal and Cyprus requiring external
assistance from the EC/ECB/IMF “Troika”, which demanded
the imposition of austerity in exchange for ongoing finance.
The scale of the shocks, and the speed at which countries
rebounded from them, varied significantly across Europe.
Central banks sought to stimulate national economies
through a combination of quantitative easing and ultra-low
interest rates, both of which boosted asset prices, including
house prices. By the onset of the next great crisis, the
Coronavirus pandemic, there were substantial concerns
about the affordability of housing in many nations.
After a long period of debate taking place in disjointed
literatures, things are, perhaps, beginning to change, and a
series of recent studies suggest a turn towards housing
within social policy and amongst scholars of poverty. In a
study of housing and poverty in Australia, Peter Saunders
(2017) calls for greater attention to the distributional and
social consequences of rising housing costs. Measuring
poverty after housing costs (AHC), he finds, leads one to
observe to a greater rise in poverty and inequality in the
pre-Global Financial Crisis years and a smaller reduction in
the period since the crisis than are observed if poverty is
measured on a before housing costs (BHC) basis. Alcántara
and Vogel (2021), more recently, examine the impact of
changes in housing costs on poverty amongst older people
in Germany over the period 1996-2016 and find that rising
housing costs are contributing to increasing poverty risks.
The growing salience of inequality and, relatedly, wealth has
resulted in newfound attention to housing as a – often the
– central component of private household wealth. A recent
special issue on social policy and wealth (Marx and Nolan,
2021) included contributions examining evolving inequalities
in housing wealth amongst young people in the decade since
the financial crisis (Dewilde and Flynn, 2021) as well as the
political consequences of housing unaffordability (Ansell and
Cansunar, 2021). In the Comparative Political Economy
literature, a recent account of twenty-first century “growth
regimes” emphasises the significance of housing market
financialisation for national growth strategies (Hassel and
Palier, 2021; Reisenbichler, 2021). Having once been
peripheral to scholarship on contemporary capitalism,
welfare states and living standards, we are perhaps
witnessing the beginnings of a turn towards housing.
This report represents the culmination of a two-year research
project funded by the Economic and Social Research
Council. In order to build bridges between these disjointed
literatures, ours has been an interdisciplinary project,
conducted by two researchers of poverty who have found
themselves turning to developments in field of housing to
make sense of those in relation to poverty in recent years
(Hick and Pomati) and a housing studies scholar who has a
longstanding interest in its links with poverty (Stephens). As
we outline in the next section, our study is a quantitative
comparative analysis of key aspects of this relationship,
drawing on a major Europe-wide dataset for the period 2004-
2018.
In this final project report, we (i) present an overview of some
of the themes we believe are fundamental to any
interdisciplinary, comparative study of housing and poverty;
(ii) outline the aims and objectives of the project and its
research questions, (iii) present descriptive data of how
housing systems in Europe are changing, (iv) provide an
overview and summarise key findings from more detailed
work conducted as part of the project that we anticipate will
be published in a series of academic journal articles, and (v)
outline a series of recommendations.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices
1. There are of course important and recent exceptions – for example, Angel (2021); Dewilde (2021).
6
How Housing Matters: Literatures and Key Debates
In setting the scene, we introduce six themes which relate to the study of housing and poverty. First, we
explain why we think comparative research can add value to understanding the relationship between
housing and poverty (“the possibilities and limits of comparative research”) before discussing some of
the key attempts to understand differences between housing systems across countries (“comparative
analysis and housing systems). Subsequently, we examine the contested position of housing tenures as
categories of analysis (“on housing tenures”).
In the final three sub-sections, we consider how EU
enlargement post-2004 and the expansion of comparative
datasets present ‘new’ challenges for analysts (“housing and
poverty in an enlarged Europe”), present trends in poverty
(“distributions and trends in poverty in Europe”) and in
house prices in Europe (“Rising house prices in Europe”).
The possibilities and limits of the
comparative approach
Comparative studies – especially of the large-N variety – do
not occupy as prominent a position in the housing studies
literature as they do in the comparative welfare state
tradition.1 Indeed, there is a certain scepticism towards
large-N quantitative comparative studies in some quarters,
often associated with an alleged over-emphasis on housing
tenure as an organising concept (e.g. Zhang, 2021: 11-12,
see also below). It is important, however, to consider both
the advantages as well as the limitations of comparative
quantitative analysis.
Two advantages of comparative study are its capacity for
verifiability and generalisability. Verifiability matters because
it is common in public and political debate to hear claims
made about, for example, rising housing costs and how
these are tipping families into poverty, or that the
financialisation of housing is associated with deteriorating
housing affordability. The quantitative comparative approach
is well-placed to assess if these claims are true and it can
thus offer an evidence-based counterpoint to accounts that
are solely reliant on theory or received wisdom.
Generalisability matters because such claims are sometimes
made in a general sense (i.e. that they are occurring
everywhere) or, conversely, are made in respect of the
supposed distinctiveness of one nation (e.g. that the German
housing system differs very substantially from that of other
nations). The large-N comparative approach can thus help to
examine whether patterns observed in one nation are
unique, rare, common or universal. For example, there has
been a substantial critical literature published in recent
years on the commodification and financialisation of
housing, and it is often claimed that these processes lead to
a deterioration in housing affordability. However, empirical
evidence that these processes are a) occurring, b) are
associated with a predictable set of housing outcomes and
that c) this is occurring in a generalised – that is, broad
based – fashion are in shorter supply than they might be.
Comparative analyses have significant potential to enrich the
evidence base and to shed light on these questions.
There are, however, also limitations. One practical obstacle
– not unique to comparative study, but a problem
nonetheless – is that the housing outcomes we might
legitimately be interested in are plural, contested, and
interact with one another in ways that are often
underexamined. While the reduction in poverty attributable
to taxes and transfers has been suggested to be ‘arguably
the single most relevant measure of welfare state
redistribution’ (Esping-Andersen and Myles, 2011: 21), the
possible housing outcomes of interest can be much more
variable, relating to housing affordability, dwelling conditions,
overcrowding, housing aspirations, and more. A lack of
agreement about the relevant dependent variables can
frustrate agreement about the patterns of change and the
dynamics that explain it.
A related limitation is that some important housing variables
are not well-measured in key datasets. A particular problem
is that housing tenure – an important variable for any
comparative study of housing – is not well-measured in a key
European dataset, the Statistics on Income and Living
Conditions survey (SILC). The benefits of the comparative
quantitative approach can be frustrated by these issues,
especially when, as in the case of the latter, the problem is
beyond the control of any analyst. What is required, then, is
working both with what we have to enrich the evidence base
while also seeking improvements in data collection in key
resources such as SILC.
On welfare regimes and housing systems
The clearest point of correspondence between comparative
housing studies and the field of social policy comes in terms
of the literature on welfare regimes and housing systems.
Esping-Andersen’s (1990) identification of Liberal,
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 7
Continental and Social Democratic welfare regimes in the
countries of Western Europe has had a deep and enduring
influence on the comparative welfare state literature.
Esping-Andersen’s work was based on an analysis of
quantitative data for 18 nations and came under challenge
from a variety of quarters – in relation to country-
classifications, extensions to new nations, coherence, and
more. The use of welfare regimes as an analytic device in
comparative European social policy as well as in poverty
scholarship more generally remains commonplace and
recent accounts continue to point to the validity of this
classification in terms of predicting the redistributive
performance of welfare states (e.g. van Kersbergen and Vis,
2014).
In the field of housing studies, Jim Kemeny (1995) offered an
account of housing system differences which, he claimed,
emerged from differences between their rental systems. In
some nations, government allowed a profit rental market to
develop ‘unhindered’ (1995: 18). But since the private
market was unable to meet families’ housing needs on a
mass basis, this forced governments to intervene to provide
limited cost renting housing, which would not compete with
for-profit housing. This ameliorated unmet housing need for
some, but resulted in unsatisfactory housing conditions for
many in the for-profit market. The result was a ‘dualist’ rental
system which, significantly, shaped household preferences
towards owner-occupation. In other nations, the government
invested in and encouraged cost rental housing with the aim
of competing with the private rental market. This competition
helped to raise floors in relation to housing quality and
moderate housing costs for renters. Eventually, and in its
purest form, a unitary rental market would be formed, where
(free) market forces did not predominate. In this model the
greater desirability of renting meant that policy did not shape
preferences towards owner-occupation in the same way, but
instead remained tenure-neutral. The dualist rental system
was identified as comprising mostly English-speaking
countries and the unitary rental system consisting of
Sweden, the Netherlands, Germany, Switzerland, Austria and
Denmark (Kemeny, 1995: 56-58). While their underlying
logics differ, there is an overlap between the nations of the
Liberal welfare regime and dualist rental systems and unitary
rental systems and Continental and Social Democratic
welfare regimes. In contrast with the work of Esping-
Andersen, Kemeny’s classification is less clearly-defined:
nations are said to be “moving towards a unitary rental
market (1995: 58; emphasis added), rather than being
classified unambiguously in this way.
In contrast to Esping-Andersen’s welfare regimes, Kemeny’s
housing systems have fared rather differently. In one sense,
their basis was never so secure: Kemeny’s work classified
only eight countries and did so on a more descriptive basis
than Esping-Andersen’s quantitative classification. Moreover,
there are a number of weaknesses to Kemeny’s thesis. One
is that although rental systems might accurately be
described as being ‘dualist’ or ‘unitary/ integrated’, it does
not follow that the entire housing system is defined by these
relationships. This is the case in the US where interventions
other than public housing are more significant (Blessing,
2016); and in the UK where in the second half of the 20th
century the key division was between social renting and
home-ownership (Stephens, 2020). Further, the schema is
really limited to explaining the development of housing
systems in countries that did opt to build subsidised housing
on scale in response to general shortages and attempts to
extend it outside this relatively small group of countries
appears incongruous, because, again, whether a (usually)
dualist rental market exists does not necessarily define the
housing system as a whole. This is particularly the case in
relation to the nations of Central and Eastern Europe, where
rates of outright ownership are very high, which jars with the
idea that national housing systems are defined by the nature
of their rental markets.
