Go to Database Directory || Go to Bibliography
Search the entire CISG Database (case data + other data)
Reproduced with permission of Vindobona Journal of International Commercial Law & Arbitration (1/2009) 157-178

Rats in the Kaleidoscope: Rationality, Irrationality, and the Economics and
Psychology of Opting in and Out of the CISG (Kaleidoscope Part II)

Lisa Spagnolo [*]

  1. Introduction
  2. Economics, Psychology and Choice of Law
    2.1     Path dependence (Unfamiliarity - Blue)
    2.2     Agency, heuristics and externalities (Learning costs - Yellow)
    2.3     Polar herds, critical mass and game theory (Blue)
                        2.3.1     Group polarization and the norm of automatic opt-outs
                        2.3.2     Game theory and the strategy of automatic opt-outs
    2.4     Iteration, networks and external shocks (Bargaining strengths, Red)
                        2.4.1     Group polarization - Environmental change leading to new norms
                        2.4.2     Game theory - Altered cost-benefits leading to new strategies
    2.5     Internalisation of learning externalities and transparency (Orange)
  3. Conclusion


In Kaleidoscope Part I, the major factors for choices of law where the CISG is concerned were described as fragments within a kaleidoscope: the rationales of unfamiliarity (blue), learning costs of becoming familiar (yellow), bargaining power (red) and substantive concerns (green). The view seen by a lawyer making such a choice differed from jurisdiction to jurisdiction depending on the size and position of the fragments from their perspective, with some views being 'pro-CISG' and others prone to 'blind' or 'automatic' opt-outs. The greenery of substantive concerns will not be revisited here. In this paper, I propose to refocus on the remaining three fragments through economic and psychological prisms.


The frameworks of neoclassical and behavioural economics are well known. Neoclassical economics proceeds on the basis that perfectly competitive markets efficiently optimize expected utility. It assumes parties make rational choices on the basis of perfect information in the absence of transaction costs and externalities.[1] [page 157] Implicitly, rational choice theory assumes actors make decisions on the basis of a cost-benefit analysis.[2] Its so-called rival behavioural economics, attacks the assumption of rational choice and perfect information.[3] Behavioural economics draws on cognitive psychology to explain why choices in the real world often seem irrational.

Rational choice theory is still useful for its predictive qualities [4] provided its limitations are acknowledged. Behavioural economics and cognitive psychology provide nuances that reflect real life complexities. I will draw on both to analyse how unfamiliarity, learning costs and bargaining strength contribute to choices of law.

2.1 Path dependence (Unfamiliarity - Blue)

The practice of 'automatic' or 'blind' exclusion of the CISG by lawyers due to unfamiliarity linked to learning costs was discussed in Part I. 'Automatic opt-outs' can be defined as exclusions from the CISG without proper consideration as to whether it is the more appropriate law for the particular transaction in question. In neoclassical economics, such decisions are irrational as they are based on imperfect information and do not involve cost-benefit analysis. They will often be suboptimal for the businesses involved. Yet, Schroeter has aptly described choice of law questions, in the minds of business people, as 'like the entrance of a distant relative at a wedding reception'.[5] Businesses are unlikely to ponder such choices in everyday transactions. [page 158] At the business coal-face, a great deal of automatic opting out occurs via general conditions in standard forms contracts drafted long ago, possibly with legal advice. Redrafting those forms with legal input for most businesses can only occur on occasional intervals, with smaller business perhaps seeking advice less frequently than larger businesses. As legal advice is a form of information cost,[6] small-medium businesses face proportionately higher information costs, and are therefore more prone to imperfectly informed decision making and concurrent efficiency losses from less-suitable choices of law.[7]

Intervals between redrafts or advice can be characterised as periods of 'path dependence', a term borrowed from behavioural economics. Path dependence occurs when people are 'locked-into' suboptimal choices because of a historical path of events.[8] By relying on what they have done in the past, businesses make path dependent choices that violate the requirement of deliberative, rational choice in neoclassical economics:[9] behaviour aptly described by Schroeter as engaging in cognitive dissonance to avoid the unwelcome guest.[10] Effectively, the decision maker chooses not to decide and ignores any potential gains from unexplored paths.[11] Of [page 159] course, one could identify this as a rational decision if it deliberately traded-off costs of legal advice against efficiency gains from the application of a more suitable law, since it might simply not be worthwhile. However, by definition, this is not what occurs in automatic opt-outs.

I have no problems about this state of affairs. What is of concern is the same 'path dependent' behaviour in lawyers in some jurisdictions. It is at the point of interval, where clients seek legal input into drafting their standard or individual contractual terms where the rationale of unfamiliarity becomes a major ethical and legal issue. In not considering whether the CISG is an appropriate choice in their client's individual circumstances, lawyers act against their client's best interests.[12] Particularly in a Contracting State, the lawyer who automatically excludes risks liability for malpractice [13] and reprimand by his professional practice body.

Would it be sufficient to argue that the practice of exclusion was dictated by law firm policy or by the firm's precedent forms of contract? The excuse is likely to fail, because the CISG offers advantages over domestic law in many situations.[14] Even where the client prefers or insists on the CISG's exclusion, as a professional advisor the lawyer has a duty to advise on the advantages forgone.[15] Should the client [page 160] ultimately choose to ignore the advice, the lawyer will have at least discharged his duties, and ensured the client made an informed choice.[16]

2.2 Agency, heuristics and externalities (Learning costs - Yellow)

Given the ethical issues and potential for liability or professional misconduct proceedings, why might some lawyers persist with automatic opt-outs?

One can view the lawyer's behaviour as simply a matter of rational choice. In high opt-out jurisdictions, lawyers might think that so few of their peers are familiar with the CISG that the risks of their clients becoming aware of their failure to consider the CISG are low.[17] They might therefore be unwilling to invest in becoming familiar with the CISG [18] or, might be willing, but just not have the time.

The thought process of the lawyer and how it affects the client's interests is not at all obvious to the client involved.[19] The client might eventually discover the inadequacy of the advice if they later engage another lawyer who appreciates the poor decision making of their predecessor, or during subsequent litigation stages.[20] Agency problems arise when interests of the principal and agent do not match precisely and an information asymmetry exists.[21] The agent can follow his own interests to the [page 161] detriment of the client.[22] If the lawyer can still earn the same fee without investing in 'start up' costs of familiarisation and not lose business, the lawyer can rationally maximize his short term gains. The client unwittingly bears the detrimental effects of the lawyer cutting corners, i.e. efficiency losses from suboptimal choices of law for many transactions. In neoclassical terms, these are market distortions in the form of negative externalities [23] that preclude optimal allocative efficiency in the legal services market because their effect is not recognised in the pricing of legal fees. Hence, in the kaleidoscope of high opt-out jurisdictions, costs are pushed out of the lawyer's choice of law 'viewfinder' and hidden in the relative obscurity of the client's sphere of gains and losses.

I do not mean to contend that all lawyers do this in any cold or calculating way. The reality is that in jurisdictions where familiarity is low, the decision-making process of the majority of lawyers is probably a function of unfamiliarity that ultimately leads to path dependence. If the lawyer is completely unaware of the CISG, he will simply not consider it at all. If he knows of it, imperfect information can still cloud his judgment. Low familiarity may mean he is simply unaware of the advantages his client will lose by discarding the possibility of the CISG, or may cause him to harbour unwarranted substantive concerns about its provisions or to exaggerate perceived uncertainty in its outcomes.[24] There may also be a related underestimation of choice of forum risks.[25] In such a jurisdiction, there may also be an element of personal reputational risk.[26]

Past choices of law are irrationally attractive due to a cognitive bias known as the status quo bias.[27] Potential losses from change are overestimated, while potential gains tend to be underestimated. Let me illustrate with an example given by Goldstein and [page 162] Gigerenzer.[28] Rats are 'neophobic' when it comes to food. At an instinctive level, rats know that everything they have eaten to date has necessarily failed to kill them, so they are extremely reluctant to eat anything new. Without wishing to compare our noble profession to rodents, lawyers may also be very reluctant to choose the CISG due to an irrationally inflated 'fear of the unknown'.[29] It is a case of 'green eggs and ham'.[30] The unfamiliarity heuristic or instinct will be heightened if the lawyer never encountered the CISG in law school, or it looks very different from local domestic law, or he has never encountered it in dispute work (or all three) [31] -- so their rule of thumb is to opt out, if the other party will let them.

