Arbitration International, 2020, 36, 233–252

doi: 10.1093/arbint/aiaa017
Advance Access Publication Date: 15 June 2020
Article

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
Prospect Theory and due process paranoia: what
behavioural models say about arbitrators’
assessment of risk and uncertainty
Rutger Metsch* and Rémy Gerbay†

ABSTRACT
The term ‘due process paranoia’ is used to describe a perceived reluctance by arbitral
tribunals to act decisively in certain situations for fear of the arbitral award being chal-
lenged on the basis of a party not having had the chance to present its case fully. This
article approaches due process paranoia from the perspective of Prospect Theory,
which is a behavioural model describing how individuals make decisions under risk
and uncertainty. The authors examine how Prospect Theory’s insight that decision
makers tend to overweight low-probability events in their decision-making (the
‘possibility effect’) affects decision-making by arbitrators when faced with the threat of
challenge to their awards on due process grounds (the ‘enforcement risk’). The article
concludes that the possibility effect is prone to contribute to an overweighting by arbi-
trators of the enforcement risk, thereby explaining the perceived tendency by tribunals
to make sub-optimal decisions when faced with due process-related complaints or
threats.

1 . I N TR OD UC T IO N
A unique feature of international arbitration is parties’ ability to participate in the se-
lection of their decision makers.1 In choosing who to nominate, experienced counsel
look to certain attributes they wish to see in candidates.2 The particular attributes
that are preferred in a given arbitration should depend on the case that counsel has

* Rutger Metsch, FCIArb, is an Associate Lecturer at Queen Mary University of London and an Associate in
the international arbitration group of Herbert Smith Freehills LLP (London). He holds an LLB in
International and European Law from the University of Groningen and an LLM in Comparative and
International Dispute Resolution from Queen Mary University of London. The authors would like to thank
Siddharth Joshi and Vanessa Naish for their invaluable help reviewing earlier drafts of this article.
†
Dr Rémy Gerbay is a Partner focusing on international arbitration at MoloLamken LLP (Washington,
DC), and an academic at Queen Mary University of London. He holds a French law degree from the
University of Lyon, a Master’s from the University of Geneva (Graduate Institute), an LLM from
Georgetown University, and a PhD from Queen Mary University of London. The views and opinions
expressed in this article are those of the authors and do not necessarily reflect the position of their law
firms.
1 See Alfonso Gomez-Acebo, Party-Appointed Arbitrators in International Commercial Arbitration (Kluwer
Law International 2016) 39.
2 ibid 40.

C The Author(s) 2020. Published by Oxford University Press on behalf of the London Court of International Arbitration.
V
All rights reserved. For permissions, please email: journals.permissions@oup.com

 233
234  Prospect Theory and due process paranoia

to defend.3 However, it is suggested that one attribute that parties will consistently
look for in arbitrators is their ability to apply established facts to objective legal

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
standards in order to reach decisions that are consistent with the requirements of
those legal standards. In other words, parties will expect an arbitrator to follow a ‘ra-
tional’4 decision-making process to reach optimal outcomes, both on issues of proce-
dure and on the merits.5
Decision-making lies at the heart of the arbitrator’s responsibilities.6 Tribunals
constantly make decisions, big and small, consequential and less consequential; rang-
ing from fixing a hearing date, to which piece of evidence should be preferred, to
whether the presented facts are convincing enough to discharge a burden of proof
and satisfy the applicable legal test.7 Perhaps, therefore, just as arbitrators are
expected to know the law, they should also be expected to understand decision-
making processes and, importantly, the behavioural decision theory models describ-
ing and predicting where decision-making tends to fall short of parties’ expectations.
As we shall see, such behavioural sciences teach us that human beings—including ar-
bitral decision makers—occasionally make decisions that are suboptimal based on the
information available to them. The use of behavioural science has become common
place in legal academia, but so far it has largely focused on other areas of the law, includ-
ing, in particular, regulatory and litigation.8 With some notable exceptions, the arbitra-
tion community has not shown much interest yet.9 This is counter-intuitive considering
that some of the highest-stake, highest-value disputes are now routinely arbitrated.10
This article aims to bring to the fore the relevance of the findings of such behav-
ioural models for international arbitration. It does so by examining one particular ex-
ample: that of due process paranoia11 (DPP). The first part of this article provides a
theoretical framework for DPP and highlights why the phenomenon is problematic

3 ibid 41.
4 For the purposes of this article, being ‘rational’ refers to the arbitrator’s behaviour adhering to the assumptions of
rational actor theory, wherein (among other assumptions) an actor pursues goals, his behaviour results from con-
scious choice, and he will choose the alternative with the highest expected utility if given a choice between
options. For a discussion and overview of these assumptions, see Kristen Renwick Monroe and Kristen Hill
Maher, ‘Psychology and Rational Actor Theory’ (1995) 16 Pol Psychol 1, 2–3.
5 Susan D Franck and others, ‘Inside the Arbitrator’s Mind’ (2017) 66 Emory LJ 1115, 1130 (stressing the
importance of the quality of decision-making to arbitration’s legitimacy).
6 Julian D M Lew and others, Comparative International Commercial Arbitration (Kluwer Law International
2003) 279.
7 For a description of the functions and powers of arbitrators, see generally Nigel Blackaby and others,
Redfern and Hunter on International Arbitration (6th edn, Kluwer Law International; OUP 2015) 309–
19.
8 Christine Jolls, Cass R Sunstein and Richard H Thaler, ‘A Behavioral Approach to Law and Economics’
(1998) 50 Stan L Rev. 1471; Avishalom Tor, ‘The Methodology of the Behavioral Analysis of Law’
(2008) 4 Haifa L Rev 237.
9 Franck and others (n 5) 1131.
10 See Stone Sweet and Florian Grisel, The Evolution of International Arbitration: Judicialization, Governance,
Legitimacy (OUP 2017) 4 (citing by way of example the Yukos arbitration, considered the ‘highest com-
pensation awarded in the history of litigation’). On the increase in the average monetary value of interna-
tional arbitrations over time, see Rémy Gerbay, ‘Is the End Nigh Again? An Empirical Assessment of the
“Judicialization” of International Arbitration’ (2014) 25 AJIL 223, 244.
11 As indicated below, DPP is defined here as the unjustified reluctance by arbitral tribunals to act decisively
in certain situations for fear of the arbitral award being challenged on the basis of a party not having had
the chance to present its case fully.
 Prospect Theory and due process paranoia  235

in practice. The second part of this article explores how behavioural science can help
us to better understand the phenomenon of DPP, looking in particular at ‘prospect

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
theory’. One specific question examined in the second section is whether the so-
called ‘possibility effect’ may help explain DPP. The article concludes that arbitrators
are inclined to make suboptimal due process decisions when fearing the challenge of
an award as a result of the human tendency to overweight small probabilities when
evaluating options with uncertain outcomes.
This grey area of arbitral decision-making should be of interest to anyone in-
volved in the business of representing parties in arbitration proceedings. Likewise, it
should be of interest to those deciding such disputes, and keen not to fall prey to
such tendencies.

