Abstract
When judges write opinions in common law cases they frequently seek to justify their decisions not merely by citing any relevant precedent or other authority but also by setting forth what we will call ‘substantive reasons’. Reasons of this kind incorporate moral, political, institutional, or other social considerations.1 In our view, substantive reasons have primacy over appeals to authority in the common law.2 Judges must rely on such reasons in cases of first impression, cases posing conflicts of precedent, and in cases involving the overruling or other renovation of precedent. Moreover, judges should and do advert to the substantive reasons behind precedents in order to interpret them and determine their scope.3
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
Summers, Two Types of Substantive Reasons: The Core of a Theory of Common Law Justification, 63 Cornell L. Rev. 707 (1978).
Id, 730–34.
Id, 730–32.
We do not canvass all the reasons why reasons are important in this article. We do assume that objectively sound reasons can be given. We also embrace intrinsic values. We are not unmindful however, that there are those who dissent (a decreasing number these days). See, e.g., Leff, Economic Analysis of Law: Some Realism About Nominalism, 60 VA. L. REV 451, 454(1974).
See n. 1 supra, 735–52.
Romano v Birmingham Ry Light and Power Co 182 Ala. 335,340–41,62 So 677,678–9 (1933).
Coleman v MacLennan, 78 Kan. 711, 741, 98 P 281, 292 (1908).
Campbell v Gruttemeyer, 222 Tenn. 133, 137–40, 432 SW2d 894, 896–97 (1968).
See n. 1 supra, 152–1’4. On the relations between the two types of reasons, see n 1 supra, 774–82.
Frostifresh Corp v Reynoso, 52 Misc. 2d 26,27–28,274 NYS 2d 757, 758–59 (Dist. Ct. 1966), rev’d on other grounds, 54 Misc. 2d 119, 281 NYS 2d 964 (App. Term, 2d Dep’t 1967).
Chase v Corcoran, 106 Mass. 286, 288 (1871).
Mercanti v Persson, 160 Conn. 468, 478, 280 A 2d 137, 142 (1971).
The literature is growing rapidly. See, for example, G. Calabresi, THE COSTS OF ACCIDENTS (1970); H. Manne ed, The Economics of Legal Relationships (1975); Posner, Utilitarianism, Economics, and Legal Theory, 8 J. Legal Stud. 103 (1978).
There is dispute over the extent to which judges implicitly use economists’ reasons. We cannot go into this here.
In this article we do not question the pervasive relevance of economic analysis to questions of means. This kind of relevance does not, however, give rise to autonomous economists’ reasons. Further, we would caution against any form of thinking that divorces goals from means here. See generally, Fuller, Memorandum, in On the Teaching of Law in the Liberal Arts Curriculum 37–43 (H. Berman ed 1956).
It simply will not do to say that economic analysis is normatively neutral or value neutral. Even the supposedly least controversial economists’ criterion includes such words as ‘better off and ‘worse off (though from the point of view of affected individuals).
The literature on economists’ reasons has become quite rich of late (and there is still more in the pipeline currently). See, for example, Baker, ‘The Ideology of the Economic Analysis of Law’, 5 Phil. & Pub. Aff. 3 (1975); Calabresi and Melamed, ‘Property Rules, Liability Rules, and Inalienability: One View of the Cathedral’, 85 Harv. L. Rev. 1089 (1972); R. Dworkin, Taking Rights Seriously, 96–100 (1978); C. Fried, Right and Wrong, Ch. 4 (1978); Michelman, ‘Norms and Normativity in the Economic Analysis of Law’, 62 Minn.> L. Rev. 1015 (1978); Michelman, ‘A Comment on Some Uses and Abuses of Economics in Law’, 46 U. Chi. L. Rev. 307 (1979); Posner, ‘Utilitarianism, Economics, and Legal Theory’, 8 J. Legal Stud. 103 ( 1978); Posner, ‘Some Uses and Abuses of Economics in Law’, 46 U. Chi. L. Rev. 281 (1979); Schwartz, ‘Economics, Wealth Distribution, and Justice’, 1979 Wis. L. REV. 799 (1979); Steiner, ‘Economics, Morality and the Law of Torts’, 26 U. Tor. L. Rev. 227 (1976); Symposium, ‘Change in the Common Law: Legal and Economic Perspectives’, 9J. Legal Stud. 189–366 (1980). See also forthcoming symposia in the Hofstra Law Review with Professors Michelman, Dworkin, Kronman, Baker and others as contributors.
