Easterbrook, Frank H --- "The Inevitability of Law and Economics" [1989] LegEdRev 2; (1989) 1(1) Legal Education Review 3

THE INEVITABILITY OF LAW AND ECONOMICS

FRANK H EASTERBROOK^*

Economics is the study of rational behaviour in the face of scarcity. Economics and law are, therefore, inseparable. The legal system, too, is about coping with scarcity. If there were an abundance of every good thing, there would be no need for law, no need for a state. Life also would be boring.

Like economics, the legal system assumes rational behaviour. It seeks to influence by the threat of sanctions, such as imprisonment or civil damages. The coercive aspect of law assumes that persons care about consequences; “a legal duty so called is nothing but a prediction that if a man does or omits certain things he will be made to suffer in this or that way by judgment of the court”, a penalty set to influence the behaviour of the “bad man” who looks only to consequences — for other sorts, the law is less important anyway. ¹ Legislatures and judges believe that people will respond to these threats by modifying their behaviour so as to minimise the sum of compliance and sanction costs; the state, for its part, tries to minimise the sum of the enforcement costs and the residual harms caused by non-compliance with law — subject to a budget constraint and the desire to equate marginal returns on the many activities in which the state is involved.

The world of the economist starts with free trade and the world of the lawyer with regulation; the two disciplines often come up with different prescriptions for social interactions. But both are all about self-interested behaviour in a world of scarcity. Take away scarcity, and both professions have no reason to exist. If there is scarcity, law cannot be understood apart from economic thought. Neither teaching nor practice nor judging can disregard the subject.

My title proclaims the “inevitability” of economic analysis in theory (teaching) and practice, and now you can see why. But you are not going to receive a screed on the glories of economic analysis, a tale of a Marxist dynamic in reverse, inevitably leading to the dictatorship of the bourgeoisie. It would be more accurate, if banal, to call this talk “The Inevitability of Instrumental Analysis in Legal Reasoning.”

For this I do not need argument. Judges care about the effects of their decisions. Often statutes call for judges to achieve specified results (such as competition or clean air), and they must have some way to know what rules will produce these results. Every time a court starts looking at the likely effects of its decision, it is engaged in instrumental analysis. If it says it is construing some rule broadly to give effect to its purpose, it is engaged in instrumental analysis. The court supposes it knows both the objective of the statute and aim to attain more of that objective — and it necessarily expresses willingness to achieve less of something else in exchange. One way of criticising such a decision in the classroom is to point out that the court overlooked some additional consequence.

Law is not a closed logical system. Every dispute worth having involves some propositions about consequences. The propositions may be true or false, though it may be hard to say which. No litigant argues before the Supreme Court, no group proposes a new law, no teacher closes the book at the end of a class without making predictions about how the rule will affect society. Few opinions or committee reports omit predictions about effects. Litigants, legislators, teachers and judges alike believe that these effects are important in determining the law. These predictions are instrumental statements: we want the law or rule because it has effects on something or someone. These predictions usually rest on a tacit economic analysis. It is best to make this analysis express so that it may be done properly.

Legal processes contain competing traditions for making instrumental arguments. One tradition assumes that valuable things are scarce — that there is conflict over resources and that people try to improve their own position, generally pursuing some rational concept of self-interest in doing so. Another tradition assumes that people behave randomly or irrationally; sometimes writers in this tradition assume that valued things are plentiful, so that if we order them moved in today’s case nothing will happen in tomorrow’s. The latter approach is not economic, and I think not useful even as a philosophic tool. Decisions based on falsehoods, such as the belief that people will not respond to legal rules in order to protect their interests, will not achieve the purposes their authors had in mind. They may backfire, achieving the opposite of the intended purpose, as laws to “protect” women by preventing them from working overtime seriously injured women by discouraging employers from hiring them for the skilled and high-paying jobs that often require long hours. Without a method that enables us to see these things, we shall be unable to achieve our objectives or understand the consequences of the rules other people propose. In this sense economic analysis is inevitable. I give you a theorem. Those who employ economic analysis will drive out of the market those who do not, because the non-users cannot achieve their instrumental objectives.

NATURE AND LIMITS OF ECONOMIC ANALYSIS OF LAW

Economics is the study of rational behaviour under constraint. All good things are scarce — from air to cars to leisure to good lecturers. People work and bid to have these. Economics is at home predicting the effects of changes in scarcity, relative demands and so on. The domain of this method extends everywhere scarcity is found — nothing about it is limited to the study of transactions in markets. ²

Laws are, or alter, constraints. One can study the effects of these as of shortages of bread or reduced prices of microchips. So too we can look at the effects of scarcity on the law. The economic analysis of law is nothing more or less fancy than these things. Yet it deals with deterrence (the price of crime), with torts (the price of negligence), with contracts (economic bargains) and with laws (contracts at a social level).

Do not think that I propose that all people are rational in the sense of computers, or concerned only about money. People choose their own ends, which may include emotional satisfactions and altruistic endeavours. In pursuing these ends, they are short of time, intelligence and information. This shortage is part of rationality and constraint. Rationality does not imply constant calculation and deliberation; we rationally form habits and acquire predispositions that release time for other endeavours. Knowledge is scarce; time itself is the commodity we cannot get more of. Economic study takes these scarcities into account; one may (and should) study them as any other scarcity. Rationality implies no more than a good fit between means and ends. It is enough, in grappling with legal questions, to treat people as if they conserve all these scarce things. When they do so they will act as if rational under the circumstances. We cannot do better in predicting than to assume rationality of groups of people. A growing body of experimental economics confirms this. ³

There are limits, of course. Some people will misunderstand all things; most people will misunderstand some things, especially the probabilities and consequences of rare events such as floods. But when decisions are made repeatedly, either people learn from experience or those who do learn drive out those who do not. The (relatively) rational calculators will set the standard (an economist would call it the price) to which the group conforms. ⁴ The stock market is a good example. Those who cannot deal astutely with uncertainty and discount future events, soon lose their money to those who can; the market as a whole then behaves as a compound of the canniest predictors and evaluators.

There are two kinds of economic analysis: normative and positive. By positive I mean the study of the legal system, as if laws were bits of data. We may look at the discretions of prosecutors, sentences and plea bargains, then say: this looks like a market, and the system makes sense if we are trying to maximise deterrence per dollar of enforcement — though not if we think of criminal law as a system of just desserts. ⁵ It makes no difference whether the actors perceive themselves as doing economic things, for economics is a way of analysing conduct rather than of describing mental states. By normative, I mean a series of prescriptions. We may look at antitrust and say: in order to maximise efficiency, the rule of law must be thus. This is harder business — to prescribe a “good” rule of law requires us to handle data about the real world, which are difficult to come by and interpret.

