Knowledge that Transforms

AST spring and summer we witnessed a vigorous exchange of views between the present Council of Economic Advisors to the President and the former chairman of that body, Professor Arthur F. Burns.' It was a lesson in humility to note that the key issue of this policy-oriented discussion was not some highly sophisticated model but the basic problem of determining the secular growth curve of the United States economy and of estimating the size of the output gap.

ing case for the introduction of Capital Budgets could be made if we could gradually evolve a concept of capital expenditures which would include both physical assets and investment in human capital, as two types of investments in economic growth. A case can be made for the introduction of Capital Budgets for such agencies as the Post Office where the existence of a chronic perpetual deficit may unduly inhibit the approval of substantial capital expenditure items which would improve efficiency. Also, in the case of FHA, FNMA, and other agencies which are required to operate programs on a self-sustaining basis, the separate financing of capital expenditure by means of bonds with redemption dates in alignment with the rate of depreciation of these assets may well be considered, provided their financial operations are geared into the Treasury program for debt management and the general economic situation. The point to be made here is that scope for the useful adoption of capital budgeting in the Federal Government is considerably more limited than proponents of the idea allow, and that, with the exceptions mentioned above, no real economic purpose would be served by the introduction of a separate Capital Budget. However, it would be highly useful if projects could have separate classifications for investment in physical assets and in human capital, but would regard all such expenditures as investments in economic growth. More education concerning the merits of the proposed Program Budget as well as the two supplementary statements is necessary, not only as regards Congress, but also -and ultimately perhaps even more important in order to create a better informed public opinion. It is essential that Congress and the general public have a clear idea of the different functions of each of these concepts. Public and Congressional approval of the necessary changes will come more quickly if the various figures shown as Budget out-turns, depending on which of the three concepts has been employed, could be reconciled without too much strain in a similar manner to the process involved when the average citizen reconciles his bank statement at the end of each month. It is felt that the adoption of these improvements would put us in a much better position to make economic policy decisions on the basis of economic and fiscal data, and would facilitate a more efficient allocation of our resources. Businessmen could make their investment decisions with the full knowledge of government programs for a number of years ahead while economists could make their recommendations with a greater degree of confidence and such recommendations would gain a greater measure of Congressional and public acceptance.

T HE Commission on Money and Credit has laid its 285-page egg 1 and gone over like a lead balloon -choose your own metaphor with both the economists and the general public. Certain of its administrative suggestions, notably those involving reconstitution of the Federal Reserve System's Board of Governors, have attracted a significant modicum of attention.2 On the substantive side, however, the Commission's main body of work appears already spurlos versen,kt, in unhappy contrast with both the National Monetary Commission of fifty years past, whose influence it was intended to rival, and the Radcliffe Report of I959,3 its closest contemporary transatlantic equivalent. This unhappy fate rather befits a series of attempted least common denominators between unreconciled and possibly unreconcilable special interests, which turned out to be meaningless verbal compromises as often as anything more. Indeed, the least uninteresting feature of the report to this reader was the triangular running battle between the predominantly sound don't rock the boat position of its text and the two accompanying sets of mutually contradictory footnote dissents. Set i, contributed primarily by the labor bloc (Lubin, Nathan, Rieve, Ruttenberg, and Thorp, with Ruttenberg the principal spokesman), stands for guaranteed full employment and a 5 per cent annual growth rate, at any cost in direct controls over everyone but organized labor, and over everything but wages. Set 2, contributed by a mixed bag of business, finance, and agricultural4 spokesmen (Black, Lazarus, Miller, Schwulst, Shuman, Thomson, and Yntema) stands for Free Enterprise in the economic aggregates -McKinley minus the gold standard.5 Rather than aim a supernumerary nail at the Commission's coffin, I propose to examine a biased sample of the professional papers submitted for the Commission's use, and to all appearances neglected by the Commission in favor of cliches and weasel words masquerading as common sense.6 The papers are to an econo-

