The focus of the paper is the effect of merger proposals on the expected profitability of the bidder and the target. We illustrate how an unsuccessful bid may increase the profitability of the target but reduce the profitability of the bidding firm, relative to the profitability of the firms before the merger offer. The profitability of a merger proposal is lowered due to learning from rejection. We use our theoretical model to explain empirical work on this issue. THEORETICAL WORK on the profitability of takeovers has rarely focused on the problem of unsuccessful merger proposals or tender offers.' This is unfortunate because the decision that is generally available to a firm is not whether it should merge with another firm, but, rather, whether it makes a merger proposal or tender offer. In this paper, we contend that failure to take account of this may bias estimates of the profitability of takeover attempts. The focus of the paper is the effect of merger proposals on the expected profitability of the bidding firm and the target firm. We describe a theoretical model to show how an unsuccessful bid may increase the profitability of the target but reduce the profitability of the bidding firm, relative to the profitability of the firms before the merger offer. This is independent of any effect of potential future takeovers on the value of the firm. Consequently, the profitability of a merger proposal will be lower than in the absence of any effect conditional on rejection. The assumption causing this result is that a target firm has private information about its profitability which is not available to the bidding firm. Assume that there is uncertainty about the profitability of the target and that the bidder offers x dollars for the target. The target accepts if its profits from not being merged are less than x and is rejected otherwise. Hence the response of the target to a merger proposal reveals information on the competitiveness or profitability of the target. We further assume that the two firms (the bidder and target) compete in a duopolistic market and play a Cournot game in quantities if a merger does not occur. (Therefore, we are restricting attention to horizontal mergers.) In the equilibrium of our model, bidding firms that suffer rejection of their *We would like to thank Peter Cramton, Norman Ireland, Alvin Klevorick and James Mirrlees and two anonymous referees for helpful comments. 'This is true for e.g. Salant, Switzer and Reynolds [1983], Davidson and Denekere [1983], Perry and Porter [1985], and Grossman and Hart [1980].