UUnniivveerrssiittyy ooff MMiiaammii LLaaww RReevviieeww Volume 57 Number 1 Article 2 10-1-2002 IIss IItt ""tthhee WWiillll ooff tthhee PPeeooppllee"" oorr aa BBrrookkeenn AArrrrooww?? CCoolllleeccttiivvee PPrreeffeerreenncceess,, OOuutt--ooff--tthhee--mmoonneeyy OOppttiioonnss,, Bush V. Gore,, AAnndd AArrgguummeennttss ffoorr QQuuaasshhiinngg PPoosstt--bbaalllloottiinngg LLiittiiggaattiioonn AAbbsseenntt SSppeecciifificc AAlllleeggaattiioonnss ooff FFrraauudd Mark Klock Follow this and additional works at: https://repository.law.miami.edu/umlr RReeccoommmmeennddeedd CCiittaattiioonn Mark Klock, Is It "the Will of the People" or a Broken Arrow? Collective Preferences, Out-of-the-money Options, Bush V. Gore, And Arguments for Quashing Post-balloting Litigation Absent Specific Allegations of Fraud, 57 U. Miami L. Rev. 1 (2002) Available at: https://repository.law.miami.edu/umlr/vol57/iss1/2 This Article is brought to you for free and open access by the Journals at University of Miami School of Law Institutional Repository. It has been accepted for inclusion in University of Miami Law Review by an authorized editor of University of Miami School of Law Institutional Repository. For more information, please contact [email protected]. University of Miami Law Review VOLUME 57 OCTOBER 2002 NUMBER 1 ARTICLES Is it "The Will of the People" or a Broken Arrow? Collective Preferences, Out-of-the-Money Options, Bush v. Gore, and Arguments for Quashing Post-Balloting Litigation Absent Specific Allegations of Fraud MARK KLOCK* The law school faculty was debating three proposed revisions to the curriculum. Twenty of the thirty faculty present ranked proposal A pref- erable to proposal B. Twenty ranked proposal B preferable to proposal C. Twenty ranked proposal C preferable to proposal A. After countless hours of debate without agreement the faculty finally reached a decision. It decided thatany decisions made would be subject to change. Having accomplished something positive, it moved on to debate the schedule for discussing the changes to decisions yet to be made. INTRODUCTION It has now been over a year since the litigation involving the 2000 presidential election ended. An avalanche of emotional commentary has subsided and given way to careful academic analysis. The University of Chicago Law Review published a symposium on the decision with some contributors claiming that heated emotions have cooled down.' But * B.A., The Pennsylvania State University, 1978; Ph.D. in Economics, Boston College, 1983; J.D. (with honors), University of Maryland, 1988; Member D.C. and Maryland Bars; Professor of Finance, The George Washington University. The author is a nonpartisan nonvoter. 1.R ichard A. Epstein, "hi Such Manner as the Legislature Thereof May Direct": The Outcome of Bush v. Gore Defended, 68 U. CHI. L. REv. 613, 613 (2001) ("After a short flurry of UNIVERSITY OF MIAMI LAW REVIEW [Vol. 57:1 have they really? Alan Dershowitz's book, Supreme Injustice, engaged in admittedly ad hominem personal attacks on the five Republican Jus- tices.2 Dershowitz not only opines that these five Justices illegally decided the result to achieve their personal ends and then concocted a legal justification-he claims to have compiled evidence that includes motive to prove his case.3 The core of his evidence is his claim that these five Justices would have decided the case differently if the parties' positions had been reversed.4 That is obviously not a testable hypothesis.' Professor Dershowitz has a reputation for provocative and flamboy- ant writing, and his book clearly displays strong rhetorical skills.6 Argu- ing that he has compiled probative evidence to support his faith, however, is as ludicrous as arguing that I have proof that the North Pole lies at the bottom of the planet. The logical reasoning is a far larger embarrassment to the legal profession than anything written by any of the Justices in any of the proceedings involving the 2000 presidential election .7 heated debate over the soundness of the decision, the nation buckled down to business as usual. The attention of the media was, to say the least, short-lived."). 2. ALAN M. DERSHOWITZ, SUPREME INJUSTICE: HOW THE HIGH COURT HIJACKED ELECTION 2000, at 110 (2001) ("The criticism I am making of' the majority justices includes a significant ad hominem component. . . . I am also accusing them of dishonesty .. . .These criticisms are directed at the justices personally ...." ). 3. Professor Dershowitz repeatedly refers to the decision and motives with labels such as lawless and improper and characterizes his arguments as based on proof and evidence. See, e.g., id. at 95 ("1 will demonstrate that by any reasonable standard of evaluation, the majority justices [decided for Bush] because of malice aforethought."). 4. Id. at 12. 5. See generally Mark Klock, Finding Randon Coincidences While Searching for the Holy Writ of Truth: Specification Searches in law and Public Policy or Cum Hoc Ergo Propter Hoc?, 2001 Wis. L. REV. 1007, 1018-22 (providing an explanation of hypothesis testing); Lee Epstein & Gary King, The Rules of Inference, 69 U. CHi. L. REV. I (2002) (criticizing inferences in legal scholarship). 