In empirical analyses, some measure of the balance
between tenures is often chosen to capture differences
between housing systems in comparative housing studies.
However, the issues we have discussed above and the sheer
diversity of tenure types that we demonstrate empirically
below, leads us to rely on welfare regimes to cluster
countries in the empirical work we present.
On housing tenures
Housing tenure is frequently treated as being a – perhaps
the – central variable in housing studies, but it is ambiguous
in terms of interpretation – theoretically, normatively and
empirically. It is recognised that tenure categories, such as
owner occupation”, “social rented”, etc, do not capture
narrowly defined and unambiguous categories that hold
meaning at all times and places. Tenure categories thus
contain contingent qualities, which can vary over time and
space. These include its legal basis which is specific to the
jurisdiction in which it is located, its financial qualities (such
as its liquidity as an asset) and its cultural symbolism (for
example, cultural preferences for home-ownership)
(Stephens, 2020). Tenure is by no means the only relevant
variable in housing studies, but as Barlow and Duncan
(1988: 229, emphasis in original) observed, “[i]n substantive
terms – that is in defining social relations of occupancy and
ownership – tenure can be important and should be
expected to have certain social and political effects”.
It is worth pausing to consider what the contingency of
tenure means for its use as a variable in comparative study.
Zhang (2021), for instance, suggests that the contingency of
housing tenure means that owner-renter distinctions should
be abandoned in cross-national work. Indeed, resistance to
tenure as an organising concept may inform dissatisfaction
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 8
with the comparative method altogether, since this tends to
rely on it – at both the country level, to account for
differences in, say, rental systems or housing system
commodification or, at the micro-level, focussing on
outcomes between households living in dwellings occupied
on the basis of different tenure arrangements.
It is worth, on this question, considering the demands
imposed by the comparative method. It is sometimes
suggested – against the prospects for comparative housing
research – that, for instance, renting in Germany comes with
a fundamentally different set of rights and obligations than
in the UK and that, therefore, analyses based on tenures
might essentially be non-comparable. We would stress at
this point a distinction between contingency and non-
comparability. It can be acknowledged that the specific
bundle of rights and obligations associated with tenure types
may vary over time and place – that is, that they are
contingent. But acknowledging this contingency does not
imply that these categories are non-comparable. Indeed, for
some questions – for example, have rents been rising for
market-rate tenants to a greater extent than wages across
Europe? – there is value in examining differences between
different institutional contexts to see whether patterns of
outcomes are broad-based (that is, generalised). In short, we
are not persuaded that differences in the substantive
meaning of housing tenure render comparative analysis
redundant though, of course, tenure differences will
inevitably provide a high-level and partial account of housing
(system) differences between and within countries.
Housing and poverty in an enlarged Europe
One of the most significant developments since the earlier
generation of comparative studies of Esping-Andersen
(1990) and Kemeny (1995) has been the expansion of the
European Union to 28 Member States (prior to the UK’s
departure) and, relatedly, improvements in data quality which
mean that comparative datasets such as the Statistics on
Income and Living Conditions survey (SILC) now includes a
greater number of countries than were considered in these
early comparative accounts. The result is that we have more
data on which to draw comparisons and test hypotheses, but
there are also new challenges since country welfare regime
and housing system classifications were constructed to
make sense of differences between a smaller number of
more similar nations. Important debates have taken places
in both the comparative welfare state and housing studies
literatures about the extent to which the logics underpinning
these county-clusters can accommodate a wider range of
nations, with lower levels of economic development.
The literature on poverty has also paid particular attention to
the consequences of EU enlargement for the analysis of
poverty. Fahey (2007: 35), for example, notes that:
‘Following the recent eastern enlargement of the EU,
the gap in living standards between the richest and
poorest Member States has greatly widened, so much
so that what is defined as the poverty threshold in the
richest Member States would count as an above
average income in the poorest Member States, and
the ‘poor’ in some states have higher living standards
than the well-off in other states.’
This has led to a debate about whether poverty needs to be
measured in a ‘less relative’ way so as to reflect these wide
differences in living standards. One way that this has been
done is to measure poverty as the enforced lack of a fixed
set of material and social deprivation items. This measure is
now included, along with the relative income (‘at risk of
poverty’) target and a measure of quasi-joblessness, as one
of the three official poverty measures of the European Union
(see Copeland and Daly, 2012, for a discussion, and Nolan
and Whelan, 2011, for a critique).
Less commented-upon in relation to welfare regime or
housing system classifications is whether the explanatory
basis of welfare outcomes is altered by the inclusion of a
wider number of nations at lower levels of economic
development. Underpinning both Esping-Andersen’s and
Kemeny’s frameworks was the idea that institutional
differences were fundamentally political in orientation,
reflecting differing policy logics (and their causes). We have
noted above that these regime differences are viewed as
having predictive power in relation to social outcomes, such
as poverty – at least in the case of the welfare regime
literature. Modernisation theory – here, the idea that
improving social outcomes might be dependent on economic
modernisation and development – has largely been cast
aside. But even The Spirit Level, an account of the
importance of economic inequality in explaining health and
social outcomes, and which argued that ‘economic growth,
for so long the great engine of progress, has, in the rich
countries, largely finished its work’, suggested that this was
only the case once GDP per capita rose above about
$25,000 (Wilkinson and Pickett, 2010: 5-9). In their own
data, that threshold divided Central and Eastern Europe on
the one hand, and Northern and Western European nations
rather neatly, with only the nations of Southern Europe
positioned across this line. And yet, the question of whether
growth in household incomes is required, at least in Central
and Eastern Europe, has not received the attention that it
deserves. That is, enlargement calls into question the
conclusion that outcome differences are explained primarily
by political, as opposed to economic, factors and warrants
further empirical work.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 9
Distributions and trends in poverty in Europe
Examining the at risk of poverty measure and the material
deprivation measure of poverty, respectively, we can see that
these suggest very different patterns of poverty across
Europe – and also display different trends over time. Figure 1
presents rates incidence of poverty in Europe on the at risk
of poverty measure (AROP) and the material and social
deprivation rate, with countries ordered in terms of their
performance on the latter. As is widely known, incidence of
material deprivation varies much more widely than does the
relative income poverty rate and is more concentrated in the
poorer nations of Central and Eastern Europe (though,
notably, to a lesser extent than was the case some years
ago). In contrast, incidence of the at risk of poverty measure
varies less widely and is less obviously patterned by
geography. It is this measure that we focus on in the
empirical work we present below.
Source: Eurostat. AROP refers to the at-risk-of-poverty rate, the proportion of the population whose equivalised, disposable
income falls below 60 per cent of the national median. Material and social deprivation refers to the proportion of the population
unable to afford five out of thirteen deprivation items.
Figure 1. Poverty rates in European nations on two measures, 2018
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 10
Figure 2. Trends in poverty and material and social deprivation over time, EU-28
Source: Eurostat.
Figure 2 examines the same two measures, this time
presenting aggregate trends over time. While the at risk of
poverty rate displays “disappointing poverty trends”
(Cantillon, 2011) – here, increasing by 0.5ppts between
2010 and 2018, the trend for material deprivation is
anything but disappointing, falling by one-third across the
whole of the EU-28 in just a four-year period between 2014
and 2018. This reflects, in absolute terms, the catch-up of
Central and Eastern Europe in terms of living standards,
which is also observable in both income and
multidimensional perspectives (Goedemé et al., 2019; Hick,
2016).
Source: Eurostat (House price index)
Figure 3. Trends in nominal house prices by welfare regime in 20 European countries, 2007-2019
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 11
Rising house prices in Europe
Finally, we present data on house prices, which have been
the source of much public debate in recent years. Eurostat
provide two sets of house price statistics. The first – the
house price index2 – captures changes in house prices for all
residential properties and Figure 3 captures the evolution of
these prices between 2007 and 2019 for 20 European
countries, clustered by welfare regime.3-4
We see in Figure 3 that house prices fell significantly
between 2007 and 2012 in the Anglophone nations.
This fall is primarily driven by dramatic falls in house prices
in Ireland; in the United Kingdom, prices also fell, but more
modestly. Prices also fall in Southern and in Central and
Eastern Europe until about 2013. In contrast, prices rose
from 2009 onwards in the Continental and Social
Democratic worlds. Thus, house prices evolved in different
directions immediately following the financial crisis and
Eurozone crisis. Since about 2013, however, we see prices
rising consistently across Europe.
2. The House Price Index (HPI) measures inflation in the residential property market. The HPI captures price changes of all kinds of residential property
purchased by households (flats, detached houses, terraced houses, etc.), both new and existing. Only market prices are considered, self-build dwellings
are therefore excluded. The land component of the residential property is included.
3. Data are unweighted cluster averages. No data for Austria, Croatia, Czechia, Greece, Italy, Poland, Portugal and Romania.
4. Throughout the report, countries are clustered into welfare regimes as follows: Anglophone: UK and Ireland; Continental : Austria, Belgium, France,
Germany, Luxembourg, the Netherlands; Social Democratic: Denmark, Sweden, Finland; Southern Europe: Cyprus, Italy, Greece, Spain, Malta, Portugal;
Central and Eastern Europe: Bulgaria, Croatia, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, Slovenia, Romania.
A second index that we are able to analyse is the owner-
occupied housing index, also provided by Eurostat. This
differs to the data analysed in Figure 3 because (i) it
considers properties bought for owner-occupation only, (ii) it
considers only new dwellings that are made available for
residential living, (iii) it includes the cost of some services
associated with buying the house – e.g. insurance and
repairs and maintenance and (iv) data are provided for a
wider range of countries (data is only missing for Greece).
The time period that we have data from this index for is also
shorter. The picture here is similar but not identical. The
most obvious difference concerns Central and Eastern
Europe, where prices are now observed to rise fairly
consistently. Reductions in the Anglophone in the years
2010 to 2013 continue to be observed, but are more
modest than in the house price index presented above.
Consistent house price rises across Europe since about
2013 continue to be observed.
From this analysis, we can see that the diversity of
experiences during the financial crisis have given way to
consistent house price increases across Europe from about
2013. But what is still unclear is how aggregate house price
rises translate into affordability problems for households.
