Negotiations, as we’ve stated repeatedly, are a special form of exchange. In this chapter, we consider another form of exchange: the auction. An auction brings potential buyers and sellers together. For example, you can advertise an item for sale and negotiate with a potential buyer(s) who responds, or you can put that item up for auction (e.g., at eBay, or any other auction outlet). Thus, auctions are an alternative to negotiating a purchase or a sale. The difference between auctions and a for-sale listing is that auctions set a fixed transaction date, thereby increasing the competition among potential buyers or sellers. Another difference is that, contrary to a for-sale listing, auctions do not involve any direct interaction between the parties, which some may find attractive—particularly those who are uncomfortable with negotiations. But there are more issues to consider than just your level of comfort with negotiating.
In this chapter, you will learn when an auction is likely to offer advantages over a negotiation and when it does not. Understanding the potential benefits and limitations of auctions is important, especially as negotiators often have the choice of negotiating with a particular individual or engaging in an auction as a way to exchange resources.
Auctions have been around for centuries. For example, early Roman texts document the auction of war spoils, food, and household goods.1
To function properly, auctions require that items under consideration can either be fully described or inspected by prospective bidders. Think of an art auction catalogue that is distributed in advance with descriptions of the artwork and certification of their authenticity. In an auction of residential real estate, you would have the opportunity to inspect a house before bidding. Participants’ willingness to bid depends on the information about the item and the terms of sale (e.g., quantity, delivery terms, etc.) they have.
AUCTIONS
Auctions are an alternative mechanism to negotiation for exchanging goods, services, or commodities. Auctions can be open bid or sealed bid. As discussed in this chapter, there are advantages and disadvantages to each type of auction.
The most common form of auction is the open ascending first-price (or English) auction, where buyers bid openly against one another (either in person, electronically, or through the facilitation of an auctioneer), with each subsequent bid higher than the previous bid. The highest bidder wins and pays the price she bids. In a variant, the second-price auction, the highest bidder wins but pays the value bid by the second highest bidder.
In a first-price sealed-bid auction all buyers simultaneously submit sealed bids so that no bidder knows the bid of any other participant.
In an open bid declining-price (or Dutch) auction, the auctioneer begins the auction with the high asking price and moves down until a buyer accepts the price.
The behavior of bidders in auctions has been extensively studied by economists and psychologists. For example, auctions are likely to result in higher average selling prices than negotiations when the costs of disclosing information are the same for buyers and sellers, when critical information is known by both parties, when parties are risk neutral, and when there are no costs to enter an auction.2 The basic reason for this effect is that auctions are an efficient mechanism to identify the counterpart with the most extreme reservation price—those sellers with the lowest reservation price when you are buying and those buyers with the highest reservation price when you are selling. Finding multiple counterparts strengthens your position, whether that interaction is a negotiation, an auction, or securing a date to the prom! In the case of an auction, the process is designed to attract as many interested parties—buyers or sellers—as possible and to get them to compete against one another.
And it works, sometimes even better than it should! Consider the behavior of bidders in one of the common auction opportunities: eBay. On this website, some goods are available to be purchased either through a buy-it-now price or by winning an auction. The availability of a buy-it-now (BIN) option offers an interesting insight into bidders’ behavior in auctions. From a rational perspective the BIN price should represent the reservation prices for bidders. There is clearly no reason to pay more in an auction than you could pay for that same item on the same site by simply clicking the BIN button.
Yet, when researchers compared the auction prices to the BIN price for the same item on the same webpage, they found some pretty crazy behavior. In 42 percent of the auctions, the winning bid exceeded the BIN price. And the overbidding was considerable: 27 percent of bidders exceeded the BIN price by more than 8 percent, and 16 percent, by 16 percent. In fact, the average net overbidding was 10 percent over the BIN price of the item.3
To illustrate how crazy auction bidding can become, consider the case of the Chicago Cows. In 1999, the city of Chicago sponsored a public art event and auction (some live and some online) of life-sized fiberglass cows decorated by local artists. The famous auction house Sotheby’s estimated that the cows would sell for $2000–$4000 each. To everyone’s surprise, including the experts, the average winning bid for the online auctioned cows was 575 percent above Sotheby’s estimate and for the live auction it was 788 percent above that estimate.4
One explanation for this behavior is the psychological process of competitive arousal—the desire to beat your competitive counterpart, even when it means exceeding your reservation price.5 This competitive emotional state, which is stronger in live auctions than in online auctions where your competitors are typically not so salient, can lead to very expensive mistakes for bidders as they vie for corporate acquisitions, managerial talent, or even life-sized fiberglass cows.