6. See Murray B. Light, Dershowitz Shows Conteniptf or Court, BUFF. NEws, Aug. 26, 2001, at F4. 7. One of Professor Dershowitz's arguments is that it is morally wrong for academics to defend the decision unless they honestly believe that the majority Justices would have voted the other way if the candidates' positions were reversed. DERSHOWITz, supra note 2, at 108. It is not obvious to me why his moral authority should supersede my religious beliefs. His position seems a bit one-sided since it implies that someone who honestly believes that they have no way of knowing what the Justices would have done in a different case, is morally wrong to defend the decision but free to criticize without sin. I strongly suspect (but cannot prove) that Professor Dershowitz fails his own test that he applies to the five Justices. I have in mind the case of N.Y. Times Co. v. Sullivan, 376 U.S. 254 (1964). This was the famous case in which the Court created new constitutional law regarding the defamation of public officials to overturn the Supreme Court of Alabama's affirmation of a defamation judgement for southern segregationists against the Times. Id. at 256. Presumably, Professor Dershowitz supports the case since it affords him legal protection for his otherwise defamatory book. My rhetorical question for Mr. Dershowitz then is, how can one intellectually reconcile the blatantly contrived result in N.Y. Tines v. Sullivan with 2002] ARGUMENTS FOR QUASHING POST-BALLOTING LITIGATION 3 A detailed analysis of Dershowitz's book is far afield of the analy- sis of this article. However, such an analysis provides some motivation because it supports two of the three facts justifying this article. First, the historical significance of the case means that analysis and commentary will continue for years.8 Second, while much has been written already, no one has previously applied the insights provided by the literature on options to the policy arguments. Litigation has significant embedded option components that were particularly acute in the context of the 2000 presidential election. An understanding of this is important for a thorough analysis. Third, Dershowitz's claim that the five Justices sub- stituted their political judgment for that of the people9 exemplifies a fairly widespread ignorance of Professor Arrow's relatively well-known Noble Prize-winning work proving that the conception of collective judgment is a construct with internally flawed logic."' What can option theory teach us about elections? Al Gore's posi- tion from November 8th to December 12th with respect to the outcome of the 2000 presidential election is perfectly analogous to an individual with an out-of-the-money option when the person with the option exer- cises some control over the riskiness of the underlying asset.I Posses- sion of the ability to create and increase risk creates a moral hazard problem for the individual with the out-of-the-money option. 2 It is his criticism of the Court in Bush v. Gore given that "[tiwo partisan wrongs do not make a judicial right"? DERSHOWITZ, supra note 2, at 8. 8. DERSHOWITZ, supra note 2, at 81 ("The majority per curiam opinion is likely to become one of the most analyzed, criticized, and defended opinions in the history of the Supreme Court."). 9. Id. at 3. 10. See, e.g., Cheryl D. Block, Truth and Probability-Ironies in the Evolution of Social Choice Theory, 76 WASH. U. L.Q. 975, 975-81 (1998) (observing the pervasiveness of Professor Arrow's proof that collective decisions cannot be made in such a manner that they will obey basic principles of rationality). 1I. November 8, 2000, was the day after the election, at which time the first ballot count was completed and showed Mr. Gore to have fewer votes than Mr. Bush. See Bush v. Gore, 531 U.S. 98, 100 (2000) (per curiam). Every subsequent ballot count led to the same result. See 5 Weeks of History, USA TODAY, Dec. 14, 2000, at 3A, available at 2000 WL 5798249. On December 13, 2000, with his option to challenge effectively extinguished by the Supreme Court, Mr. Gore finally conceded defeat. See John F. Harris & Ceci Connolly, Gore Offers Olive Branch and Finality; Concession's Grace Notes May Decide His Future, WASH. POST, Dec. 14, 2000, at Al, available at 2000 WL 29921521. 