This relationship is not straight-forward because (i) housing
affordability problems are treated as the relationship
between housing costs and household incomes, so rising
prices may not lead to affordability problems if outstripped
by the latter and because (ii) one-off sale prices for new
homeowners do not relate to the ongoing housing costs
borne by owners and renters.
Source: Eurostat (Owner-occupied house price index). Data for Greece are missing.
Figure 4. Trends in nominal prices of new owner-occupied dwellings by welfare regime, 2010-2019
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 12
The Housing and Poverty in Europe Project
Project aims and Research Questions
A central aim for our project has been to contribute towards greater integration between theoretical and
substantive themes in housing studies, the study of poverty, and social policy more broadly.
Our research project had four research questions. These
were:
How, if at all, have European housing systems changed
over the last 10 years?
What is the association between housing deprivation and
poverty, and how does this vary between countries and
over time?
What is the relationship between housing costs and
poverty, and how does this vary between countries and
over time?
To what extent does independence for young people vary
between countries and over time, and how does this
relate to their incidence of poverty?
Data and methods
This comparative study examines data from the EU Statistics
on Income and Living Conditions (SILC) survey for the period
2004 to 2018. SILC is the primary survey for understanding
living standards across Europe on a comparative basis and
enables us to link information about housing with that on
income and living standards more broadly.
We have sought to make the most of the data contained in
this survey – so analysis the full set of waves, from 2004
until 2018, the most recent wave that includes data for the
United Kingdom. In 2012, the EU-SILC included an ad hoc
module on housing conditions, providing more detailed
information to that contained in the main survey. We have
drawn on this module selectively, where the data contained
in this module sheds further light on our primary concerns.
A key issue in any study is to define the unit of analysis.
Indeed this presents a greater challenge in our own study
because while in the study of poverty the unit of analysis is
typically the individual, in housing studies this is usually the
household. This means that any study that covers both
housing and poverty must make an ‘unconventional’ choice
in at least one respect. In this report and in the work
conducted as part of this project, we follow the convention in
housing studies by reporting findings for households. The
exception to this rule is in our analysis of independent living
amongst young people, where our focus is on individuals. We
limit our focus to households where the household head is
under the age of 60 to limit the influence of between-country
variations in the share of older age groups on our results.
The measurement of housing tenure in SILC
A central challenge when working with SILC is that the
housing tenure variable – an important variable for any study
of housing – is not well-measured in that survey, which
frustrates analyses within countries that examine differences
by tenure and, between countries, by measures that focus on
the balance between tenures (on this question, see also
Dewilde, 2015; Stephens, 2016).
Amongst tenants, SILC does not distinguish between private
renting and social renting, but rather between renting at
‘prevailing or market rate’ and renting ‘at a reduced rate
(lower price than market price’ (Eurostat, 2017). This
classification is akin to the distinction between ‘profit renting’
and ‘cost renting’ that was emphasised by Kemeny (1995:
35) as being more meaningful than whether a tenant was
renting from private or public landlords. In SILC, tenants who
are renting ‘at a reduced rate’ include those ‘(a) renting
social housing, (b) renting at a reduced rate from an
employer and (c) those in accommodation where the actual
rent is fixed by law’ (Eurostat, 2017: 172). It thus captures
situations where rental costs are supressed through a variety
of mechanisms, in both the public and private rental sectors.
Tenants who rent ‘at prevailing or market rate’ include those
where such supply-side mechanisms do not apply and
includes circumstances where rental costs are paid for (in
whole or part) through housing allowances. Thus, demand-
side policies to support housing affordability, which may also
have the effect of reducing incurred housing costs, do not
influence the categorisation of whether a household rents at
a market or a reduced rate.
Moreover, and somewhat counter-intuitively, in situations
where there is no market rental category – where all rents
are restricted in some way and which might be categorised in
Kemeny’s terms as a unitary rental system – households are
categorised as paying the ‘prevailing’ rent and thus
categorised in the ‘prevailing or market rent’ category. The
consequence is that countries such as the Netherlands and
Sweden, which have significant public or social housing stock
and/or regulation of the for-profit sector, ‘almost all rental
housing is categorised as being at “market rent”’ (Stephens,
2016: 26). It has also been suggested that the proportion of
housing identified as being non-market in SILC is lower in
some countries than is indicated by national sources
(Stephens et al., 2010). In response to these issues,
Dewilde (2021) re-classifies rental housing as reduced rate
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 13
when all rental housing is listed as being at ‘prevailing or
market rent’ and these nations are known to have extensive
non-market rental sectors. We adopt that approach here too
and this leads us to classify all rental accommodation in the
Netherlands, Sweden and Denmark (except that which is
rented free) as being reduced-rate housing.
The two categories of ownership, referring to outright and
mortgaged homeownership, are more straight-forward. Data
distinguishing between outright and mortgaged homeowners
exist only in the microdata file for all countries from 2011
(and for most countries from 2010). This shortens the
observation window for analyses that make use of this
distinction. The fifth, rather ambiguous category, captures
circumstances where accommodation is provided rent free
by a private source (such as a parent) or by an employer, but
where they do not hold a title deed. It does not play an
important role in our analysis.
In summary, the way that this key variable is collected in SILC
is not unproblematic and poses challenges to analysis.
These must be borne in mind in terms of the interpretation
of some of the data and is clearly an area warranting further
scrutiny in terms of data collection, as we discuss in the
concluding section.
Our treatment of housing costs
In many analyses, measures of housing cost burdens or after
housing cost (AHC) measures of poverty do not include
mortgage principal payments within the measurement of
housing costs on the grounds that mortgage principal
payments represent investments. In one sense, this is true.
However, it is undeniable that these costs do reduce
households’ remaining income for non-housing purposes
and that they may lead household members to experience
lifestyle deprivation as a result. Moreover, omitting mortgage
principal payments leads to an almost inevitable conclusion
that renters have higher cost burdens than owners since a
significant component of owners’ actual costs are
overlooked. For these reasons, we include mortgage
principal payments in our measure of housing costs to
enable better comparison between the ongoing living
conditions of mortgage-holders and renters.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 14
ANALYSIS
Part One: Europe’s Changing Housing Systems
In Part One of our analysis, we examine how housing systems across Europe have changed in the period
2004- 2018. Our analysis is presented in three sub-sections. The first looks at changes in tenure
patterns over time.
The second shifts from a focus on the whole-of-society to
consider the position of young people in particular, who have
been the focus of so much concern in relation to housing
(e.g. Eurofound, 2019). Here we examine the extent to which
young people establish independent households and how
this has changed during this period. In the third we examine
changes in housing costs. The aim is that by looking across
these three issues, we get a sense of key housing
differences across Europe and how these are changing over
time.
Before examining how Europe’s housing systems have
changed over time, we begin by noting how the tenure
structure of European housing systems differs across
Europe. Figure 5 presents tenure balances by welfare regime
for an average of the period 2016-2018. Data are presented
in two panels. In the left-hand panel, which captures all
households where the household head is under 60, we see
that in the Anglophone world, homeownership is more
prevalent than renting, with mortgaged homeownership the
most common tenure type. The balance between owning and
renting is more even in the Continental and Social
Democratic worlds. In Southern Europe, outright ownership
is the most common tenure type, but around a third of
households are owners with mortgages and around a
quarter a market-rate renters. The pattern is different still for
the Central and Eastern European nations, where outright
ownership is much more widespread and other tenure
categories, and especially the two rental categories,
comparatively marginal. In short, tenure patterns across
Europe are highly variable, and attempts to summarise
differences between housing systems in terms of tenure
balance between, say, owners and renters, are likely to be
frustrated by the sheer diversity of experience – at least
once Southern European and Central European nations are
included.
In the right-hand panel of the figure, we restrict attention on
households with a head under the age of 60 and who are
living in poverty (BHC). The tenure balances for poor
households differ significantly from those of the population
as a whole, with much greater emphasis on rented
accommodation, and in particular market-rate renting
(especially in the Continental and Anglophone worlds). The
exception is in Central and Eastern Europe, where outright
ownership remains dominant.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 15
Figure 5. Tenure balances by welfare regime, all households and poor households
Figure 6. Change in homeownership rate between 2005-07 and 2016-18, by country and household poverty status
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 16
Section 1. Trends in homeownership
In Figure 6 below, we show how trends in homeownership
have evolved between 2005-2007 and 2016-2018.5 What is
most obviously apparent is that homeownership has fallen
almost everywhere (red bars) – only in France, Malta,
Czechia, Slovakia and Poland do we see increases, and only
in Poland is the increase in homeownership substantial. In
the other 23 countries we see declines in homeownership
and in some countries – the UK, Ireland, Denmark, Cyprus,
Spain, Estonia and Slovenia we see quite substantial
declines (of around 10 percentage points). Homeownership
rates also fall very consistently for poor households (the only
exceptions being Luxembourg and Poland), and the fall by a
greater amount in absolute terms than for all households in
much of the Continental, Southern European and Central
and Eastern European worlds. Even in a relatively short
period of time, home ownership is declining in a non-trivial
way in many countries.
To examine these changes in more depth, we restrict our
focus to households where the household head is between
40 and 55: we treat these households as having a ‘prime
age’ household head who, if they are to become
homeowners, will have done so by this stage in their
lifecourse. We restrict our attention on these households
here so that differences in the propensity of young people to
leave the family home and form independent households do
not drive our understanding of patterns of change.
Figure 7 presents comparisons between changes in
differences in the homeownership rate for the richest and
poorest income quintiles for prime-aged household heads.
These values are expressed as relativities so, for example,
values of around 100 per cent in the Anglophone world imply
an ownership rate that is approximately twice as great for the
richest quintile as it is for the poorest. The values are all
positive, consistent with higher homeownership rates for
households in the richest quintile than in the poorest, with
substantial differences in these relativities in the
Continental, Anglophone and Social Democratic worlds and
smaller differences in the Southern European and Central
and Eastern European nations.
Over the period 2005-2007 to 2016-2018, these relative
differences increase everywhere but the Anglophone
nations, with sharp increases in the Social Democratic and
Continental nations, and more modest increases in Southern
and Central and Eastern Europe.
