TRANSACTION COSTS
The expected transaction costs of securing final adjudication are the third critical element of each party’s assessment. A plaintiff must subtract from any possible recovery his likely costs of going to trial.6 Similarly, a defendant must consider not only the amount of any possible judgment but also the costs of defense. What legal fees will each side incur? What court costs? How might the lawsuit affect each party’s ability to operate its business efficiently? For example, if the hotel’s managers and staff are spending time in depositions rather than running the hotel, how much will that interfere with their work and to what extent will the hotel’s business suffer? What emotional or psychological costs could litigation have on each party? Only by understanding the costs of litigating can the parties know the net expected value of proceeding to court, and work with that expected value at the negotiation table.
In Tom’s case, Jennifer explains that litigation could get fairly expensive. Both sides might hire medical experts to testify about Tom’s concussion and its likely future effects. Jennifer would have to depose various hotel employees to try to establish a pattern of leaving trash on the loading dock. If the hotel contests the case vigorously, the litigation might drag on for a long time.
RISK PREFERENCES
Risk preferences also can affect how a party decides whether to settle or proceed to trial. Suppose that a person faces a coin toss in which he has a 50 percent chance of winning $100 and a 50 percent chance of receiving nothing. The coin toss has an expected value of $50. If the person is risk neutral, he will view a 50 percent chance of winning $100 as equivalent to receiving $50 up front. If he is risk averse, he will accept less than the expected value up front—say, $48—in order to avoid the possibility of receiving nothing. Litigants tend to be risk averse because so often the stakes are very large, but a particular disputant may in fact be a risk-preferrer. A gambler, for example, might prefer the chance of winning $100 to the certainty of receiving $55. He would prefer to gamble on getting it all, rather than settle for a lesser amount, even if the lesser amount is more than the expected value.
How do risk preferences affect negotiation? To the extent that both parties are risk averse, the zone of possible agreement is broadened because each would be prepared to accept less or pay more than the net expected value in order to avoid the gamble of going to court. Other things being equal, this should make it easier to settle the case. On the other hand, to the extent that one or both parties are risk-preferrers, the opposite might well be true.7
Figure 5
The Basic Model
This simple model—based on substantive endowments, procedural endowments, transaction costs, and risk preferences—explains why most legal disputes are settled and not adjudicated.8 When the parties have similar expectations about the opportunities and risks of proceeding to trial, settlement saves the transaction costs that would be spent securing a formal adjudication. In essence, these savings are a surplus that can be divided between the parties through settlement.
To understand how transaction costs can create such a surplus, let us simplify Tom’s case. Assume that both sides know that if Tom goes to court a judge will award him $70,000. This is a highly unrealistic assumption, as we’ll show shortly, but for now let’s assume that both sides can know this in advance. Assume further that Tom and the hotel will each have to pay $10,000 in otherwise avoidable transaction costs if the case goes to court. What will their negotiation be about?
Under these assumptions, Tom is better off settling if he receives any amount greater than $60,000 ($70,000 minus $10,000). The hotel is better off settling as long as they don’t have to pay more than $80,000 ($70,000 plus $10,000). Therefore, any settlement between $60,000 and $80,000 would make both parties better off than having the case tried. In essence, they have a $20,000 surplus to divide (see Figure 5). If they settle for $70,000, each saves its own costs. (One or the other may try to hold out for a more favorable settlement, however, and they could end up playing a game of chicken. Tom might say, “I won’t take anything less than $75,000. Otherwise I’ll go to trial.” Although the hotel would be better off paying $75,000 than going to trial, if it knew that Tom would only end up with $60,000 if the case were tried, it might not find his threat very credible.)
Why Some Cases Shouldn’t Settle
Of course, some cases shouldn’t settle: those rare cases in which a party’s interests can be served only by a complete victory, either in court or by capitulation of the other disputant. Sometimes a party’s interest in public vindication is so strong that it cannot be met without adjudication, and that interest may outweigh whatever tangible settlement options the other party can offer. Sometimes a party has a strong desire to create a lasting legal precedent in a certain area and is using litigation as a means to that end. In civil rights litigation, for example, test cases may be brought to challenge or create legal doctrine. Or in a patent dispute, a company may need to demonstrate the validity of its intellectual property to protect its core business. In such cases, the defendant may not be able to offer anything that would be better for the plaintiffs than litigating to judgment.
Sometimes a party may refuse to settle a case because it wants to establish a reputation that will deter future litigation. For example, for several years Ford Motor Company has made one take-it-or-leave-it offer to plaintiffs, correlated to Ford’s valuation of the plaintiff’s claim. If the offer is rejected, Ford litigates.9 The company would rather defend those lawsuits and establish a reputation for being willing to fight than overpay for frivolous claims. Over time, the company believes its strategy will pay off with lower total legal expenses and payments.
