From Bargaining Games, Copyright 1992, William Murrow and Company
Used by permission of the author.





People at an auction are bidding for a $20 bill. The rules require that the bids increase by a dollar each time. They started at $1 and went rapidly through $2, $3, $4, et cetera. At $15, things slowed down a bit, and only two people are now bidding. They go to $16, $17, $18 ... $19 ... $20,long pause ... $21, $22, $23. What is going on here? Why would anyone ever pay more than $20 for a $20 bill?

The bids keep going, past $30, $40, $50, $60, $70; the same two people are doing all the bidding, back and forth. All for a $20 bill! This is not a group of lunatics. (It's also not for charity.) Instead, it's a group of executives at a bargaining seminar -- and this is not such an unusual group; most groups end up bidding much more than $20 for a $20 bill. Groups regularly bid over $50, and sometimes over $100.

In the Dollar Auction, the highest bidder pays whatever he or she bid and receives the $20, just like at any other auction. The trick is that the second highest bidder also pays whatever he or she bid, but doesn't get anything. Thus, the top two bidders both pay what they bid, but only the highest of the two bidders gets the prize.


Like most auctions, the Dollar Auction generates a lot of interest for both bidders and nonbidders. In the old days, when a dollar was worth something, I ran the auction for a dollar, with no restrictions on the bids except that they exceed the previous bid. The bidding could start slowly, with early bids of a nickel, a dime, a quarter, before someone gets the idea of bidding 99 cents. Sometimes it went much faster, with someone bidding 99 cents right away.

For some reason the 99-cent bidders are almost always male. After their bids, they sit there with big, self-satisfied grins on their faces; they are clearly very proud of their little coups. For a minute, everyone else gives them a chance to bask in the glory of their creative ploy.

Invariably, however, someone bids $1, making the 99-cent bidder's smile instantly disappear. The new bidder has punctured the proud, 99-cent bidder's bubble and simultaneously presented him with a dilemma: He'll lose 99 cents if he lets things stand, but he'll have to bid more than $1 to win the $1 prize. While he's thinking about his choices, the new bidder is enjoying the fact that he has wiped a big smile off his competitor's face and everyone else buzzes with excitement. They can't help but appreciate the new bidder's ego-deflating maneuver.

The original 99-cent bidder invariably bids $1.01. Everyone breaks up laughing at the thought that someone has just bid more than $1 to get $1. Then the auction settles down to a test of wills between the last two bidders, who must negotiate how much they will lose.

They face three issues: (1) How much can I afford to lose? (2) How can I look tough enough to scare the other bidder out of bidding more? And (3) how do I get out of this predicament without looking like a total fool? One person trying to look tough often leads to the other bidder trying the same strategy. Then both reap greater and greater losses (financial and reputational) than they would if they had just stopped bidding. The only winner is the auctioneer -- me. (My favorite charities win, too.)

The bidders tend to be much more concerned with pride than money. In a group of executives several years ago, the two last bidders were the president of a real estate firm and a local judge in the federal court system. The real estate man had just completed a wonderfully profitable year; the judge collected old Mercedes-Benz automobiles. Clearly, the costs of this game (I was only auctioning $1) were insignificant to both of them.

Their bidding proceeded quite rapidly, with each trying to convince the other that he wouldn't back down. It was clear that neither of them wanted to be shown up in the bidding.

At one point one of them bid $3.00. The other's next bid was $4.25. 1 couldn't help pausing at this point, even though that is not what a good auctioneer should do. If both bidders were strictly rational, the $4.25 bid should have ended the auction. It exceeded the previous bid by more than $1.00; thus, the other bidder would only lose more by bidding any further. Instead, the $3.00 bidder almost immediately bid $4.50. Needless to say, that brought a roar from the group. In fact, the $4.25 bidder reported later that his bid was not meant to be a rational, let's-stop-the-auction-here-and-I'll-lose-more-than-you-do bid. Instead, it was just another attempt to look tough by raising the bid that much more.

In large groups (twenty or more), I always auction off several dollars, one right after the other. People have a very hard time trying to band together and collude. A group of thirtythree people, for instance, would realize a profit of only 3 cents each if they stopped the bidding at a penny and shared equally in the winnings. Someone is always tempted to bid slightly higher, and once one person bids 5 or 10 cents, everyone else is reluctant to let the defector win. Thus, someone else will also bid, especially when the auctioneer controls the bidding.

