The ‘Unrevealed Casualties’ of the Irish Mortgage Crisis: Analysing the Broader Impacts of Mortgage Market Financialisation

Richard Waldron

School of Architecture, Planning and Environmental Policy, University College Dublin, Ireland

richard.waldron@ucd.ie

Forthcoming in Geoforum, 2016, Vol. 69, Pages 53–66; doi:10.1016/j.geoforum.2015.11.005

Abstract

Since 2008, extensive research has examined the impacts of mortgage market financialisation, particularly the socio-spatial patterns of mortgage defaults and foreclosures. However, these standard statistical indicators of mortgage difficulty only capture the ‘overt casualties’ of the crisis, overlooking the mass of households who meet their mortgage commitments, but do so at considerable cost to quality of life. The impacts of the crisis on these ‘unrevealed casualties’ has received insufficient attention within the literature. As such, this article develops a framework to assess mortgage stress levels using standard and atypical indicators of mortgage payment difficulty. This framework differentiates between the ‘overt’ and ‘unrevealed’ casualties and is applied through a case study of suburban Dublin mortgagors to examine the characteristics of these groups, determine the key factors driving their mortgage stress and assess how their attitudes towards homeownership are being reshaped. The results suggest the impact of the mortgage crisis is much larger than previously considered, affecting a more diverse range of suburban households, many of whom may be one financial or non-financial trigger event away from developing a more serious payments problem. Banks’ reckless lending practices are among the strongest predictors of mortgage stress, demonstrating how the failure to adequately regulate banking practices has had detrimental financial impacts for households at the suburban scale. Attitudes toward homeownership and its investment function are increasingly negative, demonstrating how the primary rationale for homeownership expansion (i.e. its potential for wealth creation) is being revaluated by mortgaged households in the wake of the crash.

Keywords

Financialisation; Financial Crisis; Mortgage Stress; Arrears; Homeownership; Suburban Households

Introduction

In the aftermath of the Global Financial Crisis of 2008, a significant body of literature has emphasised the role of financial deregulation, the expansion and empowerment of financial markets, particularly mortgage markets, and excessive risk taking by financial institutions in driving the economic bubble that fuelled the crisis and the ensuing global recession (Aalbers, 2012, Harvey, 2011, Wolfson and Epstein, 2013). As mortgage markets were expanded, their function became transformed from markets designed to facilitate borrowers seeking credit for home purchase to markets which facilitate strategies for global investment in which homes and homeowners could be exploited for financial sector gain. This ‘high risk revolution’ within mortgage lending saw the expansion of highly-leveraged mortgages among economically more vulnerable households and those with less developed credit histories who traditionally were housed in the private or social rented sectors (Immergluck, 2011, Ashton, 2008). As lenders cultivated an increased appetite for risk accumulation as part of profit-

maximising strategies, they utilised mortgage funding (e.g. securitisation) and lending (e.g. subprime mortgages, 100% LTV loans) innovations to expand the scale and extend the reach of mortgage debt encumbrance. However, following the financial crisis, the household-level effects of this massive expansion and extension of mortgage lending have been stark. Rapidly falling house prices trapped many homeowners in negative equity, while the wider economic recession, falling incomes and rising unemployment created the conditions for mortgage default and foreclosure crises across advanced economies (Andritzky, 2014).

In Ireland, this paper’s empirical focus, the impact of the financial crisis has been particularly severe. Fuelled by deregulatory practices within the areas of economic management (Kirby, 2010), banking (Nyberg, 2011, Regling and Watson, 2010), housing (Author Reference) and urban planning (Murphy et al., 2014), one of the largest, speculative property bubbles among advanced economies emerged in Ireland. Between 2000 and 2008, the Irish housing market saw huge increases in residential construction, rapidly inflating house prices and an enormous expansion in mortgage credit, as per capita mortgage debt increased fourfold from €8,620 to €33,810 (European Mortgage Federation., 2010, DOECLG, various). However, the reliance of the Irish economy and banking sector on the domestic property market was to be exposed, with disastrous economic and social consequences. The bursting of the finance-led property bubble resulted in the bailout of the Irish banking system (€64bn),

which in itself resulted in the subsequent bailout of the Irish State by the IMF, ECB and European Commission (€67bn). Wider economic recession saw GNP decline by 8.5%, unemployment rise to 15%

and household incomes decline by 17.5% between 2008 and 2012 (CSO, 2014). Given the extraordinary levels of outstanding mortgage indebtedness, a mortgage arrears crisis quickly developed and is considered Ireland’s primary social and economic challenge (Government of Ireland, 2013). By mid-2014, 126,000 mortgages (16.5% of total residential mortgages) were in arrears and crucially, half of these cases (61,000) are in long-term arrears of more than one year. The value of accrued arrears is approximately €2.4bn or almost €27,000 per arrears case. Having failed to address the issue, the Irish banks are increasingly seeking to repossess property through the Courts system and to date some 3,865 home repossessions have occurred (Central Bank of Ireland, various).

Despite a burgeoning critical literature on the socio-spatial impacts of mortgage market financialisation, a number of key gaps are evident. Firstly, Wainwright (2010, 782) identifies “… a lacuna of geographical knowledge focusing on local-scale and national-scale studies, at present, outside of the US…” which may hinder the development of policies and practices to mitigate mortgage payment difficulties. Secondly, research has focused on the “standard statistical indicators” of mortgage stress (i.e. defaults and foreclosures) to analyse the “overt casualties of the housing- market crisis,” while overlooking the mass of households who manage to meet their mortgage commitments, but do so at considerable cost to quality of life (Forrest, 2011, 9). These ‘unrevealed casualties’ are not afforded the same level of policy concern and are rarely captured by official data. Thirdly, O’Neill et al (2010) have conceptualised ‘mortgage stress’ as a continuum of mortgage difficulties ranging from ‘struggling to make repayments on time’ to ‘home repossession’, but there remains a lack of clarity as to how to precisely define and measure households along this continuum. Fourthly, it remains unclear how mortgagors’ perceptions of homeownership and its financial benefits have been reshaped and are changing (or not) in the aftermath of the financial crisis.

Hence, this article seeks to advance the literature regarding the local level impacts and households’ experiences of mortgage market financialisation and the recession through an analysis of the extent of mortgage stress at the suburban scale in Dublin. A series of standard and atypical indicators of mortgage payment difficulty are utilised to further develop O’Neill’s et al’s (2010) framework of mortgage stress in a manner than can differentiate between the ‘overt’ and ‘unrevealed’ casualties of the crisis. This framework is then applied using household level survey data drawn from Dublin suburban mortgagors (n=433) to examine the characteristics of these groups, determine the key drivers of their mortgage payment problems and examine how their attitudes toward homeownership and its investment function are being reshaped in the wake of the crisis. Before

outlining the approach utilised, the next section situates the research within the financialisation and mortgage indebtedness literature. Thereafter, the methodological approach is outlined before it is applied in the results sections. In conclusion, the implications of the research findings for the literature and the future of Irish homeownership are discussed.

Mortgage Market Financialisation and the Experience of Mortgage Stress

The volatility created by the rising levels of defaults and foreclosures in the subprime mortgage market in the United States in 2008 fed quickly into wider financial markets, causing a seizing of the the global financial system and disrupting the flow of interest-bearing financial capital to the productive economy (Rude, 2013). Faced with the collapse of the global financial architecture, Governments intervened with extraordinary responses to stabilise national financial systems, thereby transforming the financial crisis into a fiscal crisis and producing the longest post-war recession across advanced economies. Harvey (2011) contends the crisis is rooted in the liberalisation and empowerment of finance capital from the 1980s and the creation of speculative asset bubbles, particularly in the property market, by switching capital from the primary circuit of capital accumulation (i.e. investment in the productive economy) to the secondary circuit of investments in the built environment as a means of compensating for the lack of alternative investment opportunities. A growing body of literature has examined the role of financialisation, referring to the shift in gravity of economic activity from production to finance (Foster, 2007), within mortgage markets, whereby the mortgage market is no longer just a facilitator of credit for households but has become a facilitator of global investment where homeowners and houses are viewed as assets which are financially exploitable (Aalbers, 2008, Chima and Langley, 2012, Forrest and Hirayama, 2014). Research has examined how mortgage credit liquidity was massively expanded as a result of securitisation practices and the linking of local housing/ mortgage markets to global financial markets (Gotham, 2009, Wainwright, 2009) and how mortgage lending was increasingly extended down the income/ social class gradient through mortgage product innovations, like subprime mortgages and 100% LTV loans (Scanlon et al., 2008), the increased usage of risk calculation practices, like credit scoring and risk-based pricing (Langley, 2008) and greater risk acceptance by banks (Kelly, 2014). This combination of factors served to dramatically increase mortgage indebtedness levels, particularly among households in the lowest income groups (Girouard et al., 2006), and fuelled rapid house price appreciation, which in turn led to ever rising levels of mortgage indebtedness as the mortgage market created its own expansion (Aalbers, 2009).

