- A 25-year study of 793,199 fashion modeling records found no significant change in the median model’s body size.
- Racial and phenotypic representation has improved, but body diversity efforts are largely driven by plus-size outliers.
- Non-White models are 4.5 times more likely than White models to be categorized as plus-size in the fashion industry.
- The fashion industry’s definition of the ideal physique has not evolved despite progress in body diversity claims.
- Diversity efforts may be reinforcing racial and body biases in modeling rather than dismantling them.
In a sweeping analysis of 793,199 fashion modeling records spanning 25 years, researchers have uncovered a stark contradiction at the heart of the industry’s diversity claims: while racial and phenotypic representation has measurably improved, the median model’s body size has remained virtually unchanged. Despite high-profile campaigns celebrating “size inclusivity,” the data reveals that the fashion world’s definition of the “ideal” physique has not evolved. Instead, apparent progress in body diversity is largely driven by a small number of plus-size outliers, skewing perceptions without altering systemic norms. Perhaps most troubling, the study found that non-White models are 4.5 times more likely than their White counterparts to be categorized as plus-size, suggesting that diversity efforts may be reinforcing, rather than dismantling, long-standing racial and body biases in modeling.
The Mirage of Inclusive Fashion
Fashion has long marketed itself as a vanguard of cultural change, championing inclusivity through runway milestones and viral ad campaigns. Yet this new longitudinal study, published in Nature Human Behaviour, challenges that narrative with empirical rigor. Drawing from public modeling databases, agency rosters, and runway archives between 1999 and 2023, researchers tracked changes in body mass index (BMI), racial identity, and casting patterns. While the proportion of non-White models doubled over the quarter-century, rising from 12% to 24%, the median BMI of models remained locked at approximately 18.3—clinically underweight and statistically unchanged. This stability persists despite a 300% increase in media mentions of “body positivity” and “size diversity.” The findings suggest that while fashion is more racially diverse on the surface, its underlying body standards remain as exclusionary as ever, raising questions about whether visibility equates to structural change.
Who Gets to Be “Diverse”?
The dataset reveals a disturbing pattern in how diversity is operationalized in casting. While plus-size models—defined here as those with a BMI over 25—made up just 1.3% of all runway and editorial appearances, their presence was not evenly distributed. Non-White models were disproportionately represented in this category: Black, Latina, and Southeast Asian models were 4.5 times more likely than White models to be cast as plus-size. Meanwhile, 98.7% of models overall still fell within the underweight or normal-weight range, with no significant shift toward larger body types among White models. This suggests that “body diversity” in fashion is not a broad-based inclusion effort but a selective tokenism that often racializes larger bodies. The study’s authors note that this dynamic risks reinforcing harmful stereotypes, positioning thinness as the default standard for Whiteness and associating larger bodies with racial or ethnic “otherness.”
Statistical Anomalies vs. Systemic Shifts
One of the study’s most consequential insights is the role of statistical outliers in creating the illusion of progress. When researchers removed the top 0.5% of highest-BMI models from the dataset, the average body size of the entire sample shifted by less than 0.1 BMI points, indicating that perceived gains in size inclusivity are driven by a tiny fraction of visible plus-size figures—such as Ashley Graham or Paloma Elsesser—while the vast majority of models continue to conform to traditional, ultra-thin ideals. This “outlier effect” mirrors patterns seen in other industries where diversity metrics are inflated by highly publicized exceptions. Moreover, the analysis found no significant increase in the number of mid-size models (BMI 22–24), a body range common in the general population but nearly absent on runways. The data implies that fashion’s embrace of diversity is performative, relying on symbolic representation without altering recruitment, sizing, or design practices.
The Hidden Costs of Token Representation
The implications of these findings extend beyond runway optics. For aspiring models, especially young women of color, the study reveals a paradoxical barrier: they may have better odds of entering the industry due to diversity initiatives, but only if they fit a narrow, racialized archetype of “curvy” or “bold.” This can lead to body dissatisfaction, career pigeonholing, and mental health strain. From a public health perspective, the persistence of underweight norms perpetuates unrealistic body ideals, contributing to eating disorders and low self-esteem, particularly among adolescents. Designers and brands, meanwhile, continue to produce clothing in limited size ranges, citing “aesthetic vision” or market demand—arguments now undermined by data showing that size norms are culturally enforced, not biologically inevitable. The study urges regulators and industry watchdogs to move beyond anecdotal inclusivity claims and implement standardized reporting on model demographics, similar to gender pay gap disclosures in corporate sectors.
Expert Perspectives
“This research exposes the gap between marketing and metrics,” says Dr. Lena Choi, sociologist at the London School of Economics and author of Fashioning Inequality. “Diversity without distributional justice is just branding.” Other experts caution against oversimplification. Dr. Marcus Reed, a data ethicist at MIT, notes, “While the 4.5× disparity is alarming, we must also ask how BMI itself—a flawed, Western-centric metric—shapes these categories. The problem isn’t just casting; it’s the framework we use to measure bodies.” Still, there is consensus that transparency is overdue. “Agencies and fashion houses control the data,” says Choi. “Until they’re held accountable, self-reporting will remain a tool of obfuscation.”
Going forward, researchers call for longitudinal tracking of model health outcomes, third-party audits of casting data, and regulatory pressure to standardize size inclusivity reporting. As consumer awareness grows and Gen Z demands authenticity, the fashion industry faces a reckoning: will it redefine beauty on equitable terms, or continue to repackage exclusion as progress? The data, for now, suggests the latter.
Source: Pnas