Firms do not historically call their convertible bonds as soon as conversion can be forced.A number of explanations for the delay rely on the size of the dividends that bondholders forgo so long as they do not convert. We investigate an important change in convertible security design, namely, dividend protection of convertible bond issues. Dividend protection means that the conversion value of the convertible bond is unaffected by dividend payments and that dividend-related rationales for call delay become moot. We document that call delay is near zero for dividend-protected convertible bonds. 2 This paper documents the rapid rise of callable convertibles that are dividend-protected, and the ensuing effect on convertible bond call policy. When a convertible is dividendprotected and a dividend is distributed, the number of shares to be received upon conversion increases so as to leave the bond's conversion value unchanged. We find that call delay is near zero for dividend-protected convertible bonds. The link between dividend protection and diminished call delay highlights the importance of dividends for understanding convertible bond call policy.Calling a convertible bond will force conversion provided the conversion value exceeds the call price. Ingersoll (1977a) and Brennan and Schwartz (1977) model settings in which shareholder wealth is maximized by calling to force conversion whenever feasible. Forced conversion deprives convertible bondholders of the combined value of the insurance they have as bondholders able to demand the bond's principal value at maturity (rather than the bond's conversion value at maturity) and any income advantage from the coupons they receive in excess of the dividends received in the event of conversion. Ingersoll (1977b) documents that in practice a substantial number of firms delay calling relative to this policy, reporting that on average firms wait until the conversion value exceeds the call price by 43.9%. Asquith and Mullins (1991) observe that studies on call delay are important since a failure to empirically confirm clear predictions of finance theory calls into question the validity of the models. Not surprisingly, researchers have responded to Ingersoll's analysis by considering alternate theories of convertible bond call policy. Theoretical explanations of call delay can be classified into two groups: explanations that are 3 unrelated to the firm's dividend policy, and explanations that are dependent on the firm's dividend policy.There are two explanations unrelated to dividend policy. One is that the costs of a failed call along with a required call notice period can lead firms to delay calling until a bond's conversion value exceeds its call price by a sufficient safety premium. 1 The other is that call delay can be a credible signal that management believes that the conversion value of the bond will exceed its face value at maturity and hence that there is little to be gained by forcing conversion early since bondholders will convert at maturity (Harris an...