I. So what?
In a world where gender, race and class are the de rigueur differentiators, it’s too easy to conclude: Structural problems require structural solutions. That proposition is grotesquely premature and lethal to any first-round respect for heterogeneity and complexity.
What the sample below illustrates is that the only place structural solutions thrive is in theory. Which means: The only place really-existing heterogeneities are found is in the many–and many more–empirical analyses like those below.
II. Ten illustrative examples
1. Structural racism
This is the case, for instance, with broad-brush rhetorical attacks on ‘structural racism in criminal justice’ that confuse the different scales of the American penal state (federal, state, county and city), overlook the hyperlocalism and administrative fragmentation of a criminal justice system that is not a system, and amalgamate the different practices of legislating, policing, pretrial detention, prosecution, public defence, plea negotiation and litigation, sentencing, supervising, court-mandated programming, incarceration, and sentence administration, each of which has layers of internal complexity, and may or may not produce looping ethnoracial disparities. . . .[“Structural racism”] replaces meticulous study with facile sloganeering, and pinpoint remedial action with vague calls for systemic changes that are unlikely to come about or to produce their expected results. In so doing, this vogue word betrays its ostensive purpose: to excavate the social conditions of possibility of ethnoracial justice.
https://newleftreview.org/issues/ii133/articles/loic-wacquant-resolving-the-trouble-with-race
2. Class structure
The structural primacy of class is not due to it alone governing people’s material wellbeing, since other social structures do as well, but is based instead on it being endogenously dynamic such that it generates differentiation of interests within and between class and non-class groups. . . .
[T]he structural primacy of class does not necessarily entail its political primacy, in the sense of making the abstract category of “worker” the immediate and exclusive subjective basis for class formation specifically and socialist politics more broadly. On the contrary, under certain conditions ratcheting up class struggle on the vertical dimension, i.e., scaling up class struggle beyond a segment of the working class in a single workplace to the level of politics, may require the mobilization of people on the basis of non- class subjectivities if those subjectivities are most salient conjuncturally. Building cultures of solidarity is indeed essential to class formation. But given a working class that is already highly differentiated, it is forms of solidarity that cut across differentiations within the working class on both the vertical and horizontal levels that are the building blocks for class formation that can move from the economic to the political.
3. Capitalism
a. State capitalism
The heterogeneous literature on the ‘new state capitalism’ has provoked considerable academic and popular interest in recent years, but also critique regarding how to analytically bolster the concept and enhance empirical understanding.
https://journals.sagepub.com/doi/epub/10.1177/0308518X221083986
b. Platform capitalism
With dominant perspectives towards globalized platform power and control, we argue that there has been limited analysis of platform variation. Studies of platform capitalism tend to underplay differences – in the form of governance, practice and agency across space. This leads to risks of undeveloped perspectives of unfolding platform capitalism, especially when studies seek to analyse uneven global relations. . . .
Over-globalized accounts risk underplaying the variations that occur that are essential to how platforms operate and expand across space (Wood et al., 2019). Specifically, as platforms expand across spaces and sectors, it is important to examine how they cohere with existing institutions and regulatory environments within diverse capitalist systems (Thelen, 2018). This is particularly important when analysing platforms, practices and labour outside the economies of the global north (Foster & Bentley, 2022; Panimbang, 2021), with the risk that notions of platform may ” … universalize Western platform capitalism and the ideological and cultural forces that underpin its development” (Davis & Xiao, 2021, p. 104). Global accounts of platform expansion may also overplay the power and control platform leaders actually have across all spaces and sectors. It is important to allow consideration of potential tensions, alternatives, cleavages and agency by which more diverse and inclusive forms of digital capitalism or alternative forms of economy might emerge (Arvidsson, 2019).
4. Heterogeneous income inequalities
Our results suggest that while a positive and statistically significant relationship exists between economic complexity and inequality globally, heterogeneity emerges when analysing countries by income level. Specifically, the relationship is reversed for high-income countries: the more complex the production structure, the lower the income inequality level (as found by Hartmann et al. 2017). In addition, when exploring non-linearity, results suggests a diminishing positive effect of economic complexity on inequality, resembling an inverted U-pattern. As complexity increases, the positive impact on inequality diminishes, but turning negative only at very high levels of complexity. . . .One key conclusion from our analysis is the existence of a heterogeneous relationship between economic complexity and income inequality, which underscores the importance of considering country-specific contexts.
—https://www.sciencedirect.com/science/article/pii/S0313592624002364?via%3Dihub
5. Authoritarian regimes
An important simplification of the discussion presented above is that it treats the authoritarian regime as a homogenous entity, keen on achieving one goal – its self- preservation, maximization of power and rent-seeking. Although this picture of authoritarianism is very widespread in modern scholarship (ironically, very much in the writing of academic economists), in reality, regimes are highly heterogeneous and include groups with different interests and ideas. Some of them may be stronger supporters of particularism and rejection of the West; some may embrace integration of their country into the Western world; and some may see their country as an opponent of the West but at the same time be willing to learn from the West to become more competitive. These positions can be associated with economic and political interests or with firm beliefs.
