Energy intake recommendations for infants are based on data from industrialized countries. FAO/WHO/UNU expressed the need for studies on total energy expenditure (TEE) and basal metabolic rate from developing countries covering current and changing lifestyles. For this observational study, 65 infants of differing socioeconomic status (SES) (n = 32 middle SES, n = 33 low SES) were selected in Pelotas, southern Brazil, aiming to: 1) compare TEE, minimum observable energy expenditure (MOEE), activity energy expenditure (AEE) between breast-fed infants 8.7 mo of age from middle and low SES; and 2) investigate the effect of potential mediating factors on TEE and AEE. TEE and total body water were measured with doubly labeled water, MOEE with respiration calorimetry, breast milk intake using the dose-to-the-mother deuterium-oxide turnover method, food intake using 1-d food weighing, and prevalence of overweight using BMI Z-scores. TEE adjusted for ethnicity was 257 (95% CI 232-281) kJ/(kg . d) in middle SES infants vs. 318 (95%CI 294-342) kJ/(kg . d) in low SES infants (P = 0.001). MOEE did not differ between groups and the difference in TEE was therefore attributed to AEE (P = 0.008). The effect of SES on AEE was mediated by the number of persons per bedroom (crowding). Prevalence of overweight tended to be higher in middle SES infants (P = 0.054) than in low SES infants. The difference in TEE and AEE between SES groups emphasizes the importance of an accurate description of the SES of any population in which TEE is studied and questions the extent to which TEE data from middle-class infants in transitional countries should be considered normative.