The theories that summarize the state of our scientific knowledge are no doubt necessary vehicles for that knowledge. They should also provide support for further investigations of new ideas. But neither the theories nor the ideas are immutable: we must always be prepared to abandon, modify or replace them as soon as they no longer represent reality properly. In a word, theory must be adapted to fit nature: nature cannot be adapted to fit theory.... It follows that when a scientific idea has been put forward, the aim should not be to preserve it by seeking support for it at all cost, while discarding everything that might tend to invalidate it. The opposite should govern our behavior. The facts that seem to upset a theory should be examined with the greatest care, for real progress always consists of relinquishing an earlier theory which covers fewer fact.s by another of more extensive scope.
Arthur Okun invited me to join this panel, one presumes, to present a monetarist view of inflation. This I shall do. In my judgment the current worldwide inflWtion, like virtually all of its predecessors throughout history, stems from excessive expansion of the money supply. I believe that this could have been prevented, at least in the United States, without either loss of jobs in the long run or danger of financial collapse. The only hope of restoring reasonable price stability is to cut monetary growth to the rate of growth of full employment output and to hold it there indefinitely. Implementing such a policy will not be easy but it is our only hope. Most recent literature on inflation suffers from two serious faults: it fails to examine the problem in a sufficiently long-term perspective and it takes inadequate account of delays between causal disturbances and inflationary consequences. On the first point, most writers on inflation are preoccupied with ephemeral symptoms or temporary disturbances. Events such as a rise in the minimum wage, a poor corn crop, the formation of an international oil cartel, devaluation of the dollar vis-a-vis other major currencies -the list could on and on are capable of moving the inflation indicators upward, at least temporarily. However, such alleged causes of inflation operate only sporadically; they cannot explain a sustained rise in prices. One could write a history of inflation that would explain it as a series of unrelated accidents. As social
In the March 1972 issue of this Rezview, Duncan Bailev and Charles Schotta published a study of the returns to investment in the graduate education of Ph.D. academicians in the United States. They use as a proxy for the academic income of Ph.D. faculty, the salarydata reported by the American Association of Universitv Professors (.4A4UP). The alternative income used is an estinmate of the income of bachelor's degree holders taken from an extensive survey of salaries in occupations open to bachelor's degree holders in the state of California. They conmpute both a private and a social rate of return. Their private rate of return includes as costs an estimate of the foregone income of graduate students. Their social rate of return in addition to incomne foregone by students, includes an estimate of the per unit contribution of the state of California to graduate education at the Berkeley, and Los Angeles campuses of the University of California. A second social rate of return is computed by including an estimnate of the costs of graduate school dropouts. Each of these rates of return is estimated for two through six Xyear periods of time spent in graduate school and for eleven different academic income patterns.' Bailev and Schotta explicitly omit from their calculations all incomes earned by academicians in addition to their academic -ear contract salaries. Since they are interested only in test-of-the-marketplace conclusions, they do not consider externalities or the public good aspects of research. T heir conclusions may be surmmarized as:
The bias toward understatement in reported agricultural earnings may be offset by the other sources of error which Denison mentions. Denison's footnote at this point, however, refers to another source of downward bias in agricultural earnings as a meameasure of labor quality: the preference of agricultural labor for farm life and resulting immobility. Denison chooses to include the influence of immobility in the 'ability' adjustment. The evidence relating to the contribution of ability to education earnings differentials is sufficiently fragmentary to permit Denison to come to this decision. My own interpretation of the available evidence is that ascribing two-fifths of the educational differentials in earnings to ability alone does not overstate abilitv's contribution;' I would not include in the two-fifths the influence of other nonschooling variables. My method of adjusting for agricultural immobility as well as for other sources of bias in agricultural earnings as reported is to use Victor Fuchs' estimates of educational earnings differentials based on a sample which excludes agricultural workers.
