National Academies Press: OpenBook

Papers Commissioned for a Workshop on the Federal Role in Research and Development (1985)

Chapter: Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities

« Previous: Impact of Defense Research and Development Investments on the Civilian Sector
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 265
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 266
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 267
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 268
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 269
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 270
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 271
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 272
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 273
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 274
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 275
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 276
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 277
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 278
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 279
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 280
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 281
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 282
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 283
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 284
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 285
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 286
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 287
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 288
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 289
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 290
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 291
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 292
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 293
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 294
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 295
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 296
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 297
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 298
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 299
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 300
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 301
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 302
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 303
Suggested Citation:"Measuring the Economic Impact of Federal Research and Development Investments in Civilian Space Activities." National Academy of Sciences, National Academy of Engineering, and Institute of Medicine. 1985. Papers Commissioned for a Workshop on the Federal Role in Research and Development. Washington, DC: The National Academies Press. doi: 10.17226/942.
×
Page 304

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

MEASliRING THE ECONOMI C IMPACT OF FEDERAL RES EARCH AND DE:JELOPMENT INVESTMENT IN CIVILIAN SPACE ACTIVITIES Henry R . Her~czf eld Consultant For over 27 years, the U. S . government has conducted scientific and engineering research and development (R&~) for civilian activities in ache exploration and use of outer space. Expressed in 1985 dollars, this investment has totaled over Sl5Q billion, with current federal civilian space R&D expenditures amounting to about $6 billion yearly. The space missions have been spectacular successes. Even though accounting for less than one percent of the annual IJ. S . budget, space exploration has generated a large amount of national and international attention and has captured Ache imagination of all mankind . Economic benefits to the nation were considered when Congress first created the National Aeronautics and Space Administration (NASA) in 1958. So were scientific benefits, engineering and technolo gical ~ eadership, and the s-timulatior~ o f ache educational system O ~ Nonetheless ~ economic expectations were secondary. Of primary concern to the United States was the shock to our political system created by the 1957 launching o f the Russian sate]Ll ice, Spu~cn:Ic I. We were behind and had to catch up in ache international technological and political arenas. Technologically, both the United States and the Soviet Union today are capable of comparatively advanced feats in space. These range from launching heavy equipment and sus Gaining man in a "shirt-sleeve" environment orbiting the earth for monarchs at a time, to performing intrica~ce scientific research and even manufacturing test quantities of substances previously impossible Deco create. But, politically, a race for prestige and dominance of space is still on, and much of ache investment made by the government is oriented toward keeping our image alive as a very advanced Macon on the cutting edge of new technologies. However, in times of budget deficits, there is increasing concern about the economic returns from R&D expenditures. Important questions are being raised. about the significance of the investment in space R&D. Each new space prod ect requires huge amounts of new capital. Since such expenditures should be based on economic as well 261 -

as political decisions, it is important to attempt to measure the re turns from the inves cment ~ This paper will review critically the maj or studies that have measured ache returns deco civilian space investments. As will be seen, those studies, conducted mainly in Ache m~d-1970' s, looked at a range of economic impacts, but they were not sufficiently accurate or complete to be definitive for policy decisions. However, they do provide excellent case studies of the problems that occur when economic theory and practice are applied Deco R&1) programs. This analysis of past studies should not be interpreted as indicating that the studies were useless or Chat future economic analyses should not be conducted The studies are educational in that they provide a framework for analys is that is rho t always considered by scientists and engineers in technological planning. The studies provide ins ights and ideas that can aid future go~rernment program managers in planning for private sector participation and eventual technology transfer. And, ache studies provide a benchmark for discussing ache appropriate economic criteria to be applied to future programs. DEFINITIONAL CONCERNS AND THE SCOPE OF STUDIES REVIEWED This paper will focus only on civilian space expenditures . Throughout ache his tory of space RED spending, the civilian agencies (primarily NASA) have dominated. In ache early 1980' s, military space R&D spending overtook civilian space R&D spending, arid, with the new Strategic Defense Ini~ciative, the trend will be toward an even greater share f or military space R&D . A close look at the total U. S. space budget retreats that, in addition to research and development, there also has been a very large amount of military procurement of miss iles and space equipment. That military spending represents space industry investment, even if accounting prac~cices do not label it as R&D. It is highly probable chat a good deal of military space R&D has been incorporated in other programs over ache years and not reported in the space R&D budgets. It is virtually impossible to calculate the amount of interaction between research findings from military space research and from civilian space research. There is a continuous flow of people between jobs in both sectors, and many joint research programs between NASA and the Department of Defense share research knowledge. In addition, the same aerospace companies that perform the ma] or work for NASA do closely related world for ache military. For ease of calculation, these in~cerrelationships are ignored by economic s tudies . It is easy to separate military and civilian federal spending, but that separation is artificial and hides the ir`teractions and sharing of knowledge. - 262 —

The Nary anal Aeronautics and Space Administration is ache dominant civilian space agency. There are expenditures in space R&D by the Department of Commerce (National Oceanographic and Atmospheric Administration) for weather and remote sensing satellites, as well as some space -,:ela~ced R&D performed by the Depar~cmen~c of the Interior, the Department of Agric~cure, and the Department of Energy. As these are quince small compared to ache NASA expendi~cures, they wil' not be analyzed separately in this paper. Often, the economist is asked for ache "bottom line" return-- usual~y expressed as a ratio such as lO deco ~ or 500 Deco 1, meaning ~cha~c ache benefits exceed the costs by some multiple However, s ince economic returns are not the primary reason for space investments, no economic measure or calculation possibly can encompass the entirety of the returns to space ~ nvestment . Such a calculation should no~c even be attempted for evaluating government R&D expenditures. Invariably, such numbers appear in official and unofficial documents. Most of the time they are misquoted and misused, which serves to limit further ache already limited credibility of economic studies. Most space R&D is on the dearer opment end of the spectrum. Space hardware is expensive. The engineering and fabrication of space vehicles takes the bulk of the NASA budget. Theoretical and scientific research, although an impor~cant part of the federal space effort, does not represent a very sizable fraction of the budged NASA estimates that about 91 percent of its budget is for development, and only about 9 percent for basic research. it is difficult to define the Space industry" clearly. The sector is too small and coo new to be given a separate S tandard Industrial Classifica~cion (SIC) code. ~ large portion of commercial space activity falls under SIC 3769 guided missiles and space equipment. In addition, significant portions are classified in SIC 3662, communications equipment (satellites); SIC 73, business services (computer programming); SIC 35, office equipment (computers); SIC 3S, instr',ments (measuring devices); and SIC 372, aircraft and parts. And, as more and more industries are brought into space research, "space industries" will include parsecs of almost every maj or manufac~curing and service sector. Where possible, this paper will" analyze the benefits of space R&D to the overall economy. Specific studies that have focused on the impact of space on a particular sector, region, or employment category will be noted and in some cases summarized. In-depth studies of some sectors ~ such as telecommunications ~ are far too complex to include in this review and have been discussed ex~censi~rely in other papers. Furthermore, most of the results of those studies do not lend themselves to comprehend Eve or overall measures of economic activity. - 263 -

This paper will focus on the analyses of long- te ~~u technology- enhancing and productivity impacts on the economy rather than on studies of shorty term impacts . Many studies have looked at the employment generation and income multipliers of the space program. But these factors are no' much different from those that would occur with any current government expenditure or tax cut. (Research has shown space expenditures to have their own unique geographic distribution and emp loyment impact 9 but the aggregate short- term multipliers a~e3si~nilar deco those of other government expend) cures . ~ ~ The lasting and significant economic impacts are those that enhance technology, are spun off into new privately produced products echoic increase productivity, and affect the quality of life. Once new knowledge is acquired and used it cannot be taken away, whereas short- term impacts last only as tong as the government is spending the money. ECONOMIC MEASUREMENT APPROACHES Appr IED TO SPACE R&D Three dis Inch approaches have been used deco quantify the economic impacts of space R&D. First is the adaptation of the macroeconomic production function Cody to estimate long- term effects of R&D on aggregate economic indictors such as gross national product (GNP ), employment, and inflation. The results of using this model are expressed both as ~ rate of return to a given investment and as cumulative changes affecting the indicators. Second is the microeconomic model, used to measure the direct and indirect expected benefits in the economy from a particular in~ren~cion or innovation. This model is based on a consumer surplus theory, and the resultant measure is either a benefit/cost ratio or an estimate of overall benefits . Third is the examination of data from NASA and other sources that provide evidence of the dirac~c transfer of technology from federal space R&D programs deco the pri~ra~ce sector . His tonically, technolo gy utilization, pa~cen~c licensing, and patent waiver programs have been aimed at stimulating the further development of space technology to create specific inventions with commercial potential. In addition, there are programs oriented to disseminating information about new technology to bus iness and ache general public . But, more recently, there is another List to transfer activity. Space act:vi by is maturing and is becoming an investment opportunity in its own right. Data are now available on the number and types of priorate firms that are investing in space launch vehicles and satellite systems, space-rel~ced support services, and space research and also on those companies that are planning for space manufacturing at a commercially opportune time. This indicates the growing economic impact of space activity and sets the stage for more direct and explicit economic analyses of the space industry in ache future. - 264 —

.9L-CRO ECONOMY C A:>JALY~ 1 S Me strengths and weaknesses of the macroeconomic models are well documented in the economic literature. Some au~chors, such as Griliches, 4 feet that ache macroeconomic approach has validity, and its imperfections can be cured. Others argue that the assumptions behind the model preclude any accurate result ts (see Benison and Mansfield ). Although it is possible to calculate aggregate returns from space R&33 through this model, and the results reported below have shorn very robust returns, the escapes have not stood up to statistical tests for accuracy or reliability. Me Prague of the aggregate production function approach is more in structuring thought and providing a framework for analysis than in quantifying results. The aggregate production function model assumes that a formal relationship exists between it&) expending ures and productivity. Given ache many case studies of specific new techno] ogical developments and their impacts on the economy, there is no doubt that such a relationship exists . However, the aggregate production function model skips a number of steps. Expenditures for Rho may be an early stage on the road Deco producti~ri-~y. But that stage must be followed by the creation of new knowledge, which then will lead to ideas, inventions, innovations, and, eventually, with proper marketing and distribution, to commercial products. Then, those produc~cs must be disseminated so that they can be used. It is their sale and use that translates into changes ~ n GNP and productivity. Further, some productivi~cy- enhancing inrtova~cions do not come from any formal R&D programs, j ust as certain economic benefits, such as improvements in the quality of life, are root included in standard economic measures even though they are products of R&D expenditures. Assuming that any given set of R&D expenditures results in all these missing steps is taking a giant leap without looking. From other doc~en~cation, it is irrefutable that NASA RSd) has had a positive impact on long--cerm productivity. But the macroeconomic model does not offer any new proof of this. Further, the regression equation used in this analysis shows only the changes in productivity that may be associated with changes in R&D spending. It does not determine the direction of causality. Another problem with the macroeconomic model as applied deco an Red) context is that R&D is a very small expenditure in ache economy. It has varied from a high of nearly 3 percent of GNP to a low of 2.2 2ercen~c over ache past 15 years; NASA R&D has been about 0.25 percent of GNP. Attempting to estimate empirically the impact of such a small fraction of GNP on ache GNP is very difficult, given that there is a margin for s tatis tical errors in ache GNP that is greater than the NASA R&D component. A pitchfork is being used to f ind the needle in the haystack instead o f a finely tuned stethoscope . — 265 -