Technical Change R ECENT studies have attempted to derive the relative proportions of total increase in output that are caused by growth in capital stock and by technical advance. These studies seem to have turned attention away from capital formation as a source of growth to technical change as the only important factor. Each of these papers has assumed, explicitly or implicitly, that the technical change takes place in a neutral fashion. The present paper examines this assumption and finds it to be questionable. In this light, new estimates of the importance of capital formation are found. The first important recent paper attempting to measure technical change was presented by Robert Solow.' He concluded that technical change is neutral on the average, that there was little basis of choice with his data among five possible production functions, and that about 87?2 per cent of the increase in gross output per man hour is attributable to technical change, and I2/2 per cent to increased use of capital. Other authors with various techniques have studied the relative importance of capital and of technical progress. Niitamo made a regression model of production in Finland and found most of the increase was attributable to the passage of time and to education, and little was attributable to capital formation.2 He concluded that emphasis should be taken away from capital as an aid to economic growth and placed instead on these other factors which he lumped together as the human factor. Massell made a study with improved data that used the same methods as Solow but which annlied only to the manufacturing sector.3 His results again gave the greatest importance to technical change. He said,

HIS paper presents a test of the permaT nent income theory based on a comparison of the savings-income ratios of different occupations. Specifically, I will test the hypothesis that the proportion of permanent income consumed is independent of the level of permanent income.' This is, of course, only a part of the permanent income theory the other less revolutionary part of the theory which states that consumption is a function of long-run rather than yearly income will not be discussed here. In doing this I will use the Friedman version of the theory rather than the rival Modigliani-Brumberg version.2 This is so because the Friedman version suggests that a person's idea of his permanent income is dominated by his income experience in relatively few years,3 while the Modigliani-Brumberg version leaves the length of the horizon open. Unfortunately, the data used here do not allow one to generalize about the very long-run income experience of households, and hence permit only a test of the Friedman variant. The test used is a comparison of the savingsincome ratios of different occupational groups. Since, for the period covered, occupational grouping was fairly stable, the income of an occupation furnishes a clue to the permanent income of its members. According to the absolute income theory, the higher income occupation should have the lower average propensity to consume. The relative income theory in its cross-section version also implies this.4 The permanent income theory, on the other hand, suggests that there is no tendency for the higher income occupation to have a lower average propensity to consume. This makes it possible to test the permanent income theory by seeing if the absolute income theory and the relative income theory are able to predict significantly better than a naive model. One major advantage of such a test is that, unlike many other tests, it distinguishes sharply between the permanent income theory and the (cross-section) relative income theory.

T HIS paper is concerned with the problem of measuring potential upgrading demand in the housing market. We seek to develop a means of identifying in sample survey data families who might make a housing adjustment that would provide them with more housing space or higher cost housing. Our analytical framework is a form of cross section demand analysis, with the focus on housing stocks of individual families. We assume these stocks are held because of the services they provide and that each family chooses to hold a stock that provides the flow of services and has a cost consistent with its over-all budget plan. Upgrading adjustments are assumed to arise, when, because of changing needs and ability to pay, families find themselves with less than this desired amount of housing. Our approach to measuring this demand is based on two assumptions. First, we assume that families do not adjust their housing stocks instantaneously to changing demands. Rather, a considerable amount of time may elapse before they do something, which means they are in a position of disequilibrium for some time prior to making an adjustment. Second, we assume these families have significantly less housing than those who are in equilibrium. Given these assumptions, it is our contention that a cross section demand function, calculated from data for equilibrium families, will provide the basis for measuring potential upgrading demand. Such a function would be used to calculate expected stock values for all families in a cross section sample, which would then be compared with the actual stock values reported in the data. A simple analysis of the residuals should enable us to identify the disequilibrium families since, according to our assumptions, they would appear as families who have a deficiency in housing. Measures of housing space and quality are used as the basic dimensions of demand. Our model then consists of two sets of demand equations calculated from data ' for families who appear most likely to be in equilibrium. Equilibrium is defined in terms of attitudinal, purchase plans, and mobility status information included in the data. Where the head is under 45, we assume equilibrium families are those who have no plans to buy 2 or to make major additions to their current housing and have been living in their current residence for five years or less.3 Where the head is 45 or older, equilibrium families are assumed to be those who have no plans to buy or to make major additions to their present housing.4 The model is examined in two ways. First, since its success hinges on the presence of measurable differences between the housing stocks of equilibrium and disequilibrium 5 families, we test for these differences. The equations or norms are used to calculate housing demands for disequilibrium families, which are then compared with the actual housing available to these families.6 If there are