Tuesday, May 26, 2020

A New Buchanan/Dorf Paper Explains Why We Don't See More Conflict Between Law & Economics and Originalism/Textualism

by Michael C. Dorf

In my Memorial Day essay yesterday, I approvingly referred to two recent criticisms of economists by Professor Buchanan (here and here). Readers may be wondering whether we're done criticizing mainstream economics. If so, wonder no more. We're not. Professor Buchanan (who is himself an economist by training) and I have an article forthcoming in the Cornell Law Review that explains why Law & Economics (L&E) is bunk even when it is not being invoked to promote policies that will kill tens of thousands of people.

In A Tale of Two Formalisms: How Law and Economics Mirrors Originalism and Textualism, we look at L&E alongside the other leading conservative approach to law: originalism in constitutional interpretation and textualism in statutory interpretation (collectively O&T). We address a puzzle: Given that L&E and O&T are grounded in different frameworks and prescribe different methodologies, why do they so rarely result in open conflict? The answer: Because they're both so open-ended as to work as a mask for whatever ideological priors a judge holds, and the particular people with those priors are on the political right.

That's the extremely short version. Here's the merely very short version, that is, the abstract:
Two leading schools of thought among U.S. conservative legal elites— Law and Economics (L&E) and Originalism and Textualism (O&T) —both purport to use their formalist structures to guide analysis in ways that are objective, substantially determinate, and apolitical. Because they rest on very different theoretical underpinnings, L&E and O&T should only randomly reach similar policy or legal conclusions. After all, L&E implements neoclassical economics, a theory of utility maximization, whereas O&T is a theory of semantics. Yet as practiced, L&E and O&T rarely result in conflict. What explains the missing intra-conservative clash? Despite their respective pretenses to objectivity, determinacy, and political neutrality, neither theory delivers on its promises. Economic efficiency, the linchpin of L&E, is incoherent because it relies on typically hidden but ultimately normative assumptions about preferences that would exist in an impossible world without law. O&T as it has been refined in response to devastating criticisms of earlier versions is indistinguishable from ostensibly less determinate rivals like Living Constitutionalism and purposivism. Accordingly, conservatives use L&E and O&T to obscure the role of normative priors, perhaps even from themselves. Liberals could use the same techniques for different results but heretofore generally have not, instead mostly settling for counterpunching against charges of result-orientation.
As we explain in the paper, the under-determinacy of O&T should be familiar to law-literate audiences from over a century of legal realism, but our particular critique of L&E will likely be unfamiliar to most readers. Our critique goes at neoclassical economics, which is the economics that L&E applies, and in particular at the notion of economic efficiency when used as a prescriptive standard of evaluation. We do not argue (although we agree with the objection) that neoclassical economics undervalues distributional concerns. Rather, our critique aims at efficiency itself.

As we explain at length in Part II of the article, to call some state of the world "efficient" means that supply and demand are in equilibrium (at the point at which the supply and demand curves intersect in any given market). The normative move is then to regard governmental measures that change that equilibrium as "distortions" that create waste. Our key point in Part II is that supply and demand could be in equilibrium at multiple points, depending on what one takes for granted and what one treats as a distortion. Here's a crucial piece of the analysis:
If the efficient quantity in each market is determined by where supply and demand curves cross, then it becomes rather important to know what determines the position of those curves. Demanders have money to buy things, preferences about what they like and dislike, and so on. Suppliers have access to technology, machinery, labor, and so on. Where do those underlying determinants of demand and supply come from?

If Jane currently has one million dollars in wealth, a salary of $100,000 per year, and access to many alternatives to any given good, she will decide how much of each good to buy. But why do we take those facts to be the natural baseline? If it turns out that Jane stole her wealth from Calvin, we are left with two options: first, we can say that however Jane came to possess a million dollars, she does possess it, so her demand curve is correctly measured using that baseline; or second, we could say that the correct baseline is to return the million dollars to Calvin, the rightful owner, changing both Jane’s and Calvin’s demands (in ways that cannot be presumed to offset each other).

Crucially, the quantity sold in those two situations will differ (except in the truly fortuitous case where the market demand curve changes in perfectly offsetting ways after proper ownership is restored). Which one is efficient? Both are equilibria based on supply and demand curves, but the equilibrium quantities are not the same.
Meanwhile, as we go on to argue, there is no equilibrium that reflects government inaction because economic transactions only occur against a background of laws governing contracts, property, crime, etc. Furthermore, as we explain with numerous examples,
there is no single set of laws that constitutes the bare minimum, laissez-faire approach to governing. Given that there is no bare minimum, different sets of seemingly minimalist laws are defensible as the baseline, and each set will generate its own unique supply-and-demand curves for goods and services. A society with no inheritance tax will produce more yachts than one with a substantial such tax; a society with slavery will produce more manacles than one without it; etc. Every baseline looks inefficient from the perspective of any other baseline, and every set of laws can be a baseline that tautologically determines that it is efficient. 
Accordingly, to say that a state of the world is "efficient" is essentially empty. Yet such declarations underlie nearly every conservative prescription offered in the name of L&E.

