“It’s kind of hard to sell ‘trickle-down,’” he [David Stockman] explained, “so the supply-side formula was the only way to get a tax policy that was really ‘trickle-down.’ Supply-side is ‘trickle-down’ theory.” [William Greider, “What David Stockman Said,” Washington Post (Nov. 22, 1981)
During the Eisenhower administration, critics of the Republican Party’s economic policies called them the policies of “trickle-down economics.” There was even a lyric in a Joe Glazer “folk” song about “trickle-down George” Humphrey, who was the Secretary of the Treasury. Trickle-down economics, the critics said, was based on the theory that tax breaks given to the rich would multiply investment, provide jobs, and eventually create increased income for everyone in the economy. In other words, by “giving” the rich more after-tax income, the government would foster economic growth, because the rich are more likely to invest than the poor, since any additional money in their hands would not have to be spent on necessities.
The critics resented the suggestion that the rich should receive a reduction in their tax rates. They had campaigned long and hard for the “progressive” income tax—the graduated income tax—and they were not happy with any suggestion that the reason why the American economy was not experiencing maximum economic growth was because of the graduated income tax, which in the 1950s extracted a maximum of 91 per cent of “unearned” (investment) income.
VALUE THEORY
It is one of the ironies of history that both the critics and the defenders of reduced tax rates in the highest brackets relied on the same view of income. What we call “welfare economics” was created at the turn of the century by a group of British economists, most notably A. C. Pigou, who misused the crucial economic doctrine-of marginal utility. They argued that since each additional unit of income (ounce of gold, dollar, pound sterling, etc.) is worth less to the recipient than the preceding unit of income, we must conclude that it would increase total social utility within a society to impose graduated income taxes. Why? Because the goods bought by the thousandth dollar received by a poor man are worth so much to him, whereas the goods that the millionth dollar will buy a rich man are valued very low by the recipient. The rich man will have purchased all those goods and services that were high on his value scale long before he receives his millionth dollar. Thus, concluded the welfare economists, the civil government can increase total social utility in a society by taking (say) 75 cents of that final dollar away from the rich man and transferring the money to the poor man.
It took three decades for an economist to come up with a theoretically precise rebuttal to this position. Lionel Robbins, who had been influenced by the writings of Ludwig von Mises early in his career, provided the answer. Robbins argued that while it is legitimate for an individual to compare the value to him of the first, second, or nth dollar of his own income, it is not legitimate for anyone to make interpersonal comparisons of subjective utility. We cannot make scientifically valid statements comparing the subjective value of the second dollar of income (or the millionth) in one person’s income with the subjective value of the second, third, or nth dollar of another person’s income. We cannot even make cardinal (quantitative) comparisons in our own minds—this is worth precisely this much more to me than that but only ordinal comparisons: this is my first choice, that is my next choice, and so forth.
Common sense may not accept Robbins’ conclusion, but such is often the case in matters of economic theory. Science frequently produces conclusions that are in flagrant opposition to common sense. We need to consider an example regarding interpersonal comparisons of subjective value. The millionaire may value his millionth dollar very highly, if he has some investment in mind which requires a high initial payment, or if he regards his income as a kind of measure of his value to society. On the other hand, some mystic or ascetic may not place a high value on his thousandth dollar of income in any given time period.
We do not have a quantitative measure of pleasure or utility; thus, we cannot, as scientists, make interpersonal comparisons of subjective utility. Conclusion: it is not scientifically demonstrable that total social utility within a society can be increased by taking 75 per cent of the rich man’s income in the highest tax brackets and transferring this money to a poor man (minus 25 per cent for government handling). There is no such thing, scientifically speaking, as total social utility. We cannot add up subjective utilities as if we were adding up a column of figures.
Admittedly, as policy-makers we have to make judgments concerning the advisability of particular economic programs. But Robbins’ refutation of welfare economics by means of the argument against the scientific validity of interpersonal comparisons of subjective utility cannot be limited to the narrow case of the graduated income tax. It undermines all attempts to “tally up” social utility in the name of economic science. We cannot, as economic scientists, say that any policy will increase total social utility. There is no way to measure “total social utility.” So effective is this argument that it denies to economics the legitimacy of making estimates of the total value of any aggregates. What does Gross National Product mean, anyway, if we cannot assign any value (or meaning) to the columns of figures in a GNP index? If Robbins’ thesis is correct—and since 1932, no economist has shown how it might be incorrect—then most of what we know as modern applied economics, including the formulation of economic policy, is an illusion.
Robbins had this pointed out to him by Roy Harrod, who later became Keynes’ biographer, in 1938. Incredibly, Robbins capitulated to Harrod and abandoned the obvious and inescapable logic of his earlier argument. But he could never explain where he had been incorrect. He simply wanted to maintain the status of economists as scientific advisors, so he abandoned the logic of subjectivist economics. Somehow, he and Harrod agreed, economists as scientists can make assessments of the total social utility of particular economic policies. Somehow, GNP (or other economic statistics) are meaningful They could not say exactly how, but somehow. It was a matter of faith.