—https://www.tandfonline.com/doi/full/10.1080/13507486.2025.2526055
6. Automating immigration and asylum controls
Overall, the wide range of applications for new technologies [in migration and asylum governance in Europe] implies that each one should be investigated independently, taking into consideration its development context and the unique requirements of the stakeholders who develop and use them. This report, therefore, debunks a totalising, black-and-white perception of the uses of new technologies. New technologies can be used for various purposes ranging from including migrants’ and refugees’ preferences in their settlement processes (as in the case of some preference matching tools) to profiling them through risk assessments or monitoring them through invasive tools such as electronic monitoring. While the former can benefit migrants by having a say in their migration and settlement trajectory, the latter can have extremely harmful impacts on them. It is, therefore, crucial to examine each use of new technology in its own right, considering its design and implementation processes and their legal and social impacts.
https://www.rsc.ox.ac.uk/publications/automating-immigration-and-asylum-the-uses-of-new-technologies-in-migration-and-asylum-governance-in-europe
7. Household macroeconomic narratives
The large extent of heterogeneity and fragmentation in households’ narratives has important consequences for the formation of economic expectations. Households are not only imperfectly informed about the current state of the economy (Coibion and Gorodnichenko, 2012; Mankiw and Reis, 2002; Reis, 2006) but they also systematically disagree about why the current state has been reached. Heterogeneity in narratives thus contributes to the widely-documented disagreement in macroeconomic expectations (Coibion and Gorodnichenko, 2015a; Dovern et al., 2012; Giglio et al., 2021; Link et al., 2023b; Mankiw et al., 2003). One important question for future research is to better understand the origins of the substantial heterogeneity in household narratives. While differential media exposure is likely to drive some of the heterogeneity, our experiment with exogenous variation in media exposure suggests (as discussed in Appendix 1) that traditional news media is only part of the story. A related open question revolves around the social processes that make some narratives go viral (Graeber et al., 2024c; Shiller, 2017). For instance, narratives involving corporate greed and price gouging are common among households but are neither endorsed by experts nor prominently featured in the news media, suggesting that social interactions are important.
—https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3974993
8. Gig economy
More specifically, I argue that the combination of the technological structure of gig work (nearly automatic, open-access employment, algorithm-driven work process) plus workers’ ability to choose schedules and hours yields an unusually heterogeneous labor force on a range of dimensions, especially patterns of work in other jobs and portfolios of household incomes. As a result, worker experiences are also more heterogeneous than in conventional workplaces.
https://digitalage.berlin/wp-content/uploads/2021/09/Brief-3_Schor_final.pdf
9. COVID-19
a. Preparedness and response
However, these general trends mask significant heterogeneity in responses as countries neither entered nor went through the crisis alike. . . .Overall, the pre-pandemic global outlook was heterogeneous across different geographies. . .
https://www.esm.europa.eu/publications/regional-responses-covid-19-crisis-comparative-study-economic-policy-and-institutiona
b. Worker protests
Finally, the dataset reveals significant variation between countries and regions. Political, economic, and institutional contexts clearly matter in shaping patterns of protest. Nevertheless, over-generalisation about the role of national institutional factors should be resisted, given the huge differences we found within countries. For example, a comparatively large volume of protest were identified in healthcare in India but very little in retail, and the same can be said of Nigeria; we therefore examine reports from these country cases in more detail below. Moreover, among the handful of countries reporting no protests, there is no consistent economic or institutional profile. Hence, we suggest that spikes in protest in particular sectors and countries are likely to reflect not only the national institutional context, but also contingent factors and strategic decisions made by the actors involved. To illustrate this point, we examine in more detail reports from the five countries with the highest levels of protest in the two sectors: France, India and Nigeria for healthcare, and the United States and Argentina for retail.
https://www.ilo.org/wcmsp5/groups/public/—dgreports/—inst/documents/publication/wcms_860587.pdf
10. Chinese development finance [CDF] to Africa
We contend that the impact of CDF on policy space cannot be generalised as either wholly expansive or restrictive. Instead, policy space is shaped more significantly by the internal political economies of recipient states, historical legacies, their global economic positioning and the interplay of these elements with the external financing environment. We find that, for example, countries with strong state capacity may strategically leverage CDF to implement incremental policy innovations, while more aid-dependent states often face greater constraints. Notwithstanding these general reflections, our framework and findings challenge uniform assumptions about CDF’s influence on policy space: neither divergence nor convergence between Northern and Chinese approaches results in wholesale shifts in policy autonomy. Rather, the development financing landscape, characterised by heterogeneity and ambiguity, creates opportunities for modest but meaningful policy experimentation. However, these manifest in an indeterminate way.
—-
NB. In some cases, footnote/endnote numbers have been deleted for ease of reading.
WHEN COMPLEX IS AS SIMPLE AS IT GETS 
12 thoughts on “Instead of “differentiated by gender, race and class,” why not “differentiated by heterogeneity and complexity”? Ten examples of racism, class, capitalism, inequalities, border controls, authoritarianism, COVID and more”