The Wealth Inequality of Nations. Fabian T. Pfeffer, Nora Waitkus. American Sociological Review, July 30, 2021. https://doi.org/10.1177/00031224211027800
Abstract: Comparative research on income inequality has produced several frameworks to study the institutional determinants of income stratification. In contrast, no such framework and much less empirical evidence exist to explain cross-national differences in wealth inequality. This situation is particularly lamentable as cross-national patterns of inequality in wealth diverge sharply from those in income. We seek to pave the way for new explanations of cross-national differences in wealth inequality by tracing them to the influence of different wealth components. Drawing on the literatures on financialization and housing, we argue that housing equity should be the central building block of the comparative analysis of wealth inequality. Using harmonized data on 15 countries included in the Luxembourg Wealth Study (LWS), we demonstrate a lack of association between national levels of income and wealth inequality and concentration. Using decomposition approaches, we then estimate the degree to which national levels of wealth inequality and concentration relate to cross-national differences in wealth portfolios and the distribution of specific asset components. Considering the role of housing equity, financial assets, non-housing real assets, and non-housing debt, we show that cross-national variation in wealth inequality and concentration is centrally determined by the distribution of housing equity.
Keywords: wealth, income, housing, inequality, comparison
Conclusion
While advanced capitalist societies are marked by high levels of inequality in household
wealth as well as concentration of wealth in the hands of a few, considerable variation
exists in the extent of national levels of wealth inequality and concentration. Yet, current
knowledge about national patterns and determinants of wealth inequality is limited and,
as we have argued here, will rely on fundamentally different explanatory approaches than
those developed over decades in a laborious field of research on international differences in
income inequality. International differences in income inequality tell us close to nothing about
international differences in wealth inequality, as we have shown here. In fact, many countries
that we customarily describe as comparatively egalitarian using income-based comparisons
– such as Scandinavian countries – can be classified as anything but in terms of their levels
of wealth inequality. Many countries that were henceforth thought of as similarly unequal –
for instance, Germany and Greece – are far apart from each other in terms of their level of
wealth inequality (with Germany displaying very high levels). As such, prior institutional
explanations of inequality hold little promise in elucidating the international ranking of
wealth inequality and the vast cross-national variation in wealth stratification remains in
urgent need of explanation.
This contribution takes but one first step in this direction by carefully investigating the
role of different asset components in accounting for the overall distribution of wealth. We
surmise that any potential institutional explanations of wealth inequality need to rest on a
careful consideration of the operative components of wealth. That is, we first need a clear
understanding of how the distribution of different types of assets relates to nations’ overall
level of wealth inequality and concentration. Is wealth inequality, for instance, largely a
reflection of the spread of debt, financial liabilities, and general exposure to financial markets,
as emerging theories of financialization may suggest? Or, do we best understand the degree
of wealth concentration in a given country as the concentration of capital held in real assets,
reflected, for instance, in the hoarding of wealth among a business elite? Our empirical
findings, instead, consistently point in a different direction: Cross-national differences in
wealth inequality and concentration chiefly reflect the level of inequality in and concentration
of housing equity. While simple indicators of home ownership rates, typically used to capture
the overall importance of housing assets in a given country, suggest that broader access to
home ownership may dampen wealth inequality and concentration, the overall distribution
of housing equity, of which the prevalence of home ownership is just one aspect, is the central
element accounting for overall wealth inequality. A country’s distribution of housing equity
explains its overall level of wealth inequality and concentration to a substantial degree,
including both the outlying position of the United States as well as the overall variation
across many different countries. This is not to say that the strong concentration of financial
assets and business equity at the top of the wealth distribution in most countries would be
unimportant. In fact, a focus on financial assets and business equity is likely central to the
understanding of elite closure and the continued and accelerating wealth accumulation of
the top one percent (Piketty 2014; Carney and Nason 2018). But, based on the evidence
presented here, our understanding of wealth inequality among the remaining 99 percent relies
on increased attention to the structure and dynamics of housing and mortgage markets.
Our two main findings – the non-correlation of income inequality and wealth inequality,
on the one side, and the centrality of housing equity, on the other side – are thus connected:
The reason why cross-national differences in income inequality do not predict cross-national
differences in wealth inequality is that the latter are most centrally driven by housing equity.
In turn, the distribution of housing equity, we argue, is crucially determined by financialization
and housing market dynamics, i.e., in institutional spheres outside of the labor market
and the classical realms of the welfare state. Work on comparative stratification and welfare
state regimes, therefore, will have to expand its view to these additional institutional factors
to make sense of a dimension of particularly profound and lasting inequality. Ideally, such
future work will draw on both qualitative and quantitative indicators of financialized housing
markets, such as housing and mortgage market regulations.
