Auctions Versus Negotiations

Jeremy Bulow; Paul Klemperer

The American Economic Review, Vol. 86, No. 1. (Mar., 1996), pp. 180-194.

Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28199603%2986%3A1%3C180%3AAVN%3E2.0.CO%3B2-%23

The American Economic Review is currently published by American Economic Association.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained
prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in
the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/journals/aea.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

The JSTOR Archive is a trusted digital repository providing for long-term preservation and access to leading academic
journals and scholarly literature from around the world. The Archive is supported by libraries, scholarly societies, publishers,
and foundations. It is an initiative of JSTOR, a not-for-profit organization with a mission to help the scholarly community take
advantage of advances in technology. For more information regarding JSTOR, please contact support@jstor.org.

http://www.jstor.org
Fri Oct 19 16:51:00 2007
 Auctions Versus Negotiations

Which is the more profitable way to sell a company: an auction with no reserve
price or an optimally-structured negotiation with one less bidder? We show under
reasonable assumptions that the auction is always preferable when bidders' sig-
nals are independent. For afiliated signals, the result holds under certain re-
strictions on the seller's choice of negotiating mechanism. The result suggests
that the value of negotiating skill is small relative to the value of additional
competition. The paper also shows how the analogies between monopoly theory
and auction theory can help derive new results in auction theory. (JEL D44,
G34)

There are close analogies between standard to make binding commitments, who can hold
price theory and the theory of auctions. In an an optimal auction with N buyers. This is true
absolute English auction, in which the price under standard assumptions if buyers have pri-
rises continuously until only one bidder re- vate values, common values, or something in-
mains and the seller is required to accept the termediate. No amount of bargaining power is
final bid, the sale price equals the lowest com- as valuable to the seller as attracting one extra
petitive price at which supply equals demand. bona fide bidder.
In the theory of optimal auctions the seller is Since the informational demands for com-
treated as a monopolist who can choose any puting optimal mechanisms are substantial,
mechanism, such as establishing a minimum and the computations involved are complex,
sale (or reserve) price, to maximize expected this result suggests that it will often be more
profit. As in monopoly theory, optimal auction worthwhile for a seller to devote resources to
theory assigns all bargaining power to the expanding the market than to collecting the
seller, subject to the constraint that she does information and making the calculations re-
not have access to buyers' private information quired to figure out the best mechanism.'
about an asset's value. Our analysis also has policy implications for
This paper shows how the analogies be- when the directors of a public company should
tween monopoly theory and auction theory be allowed to privately negotiate its sale. Our
can help derive new results in auction theory. result shows that a single extra bidder more
Specifically, we are able to put a fairly tight than makes up for any diminution in negoti-
bound on the value of any seller's bargaining ating power. This means that there is no merit
power: a seller with no bargaining power who in arguments that negotiation should be re-
can only run an English auction with no re- stricted to one or a few bidders to allow the
serve price among N +1 symmetric bidders seller to maintain more control of the negoti-
will earn more in expectation than a seller with ating process, or to credibly withdraw the
all the bargaining power, including the ability company from the market.'

* Bulow: Graduate School of Business, Stanford Uni- ' Similarly, in a procurement context, competitive bid-
versity, Stanford, California 94305; and Klemperer: Nuf- ding by suppliers will yield lower average prices than ne-
field College, Oxford University OX1 INF, United gotiating with a smaller number of suppliers. See R.
Kingdom. We thank colleagues at Oxford University and Preston McAfee and John McMillan (1987b) for exam-
Stanford Business School, seminar audiences, and espe- ples. More broadly, our results are supportive of the view
cially Preston McAfee, Margaret Meyer, John Roberts, that optimal regulation of an industry may be less impor-
Lawrence Summers, and our referees for valuable tant than attracting additional entry.
comments. ' Opening negotiations with additional bidders makes
VOL. 86 NO. 1 BULOW AND KLEMPERER: AUCTIONS VS. NEGOTIATIONS I81

Similarly, a seller should not accept any with a final optimal take-it-or-leave-it offer to
"lock-up" agreement that a buyer is willing the last remaining bidder. Under mild assump-
to offer in return for the seller not beginning tions, this result holds regardless of whether
negotiations with additional potential acquir- bidders' signals are independent or affiliated.4
ers. For example, in late 1993 Paramount We then show (Theorem 2) that with Nrisk-
agreed to sell itself to Viacom, knowing that neutral bidders with independent signals, it is
QVC was interested in bidding for Paramount. optimal for the seller to use the N-bidder
Paramount and Viacom agreed to terms that mechanism described above, with a final offer
gave Viacom options to buy 24 million shares that generally depends on the prices at which
of Paramount and a $100 million break-up fee the low bidders dropped out. With indepen-
in the event that any other company were to dent signals and risk-neutral bidders, there-
purchase Paramount. The boards argued that +
fore, an auction with N 1 bidders dominates
in return for effectively excluding other bid- any negotiation with N bidders.
ders, Paramount had been able to negotiate a With affiliated (but nonindependent) sig-
higher price than it could have expected in an nals an English auction plus final take-it-or-
open auction. QVC contested the terms of the leave-it offer does not maximize expected
deal, contending that holding an auction would revenue among all conceivable selling mech-
have been the appropriate way to maximize anisms, but it does maximize expected reve-
shareholder value. The Delaware courts sub- nue subject to some restrictions on the seller's
sequently agreed with QVC. Our analysis sup- choice of me~hanism.~ It therefore remains
ports that decision.' true that an auction with N + 1 bidders beats
We begin in Section I by developing the in- any standard mechanism for selling to N
tuition for our results, and informally deriving bidders.(jx7
them in the simple and familiar case of buyers
with independent private values.
We develop our general model in Section 11.
We extend Bulow and John D. Roberts's
(1989) interpretation of auctions, based on
marginal revenues, from their independent pri- Signals are affiliated if, as a bidder's signal rises, he
vate values model to this general model. We expects others' signals to rise as well, in the sense that
use this to show (Theorem 1) that an English higher values for other bidders' signals become relatively
auction with N + 1 bidders but no reserve more likely. See Paul Milgrom and Robert J. Weber
(1982).
yields higher revenue in expectation than an ' Giuseppe Lopomo (1995) shows that the English auc-
English auction with N bidders, culminating tion plus reserve price maximizes the seller's expected
profit in Milgrom and Weber's (1982) "general symmet-
ric model" among all mechanisms where losers do not
pay anything and in equilibrium the winner (if anyone) is
it harder to use negotiating tactics such as credibly threat- the bidder with the highest signal and his payments are
ening not to sell if the bidders' offers are too low. If a weakly increasing in his own signal for any realization of
board approaches many bidders, it may be legally forced other bidders' signals.
to hold an open auction and cede its power to control the The results of the one-shot seller-optimal mechanism-
form of negotiations, see e.g., Jesse H. Choper et al. design literature extend straightforwardly to dynamic
(1989). In any event, the business of a company whose games in which the seller's discount rate is at least as high
future ownership is thought to be uncertain may be dam- as the buyers', so a seller cannot do better in any multi-
aged until the question of ownership is resolved so it may period game than in the one-shot game. (Using delay is
be hard to credibly withdraw the company from the mar- simply equivalent to a commitment to not sell with some
ket. See, e.g., Andrei Shleifer and Robert W. Vishny probability-see, for example, Peter C. Cramton [I9851
(1988). and Milgrom [1987].) If the seller's discount rate is lower
'Under dominant U.S. takeover law, a company can than the buyers', then screening over time can allow the
negotiate its sale to a purchaser and decline to hold an seller to extract a larger surplus than one can obtain from
auction if (i) the board is acting in good faith to maximize a one-shot mechanism. In the extreme case in which the
shareholder value and has conducted a reasonable inves- seller does not discount the future at all and the buyers do,
tigation of whether the price is adequate, and (ii) the price then the seller should run an extremely slow "Dutch"
attained through negotiations was high enough to be worth auction, in which the price begins high and is gradually
the cost of any lock-up provisions and other prohibitions reduced, and this will extract arbitrarily close to all sur-
necessary to secure the offer. plus. We do, however, ignore any time costs of accumu-
182 THE AMERICAN ECONOMIC REVIEW MARCH I996

