This paper examines how institutional ownership (IO) affects the costs and usages of public versus private debts by borrowing firms. Considering also the cross-monitoring effects by the two types of debts, we find that increasing IO levels significantly lower bond yield spreads for firms that did not secure bank loans within the last year but increase the bank loan spreads for firms not issuing bonds over the same period. The findings indicate that public lenders perceive the presence of institutional investors as improving the conditions of the borrowing firms when they lack cross-monitoring by some existing private lenders; while private lenders are more concerned about the negative effects caused by institutional investors when the firms do not or cannot borrow from the public market. Consistent with how IO affects the bond yield spreads, higher IO firms use more bonds, especially when the firms did not secure bank loans in the previous year. For private debts, higher IO levels only weakly reduce their usages when the firms did not issue bonds but significantly reduce the usages when bonds are issued in the previous year, suggesting that the demand and substitution effects come to affect the private debt usages as well. For debt structure, higher IO firms use relatively more public than private debts. The patterns generally reflect the IO effects on individual debt usages. To address the endogeneity concern, we exploit the discontinuity of institutional ownerships at the boundary between Russell 1000 and 2000.