5. Data for Bulgaria, Malta and Romania only commence from 2007, so figures relate to changes between 2007-2009 and 2016-2018 for these nations.
Figure 7. Change in difference in homeownership rate between richest and poorest quintiles between 2005-07 and
2016-18, by welfare regime (prime-aged head only)
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 17
To examine these changes in more depth, we restrict our
focus to households where the household head is between
40 and 55: we treat these households as having a ‘prime
age’ household head who, if they are to become
homeowners, will have done so by this stage in their
lifecourse. We restrict our attention on these households
here so that differences in the propensity of young people to
leave the family home and form independent households do
not drive our understanding of patterns of change.
Figure 7 presents comparisons between changes in
differences in the homeownership rate for the richest and
poorest income quintiles for prime-aged household heads.
These values are expressed as relativities so, for example,
values of around 100 per cent in the Anglophone world imply
an ownership rate that is approximately twice as great for the
richest quintile as it is for the poorest. The values are all
positive, consistent with higher homeownership rates for
households in the richest quintile than in the poorest, with
substantial differences in these relativities in the
Continental, Anglophone and Social Democratic worlds and
smaller differences in the Southern European and Central
and Eastern European nations.
Over the period 2005-2007 to 2016-2018, these relative
differences increase everywhere but the Anglophone
nations, with sharp increases in the Social Democratic and
Continental nations, and more modest increases in Southern
and Central and Eastern Europe.
Figure 8. Changes in homeownership rate amongst households in richest and poorest income quintiles, by welfare
regime (prime-aged head only)
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 18
We now move from these relativities to show the absolute
changes for the richest and poorest quintiles. These are
presented in Figure 8. This shows that homeownership has
fallen in the poorest quintile in each welfare regime bar
Central and Eastern Europe (and even this has reduced if we
exclude Poland). Changes for the richest quintiles are less
consistent, rising in Central and Eastern Europe (though by a
more modest amount if Poland is removed), the Continental
and Social Democratic worlds and falling in the Anglophone
and Southern European worlds. What we observe across
Figures 7 and 8 is that while homeownership has declined, it
has also become more concentrated on better-off
households.
Figure 9. Changes in homeownership rates between 2005-07 and 2016-18, by level of urbanisation and welfare regime
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 19
Our final examination in relation to ownership trends is in
how they vary between urban and rural areas. A key concern
within the extant literature has been that housing
affordability problems have been a particular issue in urban
centres. In their recent account of welfare state change
Iversen and Soskice (2019) suggest that contemporary
capitalism requires the co-location of skill clusters, which
may lead to greater concentrations of wealth in major urban
areas and accentuate urban-rural divides (see also Hegedüs
and Horváth, 2015), including in relation to the cost of
housing. The Statistics on Income and Living Condition (SILC)
survey allows us to distinguish between densely-populated,
thinly-populated and intermediate areas and we examine
changes in homeownership in these areas over time.
In Figure 9, we do not find evidence of a generalised
deterioration in the position of urban households across
welfare regimes. Homeownership appears to fall for
households with prime-aged heads in urban areas in
Southern Europe and the Social Democratic nations, though
differences are only substantial in the former.
Homeownership rates rise in urban areas in Continental
world and in Central and Eastern Europe, while they fall in
the Anglophone world, but only by about as much as in
intermediately-populated areas. In short, these data provide
little evidence of a generalised concentration of the fall in
homeownership in urban areas.
While much of the literature on housing affordability in major
cities focusses on globally-important metropolitan centres
such as London, Paris, Milan and so forth, the measurement
of densely-populated areas in SILC is very much broader.
Taking the UK as an example, what constitutes a densely-
populated area includes not only London but Cardiff,
Cheltenham, Stoke-on-Trent, Bournemouth, Southend-on-Sea
and Belfast. It is perhaps not, thus, surprising that this
concern that is raised in the theoretical literature is not
borne out in our data and our findings – and those of others
relying on this data – must be read in light of the limited
sensitivity of the data when it comes to identifying
respondents in major urban centres.
Overall, this section has shown us that, while ownership
levels and trends differ across Europe, falling ownership
rates for those on lower incomes is common in many parts of
the continent and this has contributed to a concentration of
homeownership amongst better-off households in much of
Europe.
Figure 10. Percentage of young people living independently by age, period and welfare regime
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 20
Section 2. Independence for young people
A particular concern in some of the extant literature
concerns the ability of young people to leave the family home
and live independently, and the extent to which this has
changed in the years following the financial crisis (e.g.
Blanchflower, 2019). In this section we restrict our attention
to young people between the ages of 20 and 35 and explore
whether there have been changes in the timing of their
forming independent households.
To do this, and to examine how this has changed for young
people at different ages, we divide young people into three
groups according to age (20 to 25 years of age, then 26 to
30 and finally 31 to 35) and then explore trends before and
after the financial crisis by constructing four pooled cross-
sectional EU-SILC data points (2005-07, 2008-10, 2012-14
and 2016-18). Using information on household type and the
respondents’ relationship grid, we define living independently
as occurring when a young person is living in a household
without their parents, either with a partner or on their own or
with any children they themselves have.6 The analysis in this
section only focusses on individuals rather than households.
Independent living amongst young people aged 20-35
There are clear differences in terms of the propensity for
young people to live independently across Europe, as we
know from previous literature (e.g. Arundel and Ronald,
2016). Young people in the Social Democratic world have
high rates of independent living – certainly by the age of
26-30, but with two-thirds to three-quarters living
independently even between 20 and 25. A more staggered
approach is evident in the Anglophone and Continental
worlds with lower and later independent living in the Central
and Eastern and in Southern Europe. The probability of living
independently increases with age, as we would expect.
If we look at the changes over time in Figure 10, we observe
that independent living for young people aged between 20
and 25 falls everywhere, with non-trivial reductions in the
Social Democratic and Anglophone nations. There are more
marginal reductions in independent living amongst young
people aged 26 and 30 in most worlds, and there is even a
fall amongst those aged 31 and 35 in Central and Eastern
Europe and, to a lesser extent, in Southern Europe.
Equivalent figures, which we do not show here, demonstrate
that young people have become more likely to be co-resident
with their parents over the period we consider here. Thus,
independent living for young people aged 20 to 25 falls in
each of the worlds in the period in question. Moreover, not
only does independence come later in Central and Eastern
Europe and in Southern Europe, but delays in independence
effect older age groups in these worlds.
6. For the purposes of this exercise, we do not count shared living – i.e. those living with other adults, identified neither as being a parent or a partner – as
living independently. Shared living can represent a variety of circumstances. It would include living with friends but would also capture young people who
live with another family member on whom may be dependent, for example, a sibling, grandparent or an aunt or cousins. It would not be possible to
identify these broader familial relationships in the data.
Figure 11. Percentage of independent young people who are homeowners, by age, period and welfare regime
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 21
We also plot homeownership rates for young people living
independently. Here we see a substantial drop in ownership
amongst young people – including amongst the older group
aged 31-35. This is observed for all regimes bar Central and
Eastern Europe. Differences between outright and
mortgaged owners continue to play a role here, with outright
owners accounting for about half of independent young
people in Central and Eastern Europe, around 15% in
Southern Europe, and small shares elsewhere (not shown
here).
Thus, we see quite different patterns of independent living
amongst younger people aged 20 to 35 between welfare
regimes. Independent living comes early in the Social
Democratic world, with almost two-thirds to three-quarters of
young people between 20 and 25 living independently. In the
Continental and Anglophone nations, 26 to 30 is the age by
which most young people begin to live independently while in
Southern Europe and in Central and Eastern Europe,
sizeable proportions of young people aged 31 to 35 still do
not live independently.
We have observed that the period since the financial crisis
has been associated with modest reductions (delays) in
independent living. Reductions in independent living for
young people aged 20 to 25 are observable everywhere, but
delays for older age groups are also observable in those
parts of Europe where independence comes later – namely,
Southern Europe and Central and Eastern Europe. And, there
have been reductions in homeownership amongst those
younger people who do live independently, especially in the
Anglophone nations and in Southern Europe.
Figure 12. Percentage change in median housing costs, 2010-12 to 2016-18
Figure 13. Percentage change in housing costs, 2005-07 to 2016-18 (without mortgage principal)
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 22
Section 3. Housing costs
In this third section we explore the evolution of housing costs.
In contrast to the figures above, which reflect market or sale
prices of housing, in this section we explore trends in average
housing costs as they are experienced by households.
In Figure 12 we present changes in real median housing costs
by country for the three periods in question. The measure
draws on variables HH070 and HH071 in SILC and reflects
both housing costs such as rent payments, utility bills, taxes
on the dwelling as well as, for owner-occupiers, mortgage
payments. These amounts are adjusted for inflation and so
compare changes in real housing costs over time.
In Figure 12, we observe a diverse range of housing cost
trends, including within worlds. With the exception of the
Social Democratic world – here, Finland and Sweden – where
costs unambiguously rise, the other worlds have at least some
countries where real housing costs fall and indeed this is the
case in much of Southern Europe.
Omitting mortgage principal payments allows us to measure
change over a longer period (and changes the base year in the
figures). These figures are presented in Figure 13 and in this
figure we see that there are now more widespread housing
cost increases, and the increases are larger in magnitude.
This is substantially because of the initial period – 2005-07 to
2008-10, where costs increase almost everywhere, and often
by significant amounts.
Nonetheless, the data presented here leave us with something
of a puzzle. The biggest increase in housing costs in the period
analysed here occurred between 2005-07 and 2008-10.
Trends in median housing costs between 2012-14 and
2016-18 are in most nations not increasing (Ireland, the UK
and to a lesser extent Estonia are the exceptions here). And
yet as we have seen above the period of consistent house
price growth has been since 2013.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 23
Summary
From the previous discussion we draw three key conclusions.
First, homeownership has declined in much of Europe over the
past decade and a half, including for people on low incomes.
This has particularly been the case for people aged 40-55,
amongst whom we see a concentration of homeownership on
better-off households outside of the Anglophone nations.