We also note that if a seller could negotiate and the buyer's value, which is private infor-
with N bidders while maintaining the right to mation to the buyer, is drawn from a uniform
subsequently hold an English auction without distribution on [0, 11. Both parties are risk
a reserve price and with an additional bidder, neutral. It is easy to show that the optimal
the seller would always do better to proceed strategy for A in negotiating with her buyer is
directly to the auction. Thus a seller should to offer a take-it-or-leave-it price of .5; the of-
generally focus on maximizing the number of fer will be accepted half of the time, yielding
bidders, and should refuse to bargain with bid- an expected profit of .25. Seller B also has a
ders who wish to preempt the auction process.* value of 0, but differs from A in two respects:
Finally we extend our result to multiple first, she has two "serious" bidders, each with
units and show that the price-theoretic anal- private values drawn independently on [0, 11;
ogy of this extension gives an interesting re- second, she may hold only an English auction
sult about the value to firms of restricting with no r e ~ e r v eIn
. ~ this auction, the expected
competition relative to the value of expand- profit to the seller will be the expectation of
ing demand. the lower of the two bidders' values, which is
the point in the auction where the lower bidder
I. An Example with Independent Private Values will drop out. That expected profit is 113, so
the extra bidder is worth more than the reserve
We begin with a simple problem and then price.
generalize. Seller A has one "serious" poten- How can we generalize this result? The dif-
tial buyer, with a value that is at least as high ficulty can be illustrated in our numerical ex-
as the seller's. For example, A's value is zero ample. It is clear that in some cases (namely
those when the first bidder's value is greater
than or equal to .5, and the second bidder's
value is less than .5) the reserve price is worth
lating buyers, and any differences in the costs of running more ex post than the second bidder, but in all
different sales mechanisms. See Ruqu Wang (1993).
'We do not analyze how the number of bidders may
other cases the seller is better off with the extra
be affected by a firm's choice of mechanism. However, a bidder. The question is whether there is some
public auction may not only attract extra bidders through way to group the potential cases so that the
the extra publicity, but may also attract more bidders if seller with two bidders does better in expec-
bidding is costly. For example, with symmetric bidders tation within every subgrouping, and therefore
who simultaneously decide whether or not to pay the costs
of participating in an auction before learning their signals, better on the whole.
an auction with a reserve price may attract fewer bidders The most natural thing to try is to divide up
than an auction without reserve. Our analysis will speak the cases into those where the first bidder has
directly to this case. On the other hand, with sequential a value above the reserve price of .5 and those
entry of potential bidders who decide in turn whether or
not to pay the costs of acquiring a signal and then making
where the first bidder's value is below .5. At
a bid, bidders may make preemptive bids to make it un- first glance, this methodology does not work,
profitable for any future bidder to enter. In this case, there even in our simple numerical example. Con-
is no trade-off between using a reserve price and attracting tingent on the first bidder having a value below
bidders: a higher reserve price earns more money because .5, of course the seller with two bidders and
it both allows the seller to extract rents and increases the
expected number of buyers who will participate by making no reserve price will earn more than the seller
a preemptive bid more difficult. See McAfee and
McMillan (1987a). For further analysis of optimal seller
strategies with costly bidder participation, see Richard
Engelbrecht-Wiggans (1993), McAfee and Daniel Vincent Throughout the paper, an English auction is an auc-
(1992), and Shleifer and Vishny (1986). tion in which the price rises continuously until only one
We assume a single seller, interested only in expected bidder remains. At every price all bidders know how many
revenue, so in the context of selling a company we are other bidders remain active. A reserve price is a minimum
abstracting from issues such as shareholders' individual price below which the object will not be sold. (In an auc-
incentives to sell (see, e.g., Sanford J. Grossman and tion with a reserve price the seller makes a final take-it-
Oliver D. Hart, 1980; David Hirshleifer and Sheridan or-leave-it offer equal to the reserve price to the final
Titman, 1990; and Bengt Holmstrom and Barry Nalebuff, bidder, if the final bid is below the reserve price.) An ab-
1992), or management's interest in retaining control (see, solute auction is an auction with no reserve price; that is,
e.g., Milton Hanis and Artur Raviv, 1988). the seller is required to accept the final bid.
VOL. 86 NO. I BULOW AND KLEMPERER: AUCTIONS VS. NEGOTIATIONS

with one bidder (who will earn zero). But con-

tingent on the first bidder having a value be-
tween .5 and 1, expected revenue will be .5
with the reserve price and 11I24 with the extra
bidder. ''
Clearly, we need to use something other
than expected values to establish superiority
for the auction.
What we do is borrow a trick from monop-
oly theory. Consider a seller with constant
marginal costs of zero, and a linear demand
curve of p = 1 - q. How do we know that .5
is the optimal price and quantity for the seller?
If the seller chooses a quantity of .4 and a price
of .6, for example, she will earn more revenue
from the .4 buyers who actually purchase than
if she only charges .5, so there is no direct
dominance. Similarly, if a price of .4 is chosen
the seller earns less from the first .5 customers,
but there is that extra revenue from the last .l.
The way we establish superiority for the
quantity of .5 over the alternatives is by look-
ing at marginal revenues instead of prices.
Marginal revenue equals zero at a quantity of sales will be zero at a price of v(t,) = 1, in-
.5. By selling .5 units, the seller earns the same creasing linearly to one (expected) sale at a
positive marginal revenues on the first .4 units price of v (t,) = 0.
as a seller of only .4, plus some extra positive Defining revenue as price times quantity, we
marginal revenues on the next .1 sales. Selling can also derive marginal revenue in the usual
.5 is better than selling .6, because by selling way,
.5 you earn the same positive marginal reve-
nues on the first .5 units, and eliminate the neg-
ative marginal revenues on the last .l. By
looking at something like marginal revenues
we can establish the superiority of the two-
bidder auction in our initial problem.''
Assume that bidder j receives a private sig-
nal t, which is distributed with a density f (t,)
and a cumulative density F(t,), independent and graph marginal revenue against quantity
of other bidders' signals, and implies a private of expected sales. In our example, the mar-
value of v(t,). Graph value, v(t,), against ginal revenue curve from our demand curve is
' 6
quantity," q(t,) = 1 - F(t,) . In terms of our MR = 1 - 2q. Note that the sales probability
numei-ical example, the picture produced is an that is optimal for the seller with one bidder
"inverse-demand curve" of p = 1 - q. As (and the optimal take-it-or-leave-it price) can
Figure 1 shows, the quantity of (expected) be found where MR equals zero, at q = .5 and
p = .5.
Why? Just as the revenue from a take-it-or-
leave-it price can be calculated by multiplying
The expectation of the lower of two values, one of that price by the probability of sale at that
which is known to exceed 112, is 11124. price, expected revenue can also be found by
" The interpretation of auctions in terms of marginal
revenues follows Bulow and Roberts (1989). The current taking the area under the MR curve for all the
paper shows how to extend this interpretation from the values in excess of the take-it-or-leave-it price.
independent private values model to the general case. Obviously, the optimal take-it-or-leave-it price
184 THE AMERICAN ECONOMIC REVIEW MARCH 1996

must be equal to her selling price. For exam-

ple, in our numerical example, if v ( t, ) = .6 the
MR of the winner might be anywhere from .2
to 1, but on average it will be .6. This implies
that for any v (ti ) , the expected revenue of the
seller can be described as the expected MR of
the winning bidder. Averaging over all possi-
ble values for v(ti ) , therefore, the seller's ex-
pected revenue equals the expected MR of the
winning bidder.
Now add a conventional auction theorylmo-
nopoly theorylmechanism design assumption:
assume that the MR curve is downward slop-
ing. This implies that the buyer with the higher
value, who is the one who will actually win
the auction, is also the buyer with the higher
MR. If the seller's expected revenue is the ex-
pected MR of the winning bidder, and the win-
ning bidder always has the higher MR of the
two bidders, then the expected revenue from
2. THEEXPECTATION
FIGURE OF MR(t,), CONDITIONAL
ON t, 2 t,, EQUALS
THE AREAUNDER THE MR(t,)
CURVE the auction can be written as the expectation
FROM q = 0 TO q = 1 - F(t,), DIVIDED
BY 1 -
F(t,): of the maximum of the marginal revenues of
THISMUSTEQUAL~ ( t , ) the two bidders, call it