2 . D P P A S A P H E N O ME N O N
2.1 Conceptualizing DPP
DPP describes the reluctance by arbitral tribunals to act decisively in certain situa-
tions for fear of the arbitral award being challenged on the basis of a party not having
had the chance to present its case fully.12
DPP arises when three components are combined.13 First, case management deci-
sions by an arbitral tribunal that are overly attentive to due process considerations.14
The second component is a belief on the tribunal’s part that the decisions in question
are rendered necessary by the risk that the arbitral award may otherwise be set aside
and/or refused enforcement (the ‘Enforcement Risk’).15 Finally, the third element—
without which there would be no paranoia but only sensible concern for due pro-
cess—is the erroneous character of the tribunal’s belief that this level of caution is
warranted.16 This erroneous belief is caused by an inflated perception of the
Enforcement Risk.17
Practical examples of overly cautious case management decisions include:
(i) granting insufficiently justified extensions of time; (ii) accepting the belated intro-
duction of new defences/claims or of fresh evidence; (iii) acceding to last-minute
requests to reschedule oral hearings; (iv) granting overly generous disclosure orders;
and (v) accepting the filing of superfluous submissions.18
DPP operates at the intersection of conflicting duties generally imposed upon arbitra-
tors by national laws and institutional rules. On the one hand, arbitrators are required to
act fairly and impartially, to ensure that each party has an opportunity19 to present its

12 Queen Mary University of London, ‘2015 International Arbitration Survey: Improvements and
Innovations in International Arbitration’ (2015) 10 <http://www.arbitration.qmul.ac.uk/media/arbitra
tion/docs/2015_International_Arbitration_Survey.pdf> accessed 8 April 2020.
13 Rémy Gerbay, ‘Due Process Paranoia’ (Kluwer Arbitration Blog, 6 June 2016) <http://arbitrationblog.klu
werarbitration.com/2016/06/06/due-process-paranoia/> accessed 8 April 2020.
14 ibid.
15 ibid.
16 ibid.
17 ibid.
18 ibid.
19 To what extent such an opportunity should be given can vary between jurisdictions and institutional rules.
See Section 2.4.2 below.
236  Prospect Theory and due process paranoia

case20 and to render enforceable awards.21 At the same time, arbitrators are also gener-
ally required to conduct the arbitration in an expeditious manner, avoiding unnecessary

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
delay and expense.22 The first set of duties pushes arbitrators towards over-cautiousness,
whereas the second pushes them towards under-cautiousness. Arbitrators enjoy consider-
able discretionary power over the arbitral process,23 and it is incumbent upon them to
find a balance between due process and efficiency to ensure fair but expeditious
proceedings.24

2.2 DPP in context

It is important to recognize the context in which DPP arises. A party that is dissatisfied
with the outcome of an arbitration may attempt to vacate or resist enforcement of an
award on grounds related to due process violations (a ‘Challenge’). The New York
Convention includes due process violations as a reason for non-enforcement of an
award if a party was ‘unable to present its case’ at Article V(1)(b).25 Many national
laws mirror the grounds in Article V of the New York Convention in provisions relat-
ing to the annulment of arbitral awards.26 In the light of this threat, Tribunals inclined
to make robust case management decisions may hold off for fear of allegations of due
process violations. Arbitrators have a strong dislike for challenge of their awards in do-
mestic courts. Contrary to arbitration proceedings, challenges to awards tend to be
public.27 They have the potential to generate bad publicity for the arbitrators con-
cerned, which, in turn, may hurt arbitrators’ prospects of future appointments.28

20 See, for example, ICC Rules of Arbitration art 22.4; LCIA Arbitration Rules art 14.4(i); UNCITRAL
Arbitration Rules art 17.1; UNCITRAL Model Law on International Commercial Arbitration art 18;
HKIAC Administered Arbitration Rules art 13.1; English Arbitration Act 1996, s 33(1); Indian
Arbitration Act 1996, s 18.
21 See Gary B Born, International Commercial Arbitration (2nd edn, Wolters Kluwer Law & Business 2014)
1993. It is questionable whether this is a legal or a moral obligation and, especially, whether this duty is
enforceable. See Martin Platte, ‘An Arbitrator’s Duty to Render Enforceable Awards’ (2003) 20 JOIA 307.
22 See, for example, ICC Rules of Arbitration art 22.1; LCIA Arbitration Rules art 14.4; UNCITRAL
Arbitration Rules art 17.1; HKIAC Administered Arbitration Rules art 13.1; English Arbitration Act 1996,
s 33(1).
23 Blackaby and others (n 7) 309.
24 Yves Fortier, ‘The Minimum Requirements of Due Process in Taking Measures Against Dilatory Tactics:
Arbitral Discretion in International Commercial Arbitration - A Few Plain Rules and a Few Strong
Instincts’ in Albert Jan Van den Berg (ed), Improving the Efficiency of Arbitration Agreements and Awards:
40 Years of Application of the New York Convention (Kluwer Law International 1999).
25 That provision reads: ‘1. Recognition and enforcement of the award may be refused, at the request of the
party against whom it is invoked, only if that party furnishes to the competent authority where the recog-
nition and enforcement is sought, proof that: . . . (b) The party against whom the award is invoked was
not given proper notice of the appointment of the arbitrator or of the arbitration proceedings or was oth-
erwise unable to present his case; . . ..’
26 See Lew and others (n 6) 673. See also UNCITRAL Model Law art 34(2)(a)(ii); English Arbitration Act
1996, s 68(2)(a); Switzerland’s Federal Code on Private International Law PIL art 190(d); Indian
Arbitration Act 1996, s 34(2)(ii).
27 For a discussion of the position on this point in various jurisdictions, see ‘Guide to National Rules of
Procedure for Recognition and Enforcement of New York Convention Awards’ (ICC 2008) <https://li
brary.iccwbo.org/content/dr/COMMISSION_REPORTS/CR_ALL_0037.htm?
l1¼CommissionþReports> accessed 8 April 2020.
28 William W Park, ‘Arbitration in Autumn’ (2011) 2(2) Journal of International Dispute Settlement 287,
292.
 Prospect Theory and due process paranoia  237

Parties are well aware of tribunals’ concerns for due process. Some may purpose-
fully exploit these concerns to obtain a strategic advantage in the arbitration.29

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
Parties have indeed been accused of (mis-)using due process as a strategic sword,
rather than a shield protecting parties’ fundamental rights.30 Reed has noted a grow-
ing trend of invoking ‘multiple procedural complaints under the banner of due pro-
cess’ as a tool to apply pressure to tribunals.31 Similarly, Cremades has commented
that ‘[w]hen, in a party’s submissions or at the arbitral hearing, one of the lawyers
lets the expression ‘due process’ slip, the tribunal usually understands it as a threat
that, if the arbitrator does not agree to their petition, the award will subsequently be
annulled.’32 This has been described as a ‘guerrilla tactic’.33
DPP appears to have been recognized as a structural issue in international arbitra-
tion. Empirical research,34 scholarship,35 and public debate36 have recognized as
problematic a tendency by tribunals to condone disruptive behaviour by counsel due
to a concern that an award would otherwise be vulnerable to challenge. But why is
DPP a problem?

