“The Scholastic Foundation of Moral Reasoning about Economic Behavior”
by John D. Mueller
Panel M4.9 “Morality and Economics”
Association of Private Enterprise Economics (APEE) Annual Conference
Las Vegas, NV, 14 April 2014
My thanks to the Association of Private Enterprise Education (APEE) for making room in its 2014 annual conference for this timely panel on morality and economics. I’m especially grateful to Stephen J. Haessler for organizing and Dwight Lee of SMU for chairing the panel. I look forward to the discussion with our distinguished fellow panelists, David Rose, Loreen Lephardt, and Emily Lechtenberg. Steve organized and Dwight participated in a panel on my book, Redeeming Economics, at last year’s APEE conference, which was edited by Ross Emmett and published along with another panel at the History of Economics Society’s 2011 annual conference, in journal Research in the History of Economic Thought and Methodology (Biddle and Emmett 2013).
I’d like to extend the analysis in that book to outline “The Scholastic Foundation of Moral Reasoning about Economic Behavior.”
Dan Hammond of Wake Forest University, who chaired the HES panel, opened a review by remarking, “Redeeming Economics is likely to be ignored by economists”; but then gave this admirably succinct summary:
What, according to Mueller, is wrong with economics? In the simplest terms familiar to economists, there is an “equation” missing from the model. Economists have “equations” for production, consumption, and exchange, but not for the primary economic choice – the choices of persons. This missing element is a theory of distribution. Mueller argues that Thomas Aquinas had a complete economic model, with all four elements. Drawing on Aristotle and Augustine, Aquinas’s economics explained production (what is produced and how), consumption (utility), exchange (commutative justice), and distribution (production or purchase for whom).
Adam Smith, the father of modern (classical) economics, dropped two of the four equations, those for consumption and distribution. With neoclassical economics the equation for consumption was restored. But neoclassical theory has nothing to say about distribution, leaving the restoration of economics incomplete. Mueller sees this reconstruction (redemption) continuing with his book, with the efforts of other nascent neo-Scholastics and, he predicts, eventually by the profession at large. Thus Mueller himself does not think his book will be ignored. Or perhaps, if it is ignored the deficiency of economics will become evident to practitioners from their experiences doing economics. Mueller expects that economists will find their way to a neo-Scholastic economics that will preserve the best of both Smith and the neoclassicals, while restoring the theory of distribution. (Hammond 2012, 73).
Mine is the latter view: that “the deficiency of economics will become evident to practitioners from their experiences doing economics” so that “economists will find their way to a neo-Scholastic economics that will preserve the best of both Smith and the neoclassicals, while restoring the theory of distribution.”[i] In fact, I think this problem will be evident in our panel’s discussion. The technical problem is that with fewer equations than variables to be explained, the classical and neoclassical systems are “underdetermined,” thus requiring economists to adopt circular logic or empirically false assumptions (or both). I realized this by the accident of becoming a financial market forecaster, which requires spelling everything out verifiably.
But in terms that most ordinary people can understand, the scholastic, classical, and neoclassical systems presuppose three different views of human and divine nature, differing on whether man and God have free will.
Since Adam Smith essentially “de-Augustinized” economics, it’s important to understand Augustine’s theories of benevolence and beneficence, which Aquinas integrated within the scholastic natural law moral philosophy and the economic theory which prevailed for five centuries before Smith.[ii]
Both Augustine’s anthropology and theology had started from Aristotle’s insight that “every agent acts for an end”[iii] and Aristotle’s definition of love—willing some good to some person.[iv] But Augustine drew an insight that Aristotle had not: every person always acts for the sake of some person(s). For example, when I say, “I love vanilla ice cream,” I really mean that I love myself and use (consume) vanilla ice cream to express that love (and in preference, say, to strawberry ice cream or Brussels sprouts, which order reflects my separate scale of utility).[v]
So Augustine’s crucial insight is that we humans always act on not one but two scales of preference—one for persons as ends and the other for other things as means: personal love and utility, respectively. And we express our preferences for persons with two kinds of external acts, “sale or gift.”[vi] Generally speaking, we give our wealth without compensation to people we particularly love,[vii] and sell it to people we don’t, in order to provide for those we do love.[viii] Since it’s always possible to avoid depriving others of their own goods, this is the bare minimum of love expressed as benevolence or goodwill and the measure of what Aristotle called “justice in exchange.”[ix] But our positive self-love is expressed by the utility of the goods we provide ourselves, and our positive love of others with beneficence: gifts. Conversely inner hate or malevolence is outwardly expressed by the opposite of a gift: maleficence, or crime.