Expected Revenue (auction with two)

is where MR = 0. The seller may be thought

of as receiving, in expectation, the MR of the
buyer when it is positive, and zero when the This may be compared with the expectation
buyer's MR is negative. Put slightly differ- from the one-bidder mechanism,
ently, expected revenue may be thought of as
the expectation of the maximum of the MR of Expected Revenue (optimal mechanism
the bidder and zero.
Now let's move on to the problem of seller with one) = E { max(MR(t, ), 0) } .
B, who holds an auction between two bidders.
Assume that the "underbidder" has a value of Now compare the right-hand sides of the
v (t, ) and the eventual winner has a value v (t,) . two expressions above in the cases where the
We do not learn the value of v(t,) in the auc- first bidder's value exceeds the optimal re-
tion, but it is greater than or equal to v(t, ). serve price, so that MR( t, ) 2 0. It is obvious
Consider the graph with the same demand that contingent on that, the first expression is
curve and marginal revenue curve as before, larger.I2 But what about if MR(tl ) < O? Here
and mark a point on that curve as v ( t, ) ( see
Figure 2). Contingent on the underbidder's
value being v(t,), we know that the seller will I' Of course, as demonstrated above, it does not follow
earn exactly v(ti ) in the auction, but it will be that contingent on MR(tl) 2 0,
more useful to express the seller's winnings in
terms of marginal revenue. What is the ex- Expected Revenue (auction with two)
pected MR associated with the winning buyer,
> Expected Revenue (optimal mechanism with one).
conditional on the selling price being v (ti ) ? It
is obviously v (ti ) , by the same simple math as Likewise, a monopolist's actual revenue from a subset of
we use to show that the average marginal rev- the buyers sold to does not equal the sum of the marginal
enue associated with a monopolist's customers revenues of these buyers.
VOL. 86 NO. I BULOW AND KLEMPERER: AUCTIONS VS. NEGOTIATIONS 185

is where our "serious-bidder" assumption, case, the expected revenue from an auction
that both potential bidders have a value at least with N bidders and an optimal reserve price is
equal to the seller's value of zero, comes in to equal to the expectation of the maximum of
play. What is the expectation of MR(t2)? (MR(tl), MR(t2), ... , MR(tN), 0 ) while the
Since the lowest possible value of v (t,) is zero, expected revenue from an auction with N + 1
it must be that the expectation of MR(t,) bidders is equal to the expectation of the max-
equals zero. In demand-curve terms, if we set imum of (MR(tl), MR(t2), ... , MR(tN+, ) ) .
a price of zero, then total revenue, and there- Since the expectation of MR(tN+,) is equal to
fore also the average MR of all buyers, must zero, it is clear that the auction with the extra
be zero. So if MR(tl) < 0, then the second bidder yields a higher expected revenue.
expression is zero, while the first expression is We have now gone pretty far while relying
the expectation of the maximum of two terms, on only elementary mathematics. Since it is a
one of which has an expected value of zero. standard result that an auction with an optimal
Again, therefore, the first expression is larger, reserve price is an optimal mechanism if bidders
so we have now established the auction's are symmetric and risk neutral and have inde-
dominance. pendent private values and downward sloping
The need for our serious-bidder assumption MRs (John G. Riley and William F. Samuelson,
should be quite clear. Assume that there is a 1981; and Roger B. Myerson, 1981), we have
probability 1 - p that the second bidder values already shown that, under these assumptions, an
the asset below zero, and that the second bid- auction with N + 1 bidders is superior to any
der is otherwise drawn from the same distri- mechanism involving N buyers.
bution as the first bidder.l"hen the second The above discussion assumed that bidders
bidder will be worth only p times as much to have independent private values. In fact, the
the auction seller as if it were certain that the argument that the expected revenue from an
second bidder had a value above zero.14 In the absolute English auction equals the expected
limit where p approaches zero, the extra bidder MR of the winning bidder applies very gen-
would be of virtually no use, and a reserve erally. Similarly it is a very general result that
price would dominate. In our numerical ex- the expected revenue from an English auction
ample, we would require p 2 .75 for the auc- with an optimal reserve price equals the ex-
tion to be at least as good as the reserve price." pectation of the maximum of the highest bid-
It is easy to extend the analysis to compare der's MR and zero. The difficulty is that in a
a seller with N (symmetric) bidders in an auc- general model bidders' values and MRs are
tion and a reserve price to one with N 1 + not independent of other bidders' private sig-
bidders and no reserve price. By exactly the nals. Conditional on the first N bidders having
same analysis as in the-one- and two-bidder low MRs, the expected MR of the (N 1)st +
bidder is also low. Furthermore since, we will
show, to compute an English auction's ex-
pected revenue each bidder's MR must be
I' We could also assume that there is a probability of calculated based on the information that the
1 - p that the first bidder has a value below zero to main- auction will reveal, a bidder's relevant MR in
tain symmetry, but since neither sales mechanism yields
any profit when the first bidder's value is below zero, we can an N-bidder auction is different than in an N +
restrict our comparison to cases where the first-bidder's value 1 bidder auction. Nevertheless the method of
is at least zero. proof outlined above can be developed to
l 4 We assume that the seller who runs an auction can
show that an extra bidder is worth more than
demand a minimum price of zero.
I s Without the serious-bidder assumption, if there are
an optimal reserve price if either bidders' val-
enough extra bidders that in expectation the second- ues are private or bidders' signals are affili-
highest extra bidder has a value of at least zero (and their ated.16 It then follows easily that an auction
MRs are downward sloping), then it follows that the ex-
pectation of the highest MR of the extra bidder is at least
zero, so that the extra bidders are more valuable than the
reserve price in expectation, even if the seller is not al-
lowed to insist on any minimum price, and may therefore See note 4 for an informal definition of affiliation,
be sometimes obligated to sell at a loss. and the proof of Theorem 1 for the relevant implications
186 THE AMERICAN ECONOMIC REVIEW MARCH 1996

