Corn-soybean rotations are common in the Corn Belt. Farmers typically broadcast removal-based P and K fertilizer rates for both crops before corn and some apply additional starter fertilizer when planting corn. Research comparing broadcast and starter fertilization for corn seldom addresses effects on soil-test values and residual effects on soybean yield. Fifteen trials were conducted in Iowa with this objective. Fertilization treatments only for corn were a control (NF), in-furrow starter at 5-7 kg P ha^-1 and 10-14 kg K ha^-1 (SF), broadcast P and K at 49-66 kg P ha^-1 and 112 kg K ha^-1 (BF), and SF plus BF. Fertilization increased corn yield when soil-test P (STP) was < 21 mg P kg^-1 or soil-test K (STK) was < 171 mg K kg^-1. The SF and BF treatments differed at five sites testing < 17 mg P kg^-1 (values lower or borderline between the Low and Optimum interpretation classes), with BF increasing yield more than SF (0.30 to 1.48 Mg ha^-1). Applying SF in addition to BF never increased corn yield more than BF alone. Starter fertilization did not change STP or STK after corn harvest compared with NF or initial values, while BF either maintained or increased soil-test values. Soybean responded to residual fertilization at seven sites. Soybean yield was higher for all residual fertilization treatments than for NF at six sites, and at three sites yield for BF or BF plus SF were higher than for SF alone. Unexpectedly, a small amount of SF for corn often increased yield of a following soybean crop even when STP and STK were not changed. Results of fertilization strategies for the 2-year rotations showed that BF was the best option for low-testing soils but not necessarily for soils testing Optimum, for which maintenance BF currently is recommended.