Thursday, April 30, 2009

Letter to Barack # 17: A Life and Death Question of Methodology

A Life and Death Question of Methodology (Obama 17)

Dear President Obama,

The physical facts imply that it is necessary to reform our institutions. There is a threatening fact: The present relationship of our species to the biosphere is unsustainable. Buckminster Fuller and other scientists have affirmed a corresponding encouraging fact: There would be enough for everybody indefinitely if we (we humans) could learn to cooperate, to share and to use appropriate technologies. We need to learn to live simply. (See www.davidhilfiker.com). We need to learn to share. (“Poverty will end when we learn to share with the poor.” —Mother Teresa). We need to use ecological criteria when we invent and choose technologies.

Economics, and the set of institutions it studies and tends to justify, collides with the physical facts. It tends to tell us that we need to get back on the road to growth, that we need to restore consumer confidence so that consumers will buy more, and that we need to restore profitability so that investors will create more jobs. (I am paraphrasing your speeches, and I assume that in your speeches you paraphrase the advice your economics team gives you.) Economics tends to tell us that it is necessary to increase social inequality (for example formerly by means of the Bush tax cuts for people in upper income brackets, and today through reluctance to reverse them). Now more than ever, economics tends to say, at a time of economic crisis, it is necessary to reward the investing class because their expectations of profit are the sine qua non of growth, which is, in turn the sine qua non for at least trying to create that will-of-the-wisp full employment. It tells us that we need to make our automobile industry more competitive by dismissing the employees of unprofitable operations and cutting wages. From an ecological point of view, we really should be shutting the automobile industry down completely. From a human point of view, we should seek adequate incomes for all. As a half way measure we should at least be allowing only automobiles with technologies so green that they will reduce global warming. Economics tells us this even this half way measure is not now feasible because the green cars cannot be produced at affordable prices soon enough to create the mass buying needed to restart the economy.

In short, our actions are mainly driven by a logic that characterizes both the science of economics and the institutions it studies. What we should do to adjust to the physical reality we face, and what we should do to assure that every member of the human family is cared for, is overwhelmed by what we must to make our economy function. In this letter, I offer a few reflections on the present and past of economics. I am looking for ways to change it from a voice that clothes the imperatives implicit in our present institutions in the garb of science, to a voice for change.

Alan Krueger, a brilliant economist whom you recently appointed to an important post in the Treasury Department has written, “Early work in economics was primarily deductive” while the current trend is to work inductively. Today’s economists emphasize the importance of detailed empirical studies of facts. They are aware that an economic theory “is a model of reality, not reality itself.” The model is to be justified, if it is to be justified at all, by its usefulness in organizing facts. (See Krueger in Journal of Economics Education, volume 32)

Other members of your economics team, including Christina Romer and Larry Summers, have expressed similar views on methodology. Your team tends to agree with Krueger that new economics is better than old economics because it (today’s economics) is more inductive and less deductive. Krueger himself and his co-authors have shown that while deductions from the theory of supply and demand imply that raising the minimum wage will increase unemployment, inductive empirical studies show that a moderate increase of the minimum wage has no significant effect on the rate of unemployment.

We should be grateful to contemporary trends in economics for saving us from the sorts of confusion of mathematical models with real-world human behavior that led, for example, Leon Walras to write over and over again (and to put in italics) that under pure competition the satisfaction of every participant in the market is maximized. This sort of talk is tautology with no empirical content; it must be true because of the definitions used by the theorist and therefore it tells us nothing one way or the other about what happens in fact. Nevertheless, contemporary economists tend in practice, however aware they may be of the historical relativity of our institutions in principle, to naturalize our institutions and therefore to naturalize the logic that makes us an unsustainable species. They follow the bad example of John Stuart Mill by treating our socially created realities as if they were simply logical extensions of natural realities. (See the opening chapters of Mill’s Principles.) Consequently, they clothe the imperatives implicit in our present institutions in the garb of science more than they design constructive paradigm shifts.

The dichotomy “deductive vs. inductive,” or arbitrary model vs. insight into reality, or any dichotomy whatever, tends to blind us to multiple options, different interpretations, different ways of slicing reality into categories. Although all dichotomies reflect an unfortunate bias of the human brain in favor of dualities, deduction vs. induction is an especially fuzzy way to slice economic research into two categories, since—as logicians will tell you—there is not much difference between the two. Induction is deduction for which the premises are facts.

Nevertheless, Krueger’s meaning is roughly clear. He reflects a tendency among economists toward praising their current research practices as better than older ones because they are more parsimonious (i.e. recognize fewer entities) and more rigorous (i.e. more mathematical). Without using the word “inductive” we can use related although not identical terms and say that today economics is more empirical, or more econometric, or more positivistic. In the sense in which Paul Samuelson thought he was improving Alfred Marshall’s economic principles by restating them in mathematical terms in his Harvard doctoral dissertation, today’s research can be praised as more capable of stating hypotheses precisely and therefore more capable of testing them.

Any research methodology requires a series of choices. Whether the point is phrased in terms of one or another dichotomy, or even if it is phrased in a way that succeeds in categorizing the rivers of economic research flowing from universities and think tanks in non-dualistic terms, there are many alternatives to currently prestigious approaches. I do not propose to replace a tendency among your advisers and among mainstream economists generally to believe they have made the right methodological choices, with another equally overconfident tendency. But there is nonetheless one particular neo-institutionalist and critical realist alternative I do want to propose. It can be called a cultural structures approach.