with N +
1 bidders and no reserve price is vi ( T ) is a function only of t i , while in the
more profitable than any standard mechanism special case of pure common values vi ( T ) =
with N bidders. v j ( T ) b'i, j , T . So that seller revenue is
bounded, we assume v,(T ) I v * < b'j, T .
11. The General Model While tj is private information to bidder j ,
the functions v,(T ) and f (t, 1 T - , ) are common
In our general model, bidders' private sig- knowledge.
nals need not be independent, and bidders' We assume that all agents are risk neutral,
values may be either private or common, or though this is not necessary for Lemma 1 or
something intermediate.I7 Theorem 1 .
Let t, be bidder j ' s private signal about the Finally, we define
value of the asset. Without loss of generality,
normalize so that 0 I t, I 1 b'j, and normalize
the seller's value of the object to zero. We
write T to represent the vector ( t , , . . . , tN+ ) , ,
T - , to represent all of the elements of T other
than t,, and define T = T-(,+ ,,,and T - , as all
the elements of T other than t,. and
We write f (t, 1 T - ,) for the conditional den-
sity of t, given T - , , and F(t, 1 T - ,) for the prob-
ability that the jth signal is less than or equal
to t, given T - , . More generally, we write
f ( x I y ) for the conditional density of x given
y , and F ( x I y ) for this conditional distribution.
We assume f (t, 1 T - , ) is positive and finite The interpretations of the marginal revenues
for all tj and T - , . MRj and MR,. are exactly as in Section I:
Let vj(T ) be the value of the asset to bidder graphically, if we plot v j ( T ) against quantity
j as a function of the vector of signals 1 - F(tj 1 T - , ) for any bidder, varying only t j ,
T , and let & ( T ) = E,, , { v , ( T ) ) = S, we will have a downward-sloping demand
,
vj(T )f ( tN+ I T ) dtN+ be the expectation of
vj(T ) conditional on T . Higher signals imply
curve. If we think of that graph as the demand
curve of buyer j , with the quantity being the
higher expected values, so &,(T)ldt, > 0, probability that the buyer would accept a take-
hi( T )/at, r 0, and tj > ti =, v,(T ) r vi ( T ) it-or-leave-it offer at any given price if he
b'i, j , T . In the special case of private values, knew the signals of all the other bidders, then
the MR, curve is just the marginal-revenue
curve derived from that demand curve. Simi-
larly, the m, curve is derived from the graph
for our model. For the private values auction-with-two vs. of ti,(T) against 1 - F(t,lT-,), for a buyer
reserve-price-with-one case the argument is easily ex- who knows the signals of all the other buyers
tended to affiliated signals: the seller's revenue from the except the N + 1st."
auction equals the lower of the two values, which equals
the average of the values minus half their difference. Af- We maintain the following assumptions
filiation leaves the average the same but reduces the dif- throughout:
ference, so it further increases expected seller revenue
above the expected revenue from a single bidder plus a ( A . l ) Downward-Sloping MR: tj > ti =,
reserve price. In the general case, if the extra bidder's sig-
nal is affiliated with other bidders' signals, this effectively M R , ( T ) > MR, ( T ) and M R , ( T ) >
reduces the amount of private information available to MRi(T).
each bidder, thus reducing the "information rent" earned ( A . 2 ) Serious Bidders: v , ( T ) r 0 b'j, T .
by the winning bidder, and so increasing the value to the
seller of the additional bidder, relative to the value of a
reserve price.
" Our model is essentially that of Milgrom and Weber
(1982), although we do not always impose their affiliation Note that is not in general the expectation of
assumption. MR, (T) unless bidders' signals are independent.
VOL. 86 NO. I BULO W AND KLEMPERER: AUCTIONS VS. NEGOTIATIONS 187

(A.3) Symmetry: Bidders' value functions leave-it offer to the remaining bidder, equals
are symmetric, so vi ( t ,, ... , t i , ... , t,, E,-{max(MR,(T), MR2(T), . . . , MRN(T),
...) = v j ( t l ,... , tj, ... , t i , ...) V i , j , T , 0)1.
bidders' signals are symmetrically dis-
tributed, and bidders choose symmetric The proofs of these two lemmas straightfor-
strategies. l 9 wardly follow the arguments of Section I; as
with independent private values, the optimal
Assumption ( A . l ) is a standard regularity take-it-or-leave-it final offer is the maximum
condition in auction theory, analogous to of the price at which the last losing bidder
an assumption of a downward-sloping marginal- quits, and the price at which the winner's mar-
revenue curve in monopoly theory. Assump- ginal revenue would equal zero.21In the gen-
tion (A.2) ensures that every bidder is will- eral case, however, each bidder's marginal
ing to make an opening offer of zero, the revenue depends on all other bidders' signals,
seller's value, in an absolute English auc- so the optimal final offer can only be deter-
tion. Assumption (A.3) ensures that the bid- mined after all the losing bidders' signals have
der with the highest signal always wins such been inferred from the prices at which they
an auction.'O quit the auction. This can explain why it is
common for a seller to announce a reserve
A. Expected Revenue from Auctions price only at the end of the auction."

We now follow the strategy used in Section THEOREM 1: Expected revenue from an
I to develop our main theorem. All proofs are absolute English auction with N +
1 bidders
provided in the Appendix. exceeds expected revenue from an English
auction with N bidders followed by a take-it-
LEMMA 1: The expected revenue from an or-leave-it offer to the last remaining bidder if
absolute English auction with N 1 bidders + either ( i ) bidders' values are private; or ( i i )
equals E,{max(MR,(T), MR2(T), ... , bidders' signals are afJiliated.2'
MRN+1 ( T I ) 1.
Just as for our independent private values
LEMMA 2: The expected revenue from an example, the proof considers separately the
English auction with N risk-neutral bidders
followed, after the N - 1 low bidders
have quit, by an optimally chosen take-it-or-
Strictly, bidders with very low signals may be indif-
ferent about participating since they might know that they
l 9 That is, in an absolute English auction each bidder's would never meet the seller's take-it-or-leave-it price.
equilibrium strategy is to drop out of the bidding at the However, the seller can induce all bidders to participate
price he would just be willing to pay given the actual sig- at an arbitrarily small cost in expected revenue by com-
nals of the bidders who have already dropped out (in equi- mitting to foregoing the take-it-or-leave-it offer with a
librium their signals can be inferred from where they probability approaching zero and to always accepting the
dropped out) and assuming all the remaining active bid- highest bid in this event. Note also that, strictly speaking
ders (whose signals he does not know) have signals equal the rule by which the seller's final take-it-or-leave-it offer
to his own. (To see this, observe that if all other bidders will be determined must be precommitted to before the
follow this rule, a bidder is happy (unhappy) to find him- bidding. Otherwise there is in theory the possibility of
self the winner at any price below (above) this stopping other symmetric equilibria that are less profitable for the
price; in the special case of pure private values each bidder seller. For example, it is a sequential equilibrium that
just drops out at his own value.) Note that Sushi1 every bidder drops out at a certain price; if any bidder
Bikhchandani and Riley (1993) show that there may be stays, that bidder is believed to have the highest possible
other (asymmetric) equilibria. signal and is offered a very high final price.
' O If bidder signals are negatively correlated, then (A.l) 22 Of course, a seller should also not commit to a re-
is less likely to hold than with independent signals. (A.2) serve price until the end of the auction. Many auction
is less likely to apply in a common-values setting than with houses seem to commit to secret reserve prices before auc-
private values. For further discussion of the importance of tions, but there are often further subsequent negotiations
the assumptions see sections 9 and 10 of our working pa- if an object is unsold at its reserve price.
per, Bulow and Klemperer (1994a). '' Note that independent signals are affiliated.
188 THE AMERICAN ECONOMIC REVIEW MARCH 1996

