Wealth Stratification and the Insurance Function of Wealth

This thematic issue examines the insurance function as a mechanism to underlie wealth effects on various outcomes. The articles in this issue shed an innovative light on the insurance function of wealth concerning a range of topics rele‐ vant to social stratification and social policy researchers. This editorial provides an overview of the contributions of this thematic issue and highlights some gaps and remaining open questions. Altogether, the contributions suggest that wealth can provide insurance against adverse life events in various contexts. However, this insurance effect depends on welfare state characteristics, wealth portfolios


Introduction
Until the end of World War II, only an elite group of people had access to wealth and the possibility to accumulate it.Accordingly, wealth research fell into the realm of elite sociology, and wealth was considered primarily in the context of power, for example in LeBon (1939LeBon ( /1895)), Michels (1925Michels ( /1911)), Mosca (1950Mosca ( /1896)), Pareto (1955Pareto ( /1916)), or, more recently, in Mills (1956).Only since the 1950s, in times of economic prosperity and peace in the industrialized world, has wealth become a quantitatively significant economic resource for the population as a whole.
For some decades now, wealth has (again) been high on the research agenda of social stratification scholars for several reasons.The first is population aging, accompanied by a public pension retirement limit set by the modern welfare state.Due to this development, old age has become a distinct phase of life, which is much more structured by leisure and consumption than by labor market activity (Kohli, 1988).Thus, as individuals grow older, wealth increasingly determines their economic status, while income becomes less meaningful.
The second reason is the increasing importance of private provisions for old age.As a reaction to population aging and declining birth rates, welfare states nowadays reduce public pension benefits and try to set incentives for private old-age provisions.Consequently, responsibility for old-age provision is partly transferred from the welfare state to the individual actor, making the accumulation of personal wealth an even more relevant topic for the latter.Third, the growing interest in social science research on wealth correlates with the increased availability of data that enables researchers to analyze personal wealth holdings empirically.
Recent research has found that private wealth is an essential component of socioeconomic status with substantial effects on different outcomes, including education, family formation, and health (cf.Killewald et al., 2017).Less clear, however, are the mechanisms underlying these effects.Interested mainly in the impact of parental wealth on the educational outcomes of children, Hällsten and Pfeffer (2017) distinguish between three such mechanisms, which can also be applied to other outcomes.The (a) purchasing mechanism refers to the fact that wealth (similar to income) can be used to purchase goods and services related to achieving or increasing the outcome of interest-in this case, wealth is directly consumed.The (b) insurance mechanism represents the psychological benefits of having wealth and being able to make use of it, if necessary (wealth as a safety net).Finally, the (c) social norms mechanism operates through the positive association between wealth and norms and values, which favor the outcome of interest.Our thematic issue is dedicated to the insurance mechanism.
There are two scenarios of how wealth can affect behavior by acting as a safety net.First, wealth can protect individuals from the consequences of adverse events ("actual insurance function"), like job loss, divorce, illness, or reduced income during retirement.Second, wealth allows individuals to make riskier decisions because they can anticipate wealth to protect against adverse events that may occur in the future ("anticipated insurance function").From a macro perspective, the life-cycle hypothesis (Modigliani & Brumberg, 1954) suggests a substitution effect between the generosity of welfare state services and individual savings.The insurance function of wealth may have been particularly important during the Covid-19 pandemic.

Wealth as a Buffer Against the Negative Consequences of Adverse Life Events
Numerous studies indicate that wealth works as a safety net, buffering the negative consequences of adverse life events.Leopold and Schneider (2011) show that adult children's economic need triggers parental gifts.Rodems and Pfeffer (2021), as well as McKernan et al. (2009), analyze the buffering function of wealth in experiencing material hardship.Rodems and Pfeffer (2021) find that household net worth effectively buffers the risk of material hardship associated with divorce, disability, and income loss.McKernan et al. (2009) find that families with liquid assets are less likely to experience material hardship in the aftermath of an involuntary job loss, the onset of a health-related work limitation, or a parent leaving the family.Moreover, they find that the buffering function of liquid assets works most efficiently for the bottom and middle terciles of the income distribution but less for the top one.
Findings are less conclusive for subjective well-being (SWB).Smith et al. (2005) find a buffering effect of household net worth against the detrimental effects of a disability on SWB.Kuhn and Brulé (2018) do not find any buffering effect of wealth for the negative consequences of separation, death of a closely related person, unemployment, and disability on SWB.Interested in the mod-erating effect of wealth during the Covid-19 pandemic, Roll and Despard (2020) do find a buffering effect of liquid assets on the negative impacts of Covid-19-related job and income loss on financial distress.