Second, when we focus our attention on young people, we
observe the quite different timings of housing trajectories
amongst young people in Europe. While about two-thirds to
three-quarters of 20-25 year-olds in the Social Democratic
world live independently, this is the case for only about a
quarter to one-third of those in the Continental and
Anglophone worlds, and smaller proportions in Central and
Eastern Europe and in Southern Europe.
The period in question has witnessed declines in independent
living for young people aged 20-25 everywhere, but this has
been more pronounced in Central and Eastern Europe and in
Southern Europe, where a decline in independent living is
observed for young people aged 26-30 and 31-35, too. The
proportion of young people who do live independently who are
homeowners has declined too outside of Central and Eastern
Europe. There are thus substantial differences in both the
timing as well as the nature of young people’s housing
trajectories across Europe.
Third and finally, much media discussion on the housing
‘crisis’ has focussed on rising housing prices. Our focus here is
not on one-off house prices faced by owner-occupiers but
rather on monthly housing costs faced by all households. After
adjusting for inflation, we observe diverse patterns of change
between 2010-12 and 2016-18 when mortgage principal is
included. Increases are more commonly observed when
mortgage principal is omitted and 2005-7 is taken as the
reference year, with price rises between 2005-07 and 2008-
10 particularly notable.
ANALYSIS
Part Two: Housing Affordability in Europe
Parts Two, Three and Four of our report relate to analysis that is the subject of separate work packages.
Our goal here is to highlight some of the key descriptive trends rather than provide a comprehensive
overview of these work packages, which will be the focus of subsequent outputs.
In Part Two we focus on the problem of housing affordability.
The EU’s official measure of housing cost overburden
identifies households as experiencing affordability problems
when they spend more than 40% of their disposable income
on housing costs (both costs and incomes are net of housing
allowances). We present rates of housing cost overburden
for 26 European countries – Luxembourg and Denmark are
omitted because of issues with their mortgage principal
data.
In Figure 14 we present country-level rates of EU housing
cost overburden when we include mortgage payments.
These rates vary significantly across Europe, ranging from
5.4% in Malta to 54% in Greece. Moreover, while there is
substantial variation in rates of housing cost overburden,
there is no obvious patterning in terms of nations from
different welfare regimes. The countries with the five lowest
rates of housing cost overburden are from the Southern,
Central and Eastern, and Anglophone regimes, while the five
highest rates are from the Southern, Central and Eastern,
Anglophone and Continental regimes. There is, thus, a lot of
within-regime variation in levels of housing cost overburden.
One reason for this might be the rather broad measure of
housing costs captured in the SILC survey.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 24
Figure 14. Housing cost overburden rates by country, 2018
Figure 15. Housing cost overburden rates for mortgaged owners and market renters by country
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 25
Turning to differences within countries, we analyse
affordability patterns by poverty status and housing tenure.
In Figure 15, we show how cost overburden relates to two
tenure types in each nation – for mortgaged homeowners
and market-rate renters. Note that, unlike many analyses,
which omit mortgage principal payments on the grounds that
they constitute an investment, we include these here since
they also do impact on households’ non-housing disposable
income. In most countries (20/24), housing cost overburden
rates are higher for market-rate renters than they are for
mortgaged homeowners – and in many cases these
differences are substantial.
Figure 16. Housing cost overburden rate by poverty status and country
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 26
In Figure 16 below, we see that housing cost overburden is
strongly related to poverty status in every nation. That is,
while ratio measures of housing affordability, such as the EU
housing cost overburden rate, can technically identify
households all the way up the income distribution as being
cost overburdened, those on lower incomes face very
substantially elevated risks of cost overburden on this
measure.
Household-level measures of housing affordability are based
on the relationship between housing costs and household
incomes, and both dwelling and economic characteristics
prove necessary to understand its incidence. Economic
status (i.e. poverty) and some dwelling characteristics (e.g.
smaller dwellings; living in apartments rather than houses;
more recent occupancy) are closely linked with housing cost
overburden.
In analysis that we have conducted as part of our wider work,
however, we find that the position of market-rate renters has
deteriorated vis-à-vis mortgaged homeowners in a large
majority of nations across Europe between 2010 and 2018,
even after controlling for a variety of compositional variables.
That is, tenure differences appear to have become more
important over time in terms of explaining the incidence of
housing cost overburden.
ANALYSIS
Part Three: Housing Deprivation
In Part Three we consider the issue of housing deprivation, focussing in particular on the EU’s official
measure of severe housing deprivation. The EU’s statistical agency Eurostat monitors the prevalence of
a range of housing quality problems – namely, the percentage of individuals living in households with a
leaking roof, with no bath and shower and no indoor toilet, or in a dwelling considered too dark. These
are labelled housing deprivations.
Eurostat also reports a measure of severe housing
deprivation, which reflects the circumstance where
households experience at least one of these deprivations in
housing conditions and also experiences overcrowding. The
data in Part Three are from 2016, the last year for which we
have data on all components for the UK.
In Figure 17 we explore the incidence of the EU measure of
severe housing deprivation by country. The estimates in this
figure show that severe housing deprivation is, in particular,
a problem for some nations in Central and Eastern Europe
and, to a much lesser extent, in Southern Europe.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 27
Figure 17. Severe housing deprivation by country, 2016
Figure 18. Severe housing deprivation by tenure and welfare regime
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 28
Turning then to how these is experienced within countries,
we see in Figure 18 that renters – both market and reduced-
rate – are more likely to experience severe housing
deprivation than owners – both outright and mortgaged – in
each welfare regime. These tenure patterns are also
observed with a high degree of consistency at the country
level (not shown here). Severe housing deprivation is
negligible for owners in the Anglophone, Continental and
Social Democratic worlds and is significantly lower than for
renters in Central and Eastern Europe and Southern Europe
– albeit at non-negligible rates. This speaks both to
patterning by tenure but the dominance of country effects.
Figure 19. Severe housing deprivation rates by poverty status and welfare regime
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 29
Turning to the influence of economic position, we examine
differences in severe housing deprivation by poverty status
for each of the regimes considered here. In each regime,
people at risk of poverty under the relative definition face
elevated risks of severe housing deprivation (and this is
observed for each country, too). But, as before, regime
effects dominate – especially in Central and Eastern Europe,
where rates of severe housing deprivation for the non-poor
exceed those for households living in poverty in the
Anglophone world and are of a similar order of magnitude to
that of non-poor households in the Social Democratic and
Continental worlds.
ANALYSIS
Part Four: Subjective Measures
In this final empirical section, we focus in particular on the subjective evaluation of housing
circumstances. One of the challenges of comparative housing analysis, as we have noted above,
is that the relevant outcome variables are multidimensional, concerning affordability, quality, aspirations,
and more.
The analysis of subjective measures can shed new light on
housing outcomes and SILC contains a number of relevant
subjective measures. These measures are relatively
neglected but can complement what is learned from the
analysis of objective measures (see also Patsios and Pomati,
2018).
Our analysis here focusses on two substantive questions:
first, does the relative desirability of renting and owning differ
across nations? And second, do mortgaged homeowners,
who have an investment component to their housing costs,
exhibit higher satisfaction with their housing than renters
when housing cost overburdened? To answer these
questions, we draw on two subjective measures. The first is a
global measure of overall housing satisfaction, which was
asked most recently in an ad hoc module in 2012. The
second is a subjective measure of housing affordability,
which asks respondents to what extent their housing costs
are a burden. This question is contained in every year of the
survey, though to ensure comparability between the
samples, the analysis presented in Part Four is from the
2012 wave of SILC.
In the first analyses, we examine whether levels of housing
satisfaction for market-rate renters and reduced-rate renters
respectively are lower than for homeowners. The differences
below relate to coefficients from two country-specific models
run for each country individually. In the first, we present
“raw” coefficients, capturing the effect of housing tenure only
and without other controls. In the second, we adjust these
coefficients by controlling for a series of dwelling
characteristics (type of dwelling, number of rooms, size of
dwelling), conditions (presence of leak and problems in
relation to light, and overcrowding), household
characteristics (household type) and financial considerations
(household income in quintiles and housing cost as a
proportion of income). The dependent variable is a 1-4 scale
measuring housing satisfaction, with higher values
representing greater satisfaction. If the ‘contingency’ of
tenure was so great that comparison across countries was
like ‘comparing apples and oranges’, then we may expect
that these ‘raw’ coefficients to be close to zero, at least in
countries where renting is sometimes claimed to be relatively
more desirable (such as in Germany). By contrast, if there
are differences in housing satisfaction on the ‘adjusted’
coefficients – that is, after taking a wide range of controls
into account – then that would be consistent with clear
tenure-preferences in those nations.
In relation to market-rate tenants (Figure 20), we see that in
all nations bar Croatia ‘raw’ housing satisfaction rates for
market-rate tenants are lower than for mortgaged
homeowners, though differences are not significant in
Bulgaria, Romania and Croatia (where we have few cases)
and Denmark (where we have a greater number of cases,
but differences in satisfaction are very small). Dwelling and
household characteristics explain part of this difference – in
most cases between one- to two-thirds of the effect size, but
the coefficients remain negative in almost all instances (the
exceptions being Bulgaria and Estonia, in addition to Croatia,
where it was already positive). Of those countries where raw
differences were statistically significant, effect sizes for
Estonia, Hungary and Luxembourg cease to be significant at
the 95% confidence level and differences are only marginally
significant at the 90% confidence level in Lithuania and
Malta. Elsewhere, levels of housing satisfaction for market-
rate renters remains significantly lower than for mortgage
homeowners, despite controlling for a wide range of
characteristics.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 30
Figure 20. Housing satisfaction of market-rate renters vis-à-vis mortgaged homeowners
Note: Controls in ‘adjusted’ model are type of dwelling, number of rooms, size of dwelling, presence of leak and problems in
relation to light, overcrowding, household composition, household income (in quintiles) and housing cost as a proportion of
income (in quintiles).
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 31
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 32
Figure 21 provides the equivalent output for reduced-rate
renters. Here, the raw coefficient for reduced-rate renters is
lower than for mortgaged homeowners in every nation.