cases in which the highest of the first N bid- B. Auctions versus Optimal Mechanisms
ders, say bidder j , has a positive or negative
MR,, that is his value exceeds or does not ex- Lemma 3 extends to general value functions
ceed the optimal reserve price (i.e. take-it-or- Myerson's ( 1981) theorem, that with inde-
leave-it offer) that would be set contingent on pendent signals and risk-neutral bidders, any
the other N - 1 of the first N signals. As be- two mechanisms that always result in the
fore, when there would be no sale the expec- same winning bidder are revenue equivalent.
, ,
tation over t, + of MRN+ equals bidder N + (Myerson considers only common values in
1's lowest possible value, which equals or ex- which players' values are additive functions of
ceeds zero by the "serious bidder" assump- signals.) We also reinterpret Myerson's "vir-
tion. When there would be a sale, affiliation tual utilities" as marginal revenue^.^'
implies that the expectation (over tj and t, + ) ,
of MR, is greater than or equal to the expec- LEMMA 3: With independent signals and N
tation (over t,) of m,,
contingent on the other risk-neutral bidders, the expected revenue
N - 1 signals and on a sale. With either sale from any sales mechanism equals the expec-
or no sale, then, the expectation of the maxi- tation of the marginal revenue of the winning
,
mum of MR, and MRN+ exceeds in expecta- bidder, provided any bidder with the lowest-
tion the maximum of MR,and 0. possible signal expects zero surplus; the
The difference between MR, and MR,, marginal revenue of the winning bidder is
which means that affiliation reinforces our MR,(T) i f j is the winner and is taken to be
result that auctions beat negotiations, is ex- zero if the good is retained by the seller.
actly the difference that implies that with
three or more bidders an open ascending Clearly no sales procedure with voluntary
English auction is more profitable than a participation can earn greater profits than one
sealed-bid second-price auction (see Milgrom in which bidders with the lowest possible
and Weber, 1982) .24 signals expect zero surplus.26A corollary of
Lemma 3, therefore, is that the mechanism of
- - always sells to the bidder j
Lemma 2-which
for whom MR,(T) is largest if that value is
greater than zero and makes no sale other-
wise-is optimal with risk-neutral bidders and
independent signals, under our assumptions
24 In the sealed-bid auction with three bidders the bid-
(A.1) - (A.3) :
der with the second-highest signal, who determines the
price, bids his expected value assuming that he is tied with THEOREM 2: With independent signals and
the highest signal, and estimates the distribution of the N risk-neutral bidders, an optimal mechanism
third signal based on this assumption. This bid equals the
lowest-possible expected value of thewinner, say j, that for a risk-neutral seller is an English auction
is, equals the expectation (over t,) of MR,. In an open auc- followed by an optimally-chosen take-it-or-
tion the second-highest bidder chooses his dropout price leave-it offer to the last remaining bidder.
by assuming that he is tied with the highest signal and
based on the actual third signal which he infers assuming
equilibrium behaviour-see note 19. His final bid there-
Theorems 1 and 2 together imply the main
fore equals the lowest-possible actual value of the winner, point of our paper:
that is, the expectation (over t,) of MR,, and affiliation
implies the expectation of this bid exceeds& sealed bid.
In our context, the expectation (over t,) of MR, equals the
lowest expected value j could have, and if j has the lowest-
possible signal he will estimate the distribution of t N + ,
based on this. However the expectation (over t, and tN+,) 25 Special cases of Lemma 3 and Theorem 2 have been

of MR e uals the expectation (over tN+,) of the lowest independently obtained by Fernando Branco (1994) and
! q
value J could have given the actual tN+,. Affiliation im- Lopomo (1995).
,
plies the distribution of the actual tN+ stochastically dom- 26 NO sales procedure can give any type of any bidder
,
inates the distribution of t, + contingent on j having the a negative surplus, and giving the lowest type a positive
lowest poKble signal, so the expected MR, exceeds the surplus would require raising all other types' surpluses.
expected MR,. See the proof of Lemma 3.
VOL. 86 NO. I BULOW AND KLEMPERER: AUCTIONS VS. NEGOTIATIONS 189

COROLLARY: With independent signals and to all that if negotiations fail, there will be an
risk-neutral bidders, an absolute English auc- auction. Viewing the two-stage process as a
tion with N +
1 bidders is more profitable in whole, then, the seller is constrained to choose
expectation than any negotiation with N bid- among mechanisms that always lead to a sale.
ders. But clearly any optimal mechanism that al-
ways sells must always sell to the buyer with
Of course, to the extent that it is unrealistic the highest signal. Therefore, it will not be op-
to expect a seller to be able to commit as firmly timal to sell in the negotiation stage unless it
as is needed for the optimal mechanism, and is certain that the buyer's signal is greater than
to compute the optimal reserve price, the Cor- or equal to the signal of the ( N 1) th bidder. +
ollary's statement about the auction's superi- Therefore the seller should insist on a price in
ority is conservative. the negotiation phase that will only be ac-
When buyers' signals are nonindependent, cepted when a buyer gets the maximum signal
the mechanism described in Theorem 2 is not of 1, which occurs with probability zero.
optimal, and a seller who can choose any Therefore, under our assumptions, the seller
mechanism can generally extract all bidders' should not accept any high "lock-up'' bid that
surplus (see Jacques Cremer and Richard a buyer may be willing to offer in return for
McLean, 1985; McAfee et al., 1989; and not holding an auction with an additional
McAfee and Philip J. Reny, 1992). When a buyer.28
seller can extract all surplus from N bidders, it
is not hard to show that this will always dom- D. Multiple Units
inate an absolute auction with N +
1 bidders.
However, Lopomo (1995) has shown that Our model extends easily to a seller with K
with affiliated signals and risk-neutral bid- goods to sell and N 2 K symmetric bidders
ders expected revenue from the mechanism of each interested in buying one unit. With in-
Theorem 2 is higher than from any other dependent signals, the optimal sales mecha-
mechanism in which (i) losers do not pay and nism is to sell to the K bidders with the highest
(ii) in equilibrium the winner, if anyone, is the signals, provided mi (T ) r 0 for K or more
bidder with the highest signal and his pay- bidders. Otherwise, sell only to those bidders
ments are weakly increasing in his own signal for whom mi(T ) r 0.29The optimal mech-
for any realization of other bidders' signals.27 anism yields expected revenue equal to the ex-
Thus it remains true that an absolute auction pected sum of the K highest values among
+
with N 1 bidders is better than any standard m, ( T ) , ... , MRN(T) and K zeros. It is not
mechanism for selling to N bidders.

C. Negotiations Followed by an Auction This result would be unaffected by other bidders hav-
ing costs of entering the auction. However, the presence
+
A final question is: if a seller has N 1 risk- of such costs can explain why bidders may jump bid to
neutral bidders with independent signals, can deter competitors from entering; see Michael J. Fishman
she benefit by first negotiating with N of the (1988) and Hirshleifer and Ivan P. L. Png (1989). See also
Christopher Avery (1993), Kent Daniel and Hirshleifer
bidders only, reserving the fight to hold an ab- (1993), and Nils Henrik von der Fehr (1994) for related
+
solute auction among all N 1 bidders if the discussion.
negotiations failed to produce a sale? The an- 29 Optimal negotiation is in general more complex than

swer is no. in the single-unit case since determining any bidder's MR

The reason is that if the seller has the option requires knowing all other bidders' signals. One way to
achieve optimal negotiation is (i) ask each bidder i to in-
of resorting to the auction, it will be obvious dependently report his signal t, (in equilibrium all reports
will be honest), and let ? be the (K + 1)st highest signal
reported; (ii) for each i w h o ~ p o r t sa signal ip the top K
'' So, for example, the mechanism of Theorem 2 is bet- signals, compute t, such that MR,(t,, . ., , t, ,,t,, t, + ,,. . . ,
-

ter than any of the English, Dutch, first-price sealed bid t,) = 0 and sell tothis bidder if t, 2 t,, at the maximum
or second-price sealed bid auctions together with a reserve ofiJ,(t,, . . . , t, , , t, t , + , , . . . , t,)andiJ,(t,, . . . , t,-,, t,
price. See Vijay Krishna and John Morgan (1994) for an t, + ,, . . . , t,). See our paper, Bulow and Klemperer
analysis of auctions in which all bidders pay. (1994b), for a further analysis of multiple-unit auctions.
 THE AMERICAN ECONOMIC REVIEW MARCH I996

