Despite global efforts to improve maternal health, many developing countries including Pakistan have failed to achieve the target of a 75% reduction in maternal deaths by 2015. Addressing socioeconomic inequalities in access to emergency obstetric care is crucial for reducing the maternal mortality rate. This study was done to examine the time trends and socioeconomic inequalities in the utilization of cesarean section (C-section) in Pakistan during 1990–2013. We used data from the Pakistan Demographic and Health Surveys (PDHS) conducted during 1990 to 2013. All these surveys are nationally representative surveys of ever-married women aged 15–49 years with a sample size of 6611, 10,023, and 13,558 women in 1990–1991, 2006–2007, and 2012–2013, respectively, with an overall response rate of over 90%. The unit of analysis for this study was women with their most recent live birth in the five years preceding the surveys. Bivariate analyses and multivariable logistic regression models were employed to investigate the prevalence of cesarean sections according to selected sociodemographic characteristics of women. C-section rates were found to have increased during this period, with an especially significant rise from 2.7% in 1990–1991 to 15.8% in 2012–2013 with lower utilization among the non-educated women (7.5%), compared with the women with higher education (40.3%). C-section rates ranged from 5.5% in the poorest women to 35.3% in the richest women. Only 11.5% of the rural women had a C-section compared to 25.6% of the urban women. A greater likelihood of having a cesarean section was observed in the richest, highly educated, and urban-living women while there was no significant difference observed in cesarean section rates between the private and public sectors in all three surveys. To improve maternal health, routine monitoring and evaluation of the provision of emergency obstetric services are needed to address the underuse of C-section in poor and rural areas and overuse in rich and urban areas.