This understanding of economics entails an alternate view of the history of economics; hence my book begins with a “Brief Structural History of Economics,” which describes and distinguishes the scholastic, classical, and neoclassical theories, as well as the incipient “Neoscholastic” school.[x]
But the same analysis distinguishes among schools, such as the Austrian, British, Walrasian, Distributist or “neo-Thomist” schools, and even among individual economists.[xi]
Aristotle had bisected moral philosophy into ethics and politics. But scholastic philosophy and economic theory follow Thomas Aquinas by re-dividing them into three parts[xii]: Hence the middle three sections of Redeeming Economics are devoted to personal, domestic, and political prudence, or “economy.”
In each of these chapters, after re-stating and updating the scholastic economic theory, I focus on a salient practical application in which neoclassical and neoscholastic economics reach divergent empirical predictions–for example, disproving the famous claim by economist Steven D. Levitt, featured in Freakonomics, that the U.S. Supreme Court’s legalization of abortion in 1973 caused the crime rate to fall 15-20 years later, by eliminating potential criminals (Levitt and Dubner 2005, 117-144).
Actually, as I show, there is a 90% current, inverse relation between “economic fatherhood” and homicide. So legalizing abortion raised crime rates immediately and with a lag.
The final chapter concerns “divine economy,” which was Aristotle’s name for metaphysics. Three alternate world views are presupposed by (Neo-) Scholastic, classical, and neoclassical economics: biblically orthodox natural law, the Stoic and Epicurean philosophies, respectively. Though differing ultimately about immaterial realities–the existence or nature of God or the soul–these three views lead to starkly different behavior among people and starkly different predictions by economists.
In (neo-) scholastic natural law, economics is a theory of rational providence, describing how we “rational,” “matrimonial,” and “political animals” choose both persons as “ends” (which we express by our personal and collective gifts) and the scarce means to be used (consumed) by or for those persons, which we make real through production and exchange.
By dropping both distribution (the choice of persons as ends) and consumption (the choice of other things as means), Smith expressed the Stoic pantheism that viewed the universe “to be itself a Divinity, an Animal” with God as its immanent soul, so that sentimental humans choose neither ends nor means rationally; instead, “every individual…intends only his own gain…and is led by an invisible hand to promote an end which was no part of his intention.” As Larry White suggested last night in his remarks after receiving the Adam Smth Award, Smith really did view God as a sort of cosmic puppetmaster.
By restoring utility (the choice of means) but not distribution (the choice of persons as ends), neoclassical economics expressed the Epicurean materialism that claims humans somehow evolved as merely clever animals, highly adept at calculating means but having no choice other than self-gratification, since “reason is, and ought only to be, the slave of the passions,” as Hume put it. ]
I suspect–I hope–that we will disagree at least in part. But I think the model for disagreement was well expressed by Thomas Aquinas: “We must respect both parties, namely, those whose opinion we follow, and those whose opinion we reject. For both have diligently sought the truth and have aided us in this matter.”[xiii]
John D. Mueller response
“The Real Empathy Problem”
Let me briefly comment on the presentations by the other panelists. Steve Haessler, Loreen Lephardt, and Emily Lechtenberg have presented what promises to be an interesting survey of attitudes among Catholics about their understanding of Catholic social doctrine (which, I should note, is based on the scholastic economic theory that I outlined earlier).