VALUE OF ECONOMIC ANALYSIS

Why do it? Law is about efforts to influence human behaviour. To do this, or to understand how others’ efforts will succeed (or fail), we must have a system for dealing with how groups of people behave — and how groups of people interact over time. Economics provides this in a useful way. It is useful because it is consistent. It has a rigourous set of assumptions and rules, quite unlike the ad hoc utilitarianism cum moralism so common to legal teaching and thought, a “method” showing only that with enough inconsistent assumptions you can prove anything. It is useful, too, because it tells you what to look for. Life in all its fullness is far too complex for any analysis worth having. You must reduce — in physics we look at gravity without wind to find out what gravity does, and only then put wind back to find out where a ball goes when you throw it. So too with economics: we look at relationships of supply and demand excluding factors such as irrationality and nationalism.

METHODOLOGY OF ECONOMIC ANALYSIS

What do you look for? How does it work? Economic analysis arose from efforts to teach students to cope with “economic” subjects such as trade regulation and tax ⁶ and it is still largely a device for setting up and dissecting problems in the classroom and professional journals rather than providing definitive answers, but its scope has expanded to cover the legal enterprise as a whole. Useful economic analysis of law always keeps three things in mind.

1. Effects occur ex ante. Legal rules are about deterrence, not fair divisions of stakes. To view the law as concerned primarily with the distribution of spilt milk is both to miss its principal effects and to diminish one’s own ability to affect the way the world works.
2. Effects occur on the margin. People respond to rules of law (and legal penalties) by changing their behaviour, on the margin. As Milton Friedman puts it, there is no such thing as a free lunch (and if there were, it would not be worth anything). Looking at the “average” penalties, profits, and so on, is useful only if you wish to examine distributional consequences — which is useful only for ex post effects.
3. Legislation, like other contracts, is a market phenomenon. There is a market for rules of law. Sometimes this market produces useful rules; sometimes there is market failure. In studying legislation, we should not assume that every law serves the public interest. It may serve only private interests.

Let me spell out these avenues of inquiry with some examples. ⁷

Ex Ante Analysis

There is an eternal flight between pie slicers and pie enlargers. Cases that arrive in court stem from some calamity or irreproducible event. The mischance has happened; the pie is fixed (or ruined); so let us ensure fair divisions. There is a great temptation to look backwards and divide the stakes. So in a case in which the plaintiff says that its patent has been infringed, we might ask whether the patent owner needs extra income, or whether the new entrant should be allowed to grow. Ex post — given the patent — one should always favour the entrant. A patent puts a price on knowledge, which may be used without being used up, and therefore reduces efficient production. To take this ex post view for existing patents, however, has an effect ex ante; it reduces incentives to invent. The analysis is the same with rent control; fair rent today means less housing and maintenance tomorrow. Judges (and other participants in the law) must be alert to this conflict. So too they must see how people read. Rent control breeds “key money” — disguised rent, but going to the occupants, who do not produce the housing. Rent control, therefore, could both fail to reduce rent to subsequent tenants (who must pay off the initial ones) and reduce the supply of housing by reducing landlords’ income, the worst of both worlds. These adjustments are ubiquitous.

This may seem mundane. It is also called policy analysis, done in law every day. What is different about economics is, first, doing it consistently, and not invoking ex ante effects only when convenient; and secondly, doing it thoroughly, catching all the likely effects. To see the difference one need only compare two intellectual property cases in the Supreme Court of the United States. In 1963, when dealing with a small firm that copied a large firm’s lamp down to the last detail, the Court said that the right to copy things is fundamental. ⁸ It is a liberty; and we’re all against monopoly, right? It did not care about the effects of copying on the original design; we already had that. By 1979 the Court took quite a different view. It enforced a contract to pay perpetual royalties for designing a product. ⁹ Why? The Court informed us: to preserve the right incentives for making new designs. The 1979 decision was unanimous.

No analysis can be complete without an examination of ex ante economic consequences. If tort law requires the manufacturers of small private aeroplanes to compensate pilots (and passengers) under a standard of absolute liability, then the manufacturer must sell a more expensive package of plane and insurance (perversely subsidising the rich, because this package is worth more to them even though all customers pay the same price for it). The increase in cost has several consequences. Owners will keep older planes in service longer — and these planes, less safe than newer models equipped with the latest safety devices, are not only riskier to start with but become more dangerous with every hour in the air. ¹⁰ Pilots will form clubs, sharing a plane; each spends fewer hours in the air, and the reduction in experience makes each more dangerous. The same reduction in total safety may occur as commercial air transportation becomes more reliable. Scheduled air service entails only one per cent as many fatalities, per person-mile, as travel by automobile. ¹¹ Legal rules requiring air carriers to offer safer travel necessarily produce higher prices too since the methods of increasing safety are not free. Higher prices lead some persons to travel by car instead of plane (the ratio between air and road travel is very sensitive to price for trips of 1,000 kilometres or less). Unless a proposed increase in air safety is spectacular, and the diversion to road travel minimal, a legal rule meant to increase safety will have the opposite effect.

A careful treatment of effects on future conduct is essential, for disregard of them readily masks the unexpected. Gerald Frug’s contribution to this volume ¹² shows something of the problem while illustrating a difference between economic analysis and critical legal studies. Frug discusses a contract of one year’s employment at $200 per week between Jane and a male employer. Halfway through the year, Jane gets a better offer ($250 per week) and threatens to leave. After the employer says that he cannot find a replacement on such short notice (it is apparently the most important season for the business), they agree on a higher wage ($225 per week) for the same job. When the employer refuses to pay, the employee sues. Frug takes the employee’s side, on the ground that allowing the increase to $250 (or at least $225) “empowers” Jane vis a vis the employer — something Frug thinks is especially appropriate because women’s work traditionally has been undervalued and women exploited in the work place. Against this enlightened outcome Frug sets the employer’s expectation interests; but because Jane also acquires expectation interests (in the new $225 wage) and relied on the promise as well, the employer’s interests cannot trump “empowerment”.

