The current surge in inflation is forcing the working classes to dip into their savings, if available, to cover their daily expenses [1]. In addition to the consumption capacities tied to revenues, it is also draining their often limited savings, which are all the more exposed as they are deposited in low-yielding accounts. In this context, it is important to study the wealth accumulated by households, as it is a buffer against economic shocks, which is why the value of housing is not taken into account, even though ownership of the main residence also contributes to inequality (Bugeja-Bloch, 2013). Moreover, housing carries a greater share of the budget for lower-income households and is an incompressible expense (Cusset, Prada-Aranguren, Trannoy, 2021). The study of wealth and of its contribution to the social structure usually focuses on well-off households. While Piketty has eloquently shown why wealth matters as much in the upper echelons of society, this is also the case at the lower end of the social ladder. In a society that (re)places great importance on wealth, not possessing it places the households affected in a situation of social inferiority and economic vulnerability. Our aim is to identify which groups are exposed to this lack of wealth.
For this purpose, we mobilise a large-scale statistical source, largely unexploited by researchers—the Household Financial and Consumption Survey developed by the European Central Bank (Box 1). It is the only European-wise survey that integrates data on household assets, thereby making it possible to develop a measure of economic vulnerability, asset-based poverty, already mobilised by the Organisation for Economic Co-operation and Development (OECD). Asset poverty is defined as holding insufficiant liquid asset reserves (immediately available) [2] to maintain a living standard equivalent to the poverty line for at least three months (OECD, see Balestra and Tonkin, 2018).
The analysis of asset poverty by social class and age enables us to measure the extent of economic vulnerability, to identify its most concentrated zones, and question the criteria for analysing social classes as well as comparing levels of inequality between European societies.
Box 1: Data source
To perform a joint analysis of wealth, income, social class, and age in different European countries, we mobilised data from the European Central Bank’s
Household Financial and Consumption Survey. Conducted in 2017, the third wave of the survey was based on a representative sample of the population of 22 European countries. It should be noted that these surveys are criticised, by Piketty in particular, for their relative lack of reliability if compared to administrative sources, especially for high levels of assets. Small sample sizes, as well as missing values, could also raise concerns. Nevertheless, as wealthy households were over-represented in the survey and, more fundamentally, we are interested in the bottom half of social structure, the gaps we measure are less sensitive to a potential underestimation of substantial wealth.
To address the problems associated with small samples, we restricted the analysis to five countries representative of major social systems (Esping-Andersen, 1999) with substantial numbers of respondents: Finland (10 210) for the social democratic countries in Northern Europe; France (13 685) and Germany (4 942) for the continental European countries where insurance plays a predominant role in their social model; Spain (6 413) for the Mediterranean countries where family plays a vital role in the social compromises; and Ireland (4 793) as a model close to the United Kingdom in terms of the importance attached to the market. We used the European Socio-economic Groups (ESeG) to code social classes and compared their distributions in the working population with other surveys (Labor Force Survey, 2011; Adult Education Survey, 2011). No significant distortions or discrepancies were found based on the above comparison.
Context
This article builds on previous research on social insecurity (Duvoux et Papuchon, 2018), defined as an inability to project oneself positively into the future. Based on a measure of subjective poverty, the research revealed the extension of a form of social insecurity to categories commonly considered immune to monetary poverty: certain segments, particularly workers and employees in the working classes, but also the self-employed and non-proprietary pensioners. One of the results of this research was to highlight the interaction effects between certain statuses (of retirees, for example) and the possession of assets. While retirees in France are relatively well protected from monetary poverty, taking into account housing tenure status sheds light on the feeling of poverty among retired tenants, a feeling associated with a negative perception of one’s past and future social trajectory, firmly rooted in deteriorated material living conditions.