In the wake of the crash, the critical literature on financialisation has emphasised the need for research regarding the geographies of asset creation and destruction, the geopolitical consequences and socio-spatial impacts of financialisation, particularly as the impacts have not been homogenous across space (Lee et al., 2009, Martin, 2011, Engelen et al., 2010). Importantly, with respect to this paper, the literature regarding the socio-spatial impacts at the local scale has focused on the geographies of the fallout from the subprime mortgage crisis in the United States, where mortgage companies targeted individuals and neighbourhoods that historically had been ‘redlined’ and excluded from accessing mortgage credit (Ashton, 2008). A number of studies have examined the connection between subprime lending and higher rates of mortgage defaults and foreclosures (Immergluck, 2009, Agarwal et al., 2011, Schloemer et al., 2006), particularly their socio-spatial concentration among households within low-income neighbourhoods, racial minorities and borrowers of high-cost, high-risk mortgage products (Crump et al., 2008, Wyly et al., 2009, Kaplan and Sommers, 2009, Newman, 2009, Allen, 2011, Niedt and Martin, 2012, Rugh, 2015). Additionally, research has examined the negative externality effects of concentrated clusters of foreclosures within communities, including impacts on surrounding property values (Immergluck and Smith, 2006a, Schuetz et al., 2008), increased rates of crime (Immergluck and Smith, 2006b, Jones and Pridemore, 2012), the impacts on children’s school mobility (Been et al., 2011) and the costs imposed upon local municipalities (Apgar

et al., 2005). Despite this breadth of research, however, there is a much more limited range of studies examining the local level impacts of mortgage market financialisation beyond the US context, which can make it difficult to accurately identify households affected by the crisis, and in which specific ways they are affected (Wainwright, 2010, Martin, 2011). This paper contributes to the financialisation literature through an examination of the impacts of the collapse of the Irish property and banking sector at the local-suburban scale to better understand the implications of the increasing inter-connection between local housing markets and global financial markets. The focus on the suburban spatial scale is necessary for two reasons. Firstly, the suburb was the key site for the absorption of over-accumulating capital through the mortgage market in recent decades and as such is deeply enmeshed within global financial networks (Langley, 2006b, French et al., 2011). Secondly, while the impacts of the Irish mortgage arrears crisis have been examined within the national (Norris and Winston, 2011) and rural contexts (Murphy and Scott, 2014a), the suburbs have to date remained under-examined. This is somewhat surprising considering the Irish suburbs have been more deeply affected by the impacts of the property crash, particularly as recent Courts data on home repossession applications indicates a repossession crisis is looming in the commuter belt of Dublin City (Holland, 2015).

Despite the considerable body of research that has emerged from the US context regarding the impacts of mortgage market financialisation, much of this work has examined the finite events of mortgage default and foreclosure, thereby focusing on what Forrest (2011) calls the ‘overt casualties’ of the crisis. There has been much less research on the effect of financialisation and the prevalence of high-risk mortgage lending on households who have managed to avoid mortgage arrears and the threat of home repossession, but who have nonetheless been deeply affected by the crash and are bearing severe disruptions to their material comfort and quality of life. As outlined by O’Neill et al (2010, 21) rarely “…do data sets capture households that regularly meet their mortgage commitments, yet find it more and more difficult to do so...This unrevealed stress is vastly under explored” (Emphasis Added). Such households may not necessarily be in immediate risk of developing mortgage arrears and indeed may be considered a desirable customer by their bank, given that they have not missed a monthly mortgage instalment, but that is not say that their mortgage payment situation is not precarious. Indeed, such ‘unrevealed casualties’ may be struggling with their mortgage costs while trying to maintain a basic standard of living and could be one routine financial or non-financial trigger event away from developing a more serious mortgage payment problem (Fields et al., 2010). Indeed, in one of the only articles to examine the vulnerability of such households, Botein (2013) utilised an in-depth qualitative focus in the New York neighbourhood of Bedford-Stuyvesant (Brooklyn), where she determined many subprime homeowners experience mortgage payment stress long before the threat of foreclosure emerged. She finds that borrowers with high-cost, adjustable rate loans often struggled to make their monthly mortgage instalments and only did so by foregoing basic household necessities, going deeper into debt (e.g. by using expensive refinancing and pay-day loans), by liquidating life savings and household assets and by working second jobs, often at a cost to their well-being.

Part of the reason for the limited body of research regarding the impacts of the crisis on these unrevealed casualties relates to problems of how mortgage payment stress is defined and measured (O'Neill et al., 2010). Firstly, mortgage stress is closely interwoven with issues of financial and income stress which refers to broader pressures on household budgets and the two are closely related in ways that can be difficult to pre-determine. Secondly, the definition of mortgage stress varies depending on its use (e.g. inform government policy or rate the risks of mortgage-backed securities), leading to varying definitions and methods of measurement. Thirdly, data limitations exist, with inconsistency in the use of categories and thresholds for the measurement of mortgage stress. For example, foreclosure rates are the most regularly utilised measure of mortgage difficulty but can be unreliable. As outlined by Jones and Pridemore (2012), the number of foreclosures at any one time may not accurately represent a housing crisis, particularly where the number of foreclosures in one year may be low due to a large number of foreclosures in the previous year or because banks are delaying

foreclosures due to deflated house prices and are only repossessing homes at a pace at which they can be sold. Fourthly, many data series on mortgage stress analyse the impact of mortgage stress on the performance of the financial sector, rather than being concerned with the experiences of borrowers. Addressing this later point, O’Neill et al (2010) conceptualise mortgage stress “…as a circumstance in a household that could be positioned within a continuum ranging from ‘facing difficulties to make monthly repayments’ to ‘facing repossession or selling the property.” In their analysis, financial institutions define mortgage stress beginning at the point that payments are missed (overt casualties), whereas borrowers begin to define themselves as being in mortgage stress when they are struggling to make mortgage payments on time (unrevealed casualties). However, O’Neill et al (2010) do not offer precise indicators or measurement thresholds for identifying households ‘struggling to make repayments on time.’ This article addresses this gap by developing a series of measures of mortgage stress using a variety of standard and atypical indicators of mortgage payment difficulty, which can differentiate between the ‘overt’ and ‘unrevealed’ casualties, further discussed in the methodology section.

Recent literature on the ‘financialisation of everyday life’ has examined how households and individuals have increasingly been embedded within global financial markets and networks in recent decades (French et al., 2011, Hall, 2012, Hall, 2015). In particular, Langley (2006a, 2006b) demonstrates how the application of new financial technologies and practices have served to “bring forth new, investor subjectivities and financially self-disciplined subjects” and rework values around everyday borrowing and saving. Through financialisation practices, such as securitisation, the value of homeownership was rearticulated from a home capable of providing shelter and security to a commodity capable of delivering substantial returns to investors in a rising market. As such, in combination with extensive Government support for the expansion of owner-occupation, attitudes toward homeownership were reshaped to emphasise the freedom, responsibility and profitability of the tenure (Ronald, 2008). Indeed, benefits regularly, if somewhat mistakenly, ascribed to homeownership include asset building and wealth creation, increased levels of social stability, life satisfaction, enhanced civic and political participation, better physical and mental health, reduced crime rates, expanded tax bases for local services provision, improved child education performance, more stable local property markets and reduced likelihood of urban decay (Shlay, 2006, Whitehead, 2012). However, the rupture caused by the financial crisis of 2008 has called into question the financialised model of homeownership provision and has raised new questions about how households’ attitudes and behaviours toward homeownership are being shaped by the fallout from the crisis (Rohe and Lindblad, 2013, Author Reference). This paper contributes to this literature through an examination of how the attitudes of households in varying degrees of mortgage stress toward the economic benefits of homeownership and its investment function are changing in the context of the largest speculative property crash in recent history and the ongoing recessionary/ austerity conditions facing Irish households.

Mortgage Stress: From Concepts to Measurement

As stated, the core research objectives are to (1) examine the extent of mortgage stress among suburban mortgagors within the context of a major housing market crash and economic recession, (2) determine the socio-economic, property and mortgage characteristics of households struggling to meet their mortgage commitments, (3) examine what the key drivers of mortgage stress are, and (4) determine how attitudes toward homeownership investment are being reshaped as a result of the crisis. As noted above, most studies of mortgage payment stress have overlooked the implications of the crash for those who have remained current on their mortgage repayments but who are nonetheless struggling to sustain their financial position. The few studies that have attempted to explore the experiences of such financially precarious households have generally done so through the use of qualitative interviews with small samples of self-selecting respondents (Botein, 2013, O'Neill et al., 2010). While these studies have provided a significant contribution, the qualitative approach has

a number of limitations on this topic. Firstly, mortgage stress is a highly sensitive and personal issue which may influence borrowers’ willingness to discuss it. Secondly, small sample sizes and the diversity of mortgagors in terms of demographic, socio-economic and mortgage characteristics make it difficult to generate results that are replicable and generalizable. Thirdly, issues of confidentiality, anonymity and privacy are also problematic when identifying and contacting suitable participants for interview. Fourthly, as the research aims to examine the extent and depth of mortgage stress for households and to identify patterns of over-representation among specific mortgagor sub-groups, a methodological approach capable of generating a sufficiently large sample size that was representative in terms of geography, demographics and the mortgage market was necessary. A quantitative survey approach that could gather data on property and socio-economic background, as well as details of the mortgage product and lending environment, would also allow for inferences to be made regarding the causative effects of mortgage stress.

Mortgage Stress Indicators

The next step in the research design was to develop a series of indicators and measurement thresholds for the identification of mortgage stress. As noted, there remains a significant gap regarding how to identify and measure households experiencing mortgage stress, other than those who have defaulted on their mortgages or who have entered the repossession process. This article develops a more grounded set of indicators and measurement thresholds of mortgage stress by drawing on existing research on over-indebtedness (Kempson et al., 2004, Dominy and Kempson, 2003, Kempson, 2002) and reworks these indicators to apply them specifically to mortgagors. Indeed, the concept and measurement of over-indebtedness is gaining increased traction in the academic (Betti et al., 2007, Vandone, 2009) and policy (Stamp, 2009, UK. Department for Business, 2010) literatures as means of understanding how households and individuals are impacted by the economic crisis and what the implications for well-being are.