Differences are more modest in Luxembourg, Bulgaria,
Denmark and Croatia and are more pronounced in Romania,
Poland, Malta and Greece. A similar proportion of these
differences are explained by the variables controlled for in
the adjusted model as were explained in the previous
analysis (one- to two-thirds in most cases). This is
substantial, but certainly not total. These coefficients remain
negative across Europe, with the exception of Luxembourg,
though differences are not statistically significant in Bulgaria,
Cyprus, Denmark, Croatia, Ireland, Lithuania, Luxembourg,
Slovakia and the UK.
Overall, these figures provide reasonably limited support for
the idea that the relative desirability of different housing
tenures is radically different across European nations. Levels
of housing satisfaction are lower for market and reduced-rate
tenant than mortgaged homeowners in almost every nation,
though of course the magnitude of effect sizes varies. Lower
satisfaction levels amongst renters continue to be observed,
with only a small number of exceptions, after controlling for a
variety of dwelling, household and economic characteristics,
though these adjusted coefficients are not always statistically
significant. Germany, sometimes suggested to be a nation
where renting is more desirable, is not amongst the nations
with non-significant differences in housing satisfaction
between renters and mortgaged homeowners, though
Denmark and Luxembourg stand out as nations where this
argument finds greater empirical support.
In the second test we examine whether homeowners with
mortgages, who have an investment component to their
housing costs, are more satisfied than renters with their
housing, in the circumstance where these households
experience housing cost overburden. Table 1 below presents
regression coefficients from models estimating housing
satisfaction (model 1) and the subjective burden of housing
costs (model 2). The housing satisfaction variable in model 1
is measured on a 1-4 scale and takes higher values
representing greater levels of satisfaction. The reference
category is outright owners. When cost overburdened,
mortgaged homeowners have higher levels of satisfaction,
while market- and reduced-rate renters have lower levels of
housing satisfaction than outright owners, controlling for a
wide range of characteristics. However, in model 2, where
the dependent variable is measured on a 1-3 scale with
higher values implying a lower subjective burden of housing
Figure 21. Housing satisfaction of reduced-rate renters vis-à-vis mortgaged homeowners
Note: Controls in ‘adjusted’ model are type of dwelling, number of rooms, size of dwelling, presence of leak and problems in
relation to light, overcrowding, household composition, household income (in quintiles) and housing cost as a proportion of
income (in quintiles).
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 33
costs, both mortgaged homeowners and market renters
have rather similar subjective burdens when in similar
circumstances. That is, faced with equivalent housing and
financial circumstances, the subjective burden of mortgaged
owners and market-rate renters is similar, but mortgaged
homeowners are more satisfied with their circumstances
overall. This is consistent with there being an investment
component to mortgage principal payments, which are both
felt as a burden but also valued as an investment.
Table 1. Differences in two subjective measures of housing by tenure, households experiencing cost overburden
Note: Controls are: tenure, country, dwelling type, number of rooms, size of dwelling, overcrowding status, lack of toilet and bath,
dwelling too dark, age category, household composition, poverty status. Households experiencing cost overburden and with
household heads under 60 only.
Discussion and Conclusions
After a period in which debates in housing studies and Social Policy in general, and the study of poverty
in particular, appeared to be occurring in ‘disjointed literatures’, we are perhaps beginning to see the
emergence of a turn towards housing in these fields as developments in national housing systems – and
especially in relation to housing affordability – begin to impinge on wider social policy concerns. The
project that led to this report presents one attempt to build new bridges between housing and wider
social policy issues, including the study of poverty.
In this concluding section, we remind ourselves of some of
the key empirical findings that have been presented before
considering recommendations for policy and research. First,
while studies of housing systems change sometimes adopt a
long-run ‘epochal’ lens, the short window of 2004 to 2018
that we examine here has also witnessed significant change.
Homeownership has fallen in most countries in Europe,
including – in some countries especially – for low-income
households. Moreover, trends in homeownership display
signs of stratification outside of the UK and Ireland – that is,
being increasingly concentrated amongst better-off
households. When we focus specifically on young people, we
see patterns of delayed independence from the family home
and longer and later living with parents.
Second, we have demonstrated stark differences in housing
quality problems across nations on the severe housing
deprivation measure, with these problems strongly
concentrated in some of the poorer nations of Central and
Eastern Europe and, to a lesser extent, Southern Europe. In
work which will be the subject of a separate paper, we show
that the two components of the severe housing deprivation
measure – deprivation of conditions and overcrowding –
pattern rather differently and that they relate to quite
different drivers. In particular, while overcrowding is strongly
associated at the country-level with levels of wealth, the
deprivation of housing conditions is more strongly associated
with levels of relative income poverty in each nation. This
limited relationship between these items leads us to
question whether they are best combined in a composite
measure.
Third, there are also substantial variations in the incidence
of housing affordability problems on the housing cost
overburden measure, which captures households who spend
more than 40 per cent of their disposable household income
on housing costs. Indeed, the incidence of this measure is
10: 1 between Greece (54%) and Malta (5.4%), the
countries with highest and lowest incidence, respectively.
The distribution of housing affordability problems across
countries is not easily explained by welfare regime categories
and there is a lot of intra-regime variation. In our wider work,
which will be the subject of a separate paper, we show that
aggregate housing affordability problems have not
deteriorated over the period 2007-09 and 2016-18. Set
against undeniable house price rises post-2013, this
presents a puzzle. The increase in affordability problems that
we do observe, from about 2010 to 2014 coincides with the
period that incomes were supressed following the Eurozone
crisis – that is, before the substantial increase in house
prices since about 2013.
What we do observe, as noted above, is a fall in
homeownership and delayed independent living for younger
people. At present, it would appear that house price rises are
being “offset” through reductions in ownership and delayed
independence and are not observed in society-wide
deteriorations in housing affordability. There is of course no
necessity for outcomes to fall in this way, and a more
generalised deterioration in housing affordability may yet
occur – but this is not what we observe up to 2018. This
does not suggest that rising house prices are any less
significant. In his early account of new social risks – the
labour market risks arising from post-industrial society –
Giuliano Bonoli (2005) presents these as not only
threatening poverty but also ‘welfare’ losses and
‘frustrations’ in managing lifecycle transitions. Emphasising
the ‘new’ risks arising from labour markets in post-industrial
societies, housing is once again omitted from consideration.
But rising house prices also present risks – of the ability to
secure adequate accommodation at a reasonable price, and
these too have the potential to spill-over in terms of both
economic difficulties such as poverty but also in terms of
wider frustrations of unmet shelter aspirations and needs.
Fourth, in relation to differences within countries, we find
that households in poverty face heightened risks of housing
deprivation and housing affordability problems in every
nation. That said, the pattern for these two measures of
housing outcomes differ. In relation to severe housing
deprivation, the elevated risk experienced by low-income
households is consistent, but not always substantial. This
can be explained by the very low incidence of severe housing
deprivation in some of the richer nations of Europe, and the
dominance of between-country differences. In contrast, as
Figure 16 (page 26) shows, differences in housing cost
overburden between poor and non-poor households are vast
in almost every country. One criticism of “ratio” measures of
housing affordability, such as the housing cost overburden
measure is that they identify housing affordability problems
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 34
across the full range of the income distribution. This is
technically true but we find that even this measure identifies
the risk of housing affordability problems as being very
substantially greater for households in poverty.
Fifth, we find that housing outcomes vary in terms of the
extent to which they are patterned across tenures in different
nations. Renters face higher rates of severe housing
deprivation than owners in each welfare regime (and
country). In relation to housing satisfaction, we find that
renters quite consistently have lower levels of satisfaction
with their dwellings, and this remains in most nations even
when we take into account a variety of dwelling, economic
and household characteristics. This matters because a
society in which homeownership is becoming both less
accessible and more stratified might be one in which such
“frustrations” grow. These patterns are less consistently
observed in relation to housing affordability: the housing cost
overburden rate is very often but not always higher for
market-rate renters than for mortgaged homeowners.
In terms of the implications of our research, we distinguish
between implications for research and for policy. One
implication concerns explanatory models of country-level
differences in housing outcomes. The “regime” and “system”
frameworks of Esping-Andersen and Jim Kemeny were based
on institutional variation amongst mostly Northern and
Western European nations in the 1980s and 1990s, a
smaller and more homogenous set of countries than are
typically included in contemporary comparative analysis.
There were differences in terms of the levels of wealth
between these nations, but these have been dwarfed by
differences in wealth and living standards between EU
Member States following enlargement post-2004.
This matters because the “regimes” or “systems”
approaches have typically emphasised political as opposed
to economic differences in determining welfarist outcomes in
comparative terms, but the economic differences between
European nations are more stark than they were in earlier
comparative studies. Policy has evolved since these early
studies, too, and recent years have seen new attempts to
identify new dynamics of change and to identify new and
distinct set of typologies capturing contemporary institutional
differences (e.g. Stephens, 2020; Hassel and Palier, 2021).
That important housing outcomes such as severe housing
deprivation and important economic outcomes such as
material deprivation are strongly patterned across nations by
their levels of wealth calls for a reconsideration of the
importance of economic explanations in comparative studies
of both the welfare state and housing systems in accounting
for welfarist outcomes.
Second, our findings lead us to conclude that some of the
claims made in relation to the ‘contingency’ of housing
tenure and the degree to which this impedes comparative
analysis risk being taken too far. To be sure, we do not doubt
that tenure categories represent only one variable of interest
for the study of housing. Nonetheless, the strength of the
findings that we present here, and their consistency across
nations, calls into question some of the stronger claims that
are made in relation to the supposed “redundancy” of
housing tenure in explaining housing outcomes.
Third, there is a need to improve data coverage and quality in
EU-SILC, the key resource for understanding incomes,
standards of living and housing in Europe. The most striking
omission in relation to data coverage is that the UK no longer
participates in the survey and, post-2018, data for the
United Kingdom no longer appears in the SILC database.
This is all the more difficult to understand since EU
membership is not required in order to participate: Norway,
Switzerland and Serbia all participate in the survey.7 The
withdrawal of the UK from the SILC survey impedes the
ability of social scientists in the UK to conduct policy-relevant
comparative work and will harm the social scientific
community for as long as the UK’s omission continues. The
UK government and Office for National Statistics should
commit to re-joining SILC as a matter of priority.