12. See, e.g., PHILIPPE JORION, VALUE AT RISK 43 (1997). Describing the relation between moral hazard and risk when one exerts influence over the risk, Professor Jorion wrote: This government guarantee is no panacea, for it creates a host of other problems, generally described under the rubric of noral hazard. Given government guarantees, there is even less incentive for depositors to monitor their banks, but rather to flock to institutions offering high deposit rates. Bank owners are now offered what is the equivalent of a "put" option. If they take risks and prosper, they partake in the benefits. If they lose, the government steps in and pays back the depositors. As long as the cost of deposit insurance is not related to the riskiness of activities, there will be perverse incentives to take on additional risk. UNIVERSITY OF MIAMI LAW REVIEW [Vol. 57:1 well-known that moral hazard creates economic inefficiency and that rules that mitigate the moral hazard problem can improve the allocation of social resources.13 This article seeks to explore the analogy between the recent presidential election and financial options, and argues that as a matter of public policy, vague and ambiguous statutes and case law should be interpreted so as to destroy any option to challenge objective election results on the basis of technology and ballot design ex post elec- tion results. The argument is further strengthened by the assertion that the pur- pose of elections is not to determine the will of the people. Indeed, the will of the people is an internally inconsistent, illogical construct that serves no useful purpose outside of the mathematical proofs that it can- not exist.'4 Thus, it is argued that the purpose of elections is to provide a socially acceptable method of allocating political power when there is no consensus in order to end pointless and unresolvable debate so that society can move on to other business.'5 "[W]hat the people want can- not be social policy simply because we do not and cannot know what the people want."'6 Finally, it is advanced that these policy arguments are consistent with the pertinent legal authority. The Supreme Court of Florida's deci- sions in this dispute were not based on controlling legal authority, but were based on policy arguments which are not logical, and legal author- ity that is subordinate to the U.S. Constitution.'7 The U.S. Supreme id. 13. See Kenneth J. Arrow, Tite Econmics of Moral Hazard: Further Cotment, 58 AM. ECON. REV. 537, 538 (1968) [hereinafter Arrow, Moral Hazard] ("The underlying point is that ... the resulting resource allocation will certainly not be socially optimal."). 14. There is a vast literature spanning many disciplines (including ethics, philosophy, economics, sociology, psychology, and political science) on the fundamental incompatibility between voter sovereignty and rational collective choice. For an accessible description of this incompatibility, see ALFRED F. MACKAY, ARROW'S THEOREM: THE PARADOX OF SOCIAL CHOICE: A CASE STUDY IN THE PHILOSOPHY OF ECONOMICS 1-12 (1980). This literature is highly cited within the legal literature. See generally, e.g., Block, supra note 10, at 975-81. 15. It is well-known that rational individual preferences cannot be aggregated to achieve a rational social preference ordering without imposing socially unacceptable constraints such as dictatorship. See, e.g., lAIN MCLEAN, PUBLIC CHOICE: AN INTRODUCTION 25 (1987) ("[Tlhere are deep problems with all procedures of getting from many preferences to one decision."). A commonly given example of this is the fact that there is nothing inconsistent with a majority preferring A to B, another majority preferring B to C, and another preferring C to A. See, e.g., id. at 25-27. This implies that there is no such thing as a "best policy" for the government because there is no platform that another platform cannot beat. See d. at 103. Mechanisms like plurality voting, or institutional devices such as the electoral college, can serve to break this cycle and reach a decision, albeit an unstable one, because efficiency demands that a decision be made rather than endure an infinite filibuster. 16. WILLIAM H. RIKER, LIBERALISM AGAINST POPULISM: A CONFRONTATION BETWEEN THE THEORY oF DEMOCRACY AND THE THEORY OF SOCIAL CHOICE 238 (1988). 17. The Florida Supreme Court relied heavily on the "will of the people" in making its 2002] ARGUMENTS FOR QUASHING POST-BALLOTING LITIGATION 5 Court on the other hand, based its ruling on a strict application of the law,'8 but the additional public policy arguments presented here could further support and advance the Court's ruling. I. OPTIONS, MORAL HAZARD, AND THE PARADOX OF VOTING A. The Mechanics of a Call Option A call option conveys the right to purchase a security at a fixed price (which is called the strike price or exercise price) for a finite time period.'9 When the underlying security is selling in the market for more than the strike price, the option has an intrinsic value equal to the differ- ence between the price of the security and the strike price.20 This is due to the fact that the option holder could exercise the option to purchase the security at the strike price and immediately resell it at the current market price, thereby capturing the difference. Such an option is said to be "in-the-money."'"' If the underlying security is selling in the market for less than the strike price, then the option has an intrinsic value equal to zero, since it would obviously not be rational to exercise the option given that the security could be purchased for less than the