The most authoritative source on over-indebtedness indicators is the European Commission’s (2008) report “Towards a Common European Definition of Over-Indebtedness,” where a consortium of researchers reviewed the 30 most commonly used definitions and indicators of over-indebtedness by public bodies, private firms and academics within European Union member countries. The report finds the unit of measurement of over-indebtedness is typically the household and that definitions combine a range of dimensions, including economic (i.e. over-burdened in terms of financial commitments), temporal (i.e. short-term versus long-term commitments), social (i.e. the issue of financial exclusion or exclusion from participation in social/ economic life) and psychological (i.e. severe stress brought forth by over-indebtedness). Given the complexity of over-indebtedness, the report notes the phenomenon cannot be captured by a single indicator and following review of the various measures used in national contexts, the authors propose a mix of five objective and subjective indicators to measure over-indebtedness within households:

Burden of monthly commitment payments considered to be a heavy burden

High commitment payments, which push the households below the poverty threshold

Payment capacity considered to be very difficult or difficult by the household; and

Illiquidity problems (i.e. an inability to meet an unexpected expense)

Structural arrears on at least one financial commitment

Regarding the usage of subjective indicators, studies from within the economic and psychological literatures on quality of life are increasingly utilising subjective measures and peoples’ self-reported assessments of their well-being to complement more objective indicators in the analysis of income poverty and basic deprivation (Stiglitz et al., 2009, Maitre et al., 2014, Murphy and Scott, 2014b). While it is recognised that subjective measures of tolerance for debt can vary across time and between countries (Russell et al., 2011), the strength of subjective measures is their ability to enable respondents to assess their quality of life in their own terms by asking them about their life

satisfaction, happiness and psychological well-being (Hicks et al., 2013). Furthermore, research has demonstrated that subjective measures of well-being correlate well with more objective measures, such as income, employment, educational level, health and major life events (Dolan et al., 2008).

Table 1 demonstrates how these indicators have been adapted for this research and which indicators apply to the ‘unrevealed’ and ‘overt’ casualties. Negative equity was included as a variable given its prevalence in Ireland and its constraining effects regarding household mobility and job matching (Henley, 1998, Duffy, 2014). A variable regarding the psychological burden of mortgage difficulties is included due its obvious implications for quality of life and well-being (Libman et al., 2012). Indeed, extensive studies have been carried out in the UK over the last two decades by Janet Ford and colleagues into the experiences of households struggling with mortgage payment burdens and arrears (Ford et al., 2001). Such work has demonstrated a relationship between the burden of onerous mortgage commitments and negative mental health related outcomes such as feelings of shame and failure, anxiety and depression, as well as physical ailments such as tiredness, lack of diet and exhaustion (Nettleton and Burrows, 2000, Nettleton and Burrows, 1998). Burdensome mortgage commitments can also impact one’s feeling of ontological security, with negative implications for self-perception, social status and identity, personal and family relationships, future aspirations and well-being (Ford et al., 2001, 110 - 113).

Table 1 – Mortgage Stress Indicators

Mortgage Stress Survey Question Measurement

Threshold

Unrevealed

Casualties

Negative Equity Has negative equity affected (or would it affect) your ability to sell

the property?

Yes

Payment Burden If you consider the current mortgage cost, would you say it is: A Heavy Burden

Mortgage

Affordability

How much of the household’s average After Tax Monthly Income1

is spent on mortgage payments?

> 1/3 net income &

net income <€50,292

Illiquidity If the household was met with an unexpected expense (circa

€985) could it meet the expense without borrowing?

No

Overt Casualties

Restructured Is the household making reduced mortgage payments due to a

restructuring of the mortgage with the lender?

Yes

Arrears Is the household currently behind on its mortgage payments? Yes

Legal Notice Has the household received notice of legal action from the

mortgage lender regarding missed payments?

Yes

Two further measures of mortgage stress seek to ascertain the payment capacity of the

household to make their monthly mortgage commitment. The first measure assesses the affordability of the household’s mortgage using an adjusted housing expenditure to income ratio, which despite criticisms, remain a commonly adopted measure of housing affordability by national governments (Nepal et al., 2010, Rea et al., 2008). Affordability problems typically arise where a household spends more than 30% of its income on housing costs; although subtle differences can apply in differing national contexts. For example, in Ireland, the ‘Planning and Development Act 2000’ considers households to have an affordability problem if their housing payments exceed one-third of their net

1 Income is defined as after tax income derived from wages/ salaries, pension payments, payments of social welfare entitlements, income

from investments and income the household generates from other activities, like the selling of goods and services.

income (Fahey and Nolan, 2005, 85), whereas in the United States the household’s gross income tends to be more commonly used. However, the ratio approach is criticised because the normative 30% threshold does not account for those who choose to spend more of their income on their housing costs, typically higher income households, and because it overlooks whether the household has adequate income for other necessities once the housing cost has been paid (Stone, 2006). Indeed, Stone et al (2011) advocate a residual income approach to measuring affordability, which examines whether households’ income after housing costs is adequate to maintain an acceptable standard of living. While the residual approach more accurately examines the relationship between housing and non-housing expenditure, it is disadvantaged by the onerous data on non-housing costs required from households and is dependent on judgements as to what counts as necessary household expenditure (Gabriel et al., 2005). In the context of the household self-completion survey used in this study (discussed below), the residual approach is infeasible. As such, this study adopts a revised ratio measure, where a household must meet two criteria to be considered to have an affordability problem. Firstly, the household must spend more than one-third of their net income on their mortgage and, secondly, to more effectively target lower-income mortgagors, the household must have a net disposable income below the mean for a mortgaged household living in an urban area. Disaggregated data from the Irish ‘Survey on Incomes and Living Conditions’ reveals that this second threshold figure is €50,292. This revised ratio approach is similar, but not identical, to the Australian 30/40 rule, where

a household is considered to have an affordability problem where it spends more than 30% of its income on housing costs and belongs to the bottom 40% of the income distribution (Nepal et al., 2010).

The second measure, following the recommendations of the European Commission (2008) report, seeks to determine the household’s ability to withstand adverse economic shocks from their remaining financial resources. Indeed, the literature on over-indebtedness commonly utilizes the concept of illiquidity, which means that a household is unable to remedy their over-indebtedness situation without recourse to additional borrowing (ibid). As such, the measure of illiquidity seeks to determine whether a household can raise money from their immediate financial resources (e.g. savings) to address their over-indebtedness situation. Numerous studies have considered how the combination of weak financial buffers and adverse financial shocks (e.g. the birth of children, the death of spouses and the resultant loss of income) are consistently related to over-indebtedness (Betti et al., 2007, Kempson, 2002, Vandone, 2009). Following the European Commission (2008, 53) report, a household is considered to have an illiquidity problem if they are unable to meet an unexpected expense that is equivalent in value to the national at-risk of-poverty threshold (i.e. 60% of median monthly income) from their own financial resources, and this threshold is calculated independent of the size and structure of the household. This approach to measuring illiquidity has been applied in a major Irish study of financial exclusion and over-indebtedness following the financial crisis where a threshold of €985 was applied (Russell et al., 2011). To enable comparison, this is the illiquidity

measure applied in the current mortgage stress framework.

Finally, the standard statistical indicators of mortgage stress are utilised to capture households who pertain to the ‘overt casualties’ of the crisis. Households who have restructured their mortgage are also considered among the ‘overt casualties,’ as they are unable to meet the original terms and conditions of their loan. Households that are currently in arrears on their mortgage commitments and households who have received legal notification from their lender regarding their missed payments (i.e. the first step in the home repossession process) are classified as being within the ‘overt group.’

Housing Costs Survey 2012 and Sampling Strategy

In Ireland, mortgage lending, arrears and possessions data is only available at the aggregate, banking sector level and disaggregated information regarding the characteristics of troubled mortgagors is limited. As such, a ‘Housing Cost Survey’ was developed to extract data from a defined sample of mortgaged households within one local authority area within the Greater Dublin Area

(GDA). The GDA accounts for 42% of all mortgaged households in the State, as well as 40% of the country’s population and employment activity (CSO, various). The local authority of Fingal in North Dublin was selected as the survey area as it experienced intensive housing construction and mortgage market activity. Between 1994 and 2006, annual house completions in Fingal increased from 1,510 to 5,863 (+288%), accounting for one-quarter of all new housing development in the GDA (DOECLG, various). This suburban expansion was fuelled by the growth of the mortgage market, as the number of Fingal mortgaged households increased by 28% in just a four year period between 2002 (n=34,703) and 2006 (n=44,412) (Figure 1). This was the largest numeric increase in mortgage-indebted households of any local authority area in Ireland during a time of rapidly increasing house prices and mortgage debt encumbrance.

Figure 1 – Change in Mortgage Households across the Greater Dublin Area, 2002 - 2006

Source: (CSO, various)

A random sample was drawn from the 102,176 Fingal households identified from the national postal service’s GeoDirectory address database. Assuming maximum variability in the population, a margin of error of 3.5% and a confidence level of 95%, a necessary sample size of 784 responses was calculated using Cochran’s (1977) sample size formula to ensure representativeness. As GeoDirectory does not contain tenure information, the survey questionnaire had to be constructed in a manner that accommodated responses from all tenure types, including mortgagors, but also outright owners and social and private renters. As well as answering demographic and employment/ income questions, mortgaged households were asked 60 questions related to mortgage borrowing and cost issues. The survey was administered to 5,000 households between April and May 2012 and returned 914 responses, of which 433 came from mortgaged households; providing a random sample of 1% of the Fingal mortgaged population (n=43,811) (CSO, various).

The analytical approach combines both descriptive and inferential statistics. The former is used to organise and describe the profiles of households experiencing mortgage stress, while logistic regression is utilised to estimate which factors (independent variables) are most important for the prediction of whether a household is likely to experience mortgage stress (dependent variable) (Field, 2009). Logistic regression analyses are undertaken when the dependent variable is dichotomous (i.e. scored 0 or 1) and the results are expressed as odds ratios; for example the difference in the odds of a low-income household experiencing mortgage stress relative to a high-income household. An odds ratio with a value above 1.0 indicates higher odds of group membership, while a value below 1.0 indicates lower odds.These odds ratios are assigned significance values and if (p) is less than 0.05, the odds ratio can be deemed to be statistically significant in 95% of cases. Pearson Chi-Square tests are utilised to examine statistically significant relationships between the mortgage stress variable and variables relating to respondents views regarding the value of homeownership in the aftermath of the financial crisis. Again, relationships are only considered significant at an alpha level of 0.05.