The Effect of Wealth on Individual Behavior
Transfers and inheritances can affect individual behavior in various ways.Basiglio et al. (2022) show with Dutch data that individuals perceive expected inheritances as a potential increase in their wealth, leading to a reduction in their savings.Moreover, expected inheritances affect intentions to bequeath and intended choices on work versus leisure in the future.Similar findings have been reported by Lundberg (2020) for Sweden.
Wealth, both current and future expected assets, cannot only affect saving and labor market behavior but can also allow individuals to make riskier decisions, knowing that they will be financially protected in case of failure.Such decisions include educational decisions (choosing a more versus a less competitive educational track or field of study) and occupational decisions (applying for higher and more demanding versus lower and less demanding occupational positions), but also investment decisions (investing in riskier portfolios with higher payoffs versus more conservative portfolios with lower payoffs) or family decisions (timing of marriage and childbirth, divorce, fertility decisions).Previous research shows that an increase in housing wealth increases fertility rates among homeowners (Lovenheim & Mumford, 2013) as well as college enrollment rates (Lovenheim, 2011) in the US.
Several studies show that the level of individual wealth is related to risk adversity in investment behavior, though with unclarity about the direction of this relationship (Brunnermeier & Nagel, 2008;Kihlstrom et al., 1981;Paravisini et al., 2010).As to expected wealth, Greenberg (2013) finds a low-risk aversion among individuals who imagine being wealthy in the future.

Wealth and the Welfare State
A huge body of research, mostly from economics, analyses the effects of social security on individual savings behavior (e.g., Attanasio & Brugiavini, 2003;Attanasio & Rohwedder, 2003;Farley & Wilensky, 1985;Feldstein, 1983;Feldstein & Pellechio, 1979;Lefebvre & Perelman, 2020).Most of these studies find that the generosity of (public) social security partly offsets individual savings as suggested by the life-cycle hypothesis, especially at the bottom and in the middle of the income distribution.
International comparative research showed that wealth is more relevant for several individual-level outcomes-including SWB (Hochman & Skopek, 2013), health (Maskileyson, 2014), and education (Pfeffer & Hällsten, 2012)-in countries with less generous social welfare state services as compared to countries with more generous welfare state services.These findings indicate that welfare state services can moderate the relevance of wealth as private insurance.

Open Questions
Despite the numerous studies mentioned above, various open research questions remain regarding the insurance function of wealth.As to the buffering effect of wealth, few studies directly analyzed if wealth can buffer the consequences of adverse life events, how far the buffering effects differ across such events, and whether there is heterogeneity in such buffering effects across the distribution of wealth.Wealth as an additional resource to compensate for the negative consequences of adverse life events might be especially relevant in a nationwide crisis such as the Covid-19 pandemic (cf.Kuypers et al., 2022).
As to the direct effects of the insurance function of wealth on various outcomes, little research has been done on risk-taking behavior outside investment decisions, like marriage behavior, the timing of childbirth, or educational and occupational decisions.Especially interesting here is the anticipated insurance effect: Do people who expect to receive larger amounts of wealth make riskier life decisions?Or does the timing of important life-course transitions and events differ between persons who expect and don't expect to receive larger amounts of wealth?
Regarding the moderating effect of welfare state services or social security wealth on the importance of wealth as private insurance, previous research lacks large international comparative studies to analyze this relationship.Open questions also remain regarding the definition and operationalization of social security wealth.
Moreover, there remain open questions about how individuals and families manage their wealth and how wealth and its insurance function are perceived.The potential of wealth to buffer against adverse events will only affect behavior if actors assume that their wealth will remain stable or increase, but not if they fear wealth losses.Likewise, wealth can only buffer against adverse events if actors have control over their wealth.
Our thematic issue aims to shed light on these and related open questions regarding the insurance function of wealth.

