What Price a Settlement?
by Stephen Bate
Legal advice, commercial factors and decision-making in mediation.
Forecasting of case outcomes by lawyers is a key part of mediation. Are we going to win our legal case?
But what is the role of outcome prediction in settlement and how does it connect to decision-making: whether to settle or go to trial or the final hearing of an arbitration?
First, it is important to understand the value and limitations of a forecast, or what is commonly referred to as ‘advice on the merits.’
Most legal cases are decided on the facts. The winner is generally the party who satisfies the tribunal that its version of events is the more likely, though legal points (e.g. the meaning of a contract) may also affect the outcome.
A breakdown in trust is often at the heart of a commercial dispute. Lawyers are brought in and the correspondence hots up. Both sides argue that they are in the right. Each side’s version of events cannot be true. Parties to the same transaction, how can they have such different beliefs about what happened?
In discussions with the writer, the late Sir Brian Neill (former English Court of Appeal judge) said that when he heard the evidence as a trial judge, each side often appeared to be living in a different world.
There are several reasons for what may be called ‘the problem of two worlds’ in outcome forecasting. Limitations in human perception/recollection, the critical role of bias and the shortcomings inherent in prediction all contribute. Lack of access to information is often less of a factor. The information gap narrows as a case progresses, e.g. with disclosure of documents. But often this does not affect the parties’ perceptions of the likely winner.
In mediations, parties and their lawyers on both sides so often appear genuinely convinced of success at trial. Studies have shown that significant numbers of lawyers on each side will predict that their clients are likely to win. Bias affects not only parties and witnesses – but also lawyers. Their commitment to a goal may create bias and where outcomes (court/tribunal decisions) are seen as controllable, overconfidence may result.
Lawyers advise on the merits of the case. Their clients might be forgiven for thinking that the advice will predict accurately whether the case will be won or lost. But a lawyer does not undertake that their forecast will be accurate; only that reasonable care and skill is used in its preparation.
A forecast is not a prophecy. Advice on the merits should be given due weight in client decision-making – but not too much. In Superforecasting; The Art and Science of Prediction the authors Tetlock and Gardner stated, –
‘If a meteorologist says there is a 70% chance of rain and it doesn’t rain, is she wrong? Not necessarily. Implicitly, her forecast also says there is a 30% chance it will not rain.’
The authors referred to an official investigation into how the US intelligence services came to predict that Saddam Hussein had weapons of mass destruction when in fact he did not. The investigation was seen as balanced. It concluded that the services’ forecast was reasonable. Tetlock and Gardner used it to stress the difference between a reasonable and an accurate forecast.
The forecast of a case outcome is a judgment call based on available information that may not be reliable, particularly due to various biases at play in the information gathering and/or in the forecast itself.
Despite its limitations, legal advice on the merits is generally essential for informed decision-making by parties at mediation. Good advice excludes imprecise measures, such as a ‘good case’ or ‘a reasonable case’. Percentage prospects of success are helpful, bracketed if necessary; say 60-70%. The best advice includes all potential outcomes in the case, the (bracketed) percentage chances of each, and all assumptions and reservations.
In most cases, a party must succeed on several issues to win. Decision analytics are used in some cases to address these complexities, e.g. decision trees setting out the separate probabilities. Multiplying probabilities on each issue tends to reduce the overall prospects of success. Defendants like this approach. But it assumes that each issue or cause is independent, whereas some may be connected. To assume that a second plane crash is unlikely after a first one because plane crashes are rare, is to ignore possible connected causes. Similar considerations apply to the chances of picking the same card in two successive picks. The chances are not necessarily 1/52 x 1/52.
When splitting up the issues and multiplying the probabilities, sight must not be lost of the whole. Tribunals often decide cases by reference to, or taking into account, what they see as the overall merits.
How is advice on the merits best used at mediation to decide what settlement terms to accept or refuse? Positions are commonly informed by a party’s BATNA (Best Alternative to a Negotiated Agreement) and WATNA (Worst Alternative to a Negotiated Agreement). A party’s BATNA is generally to win the case in court and be paid under the judgment. The WATNA is the opposite. In simple cases, these measures may show clearly what settlement should be accepted.