hard to extend our earlier arguments to show [ A ] (the integral of marginal revenue up to
that this is less than the expected revenue from the monopoly quantity M), so collusion on
+
an absolute auction with N K bidders with the original demand curve would yield profits
independent signals (in which the final K bid- +
of (NI(N K)) [ A ] . Competitive profits on
ders pay the price at which the last excess bid- the expanded demand curve are the integral of
der quits). marginal revenues up to K, that is, [ A ] -
The analogue of this argument in traditional [B] . However, downward-sloping MR im-
price theory is informative. Consider an in- plies [B] 5 ((K - M)I((N + K) - M))([B] +
dustry with total capacity K at some constant [C]), and total marginal revenue equals zero
marginal cost c which we normalize to zero. +
at price zero so ([B] [C]) = [ A ] . So com-
Demand at a price of zero is N m K. The in- petitive profits equal [ A ] - [ B ] m [A] -
dustry has the ability to do one of two things: ((K - M)I((N + K) - M))[A] = (NI(N +
invest in a monitoring program which will en- K - M)) [A], which exceeds collusive profits.
able it to collude perfectly, or invest in an ad-
vertising campaign which will proportionally III. Conclusion
+
increase demand by a factor of ( N K)IN. In
the latter case, the industry will be perfectly A simple competitive auction with N +
competitive and will sell K units. Which op- 1 bidders will yield a seller more expected
tion is more ~rofitable? revenue than she could expect to earn by
Assuming ;hat the industry has a downward fully exploiting her monopoly selling position
sloping MR curve, the answer is that increas- against N bidders.
ing demand and remaining competitive is When a company is approached by a poten-
more valuable than colluding. The result fol- tial buyer or buyers, its options may be either
lows directly from our auction-theory model, to negotiate or to put the company up for auc-
with K units and independent private values. tion. Our analysis implies that if the board ex-
However, if N and K large enough that there pects at least one extra serious bidder to appear
is no aggregate uncertainty about valuations, in an auction, then it should generally not ne-
the argument can be made even more simply. gotiate and should directly begin an auction.
Figure 3 shows marginal revenue for the Of course, institutional considerations may
proportionately-expanded demand curve. Col- make any given situation more complex. For
lusive profits on this demand curve equal area example, if allowing many bidders access to
VOL. 86 NO. 1 BULOW AND KLEMPERER: AUCTIONS VS. NEGOTIATIONS 191

confidential financial information would cause sumed that bidders had no bargaining power
the company's value to be diminished to the in a negotiation. We therefore believe that our
eventual buyer, then one might wish to restrict basic result does not overstate the efficacy of
bidding. auctions relative to negotiations. Certainly a
But remember that our analysis assumed firm that refused to negotiate with a potential
that a seller could negotiate optimally, making buyer, and instead put itself up for auction,
credible commitments of the sort that might should be presumed to have exercised reason-
not be possible in real life, and we also as- able business judgment.

,
Write (x, T-,) for (t, , ... , t, - , x, t, + , ... , t,. , ) , that is, for the vector T but with the jth element replaced by x, and
write (x, T-,) for the vector T with the jth element replaced by x.

PROOF OF LEMMA 1:
If bidder j has the highest signal and bidder i has the second-highest signal, then bidder j will win the auction at
the price v,(t, , T- j), that is, the value i would have, if j ' s signal were t,O'. But by symmetry, v, (t, , T-,) equals
v,(t,, T-,), and

which is to say that the sales price equals the expected MR of the winning bidder, contingent on all the other signals. Because
the winning bidder has the highest MR, the result follows.

PROOF OF LEMMA 2:
As in the absolute auction, the next-to-last bidder i leaves at price Z,(t, , T-,) equals q ( t i , T-,). Let the seller choose a
take-it-or-leave-it offer for the last bidder, j , ofii,(2, T-,), where? 2 t, . (The seller infers T-, from the points where the
low bidders quit.) If ex post t, 2 1 then the seller will receive

--
If t, < i, thenAe seller will receive zero. That is, revenue equals, in expectation, MRJ(T) when t, ? L a n d zero when

tJ<A Since MR,(T) is increasing in t,, the seller maximizes expected profit by choosing 2 so that MR,(~,T-,) = 0

if MR(t, , T-,) < 0, and chooses 1 = t, otherwise. Since the winning bidder has the h i g h e s t k , the result follows.

PROOF OF THEOREM 1:

Conditional on any T-,, and on the jth signal being the highest of the first N signals, let i be such that%(;, T-,) is the

seller's optimal take-it-or-leave-it final offer (computed as in the proof of Lemma 2) when selling to the N bidders.

If t, < 2,

- - -
If t, 2 ^t, max(@,(T), ... , M R ~ T ) 0, ) = MRJ(T), so conditional on tJ 2 i and T-,, the expectation of
max(MR,(T), ... , MR,(T), O)

'O See note 19 for a full description of the equilibrium-bidding strategies.

 THE AMERICAN ECONOMIC REVIEW MARCH 1996

However, the expectation of MRj(T) conditional on f, 2 1 and T-,

= s:. I:.
=.
MR,(T )f(t,, t ~ 1 It,
+ 2 2, T-,)dt,dt~+I

f ( f ,l T-,i
f (tN+1 It, 2 l , T-,)dtN+ 1
1 - F(t,IT-,il,,=;

The inequality applies if signals are affiliated, because then the distribution of tN+ conditional on t, 2 1 and T-, sto-
chastically dominates the distribution of t,, conditional on f, = 1 and T-,. (With independent signals the inequality
,
holds with equality.) With private values, v,(?, T-,) is independent of t, + , so the inequality always holds with equality.
So conditional on any lowest N - 1 of the first N signals, T-,, and either on any t, < 1 or on t, 2 t, the expectation
ofrnax(=,(T), ... , E N ( T ) ,0 ) is (weakly) less than the expectation of max(MR,(T), ... , MRN+, ( T ) ) . Since the
inequalities are strict for a set of T-, that occurs with positive probability,

and the result follows by Lemmas I and 2

The above proof assumed risk-neutral bidders. If bidders are risk averse the expected revenue from the absolute auction is
unchanged, but the expected revenue from the N-bidder mechanism is reduced, increasing the advantage of the absolute auction.

PROOF OF LEMMA 3:
Let p, (T ) be the probability that i will receive the object, in equilibrium, let S, (t,) be the equilibrium expected surplus
to bidder i , and since we have independent signals, write f (t, ) and F ( t , ) for f (t, I r , )
and F(t, IT-,). For p, ( . ) to be an
equilibrium, it must be incentive compatible. In particular, the ith bidder, with signal t: , cannot gain by deviating to the
strategy he would use if he had signal t, so, with independent signals,

So S,(t,) has derivative dS,(t,)ldt, = ET-,((8V,(T)ldt,)p,(T)) = H , ( t , ) and S,(t,) = S,(O) + S: H,(t)dt. So

E,, {S.( i , ) ) = S.(O) + [H,(t)dtf (tt)dtt = S, (0) + (1 - F ( t t ) ) H ( t , ) d &(integrating by parts) = S. (0) +

E,, ( ( ( 1 - F ( t , i i l f ( t , ) i H , ( f , i l .
But expected seller profits can be written as the expected value of the good to the winning bidder, E T ( Z ~I =
(V, (T)p, (T )) ) , less the expected surplus of the N bidders, Z k I E,, ( S, (t, ) 1. SOexpected profits are
VOL. 86 NO. 1 BULOW AND KLEMPERER: AUCTIONS VS. NEGOTIATIONS 193