exercise price. This scenario is called "out-of-the-money."22 Options with time remaining before expiration are typically worth much more than their intrinsic value.23 This is because an option puts the holder in a situation in which losses are truncated, but gains are not. It makes no difference whether the option expires one dollar out of the money or one hundred dollars out of the money. The result is the same. But it does make a difference whether the option expires one dollar in the money or one hundred dollars in the money. Consider an option that is at the money (the security market price equals the option's strike price).24 If the value of the underlying security rises, there is a corre- decision. See Gore v. Harris, 772 So. 2d 1243, 1254 (Fla. 2000) (per curiam) ("[T]his Court, consistent with legislative policy, has pointed to the 'will of the voters' as the primary guiding principle to be utilized by trial courts in resolving election contests."). The "will of the voters" principle is derived from the Florida Constitution's vague and innocuous pronouncement that "[a]ll political power is inherent in the people." Palm Beach County Canvassing Bd. v. Harris, 772 So. 2d 1220, 1230 (Fla. 2000) (per curiam). But the laws of Florida are subordinate to the U.S. Constitution under the Supremacy Clause. See U.S. CONST. art. VI, cl.2 . 18. See Bush v. Gore, 531 U.S. 98, 111 (2000) (per curiam) ("Seven Justices of the Court agree that there are constitutional problems with the recount ordered by the Florida Supreme Court that demand a remedy."). 19. JOHN C. HULL, OPTIONs, FUTURES, & OTHER DERIVATIVES 6 (4th ed. 2000). 20. Id. at 154. 21. Id. 22. Id. 23. Zvi BODIE ET AL., ESSENTIALS OF INVESTMENTS 541-42 (4th ed. 2001) (discussing the time value of an option). 24. HULL, supra note 19, at 154. UNIVERSITY OF MIAMI LAW REVIEW [Vol. 57:1 sponding one-for-one increase in the option's intrinsic value.25 How- ever, if the value of the underlying security declines, there is no drop in the option's intrinsic value.26 As a first and fairly accurate approxima- tion, the probability of the security decreasing by certain amounts is off- set by the potential gains in the security.27 Otherwise, the expected price of the security would not equal the current price and some reason would need to exist for securities to trade at prices other than their expected value. The truncation of losses without truncation of gains is what gives additional value to options.28 An analogy to holding a call option at-the- money is the right to enter a casino or poker game in which winnings were permitted to be kept while losses were refunded in full. This right would be valuable and individuals would be willing to pay admission to enter the casino or game. Modifying the example to an in or out-of-the- money option merely modifies the analogy to a situation in which losses are only refunded beyond some deductible or gains are only kept beyond some threshold. Either way, the option with time remaining before expi- ration still has significant value due to the truncation of losses. This situation is somewhat analogous to an election because it makes no difference whether one loses by one vote or ten million votes. A loss is a loss. For an individual trailing after an initial election count, increasing the uncertainty regarding the rules of the election and the counting of votes can only help the candidate. Votes gained can poten- tially be sufficient to bring victory while votes lost make no difference in the status quo.29 A fundamental insight, which the option literature brings to this scenario, is that when losses are truncated, risk creates value.3° The greater the risk, the more valuable the option.3 This is because greater risk implies higher probabilities for large changes in the status quo, but 25. See id. (showing that since a rise in the stock price will put an at-the-money option in the money and the intrinsic value of an in-the-money option equals the stock price minus the fixed strike price, a one dollar increase in stock price will create a one dollar increase in intrinsic value for the at-the-money or in-the-money option). 26. See id. (giving equation for intrinsic value that shows that the intrinsic value is always zero for an out-of-the money option). 27. See id. at 219-26 (describing the mathematical process for stock price changes used to value options for which positive and negative changes of given size are equally likely). 28. See BODIE ET AL., supra note 23, at 542. 29. There is a difference in that the election involves a discontinuity-the gains only help if the trailing candidate overcomes a threshold, and beyond that threshold further gains do not help. But the fundamental asymmetry between the lack of harm from further losses in votes and the potential benefit from gains in votes is analogous to an option. 30. See BODIE ET AL., supra note 23, at 543 (stating that increased volatility increases option value due to limited losses). 31. Id. at 542. 