In more complex situations, the WATNA and BATNA are not enough. They are but two parts of the picture. The starting point is to make an assessment of the value of a legal claim as an asset (i.e. a chose in action) if a claimant, and to assess the contingent liability it presents if a defendant.
Advice on the merits forms the basis of that assessment. The value of a claim may be based on percentage prospects of success; e.g. 70% x £1M (claim value) = £700,000. An allowance is also made for irrecoverable legal costs if the case were won. But the percentage probabilities and consequences of losing must also be priced in as well as other factors, such as the opportunity cost of the litigation.
Decision-making in settlement – to accept or make a particular offer or to go to trial – should take into account not only the BATNA and WATNA but each potential outcome of the dispute, commercial and legal. The percentage chances of each outcome occurring should be assessed, as well as the consequences of each.
Additional outcomes may include the cost of running the litigation, winning against one party but losing against another if there are two or more defendants, any potential damage to commercial relations, and reputational consequences and risks of non-recovery under a judgment or arbitral award if the case is won.
It is critical to take all possibilities into account. In Smart Choices; A Practical Guide to Making Better Decisions, the authors state –
Some people make their decisions based on the most likely scenario, attempting to eliminate complexity by ignoring uncertainty altogether. Without bothering to make a risk profile, they just assume that the most likely chain of events will occur, determine their best choice under those circumstances, and pursue it …. Effective decision-making takes all viable possibilities into account.
A judgment should be made as to whether each outcome would be advantageous or disadvantageous for the party’s dispute objectives, and if so, to what degree e.g. by using number values. Next, there should be an assessment of all relevant considerations, with each given due weight and a balance struck: either for or against the settlement terms.
The biasing effects of decision-framing should also be borne in mind at this final stage. Studies in behavioural economics, and in the field of litigation, show that humans facing economic choices respond differently depending on the context and framing of the decision. They are often unduly risk-seeking to avoid a loss or engage in competitive bidding to the exclusion of self-interest. A cross-check should also be made for decision-making bias, such as sunk cost bias – the inclination to go to trial where significant costs have been incurred.
The decision of whether to accept the best available settlement terms is likely to be dictated by a party’s tolerance of risk. Imagining too how the world will be, with and without settlement on the terms available, can also assist.
In most cases, circumstances will favour settlement on the best available terms at or post-mediation. But in some cases, a party’s commercial interests or an opponent’s refusal to make a realistic offer require pursuit of the legal case to a final hearing, often with a Part 36 or Calderbank Offer to provide a measure of protection for legal costs.
There are numerous models, theories and books on decision-making tools in business matters. The suggestions in Benjamin Franklin’s famous letter to Joseph Priestley written in September 1772 are now modelled in what is called a ‘T-Chart’. This is a listing of matters ‘for’ and ‘against’ alternative courses of action, a brief description of the purposes or reasoning behind each of those matters and a weighting of each. Other models include Decision Matrix Analysis, Paired Comparison Analysis, Decision Trees and forms of Cost / Benefit Analysis or models for assessing returns on investment, sometimes displayed in tabular form.
Whether or not decision tools are used, structured decision-making is critical in deciding whether to settle a case or fight it. It is hoped that the thoughts in this piece will help parties and their lawyers to get the best outcomes from their disputes.
Stephen Bate, Mediator with CEDR Chambers, Barrister and Chartered Arbitrator. For more information on these topics, see his book Resolving Business Disputes / How to get better outcomes from commercial conflicts (2020) Spiramus Press.
 See e.g. Insightful or Wishful: Lawyers’ Ability to Predict Case Outcomes, Loftus et al, Psychology, Public Policy and Law (2010), Vol. 16. No.2.
 Penguin (2016), Tetlock & Gardner p. 57.
 The Organized Mind, Viking, Harmondsworth (2014); Levitin, pp. 225 and 226.
 The Law-Making Process, 7th ed., Bloomsbury (2015); Zander p. 320.
 Smart Choices; A Practical Guide to Making Better Decisions, Harvard Business Press (1998); Hammond, Keeney & Raiffa p.156.
 See e.g. Thinking Fast and Slow (2012); Kahneman; Penguin, Harmondsworth, Appendix B ref. Choices, Values and Frames; Kahneman & Teversky.