These results stress the importance of the patrimonial middle class (defined by Thomas Piketty (2013) as the individuals placed between the richest 10% and the poorest 50%) in the structuring of today’s society, as opposed to that of the beginning of the twentieth century. The latter was characterised by a very high concentration of wealth, and even more so in the large cities, where the small peasant property could not, by construction, temper the concentration observed. The approach to poverty and inequality, which focuses on their subjective and temporal dimensions, has therefore been extended by reassessment of the typology of social classes in terms of wealth (Duvoux et Papuchon, 2022) and of how it could structure expectations, possibilities, and accumulation expectations or, on the contrary, diminished prospects, depending on different class or age.
By mapping out liquid wealth (defined as total net wealth excluding property and vehicles), i.e. the savings effectively available to cope with a significant economic shock (unemployment, separation, accident, etc.), we were able to develop a measure of the relative protection provided by such liquid assets compared to income. Table 1 demonstrates how many months of income such liquid assets are equivalent to for each age and class group in the five European countries selected (data from the 2014 wave). The green zone represents a form of security, the white zone an intermediate situation and the orange zone a form of fragility.

- Table 1
- Population: People in the labor force aged 18 years or older living in Germany, Finland, Metropolitan France, Ireland, or Spain, HFCS 2, 2014 wave.
Source : Duvoux et Papuchon, 2022
The identification of this zone of fragility is consistent with OECD’s measure of asset poverty. Asset poverty is defined as an individual in a household whose liquid assets are insufficient to maintain oneself for at least three months above the income poverty line, which the OECD defines as half of the median household income for the whole population of the country concerned [3].
The liquid assets taken into account are the total liquid financial resources of the household (deposits, shares, etc.), from which is subtracted the value of non-collateral debts (overdraft and credit debts, non-mortgage loans) [4]. As for income, all measures refer to total gross household income that is available in the database for all the countries selected.
This makes it possible to establish a three-level classification: households that are ‘income poor only’ (G1); those that are not income poor but are asset poor according to the above criterion are considered to be ’economically vulnerable’ (G2). Lastly, those in both types of poverty are classified as ‘asset and income poor’ (G3). ‘Income poverty’ includes therefore group 1 and 2.
‘Asset poverty’ is the general category studied in this paper, and brings together two sub-populations: those who are asset poor but not income poor, who are ‘only’ economically vulnerable (Group 2); those who combine these two forms of fragility (Group 3). The economic vulnerability indicator, defined as households that are asset poor, but not income poor, offers an indication of the size of the populations situated in the upper halo of income poverty but still exposed to a hazard or shock.
Economic vulnerability by age and social class
Figure 1 shows that at the country level, the proportion of households in economic vulnerability is significantly higher (between 30% and 40%) than that in income poverty, which accounts for around 20% of households in all five countries except France (15%). Ireland has the highest level of relative income poverty and economic vulnerability: 22% of households are income poor, while a further 43% do not have sufficient liquid assets to maintain a living standard above the poverty line for at least three months. Finland, a Nordic country known for its egalitarian welfare model, comes second, with 39% of households in economic vulnerability, a result consistent with findings that Nordic countries are often more unequal than continental European countries in terms of wealth (Waitkus et Pfeffer, 2021). Asset poverty, which includes households that cumulatively experience both forms of poverty, is even more widespread, affecting almost or more than half of all households. These results are consistent with the orders of magnitude of OECD’s measure of asset poverty.

- Figure 1 Percentage of households in income poverty and asset poverty
- Note: In Germany, 14.8% of working population are asset and income poor (G3), 6.5% are income poor only (G1), and 31.9% are economically vulnerable (G2), i.e. asset poor only. Income poverty (G1+G3) in Germany amounts to 21.3% and asset poverty (G2+G3) to 46.7%.
Population: working population in Germany, Spain, Finland, France, Ireland
Source: HFCS 3, 2017 wave.