All it takes to win money is to have two people bid 50 cents or more. Bidders usually don't realize that -- they're too emotionally involved. Even the nonbidders tend to focus their attention on bids that exceed $1. Seeing someone bid more than $1 to win $1 is a perverse pleasure for almost everyone. But from the auctioneer's perspective, once two people bid more than 50 cents, the rest is all profit.


After the first auction, everyone sees that the second highest bidder always has an incentive to bid more than the highest bidder; the potential gain of the $1 bill offsets the cost of increasing their bid. Add a little pride and it's clear that backing out can be very difficult. Thus, many people realize that the game itself establishes forces that encourage the last two bidders to continue bidding. The optimal strategy in this game is not to bid at all -- unless you are the only bidder, or no one is willing to go beyond low bids (a truly uncertain prospect).

Second and third runs of the auction continue to generate profits (for me). People who lose money in the earlier auctions often bid again in hopes of winning back some of their losses. As we have seen with other games, one reason for this ineffective strategy is emotional: People who get involved in the bidding are more emotionally caught up in the task and have a harder time seeing its ramifications. People who haven't bid in the first auction usually don't bid in subsequent auctions. Not getting involved the first time gives them an edge in subsequent plays. They affirm one of Abraham Lincoln's famous quotes: "It is better to keep your mouth shut and be thought a fool than to open it and remove all doubt."


People often get into things and suddenly find that they're in over their heads. The Dollar Auction is about overinvolvement, how just getting started can sometimes lead to getting more and more and more involved. The United States defense system provides many examples of this phenomenon: New weapons are sometimes outdated by production time. Private-sector organizations suffer from the same problem. Many computer firms, for instance, have found that they get enmeshed in deciding whether to invest in a new product and, by the time it has been designed and produced, another firm has beaten them to the market with a product that's more advanced than theirs. This phenomenon also afflicts individuals. One scenario that might be familiar to many people is the overinvolvement that sometimes follows an initial step toward romance. Andrea's story provides a simple example.

Andrea is a young vice-president at one of the major banks in town. She's not married and is not seeing anyone regularly. She is at a cocktail party and finds that her eye is caught, almost involuntarily, by a man across the room. It also appears that she has caught his eye. Eventually they strike up a conversation; they seem to find each other attractive and interesting. He's not seeing anyone either, so when Andrea asks herself whether she should see him again, her answer is an easy one: "Why not?" So they go out for a drink. Then they go out for dinner and then a movie and all of a sudden they are seeing each other on a regular basis. While that is happening, a voice inside Andrea's head occasionally tells her that even though she's having a good time, she's not having a really great time. (He may be hearing the same kind of voice.) It's not a loud voice, so she doesn't pay much attention to it. Before you know it, though, Andrea is romantically involved; she's even beginning to feel the tug of a commitment. She discovers that it's easier to see him than to try to meet someone new. With no trouble at all, and no conscious decision on her part, Andrea has found herself involved in a romance -- but not an overwhelmingly exciting romance. She gradually gets more and more involved and it becomes harder and harder to extract herself from this relationship. She meets some of his friends; he meets some of hers; everybody soon expects to see them together at social events. If either of them is alone, friends ask where the other is.

As all this develops, Andrea's Romantic Pain Potential (RPP) increases; at the same time, her Romantic Gain Potential (RGP) drops. Before too long, RPP dominates RGP. A difficult ending becomes the next logical step. Andrea's story is not happy, but it has repeated itself often, for many, many people. And some of the dynamics of her romantic experience (not all of them) are nicely captured in the Dollar Auction.

Not surprisingly, research has shown that people who set limits for themselves before beginning the Dollar Auction do better. They are not always successful at being restrained by their stated limit, but they tend to bid less than people who don't set limits. The problem with romances and other adventures is that it's hard to pause and set limits ahead of time. Instead, we hope for everything and pay less attention to the downside.


Barry Staw, a professor at the University of California at Berkeley, and his colleagues have conducted several research projects to investigate how commitment to a failed course of action can escalate. Business students read the fictitious case of the Adams and Smith Company, which has had a long, profitable history but now looks like it's in trouble. Acting as financial vice-presidents, students are asked to invest $10 million in either the company's Consumer or Industrial products division.

After they have made their decision, time is accelerated to five years later. Half of the VPs are told that their original decisions were successful: The division that received the extra funds had turned itself around and looked as if it would continue to be successful for some time. The other VPs are told that their decisions were not successful: The division that had received the influx of money was doing worse than ever. After this news, they must invest up to $20 million more in either of the two divisions, Consumer or Industrial products; this time they can divide the money any way they like.