Mortgage Stress and the Impacts of the Financialisation of Home

The mortgage stress framework is applied to the surveyed population in Figure 2. Unexpected expenses can place pressure on household income, particularly were households have already liquidated savings or assets to make repayments, and can lead to mortgage payment difficulties (Betti et al., 2007). Utilising data from the 2008 round of the ‘Survey on Incomes and Living Conditions,’ Russell et al (2011, 95) found that 41% of Irish households could not meet an unexpected expense of €985 in 2008 without recourse to additional borrowing. In comparison, the survey results found that

50% of the total sample did not have the immediate financial resources to cope with such an expense. Forty-seven percent of respondents indicated they were experiencing negative equity, which reflects the national trend observed by Duffy (2014) at the end of 2012. For these households, negative equity has the effect of constraining labour market mobility, potentially keeps families in homes that may be no longer suitable to their needs and restricts the pool of properties available to the market (Kitchin, 2014). The Fingal survey found that 36% of mortgaged respondents experienced their mortgage as a heavy burden, which suggests these households have additional issues coping with increased financial pressures, the difficulty of managing household budgets and maintaining living standards (Libman et al., 2012). As outlined above, households were considered to have an affordability problem if they paid more than one-third of their net income on their mortgage costs and the household had a total net income less than €50,292. Twenty-one percent of respondents indicated they have a mortgage

affordability problem, which is a striking result and relates to the impacts of austerity measures, labour market contraction, the rising cost of living and the banks’ strategies of maintaining high mortgage interest rates (Weston, 2011). In terms of the mortgage stress indicators that capture the ‘overt casualties’ of the Irish crash, 13% of respondents had restructured their mortgage, while 10% were in arrears on their mortgage at the time of survey and broadly these figures are in line with the national trends (Central Bank of Ireland, various). Finally, 1% of respondents received legal notification regarding their mortgage arrears, and represent households on the cusp of repossession.

Figure 2 – Survey Responses by Mortgage Stress Indicator

To determine how well the seven mortgage stress indicators can be considered as a group, a Cronbach’s Alpha test of reliability was applied (Table 2). This test measures the reliability coefficient of a scale based on the average correlation among items (internal consistency) and the number of items. Starting with the set of seven indicators, the overall level of reliability is unsatisfactory (.553). The low initial result is due to the low number of observations carried out as part of the test (n=52). Once the ‘Legal Notice’ indicator is removed, the number of test observations increased to 409 and the Cronbach’s Alpha score increased to .647. The analysis revealed that the alpha increased if the item relating to negative equity was removed. The result demonstrates that negative equity does not fit well with the other indicators of mortgage stress, which are more closely linked to ability, or inability, to pay the mortgage. The final column presents another reliability coefficient for a set of five items, having excluded negative equity, and the Cronbach’s Alpha increases to .677, which can be considered an acceptable score for survey measures of this kind (Field, 2009, Lance et al., 2006). Respondents were then grouped based on the severity of mortgage stress. Respondents who had restructured their mortgage or were in arrears were classified as the ‘overt casualties,’ and represent 17% (n=72) of the sample. Those experiencing their mortgage as a heavy burden or who had an affordability or illiquidity problem were grouped as the ‘unrevealed casualties’ and account for 46% (n=201) of the sample. Only 37% (n=160) of respondents did not display some form of mortgage stress, demonstrating how the property crash and recession has affected a broader range of households than has been considered to date.

Table 2 - Cronbach’s Alpha Level of Reliability for Mortgage Stress

Mortgage Stress 7 Items (If deleted) 6 Items (If deleted) 5 Items (If deleted)

Unrevealed Casualties

Negative Equity .546 .677 -

Heavy Burden .538 .565 .616

Affordability .452 .580 .630

Illiquidity .547 .602 .635

Overt Casualties

Restructured .519 .601 .629

Arrears .435 .594 .633

Legal notice .539 -

Total Responses 52 409 417 Cronbach’s Alpha .553 .647 .677

0%

10%

20%

30%

40%

50%

60%

0

50

100

150

200

250

Illiquidity Negative Equity MortgageBurden

MortgageAffordability

Restructured Arrears Legal Notice

Mortgaged Households (n) Share of Total (%)

Mortgage Stress and the Characteristics of the ‘Unrevealed Casualties’

Table 3 displays the property, mortgage product, socio-economic and financial characteristics of mortgaged respondents based on whether that are found to display no mortgage stress or whether they are within the ‘unrevealed’ or ‘overt’ groups. The aim is to identify where over- or under-representation occurs among specific sub-groups of the population and determine how their profiles differ. Turning to the property related variables, respondents who belong to the ‘unrevealed casualties’ group are more likely to be First Time Buyers who bought terraced or apartment units near the peak of the Irish property bubble. Indeed, 49% of respondents who bought between 2004 and 2008 and 53% of those that bought apartments or terraced units are found within the ‘unrevealed ‘group. First Time Buyers, who are generally younger purchasers who are unlikely to have reached their full earning potential, would have had to borrow larger mortgages to purchase such units as starter homes, often with a view to trading up the property ladder. Indeed, the results demonstrate that 50% of FTBs are found within the unrevealed casualties group. Interestingly, the profile of respondents in the unrevealed group differ somewhat from the overt casualties, where respondents who are owners of detached properties (21%) and repeat purchasers (19%) are over-represented in the findings.

The household survey obtained data in relation to four variables relating to the respondents’ mortgage including loan-to-value ratio (LTV), loan term, the mortgage as a multiple of income and the mortgage product type and significant differences are observed across the three mortgage stress groupings. While the ‘No Stress’ group accounted for 37% of the sample, they were clearly over-represented among respondents with low degrees of mortgage leveraging. They accounted for 43% of respondents with low loan-to-value ratios (≤85%) and 45% of respondents who borrowed a mortgage equivalent to four times their income or less. They also were more likely to have borrowed mortgages with shorter amortisation periods (≤20 years) and were more likely to have borrowed tracker mortgages, where the monthly interest rate is index-linked to the European Central Bank’s base rate. As this base rate has been at unprecedented lows in recent years, those on tracker mortgages are paying significantly lower interests rates (approx. 1%) than those with standard variable and fixed rate mortgages (approx. 3.8% - 4.7%).

In comparison, the ‘unrevealed’ and ‘overt’ groups are more likely to have drawn down mortgages where the underwriting criteria of the lenders was applied more liberally. While the ‘overt’ group accounts for 17% of responses, they represent 28% of respondents who borrowed virtually the full purchase price of the home (LTV ≥96%) and account for 25% of those who borrowed a mortgage equivalent to five times their income or more. They are also over-represented among respondents with long amortisation schedules of 31 years or more (19%). Similarly, the ‘unrevealed group’ are over-represented among those with long mortgage terms (51%), higher mortgage debt to income ratios (56%) and also account for a larger share of borrowers with a standard variable or fixed rate mortgage products (51%). As property prices rose exponentially over the 2000s, and housing affordability became an increasing concern, many borrowers accessed the housing market by leveraging their incomes, while drawing the payments out over longer mortgage terms, so as to minimise the monthly repayments to as great an extent as possible. In this regard, the Irish banks, operating in a highly liberalised regulatory environment, eased their underwriting criteria to such an extent that they enabled the massive run up in unsustainable mortgage commitments, demonstrating how and why the Irish banks became insolvent after the housing crash (Nyberg, 2011). In addition, those Irish banks with high volumes of loss-making tracker mortgages have been effectively subsidising these losses by maintaining high interest rates for their variable-rate and fixed-rate customers (Taylor, 2015).

Table 3 – Socio-economic, property and mortgage characteristics of mortgage stressed households Independent Variables

No Stress Unrevealed Overt Total

n % n % n % n %

Year Purchase 2004-2008 78 35% 110 49% 36 16% 224 100% All Other Years 80 39% 89 44% 35 17% 204 100% Total 158 37% 199 46% 71 17% 428 100%

Accommodation Apartment/ Terraced 43 33% 69 53% 17 13% 129 100% Semi-Detached 78 36% 102 47% 37 17% 217 100% Detached 39 45% 30 34% 18 21% 87 100% Total 160 37% 201 46% 72 17% 433 100%

Purchaser First Time Buyer 80 35% 115 50% 33 14% 228 100% Repeat Buyer 80 39% 85 42% 39 19% 204 100% Total 160 37% 200 46% 72 17% 432 100%

Loan to Value ≥96% 17 24% 35 49% 20 28% 72 100% 86 - 95% 61 36% 81 48% 26 15% 168 100% ≤85% 80 43% 80 43% 26 14% 186 100% Total 158 37% 196 46% 72 17% 426 100%

Loan Term ≥31 years 30 30% 52 51% 19 19% 101 100% 21 - 30 years 74 38% 90 47% 29 15% 193 100% ≤20 years 55 42% 54 41% 23 17% 132 100% Total 159 37% 196 46% 71 17% 426 100%

Mortgage Multiple ≥5 times income 25 20% 70 56% 31 25% 126 100% of Income ≤4 times income 132 45% 125 42% 39 13% 296 100% Total 157 37% 195 46% 70 17% 422 100%

Mortgage Product Standard Variable & Fixed 79 30% 132 51% 50 19% 261 100% Tracker 77 47% 65 40% 21 13% 163 100% Total 156 37% 197 46% 71 17% 424 100%

Household Type Single 22 33% 35 53% 9 14% 66 100% Lone Parent 6 26% 9 39% 8 35% 23 100% Large Adult Household 15 32% 28 60% 4 9% 47 100% Couple with Children 84 38% 97 44% 41 18% 222 100% Couple, no children 31 45% 28 41% 10 14% 69 100% Total 158 37% 197 46% 72 17% 427 100%