There seems to be at least three drivers of this arousal: the presence of rivals, time pressure, and audience. In the case of rival bidders, people increase their willingness to bid more when there are more potential bidders, often violating their reservation prices in the process. From an economic perspective, this is exactly the wrong strategy. We will go into more details as to why when we discuss the winner’s curse later in the chapter, but for now, consider the fallacy involved in such behavior: your reservation price is determined by your alternatives, and there is no rational reason to assume that it would be influenced by the number of rivals or your ability to identity them. When the bidding winds down to just a few individuals, having a few but salient number of rivals increases peoples’ competitive arousal even further. Now you know exactly who it is you want to beat!
To avoid these effects, you need to be vigilant throughout the auction. Early on, large numbers of competing bidders are likely to encourage you to increase your reservation price. Late in the auction, the presence of a small number of rivals is likely to fuel your desire to win, resulting in your violating your reservation price and, as a result, getting less of what you want (of course, unless what you want is simply beating your rivals).6
A second factor in competitive arousal is time pressure. Deadlines, whether set by the auctioneer or by you, can narrow your willingness or perceived capacity to gather the necessary information you need to set your reservation price and map out your bidding behavior. There is often precious little time available between bids for you to assess the reasonableness of your next move. This pressure becomes stronger the closer you are to the (preset) end of the auction. If you have ever been involved in an online auction, you can sometimes physically experience the increased intensity of your emotional response as the seconds tick down to the end of the auction. The more the time pressure, the more aroused you become. The more aroused you become, the more likely you are to rely on past decision strategies that you have used—regardless of whether that strategy is appropriate in this particular situation.7 Thus, when time pressure is highest as an auction comes to a close, you are again more likely to violate your reservation price to insure that you do not lose the item to a rival.
Competitive arousal also increases in the presence of an audience. When there is an audience, even if they are not directly observing you, research has found that your performance on well-learned tasks improves and your performance on novel tasks declines. This is called social facilitation.8 So your competitive arousal would be higher at in-person auctions than at online auctions—just as was found in comparing the two auctions of the Chicago Cows. Knowing this effect, online auctions have often been designed to let you, the bidder, know how many people are following the auction. This is the virtual equivalent of creating an audience—and auction sites like it because it pushes you to keep raising your bid.
To control—or at least mitigate—excessive competitive arousal when you find yourself in an auction, consider these strategies:
• Redefine rivals as simply other parties who have similar interests and goals. This is not a war.
• Consider having an agent who is less emotionally invested than you serve as your bidder. Set clear parameters for the agent (or for yourself) in advance of the auction.
• Assess the appropriateness of your bidding strategy throughout the auction—if necessary by engaging a competent and trustworthy consultant—and stick to the strategy she outlines.
• Assess whether the time pressure you experience is because of a real deadline or an arbitrary one that you have set for yourself.
• Give serious consideration to whether you need to take this action now. Is it the case that there may be future opportunities and that the time between now and then may help you improve your preparation to insure a good deal?
• Rely on members of your team to help you diffuse the arousal that comes from being the one leading the charge.
• Throughout the auction, reaffirm the importance of your actual objective—getting more of what you want and reinforce what is a good deal for you.
WHY AUCTIONS?
What are the benefits that are unique to an auction? What advantages do auctions offer that negotiating may not? In addition to the psychological effects just discussed, the main reason to choose an auction is as a means to identify counterparts with the highest reservation price. Throughout the book we have assumed that you know with whom you are negotiating. But what if that were not the case? Reasonably, you want to identify the counterpart who has the most extreme reservation price. However, when considering potential counterparts one at a time, it is difficult to identify the one with the most extreme reservation price. After all, potential counterparts don’t go around advertising their bottom lines! In effect, this is exactly what auctions do.
But let’s explore exactly how this works. Assume for simplicity, that bidders (let’s make them buyers) are disciplined and willing to pay up to their reservation price (this is an important assumption, and we will revisit the implications of this assumption later).9 As a result, the winning bidder will be the buyer who has the highest reservation price: The auction process will automatically result in the buyers’ bidding against each other until the point is reached where the buyers with lower reservation prices drop out, leaving only the bidder with the higher reservation price still bidding. Once the bidding has exceeded the reservation price of the second-to-last bidder, the bidding stops, and the winner is the bidder with the most extreme reservation price. Notice that this bidder does not have to bid her reservation price; rather, the winning bid will be just a bit more than the reservation price of the second-highest bidder.
