The Effects of Stabilization Policy on Investment Planning

An exhaustive survey of the accuracy of forecasting occurred after the grossly erroneous forecasts of the immediate postwar period. See, for example, Sapir M. , “Review of Economic Forecasts of the Transition Period,” Studies in Income and Wealth, XI (New York: National Bureau of Economic Research, 1949), and Bratt E. C. , “A Reconsideration of the Postwar Forecasts,” Journal of Business, XXVI (April, 1953). These and similar studies point inescapably to the unreliability of forecasts. It should be mentioned that no similarly thorough analyses of subsequent forecasting performances, especially of econometric forecasts, have been published.

A case can be made for preventive discretionary action as an intelligence-gathering measure. The argument is that by taking some (almost any) action our stabilization authorities will be able to obtain more information about the state of economic activity than if no action at all is taken.

The following comments refer only to the major types of automatic stabilizers that are at present built into the economy by government policy decisions. That is, those devices from which economic instability evokes a countercyclical response, even though tax rates, exemption levels, levels of transfer payments, and so forth, remain unchanged throughout the business cycle. Another type of automatic stabilizer is sometimes proposed which would exert a compensatory influence by tying, for example, changes in tax rates, exemption levels, or transfer payment levels automatically to changes in some index of economic activity. This type of built-in stabilizer provides “formula flexibility.” Federal automatic stabilizers of this type are of minor significance.

A more complete formulation is my “Automatic Stabilizers and Business Cycle Mechanism,” Southern Economic Journal, XXVII (Jan., 1962).

The distinction used here is fully developed in Gordon R. A. , “Investment Behavior and Business Cycles,” Review of Economic Statistics, XXXVII (Feb., 1955).

See Jacoby Neil Weston J. F. , “Factors Influencing Managerial Decisions in Determining Forms of Business Financing: An Exploratory Study,” in a Conference of the Universities—National Bureau Committee for Economic Research, Conference on Research in Business Finance (New York: National Bureau of Economic Research, 1952), p. 147.

These a priori observations seem compatible with the findings of Meyer John Kuh Edwin , The Investment Decision (Cambridge, Mass.: Harvard University Press, 1957). That the desire for internal financing is less strong than formerly appears to be a proper inference from the authors' conclusion on p. 117. Also Simon Kuznets, citing Raymond Goldsmith's study of Financial Intermediaries in the American Economy since 1900 and Department of Commerce data, reports the following supporting figures for sources of funds for nonfinancial corporations [see Kuznets , Capital Formation in the American Economy (Princeton: Princeton University Press for the National Bureau of Economic Research, 1961), Table 39, p. 248]: Ratios of Internal and External to Total Funds External Funds to Total Funds Internal less Capital External Funds to Consumption Funds to Total Funds Allowances Total Funds Goldsmith Data 1946–1949 .64 .51 n.a. 1946–1956 .61 .63 n.a. Department of Commerce Data 1946–1949 .63 .48 .37 1946–1956 .58 .60 .42 1950–1956 .56 .66 .44

These data, which show a definite increase in the share of external funds relative to internal funds for the period 1950–1956, lend empirical confirmation to the assertion that internal funds financing is of lesser significance. It must be cautioned, however, that the alteration of expectations discussed in the text is undoubtedly not the sole cause of this transformation.

See Hickman Bert G. , “Comment on A. G. Hart, ‘Government Measures Designed to Promote Regularization of Business Investment,’” in a Conference of the Universities—National Bureau Committee for Economic Research, Regularization of Business Investment (Princeton: Princeton University Press for the National Bureau of Economic Research, 1954), p. 457.

The cyclical patterns of investment costs do not point uniformly to cyclical conformity. Let us first interpret investment costs as meaning the cost of funds. There is ambiguity even here. [See Modigliani Franco Miller M. H. , “The Cost of Capital, Corporation Finance and The Theory of Investment,” American Economic Review, XLVIII (June, 1958), especially pp. 261–288.] Because figures are available, let us assume that movements in interest rates and bond yields will give an indication of the conformity of cost of funds to the fluctuations in economic activity. R. A. Gordon has computed average “reference cycle patterns” for New York commercial paper rates and for high grade corporate bond yields from postwar data through 1958 [Business Fluctuations (2d ed.; New York: Harper and Brothers, 1961), p. 293]. These reference patterns do conform to the general business cycle. We may infer, therefore, that other series indicating cost of funds are at least approximately conforming and, hence, that cost of funds generally conforms to the business cycle.

We have more difficulty in determining conformity of investment costs if we mean by them the costs of physical plant and equipment. Let us adopt the U.S. Bureau of Public Roads' subindex of bid prices for “structures” in its federal aid highway construction contract prices index (see Highway Statistics) as a good approximation to plant costs. In the postwar period this subindex has roughly led economic activity by two quarters at the upper turning point (using National Bureau of Economic Research dating) and has lagged by three quarters at the trough. Thus, the series approximating plant costs is roughly conforming, indicating a countercyclical influence on plant investment. Plant investment is currently at the rate of about $13.0 billion annually and makes up roughly 25 per cent of new fixed investment of the business sector.