Yet another factor in the unreliability of macroeconomic es climates is ache long time frame needed to deve lop, marice t, and disseminate new products successfully. Although new technologies may be reported in 2 years or less, it easily can Cake 20 years for a maj or product to mature and make an impact on the economy (This point is discussed later in greater detail. ~ then most of the macroeconomic analyses reported below were conducted, in the mid~i970's, the available rime~series data for space R&13 were barely 15 years long. This alone should make one question ache robust results . FinsIly, this discuss ion has focused only on the specific problems with the macroeconomic approach to measuring R&l). The economic li~cerature reveals a host of other technical and data problems with such models. These will not be reviewed here. Mad or .Macroeconcmic Studies Using Secondary Data Sources The Midwest Research Institute Study In 1971 the Midwest Research 1 Institute CORD ~ conducted a study of the economic impact of NASA Rap. Much of the study was a subj ecti~re review of cases of successful technology transfer, but a maj or effort was wade to calculate the macroeconomic impacts. The study took a national income accounting approach, following ache methodology of Solders pioneering econometric work in analyzing the reruns deco R&D. The TORI investigators firs t looked at total national productivity changes in the economy, then subtracted those changes that were attributable directly to capital ant labor. The remaining unexplained changes (the nresidual") were scudded further to find their important components. After accounting for changes in demography, education, health, workweek length, and economies of scale, the investiga~cors at~cributed the rest of ache residual to advances in knowledge, or the R&D component . A simple least- squares regression then was used to estimate the linear relationship between gains to Red) and a weighted see of past R&D expenditures. lathe results showed that $Z07 billion would be returned to the economy through 1987 from the $29 billion NASA outlay between 19S9 and 196 9 . This bans faced to a 7 no ~ overall return to NASA, or an estimated 33 percent discounted rate of return. The MRI report was ache first such attempt to measure NASA' s returns in the aggregate . Two maj or assumptions made in that study should be questioned. The first was that NASA R&D was not sepasa~ced from any ocher R&D in the economy. In fact, what MR! did was to calculate the returns for total R&D ~ federal and private) and assume that space R&D was similar deco all other R&D, only scaled down by the proportion of NASA R&D outlays to scowl U.S. R&D outlays. — 266 —

The second ass~p lion was that Red) has an 18 -year lifetime from outlay deco "death" in the economy. After that lifetime, no re~curns were measured . Fear ~her, each R&D expenditure was cons idered to be independent of all other R&D expenditures. Giving R&D, which in these macroeconomic models is a surrogate for knowledge, a finite lifetime is unrealistic. Knowledge is never lost, it becomes the building block for further technological Cased producti~ri Gyp advances . The MRI s tudy did no t acknowledge the important, but subtle, distinction between a product's economic life cycle and the impact of society' s accumulation of knowledge. he Chase Econom.e~crics Associates Study In 197S, Chase Ec gnome tri cs ~ di d a far ma re s oph i s t icated ec gnome try c analys i s o ~ NASA spending on space R&D and the GNP. Chase analysts estima~ced a production function for the United Stances based on a potential GNP Dime series. (the potential GNP is not the she as the published figures- - it is an estimate of what ache GNP would be if full employment existed. ~ As there are several different sources of such estimates, ache choice of the particular time series will influence the results of ache analysis significantly. This was one problem with the empirical results of the Chase study, but, as described below, i_ was not ache maj or problem. Chase then calculated the res idual, the variations in the potential GNP that could not be explained by the capital and iaoor inputs . This res idual. was broken down statistically into its components us tng a regress ion technique . The independent variables used in this regression to explain changes in the residual included: NASA R&D, other R&l), industry mix, capacity utilization ranchos, and demographic factors. These variables were suggested by the earlier work of Denison and others who have done extensive analysis of the determinants of productivity.. Chase also built into the etodel various lag functions to accoun~c for the de lay be tween R&D outlays and new technology being developed. Chase experimented witch about 60 variations of the equation and chose the one that appeared "bests deco them. One problem is that with only 15 years of dime series data on NASA expenditures, the use of many independent variables in ache estimating equation restricted the ability of the model deco give accurate statistical results. Further, it is not clear that spurious and coincidental business cycle relationships witch NASA R&D were not piclced up by the shor~c acme series. lathe selection criteria of the "best" equation were not well documented, as an examination of the n t" ratios for reliability of ache regression coefficients even in Chat best equation retreated that some were marginally s ignif icant . Finally, any equation of this sort is driven not by the R&D variable, but by the capac~4cy utilization rariabie . — 267 —

lye calculated cumulative "productivity" return to NASA R&D was 14 to i. This ~cransla~ced into an annual discounted rate of return of 43 percent to NASA outlays. Chase analysts then took the results of the equation measuring producti~ri~cy enhancements and entered it into their macroeconomic s imulation model o f ache entire economy . From that model, es ~cima~ces of ache changes in future GNP, infla~cion, employment, etc., were proj ec~ced, based an hypothetical changes in NASA expenditures . Specifically, they calculated that a $l billion increment ~co NASA expenditures for each of ten years from 1975 through 1984 would augment GNP by a ctlmulati~re $83 billion by 1984. In ocher words, $10 billion of extra ~ nves~ment by NASA in space and related research and development would yield ~ . 3 times that amount in GNP benefits . Me U. S . General Accounting Office (GAO) ti performed an extensive review of the Chase study at the request of Senator foxfire . Not surpris ingly, GAG found a great deal of instability in the equation used by Chase O Although not supporting the results and not advocating the use of the aggregate production function modes for this purpose, the GAO did find the study to be interes~cing from the rieWpoine of experimental and new work in empirical economic analys is . In 1980, at NASA' s request, Chase was asked to replicate the methodology and use an updated and longer time series ~ see Cross ). The original data and choice of measurement ~cechraiques behind ache parameters also were subjected to new and more careful scrutiny. It was impossible to repeat ache earlier study exactly because all of the time series available had been updated and changed. In addition, a close took at the origins ~ variables showed that some of- them were not well cons~cructed and did not measure exactly what they were supposed to measure. Some minor modifications were made to the data for purposes of accuracy. The results of the Incests of ache equation to measure product:~rity changes from NASA R&D spending proved not to be s Scotia epically different from zero. Of course, this does nor mean that there has been no economic re scum on NASA R&~) . I t shows only that due deco ache theore~c~cal and data prob lems witch the macroeconomic model and data se~cs available, this approach to finding aggregate economic returns to R&D expenditures is difficult at best, and probably impossible. Me U. S. . Department of Labor Study Over the years, interindustsy input - output mode is have been used to measure the multiplier- type returns to various defense and other maj or national programs. With this in mind, NASA asked the U. S . Bureau of Labor Statistics to attempt to measure the productivity changes resulting from NASA expenditures at an industry level, using the production function approach and an equation for every incus try that is tied closely to aerospace production. — 268 _

From che i~icroeconomic analyses that have been conduc ted on space - related innovations (see below), it is known that many important ir.~.o~rations stem from space research and development. And, generally it is agreed ~ in spice of the paucity of data) that space has contributed to ache aggregate economic growth of the nation. The Bureau of Labor S tatistics study was intended to bridge the gap between studies of the benefits from specific innovations and studies of the entire economy by measuring the returns at an industry level, which is aggregated more than by product or firm, but is far less general than total GNP estimates. National data disaggregated to ehe industry Bevel are more susceptible to error and sta~cistical "noise " than all-indus~cry ~co~cais. The results reported below showed so much statistical variability Chat any conclusions drawn were extremely tentative . However, there were several interesting overall results on the initial runs of the model for the aggregate U. S . economy. First, overall U. S. technology change was found to be labor saving rather than neutral, the general assumption in such models. Second, the R&D component of technology change has saved capital. I-hird, the returns measured deco priorate R&D inures tments were between 15 percent and 30 percent, while returns deco government R&D were between zero and 5 percent . The equations that attempted to isolate NASA R&I) impacts di d not show the expected pos incite results . In fact, they showed no consistent pattern. For three industries, the R&D labor coefficients were greater than zero; for 16 industries, they were no~c statistically different from zero; and for the remaining 33 industries, they were less than zero. All tests were at a 90 percent confidence leered . the Bureau of Labor S tatis~cics attributed ache measured lack of pos itive returns to NASA research to several factors, most of which were not diss imilar to the problems with the Chase measures c First, overall, NASA' s capital stock changes were running counter to labor productivi~cy changes in the relatively short time period ( 196001975 studied. This was probably just a coincidence, not ~ pattern Chat implies any causality. Second, the business-cycle periods were too shore the beginning of the study was a peak and the end of ache study a trough. Third, ache NASA R&D variable may be more represent~ci~re of tlie government demand for output than of actua ~ advances in knowledge . Fourth, as with the Chase study, the capacity utiliza~cion and output variables drove the eq~ tion . A surrey Approach to Macroeconomic Analogs is: The European Space Agency S Judy In 1978, the European Space Agency (ESA) releaser a study of the economic benefits of its contract ~ see Fitussi and - 269 -