We can dig a little deeper into the decision-making process. Choice of law is a highly complex decision. Like many complex decisions, we must often make them quickly, in less than ideal conditions.[32] Rather than optimise utility, we tend to cope with complexity by simplification of decision-making strategies. Sticking to what has served us well in the past is one such strategy. Perfect or even good knowledge about the CISG involves some costs to the lawyer in terms of time, effort and sacrificed billable time. Like all humans, lawyers operate under constraints of limited time, knowledge and capability. Thus the rationality every lawyer employs is 'bounded rationality',[33]> whereby shortcuts, known as heuristics are developed to streamline decision making.[34] There is nothing wrong with that, provided the heuristics fit the environment in which the lawyer finds herself.[35] [page 163]

The automatic opt-out lawyer is making do, or 'satisficing'.[36] Like the rat, she prefers to stick to the familiarity of 'generally opting out'. Satisficing is one form of bounded rationality, employing a particular heuristic, the aspiration level, as the threshold of satisfaction.[37] Rather than search for the 'sharpest needle in the haystack', the decision maker settles for 'one which is sharp enough' for sewing.[38] Once this aspiration is achieved in a particular deal, no further thought is given to potential alternative choices of law.[39] Aspiration levels serve to reduce search costs in complex problems by limiting outcomes to a confined number considered 'OK'.[40] The outcomes sought will depend upon prior aspiration levels and how difficult or easy it was to achieve them in the past.[41] Aspiration levels can comprise a single outcome.[42] In high opt-out jurisdictions, this is likely to be a singular preference to exclude the CISG in favour of a more familiar domestic law. As an automatic preference, aspiration level or heuristic,[43] this achieves the lowest possible information search costs.[44] The lawyer's position as an agent combined with similarly low aspirations on the part of opposing [page 164] lawyers creates the perfect environment for this shortcut to flourish,[45] despite the suboptimal outcome in many cases for one or both clients.[46]

Whether by design or as a matter of less calculated satisficing, the front end lawyer operating in a highly unfamiliar jurisdiction is behaving efficiently, at least from his own perspective, since he avoids learning costs but earns the same fee.[47] If risks remain low and no other pressure comes to bear he maximises utility in the short term. But this behaviour might be far less than rational in the longer term.

Can the situation of lawyer automatic exclusion path dependence with its concomitant negative externality continue indefinitely in jurisdictions such as the U.S., Canada and Australia? I don't think so. The ability of lawyers to continue avoiding learning costs may be eroded over time. There could be a shift in the (rational) cost-benefit balance, something that might be alternatively described as a change in the environment in which the (behavioural) heuristic must work.[48] Change of this type would occur if clients begin to litigate for malpractice or professional bodies receive complaints. If one looks at the quality of some CISG cases in jurisdictions where cpmc or cptd is low,[49] then one is tempted to think this will happen sooner rather than later.[50] Yet, a much more likely prospect is the forced internalisation of learning costs due to interaction with stronger pro-CISG players. To explain how this might occur, it is necessary to locate the forces behind the current status quo in such jurisdictions.

2.3 Polar herds, critical mass and game theory (Blue and Yellow)

Game theory and the notion of group polarisation can help clarify how the norm of automatic exclusion came to be widely accepted in high opt-out jurisdictions in the first place.

2.3.1 Group polarization and the norm of automatic opt-outs

The creation of norms is described in behavioural science. Social cascades depict the process of adoption of widespread beliefs, but do not necessarily involve [page 165] deliberation.[51] 'Group polarisation' observes the tendency for groups of like-minded people after group deliberation to migrate toward a more extreme view than those initially held by individual group members.[52] Errors are 'amplified'.[53]> Group discussion has an informational influence: individuals tend to defer to views of other group members either because they are persuaded by, or wish to be part of the majority.[54] Discussion tends to be asymmetrical.[55] As pre-existing views are continuously reinforced,[56] the 'pool' of acceptable views shrinks; consequently, counterviews are repressed and aggregation of all knowledge held by group members is prevented.[57] Notably, informational influences are only effective where individuals possess little or no private information upon which they could reach their own conclusions [58] making CISG unfamiliarity of key importance in the process of group polarisation. There is safety in numbers, and sticking with the majority allows one to at least 'share the blame'[59] if proven incorrect. The second reason group deliberations lead to extreme views is the existence of social pressure, particularly in the form of reputational sanctions.[60] Even if an individual knows the group view is wrong, the individual may go with the flow rather than risk sanctions, disapproval or even hostility of the group.[61] [page 166]

The overall effect is one of a decrease in variance between members of a deliberating group and reinforcement of an extreme view which may in fact be wrong.[62] The effect is observed amongst groups of experts,[63] so law firms and the legal profession are not immune. In jurisdictions with low levels of familiarity, lawyers armed with little or no private CISG knowledge and high learning costs are prone to absorb the informational signals of the law firm majority inherent in firm precedents or pro forma contracts that invariably exclude the CISG, and the neophobic attitudes of their colleagues. Automatic exclusion of the CISG is within the 'acceptable' pool of views in the firm while pro-CISG views are not, and thus go unheard. Even lawyers armed with high levels of familiarity may deliberately choose not to stand against the majority and risk peer sanction by speaking against the 'party line', out of concern for their own reputations, epitomised by the perception that to do so would be 'sticking one's neck out'.[64] Reputational concerns are heightened by 'affective ties' between the lawyer, his firm, and the profession through physical distance, reputational capital, and promotional opportunities.[65]

Thus blindly opting out, although extreme and suboptimal, can persist within a group such as a firm or jurisdiction, even if irrational.[66] This explains why current norms of automatic exclusion are achieved and maintained as the group status quo in some jurisdictions.

2.3.2 Game theory and the strategy of automatic opt-outs

Even presuming rational choice, the need for individual lawyers to invest in learning costs to obtain CISG familiarity may contribute to a 'collective action problem'.[67] Collective action problems arise whenever others can obtain 'free rides' from another's investment, such that the investor might not recoup the cost or obtain all the benefits.[68] Free rides are also known as positive externalities [69] because they benefit [page 167] those who have not invested in the cost. Thus, there is a disincentive to invest. Are lawyers who have not yet invested in CISG familiarity rationally discouraged from doing so by an inability to gain from their investment? I will confine the following analysis to the position of an individual lawyer working within a law firm in a predominantly opt-out jurisdiction such as Canada, the US or Australia. Of course, in reality, many will be oblivious to the advantages of investment in CISG familiarity. However, in arguendo, I will presume some minimal appreciation of potential advantages.

Game theory assists in explaining collective action problems. It maps out rational choices where outcomes are contingent upon the combined choices of all decision makers.[70] CISG investment decisions are multidimensional because they yield consequences for the lawyer, law firm, clients and profession.

At a societal level, in a predominantly opt-out jurisdiction, expenditure on 'start up' costs is a good thing. Accumulation of expertise within the profession benefits all clients as better quality advice becomes more widely available. For clients, the result is more efficient choices of law [71] and fewer negative externalities.[72] Investment also improves competitiveness of the profession vis-a-vis professions in other jurisdictions to the benefit of all lawyers within it. In this sense, free rides or positive externalities are enjoyed by clients and the whole profession.

In high automatic opt-out jurisdictions, a bold lawyer might capture an innovator's 'early mover advantage'.[73] The lawyer would expect to capture some of his clients' efficiency gains through higher fees.[74] Beside ethical benefits of providing proper advice, he may perceive concurrent reduction of risks such as potential malpractice actions or professional body complaints. However, the picture is not so rosy. Even when aware of these potential benefits, the individual's latent cognitive biases may be reinforced if his peers do not value investment in CISG familiarity. Despite the irrationality of the group norm,[75] this makes the CISG a less feasible choice simply [page 168] because it requires deviation from the safety of the 'herd'.[76] The irony is that each lawyer has a professional and ethical responsibility to consider the CISG as a legal tool whenever suitable to the transaction at hand, and thus an obligation not to be blindly led by 'the mob mentality' or by his own unfamiliarity.[77] Yet in reality, deviation from firm norms risks disapproval and endangers career opportunities if the lawyer is viewed as too radical.[78] The risk of a client law suit might pale into distantly contingent irrelevance compared to the immediate disapproval of one's peers.

The firm could choose to invest in CISG learning costs directly by making training available. A firm indirectly decides whether or not to invest in the way it structures rewards and sanctions, and how it values firm skills and expertise. Organisations reward 'successes' far less than they punish 'failures'.[79] This means that 'successes' from individual investment like increased fees and improved firm skills will be undervalued and rewarded only mildly, unless the firm also decides to invest (directly or indirectly). The 'failure' of exposure to liability will be overinflated by the firm, but its perspective will differ: if it decided to invest, it probably perceived exposure to liability due to unfamiliarity; whereas if it decided against investment, it may view CISG choices of law as increasing firm exposure. The firm's reputation within a conservative profession must also be considered: it will overvalue losses and undervalue gains from changes to the status quo.