I won't attempt to summarize the balance of the article, except to say that we note both interesting parallels between L&E and O&T and at least one important difference. The difference is that while indeterminacy is baked into the efficiency concept and thus L&E, O&T are only contingently under-determinate. We are not linguistic nihilists. We acknowledge that law has constraining force. Our core point regarding O&T is that legal texts lack anything like the force that formalists claim in contested cases in courts of last resort like the US Supreme Court. Thus, as a practical matter, L&E and O&T are sufficiently indeterminate to give cover to a jurist who wants to justify any of a wide range of results in seemingly neutral and apolitical terms.

The last part of our article explores the question why liberals and progressives have for the most part not tried to play the same game. We note that the various efforts to use L&E or O&T for liberal or progressive ends typically include an acknowledgment of the substantial role of values in the process. We float a number of hypotheses for the asymmetry, including that our side is simply more honest (a claim we don't consider ourselves well positioned to evaluate) and the possibility that both L&E and O&T have at this point taken on an almost tribal significance, so that even if they can be used for liberal and progressive ends, liberals and progressives find them too distasteful to employ.

So much for the summary. Prof Buchanan will have a follow-up essay here on Thursday addressing an issue that arose in one of the workshops at which we presented a draft of the article.

8 comments:

Steve Davis said...

Proofreading comment: In the abstract to the article, I think you mean to say "world impossible" rather than "impossible world." Thus "a world impossible without law." "Without law" modifies "impossible," not "world."

Steve Davis said...


I read the article and I'm curious why you describe law and economics as a formalist doctrine. I've always thought of it as exactly the opposite of that. It's a form of legal realism, in that it describes how judges do, in fact, rule as though they are abiding by a set of substantive norms, and L&E advocates believe the fundamental norm is economic efficiency. That in a nutshell is what Posner says. You don't believe Posner is a formalist, do you? He doesn't describe himself that way -- in fact, his recent characterization of his jurisprudence, basically "Do the right thing" is the opposite of that -- and I don't see him being described that way. Do any L&E advocates actually describe themselves as formalists? Do their critics, other than you? You don't explain, as far as I can tell, why you regard this as a formalist doctrine, and why it should be seen this way, and it seems like you should. Some justification for this term is warranted here.

I think your article is somewhat under-sourced, and as a result you open yourselves to the criticism that you are mischaracterizing those you criticize, and perhaps setting up a straw man. For example, footnote 20 is supposed to support your claim that L&E advocates see efficiency as an "objective and scientific concept" -- but the citation is to a critic of L&E, not an advocate. Do any advocates actually say this? If not then this is an unfair characterization. You cite no valid source for this statement. It may or may not be true -- I don't know -- but your article doesn't SHOW that it's true, and it should.

Frank Willa said...

Professors, thank you for these posts. Unraveling the 'myth of correctness' is important as the 'school' of economics (U.of C.,to refer to one) has dominated so much of national policy to the detriment of so many. These approaches are like 'railroad tracks of ideologues'- if one doesn't fit, then use the other- they both go to the same place.
This discussion of L&E brought to mind the Judge Learned Hand 'tort test' formulation (PL>B, or which ever version); and its 'application' in the Coase Theorem.
It seems to me that this L&E application inverts and contorts the original formulation. Judge Hand was trying to assess the 'duty of care', to apply it to answer whether liability attached. This is an after the fact, or event, of harm that has occurred. Did the actor factor the potential gravity of harm, and weigh it appropriately before deciding what to do? So, when the fact finder is determining the duty owed- the requisite standard of care, the duty is less if a minor harm, but is more if a major harm.
The 'L&E' is to instead use this as a before the fact tool to see if it is 'worth it to risk' an undertaking- a la is this a good gamble? (termed as efficiency analysis)
So, the Hand test is about the tort victim- how much duty is owed, and the L&E view is about the tort feasor- what is the cost/benefit risk to me.


Spoiler alert: I was unable to post to the Professor Dorf May 18 post so here is my comment, if any one is interested.

Thank you; when I heard Sekulow I was taken aback by its stark singleness- the live audio is helpful- I hope it continues.
As you point out it is not at all so simple. I would look to - not to be too originalist- at the word 'vest'. In a current use, I note that in some jurisdictions 'vest' is a temporary event that falls to a position and is not co-extensive with an individual. Real Property is 'vested' in an estate administrator, before, it is 'vested' in an heir. It is not 'owned' by the administrator, rather the role is fiduciary in nature, and temporary in term.
Further, the founders were well aware of the English Civil Wars and did not want an executive that was coextensive with the individual. (See 'Lawfare', Oct 3, 2019 for a discussion)

Michael C. Dorf said...

