Use of adequate effective population size in maize (Zea mays L.) recurrent selection programs is important because of random genetic drift and inbreeding depression. The objectives of this study were to (i) evaluate the performance of the BS11 Cycle 0 (C0) and the BS11 Cycle 5 (C5) populations from four S₁-progeny selection programs each with a different effective population size (5, 10, 20, or 30) but with a common selection intensity of 20%, and (ii) compare the additive genetic variance among the C0 and C5 populations. Five cycles of selection were conducted by intermating 5, 10, 20, or 30 lines. One hundred thirty C5 S₁ lines from each of the selected populations (i.e., C5–5, C5–10, C5–20, and C5–30) and 100 C0 S₁lines were topcrossed to BS11 C0. The resulting half-sib progenies were evaluated at five environments in a replications-within-sets incomplete block design. The four selection programs resulted in significant increases in grain yield, reduced grain moisture, and reduced root and stalk lodging. For yield, the 10-S₁ program showed the highest gain cycle⁻¹ of 0.16 Mg ha⁻¹ followed by the 30-S₁ program with 0.13 Mg ha⁻¹ cycle⁻¹ The 5-S₁ program had a higher gain cycle⁻¹ than the 20-S₁ program. The additive genetic variance for yield did not change significantly. Heritability for yield was highest for C5–20, but no significant differences were observed among populations. These results suggest little to no advantage of using larger effective population sizes to maintain genetic variability for short-term recurrent selection.