2002] ARGUMENTS FOR QUASHING POST-BALLOTING LITIGATION 7 losses from the status quo are truncated.32 Therefore, the potential gains are not offset by potential losses and the increased risk creates value.33 This is not typically a problem with plain vanilla listed options on public equities because the individuals buying the options have no control over the riskiness of the underlying equity. Their risk is exogenous. This is not the case in all option-like scenarios, however. It is well- known that where the holder of an option exercises influence over the riskiness of the underlying asset, a moral hazard problem exists. 3n The moral hazard problem leads to socially undesirable conduct in the sense that resources are not allocated to their best use.35 The view expressed in this article is that the tactics utilized to contest the initial election results were a predictable consequence of a moral hazard problem and that this further led to a tremendous misallocation of legal resources,36 increased political risk,37 a negative impact on financial markets and the economy,38 and some loss in the credibility of the political system.39 It is further argued that as a matter of policy, arguments that an election resulted in an incorrect winner should not be considered absent a show- ing of manipulation of election results-without election tampering, the loser should not receive a free option with an inescapable moral hazard. Critics of this view can be expected to argue that the choice of a national president is far too important a decision to worry about slowing down the docket in some courts and losing some resources here and there.4" Installing either the best president or the correct one will lead to a better designation (or better alignment) of national priorities and long- run resources which will more than offset any short-term losses. This is an ends-justify-the-means argument, whereas I argue that the efficiency 32. See id. at 543 ("[E]xtremely good stock outcomes can improve the option payoff without limit, but extremely poor outcomes cannot worsen the payoff below zero."). 33. See id. ("This asymmetry means volatility in the underlying stock price increases the expected payoff to the option, thereby enhancing its value."). 34. See JORION, supra note 12, at 286-87 (giving an example of suboptimal behavior resulting from a moral hazard in an option where the option holder controls risk). 35. See STEPHEN A. Ross ET AL., CORPORATE FINANCE 597-98 (4th ed. 1996) (explaining that managers, working for stockholders who have an option to default, have incentives to invest in projects in which they should not invest, and not invest in projects in which they should invest). 36. See A Question of Trust: As Confidence in Balloting Weakens, Bush, Gore Make Matters Worse, USA TODAY, Nov. 13, 2000, at 29A (describing Gore's "utterly self-interested bid to have the vote count turn out his way at any legal cost"). 37. See id. (describing politicians' concerns over "serious danger in a protracted fight"). 38. See Peter G. Gosselin, Political Mess Raises Fears of Recession Economy, L.A. TIMES, Nov. 15, 2000, at Cl, available at 2000 WL 25918071 (reporting on economists' concern about negative impact of recounts on the economy and financial markets). 39. See A Question of Trust, supra note 36 (describing candidates' loss of credibility and crisis scenarios). 40. See Laurence H. Tribe, Let the Courts Decide, N.Y. TIMES, Nov. 12, 2000, § 4, at 15 (restating the argument that it is more important to count right than to finish counting). UNIVERSITY OF MIAMI LAW REVIEW [Vol. 57:1l of the means, or process, is the more important element at stake. This view is supported by Professor Arrow's Impossibility Theorem,4' and public choice theory.42 Advocates of the alternative view-that the Florida courts appropriately devoted resources and increased uncertainty regarding the process-must overcome at least two other problems: there is no evidence that society preferred one candidate over the other,43 and even if there was, social preferences revealed from elections are known to be irrational.44 It is not clear that any court could have done anything that would have resulted in a better president. The fact is, when one considers the normal error rates in ballot counting, the election was a statistical tie, both in the national popular vote and in Florida.45 Thus, using the plu- rality rules model of selecting one candidate as best (a model which lacks theoretical support but is assumed correct for this argument), there is no demonstrable evidence that one candidate is better than the other. That is, all systems of counting are subject to error since we cannot observe the intent of the voter at the precise moment the choice is made-at best we can only observe a ballot which is a proxy for voter intent. This election was so close that, statistically, we cannot reject the hypothesis that the difference in votes for the top two candidates was 41. Professor Arrow's work suggests that the means cannot be justified