2.3 Why is DPP a problem?

2.3.1 Delay and excessive costs
DPP is perceived as problematic because of two chief concerns of users of arbitra-
tion: delays and excessive costs.37 Arbitrators entertaining and granting (unmeritori-
ous) due process-related applications contribute directly to increased costs and delay
for obvious reasons. First, unmeritorious due process-related applications in them-
selves are wasteful. Each application will have to be argued, decided on and complied
with. Each of these stages takes time to be briefed and decided. The wasteful effects
of DPP are even more evident when the outcome of the tribunal’s decision causes to
expand unnecessarily the scope of the parties’ and the tribunal’s work in the

29 Lucy Reed, ‘Ab(use) of Due Process: Sword vs Shield’ (31st Queen Mary University and Freshfields
Arbitration Lecture, London, 27 October 2016) <http://knowledge.freshfields.com/en/Global/r/1699/
freshfields_arbitration_lecture_2016_> accessed 8 April 2020; Bernardo M Cremades, ‘The Use and
Abuse of Due Process in International Arbitration’ (Alexander Lecture, London, 17 November 2016)
<http://www.ciarb.org/docs/default-source/ciarbdocuments/events/2016/november/use-and-abuse-
of-due-process.pdf?sfvrsn¼2> accessed 8 April 2020.
30 Reed, ibid.
31 ibid
32 Cremades (n 29) 671.
33 Gunther J Horvath and others, ‘Categories of Guerrilla Tactics’ in Gunther J Horvath and Stephan
Wilske (eds), Guerrilla Tactics in International Arbitration (Kluwer Law International 2013) 3–16.
34 Mistelis and Metsch (n 12).
35 Reed (n 29); Cremades (n 29).
36 See, for instance, conferences organized on the theme of DPP: ‘Efficiency vs. Due Process Paranoia’
(Swedish Arbitration Association conference, Stockholm, 19 January 2019) <https://sccinstitute.com/
about-the-scc/news/2019/efficiency-vs-due-process-paranoia/> accessed 8 April 2020; news pieces pub-
lished by specialized journals: Corporate Disputes (Financier Worldwide, July–September 2019)
<https://www.corporatedisputesmagazine.com/the-due-process-paranoia-in-international-arbitration-
are-we-moving-towards-a-more-thoughtfully-protected-reality> accessed 8 April 2020; Eric Tuchmann,
‘Due Process Paranoia in International Arbitration: A Valid Concern?’ (CDR, 26 November 2018)
<https://www.cdr-news.com/categories/arbitration-and-adr/8813-due-process-paranoia-in-international-
arbitration-a-valid-concern> accessed 8 April 2020.
37 Gerbay (n 13).
238  Prospect Theory and due process paranoia

arbitration (ie when a tribunal—for no valid reason—allows filing of additional sub-

missions, introduction of additional evidence, production of irrelevant documents,

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
etc). In an (often) hourly fee-based sector, a single decision may result in many thou-
sands of dollars of wasted legal costs.38
Users’ concerns regarding delays and excessive costs are hardly new issues and
they cannot be overstated.39 Six out of eight of Queen Mary University’s School of
International Arbitration’s empirical surveys on international arbitration—conducted
over a span of 12 years—have highlighted delays and excessive costs as disadvantages
of the process,40 and arbitral institutions have long grappled with the best way to
control delays and costs.41

2.3.2 Suboptimal decision-making

An overly cautious approach by tribunals is problematic when concern for due pro-
cess exceeds justified caution and prevents a tribunal from acting decisively; ‘con-
cern’ may have morphed into ‘paranoia’. Put differently, every time a tribunal opts
for caution instead of procedural economy/efficiency in circumstances where the
tribunal could actually afford to be robust (because the Enforcement Risk is lower
than is believed to be by the arbitrators), the tribunal makes a wasteful decision.42
The crux of the problem of DPP lies with those decisions that are needlessly cau-
tious; decisions which, from the point of view of procedural efficiency, are
suboptimal.43
Furthermore, DPP can be a ‘self-fulfilling prophecy’ as overly cautious decisions
set a norm ‘against which future complaints of insufficient opportunity to be heard
are judged’.44 The effects of a lack of firmness by tribunals may resonate. Pinsolle
describes a situation where a party can derail the process and be rewarded for doing
so. Commenting on ‘weak tribunals’, he notes that ‘[e]xperience shows that many
parties behave far worse before Arbitral Tribunals than they would before domestic
courts simply because the perception exists that there is no sanction.’45 As such,
DPP might be part of a vicious cycle of behaviour contributing to increased costs
and expense.

38 ibid
39 ibid
40 Queen Mary University (n 12) 6 and 7 (2006 survey), 5 (2008 survey), 21 (2013 survey), 7 (2015 sur-
vey), 27 (2016 survey), 7 and 8 (2018 survey).
41 See ICC Commission Report, ‘Controlling Time and Costs in Arbitration’ (2007) <https://iccwbo.org/
content/uploads/sites/3/2018/03/icc-arbitration-commission-report-on-techniques-for-controlling-time-
and-costs-in-arbitration-english-version.pdf> accessed 8 April 2020.
42 Gerbay (n 13).
43 ibid
44 Constantine Partasides and Ben Prewett, ‘Rediscovering the Lost Promise of International Arbitration’ in
Laurent Lévy and Michael Polkinghorne (eds), Expedited Procedures in International Arbitration (ICC
2017) 111.
45 Philippe Pinsolle, ‘The Need for Strong Arbitral Tribunals’ in Andrea Menaker (ed), International
Arbitration and the Rule of Law: Contribution and Conformity, 19 ICCA Congress Series (Kluwer Law
International 2017) 823.
 Prospect Theory and due process paranoia  239

2.4 Sorting the wheat from the chaff: distinguishing justifiable due process
concerns from paranoia

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
2.4.1 The good, the fast, and the cheap
A process that is good, fast, and cheap has been described as the ‘unicorn’ of arbitra-
tion.46 Often, only one or two of these features are available, as the following dia-
gram by Risse illustrates.47

Figure 1. The ‘Magic Triangle’ of quality, time efficiency, and cost savings.

A ‘quality award’ is one that is not just correct but also enforceable.48 If the arbitral
process results in a correct and enforceable award, it is perhaps palatable that the
process is less fast and/or cheap. However, an increase in cost and expense is only
warranted if there is an actual increase in the quality of the award.
This will be the case if granting a party’s due process-related application will actu-
ally reduce the Enforcement Risk. Indeed, not every due process-related decision is
made out of ‘paranoia’. Arbitrators will encounter situations where a certain decision
would clearly result in a due process violation. It is part of the arbitrator’s mandate
to guard the parties’ right to be heard and, in situations at risk of being an obvious vi-
olation of due process, the arbitrator is required to sacrifice expense and/or duration
to protect the quality of the process.
Instead, DPP is concerned with decisions that increase cost and expense, but
without causing any actual reduction of the Enforcement Risk, because the risk itself
was minimal to start with. Such decisions include robust decisions which do not

46 See Jennifer Kirby, ‘Efficiency in Arbitration: Whose Duty Is It?’ (2015) 32(6) JOIA 689.
47 The balancing act between these features in arbitration was illustrated by William W Park’s reference to a
sign hanging in the window of a Boston shoe repair shop that read ‘“fast service”, “low price”, and “high
quality”’ and included the instruction that customers should ‘pick any two’. See William W Park,
Arbitration of International Business Disputes (2nd edn, OUP 2012) 161. This concept was later captured
in the illustration of the ‘magic triangle’ at Figure 1. See Joerg Risse, ‘Ten Drastic Proposals for Saving
Time and Costs in Arbitral Proceedings’ (2013) 29(3) Arbitration International 453, 466. For further
comment on the trade-offs running through many of arbitration’s procedural dilemma, see the ‘Balance
and Proportionality’ lecture at the American University by William W Park (full published version forth-
coming), report available at Global Arbitration Review, ‘Park on Balance and Proportionality (16 May
2019), <https://globalarbitrationreview.com/article/1193023/park-on-balance-and-proportionality>
accessed 8 April 2020.
48 Kirby (n 46) 692.
240  Prospect Theory and due process paranoia

infringe due process (or which do infringe it but not sufficiently49 as to justify a
Challenge to the award). For such a category of decisions, the benefit of the outcome

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
(an increase in quality of the award, if that) does not outweigh the cost of the deci-
sion (in the form of expense and delay).