I want to rescue some of the voluminous literature of the old-fashioned economics that was “mainly deductive.” Certain older theories legitimately claim to have insight into “reality,” as distinct from being “models” that make it a point “not to be reality.” An example is Marx’s theory of relations (Verhältnisse) of production, which is not a model but an economic theory resting on sociological and juridical foundations; that is to say, in the terms I propose, on cultural structures.

The approach I am recommending takes as first premises the constitutive rules that define the type of society we have, which Charles Taylor has characterized as a bargaining society –most notably the rules that constitute property and contracts. Vandana Shiva illustrates the fundamental character of constitutive rules dramatically when she describes how multinational companies in India promote the legal constituting of property rights in water, air, forests, traditional medical practices, and genetic codes in order to make it possible to sell them.

Unlike the older economics, which Krueger and others regard as surpassed by today’s superior parsimony and rigor, the cultural structures approach does not postulate non-physical entities (“metaphysical” entities in positivist nomenclature). For example, unlike Smith, Ricardo, and Marx it does not postulate that there is something called “value” that stands behind and causes prices. So far, we agree with the contemporary mainstream. However, unlike Milton Friedman and his followers it does not say that models used –to continue with the same example—to explain prices requires no insight into reality. Instead, like Ludwig von Mises, we think it very important to notice that a price is a contract. A price is an agreement between a buyer and a seller. Its historical conditions of possibility (to borrow a phrase from Michel Foucault) include the constitutive rules of property and contract.

A model may not be “reality itself,” but if it is a good model, or a good approach, it will not be a free-floating imaginary construction à la Friedman either. It will draw on insight into reality à la Lonergan, namely the insight that economic behavior is human behavior. In Wittgensteinian terms human behavior consists of language games. Alternatively, one can drop the idea of model altogether –since the very idea of model may connote an arbitrary construction—and begin with the premise that economic behavior is a subset of the human behavior studied by sociologists, anthropologists, and other scholars.

Contrary to what David Hume and his followers would have us believe, science advances when it achieves realistic insight into the causal powers that produce the phenomena under study. (Harré, Principles of Scientific Thinking). To continue with the same example, the study of prices advances when we observe that a price is a contract governed by ethical and legal norms. Similarly, chemistry advanced with Dalton and biology advanced with Darwin –not by running data through significance tests but by gaining insight into the causal powers of the mechanisms that produce the phenomena (mechanisms which correspond in economics to cultural structures, as I have shown in my book Understanding the Global Economy). Much of the “scientific method” studied in high schools and much of the “inferential statistics” studied in colleges reflects a superficial understanding of the natural sciences as if they were essentially about finding statistical regularities in the phenomena observed. (Compare, for example, the monetary histories of Friedman and Schwartz). On the basis of a superficial understanding social scientists are trained to do not what natural scientists historically have actually done to achieve insight into reality, but rather what a neo-Humean imagination imagines them to have done. (Harré.) Economics has been one of the fields most damaged by procedures that treat, for example, multiple regression analysis as a substitute for studies of human behavior. (Here “studies of human behavior” is meant as a generic term referring to several paradigms in social science I and other critical realists regard as more realistic than running datasets through multiple regressions, including Margaret Mead´s “customs,” Bourdieu´s “logic of practice,” Foucault´s “archaeologies” and “genealogies,” Glaser and Strauss´ “grounded theory,” Wallerstein´s world-systems approach, Goffman and Garfinkel´s micro-sociology, Patomaki´s version of post-international relations theory, Catherine Hopper´s “culturally determined behavior,” and Harré´s own anthropomorphic method. Among the more realistic mavericks in economics itself one would have to mention the German historical school, Veblen, Commons, Mitchell, Cyert and March, Drucker, Sen, feminist economists, and most Marxists.)

To the extent that economics simply accepts institutions here and now as natural and inevitable, then regardless of its mathematical sophistication and regardless of the sheer quantity of detailed empirical studies it produces, it is a “science” only in the sense that there is a “science of real estate finance,” a “science of tax return preparation,” “a science of life insurance” and a “science of banking.”

Grounding economics less ethnocentrically and more fundamentally in the basic social rules that constitute bargaining rescues some “mainly deductive” passages in older texts: Their general discussions of the human condition are not always the vague approximations of pioneers, waiting for their Samuelson to restate them in rigorous terms so that they can be empirically tested. They are often interpretations of cultural structures in historical evolution. When Adam Smith, for example, goes on and on about what is and is not natural, he is not a metaphysical essentialist; he is a liberal ethicist. Smith´s “natural justice” is precisely security of property and enforcement of contracts. Smith is quite clear that the political economy he describes is made possible by “civilization.” “Civilization” is the reign of the civil law protecting the rich against the poor that makes accumulation possible. The civil law was developed in early modern Europe drawing mainly on Roman jurisprudence, although Smith considered that China too was civilized in its own way. Smith always implicitly and frequently explicitly identifies the historical conditions of possibility of economics with the historical emergence of modern western institutions.

Now in 2009 when sustainability and social cohesion are at stake, and the odds are against them, humanity cannot afford to reject even illusions if the illusions will help to get it off the endangered species list. But dissolving what has hitherto been known as economics or political economy into the general study of human behavior requires no such pragmatic dishonesty. Realism is practical. In practice, we need realism. We need to adjust our behavior to reality. Our behavior is determined mainly by our institutions. Therefore, we need to change our institutions.

A neo-institututionalist reconfiguration of what is now known as economics, bringing it into closer touch with its own history, with the other social sciences, with law, with theology and its history, and with the humanities, would help us to change our institutions. It would help because it would be studying the logic of our institutions, and, by implication, the alternative logics we desperately need.

Peace and all good,

Howard R.