Neider~ui: ) . Tile s Judy, prepared by the Faculty of Economic Sc fences at the Louis Pas teur Univers icy, S trasbourg, France, took an entirely do fferent approach to measuring overall economic benefits from research and development in the space sector. The study team conducted an extensive se~c of interviews wit h ache maj or industrial contractors to ESA aimed at assessing the Prague to the contractors of doing advanced technological work for the space agency. Benefits were grouped into four maj or types: technological advan~cages, cormse~c~al advantages, advantages for organization and methods, and ad~ran~cages for the work factor ~ labor productivity from an increase in personnel knowhow and maintenance of a highly skilled production teams. The results were made a~ratIable in ache aggregate and by industrial sector. The overall return calcula~ced was a benefit/cos~c ratio of 2. 7 to i. Cost was derived as the value of comrades let to she various con~cracteors The time frame was 1960 to 1980 (wi th the 1977 deco 1980 benefits estimated by the respondents). Since not all economic agents knee effects of competition, interindustry rela~c~onships, etc. ~ were analyzed in this ratio, ache study team did not consider the ratio deco be an &-encompassing financial return on investment. Nevertheless, much of the literature has in~cerpre~ced the reecho as such. Again, this illustrates the problems that arise with the misuse of ~single-n,'~bern measures of economic returns. Of interest in ache ESA study is the distribution of benefits by category. Benefits due deco technology Cal gains by the contrac~cor companies were estimated to be 22.4 percent of the total. Those attributable to commercial success ~ increased sales due deco stringent qualifica~cions for space products and to better international collaboration) were 26 . ~ percent of the tomcat. he benefits from organization and methods (improved management) were 16.9 percent of the total. And, the benefits from the world factor were 34. 6 percent They also found echoic about one third of the benefits Chat contractors experienced were related directly to their space activi~cies, while about two thirds of the benefits were in other activities. This is explained by Nero factors: (~) the rela~ci~rely small part of Coccal activities of the firm devoted deco space ~ es animated to be about 3 percent and ~ 2 ~ the large amount of benefits attributed to the work factor, which makes i~c relatively easy to transfer benefits from one unit of the company deco another. Also, looking at another aspect of technology transfer, the study Indicates that one sector in particular, on-board electronics, resulted in its benefits distributed among many differen~c products. The ESA study might be described as somewhere between a macroeconomic analys is and a case or macroeconomic impact study. To the extent that it surveyed over 80 percent of ache contractors and to ache extent that it was performed in an unbiased manner, ache results — 270 —

should be as accurate, or more so, than those studies using secondary data sources. But, as the study teem acknowledged, many tnacroeconcmic and competitive factors were not analyzed. lathe es tima~ces, therefore, mus t be viewed as partial measures, perhaps biased on the positive side because of the omissions. `~OR MICROECONOMIC STIES Microeconomic analysis focuses on the economic theory Chat explains ache behavior of the fire and the supply and demand relationshi p for individual products. Benefit/cost analysis to evaluate ache success or failure of public sector proj acts is based on this theory. More often than not, individual proj ects are analyzed to determine their consumer surplus. This "surplus" is the measured value representing the additional " income " to ache consumer when the sale price of ache commodity is less than the consumer is willing deco pay. This n savoring" is the benefit in the benefit/cost ratio. The cost is the cost of producing the product. Because demand and supply curares are measured at a given point in time, these savings can be translated into streams of income over time by re-estimating demand and supply curves at different times in the life cycle of a product. Benefit and cost estimates are Sheen proj acted into the future, and each series is discounted at some specified rate, which gives the familiar benefit to cost ratio. Presumably, if the ratio is greater than unity, ache in~res~cment is worthwhile to undertake from a public policy viewpoint. Ranking several proj epics by these ratios is one way that pro orate es sometimes are set among competing projects. Many explanations and critiques of this method have been wri~cten, and it is beyond the scope of this paper deco review the Erase literature on benefi~c/cost studies. The most familiar application of benefit~cost analysis is an regulatory impacts. Using the method Deco analyze fiche imp act of R&D and new technologies poses some unique problems. For instance, benefi~c/cos~ analysis is predicated on accurate measurement of the demand and supply curares. Often, new technological innovations result in products that are so different from anything on the market that no well-defined demand curves can be constructed. Nor is there a stable supply function (which measures the costs of production), as costs and production processes Change rapidly as the product evolves (see Kochanowski and Hertzfeld ). Therefore, using this me~chodology for evaluating some R~ proj ects involves little more than advanced guesswork . Measuring appropriate costs can be another serious problem when applying benefit/cos~c analysis deco specific innovations arising from large programs such as those of NASA. Should the denominator reflect — 271 —

solely the cost of transferring the technology, the cost of the research pro; en c, the cost of the group of related pro: acts, or the entire expenditures of the agency? With some proj ects that result in unintended (or spinoff) technologies, ~ good case could be made for any of these cho ices . In such a large R&D agency as NASA, many new technologies are discovered and deve loped over time . Due to limited resources, information, and time, an economist or analyst must choose only a rela~ci~rely few to analyze. Usual y, only the "winners" are selected Of course, measuring only a sample of successful technologies cannot give a balanced picture of the agency's work. Yet it is not uncommon to see the benefits of one or several technologies touted as indicative of all of the new technological advances steaming from a particular government Rat) program. Not on: y have unsuccessful technologies been ignored, but externalities that may have either positive or negative effects on society usually are not considered. Those ex~cars~alities may be derived from either. successful or unsuccessful insto~rations. Another problem, more general to ache me~chodology than to R&D s i tuations, is the cho ice o f a proper discount rate and a repr~senta~ci~re length of time for measuring benefits. Many subj entire decisions and estimates concerning potential uses, alternative products and markets, and ache races of diffusion must be made . In spite of se: the problems with benefit/cost analysis (and this list is by no means exhaustive), economists are rela~c~vely comfortable with estimates derived from well-execu~ced benefit/cost studies. The appeal is primarily the ability deco derive measures from readily available and easily documented sources coupled with a theorem that has been a keys tone of economic analys is for over a cen~cury . But, this is a theory of comparative statics. To analyze flows over t~e adeq~a~cely, a dynamic analysis would be far more appropriate. At present, a dynamic economic theory is not developed well enough deco solve ache problem, although efforts Eye being made to apply Dynamic models deco R&D stations (see Nelson and Nelson and Winter ). Over the years' NASA has sponsored numerous benefit/cost anct related studies. Many of them have taken a cost-effectiveness approach to problems, since their major purpose was deco choose between alterna~ci~re methods of achieving a give result. The planning study done in 1970 (see Heiss and Morgenstern 9' ~ deco determine some of the economically desirable characteristics of the space cransportation system was an example of the use of this model. S ince the purpose of this paper is to report on the overall economic impacts of the space program, only two benefit/cos~c studies will be analyzed. Both attempted to measure a summation of benefits of specific technologies and technology transfer from the government deco the price economy. As such, they differ from most of the — 272 -

NASAL sponsored economic benefir/cost studies that are used in ache detailed analys is of a specific program. 1 We Ma chema~c i c a l S tilde In 1975, Mathematica, Inc.,21 studied four successful technologies than NASA had a role in developing I cryogenic insulation, integrated circuits, gas turbine engines, and NASTRAN (a computer program for analyzing structural properties of large vehicles). Each analysis required a different type of data gathering and statistical modeling. But, the studies used similar methodologies so that the results could be added to get a son for the benefits from all four innovations . The benefits measured were not the total impacts on the economy from these innovations. Instead, the study estimated the speed in ache introduction and commercial use of these inno~rattons that could be traced to the special program requirements of NASA's space R&D procurement specifications. Mathematica focused on hero measurable elements: the introduction of a new commodity and the decrease in ache costs of production for an existing good. The study was not ~ true benefit/cos~c analys is, as the cos ts were not calculated- - a handy way of skirting the difficult choice of determining acne proper yardstick for costs. It was impossible deco attribute or alloca~ce costs of developing the different technologies, particularly when NASA' s involvement was not only as a funding source of Rho, but also as a stimulus to primate firms already investing in the technology development c And, it would have been impossible to separate the specific NASA funds that went deco the technologies, since many of them crossed over various space R&D programs . S till, ache methodology used to measure benefits was cons istent witch traditional benefit/cost analysis. The results were impressive. The investigators found that, over a ~cen-year period from 1975 Deco 1984, the four technologies could be expected to return a discounted tori of $7 billion (cons~can~c 1975 dollars) in benefits that were attributable to NASA' s in~rol~rement in their development. To put that in perspective, NASA t s entire budget outlay for 1975 was $3 .2 billion. The 1971 Mathtech. Inc.. Technology Transfer Study For many years, NASA has had a formal technology transfer program to disseminate information and develop space technology for nonspace uses. In 1977, }2athtech, Inc., conducted a benefi~c/cost analysis deco measure the economic impact of nine of the more successful commercial innovations transferred from the government Deco the private domain through this program. — 273 —

In the Mathtech ~ tudy, Eve benef i ts were compared to cos ts, but, since ache analysis was oriented to the costs of further development and transfer of the innovations rather than to the costs of initial development of the technology, there was ~ tendency deco understate the true costs, making the ratio of benefits deco costs biased on the high side . Nevertheless, a valid argument under the "but for's rule can be presented to j ustify this method of assessing costs O It Is reasoned Chat if NASA's Technology lI~cilizat~on Program had not existed, then chose innovations would still be sitting on the shelves, buried in government documents, and would not have been developed comnlerciaily or would not have appeared as available products as rapidly as they did. Since ache costs of developing the initial technology would have been funded anyway because of the mission- oriented purposes of the space agency, then the true costs of bringing the innovation into the commercial world are ache costs of transferring it. Thus, "but fore the existence of the transfer program, the benefits never would have occurred. .Mathtech attempted to compensate for this overstatement by separating ache benefits to the economy that could be attributed deco the transfer program from the total measured benefits of the innovations . Where the innovation still was under development, ~ probability was assigned to ad; ust the ratios for technologies Chat had not yet reached their commercial potential. This use of a probability estimate of the commercial potential is an in~ceressis~g way of overcoming ~ serious weakness of benefit/cost analysis when applied to R&D act~,rities. That is, benefit/cost analysis is more appropriate for analyzing regulatory impacts and other government actions when they involve known, existing, well-established marke~cs for clearly identified goods and services. New technologies from an R&D program may be path breaking, both in production processes and in new products for which there are few, if any, substitutes . S tatic demand curve analyses of such RED outputs, particularly in the early s~cages of commercial applications, can distort the results greatly. The results from ache nine innovations studied by Math~cech showed a s izable variation. In ache biomedical field, the cardiac pacemaker had a benefit/cost ratio of 4 deco I, while the laser cataract tool had a ratio of 41 to I. Other biomedical results were: burn diagnosis, 8 deco I; meal systems, 6 deco I; and ~ human tissue stimulator, lO to I. For engineering innovations, benefit/cost ratios for the nickel-zinc battery were 68 to I; zinc-rich coatings, 340 to I; tracLc-train dynamics, 3 to l; and a firefighter's breathing system, 4 to I. Among the primary reasons for the wide range in ratios are theses The technology was not yet fully ready for the market; the markets for equipment that will be used by governments ~ such as the firefighter system' were smaller than for commercial products (such as paint coatings ); and the noneconomic benefits (particularly in the biomedical field) were difficult to quantify. _ 774 -