My main suggestion is that, rather than soliciting the survey participants’ subjective self-descriptions about the degree to which they are “religious,” they should be asked how frequently they participated in religious services within the past month. Though there is a well-documented tendency for Americans to overstate the frequency of their worship (Brenner 2011, Hadaway et al. 1993, and Presser and Stinsom 1998), measured with care the rate of weekly worship is still an objective fact, which has a high correlation with many other kinds of behavior.
Next, I would like to summarize the basic disagreement between the “neoscholastic” economic theory I have outlined and Prof. Rose’s ingenious effort to rest the “moral foundation of economics” on one version of neoclassical economic theory. In doing so, I’m reminded of the police sergeant in The Pirates of Penzance, who sang, “perhaps it would be wise not to carp or criticize, for it’s very evident these attentions are well meant.”
We start in agreement at least that economics rests on a moral foundation. But I think Prof. Rose needs to be more specific about what is meant by “morality.” I suggest that morality approximates “interpersonal relations,”[xiv] which as I explained in my opening remarks, neoclassical economics is unable adequately to explain, because it has no distribution function, which expresses our preferences for persons.
Prof. Rose’s statement must also be corrected on some historical points, such as his assertions that Adam Smith was the first to observe that the gains from exchange depend on the extent of the division of labor (The theory originated at least two millennia earlier with Plato’s Republic (Plato 1969 [380 B.C.], 2.369a & 2.369b, as noted by Evers 1980, 45), or that Friedrich Hayek was the first to explain that the advantages of private over socialized property arise from localized knowledge. Aquinas (1982 ), had done so seven centuries before Hayek.[xv]
But what does it matter, it might be asked, if we get the history of economic theory wrong, as long as we get the modern theory right? After all, in his ultimately successful campaign to end the history of economic thought requirement at American university economics departments, George Stigler (1969) had argued that “all that is useful and valid in earlier work is present—in purer and more elegant form—in the modern theory” and that “the history of the discipline is best left to those underendowed for fully professional work at the modern level.”
The contention of my opening remarks and of my book is that Stigler was wrong, and that getting the history of economic theory wrong inevitably leads to inferior modern theory.
This brings us to Prof. Rose’s essentially neoclassical “moral foundation of economics.” He says in his book (citing Bernard Gert) that “it is far more important to avoid behaving in evil ways than it is to behave benevolently” (152-3). I believe that he means “beneficently.” Like Adam Smith, Prof. Rose used the terms “benevolence” and “beneficence” almost interchangeably. But as I pointed out in my opening remarks, the meaning of benevolence or “goodwill” is precisely to affirm the good that, say, Prof. Rose exists and enjoys his goods, even if I do not myself provide any of those goods, (which would be beneficence, or “doing good”). This is why I may not exert the maleficent effort to deprive him of those goods.
Professor Rose’s book begins with what he calls a “thought experiment launched by the following question: If a society’s sole objective is to maximize general prosperity and it can choose its own moral beliefs, what kinds of moral beliefs would it choose?” (Rose 2011, 13-14)
I for one dissent from Prof. Rose’s essentially normative assumption. For as Aquinas noted:
If such an ultimate end either of an individual man or a multitude were a corporeal one, namely life and health of body, to govern would then be a physician’s charge. If that ultimate end were an abundance of wealth, then knowledge of economics would have the last word in the community’s government. If the good of the knowledge of truth were of such a kind that the multitude might attain it, the king would have to be a teacher. It is, however, clear that the end of a multitude gathered together is to live virtuously. For men form a group for the purpose of living well together, a thing which the individual man living alone could not attain, and good life is virtuous life. Therefore, virtuous life is the end for which men gather together…. Yet through virtuous living man is further ordained to a higher end, which consists in the enjoyment of God, as we have said above. Consequently, since society must have the same end as the individual man, it is not the ultimate end of an assembled multitude to live virtuously, but through virtuous living to attain to the possession of God.” (Aquinas 1982 , ò106-107)
Thus Professor Rose and I disagree about the highest end for an individual human or for a human community. Prof. Rose posits a society of whose “sole objective is to maximize general prosperity.” But Augustine had sliced through such notions. A miser is said to love money as his highest good, noted Augustine—yet he still parts with it to buy bread to continue living, thus showing that his deepest motive is love of self, not money (Augustine 396/397, Ch. 25).