In relation to data quality, there are a number of areas where
housing-related variables in SILC might be improved, the
most obvious and egregious of which relates to the
measurement of housing tenure, where there are important
limitations, as we have noted above. But there are other
issues, too. Inconsistencies in the measurement of
indicators of housing deprivation meant that it was not
possible to track changes in severe housing deprivation
through time. Close inspection of these indicators showed
that there were sometimes significant year-on-year shifts in
the incidence of some components that seemed implausibly
large given the relative fixity of the housing stock and left us
with variation that appeared to be more error than true
variation. This meant that we were forced to rely solely on a
single, recent wave of SILC, where we had greater confidence
in the underlying data. Uncertainties about data on mortgage
principal repayments for Luxembourg and Denmark in recent
waves led us to omit these countries from our analysis of
housing affordability. In short, the quality of data collection in
housing-related variables in SILC can be improved and
considering possibilities for improving the quality of data
collection should be made a priority by Eurostat.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 35
7. We are grateful to Jonathan Bradshaw for highlighting this point.
There are, then, implications for policy.
The first policy implication relates to attempts to tackle
severe housing deprivation, specifically. In our wider work on
housing deprivation, we show that this indicator is made up
of two very different components, which are not highly
correlated and respond to different drivers. At the risk of
oversimplification, overcrowding rates are explained by
absolute differences in wealth, while the deprivation of
housing conditions is more responsive to national relative
income poverty rates. We note that overcrowding has been
reduced to negligible levels in some nations, while the
deprivation of housing conditions seldom falls below 10%,
suggesting it has a higher “lower bound”. This in turn
suggests that reductions in severe housing deprivation in
countries currently facing significant risks is likely to be
achieved by reducing the overcrowding rate. Since this
appears strongly related to absolute differences in wealth, it
appears that economic growth – assuming this is translated
into a growth in household incomes – would lead ultimately
to a significant reduction in housing quality problems, as
captured by the EU’s severe housing deprivation measure.
The second policy implication concerns the growing
significance of housing tenure for housing affordability. We
have found evidence that the relative position of market-rate
tenants has deteriorated relative to mortgaged homeowners
in the period between 2010 and 2018, even after we take
account of the mortgage principal payments of the former.
This deterioration is quite consistent across nations and is
remains after controlling for a series of compositional
variables. One reason this matters is that households renting
at market rate are disproportionately likely to be poor. This
speaks to the need to take housing tenure more seriously as
a stratification marker in European societies and of the need
to ensure housing policy does not significantly favour
owner-occupiers at the expense of renters. This points to a
number of potential policy options, the effectiveness of
which might depend on the particular country or even city
context. One option is to make market renting more
affordable through more generous housing allowances, or
through mechanisms to control costs by controlling rent
rises, or even rent caps – although the potential unintended
consequences of such measures is well known. Assistance
to enter homeownership is another option, although
demand-side instruments tend to result in higher house
prices. Moreover, homeownership is not necessarily the most
suitable tenure for all households. A revival of forms of social
or affordable renting, especially in countries where there are
general supply shortages, is another option. It is also
important not to lose sight of the importance of a well-
functioning housing market, and that supply shortages will
tend to result in higher prices or rents.
Rising house prices risk creating growing pressures on
individuals and families to secure the kind of
accommodation that they desire without taking up a
disproportionate share of household income and pushing
them into poverty. Making sense of these developments and
how policy might respond to them requires moving beyond
the disjointed literatures of the past and building new
bridges between housing studies and scholarship on the
welfare state and on poverty.
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 36
References
Alcántara, A. L., & Vogel, C. (2021). Rising housing costs and
income poverty among the elderly in Germany. Housing Studies,
1-19.
Angel, S. (2021). Housing regimes and residualization of the
subsidized rental sector in Europe 2005-2016. Housing Studies,
1-21.
Ansell, B and Cansunar, A. (2021), ‘The political consequences of
housing (un)affordability’, Journal of European Social Policy, 31, 5,
pp. 597-613.
Arundel, R., & Ronald, R. (2016). Parental co-residence, shared
living and emerging adulthood in Europe: Semi-dependent housing
across welfare regime and housing system contexts. Journal of
Youth Studies, 19(7), 885-905.
Barlow, J. and Duncan, S. (1988) The use and abuse of housing
tenure, Housing Studies, 3(4): 219-231, DOI:
10.1080/02673038808720632.
Blanchflower, D. G. (2019). Not Working. Princeton, Princeton
University Press.
Bonoli, G. (2005). The politics of the new social policies: providing
coverage against new social risks in mature welfare states. Policy &
politics, 33(3), 431-449.
Copeland, P., & Daly, M. (2012). Varieties of poverty reduction:
Inserting the poverty and social exclusion target into Europe 2020.
Journal of European Social Policy, 22(3), 273-287.
Dewilde, C., & Flynn, L. B. (2021). Post-crisis developments in young
adults’ housing wealth. Journal of European Social Policy,
31(5), 580-596.
Dewilde, C. (2021). How housing affects the association between
low income and living conditions-deprivation across Europe.
Socio-Economic Review.
Dewilde, C. (2015), ‘What have ECHP and EU-SILC to contribute to
the comparative study of housing, Critical Housing Analysis, 2, 2,
pp. 19-26.
Eurostat (2017), ‘Methodological guidelines and description of
EU-SILC target variables’. 2017 Operation.
Esping-Andersen, G. and Myles, J. (2011), ‘Economic inequality and
the welfare state’ in Nolan, B., Salverda, W. and Smeeding, T.M.
(eds), The Oxford Handbook of Economic Inequality, Oxford, Oxford
University Press.
Esping-Andersen, G. (1990). The three worlds of welfare capitalism,
Princeton, Princeton University Press.
Eurofound (2019), Household composition and well-being,
Publications Office of the European Union, Luxembourg.
Fahey, T. (2007). The Case for an EU-wide Measure of Poverty.
European Sociological Review, 23(1), 35-47.
Goedemé, T., Trindade, L.Z. and Vandenbroucke, F. (2019), ‘A
pan-European perspective on low-income dynamics in the European
Union’, in Cantillon, B, Goedemé, T. and Hills, J. (eds), Decent
Incomes for All: Improving policies in Europe, Oxford, OUP.
Hassel, A. and Palier, B. (eds) (2021), Growth and Welfare in
advanced capitalist economies: How have growth regimes
evolved?, Oxford, OUP.
Hegedüs, J., & Horváth, V. (2015). Housing in Europe. Housing
review, 11-26.
Hick, R. (2016). The coupling of disadvantages: material poverty
and multiple deprivation in Europe before and after the Great
Recession. European Journal of Social Security, 18(1), 2-29.
Kemp, P. A. (2015). Private renting after the global financial crisis.
Housing Studies, 30(4), 601-620.
Kemeny, J. (1995), From public housing to the social market:
Rental policy strategies in comparative perspective, Routledge.
Madden, D. and Marcuse, P. (2016), In Defense of Housing,
London, Verso Books.
Marx, I., & Nolan, B. (2021). How wealth matters for social policy.
Journal of European Social Policy, 31(5), 489-495.
Nolan, B., Whelan, C. T. (2011). The EU 2020 poverty target. UCD
Geary Institute, Discussion Paper Series, WP2011, 11.
Reisenbichler, A. (2021), ‘Housing finance between social welfare
and growth strategies’, in Hassel, A. and Palier, B. (eds), Growth
and Welfare in advanced capitalist economies: How have growth
regimes evolved?, Oxford, OUP.
Patsios, D. and Pomati, M. (2018), ‘The distribution and dynamics
of economic and social wellbeing in the UK: An analysis of the
recession using multidimensional indicators of living standards
(MILS)’, Bristol, University of Bristol.
Saunders, P. (2017), ‘Housing costs, poverty and inequality in
Australia’, Housing Studies, 32, 6, pp. 742-757.
Stephens, M., Fitzpatrick, S., Elsinga, M., van Steen, G,. and
Chzhen, Y. (2010), ‘Study on housing exclusion: welfare policies,
housing provision and labour markets’, Brussels, Directorate-
General for Employment, Social Affairs and Equal Opportunities.
Stephens, M. (2020). How housing systems are changing and why:
A critique of Kemeny’s theory of housing regimes. Housing, Theory
and Society, 37(5), 521-547.
Stephens, M. (2016). The use of Esping-Andersen and Kemeny’s
welfare and housing regimes in housing research. Critical Housing
Analysis, 3(1), 19.
Stephens, M., & Van Steen, G. (2011). ‘Housing poverty’ and
income poverty in England and the Netherlands. Housing Studies,
26(7-8), 1035-1057.
Van Kersbergen, K. and Vis, B. (2014), Comparative welfare state
politics: Development, opportunities and reform, Cambridge,
Cambridge University Press.
Wilkinson, R. and Pickett, K. (2010), The Spirit Level: Why equality
is better for everyone, London, Penguin.
Zhang, B. (2021), ‘Reconceptualising housing tenure beyond the
owner-renting dichotomy: insights from housing and
financialisation’, Housing Studies. FirstView. DOI:
10.1080/02673037.2021.1961693
Housing and poverty in Europe: Examining the interconnections in the face of rising house prices 37
... Households are given a score based on the number of items they go without because of a lack of resources and our analysis is based on the average of these scores in each country (see also Nolan & Whelan, 2011, for analysis of these two poverty measures). When compared with the at-risk-of-poverty rate, this measure is much more concentrated on the nations of Central and Eastern Europe, which have lower real living standards (Hick et al., 2022). ...
... However, based on extensive initial descriptive work we decided that this was not possible because (i) there is some item missingness in some countries in some years, but also (ii) there appeared to be implausibly large year-on-year change in some items, which seemed to us to more likely capture error rather than true variation. This relates to the less-thanideal measurement of housing variables in SILC, which we have argued elsewhere are in need of greater attention by Eurostat (Hick et al., 2022). ...
... But we must accept our ability to parse the explanations empirically is limited. This is an issue given that the housing systems where outright ownership is dominant tend to be located in Central and Eastern Europe in particular, where levels of economic development are lower (see also Hick et al., 2022). To examine this further, we re-run our final multi-level moscdel as per Table 7 above, with the proportion of outright owners and GDP per capita as macro-level explanatory variables. ...