In contrast, in the negotiation situation, because the counterparts’ reservation prices are not common knowledge, you are unlikely to identify the one counterpart who has the most extreme reservation price. Thus, when negotiating the sale of an item, you might be able to push the buyer to his or her reservation price. However, you would have done even better if you had a buyer with a more favorable (to you) reservation price—an outcome for which auctions are specifically designed to achieve.
One way in which auctions also encourage parties to participate is that the auction process does not require the same level of preparation as a negotiation. You do not know who your counterpart will be in an auction on eBay, so there is little to be gained by considering the potential preferences or interests of someone who is not knowable to you. What preparation you will and should do is mostly related to the quality of the item you want to purchase or sell, your reservation price, the security of the payment process (e.g., PayPal). However, you don’t have to derive a negotiation strategy, identify the types of issues, your counterparts’ negotiation strategies, and the like.
But if auctions are so great, why have you spent the last twelve chapters learning how to negotiate? It turns out that there are some real limitations to auctions. First, auctions are not well suited for transactions with multiple issues; rather, auctions are most effective for single-issue exchanges such as price. When there is more than one issue to be considered, you still need to convert those multiple issues into a single metric. For example, you may receive multiple offers (bids) when listing your house, differing on multiple dimensions such as price, closing date, contingencies, and so forth. Thus, you still need to construct an issue-value matrix, as described in Chapter 5, that allows you to compare such diverse bids. However, such multi-issue situations complicate the auction process and make the auctions themselves more challenging. Under ideal conditions, auctions may be the most efficient means to claim value, but there is little opportunity to exchange information and, as a result, auctions are ill suited for exchanges with value creation potential. As a result, for transactions where such potential is present, you may be able to claim more through negotiations than through auctions.
Third, auctions are not well suited to situations in which the value of the item is determined by proprietary information. To be successful, auctions must provide information to potential bidders by providing detailed descriptions of the item, allowing inspections such as due-diligence in the acquisition of corporate assets. The purpose of such information is to reduce bidders’ fear of overbidding—referred to as the winner’s curse (the subject of the next section). Disclosing such information is generally in the best interest of the seller—but not always.
Consider buying a corporation—let’s say Coca Cola. To facilitate bidding, Coca Cola might consider giving potential buyers access to its trade secrets. If there were multiple potential bidders, Coca Cola’s trade secrets would no longer be secret. Thus, while providing access to its trade secrets—its primary source of value—might give bidders a better picture of the value of the company, doing so would reduce the value of Coca Cola to the eventual winner (as now many know the trade secrets), thus reducing the amount that bidders would be willing to bid.
You don’t have to have a trade secret for this to be true. When the amount of proprietary or private information is large enough, auctions lose their advantages over negotiations even when the auction is about price. Indeed, one study reports that 52 percent of companies were sold through negotiations and those that were contained the most proprietary information.10
The problem the auction creates because of the presence of multiple bidders is that simply winning the auction can itself be informative. It could be that the item is simply more valuable to the winner than it is to the other bidders. Of course, this is not a problem. However, it could be that the winner overvalued the item more than any of the other bidders and thus he was willing to bid more—and he “won”! And that is why he won. The auction reveals that he was the biggest fool in the room. This is the winner’s curse.
THE WINNER’S CURSE
The winner’s curse refers to a situation where, after winning an auction, the bidder realizes that the true value of the item is substantially more (if he were the seller) or less (if he were the buyer) than the amount bid.
To see how easy it is to experience the winner’s curse, consider a situation where the item at auction is pure common value; that is, the item has the same value to all buyers. Private-value items are goods or services that have value unique to an individual buyer or seller. The winner’s curse is a condition that occurs in common-value auctions because the differences in bidders’ reservation prices are affected by errors in estimating the true, but common, value. Thus, in a common value auction—when the item has the same value to all bidders—the winner may, in fact, be the biggest fool.
In contrast, when items with private value are auctioned differences in valuations among bidders reflect both estimation errors and differences in the underlying private values. Therefore, the buyer with the highest estimate is not necessarily the buyer who makes the highest error in estimating the value of the item. She may simply value the unique aspects of the item in a way that is different from other bidders. In such a scenario, the winner’s curse is mitigated, and bidders can bid closer to their reservation prices—confident that winning the auction does not necessarily mean that they have bid poorly.