This assumes that the employee always gets the benefits of the original bargain. Why? Is it because no employee similar to Jane will want to strike a similar bargain in the future, or because whatever the court does with today’s case will not affect tomorrow’s bargains? Either assumption would make analysis easier but not better. Suppose a court adopts a rule of law that employees always get to keep wage increases negotiated in mid-contract — or perhaps that female employees (but not male employees) get to keep such increases. What will happen? Nothing will happen if the employer and employee want a contract at will — when either can walk away on no notice. Contracts at will continually adjust as market wages change; the employee’s power to quit ensures an ability to obtain the going wage in exchange for staying. ¹³ But Jane signed a year’s contract. Why might she have done so? Perhaps Jane was learning the job; during early months her work was worth less than $200 per week (and the employer was incurring training expenses), while during later months her work was worth more than $200 per week. The year’s contract enabled the employer to agree to make the investment in increasing Jane’s skills while paying her a fixed wage; the implicit “subsidy” in early months would be repaid in later months. If Jane could collect her $200 per week (and enjoy the training) during the first six months and then switch to another employer who would pay $250 per week, she would be “empowered” (and enriched), but other women would be enfeebled. They would find employers much less willing to make similar investments in the future. Either employers would offer less training (to the special disadvantage of women) or would insist that employees pay for their own training in lower wages. Jane would be offered only $150 per week for the first six months on this job. A legal rule allowing employees to renegotiate in mid-year harms employees in the long term.

There are many other reasons why Jane and her employer might have agreed to a year’s contract. The labour market may be seasonal, with demand for Jane’s skills fluctuating. Jane may have wanted to smooth her income out over the year. Such an arrangement cannot last if, during the high season, Jane can go back to the market and earn the “spot” wage. No employer will be willing to pay more than the market wage in the low months if it must match the market wage in the high months. Or perhaps the employer’s business was concentrated during a few critical months, as agricultural work is. Frug places Jane in the fashion industry, where this is true. If the employee’s skills are applicable to many industries, the employer could be confronted with desertion at a critical moment in order to extract a higher wage. To avoid this, the employer will insist on a longer contract — and he must pay to get it. If the employee can renege on her promise, she loses the implicit year-round payment for cooperation in the peak season.

Economic analysis has offered increasingly sophisticated assessments of the cases holding that particular changes in contracts will or will not be enforced. These concentrate on the problem of opportunism — the employee (or employer) who seeks to change the terms in mid-contract simply because the other side has made expenditures in reliance on the bargain. ¹⁴ (Consider the actor who agrees to make a film for one million dollars and then, after the producer has completed half of the scenes at a cost of ten million dollars says, “either increase my pay to two million dollars, or re-shoot the movie with someone else at an extra cost of ten million dollars.”) The cases generally refuse to enforce alterations in response to opportunistic conduct but do enforce alterations in response to unanticipated events, sparing the parties the need to draft ever-more detailed contracts to handle rare events. The doctrine of consideration cannot explain the cases, but it hardly follows, as Frug believes, that they are inexplicable or that any legal argument is as good as any other. Quite the contrary, economic analysis shows that existing rules are highly functional, and that substantial departures from them would make both employees and employers worse off.

Frug might offer still another reason for the annual contract: exploitation. Perhaps the employer has the ability to force his will on Jane. If we assume this, however, we must assume that an employer compelled to offer Jane better duration terms (that is, the ability to renegotiate) would have the “power” to insist on a lower wage. If one person can exploit another, that person can take the full gain once, and once only; he cannot strangle Jane with a contract longer than she prefers, and then exploit her again with a lower wage. So long as wages and prices remain negotiable, common law rules cannot strengthen the hand of one group against another in future transactions. ¹⁵ Consider the effect of a legal rule providing that because women traditionally have been exploited in employment markets, women but not men, may renegotiate wages in mid-contract. An employer thinking that the terms of its contracts were important would prefer to hire men: for men but not women would be legally empowered to make binding promises. Formerly, married women could not make contracts. A contract is a mutually binding promise. A rule of the sort suggested by Frug, under which women would be free to abandon their undertakings as soon as someone else made a better offer, is effectively a rule disabling all women from making contracts — for it makes their promises worthless to putative contracting partners. The history of laws of this kind is too depressing to envisage a repetition in the name of “empowerment”.

One final aspect of ex ante analysis in law. Judges who disregard future effects may well come to the right conclusion, but their work will be less stable, because it can easily be attacked by analysis appealing to “fairness” and similar ex post contentions. I offer as an illustration the decision of the High Court of Australia in Breavington v Godleman. ¹⁶ The High Court considered whether to replace Australia’s prevailing rule that the law of the state in which the suit was filed applied to torts, no matter where they took place. The Northern Territory adopted a no-fault motor vehicle insurance scheme, under which recovery for injuries is certain but limited; Victoria uses a negligence system, under which recovery is less likely but more generous. Breavington filed suit in Victoria to recover for an accident that occurred in the Northern Territory. All seven Justices held, in six judgments, that the law of the place of the accident must be applied, discarding the former law in Australia.

Lex loci delicti is a plausible rule. It may serve many functions. It enables persons to know with greater certainty the consequences of their acts, and thus to plan intelligently — something especially important in designing and manufacturing products and drugs. Those who know rules also can take precautions such as obtaining insurance; and insurers will sell the product at lower cost if they, too, can assess their exposure. It cuts down on litigation about choice of law. It enables states to tailor their rules to achieve optimal deterrence and compensation. Both a high probability of a modest award (as under no-fault systems) and a lower probability of a higher award (as under negligence systems) may have the same expected value; but if the injured party with the best case can choose to sue in the state using a negligence rule (obtaining the higher award), while the party with a low chance of showing negligence can collect under the no-fault system, then net recoveries will be higher than either state envisages — a result that may discourage driving (and other risky activities) without making drivers more careful. These and other economic functions of the tort system ¹⁷ could have led to a well-supported decision to favour the law where the accident occurred.

Not one Justice of the High Court mentioned any of this. Several, however, spoke of the evils of “forum shopping” without explaining what these were. Perhaps an objection to forum shopping is shorthand notation for these kinds of considerations, but perhaps not. Who could tell? Maybe an objection to forum shopping is only an instinct for what is “fair” to defendants. But why is it the purpose of tort law to be fair to tortfeasors? Why not be fair to victims? Since fairness to victims often seems to mean higher recoveries, what is wrong with forum shopping? Sometimes forum shopping appears inevitable. Think of an injury caused by a product designed in state A, manufactured in State B, sold in State C to a resident of State D, repaired in State E, which breaks in State F gravely injuring a resident of State G, who later moves to State H, which affords him medical care at public expense. What is fair about applying State F’s law to this tort? A court that lacks a theory — as opposed to a slogan — about why one choice of law rule is preferable to another will not be able to cope with this kind of case, which presents the pressures for ex post analysis that recur in the law.