Household Income ≤€30,000 9 14% 30 45% 27 41% 66 100% €30,001 - €60,000 61 30% 107 54% 32 16% 200 100% €60,00 1 - €100,000 68 56% 49 40% 5 4% 122 100% >€100,000 19 60% 7 22% 6 19% 32 100% Total 157 37% 193 46% 70 17% 420 100%

Social Class Semi-, Unskilled & Other 37 47% 36 43% 10 12% 83 100% Non-Manual & Skilled 39 26% 72 48% 38 26% 149 100% Professional & Managerial 78 43% 82 46% 20 11% 180 100% Total 154 37% 190 46% 68 17% 412 100%

Employment Status Non Full Time Employed 9 15% 23 38% 28 47% 60 100% Full Time Employed 148 41% 173 48% 40 11% 361 100% Total 157 37% 196 47% 68 16% 421 100%

Other Financial ≥3 commitments 29 21% 73 54% 33 24% 135 100% Commitments 1-2 commitments 107 43% 106 43% 33 13% 246 100% No Commitments 22 49% 19 42% 4 9% 45 100% Total 158 37% 198 47% 70 16% 426 100%

Missed Payments on Arrears 13 9% 70 51% 55 40% 138 100% Other Financial No Arrears 146 51% 126 44% 15 5% 287 100% Commitments Total 159 37% 196 46% 70 17% 425 100%

The results reveal distinct differences in the socio-economic profiles of households found across the three groups of mortgage stress. Respondents without mortgage stress are much more likely to be households comprised of a couple without children, those on the highest incomes in the professional and managerial social classes and headed by a person in full-time employment. Conversely, single person households (53%) and large adult households2 (60%) are markedly over-represented among the unrevealed casualties, as are households on low-to-middle incomes (54%) of between €30,001 and €60,000 and those in full-time employment (48%). This is interesting as it is not

just unemployment that is driving payment difficulties, but also the prevalence of insufficient and low paid employment. Incomes declined quite dramatically from 2008 as a result of downward pay pressures and the imposition of austerity measures (e.g. new property tax and pension levy). The general cost of living has also risen, with the consumer price index increasing by 6% between January 2010 and December 2013 (Central Statistics Office, 2015). Interestingly, the profile of respondents among the overt casualties is quite different from the unrevealed group, where lone parent families (35%) and couples with children (18%) are over-represented, demonstrating the financial strain that the cost of having children can place upon households. Additionally, it is the households on the lowest incomes below €30,000 (41%), in the non-manual and skilled social class (26%) and those headed by

persons not in full-time employment (47%) who are over-represented among the overt casualties.

Given the inter-relationship between wider difficulties with financial commitments and mortgage payment problems, respondents were asked to provide details of their additional financial commitments3. Mortgage stress is clearly more prevalent among respondents with a greater number of additional financial commitments and particularly among those that are also in arrears on such commitments. Fifty-four percent of respondents with 3 or more additional financial commitments were found within the unrevealed group and 24% were among the overt group, which are both significant over-representations. Of the 138 respondents who were behind on payments on their additional financial commitments, 40% were found within the overt group, while 51% were also in the unrevealed group. Hence, the ease by which personal credit could be obtained during the Celtic Tiger years, and the conspicuous consumption this generated, is interacting with large levels of mortgage debt encumbrance and placing a serious strain on household budgets and quality of life.

Determining the Odds of Being in Mortgage Stress

While the descriptive analysis identifies patterns of mortgage stress, it does not reveal the degree of explanatory power that can be attached to each dependent variable as a predictor of mortgage stress, nor does it account for the interaction of other variables as a driver of mortgage stress. A multivariate regression analysis was conducted with the 14 independent variables run simultaneously for both the unrevealed and overt categories so that the estimates of the odds ratios are adjusted to take account of the other associations which pertain within the data (Table 4). This process enables us determine which of the explanatory factors should have the most relative importance attached to them and examine how the explanatory factors differ for both groups.

Examining the strength of associations between the unrevealed casualties group and the independent variables, respondents who borrowed mortgages equivalent to 5 times their income or more are almost 4 times more likely to be found within the unrevealed casualties group than respondent who borrowed a mortgage equivalent to 4 times their income or less (p<0.001). Demonstrating the effect that higher interest rates play in influencing mortgage stress, respondents

2 A large adult household contains 2 unrelated persons aged over 18 years or 3 or more related/ unrelated persons aged over 18 years.

3 Financial commitments include second mortgages, personal loans, credit cards, overdrafts and hire purchase agreements.

Table 4 – Logistic Regression Model of the Odds of Being in Mortgage Stress

Independent Variables Unrevealed Casualties Overt Casualties B SE Odds B SE Odds

Year Purchase All other years (Ref) 2004 - 2008 .021 .363 1.022 -.237 .952 .789 Accommodation Detached (Ref) Apartment/ Terrace .467 .468 1.595 .235 1.276 1.265 Semi-Detached .381 .383 1.464 .125 .966 1.133 Purchaser Type Repeat Purchaser (Ref) First Time Buyer -.141 .329 .868 -.145 .891 .865 Loan to Value ≤85% (Ref) ≥96% .524 .471 1.689 1.544 1.198 4.685 86-95% -.250 .326 .779 -.513 .875 .599 Loan Term ≤20 years (Ref) ≥31 years -.233 .502 .792 -1.355 1.493 .258 21-30 years .424 .344 1.528 .638 .876 1.893 Multiple Income ≤4 times income (Ref) ≥5 times income 1.378 .388 3.966*** 1.896 1.033 6.656 Current Product Tracker (Ref) Standard Variable & Fixed .555 .294 1.742 .784 .835 2.191 Household Type Couple, no children (Ref) Single .068 .504 1.070 -.236 1.424 .790 Lone Parent .224 .878 1.251 1.355 1.993 3.876 Large Adult Household 1.213 .545 3.362* 1.941 1.523 6.969 Couple with Children .960 .410 2.611* 1.825 1.149 6.202 Income Band > €100,000 (Ref) <€30,000 2.241 .833 9.406** 1.986 1.655 7.287 €30,001 - €60,000 1.357 .633 3.883* .542 1.285 1.719 €60,001 - €100,000 .439 .619 1.550 -1.934 1.201 .145 Social Class Professional & Managerial (Ref) Semi & Un-skilled & Other .026 .359 1.027 1.394 1.003 4.029 Non-Manual & Skilled .346 .337 1.414 1.377 .969 3.963 Labour Status Full Time Employed (Ref) Not Full Time-Employed .284 .591 1.328 2.623 1.092 13.772* Financial Commitments No commitments (Ref) ≥3 commitments .981 .523 2.667 1.625 1.242 5.077 1-2 commitments .056 .459 1.058 -1.805 1.167 .165 Missed Payments on Commitments None (Ref) Yes 1.583 .389 4.869*** 3.831 .830 46.092*** Constant -3.338 -5.887 Number of Observations 306 199 Nagelkerke R² .367 .814 Model X² 98.284*** 167.304***

***p<0.001, **p<0.01, *p<0.05

who have standard variable or fixed rate products are 1.7 times more likely to belong to the unrevealed group when compared to a borrower with a tracker mortgage, although the alpha value is slightly above the required significance level (p=0.059). Large adult households were over 3 times more likely to be found among the unrevealed group than a couple without children (p<0.05), while couples with children were 2.6 times more likely (p<0.05) reflecting the additional financial expense that children exert on the household budget. The largest odds ratio delivered by the regression model for the unrevealed group was the relationship with those on the lowest incomes. Respondents with incomes less than €30,000 were over 9 times more likely (p<0.01) to experience mortgage affordability, illiquidity or burden difficulties than a household with net annual income of €100,000 or more. Households with 3 or more additional financial commitments were 2.6 times more likely to be found within the unrevealed group than a household without additional commitments, although the test of significance is slightly above the required alpha value (p=0.061). However, the presence of additional arrears on financial commitments is a highly significant predictor, with households experiencing additional arrears almost 5 times more likely to be found within the unrevealed group than a household not in arrears (p<0.001).

In terms of the overt casualties group, a more limited number of predictor variables are statistically significant when all the other variables are held constant but the size of the odds ratios for these predictors are also much larger. The key predictors of whether a household is found within the overt casualties group are the head of household’s employment situation and whether the household has additional debts on which they are in arrears. A household where the head is not in full-time or self-employment is almost 14 times more likely (p<0.05) to be in the overt group than a household headed by someone in full-time employment. If the household is in arrears on their additional financial commitments, they are over 46 times more likely (p<0.001) to be found among the ‘overt casualties’ than a respondent not in arrears on their additional commitments. This finding supports O’Neill et al’s (2010) contention that analyses of mortgage payment stress cannot be separated from the household’s overall financial position. While financial stress is a broader concept than mortgage stress, it must be recognised that the two are closely related and it is likely that individual households perceive them as the same problem.

Changing Attitudes toward Homeownership in the Wake of the Crisis of Financialisation

As outlined above, minimal research has been conducted among homeowners with regard to how their attitudes and perceptions of homeownership and the investment function of housing are changing (or not) in the wake of the financial crisis. Forrest and Hirayama (2014) have considered how in the post-crisis world tensions have (re)emerged with respect to accessing homeownership. An older generation of owners are willing to see the maintenance of property values in order to protect their savings and main asset, but a younger generation of prospective purchasers, if they are not excluded from ownership as a result of more conservative lending practices, are celebrating the decline of property values to ease deposit and cost barriers. However, the results of a series of Chi-Square tests presented here demonstrate that there are considerable tensions amongst current mortgaged households with respect to the value of homeownership and its investment potential (Table 5).