From an age-category perspective, there is a striking difference in asset poverty across the five countries (Figure 2): in general, young people (18-34 years olds) face an extremely high level of asset poverty (from 60% in France to 70% in Spain), while older groups are confronted with a less, alberit significant, asset poverty, ranging from 44% in Germany to 54% in Ireland. With a gap of 25 points between the youngest and oldest groups, Spain shows the highest age-related inequality: this suggests that the critical role of the family in social policies is closely linked to a deteriorated situation for the younger generation, echoing the results of the work of Cécile Van de Velde, Camille Peugny and
Tom Chevalier). In other words, dependance on family is also found in the distribution of liquid assets. France, although characterised by lower proportions overall, still has a significant gap of 12% between the two age groups.

- Figure 2 Asset-based poverty at household level by age and country
- Note: 67.4% of 18-34 years olds in Finland are asset poor, 59.6% for 35-49 years olds and 53.1% for 50-64 years olds
Population: working population in Germany, Spain, Finland, France, Ireland
Source: HFCS 3, 2017 wave
In all socio-professional categories, asset poverty is considerably more widespread than income poverty (Figure 3). Mirroring the results established via the wealth-income ratio (Duvoux et Papuchon, 2022), there are significant differences between managers and technicians on the one hand (30% and 44% asset poor, respectively), and lower-skilled employees and workers, as well as the unemployed, on the other (from 60% to 77% asset poor). Small entrepreneurs, whose socio-economic profile is more heterogeneous than that of wage earners as it includes independent earners of various scales [5], remain a specific group, in a third position vis-à-vis capital and labour.
The articulation of the two types of poverty also reveals this social hierarchy (Figure 3). While 10% of the small entrepreneurs are affected by the cumulative effect of the two forms of poverty, this is also the case for 4% of professionals and between 6% and 7% of technicians and associate professionals, as well as skilled service employees and industrial workers. The figures are higher for low status employees (18%) and even more for the unemployed (40%).

- Figure 3: Income poverty and asset poverty by class
- Note: 9.7% of skilled employees are income poor, and 51% asset poor.
Population: working population in Germany, Spain, Finland, France, Ireland
Source: HFCS 3, 2017 wave
Wealth and income inequality patterns are also found throughout the life course (Figure 4). Crossing the two sets of information reveals several cleavages. Although it is necessary to be cautious as only net liquid assets are taken into consideration, in terms of asset poverty, young people (18-34 years old) and intermediates (35-49 years old) share the same level of economic vulnerability, in contrast to older working populations who have been able to accumulate savings as a buffer. For income poverty, the most noticeable gap seperates the young from the other two age groups. As far as the accumulation of the two forms of poverty (asset and income poor) is concerned, the sharpest divide is between the under-35s and the others. Overall, as people advance in age – insofar as the three age groups make it possible to capture this advance – it is firstly income poverty that decreases (between the 18-34 and 35-49 age groups) and then asset poverty. The relevance of the asset poverty indicator is to reveal the scope of the constellation concerned by the absence of liquid reserves, which may go hand in hand with indebtedness for purchasing the main residence.

- Figure 4 Income and/or asset poverty by age
- Note: 21.1% of 18-34 years olds are income and asset poor, 6.4% income poor only, and 42.7% economically vulnerable
Population: working population in Germany, Spain, Finland, France, Ireland
Source: HFCS 3, 2017 wave
When examining asset poverty in detail by occupational category and by age group (Figure 5), we can identify an extreme inequality that confirms the previous results. Asset poverty intensifies as one moves further down the social ladder and becomes less pronounced with age, with the previously mentioned notable exception of small entrepreneurs. There is a gap of 65 percentage points between older managers (23%) and young unemployed individuals (88%), an inequality that is even more pronounced than the one observed in terms of income poverty.

- Figure 5 Asset poverty at household level by age and class
- Note: Among skilled service employees, 69.8% of 18-34 years olds, 48% of 35-49 years olds and 47.9% of 50-64 years olds are asset poor
Population: working population in Germany, Spain, Finland, France, Ireland
Source: HFCS 3, 2017 wave
The implications of this research
This study highlights three key aspects. The first is the importance of taking wealth into consideration in the discussion of inequalities in general and class inequalities in particular. While it has been shown that wealth is an important criterion of ethno-racial inequalities in the United States (Oliver, Shapiro, 1995) and of gender inequalities in France (Bessière, Gollac, 2020), attempts to systematically articulate the level of wealth ownership and one’s social class have not yet been explored to the same extent as how important wealth is for inequalities between social groups.