2.4.2 Accurate assessment of the Enforcement Risk

To accurately assess the Enforcement Risk in any given situation, arbitrators need to
be aware of: (i) the due process standard applicable to their decision and (ii) how
courts (or ad hoc committees) have reviewed the exercise of arbitrators’ case man-
agement powers to the applicable due process standard in Challenge situations.
These two elements have been discussed comprehensively in other publications.50 A
full review of due process standards and Challenge decisions is outside the scope of
this article. For present purposes, the following observations suffice.
Broadly, under modern arbitration legislation, parties enjoy a right to equal treat-
ment and a right to be heard.51 Protection afforded by these rights includes adequate
notice of any proceedings and claims, reasonable time to present a case (and react to
the other’s case), a decision based on the evidence and submissions by the parties,
and the tribunal not deciding on the basis of grounds the parties could not address
first.52 National laws (and institutional rules) use varying language to describe the
standard of due process. The UNCITRAL Model Law stipulates in Article 18 that
the parties shall be given a ‘full opportunity’ of presenting their case.53 The Swiss
PIL provides at Article 182(3) that the tribunal must ensure ‘equal treatment’
and ‘the right to be heard’ of the parties.54 The English Arbitration Act55 and the

49 As an aside, it should be noted that due process decisions should not be conceived of as an ‘all or noth-
ing’. ‘. . . [T]he application of due process is a question which cannot always be approached in a binary
fashion, that is to say, in terms of ‘whether or not’ the due process norm in question should apply. The
application of certain due process norms is a question of degree. A prime example of this would be the re-
quirement of “adversariality” (i.e., giving a party an opportunity to present its case . . .). Often, the contro-
versial question will not so much be whether or not a party should be given an opportunity to provide its
views on a particular issue, but how much of a consultation should be allowed . . .. Namely, should there
be more than one round of submissions.’ Rémy Gerbay, The Functions of Arbitral Institutions (Wolters
Kluwer 2016) 209–10.
50 See, for example, Klaus Peter Berger and Ole Jensen, ‘Due Process Paranoia and the Procedural Judgment
Rule: A Safe Harbour for Procedural Management Decisions by International Arbitrators’ (2016) 32(3)
AI 415; Loukas Mistelis, ‘Efficiency. What Else? Efficiency as the Emerging Defining Value of
International Arbitration: Between Systems Theories and Party Autonomy’ (2019) Queen Mary
University of London Legal Studies Research Paper No 313/2019, 18–25 <https://papers.ssrn.com/
sol3/papers.cfm?abstract_id¼3372341> accessed 8 April 2020; IBA Arbitration Committee, ‘Annulment
of Arbitral Awards by State Court: Review of National Case Law with Respect to the Conduct of the
Arbitral Process’ (IBA, 2018) <https://www.ibanet.org/LPD/Dispute_Resolution_Section/Arbitration/
Projects.aspx> accessed 8 April 2020.
51 Born (n 21) 2171–80.
52 ibid 2179–80.
53 Commentary on the draft text of the UNCITRAL Model Law noted that the ‘full’ opportunity should
not be read to entitle a party to ‘obstruct the proceedings by dilatory tactics’. See Analytical commentary
on draft Text of a Model Law on International Commercial Arbitration, 25 March 1985, UN Doc A/
Cn.9, 264, 46.
54 The French Code of Civil Procedure at 16 employs a similar standard.
55 s 33(1)(a).
 Prospect Theory and due process paranoia  241

Hong Kong Arbitration Ordinance56 provide that parties must have a ‘reasonable op-
portunity’ of putting their case. This same standard is used by the ICC,57 the

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
LCIA,58 and the HKIAC.59 It is questionable whether these due process standards
actually lead to different outcomes in practice.
Courts in arbitration-friendly jurisdictions generally show deference to the arbitral
tribunal’s case management powers in Challenge situations. Indeed, commenting on
challenges to awards based on procedural irregularity, Lew, Mistelis, and Kroll have
observed that ‘the test is that of a significant injustice so that the tribunal would have
decided otherwise had the tribunal not made a mistake’.60 Similarly, commenting on
Article V(1)(b) of the New York Convention, Born notes that it only provides a ba-
sis of non-recognition ‘in cases of serious procedural unfairness which is shown to have
a material impact on the arbitral process or decision’.61 Berger and Jensen reviewed
case law relating to procedural management decisions and concluded that domestic
courts will ‘but in the most unusual of circumstances’ defer to the discretion of the
arbitral tribunal.62 In 2018, the IBA conducted a review of national case law relating
to case management decisions by arbitral tribunals in 13 jurisdictions. The study con-
cluded that ‘courts generally support the arbitration process and it is rare for an
award to be set aside for procedural reasons only’.63 In the light of these findings, it
can be concluded that the threshold to challenge awards on the basis of due process
violations is high and that courts will only rarely exercise their powers to vacate or re-
ject enforcement of an award.
While there are situations where awards have been successfully challenged on due
process grounds,64 these cases often do not relate to robust case management deci-
sions.65 Instead, they concern mostly situations where an arbitrator ‘surprised’ par-
ties, either by relying upon arguments that were not properly discussed in the
arbitration or by departing from the procedure previously set.66 For example, in
2018 the English High Court set aside an award because an arbitrator had decided a
dispute on a significantly different basis to what the parties had pleaded without
seeking the parties’ submissions on this novel point, which the court deemed to cre-
ate a substantial injustice.67 In the same year, the Hong Kong Court of First Instance
remitted an interim award to the tribunal after concluding that a party had not been

56 s 46(3)(b).
57 art 22(4).
58 art 14.4(i)
59 art 13.1.
60 Lew and others (n 6) 674 (emphasis added).
61 Born (n 21) 3403 (emphasis added).
62 Berger and Jensen (n 50) 428.
63 IBA Arbitration Committee (n 50) 2.
64 Gerbay (n 13); P v D [2019] EWHC 1277 (Comm).
65 In the English courts, a review of due process related decisions has shown that ‘[o]ut of the 110þ deci-
sions citing s.33 that have been reported since the enactment of the English Arbitration Act (1996) two
decades ago, we have not been able to find a single decision in which the English courts set aside an
award because of an overly robust case management decision.’ The authors submit that this will be the
same in other modern arbitration jurisdictions. Gerbay, ibid.
66 Gerbay, ibid.
67 RJ and another v HB [2018] EWHC 2833 (Comm).
242  Prospect Theory and due process paranoia

given a reasonable opportunity to present its case and deal with that of its opponent
when a tribunal had decided liability on a basis not advanced by the claimant.68

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
As a general observation, it is welcome that the courts act in situations where seri-
ous due process violations occur in order to safeguard the integrity of the arbitral
process. It is not denied that due process violations occur at times. Rather, it is sub-
mitted that such situations are rare and the threat of a successful Challenge is limited
to serious infringements of due process. It is highly unlikely that an arbitrator who
makes a robust case management decision opens the award up to Challenge on the
basis of due process grounds. DPP implies that arbitrators overestimate the probabil-
ity of such a Challenge: the fear of Challenge of an award when the arbitrator acts
decisively is not justified.