Lice 2rincl~al advantage of this type of benefit/cost analysis is chat s imilar methods were applied to all innovations, and the re suits can be compared. The maj or drawback is that the study was done only once, in 1977. No follow-up was attempted, and, therefore, these one-~cime estimates for a limited set of successful and potentially successful technologies cannot be verified. Now, eight years iater, it would be interesting Deco review these estimates and update the dance. An updated study would not only check the methodology, it would begin to shed some light on the value of doing these types of studies and would improve the methodologies for future ~cechuc,logy assessments . Other Microeconomic S tudies As discussed above, many other types of benefit/cost studies have been conducted by NASA. The two described- were singled out because. they used a formal consumer surplus model to measure benefits and costs, and because they used similar methodologies for different cechnologies, which enabled the results to be sunned toge~cher. Hundreds of technology assessment and technology forecasting studies have been conducted by NASA for specific aerospace technologies . Many were subj ec~ci~re, quali~cati~re evaluations rather than fo real economic analyses . Because many space R&D outputs are at the cutting edge of technology, subj active methodologies were the only ones that could be applied at the time. It is ~forturtate that the results of those analyses could not be aggregated to give a broad picture of the imp ac ~ of space- related technology on the economy. Of current interest to NASA is the use of space for new research arid commercial ventures . A sizable library of bus iness and financial analyses of numerous candidate technologies for private investments in space is accumulating. Some of ache information is proprietary and some can be found in the open literature. A later sec~cion of this paper, which looks at the types of economic analyses of curren~c importance in the space R&D community, will discuss such studies. ACTUAL MEASURES OF NASA'S TECHNOLOGICAL AND i:CONOt£IC IMPACT The National Air and Space Administra~cion keeps records. Some of them include actual counts of the uses and benefits of various space technologies; while. some reflect results of surreys of users of NASA technology. This section will review analyses of these duct bases, which are maintained primarily by three different NASA technology transfer offices: (~) the Patent waiver Office, (2) the Patent Licensing Section of the Office of the General Counsel, and (3) the Technology Utilization Office. The common link between scheme offices is their ob: ective of s~cimula~cing private use of government-developed space technology .

At best, the scary statistics described below present a `general picture of ache amount and variety of commercial activity spun off, or derived, from space technology. Otrec~c comparisons of these data should be avoided, as each program has different users, different goals, and different institutional limitations. Definitions of industries, of what constitutes commercial success, and of measurement values vary considerably among programs. Evidence on the Time Span for Technology Transfer to the Private Sector Data has been collected by NASA from several of its technology transfer programs on the length of time it takes for inventions and innovations to be used commercially. Through its "new technology clause, ~ NASA requires every contractor to report discoveries and inventions. A correlation analys is of the annual NASA R&~) contract outlays witn the aggregate total of inventions reported ~ legged two years ~ showed a correlation of 0.97. In other words, on the average, the-e is a two-year time span from the first award of funds deco a contractor unfit irrventions are reported to ache agency. Of course, the idea for the invention may predate the funding, j ust as the development of the new technology into a useful product may cake many more years or may not even occur. Me two-year lag is evidenced further by data collected through the NASA patent waiver system. (As discussed below, a waiver may be granted upon application to NASA when a con~cractor desires permission deco seek a patent on a specific invention discovered under contract. ) An analys is of the data for 87 inventions indicated that the time span be - -een the first award of NASA funds deco ache contractor and ache date of first reported commercial us,3a~rerages 5.2 years, with a range of ~ to 15 years ~ see Her~czfeld ~ . However, the same set of tla~ca shows that ache time between the NASA grant of the patent waiver and ache first commercial use averages 3 . ~ years . Therefore ~ the difference- - j ust over two years - - indicates the same ~cime lag as is shown by the aggregate data analysis for ache period between award of funds and the report to the agency of new technology . Many new technologies, particularly those that result in. new products, can take 20 or more years to become diffused throughout the economy . The transmiss ion of data and other information from satellites to the ground was demonstrated in the lance 1950's. But, it cook unfit the lance 1970's before the satellite communications industry was developed fully. During the 1960' s, many experimental go~rernmen~c communications satellites were launched. By ache late 1970's, private companies were designing, constructing, purchasing launch services for, and operating communications sat" 1li~c~s and — 276 —

re laced "erres trial equipment . Financing and insurance for these proj ects also was from private sources . The communications industry has been in an ideal si~cu=~=on to develop new space products quickly. Large firms with fa~rarable financing arrangements already existed in ache industry; ache consumer product was in place with an established demand and distribution network; and the industry already employed many highly skilled scientific and technical personnel. Even with all of these favorable factors, 20 years elapsed between technological demans~cration' and economic diffusion. Patent Starch sties Z~i Griliches, 24 among others, has suggested using patent statistics as an indicator of the output of research and development expenditures. Although they are only a partial measure of R&D output, patent statistics do provide an easily available and documentable data set to analyze. AL important note must be added concerning the definition of "commercial success n as it has been applied to patent -use statistics. This term is defined as the use of the invention or innovation by a firm in its production process or by the marketing of a new product based on the new technology. Success may be measured by production cost reduction or by an increase in sales. Success does not imply profits, general use in the economy, or diffusion of the product. Basically, it means that the reporting firm developed the technology to the point where it was used in a production process or put it on the marice~c for sale. NkSA Patent waivers Sec~c~on 305 (a) of the National Aeronautics and Space Act of 195B, as amended (42 USC 2457~, requires that inventions made by contractors under RED contracts with NASA "shall be the exclus Eve property of the United S tares . . . unless the Adminis trator waives all or any part of the rights of ache U. S. co such invention.... n A Pore recent law, PL 96-517 (35 USC 200 et seq. ), effective July I, 1981, allows a contractor who meets the definition of a small business or nonprofit organization the first option to retain Circle to inventions made under contract, provided Ache contractor follows certain procedures described in the law. Such patent waivers can take ewe forms: advance or specific waivers . Advance waivers are granted to a contrac tor upon application, usually in the early stages of the contract before any inven~cions are discovered. The contractor then has blanket permission to apply for his own patents on those inventions, although he must still. report findings to NASA, and ache government retains royalty- free use of the inventions . Specific waivers are granted on particular inventions after an application and subsequent review by the agency. — 277 —

The author tsee Hertzfeld25) conducted an analysis of the industrial distribution of NASA' s specific patent waivers for the years 1961 to 'S75. The commercialization rate (to Cal commercialized inventions divided by total waivers ~ by industry appeared relatively constant, averaging 20.8 percent and having a range between 15 and 26 percent, with the exception of the n al.1 other' category. However, the distribution of inventions among industries shows much greater variation. Electrical machinery, communications equipment, and ~nstrumen~cs together accounted for over 69 percent of all commercialized specific waivers . The transportation equipment incus try (which accounts for approximately 60 percent of all NASA outlays) had only 7.3 percent of commercialized patent waiver innovations. This may be explained by the nature of ache industry itself. Transportation equipmen~c, particularly the space segment, is oriented primarily toward governmen~c business. It is not as fast growing as the electronics indus~cry, and the value of patents to ache firms may be less. In addi~cion, various corporations within indus~cries have different policies toward patenting inventions. Some take out every patent they can au~coma~cically, while others take ache pos itiorl that pa~cen~cs are not worth ache effort and publicity O Further analys is would be needed to explore the reasons why the industrial distribution of patent wai Hers of NASA technology is so different from the dis tribution of NASA R&D expenditures . In the years before 1975, a nuder of studies were conducted on ache com~ercializat~gn of indentions from NASA2~atent waivers ~ see Watson and tolman, Solo, Harbridge House, Kasko~rich 5 and Wrights ). It is interesting so notice that the reported rate of commercialization of such inventions has increased steadily over time . This could indicate one of two things: Et ther the space technology is maturing and more commercial uses are being found for earlier discoveries, or ache study and data collection methodologies have improved. Corer the past ten years (1975 to 1985), ache race of commercialization of specific waivers has stabilized at a rate of 18 to 20 percent ~ see Hertzfeld 1 ) . This may have occurred because some maj or contractors have shifted Deco applying for advance waivers rather than specific waivers. In recent years, the administrative process of granting advance waivers has been s imp lifted and streamlined no the" these waivers are granted almost automatically. Me ferreting of the rate of commercialization also may be due to the fact thatch since 1975, the survey ins trumen~c has been more uniform anti data analysis has been more consistent. NASA Licenses Mach exclusive and nonexclusive licenses are issued by NASA o During the 1960' s and 1970' s, the policy was Deco issue nonexclusive licenses, asked grant an exclusive license only rarely. However, NASA now recognizes Chat businesses often need property — 2/S —

right in an invent ion to secure financing and that they also need the incentive given by a limited monopoly right so Chat a return on the investment can be accumulated before competitors enter the market. Therefore, NASA has changed its policy and now will grant exclusive licenses more frequently. Currently ( 19 84 ) , NASA owns 43, 23 8 patents and patent app lications that are available for licensing. Where are 339 nonexclusive licenses in force on 217 patents, plus 60 exclusive licenses. ~ current analyst s of the commercial success of NASA- issued licenses is not available. But the anchor did analyze licenses for the years 1959 to 1979 ~ see Her~czfeld ~ . At that time, NASA owned 3, 512 patents and had issued S23 licenses on 218 patented inventions . Of those, 197 inventions were ache subj ect of 502 nonexclusive licenses, and 21 held exclusive licenses. Only 54 of the 3, 512 NASA-patented inventions had been commercia3.ized by the end of 1973. This relatively tow rate of t. 5 percent can be explained by a variety of factors. First, NASA patents only a few of ache inventions reported each year. Commercial potential is only one o f many criteria used ~co select the inventions that get patents . Second, the granting of nonexclus ive licenses discouraged many potential license applicants. rnird, many uses of these inventions in the commercial sector may be undocumented and unreported by the formal system. Over the years, NASA has not gone to court to protect its domestic patents from unauthorized domestic use. ~ Therefore, rather than focus on the tow rate of commercialization of licenses when compared to all patents, i~c is more appropriate to compare the 54 commercialized inventions to ache total of t97 inventions that were attractive enough for bus iness to want to develop into commercial products. Using this ratio, the commercialization rate is just over 27 percent for nonexclusive patent licenses. Inf orn~a~c i on ~ is s eminat i on The Technology U~cilization Office publishes the Tech Brief Journal, a quarterly volume containing short statuaries of NASA- sponsored technologies and directing the reader to a contact for more information. This data bank not only includes new inven~cions, it al so l is ts manna l s and information about newly discovered processes . The Tech Brief Journal has been distributed without cost by the government. The Office of Management and Budget had placed a distribution ceiling of 75, OOO copies on the ; ournal . As of January 1985, Chat ceiling was lifted and the subscription rate Disc is still free, but no longer printed at government expense ~ began growing - 279 -

