Inside arbitrators' minds

20 January 2011

Manuel Conthe

Christopher Nolan's "Inception" gets "inside the architecture of the mind". Can we do the same with arbitrators?

Arbitrators are as likely as any other profession to fall victim to “cognitive biases” that influence their decision making, says Manuel Conthe, an arbitrator and of counsel at Bird & Bird in Madrid.

Can inflated demands by claimants increase the damage awards granted by arbitrators? Can arbitrators, once an accident or mishap has happened, exaggerate its predictability and overestimate the standard of care that respondents should have applied to avoid it? Does providing identical additional information to claimant and respondent make them less likely to settle?

Most arbitrators would probably reply negatively to all three questions. But unless we regard ourselves as superior to professional judges and other professions, there is a good chance that these answers may be wrong: arbitrators and parties alike may fall prey, inadvertently, to a number of empirically documented ’cognitive biases’.

The seminal work on how people make decisions and commit systematic mistakes was carried out in the 1970s by two Israeli psychologists, Amos Tversky and Daniel Kahneman. Tversky died in 1996 but Kahneman was awarded the 2002 Nobel Prize for Economics for their collaborative work. Their experiments confirmed that, just as we suffer well-known ’optical illusions‘ (for example the Müller-Lyer illusion, in which two segments of identical length are perceived shorter or longer depending on whether they are drawn with arrows or forks at the end), our abstract thinking and decisions are also distorted by ’cognitive biases‘. When Kahneman and Tversky’s approach was embraced by financial economists, ’behavioural finance‘ was born; and when its scope became broader and more interdisciplinary, ’behavioural law & economics‘ emerged.

In a recent programme presented at the International Bar Association’s Annual Conference in Vancouver at the initiative of US arbitrator José Astigarraga, I mentioned three cognitive biases that stand out as potential threats for arbitrators: ’anchoring‘, ’hindsight bias‘ and ’confirmation bias’.

The anchoring effect: the more you ask the more you get?

In an experiment described in their famous study “Inside the Judicial Mind”, Guthrie, Rachlinksi and Wistrich presented a notional personal injury federal lawsuit - in which the plaintiff had been struck by a truck owned by a package delivery company from a different US state - to 116 federal judges attending a conference. They described how, after being hospitalised for several months, the plaintiff was confined to a wheelchair for life, unable to use his legs. He had requested damages for lost wages, hospitalisation, and pain and suffering, but had not specified an amount.

To one group of 66 judges the researchers posed a direct question: “How much would you award the plaintiff in compensatory damages?”

To a second group of 50 judges they gave additional information: “the defendant has moved for dismissal of the case, arguing that it does not meet the jurisdictional US $75,000 minimum for a diversity jurisdiction case (that is, a case featuring parties from different US states)”. After asking this group to rule on the motion, they posed the question: “If you deny the motion, how much would you award the plaintiff in compensatory damages?”

Since under US standards, the personal injuries sustained by the plaintiff were, quite plainly, compensable in an amount well in excess of the US$75,000 minimum, the defendant’s motion to dismiss federal jurisdiction should have been considered irrelevant in determining the actual compensatory damages granted to the plaintiff. In practice, however, the first group of judges awarded the plaintiff average damages of US$1.25 million, while the second awarded a significantly lower US$882,000.

The experiment confirmed the so-called ’anchoring effect‘: when making estimates of any variable (for example, the amount of compensatory damages or market value of a house), we are inadvertently attracted to any figure or ’anchor‘ which looms prominent in our mind, even if totally irrelevant or unrelated to the task at hand.

Thus, although many professional judges and arbitrators may subscribe to the ’boomerang theory‘ - over-the-top demands backfire, since they make claimants appear greedy - empirical research lends credence to the prosaic view that “the more you ask, the more you get”.

By the same token, one of the speakers at the Vancouver programme, US arbitrator David W Rivkin told Global Arbitration Review afterwards that “I am always pleased if an article appears in GAR that shows arbitrators that the client I represent is not the only claimant out there seeking billions of dollars in damages.”

Hindsight bias or ‘I knew it all along’

Abundant research has demonstrated the difficulty of assessing ex-post (after a crisis or event has taken place) the level of care or information that would have been regarded as reasonable ex-ante (before the event). We have a spontaneous tendency to overestimate the predictability of past events once they have occurred. This ‘hindsight bias’ -  sometimes referred to as ’Monday morning quarterbacking‘, ’20/20 hindsight vision‘ - does not refer to the fully rational process of using known outcomes to update our beliefs about future events, but describes the tendency to use these outcomes to assess their predictability at some earlier time, before they took place. In other words: "I knew it all along".