3. H U MA N F I R S T , A R B IT R A T O R S E CO N D : B EH A V IO UR A L M O D E L S
A N D A R BI TR A TOR S’ A P P RO A C HE S TO R I SK
Section 2 of this article has demonstrated that DPP exists and that there is no good
reason why it should exist. The remainder of this article focuses on the ‘why’; what
causes arbitrators to fall prey to DPP.

3.1 Modelling arbitrator decision-making

There are various theories predicting and explaining how decisions are made in the
realms of decision-making under risk and uncertainty.69 Decision-making under risk
refers to a situation where the decision maker is aware of the probability distribution
over the different outcomes. In contrast, decision-making under uncertainty refers to
a situation where the probability distribution over various outcomes is not known
but could be estimated (imprecisely) by the decision maker.70
One approach is to view a decision from the perspective of an integration
model.71 In this context, an integration model broadly means that the value of an op-
tion is estimated by multiplying some function of probabilities by some function of
the outcomes of the option. For example, a coin flip that pays £100 if it lands heads,
and £0 if it lands tails, can be valued as worth £50 to the decision maker: 50 per cent
chance of ending up with the outcome ‘heads’ with a pay-off of £100 plus 50 per
cent chance of the coin landing tails with a pay-off of £0. This expected value of the
flip is a useful decision-making tool and is based on the logic that if this same coin
flip were repeated a large number of times, the average return per flip would be in
the region of £50. If a decision maker was offered to buy a right to this coin flip for,
say £45, this would objectively be a good investment because, on average, the deci-
sion maker would make a £5 profit per flip.

68 P v M [2018] HKCFI 2280.

69 For a good overview (including, eg ‘regret theory’, ‘disappointment theory’, ‘decision affect theory’, ‘sa-
lience theory’, ‘decision field theory’, or ‘fuzzy trace theory’), see generally Craig Fox, Carsten Erner and
Daniel Walters, ‘Decisions Under Risk: From the Field to the Laboratory and Back’ in Gideon Keren and
George Wu (eds), The Wiley Blackwell Handbook of Judgment and Decision Making (1st edn, Wiley-
Blackwell 2016).
70 ibid 44.
71 ibid 64.
 Prospect Theory and due process paranoia  243

Outcomes in decision-making models may also be presented in utility values

rather than in monetary values. A utility value is a numerical representation of a per-

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
son’s preferences72 and represents the psychological value or satisfaction a certain
outcome generates for that person. By using utility to represent the value of an out-
come, it is possible to compare and decide between options that may not have a
monetary value (eg, a decision between a job that offers more days of annual leave
and a job that is closer to home) or to account for phenomena such as the diminish-
ing marginal value of money73 when weighing different options.
Using utility values, integration models can serve as a model to analyse certain
decisions by arbitrators. For example, a due process-related decision by an arbitrator
can be considered a decision under uncertainty. After all, the arbitrator knows the
possible outcomes of the decision but does not know the precise probabilities of
these outcomes occurring. Below is an illustration of what such an analysis might
look like (on a high-level basis).
Imagine that a party to an arbitration (the ‘applicant’) requests to introduce new
evidence at the last minute without a convincing reason why the evidence could not
have been introduced at an earlier stage of the proceedings. Assume that the arbitra-
tor has the choice to: (i) refuse the request or (ii) grant the request.
Focusing on the first option, the possible outcomes of refusing the request are
two-fold: (a) the decision may be challenged by the applicant and the Challenge is
successful74 or (b) the decision does not lead to a successful Challenge of the award
by a party (either because no Challenge is made by the applicant or a Challenge is
made but does not succeed).
The arbitrator does not know the precise probability distribution of either of
these two options. However, he can make an (imprecise) assessment: eg ‘very likely’
(corresponding to a high percentage chance) that outcome (a) will occur; and
‘highly unlikely’ (corresponding to a low percentage chance) that outcome (b) will
occur.
In terms of pay-offs, a successful Challenge by one of the parties can be said to re-
sult in a significant amount of negative utility for the arbitrator (ie it would feel like a
considerable loss). This is because of factors such as harm to the arbitrator’s reputa-
tion when a decision is successfully challenged, delay to the arbitration because the
case is remitted to the tribunal, etc. If the award is not successfully challenged,
the decision can be said to result in a small amount of positive utility for the arbitra-
tor (ie it would feel like a gain). This is because of factors such as there being less
disruption to the arbitral process, it feeling like the right thing to do, etc.75

72 Peter C Fishburn, ‘Utility Theory and Decision Theory’ in J Eatwell and others (eds), Utility and
Probability (1st edn, W.W. Norton & Co. 1990) 303.
73 That is, the satisfaction experienced from an increase in wealth decreases as a person’s overall wealth
increases: a person will experience more happiness from a change from 0 to 1 million in the bank than
from 19 to 20 million in the bank. The same can be said for other products: when hungry, the utility de-
rived from the first slice of pizza will be more than from the 20th slice of pizza.
74 For the sake of simplicity and clarity of the example, the separate probability events of whether a party ac-
tually initiates the Challenge (yes or no) and, after, whether that Challenge is successful (yes or no) are
combined into one outcome.
75 This article, and the pay-off structures and estimates below, assumes that the arbitrator’s preferences are
aligned with his duties (ie to make the legally correct and procedurally efficient decision). An
244  Prospect Theory and due process paranoia

The combination of these factors creates a situation where there is a very high
probability of a small gain and a very small probability of a great loss if the arbitrator

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
makes a robust due process decision. When assigning arbitrary (though relative) pay-
offs and probabilities to these outcomes, the consequences of this decision from the
arbitrator’s perspective can be represented in tabular form as follows (Table 1).

Table 1. Pay-offs of successful or unsuccessful Challenges and their

probabilities
Outcome Utility Probability (%)

Successful Challenge 100 (significant loss) 1

No successful Challenge þ2 (small gain) 99

However, the arbitrator might also choose option (ii) and grant the belated re-
quest for the introduction of new evidence. In that situation, there is no risk of the
award being challenged on the basis of the decision. Instead, it can be said that—
from the arbitrator’s perspective—there is a certain (100 per cent) outcome with a
small negative pay-off resulting from the grant of the request. This is because of fac-
tors such as the delay to the arbitral process, the frustration from the other party, the
arbitrator feeling that he made the ‘wrong’ decision, etc. For present purposes, it is
assumed that the negative pay-off of this option is 3 utility points.
It becomes clear that the arbitrator faces a choice between a risky option (grant-
ing the request) that will likely lead to a small gain, but could potentially result in a
significant loss, and a non-risky option (rejecting the request) that will surely lead to
a small loss. When factoring in the assumed probabilities and pay-offs above, the
choices are:

• Option (i): make a robust due process decision. There is then a 1 per cent
chance of a 100 pay-off, and a 99 per cent chance of a þ2 pay-off, amount-
ing to an expected utility value of þ0.98.76
• Option (ii): grant the request. There is then a certain pay-off of 3 utility
points.

