of quantitative living standard increases to improvements in the quality of living is fundamental and far-reaching, not superficial or ephemeral.

During the past five years aggravated by the war in Southeast Asia, our basic long range economic problem has changed. Our problem of stimulating demand to increase production and provide more jobs is rapidly shifting toward a problem of stabilizing excessive demand for services, money, goods and manpower and increasing productivity so that we may keep costs and prices under better control and make the qualitative progress that we desire.

Now that we have sampled full employment, the economic objective of the last forty years, we have come to realize that inflation can become as chronic a problem as a recession. I have confidence in our ability to maintain reasonable levels of employment and production over the long pull. Indeed, we can now begin to divert more of our individual and collective energies to reducing the costs of achieving that goal.

Public interest is aroused, especially among the young. Their own personal needs for economic security have been generally satisfied. They've turned now to more qualitative demands to satisfy public needs, and, for example, have catapulted "the correction of environmental pollution"-in the broadest sense into a top spot among our major social and political issues. In addition, businesses are coming to recognize their own legitimate social responsibilities to help satisfy these qualitative demands. They are coming to view "social performance" as a relevant complement to "profit performance."

We must recognize, however, that social, qualitative progress has its costs. In the generally fully employed economy of the Seventies, this means that hard choices must be made-we must say "no" to some demands if we are to satisfy others. We can't do everything at once. Priorities must be established, not left to chance or the naive assumption that none are needed.

Establishing and maintaining these priorities will sorely test the skills of public and private managers. For public decision makers, there will be endless demands for "immediate action" on a host of individual and community problems in the face of limited financial resources and taxpayers' revolts. For private business managers there will be the never-ending price-quality demands of customers, the incessant wage demands of employes, the persistent profits demands of stockholders and the new and growing social performance demands of the public.

Resolving these apparently conflicting demands will require the dedication of countless individuals and businesses. Achieving the desired results of increased productivity and qualitative progress will require determination and public and private cooperation, leadership and planning.

CAN INDUSTRIAL R. & D. LIVE UP TO THE CHALLENGE OF THE SEVENTIES? The social and economic hallmarks of the Seventies, then, will be productivity and quality. They suggest the avenues by which industrial-military research and development can make its contribution to progress. Does industrial R & D contribute to economic productivity or growth potential? Robert Solo, in an important article in 1962 [6], demonstrated that rising expenditures on research and development seemed to bear little relationship to overall economic growth. If anything, R & D outlays to that time appeared to be detrimental to the pace of economic development.

Solo founded his position on the following points: First, a comparison of generally accepted measures of growth (such as GNP or output per man-hour) with spending on R & D showed no relationship between the two. From 19531962, spending on R & D skyrocketed, but the index of output per man-hour trended downward. Second, a large part of R & D was directed to outer space exploration and advanced weaponry and thus was not immediately growthoriented at all. Third, in an economy of growing labor scarcity, increased spending on research and development required a reduction in the quantity and quality of labor devoted to civilian purposes in private industry.

My own investigation of the relationship between economic growth and R & D spending through 1967 confirms Solo's initial findings. Exhibit I compares the rates of real growth in total industrial research and development, aircraft and missiles, and electrical and communications equipment R & D with rates of growth in real GNP, output per man-hour and real per capita disposable income from 1957 to 1967. Beyond the simultaneous cyclical movements that appeared to affect the data early in the period, there is no apparent relationship between increased spending on R & D and quantitative economic progress.

A more technical evaluation of the relationship between these growth rates based on accepted statistical methods implies that, to the extent that a measurable relationship exists, it is negative! Additional spending on R & D may have actually retarded economic development."

In any case, it's clear that Solo's original contention still stands: "The evidence does [s.i.c.] show that the effect of the national R. & D. effort on economic growth cannot be taken for granted."3 Certainly a more detailed investigation than that presented here might reveal a complicated structure of lead-lag timing that relates industrial-military R & D to national economic development. However, proof of the existence of such a structure is the burden of those who assert that R & D must necessarily accelerate economic progress.

Of course, there have been tremendous benefits of industrial R & D to individual companies or industries. In fact, some foreign observers have attributed most of our postwar economic success to industrial research and development. Total federally sponsored R & D has probably had a significant effect on economic development. But the empirical evidence implies that these company and national benefits may have been offset by undetermined costs related to R. & D. On this score, reference to the research on the sources of productivity change by Jorgenson and Griliches is helpful.