The relationship between mortgage stress groups and perceptions regarding the expense of buying a home relative to renting displays a highly significant relationship (p=0.000). Households that had experienced ‘No Mortgage Stress’ following the crash were overwhelmingly positive (72%) in their view of homeownership as superior to renting in cost terms, while only 7% considered renting to be cheaper. In comparison, support for ownership was markedly less positive among the unrevealed (56%) and among the overt (47%) groups, while approximately one-fifth of respondents in both groups identified renting as cheaper. Upon reflection, it is perhaps surprising that half of respondents who

had either restructured their mortgage or were in arrears still considered ownership in positive or neutral terms relative to renting. A possible explanation may be that those households in the deepest difficulties with mortgage payment, and for whom the possibility of repossession is strongest, are more acutely aware of the downsides of the private and social rental sectors. These have been well documented in the Irish case and include insecurity of tenure, a trend toward short-term leases, an unregulated approach to private rent setting, a very large social housing waiting list (c.89,000 households) and a historic trend of poor investment in social housing provision (Punch, 2005, Sirr, 2014). That said, it is notable that almost one-fifth of total respondents felt that renting was superior value to owning one’s home, which at least suggests a changing attitude toward tenure in Ireland following the financial crisis. Indeed, at the national level there has been a 65% increase in the number of households renting privately between 2006 and 2011 (CSO, various).

Table 5 – Relationships between Mortgage Stress and Perceptions of Housing and the Economy

No Stress

Unrevealed

Overt

Total

Notes

n % n % n % n %

Over time buying your home works out less expensive than paying rent: Strongly Agree/ Agree 114 72% 111 56% 33 47% 258 60% X²= 23.606

Neutral 34 21% 41 21% 22 31% 97 23% p=0.000

Strongly Disagree/ Disagree 11 7% 45 23% 16 22% 72 17%

Total 159 100% 197 100% 71 100% 427 100%

Owning your home can be a risky investment: Strongly Agree/ Agree 80 50% 126 63% 51 72% 257 60% X²= 17.212

Neutral 31 19% 39 20% 13 18% 83 19% p=0.001

Strongly Disagree/ Disagree 48 31% 33 17% 7 10% 88 21%

Total 159 100% 198 100% 71 100% 428 100%

The extent to which the economic crisis since 2008 has impacted your life: Very Positive/ Positive 3 2% 12 6% 3 4% 18 4% X²= 33.792

Neutral 49 31% 21 11% 5 7% 75 18% p=0.000

Very Negative/ Negative 107 67% 166 83% 63 89% 336 78%

Total 159 100% 199 100% 71 100% 429 100%

The responses to the second question regarding the riskiness of homeownership as an investment indicate that many mortgagors have been cautioned by their experience of ownership over the Irish property bubble (p=0.001). Across the total sample, 60% of respondents considered owning one’s home to be a risky investment and this rose to 72% and 63% among the overt and unrevealed groups respectively and declined to 50% among the no stress group. The result reveals that one of the central premises behind neoliberal narratives of the benefits of expanding homeownership, namely that ownership is a relatively safe and secure means of building wealth for the future and one’s family, is not accepted by the majority of respondents. Even among mortgagors that are relatively unscathed by the property/ financial market crash, more respondents indicated ownership was risky (50%) than those who felt it was not an investment risk (30%). Indeed, such a negative view of the safety of housing as an investment class could be the result of an increasing awareness of the lack of security around full time work, as well as the fact that associated costs of

ownership (e.g. insurance, maintenance) are high and rarely considered when purchasing, as well as the fact ownership increasingly requires larger financial resources and competency.

Finally, respondents were asked to assess the extent to which the economic crisis since 2008 had impacted their lives as a whole and again the result displays a highly significant statistical relationship with the categorisation of mortgage stress (P=0.000). Considering the catastrophic collapse of the Irish banking sector, the severity of the economic contraction, the prevalence of unemployment and insecure employment and the huge falls in household income and social service provision levels, it is hardly surprising that 78% of respondents across the full sample felt their lives had been negatively or very negatively impacted. Among the ‘no stress’ group the result falls to 67%, while among the unrevealed and overt groups its rises to 83% and 89% respectively. While the result demonstrates mortgagors have, across the board, experienced negative outcomes following the crash, the result is much stronger among those who were most exposed to the adoption of liberalisation practices within the mortgage market, bearing in mind that the unrevealed and overt groups were much more likely to have been advanced mortgages underpinned by imprudent underwriting criteria.

Conclusions

The research presented here analyses the manner in which a crisis originating in global financial markets, itself a result of financialisation processes within the mortgage market, has impacted households at the local spatial scale in a suburban context. It establishes a framework for assessing the extent of households’ mortgage payment stress in the aftermath of one of the largest speculative property market crashes in recent history. Building upon Forrest’s (2011) observation that research to date has focused on the ‘overt casualties’ of the crisis (i.e. those facing mortgage arrears and repossession), this study developed a framework for the identification and measurement of the ‘unrevealed casualties’ of the crash, or those households have managed to maintain their mortgage commitments but are struggling to do so. Drawing upon O’Neill et al’s (2010) conceptualisation of mortgage stress as a continuum of escalating mortgage payment pressures, this research utilised both objective and subjective measures of financial difficulty to develop a more robust framework for identifying and measuring mortgage stress. Crucially, this framework could categorise respondents based on the severity of their mortgage payment difficulties and as such contributes to the emerging literature examining the actual lived experiences of the impacts of mortgage market financialisation at the household scale. Better understanding of the local implications of the financial crisis should further illuminate how the local and global spheres are intertwined with regard to financial investment, speculation and disinvestment, demonstrate how increasing financial market volatility translates into geographic instability, and inform discussion about how best to re-regulate post-crisis financial systems and markets (Martin, 2011).

The research highlights that the full impacts of the financial/ property market crash, the ensuing recession and austerity measures, have affected a much larger and more diverse range of households than has been considered to date. A mass of households live in a financially precarious situation, where they may be just one financial or non-financial blow away from developing a more serious mortgage payments crisis, which can result in arrears, damaged credit ratings, bankruptcy and the loss of the family home. The results demonstrate that there are degrees of vulnerability as a result of the financial crisis and that the profiles of households affected differ. The overt casualties were more likely to be low-income households, with very highly leveraged mortgages, additional financial commitments and have a vulnerable employment profile. The ‘unrevealed casualties’ were also likely to have highly leveraged mortgages, but in other respects were quite distinct. They were more likely to be first time buyers, who bought during the peak years of the property bubble using more expensive standard or fixed rate mortgage products. They were more likely to be single person or large adult

households with low-to-middle incomes and were more likely to be headed by someone in full-time employment. The results also demonstrate that in the immediate aftermath of the crash, mortgagors’ perceptions of the superiority of ownership over rental, in terms of cost and its investment potential, had become more negative, particularly among those who demonstrated greater levels of mortgage payment stress. However, further research is needed as to whether the long-term cultural preference for homeownership remains strong, particularly considering the fact that property values in Ireland have begun to rebound. Furthermore, the attitudes explored here relate primarily to the economic considerations of ownership and further analyses of how perceptions of the social benefits of homeownership, for example greater social involvement and residential satisfaction, have been reshaped following the crash are needed.

The most interesting research finding is the relationship between mortgage stress and the underwriting criteria of the banks. There is little doubt that this relationship relates to the manner by which liberalisation and deregulatory practices were introduced within the Irish mortgage market, and wider financial sector more generally, over the 2000s and demonstrates that one of the key drivers of mortgage stress for households was the profligacy and excessive risk taking that was endemic among financial institutions during the Irish Celtic Tiger (Regling and Watson, 2010). When all property, mortgage and socio-economic factors are controlled for, the key drivers of mortgage related stress for the unrevealed casualties of the crash are the mortgage debt-to-income ratio, the number or additional financial commitments and the presence of arrears on those commitments. The research findings provide empirical support for the recent introduction of stricter controls with regard to mortgage lending by the Central Bank (2015), where mortgage lenders may only lend at loan-to-value ratios of 80% (or 90% for first time buyers) and mortgage-to-income ratios of 3.5.

It is acknowledged that the measurement of mortgage stress is experimental in nature, given that this is the first time a detailed set of indicators have been proposed to capture households who are struggling to meet their mortgage commitments, as well as those who are in arrears or facing home repossession. As such, a number of assumptions have been made with regard to the application of some indicators, particularly the affordability and illiquidity indicators. However, the key strength of the mortgage stress framework is its flexibility and these measures could easily be adapted where national contexts differ. For example, the framework has used an adjusted ratio measure of affordability, but there is no reason that alternative measures of affordability cannot be applied as part of the framework. Indeed, Stone et al (2011) have outlined in detail a residual income approach to measuring housing affordability (i.e. where households are considered to have an affordability problem where they cannot meet their non-housing needs at some minimum level of adequacy after paying for housing). This could be a more precise means of measuring the affordability element of the mortgage stress framework, although it requires a socially-defined and accepted standard of adequacy for non-housing expenditure in differing national contexts. Considering the inter-relationship between mortgage payment difficulties and wider financial difficulties, a broader range of financial indicators could also be considered for addition to the framework. One manner by which the framework could be improved would be to weight the specific indicators based on their perceived severity. For example, if households were asked to rank measures of mortgage stress on a 1 to 10 point scale, then the individual contribution of each indicator to overall mortgage stress could be determined. With such information, the mortgage stress framework could be transformed into a mortgage stress index, with each respondent assigned a mortgage stress score which would enable a greater range of analytical procedures be applied.