When assessed objectively, the arbitrator should choose option (i) because it has a
higher expected utility value (0.98 vs 3).
A visual representation of this decision is provided in (Figure 2).
The above example is for illustration purposes only and is simplified to demon-
strate how decision analysis models and theories can be used to analyse arbitrator
decisions. For present purposes, it is key to appreciate that optimal decision-making
under uncertainty is influenced mainly by two factors: the probability of an outcome
unscrupulous arbitrator who, for example, is interested in raising his chargeable hours would perhaps re-
ceive a positive utility pay-off if the procedure is unnecessarily dragged out, or would not experience ‘satis-
faction’ from not disrupting the arbitral process or ‘doing the right thing’.
76 (Pay-off of successful challenge * probability of occurrence) þ (pay-off of no successful challenge * proba-
bility of occurrence) ¼ (100 * 0.01) þ (2 * 0.99) ¼ 1 þ 1.98 ¼ 0.98.
 Prospect Theory and due process paranoia  245

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
Figure 2. Decision tree of expected pay-offs from arbitrator’s grant or refusal of request.

occurring and the pay-off of that outcome. Consequently, if the perception of proba-
bilities and outcomes is distorted, decision-making may not be optimal.

3.2 Prospect Theory: the possibility effect and the fourfold pattern
In the example above, it is possible on the basis of the given probabilities and pay-
offs to identify what the ‘correct’ decision is from the point of view of maximizing
utility.
It may be tempting for counsel involved in arbitration proceedings to try to pre-
dict what a decision maker might actually do when faced with such a situation by
plotting the ’rational’ course of action on the basis of these probabilities and pay-offs.
However, there is ample evidence and scholarship demonstrating that human behav-
iour often does not accord with the predictions of prescriptive economic theories of
decision-making which presuppose that the individual is a rational actor seeking to
maximize pay-offs and consistently makes decisions in accordance with mapped ra-
tionality axioms.77 Arbitrators are not immune to these flaws in humans decision-
making: Franck and others concluded on the basis of the ‘first-ever psychological ex-
periment involving international arbitration’ that ‘arbitrators often make intuitive and
impressionistic decisions rather than the fully rational and deliberative decisions that
might be normatively desirable’.78
Often, such ‘deviations from rationality’ occur in predictable and systematic ways
that are studied and highlighted by descriptive79 decision-making theories. The domi-
nant theory80 is cumulative prospect theory (‘Prospect Theory’)81: a descriptive theory

77 Daniel Kahneman and Amos Tversky, ‘Prospect Theory: An Analysis of Decision under Risk’ (1979)
47(2) Econometrica 263.
78 Franck and others (n 5) 1166.
79 That is, describing how people actually make decisions rather than prescribing how people should make
decisions.
80 At least according to the 2002 Committee for the Prize in Economic Sciences in Memory of Alfred
Nobel.
81 See Kahneman and Tversky (n 77) 263–92; Amos Tversky and Daniel Kahneman, ‘Advances in Prospect
Theory: Cumulative Representation of Uncertainty’ (1992) 5 J Risk Uncertain 297.
246  Prospect Theory and due process paranoia

of decision-making under risk or uncertainty posited by Tversky and Kahneman.82

A full discussion of Prospect Theory is complex and outside of the scope of this article.

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
Instead, the focus will be on two elements of Prospect Theory that may help explain
DPP: the possibility effect (and the related concept of decision weights) and the
fourfold pattern of risk attitudes.

3.2.1 The possibility effect

In very simple terms, the possibility effect denotes the phenomenon that very small
probabilities ‘feel’ higher to a decision maker than they actually are.
The existence of the possibility effect is linked to experimental evidence that deci-
sion makers do not usually treat probabilities linearly.83 Consider the following ex-
ample given by Kahneman and Tversky (and borrowed from Zeckhauser)84:

Suppose you are compelled to play Russian roulette, but are given the oppor-
tunity to purchase the removal of one bullet from the loaded gun. Would you
pay as much to reduce the number of bullets from four to three as you would
to reduce the number of bullets from one to zero? Most people feel that they
would be willing to pay much more for a reduction of the probability of death
from 1/6 to zero than for a reduction from 4/6 to 3/6.

In the above example, the 1/6 change in probability is objectively as big a difference
when it alters the probability from 1/6 to 0 or from 4/6 to 3/6. It should, therefore,
be worth an equal amount of money to the decision maker. Nonetheless, it is gener-
ally valued differently in practice, indicating that the perception of a change in proba-
bilities is non-linear. Prospect Theory asserts, on the basis of experimental evidence,
that human decision makers facing a risky prospect assign decision weights to out-
comes that are not identical to the probabilities of the outcome.85 In other words:
the subjective weighting of an x per cent probability is different from what the actual
probability would suggest.
This phenomenon is especially pronounced at the extreme ends of the probability
scale, namely for very low-probability events (which will almost certainly not occur)
and for very high-probability events (which will almost certainly occur). In reality, a
decision maker ‘overweights small probabilities and underweights moderate and high
probabilities’.86 On the low end of the probability scale (5 per cent chance or less),
decision makers appear not to ‘weight rare events according to their actuarial chan-
ces of occurring’ but ‘tend to overestimate the chance that rare events will occur’.87
This phenomenon is referred to as the ‘possibility effect’. The related phenomenon

82 Perhaps best known through it featuring in the bestseller by Daniel Kahneman, Thinking, Fast and Slow
(1st edn, Farrar, Straus and Giroux 2013).
83 Richard Gonzalez and George Wu, ‘On the Shape of the Probability Weighting Function’ (1999) 38
Cogn Psychol 129, 131.
84 See Kahneman and Tversky (n 77) 283.
85 Kahneman (n 82) 312.
86 Tversky and Kahneman (n 81) (emphasis added).
87 Zach Burns, Andrew Chiu and George Wu, ‘Overweighting of Small Probabilities’ in James J. Cochran
(ed), Wiley Encyclopaedia of Operations Research and Management Science (Wiley, New York 2010) 1.
 Prospect Theory and due process paranoia  247

at the other extreme of the probability scale, whereby individuals underweight the
chance of an almost certain event (95 per cent chance or more), is referred to as the

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
‘certainty effect’. In very simplistic terms: an event that is highly unlikely ‘feels’ much
more likely to occur than its probability would suggest and an event that is highly
likely (to the point that it is almost certain) ‘feels’ less likely to occur than its proba-
bility would suggest.
Prospect Theory accounts for this overweighting of low probabilities and under-
weighting of high probabilities by using decision weights instead of probabilities in
its analysis of decision-making under risk and uncertainty.88 Decision weights can be
described as ‘a person’s subjective interpretation of an objective probability’89 and
capture ‘what may be called chance attitude’.90 Kahneman and Tversky measured the
relation between decision weights and probabilities through a series of experiments
wherein subjects had to choose between given sure gains/losses and a risky pros-
pect.91 They found the following decision weights associated with a given
probability.92