immediately By August 198S, it was close to 100, 000 copies- -a 33 percent jump in less than ~ year (see Ault34~. Lois is indicative of the demand for such info~u~at' on by businesses . The Denver Research Institute (DRI), 3S under contract to NASA, has cad lected and analyzed data about fires that request Technical Support Packages (TSP' s), which include de~cailed documenta~cion of the innovations summarized in ache Tech Brief Journa1. A random sample of 358 requesters of TSP's in 1978 was taken. The requesters included all sizes of fi `~=s in all ma] or industrial sectors . Slightly less than half of the users reported benefits from the TSP request. The analysis showed that firms In several industries, including electronic components, communications, utilities, and business services, reported high rates of use of the information and more successes than failures in applying the information to bus iness situations . Of those reporting benefits from the TSP' a, 61 percent attributed the benefits to having informa~cton about new processes or products sooner than they otherwise could have acquired ito Only one loser reported increased revenues, but 14. percent reported cost savings. Eighty- seven percent of Ache users that reported corporate benefits from the information received found that Chose benefits were measurable within s ix mor.~chs, and ~ for 98 percent, within one year c Firms are more likely ~ca digest and use information Chat relates ~co their production process technology more quickly than information rela~ced to developing and marketing new products. The ORI data support the hypo~chesis ~chat information programs such as the Tech Brief Journal and follow-up TSP' s are extremely useful to bus inesses in improving produc~ci~rity and enhancing technological deve lopment in production processes . Measuring and evaluating the ir benefits deco the economy directly is virtually impossible. But, from ache rapid rise in demand for the Tech Brief Journal and the data on information use from DRI, it is clear that the information is used by industry and has a positive impact on the economy. )ther Economic Impact S tudies stably missing from the above discussion are the studies that ocument the "b is" technologies Chat are a direct or indirect result f the space R&~) program. lathe most obvious examples are satellite ommunicat~ ons, weather and remote sensing satellites, private space Bunch vehicles, the commerciatizati;on of space through private use the shuttle, and such new materials as carbon/graphite composites. detailed discuss ion of ache specific impacts of these technologies , beyond the scope of this paper. Most of such analyses' a few of rich are mentioned below, stand alone and do not lend themselves to ;gregate measures. - 280 —

Summarized briefly, the telecommunications industry has been altered radically as a result of satellite communications technology. Long distance communications have become less expens ive, and new services are being offered. Expressed in constant 1982 dollars, the charge for the yes tar C-Band satellite in 1972 was $650, 000 per transponder per year. The equivalent charge for ~ transponder on ache GTE Spacenet satellite in 1982 (with both C°Band lad Ku-Band a~raitable) was $280, 000 per transponder per year. Looked at another way, since 1965 the cost of living index has trebled while the unit service charge on ache Intelsat satellites has fallen 90 percent. This remarkable reduction in cost reflects ~ number of factors, including more efficient systems that have resulted from new technologies, and larger satellites that spread fixed costs, such as launch operations, over more transponders. Another, most likely temporary, factor in the lower costs is the current oversupply of sate Alit communications capacity. Not only has the supply side of communications changed but improved products and services have increased the demand for communications satellites. Between 1966 and 1982, the global traffic, expressed in half-circuits grew from virtually zero to over 51,000. Associated with this direct inves~en~c in space hardware has been the development of a rapid and competitive market for earth receiving stations, including antennae, sophisticated electronic equipment, and support services. Although it is likely Chat the macroeconomic and interindustry analyses recorded some of tints growth, it is unlikely that they coul d have accounted fully for its impact. Not only has it changed the industrial struc~cure of ache industry and promoted competition where formerly there was none, but it also has created difficult pa litical and social problems. For ins~cance, direct broadcast television, which is capable of broadcasting over wide reception areas, will create sens incite issues for small countries whose governments j esiously control scheduling and coneen~c of programs within their borders. The same technology that permits coverage of large areas will enable education and information to be made available ~co remove populations previously inaccessible to T,t broadcasts. Ano ther examp le of an economic analys is o f a particular NASA R&I) program was a series of studies conducted by Econ, Inc.9 for NASA during the mid-1970' s that attempted to measure the impact al; remote sensing satelli~ces on agriculture (see Bradford and Kelejian 7~. The studies measured the gains from having better and more generally available knowledge of current farm crop production, particularly of wheat. By s~cudying ache historical price fluctua~cions of the crop and by analyzing the potential dampening of those fluctuations due deco less uncertain~cy about near- term total crop production, the benefits of decreased speculation in ache futures otarice~c, and, therefore, more stable prices, could be assessed. The results did show ache expected trends, but were controversial because of the study' s - 281 -

assumptions concerning the industrial structure of the agricultural sector. - Specific applications of the remote sensing satellites led to better information about the spread of crop blight, snow melt, land use, pollution monitoring, and other data previously unobtainable or very expensive. In economics, calculating the value of information is a difficult and not weil-developed part of theory. These attempts were exploratory. Although case studies revealed significant benefits, some characteristics of the satellite system itself, such as the inability to returns pic~cures on cloudy days, malce it difficult to predict exactly when and what information would be available, and it is, therefore, difficult to predict future benefits and their value . The Bust can be extended. Meteorology satellites, search and rescue instruments on satellites, navigation aids, and other new space - related technologies have .been developed into both commercial products and government capabilities echoic affect health, safety, and the economy. Specific analyses have measured the impacts of these and other products. But ache methodologies have varied so greatly chat it is impossible Deco aggregate the results. Over the years, NASA has conducted numerous cost-effectiveness and engineering studies oriented to detailed dectsionma~ing. Although these may add to our information about government programs, they tend not to fit into the productivity/impact analyses that this paper is reviewing. Similarly, the various regional and short-term economic multiplier analyses that have been conducted add little to our information about the major tong-te~ benefits from space R&D that translate eventually into improved products, improved manufacturing processes, and impacted productivity. CURRENT TRENDS IN "JUSTIFYING" SPACE RESEARCH AND DEVELOPMENT Previous sections of this paper have focused on the economic studies of space R&D during the mid-1970's, which was the most active time for conduc~cing such analyses. Earlier, NASA had a large budget due to the Apollo mission. As that mission ended, and ache shuttle became ache only major project (with a much more spread out and leered budge~c allocation), funding for NASA fell. Economic pressure deco allocate funds more efficiently among projects and to justify new R&D programs was felt for ache first time in NASA history. Congress and the Office of Management and Budget asked pointed questions about what economic returns could be expected from additional expenditures. In an effort to answer those questions, the aggregate impact studies were conducted. - 282 -

We space progr=~:n does not exist outs ide of the political realm . .~£any of the economic analyses reflected the political expediency of showing Ugood" returns. In some respects, they performed their function. Right or wrong, some still associate the figures " 14 Deco 1 returns n and "43 percent per year" with NASA expenditures . But, most now recognize that these figures do not represent the true ~ralue-- both economic and noneconomic ~ -of the expenditures . A factor overlooked often is that there is no a priori reason for any government mission-orien~ced project to show a positive economic return on investment. If "he mission was successful in its own right (for example, landing on the moon, Caking pictures of Mars, epic. ), then that should be enough . Further, us ing government national income accounting methodology, all government R&D expenditures are valued at the time they are spent. They are not considered inures tments wi th ~ future re turn. Then, it should be no surprise that most analyses that compare government returns to R&D expenditures with priorate returns to R&D in~res~cments show that the priorate returns are much higher. I ~ would be surpris ing if that were not true . The macroeconomic impact studies of R&I) have always measured the his Conical average returns to the economy. For policy purposes, ache Ideal measure is the marginal recurs--what the next dollar spent will yield in the future. So far' economists have been unable to isolate the marginal return from space R&D. SPACE AS AN INDUSTRY There is no such aching as ~ space "industry. ~ On ache supply side, there is, of course, ache SIC 376 (guided missiles and space equipment) Chat is dedicated to manufacturing hardware for space. But, a large amount of space work is done in other sectors. Communications equipment, e iectronic components, ins truments, business services (software), universi~cies and nonprofit organizations, and professional services (engineering and R&D labs) all make significant contributions to space technology. Unfortunately, for measurement purposes, the tomcat quantity of space - refaced output from these industries is small in comparison Deco ache total sales in these sectors. Communications equipment, for eX?'nple' includes the production of radio and TV receivers, C8 sets, and various other electronic components. Satellites may account for only a few percentage points of tomcat sales. Therefore, it is very difficult deco iso] ace apace components in the aggregate industrial figures that are available. The demand for space technology is felt throughout ache economy. As shown by various macroeconomic analyses, ache direct and indirect uses of the products of space technology have an impact on the quality and productive ty of almost all sectors of the economy. - 283 -

Mos ~ demand for space equipment comes from the go~rernmen~c . In the United States, Actual space R&D costs approximately $1S billion per year, with military expenditures accounting for slightly more than civtItan expenditures. Allis does not include the mill tary procurement of guided miss)] es and space equipment, which is a ~ arge stem. ~ Private expenditures for R&D and space equipment (satellites, services, etc. ~ are estimated to be $2 billion to S3 billion annually. Including foreign demand for similar equipments the total domestic spending on space is approximately $20 billion annually in a U. S . economy that has ~ GRIP of close to $4 ~crillion. lotus, space spending amounts deco just over one half of one percent of the total output of the U. S . economy. Today, it is so small and diffused that it really cannot be class if fed as a separate incus try . TRENDS IN THE ECONOMIC S lRUCTURE OF SPACE ACTlYIT7ES lye past few years blare been ~ period of limited transi~cion and adjustment for spare acti~ri~cies . Until ache 1930' s, the government, for all practical purposes, had sole control over space investment' research, hardware, and facilities. Now, private firms are emerging and Caking a place alongside the government as both suppliers and users of space. Over 350 firms are involved in space business; some are small and entrepreneurial in spirit and some are divisions of the giant aerospace companies. Although, theoretically, it now is possible to put a payload into space through private companies without using any government facilities, there stir, are a number of government regulations that must be followed, as well as various licenses that have to be- obtained In reati~cy, however, i~c is, and for the near future wail continue to be., almost impose ible to get into space without the go~rernment's d~rec~c in~rolvemen~c. Cat is not recognized clearly by many is that any private firm will be in a partnership of some form with the government and dependent upon government infrastructure to launch, operate, and maintain facilities in space. While things are changing slowly, ache government still is the dominant supp lier of space equipment, facilities ., and is~frastn~cture. And, the government is the principal user of space for research, scientific exploration, defense, and provision of such public services as meteorological data. Two current economic trends in the space industry are often confused. The first is the transfer of current government acti~ri~cies to priorate ownership. This is called privatization. Lee second is the development, through a partnership of government and industrial R&D, of new technologies that are expected to lead to marlcetabie products. This is called conunerciaQ ization. - 284 —