REFERENCES Economics, Spring 1980, 11( I ) , pp. 42-

64.
Avery, Christopher. "Strategic Jump Bidding Harris, Milton and Raviv, Artur. "Corporate
and English Auctions." Discussion paper, Control Contests and Capital Structure."
Kennedy School of Government, Harvard Journal of Financial Economics, 1988,
University, 1993. 20( 1/2), pp. 55-86.
Bikhchandani, Sushil and Riley, John G. Hirshleifer, David and Png, Ivan P. L. "Facilita-
"Equilibria in Open Auctions." Unpubli- tion of Competing Bids and the Price of a
shed manuscript, University of California, Takeover Target." Review of Financial
Los Angeles, September 1993. Studies, 1989, 2(4), pp. 587-606.
Branco, Fernando. "Common Value Auctions Hirshleifer, David and Titman, Sheridan. "Share
with Independent Types." Discussion pa- Tendering Strategies and the Success of
per, Universidade Catolica Portuguesa and Hostile Takeover Bids." Journal of Politi-
Banco de Portugal, July 1994. cal Economy, April 1990,98(2), pp. 295-
Bulow, Jeremy I. and Klemperer, Paul D. ''Auc- 324.
tions vs. Negotiations." National Bureau of Holmstrom, Bengt and Nalebuff, Barry. "TO the
Economic Research (Cambridge, MA) Raider Goes the Surplus? A Reexamination
Working Paper No. 4608, January 1994a. of the Free-Rider Problem." Journal of
. "Rational Frenzies and Crashes.' ' Economics and Management Strategy, No-
Journal of Political Economy, February vember 1992,1(1), pp. 37-62.
1994b, 102(1), pp. 1-23. Krishna, Vijay and Morgan, John. "An Analysis
Bulow, Jeremy I. and Roberts, D. John. "The of the War of Attrition and the All-Pay Auc-
Simple Economics of Optimal Auctions." tion.'' Discussion Paper, Pennsylvania State
Journal of Political Economy, October University, September 1994.
1989, 97(5), pp. 1060-90. Lopomo, Giuseppe. "Optimality and Robust-
Choper, Jesse H.; Coffee, John C., Jr. and Morris, ness of the English Auction." New York
C. Robert, Jr. Cases and Materials on Cor- University, Stern School of Business Work-
porations, 3rd Ed. Boston: Little, Brown, ing Paper No. EC-95-03, January 1995.
1989. McAfee, R. Preston and McMillan, John. ''Auc-
Cramton, Peter C. "Sequential Bargaining tions with Entry." Economics Letters,
Mechanisms," in Alvin Roth, ed., Game 1987a, 23(4), pp. 343-47.
theoretic models of bargaining. New York: . "Auctions and Bidding." Journal of
Cambridge University Press, 1985, pp. Economic Literature, June 1987b, 25(2),
149-80. pp. 699-738.
Cremer, Jacques and McLean, Richard. "Opti- McAfee, R. Preston; McMillan, John and Reny,
mal Selling Strategies under Uncertainty for Philip J. "Extracting the Surplus in the
a Discriminatory Monopolist when De- Common-Value Auction.' ' Econornetrica ,
mands Are Interdependent." Economet- November 1989,57(6), pp. 1451-59.
rica, March 1985,53(2), pp. 345-61. McAfee, R. Preston and Reny, Philip J. "Corre-
Daniel, Kent and Hirshleifer, David. "A Theory lated Information and Mechanism Design."
of Costly Sequential Bidding." Manuscript, Econornetrica, March 1992, 60(2), pp.
Graduate School of Business, University of 395-421.
Chicago, 1993. McAfee, R. Preston and Vincent, Daniel. "Up-
Engelbrecht-Wiggans, Richard. "Optimal Auc- dating the Reserve Price in Common Value
tions Revisited." Games and Economic Be- Auctions." American Economic Review,
havior, April 1993,5(2), pp. 227-39. May 1992,82(2), pp. 512-18.
Fishman, Michael J. "A Theory of Pre-emptive Milgrom, Paul. "Auction Theory," in Truman
Takeover Bidding." Hand Journal of Eco- F. Bewley, ed., Advances in economic the-
nomics, Spring 1988,19(1), pp. 88-101. ory. New York: Cambridge University
Grossman, Sanford J. and Hart, Oliver D. "Take- Press, 1987.
over Bids, the Free-Rider Problem, and the Milgrom, Paul and Weber, Robert J. "A Theory
Theory of the Corporation." Bell Journal oj' of Auctions and Competitive Bidding."
194 THE AMERICAN ECONOMIC REVIEW MARCH 1996

Econometrica , September 1982,50(5), pp. terest.' ' Rand Journal of Economics,

1089- 122. Autumn 1986,17(3), pp. 293-309.
Myerson, Roger B. "Optimal Auction Design." . "Value Maximization and the Ac-
Mathematics of Operations Research, Feb- quisition Process." Journal of Economic
ruary 1981,6(1), pp. 58-73. Perspectives, Winter 1988,2( 1), pp. 7-20.
Riley, John G. and Samuelson, William F. von der Fehr, Nils Henrik. "Predatory Bidding
"Optimal Auctions." American Eco- in Sequential Auctions." Oxford Economic
nomic Review, June 1981, 71 ( 3 ) , pp. Papers, 1994,46, pp. 345-56.
381-92. Wang, Ruqu. "Auctions versus Posted-Price
Shleifer, Andrei and Vishny, Robert W. "Green- Selling." American Economic Review, Sep-
mail, White Knights, and Shareholders' In- tember 1993,83(4), pp. 838-51.
http://www.jstor.org

LINKED CITATIONS
- Page 1 of 6 -

You have printed the following article:

Auctions Versus Negotiations
Jeremy Bulow; Paul Klemperer
The American Economic Review, Vol. 86, No. 1. (Mar., 1996), pp. 180-194.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28199603%2986%3A1%3C180%3AAVN%3E2.0.CO%3B2-%23

This article references the following linked citations. If you are trying to access articles from an
off-campus location, you may be required to first logon via your library web site to access JSTOR. Please
visit your library's website or contact a librarian to learn about options for remote access to JSTOR.

[Footnotes]

1
Auctions and Bidding
R. Preston McAfee; John McMillan
Journal of Economic Literature, Vol. 25, No. 2. (Jun., 1987), pp. 699-738.
Stable URL:
http://links.jstor.org/sici?sici=0022-0515%28198706%2925%3A2%3C699%3AAAB%3E2.0.CO%3B2-9

2
Value Maximization and the Acquisition Process
Andrei Shleifer; Robert W. Vishny
The Journal of Economic Perspectives, Vol. 2, No. 1. (Winter, 1988), pp. 7-20.
Stable URL:
http://links.jstor.org/sici?sici=0895-3309%28198824%292%3A1%3C7%3AVMATAP%3E2.0.CO%3B2-G

4
A Theory of Auctions and Competitive Bidding
Paul R. Milgrom; Robert J. Weber
Econometrica, Vol. 50, No. 5. (Sep., 1982), pp. 1089-1122.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28198209%2950%3A5%3C1089%3AATOAAC%3E2.0.CO%3B2-E

5
A Theory of Auctions and Competitive Bidding
Paul R. Milgrom; Robert J. Weber
Econometrica, Vol. 50, No. 5. (Sep., 1982), pp. 1089-1122.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28198209%2950%3A5%3C1089%3AATOAAC%3E2.0.CO%3B2-E

NOTE: The reference numbering from the original has been maintained in this citation list.
http://www.jstor.org

LINKED CITATIONS
- Page 2 of 6 -

6
Auctions versus Posted-Price Selling
Ruqu Wang
The American Economic Review, Vol. 83, No. 4. (Sep., 1993), pp. 838-851.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28199309%2983%3A4%3C838%3AAVPS%3E2.0.CO%3B2-0

7
Updating the Reserve Price in Common-Value Auctions
R. Preston McAfee; Daniel Vincent
The American Economic Review, Vol. 82, No. 2, Papers and Proceedings of the Hundred and Fourth
Annual Meeting of the American Economic Association. (May, 1992), pp. 512-518.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28199205%2982%3A2%3C512%3AUTRPIC%3E2.0.CO%3B2-9

7
Greenmail, White Knights, and Shareholders' Interest
Andrei Shleifer; Robert W. Vishny
The RAND Journal of Economics, Vol. 17, No. 3. (Autumn, 1986), pp. 293-309.
Stable URL:
http://links.jstor.org/sici?sici=0741-6261%28198623%2917%3A3%3C293%3AGWKASI%3E2.0.CO%3B2-F

8
Share Tendering Strategies and the Success of Hostile Takeover Bids
David Hirshleifer; Sheridan Titman
The Journal of Political Economy, Vol. 98, No. 2. (Apr., 1990), pp. 295-324.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28199004%2998%3A2%3C295%3ASTSATS%3E2.0.CO%3B2-Y