Table 2. Decision weights assigned to given probabilities

Probability (%) 0 1 2 5 10 20 50 80 90 95 98 99 100

Decision weight 0 5.5 8.1 13.2 18.6 26.1 42.1 60.1 71.2 79.3 87.1 91.2 100

Table 2 accounts for the possibility effect. A 1 per cent probability, for example, is
attributed much more weight in decision-making (5.5) than the probability would
suggest. Also, the table shows the non-linear treatment of probabilities: the subjective
difference between 0 per cent and 5 per cent chance will feel greater than that be-
tween 5 per cent and 10 per cent chance because the 0–5 per cent change creates a
possibility that did not exist before. This phenomenon is represented in the table: 0–
5 per cent has a decision weight delta of 13.2 and 5–10 per cent has a (much lower)
decision weight delta of 5.4. Interestingly, the risk attitudes with respect to gains
were also observed in situations where high monetary incentives were offered and
test subjects were able to earn up to three times their monthly salary.93
The certainty effect is demonstrated on the other end of the table. A 99 per cent
probability is attributed much less weight in decision-making (91.2) than the proba-
bility would suggest. Also, the subjective difference between 95 per cent and 100 per
cent chance will feel greater than that between 95 per cent and 90 per cent chance
because the 95–100 per cent is a qualitative difference from certainty to uncertainty

88 Kahneman and Tversky (n 77).

89 Kevin Weinfurt, ‘Decision Weights’ in Michael Kattan (ed), Encyclopedia of Medical Decision Making
(Sage 2009) 362.
90 Amos Tversky and Peter Wakker, ‘Risk Attitudes and Decision Weights’ (1995) 63(6) Econometrica
1255.
91 Kahneman and Tversky (n 77) 263–67.
92 Kahneman(n 82) 315.
93 Steven J Kachelmeier and Mohamed Shehata, ‘Examining Risk Preferences under High Monetary
Incentives: Experimental Evidence from the People’s Republic of China’ (1992) 82(5) Am Econ Rev
1120.
248  Prospect Theory and due process paranoia

and the 95–90 per cent is not. The use of decision weights captures this difference:
95–100 per cent has a decision weight delta of 20.7 and 90–95 per cent has a (much

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
lower) decision weight delta of 8.1.
The possibility and certainty effect and the non-linearity of the experience of
probability feel intuitive when presented with examples94:

when a loved one is wheeled into surgery, a 5% risk that an amputation will be
necessary is very bad – much more than half as bad as a 10% risk. . . . The psy-
chological difference between a 95% risk of disaster and the certainty of disas-
ter appears to be even greater: the sliver of hope that everything could still be
okay looms very large.

In Prospect Theory, to account for these differences in the weighting of probabilities

to predict (or explain) decisions, the value of an option is determined by multiplying
the value of the outcome by a decision weight rather than by the given probability.95
For example, we have seen in the coin flip discussed earlier—where heads pays £100
and tails pays £0—that an expected value calculation could value the option at £50
by adding 50 per cent of £100 and 50 per cent of £0. However, if the same situation
is analysed from the perspective of Prospect Theory and decision weights are used
for the calculation rather than the given probabilities,96 the coin flip would be valued
at £42.1. In other words, the decision weight corresponding with a 50 per cent prob-
ability (according to the experimental evidence compiled in Table 2) is 42.1, and the
decision weight 42.1 multiplied by £100 plus the decision weight 42.1 multiplied by
£0 is £42.1. Even though the ‘objective’ expected value of the coin flip is still £50, the
‘subjective’ expected value is only £42.1. It would still objectively be a good invest-
ment to pay £45 to buy the right to the flip, but the ‘subjective’ expected value sug-
gests that the decision maker would not be keen to do so. The decision maker
would, therefore, be inclined to make a risk-averse choice.
However, if the decision maker is offered a ticket with a 1 per cent chance of a
£100 return, the ‘objective’ expected value is £197 but the ‘subjective’ expected
value is £5.5.98 If the right to participate in this raffle is sold for £2, this is objec-
tively a bad investment leading to an, on average, £1 loss per time the decision
maker plays. However, the decision maker’s subjective experience of the risk sug-
gests (due to the overweighting of the low probability as a result of the possibility
effect) that he would be inclined to buy the ticket and, therefore, make a risk-
seeking choice.
Of course, the observations above are based on the general tendencies shown by
the average decision maker: actual risk attitude may vary from person to person and

94 Kahneman (n 82) 312.

95 Kahneman and Tversky (n 77) 280.
96 Though outcomes are valued according to their monetary pay-offs rather than in utility value, which
would be the usual approach when applying Prospect Theory.
97 1% * £100 þ 99% *£0 ¼ £1.
98 The decision weight corresponding to 1% is 5.5. 5.5 * £100 þ 91.2 (the decision weight corresponding to
99%) * £0 ¼ £5.5.
 Prospect Theory and due process paranoia  249

the above should be seen as indicators of human predispositions rather than an exact
decision-making tool.

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
3.2.2 The fourfold pattern
Kahneman and Tversky generalized such observations and combined them into the
fourfold pattern of risk attitudes, which they considered ‘the most distinctive implica-
tion of Prospect Theory’.99 The fourfold pattern uses the findings of Prospect
Theory to predict a decision maker’s approach to risk in high- and low-probability
situations and looks as follows (Figure 3).100

Figure 3. The fourfold pattern of risk attitudes.

The fourfold pattern almost reads like a cheat sheet of risk attitudes and helps to ex-
plain situations that are encountered regularly, such as buying lottery tickets (lower left
quadrant) and insurance (lower right quadrant) for higher prices than their expected
values would suggest.101 The most interesting legal application of the fourfold pattern
was offered by Guthrie, who offered the fourfold pattern (lower left quadrant) as an ex-
planation why claimants bring frivolous claims and, relatedly, why (lower right quad-
rant) respondents are likely to settle such claims on terms advantageous to the
claimant; hypotheses that he later confirmed in an experimental set-up.102

3.2.3 Does the possibility effect cause DPP?

The information that there is a high threshold for a successful Challenge to an award
on the basis of due process grounds (and, by corollary, the low probability of the

99 Tversky and Kahneman (n 81) 306.

100 Kahneman (n 82) 317.
101 Tversky and Kahneman (n 81) 316.
102 Chris Guthrie, ‘Framing Frivolous Litigation: A Psychological Theory’ (2000) 67 U Chi L Rev 163.
250  Prospect Theory and due process paranoia

Enforcement Risk) is available to any person with access to an arbitration textbook.

Arbitrators, who are usually appointed in part because of their legal and/or arbitra-

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
tion expertise, would (and should) be aware of this information too.
Anecdotally—on the basis of personal experience, the interviews for the 2015
Queen Mary Survey, and conference discussions—arbitrators are indeed aware that
the odds of an award being successfully challenged for due process reasons are ex-
tremely low. Nonetheless, and despite some arbitrators perhaps thinking that they
themselves are immune to DPP,103 DPP still appears to be an issue within the field
of international arbitration today.104
These observations imply that there is a discrepancy between arbitrators’ objec-
tive evaluation of the Enforcement Risk on the one hand and their behaviour on the
other hand: even though arbitrators’ academic evaluation of the Enforcement Risk
may be risk-neutral, their translation of these beliefs into action appears to lead to
risk-averse behaviour in practice. In other words, it appears that an arbitrator pre-
scribing how a due process-related decision should be made would arrive at a differ-
ent outcome than when it is described how that arbitrator actually makes the
decision in practice.
This observed behaviour corresponds with predictions by a Prospect Theory
model of decision-making under uncertainty. When looking at the two input ele-
ments of the fourfold pattern, ie the existence of a very low or a very high probability
of an event occurring and a (substantial) loss or gain arising as a result of the event,
due process decision-making by arbitrators certainly seems to fit within the frame-
work of analysis.
Although we do not have specific data available,105 the chance of a robust due
process decision leading to a (successful) Challenge in any pro-arbitration jurisdic-
tion is likely below 5 per cent, especially when the success rate is adjusted to exclude
cases involving egregious due process violations (which are situations that arbitrators
should be able to recognize and act upon in accordance with the applicable due pro-
cess standard) and only accounts for ‘borderline’ or ‘robust’ due process decisions
(and it is quite likely that the Challenge success rate is below 5 per cent even if such
an adjustment is not made). The probability of a successful Challenge is, therefore,