Yet the real difficulty with Professor Rose’s theory is positive, not normative: the neoclassical economic theory it presumes cannot adequately describe either his or my normative goal. For Professor Rose goes on to make the further assertion, that Augustine’s theory of the asymmetry between benevolence and beneficence (with which Prof. Rose agrees) is equivalent to saying that “we have a great ability to reduce others’ utility, but we have comparatively little ability to increase others’ utility” (152-3).
Consider Prof. Rose’s attempt to describe interpersonal relations by using a utility function but no distribution function:
B considers undertaking an opportunistic action (x1) that promotes his welfare at the expense of A. Because we assume that B has a capacity for empathy, B understands that action x1 will harm A and forms an expectation of the effect of the action. In B’s estimation, A’s harm will be E[ΔUA(x1)] (the perceived harm). If B cares about A’s welfare, then θA > 0 and the reduction in A’s welfare will also produce a reduction in B’s welfare equal to θA E[ΔUA(x1)] (the relevant perceived harm). (Rose 2011, 104)
To which the obvious response is “prove it.” He can’t. Prof. Rose says that his basic insight is what he calls a “thought experiment,” leaving empirical testing to the future. But I maintain that he cannot test, let alone prove the statement I just quoted, for at least two reasons.
First, Prof. Rose’s theory is unscientific for the reasons Lionel Robbins outlined in 1932 while criticizing the early neoclassical welfare theorists of the British School: the theory is not independently verifiable (or falsifiable), because “there is no means of testing the magnitude of A’s satisfaction compared with B’s…. It cannot be justified by appeal to any kind of positive science.” (Robbins 1932; quotation is from 1935 second edition).
In other words, how can A (David Rose) experience B’s (John Mueller’s) utility–in his conception, feel my feelings? He can’t. What’s worse, suppose I am B in Prof. Rose’s example, and I happen to care about A, David Rose: This means that our utility functions are interdependent. So A (David Rose) must not only experience B’s (John Mueller’s) utility (interpreted, again, as feeling my feelings) but A also experiences B experiencing A’s utility from experiencing B’s utility–in short, an infinite regress. Now, I grew up in a family of 14, in which case the relevant measure of what Gary Becker called “social income” is anywhere between 1 and 14 times actually observable income. That way lies madness–but not economics. The situation is simply not modelable.
The second point is that these difficulties arise because David Rose’s neoclassical model is radically “underdetermined,” thanks to omission of Augustine’s distribution function: there are more unknown variables to be explained than explanatory equations. Lionel Robbins’s objection applies to the British School’s notion of cardinal utility–utility interpreted as a thing. But it does not apply to the Austrian or Walrasian schools, which understand that utility is essentially an order of preference, not a thing. But these other schools do share the debility that they are underdetermined.
This is why Prof. Rose must either substitute assumptions about my behavior for my actual behavior (in which case his predictions are liable to be empirically false); or his predictions will be unfalsifiable, as in the case of measuring interacting utility functions in an altruistic family of 14.
The logically consistent and empirically testable (neo-)Scholastic hypothesis is that crime, like love, is essentially not a weighing of utilities but a weighing of persons: this, as I suggested, is what makes it a moral decision.[xvi] In the (neo-)scholastic economic theory, each person experiences only his own utility, not someone else’s, thus preventing multiple counting of the same utility or income. Just as love means distributing some good to some person and selfishness means distributing all of one’s goods to oneself (giving everyone else a zero significance in that distribution), a crime consists in depriving some person of a good that belongs to him or her—giving that person a negative significance in the distribution of goods. (Mueller 2010, 109-110).[xvii] This view is empirically testable (or falsifiable), and unlike Prof. Rose’s neoclassical model, the neoscholastic theory withstands that test.