11
The Simple Economics of Optimal Auctions
Jeremy Bulow; John Roberts
The Journal of Political Economy, Vol. 97, No. 5. (Oct., 1989), pp. 1060-1090.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28198910%2997%3A5%3C1060%3ATSEOOA%3E2.0.CO%3B2-8

17
A Theory of Auctions and Competitive Bidding
Paul R. Milgrom; Robert J. Weber
Econometrica, Vol. 50, No. 5. (Sep., 1982), pp. 1089-1122.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28198209%2950%3A5%3C1089%3AATOAAC%3E2.0.CO%3B2-E

NOTE: The reference numbering from the original has been maintained in this citation list.
http://www.jstor.org

LINKED CITATIONS
- Page 3 of 6 -

28
A Theory of Preemptive Takeover Bidding
Michael J. Fishman
The RAND Journal of Economics, Vol. 19, No. 1. (Spring, 1988), pp. 88-101.
Stable URL:
http://links.jstor.org/sici?sici=0741-6261%28198821%2919%3A1%3C88%3AATOPTB%3E2.0.CO%3B2-L

28
Facilitation of Competing Bids and the Price of a Takover Target
David Hirshleifer; I. P. L. Png
The Review of Financial Studies, Vol. 2, No. 4. (1989), pp. 587-606.
Stable URL:
http://links.jstor.org/sici?sici=0893-9454%281989%292%3A4%3C587%3AFOCBAT%3E2.0.CO%3B2-9

28
Predatory Bidding in Sequential Auctions
Nils-Henrik Morch von der Fehr
Oxford Economic Papers, New Series, Vol. 46, No. 3. (Jul., 1994), pp. 345-356.
Stable URL:
http://links.jstor.org/sici?sici=0030-7653%28199407%292%3A46%3A3%3C345%3APBISA%3E2.0.CO%3B2-4

29
Rational Frenzies and Crashes
Jeremy Bulow; Paul Klemperer
The Journal of Political Economy, Vol. 102, No. 1. (Feb., 1994), pp. 1-23.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28199402%29102%3A1%3C1%3ARFAC%3E2.0.CO%3B2-Q

References

Rational Frenzies and Crashes

Jeremy Bulow; Paul Klemperer
The Journal of Political Economy, Vol. 102, No. 1. (Feb., 1994), pp. 1-23.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28199402%29102%3A1%3C1%3ARFAC%3E2.0.CO%3B2-Q

NOTE: The reference numbering from the original has been maintained in this citation list.
http://www.jstor.org

LINKED CITATIONS
- Page 4 of 6 -

The Simple Economics of Optimal Auctions

Jeremy Bulow; John Roberts
The Journal of Political Economy, Vol. 97, No. 5. (Oct., 1989), pp. 1060-1090.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28198910%2997%3A5%3C1060%3ATSEOOA%3E2.0.CO%3B2-8

Optimal Selling Strategies under Uncertainty for a Discriminating Monopolist when Demands
are Interdependent
Jacques Crémer; Richard P. McLean
Econometrica, Vol. 53, No. 2. (Mar., 1985), pp. 345-361.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28198503%2953%3A2%3C345%3AOSSUUF%3E2.0.CO%3B2-6

A Theory of Preemptive Takeover Bidding

Michael J. Fishman
The RAND Journal of Economics, Vol. 19, No. 1. (Spring, 1988), pp. 88-101.
Stable URL:
http://links.jstor.org/sici?sici=0741-6261%28198821%2919%3A1%3C88%3AATOPTB%3E2.0.CO%3B2-L

Facilitation of Competing Bids and the Price of a Takover Target

David Hirshleifer; I. P. L. Png
The Review of Financial Studies, Vol. 2, No. 4. (1989), pp. 587-606.
Stable URL:
http://links.jstor.org/sici?sici=0893-9454%281989%292%3A4%3C587%3AFOCBAT%3E2.0.CO%3B2-9

Share Tendering Strategies and the Success of Hostile Takeover Bids

David Hirshleifer; Sheridan Titman
The Journal of Political Economy, Vol. 98, No. 2. (Apr., 1990), pp. 295-324.
Stable URL:
http://links.jstor.org/sici?sici=0022-3808%28199004%2998%3A2%3C295%3ASTSATS%3E2.0.CO%3B2-Y

Auctions and Bidding

R. Preston McAfee; John McMillan
Journal of Economic Literature, Vol. 25, No. 2. (Jun., 1987), pp. 699-738.
Stable URL:
http://links.jstor.org/sici?sici=0022-0515%28198706%2925%3A2%3C699%3AAAB%3E2.0.CO%3B2-9

NOTE: The reference numbering from the original has been maintained in this citation list.
http://www.jstor.org

LINKED CITATIONS
- Page 5 of 6 -

Extracting the Surplus in the Common-Value Auction

R. Preston McAfee; John McMillan; Philip J. Reny
Econometrica, Vol. 57, No. 6. (Nov., 1989), pp. 1451-1459.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28198911%2957%3A6%3C1451%3AETSITC%3E2.0.CO%3B2-O

Correlated Information and Mecanism Design

R. Preston McAfee; Philip J. Reny
Econometrica, Vol. 60, No. 2. (Mar., 1992), pp. 395-421.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28199203%2960%3A2%3C395%3ACIAMD%3E2.0.CO%3B2-4

Updating the Reserve Price in Common-Value Auctions

R. Preston McAfee; Daniel Vincent
The American Economic Review, Vol. 82, No. 2, Papers and Proceedings of the Hundred and Fourth
Annual Meeting of the American Economic Association. (May, 1992), pp. 512-518.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28199205%2982%3A2%3C512%3AUTRPIC%3E2.0.CO%3B2-9

A Theory of Auctions and Competitive Bidding

Paul R. Milgrom; Robert J. Weber
Econometrica, Vol. 50, No. 5. (Sep., 1982), pp. 1089-1122.
Stable URL:
http://links.jstor.org/sici?sici=0012-9682%28198209%2950%3A5%3C1089%3AATOAAC%3E2.0.CO%3B2-E

Optimal Auctions
John G. Riley; William F. Samuelson
The American Economic Review, Vol. 71, No. 3. (Jun., 1981), pp. 381-392.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28198106%2971%3A3%3C381%3AOA%3E2.0.CO%3B2-T

Greenmail, White Knights, and Shareholders' Interest

Andrei Shleifer; Robert W. Vishny
The RAND Journal of Economics, Vol. 17, No. 3. (Autumn, 1986), pp. 293-309.
Stable URL:
http://links.jstor.org/sici?sici=0741-6261%28198623%2917%3A3%3C293%3AGWKASI%3E2.0.CO%3B2-F

NOTE: The reference numbering from the original has been maintained in this citation list.
http://www.jstor.org

LINKED CITATIONS
- Page 6 of 6 -

Value Maximization and the Acquisition Process

Andrei Shleifer; Robert W. Vishny
The Journal of Economic Perspectives, Vol. 2, No. 1. (Winter, 1988), pp. 7-20.
Stable URL:
http://links.jstor.org/sici?sici=0895-3309%28198824%292%3A1%3C7%3AVMATAP%3E2.0.CO%3B2-G

Predatory Bidding in Sequential Auctions

Nils-Henrik Morch von der Fehr
Oxford Economic Papers, New Series, Vol. 46, No. 3. (Jul., 1994), pp. 345-356.
Stable URL:
http://links.jstor.org/sici?sici=0030-7653%28199407%292%3A46%3A3%3C345%3APBISA%3E2.0.CO%3B2-4

Auctions versus Posted-Price Selling

Ruqu Wang
The American Economic Review, Vol. 83, No. 4. (Sep., 1993), pp. 838-851.
Stable URL:
http://links.jstor.org/sici?sici=0002-8282%28199309%2983%3A4%3C838%3AAVPS%3E2.0.CO%3B2-0

NOTE: The reference numbering from the original has been maintained in this citation list.