V. Pareto, Manuel D’Economie Politique (2nd ed 1927). Of course a Pareto-superior reallocation may not be what economists call ‘Pareto optimal’. It would be so only if no further reallocations could make someone better off without making anyone worse off. There are many variants of Pareto criteria. See, e.g., G. Calabresi and P. Bobbitt, Tragic Choices (1978).
We are not unmindful that this example is unrealistic in several respects. It nonetheless serves our purpose well, including the purpose of stressing the rarity of genuine Pareto-superior reasons (without compensation).
It is not even necessary in order for a judicial decision to bring about a Pareto-superior allocation that it do so by facilitating free market exchanges among the parties or others who might be affected by the decision. Indeed, a Pareto superior move is theoretically possible even though it involves no voluntary bargains or exchanges whatever. All that is required by the Pareto criterion is that (i) there be a reallocation (or redistribution) of goods and resources, and (ii) that no one affected by this reallocation be worse off and at least one be better off in relation to antecedent preference rankings. All that a judge needs to know is, for example, that the parties would agree to reallocation were it not for some impediment, e.g., that one of them has erroneous beliefs, for example, about what his initial stock of goods and resources is. In such a case, the court can order the relevant reallocation.
Perhaps further elaboration is called for. In general, we have an intuitive notion of what is yours and what is mine, but that may not be good enough, for a Pareto reason requires resolving close issues and doing narrow calculations in order to see whether the fundamental constraint on the availability of such a reason is violated, viz., that no one be worse off by the decision. Consider our car dealer case. We built assumptions into that case designed to avoid these problems. But imagine a slightly different case, one in which there is no settled ban on protected groups such as the churches selling off rights to bring actions for violating the Sunday law against sales of cars. There would then be more voluntary transactions (for the costs of such transactions — risk of legal sanctions — would drop). Even the church groups could get into the bargaining act and offer to pay merchants to stay closed. The merchants would stay closed if the amount offered by church goers exceeded the expected profit to be made from Sunday car sales. There is a catch, however. Before the court renders its decision, there were restrictions on Sunday sales of cars. Those who would be willing to pay something to keep those restrictions in place — the churches — benefitted from such laws in a simple way: They were not faced with a situation in which they would have to pay anything to preserve the restriction. The question is: Do we say that the churches, prior to the court’s decision, have a right to be free from Sunday car sales (or at least that they were the beneficiaries of the law) so that this right is to be included in their ‘trading base’ for purposes of the initial allocation? Or, are we to say that the churches were merely the beneficiaries of an externality set up by the restrictions and so no better off in relation to their shares before or after the court’s decision to remove the restrictions? If the former, the churchgoers would suffer a loss from judicial removal of the ban on Sunday car sales (assuming they cared) and hence no Pareto-superior reason would be available to the court. Nothing in economic theory can determine this matter one way or another. But if the very availability of a Pareto reason depends on such a determination, then in a sense, the Pareto reason cannot have justificatory force unless its presupposition is justified. But such a justification ofthat presupposition—the criteria we set up for determining allocations—will not come from economic theory nor from any of the values typically implicated by economic theory. This fact calls in serious question the justificatory autonomy of Pareto reasons. Now, one might think that a simple way around this difficulty would be to let the existing laws of property serve to determine who has or gets what under initial and resulting allocations. But of course, that maneuver won’t work, as the foregoing example shows. Consider also cases in which property law is to be changed or is indeterminate or consider all those transactions, say in illegal goods, that do represent economic activity but in which the ‘owners’ of exchanged goods are not, for purposes of property law, considered to be owners or to have rights in those goods. But even if existing property law is taken to determine initial and resulting allocations, the justification for use of this criterion will be no better than the justification for that property law, and such justification will not rest on solely economic analysis, but often on policy-oriented, state-ordered distributions and redistributions. One might be tempted to think a Kaldor reason—to be discussed later—is immune from such difficulties. But that would be a mistake. Indeed, such reasons could not avoid questions of initial allocations at least insofar as a determination of initial allocations will be relevant to determining (and setting limits on) a person’s willingness to pay or what the aggregate gains and losses from the court’s decision will be. And depending on our criteria, answers to such questions may be very different.