If neither the lawyer nor firm knows in advance with absolute precision whether the other will decide to invest, then the result is a typical game.[80] In the well known Prisoner's Dilemma, two prisoners must independently decide whether to remain silent or testify against the other prisoner (hoping the other stays silent).[81] In the [page 169] Herder Problem, two farmers can choose to place further cows on a communal meadow, or act for the greater good by refraining.[82] The 'choice of law game' is notionally played each time the lawyer drafts a choice of law clause. Interests of the lawyer and firm overlap resulting in a non-zero sum game.[83] Although relevant decisions are made by the lawyer and his firm, payoffs are borne by clients and the profession. The following table sets hypothetical values for the purposes of illustration: for the lawyer on the left side of the bracket; the firm in the centre; and clients and profession on the right side.

Table 1: Opt-Out Dominant Jurisdiction:
        INVEST       NOT INVEST
      INVEST       (3, 1,+20)

      (0, 3,+20)
      NOT INVEST       (2, 0,-20)

      (1, 2,-20)

Cooperative investment by both lawyer and firm is optimal if we include externalised effects for clients and the profession since this produces the best outcome for society as a whole (+24 units). This is also the best outcome from the collective perspective of just the firm and lawyer (+4 units). Nonetheless, when each decision maker looks at the decision self-interestedly, they will probably conclude that the investment is not a [page 170] good idea. From the firm's perspective, non-investment is better no matter what the lawyer decides. If the lawyer invests, the firm is better off not investing (3 units vs. 1 unit); the same is true if the lawyer does not invest (2 units vs. 0 units). Thus, non-investment is the firm's optimal strategy.

The lawyer might prefer to invest, but fears the consequences if the firm does not do likewise in the form of firm undervaluation and sanctions. For the lawyer, if the firm invests, he is slightly better off investing (3 vs. 2 units), but if the firm does not, he is worse off having invested (0 vs. 1 unit). Investment is the riskier strategy for the lawyer. It can either go very well or very badly yielding between 3 to 0 units (as compared with 2 to 1 units for non-investment). If he thinks the firm is unlikely to invest, a reasonable guess given the prevailing norm in the jurisdiction he is, on average, better served by non-investment. For example, if he anticipates the likelihood of firm investment at just 30%, his expected gains from investment are just 0.9 units compared with 1.3 units from non-investment.[85] A rational individual lawyer would not invest in learning costs. The probable outcome of this game is that neither firm nor lawyer will invest a suboptimal result from their collective perspectives and for society.[86]

Unlike the prisoners, who cannot communicate and have no second chances,[87] the farmers replay their game every year. The vast difference is in repetition and communication.[88] The farmers can cooperate to prevent overgrazing. Likewise, our choice of law game is repeated with each choice of law decision so the possibility of cooperation and learning arises.[89] The lawyer can seek to influence payoffs by persuading peers of the irrationality of automatic CISG exclusion.[90] Unfortunately, firms probably do not 'communicate' in the cooperative game theory sense.[91] In trying to persuade peers, the lawyer might incur the very sanctions he seeks to change. Consequently, conditions are ripe for 'excess inertia', despite the fact that societal [page 171] benefits outweigh costs.[92] Without cooperation, one might conclude high exclusion jurisdictions are doomed to self-perpetuating automatic opt-outs.

2.4 Iteration, networks and external shocks (Bargaining strength, Red)

In the long-term, it is doubtful that the decision not to invest in learning costs will continue to be viable. International trade by its nature involves interaction between lawyers from various jurisdictions. Predominantly opt-out jurisdictions cannot completely inoculate themselves from all CISG exposure. The existence of stronger bargaining parties from a pro-CISG jurisdiction like China can effectively substitute for communication and cooperation in iterated games and slowly change the group's environment. As more lawyers in a predominantly opt-out jurisdiction are newly exposed to pro-CISG counterparties, a 'network effect'[93] of increasing returns can slowly arise and CISG familiarity may be increasingly valued as more lawyers become marginally familiarised in the course of practice. Nevertheless, how does a firm or entire profession arrive at a new norm?

2.4.1 Group polarization - Environmental change leading to new norms

External shocks can change a group view. At some point after an external shock the group will reach a critical mass or 'tipping point'[94] whereby a new view prevails as the norm.[95] A paradigmatic shift can be triggered by changes to incentives or perspectives.[96] In essence, this is exactly what happens when encounters with a stronger pro-CISG jurisdiction such as China makes familiarity increasingly valuable. The process of group deliberation begins again but this time gradually reversing the [page 172] previous view and arriving at a new one. Societal cascades may spread the new norm across the jurisdiction by similar processes, even without deliberation.[97]

Like game theory, behavioural science emphasises the frequency of repetition.[98] Obviously the frequency with which jurisdictions deal with counterparties from pro-CISG jurisdictions affects the speed with which network effects lead to revaluation of CISG skills. In turn, this affects the time taken to reach the point of critical mass at which blind opt-outs are rejected in favour of a new norm. Indeed, in a decisionmaking environment of frequent exposure to pro-CISG counterparties, decision makers will rapidly appreciate the long-term danger of losing business if they dared not invest in CISG 'start up' costs.[99] The profession is also more likely to perceive unfamiliarity as a threat to its competitiveness.[100]

2.4.2 Game theory - Altered cost-benefits leading to new strategies

When forced to deal with the CISG, even at low levels, firms might slowly adjust sanctions and incentives because they are forced to invest. Investors are assured of recapture through higher fees and more frequent CISG transactions; the cost-benefit structure is altered leading to new rational strategies. Consequently, the hypothetical risk-reward structure under the new norm in a jurisdiction having frequent contact with pro-CISG jurisdictions might look more like Table 2.

Table 2: After Frequently Dealing with Stronger
        INVEST       NOT INVEST
      INVEST       (3, 3,+20)

      (2, 1,+20)
      NOT INVEST       (2, 2,-20)       (0, 0,-20)

As before, from a societal perspective, the best solution is for both the lawyer and the firm to invest (total benefit of 26 units). Moreover, it is now clear to both the lawyer and firm that they should invest. For the firm, investment is the best decision, [page 173] irrespective of the lawyer's decision, since it will enhance gains in either case. For the lawyer, regardless of the direction the firm takes, investment yields better results than non-investment due to a changed reward structure.[101] As reformulated after the external shock, there is a new solution to the game. If all act rationally, the decisions will converge upon a solution collectively optimal for the decision makers themselves, and for society.

Pro-CISG China, with its (still) massive economic power, will inevitably change the choice of law landscape, whether viewed from the perspective of global or limited rationality.[102] There are signs that in some European nations, although opt-outs are still prevalent, the decision-making environment may already be changing,[103] although for reasons of survey design, this might not always be obvious from the data.[104] It might even be the case that in the U.S., opt-out behaviour is slowly beginning to change.[105] Additionally, there are other opt-out predominant jurisdictions in which lawyers do not favour opting out, but have so far had little choice because of the superior bargaining power of counterparties from predominantly opt-out nations.[106] It will take little for social cascades and network effects to unfold in those jurisdictions. My argument is that the decision-making environment that, in some jurisdictions, has [page 174] fostered externalisation, path dependence or the satisficing heuristic of automatic opt-outs is slowly coming to an end. The bargaining strength of pro-CISG China may be just the catalyst needed to tilt the decisional environment in the direction of more efficient choices of law.

2.5 Internalisation of learning externalities and transparency (Orange)

If, with a turn of the kaleidoscope, we bring into focus the fact that superior bargaining strength is sometimes held by pro-CISG counterparties whose standard terms either directly or indirectly opt in, we find that other factors in the individual lawyer's decision are profoundly altered. Moreover, the identity of those privileged to see the view might actually change.

The lawyer who has to date been 'satisficing' is suddenly faced with a stark choice. She could still insist on the opt-out anyway and allow her client to bear, in addition to background efficiency losses, a worse bargain, which the party in the superior position will exact in terms of price,[107] other contractual terms, or even loss of the deal. Alternatively, she could bite the bullet and invest in the time and effort of becoming familiar with the CISG.

Thus, pro-CISG bargaining strength is extremely powerful because it has the power to make more transparent to the client the real cost of sticking to the lawyer's preferred 'path'. The client faced with disadvantageous terms suddenly has reason to question why it is so important not to use the CISG. Pressure from a client asking for an explanation and potential embarrassment at his own inability to give a professionally competent answer, might prompt the lawyer to reconsider investment in learning costs.