It is not my purpose to criticise the High Court, for Breavington is not inferior in any way to the analysis which courts in the United States apply to questions of this sort. Indeed, I rather prefer Breavington to American cases, for a majority of the states in the United States has abandoned the lex loci rule in favour of the “choice-influencing considerations” approach sketched by the American Law Institute. The development of American conflicts law has been driven by consistent yielding to the siren of “fairness” — at each turn choosing the rule that maximises plaintiffs’ recovery, in large part because courts lack a theory about choice of law and the consequences of tort doctrines. The upshot includes products that are uninsurable (and hence not made), even though they have substantial benefits. This “modern” approach has come under increasing criticism by commentators who point out economic (and other) adverse consequences of an approach so foggy that it is no rule at all. ¹⁸ A court without a theory of consequences is a court without a stable law. If Australia is to follow in the path of American choice of law decisions, it ought to have a clear view of what it is getting.

Marginal Analysis

If we wish to know the effects of rules and decisions, where shall we look, and for what? Economists say, look at the margin. This means incremental effects. To know whether the death penalty is a useful punishment, look at the change in the volume of murders that accompanies a change in punishment, not at whether there is “a lot” of murder with or without capital punishment. It is common to say, “criminal law does not deter; look at all the crimes.” This is like saying that a higher price does not discourage the sale of cigarettes; they are addictive, and look at all the sales. Rates of change are more important. It turns out, much to most people’s surprise, that the sales of cigarettes and liquor fall off faster when prices rise than do sales of yams or automobiles. ¹⁹ People substitute from cigarettes toward chocolate faster, when the price of cigarettes rises relative to chocolate, than they substitute from soda pop to chocolate. Even though it is hard to give up cigarettes, it turns out to be relatively easy to cut back on smoking; and for some persons a change in price makes it worthwhile to give up altogether. The sales of cigarettes are high, and nicotine is habit forming; but it would be a grievous mistake to assume that law has a limited influence over such things just because, on average, it is hard for a given person to kick the habit. That is a lesson with general application.

To return to patent law: it was common for a long time to say that a restriction on the use of patents or the scope of coverage did no harm, because substantial profits could still be made from exploiting patents. This looked at the average return. It ignored those things that will not be invented because a reduction in return is bound to make the venture not worthwhile for some people or firms. Effects on the margin tell us that we will get less invention.

In general, concentration on marginal effects forces us to look at substitution. If we make innovation less attractive relative to something else, we will get not only less innovation but more of the something else. Substitution is ubiquitous. If we imprison hardened criminals, we incapacitate them. But this also opens up new vistas of crime. Good burglars crowd bad ones out of the market in crime. If we imprison all the best burglars, we will not wipe out burglary; we will simply have more amateurs, who find that the returns to crime have just gone up. ²⁰ We should never stop analysing the immediate effects of a legal rule; we must look at how people adjust to the rule.

So if we tell the National Highway Transportation Safety Agency that it cannot repeal a rule requiring all automobiles to have passive restraints (air bags) without having a good reason for changing its mind — as the Supreme Court did four years ago ²¹ — we have made regulations more durable on the margin. Durable goods are worth more, so interest groups fight harder, both to obtain beneficial regulations and to o p pose costly ones. Agencies devote more time to each rule. If they cannot change their minds as easily, effects both good and bad are locked in. Agencies, therefore, study the effects more fully before acting. The Supreme Court’s decision turns out not to encourage regulation since it makes a return to laissez faire harder. Instead, it may discourage regulation by reducing the number of projects an agency can handle and by postponing the implementation of any one regulation.

Another example can be taken from a case decided under s46 of Australia’s Trade Practices Act 1975. This statute is designed, as I understand it, to forbid devices by which monopolists (or firms with substantial market power) reduce the effectiveness of their competitors — by raising rivals’ costs of production or by reducing the rate at which they increase their production in response to increases in price. Broken Hill Proprietary (BHP) is the only Australian manufacturer of “Y-bar”, a steel component of a particular kind of fence. It uses Y-bar to make fences, which it sells, but it does not sell Y-bar to others who wish to make fences. A manufacturer of fence wire sued under 546, seeking a supply of Y-bar at “reasonable” (perhaps meaning, competitive) prices, and it lost on the ground that because there is no separate “market” in Y-bar, the statute does not apply. ²²

Surely Y-bar can be sold separately. BHP sells it to subsidiaries, and some has been imported. There is a demand for it as a separate product (hence the litigation). As an economic matter, supply consumed internally is part of a market. The question is not so much whether to apply the name “market”, but why BHP declines to sell. If it is just trying to collect the monopoly price attributable to its market power, this attracts rather than excludes entry and so would not violate 546 (if, as I am assuming, that statute deals only with exclusionary practices). If there are efficiencies in vertical integration, things would be the same. And maybe the kind of fence made from Y-bar competes with other kinds of fences, so that BHP has no market power even if it is the sole manufacturer of Y-bar. But maybe something else is going on. In the United States, the Aluminium Co of America (Alcoa) — at a time when it held a monopoly of aluminium manufacture — employed a device similar to BHP’S for selling consumer goods and was branded a monopolist on that and other grounds. ²³ Many of Alcoa’s industrial customers demanded “virgin” aluminium. If it sold virgin aluminium to firms so that they could manufacture pots and pans, they might, instead, sell the ingot in Alcoa’s primary markets. It could sell consumer products made from aluminium at low prices (reflecting consumers’ ability to substitute stainless steel pots and pans) without jeopardising its ability to obtain a high price for virgin ingot. So Alcoa made the pots and pans itself, refusing to sell ingot at a price low enough to enable would-be rivals to manufacture consumer goods at similar prices. This led to howls of protest from rival manufacturers of consumer goods because the implicit price of the aluminium in Alcoa’s consumer goods was well below the price at which Alcoa sold aluminium ingot. This sort of price discrimination not only enhances a monopolist’s profits but also eliminates rivals from the market. If this is what BHP is doing, then the case under 546 appears in a different light. Far be it from me to say what 546 does or does not prevent. I imply no view whatever on questions of Australian law. My point is only that economic inquiry puts the questions in focus so that answers may be given with knowledge of the consequences.

One final example: in criminal law, courts often exclude evidence that was unlawfully obtained. They give, as a rationale, that the exclusion will deter wrongful conduct by the police. Wanting to get convictions, the police will obey the rules. This is an ex ante argument for a rule of law, based on the influence it has at the margin. So far, so good. But we cannot stop. The exclusion of evidence reduces the number of convictions (if only by making investigations harder to conduct, so police will turn fewer suspects over to the prosecutors). A reduction in the number of convictions has effects of its own. Judges, no less than police, seek to maintain the level of deterrence. They may do this by increasing sentences in the event of conviction. If convictions become scarcer (or harder to obtain in the kinds of cases, such as drug prosecutions, most affected by exclusionary rules), an attempt to maintain constant deterrence must mean higher sentences elsewhere. So the criminals whose rights were respected (and who, therefore, cannot exclude any evidence) pay in higher sentences for the invasion of the rights of other prisoners.