For equipment costs, however, using the price index of the machinery and motive products component of the wholesale price index (Bureau of Labor Statistics, Whole-sale Prices and Price Indexes) as an approximation, we note no cyclical pattern at all. Rather machinery and motive products prices have, until quite recently, risen continuously. Thus, a greater emphasis on cost of physical capital considerations in the investment decision would have no appreciable cyclical consequence as concerns equipment purchases, which constitute about three-fourths of total business expenditures on new fixed investment.

Irwin Friend makes this interesting point in his “Summary and Appraisal,” in Conference on Research in Business Finance (“Special Conference Series III” [New York: National Bureau of Economic Research, 1952]), p. 328.

We should not necessarily expect that empirical findings will corroborate this deductive judgment. Hickman W. Braddock , The Volume of Corporate Bond Financing since 1900 (New York: National Bureau of Economic Research, 1953), p. 168, has observed that the relationship between bond and stock financing is closely governed by their relative costs. (Note also Hickman's warning, ibid., pp. 178–179.) Since interest rates were high relative to the costs of equity funds during the late forties and fell relative to equity costs during the first half of the 1950's [ Shapiro Eli Mendelson Morris , “A Decade of Corporate Capital Investment: 1946–1955” in Freeman Ralph E. (ed.), Postwar Economic Trends in the United States (New York: Harper and Brothers, 1960), p. 332], we would anticipate that bond financing would have become less important in the later period.

Kuznets (op. cit., Table 48, p. 278) records that this in fact did happen, although his evidence is conflicting. He notes for all nonfinancial corporations, citing Goldsmith and the Department of Commerce data, that the proportion of net bond and note issues to total external financing fell for the period 1950–1955 relative to 1946–1949. (It is somewhat reassuring to notice, however, that the proportion of net stock issues to external financing was constant over the two periods.) Contrastingly, Goldsmith's figures show that the proportion of net bond and note issues to total long-term external financing rose from 1946–1949 to 1950–1955; but the Department of Commerce data lead to the opposite conclusion. We should also relate that the discussion in Shapiro and Mendelson (op. cit.), pp. 320–36, gives conflicting testimony.

Nevertheless, there is some evidence in support of the contention that debt financing has grown in relative stature as a result of the change in long-term expectations. Thus, Creamer Daniel Dobrovolsky Sergei P. Borenstein Israel , Capital in Manufacturing and Mining (Princeton: Princeton University Press for the National Bureau of Economic Research, 1960), Table 51, pp. 162–63, note that new bond issues as a per cent of new stock issues plus retained net earnings (i.e., equity financing) rose from the period 1946–1949 to 1949–1954 from 27.1 per cent to 36.7 per cent (for their so-called positive business cycles) and from 1944–1948 to 1948–1953 from 20.4 per cent to 31.4 per cent (for inverted business cycles). These data, for manufacturing and mining only, clearly suggest that debt financing has increased in relative importance in the later period. It is difficult to see, however, how these data on manufacturing and mining can be reconciled with the conflicting data on all corporations, since the manufacturing and mining sector dominates all corporate activity.

There seems to be no empirical or survey evidence on the length of the pay-out period. Wilson Both T. [“Cyclical and Autonomous Inducements to Invest,” Oxford Economic Papers, V (March, 1953), 75] and Youngson A. J. [“Investment Decisions, Trade Cycle, and Trend,” Oxford Economic Papers, VI (Sept., 1954), 295] believe that the length of the pay-out period has an appreciable impact on the rate of fixed investment. They suggest that the pay-out period is of greater length than the duration of a single business cycle (two to four years) and imply that it may even extend over several cycles. They do not express an opinion concerning pay-out lengths in different expectational situations. There are, however, precedents for the statement that improved long-term expectations will tend to lengthen the pay-out period. See, for example, Bert G. Hickman's commentary on Hart (cited above); his “Capacity, Capacity Utilization, and the Acceleration Principle,” in Conference on Research in Income and Wealth, Problems of Capital Formation: Studies in Income and Wealth, XIX (Princeton: Princeton University Press for the National Bureau of Economic Research, 1957), especially pp. 423–29; and Jacoby Neil Weston J. F. , “Financial Policies for Regularizing Business Investment,” in a Conference of the Universities—National Bureau Committee for Economic Research, Regularization of Business Investment (Princeton: Princeton University Press, 1954), p. 446. The ideas in this paragraph evolved from suggestions in these sources.

Liquidity condition would be given more weight in decisions, leading to postponement of investment and production decisions and to reduction of immediate production and capital formation. See Shackle G. L. S. , Expectations in Economics (Cambridge: The University Press, 1952), p. 75, and Hastay Millard , “The Cyclical Behavior of Investment,” in Regularization of Business Investment (see note 12), p. 30.