It seems unfortunate that one of the most ambitious theoretical and empirical studies on
the determinants of wealth inequality, Piketty’s Capital (2014), also mostly disregards the
role of housing as a driver of wealth inequality (see also Bonnet et al. 2014; Fuller et al. 2019;
Rognlie 2015), and the proposed “rule” of growing wealth inequality (r > g) at best discounts
the importance of a careful analysis of the institutional determinants of wealth inequality
(see also Acemoglu and Robinson 2015). An alternative, theoretically ambitious effort that
focuses on the role of housing may, instead, naturally align with the rapidly expanding
literature on financialization that has forcefully argued for the central role of mortgage
lending. At the backdrop of the findings presented here, one way to bring the literature on
financialization and the literature on wealth into closer conversation would be to establish a
clear empirical link between different lending regimes and the structure of national housing
markets. Doing so would also promise to ameliorate the surprising disconnect between the
scholarships on wealth and debt (see also Dwyer 2018). The comparative study of lending
regimes is at an early stage but has produced some interesting initial insights: For instance,
in a comparison of the mortgage debt structure in six European countries, van Gunten and
Navot (2018) show that differences in the distribution of mortgage debt is best captured by
the degree of credit intensity, i.e., the expansion of credit among those already holding it,
rather than differences in mortgage market participation (which also makes the distribution
of mortgage credit largely independent from national home ownership rates). This pattern
chimes well with our finding of the dominant role of the distribution of housing equity,
rather than home ownership rates, in explaining overall wealth inequality. However, in
the U.S., mortgage debt has also expanded into new population groups as the “predatory
inclusion” of minority households grew through new and exploitative mortgage products
(Rugh and Massey 2010; Taylor 2019). Future research should thus expand its comparative
range to understand different modes of housing market financialization (see also Blackwell
and Kohl 2018). Some of this research may also pursue a meso-level approach, popular in
some financialization studies, to compare the role of banks and asset management firms, the
real estate industry, or other intermediaries involved in expanding and intensifying mortgage
credit (Baradaran 2017; Jorda et al. 2016; Taylor 2019; Braun 2020).
To pursue an explanatory agenda, comparative wealth research will also be able to fruitfully
draw on research on recent housing markets dynamics. For instance, Adkins et al.
(2020) proposes property price inflation as the foundation of a new logic of inequality: Having
access to home ownership in areas experiencing such inflation determines individuals’
economic well-being over and above their employment. The extent to which homes out-earn
the individuals who own them, of course, also varies vastly within countries. Geographic
polarization of home ownership and housing prices has been documented in several countries
(e.g., Levin and Pryce 2011; Baldenius et al. 2020), in some taking the shape of run-away
home values in “superstar” cities, where transnational wealth elites store and invest vast
fortunes and drive up home prices in the process (Fernandez et al. 2016). Outside of these
zones of wealth storage and accumulation, asset prices are depressed and yield lower wealth
returns, for instance, in U.S. minority neighborhoods (Killewald and Bryan 2016; LaBriola
2020). Future research may seek to relate national-level wealth inequality and concentration
to regional and other spatial inequalities within countries. Recent contributions that
have pursued similar questions in the context of the income distribution in the U.S. have
shown that national-level trends in income inequality are the main driver of regional income
inequality (Manduca 2019) and that the distribution of income across and within U.S. geographies
has large, causal effects on the economic well-being of the next generation (Chetty
and Hendren 2018). If the variation in local housing markets is at least as large as that in
local labor markets, one may hypothesize that geographic variation in wealth levels and inequality
may be even more pronounced and consequential for the distribution of opportunity
among the next generation. For most nations, this vital analysis of within-country variation
in wealth levels, inequality, and persistence, however, awaits the development of a new data
data infrastructure to assess the distribution of wealth at the sub-national level, for instance,
based on full-population tax data or other administrative records. Finally, complementary
to a focus on recent housing market dynamics, a comparative-historical approach to uncover
the institutional foundations of countries’ housing and mortgage markets can draw on recent
work that not only documents high long-term wealth returns on housing (Jorda et al. 2019;
Blackwell and Kohl 2019) but also great cross-national variation in housing price trajectories
(Knoll et al. 2017). We remind the reader that our data are chiefly drawn from the period
following the Great Recession. And although our stability analyses based on immediate
pre-recession measures for a few countries suggest that our main conclusions are stable, we
believe that the cross-national variation in the impact of the housing crisis provides new
analytic opportunities.