Any effort to maintain “uniform” sentences will have a similar effect. The federal courts in the United States are about to embark on uniform sentencing under a set of guidelines. These guidelines inevitably will depart from the preferences of many judges and the expectations of many litigants — after all, these diverse preferences explain why we now have non-uniform sentences. If the guidelines set some ranges too high, these defendants will refuse to settle by pleading guilty. It will take more time to prosecute each of them; and to maintain the level of deterrence the state must prosecute more people in other categories and fewer in the categories with the newly increased sentences. The anticipated sentence (viewed from the perspective of someone about to commit a crime) may very well go down even as the imposed sentences go up; so deterrence may fall as retribution rises. Such is life in a world where margins, rather than averages, matter. We should take account of these things before tampering with legal rules that have lasted a long time.

The Meaning of Legislation

There is a comfortable tradition that legislation is the way to achieve the public interest. That tradition has implications for interpretation. Judges can see the direction in which the statute points and can get more — when they are sure more is a good thing. That means private rights of action freely implied, remedial legislation liberally construed, and many other canons.

Maybe, though, laws are like TV sets — commodities to be bought and sold. There is a developing recognition that laws respond to the interests of organised groups — that discrete minorities are better at securing laws than are diffuse majorities. Much American constitutional law is based on the proposition, stated in the Carolene Products case, ²⁴ that “discrete and insular” minorities need special protection. Yet today in the United States minorities, having the ballot, assemble coalitions to obtain substantial transfer payments in their favour, using their ability to vote as a bloc. ²⁵ The same effect occurs with less-discrete minorities: lobbying groups, pro and con, on gun control, abortion and similar issues. Economic interests organise in the same way. Gains and losses can be quantified in money; and these coherent interest groups may be very good at achieving what they want. The power of the cohesive minority to obtain legislation — even legislation that injures the rest of us more than it helps them — is a feature of any democratic form of government. We must understand it in order to appreciate the significance of laws. ²⁶

Once we see that law may be a compromise among competing interest groups, nothing is sacred about the direction in which a law points. The stopping point may simply tell us the amount of protection that has been purchased. ²⁷ Compromises have no spirit; they just are. This suggests that one cannot take a law and implement its purpose; there will be a complex set of decisions. This is something the Supreme Court of the United States is beginning to recognise. ²⁸

There is something special about legislation, however. If it is bought, or at least rented, the transaction is not with money or in the open. Goods and services are sold in liquid markets; everyone can choose how much of each item he or she wishes to have, and no-one’s choice precludes someone else from making a different choice. A political choice is a collective choice, obtained not with money but with “support”. Coordinated groups can furnish support, even though they have less to gain than larger majorities have to lose. So laws may turn out to be perverse. This does not mean only subsidising tobacco growers at the same time as the government spends many millions trying to persuade people not to smoke — though the United States does both. It means not only milk subsidising prices — transfers from babies to cows; it also means entry control and many other kinds of laws. There is simply no necessary correlation between what is good for the interest group and what is good for the nation as a whole.

Empirical work shows that even pollution laws, which seemingly have good free-rider justifications, may be arranged to stifle new entry. One rule in the United States is called “prevention of significant deterioration” — if the local air is clean, we will not allow it to become significantly dirtier, even though areas with air that is just a little dirty will not be made cleaner. Such rules are the Rust Belt’s revenge on the Sunbelt, making it cheaper to pollute where you are and costly to open new plants elsewhere. ²⁹ It is much more expensive in the United States today to start new (clean) plants than to run old (dirty) ones harder. That’s not all. Another environmental rule requires emitting plants to reduce pollutants by some percentage rather than to hit a fixed maximum release objective. This rule gives a big economic advantage to high sulphur coal (from the midwest) over clean coal (from the west). The clean coal is more costly, yet the user cannot save on pollution control equipment even though burning clean coal with no scrubbers yields less pollution than burning dirty coal with the best scrubbers. I tried a case last summer in which a company avoided buying low sulphur coal because the Environmental Protection Agency virtually forced it to burn dirty local coal — it could discharge pollutants galore from local coal at older generating stations but had to use very expensive equipment to clean up the clean coal further in its newer stations. ³⁰

The recognition that laws may be products, and that the market may be perverse, suggests great care is needed in statutory interpretation. Judges must not give interest groups more than they bought. They should look for the deal, perhaps trying to flush deals into the open. Removal of rose-coloured glasses affects everything we do as lawyers.

LIMITATIONS OF ECONOMIC ANALYSIS

It is time to stress some of the limitations of economic analysis. Like any other tool, economics may be harmful if misused.

Roles of Assumptions and Data

You may hear that economists are conservative ideologues. On some things, like minimum wages and price control, economic views track the conservative agenda. Some economists are ideologues. But the ideology runs the political spectrum. ³¹ Many, from Keynes to Samuelson, are very liberal; John Kenneth Galbraith is a good example of the liberal economic ideologue. Economics is a method, not a result. It tells you what to ask and look for, but if you would evaluate the effects of laws using economics you must have data.

Think of economics as professional scepticism. ³² Does someone say that a law is in the public interest? Show me. Does someone claim that a market failure justifies regulation? How do you know the regulation will not be worse? Economists want a comparative judgment. Do not compare the imperfect market against a perfect government. Government has its own costs — and fewer self-correcting mechanisms, such as competition. Bureaucrats do not suffer for mistakenly refusing to approve new and useful drugs in the way that pharmaceutical companies suffer for introducing bad drugs. This introduces a bias into administration, of which we must be aware in interpreting legal rules. ³³

I edit a journal that contains lots of data, the Journal of Law & Economics. One of its successes is in antitrust law. Once it was thought that any concentration, such as four firms with fifty per cent of the market, was dangerous. We have seen a change in antitrust policy throughout the liberal democracies because that belief was shown to be false. Profits are not higher in concentrated industries. They are higher only for the largest firm in the market which implies growth of the most efficient. ³⁴ Size comes from ability to please the most people. The opinion of the whole antitrust profession, not just of people at the University of Chicago, has changed in the last ten years. There is now no difference of opinion between Chicago and Donald Turner, who as head of the antitrust division during the Johnson administration brought many of the cases economists loved to hate. Data and arguments made a difference in a field that from the start had made policy on the basis of instrumental arguments.