In a controversial article [2], Jorgenson and Griliches show that when real factor outputs and inputs, including both capital and labor, are measured correctly, almost all of the growth in output can be accounted for by growth in input. Of the 3.59% annual rate of growth in real output from 1945-65, 3.47% can be accounted for by increased input, leaving only .10% to be explained by increased productivity. They conclude that advances in knowledge-human capital accumulation, the accepted source of productivity growth-are governed by the same economic laws as other capital accumulation-growth benefits from the accumulation of knowledge are not costless. These results imply that the social rates of return from research and development investments are similar to the returns from other types of investment." This further implies that "transference" or "spillover" effects of industrial-military R & D (see below) must be included in their benefits package if the social benefits of these investments are to match the social benefits of other types of investments. If these benefits are not forthcoming, then industrial-military R & D will yield lower social returns than other investments.

If the contribution of industrial-military R & D to productivity growth is suspect, then perhaps it can meet the challenge of the Seventies by contributing to qualitative improvements in living conditions. Certainly there has been a tremendous quality impact on the private economy from defense related research in the past. There have been numerous developments in techniques of economic analysis, production and inventory management, air and water transportation, chemicals, metals, electronic components and other materials, not to mention the most popular examples of Teflon, transistors, and computers. But can we expect the same degree of "spillover" or "transferability" in the future?

A negative answer to this question is motivated by several considerations. Much of the industrial-military R & D is related specifically to high technology space exploration. The gap between this research and its private industrial application is much greater than it has been in the past and it is widening rapidly. The speed of scientific and technological advance through aero-space research is much faster than our ability to assimilate it for private development; witness, for example, the impending problems of the supersonic transport. It may

2 For example, a standard linear regression of growth in real per capita disposable income (PDI) on growth in real aircraft and missile R & D (AMRD) for 1956-67 yields the following results:

PDI 3.19.10 AMRD

t= (3.62) (-1.15)

R2.13. D.W.=1.51

In terms of annual real dollar flows, the results are:

PDI 1701.4 -82.87 AMRD+86.92 (time trend)

t (45.94) (-2.63)

R2.93, D.W. 1.01

(6.23)

The negative coefficient for AMRD is indicative of the negative relationships estimated. The details of other experiments are given in the appendix.

3 Solo [6]. p. 50.

4 Jorgenson and Griliches. [2], p. 54.

5 Ibid., p. 56.

take the next decade of applied research just to develop and apply our existing storehouse of technical knowledge.

Finally, except for a limited number of cases where scientific developments have had direct and immediate industrial applications, space and military R & D represents an untapped reservoir of opportunities. It would seem that the companies that have made these scientific achievments would be best able to exploit them for private social benefits, but such may not be the case. In part because of the well-known inefficiencies of the federal government contracting and production system, many of these companies have had difficulty in developing the creative management expertise necessary to exploit their scientific developments under the cost and price competition of the market place and for the improved quality of life of the general public. As Exhibit II indicates, from 1957-1967 over 50% of industrial R & D was financed by the federal government. In aircraft and missiles and electronics and communication equipment, the figures were roughly 80 and 60%. Significant differences in management and production techniques between space-military and civilian business organizations make it increasingly difficult for the former to communicate the fruits of their research to the latter in an understandable manner that facilitates transferance of military research to private social benefits.

I conclude, therefore, that the present scope, direction and conduct of industrial research and development, especially military-space R & D, will seriously impair its effectiveness in meeting the productivity and quality challenges of the Seventies.

NEEDED: A NEW STRATEGY FOR RESEARCH AND DEVELOPMENT SPENDING

In the Seventies, the American economy will be characterized by a scarcity of qualified labor, continued inflationary pressures and relatively tight money. The resulting strains on our human, natural and financial capital will be heightened, as they have been of late, by the differences in attitudes between young and old. "Prosperity-scarred" youth are optimistic about their own personal economic security and pessimistic about our collective chances for qualitative social development. "Depression-scarred" older people tend to be pessimistic about their personal economic security and optimistic about our ability to solve our pressing social problems if only we are patient and employ reason and understanding in their solution.

In this environment, mental manpower will not be a free commodity. Its appliration to the solution of one set of problems means that another set of problems must go untreated, unless there is a transferance effect. For industrial-military research and development, this effect will be severely limited in the future. Consequently, we must reevaluate and redirect our research efforts if they are to be useful in meeting the social and economic challenges ahead.