People whose previous decisions were unsuccessful pour more money into that same, failing division. Not only that, when they also face a serious personal threat (that is, losing their job as financial VP), they commit even more to the unsuccessful division. Commitments do not escalate, however, when decision makers know that they will be throwing good money after bad. Only when they have hope of escaping the costs of previous poor decisions do decision makers escalate their commitments.

People escalate and commit themselves to a course of action when (1) they think that future gains are possible, (2) they are optimistic that they can turn a project around, (3) they're publicly committed or identified with the project, or (4) they can get much of their investment back if the project fails.

Most of these forces come into play in the Dollar Auction Game. Bidders drop out if they don't mind taking a loss or if they are less ego-involved in the bidding and don't feel that they have to win at all costs. They continue when they think they can still gain by bidding, and when they think they can still be a winner.

Escalating commitment is insidious -- it creeps up on you and, before you know it, you're much more committed than you want to be. Stories of how innocent people have gotten involved in the Mafia are an example drawn from the popular media. The history of the Vietnam War is another example: Once the first serviceman was killed overseas, America's involvement in the war escalated. With each successive casualty, involvement increased more, even though early intelligence reports indicated that waging a successful war in Southeast Asia was highly unlikely.

Investment fiascoes provide numerous examples. People often buy a stock when they think it's a good deal and, when it goes down, they often buy more of the same stock, since now it's an even better deal. Commitment and investment can easily escalate, and mean serious financial losses.

There are times when you get involved in something and, all of a sudden, you find that you have become over involved. It's a kind of commitment that seems to have a life all its own; you don't really feel like you're in control of the situation at all. Examples abound. A friend once told me this true story: A couple he knows moved to a new town and joined the synagogue in their neighborhood. The couple felt like they had paid quite a lot to join. They began going regularly to the synagogue's social get-togethers and, after attending several, complained that they were having a terrible time, week after week. When asked why they didn't stop going, they replied, "Well, we paid all this Money to join. We don't want it to go to waste."


The Dollar Auction is unlike typical auctions. If we simply auctioned off a dollar bill, the bids would certainly go to 99 cents. Someone might even bid a dollar if they got some value from winning. But the introduction of a second bidder who must also pay contributes to a tremendous increase in the bids.

The key elements that lead to escalating commitment are (1) free choice at the very beginning and (2) negative consequences, which (3) you think you have the opportunity to overcome. It's hard to stop a commitment that's escalating.

Several years ago I owned a sporty, imported car. It was a wonderful car to drive. Unfortunately, it was not a wonderful car to fix. When it had accumulated a few miles, it needed more and more repairs. The difficulty was that my commitment to the car escalated every time I paid a new repair bill. I freely chose to buy the car and to have it fixed when it broke down. Thus, I fulfilled step number one in escalating commitment. When things broke, I paid quite a bit to get the car in working order again. I implicitly thought of each of these repair bills as investments ("Now that I've fixed this, it's bound to run well."). The more I spent, the more committed I became. Negative consequences, element number 2, were more and more fulfilled the more I paid for repairs. The escalation continued, of course, because the longer I kept the car, the more it broke down, the more I paid for repairs, and the more committed I became.

Finally, I hit my mileage limit for a car -- after a car hits 80,000 miles, I feel like I should sell it. I have a hard time trusting the car to work on long trips when it has so many miles, so I invoke my informal rule that this is the time to sell. Thank goodness for the rule -- without it, I may have gotten even more committed to that car. In the end, I developed a real love-hate relationship with it, possibly like Andrea and her romance or investors and their failing stocks.

The moral of this story is to beware of escalating commitment. It's a disease that can creep up and grab you and not let go. The crux of the problem is the commitment's sunk costs:

Once we've invested, whether it's time, effort, or money, we value the investment and have a tremendously hard time abandoning it. It may be water over the dam or spilled milk, but if we feel that the water or the milk cost us a lot, we have a hard time ignoring it. We don't think of sunk costs as sunk. So we're susceptible to pouring good money (or effort or time) after bad, It's very hard to avoid.

One last anecdote about the Dollar Auction. I described the Dollar Auction to a German friend who is a dealer in antique books and prints. He attends art auctions all the time. In fact, that is a major part of his work; he buys and sells at auctions around the world. He said that the rules of the Dollar Auction are often used by German charities. That way they raise more money. He also said that they call it "An American Auction."