Consider the debate over takeovers. Those who want to regulate, slow down or forbid takeovers will say that markets value the short run so that takeovers penalise long-term planners and injure the economy. These claims have implications. Do firms taken over have unusually high investment in research and development? No; it turns out that they have less than half the national average investment! ³⁵ Do firms that defeat takeover bids prosper? No; their stock price falls relative to the market, implying that they are not gems in the rough. ³⁶ Economics directs our attention to the right questions in a field dominated by claims about consequences, and economic data provide answers. It is the data, and not untested theories, on which we ultimately must rely. ³⁷

Testing propositions is the proof of any pudding. A dispute arose at the beginning of the Reagan administration about decontrolling the price of natural gas and the entry of new firms into the market. Economists said that decontrol would lead within a short time to lower prices as people found more gas; self-styled “consumer advocates” said decontrol would lead to much higher prices as vicious firms gouged consumers. What happened? The price of gas fell. The same debate occurred about airline prices, with the additional claim that only regulation protected safety. Australia furnished a test, with state-owned and private carriers competing and the private carrier consistently performing more efficiently; ³⁸ so did California and Texas in the United States, demonstrating how prices fell as competition increased. What happened with deregulation? The price of air travel fell. Travel was as safe as ever, with the new entrants as safe as the established firms. ³⁹ The verification of economic premises in such large scale tests is a basis on which we can extrapolate to other disputes.

The Domain of Economics

Economists are imperialists. They moved from markets for goods to markets for crime. They conquered torts and securities and have invaded civil procedure and family law. This should come as no surprise. Economics is the study of maximisation under constraint. Constraint — scarcity — is everywhere. The tools go along.

They are congenial tools in a liberal republic. Economics treats people as autonomous, able to decide for themselves. Certainly these were the premises on which the founders of Australia, New Zealand and the United States operated. The moral basis of economic analysis is the Pareto criterion: any transaction that improves at least one person’s lot and makes no-one worse off is desirable. There is widespread agreement on this as an ethical rule, and it describes the sorts of voluntary exchanges with which economic inquiry is concerned. (If the exchange does not make at least one party better off, why do we observe it? If it makes one party worse off, that party would balk. Hence voluntary transactions between adults satisfy the Pareto criterion, and likely make all participants better off.) Although transactions often leave someone worse off in retrospect (the buyer of stock who is disappointed when the price falls; the spouse disappointed by the chosen partner), they are beneficial ex ante, which is the right time frame. One cannot do much about the way things turn out without stifling agreements that everyone desires at the outset.

It is hard to escape the use of these tools even in constitutional law. North Carolina enacted a statute providing that professional fund-raisers who charged charitable organisations for their services more than 20 per cent of the donations they collected were presumptively charging too much, and that fund-raisers who charged more than 35 per cent had to bear a heavy burden of justification. The law also required fund-raisers to obtain licenses and to disclose to potential donors their customary fee. The Supreme Court of the United States held that this was unconstitutional as an abridgement of free speech. ⁴⁰ It did so on wholly economic grounds.

First, it observed that money is speech and speech money. Charities are “causes”; what they do and what they say is inseparable. To speak they need money; to raise money they proselytise. Next, the Court saw that charities will use cost effective means to raise money. They do not hire professional fund-raisers unless the net income exceeds that obtainable from amateurs. To regulate professional fund-raisers is to make them less readily available, making the speakers’ options less attractive. Third, the Court noted that laws requiring fund-raisers to disclose information they would prefer not to, diminishes the effectiveness of their speech on the margin, and indirectly the amount of money charities obtain with which to speak. Failure to disclose will not mislead the donors; they need only ask for whatever information concerns them. Finally, the Court reasoned that licensing rules retard entry into the business, cutting down the number of organizations (perhaps creating market power) and raising their costs (compliance costs money) even if competition continues. Charities are worse off.

Of course, all of these things happen only if both the charities and the donors are rational economic actors. If professional fund-raisers are sharpsters while charitable organisations are dullards, then the charities will be deceived. If putative donors are not capable of deciding which organisations to support, nor of noting the difference between professional and altruistic solicitors, nor of seeing that their favourite charities obtain only the donations net of fundraising costs, then they will be misled. North Carolina’s rationale for regulation was one of selective rationality: professional fund-raisers would be more coldly calculating than charities and donors, so the latter needed protection. The Supreme Court rejected this assumption of selective rationality, leading immediately to a conclusion that the law is unjustified. ⁴¹

Does this imply that all judicial decisions are economic decisions and that judges just vote according to their agendas? No indeed. Legal decisions must first conform to a political theory, and the theory underlying both of our nations is one of divided and limited authority. Judges are supposed to implement laws rather than invent them. There are good economic reasons for this. Divisions within the government are substitutes for the sort of checks that competition in markets provides. ⁴²

Adhering to the allocation of powers is vitally important for any approach to the economic analysis of law. Economics is instrumental reasoning. We must start by asking when instrumental arguments are appropriate. Unless there is a grant of authority to employ such reasoning, it is inappropriate. Ours is a government in which each branch has limited power, so the first inquiry is the scope of the grant. The grant may even be anti-economic. The Constitution of the United States is not designed to make government easy but to constrain it. A parliamentary system is supposed to be more “efficient”, but because changes in government can bring sudden swings in policy it, too, may confound the government in the long run. It is hard to say that either the system in the United States or the parliamentary system yields an efficient, powerful, government — compared, say, to the self-perpetuating system in the Soviet Union (or a modern university!).

Statute and Administrative Law

For clear statutes and much of administrative law, instrumental reasoning has no role in the courts, though it does have a major role in the political branch of the government and the operational agencies. The political branch may have made the instrumental choices which bind courts and agencies. Or it may have delegated the making of these choices. So honest judges implement rules they detest — all because of the limited authority of a judge bound to carry out someone else’s decision.

It is essential to learn whether a statute is a delegating statute or a rule-creating statute. The principal antitrust laws in both Australia and the United States refer to competition but do not tell a court how to achieve it. The court, therefore, has little choice but to devise an economically-oriented common law. ⁴³ On the other hand, there is little role for instrumental inquiry in environmental cases, where the decision may be anti-economic. One need only think of the decision, sustained in Cotton Dust, to regulate toxic substances without regard to costs and benefits. ⁴⁴ There is especially little role in anti-discrimination statutes, expressly designed to override the results of market interactions. Other cases, like securities laws, may employ instrumental inquiries to measure damages but not to fix the amount of disclosure. The appropriate domain of economics follows from the careful, statute-by-statute inquiry into the grant of authority to courts and agencies. (From an economic perspective, courts are simply the administrative agencies with the most diverse portfolios.)