Yet, as discussed in relation to game theory analysis, it is the iteration of this pressure which is the most important characteristic of bargaining power in the hands of a pro-CISG jurisdiction with superior economic power. When the isolated embarrassment of a single less advantageous deal has passed and our satisficing lawyer is faced time and again with similar situations, the prior sense of inadequacy is more likely to be actively addressed.

One can only 'fudge it for so long. Repeated interactions with stronger pro-CISG parties reveal to clients the true expense of unfamiliarity. It follows that this erodes the ability of lawyers to segregate the negative impact of unfamiliarity from pricing of legal fees. [page 175]

Transparency increases the likelihood that an individual lawyer who fails to invest in learning costs will simply become uncompetitive in a global legal market.[108] Internalisation becomes increasingly attractive if firms begin to notice the trend and adjust their risk-reward structures accordingly. Through this new perspective, the previously 'insurmountable' learning costs may relatively speaking, seem to magically shrink.

Pressure from clients as a by-product of transparency may reinforce the process of reaching 'critical mass'. If that happens, then jurisdiction by jurisdiction, we may see the formerly automatically excluding 'rats' opting out of the sinking ship of automatism in choice of law. That is precisely why bargaining power is so important; it largely alters the hue of learning costs in high opt-out jurisdictions from an optional risky strategy to a basic training requirement necessary for professional survival.[109] It has the key potential to shift the scales in jurisdictions like the US by changing the decision-making environment from one in which automatic opt-out satisficing is feasible to one in which it is not.

Lawyers will no doubt still 'satisfice', but at a new aspiration level. As lower aspiration levels become more difficult to attain, they are generally altered.[110] Under the pressures described above, the new heuristic must involve consideration of the CISG on a case-by-case analysis, perhaps reduced to some key elements, unless there is no choice due to bargaining position. This necessarily involves greater familiarity. If the number of jurisdictions in which lawyers are less prone to opting out slowly climbs, we can expect lawyers in remaining jurisdictions to feel increasing pressure to revise their heuristic toolbox to fit the new environment.[111] Repeated encounters with [page 176] the CISG will be the key to alerting individuals and the profession to the need to raise their aspirations to seriously consider the CISG.[112]


The CISG was never intended to 'claim a monopoly over international sales law'.[113] Yet, its underutilisation in some jurisdictions is due to factors apart from its true value for many transactions and clients. The market failure identified here is not the failure to adopt the CISG universally, but the failure of some lawyers to properly consider it as a potential choice of law. Automatic exclusion of the CISG in some jurisdictions is a form of path dependence reinforced by peer groups which maintain the status quo. Their existence is dependant on environments of CISG unfamiliarity and high learning costs, which are conducive to the satisficing behaviour of blind exclusion of the CISG.

My contention is that, in the long-term, path dependent automatic opt-outs are unsustainable. Even low level exposure to pro-CISG counterparties carries the potential for network effects that can slowly increase the value placed on CISG skills, gradually alerting clients to efficiency costs borne by them and awakening firms to changed conditions in the market for legal services. Frequent or repeated exposure may act as an external shock, leading to a tipping point within firms and across the entire profession. In the new environment, it could become imperative for competitive survival to invest in CISG knowledge, and eschew blind opt-outs. New heuristics will be required, and satisficing will need to take place at a more sophisticated aspiration level, resulting in more efficient choices of law for clients.

There is no doubt that China can be a potential catalyst in the correction of this market failure. China's relatively strong economic position and similar trends in ASEAN nations could ultimately serve to slowly force CISG exposure on more reluctant jurisdictions such as the U.S., Canada and Australia, where environments of unfamiliarity and high learning costs still harbour automatic opt-outs. In terms of the kaleidoscope, the 'bargaining strength' fragment might perform a more powerful role at the drafting end than forced familiarisation through CISG dispute work. This is particularly significant in jurisdictions with little CISG dispute work, or where firm or professional structures preclude such exposure from filtering through to drafters.[114] In jurisdictions suffering both problems, where avoidance of high learning costs is well hidden, the influence of bargaining strength in re-colouring previously endless views of blue and yellow is particularly welcome. [PAGE 177]

There will be much structural change by the time we escape our current worldwide economic malaise. One would hope that as we emerge, the advantages inherent in having a frequently utilised uniform international sales law will have become increasingly apparent for the benefit of individual businesses and the world economy. If you are a gambler, or a rat, put your money on the red fragment. [page 178]


** Kaleidoscope Part I appears in this volume of the Vindobona Journal of Internatoinal Commercial Law and Arbitration (Vindobona Journal): "A Glimpse Through the Kaleidoscope: Choices of Law and the CISG" (Kaleidoscope Part I) (2009) 13 Vindobona Journal 135.

* Faculty of Law, Monash University, Australia. All websites last checked 16 March 2009 unless otherwise stated. This paper was presented 2 April 2009 at the MAA 'Issues on the CISG Horizon: Conference in Honour of Peter Schlechtriem (1933-2007)' in Vienna. I wish to thank Kwame Mfodwo and Patrick Emerton for their (always) helpful discussions, and my three little rats, Matt, Kier and Ellanya.

1. Arrow, K. J., Existence of An Equilibrium for a Competitive Economy, in Arrow, K. J. (ed), Collected Papers of Kenneth J. Arrow, 1984, Basil Blackwell Vol. 2: General Equilibrium 58, at p. 60 [hereinafter Collected Papers]; Stigler, G. J., The Theory of Price, 4th ed, 1987, MacMillan, New York, at pp. 43-74, 320-28; Stiglitz, J. E., Principles of Microeconomics, 1993, Norton, New York, at pp. 493-502; Korobkin, R. B. and Ulen, T. S., "Law and Behavioural Science: Removing the Rationality Assumption from Law and Economics" (2000) 88 California Law Review 1051, at pp. 1080 fn 104, 1055, 1060-66.

2. Korobkin, R. B. and Ulen, T. S., supra fn 1, at 1082; Sunstein, C. R., "Cognition and Cost-Benefit Analysis" (2000) 29 Journal of Legal Studies 1059; Stiglitz, J. E., supra fn 1, at pp. 28-29.

3. Simon, H. A., "A Behavioral Model of Rational Choice" (1955) 69 Quarterly Journal of Economics 99, at p. 101; Simon, H. A., "Rational Choice and the Structure of the Environment" (1956) 63(2) Psychological Review 129, at p. 129; Tversky, A. and Kahneman, D., "Prospect Theory: An Analysis of Decision under Risk" (1979) 47 Econometrica 263; Simon, H. A., "Rational Decision Making in Business Organizations" (1979) 69(4) American Economic Review 493; Tversky A. and Kahneman, D., "Judgment Under Uncertainty: Heuristics and Biases" in Kahneman, D., Slovic, P. and Tversky, A. (eds), Judgment Under Uncertainty, 1982, Cambridge, 1, at p. 3; Tversky, A. and Kahneman, D., "Rational Choice and the Framing of Decisions" in Hogarth, R. M. and Reder, M. W. (eds), Rational Choice: The Contrast between Economics and Psychology, 1987, U. Chicago, 67, at pp. 88, 91; Gigenzer, G., and Goldstein, D. G., "Reasoning the Fast and Frugal Way: Models of Bounded Rationality" (1996) 103 Psychological Review 650, at p. 650; Jolls, C, Sunstein, C. R. and Thaler, R. H., "A Behavioural Approach to Law and Economics" in Sunstein, C. R. (ed), Behavioural IMW and Economics, 2000, Cambridge, 13. at p. 14; Kahneman, D., "Maps of Bounded Rationality: Psychology for Behavioral Economics" (2003) 93 American Economic Review 1449, at pp. 1454-56.

4. Korobkin, R. B. and Ulen, T. S., supra fn 1, at p. 1074. Contra, Gigenzer, G., and Goldstein, D. G., supra fn 3, at p. 650: arguing the 'rationality of the Enlightenment' is replaced by the 'heuristics and biases program'; Thaler, R. H., "The Psychology of Choice and the Assumptions of Economics" in Thaler, R. H. (ed), Quasi Rational Economics 1991, Russell Sage Foundation, 137, at p. 163.

5. Schroeter, U. G., "Creation and Acceptance of Uniform Private Law in Europe: Lessons from the Application of the United Nations Convention on Contracts for the International Sale of Goods (CISG) for a European Contract Law" (2007) 14 Jahresheft der Internationalen Juristenvereinigung Osnabrück 35, at 1(b)(bb) (translation by Sylvester Urban): likening reactions to choice of law questions to reactions to a relative who 'lets everyone know in a loud voice his highly unpleasant experiences with divorces and the financial consequences thereof.