Finally, some cases don’t settle because one or both parties is using the suit for larger strategic or corporate ends. In some corporate takeover situations, for example, the target company will file a lawsuit in an attempt to deflect or defend against a hostile takeover bid. The goal is not so much to win the battle as to win the larger war for control of the company. The suit itself may be over some relatively insignificant thing, but the target company uses the suit to drop the share price and block the takeover. The parties aren’t likely to settle in such instances.
LITIGATION DYNAMICS
The basic economic model of litigation and settlement explains why most cases settle: if the parties’ expectations about the value of going to court converge, why bother actually taking the case to trial? And most cases do settle, as we have noted. But the settlement process is typically very inefficient, for two reasons.
First, even when cases settle, they often settle late rather than early, and this leads to unnecessarily high transaction costs. Legal disputes become trench warfare rather than exercises in problem-solving. Each side takes extreme positions and refuses to compromise, even though each side knows that ultimately a settlement is likely. Time is wasted, relationships are damaged, and in the end the case is still settled on the courthouse steps. By that point the parties have already spent a great deal on the dispute resolution process.
Second, the settlements reached in the litigation process typically ignore the possibility of finding value-creating trades other than saving transaction costs. Although the litigation game includes the evaluation of the legal opportunities and risks, it does not usually incorporate a broad consideration of the parties’ interests, resources, and capabilities. As a consequence, the parties may never discover possible trades that could have left both sides better off.
The Distributive Challenges
The distributive aspects of bargaining often preoccupy disputing parties. The litigation game is complex and fluid. The two sides seldom have perfectly convergent expectations about the value of going to court, and each is constantly trying to influence the other’s perceptions of that value through moves and countermoves in the litigation process. Because successfully shaping such perceptions can confer real distributive benefits, the parties may escalate the conflict and ultimately become locked
into an adversarial and destructive dynamic that neither can then easily change unilaterally. The result is a war of attrition rather than a search for ways to resolve differences efficiently. Here we explore these strategic complexities that can often keep litigants from creative problem-solving.
UNCERTAINTY ABOUT THE OUTCOME OF LITIGATION: “WHAT’S THE CASE REALLY WORTH?”
The first complication is that litigants cannot know with certainty what a court will do in a given case. Unlike our simple example with Tom, in most cases neither side is sure about what the plaintiff would actually receive if the case proceeded to trial. But even with such uncertainty the parties can have similar assessments of the probability distribution of possible outcomes. For example, before the trial neither side can know with certainty that a jury will find the Big Apple Hotel negligent. But they might nevertheless agree on the probabilities.
Consider the simple example illustrated in Figure 6. Both Tom and the hotel expect that there is a 30 percent chance of the jury deciding that the hotel was not negligent, which would lead to no payment ($0) by the hotel. They also agree that there is a 70 percent chance of the defendant being found negligent and paying $100,000. They share expectations about the case, and thus they should agree that the value of the outcome is $70,000, not considering transaction costs. (Multiply each probability by its associated outcome and then add the results. Thirty percent multiplied by zero is zero; 70 percent multiplied by $100,000 is $70,000. Thus, the expected outcome is $70,000.) The expected value is simply the sum of possible outcomes, each weighted by the odds that a particular outcome will in fact be the result.
Of course, often the parties do not perceive the relevant probabilities identically. Instead, each has his own expectations about the likelihood and consequences of various trial events occurring, and each has a private estimate of the value of the expected outcome. The parties’ differing decision trees could look something like Figure 7. There, each party has a different assessment of the probability that the hotel will be found negligent. Although in this example their expectations about damages are similar, their sense of the value of the expected outcome differs dramatically: Tom thinks the case is worth $90,000, and the hotel thinks the case is worth only $10,000.
Figure 6
Figure 7
In most cases, this sort of decision tree will have many branches and sub-branches. Will the hotel be found negligent? Will Tom be found contributorily negligent? If so, what is the comparative negligence of the two parties? What is the range of possible damages? What is the probability of each award? For the parties to reach convergent expectations about a trial outcome, they will need to discuss their differing answers to these questions and figure out ways to bridge the gaps between them.
The parties may have divergent expectations for a number of reasons. First, they may know different facts. Tom may know the name of a person who observed how careless the hotel’s employees were with trash around the loading dock. The insurance company lawyer, at least initially, may not know about this potential witness. Of course, before the trial takes place, through discovery and disclosure, this should become common knowledge. But in some circumstances one or both parties may hold private information that influences their expectations about litigation. Second, and more commonly, the parties may have different interpretations or perceptions of the same facts. Third, they may have different assessments of the relevant law and how it would be applied to the facts. Finally, especially with respect to damages, a jury has a great deal of discretion. The outcome can turn on the jury’s subjective impressions of the people involved in the case.
In Tom’s case, consider the following uncertainties, about which the parties may have very different views. Is Tom lying about the existence of a newspaper on the loading dock? Whether he is lying or not, is a jury likely to believe him? Will he be a credible witness? What about Tom’s conversations with the hotel manager, alerting him to the recurring problem of trash on the loading dock? Suppose the manager denies that such conversations ever occurred. Who will the jury believe, Tom or the manager? How will the manager fare as a witness? And what medical expenses is Tom likely to incur in the future? What damages for pain and suffering is a jury likely to give Tom if they find in his favor? Because the range of possible outcomes is very broad and there is no easy way to assess the odds of particular outcomes, opposing counsel can reach very different conclusions about the expected value of the case.