We believe that future wealth research stands to learn a lot from a focus on countries at
either end of the international ranking of wealth inequality. As some of the most wealthegalitarian
countries in our analysis, post-socialist nations and their radical shift in home
ownership regulations during market transition provide promising analytic opportunities
(Marcuse 1996; Zavisca 2008; Tsenkova 2017; Song and Xie 2014; Xie and Jin 2015). At the
same time, we expect our results to trigger additional interest in analyzing countries with
the highest level of wealth inequality and concentration. Likely, the unfortunate leadership
position of the U.S. in the international ranking of wealth inequality will not come as a
surprise to most comparative stratification scholars; the degree to which the U.S. outranks
its peer countries in terms of wealth concentration may. We have gone to great lengths
to rule out that the high wealth concentration estimate for the U.S. is simply a product
of (putatively) superior data quality. It is also not exclusively a reflection of deep racial
inequalities in wealth; even among white U.S. households the level of wealth concentration
is exceptional in comparative perspective. The next two most wealth-unequal countries in
our analysis, Sweden and Norway, in contrast may cause more surprise and critique – even
though we are not the first to document high wealth levels for these countries (e.g., Roine
and Waldenstroem 2009; Jaentti et al. 2013). After all, comparative stratification research
has long and rightfully held up Scandinavia as the egalitarian poster-child based on its
national income distributions. The analysis of wealth considerably complicates this image
and invites scholars to revisit the assessment of Scandinavian egalitarianism. High wealth
stratification in Scandinavian countries may well be a long-term reflection of its much less
egalitarian history (see e.g., Piketty 2020) as well as the more recent neo-liberal turn in their
politics (Fagerberg et al. 1990; Ryner 1999). Critics may still wonder whether high wealth
inequality takes on fundamentally different social significance in a context with comparatively
generous systems of public insurance that may make wealth less central to maintaining more
stable lives. In contrast, we submit that wealth inequality in such contexts is still highly
consequential for a range of outcomes, in particular, for the intergenerational reproduction
of inequality: Recent contributions have highlighted the independent role of wealth in the
distribution of educational opportunity and the intergenerational transmission of advantage
in Sweden and Norway (Haellsten and Pfeffer 2017; Adermon et al. 2018; Hansen 2014;
Galster and Wessel 2019).
At the same time, concerns about the public insurance context of different wealth inequality
regimes do point to an important area for future research: As acknowledged before,
the inclusion of (estimated present values of) public pension entitlements is certain to provide
lower estimates of inequality in Scandinavia and other contexts. We have pointed out
that our analysis, in line with most other wealth research, applies a definition of net worth
that does not include public pensions nor most other forms of employer-provided pensions.
We have focused on assets available to working-age households. Unlike the marketable assets
included in our analyses, pension wealth is inaccessible (to varying degrees depending on the
type of pension) to households until older ages. Measures of wealth that include the present
values of pensions, i.e. “augmented net worth,” thus shift the analytic question.13 Although
harmonized measures of augmented net worth will be enormously difficult to construct for
a broad range of countries given cross-national differences in pension systems, future comparative
studies of augmented net worth inequality may provide a different country ranking.
Institutional explanations of such ranking will likely also profit from direct connections to
the literatures reviewed here as the financialization of pension systems complements that of
housing markets (Dixon 2008; Schwartz 2012; van Gunten and Kohl 2020).
Finally, we are convinced that the analysis of wealth inequality stands to gain from
future expansion of its comparative scope to other national contexts (see also Davies 2008).
As typical of most “medium-N” and “large-N” cross-national comparisons, our sample of
countries is a reflection of data availability, which in turn is based on various historical and
political contingencies that prohibit inference to other countries (see Ebbinghaus 2005). In
this sense, we provide an initial descriptive approach that awaits expansion to other countries
as the availability of LWS and other wealth data continues to expand (see Killewald et al.
2017; Zucman 2019). The findings reported here may also facilitate the meaningful selection
of a smaller number of comparative cases (Ebbinghaus 2005) that, in a “small-N” comparison,
would help elucidate the institutional foundations of distinct housing markets and their
relationship to overall wealth. The inability to draw firm causal conclusions based on either
type of comparative approach should not keep us from taking the next significant step in
filling the lacuna of evidence on the potential sources of national levels of wealth inequality.
Posts
No comments:
Post a Comment