6. Information costs and costs of expertise, when averaged across the number and value of transactions involved, may be relatively much higher for small or medium businesses than for larger businesses. Linarelli, J., "The Economics of Uniform Laws and Uniform Lawmaking" (2002) 48 Wayne Law Review 1387, at p. 1402; Castellani, L., "Promoting the Adoption of the United Nations Convention on Contracts for the International Sale of Goods (CISC)" (2009) 13 Vindobona Journal 241, at p. 244.

7. Thus, the role of information and transaction costs is vital in ensuring efficient market outcomes. Coase, R. H., "The Problem of Social Cost" (1960) 3 Journal of Law and Economics 1; Farrell, J. and Saloner, G., "Standardization, Compatibility, and Innovation" (1985) 16 Rand Journal of Economics 70.

8. Tversky, A. and Kahneman, D., "Prospect Theory" supra fn 3; Tversky, A. and Kahneman, D., "Judgment Under Uncertainty" supra fn 3; Tversky, A. and Kahneman, D., "Rational Choice" supra fn 3; David, P. A., "Clio and the Economics of QWERTY" (1985) 75 American Economic Review 332, at p. 332: defining path dependence and arguing its relevance in standard keyboard design; Arthur, W. B., "Competing Technologies, Increasing Returns, and Lock-in by Historical Events" (1989) 99 The Economic Journal 116; Liebowitz, S. J. and Margolis, S. E., "Path Dependence, Lock-in and History" (1995) 11 Journal of Law, Economics and Organization 205, at pp. 205-06. For a contrary account of the QWERTY story, see Liebowitz, S. J. and Margolis, S. E., "The Fable of the Keys" (1990) 33 Journal of Law and Economics 3, at pp. 6-21: claiming a lack of evidence to support the conclusion that the alternative Dvorak keyboard design was superior. On path dependence in relation to uniform law, see Linarelli, J., supra fn 6, at pp. 1394-95, 1398, 1402. The term 'path dependence' in relation to CISG opt-outs is also mentioned in Gopalan, S., "A Demandeur-Centric Approach to Regime Design in Transnational Commercial Law" (2008) 39 Georgetown Journal of International Law 327, at p. 330.

9. However, weaker forms of path dependence, where inefficiency becomes knowable only ex post or change is not feasible after the initial choice is not irrational, are compatible with rational choice theory. Liebowitz, S. J. and Margolis, S. E., "Path Dependence" supra fn 8, at p. 206.

10. Schroeter, U. G., supra fn 5.

11. In behavioural economics, this is also referred to as 'status quo bias'. Korobkin, R. B., "Behavioural Economics, Contract Formation and Contract Law" in Sunstein, C. R., supra fn 3, at p. 137; Thaler, R. H., supra fn 4, at p. 143: stating 'losses loom larger than gains'; Sunstein, C. R., supra fn 3, at p. 4; Korobkin, R. B. and Ulen, T. S., supra fn 1, at pp. 1069, 1095, 1111-12; Kahneman, D., supra fn 3, at p. 1457; Sunstein, C. R., supra fn 2, at p. 1068: describing the combination of tendencies to overvalue Moss from the status quo' and to ignore or undervalue gains; Liebowitz, S. J. and Margolis, S. E., supra fn 8, at p. 205. See also Hiscock, M., "The Role of International Conventions in International Business Transactions" presented at the 2008 International Trade Law Symposium, 11-12 April 2008, Canberra, at p. 5: similarly regarding CISG.

12. Contra, Korobkin, R. and Guthrie, C, "Psychology, Economics and Settlement: A New Look at the Role of the Lawyer" (1997) 76 Texas Law Review 77, at p. 81: arguing that the presence of lawyers in litigation settlements improves efficiency.

13. Ferrari, P., "General Report" in Ferrari, F. (ed), National Impact, The CISG and its Impact on National Legal Systems, 2008, Sellier, 413, at p. 427 [hereinafter National Impact]; Spagnolo, I,., "The Last Outpost: Automatic CISG Opt Outs, Misapplications and the Costs of Ignoring the Vienna Sales Convention for Australian Lawyers" (2009) 10(1) Melbourne Journal of International Law 141, at pp 146, 163-65. See also Brand, R. A., "Uni-State Lawyers and Multinational Practice: Dealing with International, Transnational, and Foreign Law" (2001) 34 Vanderbilt Journal of Transnational Law 1135, at pp. 1162-66; Brand, R. A., "Professional Responsibility in a Transnational Transactions Practice" (1997) 17 Journal of Law and Commerce 310, at p. 341; Saiegh, S., "The Business Lawyer's Perspective" in Flechtner, H. M., Brand R. A. and Walter, M. S. (eds), Drafting Contracts under the CISG, 2008, Oxford, at p. 254; Dodge, W. S., "Teaching the CISG in Contracts" (2000) 50 Journal of Legal Education 72, at p. 74. Cf. Andersen, C. B. "United Kingdom" in Ferrari, F. (ed), National Impact, 303, at p. 305.

14. For a summary, see Spagnolo, L., supra fn 13, at pp. 149-59.

15. Butler, A. E., "The International Contract: Knowing When, Why, and How to 'Opt Out' of the United Nations Convention on Contracts for the International Sale of Goods" (2002) 76(5) Florida Bar Journal 24.

16. Ibid.

17. Arrow, K. J., "Information and Economic Behavior" in Arrow, K. J. (ed), Collected Papers, supra fn 1, at p. 150: explaining the economic role of ethics in professional situations of information asymmetry.

18. Gillette, C. P. and Scott, R. E., "The Political Economy of International Sales Law" (2005) 25 International Review of Law and Economics 446, at p. 478: noting that 'attorneys have incentives to avoid learning about novel law' and therefore will opt out of even an 'optimal' novel law; Gopalan, S., supra fn 8, at pp. 328-29: arguing public conflicts arguments cannot explain failures of private law agreements like the CISG. However, when party choices are considered in light of lawyer-client agency, conflicts of interest are arguably highly important in explaining opt out practices. Gopalan argues, referring to Gordon's Florida study, that the CISG is a 'failed' convention compared with the 'highly successful' UNIDROIT Principles. I would not disagree with respect to the UNIDROIT Principles' success, but the dichotomy is overstated. Unfortunately, the paper does not refer to other studies discussed in Spagnolo. L., Kaleidoscope Part I, supra fn +, at fns 2-6. In an article perhaps not available to Prof. Gopalan at the time of writing, it was reported that US practitioners were far less familiar with the UNIDROIT Principles than the CISG, and chose to opt fully or partially into the CISG slightly more often than the UNIDROIT Principles: Fitzgerald, P. L., "The International Contracting Practices Survey Project" (2008) 27 Journal of Law and Commerce, at pp. 33-34, 64, 67 (forthcoming), available at: <http://ssrn.com/abstract=127382>.

19. Kahan, M. and Klausner, M., "Path Dependence in Corporate Contracting: Increasing Returns, Herd Behavior and Cognitive Biases" (1996) 74 Washington University Law Quarterly 347, at p. 353: clients have imperfect information about the quality of the lawyer's work.

20. These involve monitoring costs. See generally Stiglitz, J. E., supra fn 1, at pp. 551, 567-72. See also Walt, S., "Novelty and the Risks of Uniform Sales Law" (1999) 39 Virginia Journal of International Law 671, at p. 687.

21. Stiglitz, J. E., supra fn 1, at pp. 551, 567-72; Winter, S. G., "Satisficing, Selection and the Innovating Remnant" (1971) 85 Quarterly Journal of Economics 237, at p. 237; Kahan, M. and Klausner, M., supra fn 19, at pp. 353-55; Engel, C, "The Behaviour of Corporate Actors -- A Survey of the Empirical Literature" (2008), available at: <http://ssrn.com/abstract=1135184>.

22. Kahan, M. and Klausner, M., supra fn 19, at pp. 353-55; Walt, S., supra fn 20, at pp. 685-88: considering but ultimately dismissing the possibility that opt-out rates might be influenced by lawyers seeking to maximise future litigation fees.

23. See Stigler, G. J., supra fn 1, at p. 121; Samuelson, P. A., Economics: An Introductory Analysis, 7th edn, 1967, McGraw-Hill, New York, at pp. 450-451, 743; Stiglitz, J. E., supra fn 1, at p. 179.

24. See discussion in Spagnolo, L., Kaleidoscope Part I, supra fn +, at 2.4.

25. If a choice of forum clause fails to attain the intended result, it may affect the efficacy of the choice of law. The risks are likely to be underestimated due to the human tendency to underestimate probabilities of failure where events involved are disjunctive; for example, where one failure alone will cause all intended outcomes to be lost. In cognitive psychology, this error is known as the cognitive bias of 'anchoring.' Tversky A. and Kahneman, D., "Judgment Under Uncertainty" supra fn 3, at p. 16. See also Jones, G. T., "Designing Heuristics: Hybrid Computational Models for Teaching the Negotiation of Complex Contracts" (2009), available at: <http:ssrn.com/abstract=1370525>: regarding loss of other optimal outcomes due to tendencies to confine contractual negotiations to a small range of possibilities.