INFLUENCING PERCEPTIONS IN A DYNAMIC GAME: “YOUR CASE IS A DOG”
Litigation does not simply involve the dispassionate assessment by each party of a fixed set of facts under a given legal regime. Instead, it is a dynamic process in which counsel for each side is constantly trying to shape the other party’s perception of what might happen at trial. The odds are not fixed at the outset. Because of moves and countermoves during the pretrial process, an astute lawyer may be able to improve her client’s chances of winning: new documents may be discovered; a partial summary judgment motion may be granted throwing out part of the case; careful questioning during a deposition may undermine a witness’s credibility; a persuasive expert witness may be hired.
In addition to working to change the odds, each side tries to influence the other’s perceptions of the likely outcome, and what might be an acceptable settlement. Puffing and exaggeration are commonplace. Parties stake out extreme positions, hoping to signal their confidence and expectations. Lawyers will often attack and belittle the other side’s case, trying to shift the other side’s subjective assessment of the litigation. These litigation moves are a common part of the negotiation process.
Changing the other side’s perceptions of its litigation alternative has important distributive consequences. Imagine that Tom’s lawsuit against the Big Apple Hotel proceeds to the discovery phase and that Jennifer uncovers a document in the hotel’s files showing that the original blueprints for the construction of the loading dock called for a reinforced safety railing. In all likelihood she will bring this information with her to her next negotiation with the hotel and wave it in front of the other side’s attorney, hoping to influence the hotel’s perception of Tom’s likelihood of success should the case go before a jury. Will she succeed? Who knows? But most likely she will try.
We cannot overstate the importance of these dynamics. Lawyers and clients constantly base their negotiation strategies on the possibility that a litigation move will change the value, or the perceived value, of the case at hand. The litigation process can become all-consuming, with lawyers and clients focusing exclusively on influencing the possible outcome of the case and ignoring what it will cost in real dollars to do so. Moreover, the fluid nature of the game makes the process of valuing the net expected outcome of a legal dispute very difficult. Litigation is not a game where you know the odds—where you pay a dollar to flip a coin for the chance at two dollars. Instead, it is a game that often feels—at least to clients—as if you must pay some undisclosed and uncertain amount in order to have some undisclosed and uncertain odds of winning an uncertain prize.
INFLUENCING TRANSACTION COSTS: “WE CAN HURT YOU WORSE THAN YOU CAN HURT US”
As we have seen, a party’s reservation value in a legal dispute depends, in part, on the transaction costs that the party thinks he is likely to incur if he proceeds with litigation.10 If the other side can change these costs, it can alter the perceived net value of going to court.
Some transaction costs are fixed, such as the filing fee a court imposes for initiating a claim. But many transaction costs are linked to the parties’ behavior. For example, various litigation expenses—including attorney’s fees, deposition costs, discovery burdens, and other out-of-pocket expenses—can vary widely depending on what the other side chooses to do. This can be critically important. In litigation, one party may be able to impose substantial transaction costs on the other at very little cost to itself. In discovery, for example, it might cost Attorney A very little to send Attorney B a long list of
written interrogatories, but it may take Attorney B hours of work to answer them. Of course, Attorney B could retaliate with a similar list, thereby imposing costs on Attorney A. But at the start of a lawsuit neither side knows what choices each side will make along the way, and how those choices will affect transaction costs.
The result may be a war of attrition. The goal is to impose such a great burden that the other side gives in, but the reality can be that both sides stagger under the weight of mounting transaction costs that lead ultimately to a lose-lose outcome. Although starting the war in hopes of winning it might be a rational move for either side, the collective outcome is irrational.
The temptation to wear the other side down may be especially great when the parties face different costs or have different resources. An elegant experiment conducted by Richard Zeckhauser illustrates what can happen.11 Zeckhauser asked subjects to divide $2 between them. In the event they failed to agree on a split, neither party would receive any money. To no one’s surprise, in this version of the game virtually all of the pairs hastened to split the $2 evenly. But in a second version of the game, the subjects again were asked to divide the $2, but for every minute that elapsed one party was taxed five cents while the other party was taxed ten cents. In this situation, most people intuit that the bargainer taxed five cents a minute has a degree of leverage over his counterpart by virtue of these asymmetrical costs. And while many pairs in this version will quickly agree to split the $2 evenly, many more do not. Often, the bargainer taxed five cents a minute will try to exploit his apparent leverage by asking for more than $1. What is fascinating about Zeckhauser’s results in this version of the game is that the party with the purported leverage typically does worse on average than when the parties face symmetrical costs or no costs at all. The attempt to exploit asymmetrical costs often induces stubbornness by the other side, and both sides typically end up with much less than $1.
Beyond Winning Page 13