Common law

Federal courts in the United States are bound by the Erie Doctrine, ⁴⁵ so the economics of tort and contract law has little role except to the extent federal judges and law journals may make suggestions to the states. But we can consider instrumental questions about procedure, such as the meaning of harmless error. The meaning of the rule, and the procedures used to redress errors concerning that rule, are inextricably linked. We cannot know whether to enforce a rule vigorously or instead to take a relaxed view of error without knowing how valuable the rule is, what other means there are to secure adherence to it, and what the full consequences of any level of error will be. ⁴⁶ And we can always take into account the ease with which judges can destroy wealth. We must learn that the implication of warranties of habitability and the like will not improve the lot of the poor. You cannot transfer wealth unless you control price, quantity and quality; if you can control only quality, the price will rise and the quantity fall. A judicial decree saying, “all housing must be suitable for the middle class” produces housing that only the middle class can afford. A decision annulling a contract or warranty, on the ground that it is a printed term or some similar ground, does not affect the price; it simply reduces the number of options available and may eliminate a beneficial one. Knowing the inability to restructure society through the judicial process should make us appropriately modest. And in recent terms the Supreme Court of the United States has been quite modest about its ability to improve life by striking clauses out of contracts — sustaining, for example, clauses requiring the arbitration of antitrust claims in international transactions, and of domestic securities claims. ⁴⁷

Constitutional law

Here there is usually no role at all. Constraints on the acts of other branches come from political rather than economic theory. Efforts to introduce economics are unwarranted. Mathews v Eldridge ⁴⁸ defines due process as economic-cost benefit analysis. But there is no warrant for this; due process had a different meaning historically and there is no reason to think that the Constitution deputises courts to take the active, policy-making role. ⁴⁹

CONCLUSIONS

Economics is a way to think before it is a way to act. Sometimes it is not a warrant for any action. But we live in a world where consequentialist arguments dominate legal debate. If consequences matter, we need a method to derive consequences and assess their costs on other people. No-one can give useful answers to hard questions without being able to think through the problems that economics sets. So I close where I began: economics is not an addition to law, a strange outside force. Economics is an integral part of the study of legal rules and the rule-making process. The only question is whether we do this well or poorly.

---

^[*] Judge, United State Court of Appeals for the Seventh Circuit. University of Chicago Law School.

© 1989. [1989] LegEdRev 2; (1989) 1 Legal Educ Rev 3.

^[1] OW Holmes, The Path of the Law (1897) 10 Harv L Rev 458, at 460–61.

^[2] See J Hirshleifer, The Expanding Domain of Economics (1985) 75 Am Econ Rev 53. For applications of economics outside the customary field, see GS Becker, A Treatise on the Family (Cambridge: Harvard University Press, 1981); RA Posner, The Economics of Justice (Cambridge: Harvard University Press, 1981). Contrast RH Coase, Economics and Contiguous Disciplines (1978) 7 J Legal Stud 201.

^[3] For example, S Schwab, A Coasean Experiment on Contract Presumptions (1988) 17 J Legal Stud 237; E Hoffman & M Spitzer, Experimental Law and Economics: An Introduction (1985) 85 Colum L Rev 991. Both of these articles not only report the results of experiments but also discuss similar literature.

^[4] See GS Becker, The Economic Approach to Human Behaviour (Chicago: University of Chicago Press, 1976) at 1–68; AA Alchian, Uncertainty, Evolution, and Economic Theory (1950) 58 J Pol Econ 211. It is not altogether fanciful to treat this view of group rationality as a dose kin to evolutionary theory in biology. See RR Nelson & SG Winter, An Evolutionary Theory of Economic Change (Cambridge Massachusetts: Belknap Press, 1982).

^[5] See FH Easterbook, Criminal Procedure as a Market System (1983) 12 J Legal Stud 289; See also RA Posner, An Economic Theory of the Criminal Law (1985) 85 Colum L Rev 1193.

^[6] See generally EW Kitch ed, The Fire of Truth: A Remembrance of Law and Economics at Chicago 1932–1970 (1983) 26 JL & Econ 163.

^[7] For an overview See RA Posner, Economic Analysis of Law, 3rd ed (Boston: Little Brown, 1986); R Cooter & T Ulen, Law and Economics (Glenview Illinois: Scott, Foreman, 1988); CJ Goetz, Cases and Materials on Law and Economics (St Paul. West, 1984); AM Polinsky, An Introduction to Law and Economics (Boston: Little Brown, 1983); P Burrows & CG Veljanovski eds, The Economic Approach to Law (London: Butterworths, 1981); WZ Hirsch, Law and Economics: An Introductory Analysis (New York: Academic Press, 1979) and many other sources. For more along the lines of the text, including a complaint by Laurence Tribe about the use of economics in law, see FH Easterbrook, The Supreme Court 1983 Term — Foreword: The Court and the Economic System (1984) 98 Harv L Rev 4; LH Tribe, Constitutional Calculus: Equal Justice or Economic Efficiency? (1985) 98 Harv L Rev 592; FH Easterbrook, Method, Result and Authority: A Reply (1985) 98 Harv L Rev 622.

^[8] Sears, Roebuck & Co v Stiffel Co [1964] USSC 68; (1963) 376 US 225.

^[9] Aronson v Quick Point Pencil Co [1979] USSC 35; (1979) 440 US 257.

^[10] PW Huber, Liability: The Legal Revolution and its Consequences (New York: Basic Books, 1988) at 161–70.

^[11] RB McKenzie & WF Shughart 11, Deregulation and Air Travel Safety (1987) 11 Regulation (Nos 3/4) 42.

^[12] GE Frug, A Critical Theory of Law [1989] LegEdRev 5; (1989) 1 Legal Educ Rev 43.

^[13] On the consequences of employment at will, see RA Epstein, In Defense of the Contract at Will (1984) 51 U Chi L Rev 947; DL Martin, The Economics of Employment Termination Rights (1977) 20 JL & Econ 187.

^[14] See for example, VA Aivazian, MJ Trebilcock & M Penny, The Law of Contract Modifications: The Uncertain Quest for a Benchmark of Enforceability (1984) 22 Osgoode Hall LJ 173; TJ Muris, Opportunistic Behavior and the Law of Contract (1981) 65 Minn L Rev 521.

^[15] See A Schwartz, Justice and the Law of Contracts: A Case for the Traditional Approach (1986) 9 Harv JL & Pub Pol’y 107; RA Epstein, The Social Consequences of Common Law Rules (1982) 95 Harv L Rev 1717.

^[16] [1988] HCA 40; (1988) 80 ALR 362.

^[17] On which see WM Landes & RA Posner, The Economic Structure of Tort Law (Cambridge: Harvard University Press, 1987); S Shavell, Economic Analysis of Accident Law (Cambridge: Harvard University Press, 1987).

^[18] L Brilmayer, Governmental Interest Analysis: A House Without Foundations (1985) 46 Ohio St LJ 459; L Brilmayer, Related Contacts and Personal Jurisdiction (1988) 101 Harv L Rev 1444; JH Ely, Choice of Law and the State’s Interest in Protecting its Own (1981) 23 Wm & Mary L Rev 173; Posner, supra note 7, at 553–54.