by the goal of reaching society's preferred decision because such a concept cannot exist. Professor Arrow gives the following interpretation of his theorem: If we exclude the possibility of interpersonal comparisons of utility [weighing some votes more than others], then the only methods of passing from individual tastes to social preferences which will be satisfactory and which will be defined for a wide range of sets of individual orderings are either imposed or dictatorial. KENNETH J. ARROW, SOCIAL CHOICE AND INDIVIDUAL VALUES 59 (2d ed. 1963) [hereinafter ARROW, SOCIAL CHOICE]. If rational collective social preferences do not exist, then we should at least use a decisive process. Negotiating over the rules after the contest is certainly not decisive. 42. Cf DENNIS C. MUELLER, PUBLIC CHOICE II, at 53 (1989) (explaining that the cost of time lost through decision making is an important consideration in public choice). 43. Aside from the fact that the margin of victory was lower than the potential error rate ("potential" describes the error rate because we do not know whether voters whose ballot did not register a vote chose to choose no candidate), consider that only approximately 100 million votes were cast in a nation with a population exceeding 281 million. 44. Given the votes for minor candidates, it is fair to say that a majority of voters voted against both Mr. Bush and Mr. Gore. Thus, we would have a president who a majority of voters voted against regardless of which candidate won the legal battle. 45. The initial count gave Mr. Bush a lead in Florida of 1,784 out of 5,816,482 votes that were cast for the two top candidates. See Bush v. Palm Beach County Canvassing Bd., 531 U.S. 70, 73 (2000) (per curiam). This margin is 0.03%. When considering votes for minor candidates, the margin is smaller. Nationally, Mr. Gore lead Mr. Bush by 232,895 out of 97,451,927 votes that were cast for the two of them. See Dan Balz, Bush's Florida Lead Shrinks to 300, WASH. POST, Nov. 15, 2000, at Al, available at 2000 WL 25428401. This margin is 0.2%. Again, considering votes for minor candidates, the margin is smaller. Nationally, two percent of all ballots do not register a vote for president. See Bush v. Gore, 531 U.S. 98, 103 (2000) (per curiam). 2002] ARGUMENTS FOR QUASHING POST-BALLOTING LITIGATION 9 zero. Additionally, advocates of a prolonged process must overcome the well-known fact that our voting system does not necessarily select either the most desirable or the least objectionable candidate.46 From a public choice perspective, our society has decided that it is less inefficient to have a system for selecting elected officials that is clear, final, and socially acceptable, than to have a system which expends more resources debating who should select the best officials and by which method. This is an argument against giving a losing candidate the option to litigate over the rules and standards for interpreting ballots after the election. B. Moral Hazard Moral hazard can be described as a situation in which an individual has been insulated from the economic consequences of his actions.47 The classic example is the case of an insured individual.4" A person with full coverage for all automobile losses lacks the economic incentive to drive as carefully as he otherwise would were there no insurance.49 Likewise, an individual with full medical coverage lacks the economic incentive to control costs, find the best prices, avoid frivolous treat- ments, and exercise a degree of prudence which would be exercised in the absence of insurance." In the insurance context, the moral hazard problem exists wherever the event against which insurance is taken out is at least partially within the control of the individual.5 The wide- spread use of copayments and deductibles is an effort to mitigate the moral hazard problem.52 Professor Arrow, a Nobel laureate, is among those who have noted the economic importance of the moral hazard problem: There is one particular case of the effect of differential information on the workings of the market economy (or indeed any complex economy) which is so important as to deserve special comment: one agent can observe the joint effects of the unknown state of the world 46. See McLEAN, supra note 15, at 156 (explaining why plurality voting fails all desirable criteria for voting systems abysmally). 47. See KENNETH J. ARROW, The Organization of Economic Activity: Issues Pertinent to the Choice of Market Versus Nonntarket Allocation, reprinted in 2 COLLECTED PAPERS OF KENNETH J. ARROW 143 (1983) [hereinafter ARROW, Economic Activity]. 48. Id. 49. Id. 50. Id. 51. KENNETH J. ARROW, General Economnic Equilibrium: Purpose, Analytic Techniques, Collective Choice, reprinted in 2 COLLECTED PAPERS OF KENNETH J. ARROW 222 (1983) [hereinafter ARROW, Econonic Equilibrium]. 52. Cf Arrow, Moral Hazard, supra note 13, at 538 (describing healthcare rationing as a reasonable method for improving the allocation of resources in the presence of moral hazard).