To illustrate this point, let’s consider an extreme example, the auctioning of a sealed envelope containing cash: a pure common-value item. Obviously, before placing your bid, you would like to know how much cash there is in the envelope. Assume that the seller provides each of the potential bidders with the following information: each bidder receives an individual note from the seller reflecting an amount equal to the sum of the true dollar value plus a random number drawn from a uniform distribution with a mean of zero and a range of −3 to +3.11 So obviously the information provided to each bidder reflects the true amount of cash (since the average value added is zero), but a random error.
Your note says $6.60. What you know for sure is that the envelope contains between $3.60 and $9.60. What is the most you would bid—that is, what is your reservation price? Is it $6.60?
If bidders were to set their reservation prices equal to what their respective information says the expected value of the item is, the winner would be the one whose estimate had the most positive error. Thus, the winner of the auction is most likely to be the person whose individual report had the highest random number. In more day-to-day terms, the winner is likely to be the person with the most optimistic assessment of the value of the item, regardless of whether that item is an envelope, a piece of real estate, a company, or a fiberglass cow).
To avoid this problem, you should set your reservation price lower than your information suggests it is. But by how much? Well, if you want to be absolutely sure never to lose money, you could set your reservation price at $3.60 (that is the expected value minus the maximum potential error contained in your information). Although this strategy will avoid losses, it also is unlikely that you will win the auction. So realistically you may want to set your reservation price somewhere between $3.60 and $6.60. But where? The mathematical computation is quite involved, but we can give you some intuition about how to think about this.
First, consider that you were to set it at $4.50. Would that number change if the number of potential bidders were different? In other words, if you were to set the reservation at $4.50 with 40 bidders, would you set it higher, lower or the same if there were 4,000 bidders? Intuitively, most would either set their reservation price the same or higher, as the number of bidders increases. As it turns out, that intuition is leading you astray. The correct answer is that you should set your reservation price lower as the number of bidders increases.
Consider what appears to be an unrelated question. Who is taller: the tallest Chinese male or the tallest Swiss male? Well, what do we know about the relative height of the average Swiss as compared to the average Chinese? As it turns out, the average Swiss male is 5'9" while the average Chinese male is 5'6". So, you might be tempted to choose the Swiss option to our question. But you would be wrong. Think about the population of Switzerland (about 7.6 million) compared to the population of China (about 1.3 billion). Now, we are not interested in the height of the average male but in the most extreme (i.e., tallest) male in each of these countries. If we sample 7.6 million times compared to sampling 1.3 billion times, in which would we be more likely to find the more extreme? Clearly, we would find the tallest male among the Chinese even though as a population, they are shorter. In fact, Yao Ming who plays basketball for the Houston Rockets is reported to be 7'5", and a man from the inner Mongolia region of China, is nearly 7'9" and has been confirmed by the Guinness Book of World Records as the tallest naturally growing person in the world.12
This analogy is useful for understanding the auction of the envelope containing cash because it shows that, in that scenario, the expected value of the error of the bidder receiving the highest feedback increases with the number of bidders. So if there are only 40 of you, it is unlikely that anybody got the +3.00. After all, there are only 40 possible draws on the random numbers between +3 and −3. However, if there were 4000 draws, the likelihood that somebody got a +3.00 is much higher, approaching certainty as the number of draws from the −3.00 to +3.00 distribution increases. Thus, to avoid this problem, you should reduce your reservation price as the number of potential bidders increases. Yet, in practice, bidders expect to have to pay more—and thus set their reservation prices higher—as the number of bidders increases. And, because of competitive arousal, they actually bid more if they know that there are many other bidders.
Note that the winner’s curse is a problem no matter how relatively big or small the winning bid. Not only do people on eBay risk suffering from the winner’s curse, so do CEOs of large corporations who acquire other companies. In a study of eighty-two mergers that occurred between 1985 and 2009 that had at least two concurrent bids, the performance of the eventual winners and the eventual losers was compared in the months and years prior to and after the acquisition. There was no difference in stock-market performance prior to acquisition; however, after the acquisition, those companies that did not win the acquisition clearly outperformed the companies that won the merger contest. The most likely explanation for this result is that the winners won by paying too much. They suffered the winner’s curse, and their relatively lower stock price in subsequent months and years reflected their overpayment. In fact, over the next three years after the acquisition, losers outperformed winners by 50 percent.13
The existence of the winner’s curse makes it critical for each party to consider the insight about the value of their bid if they were to win the auction. Each party should estimate the value of the item if they were to win. For example, buyers should estimate the value of the item if the sellers were to accept their offer; and the sellers should estimate the value of the item if the buyers were to accept theirs. Remember our earlier mention of the Groucho Marx remark: “I don’t want to belong to any club that will accept people like me as a member.” If a club were to accept him as a member, it obviously accepted all sorts of riff-raff, and who would want to belong to a club that had such low standards?