Of course, the churches could not reinstate the restriction later (once lifted) for there would then be losers (the car dealers) and Pareto analysis would not permit this. But the question posed in the text introduces a further fundamental kind of indeterminacy in such economic analysis: Suppose the judge knows now of a possible future change of mind by parties to be affected now. Should he take this into account now in determining whether there will be any losers from a contemplated decision?
We suggested earlier that one way in which a court decision could work a Pareto-superior allocation was that the court might order a reallocation that the parties would themselves bargain to but for one market imperfection or another that can’t be overcome ‘cost-effectively’ in such a way that the indicated bargaining will actually occur. But what about the condition, ‘would bargain to’? One is inclined to ask, ‘would bargain to’ under what conditions of information? Depending on how much information we assume the parties would have, we will get markedly different results. If we assume that the parties have too much information we may find that they would bargain differently than they would had the market simply taken its natural course, or we may even find that their whole ordering of preferences would have been different. If the parties have too little information, our results will again not track their real, informed preferences, even those that are de facto.
This may be thought to cast doubt on the very idea of an economist’s reason. For when it comes to dealing with possible social goals such as net increases in psychological satisfaction, economists are head over heels in value theory, historically a well established branch of philosophy. This remains true even though economists adopt their own names (those of famous economists!) for the relevant value conceptions. More important, it remains true even though the reallocations of resources that their science is concerned with characteristically serve some general types of values.
The problem of trying to define in a fully satisfactory way a net increase in want realization is not one we can undertake here.
To evince respect for the choices of others is something else again, but we cannot go into the complexities of this.
The problem of defining a net increase in psychological satisfaction is not identical to the problem of defining a net increase in want realization. At least one can imagine computing the intensity and number of units of psychological satisfaction of different persons and multiplying accordingly. How one might do this with respect to wants, however, escapes us.
It may be noted that increased psychological satisfaction may derive not only from want realization. Third parties may become satisfied from an occurrence without having antecedently entertained a relevant want.
Of course, if the Pareto criterion has already been adopted in a relevant common law precedent, then it would become a relevant part of the justificatory analysis, even in a ‘no change’ case.
We should distinguish two senses of’loser’. A person may be a loser in court without being a loser in relation to what he had initially. These are not, of course, identical. A property claimant, for example, may ‘lose’ in court, but it may be true that he never really had a good claim, and hence is not a loser in relation to what he had initially.
It is sometimes suggested that if courts were to require winners to pay compensation to all losers in, for example, overruling cases, after the compensation is paid, we would have a state of affairs that would be equivalent to a Pareto-superior one (provided costs of exacting and paying the compensation did not wipe out the gain). In this sense, there might be many Pareto-superior reasons. As we will later see, however, economists do not, qua economists, even claim that such losers are actually to be compensated. See also n. 50 infra.
Of course, there are borderline cases and we do not claim that all common law cases fall neatly within our four categories.
We name this reason for the economist Lord Kaldor. See Kaldor, Welfare Propositions of Economics and Interpersonal Comparisons Of Utility, 49 Econ. J. 549 (1939), in which the ‘Kaldor criterion’ (gainers gain more than losers lose) is set forth. Today, the criterion is also sometimes called the ‘Kaldor-Hicks criterion’. See Hicks, Foundations of Welfare Economics, 49 ECON. J. 696 (1939).