Content and Contributions of the Thematic Issue
Our issue includes six contributions.Two of them are large international comparative studies: One compares 27 European countries (Heidenreich & Broschinski, 2023) and the other one looks at 17 European countries and the US (Rapp & Humer, 2023).The remaining four are single-country studies carried out with data from Germany, Italy, and the UK.Five of the six studies carry out quantitative data analyses, and one study (Carmichael, 2023) decided on a qualitative research design.Three out of six studies measure wealth in terms of housing wealth (homeownership/tenure status and housing value; see Althaber et al., 2023;Bedük, 2023;Heidenreich & Broschinski, 2023), one study measures it in terms of financial wealth (Rapp & Humer, 2023) and another creates an index to directly measure the insurance capacity of wealth (Gritti et al., 2023).One study investigates high-net-worth individuals (Carmichael, 2023).
The first two contributions test the insurance function of wealth each for a single country but under very different circumstances: Bedük (2023) tests the insurance function of wealth against job loss in the UK, while Gritti et al. (2023) test the insurance function of wealth against the consequences of the Covid-19 pandemic in Italy.Bedük (2023) examines the effect of job loss on several outcomes and its moderation by wealth in the UK with household panel data from 1991 to 2008.The author uses homeownership status and housing values as wealth measures and earnings, net household income, relative and absolute poverty, and life satisfaction as outcomes.He finds that renters have a higher risk of job loss than owners, while housing values do not matter.For the effect of job loss on most examined outcomes, he similarly finds greater differences between renters and owners as compared to the differences across housing value percentiles.Also, he finds a distinct moderating effect of the housing value on poverty.Gritti et al. (2023) analyze the impact of the Covid-19 pandemic in Italy on individuals' psychological and socioemotional responses-measured as dispositional optimism.To operationalize the insurance function of wealth as directly as possible, they create an "insurance capacity" index.This index combines respondents' capacity to cover their financial obligations and afford their basic necessities in case of a shortage of income with their current housing situation.The authors then analyze the relationship between the pandemic and dispositional optimism across groups of individuals with different levels of insurance capacity.They find slightly higher optimism for individuals with a higher insurance capacity.Overall, however, their findings show only weak support for the insurance function of wealth in the socioemotional sphere.
The next contribution tests differences in the insurance function across country contexts.Heidenreich and Broschinski (2023) compare the insurance function of wealth against unemployment across 27 European countries.The authors use the EU-SILC data to examine homeownership as a form of wealth that can be used as insurance against life risks.They look at short-term unemployment's effect on households' perceived financial situation and find that debt-free homeownership reduces financial stress among homeowners due to unemployment compared to tenants and owners holding debt.The authors use the cross-national comparison of EU-SILC to examine if social protection regimes moderate the effect of homeownership and outstanding mortgage payment.Against their expectation, they find that this difference in perceived financial stress between debt-free owners and owners who still pay off mortgages and tenants is larger in countries with high unemployment benefits.
In the next contribution, Rapp and Humer (2023) evaluate directly how welfare states and family transfers compensate for vulnerability in European countries and the US.The authors propose a measure of vulnerability, which besides asset poverty also takes into account buffering by public insurance programs and the possibility of receiving financial assistance from relatives or friends.The measure is derived from and applied to a sample of 17 European countries and the US based on data from the Household Finance and Consumption Survey (HFCS) and data from the Survey of Consumer Finances (SCF), respectively.Results show that while asset poverty in the US is lower than in most European countries, households are less vulnerable in the latter due to higher cushioning through insurance systems.Help through social networks is substantial in several countries, yet may not be available to its full extent when shocks are distributed broadly across the population.Taking into account the insurance function of wealth (and private transfers) in the measurement of poverty thus allows a different assessment of poverty rates in countries with different welfare state systems.
The two concluding contributions evaluate the strategies of how individuals and families manage assets to maintain their wealth and its insurance function.Althaber et al. (2023) assess how within-couple income and wealth inequalities affect couples' money management strategies.Carmichael (2023) evaluates de-risking strategies of high-net-worth individuals in the UK.Althaber et al. (2023) evaluate how income and wealth inequalities among couples are associated with money management.Using data from the German Socio-Econonomic Panel, they find that couples with unequal income are more likely to pool their money.In contrast, similar-income couples are more likely to manage money independently.Yet, they find the opposite within couple wealth inequality: Couples with unequal wealth are more likely to manage money independently, while couples with similar levels of wealth are more likely to pool their money.Both patterns are independent of which partner has more income or wealth.Carmichael (2023) analyzes how high-net-worth individuals perceive the insurance provided by their wealth and their strategies to maintain it during their future retirement based on qualitative interviews with 35 individuals in the top 5% of the net worth distribution in the UK.In contrast to many other studies, the author finds that high-net-worth individuals perceive and fear risks that may jeopardize a comfortable retirement.Thereby, she challenges the view that wealth releases individuals from the psychological burden of worrying about retirement.Carmichael identifies two main causes of worries: (a) the risk of making bad decisions due to emotions and (b) the fear of missing out on the most profitable investments and the pressure to keep up with others' wealth.To deal with these worries, high-net-worth individuals rely on two strategies: (a) relying on outside expert advice and (b) preservation through expense/debt reduction and tax reduction strategies.

