How Is Racism Structured in the Present Day?

Present-day racism is not limited to personal prejudice. It is also embedded in rules, budgets, professional routines, algorithms and markets that distribute security, income, health and political power. A policy can reproduce racial inequality without naming race or requiring every participant to hold racist beliefs.

The structure has three layers. Historical allocation includes conquest, slavery, segregation, dispossession and immigration controls that determined who could own land, accumulate wealth or receive state protection. Institutional reproduction occurs when schools, employers, lenders, hospitals and courts build decisions on those unequal starting conditions. Individual enforcement includes harassment, exclusion and violence, whether intentional or enabled by institutional impunity.

Sociologist Eduardo Bonilla-Silva described this adaptation after formal desegregation:

“The crux of the problem is that the main social practices and mechanisms behind contemporary racial disparity are not overtly racial.” — Eduardo Bonilla-Silva, 2018, Racism Without Racists, fifth edition.

This distinguishes systemic racism from a theory claiming that every disparity has one cause. Age, class, disability, gender, geography and migration status interact with racial classification. Researchers therefore compare similarly situated groups, audit decision-makers and trace mechanisms rather than treating race as biological destiny.

Read the archive’s linked explanations of modern racism, ongoing exploitation and the relationship between racial ideology and present-day institutions.

The strongest evidence combines outcomes with direct tests. In US employment field experiments synthesised by Lincoln Quillian and colleagues in 2017, white applicants received 36% more callbacks than African American applicants and 24% more than Latino applicants; the meta-analysis covered 24 studies conducted from 1989 to 2015, and found no decline in anti-Black hiring discrimination over that period.

Housing carries earlier policy into present wealth. The US Federal Reserve’s 2022 Survey of Consumer Finances, published in 2023, estimated median family wealth at $285,000 for non-Hispanic white families, $44,900 for Black families and $61,600 for Hispanic families—ratios of about 6.3:1 and 4.6:1, respectively. These figures are medians, not estimates that race itself mechanically causes wealth.

Criminal justice shows unequal exposure to state coercion. Mapping Police Violence counted 1,247 people killed by US police in 2023; its 2023 population-adjusted rates were 5.9 per million for Black people and 2.3 per million for white people, approximately 2.6 times as high for Black residents. Definitions and incomplete official reporting make exact totals source-dependent.

In health, US maternal mortality records for 2022 showed 49.5 deaths per 100,000 live births among non-Hispanic Black women, against 19.0 among non-Hispanic white women, according to the National Center for Health Statistics in 2024: a 2.6-fold difference. Clinical access, environmental exposure, income, stress and unequal treatment contribute; racial categories do not identify a genetic cause.

There is no credible single estimate of “how much racism” exists. Scholars quantify separate outcomes, using methods with different strengths.

Domain and source	Data year or study period	Estimate	What it establishes
Federal Reserve, Survey of Consumer Finances	2022	Median wealth: white $285,000; Black $44,900; Hispanic $61,600	Scale of wealth inequality, not one exclusive cause
Quillian et al., hiring meta-analysis	1989–2015	White callback advantage: 36% over Black, 24% over Latino applicants	Direct differential treatment in matched field tests
NCHS maternal mortality	2022	Black 49.5, white 19.0 deaths per 100,000 births	Population disparity requiring causal investigation
Mapping Police Violence	2023	Black 5.9, white 2.3 deaths per million	Unequal population-level risk under its inclusion rules
Williams Institute	2020–2022	1.2 million US LGBT adults identified as nonbinary; 42% were people of color	Intersectional population estimate, not a racism rate

The table separates discrimination estimates, which test differential treatment, from disparity estimates, which measure unequal outcomes. Statistical adjustment can test proposed explanations, but adjustment is not neutral when variables such as neighborhood, income or criminal record are themselves shaped by earlier discrimination.

International comparisons are difficult because states classify race differently. France generally restricts official race statistics; Brazil’s census uses self-identified color categories; the United Kingdom records multiple ethnic groups. Those systems reveal different parts of modern racism and can conceal others.

Critics dispute the concept on definitional, causal and political grounds. Some argue that “racism” should mean intentional hostility or explicitly unequal law; under that definition, race-neutral institutions cannot themselves be racist. Others attribute disparities primarily to class, family structure, crime patterns, geography or individual behavior. Methodological critics correctly note that administrative categories are imperfect, observational studies face confounding and police-death databases use differing inclusion criteria.

A narrower legal definition appears in Chief Justice John Roberts’s opinion in Parents Involved in Community Schools v. Seattle School District No. 1:

“The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.” — Chief Justice John G. Roberts Jr., 2007, plurality opinion.

That color-blind position is disputed because formally identical treatment can preserve unequal conditions created by state action. US redlining, for example, involved federal underwriting and local segregation; removing explicit racial maps did not redistribute the property appreciation, school funding or environmental safety accumulated under them.

Political disputes also concern remedies. Opponents of affirmative action, reparations or targeted public-health programs may accept historical racism while rejecting race-conscious correction. In 2023, the US Supreme Court’s decisions in Students for Fair Admissions v. Harvard and Students for Fair Admissions v. University of North Carolina restricted race-conscious university admissions under Title VI and the Fourteenth Amendment. Disagreement over remedies does not erase the measured hiring, wealth, policing or health gaps; it changes how their causes and constitutional significance are interpreted.

Effective analysis asks who designed a rule, what inputs it uses, whose interests it protects and whether its effects persist across generations. Institutions should publish disaggregated data, commission matched-pair audits, assess algorithms for differential error rates and preserve records that permit independent scrutiny. Data collection must include privacy safeguards and avoid treating administrative race categories as biological facts.

Policy responses differ by mechanism. Hiring discrimination supports structured recruitment and enforcement. Segregated housing and inherited wealth require fair-lending enforcement, affordable housing, tax reform and forms of restitution. Unequal maternal mortality requires access to care alongside investigation of clinical treatment and environmental risk. Police disparities require reliable death reporting, enforceable use-of-force rules and accountability independent of the agencies involved.

Historical repair is not identical to generic poverty policy. Universal programs can reduce deprivation, but they may leave racial asset gaps intact when benefits track existing property, employment or citizenship. Targeted remedies require clear eligibility, democratic authorization and periodic evaluation.

The central conclusion is empirical rather than rhetorical: present-day racism is structured when institutions repeatedly convert inherited racial hierarchy into unequal exposure to opportunity, surveillance, illness and premature death. The relevant question is not only whether an official expresses prejudice, but whether a system predictably distributes benefits and harms along racialized lines. See ongoing exploitation for the international economic dimension.

• Federal Reserve: Changes in U.S. Family Finances from 2019 to 2022
• Quillian et al.: Meta-analysis of field experiments on racial discrimination in hiring
• National Center for Health Statistics: Maternal Mortality Rates in the United States, 2022
• Mapping Police Violence: 2023 Police Violence Report
• National Academies: Communities in Action—Pathways to Health Equity
• OHCHR: International Convention on the Elimination of All Forms of Racial Discrimination