This does not mean, of course, that government or the private sector should eliminate basic scientific research, but that we must understand explicitly how much human, physical and financial capital is devoted to its conduct and why. Many private corporations already seem to be reappraising their own R & D programs for these same reasons. The allocation of resources to basic research must be done explicitly and in accordance with clearly understood priorities, not left to chance or the naive assumption that no priorities are needed. The social costs and benefits of applied research in the areas of urban renewal, environmental control, welfare, low income housing, etc., must be carefully weighed against the costs and benefits of research in the interests of national security and space exploration. If the former set of problems should head our list of priorities (given some minimal national security needs), and I think they should, then addressing them directly is far more efficient than relying on nonexistent or increasingly difficult to obtain spillover from research that is directed elsewhere.

At the federal level, Congress can directly alter the flow of federal financing for R & D to the solution of different sets of problems. Perhaps this could be aided by public referenda on priorities for public expenditures, especially in "quality of life" areas. At the private level, the companies doing the research can be provided with incentives (positive and negative) to develop and/or communicate the private industrial applications of their military-space research. All this may well require a change in the current government contracting and production system to one incorporating more of the elements of private costprice competition. Government agencies should play a much greater role in the

communication of research results to private industry. In fact, the program of defense-space R & D should have built in a parallel or concurrent program to develop and exploit the breakthroughs for general public social benefit. Finally, civilian oriented industry has the responsibility to capitalize on this research. Its profit and social performance objectives can be efficiently served— to the nation's benefit-by exploiting this storehouse of scientific knowledge.

EXHIBIT I. RATES OF GROWTH IN SPENDING ON INDUSTRIAL-MILITARY R. & D. AND SELECTED ECONOMIC INDICATORS

EXHIBIT II.-PROPORTIONS OF RESEARCH AND DEVELOPMENT EXPENDITURES THAT ARE FEDERALLY FINANCED

EXHIBIT III.-STATISTICAL RELATIONSHIP BETWEEN INDUSTRIAL-MILITARY R. & D. AND ECONOMIC GROWTH (LINEAR REGRESSION COEFFICIENTS)

APPENDIX: STATISTICAL RELATIONSHIPS BETWEEN INDUSTRIAL RESEARCH AND DEVELOPMENT AND ECONOMIC GROWTH

A number of attempts were made to estimate a statistical relationship between spending on R & D and economic growth. In general they validated the conclusions in the text. The results of these attempts are presented in Exhibit III, where the following definitions are used:

GNP-Gross National Product,

PDI-per capita disposable personal income,
IOPM-index of real output (man-hour),

IIP-index of industrial productivity,

IRD-industrial R & D spending (deflated by the implicit GNP deflator), AMRD-aircraft and missile R & D (deflated by the implicit GNP deflator). A bar over a variable refers to the rate of growth.

Some simple distributed lags were also tried. The results warranted no change in the conclusions implied in Exhibit III.

BIBLIOGRAPHY

1. Economic Report of the President, 1970, U.S. Government Printing Office. 2. Jorgenson, D. W. and Z. Griliches, "The Explanation of Productivity Change", Survey of Current Business, U.S. Department of Commerce, Vol. 49, No. 5, Part II, May 1969, Reprinted with corrections from The Review of Economic Studies, Vol. XXXIV (3), No. 99, July 1967.

3. National Science Foundation, "Funds for Performance of Research and Development in American Industry", Reviews of Data on Research and Development, Selected issues 1959, 1960, 1962, (NSF 60-81, No. 24, December 1969), (NSF 61-51, No. 30, September 1961), (NSF 63-37, No. 40 September 1963). 4. National Science Foundation, "Research and Development in American Industry", Reviews of Data on Science Resources, Selected issues 1963-1967, (NSF 64-26, No. 1, December 1964), (NSF 66-6, No. 7, January 1966), (NSF 66-33, No. 10, December 1966), (NSF 68-5, No. 12, January 1968), (NSF 6912, No. 17, February 1969).

5. National Science Foundation, "Research Funds Used in the Nation's Scientific Endeavor", Reviews of Data on Science Resources, NSF 65-11, No. 4, May 1964.

6. Solo, Robert A., "Gearing Military R & D to Economic Growth", Harvard Business Review, Vol. 40, No. 6, November-December 1962.