For a classic statement of this aversion, see L. Robbins, An Essay on the Nature and Significance of Economic Science (1935). See also Harsanyi, Cardinal Welfare, Individualistic Ethics, and Interpersonal Comparisons of Utility, 63 J. Pol. Econ. 309 (1955).
See sources cited in n. 13 supra.
We do not hold that common law courts are not interested in net increases in the satisfaction of individuals! Courts do decide cases on the basis of such notions as ‘freedom of contract’, ‘family harmony’, ‘public convenience’, and ‘smooth labour-management relations’.
See also Jean v Foto in the text.
The Upland v Downland case introduces what some economists call the ‘cheaper cost avoider’ principle: Downland is the cheaper cost avoider so Downland must bear the cost. This terminology, however, is misleading. It will not do merely to impose the cost on the cheaper cost avoider (Downland) without further inquiry into gains. In fact, the imposition of such cost on Downland would not be justified on economic analysis unless the benefit to Upland of blacktopping his driveway exceeded the cost to him of putting in the blacktopping and the external harm (or cost of avoiding such harm) to Downland. Otherwise, people like Upland could engage in activities of very low benefit to themselves and at the same time impose huge external costs on other parties merely because those other parties happened to be those who could most cheaply avoid the external harm generated by such activities. That would be the case here if the value to Upland of the blacktopping were only a few dollars more than the cost of its installation, plus the costs of most cheaply avoiding resulting harm.
Why a ‘monopoly’ case? Of course, at least in the United States, such a case would qualify in a somewhat extended sense as a ‘common law’ case. But more pertinently, the case illustrates an economists’ reason in its most robust and indisputable form.
See Brown, Toward an Economic Theory of Tort Liability, 2 J. Legal Stud. 279 (1973). See also, Coons, ‘Approaches to Court Imposed Compromise’, 58 Nw. L. Rev. 750 (1964).
See our discussion of the Jean v Foto case in the text.
See n. 21 and accompanying text supra.
For a tiny sampling of the flavour of such analysis, see the price fixing case in the text.
The willingness to pay criterion is deployed over a wide range of contexts in R. Posner, The Economic Analysis of Law (2nd ed. 1977).
An alternative would be to apply the Kaldor analysis directly to determine satisfaction, contentment, utility, or whatever. To do this a judge would, of course, have to adopt some uniform unit of measurement (among other things).
See text between nn. 52 and 53 infra.
See, e.g., Baker, The Ideology of the Economic Analysis of Law, 5 Phil. & Pub. Aff. 3 (1975).
Utilitarianism, Economics and Legal Theory, 8 j. of Legal Stud. 103, 119 (1978).
A further problem with the willingness to pay criterion arises because of the ‘Kelman effect’ which, however, we do not explore here. See Kelman, Consumption Theory, Production Theory, and Ideology in the Coase Theorem, 52 So. Calif. L. Rev. 669 (1979).
In most cases there will be additional positive costs of transferring compensation and (sometimes) of enforcing such transfers. Thus, in the Edgar v Betsy case, the likely cost would probably be small (if Betsy has $150 of readily disposable cash) — she can simply write Edgar a check. But in the price fixing case, the cost of compensation (and of providing it in such a way that the primary [justifying] gains to consumers aren’t nullified) may be enormous. So in general, we cannot say that for every Kaldor transfer ordered without compensation there is a corresponding transfer (in which compensation is required) that is Pareto-superior and achieves the same level of net gain overall.
See Kaldor, Welfare Propositions of Economics and Interpersonal Comparisons of Utility, 49 Econ J 549 (1939) and I. Little, A Critique of Welfare Economics (1955).
See section II, 3 of text.
See further on autonomy, text accompanying n. 26 supra.
See section II, 3 of text.