Article
Full-text available
Concerns about the quality of housing feature prominently in academic and policy discussion on housing, yet there is little agreement about how housing deprivation should be measured or monitored. In empirical studies, measures of housing deprivation are typically examined for one of two purposes—either to compare incidences of housing quality problems for different groups, which typically leads to an examination of performance of different measures of housing deprivation, or as dependent variables to examine competing theories about what explains cross-national variation in such problems, which typically ignores these measurement considerations. Our paper seeks to analyse measurement and theory jointly, focussing in particular on the EU’s severe housing deprivation measure, and its subcomponents—overcrowding and housing conditions problems. In descriptive analysis, we show that the two components of the severe housing deprivation measure are weakly related and pattern differently across nations and that the aggregation rule of the main measure has a substantial influence on observed incidences of this problem. We subsequently construct multi-level regression-based models and demonstrate that the two components have quite different determinants. Our paper has implications for the measurement of severe housing deprivation in Europe, for theories that seek to account for differences in housing outcomes, and for policy that seeks to tackle housing deprivation problems.
... These questions are ripe for examination in comparative studies and are, we argue, of relevance for scholars of the welfare state and poverty as well as those of housing studies. This paper is part of an interdisciplinary project examining the links between housing and poverty (and welfare states more broadly), which seeks to move beyond the "disjointed literatures" of housing, welfare states and poverty (Hick, Pomati and Stephens, 2022a). This paper, and our wider work, falls within the intellectual project called for by Kemeny 1992Kemeny 13, 2005 in relation to the need to better-integrate housing studies within wider disciplines. ...
... In substantive terms, the growth in house prices in the period since the Global Financial Crisis has led to newfound interest in, and concern with, developments in relation to housing, including amongst scholars of the welfare state and poverty. The threat that rising house prices presents to living standards makes this an opportune moment for greater integration between these disjointed academic fields (see also Hick, Pomati and Stephens, 2022a). ...
... By contrast, Kemeny's housing systems have come under greater challenge, because differences between rental systems do not reflect the housing systems of a wider group of nations (e.g. the nations of Central and Eastern Europe) and because the focus on rental systems as being the core dynamic explaining contemporary housing systems in the countries in the original typology has been questioned (Stephens 2020a). This suggests that, of the available typologies, clustering countries by welfare regimes may prove more fruitful for empirical researchers working with large numbers of cases than attempting to classify nations by housing systems (see Hick, Pomati and Stephens, 2022a), though the practice of relying on welfare regime classifications to study differences in housing is not uncontroversial (see Stephens and Hick, forthcoming for a discussion). Within the field of housing studies, there has been a substantial growth in scholarship on the financialization of housing, which is argued to be a newer process shaping housing institutions. ...
Article
Full-text available
How has housing wealth inequality changed for young-adult households in the post-financial crisis period, and what is driving such change? We chart a path for subsequent studies by analysing the previously unexamined post-crisis housing wealth profile of young adults via different angles and using multiple inequality measures. Using household micro-data for 11 European countries (Household Finance and Consumption Survey, 2010–2017) and the United States (Survey of Consumer Finances, 2010–2016), we find that the accumulation of housing assets for 22–44 year olds is unevenly concentrated among high-income homeowners, over and above what would be expected given the well-known decline in homeownership. We describe and assess several potential drivers for these wealth profile changes, finding that the current explanations offered in the literature do not adequately account for the unequal wealth profile of young people. We conclude that a mix of dynamics, including housing market volatility, housing market configurations leading to uneven capital gains and losses, and the increased social selectivity of homeownership intersect to shape the ways that young adults navigate the housing market in post-crisis times.
Article
Full-text available
Residualization refers to the process whereby publicly subsidized rental housing moves towards a position in which it provides only a safety net for low-income households. In this paper we quantify residualization based on the income profile of households in the below market rates (BMR) rental sector. Various residualization indicators for 12 European countries from EU-SILC data are calculated. First, we explore the effect of the size of the BMR sector and the cost of alternative tenure types on residualization. Second, we investigate if there are similar trends of residualization from 2005 to 2016. We find that decreases in the share of the BMR sector are associated with significant increases in residualization. Increases in rent differences between the BMR and market rate rental sector are related to larger degrees of residualization. We further observe a rising level of residualization for most countries. However, countries with allocation systems that aim at broader income groups still display the lowest degree of residualization.
Article
Full-text available
In order to explain an emerging trend towards deteriorating living conditions amongst low-income households across several (West-)European countries, it makes sense to investigate domains of socio-economic regulation that impact on expenditures, rather than incomes. I focus specifically on the domain of housing. Multilevel analyses for 28 countries (EU-SILC) reveal that ‘redistributive’ housing policies such as rental market regulation and housing allowances weaken the cross-sectional ‘between-country’ positive association between a low income and living conditions-deprivation, whilst also benefiting living conditions across the population. Regarding changes over time, increased uptake of housing allowances throughout the Great Financial Crisis (2008/9) seems to have shielded in particular renters from deteriorating living conditions, and might have compensated for declining availability of social housing. Higher house prices and price volatility, indicating housing market financialization, are associated with increased living conditions-deprivation for renters and low-income owners, both cross-sectionally between countries and within countries over time.
Article
Full-text available
This article critiques Kemeny’s theory of housing regimes to explain housing systems change. Power balances mediated through institutional structures are underlying causes of housing regimes in Kemeny’s schema in which the design of cost-rental sectors defines whole housing systems. However, the distinctive “unitary” systems Kemeny identified in Germany and Sweden are breaking down as economic failure prompted reforms to wider welfare systems, whilst mature cost-rental sectors were unable to maintain supply without subsidies. These mis-specifications in the theory have been exacerbated by the rise in unorthodox monetary policy. As poverty rates have risen, so the boundaries of possibility have shrunk, rendering “housing for all” approaches problematic and heralding more acute policy trade-offs. Nonetheless, policy choice and institutional differences counterbalance forces of convergence. Understanding system change requires theories of the middle range to be extended upwards to capture high-level forces of convergence and downwards to capture institutional detail that explains the difference.
Article
Full-text available
This paper examines the impact of the Great Recession on material poverty and multiple deprivation in Europe. Applying as its conceptual framework Poverty as Capability Deprivation (Hick 2014), which is one specification of Amartya Sen's capability approach, and employing the Alkire-Foster adjusted headcount measure, the paper draws on data from the EU Survey of Income and Living Conditions to present a multidimensional poverty analysis of 24 EU Member States at four time points: 2005, 2008, 2011 and 2013. The analysis shows that the pre-crisis period was associated with substantial reductions in multidimensional poverty in Europe, with the largest reductions occurring in the poorest Member States. However, the Southern European countries largely failed to benefit from these pre-crisis poverty reductions and, when the crisis hit, experienced the largest increases in multidimensional poverty in Europe. These patterns reflect a changing geography of poverty within the European Union, increasingly concentrated away from the East, and towards the South.
Article
The enormous growth in house prices in Europe since the 1990s has led to increasing concerns about the affordability of housing for ordinary citizens. This article explores the relationship between housing affordability – house prices relative to incomes – and the demand for redistributive and housing policy, using data drawn from European and British social surveys and an analysis of British elections. It shows that, as unaffordability rises, citizens appear in aggregate to become less supportive of redistribution, interventionist housing policy and left-wing parties. However, this aggregate rise, driven by the predominance of homeowners in most European countries, masks a growing polarization in preferences between renters and owners in less affordable regions.
Article
This special issue looks at wealth in relation to social policy from a variety of perspectives. The articles all shed an innovative light on wealth in relation to a range of topics relevant for social policy researchers. This introduction provides an overview of the papers in this special issue and then highlights some of the gaps and shortcomings that remain. We conclude with some reflections on what this means for the future of social policy and research on it.
Article
Housing tenure has often been taken-for-granted as consisting of dichotomous tenure types of owning and renting. This article critiques the owning-renting dichotomy through the lens of housing finance. It critically engages with three housing research programs wherein the owning-renting dichotomy is deep-seated: (a) the bundle of rights thesis, (b) comparative housing and welfare research, and (c) the ideology of housing. For each of them, the article first provides a brief recapitulation of the literature. It then explicates how they are constrained by the owning-renting dichotomy and why abandoning this dichotomy is necessary to transcend their limitations. Based on this, the article reveals a dialectical relation between owning and renting - the binary-oppositional attributes of owning and renting can be resolved with the changing relation of the occupant to financial markets. The article further proposes to re-conceptualize housing tenure as a relation of the occupant to financial markets and discusses the implications for reframing housing studies.
Chapter
This chapter examines trends at the lower tail of the EU-wide distribution of disposable household incomes. Using EU-SILC data from 2008 to 2014, it compares income levels across countries, after accounting for average price differences, focusing on levels and trends of the EU-wide low-income proportion and the EU-wide low-income gap. The analysis reveals that living standards in the new EU Member States have improved in comparison with the EU-wide median, while living standards in the crisis-hit Southern European countries have deteriorated. This indicates an important change in the composition of the bottom of the pan-European income distribution. It is also observed that no country succeeded in reducing the EU-wide low-income proportion while also reducing the at-risk-of-poverty rate. The results emphasize the need of a dual perspective on solidarity, a national and a pan-European, and the importance of reflecting further on the need of mutual insurance and true solidarity across borders.
Article
Housing costs have long been recognised as a factor contributing to poverty, and poverty researchers have estimated poverty using income before and after deducting housing costs. This paper examines the treatment of housing in the literatures on poverty and housing and applies the before and after housing costs approach to examine the extent of both poverty and income inequality in Australia and how they have changed since the early 2000s, focusing on the role of housing costs in the periods before and after the global financial crisis (GFC). Account has been taken of changes in the income measure used by the Australian Bureau of Statistics in its household income surveys as these have been shown to have a marked impact on empirical estimates. The findings indicate that taking account of housing costs leads to greater increases in both poverty and inequality between 2003–2004 and 2007–2008 and to smaller reductions in both since the GFC in 2007–2008.