The Golden Rule of accumulation in an economy with an endogenously determined growth rate of labor has been examined by Eric Davis. Davis finds that if the growth rate is an increasing function of per capita net income, steady-state per capita consumption is maximized not at the equality of the propensity to save and capital's share of output, but at a propensity to save smaller than capital's share of output and an interest rate which exceeds rather than equals the growth rate. The supply of labor in an economy is affected by both the growth and participation rates of its population. Davis' results and the traditional Golden Rule depend upon the implicit assumption that the latter rate is constant. Davis calls for a more flexible treatment of the participation rate, allowing the possibility of optimal unemployment in a theory of optimal savings. This note examines a simple model with a varying participation rate. It is seen that one common assumption describing the participation rate leads to a recommendation opposite that of Davis.
In recent years, several astute observers of the federal have argued that the separation of executive and legislative branch powers makes the formulation and execution of a coherent set of federal policies -including economic policies-nearly impossible. James Sundquist has suggested that, while the separation of powers has always hampered presidents' efforts to translate their programs into action, recent trends-the disintegration of political parties, haphazard selection of presidential candidates, and congressional self-assertiveness, combined with fragmentation of congressional leadership -have brought us to a ... crisis of competence in government (p. 531). Barry Bosworth, citing many of the same weaknesses, alleges that ... the economic and political system we have created may make the task of leadership virtually impossible (p. 70). Going farther, Lloyd Cutler advocates constitutional reform reducing the separation of powers and allowing a president to form a able to lay out and implement its policy, unimpeded by the Congress (pp. 126-27). This paper sounds a more hopeful note. It argues that, since the passage of the Budget Reform Act of 1974, the Congress has made enormous strides in its ability to consider and act on major questions of budgetary and fiscal strategy. The new procedures have given a president who has a well-articulated economic program a forum for debate and decision that did not exist before. They have also given a Congress that finds the president's program wanting a mechanism for choosing an alternative. But the new procedures have brought their own problems. There simply is not enough time to make the major strategic decisions and to continue the Congress' traditional role of annual appropriations and minute examination of detailed spending and taxing legislation. If the Congress is not to collapse under the stress of decision overload, choices will have to be made in less detail or with less frequencyor both.
In a recent article in this Review, Jeffrey Schaefer reports an important discovery that runs counter to currently accepted views. He finds that clothing exemptions in the New Jersey tax law actually reduce sales tax progressivity.' It would be of great value if Schaefer's results could be generalized to a wider geographical area. Other state legislatures considering adopting or revising the sales tax, for example, could be spared the mistake of exempting clothing in the belief that such an exemption pushes the sales tax toward progressivity if, in fact, the opposite is true. This comment extends Schaefer's analysis to the whole of the United States. We derive progressivity-regressivity indexes for a sales tax which excludes clothing from taxation for 1) all urban areas in the United States, and 2) all urban and rural areas combined. The empirical information for these areas comes from the most recent survey of consumer expenditures and income for the United States by the Bureau of Labor Statistics (1964 and 1966). Following Schaefer's definitions, a sales tax is considered to be progressive if the effective rate of taxation increases as the ability to pay increases. If the effective tax rate declines as the ability to pay increases, the tax is regressive; and if the rate remains approximately constant as ability to pay changes, the tax is proportional. An effective way to test whether a sales tax is progressive or not is to derive the elasticity of the tax base with respect to the measure of ability to pay. We can use Schaefer's regression equation to derive the progressivity-regressivity index of alternative sales tax bases:
The authors suggest a general, heuristic method, based on elementary notions of price theory, for dealing with the question of anticipated changes in models of perfect foresight. The method is illustrated for the case of anticipated shocks to the state variable; in this context the standard "shadow price continuity condition" fails, while the method presented here applies. After presenting the general solution, it is used to analyze various examples of preannounced changes.
the current stock of women economists in academia, with emphasis on women with new Ph.D.s in economics; 2) are the flows of women economists into the various faculty levels (a) in line with the proportion of women in the relevant stock, and (b) enough greater than the past pattern to suggest affirmative action is occurring; and 3) can the revolving door syndrome be quantified, and is it affecting women disproportionately? These questions pick up the problem at the point of production of Ph.D.s. No analysis is made of the prior issues of reducing barriers to filling the pipe line with women earning Ph.D.s, or of the contributing issues related to encouragement of women after employment