Chile commercialization represents po~cen~cial real growth to the economy through maj or technological breakthroughs, privatization s bands mainly for a dis tr~butional change be Been government and industry and does not necessarily imply that new industries or products will be forthcoming . However, in the near term, privatization does mean a s ignif~can~ shift toward economic market orienta~cion from ~craditionalt y governmen~cal functions. To the ex~cen~c that products are developed and sold in addition to the expected government uses and purchases, then some economic growth could occur from pr:vatiza~cion. For instance, Ache Lands remote sensing satellite system has recently been transferred from the National Oceanic and Atmospheric Administration (NOAA) deco a private company, Eosat. A large government subsidy is involved over the next seven years, so that an evaluation of the true profit position of the company is many years away. Eosat will sell back deco the government the same products that now are produced by the government. To the extent that the company can (~) reduce the cost of operations9 (2) develop new products for the government, (3) open new priorate markets, and (4) develop new products that nearer would have been developed by government managers, the privatization of the Lands at system will reap long-~cerm economic benefits . Given that the primary maricet:s for Lands at products are governmental, that new foreign competition is developing ~ for example, French and Japanese satellite remote sensing systems ), and chat there is a demonstrated need for additional R&~:) in sensing systems, the probability ~chat the privatiza~cion of the Landsat system will yield economic benefits is questioned by many. If Eosat is not profitable, will the U. S . go~rerr~ent be prepared to let ache company go out of business? Continui~cy of data is impor~can~c for the many government uses of the system, and it is likely that some form of bailout would occur through either additional subsidies or takeover of the system. Whether this is true privatization and whether net economic benefits will accrue are questionable presumptions. S imilar problems exist in ocher pri~ra~cization attempts . The recent stimulation by the U.S. government of private launch companies is an example. A more complete description of the growing international competition for launch vehicles is provided in a separate paper by the author (see Her~czfeld ). The major conclusion of that paper is that the ~competition" in the launch vehicle industry is between governments, not between private parties. With ache exception of certain Marx niches, such as the launch of communications sa~celli~ces to geosynchronous orbit, the commercial side of ache business is limi~ced. And, even where new launchers have been built, as with the European Space Agency' s Ariane rocke C, a large government subs idy was provided . - 285 —

The importance and influence of political considerations on economic decisions cannot be underestimated. Several years ago, the weather forecasting satellite system operated by NOM was suggested as a good system for the government to sell to the private sector. The proposal nearer got very far because the groups that would have been affected, both within ache government and outside, protested loudly . Commercial iza~cion of space has the potential for true economic growth and benefits. But, the private sector's success in the use of the space environment for manufacturing cannot be a reality until the next century, if ever. A few near-~cerm successful proj ects are possible, but for any real benefits, given the enormous up- front capital investments required for space industry al R&D, it will take many false stares, many new proj ects, and a working partnership be tween go~rernmen~c and incus try . (red the basic decade, NASA has begun the commercialization process by stimulating and encouraging industry to lice NASA infrastructure and facilities to develop R&D projects. Such programs as the j oint endeavor agreements will give private companies space on the shuttle and the use of certain NASA facilities and services free of charge if they show that they can develop promising R&D projects and if they commit resources to the project. Each agreement is negotiated individually. The government's objectives are to give title to the companies for any inventions discovered and to find ways to encourage the commercial use of the fruits of space research conducted by industry. The doors have been open in the government for over six years, but, as of September 1985, only eight joint endeavor agreements had been signed (see Yadvish ). Fourteen other research and technical exchange agreements witch priorate companies interested in space from the commercial potential have been negotiated. This is a modest and slow beginning. Space is a new frontier and a risky investment, however, and it whit cake time before many companies are willing deco risk the time and expense of doing research in space. The first and most discussed NASA/industry joint endeavor is the McDonnell Douglas/Johnson and Johnson electrophoresis research on new pharmaceuticals (signed in 1980) . If the proj ect is analyzed carefully, ~ picture far different from the newspaper depictions emerges . First, it is not a new developmen~c. More than eight years ago, NASA funded McDonnell Douglas under a contract from the Marshall Space Flight Cen~cer deco explore the feasibility of creating new pharmaceuticals in space. This indicates that the idea of manufacturing drugs in space existed years before the agreement, and it also means that even though private industry has interested a good deal of its own money into research and hardware for the current experiments, an least some of ache initial stimulus came from the government . - 986

Second, as with all research proj ects, it has had its technical problems as well as it successes. Scaling up deco full production from research machines often is quite troublesome- - even with traditional incus tries . Third, the proper infrastructure must be in place in space: power, reliable ~cransportation, repair facilities, etc. At present, they are not adequate. It is interesting deco note that last year (1985), Johnson and Johnson pulled out of the projec~c. The reasons have not been disclosed full y. Currently, McConnell Douglas is searching for another partner while continuing its research efforts under the NASA agreement . Will this effort be a failure? I think not. But, traditional business tools such as discounted cash flows and return on investment app lied to tile costs in research and subsequent production, distribution, and eventual profits will not show this to be a prudent business ~ren~cure with a rate of return equal to or greater than other opportunities for corporate investments . This research proj act will be cons idered a success because of other factors, including good publicity and advertising, on- the-j ob training and learning of cutting- edge technology for employees, and the potential for improving terrestrial pharmaceutical production processes. In fact, out o f this effort, the electrophores is process machinery available for terrestrial production has been improved significantly ~ see Pake4 ~ . McDonnell. Douglas claims that earth-based production using this process is mired, and space experiments have improved efficiency over 700 percent.4 If valuable drugs result from space research, there will be a big incentive to find a terres~crial method of producing Chose drugs. Eliminating ache risk and cost involved in transportation to and from space would j ustify added resources for improving t~rres~crial equipment. To recoup the expenses of more than eight years of research, ache hardware cos ts for space manufacturing, and the ongo ing operational costs of a space processing facility would require very large returns on not just one drug, but on a portfolio of at East 10 to 20 drugs. I~ will take years jute to get Food and Drug Administra~cion and other approvals on drugs that are not yet produced even in sufficient quantity to be tested. It will Cake even more time to advertise the drugs, to have doctors prescribe them, and to have them reach a large share of the potential market. There is also a risk that new gene~c~c engineered compounds will be developed to compete with ache space drugs. And, there is the obvious problem of developing a working, reliable space production facility with enough power and sufficiently cheap transportation to operate it. _ 987 —

There are several demand- related points that are important for space commercialization. The future is uncer~cain, the investment large, and the risks high. Added to all the normal marketing uncer Dainties that any new product faces are the unknowns of space and the risks that an accident will make ache entire opera~cion . useless. Space hardware rarely can be reco~rered--there is little salvage Prague. In addition, there is no reason to believe that space research as performed currently will follow a path significantly different from that of past space research. Economic benefits and returns have been primarily from spinoff technologies--chose ~cha~c are indirect and are the result of learning unexpected things from research rather than the direct goals of the research itself. Spinoffs are unpredictable. Such research undoubtedly will make valuable contributions to society, but they may be in areas we cannot foresee todla3t. A related point should be considered: Had it not been for space research, would the many technological advances that can be traced to the space program have occurred? And9 if they did occur, has the space research effort accelerated the p - ocess ? Space activity should be viewed as a stimulus for new R&I) effor, s. Any commercial products that materialize in the future will be valuable to industry, but cannot and should not be measured as a benefit" today. In summary, even with. all the talk about space commercialization, only one space industry- - satellite communications ° -has been profitable. And tics exponential growth during the 1970' s is ferreting off, due mainly to ache growth of fiber optic cables, the launching of larger than needed satellites ? and the s lower than expected development of direct broadcast television. Most of the other commercialization proj ects are in the early R&D stages . History indicates that it will take an average of 20 years before canister products from these R&D efforts are ready for the mass market. Nevertheless, space research and developments programs will yield returns in the for of spinoff benefits. Exactly what they will be and how much they will contribute is ~known, but the fact that it will happen is predictab le . RECENT CHANGES IN ECONOMIC EVALUATIONS OF SPACE ACTIVITY As described above, space R&D is as much a political activity as an economic activity. Civilian space programs have been in existence for nearly 30 years. They are maturing, and commercial opportunities are developing as business and industry learn and par~cicipate. Recent political trends are away from government ownership and toward entrepreneurship . lye President' s Space Policy4 underscores -this trend by promoting policies to stimulate both privatization and commercialization of space. - 288 -

The economic Gnat yses of space activities have shifted from j us tifying addi~cior.al government funding for NASA programs . Now, bus iness and financial analyses chat demonstrate that companies can make a profit in space, open renew marketing frontiers, and s~cimula~ce employment and income in the United S Gates are in vogue . The economic j ustification for continued government support for such civilian space activities as the space station is centered around ache potential for manufacturing and Ache support of commercial uses of space facilities. (Of course, now, as in the panic, the economic reasons are only one part of the justification process, which also includes scientific, engineering, exploratory, and technological leadership, plus inte~`a~cional competition, as reasons. The business and financial analyses are no better or worse from technical viewpoint than the macroeconomic and benefit/cost studies used previously for sho~ir.g potential retuners from space R&D. The business models have their own limitations, due primarily to ache short- tens focus of their ass~Tap~cions . Discounted cash flows are ex~cremely sensitive to heavy up-front capital in~res~cment, interest races, and immediate marice~c sales. Space R&D involves ail of ttse negatives for this analysis--1 ong time frames, expensive equipment, and a high degree of risk. It is difficult to j ustify longer term investments using these models because they are geared to marketing efforts more than to R&D efforts. They can he adjusted through a variety of mechanisms ( see Mechlin and Berg4 and Hodder and Riggs" ). But7 by and large' these models will not show space Rid to be a good inns tmen~c for corporations when compared to availab be alternatives . WHAT HAS BEEN LEARNED ABOUT: ME ECONOMIC IMPACTS OF SPACE R&D ANI) ABOUT MEASURING ME IMPACTS CHARACTERISTICS OF SPAC}: RESEARCH THAT CAN BE MEASURED Every age and time has its imaginative and awesome technological achievements . Landing ~ man on the moon, ache mos ~ mind capturing 0 f all of our space accomplishments, ranks eas fly with the building of the Gothic cathedrals of the Middle Ages, ache Great Wall of China, and the ~cranscontinental railroads of the 19th cen~cury. In their time, each of the pro] ecus was without precedent . The investment capital had deco come from a central governmental (or religious ) source with a surplus of funds. New technology had to be developed and perfected. And, party as a result of their investments, the nations that were able to support those activities maintained and increased their po li~cical and economic leadership . All of the great eng' peering proj ects had maj or economic impac ts: The money to build the structures was diverted from other uses; the wages paid to those employed by the proj acts had direct and - 289 -

