Cross-National Support for the Welfare State Under Wealth Inequality

• Published date: 01 November 2023
• DOI: http://doi.org/10.1177/00104140231168364
• Author: Amalie Sofie Jensen,Andreas Wiedemann
• Date: 01 November 2023
• Subject Matter: Articles

Article

Comparative Political Studies

2023, Vol. 56(13) 1959–1995

© The Author(s) 2023

Article reuse guidelines:

sagepub.com/journals-permissions

DOI: 10.1177/00104140231168364

journals.sagepub.com/home/cps

Cross-National Support

for the Welfare State

Under Wealth Inequality

Amalie Sof‌ie Jensen

1

and Andreas Wiedemann

2



Abstract

Wealth is often more unequally distributed than income, and there are considerable

differences across countries. In this paper, we argue that wealth inequality helps

explain cross-national variation in support for (and the size of) the welfare state

because assets serve as private insurance. When wealth, particularly liquid assets, is

unequally distributed across the income spectrum and high-income groups hold

most assets, strong reinforcing preferences in favor of or against social policies result

in antagonistic welfare politics and less government spending. When assets are more

equitably distributed across the income spectrum, cross-cutting preferences

emerge as more people support either insurance or redistribution. Welfare politics

is consensual and facilitates a broader welfare coalition and more government

spending. We analyze original cross-national survey data from nine OECD

countries and provide evidence in support of our argument. Our f‌indings suggest

that wealth inequality reshapes the role of income in structuring welfare politics.

Keywords

social welfare programs, political economy, European politics, wealth

inequality

1

University of Copenhagen, Copenhagen, Denmark

2

Princeton University, Princeton, NJ, USA

Corresponding Author:

Andreas Wiedemann, Department of Politics and School of Public and International Affairs,

Princeton University, Princeton, NJ 08544, USA.

Email: awiedemann@princeton.edu

Income inequality has grown considerably across and within countries in

recent decades. Much less is known about how wealth is distributed across

countries or, more importantly, among individuals. Recent work based on

surveys and administrative tax data has shown that wealth is even more

unequally distributed than income (e.g., Chwieroth & Walter., 2019;Fuller

et al., 2020;Kuhn et al., 2020;Zucman, 2019). In China, Europe, and the

United States, the top 10% of the population own more than 70% of the total

wealth. The share of wealth held by the top 1% in the United States grew from

25% to 30% in the 1980s to around 40% in 2016 (Piketty, 2014;Saez &

Zucman, 2016). These numbers mask the fact that a growing share of

households live paycheck to paycheck and set aside little or no savings to

protect them from f‌inancial shocks (Kaplan et al., 2014). For example, Lusardi

et al. (2011) f‌ind that over 25% of American survey respondents, including

high-income respondents, could not come up with $2000 within 30 days for

unanticipated expenses such as a major car repair or a large medical co-

payment; 19% could only cover such expenses through payday loans or

selling items at pawnshops. Almost half of all US respondents indicated that

they could “certainly not”or “probably not”come up with the f‌inancial

resources to address a f‌inancial shock of this magnitude. Hacker et al. (2014)

expand the notion of economic risk to include income losses, unexpected

medical expenses, and limited assets to weather f‌inancial shocks, and doc-

ument that the share of Americans that are insecure by that metric has in-

creased steadily since the mid-1980s.

These patterns are not limited to the United States. Kaplan et al. (2014)

show that households living paycheck to paycheck (with very little savings)

represent about 30% of the population in Canada, Germany, and the United

Kingdom and 20% or less in Australia, France, Italy, and Spain. Nor are these

patterns limited to low-income households. The same study estimates that

two-thirds of American households that live paycheck to paycheck hold a

large amount of wealth in assets such as housing or retirement accounts but

have very little or no liquid wealth. Thus, studies of households’f‌inances

should measure multiple components of wealth and degrees of liquidity rather

than just total wealth to fully understand wealth inequality (Pfeffer & Waitkus,

2021).

Yet we know little about the policy implications of this uneven distri-

bution of different types of wealth acrossandwithincountries.Inthispaper,

we are particularly interested in how it affects support for the welfare state.

Most political economy models that consider material self-interest as the

foundation of social policy preferences are based on individuals’current or

expected income (Meltzer & Richard, 1981;Romer, 1975;Rueda &

Stegmueller, 2019) and economic shocks that negatively affect people’s

incomes (Rehm, 2016;Rehm et al., 2012). More recently there has been an

increasing focus on how wealth—both illiquid housing equity and liquid

1960 Comparative Political Studies 56(13)

assets (Ansell, 2014;Hacker et al., 2014;Hariri et al., 2020)—,participation

in f‌inancial markets (Chwieroth & Walter., 2019;Margalit & Shayo, 2021),

and access to credit and indebtedness (Ahlquist & Ansell, 2017;Markgraf &

Rosas, 2023;Wiedemann, 2021,2022)inf‌luence people’s policy prefer-

ences, fairness perceptions (Scheve & Stasavage, 2016), and their incentives

to opt out of public goods provision in light of private alternatives

(Busemeyer & Iversen., 2020). These arguments have helped advance our

understanding of redistributive politics, but often fail to take the joint

distribution of income and wealth into account (see Fuller et al., 2020,foran

important exception). Assets are an important aspect of people’ssocial

policy preferences because they allow them to privately insure against the

negative consequences of economic shocks such as unemployment. People

with large amounts of assets may be less likely to supportthe welfare state, in

particular social insurance. What matters most, however, is how the ability to

self-insure is distributed across the income spectrum.

In this paper, we argue that wealth inequality—and in particular the degree

to which income and assets holdings are correlated—and the degree of asset

liquidity help explain variation in support for social policies across individuals

as well as variation in the size of the welfare state across countries. At the

micro level, we argue that the joint income–asset distribution yields four

stylized groups with distinct social policy preferences. Most prior research has

focused on the f‌irst two groups: what we call the “economically precarious”

group with low incomes and low assets (which strongly supports the welfare

state) and the “truly wealthy”group with high incomes and high assets (which

strongly opposes it). But the two cross-pressured groups have been largely

overlooked: those with high incomes and low assets (“income buffered”)or

low incomes and high assets (“asset buffered”). These groups are moderately

supportive of the welfare state’s insurance or redistribution function and play

an important role in welfare state politics. For example, high-income people

with few liquid assets have different social policy preferences than high-

income people with lots of private savings since the former are more vul-

nerable to income losses than the latter. Only focusing on income misses this

dimension.

These predictions have macro-level implications for the political coalitions

behind the welfare state. The sizes of the four groups vary considerably across

countries as a function of wealth inequality across income groups. The

correlation between income and assets determines the strength of the political

cleavages between economically precarious and truly wealthy voters. When

assets are unequally distributed across the income spectrum, such that low-

income people have few assets and high-income people have many, pref-

erences in favor of and against social policies are reinforcing and will result in

antagonistic redistributive politics and, ultimately, a smaller welfare state. By

contrast, a weaker correlation between income and assets—and a more

Jensen and Wiedemann 1961