Conclusion and Avenues for Future Research
Wrapping up the findings of the six articles in our thematic issue, we can conclude: 1. Wealth can insure against different kinds of negative life events.The articles in this thematic issue found wealth to insure against some of the negative consequences of unemployment and, to a lesser extent, the negative consequences of the Covid-19 pandemic.2.Not all wealth components exert an insurance function.In line with previous findings, the articles in our issue indicate that housing wealth can function as private insurance.This is especially true if the housing asset is free of debt.3.As expected and lining up with previous findings, the function of wealth as insurance seems to be more relevant in less generous welfare states than in more generous ones.Indeed, wealth appears to be an important supplement or compensation for low coverage by the social welfare state.4. Wealth is usually assumed to be pooled among couples.However, this is not necessarily the case.While couples with similar levels of wealth indeed pool their wealth resources, couples with different levels of wealth don't.This has important implications for the insurance function of wealth, implying that individuals (at some point) have access to it.5.While almost all articles show that wealth can buffer the consequences of negative events, referring to its actual insurance function, one article suggests that this might not necessarily be the case for its anticipated insurance function.Carmichael (2023) shows that very wealthy individuals feel uneasy and concerned about the future insurance capacity of their actual wealth.
Most of our questions about the insurance function of wealth formulated in the introduction have been tackled by one or several of our contributions.Only indirectly addressed by the articles in our thematic issue is the anticipated insurance function of wealth.We still do not know if people who expect to receive larger amounts of wealth make riskier life decisions, or if the timing of important life-course transitions and events differs between persons who expect and don't expect to receive larger amounts of wealth.This may be due to the high demands on the data to answer such questions.
To answer these questions, we need data on the decision behavior of individuals combined with information on their parents' wealth holdings.So far, only very few data sets provide such information, among them the German Socio-Economic Panel Study and the US-American Panel Study of Income Dynamics.Moreover, we still know little about how far the buffering effect of wealth on the consequences of negative life events differs across different life events and different outcomes and if this differs across countries.Also, while we have seen a good example of how to operationalize social welfare state benefits, the definition and operationalization of social security wealth still offer much potential for future research.In addition, the contributions in this thematic issue have only considered the insurance function in Europe and the US.Yet, the insurance function may work very differently in other countries, particularly in countries of the Global South.Althaber et al. (2023) raise new questions on wealth management affects its insurance function.If wealth is controlled by only one partner in a couple, this will affect who may profit and who is protected by the insurance function of wealth.This does not only apply to the couple but also to their extended network.Moreover, this raises the question of whose wealth insures against adverse events.The wealth of individuals, the wealth of the couple, or the wealth of the extended family?To answer this question, analyzing the insurance function of wealth within network structures may be fruitful.
Likewise, Carmichael (2023) raises new questions on how the fear of losing one's wealth affects the "anticipated" insurance function.If individuals or families are afraid to lose their wealth, this may stop them from making risky decisions.Here the question is whether this fear also emerges for individuals with moderate levels of wealth, whether it also affects other aspects than retirement planning, and whether the fear of wealth losses is also present among children.For example, while the fear of losing wealth may affect the investments of high-net-worth individuals, it may not affect the educational or career decisions of their children.
Another task for future research would be to test whether it is actually wealth that provides the buffering against adverse events or whether the variance in the consequences of adverse events can be attributed to other factors that cause both wealth and the outcomes of interest (e.g., race/ethnicity, or other dimensions of the socioeconomic status).In other words, are the effects of adverse events causally moderated by wealth (Bansak, 2021)?Depending on this, policy recommendations will be different because if the insur-ance function of wealth is not a causal moderation, we would observe the same heterogeneity in the consequences of adverse events, even if wealth were more equally distributed.
We hope that the articles in our thematic issue will contribute to the continued and increasing collection and provision of high-quality wealth data and that researchers will take up the cause of answering the still open and newly opening research questions on the insurance function of wealth.