^[19] GS Becker, M Grossman & KM Murphy, An Empirical Analysis of Cigarette Addiction (unpublished paper, May 1987); see generally GS Becker & KM Murphy, A Theory of Rational Addiction (1988) 96 J Pol Econ 675. The pattern holds for alcohol too. See D Coate & M Grossman, Effects of Alcoholic Beverage Prices and Legal Drinking Ages on Youth Alcohol Use (1988) 31 JL & Econ 145; H Saffer & M Grossman, Beer Taxes, the Legal Drinking Age and Youth Motor Vehicle Fatalities (1987) 16 J Law Stud 351. Much evidence shows that other “addictive” substances also can be discarded when the users decide to do so. Almost all of the American servicemen who used heroin in Vietnam gave up the drug on returning to the United States. See generally the extensive survey of the literature in J Orford, Excessive Appetites: A Psychological View of Addictions (Chichester New York: Wiley, 1985).

^[20] I Ehrlich, On the Usefulness of Controlling Individuals: An Economic Analysis of Rehabilitation, Incapacitation, and Deterrence (1981) 71 Am Econ Rev 307.

^[21] Motor Vehicle Manufacturers Association v State Farm Mutual Insurance Co [1983] USSC 148; (1983) 463 US 29.

^[22] Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co (1988) 78 ALR 407. The case has since been reversed by the High Court of Australia which did not discuss the line of argument advanced in the text. See [1989] HCA 6; (1989) 83 ALR 577; 63 ALJR 181.

^[23] United States v Aluminium Co of America (1945) 148 F 2d 416.

^[24] United States v Carolene Products Co (1938) 58 S Ct 978.

^[25] See generally BA Ackerman, Beyond Carolene Products (1985) 98 Harv L Rev 713. I put to one side the question whether the Supreme Court got anything right in Carolene Products. It quite obviously misunderstood the purpose and effects of the statute, uncritically treating it as public interest rather than private-interest-group legislation. See GP Miller, The True Story of Carolene Products [1987] Sup Ct Rev 397.

^[26] See FH Easterbrook, Statutes’ Domains (1983) 50 U Chi L Rev 533; FH Easterbrook, The Role of Original Intent in Statutory Construction (1988) 11 Harv JL & Pub Pol’y 59.

^[27] “Purchased” does not imply bribery or any other illegal act. It means organised political activity — contributions, turning out volunteers, monitoring public positions and engendering support, finding sympathetic persons to be candidates (even the most crass interest group wants true believers to be candidates; it is never wise politics to elect a cynic, whose loyalties may shift). The methods by which political support may be created are many. Clever interest groups use them all.

^[28] See Rodriguez v United States [1987] USSC 36; (1987) 107 S Ct 1391, at 1393; see also KMart Corp v Cartier Inc [1988] USSC 96; (1988) 108 S Ct 1811, at 1834–35 (Scalia J concurring and dissenting).

^[29] See BP Pashigian, Environmental Regulation: Whose Self-Interests Are Being Protected? (1985) 23 Econ Inq 551.

^[30] Northern Indiana Public Service Co v Colorado Westmoreland Inc (1987) 667 F Supp 613, affirmed [1988] USCA7 366; (1988) 845 F 2d 1024. See also BA Ackerman & WT Hassler, Clean Coal/Dirty Air (New Haven: Yale University Press, 1981).

^[31] See S Rose-Ackerman, Progressive Law and Economics — and the New Administrative Law (1988) 98 Yale LR 341.

^[32] It goes with the territory. See, in addition to any randomly selected essay by Holmes, RA Posner, The Jurisprudence of Skepticism (1988) 86 Mich L Rev 827.

^[33] See FH Easterbrook, Monopoly, Manipulation, and the Regulation of Futures Markets (1985) 59 1 Bus S103, S115–16, S121–23.

^[34] I collect a lot of the studies in FH Easterbrook, Workable Antitrust Policy (1986) 84 Mich L Rev 1696, at 1698 n 5.

^[35] One accessible summary of the data appears in J Pound, K Lehn & G Jarrell, Are Takeovers Hostile to Economic Performance? (1986) 10 Regulation (Sept/Oct) 25.

^[36] For example, FH Easterbrook & GA Jarrell, Do Targets Gain from Defeating Tender Offers? (1984) 59 NYUL Rev 277.

^[37] Although I do not want to demean analytical work, which has made great headway. See A Schwartz, The Fairness of Tender Offer Prices in Utilitarian Theory (1988) 17 J Legal Stud 165.

^[38] DG Davies, The Efficiency of Public versus Private Firms: The Case of Australia’s Two Airlines (1971) 14 JL & Econ 149.

^[39] Evaluating the safety effects of deregulation is a difficult task, since improvements in safety built in by the manufacturer are extrinsic to the way the planes are used. For the best available effort to separate one from the other, see McKenzie & Shughart, supra note 11.

^[40] Riley v National Federation of the Blind [1988] USSC 156; (1988) 108 S Ct 2667.

^[41] One wonders why, in other parts of the law, courts are so willing to indulge the assumption that consumers are hapless fodder for predatory businesses. But consistency is not to be expected, and the patterns of inconsistency are beyond the scope of this essay.

^[42] See RA Posner, The Constitution as an Economic Document (1987) 56 Geo Wash L Rev 4.

^[43] See Business Electronics Corp v Sharp Electronics Corp [1988] USSC 75; (1988) 108 S Ct 1515, at 1523–24.

^[44] American Textile Manufacturers Institute v Donovan [1981] USSC 147; (1981) 452 US 490.

^[45] Erie Railroad Co & Tompkins [1938] USSC 94; (1938) 304 US 64; See generally, TW Merrill, The Common Law Powers of the Federal Courts (1985) 52 U Chi L Rev 1.

^[46] See United States ex re1 Miller v Greer (1 986) [1986] USCA7 262; 789 F 2d 438, at 448–57, reversed [1987] USSC 156; (1987) 107 S Ct 3102.

^[47] Ricketts v Adamson [1987] USSC 129; (1987) 107 S Ct 2680; Shearson / American Express Inc v McMahon [1987] USSC 158; (1987) 107 S Ct 2332; Town of Newton v Rumery [1987] USSC 34; (1987) 107 S Ct 1187; Evans v Jeff D [1986] USSC 123; (1986) 475 US 717; Mitsubishi Motors Corp v Soler Chrysler-Plymouth Inc [1985] USSC 203; (1985) 473 US 614.

^[48] [1976] USSC 20; (1976) 424 US 319.

^[49] See FH Easterbrook, Substance and Due Process [l982] Sup Ct Rev 85, at 109–15.