Thus I conclude with the claim that if Professor Rose (or anyone else) wishes to discover the “moral foundation of economic behavior,” he must employ an updated version of “the scholastic foundation of moral reasoning about economic behavior.”
[i] I use the term “redeem” in the sense of “fulfill (an earlier promise or pledge).” If, as I believe, the next phase in economics is Neoscholastic, it will have fulfilled its original promise.
[ii] In the book I typically use the term “final distribution,” to distinguish it from “distribution” as the term has been used since Adam Smith. The original scholastic theory of distribution comprises Augustine’s theory of personal distribution–gifts and their opposite, crimes–and in every social community (like a family or political community), what Aristotle called “distributive justice.” Smith conflated what is more properly called “compensation” or “justice in exchange” with distribution properly so called, by introducing the assumption that “every individual … intends only his own gain.” (Smith 1966 (1776), Wealth of Nations, IV.ii.9, accessed on 19 September 2009 from http://www.econlib.org/library/Smith/smWN13.html#IV.2.9. I recount Smith’s oversimplification of the scholastic economic theory he had been taught by his teacher Frances Hutcheson in Mueller 2010 chapter 3.
[iii] This concise formulation seems to be that of Aquinas, paraphrasing Aristotle’s Physics, ii, 5: Summa 1-II Q1. A2: http://www.newadvent.org/summa/2001.htm; cf. http://classics.mit.edu/Aristotle/physics.2.ii.html.
[iv] Aristotle (1932 [335-322 B.C.]), p. 2, p. 4, pp. 102-103.
[v] Augustine also introduced the important distinction between “private” goods like bread, which inherently only one person at a time can consume, and “public” goods (like national defense, enforcement of justice, or even this panel) which, at least within certain limits, many people can simultaneously enjoy, because they are not “diminished by being shared” (i.e., scarce) Augustine (395-396), viii, 19, p. 146. Private goods are now sometimes called “rival” goods. The formulation “diminished by being shared” is from Augustine (396/397), I, 2.
[vi] Augustine (395/396), p. 131.
[vii] To be more precise, love with both benevolence and beneficence.
[viii] Or rather, love only with benevolence but not beneficence.
[ix] Aristotle 1954 [c. 350 BC], V, v; pp. 117-122.
[x] The simplified version merely lists “yes” or “no” to denote the presence or absence of each fundamental element of economics. The three schools of neoclassical economics originated of course with Jevons 1871, Menger 1976 , and Walras 1954 ).
I presented the (neo-)scholastic system and contrasted it with subsequent classical and neoclassical revisions in Redeeming Economics (Mueller 2010) as follows (pp. 375, 400, 416):
(1) CKi + CLi = YiDii/SDij [final distribution function],
where CKi, and CLi represent the use (“consumption”) by Person i of the services of his or her human capital, Li, and nonhuman capital, Ki; Yi is total compensation (labor and property income) of Person i; Dii is the significance of Person i to himself, and SDij is the significance of all persons to Person i.
For clarity and simplicity later on, we will define
(5) Yi = rKi+ wLi
meaning that Yi is the total net factor compensation (labor and property income) of Person i; and
(6) Ti = Yi – YiDii/dDij.
By substituting (5) and (6), (1) may therefore be restated as
(1a) CKi, + CLi = Yi – Ti.
This makes clear that the difference between Person i’s total consumption, CKi + CLi, and total compensation, Yi, is equal to Ti—(net) personal, domestic, and political “transfer payments” from Person i to other persons. Transfer payments comprise any income not received as compensation for contributing to current production. “Net” means that personal gifts made are offset by gifts received, while taxes are treated as transfers paid to the government and balanced against government transfers received.
For a purely selfish person, the distributive share Dii/SDij is 100 percent; for a person who makes gifts to others, less than 100 percent; for a criminal, more than 100 percent; and for the victim of crime (or abortion), less than zero percent.