Note that any such direct measure would have to compare interpersonal satisfactions. Although it may in principle be possible to determine and net the number and intensity of satisfactions (unlike in the case of want realization), factual inquiries of this nature are extraordinarily difficult to make, and it is fairly certain that in many circumstances neither the data nor an appropriately non-arbitrary measure would be available.
It is logically possible that independent values of quite different kinds might be served by Kaldor reallocations. We are framing our discussion, however, as a search for values that might be thought of as more or less characteristically associated with such reallocations. But by ‘characteristically’ we do not envision either logical connection or perfect correlation. This is not to say that quite different kinds of values cannot be realized in particular cases by such reallocations, but whether they are and what they will be, will turn on the quite specific facts of the specific cases. If the specific facts of the case indicate that a certain Kaldor reallocation would serve a valuable independent goal, this would favour the reallocation. More importantly, depending on what wants people happen to have, a Kaldor reallocation in a particular case might disserve the goal in question. In that case, we would have a reason (perhaps) to render a decision that precisely was not Kaldor efficient. What seems distinctive about Kaldor reasons is that a goal, increasing net social gains generally delimited by the willingness to pay criterion, is invariant across such reasons. If any goal that has a chance of being intrinsically prizeable characteristically, but perhaps not perfectly, tracks the invariant goal of maximizing ‘social gains’, it is increased satisfaction, happiness, contentment, etc.
See Posner, Utilitarianism, Economics and Legal Theory, 8 J. Legal Stud. 103 (1978).
Note, however, that an economists’ reason may be available (in certain circumstances) which supports a decision that is not the one among the available alternatives which maximizes resource worth the most. For example, suppose that there were three possible decisions, Dl, D2, and D3, that a court could render in a particular case. Let the only affected resource be R. It might turn out that Dl maximizes the worth of R relative to D2 and D3, but that D2 maximizes the worth of R relative to D3, i.e., that D2 will result in a greater increase in the worth of R (or less of a diminution in the worth R) than would D3. Finally, suppose Dl, the one of three which will increase the worth of R the most, must be rejected because of decisive reasons not derived from economics. D2 might be immune to such attacks. If so, it would still argue in favour of D2 over D3 that the former increased the worth of R more than did D3, i.e., if Kaldor reasons in fact have any independent and objective justificatory significance. Therefore, an economists’ reason of this sort can be given to support a decision which does not maximize resource worth relative to all the other possible decisions. Thus, the general decisional criterion which mediates the relevance of Kaldor reasons to all possible decisions is best formulated as follows: Ceteris paribus, choose that decision that will increase the worth of affected resources the most.
Prior to any transfer, the resource or right involved has its maximum worth, i.e., there is some amount more than which no one will pay to obtain the resource, for the value of the resource cannot be affected by transferring it or awarding it to anyone. Prior to any transfer, the resource already had its maximum worth, i.e., there is some amount more than which no one will pay to obtain it. The existence of this amount and thus of its maximal worth remains constant through any transfers, freely bargained or court ordered.
See Posner, n. 57 supra.
Even if the virtues involved are intrinsically valuable, it is not clear that a Kaldor reallocation or pursuit of wealth maximization substantially promotes them or is the best way to promote them.
For discussion of these other reasons, see the article in n. 1, supra.
Professors Kenneth Arrow and Frank Michelman have pointed this out to us. What we offer here addressed to this issue is necessarily abbreviated.
Of course in our view Kaldor reasons a fortiori do not swallow up tightness reasons’, even if for some purpose we may ‘factor’ them into the calculus. But we cannot go into this here.
We mean only two things by the words ‘social goal’: (1) that in general the goal is of a kind that may be pursued within the society on a broad social level and (2) that it is an appropriate goal forjudges to try to serve, at least in some contexts. The general efficacy of court decisions as means of serving such goals is a complex matter we cannot go into here.
See section IV, 1 of the text, supra.
Seen 1 supra, 735–51.
See n 1 supra, 752–53.
See text preceding n. 64 supra.