26. Although this concern is irrelevant to the client's interests, it may play a part in decisions in high opt-out, low familiarity jurisdictions. See discussion infra, at 2.3.

27. Kahan and Klausner, supra fn 19, at pp. 355-58. Risk aversion and status quo biases are strong predictors of choices, due to the endowment effect, and may cause path dependency. Kahneman, D., supra fn 3, at p. 1459. Unfamiliarity levels can exacerbate the problem via accessibility biases in decision making. Ibid, at p. 1459: noting interrelatedness of cognitive biases and externalities in standardisation of suboptimal standard contract terms. See also supra fn 11.

28. Goldstein, D. G. and Gigenzer, G., "The Recognition Heuristic, How Ignorance Makes Us Smart" in Gigenzer, G., Todd, P. M. and the ABC Research Group (eds), Simple Heuristics that Make Us Smart, 1999, Oxford University Press, 37, at p. 40.

29. See Spagnolo, L., Kaleidoscope Part I, supra fn +, at fn 83; Simon, H. A., "A Behavioral Model of Rational Choice" supra fn 3, at p. 113; Elster, J., Ulysses and the Sirens: Studies in Rationality and Irrationality, 1979, Cambridge University Press, Cambridge, at p. 66.

30. Parents and those with long memories will recall the delicious but visually unappetising fictitious dish from the Dr. Seuss story Green Eggs And Ham, in which Sam urges, 'Try them! Try them! And you may. Try them and you may, I say.' Seuss, Green Eggs and Ham, 1960, Beginner Books, New York.

31. See Spagnolo, L., Kaleidoscope Part I, supra fn +, at 2.2.

32. Gigenzer, G. and Todd, P. M., "Fast and Frugal Heuristics, The Adaptive Toolbox" in Gigenzer, G., and Todd, P. M., supra fn 28, at p. 5; Korobkin, R. B. and Ulen, T. S., supra fn 1, at p. 1077.

33. Simon, H. A., "Rational Decision Making in Business Organizations" supra fn 3, at p. 502: stating that 'rationality is bounded when it falls short of omniscience'; Simon, H. A., "A Behavioral Model of Rational Choice" supra fn 3, at 101: describing how all organisms suffer from limited capacities to obtain and process information; Simon, H. A., "Rational Choice and the Structure of the Environment" supra fn 3, at p. 129: describing 'approximate rationality' as closer accounts of observable conduct than classical maximizing rationality; Korobkin, R. B. and Ulen, T. S., supra fn 1, at p. 1075.

34. Korobkin, R. B. and Ulen, T. S., supra fn 1, at pp. 1069 and 1076; Gigenzer, G. and Todd, P. M., supra fn 32, at p. 5: describing human dependence on an 'adaptive toolbox' of 'frugal heuristics' to enable complex decision making in conditions of bounded rationality.

35. Gigenzer, G. and Todd, P. M., supra fn 32, at pp. 13, 18: utilising the term 'ecological rationality' to emphasise this aspect of bounded rationality, i.e., the match between heuristics and the decision-making environment; Simon, H. A., "Rational Choice and the Structure" supra fn 3, at pp. 129-130.

36. Simon argues that, rather than searching for all alternatives and completely ordering preferences as demanded by rational choice theory, 'satisficing' behaviour only partially orders alternatives by setting an aspiration level of viable alternatives. This facilitates 'limited' rationality under constraints of knowledge and capacity, as opposed to global rationality, which optimizes outcomes under conditions of perfect information and limitless capacity. Simon, H. A., "A Behavioral Model of Rational Choice" supra fn 3, at pp. 104-113; Simon, H. A., "Rational Choice and the Structure" supra fn 3, at pp. 129, 136; Simon, H. A., "Rational Decision Making" supra fn 3, at p. 501; Korobkin, R. B. and Ulen, T. S., supra fn 1, at p. 1075; Gigenzer, G. and Todd, P. M., supra fn 32, at pp. 14-18.

37. Simon, H. A., "Rational Decision Making" supra fn 3, at p. 501.

38. Eisenberg, M. A., "The Limits of Cognition and the Limits of Contract" (1995) 47 Stanford Law Review 211, at p. 214.

39. There is one line of thought that 'optimizing' behaviour still occurs within constraints in relation to the search for information. Stigler, G. J., "The Economics of Information" (1961) 69 Journal of Political Economy 213. However, this would require greater computational skills than bounded rationality contemplates. See also, Simon, H. A., "Rational Decision Making" supra fn 3, at p. 503; Gigenzer, G. and Todd, P. M., supra fn 32, at pp. 10-11; Winter, S. G., supra fn 21, at pp. 239-41.

40. Jones, G. T., supra fn 25, at p. 3.

41. Simon, H. A., "A Behavioral Model of Rational Choice", supra fn 3, at p. 113; Simon, H. A., "Rational Choice and the Structure of the Environment" supra fn 3, at p. 137.

42. Simon, H. A., "Rational Choice and the Structure" supra fn 3, at pp. 129, 136; Simon, H. A., "Rational Decision Making" supra fn 3, at p. 501; Gigenzer, G. and Todd, P. M., supra fn 32, at pp. 14-18.

43. Simon, H. A., "Rational Decision Making in Business Organizations" supra fn 3, at p. 507: discussing the 'clear [...] fit' between satisficing theories of problems solving and psychological selectivity utilising heuristics. See also Jones, G. T., supra fn 25, at pp. 4 ff: using a computational model to evaluate various heuristics in complex contractual negotiations.

44. In the case of the CISG, in jurisdictions prone to 'automatic' opt-outs, the aspirational level at which lawyers are satisficing is set at the benchmark of 'generally opt out', and provided this is agreeable to the counterparty's lawyer, any other possibility need not be considered.

45. The existence of a suitable environment to the operation of a heuristic device in decision making under bounded rationality is highly important to the feasibility of the heuristic. See supra fn 35 and accompanying text.

46. See Korobkin, R. B. and Ulen, T. S., supra fn 1, at p. 1075.

47. Ibid, at p. 1076.

48. Gigenzer, G. and Todd, P. M., supra fn 32, at p. 32: referring to external factors that may help to select heuristics.

49. Spagnolo, L., Kaleidoscope Part I, supra fn +, at 2.2(B): discussing 'cases per million capita' and 'cases per trillion trade dollar'.

50. I am not aware of any instances in which such matters have not been settled.

51. Similar processes occur as cascade effects, but deliberation appears to result in more extremist views. Sunstein, C. R., "Deliberative Trouble? Why Groups Go to Extremes" (2000) 110 Yale Law Journal 71, at pp. 75, 85; Kahan, D. M., "Gentle Nudges vs. Hard Shoves: Solving the Sticky Norms Problem" (2000) 67 University of Chicago Law Review 607, at pp. 614-15; Isenberg, D. J., "Group Polarisation: A Critical Review and Meta-Analysis" (1986) 50 Journal of Personality and Social Psychology 1141, at p. 1141.

52. Sunstein, C. R., "Deliberating Groups Versus Prediction Markets (Or Hayek's Challenge to Habermas)" (2007), available at: <http://ssrn.com/abstract_id=956189>, at pp. 2, 4-5; Kahan, D. M., supra fn 51, at p. 614; Isenberg, D. J., supra fn 51, at p. 1141.

53. Sunstein, C. R., supra fn 52, at pp. 2, 4-5; Kahan, D. M., supra fn 51, at p. 614: noting that the group view tends to a more extreme rather a median view; Engel, C, supra fn 21, at p. 12.

54. Isenberg, D. J., supra fn 51, at pp. 1144-45; Sunstein, C. R., supra fn 51, at p. 78; Sunstein, C. R., supra fn 52, at pp. 2-5.

55. Isenberg, D. J., supra fn 51, at p. 1141; Engel, C, supra fn 21, at p. 13; Sunstein, C. R., supra fn 51, at pp. 75, 85.

56. Referred to as the 'common-knowledge effect:' Sunstein, C. R., supra fn 52, at pp. 10-13. Thus, group views are based on incomplete information. Gigone, D. and Hastie, R., "The Common Knowledge Effect: Information Sharing and Group Judgements" (1993) 65 Journal of Personality and Social Psychology 959, at p. 959; Stasser, G., Taylor, L. A. and Hanna, C, "Information Sampling in Structured and Unstructured Discussions of Three- and Six-Person Groups" (1989) 57 Journal of Personality and Social Psychology 67; Sunstein, C. R., supra fn 52, at p. 19.