indirect income and employment effects; and the learning and new technological capabilities enhanced crude and national prestige. But, like ehe space program, most of those great feats had to amass their support from ~ coalition of political, social, and economic institutions . Economic impact arguments alone foul d not have been enough of a stimulus to see them through to completion. Even proj ects like the railroads had social and defense obj ectives (opening the territory of the rest and being able to transport: supplies to troops in remote sections of the country). Go~renmen~c subs idles built and kept such mat or proj ects going. Because of the subsidies and monopoly gran~cs, the operating companies realized profits. Without support from public sources, the time required to recover in~rest~nents would have been long, and the economic profits, if they materialized, would not have been realized quickly. Outer space is a national and international resource of great potential value. Rancher than concentrate on the economic impacts that are identifiable directly, one can view space as a national laboratory . Rarely does our sys rem question the val ue of having government research institutions devoted to specific types of research. The National Institutes of Health, ache National Bureau of S tandards, and the na~cional laboratories - ° Los Alamos, Brookhaven, Lawrence Li~'e~more, etc . - ~ are examples of public research facilities that have long- term co.~'mitmen~cs to continued large - scale research efforts . The space shuttle and ache future space station are similar deco these long- Berm research efforts . We tend deco justify them with projections of economic benefits that will emerge from curren~c and p fanned research . But, that is not the real reason to build scheme large- scale space facilities . lathe maj or reasons are to crevice 8 national resource and to preserve our technological leadership. The short-term political and social benefits may or may rot outweigh ache longer term economic returns. We do not and cannot know what those economic recut ns will be . The returns may well be in new products and improved processes, or the current space proj ects may create an infrastructure that will lead to other space pro] ects and improvements in our economic system no~c even thought of today. The railroads and interstate highway system created a new type of infrastructure. While they Made transpor~c of materials and products cheaper and faster, they also changed the shape of our ci~cies and, subse~sentl~y, our way of living. These were economic effects hardly dreamed. of by the governmen~c planners or by the industries and population involved at the onset of railroad and highway development. The nation can and should think about the ultimate effects of creating a new infrastructure in space, even though measuring the magnitude of the long- term economic impacts is nearly impose ible . About all tha~c can be said accurately is that the impacts will occur, and they will be sizable. — 290 -

S ~and=~<i economic models are devised to quantify what can be measured. That should be measured is entirely different: We need ~co. develop measures of the. value of a national resource. We should attempt to look at the political Prague of the space program, and it is extremely important to Ace the social impact how it has and will change society and our way of fitting. Now that space is a maturing sectors and there are opportunities for business to participate directly in research in space, there is an Pancreas ing need to measure this invo Element . In general, priorate activities in space are carry ed out in partnership with the government here is a need to look arc go~rernmen~c financial and institutional policies and measure the speed and effectiveness in opening space for commercial opportuni~cies. S tudies of the economic and political importance of international co~nge~cition in space R&D are inheres ring and potentially useful for formulating government policy. These issues raise sensitive questions, since many nations are proposing space research acttvi~cies and s ince the Uni Iced S tates and the US SR no longer share a monopo by on access to space. The true economic (as compared to political or pres tige ~ Prague of space commercial activity has never been measured accurately . Finally, a much better understanding of the government's rote as a s simulator of priorate P.&D is needed . lathe relationship of federal - space R&D as a complement to or substitute for private research is complex. Better economic measures could help to shape government R&D policy toward projects ~ ~t will both provide needed public goods and s ~ imniate further private investment leading to commercial success . ECONOMIC MEASURES OF SPACE R&D TO BE AVOIDED First, no economic study should attempt deco put a "bottom lines ratio or return on space R&D investments. There is no such number in existence- - it lives only in the uncharted world of general equilibria theory. All such namers that have been used as representative of a total return deco space R&D actually have measured partial returns . And, all such numbers are products of economic models witch many limiting ass~np~cions. Even when these assumptions and qualifications have been laid out carefully, ache existence of the number is an at~cractive bait to those politicians and others who need to ; ustify space R&D . Once a n total" returns number is used, it finds its way into misuse quickly. Second, studies Chat contrast the retie Its from government space R&D with ache returns from priorate it&l) should be avoided. There is no a priori reason why goverr~men~c it&l) must have a measurable GNP or producti~ri~cy return. For technical measurement reasons - - such as ache fact that government accoun~cing standards treat all government _ ~ 9 1 —

expenditures, include ing R&D, as current spending with no imputed investment returns - - the measured retunes to government Rho should. be smaller than those to private R&D. And, for the obvious reason that investment of federal funds for R&D is undertaken because of a national mission-orier~ted need not translatable directly into economic profits, the returns may not be picked up by standard economic measures. Therefore, it is of no surprise that comparative studies show returns to government R&D hovering around 5 percent and those of private R&D arc nearly 40 percent. Third, studies that have poorly defined ob: ectives are particularly susceptible to error when dealing with R&D . S ince R&D is a very general Ices =, encompassing many different activi~cies 9 the evaluation models frequently may not fit the questions being asked. Research activities produce knowledge, while development is aimed at useful end products . A process - related improvement may occur quickly and be hidden from direct measurement, while a new product Chat is eas ier to observe and measure nay take a long time to reach a marice and be diffused through the economy. Different economic models and data have to be used in each case. Or, as described above, analyzing go~rernmen~c R&D should be handled quite differently from analyzing returns to private R&~) ~ Frequently, those differences are no ~ understood, and the same model is used for all R&D measurements . Fourth, new studies should not repeat the mistakes of earlier analyses. Over the years, the sophistication of the economic analyses of space R&D has improved. Much is stilt unknown, and new methods, models' and data are being tested to measure returns to R&~. We can expect to learn more about the benefits and impacts of space R&D, but economists still are a long way from providing comprehens ire and accurate measures . CONCLIJS ION In the near term, economists have deco regain ache conf idence of go~rerT=ent and priorate pc,licymalcers by performing analyses that can be put deco practical use. The macroeconomic measures make some public relations efforts easier, but they do Tickle for ache decisios~maker. Fisst, a balance must be reached between economic and political objec~cives. Models and measures should be constructed to meet this balance, and the nonquantifiable returns should be recognized. Unders~andirtg the different economic and political objectives will begin to put space R&D into the proper perspective . S ince almost all space R&D is government supported, the traditional economic models built on the operating assumption of a freely competi~ci~re market must be modified. In current practice, this rarely is done. The only studies of the purely economic returns to space R&D are those that deal witch private business ventures in space and where — 292 —

profits are ~ he sole measure of success . However, even for a company, more may be involved than short- term profits, particularly in space R&D. The company may wish to keep its technical employees up with state-of- the-art scientific knowledge Deco improve other lines of bus iness . Or, it may seek ache publicity and advertising value of conductor research and business in space. Or. it may be entrepreneurial in spirit and willing deco persevere witch a research proj ece over a long period before profits are measured. In other words, economic analys is of R&D investments even in the "pure " case of a profit-oriented company may involve intangibles that to not fit easily into a standard economic measurement framework. Most often, business analyses of space R&D investments are held as proprietary by a company. It is unlikely that any aggregate impact or benefit figures will be available for general public evaluation. The most promising type of economic study for measuring returns to space R&D is ache documentation of ac~cual cases, based on surveys of ache agencies, companies, and users intoned directly with the space technologies. Such studies provide one way Chat specific economic questions can be formulated and answered with relative clarity. As a structured series of case studies is built, further theoretical work to integrate the findings and point toward aggregate models and measures should be encouraged. . 293 -

REFERENCES US S. Senate, Committee on Co~erce, Science, and Transportation. Space Law, Selecred Basic Documents. Second edition. Washington, DC: U. S. Government Printing Office, December 1978 . 2. Michael K. Evans. The Economic Impace of NASA R&D Spending. Prepared under NASA contract NASTY 27410 Philadelphia: Chase Econometric Associa~ces, Inc., April 1976. U. S . Congress, General Accounting Office. ~NASA Report May Overstate the Benefits of Research and Development Spending." Report of ache Comptroller General, PAD- 78 ° is, October 1977 . 4. Z~i Griliches. "Issues in Assessing ache Contribution of Research and Development to Productivity Growth, n Bell Journal of Economics, Voi. ~ O. No. I(Spring 1979), pp. 92-~16. Edward F. Denison. Accounting for Slower Economic Growth, The Uniced States in Che 1970's. Washington, OC: Brookings Institution ~ 1979 . Edwin Mansfisid. "Contribution of R&D to Economic Growth in the United States," Science, Sol. 17S, No. 4021~February 1972), p. 477. Economic Impact of Scimuiared Technological Activi ty . Prepared under NASA contract NASH- 2030 . Midwest Research Institute, November 1971. 8. Robert M. Solow. "Technical Change and the Aggregate Production Function, n Review of Economics and Statistics, Vol. 39(August i957), p. 312. Extracts, op. cit. 10. Edward F. Denison. Accouncing for United Stares Economic Growth , 1929 - 1969 . '~ashing~con, OC: Brookings Ins ti ~cu~cion, 1974. 11. U. S. Congress, Genera1 Accounting Office, op. cit. 12. David M. Cross. The Economic Impact of NASA R&D Spending, An Update . Prepared under NASA contract NASH- 3345 . Philade lphia: Chase Econometric Associates, March 1980. 13. Impact of Government and Private R&D Spending on Factor Produc rivi ty in Space ~f~uf ac luring. Prepared under NASA grant W- 14539 . Washington, DC: U. S . Deparemen.t of LLabor, Bureau of Labor S tatis tics, July 1980 . — 294 —

14 . F. P Fitussi . Economic Benefit us of ESA Contracts . Paris: European Space Agency, July 1978; 15. Goetz Niederau. Economic Benefits of USA Coneracts. Paris: European Space Agency, 1980. 16. Paul Kochanowski and Henry R. Hertzfeld. "Often Overlooked Factors in Measuring the Rare of Rescues. to Government R&D Expenditures,- Policy Analysis, Val. 7, No. 2(Spring 1981), pp. t53-16?. . Richard Nelsort. "Research on Productivity Growth and Productivity Differences: Dead Ends and New Departures, ~ JourI:a] of Economic Literature, Vol. XIX(September 1981), pp. 1029 - 1064. 18. Richard Nelson and Sidney Winter. "The Sch~s~peterian .Tradeaff Re~isi~c~d," American Economic Review,.Vol. 72, No t(March 1982), pp . 114- 132 ~ 19. Klaus P. Heiss and Oskar Morgenstern. Economic Analysis of New Space Transporta~cion Systems. Prepared under NASA contract NASW- 2081. Princeton: Hathematica, Inc ., May 1971. 20. Klaus P. Heiss and Oscar Morgenstern. Economic Analysis of cbe Space SE,u~2e System. Prepared under NASA contract NASW- 2081. Princeton: Ma~chematica, Inc., January 1972. 21. Quantifying the Benefits to ~c]2e National E:conomy from Secondary Applications of MESA Technology. NASA CR-2674. Princeton.: Mathematica Inc., March 1976. 22. A Cose-Benef:e Analysis of Selected Technology Utilization Off ice Programs . Prepared under NASA contract NASH- 2?31 . Prince ~on: Math~cech, Inc ., November 1977 . 23. Henry R. Her~tzfeld. "The Impact of NASA Research and Development Expenditures on Technology Innovation and the Economy. ~ In Proceedings of the International Coi loqulum on the Economic Effects of Space and Advanced Technologies. ESA SP-151. Paris: European Space Agency, September 198Q . 24. Griliches, op. cite. 2S. Henry R. Hertzfe3~d. "NASA Patent Waivers and Licenses. Unpublished manuscript, January 1980. . D. Watson and M. Holman. An Evaluation of the Pa~cen~c Policies of ache National Aeronautics and Space Adminiscrarion. Washington, DC: George Washington University, 1966. - 295 —