It might be thought that a judge simply cannot give a reason with justificatory force without putting it in some such general form envisioned in the text. But this is false. If we don’t know what the general (aggregate) effects will be, it is still a reason for decision that a contemplated decision will increase one persons’s satisfaction. It might also be thought that a reason a judge gives must always be in the form of a reason in support of a rule for a class of cases. This, too, is false, even with respect to goal reasons, for a judge may sufficiently know likely goal-serving effects of a projected decision to justify so deciding in the particular case yet not be able to identify sufficiently well the combination of general goal serving factors required to formulate a general rule.
One issue that we want to raise briefly (but cannot treat here in any detail) concerns the character of the economic generalizations supporting the predictions often ingredient in an economist’s reason. Some might object to calling such generalizations ‘causal’ or ‘empirical’; rather, it might still be said that these generalizations are really either tautological or definitional. The plausibility of this claim seems to depend, in the first instance at least, on what one takes the content of the relevant economic generalizations to be. For example, one economic generalization — that if price fixing is removed then, without more, the worth of affected resources will be maximized — could be interpreted as asserting only what is generally or likely to be the case and would not necessarily be interpreted as containing implicitly all of the assumptions, definitions, and boundary conditions that it would have to contain under some relatively formal presentation of micro-economic theory and welfare economics in order for it to assert an unqualified generalization (in the logician’s sense). On the other hand, when one states a corresponding generalization in which all such assumptions and definitions are made quite explicit, they do seem to become more like tautologies. This would appear to hold for our pricefixing generalization, especially if we built into it the assumptions that, e.g., all affected parties will be rational (in the game theoretic sense), that there will be no unforeseen transaction or information costs (i.e., that consumers will learn of the lower, competitive prices), etc. From an epistemological point of view, a switch to the more explicit and guarded formulation has its price, for with each new explicit assumption and condition, the less clear it is that the generalization in question actually applies to the particular case at hand (even though on its own, the guarded generalization may be more certain). From a practical point of view, it would probably be best for the judge to work with generalizations in which all relevant theoretical assumptions were explicit, for in this way he will be aware of precisely what he has to know about actual social conditions in order for his reason and its ingredient causal prediction to be sound.
See n. 1 supra, 735–51.
Seen. 1 supra, 752–73.
Schwartz, On the Utility ofMacKay’s Comparisons, J. of Phil. 549 (1975).
See, e.g., R. Posner, The Economic Analysis of Law (2nd ed 1977).
For a more extended effort to canvass the ‘legal’ value of rules, see Summers, Working Conceptions of the Law 1 Law and Philosophy 301 (1982).
See discussion of Jean v Foto in the text.
See discussion ofEdgar v Betsy and of Jean v Foto in the text.
‘[T]he most desirable kind of rule ... is a rule which wears both a right situation reason and a clear scope criterion on its face’. K. N. Llewellyn, ‘Grand Style and Formal Style Rules’, in W. Twining, Karl Llewellyn And The Realist Movement, 495 (1973).
See, e.g., R. Posner, The Economic Analysis of Law 18 (2nd ed. 1977) and remark in parenthetical text accompanying n. 38 supra.
Of course, Kaldor may already be ‘built in’. Compare n. 29 supra.
See text supra.
Finally, we are painfully aware that we have not been able to treat in any kind of detail a number of important issues in value theory. These include the basic questions of whether voluntary choice or psychological satisfaction is intrinsically good and of what it is ultimately for a reason to be objectively sound or have genuine justificatory force. Here, we have only been able to identify those points at which economic analysis in the common law makes contact with value theory proper and to indicate what our intuitions on these issues are.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Summers, R.S., McRoberts, W.G., Goodhart, A.L. (2000). Economists’ Reasons for Common Law Decisions. In: Essays in Legal Theory. Law and Philosophy Library, vol 46. Springer, Dordrecht. https://doi.org/10.1007/978-94-015-9407-3_16
Download citation
DOI: https://doi.org/10.1007/978-94-015-9407-3_16
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-5481-4
Online ISBN: 978-94-015-9407-3
eBook Packages: Springer Book Archive