57. Gigone, D. and Hastie, R., supra fn 56, at pp. 960, 966 ff. Consequently, groups where disagreements are aired make more accurate decisions. Ibid, at p. 967.

58. Sunstein, C. R., supra fn 51, at pp. 81-83; Sunstein, C. R., supra fn 2, at p. 1066.

59. Kahan, M. and Klausner, M., supra fn 19, at pp. 356-58: explaining that standard terms enable lawyers to 'share the blame' if the choice is proven ill-advised, whereas customization carries a far greater reputational risk for the individual who must shoulder any error alone.

60. Sunstein, C. R., supra fn 51, at p. 78; Sunstein, C. R., supra fn 53, at pp. 2, 4-5.

61. Sunstein, C. R., supra fn 53, at pp. 2, 4-5, 19.

62. Ibid, at pp. 2, 4-5.

63. Ibid, at pp. 4-5. On experiments demonstrating framing biases even amongst experts; see Thaler, R. H., supra fn 4, at pp. 158 (citing a study on physicians by McNeil et al (1982)), 137: relating a study by Allais and Ellsberg on economists and statisticians.

64. See Sunstein, C. R., supra fn 2, at p. 1067: observing that if an individual member knows the alarmist view held by the group is unwarranted, the member 'may not voice [...] doubts about whether the alarm is merited'.

65. Sunstein, C. R., supra fn 51, at pp. 91-92.

66. Kahan, M. and Klausner, M., supra fn 19, at p. 348.

67. Goetze, D., "Comparing Prisoner's Dilemma, Commons Dilemma, and Public Goods Provision Designs in Laboratory Experiments" (1994) 38 Journal of Conflict Resolution 56, at p. 56: defining 'collective action dilemmas' as situations where 'incentives confronting individuals in a group may not be consistent with achievement of collective ends that would benefit everyone'.

68. Bar-Grill, O., "The Behavioural Economics of Consumer Contracts" in Bar-Grill, O. and Epstein, R. A., Consumer Contracts: Behavioural Economics vs. Neoclassical Economics (2007), available at: <http://ssrn.com/abstract=982527>. See also Walt, S., supra fn 20, at pp. 692-97: discussing collective action problems from the perspective of the need to develop CISG caselaw, and the costs and inability to capture the benefits of such clarification in the case of the individual parties involved.

69. 'Externalities' are '[c]osts or benefits related to a good or service that fall on others besides buyers and sellers of that particular good or service'. Waud, R. N. and Hocking, A., Microeconomics, (Australian edn.), 1988, Harper and Row, Sydney, at G4. See also Coase: supra fn 7; Stigler, G. J., The Theory of Price, supra fn 1, at pp. 118-20.

70. Poundstone, W., Prisoner's Dilemma, 1992, Oxford University Press, Oxford, at p. 6; Cole, D. H. and Grossman, P. Z., "Institutions Matter! Why the Herder Problem is not a Prisoner's Dilemma" (2008), available at: <http://ssrn.com/abstract=1114541>. See also Poundstone, ibid, passim: explaining the origins of game theory; Arrow, K. J., supra fn 1, at pp. 69-70; Thaler, R. H, supra fn 4, at p. 137; Elster, J., supra fn 29, at pp. 18-21.

71. Kahan, M. and Klausner, M, supra fn 19, at pp. 350-53: discussing positive learning externalities from certainty generated by past use of boilerplate terms, and positive network externalities accruing from their future use.

72. See discussion supra, at 2.2.

73. Farrell, J. and Saloner, G., supra fn 7, at pp. 73, 82; Bar-Grill, O., supra fn 68, at p. 8, fn 19.

74. Players are assumed to be completely logical and self-interested in game theory. Poundstone, W., supra fn 70, at pp. 6, 37, 44.

75. The irrationality of the crowd can reduce even the most informed market participant's ability to make rational choices. Bar-Grill, O., supra fn 68, at p. 30. This again highlights the importance of taking into account the way the decision maker perceives the world. Simon, H. A., "A Comparison of Game Theory and Learning Theory" (1956) 21 Psychometrika 267, at p. 271.

76. For discussion of the reputational issues facing lawyers in the choice between standard and customized contract terms, see Kahan, M. and Klausner, M., supra fn 19, at pp. 356-58: explaining how the safety of numbers allows lawyers choosing suboptimal but standard terms to 'share the blame' when the suboptimality is revealed.

77. Spagnolo, L., supra fn 13, at pp. 162-64; Arrow, K. J., supra fn 17.

78. Kahan, M. and Klausner, M., supra fn 19, at pp. 356-58: explaining how the status quo bias and endowment effect may contribute to path dependency in standard contract terms. See also Engel, C, supra fn 21, at p. 12.

79. Kahan, M. and Klausner, M., supra fn 19, at pp. 356-58; Engel, C, supra fn 21, at p. 18. Note that taking such cognitive biases into account, the model in this paper is not strictly rational, but instead a hybrid (and hopefully a more accurate depiction).

80. Arguably, in reality, decisions are not made simultaneously, and thus game theory is incapable of capturing more dynamic scenarios. Granovetter, M., "Threshold Models of Collective Behavior" (1978) 83 American Journal of Sociology 1420, at pp. 1441-42: arguing that while game theory permits utility inferences, threshold models do not depend on homogeneity of preferences or simultaneous decision making.

81. The Prisoner's Dilemma involves two prisoners suspected of the same crime, against each of whom the police have enough evidence to secure a conviction for a minor crime. To elicit evidence for a major conviction, the penalty system has been tailored to reward a prisoner giving evidence against the other prisoner, but only if the other prisoner remains silent, and does not also provide such evidence. If both remain silent each will receive a 3 year sentence; if one gives evidence and the other does not, the silent prisoner receives a 10 year sentence and the prisoner giving evidence goes free; whereas if they both give evidence, each receives a 6 year sentence. Cole, D. H. and Grossman, P. Z., supra fn 70, at p. 3; Baird, D. G., Gertner, R. H. and Picker, R. C, Game Theory and the Law, 1994, Harvard University Press, Cambridge, at p. 33; Poundstone, W., supra fn 70, at pp. 116-18; Elster, J., supra fn 29, at p. 21.

82. Hardin, G., "The Tragedy of the Commons" (1968) 162 Science 1243, at p. 1244: attributing it to a pamphlet by Lloyd, W. F., Two Lectures on the Checks to Population (1833). In its game theory version, there are two herdsmen with access to an open meadow. If they graze equal numbers of cattle and both do not add extra cattle, total profits are 20 units, each farmer reaping 10 units. If one adds extra cattle but the other does not, total profits decrease to 10 units, but the self interested herdsman reaps 11 units profit while the other farmer suffers a 1 unit loss. If they both add extra cattle, the total profit for all and for each will be zero: See Goetze, D., supra fn 67, at p. 78; Cole, D. H. and Grossman, P. Z., supra fn 70, at pp. 5-6.

83. As opposed to zero-sum games where the loss of one player amounts to the gain of the other. Poundstone, W., supra fn 70, at p. 51 (defining zero-sum games).

84. The table depicts an asymmetric hybrid game akin to a mirror image 'bully' game: the lawyer faces preferences from the standard 'stag' game whereby he prefers cooperation (investment by both) but fears defection (non-investment by the firm); the firm faces preferences from the standard 'deadlock' game, whereby it prefers to defect (not invest) no matter what. See Poundstone, W., supra fn 70, at pp. 221 ff; Baird, D. G. et al, supra fn 81, at pp. 36-37.

85. For example, if he estimates the likelihood of firm investment at 30%, the average gain from investment is 0.9 units (3 x 0.3), whereas the gain from non-investment is on average 1.3 units ((2 x 0.3)+(1 x 0.7)). This reasoning employs a mixed strategy approach. See Baird, D. G. et al, supra fn 81, at p. 37; Korobkin, R. B. and Ulen, T. S., supra fn 1, at pp. 1062-64, 1084.

86. Referred to as a 'Nash equilibrium.' Baird, D. G. et al, supra fn 81, at p. 310; Cole, D. H. and Grossman, P. Z., supra fn 70, at p. 4; Poundstone, W., supra fn 70, at pp. 98-99: stating that such solutions exist in all finite two player games such that each player would have 'no regrets [ ... ] given how the other player(s) played'.

87. Cole, D. H. and Grossman, P. Z., supra fn 70, at pp. 2, 4.

88. Alcock, J. E. and Mansell, D., "Predisposition and Behavior in a Collective Dilemma" (1977) 21 Journal of Conflict Resolution 443, at p. 444; Cole, D. H. and Grossman, P. Z., supra fn 70, at p. 2.