Hindsight bias may lurk in the traditional principle of ’res ipsa loquitur‘ - a thing speaks for itself - especially if its interpretation relieves the claimant from the burden of proving that harm was due to the defendant’s negligence.

In a classic case of hindsight bias, shortly after the 1929 stock market crash a New Jersey court held a trustee liable for failing to sell stock before the crash, since “it was common knowledge, not only among bankers and trust companies, but the general public as well, that the stock market condition at the time of the testator’s death was an unhealthy one, that values were very much inflated and that a crash was almost sure to occur.”

In a later case, an Alabama court found trustees liable for investing in a company in spite of “red flags” suggesting poor performance and for selling securities at the “bottom of the market”. How the trustees in such cases were supposed to have known that the red flags were more predictive than the positive signals or that a stock price had actually reached bottom is unclear.

More recently, the international financial crisis dramatically changed the market value of many financial contracts, which banks and their clients – including households and small and medium-sized enterprises - had entered into before the crisis. Examples were interest rate swap contracts offering protection against interest rate hikes. This led to a flurry of judicial and arbitral disputes, many of which depended essentially on whether there had been genuine misrepresentation or fraudulent reticence by financial intermediaries or just unexpected bad luck for their clients.

The practical difficulty of such judgments was compounded in the European Union by the entry into force, in late 2007, of a new directive on financial instruments, which broadened the disclosure obligations for financial intermediaries selling complex financial products to non-professional clients. Thus, while in some cases banks may have actually mis-sold financial products to their clients, in other cases what today looks retrospectively like fraud may have been nothing more than misplaced optimism.

In a likely illustration of such post-crisis hindsight bias in the political realm, on September 28, 2010 Iceland’s Parliament voted to press criminal charges against the former prime minister Geir Haarde for his alleged negligence in office in the events leading up to the banking collapse in 2008.

Hindsight bias may also be relevant in investment arbitration, since it is at the root of the ’obsolescing bargain‘ theory advanced years ago by US economist Raymond Vernon to explain what he said was the tendency of Latin American countries to renege on previous commitments made to foreign investors.

According to the theory, new investment projects in risky sectors or regions will only be undertaken if their ex-ante potential profitability is high enough to compensate risks. If the risks materialise and the project fails, investors will lose money but no political body will complain or even notice. But if, against adverse odds, the project succeeds, its high ex-post profitability will be regarded with suspicion by the government – which was probably not in office when the risky venture was launched. Eventually it may be taxed or otherwise regulated away. In retrospect it is not easy to see the odds surmounted by successful undertakings.

First impressions and confirmation bias

As early as the 17th century Sir Francis Bacon wrote: “The human understanding, when it has once adopted an opinion [...] draws all things else to support and agree with it. And though there be a greater number and weight of instances to be found on the other side, yet these it either neglects and despises, or else by some distinction sets aside and rejects; in order that by this great and pernicious predetermination the authority of its former conclusions may remain inviolate”. Bacon further wrote that “the first conclusion colours and brings into conformity with itself all that come after” - introducing the idea of what pyschologists now call 'confirmation bias" or "belief persistence".

The overwhelming influence of first impressions makes it particularly challenging for judges, juries and arbitrators to ’maintain an open mind‘ during the evidence-presentation phase of legal proceedings. Indeed, we all have a natural tendency to construct a mental model or ’story’ of ’what really happened’ as a case develops. We continuously refine those stories as new evidence is presented, however, once we have constructed our story, it may strongly influence how we interpret new information. Thus, if one has come to believe that a respondent is guilty or innocent, further evidence that is open to various interpretations may be seen as supportive of that belief.

As a former chairman of Spain's Securities and Exchange Commission, I can well imagine that Bernard Madoff’s Ponzi scheme remained hidden for years probably because the ’halo effect’ from his high reputation in Wall Street led the US Securities and Exchange Commission and investors to neglect the red flags that otherwise might have alerted them to the ongoing mischief.

Many happily married couples, meanwhile, will attest how reading exactly the same newspaper article can result in huge convergence of views. Discovery and document production can have the same effect - fostering, not reducing, divergence.

Are arbitrators immune?

Studies show that, except when depressed, most people – whether drivers, students or financial analysts - consider themselves better than average. In this sense they are like the occupants of the fictional Lake Wobegon, a city where, according to author Garrison Keiler, “all the women are strong, all the men are good looking, and all the children are above average”.

Many arbitrators may consider themselves immune to these subconscious threats to their impartiality or open mind. But, somewhat paradoxically, the more they feel free of biases, the more likely they are to fall under their spell.