Finally, after numerous failed approaches by Government and the financial sector to tackle the Irish mortgage crisis (Author Reference) it appears that recovery in the value of property, particularly in Dublin, is driving the Irish banks to enforce legal repossession of properties in mortgage arrears. While the number of repossessions through the courts have been low to date, as of the beginning of 2015 the Irish banks have lodged 7,101 civil bills for repossession across the State’s 26 circuit courts and the majority of these relate to cases in Dublin City and its commuter belt (Holland,

2015). In addition, the Government have revised Ireland’s bankruptcy code, part of which involves home repossession, and have introduced changes to maximise efficiencies in the Courts system of repossession (Expert Group on Repossessions, 2013). On a practical level, the mortgage stress framework identifies where the greatest vulnerabilities lie in the Irish mortgage market at present, in particular pointing out which categories of mortgagor are likely to experience the threat of home repossession to the greatest extent and where the greatest resources will be required to counteract the impact of forced eviction.

References

AALBERS, M. B. 2008. The Financialization of Home and the Mortgage Market Crisis. Competition & Change, 12, pp 148 - 166.

AALBERS, M. B. 2009. The sociology and geography of mortgage markets: Reflections on the financial crisis. International Journal of Urban and Regional Research, 33, 281-290.

AALBERS, M. B. (ed.) 2012. Subprime Cities: The Political Economy of Mortgage Markets West Sussex: Wiley-Blackwell.

AGARWAL, S., AMBROSE, B. W., CHOMSISENGPHET, S. & SANDERS, A. B. 2011. Thy Neighbor's Mortgage: Does Living in a Subprime Neighborhood Impact Your Probability of Default? Real Estate Economics, pp 1 - 22.

ALLEN, R. 2011. Who Experiences Foreclosures? The Characteristics of Households Experiencing a Foreclosure in Minneapolis, Minnesota. Housing Studies, 26, 845-866.

ANDRITZKY, J. R. 2014. Resolving Residential Mortgage Distress: Time to Modify? Wahington DC: International Monetary Fund.

APGAR, W. C., DUDA, M. & GOREY, R. N. 2005. The Municipal Costs of Foreclosure: A Chicago Case Study. Housing Finance Policy Research Paper Number 2005-1. Minnesota: Homeownership Preservation Foundation.

ASHTON, P. 2008. Advantage or Disadvantage? The Changing Institutional Landscape of Underserved Mortgage Markets. Urban Affairs Review, 43, 352-402.

BEEN, V., ELLEN, I. G., SCHWARTZ, A. E., STIEFEL, L. & WEINSTEIN, M. 2011. Does losing your home mean losing your school?: Effects of foreclosures on the school mobility of children. Regional Science and Urban Economics, 41, 407-414.

BETTI, G., DOURMASHKIN, N., ROSSI, M. & YIN, Y. P. 2007. Consumer over-indebtedness in the EU: measurement and characteristics. Journal of Economic Studies, 34, pp 135 - 156.

BOTEIN, H. 2013. From Redlining to Subprime Lending: How Neighborhood Narratives Mask Financial Distress in Bedford-Stuyvesant, Brooklyn. Housing Policy Debate, 23, 714-737.

CENTRAL BANK OF IRELAND 2015. Information Note: Restrictions on residential mortgage lending Dublin: Central Bank of Ireland.

CENTRAL BANK OF IRELAND various. Residential Mortgage Arrears and Repossession Statistics. Dublin: Central Bank of Ireland

CENTRAL STATISTICS OFFICE. 2015. Survey on Income and Living Conditions (SILC) - 2013 Results [Online]. Dublin: Central Statistics Office. Available: http://www.cso.ie/en/releasesandpublications/er/silc/surveyonincomeandlivingconditions2013/#.VSUItfnF9CA.

CHIMA, O. R. & LANGLEY, P. 2012. Putting Humpty Dumpty Back Together Again: Financialisation and the Management of the Subprime Mortgage Crisis. Global Society, 26, 409-427.

COCHRAN, W. G. 1977. Sampling Techniques, New York, John Wiley & Sons. CRUMP, J., NEWMAN, K., BELSKY, E. S., ASHTON, P., KAPLAN, D. H., HAMMEL, D. J. & WYLY, E. 2008.

Cities destroyed (again) for cash: Forum on the U.S. foreclosure crisis. Urban Geography, 29, 745-784.

CSO 2014. Survey on Income and Living Conditions (SILC) 2012. Dublin: Central Statistics Office.

CSO. various. Census of Population - Small Area Population Statistics [Online]. Dublin: Central Statistics Office.

DOECLG various. Annual Housing Statistics Bulletin. Dublin: Department of Environment, Community & Local Government,.

DOLAN, P., PEASGOOD, T. & WHITE, M. 2008. Do we really know what makes us happy? A review of the economic literature on the factors associated with subjective well-being. Journal of Economic Psychology, 29, 94-122.

DOMINY, N. & KEMPSON, E. 2003. Can't Pay or Won't Pay: A review of creditor and debtor approaches to the non payment of bills. Bristol: Personal Finance Research Centre, Univerisyt of Bristol & Lord Chancellor's Department

DUFFY, D. 2014. Updated Estimates on the Extent of Negative Equity in the Irish Housing Market. Dublin: Economic and Social Research Institute.

ENGELEN, E., KONINGS, M. & FERNANDEZ, R. 2010. Geographies of Financialization in Disarray: The Dutch Case in Comparative Perspective. Economic Geography, 86, 53-73.

EUROPEAN COMMISSION 2008. Towards a Common Operational European Definition of Over-Indebtedness. Brussels: European Commission.

EUROPEAN MORTGAGE FEDERATION. 2010. Hypostat 2009: A Review of Europe's Housing and Mortgage Markets. Brussels: European Mortgage Federation.

EXPERT GROUP ON REPOSSESSIONS 2013. Report of the Expert Group on Repossessions. Dublin: Department of Justice and Equality.

FAHEY, T. & NOLAN, B. 2005. Housing Expenditures, Housing Poverty and Housing Wealth. In: NORRIS, M. W. & REDMOND, D. (eds.) Housing Contemporary Ireland - Policy, Society and Shelter. Dublin: Institute of Public Administration.

FIELD, A. 2009. Discovering Statistics using SPSS - Third Edition, London, SAGE Publications Ltd. FIELDS, D., LIBMAN, K. & SAEGERT, S. 2010. Turning everywhere, getting nowhere: experiences of

seeking help for mortgage delinquency and their implications for foreclosure prevention. Housing Policy Debate, 20, 647 - 686.

FORD, J., BURROWS, R. & NETTLETON, S. 2001. Home Ownership in a Risk Society, Bristol, The Policy Press.

FORREST, R. 2011. Households, homeownership and neoliberalism. In: FORREST, R. & YIP, N.-M. (eds.) Housing Markets and the Global Financial Crisis. Cheltenham, Uk: Edward Elgar Publishing Ltd.

FORREST, R. & HIRAYAMA, Y. 2014. The financialisation of the social project: Embedded liberalism, neoliberalism and home ownership. Urban Studies.

FOSTER, J. B. 2007. The financialization of capitalism. Monthly Review, 58, 1-12. FRENCH, S., LEYSHON, A. & WAINWRIGHT, T. 2011. Financializing space, spacing financialization.

Progress in Human Geography, 35, 798-819. GABRIEL, M., JACOBS, K., ARTHURSON, K., BURKE, T. & YATES, J. 2005. Conceptualising and Measuring

the Housing Affordability Problem. Sydney: Australian Housing and Urban Research Institute (AHURI).

GIROUARD, N., KENNEDY, M. & ANDRE, C. 2006. Has the Rise in Debt Made Households More Vulnerable? OECD Economics Department Working Paper No. 535. Paris: OECD.

GOTHAM, K. F. 2009. Creating Liquidity out of Spatial Fixity: The Secondary Circuit of Capital and the Subprime Mortgage Crisis. International Journal of Urban and Regional Research, 33, 355-371.

GOVERNMENT OF IRELAND. 2013. Resolving the Mortgage Arrears Crisis [Online]. Available: http://www.merrionstreet.ie/en/News-Room/Releases/government-statement-resolving-the-mortgage-arrears-crisis.html [Accessed 16/01/2015.

HALL, S. 2012. Geographies of money and finance II: Financialization and financial subjects. Progress in Human Geography, 36, 403-411.

HALL, S. M. 2015. Everyday family experiences of the financial crisis: getting by in the recent economic recession. Journal of Economic Geography.

HARVEY, D. 2011. Roepke Lecture in Economic Geography—Crises, Geographic Disruptions and the Uneven Development of Political Responses. Economic Geography, 87, 1-22.

HENLEY, A. 1998. Residential Mobility, Housing Equity and the Labour Market The Economic Journal 108, 414 - 427.

HICKS, S., TINKLER, L. & ALLIN, P. 2013. Measuring Subjective Well-Being and its Potential Role in Policy: Perspectives from the UK Office for National Statistics. Social Indicators Research, 114, 73-86.

HOLLAND, K. 2015. Banks attempt to repossess 7,000-plus homes. The Irish Times, 09/03/2015. IMMERGLUCK, D. 2009. Foreclosed: High Risk Lending, Deregulation and the Undermining of America's

Mortgage Market, Ithaca, NY, Cornell University Press. IMMERGLUCK, D. 2011. From risk-limited to risk-loving mortgage markets: origins of the U.S. subprime

crisis and prospects for reform. Journal of Housing and the Built Environment, 26, 245-262. IMMERGLUCK, D. & SMITH, G. 2006a. The external costs of foreclosure: The impact of single family

mortgage foreclosures on property values. Housing Policy Debate, 17, pg 57 - 79. IMMERGLUCK, D. & SMITH, G. 2006b. The Impact of Single-family Mortgage Foreclosures on

Neighborhood Crime. Housing Studies, 21, 851 - 866. JONES, R. W. & PRIDEMORE, W. A. 2012. The Foreclosure Crisis and Crime: Is Housing-Mortgage Stress

Associated with Violent and Property Crime in U.S. Metropolitan Areas? Social Science Quarterly, 93, 671-691.