Steve: Thanks for the thought on the typo. I get what you're saying, but I think it's

As for formalism: The reason we call this a tale of TWO formalisms is that economic formalism is quite different from legal formalism. That's why Posner could be a critic of legal formalism. (An aside about Posner: By the end of his career, he was a simple all-things-considered pragmatist.) There is an empirical version of economics, but neoclassical economics--which, we make clear, is the brand we critique-- is so formalistic that it has become the butt of jokes for that reason ("first, assume a hammer"). What we mean by saying that economics is a brand of formalism is that it is based on formal inferences. An econometrician who wants to predict the impact of, say, raising the minimum wage on employment, will look at data. A neoclassical economist will draw supply-and-demand curves.

But there's no harm in our adding a footnote to that effect to explain why and how we (in my view uncontroversially) regard neoclassical economics and its concept of efficiency as formalistic.

As to sourcing, I'll leave to the law review editors the decision whether we need more. My view is that law review articles contain too high a ratio of citations to analysis in general.

Joe said...

As to the third comment, the last point also brings to mind this:

https://www.theatlantic.com/ideas/archive/2020/05/nondelegation-doctrine-orliginalism/612013/

Neil H. Buchanan said...

This is in response to Steve Davis's second comment. in addition to Mike's response, with which I fully agree:

(1) Thank you, because this comment makes one of our key points, which is that O&T formalism is -- or at least ought to be -- quite different from L&E formalism. Even if one rejects the term formalist to describe L&E, our point is not about labeling but that O&T and L&E should not align, but their adherents make them align. So again, thank you for that.

(2) Although fn20 quotes Purdy's critique of Posner's foundational work in L&E, Pudy quotes Posner extensively and faithfully. Within the parenthetical, the "acid bath" metaphor is Posner's and in quotation marks. If a law review editor concludes that Purdy misrepresented Posner, I'll be surprised but will happily provide other sources as needed.

More to the point, fn20 supports this sentence: "The second claim is that efficiency is an objective and scientific concept, thus inoculating L&E adherents against the tendency to substitute personal normative priors for cold-blooded logic." I would be absolutely delighted if an adherent of L&E were to renounce the assertion that efficiency is objective and scientific. If they actually did that, they'd undermine our characterizations in the paper, but our larger goal -- that everyone should stop pretending to be objective and argue our points on equal ground -- would be a big step closer to reality.

And for that very reason, I'm not holding my breath.

Nico Banks said...

A Tale of Two Formalisms powerfully critiques any economist who believes that neoclassical economic theory is sufficient to determine the just distribution of goods and services in a society. As the article rightly argues, any neoclassical model that purports to calculate a just macroeconomic equilibrium--meaning the just distribution of goods and services for a whole society--must make assumptions about what the law should be, and those assumptions will be question begging unless they are informed by some actual moral theory (i.e., not neoclassical economics).

But L&E theorists are often concerned with marginal outcomes rather than calculating a macroeconomic equilibrium. When addressing marginal issues, an L&E theorist need not make normative assumptions about what property laws should be. Rather an L&E theorist can determine the efficient marginal outcome holding everything else--including the existence of current property laws--constant.

Do you agree that L&E is not subject to your critique when it is used for marginal decision making, rather that determining a macroeconomic equilibrium?

Here's an example, in case my argument was not clear in the abstract: suppose Judge is faced with the question of whether she should allow Buyer to buy Widget from Seller, when Buyer and Seller have both agreed on a price, but the price strikes Judge as grossly unfair. If Judge uses L&E analysis, it does not matter that Buyer and Seller would not have agreed to the deal under a different set of property laws that may have altered their wealth or willingness to pay. It also does not matter whether the existing property law regime is just. Rather, Judge would permit the transaction solely because, under existing current property laws, that marginal transaction makes both parties better off without making anyone else worse off (assuming that both parties agreed to the transaction rationally, which is an unrealistic but presently unattacked assumption).

Nico Banks said...

I just realized that someone could reply to my comment by saying “Nico... you can’t hold property laws ‘constant’ when you’re applying a theory whose purpose is to determine the property law.” Fair point. But, I still think that an L&E theorist does not need to make any assumptions about the justness of the property laws that lead to the present distribution of goods in order to argue “the present distribution of goods could be improved by allowing X and Y to enter into a mutually agreed upon transaction because it would make both of them better off without making anyone else worse off.” Thus, in the Jane hypothetical, an L&E theorist need not determine whether Jane acquired her wealth justly (e.g. by theft or not) in order to determine whether her use of her now-acquired wealth is efficient. Determining whether wealth was justly acquired and determining how it should be used in the future are two different questions. L&E is much better suited to answering the latter than the former.