Ch. 3: Auctions in Corporate Finance 117
then infer that the gain from participating in the auction is low (they are not likely to
win in the final English auction). Since participation in the auction requires a bidder
to spend resources to determine her own value, the second bidder can be discouraged
from even entering the auction.Fishman (1989)extends this initial work by including
the possibility of non-cash offers.
Fishman’s (1988)model works as follows. The value of the target assets to the bidders
depends in part on the realization of a state of nature which is observed only by the
target. Conditional on the target’s information, the value of the assets to the bidders is
increasing in the bidders’ independent private signals. The means-of-payment can be
either cash or debt that is backed by the target’s assets. Each bidder has to incur some
cost to learn the private signal. Bidder 1 identifies a target by accident, and then incurs
some cost to learn his signal (bidding is assumed to be not profitable if the true signal
is unknown). If bidder 1 submits a bid, the target is “put in play”, and a second bidder
is aware of the target. This bidder then decides whether or not to compete for the target
and incur the cost of learning her signal.
There is a stand-alone value of the target that is public information, and the target
rejects all bids below this value. Since the bidders do not know completely how much
the target assets are worth to them (recall that the target privately observes part of this
information), bidders could end up overpaying for the target. Paying with debt mitigates
the overpayment because the value of the debt is contingent on the value of the target
assets (since the debt is backed by these assets). However, if bidder 1 draws a high
private signal, a cash offer—though costly—will separate it from a bidder with low
signal: the latter will prefer to pay only with debt since his own private signal is not
sufficiently high. Thus, by bidding with cash, the first bidder can signal to the second
bidder that the latter’s likelihood of winning the ensuing auction is low: hence, the
second bidder may decide not to incur the cost of learning her signal. This, then, is a
“pre-emptive” bid. On the other hand, if the first bidder’s signal is low, bidding high with
a cash offer is too costly. Thus, such a bidder would decide not to preempt and instead
bid with debt to mitigate the potential loss from buying a target with low synergy. Notice
that one prediction of this model is that more competing offers should be forthcoming
with non-cash offers than with cash offers.^30
4.2.5. Modelling auctions of companies
Hansen (2001)reviews the formal auction process used for selling private companies
and divisions of public companies. The model explains the common practices of limit-
ing the number of bidders and limiting the disclosure of information to bidders (Boone
and Mulherin, 2006a), even though theory suggests that both practices would reduce
(^30) Betton and Eckbo (2000)find that the average offer premium in successful single-bid contests is greater
than the average offer premium in thefirstbid in multiple-bid contests. This what consistent with preemptive
bidding.