In addition to those economic effects, powerful psychological effects also contribute to the winner’s curse. In Chapter 7, we discussed first offers. In general, we suggested that making an extreme first offer was a beneficial strategy. In contrast, in some situations—and auctions may clearly be one of those, it may be best to make a modest first offer.
While, on average, you gain a strategic advantage in making an extreme first offer, there are times when you can gain considerable strategic advantage by making a more modest initial offer. Probably the most common example of when you might want to make an offer that is perceived by your potential counterparts as reasonable—or even low—is when you are trying to start an auction.14 A low asking price is likely to lure in more bidders, increasing the likelihood that bidders with the most extreme reservation prices will be in the auction.
This was a strategy that many real estate agents used during the height of the dotcom mania in Silicon Valley. The residential real estate market in Silicon Valley was very hot. There were lots of buyers and very little available inventory. So, one strategy might be to list the available houses at very aggressive levels. Certainly some agents and sellers followed this strategy. In contrast, a number of other agents and sellers offered their homes at what appeared to be relatively reasonable prices. The intent of these more reasonable listing prices was to induce many buyers to make an offer.
In addition to the artificially low asking price, it was also common practice for the seller to demand that buyers submit their bids in a relatively short time horizon. Sellers would often set a specific time and date when they would consider offers and make their decisions. As a potential buyer, you were under considerable pressure to get your offer to the seller by the deadline—and to make an offer that would attract the attention of the seller. Often that offer would include not only a bid that exceeded the seller’s listing price but also such other inducements as offering the seller full-cash offers, stock options, automobiles, computers—anything that might make the bid unique and attractive to the seller. At the appointed time, some of the sellers would simply accept the most attractive offer while others would inform the buyers of where their bid stood and the numbers of bidders in contention. Then the sellers would invite the bidders to make another offer for a final round. Although some of the bidders might drop out after the first round, there were many others who improved their very generous initial offers. Those who stayed were either those with higher reservation prices, or those most affected by competitive arousal. Either way, this was good for the sellers.
While perhaps not consciously—or, more likely, with great intention—sellers were systematically lowering the barriers to entry for bidders by setting low listing prices.15 Then, they took advantage of buyers’ desire to finish what they started. That is, once a buyer made an initial bid, making the second bid was much easier. Also, given that the buyer had already indicated interest by making the first bid, a second bid seemed a much more reasonable strategy to justify the time and effort that had already been expended in making the first bid. This tendency was exacerbated by knowing that others were bidding for the same property. Not only did this arouse the buyers’ competitive juices; it also reaffirmed that other people found the property attractive.
You are considering participating in an auction. What should you do? When should you avoid getting into an auction? You should avoid an auction when there are many bidders on your end of the deal, when there are multiple issues that you and your counterparts can value differently (a large integrative potential), or when your desire to win is extreme. On the other hand, when should you consider an auction? When there is only a single issue—particularly price; when there is little risk of disclosing information to potential bidders that will affect their assessment of the value of the item, when preparation time is extremely limited (allowing you to only focus on your reservation price), when there are a large number of parties on the other side, and when the identity of the ideal counterpart is unknown. In such scenarios, auctions can be extremely beneficial to you.
SUMMARY
Under specific conditions, auctions can get you more of what you want. First and foremost, you should consider auctions if you have a particularly strong aversion to negotiation, cannot identify counterparts that are likely to have a high reservation price (if you are the seller) or a low reservation price (if you are the buyer), or simply do not have the time to prepare adequately and go through the negotiation process. But still, auctions have their limitations.
• Auctions are very effective in negotiations when there is only a single issue such as price, the product or service is fairly standardized, or the product or service has broad appeal. In those circumstances, both the economics and the psychology of the auction process work to your favor: consider attracting multiple bidders with a low opening price (you can specify a reserve to mitigate the downside if too few bidders show up), set a reasonably tight deadline, and favor an open outcry auction process to maximize the psychological factors to maximize competitive arousal.
• Auctions are less well suited to situations that have a large (and complex) value-creation potential, and when the due diligence that bidders require involves proprietary information that if made broadly available would reduce the value of the object to the eventual winner. In those situations, negotiations are likely to get you more of what you want.
• When participating in an auction, beware of competitive arousal. Just as in any negotiation, set and respect your bidding limits. As a general rule, such limit should only be revised if you learn something that you could not have known when preparing for the auction.