Policy Making
7. Doubts about the model

The Tinbergen model is one way to think about the role of the economist as expert in a democratic and market-oriented society. Its goal of creating rational economic policy on the basis of pre-existing tastes, objective constraints, and available policy tools fits comfortably with the prejudices and personalities of many economists. Public choice theory serves for some economists to assuage discomfort with its highly abstract and apolitical assumptions. Other economists express grave doubts about the general validity of the model.

Beyond Plato's vision for the philosopher-king and Herbert Stein's three-line demolition of the "economic policy advisor," two doubts we've already mentioned, reasons for doubting the validity of the Tinbergen model are in fact many and distinguished. Here are three prominent ones:

  • The Hirschman Critique. A.O. Hirschman, a distinguished economist, political scientist, and philosopher long at Princeton University's Institute for Advanced Study, responds that "tastes" are not the same as "values," and that values are a thing worth arguing for, even if you are hired for pay to be an expert economist. In other words, both tastes and values can change in the short run through dialogue and personal reflection, and should. Take an example in "institutional policy." If you want to live in a society with less racial segregation (that is, you value integration) then you, the economist-expert, should not sell your soul to the politician who wants advice on how to create more racial segregation and divisiveness in your society. "Tastes" for segregation may be expressed by a politician with aspirations but the economist-expert with an opposing ethical view will at minimum engage the politician in dialogue, and attempt to re-orient his tastes-or, even better, his "values".
  • The Burke Critique. Kenneth Burke (1897-1993) was one of the most distinguished and prolific of American philosophers, cultural and literary critics of the 20th Century. Nearly single-handedly he revived the study of "rhetoric" with a spate of books he published in the 1940s and 1950s. Of the validity of the Tinbergen model Burke would have had a lot to say (he always had a lot to say). But the biggest doubt he'd raise would concern what Burke called in A Grammar of Motives the "agent-to-action ratio," a matter of scale. In the Tinbergen model one is led to believe that the economist as agent has a direct and large effect on the actions (the votes) of politicians, like a monopolist. Not so. By the time a bill has become a law an economist will hardly recognize the remains of his or her idea. This is a result of bipartisan compromise and a host of other variables we needn't mention here. The Burkean critique amounts to this: the expert works on a scale that is too small relative to all policy making activity in the national economy to have much effect on it. It is worth mentioning in this context that after a few months the formidable Milton Friedman quit his job as Chairman of Reagan's Council of Economic Advisors on grounds that Reagan wouldn't listen to him. "Reagan wasn't conservative enough in policy matters," Friedman complained, confirming the relevance of both the Burke and Hirschman critiques.
  • The Kuran Critique. In a widely acclaimed book, Private Truths, Public Lies (Harvard University Press, the University of Southern California economist Timur Kuran has argued that preferences "revealed" are not what they seem. In fact, he says, citing everything from the outing of homosexuals to the rapid collapse of apartheid in the U.S. and South America, what we say we prefer in public may be a plain-faced lie about what we actually prefer in private. Like Hirschman's subtle work in a little-read book called Shifting Involvements, the Kuran critique should restore sobriety to any economist or politician believing he's got his finger on the pulse of others' "ultimate policy goals."

A fourth important but less well known way to criticize the Tinbergen model might be called the Atkinson Critique, which we name after the great British economist A.B. Atkinson, Warden of Nuffield College, Oxford. Against the Tinbergen model Atkinson would say that the political values of economists are in reality "surfacing in their own models," particularly at times when real world equilibria are drifting toward an unwanted extreme. The values of economists should therefore enter their formal models of the political process endogenously-"inside" the model.

Atkinson gives the example of economists alarmed by the threat of a larger welfare state (no random example). Suppose in a public choice model that the long-run tax rate is with "full [public] confidence" tending toward a high tax rate/basic income guarantee equilibrium. "Alarms are then raised, by economists . . . about the hazards of the welfare state. This causes a large instantaneous drop in public confidence," leading the economy toward a "zero tax, zero welfare state equilibrium." (A.B. Atkinson, The Economic Consequences of Rolling Back the Welfare State [CES Munich and M.I.T. Press, 1999], pp. 128-130.) Atkinson presumes that the populace actually listens to economists, and that the economists, for their part, have attempted to address the populace. Both of these assumptions require further investigation, as Burke and Kuran would note. But the point here is that to Atkinson's way of thinking, unlike Tinbergen's, the values of economists ought to be a variable in their own models of the political process.

Still, a majority of today's mainstream economists do not have a problem with the Tinbergen model. They see it as a plausible extension of a rational choice theory they already espouse.

Ziliak: Arjo, you advise government as an economist-expert. In fact, like Jan Tinbergen, you are a Dutch economist who advises Dutch politicians on the Dutch economy. Do you see a reflection of yourself and your employers in the Tinbergen model?

Klamer: Yes and no. I think Tinbergen was right to emphasize the distinction between tastes and constraints, between preferences and budgets. The distinction is crucial because, as we've been saying all along in this book, you can't always get what you want. There's no such thing as a free lunch. Tinbergen was right to emphasize this point, which you only hear from economists. But at the same time I think a lot of economists try to steer their politicians' preferences toward different ultimate goals. I certainly do. Still, at the end of the day, the economist has to show how to rationally achieve those goals.

Ziliak: Do you also believe that preferences sometimes get shaped the other way around-politicians shaping economists? I am asking if you think that some economist-experts fall so deeply enchanted with the White House or their nation's center of power that they change their rhetoric and beliefs just to fit into the current administration?

McCloskey: I have an opinion on that question. Yes, I say, some economists do change their rhetoric and beliefs in order to better fit with the ultimate policy goals of a politician. I've seen it, and I'm not in favor of it. Look, since the 1960s in the United States academic economists have with other scientists participated in a race to fetch big government grants and jobs. Good for research and university status, you say. But if a liberal Democrat is in office, the grant-writing economist tries to sound like a liberal Democrat. He gets the grant. Now he has to make that liberal Democrat happy. Eventually, though not every time, the economist becomes a liberal Democrat. You think my example is outdated? Well, nowadays a grant proposal for a grant with Health and Human Services sounds more like a neoconservative manifesto than a scientific inquiry into poverty and well-being.

Ziliak: So one way to change the Tinbergen model is to change the linear, top-down flow of "preferences" from politicians to ultimate goals. Instead, the flow would go from economist-experts to politicians, and from politicians to economist-experts; it could also have a time dimension, suggesting how individual economists change what they take to be ultimate policy goals over the course of the political business cycle.

McCloskey: If you ask me, those kinds of changes would improve upon Tinbergen's still useful model. I say this without hypocricy. I am the Jan Tinbergen Distinguished Professor of Erasmus University, in Rotterdam. But the great Tinbergen worked at the height of a faith in a certain kind of instrumental rationality that is not ours.

Maria: I guess you wouldn't have written a book on dialogical economics if you didn't think that dialogue mattered to the ultimate goals of economic policy.

Klamer, McCloskey, and Ziliak [simultaneously]: You're certainly right about that!

The notion of instrumental rationality derives from the philosophy of David Hume and has been refined or refuted by neoclassical economists, game theorists, cognitive psychologists, and philosophers. Rational economic agents are said to be instrumentally rational when they choose an optimal mix of means to achieve ends which are already given. By contrast, communicative rationality is a broader notion of rationality acknowledging the pivotal role that dialogue can have on both the means and the ends.