Damian Curran

Decision analysis is a useful tool for a litigator to communicate process and risk, assess how much a case is ‘worth’, and provide a robust justification for settlement offers or acceptance. However the accuracy of the analysis is only as good as the information about the merits of a case available to the litigator at the time the analysis is conducted. That information is revealed over time, as so the output of the decision analysis based on that information will vary over time also. This post explores that observation and the implications for practitioners, clients and funders using decision analysis in litigation.

(I have previously discussed the basic concepts, the potential of using decision analysis in litigation, and links to helpful literature, here.)

The interaction between case information and expected value

The output of decision analysis in litigation is the ‘expected value’ of the case. The expected value is the sum of many smaller judgement calls on a particular limb of the case. (For example, the probability that the defendant will be found to have a duty of care to the defendant.) These questions are assigned a probability between 0 and 1. These inputs are in turn multiplied through the entire model and summed to produce a single ‘expected value’.

Whilst the expected value can be thought of as the ‘value’ of the litigation, this valuation exercise is different from the valuation of other assets. In a real estate valuation, for example, the information available to the valuer remains relatively static. It is based on long historic market prices, and somewhat objective, static data about the property itself – such as location, layout, floor plan and planning restrictions.

But litigation is far more volatile. The inputs into the expected value calculations in litigation are based on the information available to the litigator at the time of the analysis. As I explore further below, that information is not static. It is revealed rapidly as the proceedings progress. Therefore, the probability inputs, and the expected value output, should be expected to change just as rapidly.

Illustrative examples of the revelation of information during litigation

To illustrate the revelation of information about a case over the course of litigation, and how this affects its expected value, consider the hypothetical examples set out in the following graphs.

In Example 1 we see the incremental gathering of information which helps the plaintiff’s case, increasing the probabilities of successfully establishing the claims and the corresponding increase in the expected value of the litigation over the course of the proceedings. This example starts with a lawyer reviewing documents from her client for the purposes of a memo of advice prior to commencing litigation. At this stage, she judges that the prospects of establishing a breach of contract are 50%, because there seems to be acceptable arguments for and against, and without further knowledge of the other party’s alleged case, documents and witnesses, it is not prudent to give a higher assessment.

But, some months later, the client has commenced proceedings, the other party’s defence has been received, further documents have been reviewed, and more in-depth interviews with witnesses have been conducted. It is now more apparent that the allegation of breach is quite strong, there is good evidence to support it and the alleged defences are questionable. Asked again what the prospects of establishing a breach are, and the lawyer’s assessment shifts upward to 75%. There is a corresponding increase in the expected value of the litigation as a whole. As the case progresses further still, up to the eve of the hearing, there is a further upwards shift in confidence in her client’s claim and a corresponding increase in the expected value.

These adjustments are set out in the graph above. It shows is the expected value at each stage, the new information available to the lawyers at that stage, and the effect that that new information has on the probabilities and expected value outputs.

Further possibilities are explored in Example 2. In this example, a claim was initially valued quite highly, but then the prospects of success dipped when strong, new and unexpected defences were raised by the other party. These defence were then further corroborated by documentation during the discovery phase. However, during the hearing there was a favourable turn, when the witnesses for the other party made unexpected concessions which the judge appeared to look unfavourably upon. These fluctuations over the life of the case are reflected in the expected value graph for example 2.

General observation about the revelation of information in the course of litigation

We could create a multitude of similar hypotheticals. More generally though, these examples illustrate a broader observation about the nature of litigation – that its procedures cause information about the claims and defences, both helpful and unhelpful, to be revealed over time.

That revelation of information includes obviously the discovery / disclosure process, which is explicitly designed to extract such information from the opposing party. However it also takes on more subtle forms, including:

1. Pleadings and particulars, where the nature and specifics of defences and counterclaims are often brought to the attention of the other party for the first time;
2. Legal research, where strength of legal arguments are refined, often not until deep into a case;
3. Lawyer time spent considering a claim. A lawyer subsumed in a matter over many months (or years!) will inevitably develop a more nuanced view of the merits of its constituent limbs;
4. Expert evidence, where independent evidence on quantum and liability is first clarified, often deep into the litigious process;
5. Time before a judge, including preliminary hearings, where the judges views on particulars arguments or the veracity of witnesses can be tested in advance of a substantive hearing; and
6. Time with witnesses, including during evidence preparation, where the veracity and performance of a witness can be better assessed.

Practical implications

This observation – that litigious procedures cause information about a case to be revealed over time, and hence the expected value of the case changes – has practical implications for clients, lawyers and litigation funders.

Clients should remember that their case is always evolving and the views of their lawyer on the merits of their case is unlikely to be static. At different stages of the proceedings, clients should prompt their lawyers to provide updated views on settlement ranges and offers of compromise. They can use updated valuation assessments to decide whether their litigation expense strategy needs to be amended. A positive awareness of this evolution may also elicit a little sympathy for the lawyers who need to explain to their client’s senior management why their initial assessment of a set of claims has changed from what it was 12 months earlier.

Lawyers should be proactive. By properly communicating to the clients the evolving nature of litigation, they can do their client a great service. They may prompt offers of settlement at different stages of the proceedings and may encourage a client to do a revised assessment prior to, say, a mediation. They will also help guide a client’s expectations of likely outcomes.

Litigation funders may also be affected. They should ensure that they base their funding decisions on up to date assessments of merits and likelihood of success, knowing that these assessments evolve over the course of the proceedings.