Objective: To determine the influence of socioeconomic factors on disease activity in a Latin American (LA) early rheumatoid arthritis (RA) multinational inception cohort at baseline.Methods: Clinical evaluation, ethnicity, socioeconomic status (SES), 4‐variable Disease Activity Score in 28 joints using the erythrocyte sedimentation rate (DAS28‐ESR), Health Assessment Questionnaire (HAQ) disability index (DI), and erosions were recorded in 1,093 patients with early RA (<1 year from onset). Multivariate analyses evaluated influences of sex, age, marital status, education, medical coverage, SES, and ethnicity on HAQ DI, DAS28‐ESR, and presence of erosions.Results: Ethnicities included 43% Mestizo, 31% Caucasian, 19% African LA, 4% Amerindian, and 3% other. Fifty‐eight percent were of low/low‐middle SES, 42% had <8 years of education, 21% had no medical coverage, median disease duration was 6 months (25th, 75th percentiles 4, 9 months), median HAQ DI score was 1.25 (25th, 75th percentiles 0.63, 2.00), median DAS28‐ESR score was 6.2 (25th, 75th percentiles 4.9, 7.2), and 25% had erosions. Women and Mestizos, African LA, and Amerindians had earlier onset than men or Caucasians (P < 0.01). When adjusted by country, the analysis of covariance model showed that low/low‐middle SES, female sex, partial coverage, and older age were associated with worse HAQ DI scores; only low/low‐middle SES was associated with higher DAS28 scores. Statistically significant differences were found in HAQ DI and DAS28 scores between countries. When excluding country, low/low‐middle SES, female sex, and no coverage were associated with worse HAQ DI and DAS28 scores, whereas separated/divorced/widowed status was associated with worse HAQ DI scores and age was associated with worse DAS28 scores. Logistic regression showed that older age, no coverage, and the Amerindian and other ethnic groups were associated with erosions.Conclusion: We compared early RA patients from the main LA ethnic groups. Our findings suggest that low/low‐middle SES is important in determining disease activity. A more genetic‐related background for erosions is possible.