103 This is reminiscent of the overconfidence/egocentric bias study conducted on a group of 167 federal
magistrate judges in the USA, which found that 87.7% of the judges believed that at least half of their
peers had higher reversal rates on appeal. See Chris Guthrie, Jeffrey Rachlinski and Andrew Wistrich,
‘Inside the Judicial Mind’ (2001) Cornell Faculty Publications, Paper No 814 <https://scholarship.law.
cornell.edu/cgi/viewcontent.cgi?article¼1734&context¼facpub> accessed 8 April 2020. Similarly, arbi-
trators too are subject to this bias. ‘Arbitrators, like judges, are fallible. Arbitrators, like judges, make intu-
itive decisions that they might, or might not, override with deliberation. Arbitrators, like judges, are
influenced by anchoring, framing, representativeness, and egocentric bias. In short, arbitrators are like
judges, and arbitral decisionmaking is like judicial decisionmaking. Whether appointed by the state and
appearing in robes, or selected by parties and appearing in business suits, adjudicators are human beings,
and human beings make predictable judgment and decisionmaking errors’. See Franck and others (n 5)
1177.
104 As discussed in Section 2.3 above.
105 Gerbay (n 13); Remy Gerbay, ‘Due Process Paranoia (Part 2): Assessing the Enforcement Risk under
the English Arbitration Act’ (2017) <http://arbitrationblog.kluwerarbitration.com/2017/02/20/due-
process-paranoia-part-2-assessing-the-enforcement-risk-under-the-english-arbitration-act/> accessed 8
April 2020.
 Prospect Theory and due process paranoia  251

expected to be squarely within the possibility effect range and Prospect Theory sug-
gests that arbitrators will be inclined to overweight the possibility that the low-

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
probability event of a successful Challenge occurs.
If a Challenge is successful, that would be experienced by the arbitrator as a
great loss. There will be several elements contributing to the feeling of incurring a
loss, including the cost and delay to the arbitration. However, arguably the biggest
perceived loss by the arbitrator will be the loss of reputation if the Challenge is suc-
cessful, especially if the Challenge is based on due process grounds. Moreover,
these reputational effects will likely ‘outlive’ the case and might resonate in a rela-
tively small arbitration community where work is often based on referrals from
that same community.
When confronted with a low probability event that might lead to a substantial
loss, the fourfold pattern (lower right quadrant) implies a tendency for decision mak-
ers to show risk-averse behaviour. In practice, the risk-averse choice is to grant the
due process-related request and thereby exclude the possibility of the substantial
loss. The choice, however, comes at a cost with respect to other aspects of the arbi-
tration, such as higher arbitration costs and delay to the arbitration.
When reviewing the example of an arbitrator facing a due process-related decision
discussed earlier from the perspective of Prospect Theory, it is possible to re-
evaluate the decision by replacing the given probabilities with their associated deci-
sion weights. The table of outcomes for a robust due process decision then looks as
follows from the arbitrator’s perspective (Table 3).
Table 3. Utility of successful or unsuccessful challenges and their decision
weights
Outcome Utility Decision weight

Successful Challenge 100 (significant loss) 5.5 (1% probability)

No (successful) Challenge þ2 (small gain) 91.2 (99% probability)

The expected utility of this option can be calculated as (5.5 * 100) þ (91.2 * 2)
¼ 3.7. Still assuming that granting the due process request will have a negative
pay-off of 3, the (subjectively) ‘optimal’ decision from the arbitrator’s perspective
is then to grant the request.
For that reason, whereas objectively the best option is to make a robust due pro-
cess decision (on the basis of the facts and pay-offs as set out at Section 3.1), the
possibility effect and related overweighting of the low probabilities demonstrate that
it is more likely that the arbitrator will be inclined to grant the due process request.
This hypothesis corresponds with the approaches to due process decisions that are
observed in practice.
It is important to keep in mind that this example is not meant to be an exact rep-
resentation of a due process decision in practice: the outcomes and probabilities are
arbitrary and may vary per decision, arbitrators might have different risk attitudes
based on personal traits, or arbitrators may be more or less sensitive to the down-
sides of having a decision successfully challenged. Instead, this example is used to
demonstrate that the average arbitrator will have a tendency to make a risk-averse
252  Prospect Theory and due process paranoia

choice in a similar situation, by virtue of being human. For that reason, whereas it is
not possible to conclude that DPP is caused by the possibility effect, it can be said

Downloaded from https://academic.oup.com/arbitration/article-abstract/36/2/233/5857622 by Max Planck Institute Luxembourg user on 06 July 2020
with a degree of confidence that the possibility effect contributes to DPP.

4 . C O N C L U S I O N — T H E B R O AD E R RE L E V AN C E O F P R O S P E C T
T HE O R Y F O R A R B I T R A T IO N
This article has attempted to show how concepts of (behavioural) decision theory
and modelling of decision-making can be used to explain and analyse arbitrator deci-
sions. It did so by applying Prospect Theory to the tendency perceived in arbitral tri-
bunals to make ‘paranoid’ due process decisions. Prospect Theory and particularly
the possibility effect could also explain other phenomena encountered in interna-
tional arbitrations. For example, it might help explain why parties keep challenging
awards even though the chance of success is so slim,106 or why they keep challenging
arbitrators when such challenges are so rarely granted.107 In each of these instances,
it is a low-probability, high-gain event (akin to a lottery) and, therefore, induces risk-
seeking behaviour.
The purpose of this article was two-fold: first, to bring the fascinating Prospect
Theory to the fore; and, second, to better understand how arbitrators may have a
natural tendency to make suboptimal due process decisions, thereby unconsciously
feeding the phenomenon of DPP. Understanding how these processes work—and
that tribunals are perhaps not ‘weak’ but just human—might be a first step to cor-
recting them.

106 In the major arbitration seats, the success rate of challenges to arbitral awards is very low. For England,
see <https://www.judiciary.uk/wp-content/uploads/2015/10/Eder-Speech-Dec-2014.pdf> accessed 8
April 2020. For the USA, see <https://www.alston.com/-/media/files/insights/publications/2015/10/
your-arbitral-awards-are-safe-us-courts-are-defere/files/arb102715cm/fileattachment/arb102715cm.
pdf> accessed 8 April 2020. For Switzerland, see Felix Dasser, ‘International Arbitration and Setting
Aside Proceedings in Switzerland: A Statistical Analysis’ (2007) 25 ASA Bull 444, 454.
107 A review of the outcome of challenges to arbitrators in ICC arbitration between 2001 and 2011 shows
that the rate of success varied between a low of 3.8% and a high of 13% only. Gerbay (n 10) 233.