Socioeconomic position (SEP) is a potential correlate of sedentary behavior in adolescents. The aim of this study was to examine the associations between SEP and self-reported and objective measures of sedentary behavior in adolescents, using a life-course approach. Data from the 1993 Pelotas (Brazil) Birth Cohort Study were analyzed (N = 5249). Cross-sectional and longitudinal associations between multiple SEP indicators (maternal education, family income, SEP composite, cumulative family income) at birth, 11, 15 and 18 years, and five sedentary behavior outcomes (≥ 4 h/day screen time; ≥ 4 h/day TV; ≥ 2 h/day computer; ≥ 2 h/day video game; ≥ 12.7 h/day objectively measured sedentary time) at 11, 15 and 18 years, were examined. In cross-sectional analyses, higher SEP was positively associated with more screen time at ages 11 and 15 years. There was a consistent and positive association between higher SEP with time spent using a computer, and with sedentary time assessed through accelerometry. SEP at birth had a positive and direct effect on screen, computer and total sedentary time at 18 years. Participants in the highest cumulative income group had higher odds of high sedentary behavior in screen (OR: 2.40; 95% CI: 1.50–3.54), computer (OR: 7.35; 95% CI: 4.19–12.89) and total sedentary time (OR: 5.40; 95% CI: 3.53–10.35), respectively, compared with their counterparts with lower cumulative income. Our findings showed that SEP is an early determinant of sedentary behavior in adolescents.