Nowadays, it is either considered as one criterion among others, or rejected in favour of social hierarchy approaches centred on the sole criterion of wealth. In fact, some sociologists such as Mike Savage or Adkins, Cooper and Konings (2020) in Australia propose to integrate wealth into the analysis of classes, even seeking to put forward a typology of these classes based on wealth alone. We believe that occupational status remains a determining factor, but it must be analyzed in conjunction with levels of economic resources, which, in the case of wealth, results in considering differentiation based on one’s position in the life cycle.
To understand class, we must pay joint attention to both labour and wealth. This may be regarded as a truism with regard to the classical analyses of social classes centered on the conflict between capital and labour. Nevertheless, it implies liberating oneself from cumbersome disciplinary boundaries, as illustrated by the small amount of research crossing these dimensions. Moreover, wealth is now more widely distributed, at least to a ’patrimonial middle class’ (Piketty, 2013). Measuring economic vulnerability precisely enables us to mark the limits of this secular process of democratising access to wealth and to identify the categories placed in a vulnerable situation due to the renewed role of wealth in the class structure (Chauvel et al., 2021).
The second implication lies in the question of how relevant it is to provide a high figure for the economically vulnerable population. In France, asset poverty concerns 46% of the population, 36% of whom are economically vulnerable and 10% in both income and asset poverty. Is this figure excessive? Beyond the precise number, the order of magnitude is significant. It is not, however, inconsistent with other indicators. Recent studies seek to circumscribe the forms of accumulation between different dimensions of poverty (Blasco, Picard, 2021) and result, by construction, in more restrictive visions of social disadvantage. In contrast, our approach emphasizes the existence of a threshold of fragility around the 4th decile of the population. As outlined by Michèle Lelièvre, several studies concur on the point that material difficulties, exposure to economic circumstances, and the weight of constrained expenses in the budgets are particularly strong up to these levels (Duvoux et Lelièvre, 2021, p.140).
Thirdly, there are changes in the conventional comparisons of societies, particularly in Europe, where classical typologies of countries continue to be used as references. As shown by the case of Finland in our study, but also by the case of Sweden in more systematic comparisons (Pfeffer et Waitkus, 2021), the levels of income inequality and social protection that generally underpin the construction of welfare state models are not aligned with those of wealth inequality. The differentiation between egalitarian northern European models, intermediate continental European models and southern European models with a low level of protection needs to be supplemented by an examination of the public and private buffers provided by social ownership (Castel, 1995) of the state or insurance on the one hand and private ownership on the other. This is particularly relevant given that social state reforms open up spaces for leveraging private savings to mitigate shocks or maintain living standards in retirement. Our study highlights France’s favourable position in international comparisons in terms of asset poverty as well as income poverty, which does not alter the massive scope of economic vulnerability and, more importantly, asset poverty.
Conclusion
This article proposes to reassess rather radically the levels of diffusion of social difficulties measured in different countries. Taking wealth into consideration not only at the top but also at the bottom of society offers a better perspective on the extent of social insecurity. This is particularly the case in countries such as France, which have managed to limit income poverty through state intervention. Such an approach also suggests that we should question what for us seems obvious in the ranking of countries by their greater or lesser degree of inequality.
Measuring inequality in this way also helps to capture the permanence of class inequalities. Differences in exposure to economic insecurity are one way of measuring them and one that has the advantage of broadening the spectrum of criteria taken into account. Undoubtedly, several pieces of information should be integrated to analyse a social position. It is not possible to do everything with the data we have mobilised. However, at the very least, the socio-professional group, wealth, income and age are all salient elements, all of which indicate that they play a role in a society that, whilst remaining a wage-earning society, is nevertheless once again marching towards a wealth-owning society.