KAPLAN, D. & SOMMERS, G. 2009. An analysis of the relationship between housing foreclosures, lending practices, and neighborhood ecology: Evidence from a distressed county. Professional Geographer, 61, 101-120.

KELLY, S. 2014. Light-Touch Regulation: The Rise and Fall of the Irish Banking Sector. In: MACLARAN, A. & KELLY, S. (eds.) Neoliberal Urban Policy and the Transformation of the City: Reshaping Dublin. Basingstoke: Palgrave Macmillan.

KEMPSON, E. 2002. Over-indebtedness in Britain. Bristol: Department of Trade and Industry. KEMPSON, E., MCKAY, S. & WILLITTS, M. 2004. Characteristics of families in debt and the nature of

indebtedness. Department of Work and Pensions Research Report 211. Leeds: Corporate Document Services.

KIRBY, P. 2010. Celtic Tiger in Collapse: Explaining the Weaknesses of the Irish Model, Basingstoke, Palgrave Macmillan.

KITCHIN, R. 2014. The geographical distribution of negative equity of post-2001 built residences in Ireland [Online]. Ireland After Nama. Available: https://irelandafternama.wordpress.com/2014/01/29/the-geographical-distribution-of-negative-equity-of-post-2001-built-residences-in-ireland/ [Accessed 26/03/2015.

LANCE, C. E., BUTTS, M. M. & MICHELS, L. C. 2006. The Sources of Four Commonly Reported Cutoff Criteria: What Did They Really Say? Organizational Research Methods, 9, 202-220.

LANGLEY, P. 2006a. The making of investor subjects in Anglo-American pensions. Environment and Planning D: Society and Space, 24, 919-934.

LANGLEY, P. 2006b. Securitising Suburbia: The Transformation of Anglo-American Mortgage Finance. Competition & Change, 10, 283-299.

LANGLEY, P. 2008. Sub-prime mortgage lending: a cultural economy. Economy and Society, 37, 469-494.

LEE, R., CLARK, G., POLLARD, J. & LEYSHON, A. 2009. The remit of financial geography-before and after the crisis. Journal of Economic Geography, 9, pp 723 - 747.

LIBMAN, K., FIELDS, D. & SAEGERT, S. 2012. Housing and Health: A Social Ecological Perspective on the US Foreclosure Crisis. Housing, Theory and Society, 29, 1-24.

MAITRE, B., RUSSELL, H. & WHELAN, C. T. 2014. Trends in Economic Stress and the Great Recession in Ireland: An Analysis of the CSO Survey on Incomes and Living Conditions. Dublin: ESRI & Dept of Social Protection.

MARTIN, R. 2011. The local geographies of the financial crisis: from the housing bubble to economic recession and beyond. Journal of Economic Geography, 11, 587-618.

MURPHY, E., FOX-ROGERS, L. & GRIST, B. 2014. The political economy of legislative change: Neoliberalising planning legislation. In: MACLARAN, A. & KELLY, S. (eds.) Neoliberal Urban Policy and the Transformation of the City: Reshaping Dubli. Basingstoke: Palgrave Macmillan.

MURPHY, E. & SCOTT, M. 2014a. ‘After the crash’: Life satisfaction, everyday financial practices and rural households in post Celtic Tiger Ireland. Journal of Rural Studies, 34, 37-49.

MURPHY, E. & SCOTT, M. 2014b. Household vulnerability in rural areas: Results of an index applied during a housing crash, economic crisis and under austerity conditions. Geoforum, 51, 75-86.

NEPAL, B., TANTON, R. & HARDING, A. 2010. Measuring Housing Stress: How Much do Definitions Matter? Urban Policy and Research, 28, 211-224.

NETTLETON, S. & BURROWS, R. 1998. Mortgage Debt, Insecure Home Ownership and Health: An Exploratory Analysis. Sociology of Health & Illness, 20, 731-753.

NETTLETON, S. & BURROWS, R. 2000. When a Capital Investment Becomes an Emotional Loss: The Health Consequences of the Experience of Mortgage Possession in England. Housing Studies, 15, 463 - 478.

NEWMAN, K. 2009. Post-industrial widgets: Capital flows and the production of the urban. International Journal of Urban and Regional Research, 33, 314-331.

NIEDT, C. & MARTIN, I. W. 2012. Who Are the Foreclosed? A Statistical Portrait of America in Crisis. Housing Policy Debate, 23, 159-176.

NORRIS, M. & WINSTON, N. 2011. Housing Wealth, Debt and Stress Before, During and After the Celtic Tiger. In: FORREST, R. (ed.) Housing Markets and the Global Financial Crisis. Edward Elgar.

NYBERG, P. 2011. Misjudging Risk: Causes of the Systemic Banking Crisis in Ireland. Report of the Commission of Investigation into the Banking Sector in Ireland. Dublin: Government of Ireland.

O'NEILL, P., DUANTE-CAMACHO, O., CASIRO, J., GWYTHER, G., PHIBBS, P., BRYAN, D., RAFFERTY, M. & ALLON, F. 2010. The experience of mortgage distress in Western Sydney [Online]. Penrith, N.S.W.: University of Western Sydney. Available: http://www.uws.edu.au/__data/assets/pdf_file/0019/140536/mortgage_distress_report-webversion_hires.pdf.

PUNCH, M. 2005. Uneven development and the private rental market: Problems and prospects for low income households. In: NORRIS, M. W. & REDMOND, D. (eds.) Housing Contemporary Ireland: Policy, Society and Shelter. Dublin: Institute of Public Administration.

REA, W., YUEN, J., ENGELAND, J. & FIGUEROA, R. 2008. The dynamics of housing affordability. Labour and Income, 20, 37-48.

REGLING, K. & WATSON, M. 2010. A Preliminary Report on the Sources of Ireland's Banking Crisis. Dublin: Government of Ireland.

ROHE, W. & LINDBLAD, M. 2013. Reexamining the Social Benefits of Homeownership after the Housing Crisis. Homeownership Built to Last: Lessons from the Housing Crisis on Sustaining Homeownership for Low-Income and Minority Families. Harvard Business School, Massachusetts.

RONALD, R. 2008. The ideology of homeownership: Homeowner societies and the role of housing, Basingstoke, Palgrave.

RUDE, C. 2013. The Role of the Federal Reserve: Lender of Last Resort. In: WOLFSON, M. H. & EPSTEIN, G. A. (eds.) The Handbook of The Political Economy of Financial Crises. New York: Oxford University Press.

RUGH, J. S. 2015. Double Jeopardy: Why Latinos Were Hit Hardest by the US Foreclosure Crisis. Social Forces, 93, 1139-1184.

RUSSELL, H., MAITRE, B. & DONNELLY, N. 2011. Financial Exclusion and Over-indebtedness in Irish Households. Dublin: Dept. of Community, Equality and Gaelteacht Affairs & ESRI.

SCANLON, K., LUNDE, J. & WHITEHEAD, C. 2008. Mortgage Product Innovation in Advanced Economies: More Choice, More Risk. European Journal of Housing Policy, 8, 109 - 131.

SCHLOEMER, E., LI, W., ERNST, K. & KEEST, K. 2006. Losing Ground: Foreclosures in the subprime market and their cost to homeowners. Washington D.C: Centre for Responsible Lending.

SCHUETZ, J., BEEN, V. & ELLEN, I. G. 2008. Neighborhood effects of concentrated mortgage foreclosures. Journal of Housing Economics, 17, 306-319.

SHLAY, A. B. 2006. Low-income Homeownership: American Dream or Delusion? Urban Studies, 43, 511-531.

SIRR, L. (ed.) 2014. Renting in Ireland: The Social, Voluntary and Private Secctors, Dublin: Institute of Public Administration.

STAMP, S. 2009. A Policy Framework for Addressing Over-Indebtedness. Dublin: Combat Poverty Agency.

STIGLITZ, J., SEN, A. & FITOUSSI, J.-P. 2009. Report by the Commission on the Measurement of Economic Performance and Social Progress. Paris: The Commission on the Measurement of Economic Performance and Social Progress.

STONE, M. E. 2006. A Housing Affordability Standard for the UK. Housing Studies, 21, 453-476. STONE, M. E., BURKE, T. & RALSTON, L. 2011. The residual income approach to housing affordability:

the theory and the practice. Melbourne: Australian Housing and Urban Research Institute. TAYLOR, C. 2015. Competition best chance for hard-pressed SVR mortgage holders. The Irish TImes,

01/04/2015. UK. DEPARTMENT FOR BUSINESS, INNOVATION AND SKILLS, 2010. Over-indebtedness in Britain:

Second follow-up report. London: Department for Business, Innovation and Skills,. VANDONE, D. 2009. From Indebtedness to Over-Indebtedness. In: VANDONE, D. (ed.) Consumer Credit

in Europe. Berlin: Springer-Verlag. WAINWRIGHT, T. 2009. Laying the Foundations for a Crisis: Mapping the Historico-Geographical

Construction of Residential Mortgage Backed Securitization in the UK. IJURR, 33, 372-388. WAINWRIGHT, T. 2010. 'It's crunch time': the 'lost' geographies of the crisis. Environment and Planning

A, 42, 780-784. WESTON, C. 2011. Elderfield orders banks to stop hiking variable rate mortgages. The Irish

Independent, 14/10/2011. WHITEHEAD, C. 2012. Owner-occupation in an increasingly uncertain world. In: RONALD, R. &

ELSINGA, M. (eds.) Beyond Homeownership: Housing, Welfare and Society. Routledge: Oxon. WOLFSON, M. H. & EPSTEIN, G. A. (eds.) 2013. The Handbook of The Political Economy of Financial

Crises, New York: Oxford University Press. WYLY, E., MOOS, M., HAMMEL, D. & KABAHIZI, E. 2009. Cartographies of race and class: Mapping the

class-monopoly rents of American subprime mortgage capital. International Journal of Urban and Regional Research, 33, 332-354.