Positivism as Cost-Benefit's Tragic Flaw

Writing Sample
Grade awarded: A

"Positivism as Cost-Benefit's Tragic Flaw"

Clinton T. Brass

SPP 585
Professor Rick Hall
September 17, 1995

The Economist, editorial, April 1, 1995.

"Stay cool."

Letter to the editor, The Economist, April 22, 1995.

"You make the extraordinary claim that because we are unable to predict the effects of global warming, we are well advised to carry on as we are."

In his article "Cost Benefit Analysis and the Art of Motorcycle Maintenance," Baruch Fischhoff cites some examples of how cost-benefit analysis can be misused.¹ Among them, Fischhoff mentions the problem of costs and benefits that do not quantify easily or that are difficult to measure. Citing another author, he explains that "many analysts adopt the easiest approach of all for dealing with hard-to-measure costs and benefits: they simply omit them."² While Fischhoff proposes a plausibly effective countermeasure against this tendency, I submit that a root problem of positivism is deeply ingrained in our scientific culture. So deeply, in fact, that cost-benefit analysis will probably suffer from positivism until the culture of science and, ipso facto, economics change substantially.

What is positivism? Embraced by French philosopher Auguste Comte in the early-middle nineteenth century, positivism became a powerful philosophical movement in the Western world. According to the Encyclopedia of Philosophy,

characteristic theses of positivism are that science is the only valid knowledge and facts the only possible objects of knowledge; . . . [that] positivism denies the existence or intelligibility of forces that go beyond facts and the laws ascertained by science . . . [and] opposes any procedure of investigation that is not reducible to scientific method.³

So influential was the new philosophy that in the last decade of the nineteenth century, Austrian physicist Ernst Mach championed a more critical form of positivism that directly influenced philosophers and scientists like Ludwig Wittgenstein and Rudolf Carnap. Such crucial concepts as the verifiability principle in the philosophy of science found rich soil in positivism.

What does this have to do with cost-benefit analysis? As Nobel-Prize winning physicist Steven Weinberg put it, this positivist imperative permeated scientific thought and created a bias which "demands not only that science must ultimately test its theories against observation . . . but that every aspect of our theories must at every point refer to observable quantities" (italics added).⁴ If we take this statement to be a broadly applicable analog for the "hardest" social science -- economics -- positivism's threat to the balance of cost-benefit analysis becomes clear.

Let me mention two examples. First, the problem of what causes long-term economic growth has confounded economists for the last forty years. According to The Economist, "Until recently, economics had little of interest to say about economic growth. . . . even by their standards, they have been terribly ignorant about it."⁵ Economist Robert Solow devised an ad hoc, "neoclassical" theory of economic growth in the 1950s by accounting for growth by adding up easily observed, easily measured quantities of capital (machinery), workers, etc. He then subtracted the growth due only to increases in scale (of plant, equipment, workforce, etc.) and was left with a "residual" number that could be "attributed" to technology and human skills and ideas. However, as The Economist says, this "so-called theory is inadequate -- so much so that its teachings have had virtually no influence on policy-makers." The data Solow analyzed was only of tangential relevance to the question he was asking, but unfortunately he had to be empirical to get any attention from colleagues and policy-makers.

Using common sense, most economists would have guessed that the primary determinants of long-term, per capita economic growth were technological innovation and development of "human capital," not a residual after controlling for increases in "factor inputs." But positivism helps to prevent any such "unobserved" conclusions from influencing policy-makers.

For a second example, please refer to the quotations that introduced this essay. In its editorial regarding the recent Berlin summit on climate change, The Economist took the line that because the phenomenon of "global warming" was not proven and proponents for action could not predict its effects, the world's leaders should adopt a wait-and-see stance. Clearly the editorial takes a positivist stance. However, the letter to the editor responded with logic, not empirical data: "If any complex system [the environment] is operating satisfactorily, it is stupid to think that an eight-fold change in one of the parameters [carbon dioxide level] should be a matter of indifference. For The Economist to reassure us that there is no need to worry, since we do not know what the outcome will be, is mind-boggling."

Thus, we return to Fischhoff. If economic analysts tend to omit or discount those factors in cost-benefit analysis that are qualitative or difficult to measure, inevitable miscalculations of costs and benefits will result. The net effect of positivism is to weight factors according to the hardness of their data, not necessarily their logic, relevance, or importance.

Fischhoff concludes by arguing convincingly that cost-benefit analysis can and should be a powerful tool to motivate critical thinking in government, industry, and among the public. Unfortunately, because of positivism's huge influence on science and economics, I believe cost-benefit analysis will virtually always bring with it the tragic flaw of disproportionately weighting the quantitative and measurable while paying too little attention to the qualitative and difficult-to-measure. Fischhoff's proposals to deliberately weight qualitative aspects of policy problems and avoid scientific hubris have some promise, but they probably do not alone have the strength to overcome a century of positivist thought.

Endnotes:

¹ Baruch Fischhoff, "Cost Benefit Analysis and the Art of Motorcycle Maintenance," Policy Sciences 8 (1977): 177-202. [Back to text.]

²Ibid., p. 186. [Back to text.]

³Encyclopedia of Philosophy, Paul Edwards, ed., 1967, s.v. "Positivism," by Nicola Abbagnano. [Back to text.]

⁴Steven Weinberg, Dreams of a Final Theory (New York: Vintage, 1993), pp. 174-5. Weinberg goes on to complain that this form of positivism naively impeded acceptance of the theory of atoms at the beginning of the twentieth century and continues to unjustifiably harass theoretical physics. Indeed, modern day string theory cuts totally against the positivist grain. [Back to text.]

⁵"Explaining the Mystery," The Economist, January 4, 1992, p. 15. [Back to text.]

[Back to Writing Samples Index]
[Back to Home Page]