(2) Ui = f(CKi, CLi) [utility function],
where Ui is the ranking by Person i (“utility”) of CKi, and CLi, the units consumed in use by Person i of the services of his or her nonhuman goods, Ki, and human capital, Li, respectively. In reality, K and L are not two goods but two classes of goods consumed: (K1, K2, . . . ,Kn) and (L1, L2, . . . ,Ln). Scarcity implies that the value of each unit consumed declines as the number of units increases (dU/dC<0: “declining marginal utility”) and that goods are “used up”—that is, rendered unusable—by consumption (for example, CKi = -DKi).
(3a) dKi = f1(Ki, Li) [production function for nonhuman capital];
(3b) dLi = f2(Ki, Li) [production function for human capital];
where DKi is the change in the stock (production) of nonhuman goods and dLi the change in the stock of “human capital,” owned by Person i.
(4) PKdKi+PLdLi = rKi+wLi, where PK and PL are the unit prices of K and L, respectively, w labor compensation per unit of L, r property compensation per unit of K. (PL is a market price only in a slave-owning society, like ancient Athens or the antebellum American South.)
To summarize: The neoclassical economists restored the utility function (equation  above). They restored the two-factor production function (3a) and (3b). But until about 1960, they interpreted both human and nonhuman capital as being limited to tangible factors. The neoclassical economists followed Adam Smith in ignoring the distribution function in theory, but in practice they have assumed that everyone is purely selfish, thus adding the restrictive assumption Dii/SDij = 1. As with Adam Smith, this special assumption collapses equation (1) into:
(1b) CKi +CLi = Yi.
It also means of course that there are no personal gifts, crimes, common goods, or distributive justice:
(6a) Ti = 0.
[xi] For example, I note my own migration from the Chicago School as of Mueller 1996 to the Neo-Scholastic School in Mueller 2010.
[xii] Aquinas, T. (1981 [1265-72]) II-II Q47 A11 contra and corpus, http://www.newadvent.org/summa/3047.htm#article11, and Q50 A3, http://www.newadvent.org/summa/3050.htm.
[xiii] Aquinas paraphrasing Aristotle in his Commentary on Aristotle’s Metaphysics 1073a14-1073b17 (tr. By Rowan)
[xiv] If we accept that God is a person (or, in the Christian view three persons in one God), this definition covers both economic and religious behavior.
[xv] Aristotle had pointed out two general advantages of private property over Plato’s proposed communal ownership of all property: greater social peace and productivity. Aristotle (1962 [c. 350 BC], Politics II, 5; 63.) Aquinas added the third, greater order resulting from the efficient use of specialized knowledge. Aquinas (1981 [1267-72]. II-II Q66 A2, retrieved from http://www.newadvent.org/summa/3066.htm#article2 on 27 January 2010.
[xvi] Crime, or any unjust act, is the reverse of love. Rather than a gift or voluntary transfer payment, it is an involuntary transfer payment exacted from its rightful owner. In both cases, the motivation of the transfer depends essentially on a weighing of persons, not a weighing of utilities. In gifts (voluntary transfers), the significance of the other person is either positive (for someone who receives a gift) or zero (for someone who doesn’t). In the case of a crime, the criminal gives himself a positive and the victim a negative significance. If I take what belongs to you against your will, I am giving myself a positive significance in a distribution that exceeds 100 percent of my own resources and giving you a negative significance in the “distribution.” I may take something from you, or I may destroy something belonging to you. Just as loving one other person half as much as oneself is mathematically equivalent to loving one-and-a-half persons equally, increasing one’s wealth by half through stealing from another person is mathematically equivalent to loving “two-thirds of a person” equally with oneself. But the number of persons loved is always greater than zero, because one always loves oneself. Note that this analysis provides an objective basis for defining a crime, one that is unaffected by whether the crime (for example, slavery) is made legal by positive human law. In fact, it provides a formula for measuring the economic harm done by crime or exploitation.
Thus, it is clear that pure selfishness, which has been the assumption of classical and neoclassical economists since Adam Smith, is just that—an assumption, and in fact a special case. It is a single point on a continuum (depicted in the chart “Augustine’s Distribution Function” that ranges from great generosity to others to great crimes against others (Mueller 2010, 144-145).