89. This is the case in theory and in practice in situations of communal resources throughout history. Cole, D. H. and Grossman, P. Z., supra fn 70, at p. 9.

90. Richards, R., "Reciprocity and Shared Knowledge Structures in a Prisoner's Dilemma Game" (2001) 45 Journal of Conflict Resolution 621, at p. 621; Cole, D. H. and Grossman, P. Z., supra fn 70, at p. 8.

91. Kahan, M. and Klausner, M., supra fn 19, at p. 352, fn 15: remarking that law firms do not tend to perform coordination.

92. Farrell, J. and Saloner, G., supra fn 7, at pp. 70-72, 79-80; Liebowitz, S. J. and Margolis, S. E., "The Fable of the Keys" supra fn 8, at p. 3.

93. A network effect occurs when 'one consumer's value for a good increases when another consumer has a compatible good'; the classic example is the telephone. Farrell, J. and Saloner, G., supra fn 7, at p. 70; Katz, M. L. and Shapiro, C, "Network Externalities, Competition, and Compatibility" (1985) 75 American Economic Review 424, at p. 424; Arthur, W. B., "Positive Feedbacks in the Economy" (1990) 262 Scientific American; Katz, M. L. and Shapiro, C, "Systems Competition and Network Effects" (1994) 8(2) Journal of Economic Perspectives 93; Leibowitz, S. J. and Margolis, S. E., "Network Externality: An Uncommon Tragedy" (1994) 8 Journal of Economic Perspectives 133, at p. 135; Lemley, M. A. and McGowan, D., "Legal Implications of Network Economic Effects" (1998) 86 California Law Review 479, at p. 483. Arguing network effects of boilerplate terms, see Kahan, M. and Klausner, M., supra fn 19, at pp. 348-49: citing the choice of New York law in bond indentures. See also Walt, S., supra fn 20, at p. 688; Gopalan, S., supra fn 8, at p. 330: mentioning network effects in relation to CISG opt-outs.

94. Kahan, M. and Klausner, M., supra fn 19, at p. 349.

95. Sunstein, C. R., supra fn 51, at pp. 81, 82, 84: describing the point of critical mass, and 'epidemic' nature of change; Granovetter, M., supra fn 80, at pp. 1441-42: explaining 'thresholds' models of collective behaviour.

96. Sunstein, C. R., supra fn 51, at pp. 95-96.

97. Sunstein, C. R., supra fn 53, at pp. 13-20.

98. Sunstein, C. R., supra fn 51, at pp. 75 and 95; Isenberg, D. J., supra fn 51, at p. 1141.

99. Korobkin, R. B. and Ulen, T. S., supra fn 1, at p. 1071: arguing that occasional suboptimal decisions do not cause a business to become insolvent; Engel, C, supra fn 21, at p. 5: citing survival after extended poor corporate decision making.

100. Ibid.

101. The strategy is enhanced if, as is likely in the new environment, the lawyer perceives that the likelihood of the firm investing is greater than 50%.

102. That is, rational choice or satisficing heuristic.

103. Authors anecdotally report certain trends in Italy and Germany respectively. See Torsello, M., "Italy" in Ferrari, F. (ed), National Impact, supra fn 13, 187, at pp. 189, 190, 195-99: reporting that while opting out is still prevalent in Italy, many specialist drafters are now choosing not to opt out; Magnus, U., "Germany" in Ferrari, F. (ed), National Impact, supra fn 13, at p. 146: reporting that in Germany an increasing number of business associations no longer generally recommend opting out, and that opting out is no longer the norm for standard forms.

104. See, Spagnolo, L., Kaleidoscope Part I, supra fn +, at fn 66.

105. Flechtner, H. M., "Changing the Opt-Out Tradition in the United States" paper presented at 'Modern Law for Global Commerce' Congress to Celebrate the 40th Annual Session of UNC1TRAL, Vienna, 9-12 July 2007, available at: <http://www.uncitral.org/pdf/english/congress/Flechtner.pdf>: noting pressure in the US exerted by globalised legal services market and observing a 'change' in queries he received from practitioners regarding CISG from purely litigious to front end (drafting/choice of law) queries). Note the change in numbers of lawyers 'generally opting out' in the US from 71% in 2004 to 61% or 55% in 2006-7. See Koehler, Philippopoulos and Fitzgerald in Spagnolo, L., Kaleidoscope Part I, supra fn +, at fn 2.

106. Rozehnalova, N., "Czech Republic" in Ferrari, F. (ed), National Impact, supra fn 13, at p. 108: noting that those Czech lawyers familiar with the CISG come under pressure to exclude the CISG by foreign counterparties who 'require the application of their own national law'; Veytia, H., "Mexico" in Ferrari, F. (ed), National Impact, supra fn 13, at pp. 235, 237, 239, 248: stating that Mexican lawyers often deal with US counterparties who opt out of the CISG, or use standard form contracts drafted in the US which opt out of the CISG due to their firm's position as integrated within US-based law firms, with the result that their Mexican-based lawyers are heavily influenced or controlled from New York. Both Veytia, in relation to Mexico, and Mozina, in relation to Slovenia, state that lawyers in those jurisdictions are more likely to accept exclusions of the CISG proposed by counterparties for reasons related to economic bargaining strength as well as what could be termed a legal inferiority complex that, at least in the Mexican case, reflects a wider societal aspect of cultural identity. Ibid, at pp. 235, 237, 239, 248; Moz ina, D., "Slovenia" in Ferrari, F. (ed), National Impact, supra fn 13, at pp. 265, 272 fn 26.

107. Flechtner, H. M, supra fn 105; Lookofsky, J., "Online with Al K" in Andersen, C. B. and Schroeter, U. G. (eds), Sharing International Commercial Law Across National Boundaries: Festschrift for Albert H. Kritzer on the Occasion of his Eightieth Birthday 2008, Wildy, Simmonds and Hil), 287, at p. 291: arguing that few, if any, 'well-positioned' Chinese traders would be willing to contract on the basis of a foreign trader's domestic law 'without additional compensation'.

108. Ultimately I therefore favour the view that in the longer term, some persistent errors are eliminated by means of evolutionary competitive forces favouring behaviour closer to (while perhaps not perfectly) rational. Cf. see generally Tversky, A. and Kahneman, D., "Rational Choice and the Framing of Decisions" supra fn 3, at pp. 89-91; Korobkin, R. B. and Ulen, T. S., supra fn I, at 1071. Taking the middle ground, see Thaler, R. H., supra fn 4, at pp. 158-159. Interestingly, Tversky and Kahneman list situations of transparency as more likely to give rise to rational behaviour. Tversky, A. and Kahneman, D., "Rational Choice and the Framing of Decisions" supra fn 3, at p. 88. The above argument poses competitive forces as likely to lead to transparency and ultimately improved (though not perfect) rationality. See also Winter, S. G., supra fn 21, at p. 245; Liebowitz, S. J. and Margolis, S. E., "The Fable of the Keys" supra fn 8, at p. 4.

109. The attaining of a 'tipping point' or 'critical mass' resolves what might earlier be characterised as a coordination problem, by substituting a competitive imperative in place of communication between decision-making peers. In relation to co-ordination problems, see supra fn 95; Linarelli, J., supra fn 6, at p. 1413: listing reasons for the persistence of path dependence as 'information costs, costs of cooperation or coordination'. See also Baird, D. G. et al, supra fn 81, at p. 191.

110. Simon, H. A., "A Behavioral Model of Rational Choice" supra fn 3, at pp. 104-13.

111. It is well known that, while heuristics are useful, they often lead to 'severe and systematic errors'. Tversky A. and Kahneman, D., "Judgment Under Uncertainty" supra fn 3, at p. 1; Kahneman, D., supra fn 3, at p. 1449; Sunstein, C. R., supra fn 53, at p. 9 ('severe blunders').

112. Repetition of encounters with the CISG can be expected to correct the 'availability bias' whereby lawyers may be unable to recall instances in which CISG familiarity was important from personal experience. See Tversky A. and Kahneman, D., "Judgment Under Uncertainty" supra fn 3; Sunstein, C. R., supra fn 2, at pp. 1065-66; Thaler, R. H., supra fn 4, at p. 157: arguing that this is frequently due to lack of repetition and feedback.

113. De Ly, F., "Sources of International Sales Law: An Eclectic Model" (2005) 25 Journal of Law and Commerce 1, at p. 3.

114. See discussion in Spagnolo, L., Kaleidoscope Part I, supra fn +, at 2.2(B).

©Pace Law School Institute of International Commercial Law - Last updated January 14, 2010
Go to Database Directory || Go to Bibliography