27 . Robert Solo . "Patent Policy f or Government Sponsored Research and Development, n IDEA, Sol. 1041966), pp. 143-199. 28 . Government Parent Policy Scudy. S tudy conducted for the Committee on Goversmen~c Patent Policy of the Federal Council on Science and Technology. Boston: Harbridge House, May i968. 29. Nicholas S. Kaskovich. The ~Yoni~coring of NASA Patent Licensing . and Waiver Act~vi~cy--lmp~ications for Experimental Design. Evanston, ILL Northwestern University, September 1974. 30. Philip Wright. A F'oliow-up Study of Selected NASA Patent 'waiver Cases. Philip Wright Associates, September 1974. 31. Hertzfeld, January 1980, op. cit. 32. Fiscal Year Statistics of the NASA Parent Program', 1959-1984. Washington, DC: National Aeronautics and Space Administration, Office of the General Counsel, 1985. 33 Hertzfeld, January 1930, op. cit. 34. Personal co. l''~unication from Leonard Aunt, Deputy Director, Technology Utilization Minis ion, National Aeronautics and Space Administration, September 198S. 35. Denver Research Institute. "Survey of users of Technical Suppor. Packages, 1978." Unpublished data, Hay i979. 36. Trade ~ Righ-TechnoJogy Products, The Space Products Infuse: Markers , Industrial S~cructure and Government Pol icies . Paris: Organization for Economic Cooperation and Deve iopment, Directorate for Science, Technology, and Industry, March 1985 . 37. David F. Bradford and Harry H. Keynesian. The ttaJue of Info rmacion for Crop Forecasting with Bayes~an Speculators: Theory and Empirical Results. Prepared under NASA contract NASH- 2S58 . Princeton: Econ, Inc ., 1974 . 38. Henry R. Hertzfeld. haunch Vehicles sad the Commercial Uses of Outer Space, n Space Policy, Sol. I, No. 4~November 1985) ~ pp. 379 - 389 . - 39. Personal comnrunica~cion from John Yiddish, Chief, Financial Analysis Branch, Office of Commercial Programs, National Aeronautics and Space Administration, October 1985. 40. Fe~sibiiity of Cor~ercia] Space Manufacturing, Production of Pharmaceu~cica~s. St. Louis: McDonnell Douglas Corporation, November 1973. — 296 —

o 41. George Pake. "Experimenting in Space,' 'Washington University Magazine, ( Spring 1984), pp . 24- 25 . 42 . El eccrophores~s Operations in Space . Douglas Corporation, March 19 84 . S ~ . Louis: McI)onnel1 43 . National Space Pol icy. Washington f DC: Me White House, Executive Office of the President, July 1982. 44 . George Mechlin and Daniel Berg . n Evaluating Research- -ROI is No Enough, n Harvard Busyness Review, Vol . 80, No. 5(September/ October 19 80 ), pp . 9 3 - 99 . 45 . J ~ E . Hodder and H . E ~ Riggs ~ " Pitfalls in Evaluating Risky Projec~cs, " Rarvard Business Review' Sol . 85, No. [(January/ February 1985 ), pp . 128 -135 ~ 297 -

ADDITIONAL BIBLIOGRAPHY Robert 0. Bartlett. "A Survey of Economic Factors of Research and Developmen~c: National Productivity Impact of NASA Expenditures. " Unpublished manuscript, George Washington University, Spring 1981. Edward F. Denis on. "Explanations of Declining Produc~ci~rity Growth, Survey of Current Business, Vo: . S9, No . 8 (August 1979), pp . l-24. Denver Research Institute. Tipsier Oriented R&D and the Advancement of Technology: The Impact of LANA Coneriburions. Prepared for NASA. Denver: Denver Research Institute, May 1972. Samuel Doctors. The Role of Pedera] Agencies in Technology Transfer. Cambridge, MA: MIT Press, 1969. L. 8. Early, C. J. Reber, and PO M. Caughran. "Economics of Communications Satellite Systems--1976 ~ Acta AsCror.aurica, Sol. 5, No. 3-4(March 1978), pp. 261~273. Earth Satellite Corporation and Abe Associates, Inc. A Slurp co Examine the Mechanism-c to Carry Out the Transfer of Civil Land Remote Sensing Systems co the Private Sector. Prepared under NOAA contract NAB 83 - SAC-00679 . Bethesda ~ ED: EarthSat, March 1983. Ebony Inc . The Effects of Research and Developmene on the Uni ted States Teiecomawnicatzons Industry: An Econometric Analysis. Princeton: Econ, Inc., May 1977. Joel S. Greenberg. Co~ercia~izarion. of Che Land RemoCe Sensing System: An Examination of Mechanisms and Issues. Prepared under NOAA contract NA- 83 - SAC-00658 . Princeton: Econ, Inc ., April 1983. George A. Hazeirigg. The Deve~opmene arid CommercializaCion of Communication Sate] li He Technology by Che Uni ted States . Prince~con: Econ, Inc., 1982. Henry R. Hertzfeld. Measuring NASA'S Impact on the Economy. Speech before ache Technology Transfer Society, June 1981. F. Douglas Johnson. NASA Tech Brief Program: A Cost-Benefit EvaJuar~on. Prepared under NASA contract NASW-2892. Denver: Denver Research Institute, May 1977. - 298 —

. Douglas ~ Johnson and Martin Kokus. it Spry of Cost-Benefit Studies, .VASA Technology Utilization Progr=,n. Prepared under MESA contract NASW- 3021. Denver: Denver Research Insti~cu~ce, December 19 77 . Joseph I^bow, Chief, Waiver Branch, NASA Inventions and Contributions Board, Oc sober 19 8 5, personal communication . John M. Logsdon. "NASA at a Crossroads: One Obser~er's Perspective. n Speech presented at George Washington University, July 1983. John M. Logsdon. "U. S. Initiatives in Space Commercialization. n Speech presented at the International Astronomical Federation Congress, October 1984. . W . Lorbe' and R . H . Drake . The Economic Benefi es of Spa~- Developmene. IA- 9297 -MS . Los Alamos National Laboratory, March 1982. McDonnell Douglas Astronautics Company East. Feasibility Study of Co~ercia] Space Manufac~curing. Prepared under NASA contract NAS 8- 31353 . St. Louis: McDonnell Douglas Corporation, January 1977 . National Academy of Public Administration. Encouraging Business Ven cures in Space . Prepared under contract to NASA. Washington, DC: National Academy of Public Administration, May 1983 . National Aeronautics and Space Administration. A Cost-B'enef i Eval cation of the I-andsar Fo2 Jow-on Operational System. Greenbett, f1D: National Aeronautics and Space Administration, Goddard Space Flight Center, March 1977. National Aeronautics and Space Adntinistra~on. NAUSEA Ann up] Procurement Report . Uashing~con, DC: National Aeror~au~cics and Space Administration, Office of Procurement, 1961 through 1984, · ~ - 1nc. "us eve . National Aeronautics and Space Adminis tration . NASA Commercial Space Policy. Washington, DC: Na~cional Aeronautics and Space Administration, October 1984. National Aeronautics ant Space Adminis "ration. A Surveyor of Space Applications. Washington, DC: National Aeronautics and Space Adminis~cration, Office of Technology Ueiliza~cion, April 1967. National Aeronautics and Space Administration, Office of External Relations . Spinoff . Washington, DC: U. S . Government Printing Office, July 1984. - 299 —

National Science Foundation. Science Ind ~ caters 1982 . Washington, DC: U. S . Government Printing Office, 1983 . . Lloyd Oar and David Jones . An Induserza] Breakdown of NASA £xpen d i Cures . Prepared under NASA grant NS2 15 - 003 - 067 . Bloomington, IN: Indiana University, November i969. Rockwell International. Space Industrialization. Prepared under NASA contract NASS- 32198 ~ Downey, CA: Rockwell International Space Di~risior~, April 1978. Rockwell International. The Space Shu~ceJe Program, Nationwide Invo2veruent. Downey, CA: Rockwell International, October 1980. Science Applications Inc. Space Induscriaiiza~ion. Prepared under NASA contract NAS8-32197. Huntsville, AL: Science Applications Inc., April 19780 U. S. Congress, Office of Technology Assessment. Civilian Space Policy and Applications. OTA-STI-177. Washing~con, I)<:: U.SO Government Printing Office, June 1982. U. S. Congress, Office of Technology Assessment. International Cooperarion and Compete cion In Civilian Space Acrivi ~ ~ es . OTA° ISC. 239 . Washing~con, DC: U ~ S . Government Printing Office, July 1985. U. S . Congress, Subcommi~c~cee on Space Science and Applications . United Stares Civilian Space Progr=~: Applications Satires. Uashing~con, DC: U. S . Government Printing Office, May 1983. U. S . House of Represen~catives ,, Commi~ctee on Science and Astronautics . For the Benefit ~ of Al] Hankind, The Prace~cat" Returns from Space Investmen~c. Washington, 1)C: U. S . Go~rernmen~c Printing Office, October 1972. U.S. Senate, Committee on Aeronautical and Space Sciences. Hearing: Space Program Benefits. Washington, OC: U.S. Government Printing Office, April 1970. U. S . Senate, Committee on Commerce, Science, and Transportation. policy and begat Issues Involved in ebe Cor~mercia~izat~on of Space. Washington, DC: U. S . Government Printing Office, September 1983. U.S. Senate, Committee on Co~erce? Science, and Transpor~cation. Insurance and the Co~erciaiiza~cior2 of Space. Washington, DC: U. S . Go~rers~ment Printing Office, March "L985. _ 300 —

Next: Federal Funding of Research and Development in Transportation: The Case of Aviation »
Papers Commissioned for a Workshop on the Federal Role in Research and Development Get This Book
×
Buy Paperback | $95.00
MyNAP members save 10% online.
Login or Register to save!
  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!